Skip to main content

Full text of "1961-1962 Year Book - Carnegie Institution of Washington"

See other formats


-/?  •3>-rSrV(?  £&       


^~sy    *&V     J>y*.*  c*»*'<-    fZ?'^ 


CARNEGIE 
INSTITUTION 
OF  WASHINGTON 


Year  Book  61 


Digitized  by  the  Internet  Archive 

in  2012  with  funding  from 

LYRASIS  Members  and  Sloan  Foundation 


http://www.archive.org/details/yearbookcarne61196162carn 


CARNEGIE 
INSTITUTION 
OF  WASHINGTON 


Year  Book 


July  1,  1961  -  June  30,  1962 


61 


'"**'  ^^"^gaJOOawoooornm.  K^^rftSX'i^'^SSK^K^Z^^SSX 


Sixtieth  Anniversary 


Library  of  Congress  Catalog  Card  Number  3-16716 
Garamond  Press,  Baltimore,  Maryland 


Contents 


page 

Officers  and  Staff  v 

Report  of  the  President  1 

Reports  of  Departments  and  Special  Studies 

Mount  Wilson  and  Palomar  Observatories  3 

Geophysical  Laboratory  51 

Department  of  Terrestrial  Magnetism  209 

Committee  on  Image  Tubes  for  Telescopes  295 

Department  of  Plant  Biology  303 

Department  of  Embryology  367 

Department  of  Genetics  435 

Bibliography  477 

Administrative  Reports  479 

Report  of  the  Executive  Committee  481 

Report  of  Auditors  483 

Abstract  of  Minutes  of  the  Sixty-Fourth  Meeting  of  the 

Board  of  Trustees  497 

Articles  of  Incorporation  499 

By-Laws  of  the  Institution  503 

Index  507 


in 


President  and  Trustees 


PRESIDENT 

Caryl  P.  Haskins 


BOARD  OF  TRUSTEES 

Barklie  McKee  Henry 
Chairman 

Henry  S.  Morgan 
V  ice-Chairman 

Garrison  Norton 
Secretary 

Robert  Woods  Bliss1 
Amory  H.  Bradford 
Omar  N.  Bradley 
Vannevar  Bush 
Walter  S.  Gifford 
Carl  J.  Gilbert 
Crawford  H.  Greenewalt 
Caryl  P.  Haskins 
Barklie  McKee  Henry 
Alfred  L.  Loomis 
Robert  A.  Lovett 
Keith  S.  McHugh 
Margaret  Carnegie  Miller 
Henry  S.  Morgan 
Seeley  G.  Mudd 
William  I.  Myers 
Garrison  Norton 
Richard  S.  Perkins 
Elihu  Root,  Jr. 
William  W.  Rubey 
Henry  R.  Shepley 
Charles  P.  Taft 
Juan  T.  Trippe 
James  N.  White 
Robert  E.  Wilson 


i  Died  April  19,  1962. 


Trustees  continued 


AUDITING  COMMITTEE 


Keith  S.  McHugh,  Chairman 
Alfred  L.  Loomis 
Juan  T.  Trippe 


EXECUTIVE  COMMITTEE 


RETIREMENT  COMMITTEE 


Henry  S.  Morgan,  Chairman 
Amory  H.  Bradford 
Walter  S.  Gifford 
Caryl  P.  Haskins 
Barklie  McKee  Henry 
Robert  A.  Lovett 
Garrison  Norton 
James  N.  White 
Robert  E.  Wilson 


Omar  N.  Bradley,  Chairman 
Henry  S.  Morgan 
Garrison  Norton 
James  N.  White 


COMMITTEE  ON  ASTRONOMY 


FINANCE  COMMITTEE 


James  N.  White,  Chairman 
Walter  S.  Gifford 
Alfred  L.  Loomis 
Henry  S.  Morgan 
Richard  S.  Perkins 
Elihu  Root,  Jr. 


Seeley  G.  Mudd,  Chairman 
Amory  H.  Bradford 
Crawford  H.  Greene  wait 
Elihu  Root,  Jr. 


COMMITTEE  ON  BIOLOGICAL  SCIENCES 

Alfred  L.  Loomis,  Chairman 
Margaret  Carnegie  Miller 
William  I.  Myers 
Charles  P.  Taft 


NOMINATING  COMMITTEE 


Amory  H.  Bradford,  Chairman 
Barklie  McKee  Henry 
Richard  S.  Perkins 
Charles  P.  Taft 


COMMITTEE  ON  TERRESTRIAL  SCIENCES 

Juan  T.  Trippe,  Chairman 
Barklie  McKee  Henry 
Richard  S.  Perkins 
Robert  E.  Wilson 


VI 


Former  Presidents  and  Trustees 


PRESIDENTS 

Daniel  Coit  Gilman,  1902-1904  Robert  Simpson  Woodward,  1904-1920 

John  Campbell  Merriam,  President  1921-1938;  President  Emeritus  1939-1945 
Vannevar  Bush,  1939-1955 


TRUSTEES 

Alexander  Agassiz 

1904-05 

Henry  Cabot  Lodge 

1914-24 

George  J.  Baldwin 

1925-27 

Seth  Low 

1902-16 

Thomas  Barbour 

1934-46 

Wayne  MacVeagh 

1902-07 

James  F.  Bell 

1935-61 

Andrew  W.  Mellon 

1924-37 

John  S.  Billings 

1902-13 

Roswell  Miller 

1933-35 

Robert  Woods  Bliss 

1936-62 

Darius  O.  Mills 

1902-09 

Lindsay  Bradford 

1940-58 

S.  Weir  Mitchell 

1902-14 

Robert  S.  Brookings 

1910-29 

Andrew  J.  Montague 

1907-35 

John  L.  Cadwalader 

1903-14 

William  W.  Morrow 

1902-29 

William  W.  Campbell 

1929-38 

William  Church  Osborn 

1927-34 

John  J.  Carty 

1916-32 

James  Parmelee 

1917-31 

Whitefoord  R.  Cole 

1925-34 

Wm.  Barclay  Parsons 

1907-32 

Frederic  A.  Delano 

1927-49 

Stewart  Paton 

1916-42 

Cleveland  H.  Dodge 

1903-23 

George  W.  Pepper 

1914-19 

William  E.  Dodge 

1902-03 

John  J.  Pershing 

1930-43 

Charles  P.  Fenner 

1914-24 

Henning  W.  Prentis,  Jr. 

1942-59 

Homer  L.  Ferguson 

1927-52 

Henry  S.  Pritchett 

1906-36 

Simon  Flexner 

1910-14 

Gordon  S.  Rentschler 

1946-48 

W.  Cameron  Forbes 

1920-55 

David  Rockefeller 

1952-56 

James  Forrestal 

1948-49 

Elihu  Root 

1902-37 

William  N.  Frew 

1902-15 

Julius  Rosenwald 

1929-31 

Lyman  J.  Gage 

1902-12 

Martin  A.  Ryerson 

1908-28 

Cass  Gilbert 

1924-34 

Theobald  Smith 

1914-34 

Frederick  H.  Gillett 

1924-35 

John  C.  Spooner 

1902-07 

Daniel  C.  Gilman 

1902-08 

William  Benson  Storey 

1924-39 

John  Hay 

1902-05 

Richard  P.  Strong 

1934-48 

Myron  T.  Herrick 

1915-29 

William  H.  Taft 

1906-15 

Abram  S.  Hewitt 

1902-03 

William  S.  Thayer 

1929-32 

Henry  L.  Higginson 

1902-19 

James  W.  Wadsworth 

1932-52 

Ethan  A.  Hitchcock 

1902-09 

Charles  D.  Walcott 

1902-27 

Henry  Hitchcock 

1902-02 

Frederic  C.  Walcott 

1931-48 

Herbert  Hoover 

1920-49 

Henry  P.  Walcott 

1910-24 

William  Wirt  Howe 

1903-09 

Lewis  H.  Weed 

1935-52 

Charles  L.  Hutchinson 

1902-04 

William  H.  Welch 

1906-34 

Walter  A.  Jessup 

1938-44 

Andrew  D.  White 

1902-03 

Frank  B.  Jewett 

1933-49 

Edward  D.  White 

1902-03 

Samuel  P.  Langley 

1904-06 

Henry  White 

1913-27 

Ernest  0.  Lawrence 

1944-58 

George  W.  Wickersham 

1909-36 

Charles  A.  Lindbergh 

1934-39 

Robert  S.  Woodward 

1905-24 

William  Lindsay 

1902-09 

Carroll  D.  Wright 

1902-08 

Under  the  original  charter,  from  the  date  of  organization  until  April  28,  1904,  the  following  were 
ex  officio  members  of  the  Board  of  Trustees :  the  President  of  the  United  States,  the  President  of  the 
Senate,  the  Speaker  of  the  House  of  Representatives,  the  Secretary  of  the  Smithsonian  Institution, 
and  the  President  of  the  National  Academy  of  Sciences . 


Vil 


Staff 


MOUNT  WILSON  AND 
PALOMAR  OBSERVATORIES 


813  Santa  Barbara  Street 
Pasadena,  California 

Ira  S.  Bowen,  Director 

Horace  W.  Babcock,  Asst.  Director 

Halton  C.  Arp 

William  A.  Baum 

Armin  J.  Deutsch 

Olin  J.  Eggen 

Jesse  L.  Greenstein 

Robert  F.  Howard 

Robert  P.  Kraft 

Guido  Munch 

J.  Beverley  Oke 

Allan  R.  Sandage 

Maarten  Schmidt 

Otto  Struve 

Olin  C.  Wilson 

Fritz  Zwicky 


DEPARTMENT  OF 
TERRESTRIAL  MAGNETISM 


GEOPHYSICAL  LABORATORY 


2801  Upton  Street,  N.W. 
Washington  8,  D.  C. 


Philip  H.  Abelson,  Director 
Francis  R.  Boyd,  Jr. 
Felix  Chayes 
Sydney  P.  Clark,  Jr.1 
Gordon  L.  Davis 
Gabrielle  Donnay 
Joseph  L.  England 
Hugh  J.  Greenwood 
Thomas  C.  Hoering 
Gunnar  Kullerud 
Patrick  L.  Parker2 
J.  Frank  Schairer 
George  R.  Tilton 
Hatten  S.  Yoder,  Jr. 


52 U  Broad  Branch  Road,  N.W. 
Washington  15,  D.  C. 

Merle  A.  Tuve,  Director 
L.  Thomas  Aldrich 
Ellis  T.  Bolton 
Roy  J.  Britten 
Bernard  F.  Burke 
Dean  B.  Cowie 
John  W.  Firor3 
Scott  E.  Forbush 
W.  Kent  Ford,  Jr. 
Stanley  R.  Hart4 
Norman  P.  Heydenburg 
Brian  J.  McCarthy 
Richard  B.  Roberts 
T.  Jefferson  Smith5 
John  S.  Steinhart 
Georges  M.  Temmer 
Harry  W.  Wells6 


1  Resigned  June  30,  1962. 

2  Appointed  September  1,  1961. 
s  Through  September  15,  1961. 

4  From  September  1,  1961. 

5  From  June  1,  1962. 

6  On  leave  of  absence  to  serve  as  State 
Department  Scientific  Attache"  sta- 
tioned at  Rio  de  Janeiro,  Brazil, 
through  April  30,  1962. 


Vlll 


Staff  continued 


DEPARTMENT  OF  PLANT  BIOLOGY 

Stanford,  California 

C.  Stacy  French,  Director 
Jeanette  S.  Brown 
David  C.  Fork 
William  M.  Hiesey 
Harold  W.  Milner 
Malcolm  A.  Nobs 


DEPARTMENT  OF  EMBRYOLOGY 

115  West  University  Parkway 
Baltimore  10,  Maryland 

James  D.  Ebert,  Director 
David  W.  Bishop 
Bent  G.  Boving 
Robert  K.  Burns1 
Robert  L.  DeHaan 
Irwin  R.  Konigsberg 
Elizabeth  M.  Ramsey 
Mary  E.  Rawles 


DEPARTMENT  OF  GENETICS 


Cold  Spring  Harbor 
Long  Island,  New  York 

Berwind  P.  Kaufmann,  Director1 
Elizabeth  Burgi 
Helen  Gay 
Alfred  D.  Hershey 
Barbara  McClintock 
Margaret  R.  McDonald 


i  Retired  June  30,  1962. 


IX 


Staff  continued 


OFFICE  OF  ADMINISTRATION 

1530  P  Street,  N.W.,  Washington  5,  D.  C. 

Caryl  P.  Haskins  President 

Edward  A.  Ackerman  Executive  Officer 

Ruth  L.  McCollum  Assistant  to  the  President1 

Marjorie  H.  Walburn  Acting  Assistant  to  the  President2 

Ailene  J.  Bauer  Director  of  Publications 

Lucile  B.  Stryker  Editor 

James  W.  Boise  Bursar;  Secretary-Treasurer  Retirement  Trust 

Kenneth  R.  Henard  Assistant  Bursar;  Assistant  Treasurer  Retirement  Trust 

Donald  J.  Patton  Administrative  Associate 

James  F.  Sullivan  Assistant  to  the  Bursar 

Richard  F.  F.  Nichols  Executive  Secretary  to  the  Finance  Committee 

Marshall  Hornblower  Counsel 


Staff  Members  in  Special  Subject  Areas 

Tatiana  Proskouriakoff 
Anna  0.  Shepard 


1  Retired  June  30,  1962. 

2  Effective  from  May  21,  1962. 


Staff  continued 


RESEARCH  ASSOCIATES 

Carnegie  Research  Associates 

William  A.  Arnold 

Oak  Ridge  National  Laboratory 

J.  D.  McGee 

Imperial  College  of  Science  and  Technology,  University  of  London 

Jan  H.  Oort 

Leiden  Observatory,  The  Netherlands 

Paul  Ramdohr 

Heidelberg  University 

C.  E.  Tilley 

Cambridge  University 

Evelyn  M.  Witkin 

State  University  of  New  York 


Research  Associates  of  Carnegie  Institution  of  Washington 

Louis  B.  Flexner 

University  of  Pennsylvania 

John  H.  Holland 

Logic  of  Computers  Group,  University  of  Michigan 

Peter  Milner 

The  Department  of  Psychology,  McGill  University,  Montreal 

Harry  E.  D.  Pollock 

Carnegie  Institution  of  Washington 

Donald  L.  Richards 

The  Cooley  Electronics  Laboratory,  University  of  Michigan 


XI 


The  Report  of  the  President 


I  look  upon  the  Carnegie  Institution  as  the  most  interesting  effort  the  world 
has  known  for  the  development  of  a  national  interest  in  research. 

Henry  S.  Pritchett 

in  a  letter  to  Major  Henry  L.  Higginson,  May  1904 


Without  the  degree  of  liberty  which  culture  demands  even  a  perfect  society  will 
be  no  better  than  a  jungle.  For  this  reason  all  authentic  creation  is  a  gift  to 
the  future. 

Albert  Camus 

"Y Artiste  et  son  temps' ' 

Actuelles  II,  chroniques  1948-1953 


The  difference  is  infinitely  small  between  a  system  of  labour  which  leads  men 
to  discover  the  beauty  of  the  world  and  one  which  hides  it  from  them.  But  this 
infinitely  small  difference  is  real,  and  no  effort  of  the  imagination  can  bridge  it. 


Simone  Weil 

"Cette  guerre  est  une  guerre  de  religions'7 

Ecrits  de  Londres  et  dernieres  lettres 


THIS  YEAR  MARKS  THE  SIXTIETH  ANNIVERSARY  OF  THE  CARNEGIE 
Institution  of  Washington.  Sixty  years  ago,  in  1902,  Andrew  Carnegie 
transmitted  to  a  newly  elected  Board  of  Trustees  a  deed  of  trust  conveying 
the  sum  of  ten  million  dollars  "to  found,  in  the  city  of  Washington,  an 
Institution  which  with  the  cooperation  of  institutions  now  or  hereafter 
established,  there  or  elsewhere,  shall  in  the  broadest  and  most  liberal 
manner  encourage  investigation,  research,  and  discovery.  .  .  ."  At  the  end 
of  January  in  that  year,  the  Trustees  elected  Daniel  Coit  Gilman,  fresh 
from  the  career  for  which  he  was  already  noted  as  president  of  the  Johns 
Hopkins  University,  as  first  president  of  the  Carnegie  Institution,  and 
resolved  "to  promote  original  research  by  systematically  sustaining  projects 
of  broad  scope  that  may  lead  to  the  discovery  and  utilization  of  new  forces 
for  the  benefit  of  man  .  .  .  projects  of  minor  scope  that  may  fill  in  gaps 
of  knowledge  of  particular  things  or  restricted  fields  of  research  .  .  .  admin- 
istration of  a  definite  or  stated  research  under  a  single  direction  by  compe- 
tent individuals." 

It  was  not  the  first  of  Andrew  Carnegie's  great  philanthropic  gifts.  Far 
from  it  indeed.  In  the  last  decade  of  the  closing  century  in  Pittsburgh  he 
had  established  the  Carnegie  Institute  with  its  natural  history  museum,  its 
music  hall,  and  its  department  of  fine  arts,  and  had  made  possible  the 
Carnegie  Institute  of  Technology,  grown  now  to  front  rank  among  the 

3 


4  CARNEGIE     INSTITUTION     OF      WASHINGTON 

scientific  and  technical  universities  of  the  nation.  In  the  opening  years  of 
the  new  century  he  had  established  the  Carnegie  Trust  for  the  Universities 
of  Scotland,  and  the  Carnegie  Dunfermline  Trust  in  benefit  of  his  native 
town.  Nor  was  it,  by  many  removes,  to  be  the  last.  There  were  to  follow 
the  Carnegie  Foundation  for  the  Advancement  of  Teaching,  the  Carnegie 
Endowment  for  International  Peace,  Carnegie  Hero  Funds  in  no  less  than 
eleven  countries,  and  finally,  in  culmination,  the  Carnegie  Corporation  of 
New  York.  And  long  before  all  of  them — indeed  well  before  the  publication 
of  his  pioneering  " Gospel  of  Wealth"  in  the  North  American  Review  in 
1889 — he  had  initiated  that  career  of  benefactions  which  was  to  be  so 
profoundly  influential  in  all  the  subsequent  shaping  of  American  philan- 
thropic tradition  with  the  gift  of  a  library  to  his  native  Dunfermline. 

But  the  establishment  of  the  Carnegie  Institution  of  Washington  marked 
a  new  direction  in  the  kinds  of  institutions  made  possible  by  Mr.  Carnegie's 
gift.  In  fact,  it  established  a  new  kind  of  institution  for  America — the  first 
to  be  devoted  wholly  and  completely,  in  intent  and  in  philosophy,  to  the 
ideal  of  research  scholarship  over  wide  fronts  of  science  in  its  broadest, 
most  unfettered,  most  completely  uncommitted  aspect.  This  was  a  novel 
concept,  and  quite  obviously,  from  some  of  the  records  of  the  time,  one 
neither  everywhere  comprehensible  nor  even  everywhere  palatable  in  a 
youthful  nation  with  a  strongly  established  pragmatic  tradition.  It  repre- 
sented, indeed,  a  notably  original  idea,  which  six  following  decades  have 
shown  to  be  both  great  and  enduring. 

Four  years  after  the  establishment  of  the  Institution,  it  had  been  granted 
a  new  Charter  by  special  Act  of  Congress  and  had  been  organized  into  no 
less  than  fourteen  departments,  representing  as  many  subjects.  Over  the 
next  five  years,  definitive  judgments  were  made  as  to  where  and  how  the 
Institution  could  work  most  effectively.  One  of  them  made  during  these 
years  of  experiment  and  trial  was  to  prove  crucial.  It  involved  the  decision 
to  concentrate  the  resources  of  the  Institution  primarily  on  the  research  of 
its  various  departments ;  to  make  of  it,  in  essence,  an  operating  rather  than 
a  granting  scientific  organization.  By  1911,  its  endowment  more  than 
doubled  by  subsequent  additions  by  Mr.  Carnegie,  its  departments  firmly 
established  but  now  reduced  to  ten,  the  Institution  was  molded  to  the 
purpose,  and  had  taken  on  essentially  the  form  of  organization,  that 
characterize  it  to  this  day.  Through  the  following  years  new  departments 
have  arisen,  departments  have  been  consolidated,  and  some  departments 
have  been  closed,  as  the  needs  and  the  research  frontiers  of  each  decade 
have  dictated.  Whole  fields  that  were  represented  in  the  Institution  in  1911, 
like  economics  and  sociology,  historical  research,  meridian  astrometry, 
nutrition  in  the  medical  sense,  no  longer  are  included  in  its  program  as  the 
resources  of  the  nation  in  those  areas  have  strengthened  and  enlarged. 


REPORT   OF   THE    PRESIDENT  5 

Other  fields  not  represented  then  but  now  on  the  frontiers  of  research,  like 
modern  embryology,  molecular  and  cellular  biology,  the  study  of  the 
mechanisms  of  photosynthesis,  have  been  included  in  its  purview  in  more 
recent  years.  Today  there  are  five  instead  of  ten  departments  in  the  Insti- 
tution. Most  originated  in  planning  going  back  to  the  very  beginning, 
though  the  work  they  conduct  today,  under  the  same  general  titles  with 
which  they  began,  has  expanded  far  beyond  the  original  concepts  embodied 
in  those  rubrics,  and  may  have  wandered  far  afield  from  them  as  well.  The 
Department  of  Terrestrial  Magnetism  was  founded  in  1904,  the  Geophysical 
Laboratory  in  1906,  and  a  Desert  Laboratory,  later  to  become  the  Division 
and  then  the  Department  of  Plant  Biology,  appeared  in  1903.  A  Solar 
Observatory  for  Mount  Wilson  was  planned  as  early  as  1902.  Studies  of  the 
sun  remain  at  the  pioneering  fringes  of  investigation  in  that  part  of  the 
Institution  to  this  day.  But  now  the  Solar  Observatories  have  metamor- 
phosed to  the  complex  of  giant  telescopes  included  in  the  Mount  Wilson 
and  Palomar  Observatories,  operated  jointly  with  the  California  Institute 
of  Technology.  To  the  intensive  program  of  solar  investigations  of  which 
George  Ellery  Hale  dreamed  and  which  he  initiated  with  his  striking 
discoveries  of  magnetic  fields  in  the  sun  have  been  added  a  goodly  share  of 
the  world's  most  important  findings  about  the  farthest  reaches  of  the 
celestial  universe. 

But  through  all  the  years  the  major  philosophies  of  the  Institution  and 
one  major  feature  of  its  organizational  pattern  have  stood  constant,  tested 
and  retested  in  situation  after  situation  and  proved  as  fresh  and  relevant 
today  as  when  they  were  conceived.  The  decision  made  at  the  outset  that 
flexibility  and  effectiveness  in  the  kind  of  research  to  which  the  Institution 
is  dedicated  can  best  be  achieved  through  a  series  of  rather  small  unit 
laboratories,  each  mobile  and  relatively  independent,  each  able  to  seize  the 
initiative  in  new  and  appropriate  fields  as  they  appear,  yet  all  sufficiently 
connected  so  that  they  may  be  of  mutual  assistance  as  the  needs  arise,  was 
a  remarkable  one,  both  for  its  uniqueness  at  the  time  and  for  the  subtlety 
of  the  vision  that  dictated  it.  Over  the  decades,  as  research  has  burgeoned 
in  the  nation  and  groups  devoted  to  research  have  multiplied,  many  other 
experiments  in  organizational  form  have  been  tried.  But  it  is  especially 
interesting  that  some  of  the  most  modern  thinking  and  experimenting  in 
organization  for  research,  in  this  country  as  well  as  abroad,  has  returned  to 
precisely  this  pattern  as  one  of  the  most  effective  in  exploring  the  dynamic 
frontiers  of  scientific  knowledge. 

Organization,  however,  is  only  a  framework,  vital  but  at  last  only 
supporting.  Most  significant — and  most  truly  enduring — have  been  the 
elements  of  philosophy  and  purpose  which  inaugurated  the  Institution  and 
which  have  remained  unchanged  through  all  the  years :  the  philosophy  that 


6  CARNEGIE     INSTITUTION     OF      WASHINGTON 

all  its  resources,  all  its  deepest  purposes,  are  centered  in  the  creative 
individual,  whatever  be  his  field,  that  in  the  truest  sense  he  is  the  uncom- 
mitted investigator,  suitably  endowed  and  suitably  protected,  whose  time, 
quite  literally,  is  bought  by  the  Institution  and  then  returned  as  uncon- 
strained endowment.  And  with  this  goes  the  philosophy,  equally  deep-seated 
and  equally  important,  that  this  freedom  from  fixed  commitment  applies 
to  fields  of  endeavor  as  well  as  to  men:  that  high  mobility  within  specific 
fields,  that  the  unfettered  crossing  of  fields,  that  the  fashioning  of  uncon- 
ventionally wide-ranging  programs,  are  subject  only  to  the  limitations 
imposed  by  Nature  and  by  the  judgment  of  gifted  and  discriminating 
investigators,  and  that  making  this  mobility  and  this  flexibility  possible  is 
a  principal  objective  of  the  Institution. 

Over  the  years  that  philosophy,  and  the  programs  that  have  followed 
from  it,  have  led  to  many  pioneering  practical  discoveries  within  the 
Institution.  The  elucidation  of  the  genetic  principles  underlying  the 
development  of  hybrid  corn,  first  accomplished  by  Shull  in  the  Department 
of  Genetics  at  Cold  Spring  Harbor  working  with  East  at  Harvard,  provided 
the  fountainhead  for  an  agricultural  innovation  which  by  1952  was  esti- 
mated to  have  brought  an  economic  gain  for  the  United  States  of  almost 
forty  billion  dollars.  For  many  of  the  predominantly  agricultural  countries 
of  the  world,  moreover,  the  technique  of  hybrid  corn  has  provided  one  of 
those  basic  resources  which,  as  Galbraith  has  recently  pointed  out,  is  in 
the  truest  sense  a  fundamental  contribution  to  their  economic  strength — an 
advance  of  really  general  application.  At  the  same  Department,  during  the 
second  world  war,  studies  of  mutations  occurring  under  X-ray  bombard- 
ment in  the  famous  mold  Penicillium  resulted  in  the  development  of  a 
strain  of  that  fungus  which  produced  three  to  five  times  as  much  of  the 
vitally  needed  penicillin  as  the  highest-yielding  strains  then  known. 

In  1925,  fully  fifteen  years  before  the  intensive  research  on  radar  for 
combat  in  the  second  world  war,  Breit  and  Tuve  at  the  Department  of 
Terrestrial  Magnetism,  experimenting  with  a  modified  Navy  transmitter, 
produced  radio  pulses  and  for  the  first  time  observed  their  echoes  from  the 
ionosphere.  In  the  course  of  those  experiments,  moreover,  they  detected  a 
curious  interference  of  normal  echoes  by  passing  planes — prophecy  of  the 
field  of  radar.  At  the  Geophysical  Laboratory,  Day  and  Shepherd  early 
undertook  studies  in  the  field  of  low-expansion  quartz  glasses  that  proved 
basic  to  the  evolution  of  Pyrex — a  program  which  during  the  first  world 
war  supplied  the  United  States  with  ninety-seven  per  cent  of  its  require- 
ments for  optical  glass.  In  1935  a  modified  formula  for  annealing  that  same 
Pyrex  glass  proved  fundamental  to  the  manufacture  of  the  mirror  for  the 
two-hundred-inch  telescope  on  Palomar  Mountain.  Later,  in  the  same 
laboratory,  studies  by  Morey  on  lanthanum  and  borate  glasses  of  high 


REPORT    OF   THE    PRESIDENT  7 

refractive  index  led  to  a  whole  new  family  of  glasses  of  great  importance  in 
the  manufacture  of  photographic  lenses — a  development  having  important 
implications  for  the  second  world  war.  In  the  Geophysical  Laboratory, 
again,  Rankin  and  Wright  as  early  as  1915  were  able  to  solve  the  age-old 
riddle  of  cement,  and  their  classic  work  has  served  ever  since  as  a  guide  in 
the  chemical  aspects  of  the  cement  industry.  From  the  same  laboratory  in 
later  years  have  come  new  refractories  for  the  steel  industry,  studies  of 
natural  geothermometers  and  geochronometers  of  fundamental  concern  to 
practical  mining  and  oil  prospecting  as  much  as  to  fundamental  geology, 
and,  as  recently  as  1959,  synthetic  diamonds  produced  with  new  substrates 
and  under  new  conditions  of  pressure. 

Such  practical  innovation  within  the  Institution  has  not  been  confined 
to  the  substantive  aspects  of  its  concerns.  In  both  world  wars  the  Institution 
played  a  major  role  in  initiating  forms  of  research  organization  for  armed 
conflict.  In  the  first  war,  the  scientific  and  technical  role  of  the  Institution 
overshadowed  its  organizational  one.  But  in  World  War  II,  through  its 
President,  the  Institution  served  as  a  core  of  thinking  and  effort  from  which, 
in  the  following  war  years,  the  Office  of  Scientific  Research  and  Develop- 
ment was  to  grow  and  to  assume  the  lead  in  civilian  scientific  and  technical 
military  development  in  the  nation.  Through  its  activity  and  its  influence, 
a  preponderant  share  of  all  the  major  scientific  and  technical  advances  in 
the  military  art  were  achieved,  from  radar  to  modern  submarine  detection 
to  proximity  fuzes  to  nuclear  weapons  to  new  and  improved  prosthetic  aids 
for  the  war  wounded  and  the  war  blinded. 

But  as  critical  as  the  technical  findings  developed  from  its  activities,  and 
in  the  final  analysis  perhaps  more  enduring,  was  the  dramatic  and  conclusive 
demonstration  of  the  crucial  role  that  science  as  a  whole  must  play  in  our 
national  life  in  the  years  to  come,  in  formal  peace  as  in  formal  war.  Experi- 
ments in  the  organization  of  science  were  initiated  in  the  O.S.R.D.  which 
were  ultimately  to  find  fruition  in  such  government  instruments  for  the 
furtherance  of  scientific  development  throughout  the  nation  as  the  Office 
of  Naval  Research  and  later  the  National  Science  Foundation,  and  in  such 
bodies  as  the  Atomic  Energy  Commission,  whose  present  organizational 
patterns,  first  tested  in  the  Manhattan  Project  of  the  Army  Corps  of 
Engineers,  were  likewise  pioneered  in  the  O.S.R.D.  They  were  reflected,  too, 
in  such  special  resources  of  military  thinking  and  planning  as  the  Rand 
Corporation,  founded  shortly  after  the  close  of  the  war.  Before  those 
wartime  demonstrations  of  the  crucial  role  of  science  and  technology  in  the 
very  web  of  our  national  life  had  been  made,  the  greater  part  of  the  scientific 
activity  of  the  nation  was  prosecuted  outside  the  sphere  of  government  and 
of  public  funds.  Today,  probably  sixty-five  per  cent  of  the  total  research  of 
the  nation  is  supported  by  federal  funds,  and  the  proportion  is  continuing 


8  CARNEGIE     INSTITUTION     OF     WASHINGTON 

to  grow.  It  is  a  dramatic  demonstration  of  how  deeply,  in  the  public  view, 
the  scientific  and  technical  development  of  the  nation  has,  in  fact,  become 
the  whole  nation's  concern.  This  situation  has  brought  its  own  problems,  of 
a  wholly  new  order  of  scope  and  depth.  They,  too,  must  be  important 
concerns  in  the  future  for  the  Carnegie  Institution. 


To  have  initiated  such  practical  contributions  to  the  public  welfare  on 
the  scientific  and  technical  fronts,  to  have  participated  actively  and  sig- 
nificantly in  the  initiation  of  major  currents  of  scientific  history  whose 
sweep  has  now  carried  us  to  realms  far  beyond  what  was  remotely  imagined 
even  twenty  years  ago,  to  have  pioneered  forms  of  organization  that  are 
today  in  the  furnace  of  national  trial  and  test,  sum  to  considerable  useful 
achievement,  and  might  be  thought,  in  and  of  themselves,  to  justify  the 
vision  upon  which  the  Institution  was  founded  and  through  which  it  lives 
today.  Yet,  in  one  sense,  they  represent  mere  by-products,  mere  projecting 
iceberg  tips,  as  it  were,  of  that  vision,  indicators  only  of  the  submerged 
seven-eighths.  That  seven-eighths  lies  in  the  kingdom  of  the  mind.  It  lies 
in  that  devotion  to  deeper  patterns,  the  symmetries,  the  lights  and  shades 
of  Nature,  wherever  the  search  may  lead,  to  which  the  Institution  was 
originally  dedicated,  and  which,  undeviatingly,  it  pursues  today. 

That  seven-eighths  too  has  been  productive  of  striking  innovations  in  its 
own  realm,  and  these,  possibly  in  a  truer  sense  than  the  practical  "firsts," 
stand  as  proper  signatures  of  the  Institution.  They  range  over  many  fields. 
While  the  thinking  which  underlay  the  famous  Michelson-Morley  experi- 
ment on  "ether  drift"  was  yet  fresh,  Professor  Michelson,  holder  of  the  first 
Nobel  prize  in  the  natural  sciences  to  be  awarded  in  America,  within  the 
Institution  repeated  the  experiment  with  an  accuracy  hitherto  unattained, 
giving  strong  support  to  the  theory  of  relativity,  itself  still  at  the  stage  of 
question  and  of  doubt.  Within  the  Institution,  too,  Michelson  repeated 
with  greater  refinement  that  classic  work  that  he  had  first  undertaken  as 
Ensign  A.  A.  Michelson  of  the  United  States  Navy,  determining  the 
velocity  of  light  with  a  new  precision,  first  across  a  path  between  the  peaks 
of  Mount  Wilson  and  Mount  San  Antonio,  then  in  one  defined  by  a  mile- 
long  line  of  evacuated  pipe  at  the  Irvine  Ranch  in  southern  California.  At 
the  Mount  Wilson  and  Palomar  Observatories  Hale's  pioneering  discovery 
that  sunspots  mark  strong  magnetic  fields  has  been  followed  in  more  recent 
years  by  studies  of  solar  magnetism  of  unprecedented  refinement,  and  by 
the  discovery,  among  the  stars,  of  the  most  intense  magnetic  fields  ever 
observed  in  any  astronomical  body.  Hubble's  studies  of  the  phenomenon  of 
the  redshift  in  stellar  spectra  led  to  the  theory  of  the  expanding  universe, 
culminating  dramatically  a  year  ago  in  the  measurement  of  the  redshift  of 


REPORT   OF   THE    PRESIDENT  9 

by  far  the  most  distant  celestial  object  yet  recorded.  At  the  Observatories, 
too,  Baade's  studies  of  the  structure  and  stellar  composition  of  galaxies, 
with  those  of  others,  have  suggested  concepts  of  stellar  evolution,  of  growth 
and  decay,  undreamed  of  as  little  as  a  quarter  century  ago. 

At  the  Department  of  Terrestrial  Magnetism  a  series  of  conferences  on 
theoretical  physics,  held  shortly  before  the  second  world  war  in  cooperation 
with  the  George  Washington  University,  among  other  things  stimulated 
the  suggestion  that  the  source  of  energy  in  the  sun  and  the  stars  is  a  nuclear 
reaction  involving  carbon — a  notion  leading  within  the  next  year  to  a 
classical  model  of  the  hydrogen-helium  reaction  now  familiar  as  one  of  the 
accepted  sources  of  stellar  energy,  ancillary  to  the  hydrogen-deuterium- 
helium  reaction  recognized  in  recent  years  as  more  important.  In  the 
Geophysical  Laboratory  studies  of  the  biochemistry  of  ancient  sediments 
have  given  new  dimensions  to  our  concept  of  the  age  of  terrestrial  life,  while 
studies  of  the  artificial  synthesis  of  amino  acids  from  inorganic  components 
under  a  variety  of  physical  and  chemical  conditions,  besides  shedding  new 
light  on  the  probable  modes  of  the  origin  of  life  on  earth  and  the  nature  of 
its  chemical  environments,  have  also  carried  important  theoretical  impli- 
cations for  our  notions  about  the  existence  of  life  on  other  planets. 

At  the  Department  of  Plant  Biology,  work  on  photosynthesis  has  pro- 
duced suggestive  insights  about  that  critical  step  which,  with  all  the  research 
that  has  been  brought  to  bear  for  the  last  half-century,  still  eludes  our 
understanding — the  initial  process  by  which  the  energy  of  light  is  used  in 
the  fixation  of  carbon  dioxide.  It  has  brought  suggestions,  too,  about  that 
further  mystery,  still  elusive,  of  what  it  is  about  the  chloroplast  that  enables 
it  alone,  when  intact,  to  bring  this  about,  whereas  extracted  chlorophyll 
itself  will  not.  And  in  that  Department,  too,  investigations  of  many  years' 
duration  have  illuminated  the  detailed  bases  of  plant  evolution — of  the 
roles  of  mutation  and  selection,  of  the  development  of  ecological  races  and 
of  speciation — and  have  revealed  the  often  enormously  complex  and 
exquisitely  coordinated  detail  of  the  evolutionary  patterns  they  compose, 
at  the  levels  both  of  form  and  of  physiological  function. 

In  three  laboratories  of  the  Institution — the  Department  of  Terrestrial 
Magnetism,  the  Department  of  Embryology,  and  the  Department  of 
Genetics  at  Cold  Spring  Harbor — investigations  of  cellular  metabolism  and 
development,  of  cellular  differentiation,  and  of  the  mechanisms  of  heredity 
at  the  molecular  level  have  brought  striking  new  knowledge  of  the  detailed 
ways  in  which  the  materials  of  heredity  and  of  development  interact  at  the 
level  of  the  cell  nucleus  and  of  its  cytoplasm,  at  the  level  of  the  germinal 
cell  and  of  the  body  cell  of  the  plant  or  animal,  at  the  level  of  differentiation 
and  development  of  the  individual  organism,  and  at  the  level  of  its  heredity. 

Such  discoveries  and  results  are  but  scattered  samples  taken  from  a  rich 


10  CARNEGIE     INSTITUTION     OF      WASHINGTON 

matrix  of  sixty  years  of  Institution  work.  But  they  are  fair  examples  of  its 
most  typical  fruit — the  truest  product  of  the  philosophy  in  which  it  was 
founded  and  through  which  it  lives.  It  may  well  be  said  that  all  else  is  in 
one  sense  by-product. 


In  the  seventh  decade  of  the  twentieth  century,  it  is  hard  to  recast  the 
scientific  and  technical  America  in  which  the  Carnegie  Institution  was 
founded  in  1902.  In  the  America  of  1902,  few  if  any  corporations  in  the 
United  States  could  boast  over  sixty  thousand  stockholders.  The  American 
Telephone  and  Telegraph  Company,  as  example,  admitted  to  less  than 
eight  thousand.  A  third  of  all  the  manufactured  products  of  the  country 
were  produced  by  partnerships  or  by  individual  proprietors.  Speech  had 
been  transmitted  by  wireless,  but  the  Fleming  valve  was  still  to  be  produced, 
and  the  first  audion  was  not  to  be  developed  for  eight  more  years.  The  first 
aerial  flight,  the  twelve-second  achievement  of  Orville  Wright  at  Kitty 
Hawk,  was  not  to  occur  until  the  following  year.  The  earliest  motion  picture 
to  tell  a  connected  story,  The  Great  Train  Robbery,  was  yet  to  be  produced. 
A  large  proportion  of  such  great  technical  industries  of  today  as  the  movie 
and  the  aircraft  industries  had  not  been  born,  and  even  the  technical  prin- 
ciples underlying  the  television  industry  were  not  yet  remotely  conceived. 

The  independent  industrial  laboratory  had  been  pioneered  some  years 
earlier  by  the  Arthur  D.  Little  Company,  but  the  concept  of  such  a  labora- 
tory within  an  industry  had  just  been  formulated  and  put  into  practice 
with  the  establishing  of  the  General  Electric  research  laboratory  in  1901 
and  of  that  of  the  du  Pont  Company  in  the  same  year  the  Institution  was 
founded.  Of  all  the  great  complex  of  industrial  laboratories  that  were  to 
transform  the  nature  of  American  industrial  science  and  technology  in  the 
twentieth  century,  not  one  other  had  yet  appeared. 

For  the  scope  of  science  in  that  day,  it  is  worth  noting  that  in  genetics 
it  was  only  two  years  before  that  the  work  of  Gregor  Mendel  had  been 
rediscovered  and  its  significance  truly  appreciated  by  Hugo  de  Vries  and 
Correns  and  von  Tschermak-Seysenegg.  The  very  notion  that  some  genetic 
characteristics  can  be  dealt  with  in  crosses  in  numerical  ratios  was  still 
unfamiliar,  while  ideas  of  genetic  linkage  and  dominance,  or  the  notion  of 
the  linear  array  of  genes,  was  still  almost  a  decade  away.  Indeed,  there  was 
no  proper  science  of  genetics  at  all,  and  the  word  gene  itself  had  yet  to  be 
coined.  In  astronomy,  it  is  probably  fair  to  say  that  the  entire  known 
universe  was  thought  to  lie  within  our  own  Galaxy.  By  contrast,  within  the 
range  of  the  two-hundred-inch  Hale  telescope  today  lie  perhaps  a  thousand 
million  such  galaxies. 


REPORT    OF   THE    PRESIDENT  H 

Only  seven  years  before  the  Institution  was  established,  Wilhelm 
Roentgen  had  given  the  first  demonstration  of  the  X  rays  that  bear  his 
name,  and  the  first  Nobel  award  in  science  had  gone  to  him  for  that  dis- 
covery only  a  year  before  the  founding  of  the  Institution.  The  electron  had 
been  discovered  by  J.  J.  Thomson  but  five  years  earlier,  and  radium  and 
thorium  had  been  isolated  by  the  Curies  only  four  years  before.  Max  Planck 
had  advanced  the  quantum  theory  in  the  year  preceding  the  founding  of 
the  Institution.  And  the  special  theory  of  relativity  was  not  to  appear  for 
three  more  years.  The  Institution  was  five  years  old  when  the  first  Nobel 
award  in  science  to  be  made  in  the  United  States  came  to  Albert  Michelson. 

In  the  world  of  technology,  plastics,  synthetic  fibers,  vitamins,  anti- 
biotics, all  were  unknown.  And  in  practical  medicine,  it  is  striking  that  the 
national  death  rate  from  influenza  and  pneumonia  was  reckoned  at  one 
hundred  and  eighty-two  per  one  hundred  thousand  of  the  population — a 
figure  to  be  reduced  to  thirty-nine  forty-eight  years  later.  In  the  same 
period  deaths  from  scarlet  fever  fell  from  more  than  eleven  per  thousand  to 
a  total  of  sixty-eight  for  the  entire  country.  It  is  worth  recalling  that,  when 
Lord  Lister,  scientific  disciple  of  Pasteur  to  whom  the  whole  concept  of 
antisepsis  and  sterilization  in  medical  practice  may  be  said  to  have  been 
due,  died  in  1912,  the  Institution  was  already  completing  its  first  decade. 
Such  was  the  world  scene  of  science  and  technology  within  which  the 
Institution  took  its  place. 


In  1902  science  and  technology  were  already  familiar  concerns  within  the 
federal  government.  They  were  indeed  concerns  as  old  as  the  nation  itself. 
It  was  Thomas  Jefferson  who  as  Secretary  of  State  in  1790  submitted  a 
"Report  ...  on  the  Subject  of  Establishing  the  Uniformity  of  the  Weights, 
Measures,  and  Coins  of  the  United  States,"  and  who,  upon  recommendation 
of  the  American  Philosophical  Society,  transmitted  to  the  Congress  a 
proposal  for  the  establishment  of  a  United  States  Coast  Survey,  which  was 
set  up  within  the  Treasury  Department  seventeen  years  later.  And  it  was 
John  Quincy  Adams,  when  he  was  Secretary  of  State,  who  personally 
prepared  for  the  Congress  a  similar  report  upon  weights  and  measures.  It 
was  Adams,  too,  who  led  the  fight  to  accept  the  bequest  from  James 
Smithson,  who  had  died  in  1829,  to  found  the  organization  that  was  to  grow 
to  the  Smithsonian  Institution  of  today.  The  establishment  of  the  Depart- 
ment of  Agriculture  dated  from  Civil  War  days,  contemporary  with  the 
passage  of  the  Morrill  Act.  So  also  did  the  National  Academy  of  Sciences, 
from  whose  recommendations,  somewhat  later,  were  to  follow  the  Geo- 
logical Survey  and  the  Weather  Bureau. 


12  CARNEGIE     INSTITUTION      OF      WASHINGTON 

These  early  involvements  of  the  federal  government  in  science  and 
technology,  however,  gave  little  hint  of  the  massive  and  commanding  role 
it  would  play  on  the  national  scene  in  little  more  than  half  a  century.  Even 
at  the  end  of  the  fourth  decade  of  the  twentieth  century  the  total  federal 
research  program  is  estimated  to  have  cost  annually  only  about  one  hundred 
million  dollars — less  than  the  annual  budget  for  the  National  Science 
Foundation  alone  in  1962.  Twenty  years  later,  however,  yearly  federal 
expenditures  for  research  and  development  had  grown  to  over  a  billion 
dollars  out  of  a  total  estimated  national  commitment  of  about  three  billion. 
By  1960  the  national  total  had  climbed  to  fourteen  billion  dollars  or  more, 
of  which  the  federal  government  supplied  some  nine  billion.  Today  it  may 
have  reached  sixteen  to  eighteen  billion.  The  budget  of  the  National  Science 
Foundation  for  scientific  research  and  related  activities  as  submitted  to  the 
Congress  for  1963  will  total  one  hundred  and  sixty-five  million  dollars,  while 
the  Department  of  Defense  is  expected  to  spend  about  seven  billion  dollars 
on  research  and  development,  the  National  Aeronautics  and  Space  Admin- 
istration about  two  and  one-half  billion,  the  Atomic  Energy  Commission 
approximately  another  one  and  one-half  billion.  The  total  government  funds 
spent  in  research  and  development  in  1963  are  expected  to  reach  almost 
twelve  and  one-half  billion  dollars,  of  which  expenditures  for  research  alone 
may  attain  to  one  and  one-half  billion  dollars,  as  compared  with  approxi- 
mately one  billion  for  the  present  year. 

It  has  been  calculated  that  the  total  funds  expended  for  research  and 
development  in  the  United  States  over  the  past  decade  have  increased  at 
approximately  fifteen  per  cent  per  year,  leading  to  a  doubling  of  volume 
every  five  years.  If  the  present  rate  of  increase  of  our  expenditures  in  the 
field  were  to  continue,  indeed,  our  projected  monetary  support  of  research 
and  development  in  their  current  definition  could  formally  exceed  our  total 
governmental  budget  before  1975,  and  could  exceed  our  gross  national 
product  before  the  end  of  the  century — a  reflection,  however  hypothetical, 
that  vividly  illuminates  the  scientific  and  technical  dynamism  and  the 
scientific  and  technical  problems  with  which  we  live.  How  different  is  this 
scene  from  that  upon  which  the  Institution  entered ! 

The  implications  of  this  astonishing  vista  are  many.  One  is  the  degree  to 
which,  with  almost  explosive  suddenness  since  World  War  II,  science  and 
technology  have  been  universally  recognized  as  of  major  national  concern. 
Another,  of  course,  reflects  the  depth  and  intensity  of  technological  compe- 
tition in  the  world  and  our  own  needs  in  national  defense.  A  third  mirrors 
both  the  rate  of  population  growth  and,  most  pointedly,  the  growth  of 
wealth  in  the  United  States.  And  the  climates  in  which  these  expenditures 
on  both  the  private  and  the  public  fronts  have  taken  place  and  the  govern- 
mental patterns  through  which  they  are  effected  in  the  public  sector — 


REPORT    OF   THE   PRESIDENT  13 

patterns  at  present  in  perhaps  their  most  active  phases  of  evolution  and  of 
adjustment — make  a  compelling  chapter  in  the  history  of  development  both 
of  American  scientific  enterprise  and  awareness  and  of  American  political 
institutions,  and  reveal  much  about  their  nature. 

All  these  factors — the  vast  increase  in  the  volume  of  our  scientific  and 
technical  resources,  in  human  and  in  monetary  terms  and  in  terms  of 
scientific  and  technical  facilities,  the  pressing  demands  of  overriding  national 
objectives,  economic  and  military,  the  consequent  larger  and  larger  partici- 
pation of  federal  resources  in  the  total  funding  of  the  national  research  and 
more  especially  of  the  national  technical  effort — have,  not  unnaturally,  had 
profound  impacts  on  our  thinking  about  science  generally.  Bit  by  bit  they 
may  have  led  to  some  subtle  changes,  perhaps  well-nigh  unconscious  ones, 
in  our  conception  of  the  ways  in  which,  typically,  the  frontiers  of  truly  new 
scientific  knowledge  are  pushed  back.  This  evolution  could  carry  implica- 
tions grave  enough  to  warrant  serious  thought. 

In  all  the  years  of  American  scientific  research,  from  the  times  of  Josiah 
Willard  Gibbs  to  those  of  the  second  world  war,  we  were  accustomed  to 
think  of  the  great  advances  in  scientific  thought,  of  the  initiation  of  its 
great  new  directions,  as  being  predominantly  the  product  of  individual 
genius,  working  in  environments  which,  however  modest,  and  in  part 
perhaps  because  of  that  very  modesty,  were  especially  adapted  for  flexi- 
bility, for  absence  of  constraint,  for  a  maximum  of  freedom  in  concept  and 
in  execution.  We  thought  of  the  outstanding  scientific  conquest  as  typically 
an  achievement  of  extraordinary  brilliance,  originality,  and  insight  in 
individual  innovation,  giving  significant  new  dimensions  to  its  time,  and 
ideally  climaxing  a  career  of  unfettered  scholarship.  We  did  not  particularly 
conceive  research  in  this  sense  as  the  composite  product  of  large  numbers 
of  men  working  in  numerous  and  highly  organized  groups. 

Since  the  second  world  war,  however,  following  the  spectacular  demon- 
strations of  technical  conquest  wrought  by  great  organizations,  of  which  the 
Manhattan  Project  was  but  the  forerunner,  we  have  sometimes  been 
inclined  by  analogy  to  conceive  of  pioneering  research  for  basically  new 
ideas  in  rather  similar  terms — inclined,  perhaps,  to  more  than  half  believe 
that  in  the  contemporary  world  it  too  may  require  such  teams.  It  is  then 
only  logical  to  reason  that  if,  at  this  stage  of  the  world's  scientific  develop- 
ment, pioneering  scientific  research  critically  depends  upon  the  large-scale 
efforts  of  highly  organized  and  massively  implemented  teams,  its  effective- 
ness may  be  roughly  proportionate  to  the  material  resources  bestowed  upon 
it — and  that  cost  and  magnitude  themselves  may  provide  an  important 
index  of  scientific  significance.  We  have  even  been  tempted  at  times  to 
imagine  that  the  speed  and  effectiveness  with  which  new  scientific  frontiers 
are  breached  may  be  a  simple  function  of  numbers  of  men  and  rates  of 


IJf.  CARNEGIE     INSTITUTION      OF      WASHINGTON 

expenditure,  and  to  expect  that  the  attainment  of  new  scientific  vision  in 
an  area  of  basic  research  may  be  accelerated  in  direct  proportion  to  the 
size  of  teams  and  the  amounts  of  money  committed  to  the  search. 

This  philosophy,  so  directly  derived  from  the  demonstrated  course  of 
practical  achievement,  appeals  especially  to  that  keen  pragmatic  instinct 
that  has  run  like  a  golden  thread  through  all  the  fabric  of  our  development 
as  a  nation,  and  to  the  genius  for  organization  which  has  so  long  been  one  of 
our  most  pronounced  national  characteristics.  Nor  is  there  lack  of  evidence 
that  at  first  sight  seems  to  confirm  the  idea.  It  is  patent  today  that  the 
physical  equipment  required  on  the  frontiers  of  research  in  many  of  the 
sciences,  especially  those  of  the  greatest  conceptual  maturity,  is  massive, 
complex,  and  expensive,  and  requires  the  collaboration  of  sizable  teams  in 
designing  it,  in  manipulating  it,  and  in  gathering  data  with  it  if  truly  new 
information  is  to  be  obtained.  The  productiveness  of  research  in  many  such 
fields  since  men  and  money  have  made  possible  the  design  of  powerful  new 
tools  and  massive  teams  have  been  assembled  to  operate  them  gives  vivid 
testimony  to  how  powerful,  and  indeed  how  indispensable,  resources  of  this 
kind  may  be  in  some  of  the  most  highly  developed  fields  of  science. 

Yet  in  a  deeper  sense  this  judgment  may  harbor  a  considerable,  and 
sometimes  a  positively  dangerous,  misconception,  especially  when  it  is 
assumed  that  great  teams  and  high  costs  are  prerequisites  for  the  setting  of 
new  directions  in  scientific  thought.  A  part  of  that  misconception  doubtless 
stems  from  a  failure  to  demark  sufficiently  two  general  approaches  in 
research,  which,  though  they  are  complementary  and  often  intergrade,  yet 
have  certain  characteristics  and  pose  certain  requirements  that  are  quite 
distinct.  In  one  the  basic  ends  of  the  investigation  are  generally  evident,  if 
not  wholly  clear  in  detail,  at  or  near  its  beginning.  The  preeminent  challenge 
to  the  investigator  is  to  chart  the  road  toward  his  goal — mapping  it, 
projecting  it,  building  it,  all  that  it  may  approach  a  citadel  already  at  least 
dimly  visible  on  the  horizon.  The  other  general  kind  of  research  may  begin 
without  specific  ends  or,  indeed,  without  consciously  conceived  objectives 
of  any  kind.  Its  driving  motive  is  likely  to  be  pure  curiosity,  the  winning 
from  Nature  of  deeply  new  knowledge,  of  knowledge  won  wholly  for  its 
own  sake.  The  talents  and  the  training  demanded  by  these  two  kinds  of 
research,  and  the  difficulty  of  the  scientific  challenges  posed  by  each,  are 
often  much  the  same.  At  one  end  of  a  spectrum  of  research  they  intergrade, 
and  any  distinction  attempted  between  them  becomes  formal  and  unreal. 
At  their  extremes,  however,  the  challenges  they  present  are  undoubtedly 
quite  different,  often  to  be  met  in  widely  divergent  ways.  Above  all,  whereas 
research  programs  of  the  first  kind  can  frequently  be  visualized  in  a  general 
way  ahead  of  time,  and  so  planned  intelligently,  the  same  is  rarely  true  in 
the  second  type  of  research.  A  very  large  share  of  the  concerns  of  such  a 


BEPORT    OF   THE    PRESIDENT  15 

great  team  effort  as  was  involved  in  the  program  of  the  Manhattan  Project, 
for  instance,  fell  into  the  former  category.  The  deeply  underlying  theoretical 
knowledge,  the  unexpected  and  radically  new  ideas  about  Nature,  on  which 
the  whole  program  of  the  Project  was  based  and  on  which  it  turned,  had 
been  achieved  by  investigators  like  Meitner  and  Hahn  and  Strassmann  in 
Europe  in  1938,  by  such  individuals  as  Rutherford  and  his  colleagues  at 
Cambridge  in  1914.  They  had  been  won  through  research  of  the  second 
kind,  conducted  by  a  very  few  gifted  scientists  working  in  the  settings  we 
have  traditionally  visualized  as  consonant  with  the  finest  of  individual 
creative  effort. 

It  is  no  accident  that  today  we  sometimes  make  these  distinctions  less 
clearly  than  we  might.  At  a  very  deep  level  it  may  be  a  consequence  of  our 
peculiar  history  and  circumstances.  Throughout  our  earlier  years  as  a 
technically  developing  nation  we  were  able  to  rely  on  the  older  countries  of 
Europe  for  basic  ideas  on  which  to  build  our  applications  as  implicitly,  and 
often  as  unconsciously,  as  we  relied  upon  the  British  navy  for  the  protection 
of  our  seas.  It  was  both  natural  and  adaptive  that  the  kind  of  scientific  and 
technical  contributions  at  which  we  early  became  most  adept  and  developed 
most  highly,  and  to  which  perhaps  we  initially  attached  greatest  attention 
and  attributed  greatest  value,  should  have  involved  the  brilliantly  organ- 
ized, the  meticulously  careful  development,  often  undertaken  on  the  boldest 
and  most  breathtaking  scale,  of  basic  ideas  that  had  been  conceived  abroad. 
Today  such  ideas  are  much  more  often  drawn  from  our  own  resources.  But 
historically  our  first  attachment  was  to  their  execution  rather  than  to  their 
generation.  And  so  it  is  not  surprising  that  we  sometimes  fail  to  distinguish 
innovation  from  execution,  and  have  not  always  recognized  the  limitations 
within  which  we  can  extrapolate  experience  from  one  kind  of  activity  to 
the  other. 


But  there  is  more  to  the  matter  than  this.  For  it  is  demonstrably  true 
that  gains  in  our  knowledge  of  Nature  as  new  and  fundamental  and  unex- 
pected as  any  in  the  world  can  come,  unbidden,  from  the  investigations  of 
great  teams  for  research  and  development  in  many  areas.  As  our  resources 
for  team  research  grow  in  the  coming  years,  we  can  properly  expect  the 
rate  at  which  such  new  knowledge  is  revealed  to  increase  also — if  not 
proportionately,  at  least  very  substantially.  And  so  we  should  not  fail  to 
ask  an  implied  question  of  great  importance.  The  philosophy  that  envisaged 
the  environment  of  brilliant,  original,  unfettered  individual  research  as  the 
milieu  in  which  the  great  new  directions  of  scientific  thought  were  born  and 
nourished,  the  philosophy  which  has  had  such  confirmation  in  recent 
scientific  history,  was  itself  developed  in  the  days  of  scarcity  in  science — 


16  CARNEGIE     INSTITUTION     OF     WASHINGTON 

scarcity  not  only  of  material  wealth,  but  especially  scarcity  of  scientific 
workers.  Now  we  live  and  work  in  a  nation  committed  to  an  unparalleled 
rate  of  growth  in  the  material  resources  for  research,  and  in  a  world  in 
which  perhaps  eighty  per  cent  of  all  the  scientists  who  have  ever  lived  are 
our  contemporaries.  Is  it  possible  that  the  philosophy  itself  was  adjusted  to 
the  needs  of  other  times ;  that  it  is  not  relevant  to  an  era  of  plenty?  May  it 
actually  be  true  today  not  only  that  major  advances  in  new  knowledge,  the 
setting  of  radically  new  scientific  directions,  can  be  achieved  in  the  environ- 
ment of  great  and  highly  organized  research  teams,  but  also  that,  in  practice, 
such  environments  are  indeed  essential,  or,  at  any  rate,  the  most  favorable, 
to  the  process?  Is  it  possible  that  we  are  witness  to  a  profound  revolution 
in  the  very  character  of  research  itself?  Is  it  possible  that  the  small  and 
mobile  groups  to  which  we  earlier  looked  for  some  of  the  most  significant 
scientific  innovations,  the  groups  which  in  the  past  characteristically  had 
an  influence  on  scientific  progress  out  of  all  proportion  to  their  numbers  or 
their  social  cost,  can  no  longer  in  our  day  provide  such  significant  approaches 
to  the  unknown? 

Such  a  radical  query,  of  course,  bears  profoundly  on  the  whole  philosophy 
of  research.  It  is  far  more  than  a  practical  question.  It  touches  some  of  the 
deepest  wellsprings  of  scientific  faith.  It  touches  belief  in  the  very  nature 
and  effectiveness  of  the  individual  search  for  truth  in  our  time.  In  subtle 
ways  it  touches  on  the  nature  of  scientific  truth  itself.  It  is  an  important 
question  for  the  Carnegie  Institution,  deeply  committed  to  the  faith  that 
the  distinguished,  unfettered  individual  can  bring  unique  gifts  to  his 
society,  and  deeply  committed,  too,  to  belief  in  the  uniqueness  and  the  im- 
portance of  the  influence  which  a  community  of  independent  scholars  can 
exercise  on  scientific  progress. 

For  a  question  of  such  magnitude  and  gravity,  abstract  analysis  will  not 
suffice.  Contemporary  evidence  alone  can  give  convincing  answers.  Have 
the  recent  great  advances  in  our  knowledge  of  the  universe  and  of  our  own 
more  immediate  environment,  the  original  ideas  of  scientific  stature 
achieved  in  the  last  few  years  which  promise  to  open  truly  novel  avenues  of 
thought  for  the  future — have  these  been  necessarily,  or  even  primarily, 
associated  with  the  massive  programs  of  great  teams?  Or  do  the  basic 
contributions  of  small  and  mobile  research  groups  continue  in  our  day  to 
have  their  old  significance? 


Such  an  abundance  of  evidence  springs  to  mind,  provided  by  striking 
advances  no  more  than  a  half-dozen  years  old,  in  so  many  regions  of  scientific 
inquiry,  that  its  very  selection  poses  a  problem  and  must  necessarily  be 
arbitrary.  But  three  outstanding  areas  of  recent  investigation  are  particu- 


REPORT   OF   THE    PRESIDENT  17 

larly  interesting  to  consider  from  this  standpoint,  because  their  environ- 
ments and  circumstances  span  such  an  extraordinary  range  of  magnitude 
and  character  and  form. 

The  first  example  may  comprehend  that  immense  complex  of  research  and 
development  dedicated  to  the  placing  of  man  in  outer  space  and  ultimately 
on  the  moon  or  on  neighboring  planets,  its  present  great  achievements  in  our 
country  vividly  symbolized  by  the  voyages  of  Shepard  and  of  Grissom,  of 
Carpenter,  Glenn,  and  Schirra.  The  second  is  of  quite  a  different  kind.  It 
involves  an  achievement  in  astronomy  of  the  year  just  past  which  in  the  stag- 
gering distances  with  which  it  deals  emphasizes  anew  what  a  thin  terrestrial 
shell  is  the  outer  space  so  far  entered  by  man.  It  is  the  identification  of  what 
has  proved  to  be  by  far  the  most  remote  celestial  object  ever  discovered  in  the 
heavens — an  object  certainly  billions  of  light  years  distant  from  us — and 
the  measurement  of  the  redshift  of  its  spectrum.  The  third  selected  area  of 
advance  may  in  some  ways  be  the  most  profound  of  all,  though  it  is  far  from 
the  best  known.  It  includes  the  experimental  evidence  so  brilliantly  obtained 
in  the  last  few  years,  and  the  reasoning  directing  the  search  for  it,  indicating 
beyond  reasonable  doubt  that  the  information  governing  the  inheritance  of 
all  the  qualities  of  living  things  is  structurally  graven  on  the  chromosomes 
within  their  germ  cells  in  the  form  of  a  genuine  code.  It  includes,  as  a  climax, 
the  demonstration  of  the  general  nature  of  that  code,  which  the  year  just 
past  has  witnessed.  These  findings  may  well  mark  the  greatest  single  ad- 
vance in  genetics  since  the  demonstration  five  decades  ago  that  the  genes  of 
heredity  lie  in  the  chromosomes  in  a  linear  array. 

These  three  advances  in  natural  knowledge  bear  much  resemblance  in 
certain  fundamental  qualities.  All  have  won  important  and  striking  new 
knowledge.  In  all  of  them,  the  research  for  that  knowledge  has  included  a 
variety  of  scientific  disciplines  apparently  far  removed  from  the  main 
concern — in  the  case  of  the  third  as  far  removed  as  crystallography  seems  to 
be  from  conventional  genetics.  Profoundly  new  directions  of  thought  have 
resulted  from  all  three.  Possibly  the  third  has  produced  the  most  thoroughly 
revolutionary  new  insights.  The  first  has  brought  a  sense  of  liberating  con- 
quest and  a  wealth  of  first-hand  information  about  regions  known  hitherto 
only  palely  and  at  second  hand. 

But  in  many  features  of  the  modes  and  environments  of  research  char- 
acterizing them,  the  three  examples  diverge  about  as  much  as  scientific 
activities  can  differ.  The  contrast  is  particularly  vivid  when  cast  in  terms  of 
the  parameters  under  special  consideration:  the  relative  size  of  the  efforts, 
the  sheer  volume  of  human  and  material  sources  brought  to  bear,  the  kinds 
and  degrees  of  organization.  The  enormous  magnitude  of  the  space  program 
and  the  tremendous  cooperative  efforts  currently  involved  in  its  prosecution 
and  planned  for  the  future  need  little  emphasis.  In  this  respect,  indeed, 


18  CARNEGIE     INSTITUTION     OF      WASHINGTON 

Project  Apollo  is  much  in  the  tradition  of  a  Manhattan  Project,  though 
yet  bolder  in  both  variety  and  scale.  It  is  estimated  that  by  the  close  of  the 
budget  for  1963  the  National  Aeronautics  and  Space  Administration  will 
have  spent  more  than  four  thousand  millions  of  dollars  for  the  conduct  of 
research  and  development.  For  research  facilities  alone  it  will  have  expended 
more  than  eight  hundred  and  twenty  millions.  Behind  the  great  individuals 
who  have  manned  the  space  vehicles,  and  have  recorded  and  analyzed  the 
data  of  research,  and  who  will  do  so  in  the  future,  lie  the  years  of  develop- 
ment on  a  scale  of  unprecedented  magnitude  and  the  immense  organizations 
required  for  its  successful  prosecution.  Behind  the  fashioning  of  the  tools 
the  final  explorers  command  lie  combinations  of  highly  specialized  dis- 
ciplines and  intricate  techniques  of  the  most  varied  kind — chemical,  elec- 
tronic, mechanical — ranging  from  the  arts  of  propulsion  engineering  to 
those  of  miniaturization.  It  is  interesting  to  notice  in  this  connection  that 
the  cast  of  the  effort  at  present  is,  as  it  perforce  must  be,  importantly 
oriented  about  the  design  and  use  of  tools.  In  considerable  measure  it  is 
basically  an  engineering  effort — perhaps  the  most  exciting  and  compelling 
engineering  effort  of  this  century. 


Shortly  after  the  second  world  war,  when  instruments  of  radio  detection 
were  being  put  to  a  new  use  in  the  service  of  astronomy,  several  surveys  of 
the  skies  were  undertaken  to  detect  and  locate  the  positions  of  celestial 
bodies  that  were  emitters  of  radio  waves.  The  equipment  then  available, 
however,  was  relatively  poor  in  both  resolution  and  accuracy.  It  could  not 
effectively  complement  the  far  more  precise  tools  of  optical  astronomy. 
Resolution  and  precision  were  often  too  low  to  permit  a  reliable  identifica- 
tion of  radio  sources  with  corresponding  objects  observed  optically,  though 
sometimes  they  were  suspected  to  be  the  same.  As  the  techniques  of  radio 
astronomy  sharpened,  however,  as  larger  dishes  were  built  and  manned  and 
put  into  use,  both  penetration  and  resolving  power  improved  greatly.  At  the 
radio  observatory  of  the  Cavendish  Laboratory  in  England  and  at  the 
observatory  of  the  California  Institute  of  Technology  at  Bishop  in  the 
Owens  Valley,  instruments  of  outstanding  capacity  were  built.  During  1959 
and  1960  two  fresh  surveys  of  the  skies  were  undertaken  with  them:  in 
Cambridge  at  169  and  189  centimeters,  in  California  at  about  a  sixth  that 
wavelength  (31.2  cm).  In  the  course  of  these  surveys  the  celestial  positions 
of  certain  emitters  of  radio  waves  were  determined  with  a  new  precision.  So 
precise  was  the  location  of  one  of  these  objects,  indeed,  that  the  two- 
hundred-inch  Hale  telescope  could  be  brought  to  bear  upon  it.  The  peculiar 
color  characteristics  of  the  object  suggested  that  it  might  include  a  pair  of 
galaxies  in  collision,  and  so  might  be  expected  to  have  one  or  more  emission 


REPORT    OF   THE    PRESIDENT  19 

lines  in  its  spectrum.  And  so  it  happened  that  a  prescient  astronomer  of  the 
Mount  Wilson  and  Palomar  Observatories  was  able  to  obtain  two  spectra  of 
the  visible  light  from  this  source  and  to  measure  the  degree  of  redshift  in 
them.  At  the  same  time  another  observer,  obtaining  multicolor  photometric 
observations  of  two  of  the  fainter  galaxies  of  the  same  cluster  and  construct- 
ing their  curves  of  continuous  emission,  confirmed  this  measurement  of  red- 
shift.  It  corresponded  to  a  recession  velocity  of  nearly  half  the  speed  of 
light.  This  heavenly  body  defines  a  new  boundary  for  the  universe  compre- 
hended within  human  ken.  It  marks  by  far  the  most  searching  probe  into 
unplumbed  reaches  of  space  that  the  mind  and  hand  of  man  have  yet 
accomplished,  ranging  certainly  to  the  order  of  several  billion  light  years. 
When  it  is  recalled  that  a  single  light  year  amounts  to  almost  six  million 
million  miles — about  sixty-three  thousand  times  the  distance  of  our  own 
world  from  the  sun — it  makes  the  orbits  of  earth  satellites,  spectacular  as 
they  are,  yet  appear  as  comparatively  near-neighborhood  adventures. 

Perhaps  the  greatest  ultimate  significance  of  this  achievement  will  lie  in 
the  contribution  it  can  make  to  our  ideas  about  the  basic  nature  of  the 
universe.  Indeed,  this  newly  determined  point  of  distance,  so  far  beyond  any 
other  yet  obtained,  has  already  offered  suggestive  evidence  on  the  great 
question  of  whether  our  universe  is  a  continuously  expanding  one,  or  a 
universe  in  which  the  continuous  creation  and  destruction  of  matter  stand 
in  equilibrium,  or  whether  the  universe  in  fact  may  experience  alternate 
expansion  and  contraction  extending  over  astronomic  periods  of  time. 

In  sharp  contrast  to  the  first  example,  the  planning  of  these  observations, 
their  confirmation,  and  the  deductions  from  them  were  not  the  work  of  great 
teams  of  highly  coordinated  technical  workers.  These  were  the  fruits  of 
observations  and  calculations  made  by  a  few  individuals  laboring  in  relative 
solitude,  the  fruits  of  work  of  a  relative  handful  of  gifted  astronomers. 
Perhaps  never  in  science  has  the  work  of  individuals  been  more  clearly 
identifiable.  The  contrast  with  the  first  example  is  sharp. 

Yet  behind  this  classical  achievement  of  gifted  individuals  lay  many 
decades  of  research  and  engineering  focused  on  the  design  of  the  powerful 
modern  tools  of  optical  and  radio  astronomy.  Without  them  the  achieve- 
ment itself  would  have  been  quite  impossible.  These  tools,  like  those  in- 
volved in  the  space  effort,  were  the  products  of  hands  and  minds  and  toil  in 
literally  hundreds  of  specialized  skills.  And  it  was  not  skill  and  art  that  alone 
were  brought  to  bear,  but  with  them  the  magnificent  resources  of  intellect 
and  materials  and  time  and  research  that  gave  them  scope  and  effectiveness. 
The  achievement  itself  dramatically  underlines  how  significant  and  how 
essential  the  gifted  and  untrammeled  individual  investigator  is  today  on 
some  of  the  most  advanced  frontiers  of  the  physical  sciences.  It  was  pri- 
marily focused  on  the  gathering  and  the  interpretation  of  information  about 


20  CARNEGIE     INSTITUTION      OF      WASHINGTON 

nature,  not  on  the  design  of  tools.  Yet  its  success  depended  in  turn  on  a 
panoply  of  instruments  brought  to  perfection  in  other  times  and  other 
places,  the  development  of  which  had  required  a  structure  of  science  and 
technology  of  whose  cumulative  magnitude  and  scope  no  scientist  of  an 
earlier  generation  could  have  had  the  faintest  dream. 


The  third  example  embodies  yet  a  different  pattern.  It  would  be  hard  to 
imagine  a  more  fundamental  or  more  sweeping  discovery  than  one  elucidat- 
ing, at  a  deeper  level  than  had  hitherto  been  imagined,  the  manner  in  which 
the  information  governing  all  the  qualities  of  inheritance  may  be  recorded 
and  stored  in  the  chromosomes  of  plants  and  animals  and  men — stored  with 
such  extraordinary  effectiveness  and  such  enduring  stability  that  there  are 
organisms  living  today  whose  hereditary  characteristics  have  been  main- 
tained more  durably  than  the  very  rocks  within  whose  strata  the  fossils  of 
their  remote  ancestors  are  preserved.  Yet  in  terms  of  magnitude  the  human 
and  the  material  resources  committed  to  that  search,  by  comparison  with 
the  preceding  illustrations,  have  been  positively  minuscule. 

In  1953  Linus  Pauling  and  Robert  Brainard  Corey  at  the  California 
Institute  of  Technology  suggested  that  the  molecular  structure  of  the  unit 
of  heredity,  the  "molecule"  of  deoxyribonucleic  acid,  might  consist  of  chains 
of  polynucleotides  intertwined  in  the  form  of  a  helix,  with  four  characteristic 
bases,  the  purines  adenine  and  guanine  and  the  pyrimidines  thymine  and 
cytosine,  attached  to  them  and  projecting  outward,  while  phosphate  groups 
were  oriented  to  the  center.  There  were  features  of  this  model  which  con- 
flicted with  experimental  evidence,  notably  that  it  was  hard  to  reconcile  the 
fact  that  DNA  is  an  acid  with  the  existence  of  bases  lying,  as  it  were,  on  the 
outside  of  the  molecule.  But  the  model  involved  one  very  great  idea  which, 
though  it  was  not  widely  credible  in  terms  of  that  particular  construction, 
yet  was  to  prove  fundamental  to  all  further  thinking  on  the  matter.  It  was 
the  idea  that  the  biological  specificity  of  the  unit  of  DNA,  on  which  its 
power  of  determining  inheritance  must  rest,  must  inhere  in  the  sequence  of 
occurrence  of  these  bases  along  the  molecular  chain  and  the  suggestion  that 
the  periodic  distances  at  which  these  bases  occur  might  be  of  the  right  order 
to  permit  them  to  order  the  sequence  of  amino  acids  in  the  construction  of  a 
protein.  This  was  a  most  important  foundation  upon  which  to  rear  what 
would  prove  a  truly  extraordinary  arch  of  reasoning.  But  for  long  even  the 
idea  that  the  nucleic  acid  structure  could  be  locally  specific  was  resisted.  Until 
that  idea  had  been  widely  accepted,  its  more  detailed  consequence  could 
hardly  gain  effective  credence.  Both  these  developments  were  made  possible 
by  a  second  great  idea,  which  might  be  likened  to  a  keystone  of  the  arch. 

This  critical  idea  was  provided  by  J.  D.  Watson  when,  in  a  flash  of  insight 


REPORT   OF   THE    PRESIDENT  21 

reminiscent  of  Kekule's  vision  of  the  structure  of  the  benzene  molecule  that 
came  to  him  in  a  London  bus  almost  a  hundred  years  ago,  he  imagined 
the  consequences  of,  in  effect,  turning  the  model  inside  out,  pointing  the 
bases  inward,  and  pairing  the  purine  molecules  with  the  smaller  pyrimidines. 
Highly  significant  correspondences  with  nature  were  achieved  by  this 
remarkable  insight.  The  first  and  fundamental  rule  of  the  composition  of 
deoxyribonucleic  acid,  namely  that  it  incorporates  purines  and  pyrimidines 
in  equal  ratio,  was  given  a  rational  basis.  And  the  contradiction  between 
the  acidic  nature  of  DNA  and  its  presumed  outwardly  pointing  bases,  which 
had  plagued  the  model  of  Pauling  and  Corey,  was  resolved.  But  there  were 
impressive  difficulties  to  be  met  also.  The  idea  that  the  bases  were  outward- 
pointing  had  not  resulted  simply  from  neglecting  the  alternative  that  they 
might  point  inward.  That  possibility,  indeed,  had  been  carefully  examined  in 
formulating  the  earlier  model.  But  it  had  been  concluded  that  such  a 
structure  was  not  possible.  For  the  new  model  to  be  convincing,  the  physical 
possibility  of  such  an  arrangement  had  to  be  demonstrated,  and  the  details 
of  the  linkages  between  the  purines  and  pyrimidines  had  to  be  worked  out — 
formidable  tasks  requiring  concepts  and  techniques  familiar  to  those  dealing 
with  the  structure  of  crystals. 

And  so  it  was  that,  also  in  1953,  Watson  and  F.  H.  C.  Crick,  working  in 
the  Molecular  Biology  Unit  of  the  British  Medical  Research  Council 
adjacent  to  the  Cavendish  Laboratory  at  Cambridge,  announced  their 
brilliant  hypothesis  of  the  structure  of  the  unit  of  heredity,  of  the  "molecule" 
of  deoxyribonucleic  acid,  as  a  pair  of  "ribbons"  wound  in  the  form  of  a 
double  helix  around  a  common  axis  and  linked  by  the  four  bases,  the  purines 
adenine  and  guanine  and  the  pyrimidines  thymine  and  cytosine,  paired  in  a 
highly  specific  fashion.  The  model  of  Pauling  and  Corey  had  suggested  that 
the  bases  could  not  be  packed  in  the  center  of  the  molecule.  The  new  model 
proved  that  indeed  they  could,  and  from  that  demonstration  came  perhaps 
the  most  significant  idea  in  the  whole  chain — the  concept  of  base  pairing 
itself,  and  with  it  the  associated  and  important  notion  that  a  maximum  of 
four  kinds  of  base  pairs  could  be  involved.  The  beauty  and  credibility  of  the 
model  gave  firmness  and  emphasis  to  the  earlier  idea  that  the  biological 
specificity  of  the  unit  of  heredity  must  derive  in  large  measure  from  the 
ordering  of  the  pairs  of  bases  along  the  chain  of  the  deoxyribonucleic  acid. 

All  together,  three  biological  consequences  stemmed  directly  from  the 
model,  which  must  rank  among  the  most  important  advances  of  our  age  in 
the  understanding  of  the  fundamental  nature  of  earthly  life.  First,  the  model 
allowed  the  extraordinary  phenomenon  of  the  replication  of  the  genetic 
pattern  which  occurs  at  every  division  of  every  living  cell — the  mechanism 
fundamental  to  the  very  process  of  the  growth  and  multiplication  of  life  on 
earth — to  be  understood  consistently  for  the  first  time.  Second,  the  nature 


22  CARNEGIE     INSTITUTION     OF      WASHINGTON 

of  the  phenomenon  of  the  sudden  changes  in  inheritance  which  we  call 
mutation,  intensively  studied  since  the  days  of  de  Vries  but  never  under- 
stood in  their  fundamental  molecular  mechanisms,  now  for  the  first  time 
became  comprehensible  at  that  level,  in  terms  of  known  changes  in  bases 
which  could  result  in  alterations  of  their  sequence  to  produce  such  changes. 
Third,  and  greatest  of  all,  perhaps,  was  the  full  rationalization  of  the  key 
concept  that  biological  specificity  in  inheritance  must  in  large  part  derive 
from  the  sequential  ordering  of  the  bases  in  the  nucleic  acids. 

This  third  great  consequence  was  to  lead  to  a  scientific  vision  of  new  and 
unexpected  dimensions.  That  vista  was  provided  by  the  idea  that  genetic 
information  might  in  fact  be  coded  in  the  DNA  molecule  in  the  form  of  a 
linear  message  for  which  the  four  permissible  combinations  of  bases  might 
serve  as  alphabet,  in  a  manner,  indeed,  reminiscent  of  the  coding  of  a 
message  on  the  punched  tape  of  a  computer.  This  radical  concept  was  first 
examined  in  detail  by  the  astrophysicist  Gamow  in  1954.  Although  the 
precise  form  of  the  code  suggested  at  that  time  has  since  proved  incorrect, 
the  basic  idea  has  become  established  as  one  of  the  great  theoretical  ad- 
vances in  our  view  of  the  nature  of  the  living  world.  And  so  was  posed  the 
pointed  question:  if  such  a  code  exists,  what  is  its  specific  nature? 

It  is  that  question  which  theoretical  and  experimental  work  of  the  past 
two  years  has  done  much  to  answer.  An  important  share  of  the  answer,  like 
the  original  question,  has  come  once  again  from  the  laboratory  of  the  Unit 
for  Molecular  Biology  at  Cambridge ;  other  critical  parts  have  followed  from 
several  American  university  laboratories,  from  the  National  Institutes  of 
Health,  from  the  Carnegie  Institution  of  Washington.  Suffice  it  to  say  that 
preponderant  evidence  suggests  that  the  code  employs  words  containing 
very  few  "letters,"  probably  not  more  than  three. 

A  virus  may  include  within  its  single  chromosome  something  of  the  order 
of  a  hundred  thousand  base  pairs.  A  billion  pairs  of  bases  may  be  included 
within  the  total  store  of  information  of  our  own  chromosomes.  It  is  a 
startling  concept  that  if  the  DNA  strands  from  all  the  cells  in  a  single  human 
body  were  uncoiled  their  total  length  might  well  span  the  solar  system. 
There  is  ample  opportunity  for  diversity  in  the  ways  that  the  elements  of 
the  code  can  be  combined. 

With  this  conceptual  advance,  carrying  the  implication  that  one  of  the 
basic  challenges  offered  by  the  problem  of  heredity  might  lie,  in  effect,  in  the 
decoding  of  a  script,  progress  in  meeting  that  challenge  has  come  with 
remarkable  speed.  What  may  well  prove  to  be  a  Rosetta  stone  has  been 
provided  by  the  development  of  methods  of  accomplishing  protein  synthesis 
in  cell-free  systems  under  the  influence  of  artificial  ribonucleic  acids  com- 
posed of  only  two  bases  in  known  ratios  and  therefore  containing  specified 
code  words  in  known  frequencies.  The  composition  of  the  resulting  protein 


REPORT   OF   THE    PRESIDENT  23 

should  yield  the  key  to  code  " letters"  in  terms  of  the  ratios  of  specific  amino 
acids  corresponding  to  them.  Another  highly  promising  approach  involves 
techniques  for  investigating  the  coupling  between  the  base-pair  patterns  of 
the  deoxyribonucleic  acid  of  an  organism  and  the  "messenger  RNA"  of 
related  forms,  which  may  differ  in  their  coding  only  in  relatively  minor,  but 
specific  and  determinable,  particulars.  The  current  year  sees  work  of  this 
kind  at  a  peak  of  activity.  With  wing-swift  speed,  a  whole  new  area  in  our 
understanding  of  the  basic  mechanisms  of  heredity  at  the  molecular  level  is 
being  exploited. 


Here,  then,  are  three  genuinely  great  advances  marking  the  technical  and 
scientific  progress  of  the  last  three  years.  In  a  profound  sense  all  three  are 
typical  of  their  age,  and,  for  a  variety  of  reasons,  could  not  have  occurred  at 
any  earlier  time.  Obviously  neither  space  exploration  nor  the  astronomical 
investigations  of  the  new  "edge  of  the  universe"  now  within  our  ken  could 
have  been  achieved  with  the  tools  of  any  other  era.  The  peculiar  modernity 
of  the  third  example  involves  especially  a  yet  different  circumstance.  For  the 
very  idea  that  the  information  of  inheritance  may  be  recorded  as  a  code  is 
peculiarly  consonant  with  our  age — perhaps  so  characteristic  that  it  should 
be  treated  with  a  caution  doubled  by  this  very  fact.  In  the  nascence  of 
primitive  biological  thought  fire  was  a  living  thing,  dangerous  and  bright, 
and  the  expression  "vital  fires  within  us"  remains  to  remind  us  how  much  we 
once  thought  of  life  as  the  "inhabiting  property"  of  something  that  was 
obviously  dynamically  alive.  In  an  age  when  the  frontiers  of  engineering 
exploration  concerned  pumps  and  hydraulics  the  mechanism  of  the  circula- 
tion of  the  blood  was  a  fascinating  and  fertile  subject  of  physiological 
speculation  and  of  physiological  research.  For  the  age  of  Descartes,  strings 
and  pulleys  provided  compelling  images  for  the  mechanisms  of  life,  and 
images  of  clockwork  for  the  mind.  In  the  early  nineteenth  century,  domi- 
nated by  the  vision  of  steam  power  engineering,  energy  transformations 
seemed  among  the  most  important  aspects  of  life,  and  the  rise  of  large-scale 
electrical  power  engineering  in  the  latter  part  of  the  nineteenth  and  the 
early  twentieth  century  reinforced  the  vision.  Then,  in  our  own  era,  with  its 
emphasis  on  small-current  engineering  and  the  modulated  control  of  gigantic 
mechanical  and  electrical  processes,  the  aspects  of  living  processes  included 
under  the  rubric  of  Cybernetics  have  occupied  a  center  of  the  stage.  Studies 
of  those  fascinating  properties  of  living  systems  involving,  in  all  their  varied 
and  exquisitely  elaborate  mechanisms,  the  maintenance  of  homeostasis,  the 
preservation  of  balance  in  dynamic  systems,  have  held  a  special  attraction 
for  our  time.  And  in  our  immediate  day,  when  communication  of  new  orders 
of  content  and  of  speed,  and  with  it  the  massive  processing  of  information, 


24  CARNEGIE     INSTITUTION     OF      WASHINGTON 

so  dominates  our  lives,  when  we  are  inevitably  so  much  concerned  with  the 
coding  of  information  and  the  unraveling  of  such  codes,  it  is  scarcely 
surprising  that  a  natural  process  operating  upon  those  principles,  which  has 
evidently  been  central  to  the  evolution  of  all  life,  as  no  doubt  it  was  also  in 
its  origin,  should  only  now  have  so  powerfully  focused  our  attention  as  to  be 
on  the  threshold  of  solution.  It  follows,  too,  that,  just  as  each  of  the  earlier 
interpretations  of  living  processes  subsequently  gave  central  place  to  its 
successor  but  left  the  residue  of  its  own  unalterable  truth  to  contribute 
permanently  to  our  basic  understanding,  we  must  be  prepared  to  accept — 
and  indeed  to  welcome — the  same  fate  for  the  concept  of  genetic  coding. 


The  likenesses  uniting  these  three  examples,  then,  lie  deep.  It  would  be 
hard  to  select  the  most  significant  among  them,  though  in  the  achievement 
of  particular  new  insights  the  second  and  especially  the  third  may  pre- 
dominate. What  now  of  the  parameters  of  scale,  of  magnitude  of  the  re- 
sources committed,  of  the  extent  of  organization  of  the  work,  as  criteria  of 
its  significance?  Here  it  would  be  difficult  to  imagine  wider  contrasts. 

At  every  point  in  the  extraordinary  conceptual  development  that  marks 
the  third  example,  the  commitment  to  it  in  terms  of  numbers  of  workers,  in 
terms  of  material  resources,  was  extraordinarily  modest.  The  Unit  for 
Molecular  Biology  of  the  Medical  Research  Council  at  Cambridge  began 
with  two  crystallographers.  Ten  years  later,  when  its  revolutionary  dis- 
coveries were  well  launched,  it  numbered  perhaps  a  dozen  workers  and  was 
housed  in  a  temporary  building  behind  the  Cavendish  Laboratory  and  in 
various  University  rooms — a  very  minimum  of  space.  It  was,  indeed, 
superbly  instrumented  for  its  task.  But  such  instrumentation  was  in- 
credibly modest  in  both  mass  and  cost  compared  with  that  required  in 
either  of  the  other  fields.  In  that  free  and  flexible  atmosphere,  built  about 
the  largely  unfettered  efforts  of  a  few  gifted  individuals  working  within  a 
minimum  of  formal  organization,  have  been  made  some  of  the  most  im- 
portant advances  in  man's  concept  of  his  world  and  of  himself  possible  to 
the  twentieth  century.  It  is  striking  to  compare  this  situation  with  that  in 
which  the  exploration  of  space  must  go  forward. 


This,  then,  is  the  character  of  the  contemporary  evidence.  Such  contrasts 
of  size  and  structure  and  organization  in  the  modes  of  some  of  the  most 
significant  assaults  on  the  frontiers  of  natural  knowledge  in  this  decade 
strongly  suggest  that  these  parameters,  broadly  considered,  bear  little 
direct  relation  to  their  scientific  significance.  They  inspire  compelling  re- 
flections about  the  continuing  effectiveness,  in  our  own  day,  of  the  scale  and 


REPORT   OF   THE    PRESIDENT  25 

the  pattern  and  the  philosophy  of  research  to  which  the  Carnegie  Institution 
is  so  deeply  committed.  It  seems  abundantly  clear  that  the  essential  qualities 
and  requirements  of  inquiry  at  the  very  frontiers  of  man's  knowledge  of  his 
universe  do  not  now,  and  in  all  probability  will  not  in  the  foreseeable  future, 
differ  significantly  from  those  of  our  classical  scientific  past.  Such  inquiry 
will  surely  continue  to  bear  the  unmistakable  stamp  of  the  gifted  and  un- 
trammeled  individual,  whatever  may  be  the  scale  of  resources,  in  knowledge, 
in  tools,  in  human  and  material  support,  which  he  may  require. 

Bronowski  has  pointed  out  that  perhaps  the  most  fundamental  discovery 
of  the  scientific  age  was  that  Nature  was  to  be  approached  and  won,  not  by 
attempting  to  outwit  her  by  magic,  as  many  a  medieval  alchemist  had 
imagined  reflecting  a  prevailing  climate  of  his  time,  but  rather  by  discover- 
ing the  true  quality  of  natural  laws  and  taking  care  to  work  within  them.  It 
is  easy  to  forget  how  tremendous  was  that  change  of  view,  how  much  of  trial 
and  vision  was  comprehended  within  Newton's  simple  admonition  that 
"science  must  be  kept  free  from  occult  influences."  The  atmosphere  of  true 
research  is  still  as  it  was  when  that  great  advance  of  philosophy  was  made, 
still  the  atmosphere  in  which,  as  Lionel  Trilling  has  recalled,  Faraday  re- 
fused to  be  called  physicist,  holding  the  term  too  narrowly  imprisoning  a 
chamber  for  his  life's  commitment.  These  are  the  dimensions,  whatever  be 
the  nature  of  the  structures  in  which  they  are  embedded,  which  still  evoke 
the  great  advances  of  today. 


In  the  central  context  of  discovery,  it  seems  clear  that  the  magnitude  and 
organization  of  a  research  effort  may  be  the  least  meaningful  of  parameters 
in  any  fundamental  or  enduring  sense.  One  may  indeed  think  of  the  large 
and  the  small  research  enterprises  in  our  society  as  essentially  symbiotic, 
each  fulfilling  its  specific  role — one  more  example  of  the  rich  diversity  by 
which  we  live. 

The  relation,  however,  is  actually  more  subtle.  The  responsibility  that 
devolves  upon  small  and  mobile  groups  dedicated  to  the  exploration  of  new 
frontiers  is  clearly  greater  in  our  own  clay  than  merely  that  of  one  component 
in  a  many-hued  panoply  of  research.  At  least  one  aspect  of  the  relation  is  far 
more  serious,  and  wears  a  significance  which  must  inevitably  sharpen  further 
in  the  coming  years.  It  is  not  only  important  that  the  small  and  mobile  re- 
search group  be  maintained  and  strengthened  to  ensure  continuing 
advance  along  those  remote  boundaries  of  natural  knowledge  so  vital  to  our 
spiritual  as  well  as  to  our  material  well-being.  It  is  not  only  important  be- 
cause, in  such  a  massive  and  highly  advanced  technical  and  engineering 
society  as  our  own  is  today  and  must  even  more  become  tomorrow,  the 
scientific  "leverage"  of  such  pioneering  groups  must  inevitably  increase.  It 


26  CARNEGIE     INSTITUTION     OF      WASHINGTON 

is  a  further  and  a  significant  truth  that,  while  climates  that  foster  innovation 
can  be  maintained  in  the  midst  of  complex  and  highly  organized  technical 
undertakings,  preserving  them  intact  is  no  common  or  easy  achievement.  It 
requires  a  particular  determination,  an  extraordinary  persistence  of  vision 
and  pertinacity  of  will,  an  unusual  sensitivity  and  skill,  to  sustain  conditions 
favorable  to  original,  exploratory  research  on  remote  and  far-flung  frontiers 
of  the  mind  in  massive  working  environments  over  considerable  periods  of 
time,  undeflected  by  all  the  immediate  demands  that  architecting  to  known 
ends  in  those  environments  inevitably  imposes,  in  some  multiple  proportion 
of  intensity  to  scale.  Without  the  sustaining  view  that  small  and  mobile 
groups  attaining  great  discoveries  can  offer,  without  their  inspiration,  the 
task  must  become  doubly  difficult.  These  circumstances  may  define  for  the 
small  and  mobile  group  the  most  demanding  and  important  of  all  its 
functions — the  heavy  responsibility  of  the  keeper  of  a  vision — the  vision  of 
the  creating  individual. 

In  the  future  that  responsibility  may  well  become  not  only  wider  but  yet 
more  challenging.  For  it  is  abundantly  evident  that  science  and  technology, 
in  the  world  as  a  whole  as  well  as  in  our  own  nation,  have  entered  phases  of 
development  in  our  day  so  different  in  scale  and  complexity  from  their 
beginnings — or  from  what,  incidentally,  the  newly  developing  nations  of  the 
world  may  confront  or  may  require  in  their  own  immediate  futures — as  to 
differ  essentially  in  kind.  As  Pierre  Teilhard  de  Chardin  has  written  with 
sensitive  perception,  "The  Earth  is  covering  itself  not  merely  by  myriads  of 
thinking  units,  but  by  a  single  continuum  of  thought,  and  finally  forming  a 
functionally  single  Unit  of  Thought  of  planetary  dimensions."  An  important 
aspect  of  the  qualitative  growth  of  contemporary  science,  of  course,  inheres 
in  its  essentially  additive  nature,  in  the  formidable  integration  of  knowledge 
and  of  thought  characteristic  of  a  pursuit  where  discoveries  in  one  field  may 
in  the  span  of  a  few  months  alter  the  entire  basis  against  which  thinking  in 
very  different  areas  must  be  projected.  Another  concerns  almost  the  op- 
posite situation.  The  significance  of  great  research  is  largely  measured  by 
the  impact  of  its  results  over  a  wide  range  of  frontiers  of  inquiry,  demanding 
the  widest  and  swiftest  communication  possible  and  challenging  human 
intellectual  capacities  for  assimilation  and  generalization  to  their  limits. 
But  the  processes  of  research  bring  heavy  demands  on  quite  opposite 
qualities — on  extraordinarily  detailed  knowledge  of  a  single  field,  on  that 
supreme  mastery  of  all  its  coordinates  down  to  the  most  minute,  developed 
over  long  periods  of  years,  which  so  often  is  prerequisite  to  significant  and 
sustained  advance.  In  the  past,  science  has  been  able  to  reconcile  these  two 
quite  opposite  requirements  in  tolerable  fashion.  With  increase  of  scale  the 
problem  takes  on  new  dimensions. 

Science  in  the  last  decades  has  responded  to  the  challenge  with  enormously 
increased  sophistication,  with  vastly  expanded  organization  and  integration 


KEPORT   OF   THE    PRESIDENT  27 

of  knowledge,  with,  indeed,  quite  a  new  development  of  recent  years,  the 
field  of  research  on  research  itself.  But  as  science  has  matured  in  its  modes  of 
cultivating  the  whole  vast  field  of  its  thought,  as  its  power  has  grown  to  enter 
and  occupy  new  areas  of  research  in  force  so  soon  as  the  first  hint  of  them 
appears,  these  very  qualities  have  brought  novel  and  troubling  consequences 
for  the  gifted  individual,  particularly  for  the  gifted  young  research  student 
just  entering  upon  his  life's  work,  upon  whom  so  much  of  the  future  de- 
pends. As  A.  B.  Pippard,  among  others,  has  pointed  out  dramatically,  the 
legions  of  investigators  can  now  be  mobilized  with  such  speed  and  effective- 
ness at  a  new  and  attractive  breach  in  the  frontier  of  knowledge  that, 
particularly  if  the  area  offers  a  promise  of  practical  benefit,  a  green  and 
fertile  intellectual  valley  can  be  reduced  to  aridity  for  the  innovator  within 
less  than  the  working  life  of  a  generation  of  young  scientists.  The  conse- 
quences incident  to  such  swift  and  locustlike  invasions,  however  effective 
and  profitable  they  may  be  for  a  technical  society  in  the  large,  can  be  dis- 
couraging to  vulnerable  individuals,  and  they  bear  at  precisely  the  points  of 
talent  and  dedication  most  precious  to  us.  There  can  be  no  more  urgent 
imperative  than  the  creation  of  opportunity  for  individuals  faced  with  this 
dilemma  to  address  themselves  once  again  to  wholly  new  fields  of  inquiry. 
This  too  lies  peculiarly  in  the  domain  of  small  and  mobile  and  basically 
highly  uncommitted  research  groups. 


What,  in  final  essence,  is  the  deepest  meaning  of  the  scientific  way?  In  the 
profoundest  sense,  what  is  the  meaning  of  the  individual  human  life  dedi- 
cated to  it?  Within  the  scientific  context,  as  well  as  outside  it,  what,  at  last, 
are  people  for?  A  generation,  perhaps  even  a  decade,  ago  such  a  question 
was  all  but  unasked  by  most  Americans.  Certainly  it  was  all  but  unasked  in 
1902.  Even  if  put,  in  that  day,  it  would  have  appeared  to  many  not  only 
irrelevant  but  quite  possibly  sinister.  But  in  a  world  with  a  population 
estimated  at  nearly  three  billion  and  predicted  by  conservative  demog- 
raphers to  reach  almost  four  billion  by- 1980  and  to  attain  nearly  seven 
billion  by  the  turn  of  the  century,  the  question  wears  quite  a  different 
aspect.  In  our  own  nation,  with  a  population  now  over  one  hundred  and 
seventy  million  and  destined  perhaps  to  reach  two  hundred  and  twenty 
million  by  1975,  the  revolutionary  consequences  of  this  flood  tide  upon 
every  facet  of  the  world  we  know  demand  no  emphasis.  It  must  profoundly 
affect  every  circumstance  of  our  society,  of  its  organization  and  its  function. 
It  must  affect  the  individual's  inner  view  of  himself  and  his  conception  of  his 
relation  to  his  universe,  his  understanding  and  his  reach  in  his  own  physical 
world,  and  much  else  besides. 

The  rate  of  growth  of  the  scientific  effort  today  considerably  exceeds  that 
of  the  population  as  a  whole.  Inevitably,  it  would  seem,  it  must  change  after 


28  CARNEGIE     INSTITUTION     OF      WASHINGTON 

two  or  three  more  periods  of  doubling.  But  in  absolute  terms  it  would  seem 
beyond  reasonable  doubt  that  the  legions  of  technically  trained  people  in  the 
future  will  vastly  exceed  in  numbers  those  now  active,  even  as  these  in  turn 
so  vastly  exceed  the  numbers  of  only  a  few  decades  ago.  Great  technical  and 
engineering  efforts  will  be  ready  and  available  to  confer  rich  meaning  on  the 
lives  of  many.  In  massive  and  compelling  developmental  undertakings  op- 
portunities will  continue  to  be  provided  to  great  numbers  of  active  minds  to 
labor  for  ends  not  only  dramatic,  not  only  economically  and  socially  adap- 
tive, but  as  creative  and  as  meaningful  in  our  times  as  the  tasks  of  the 
builders  of  Chartres  or  of  the  Parthenon  must  have  been  in  theirs.  Pippard 
has  presciently  pointed  out  that,  if  the  field  of  technology  is  to  prove 
sufficiently  magnetic  to  attract  first-class  intellects  to  it,  opportunities  for 
the  dramatic  and  the  spectacular,  outlets  for  the  moral  impulse  to  share  in 
socially  significant  undertakings,  the  sheer  intellectual  quality  of  the  under- 
takings themselves,  must  provide  the  motivations.  Among  the  great  and 
challenging  technical  and  engineering  undertakings  of  our  time,  all  three 
motivations  are  presented  on  a  scale  the  world  may  never  have  experienced 
before. 

But  there  will  be  other  scientific  workers,  too,  of  other  and  less  specially 
identifiable  tastes  and  talents,  hostages  to  a  more  distant  future.  For  them 
the  requirements  will  be  quite  different.  Perhaps  the  deepest  question  the 
times  can  pose  for  them,  and  as  well  the  most  poignant  for  all  man's  spirit- 
ual welfare,  will  be  this.  In  a  society  as  densely  packed,  as  intricately 
organized,  as  highly  urbanized,  as  our  own  must  inevitably  become  in 
future  years,  can  small  and  mobile  enclaves  of  thoughtful  and  imaginative 
men  and  women  continue  to  maintain  integrity  and  distinctive  freedom 
within  the  greater  society?  On  their  ability  to  do  so  in  the  broadest  context 
will  depend  in  no  small  measure  the  fate  of  the  individual  and  of  those  goals 
and  motivations  through  which  in  the  past  we  have  lived  and  taken  our 
national  being.  In  a  very  real  sense  their  persistence  alone  can  effectively 
preserve  the  priceless  jewel  of  the  opportunity  for  quietness  and  temporary 
solitude  which  in  our  past  has  been  so  vital  a  nursery  for  individual  American 
greatness  as  well  as  for  that  of  our  society  as  a  whole.  For  it  is  the  gifted, 
unorthodox  individual  in  the  laboratory  or  the  study  or  the  walk  by  the 
river  at  twilight  who  has  always  brought  to  us,  and  must  continue  to  bring 
to  us,  all  the  basic  resources  by  which  we  live.  His  position  must  be  guarded 
and  honored  and  implemented  with  every  resource  that  we  can  muster,  now 
and  in  the  future,  for  he  is  irreplaceable.  This  matter  too,  and  all  the  circum- 
stances attendant  upon  it,  must  be  a  central  and  abiding  concern  through 
all  the  coming  years  for  the  Carnegie  Institution  of  Washington.  As  Chaucer 
said  six  hundred  years  ago,  so  may  we  today:  "Out  of  the  old  fields  cometh 
the  new  corn." 


The  Year  in  Review 

It  is  fascinating  to  compare  the  Institution  of  approximately  sixty  years 
ago  with  that  of  today.  There  was,  of  course,  very  little  to  report  from  the 
first  year  or  two  of  the  Institution's  existence,  which  was  spent  in  a  search 
for  profitable  lines  of  endeavor  and  experiments  with  organization  toward 
that  end.  As  early  as  1904,  however,  the  lines  the  Institution  was  to  follow 
for  some  years  were  discernible,  and  the  report  for  the  year  1905  (Year  Book 
4)  describes  the  nature  of  the  Institution's  work  in  nearly  all  of  the  broad 
fields  in  which  it  was  active  during  1961-1962.  Some  glimpses  of  these  early 
activities,  set  alongside  typical  activities  in  our  several  fields  for  1961-1962, 
give  a  most  illuminating  view  of  the  progress  of  the  Institution,  and  indeed 
of  science  in  the  United  States. 

In  1905  the  resources  and  objectives  of  the  Institution  were  much  more 
widely  dispersed  than  they  are  today.  The  total  budget  for  that  year  was 
$586,000,  a  little  more  than  half  of  which  was  allotted  to  ten  " Departments 
of  Investigations"  which  included  the  forerunners  of  all  the  Institution's 
present  fields  except  embryology.  Among  the  Departments  were  several  that 
have  since  been  terminated  (Marine  Biology,  Economics  and  Sociology, 
History,  Nutrition,  and  Horticulture).  Half  of  the  total  budget  for  the 
Departments  ($302,700)  went  to  the  Solar  Observatory  on  Mount  Wilson, 
which  was  under  construction  in  that  year.  In  addition,  43  individuals  or 
organizations  outside  the  Institution  received  grants  to  the  sum  of  $130,625 
in  the  fields  of  anthropology,  archaeology,  astronomy,  bibliography,  botany, 
chemistry,  geology,  history,  paleontology,  philology,  phonetics  and  linguis- 
tics, physics,  and  zoology.  The  Institution  also  had  in  1905  a  program  of 
subsidizing  outside  publications  of  "meritorious  works  which  would  not 
otherwise  be  readily  printed."  Nearly  $30,000  was  expended  in  1905  for  this 
purpose  and  for  the  publication  of  works  written  within  the  Institution 
itself. 

By  contrast  the  Institution's  budget  for  1961-1962  was  $2,848,480,  all  of 
which  was  spent  upon  the  six  operating  Departments  that  have  been 
maintained  in  recent  years.  Except  for  departmental  fellowships  the 
Institution  made  no  outside  grants  and  did  not  subsidize  publication  for 
works  written  outside  the  Institution.  While  a  great  variety  of  subjects  was 

29 


30  CARNEGIE     INSTITUTION      OF      WASHINGTON 

under  investigation  within  the  Institution  in  1961-1962,  research  was  under- 
taken in  a  better  organized  and  more  purposeful  manner. 

Four  of  the  more  promising  lines  of  research,  as  viewed  by  the  President 
and  Trustees  of  the  Institution  in  1905,  lay  in  the  work  of  its  Solar  Ob- 
servatory, in  its  Department  of  Terrestrial  Magnetism,  in  geophysical  re- 
search, and  in  biological  investigations.  With  rather  remarkable  perception 
the  importance  of  fundamental  research  in  the  physical  and  biological 
sciences  is  commented  upon  in  the  1905  report.  The  Solar  Observatory  is 
described  as  ranking  among  Institution  projects  "first  in  order  of  cost  for 
initial  construction  and  equipment.  This  cost,  however,  is  no  more  than 
commensurate  with  the  magnitude  of  the  problem  attacked.  .  .  ."  Of  the 
biological  investigations,  including  those  of  the  Station  for  Experimental 
Evolution  and  the  Desert  Botanical  Laboratory,  which  was  the  predecessor 
of  the  Department  of  Plant  Biology,  the  report  noted  that  fundamental 
research  in  plant  and  animal  biology  "for  a  series  of  years  can  hardly  fail  to 
yield  results  of  signal  practical  and  theoretical  value." 


The  Department  of  Genetics 

In  our  series  of  "then  and  now"  snapshots  it  is  appropriate 

to  begin  with  the  Department  of  Genetics,  whose  prede- 
1905         cessor  in  1905  was  the  Station  for  Experimental  Evolution, 

one  of  the  most  active  parts  of  the  Institution  in  that 

year. 
Even  though  the  Station  for  Experimental  Evolution  at  Cold  Spring 
Harbor  had  been  in  existence  for  only  a  little  more  than  a  year,  a  year  of 
very  full  activity  was  reported.  Following  the  inspiration  of  Hugo  de  Vries, 
who  had  given  the  dedication  lecture  at  the  Station  the  year  before,  C.  B. 
Davenport  described  the  long-range  objectives  of  the  Station's  work.  "The 
factors  of  evolution  are  three — variation,  inheritance,  and  adjustment. 
Studies  may  be  made  on  any  one  of  these  factors  or  on  all  three  together; 
as  a  matter  of  fact,  they  can  hardly  be  studied  wholly  independently. 
.  .  .  Since  studies  in  inheritance  have  been  relatively  neglected.  .  .  our  first 
efforts  have  been  directed  primarily  toward  such  studies."1  Already  five 
principal  investigators  and  the  Director,  Dr.  Davenport,  had  commenced 
their  programs  of  research. 

From  a  modern  point  of  view  the  range  of  the  work  undertaken  was 
astonishing.  It  was  described  as  "investigations  into  inheritance  and 
variability"  of  plants,  insects,  and  other  invertebrates;  "investigations  upon 

1  Year  Book  4,  p.  87. 


REPORT    OF   THE    PRESIDENT  31 

aquatic  vertebrates";  "studies  on  inheritance  in  domesticated  animals"; 
and  "investigations  into  the  cytological  basis  of  heredity."  Experiments 
were  in  progress  on  eight  beetle  species,  three  species  of  moth,  flies,  aphids, 
crickets,  bees,  and  snails  (Helix  nemoralis).  The  brown  trout  and  several 
killifishes  (Fundulus  sp.)  were  studied,  and  the  Station  experimented  with 
goats,  sheep,  and  cats.  During  the  year  George  H.  Shull  became  well  started 
on  the  research  which  led  to  his  later  valuable  knowledge  of  maize  re- 
production. But  in  1905  he  was  searching  for  suitable  material  for  experi- 
ment, and  had  a  garden  of  81  different  species  of  biennials,  perennials,  and 
annuals.  Along  with  this  search  he  conducted  a  variety  of  experiments, 
which  included  investigation  of  the  inheritance  of  seed  weights  in  beans 
(repeating  W.  Johannsen's  experiments)  and  the  vegetative  habits  of 
Russian  sunflowers  (Helianthus  annuus)  and  other  species.  He  had  also 
begun  his  observation  of  the  characteristics  of  maize.  The  particular 
character  chosen  for  study  in  1905  was  the  number  of  rows  on  the  maize  ear. 
Although  the  importance  of  cytological  research  was  recognized,  the 
year's  effort  failed  to  devise  even  a  suitable  experiment.  The  report  ob- 
served, "The  results  of  the  last  three  years  confirm  the  belief  in  the  im- 
portance of  the  chromatic  material  in  inheritance.  This  chromatic  material 
exhibits  a  bewildering  complexity  and  diversity  scarcely  less  than  that  of 
adult  organisms."2 

It  is  interesting  to  find  in  1961-1962  two  lines  of  investi- 
gation which  were  at  a  germinal  stage  in  1905.  Experiments 
1961-1962       with  maize  are  still  productive  of  fundamental  results,  and 
cytological  research  using  flies  (now  the  familiar  Drosophila) 
formed  an  important  part  of  the  departmental  program. 
Thus  in  one  way  or  another  these  lines  have  held  some  of  the  departmental 
attention  for  more  than  56  years. 

The  approach  of  the  Department  in  1961-1962,  however,  was  a  vastly 
different  enterprise.  In  a  sense  Barbara  McClintock's  methods  of  working 
with  maize  genes  are  lineal  descendants  of  the  variation  and  inheritance 
techniques  that  Shull  was  commencing  to  pioneer  by  counting  rows  of 
kernels  on  ears.  But  in  Dr.  McClintock's  hands  these  methods  have 
become  highly  sensitive  and  one  of  the  sharpest  tools  in  modern  genetics. 
She  has  made  them  a  match  for  other  sharp  new  tools  heavily  dependent  on 
chemistry  and  physics.  For  more  than  a  dozen  years  she  has  been  interested 
in  the  elements  associated  with  genes  that  activate,  control,  suppress,  or 
regulate  genie  action.  Her  work  during  these  years  has  revealed  the  presence 
in  maize  of  two  controlling  systems,  an  Activator  (Ac)  system,  whose  pres- 
ence or  absence  is  associated  with  the  appearance  or  nonappearance  of 
mutations  of  a  particular  gene,  and  the  Suppressor-mutator  system  (Spra), 

2  Year  Book  4,  P-  94. 


32  CARNEGIE     INSTITUTION      OF      WASHINGTON 

which  causes  a  varied  expression  of  the  action  of  a  single  gene  as  observed 
in  somatic  cells.  In  her  research  this  observation  has  been  associated 
especially  with  the  appearance  of  the  reddish-blue  pigment  anthocyanin. 
Depending  on  its  phase,  the  Suppressor-mutator  element  may  either  inhibit 
or  activate  the  gene  expression  which  results  in  the  formation  of  antho- 
cyanin in  maize  leaves  or  kernels. 

A  second  theme  of  Dr.  McClintock's  work  through  these  years  has  been  a 
search  for  evidence  that  even  the  fine  structure  of  inheritance  is  basically 
similar  for  all  forms  of  life.  In  a  much  more  general  way  Davenport  and 
others  started  with  the  same  hypothesis  at  the  Station  for  Experimental 
Evolution,  attempting  to  observe  genetic  expression  in  many  forms  of  life. 
Dr.  McClintock's  first  experimental  evidence  on  the  similarity  of  operation 
of  genie  control  elements  in  different  forms  of  life  was  reported  in  1950. 3 
In  that  year  she  observed,  " Because  the  same  types  of  mutability  as  those 
observed  in  maize  have  been  described  for  a  wide  variety  of  organisms,  it  is 
probable  that  the  same  events,  involving  the  same  chromosome  materials, 
may  occur  in  all  organisms.4 

During  the  year  1961-1962  Dr.  McClintock  continued  to  examine  the 
parallels  between  the  gene-control  systems  in  maize  and  bacteria.  She 
observes  in  her  report  that  both  organisms  have  gene-control  systems 
composed  of  an  " operator"  element  directly  controlling  genie  activity  ad- 
jacent to  the  structural  gene  and  a  "regulator"  element  acting  upon  the 
operator  element.  Other  investigators  have  shown  that  the  position  of  the 
regulator  element  on  the  bacterial  chromosome  may  differ  for  individual 
systems.5  It  may  be  near  to  or  removed  from  the  locus  of  the  operator 
element.  Dr.  McClintock's  work  during  the  year  confirmed  her  hypothesis 
that  there  is  a  high  probability  that  genie  control  systems  in  maize  and 
bacteria  act  in  similar  fashion.  She  concludes  her  report  by  stating  that 
her  findings  "are  sufficiently  extensive  to  leave  no  doubt  that  a  two-element 
system  of  control  of  gene  action,  composed  of  an  operator  element  at  the 
locus  of  the  gene  and  a  regulator  element  located  elsewhere,  may  arise 
at  a  gene  locus  that  initially  carried  the  regulator  of  the  system."  It 
would  appear  that  one  more  link  has  thus  been  added  to  the  gradually 
extending  chain  of  evidence  on  basic  similarities  for  many  forms  of  life  at 
the  cellular  level. 

A  second  field  of  departmental  interest  in  1905  survived  to  1961-1962. 
This  was  the  application  of  cytology  to  genetics,  which  was  considered,  but 
only  futilely  explored,  in  1905.  Indeed,  successful  development  of  this  field 

3  Proceedings  of  the  National  Academy  of  Sciences,  36,  344-355,  1950. 

4  Year  Book  49,  p.  165,  1950. 

6  F.  Jacob  and  J.  Monod,  On  the  regulation  of  gene  activity,  Cold  Spring  Harbor  Symposia  on 
Quantitative  Biology,  26,  193-209,  394-395,  1961,  presented  completely  for  the  first  time  evidence 
on  the  operator  and  regulator  elements  in  bacteria. 


REPORT   OF   THE    PRESIDENT  33 

actually  was  postponed  for  more  than  15  years  after  1905,  when  in  the 
1920's  the  work  of  John  Belling  finally  laid  the  foundations  for  modern 
cytogenetics.  This  work  was  continued  in  1961-1962  in  the  research  of 
Berwind  P.  Kaufmann,  Helen  Gay,  Margaret  McDonald,  and  their  associ- 
ates. The  general  objectives  of  the  group  bore  some  resemblance  to  the 
crudely  stated  convictions  about  the  importance  of  cytology  in  the  1905 
report.  The  group  continued  its  work  of  nearly  two  decades,  charting  the 
changes  occurring  in  the  organization  of  chromosomes  and  cytoplasmic 
organelles  as  cells  in  higher  organisms  grow  and  differentiate.  Their  methods, 
however,  were  a  world  apart  from  those  of  1905,  including  as  they  did 
electron  microscopy,  fluorescent  microscopy,  enzyme  chemistry,  and  bio- 
chemically specific  stains.  In  addition,  they  had  at  their  disposal  the  vast 
knowledge  that  has  accumulated  over  40  years  on  the  genetic  characteristics 
of  Drosophila  flies,  which  continued  to  be  one  of  the  objects  of  their  ob- 
servations. A  second  material  for  study  has  been  the  plant  Tradescantia 
(spiderwort  family),  which  offers  a  very  favorable  opportunity  for  cyto- 
plasmic study  during  microsporogenesis.6  Of  particular  interest  has  been 
the  effort  of  this  group  to  approach  the  problems  of  charting  the  submicro- 
scopic  organization  of  chromosomes  by  means  of  "enzymatic  dissection." 

All  these  techniques  were  employed  during  the  year,  adding  to  the  results 
obtained  in  other  years.  Experiments  were  conducted  on  the  mutagenic 
properties  of  deoxyribonuclease  when  introduced  into  Drosophila.  An 
enzyme  analogue,  5-bromodeoxyuridine,  was  added  to  the  list  of  mutagenic 
agents  employed  on  both  Drosophila  and  Tradescantia.  Perhaps  the  most 
interesting  results  from  this  group's  program  during  the  year  were  two 
discoveries:  (1)  The  finding  that  direct  chromosomal  breakage  occurs  in 
Tradescantia  root  tips  in  the  presence  of  5-bromodeoxyuridine.  This  enzyme 
analogue  acts  by  modifying  the  base  sequences  in  nucleic  acid  rather  than 
the  phosphate-sugar  helices  attacked  by  deoxyribonuclease.  (2)  The 
observation  that  Golgi  bodies,  one  of  the  types  of  cytoplasmic  organelle, 
exhibit  different  forms  in  the  progression  of  microsporogenesis  in  Trades- 
cantia. 

A  third  activity  important  to  the  1961-1962  Department  was  not  even 
dreamed  of  in  1905.  It  is  represented  in  the  work  of  Alfred  D.  Hershey  and 
his  associates,  who  are  gradually  charting  the  molecular  structure  of  the 
viral  chromosome.  Dr.  Hershey's  work  illustrates,  more  than  anything  else 
in  the  Department,  the  observation  made  by  M.  Demerec  as  early  as  1942 
that  "From  the  purely  biological  science  of  early  days,  genetics  has  de- 
veloped into  a  science  where  cooperation  with  physics,  chemistry,  and 
mathematics  is  essential."7  Hershey  and  his  associates  observe  in  their 

6  Microspore  =  pollen. 

7  Year  Book  41,  p.  171. 


34  CARNEGIE     INSTITUTION      OF      WASHINGTON 

report  of  this  year  that  methods  have  been  devised  in  recent  years  to 
characterize  and  differentiate  among  different  types  of  deoxyribonucleic 
acid  (DNA)  molecules.  Among  these  methods  are  optical  analysis  of 
thermal  denaturation,  chromatographic  analysis,  measurement  of  fragility 
and  buoyant  density,  and  specific  enzymatic  tests.  But  these  tests  do  not 
give  information  about  molecular  structure,  which  remains  a  more  or  less 
"plausible  inference."  Hershey's  objective  is  to  remove  genetics'  dependence 
on  inference  for  its  concepts  of  molecular  structure  of  genetic  material.  To 
this  end,  he  and  his  associates  are  experimenting  with  the  DNA  of  several 
types  of  bacteriophage.8  He  considers  these  DNA's  to  be  favorable  material 
for  experiment  because:  (1)  they  can  be  isolated  in  a  molecularly  homo- 
geneous state,  permitting  correlation  between  structure  and  biological 
function ;  (2)  their  synthesis  can  be  studied  in  infected  cells  that  have  been 
proved  suitable  for  metabolic  study  in  the  past;  and  (3)  present  intensive 
study  of  the  genetics  of  a  few  bacteriophage  species  gives  valuable  refer- 
ence points  for  physical  and  chemical  findings.  He  considers  his  current 
work  at  least  in  part  "exploratory." 

Several  interesting  results  ensued  from  Dr.  Hershey's  exploration  of 
physical  techniques  in  measuring  molecular  weight  during  the  year.  In  one 
he  established  the  molecular  weight  of  the  DNA  of  a  bacteriophage  known 
as  T5  by  first  establishing  an  ingenious  pair  of  "scales"  by  analyzing  DNA 
fragments  of  another  phage  (T2).  One  scale  is  established  by  determining 
sedimentation  constants9  of  fragments  of  labeled  T2  DNA  as  separated  by 
column  chromatography.10  The  other  was  obtained  from  fragility  tests  that 
measured  the  rate  of  breakage  of  T2  DNA  fragments  of  a  given  sedimenta- 
tion coefficient  when  stirred  in  a  mixer  at  a  given  speed.  The  sedimentation 
coefficient9  and  the  fragility  index  of  T5  DNA  were  then  determined.  By 
comparison  with  the  T2  "scales"  a  molecular  weight  of  84  million  was 
determined.  The  T5  DNA  matched  very  closely  fragments  of  T2  DNA  in 
one  sedimentation  coefficient  range  (48.5-49.5). 

By  similar  techniques  Dr.  Hershey  also  brought  to  light  during  the  year 
some  interesting  molecular  characteristics  of  the  DNA  of  phage  lambda, 
which  was  found  to  have  astonishingly  different  molecular  properties  from 
other  well  known  DNA's.  Of  particular  interest  was  a  broad  range  of 
denaturation  temperatures,  like  that  of  bacterial  DNA's  and  contrasting 
with  an  exceedingly  narrow  range  typical  of  other  phage  DNA's.  On  one 
hand  these  and  other  properties  suggest  a  marked  tendency  of  the  molecules 
to  interact  with  each  other,  and  on  the  other,  a  remarkable  differentiation 
in  structure  along  their  lengths.   These  exceptional  properties  may  be 

8  Bacteriophage — any  of  a  number  of  intracellular  virus  parasites  of  bacteria. 

9  Measure  of  the  rate  of  precipitation  of  particles  in  suspension  in  a  solution  when  centrifuged. 

10  Chromatography — a  method  of  separating  and  analyzing  chemical  substances  by  inducing 
differential  migration  and  adsorption  from  solution  in  a  porous,  insoluble,  sorptive  medium. 


REPORT    OF   THE    PRESIDENT  35 

related  to  each  other  and  to  some  of  the  well  known  biological  peculiarities 
of  phage  lambda. 

By  infecting  bacteria  with  isotopically  labeled  phage  particles  and  by 
labeling  DNA  synthesized  in  the  bacteria  after  infection,  Dr.  Hershey  and 
Dr.  F.  R.  Frankel  have  determined  that  cells  subjected  to  such  infection 
always  contain  a  considerable  fraction  of  their  total  DNA  in  a  form  indis- 
tinguishable from  that  found  in  finished  phage  particles.  They  note  that 
this  points  to  a  mechanism  for  the  preservation  and  determination  of 
molecular  length  that  operates  continuously  during  DNA  replication,  not 
only  at  some  terminal  stage  in  the  formation  of  the  phage  particle.  This 
conclusion  is  considered  significant  evidence  bearing  upon  several  hypoth- 
eses about  genetic  mechanisms. 


The  Department  of  Plant  Biology 

The  Department  of  Plant  Biology  also  has  developed  from 
an  operation  under  way  in  1905.  The  Desert  Botanical 
1905  Laboratory  was  active  that  year,  located  at  Tucson,  Arizona. 
The  program  in  1905  was  not  as  varied  as  that  of  the  Station 
for  Experimental  Evolution.  Twelve  investigators  were 
associated  with  the  Laboratory  in  that  year,  most  of  them  as  recipients  of 
grants.  As  might  be  expected,  their  investigations  were  heavily  weighted 
toward  the  characteristics  of  arid-region  plants,  especially  transpiration11 
and  water-conducting  mechanisms.  A  substantial  amount  of  attention  was 
paid  to  the  character  of  plant  environment,  as  in  D.  T.  MacDougal's 
observations  of  soil  temperature  and  B.  E.  Livingston's  study  of  the 
relations  of  desert  plants  to  soil  moisture  and  evaporation.  More  typical, 
however,  was  F.  E.  Lloyd's  study  of  correlation  between  stomatal12  action 
and  transpiration  in  certain  types  of  desert  plants.  (No  positive  correlation 
was  observed.)  But  along  with  these  was  displayed  at  least  a  secondary 
interest  in  what  later  became  biochemistry  and  biophysics.  For  example, 
A.  L.  Dean  conducted  an  " Investigation  of  the  proteolytic  enzymes  of 
plants"  and  W.  T.  Swingle  received  a  grant  for  an  "Investigation  of  electro- 
magnetic and  electrostatic  effects  on  lines  of  force  found  in  living  plant 
cells."  No  conclusive  results  were  reported  from  the  latter  study,  but  Dean 
reported  finding  an  ereptic  enzyme13  in  all  tissues  of  a  species  of  bean 
(Phaseolus  vulgaris). 

11  Transpiration — the  escape  of  water  vapor  from  living  plants. 

12  Stomata — minute  pores  in  the  epidermis  of  plants,  through  which  gases  and  water  enter  or 
escape  from  the  plant. 

13  A  type  of  enzyme  that  breaks  down  proteoses  and  peptones,  as  in  the  intestinal  tract  of 
animals. 


86  CARNEGIE     INSTITUTION     OF      WASHINGTON 

Most  interesting  about  the  program  of  the  Desert  Botanical  Laboratory 
in  1905  was  the  complete  absence  of  any  attention  to  the  problems  of  photo- 
synthesis, which  have  since  become  a  major  preoccupation  of  the  Depart- 
ment of  Plant  Biology.  Although  the  basic  physical-chemical  relations  of 
photosynthesis14  had  been  suggested  sixty  years  before,  there  was  no  hint 
of  the  importance  of  these  problems  in  the  1905  program.  Th.  W.  Engelmann 
in  1887  discovered  that  light  absorbed  by  pigments  other  than  chlorophyll 
also  produced  photosynthesis,  more  than  fifteen  years  before  the  establish- 
ment of  the  Laboratory.  Even  during  the  year  of  the  1905  report,  the 
English  plant  physiologist,  F.  F.  Blackman,  demonstrated  that  photo- 
synthesis includes  at  least  one  "dark"  reaction  not  initiated  by  light. 

The  interest  of  the  Institution  in  photosynthesis  actually  began  six 
years  later,  in  1911,  when  H.  A.  Spoehr  came  to  the  Department  of  Botan- 
ical Research  at  Tucson,  which  succeeded  the  Desert  Botanical  Laboratory. 
Spoehr  first  came  to  the  Institution  to  study  the  "chemical  physiology"  of 
plants  but  very  soon  became  immersed  in  the  problems  of  photosynthesis, 
an  interest  he  maintained  actively  until  his  retirement  in  1950.  Just  as 
intensively  as  in  Spoehr's  time  the  Department  of  Plant  Biology  today 
applies  its  research  efforts  to  the  great  problem  of  unraveling  the  complex- 
ities of  photosynthesis. 

The  work  of  the  Department  in  1961-1962  on  photosynthesis 
still  centers  on  a  problem  the  general  outlines  of  which 
1961-1962      emerged  in  Engelmann's  time :  the  exact  function  of  the  two 
sets  of  pigments,  chlorophyll  and  the  accessory  pigments, 
both  of  which  induce  photosynthesis.  It  is  now  supposed 
that  photosynthesis  comprises  at  least  two  photochemical  events,  one  driven 
by  chlorophyll  a,  the  other  by  the  accessory  pigments.  Two  discoveries 
made  about  1955  provided  some  evidence  for  this  hypothesis.  One  discovery 
was  Blinks'  chromatic  transient  effect,  a  momentary  change  in  photo- 
synthetic  rate  observed  when  light  absorbed  by  chlorophyll  is  changed  to  a 
color  absorbed  by  accessory  pigments.  The  other  was  Emerson's  enhance- 
ment effect.   In  this  effect  photosynthesis   resulting  from  wavelengths 
absorbed  by  chlorophyll  a  alone,  when  augmented  by  wavelengths  absorbed 
through  accessory  pigments,  is  increased  more  than  would  be  predicted 
from  the  simple  sum  of  the  effects  from  both  radiations  presented  separately. 
A  major  effort  is  now  being  made  in  the  world  of  research  to  define  the 

14  Joseph  Priestley  demonstrated  the  production  of  "good  air"  (oxygen)  by  plants  in  1772; 
Jan  Ingenhousz  in  1778  showed  that  the  effect  noted  by  Priestley  resulted  from  the  influence  of 
sunlight;  Jean  Senebier  noted  in  1782  that  "bad  air"  (carbon  dioxide)  was  a  necessary  input; 
Lavoisier  determined  the  composition  of  carbon  dioxide  in  1784;  Nicolas  de  Saussure  showed 
precisely  in  1804  that  water,  light,  and  carbon  dioxide  were  inputs,  and  oxygen  plus  organic 
matter  outputs;  Julius  Mayer,  through  his  concepts  of  the  conservation  of  energy,  in  1845  sug- 
gested the  place  of  sunlight  and  vegetative  organisms  in  chemical  action  taking  place  on  a  global 
basis  at  the  earth's  surface. 


REPORT   OF   THE    PRESIDENT  37 

nature  of  these  two  essential  photochemical  reactions  and  relate  them  to  the 
chain  of  events  in  photosynthesis  that  results  in  oxygen  evolution  and 
carbon  dioxide  reduction.  As  throughout  the  long  history  of  research  in 
photosynthesis,  ingenious  theories  currently  exist  to  explain  in  detail  most 
of  the  known  effects.  Generally  considered,  each  investigator  has  his  own 
favored  concept  of  the  process,  and  the  different  hypotheses  are  not  entirely 
compatible  with  one  another.  Further  experiments  and  more  comprehensive 
concepts  are  still  needed  for  an  adequate  understanding  of  photosynthesis. 

At  the  Department  of  Plant  Biology,  C.  Stacy  French  and  his  associates 
continued  their  efforts  to  provide  experimental  evidence  on  the  exact 
functions  of  the  different  plant  pigments. 

A  year  ago  they  found  in  a  red  alga  (Porphyridium  cruentum)  that 
chlorophyll  a  but  not  the  accessory  pigment,  phycoerythrin,  produces  a 
chemically  unidentified  substance  that  rapidly  consumes  oxygen.  Some  of 
it  is  left  over  after  a  light  exposure,  as  is  demonstrated  by  the  temporarily 
accelerated  rate  of  oxygen  uptake  after  an  exposure  to  light  absorbed  by 
chlorophyll  a.  This  material  is  also  believed  to  be  an  intermediate  in  the 
process  of  photosynthesis. 

This  year  the  persistence  of  the  chemically  unidentified  material  previ- 
ously formed  by  illumination  of  chlorophyll  a  was  measured  by  French  and 
Jeanette  Brown.  This  was  done  by  observing  the  increased  oxygen  pro- 
duction of  the  algae  upon  exposure  to  individual  flashes  of  light  at  the 
wavelength  absorbed  by  phycoerythrin.  The  presence  of  the  material 
enhances  the  oxygen  evolution  by  a  light  flash  that  activates  phycoerythrin. 
The  half-life  of  the  material  measured  in  this  way  was  found  to  be  about 
18  seconds  under  certain  conditions.  By  contrast,  preillumination  by 
phycoerythrin-absorbed  light  did  not  enhance  oxygen  production  when 
chlorophyll  a  was  subsequently  activated. 

Another  series  of  experiments,  made  this  year,  shows  even  more  complex 
relations  between  the  effects  of  different  pigments  of  green  leaves.  The  story 
began  about  eighty  years  ago,  when  Engelmann  found  traces  of  oxygen 
evolution  from  isolated  chloroplasts.  This  effect  was  further  investigated  by 
Molish  early  in  this  century,  but  since  then  the  reaction  has  had  very  little 
attention  until  recently,  no  doubt  owing  to  R.  Hill's  discovery  in  1937  that 
the  addition  of  oxidants  such  as  ferricyanide  greatly  increases  the  amount 
of  oxygen  produced.  An  avalanche  of  papers  on  the  Hill  reaction  followed, 
and  experiments  with  the  evolution  of  oxygen  from  within  chloroplasts 
without  added  substances  have  been  all  but  abandoned. 

In  the  past  year,  however,  Y.  de  Kouchkovsky  of  the  Centre  National  de 
la  Recherche  Scientifique,  Gif-sur-Yvette,  France,  and  David  C.  Fork  of 
the  Department  of  Plant  Biology,  have  reexamined  this  effect  with  greatly 
improved  methods.  The  work,  started  independently  at  the  two  laboratories, 


38  CARNEGIE     INSTITUTION     OF      WASHINGTON 

was  continued  as  a  collaborative  effort  during  Dr.  Fork's  visit  to  Gif-sur- 
Yvette  in  March  1962. 

By  measuring  oxygen  exchange  of  Swiss  chard  chloroplasts  Fork  showed 
that  it  is  possible  to  distinguish  four  separate  effects  of  light,  each  with  its 
characteristic  action  spectrum.  They  are: 

1.  The  evolution  of  oxygen  from  chloroplasts  without  added  oxidants  is 
driven  most  effectively  by  light  having  a  wavelength  of  650  millimicrons 
(red).15  This  corresponds  to  the  absorption  peak  of  chlorophyll  b  in  chloro- 
plasts, thereby  showing  that  chlorophyll  b  is  more  effective  than  chlorophyll 
a  in  this  reaction.  A  shoulder  on  the  curve  of  the  action  spectrum,  however, 
shows  that  at  least  one  of  the  three  forms  of  chlorophyll  a  is  also  active. 
This  oxygen  production  within  the  chloroplast  goes  rapidly  for  only  a  few 
seconds,  then  its  rate  drops  to  a  very  low  value.  Storage  in  the  dark  revives 
the  ability  to  evolve  oxygen.  Apparently  light  consumes  some  material 
found  in  chloroplasts  which  is  restored  in  darkness. 

2.  Dr.  Fork  found  the  recovery  process  to  be  strongly  accelerated  by 
exposure  to  far-red  light.  A  wavelength  of  about  730  millimicrons  was  most 
effective  for  this  purpose.  This  wavelength  suggests  identity  with  phy to- 
chrome,  a  substance  which,  though  present  in  very  small  amounts,  controls 
many  plant  responses.  In  addition  to  the  730-millimicron  peak,  however, 
the  action  spectrum  for  the  regeneration  of  the  chloroplasts'  ability  to 
evolve  oxygen  also  has  a  peak  in  the  blue  wavelengths  which  does  not 
activate  phytochrome. 

3.  Ferricyanide  [K3Fe(CN)e],  when  added  to  chloroplasts,  substitutes  for 
the  natural  oxidant  substance  responsible  for  photoproduction  of  oxygen. 
The  rate  of  oxygen  evolution  remains  for  long  light  exposures,  and  the  action 
spectrum,  which  peaks  at  678  millimicrons,  shows  that  chlorophyll  a  is 
more  effective  than  chlorophyll  b  when  ferricyanide  is  present. 

4.  A  very  specific  inhibitor  for  oxygen  production  by  chloroplasts  is  the 
herbicide  DCMU.16  When  this  poison  is  added  to  chloroplasts  the  photo- 
consumption  of  oxygen  can  be  measured  without  interference  by  oxygen 
evolution  and  shows  a  maximum  efficiency  at  wavelength  690  millimicrons 
(red). 

Four  different  action  spectra  have  thus  been  measured  for  oxygen 
exchange  in  isolated  chloroplasts.  French  raises  the  question  of  the  exact 
function  of  each  pigment  in  these  various  photoprocesses.  He  says  that  the 
answer  is  clear  for  chlorophylls  a  and  b  (678-  and  650-millimicron  peaks) : 
they  are  concerned  with  oxygen  evolution.  But  it  is  not  yet  known  why 
chlorophyll  b  is  more  effective  than  chlorophyll  a  for  the  reaction  within  the 

15  One  millimicron  =  10-6  millimeter. 

16  3-(3,4-Dichlorophenyl)-l,l-dimethylurea;  manufactured  by  E.  I.  du  Pont  de  Nemours  and 
Company. 


REPORT    OF   THE   PRESIDENT  39 

natural  chloroplast  whereas  the  reverse  is  true  when  ferrieyanide  is  added. 

The  two  action  spectra  with  peaks  at  730  and  690  millimicrons  are  more 
obscure.  They  do  not  necessarily  indicate  that  there  are  active  pigments 
with  absorption  maxima  at  either  wavelength.  Instead,  spectra  may  result 
from  the  activation  of  two  pigments  whose  reactions  either  reinforce  or 
counteract  each  other.  In  both  cases  the  action  spectra  maxima  may  differ 
greatly  from  the  absorption  maxima  of  the  reacting  pigments.  These  are 
interesting  subjects  for  further  investigation. 

Ellen  C.  Weaver  started  an  attack  on  the  problems  of  photosynthesis 
with  an  intriguing  and  promising  new  technique,  that  of  electron  para- 
magnetic17 resonance  (EPR)  spectroscopy.  She  notes  in  her  report  the  well 
established  fact  that  illuminated  chlorophyll-containing  material  has  a 
higher  level  of  unpaired  electrons  than  material  in  the  dark,  suggesting  that 
some  phase  of  photosynthesis  proceeds  by  single-electron  transfers.  Even 
though  several  research  groups  outside  the  Institution  had  employed  this 
new  technique  (about  six  years  old)  in  studying  photosynthesis,  no  rigorous 
demonstration  had  yet  been  made  that  electron  resonance18  had  a  direct 
connection  with  photosynthesis. 

Dr.  Weaver  set  out  during  the  year  first  to  determine  whether  or  not  the 
established  resonance  was  associated  with  chlorophyll.  She  observed  two 
distinctly  different  light-induced  resonances.  One  is  the  R  (rapid-decaying) 
signal,  seen  only  when  cells  are  illuminated.  The  other  may  persist  for  hours 
in  the  absence  of  light,  and  it  is  designated  the  S  (slow-decaying)  signal. 
Using  a  yellow  mutant  (no  chlorophyll)  of  the  fresh-water  alga  Chlamy- 
domonas  reinhardi,  Dr.  Weaver  obtained  no  R  signals  in  EPR  observation, 
suggesting  that  the  R  signal  is  ascribable  to  chlorophyll.  She  also  discovered 
by  using  dilute  cell  suspensions  that  680-millimicron  light  (near  the  absorp- 
tion peak  for  chlorophyll  a)  was  the  most  effective  for  producing  the  R 
signals.  Another  interesting  result  is  her  discovery  that  the  amplitude  of 
the  R  signal  has  a  strictly  linear  (proportional)  relation  to  light  intensity 
for  the  wavelengths  least  absorbed  by  chlorophyll,  whereas  wavelengths 
most  strongly  absorbed  by  chlorophyll  have  no  linear  relation  to  light 
intensity  (assuming  low  light  levels  in  both  cases).  Dr.  Weaver's  tentative 
conclusion  from  these  observations  is  that  the  R  signal  is  associated  with 
chlorophyll  and  arises  from  the  "primary"  act  of  photosynthesis. 

Dr.  Weaver  also  discovered  that  any  inhibition  of  oxygen  evolution,  as 
by  DCMU  or  by  limiting  the  manganese-ion  concentration  in  the  growing 
medium,  will  produce  an  enhanced  R  signal.  This  suggests  that  if  the 

17  Paramagnetic — atoms  having  spin  systems  with  magnetic  moment  (or  materials  containing 
those  atoms)  are  paramagnetic. 

18  Electron  resonance — a  property  of  unpaired  electrons,  whereby  precession  of  the  spinning 
electron  may  be  inferred  when  it  is  subjected  to  an  electromagnetic  field  at  a  specific  frequency, 
as  in  EPR  spectroscopy. 


40  CARNEGIE     INSTITUTION      OF      WASHINGTON 

pathway  of  the  electrons  is  in  any  way  obstructed  the  net  level  of  unpaired 
spins  rises.  The  result  indicates  further  that  the  alteration  of  photosynthetic 
processes  other  than  oxygen  evolution  may  provide  a  fruitful  field  for 
experiment  using  the  EPR  spectroscopic  technique.  Interestingly,  the  S 
(slow-decaying)  signal  is  not  altered  by  blocking  the  oxygen  evolution 
pathway  with  DCMU,  but  manganese  starvation  reduces  that  signal  to  an 
extremely  low  level.  It  is  thought  that  this  result  may  be  correlated  with  a 
lack  of  plastoquinone,19  previously  determined  elsewhere  to  be  a  necessary 
and  apparently  universal  factor  in  the  oxygen  evolution  of  green  plants. 

Dr.  Weaver  has  thus  presented  evidence  that  chlorophyll  is  the  source  of 
one  type  of  free  electrons  in  an  intact  photosynthetic  organism  and  that 
plastoquinone  is  the  site  of  another  type.  She  has  also  demonstrated  the 
correlation  of  the  two  types  of  signals  with  the  evolution  of  photosynthetic 
oxygen.  The  method  and  her  results  are  of  more  than  usual  interest,  because 
photosynthesis  is  essentially  a  photoreduction  process  when  viewed  in  a 
highly  general  way,  that  is,  the  transfer  of  electrons  from  one  substance 
to  another. 

Although  photosynthesis  still  presents  an  awesome  complexity  to  those 
investigating  it,  studies  like  those  of  French,  Fork,  Brown,  and  Weaver 
examining  the  effects  of  light  on  metabolic  reactions  are  continually 
changing  concepts  of  how  synthesis  takes  place  and,  step  by  step,  are  build- 
ing a  more  complete  understanding  of  this  vastly  important  phenomenon. 

Another  field  in  plant  biology,  experimental  taxonomy,  can  trace  its 
origin  to  the  activities  of  the  1905  Desert  Laboratory.  Again,  however,  the 
diffuse  approach  of  1905  is  gone.  William  M.  Hiesey  and  his  associates  note 
in  the  1961-1962  report  that  current  developments  in  precise  techniques 
have  greatly  extended  the  horizon  of  this  field.  Instead  of  the  compart- 
mentalizing of  botanical  study,  which  was  commencing  in  1905,  they  see 
"a  truly  integrated  plant  science  whereby  contributions  from  the  various 
specialized  fields,  including  taxonomy,  ecology,  cytology,  genetics,  physi- 
ology, developmental  morphology,  and  biochemistry,  can  be  incorporated 
in  a  panoramic  view  of  plant  relationships  and  evolution."  Their  goal  is  an 
integrated  understanding  of  the  chain  of  mechanisms  that  determine  plant 
evolution,  including  the  genetic  and  the  biochemical.  For  a  number  of  years 
plants  of  the  genus  Mimulus20  had  been  used  for  comparative  growth 
studies  of  altitudinal  effects  at  the  Stanford,  Mather,  and  Timberline 
stations.  More  recently  the  races  of  one  species,  Mimulus  cardinalis,  have 
been  subjected  to  controlled  growth  chamber  experiments. 

During  the  year  Harold  W.  Milner  made  some  particularly  interesting 

19  Plastoquinone — quinone  found  in  chloroplasts ;  the  structure  of  this  compound  is  given  in 
figure  31  of  the  report  of  the  Department  of  Plant  Biology. 

20  The  garden  "monkey  flower"  belongs  to  the  Mimulus  genus. 


REPORT   OF   THE    PRESIDENT  J^.1 

studies  of  the  photosynthetic  rates  of  six  races  of  the  species  originating  in 
diverse  climates  and  altitudes.  Among  the  variations  observed  were  a  60 
per  cent  difference  among  the  races  in  the  light  intensity  required  to  saturate 
photosynthesis  at  high  temperature,  and  a  100  per  cent  difference  at  a  very 
low  temperature  (0°C).  Significant  variance  in  photosynthetic  rate  at 
extreme  temperatures  also  was  observed,  as  well  as  disparate  abilities  to 
maintain  a  high  rate  of  photosynthesis  over  a  long  period.  From  these  and 
other  results  one  may  conclude  that  climatic  races  within  the  same  species 
may  show  differential  patterns  of  response  undoubtedly  linked  with  vari- 
ations in  internal  physiology. 

During  the  year  an  important  step  was  taken  toward  establishing  tissue 
cultures  from  Mimulus  plants,  so  as  to  make  quantitative  measurements  of 
growth  and  photosynthetic  rates  in  tissue  cultures  similar  to  those  for  whole 
plants.  By  examining  the  physiological  requirements  of  tissue  from  various 
plant  organs,  it  should  be  possible  to  localize  the  site  of  physiological 
differences  within  the  plant. 

In  addition,  the  group  extended  its  work  during  the  year  to  species  of 
Solidago  (goldenrod),  particularly  in  the  collaborative  work  of  Malcolm 
Nobs  of  the  Department  working  at  the  Institute  of  Plant  Systematics  and 
Genetics  at  Uppsala,  Sweden.  The  same  type  of  difference  in  response  to 
light  intensity  was  observed  between  two  races  of  Solidago  virgaurea:  one  a 
shade-loving  race  from  Sweden  and  the  other  an  alpine  race  from  Norway. 
The  alpine  race  has  a  much  higher  requirement  for  light  saturation  than 
the  shade  race,  and  its  chloroplasts  remain  normal  at  light  intensities  that 
cause  the  disintegration  of  those  from  the  shade  race. 


The  Department  of  Terrestrial  Magnetism 

The  Department  of  Terrestrial  Magnetism  was  also  among 
the  active  Departments  of  the  Institution  in  the  year  1905. 
1905         The  work  of  the  Department  in  1905  faithfully  followed  its 
name,  although  a  wide  range  of  projects  was  reported,  with 
activity  on  an  almost  worldwide  basis.  A  major  preoccupa- 
tion of  the  Department  during  that  year  was  an  effort  to  start  a  systematic 
series  of  magnetic  observations  on  most  parts  of  the  globe,  which  at  that 
time  were  informational  blanks.  L.  A.  Bauer,  Director  of  the  Department 
in  that  year,  stated,  "our  progress  with  regard  to  the  great  and  principal 
facts  of  the  earth's  magnetism  will  be  at  a  standstill  unless  a  magnetic 
survey  of  the  whole  globe  be  undertaken  immediately."  Toward  that  end  a 
wooden  sailing  vessel,  the  brig  Galilee,  had  been  manned  and  outfitted,  and 
had  undertaken  trial  runs.  This  was  the  beginning  of  a  program  that 


42  CARNEGIE     INSTITUTION      OF      WASHINGTON 

continued  for  almost  25  years  thereafter,  in  which  the  sailing  vessels  Galilee 
and  Carnegie  logged  more  than  400,000  miles  to  undertake  magnetic  and 
other  scientific  observations  in  every  ocean  area  of  the  globe.  It  ended  only 
with  the  accidental  destruction  by  fire  of  the  Carnegie  at  Samoa  in  Novem- 
ber 1929. 

A  very  extensive  land  survey  program  also  was  being  initiated  for 
magnetic  observations  in  1905.  Many  of  the  islands  of  the  West  Indies  were 
covered  in  that  year,  and  arrangements  were  being  completed  for  observa- 
tions on  the  South  Pacific  Islands  and  in  Canada,  Mexico,  Central  America, 
South  America,  and  China.  Cooperative  arrangements  for  observations  and 
research  were  maintained  with  several  German  scientific  institutions  and 
with  the  St.  Petersburg  Academy  of  Sciences  in  Russia. 

Besides  its  primary  program  on  the  study  of  and  basic  data  collection  for 
terrestrial  magnetism,  the  Department  in  1905  organized  and  participated 
in  the  program  of  observing  the  solar  eclipse  of  that  year,  and  it  began 
cooperating  with  the  Institution's  Solar  Observatory  in  the  study  of  several 
solar  phenomena. 

The  Institution  in  1905  also  expressed  a  substantial  interest  in  physics 
research,  but  entirely  through  a  program  of  grants  to  fourteen  American 
physicists.  Among  the  grants  were  several  for  studies  of  emission  spectra 
and  a  study  of  the  theory  of  light. 

Although  the  emphasis  so  prominent  in  the  1905  program 
of  the  Department  of  Terrestrial  Magnetism  was  continued 
1961-1962       until  the  early  1930's  with  relatively  slight  changes,  the 
program  of  1961-1962  in  the  Department  was  a  much  differ- 
ent one.  The  principal  activities  reminiscent  of  the  earlier 
days  of  the  Department  came  in  the  research  of  Scott  E.  Forbush,  but  again 
in  an  environment  strikingly  different  from  that  of  the  first  twenty-five 
years  of  the  Department.  Forbush's  principal  investigations  during  the 
report  year  were  devoted  to  the  intensity  of  the  charged  particles  in  the 
Van  Allen  trapped-radiation  belt  adjacent  to  the  earth,  as  recorded  during 
the  transits  of  the  satellite  Explorer  VII  through  the  belt  between  1959  and 
1960.  He  also  had  under  way  studies  examining  the  southward  shift  of  the 
auroral-zone  current  system  during  magnetic  storms  in  its  probable  associ- 
ation with  particles  coming  from  the  outer  Van  Allen  belt. 

The  bulk  of  the  Department's  varied  and  imaginative  research  in  1961- 
1962,  however,  derived  from  applying  the  techniques  of  physics  to  a  wide 
variety  of  geophysical  and  biological  problems.  They  ranged  from  the  ex- 
amination of  the  interior  of  living  cells  to  charting  the  hydrogen  clouds  of 
our  Galaxy. 

Perhaps  the  most  significant  results  to  emerge  from  the  year's  work  were 
from  a  quarter  that  could  hardly  have  been  envisioned  as  associated  with 


REPORT   OF   THE    PRESIDENT  /$ 

the  Department  even  twenty  years  ago.  They  came  from  the  work  of  the 
Biophysics  Section  (E.  T.  Bolton,  R.  J.  Britten,  D.  B.  Cowie,  B.  J.  Mc- 
Carthy, J.  E.  Midgley,  and  R.  B.  Roberts)  on  the  fine  structure  of,  and 
biochemical  processes  taking  place  within,  bacterial  and  other  cells. 

As  the  end  of  the  report  year  approached,  the  Section  was  engrossed  in 
some  striking  experiments  involving  "messenger"  ribonucleic  acid  (RNA). 
This  type  of  RNA  contains  nucleotide21  sequences  complementary  to  those 
in  the  appropriate  DNA  which  provides  the  genetic  information.  In  follow- 
ing a  lead  provided  by  E.  K.  F.  Bautz  and  B.  D.  Hall  at  the  University  of 
Illinois  it  was  discovered  that  single-stranded  DNA  could  be  immobilized 
in  agar  and  complementary  RNA  could  be  caused  to  hybridize  with  it 
through  the  formation  of  hydrogen  bonds.  By  washing,  the  immobilized 
hybrid  DNA-RNA  combination  was  freed  of  other  contaminating  RNA. 
The  hybridized  RNA  could  then  be  reclaimed,  in  a  state  of  high  purity,  by 
dissociation  of  the  hydrogen  bonds,  and  could  be  chemically  analyzed. 

With  this  simple  and  effective  new  procedure  it  has  been  possible  to 
demonstrate  that  the  DNA-like  RNA  comprises  about  1  per  cent  of  the 
total  RNA  of  bacterial  cells  and  that  it  has  a  half-life  during  active  syn- 
thesis of  approximately  2  minutes.  On  the  assumption  that  this  RNA  is  in 
fact  the  active  template  for  protein  synthesis,  the  measurements  of  its 
quantity  and  half-life  show  that  a  single  molecule  acts  catalytically  for  the 
synthesis  of  many  polypeptide22  chains. 

Further  work  has  revealed  that  the  method  can  be  used  to  exploit  the 
specificity  inherent  in  the  hybridization  process,  which  depends  upon  long 
regions  of  complementary  nucleotide  sequences  in  molecules  of  RNA  and 
DNA.  Thus,  RNA  from  bacteriophage  T2  will  hybridize  well  with  DNA 
of  the  genetically  closely  related  phage  T4  but  not  with  the  apparently 
unrelated  T7  DNA.  Several  species  of  bacteria  have  also  been  tested,  and 
cross  reactions  have  been  found  to  occur  to  a  greater  or  lesser  degree  in 
accord  with  accepted  taxonomic  relationships.  Thus,  the  method  has  made 
feasible  a  quantitative  chemical  analysis  of  the  amount  of  genetic  informa- 
tion held  in  common  among  species. 

Since  the  method  is  a  general  one,  applying  to  the  DNA  of  all  species  and 
tissues,  it  can  be  used  in  studies  of  the  transcription  of  genetic  information 
and  of  differentiation,  two  of  the  key  subjects  of  modern  biology. 

During  the  year  the  Biophysics  Section  also  contributed  a  new  hypothesis 
about  the  code  associated  with  the  role  of  nucleic  acid  in  specifying  the 
order  of  amino  acids  in  protein.  The  prevailing  hypothesis  interprets  ex- 
perimental findings  in  terms  of  a  "three-letter"  or  triplet  code.  The  experi- 

21  Precursor  of  or  decomposition  product  from  nucleic  acid,  composed  of  a  nitrogenous  base,  a 
ribose  sugar,  and  phosphoric  acid. 

22  Peptides  are  proteins  linked  by  amide  (RCO-NHR/),  or  "peptide,"  bond. 


44  CARNEGIE     INSTITUTION      OF      WASHINGTON 

ments  of  the  Section  lead  its  members  to  believe  that  a  two-letter  or  doublet 
code  eliminates  the  major  failing  of  the  triplet  code,  which  implies  an  un- 
realistically  high  uridylic  acid23  content  for  the  "template"  material  of 
protein  synthesis.  The  doublet  code  apparently  provides  a  good  correlation 
between  the  amino  acid  composition  of  the  bacterially  synthesized  protein 
and  the  nucleotide  composition  of  the  RNA  templates  on  which  it  is  formed. 


Other  fields  in  which  the  techniques  of  physics  are  being  applied  by  Staff 
Members  of  the  Department  are  seismological  exploration  of  the  earth's 
crust,  radioactive  dating  of  rocks,  radio  astronomy,  and  the  development 
of  image  tubes  for  use  in  astronomical  studies. 

It  is  of  particular  interest  that  all  these  programs  in  one  respect  or  another 
are  cooperative,  carrying  on  the  tradition  of  joint  investigations  or  joint 
enterprise  which  was  started  and  even  widely  used  in  the  earliest  days  of 
the  Department.  As  Merle  Tuve,  the  Director  of  the  Department,  observes 
in  the  introduction  of  his  1961-1962  report,  "  'cooperation'  .  .  .  has  many 
very  different  aspects  in  the  current  work  of  the  Department,  but  in  each 
case  it  represents  a  situation  where  there  is  special  usefulness  in  our  freedom 
of  initiative  and  recognition  of  the  infectious  characteristic  of  personal 
enthusiasm."  To  some  extent,  the  same  thing  might  have  been  said  for  the 
programs  in  biophysics  and  geomagnetic  studies. 

A  good  example  of  the  Department's  cooperative  approach  is  shown  in 
its  radio  astronomy  program.  With  the  support  of  the  National  Science 
Foundation  a  new  Carnegie  Radio  Astronomy  Station  will  soon  be  estab- 
lished in  Argentina.  Parts  for  a  major  instrument,  a  parabolic  antenna 
nearly  100  feet  (30  meters)  in  diameter,  are  now  being  manufactured  in 
this  country  and  will  be  shipped  to  Buenos  Aires  for  assembly  there  during 
1962-1963.  The  Argentinian  National  Council  for  Scientific  and  Technical 
Investigations  and  the  Research  Council  of  the  State  of  Buenos  Aires  have 
created  a  new  National  Institute  of  Radio  Astronomy  to  participate  in  the 
construction  and  operation  of  the  station.  Later  operation  will  be  a  cooper- 
ative venture  among  the  Carnegie  Institution,  the  University  of  Buenos 
Aires,  and  the  University  of  La  Plata.  Invitations  will  be  extended  to 
astronomers  in  other  institutions  in  South  America  to  participate  in  the 
research  program.  Some  fellowships  are  being  offered  by  the  Institution  to 
bring  students  and  professional  research  men  interested  in  radio  astronomy 
to  this  country  for  training  in  the  use  of  parabolic  antennas  and  for  ac- 
quiring educational  background  in  radio  astronomy. 

The  observational  program  in  radio  astronomy  using  the  Department's 

23  Uridylic  acid — a  nucleotide;  technically  uracil  (2,6-dioxypyrimidine)  +  D-ribose  sugar  -f- 
phosphoric  acid. 


REPORT   OF   THE    PRESIDENT  £5 

instruments  also  continued  during  the  year.  Observations  of  the  hydrogen 
gas  content  at  the  center  of  our  Galaxy  confirmed  previous  observations  at 
Leiden,  the  Netherlands,  and  Sydney,  Australia,  that  the  motions  of  hy- 
drogen close  to  the  Galactic  center  are  complex,  and  that  the  hydrogen  gas 
not  only  is  rotating  about  the  center  of  the  Galactic  mass  but  also  is  ex- 
panding. Because  of  its  latitudinal  position,  the  Derwood,  Maryland, 
Station  of  the  Department  was  able  to  extend  observations  nearly  20° 
farther  south  along  the  Galactic  plane  than  the  Dutch  station. 

The  Department  also  decided  during  the  year,  after  considerable  experi- 
ment, to  begin  construction  of  an  interferometer  array  from  parabolic 
dish  antennas,  to  be  able  to  obtain  precise  positions  of  radio  noise  sources 
in  the  sky.  A  30-meter  dish  closely  following  the  design  of  the  Argentinian 
radio  telescope  is  now  being  constructed  at  Derwood  and  will  be  used  with 
the  existing  60-foot  parabolic  antenna  as  a  two-element  interferometer. 
These  two  antennas  will  be  employed  in  experiment  to  evaluate  the  po- 
tentialities of  such  a  system  in  determining  precise  radio-star  positions. 

A  second  cooperative  venture  of  the  Department  in  the  area  of  astro- 
nomical study  has  been  the  work  of  the  Committee  on  Image  Tubes  for 
Telescopes,  of  which  Merle  Tuve  is  chairman.  In  this  the  Department  has 
collaborated  with  the  Mount  Wilson  and  Palomar  Observatories,  the  Lowell 
Observatory,  the  National  Bureau  of  Standards,  and  the  United  States 
Naval  Observatory  to  develop  electronic  image  tubes  for  magnifying  signals 
received  on  optical  telescopes.  This  work  has  also  been  supported  in  large 
part  by  generous  grants  from  the  National  Science  Foundation. 

During  the  year  the  Committee  continued  the  testing  of  tubes  manu- 
factured experimentally  upon  its  order  by  the  International  Telephone  and 
Telegraph  Corporation  Laboratories  and  by  the  Radio  Corporation  of 
America.  The  tests  conducted  were  largely  undertaken  by  W.  K.  Ford,  Jr., 
of  the  Department.  The  tubes  proved  to  have  better  operating  character- 
istics than  the  Committee  had  hoped  for  only  three  years  ago.  Telescope 
observations  were  made  at  the  Lowell  Observatory  with  the  tubes  to  ex- 
amine their  reliability  and  effectiveness,  and  laboratory  investigations  were 
conducted  to  distinguish  among  the  relative  merits  of  the  several  tubes. 
On  the  basis  of  the  spectrographic  tests  from  telescope  observations  and 
the  laboratory  tests,  the  Committee  believes  that  the  two  types  of  tubes 
recently  examined  (mica-window  and  cascaded)  will  have  wide  application 
in  astronomy  because  of  their  advantages  over  conventional  photography. 
Development  will  be  continued,  again  with  the  support  of  the  grant  from 
the  National  Science  Foundation. 

A  major  project  of  the  seismic  studies  group  in  the  Earth's  Crust  Section 
of  the  Department  (J.  S.  Steinhart,  L.  T.  Aldrich,  M.  A.  Tuve,  and  associ- 
ates) was  an  intensive  study  of  the  earth's  crust  in  Maine,  in  which  col- 


46  CARNEGIE     INSTITUTION     OF      WASHINGTON 

leagues  from  the  University  of  Wisconsin,  Princeton  University,  Penn- 
sylvania State  University,  the  University  of  Michigan,  and  the  Woods  Hole 
Oceanographic  Institution  participated,  and  the  United  States  Coast  Guard 
assisted  in  detonating  explosions  in  the  Gulf  of  Maine  in  July  1961. 

The  data  obtained  from  the  explosions  have  since  been  the  subject  of 
appraisal  to  determine  the  application  of  explosion  seismology  to  designation 
of  crustal  structures.  This  is  a  very  real  geological  problem,  because  the 
traditional  conception  of  the  earth's  crust  as  one  or  more  horizontal  layers 
of  constant  seismic  wave  velocity  has  appeared  inadequate  for  more  than  a 
decade.  Efforts  to  find  the  proper  reflections  from  the  surfaces  of  the  sup- 
posed layers  have  been  unsuccessful;  and  laboratory  measurements  of 
seismic  velocities  in  various  rock  types  contradict  the  layer  hypothesis. 
Field  evidence  suggests  significant  lateral  as  well  as  vertical  differences  in 
structure.  Several  models  that  might  conform  to  the  seismic  results  received 
from  the  explosions  were  therefore  constructed. 

On  the  basis  of  these  models  it  seems  fairly  certain  that  in  Maine  the 
Mohorovicic  discontinuity24  lies  at  36  ±  3  kilometers  below  the  surface. 
The  most  likely  models  suggest  that  the  upper  3  kilometers  of  the  crust  is 
granitic  and  that  below  the  granite  the  percentage  of  gabbro25  increases  at 
a  rate  that  maintains  a  steady  gradient  in  seismic  wave  velocity  change  to 
a  depth  of  about  20  kilometers.  These  findings  are  of  interest  geologically 
in  that  they  postulate  appreciably  less  granitic  material  than  is  customarily 
thought  to  be  in  a  continental  crust. 

The  radioactive  dating  group  is  not  only  interinstitutional  but  also 
interdepartmental  (L.  T.  Aldrich  and  S.  R.  Hart  of  the  Department  of 
Terrestrial  Magnetism,  G.  L.  Davis,  G.  R.  Tilton,  and  B.  R.  Doe  of  the 
Geophysical  Laboratory  and  associates).  During  the  year  the  Department 
of  Terrestrial  Magnetism  members  of  the  group  participated  in  an  exchange 
program  with  the  Geological  and  Mineralogical  Institute  of  the  University 
of  Kyoto. 

Dr.  I.  Hayase,  of  the  University  of  Kyoto,  spent  part  of  the  year  at  the 
Department  becoming  familiar  with  its  techniques  of  measuring  mineral 
ages.  In  the  course  of  his  visit  he  analyzed  samples  collected  in  Japan.  The 
data  were  of  interest  as  the  first  measurement  of  the  kind  from  Japan.  They 
showed  no  contradictions  between  the  isotopically  determined  ages  and 
ages  implied  by  geological  structure.  They  also  showed  discordances  between 
rubidium-strontium  and  potassium-argon  age  determinations  commonly 
enough  to  indicate  a  complex  geological  history  for  the  Islands. 

As  a  second  part  of  the  exchange,  L.  T.  Aldrich  of  the  Department  is  now 

24  A  phenomenon  recorded  in  the  changing  speed  of  seismic  waves  at  certain  depths. 

25  A  granitic  rock  formed  of  plagioclase  (light-colored)  feldspar  and  a  monoclinic  pyroxene  like 
augite  (dark-colored). 


REPORT   OF   THE    PRESIDENT  Jfl 

in  Kyoto  as  a  visiting  professor  at  the  University.  He  is  assisting  in  the 
establishment  of  a  complete  laboratory  for  the  measurement  of  mineral 
ages.  To  facilitate  this  work  the  Department  constructed  and  shipped  to 
the  University  a  mass  spectrometer26  which  Dr.  Aldrich  now  has  in  opera- 
tion at  the  Institute  there.  It  is  expected  that  the  spectrometer  will  serve 
as  a  model  for  similar  equipment  to  be  built  elsewhere  in  Japan.  We  hope 
that  this  particular  interinstitutional  collaboration  will  continue  indefinitely. 

The  Geophysical  Laboratory  members  of  the  group  also  worked  with  a 
staff  member  of  the  Geological  Survey  of  Finland,  O.  Kouvo,  on  the  dating 
in  two  orogenic  (mountain-building)  belts  in  Finland:  the  Karelian  belt 
extending  from  southeastern  Finland  northwesterly  to  Finnish  Lapland, 
and  the  Svecofennian  extending  east-west  in  southern  Finland.  It  is  gener- 
ally believed  by  geologists  that  the  Svecofennian  belt  is  older  than  the 
Karelian.  The  radioactive  dating  work,  however,  gives  strong  evidence  that 
the  intrusion  of  igneous  rocks  occurred  about  1.9  billion  years  ago  in  both 
orogenic  belts,  and  the  two  orogenies  therefore  are  approximately  con- 
temporary. 

The  radioactive  dating  group  has  also  compiled  a  new  map  of  age  dis- 
tribution in  crystalline  basement  rocks  of  North  America.  This  shows  one 
belt  of  rocks,  ranging  from  0.9  to  1.2  billion  years  old,  extending  from 
Labrador  to  Texas;  another,  1.2  to  1.55  billion  years  old,  occupying  a  large 
part  of  the  central  and  southwestern  part  of  the  country;  a  third,  2.0  to  2.8 
billion  years  of  age,  from  the  Rocky  Mountains  northeastward  over  the 
Laurentian  Shield  to  Quebec;  and  still  another,  1.55  to  2  billion  years  old, 
in  Alberta  and  northwestern  Canada.  A  picture  of  the  geographical  differ- 
entiation of  ancient  rocks  in  North  America  is  thus  beginning  to  emerge. 

In  cooperation  with  the  University  of  Basel,  Switzerland,  the  Department 
completed  the  installation  of  a  polarized  ion  source  in  the  departmental 
accelerator  during  the  year.  It  consists  of  a  discharge  tube  for  the  production 
of  atomic  hydrogen,  diaphragms  and  pumps  for  defining  the  atomic  beam, 
a  quadrupole  magnet  for  selecting  and  focusing  the  atoms  having  the  de- 
sired orientation,  an  ionizer  for  the  atomic  beam,  and  a  device  for  pre- 
accelerating  and  focusing  the  ionized  atoms.  The  machine  was  operated 
successfully.  It  is  planned  to  use  the  polarized  deuteron  beam  in  the  study 
of  a  number  of  nuclear  reactions,  thus  returning  the  Department  more 
directly  to  the  field  of  nuclear  physics  than  at  any  time  since  the  end  of 
World  War  II.  For  more  than  fifteen  years  after  the  mid- 1 920 's  the 
Department  maintained  a  pioneering  effort  in  nuclear  physics,  operating 
one  of  the  first  accelerators  in  this  country. 

26  Mass  spectrometer — an  instrument  for  determining  the  masses  of  atoms  or  molecules  in  a 
gas,  liquid,  or  solid.  In  it  a  beam  of  ions  is  directed  through  electric  and  magnetic  fields  so  as  to 
produce  a  mass  spectrum  identifiable  by  an  electrical  detector. 


48  CARNEGIE     INSTITUTION      OF      WASHINGTON 

The  Geophysical  Laboratory 

Other  than  the  Terrestrial  Magnetism  program,  geophys- 
ical research  in  1905  was  not  carried  on  within  the  premises 
1905         of  the  Institution  but  nonetheless  was  considered  an  im- 
portant part  of  the  total  program.  It  was  the  type  of  project 
that   President  Woodward   advocated   continuing,   in  his 
" Suggestions  Concerning  Pending  Problems  of  the  Institution."27  Indeed, 
a  large  part  of  the  total  geophysical  program  in  that  year  was  carried  on 
in  close  collaboration  with  the  United  States  Geological  Survey  in  Wash- 
ington, thus  commencing  a  friendly  professional  relation  that  has  continued 
ever  since.  The  two  principal  investigators  of  that  year,  Arthur  L.  Day  and 
G.  F.  Becker,  held  appointments  in  the  Survey  even  though  a  substantial 
proportion  of  Dr.  Day's  time  was  spent  on  Institution  projects.  Included 
was  a  three-month  visit  by  Dr.  Day  to  Europe  for  the  purpose  of  studying 
laboratory  equipment  for  geophysical  research  and  making  an  inventory 
of  European  research. 

Becker's  research  was  concerned  entirely  with  an  effort  to  determine 
experimentally  the  relation  between  stress  and  strain.  The  main  part  of  his 
apparatus  was  a  3-inch  tube  480  feet  long  erected  in  the  Washington  Monu- 
ment, within  which  steel  tapes  were  suspended.  He  made  some  observations 
by  means  of  this  equipment  during  the  year.  In  another  project,  F.  D. 
Adams  of  McGill  University  conducted  experiments  on  the  cubic  compressi- 
bility, the  modulus  of  shear,  and  the  flow  of  rocks,  in  which  hundred-ton 
pressures  were  used. 

The  heart  of  the  1905  program,  however,  lay  in  the  work  of  Dr.  Day. 
Much  of  his  time  was  spent  in  setting  up  his  newly  designed  laboratory 
equipment.  It  comprised,  among  other  apparatus,  a  furnace  capable  of 
reaching  2100°C  in  oxidizing  or  reducing  atmospheres,  a  large  electric 
furnace  in  which  pressures  up  to  500  pounds  or  a  vacuum  could  be  main- 
tained, and  a  water-pressure  plant  capable  of  reaching  2000  atmospheres. 
A  similar  plant  capable  of  reaching  3000  atmospheres  was  under  construc- 
tion. Dr.  Day's  research  included  the  completion  of  a  three-year  investi- 
gation of  the  lime-soda  feldspar28  group  of  rocks.  His  results  showed  "that 
the  lime-soda  feldspars  form  a  continuous  series  of  mixed  crystals  capable 
of  stable  existence  in  any  proportion  of  the  two  component  minerals. " 
Experimental  proof  of  this  isomorphism  was  established  by  correlating 
melting  points  with  change  in  the  mixes  of  the  two  components.  Experi- 
ments also  were  conducted  on  wollastonite  (CaSi03),  determining  for  the 
first  time  the  exact  temperature  of  crystallization  of  this  mineral  as  found 
in  nature. 

27  Year  Book  4,  pp.  28-29. 

28  Feldspar  is  one  component  of  granite. 


REPORT   OF   THE    PRESIDENT  1^9 

Inspired  by  his  thought  on  silicates,  Dr.  Day  already  was  looking  toward 
the  future,  as  he  mentioned  two  practical  problems  to  which  his  laboratory 
later  contributed  most  significantly.  He  notes  that  "the  study  of  lime-silica 
mixtures  is  fundamental  in  the  preparation  of  Portland  cement.  Questions 
of  technical  interest  in  glass  manufacture  reappear  everywhere  in  handling 
silicate  solutions."29  He  concluded  in  a  satisfied  vein,  "grave  doubts  were 
entertained  as  to  the  feasibility  of  handling  physical  phenomena  at  high 
temperatures  with  anything  like  the  certainty  attained  at  ordinary  tem- 
peratures, but  the  experience  of  this  first  year  has  justified  the  effort  ...  ." 
If  Dr.  Day  could  look  in  on  the  Geophysical  Laboratory  of 
today  he  should  feel  greatly  gratified,  both  because  his 
1961-1962      beginning  work  in  1905  accurately  forecast  a  direction  and 
method  of  research  that  continues  to  be  highly  productive 
after  nearly  sixty  years  and  because  of  the  enormously  great 
range  and  resolving  power  of  the  methods  now  in  use. 

The  techniques  upon  which  the  Laboratory  depend  have  become  enor- 
mously more  powerful  and  more  sensitive  than  in  Dr.  Day's  time.  The 
3000-atmosphere  pressures,  which  were  tremendous  to  Dr.  Day,  have  been 
succeeded  in  1961-1962  by  pressures  of  100,000  atmospheres.  Moreover, 
these  elevated  pressures  can  be  employed  in  combination  with  almost  any 
temperature  needed  in  geophysical  experiment.  In  Dr.  Day's  1905  experi- 
ments, high  temperatures  could  be  accompanied  by  a  pressure  of  only  a 
few  hundred  pounds.  The  present-day  Laboratory  has  firm  grasp  of  these 
tools,  and  it  applies  them  to  the  whole  range  of  problems  on  the  frontiers 
of  modern  geology.  From  the  first  explorations  of  the  potentiality  of  these 
geophysical  techniques  it  has  arrived  at  the  full  power  of  applying  them  to 
revelation  of  the  earth's  interior  and  its  history.  Furthermore,  the  capacities 
of  the  Laboratory  now  include  a  wide  variety  of  techniques — beyond  those 
of  high  temperature  and  high  pressure — taken  from  modern  physics, 
chemistry,  and  mathematics.  The  Department  of  1961-1962  included  work 
in  experimental  petrology,  statistical  petrology,  crystallography,  ore 
minerals,  meteorite  analysis,  geothermal  calculations,  the  ages  of  rocks  and 
minerals,  and  organic  geochemistry. 

Among  the  numerous  investigations  carried  on  in  these  fields  in  1961- 
1962,  three  will  be  described  briefly  to  illustrate  more  in  detail  the  charac- 
teristics of  research  at  the  Geophysical  Laboratory.  These  are  experimental 
petrology,  in  which  much  of  the  work  this  year  was  focused  on  pyroxene 
minerals,  and  emphasized  the  study  of  phase  equilibria30  at  higher  pressures; 

29  Later  work  of  the  Laboratory  made  fundamental  contributions  to  the  technology  of  both 
industries. 

30  In  chemical  terms,  any  crystalline  compound  or  liquid  is  a  phase ;  hence,  a  mineral  separated 
from  a  rock  is  also  a  phase.  Assemblages  of  phases  (or  minerals)  which  do  not  melt  or  react  at  a 
particular  temperature  and  pressure  are  said  to  be  at  equilibrium.  Study  of  these  mineral  equi- 
libria is  a  means  of  understanding  the  conditions  of  formation  of  rocks. 


50  CARNEGIE     INSTITUTION      OF      WASHINGTON 

the  mineralogy  of  meteorites;  and  organic  geochemistry,  including  analysis 
of  Precambrian  carbonaceous  materials. 

The  program  of  studying  the  mineralogical  composition  of  meteorites, 
begun  last  year,  continued  to  produce  most  interesting  results.  Particularly 
relevant  as  a  preview  of  the  solid  matter  to  be  found  in  the  spatial  environs 
of  the  earth,  the  meteorites  studied  are  continuing  to  yield  mineralogical 
surprises.  P.  Ramdohr  and  G.  Kullerud  examined  more  than  a  hundred 
stony  meteorites  during  the  year,  rinding  in  them  fourteen  new  minerals 
thought  to  be  observed  for  the  first  time  anywhere.  Only  one  of  them  has 
been  given  a  name,  the  others  being  referred  to  simply  by  letters  of  the 
alphabet  for  the  time  being.  Because  they  occur  in  amounts  too  small  to 
permit  performance  of  standard  chemical  analyses  or  X-ray  powder 
diffraction  studies,  the  component  elements  in  only  two  have  been  identified. 
These  were  a  nickel-iron  sulfide  [(NiFe)2S]  called  the  Henderson  phase,  and 
a  colorless  mineral  of  spinel31  type  (Mg2Ti04).  Several  of  the  remaining 
twelve  minerals  are  thought  to  be  sulfides,  and  one,  having  an  hexagonal 
layered  structure,  seems  to  be  a  compound  of  iron,  carbon,  and  sulfur.  One 
is  thought  to  contain  arsenic.  The  electron  probe  is  considered  to  have 
promising  potentialities  for  assisting  in  the  chemical  identification  of  these 
minerals.  Another  method  of  identification  of  the  new  phases  is  synthesis, 
once  the  major  constituents  are  surmised  from  deductions  about  the  origin 
of  the  minerals.  Ramdohr  and  Kullerud  state  that  their  efforts  in  this 
direction  are  increasingly  successful. 

Ramdohr  and  Kullerud  also  made  a  number  of  observations  on  distinctive 
structural  and  textural  phenomena  in  meteorites.  They  include  evidences  of 
mechanical  distortion  and  crystallization  in  many  meteorites,  evidence  of 
spontaneous  melting  in  the  interior  of  many,  and  the  effects  of  terrestrial 
weathering,  which  may  yield  products  that  may  be  mistaken  for  primary 
components.  Magnetite  (Fe304)  frequently  may  be  such  a  product.  In 
another  set  of  analyses  on  meteorites,  S.  P.  Clark,  Jr.,  has  identified  an 
unknown  mineral  in  tektites  (glassy  bodies  probably  of  meteoric  origin)  as 
schreibersite  (Fe3P).  He  concludes  that  the  content  of  minor  elements  in 
meteoric  bodies,  like  sulfur,  phosphorus,  or  carbon,  should  be  helpful  in 
identifying  the  number  of  meteoric  falls  in  complex  fields  like  those  of 
southeast  Asia  or  Australia.  Presumably  the  minor  elements  would  be  the 
same  in  each  fall  but  would  differ  in  separate  falls. 

Another  development  of  1961-1962  meriting  special  mention  is  the  study 
of  phase  equilibria  at  high  pressures.    This  study  has  extended  over  sev- 

31  Spinel  is  typically  magnesium  aluminate  (MgO-Al203),  but  it  has  a  wide  variety  of  forms 
containing  ferrous  iron,  manganese,  ferric  iron,  and  chromium.  It  may  be  red,  yellow,  green, 
black,  or  some  other  color.  A  general  formula  is  R"0  -R/'^Os,  where  R"  may  be  one  of  the  bivalent 
metals,  magnesium,  zinc,  manganese,  iron,  nickel,  cobalt,  or  cadmium,  and  R"'  may  be  trivalent 
aluminum,  cobalt,  iron,  chromium,  or  gallium. 


REPORT   OF   THE   PRESIDENT  51 

eral  years  and  has  drawn  increasing  effort  by  Laboratory  Staff  Members. 

Geochemical  studies  at  pressures  up  to  100,000  atmospheres  have  per- 
mitted geologists  to  take  a  fresh  approach  to  various  problems  that  have 
been  the  subject  of  spirited  theoretical  discussion  for  decades.  Is  the 
Mohorovicic  discontinuity  a  phase  change  from  basalt  to  eclogite?32  Is  it 
the  same  under  the  continents  as  under  the  oceans?  What  is  the  mineralogy 
of  the  earth's  mantle?33  Can  the  various  types  of  basaltic  lava  be  related  to 
variations  in  the  melting  of  mantle  rocks  at  different  depths  and  pressures? 
What  temperatures  are  present  in  the  lower  mantle  and  core?34  As  yet  none 
of  the  questions  can  be  fully  answered,  but  the  high-pressure  studies  of  the 
last  five  years  have  contributed  to  an  understanding  of  all  and  promise  to 
contribute  far  more. 

The  Mohorovicic  discontinuity  continues  to  be  one  of  the  more  absorbing 
geological  problems.  High-pressure  high-temperature  experiment  with  the 
synthesis  of  rocks  expected  at  the  depths  of  the  discontinuity  has  given 
some  indication  of  the  rocks  to  be  found  there.  Under  the  continents  they 
are  principally  basalt  and  eclogite.  Basalt  is  transformed  by  high  pressure 
to  the  denser  eclogite.  Eclogite  consists  essentially  of  jadeite-bearing 
pyroxene  and  pyrope-bearing  garnet.35  Both  jadeite  [NaAl(Si03)2]  and 
pyrope  (Mg3Al2Si30i2)  are  high-pressure  phases,  and  their  pressure- 
temperature  fields  of  stability  have  been  established  in  recent  years  at 
elevated  temperatures.  Significantly,  both  the  reactions  leading  to  the 
formation  of  jadeite  and  pyrope  take  place  in  a  relatively  narrow  pressure- 
temperature  range.  The  experimental  results  now  indicate  that  in  the  depth 
range  50  to  100  kilometers  in  the  mantle,  where  basaltic  lava  is  believed  to 
form,  the  mineral  assemblage  will  be  characteristic  of  eclogites.  The  experi- 
mental data  for  the  transition  fit  reasonably  well  the  hypothesis  that  the 
continental  Mohorovicic  discontinuity  is  a  basalt-eclogite  transition.  The 
nature  of  the  discontinuity  under  the  oceans  apparently  is  different  from 
the  continental,  and  is  a  challenging  question  for  future  thought  and 
experiment. 

Additional  experimental  data  for  constructing  concepts  of  the  earth's 
mantle  and  crust  are  being  obtained  in  quantity  at  the  Laboratory  from  an 
examination  of  the  melting  relations  of  silicates  at  high  pressure.  As  a  result 
the  present-day  conceptions  of  reactions  by  which  basalts  form  in  the 
partial  fusion  of  mantle  rocks  are  wholly  different  from  those  of  earlier 
workers.  The  system  of  petrology  developed  by  N.  L.  Bowen  and  others 

32  A  dense  rock  equivalent  in  composition  to  basalt,  found  in  association  with  Russian  and 
South  African  diamond  pipes,  and  occurring  in  rocks,  elsewhere  on  the  earth's  surface,  thought 
to  originate  from  deep  in  the  earth's  mantle. 

33  That  part  of  the  earth's  interior  between  the  Mohorovicic  discontinuity  and  the  core. 

34  The  core  is  thought  to  commence  at  a  depth  of  about  2900  kilometers. 

35  Garnet  has  the  general  formula  R"R'"(Si04)3,  where  R"  may  be  bivalent  iron,  magnesium, 
manganese,  or  calcium,  and  R'"  may  be  trivalent  iron,  aluminum,  or  chromium. 


52  CARNEGIE     INSTITUTION      OF      WASHINGTON 

earlier  at  this  Laboratory  from  experiments  at  atmospheric  pressure 
successfully  explained  many  characteristics  of  igneous  rocks.  It  now  is  clear, 
however,  that  pressures  as  low  as  10,000  to  20,000  atmospheres  produce 
very  pronounced  changes  in  crystal-liquid  equilibria  in  silicate  rock  systems. 
Even  though  the  data  on  phase  relations  at  high  pressures  still  do  not 
permit  the  construction  of  a  system  of  petrology  for  the  lower  crust  and 
upper  mantle  of  the  earth,  answers  to  some  important  questions  are  being 
obtained. 

One  of  the  intriguing  questions  concerned  the  formation  of  silica-saturated 
basalt  rocks  of  the  crust  from  silica-undersaturated  mantle  rocks.  F.  R. 
Boyd,  Jr.,  and  J.  L.  England  experimented  during  the  year  with  the  melting 
of  pyrope  garnet,  thought  to  be  an  important  constituent  of  the  mantle,  at 
pressures  prevailing  where  basalts  are  considered  to  be  formed.  They  found 
that  pyrope  garnet  melts  incongruently  at  these  pressures  to  spinel  and 
liquid.  This  melting  relationship  could  explain  the  formation  of  the  silica- 
saturated  basalts,  such  as  are  found  in  Hawaii,  from  the  typical  minerals 
assumed  to  be  in  the  upper  mantle. 

Continuing  the  experiments  reported  in  Year  Book  60,  H.  S.  Yoder,  Jr., 
and  C.  E.  Tilley  examined  the  possible  origin  of  alkali  basalt  and  tholeiitic 
basalt,  two  groups  of  rocks  that  are  very  important  components  of  the 
earth's  crust.  Their  previous  experiments  with  natural  rocks  and  synthetic 
mineral  systems  established  that  the  same  magma  (liquid  rock),  depending 
on  pressure,  could  yield  both  types  of  basalt.  They  now  have  suggested 
mechanisms  whereby  both  alkali  and  tholeiitic  basalt  may  be  generated 
from  an  eclogitic  liquid  deep  within  the  earth's  mantle. 

Sydney  P.  Clark,  Jr.,  J.  F.  Schairer,  and  John  de  Neufville  have  attacked 
the  same  problem  with  a  different  approach.  They  also  believe  that  it  is 
necessary  to  examine  critically  important  systems  of  minerals  in  their 
entirety  under  pressure  before  inferences  about  melting  and  solidification 
within  the  mantle  can  be  drawn  with  confidence.  They  chose  to  examine 
the  important  but  complicated  quaternary  system36  that  includes  among 
its  phases  the  oxides  spinel  and  corundum,  forsterite  (Mg2Si04),  diopside 
(CaMgSi206),  pyrope  garnet,  various  forms  of  silica,  and  still  other  minerals. 
Their  experiments  were  conducted  at  atmospheric  pressure  and  at  a  pressure 
of  20,000  atmospheres.  Their  observations  showed  a  range  of  solid  solution 
in  pyroxene  minerals  (e.g.,  diopside  and  others)  at  high  pressures  that  is 
far  more  extensive  than  in  the  same  system  at  atmospheric  pressure. 
Pressure  therefore  undoubtedly  produces  profound  changes  in  the  melting 
relations  within  at  least  this  mineral  system.  For  some  of  the  compositions, 
the  system  at  20,000  atmospheres  is  not  even  qualitatively  similar  to  the 

36  Quaternary  system — a  system  of  phase  relations  among  minerals  having  four  end  members, 
schematically  expressible  in  a  tetrahedral  diagram. 


REPORT    OF   THE    PRESIDENT  53 

system  at  atmospheric  pressure,  as  in  the  appearance  of  quartz  on  the 
liquidus  above  1000°C.  The  experiments  showed  that  this  system  is  well 
suited  to  the  study  of  the  complex  chemical  equilibria  at  high  pressures. 
Further  study  should  yield  important  contributions  to  the  petrology  of  the 
earth's  rocks  at  depth. 

Heat  is  the  source  of  energy  for  most  geological  processes,  and  knowledge 
of  the  temperatures  at  the  depth  of  the  core-mantle  boundary37  in  the  earth 
is  of  fundamental  importance.  It  is  probable  that  the  temperature  at  this 
depth  is  not  far  below  the  minimum  melting  temperature  of  rocks  in  the 
lower  mantle  and  not  far  above  the  solidifying  temperature  of  the  iron- 
nickel  alloy  believed  to  comprise  the  outer  core.  Data  obtained  at  low  pres- 
sures showed  that  the  slopes  of  silicate  melting  curves  were  two  to  five 
times  greater  than  the  slopes  of  the  melting  curves  of  most  metals.  However, 
recent  results  for  diopside  and  a  few  other  silicates  at  pressures  up  to  50,000 
atmospheres  yield  diagrams  with  slopes  having  a  pronounced  curvature. 
Extrapolation  of  the  diopside  data  to  a  pressure  at  the  core-mantle  bound- 
ary38 indicates  that  a  temperature  of  3750°C  would  be  required  to  melt 
diopside  there.  Similar  extrapolation  of  data  on  the  melting  of  iron  indicates 
a  temperature  of  5200°  at  the  boundary.  Although  the  uncertainties  in  these 
extrapolations  are  very  numerous,  it  is  interesting  that  the  estimates  are 
close.  They  are  furthermore  in  rough  agreement  with  estimates  made  by 
other,  equally  uncertain,  methods. 

An  elegant  example  of  the  application  of  the  sensitive  and  powerful 
modern  research  techniques  in  geophysics  to  a  problem  that  scarcely  could 
have  been  touched  even  a  decade  ago  is  shown  in  the  identification  of  com- 
pounds characteristic  of  life  from  ancient  rocks.  P.  H.  Abelson  and  P.  L. 
Parker  have  isolated  fatty  acids  from  rocks  as  old  as  500,000,000  years.  This 
is  the  oldest  known  occurrence  of  these  substances.  Among  the  compounds 
identified  were  the  saturated  acids  myristic  [CH3(CH)i2C02H],  palmitic 
[CH3(CH)14C02H],  and  stearic  [CH3(CH)16C02H],  the  last  being  the 
most  abundant.  Although  the  quantities  found  are  minute  (10  micrograms39 
per  gram  of  organic  carbon),  gas-liquid  chromatography  permits  isolation, 
identification,  and  quantitative  measurement  of  the  individual  acids,  even 
when  major  amounts  of  impurities  are  present. 

The  same  fatty  acids  were  isolated  from  recent  sediments.  Palmitic  acid 
was  the  major  component  in  the  young  rocks,  being  as  much  as  ten  times  as 
abundant  as  stearic  acid,  the  more  abundant  in  the  old  rocks.  Thus  although 
the  fatty  acids  in  very  young  and  in  old  rocks  are  qualitatively  similar  a 
puzzling  quantitative  difference  has  been  noted. 

37  Postulated  to  be  at  a  depth  of  about  2900  kilometers. 

38 1,400,000  atmospheres. 

39  A  microgram  is  0.000001  gram. 


54  CARNEGIE     INSTITUTION      OF      WASHINGTON 

T.  C.  Hoering  has  investigated  two  important  aspects  of  the  geochemical 
record  of  very  early  life  on  earth.  He  has  studied  some  of  the  earth's  oldest 
sedimentary  rocks,  which  contain  structures  geologists  consider  related  to 
algal  activity.  He  has  measured  stable  carbon  isotope  ratios,  C13/C12,  in 
coexisting  carbonates  and  reduced  carbon  obtained  from  these  specimens. 
He  has  found  that  the  isotopes  have  been  fractionated  into  a  C13-enriched 
carbonate  phase  and  a  C13-depleted  reduced  carbon  phase.  The  amount  of 
this  fractionation  is  nearly  identical  to  that  found  in  contemporaneous  algal 
cells  and  their  associated  carbonates.  The  magnitude  of  the  effect  is  also 
similar  to  that  found  between  limestones  and  coals  of  all  geological  ages. 
Such  isotope  fractionation  is  caused  by  a  slightly  different  rate  of  photo- 
synthesis for  molecules  of  carbon  dioxide  containing  C12  as  compared  with 
those  containing  C13. 

The  samples  examined  include  a  limestone  from  the  Belt  Series  of  Glacier 
Park,  Montana,  with  a  minimum  age  of  1.2  billion  years,  the  Randville 
dolomite  of  Crystal  Falls,  Michigan,  with  a  minimum  age  of  1.5  billion 
years,  and  the  Bulawayan  limestone  of  Southern  Rhodesia  with  a  minimum 
age  of  2.7  billion  years.  In  these  rocks,  which  are  among  the  oldest  known 
sedimentary  rocks,  the  isotopic  evidence  is  consistent  with  the  presence  of 
photosynthetic  algae  in  the  very  early  Precambrian  era. 

The  second  study  by  Hoering  was  on  the  reduced  carbon  of  Precambrian 
sedimentary  rocks.  A  successful  effort  was  made  to  extract  and  partially 
identify  organic  molecules  from  them.  By  means  of  a  number  of  chemical 
degradations  he  was  able  to  liberate  soluble  fractions  from  the  insoluble 
"fabric"  of  the  reduced  carbon.  The  fractions  were  analyzed  with  the  aid  of 
ultraviolet  spectroscopy  and  chromatography.  The  results  indicate  that  the 
insoluble  reduced  carbon  may  be  related  to  the  kerogen  of  more  recent 
rocks.  Kerogen  is  produced  by  interactions  of  organic  products  of  cells  when 
deposited  in  sediments  deprived  of  oxygen.  Thus  additional  evidence  has 
been  produced  pointing  to  the  existence  of  life  in  very  early  Precambrian 
times,  more  than  two  billion  years  ago. 


Mount  Wilson  and  Palomar  Observatories 

In  1905  the  astronomical  activities  of  the  Institution  were 

mainly  those  of  the  Solar  Observatory,  whose  building  and 

1905         equipment  were  under  construction  during  the  year,  with 

view  to  completion  in  1906.  Mount  Wilson  was  considered  an 

especially  favorable  site  because  "The  unusually  favorable 

atmospheric  conditions  which  prevail  day  and  night  at  the  site  of  the 

observatory  have  attracted  the  attention  of  astronomers  and  astrophysi- 


REPORT   OF   THE   PRESIDENT  55 

cists  generally. "40  "It  has  been  been  found  that  the  average  night-seeing  is 
exceedingly  good,  while  the  low  wind-velocity,  coupled  with  the  trans- 
parency of  the  atmosphere,  afford. . .  advantages  which  should  render  Mount 
Wilson  an  ideal  site  for  the  5-foot  reflector."  George  E.  Hale,  the  Director, 
defined  his  purposes  as:  "(1)  The  investigation  of  the  sun  (a)  as  a  typical 
star,  in  connection  with  the  study  of  stellar  evolution:  (b)  as  the  central 
body  of  the  solar  system,  with  special  reference  to  possible  changes  in  the 
intensity  of  its  heat  radiation,  such  as  might  influence  the  conditions  of  life 
upon  the  earth.  (2)  The  choice  of  an  effective  mode  of  attack,  involving  (a) 
the  application  of  new  methods  in  solar  research;  (6)  the  investigation  of 
stellar  and  nebular  phenomena,  especially  such  as  are  not  within  the  reach  of 
existing  instruments ;  and  (c)  the  interpretation  of  these  celestial  phenomena 
by  means  of  laboratory  experiments."  He  was  at  this  time  already  con- 
sidering the  design  of  "a  large  reflecting  telescope  and  of  new  types  of 
instruments."  He  also  looked  forward  to  "The  furtherance  of  international 
cooperation  in  astrophysical  research  through  the  invitation  to  Mount 
Wilson,  from  time  to  time,  of  investigators  especially  qualified  to  take 
advantage  of  the  opportunities  afforded.  .  .    ." 

A  large  part  of  Dr.  Hale's  report  in  1905  was  necessarily  devoted  to  a 
statement  on  the  numerous  construction  projects  that  had  absorbed  his 
attention  during  the  year.  They  included  ten  buildings  on  Mount  Wilson, 
and  the  Pasadena  office  and  shop,  which  were  constructed  on  land  given  by 
citizens  of  Pasadena.  Dr.  Hale  nonetheless  found  time  not  only  for  instru- 
ment testing  but  also  for  an  observing  program  and  planning  a  future 
research  program.  Daily  direct  photographs  of  the  sun  on  a  scale  of  6.7 
inches  to  the  solar  diameter  were  taken  on  the  Snow  telescope.  Observations 
were  made  to  test  an  hypothesis  of  Dr.  Hale's  about  the  relation  of  calcium 
vapor  to  the  faculae  and  plages41  of  the  sun.  Some  experimental  study  of 
the  spectra  of  sunspots,  plages,  and  the  chromosphere  was  undertaken  for 
instrument  design.  Photographs  were  also  taken  of  bright  stars  with  a  long- 
focus  grating  spectrograph.  They  included  a  photograph  of  the  blue  region 
of  the  first-order  spectrum  of  Arcturus,  which  required  an  exposure  of  14 
hours  on  three  successive  nights. 

Visiting  investigators  had  already  found  their  way  to  Mount  Wilson.  E. 
E.  Barnard  of  the  Yerkes  Observatory  photographed  the  southern  part  of 
the  Milky  Way,  described  by  Dr.  Hale  as  "a  most  important  contribution 
to  our  knowledge  of  the  structure  of  the  Milky  Way  and  of  the  remarkable 
nebulae  within  it."   The  Smithsonian  Institution  also  sent  an  expedition 

40  Year  Book  4,  p.  25. 

41  Faculae — small  irregular  bright  patches  in  the  photosphere  (visible  disk)  of  the  sun,  sur- 
rounding sunspots. 

Plages — faculae  of  the  chromosphere,  which  is  the  outer  layer  of  the  sun's  "atmosphere," 
extending  to  a  height  of  several  thousand  kilometers  from  the  visible  disk. 


56  CARNEGIE     INSTITUTION     OF      WASHINGTON 

to  the  mountain  for  observing  solar  radiation,  directed  by  C.  G.  Abbot. 
Other  astronomical  work  supported  by  the  Institution  in  1905  included 
the  compilation,  by  Lewis  Boss  of  the  Dudley  Observatory,  Albany,  New 
York,  of  a  Preliminary  General  Catalogue  of  Stars  for  the  6000  stars  visible 
to  the  naked  eye.  Also  included  were  grants  to  Simon  Newcomb  of  Washing- 
ton, D.  C.,  for  an  " Investigation  of  the  mean  motion  of  the  moon,"  and  a 
rather  enigmatical  grant  "To  aid  investigations  in  mathematical  astronomy, 
statistical  methods,  and  economic  science."  The  economic  science  part  was 
never  reported  upon,  either  in  1905  or  in  the  four  succeeding  years  when 
Dr.  Newcomb  held  sequel  grants. 

The  program  of  1961-1962  at  the  Observatory,  as  for  each  of 

the  four  preceding  Departments,  differed  vastly  from  its 

1961-1962       ancestor  of  sixty  years  ago.  The  primary  emphasis  on  solar 

studies  gave  way  in   1918  to  a  more  general  astronomy 

program  with  the   completion  of  the   100-inch  telescope. 

Nonetheless,  solar  observation  and  solar  study  have  continued  to  the  present 

day,  but  with  gradually  decreasing  emphasis.  A  major  change  came  in  1958- 

1959,  with  a  decision  to  drastically  curtail  routine  observations.  Since  then 

solar  studies  have  centered  on  the  sun's  magnetic  fields,  of  which  daily 

observations  are  made  with  the  aid  of  the  solar  magnetograph  originated 

and  developed  by  H.  D.  and  H.  W.  Babcock  of  the  Observatories.  Daily 

solar  magnetograms  have  been  made  since  1957. 

During  the  1961-1962  year  R.  F.  Howard  commenced  an  extensive  study 
of  the  accumulated  magnetograms  to  classify  magnetic  regions,  and  cor- 
related them  with  optical  and  radio  phenomena.  He  has  already  obtained 
the  very  interesting  finding  that  the  unipolar  magnetic  (UM)  regions  of  the 
sun  correlate  in  their  position  with  calcium  absorption  phenomena  observed 
spectroscopically.  It  may  thus  now  become  possible  to  extend  observation 
of  UM  regions  backward  for  50  years  or  more,  using  the  Observatory's 
extensive  collection  of  spectroheliograms  showing  the  absorption  lines  of  the 
elements. 

Responding  somewhat  to  the  explosion  of  national  interest  in  inter- 
planetary space,  the  year  was  also  marked  at  the  Observatories  by  renewed 
attention  to  the  planets,  which  have  been  subject  to  recurring  study  at  the 
Observatories  in  the  past.  It  has  seemed  important  to  press  ground-based 
observations  like  those  that  can  be  undertaken  at  the  Observatories  to  the 
limits  made  possible  with  new  photometric  and  infrared  techniques,  be- 
cause information  about  the  planets  can  be  acquired  by  these  techniques  at  a 
cost  of  much  less  effort  and  money  than  by  observations  from  rockets.  G. 
Munch,  with  the  collaboration  of  H.  Spinrad  of  the  Jet  Propulsion  Labora- 
tory of  the  California  Institute  of  Technology,  and  R.  Younkin  of  the  same 
laboratory,    began   studies    of  the   spectra    of    the   major   planets.    Two 


REPORT    OF   THE    PRESIDENT  57 

lines  of  the  hydrogen  molecule  were  found  in  the  spectrum  of  Saturn,  pro- 
viding the  first  firm  evidence  of  the  presence  of  hydrogen  in  the  atmosphere 
of  that  planet.  Spinrad  also  analyzed  high-dispersion  spectra  of  Venus, 
finding  evidence  of  large  changes  in  the  temperature  of  the  atmosphere  of 
Venus.  B.  Murray  of  the  California  Institute  of  Technology  continued 
studies  of  the  photoelectric  colorimetry  of  the  moon  with  the  Mount 
Wilson  facilities. 

A  major  part  of  the  Observatories'  program,  however,  has  been  devoted  to 
a  study  of  the  masses,  luminosities,  surface  temperatures,  and  chemical 
composition  of  stars,  and  the  variation  of  luminosity  and  surface  tempera- 
ture with  age.  During  recent  decades  these  have  been  among  the  major 
problems  in  astronomy.  Even  though  such  research  has  become  increasingly 
important  with  time,  steps  toward  the  modern  understanding  of  these 
phenomena  date  back  to  the  early  years  of  this  century.  The  first  important 
step  was  taken  by  E.  Hertzsprung  and  H.  N.  Russell,  when  they  plotted  a 
diagram  of  the  absolute  magnitude  of  stars  in  the  solar  neighborhood  against 
their  surface  temperatures  as  indicated  by  spectral  class  or  color.  They 
found  that  most  stars  fall  in  a  narrow  diagonal  band  on  their  diagram,  the 
very  hot  giants  being  at  one  end  and  the  cool  dwarfs  at  the  other.  Later 
theoretical  investigations  based  on  nuclear  physics  showed  that  the  fusion  of 
hydrogen  into  helium  was  an  important  source  of  energy  for  most  stars42 
and  that  stars  obtaining  their  energy  from  this  reaction  logically  fall  on  the 
color-magnitude  diagram  in  the  narrow  "main  sequence"  band  noted  by 
Hertzsprung  and  Russell. 

During  the  second  world  war,  Walter  Baade  of  the  Observatories  made  a 
detailed  investigation  of  the  stellar  content  of  the  Andromeda  galaxy.  He 
found  that  the  brightest  stars  in  its  spiral  arms  are  very  hot  giants  similar  to 
those  in  the  solar  neighborhood,  which  also  is  on  a  spiral  arm.  In  the  nucleus, 
however,  the  brightest  stars  were  cool  red  giants.  To  differentiate  these, 
Baade  introduced  the  concept  of  Population  I  (younger)  stars  typically  on 
the  spiral  arms  and  Population  II  (older)  stars  typically  at  galactic  centers. 

Theory  then  predicted  that,  as  the  hydrogen  fuel  approaches  exhaustion 
in  a  stellar  core,  a  star  expands  greatly  but  cools  and  thereby  moves  off  the 
narrow  main-sequence  band  in  the  color-magnitude  diagram  and  becomes  a 
red  giant.  Since  the  brightest  stars  use  their  fuel  most  rapidly  this  change 
starts  at  the  upper  end  of  the  main  sequence  and  moves  down  the  sequence 
with  time.  Obviously,  the  hot  giants  in  the  solar  neighborhood  and  in 
galactic  spiral  arms  indicate  a  population  of  stars  that  have  formed  re- 
cently, whereas  the  red  giants  in  a  galactic  nucleus  represent  a  population 
of  old  stars.  Theory  permits  one  to  go  even  further  and  fix  the  age  of  a  group 

42  The  hydrogen-helium  reaction  is  now  considered  ancillary  to  the  hydrogen-deuterium-helium 
reaction. 


58  CARNEGIE     INSTITUTION     OF      WASHINGTON 

of  stars  by  observing  the  magnitude  at  which  stars  are  just  beginning  to 
move  off  the  main  sequence.  Color-magnitude  diagrams  have  been  con- 
structed for  a  large  number  of  globular  and  galactic  clusters43  by  A.  R. 
Sandage,  H.  C.  Arp,  and  W.  A.  Baum  of  the  Observatories,  and  many 
others.  Ages  of  a  few  million  up  to  ten  billion  or  more  years  have  been  found. 

With  the  aid  of  high-dispersion  spectra  it  has  become  possible  recently  to 
make  detailed  quantitative  chemical  analyses  of  stellar  atmospheres.  The 
first  studies  of  the  sun  and  of  bright  nearby  stars  indicated  a  surprising  uni- 
formity of  chemical  composition.  When  these  measurements  were  extended 
to  some  of  the  distant  older  clusters,  however,  it  was  found  that  their  stars 
were  deficient  in  the  heavier  metallic  elements,  often  by  factors  of  100  or 
more  compared  with  the  stars  near  the  sun.  Since  most  of  the  strong 
metallic  lines  fall  in  the  ultraviolet  ( U)  region  of  the  spectrum,  a  star  of  high 
metallic  content  exhibits  a  depressed  U  region  compared  with  the  blue  (B) 
or  green-yellow  (V,  " visible' ')  spectral  regions.  Within  the  past  two  years 
astronomers  at  the  Observatories  have  found  it  possible  to  fix  the  metallic 
content  from  a  comparison  of  the  magnitudes  of  a  star  measured  in  the 
U,  B,  and  V  regions.  This  makes  feasible  the  extension  of  abundance 
determinations  to  stars  far  too  faint  for  detailed  spectrum  analysis. 

In  general,  old  stars  such  as  those  in  the  globular  clusters,  or  high- velocity 
stars,44  which  presumably  were  formed  at  the  same  time  as  those  in  the 
galactic  nucleus,  are  metal  deficient  compared  with  the  younger  stars. 
Theory  suggests  that  metals  are  formed  late  in  the  evolution  of  a  star,  after 
the  hydrogen  fuel  has  been  exhausted  in  the  stellar  core  and  the  central 
temperature  has  increased  to  many  times  that  of  stars  on  the  main  sequence. 
Therefore  the  metal-containing  material  in  recently  formed  stars  has  gone 
through  one  or  more  earlier  generations  of  stars  in  which  the  metals  are 
formed  and  then  blown  off  into  space  either  in  a  gradual  flow45  or  explosively 
in  a  nova  or  supernova  outburst. 

Obviously,  a  project  to  understand  stars  will  require  decades  for  comple- 
tion as  well  as  investigation  by  many  astronomers  at  a  number  of  observa- 
tories. The  Institution  can  take  pride  not  only  in  the  participation  of  the 
Observatories  in  the  grand  conception  of  such  a  project  but  also  in  their 
preeminent  position  as  contributor  of  observational  data  leading  to  widen- 
ing views  of  the  universe  which  these  studies  are  providing.  During  the  year 
1961-1962  the  staff  of  the  Observatories  skillfully  exploited  the  wonderful 
instruments  at  their  disposal  to  give  us  further  insights  on  this  frontier  of 

43  A  globular  cluster  is  a  group  of  many  thousands  of  stars  arranged  in  a  regular  form  showing 
spherical  symmetry.  Many  are  located  outside  the  plane  of  the  Milky  Way.  A  smaller  group  of 
stars  always  found  near  the  plane  of  the  Milky  Way  is  known  as  a  galactic  cluster. 

44  A  high-velocity  star  is  a  star  that  is  moving  about  the  galactic  nucleus  with  a  velocity 
markedly  different  from  that  of  the  sun. 

45  As  studied  by  A.  Deutsch  of  the  Observatories.  See  Year  Book  59,  p.  8,  and  other  year  books. 


REPORT    OF   THE    PRESIDENT 


59 


logT  4.5 


M 


bol 
-6 

-4 
-2 

0 
+2 
+4 
+6 
+8 
♦10 
♦12 


o 

O    o         ° 
SUPERGIANTS 


G  ft  NTS 


UBGIANTS 


BLUE 
STARS 


WHITE 
STARS 


YELLOW 
STARS 


RED 
STARS 


-8 

(BRIGHTEST) 

-6 

-4 
-2 

0 

+2 

+4 

+6 

+8 

+10 

+12 

(FAINTEST) 


Generalized  Hertzsprung-Russell  diagram  of  star  color-magnitude  relation.  Log  T  —  logarithm  of 
temperature,  degrees  Kelvin;  Mboi  =  bolometric  magnitude,  as  measured  from  calculated  total 
energy  emission.  (Adapted  from  Cecilia  Payne-Gaposchkin,  Introduction  to  Astronomy,  Prentice- 
Hall,  New  York,  1954.) 


60  CARNEGIE     INSTITUTION     OF      WASHINGTON 

the  universe.  Among  the  results  were  new  knowledge  about  the  differences 
in  chemical  composition  among  stars,  the  correlation  of  chemical  composi- 
tion and  star  movement,  a  determination  of  the  time  of  formation  of  the 
Galaxy  in  which  we  are  located,  and  new  evidence  on  the  expansion  of  the 
cosmos. 

The  staff  of  the  Observatories  participated  in  detailed  chemical  investiga- 
tions of  a  number  of  stars.  J.  Greenstein  and  R.  A.  Parker  of  the  Observa- 
tories have  collaborated  with  G.  Wallerstein  of  the  University  of  California, 
H.  L.  Heifer  of  the  University  of  Rochester,  and  L.  Aller  of  the  University  of 
Michigan  to  study  one  group  of  three  red  giant  stars.  They  found  that  the 
common  metals  were  only  1/500  as  abundant  in  this  group  as  in  the  sun,  and 
the  heavy  elements  strontium,  zirconium,  barium,  cerium,  and  europium 
were  deficient  by  a  factor  of  25,000.  Considering  the  deficiencies,  they 
estimate  that  these  stars,  which  are  part  of  our  Galaxy,  probably  condensed 
within  a  few  hundred  million  years  after  the  formation  of  the  Galaxy.  In 
investigating  several  dozen  peculiar  B  and  A  stars,46  J.  Jugaku,  W.  L.  W. 
Sargent,  and  L.  T.  Searle  found  that  the  abundances  of  individual  elements 
vary  erratically  compared  with  neighboring  elements  in  the  periodic  table, 
often  fluctuating  by  factors  of  100  or  more.  Elements  found  to  have  marked 
over-  or  underabundance  in  certain  stars  are  beryllium,  carbon,  nitrogen, 
oxygen,  silicon,  phosphorus,  and  mercury. 

For  some  years  a  group  of  stars  have  been  recognized  and  designated  as 
subdwarfs  because  they  lie  appreciably  below  the  main  sequence  on  the 
color-magnitude  diagram.  Early  studies  showed  that  they  were  metal- 
deficient,  therefore  old,  stars.  They  have  high  velocities  considered  in 
relation  to  the  sun.  To  learn  more  about  these  rather  rare  stars,  A.  Sandage 
and  C.  T.  Kowal  have  started  a  program  for  the  photoelectric  observation  of 
the  ultraviolet-blue-visible  magnitudes  of  the  high-velocity  stars  given  in 
the  Giclas  Proper  Motion  Catalogue.  More  than  100  new  metal-deficient 
subdwarfs  have  been  discovered  among  the  700  stars  observed  thus  far. 
Spectroscopic  studies  by  Greenstein  and  by  Sandage  of  an  enlarged  sample 
of  these  subdwarfs  confirmed  the  high  velocity  of  all. 

In  a  further  effort  to  obtain  information  on  the  relation  of  the  subdwarfs 
to  other  principal  groups  of  stars,  O.  J.  Eggen  and  Sandage  studied  the  effect 
of  "line  blanketing"47  on  the  position  of  a  star  in  the  color-magnitude  dia- 
gram. The  results  were  of  special  astronomical  interest,  for  Eggen  and 
Sandage  found  that  if  proper  correction  is  made  for  line  blanketing  the  sub- 
dwarf  stars  move  into  the  same  position  as  normal  dwarf  stars  on  the  main 

46  The  accepted  spectral  classification  of  stars  designates  them  by  arbitrary  letters  as  O,  B,  A, 
F,  G,  K,  and  M.  O  and  B  stars  have  the  highest  temperatures,  and  M  the  lowest. 

47  Line  blanketing  refers  to  the  situation  in  which  the  abundance  in  a  star  of  the  metallic 
elements  having  strong  absorption  bands  in  the  ultraviolet  is  so  great  that  it  causes  an  appreciable 
deficiency  in  the  spectral  region  compared  with  other  parts  of  the  spectrum  of  the  star. 


REPORT    OF    THE    PRESIDENT  61 

sequence  of  the  color-magnitude  diagram.  Thus  another  addition  was  made 
to  our  understanding  of  the  wonderful  order  which  astronomers  have  been 
slowly  illuminating  with  the  aid  of  modern  instruments. 

Eggen,  D.  Lynden-Bell,  and  Sandage  also  studied  the  orbits  around  the 
nucleus  of  our  Galaxy  of  a  large  number  of  dwarf  stars,  including  both  the 
normal  and  subdwarf  types.  They  find  a  close  correlation  between  metal 
deficiency  and  the  eccentricity  and  angular  momentum  of  the  stellar  orbit. 
They  interpret  this  as  indicating  that  metal-deficient  stars  were  formed  in 
an  early  period  while  our  Galaxy  was  rapidly  contracting.  From  the  age  of 
these  stars  they  were  able  to  fix  the  time  of  formation  of  our  Galaxy  out  of 
the  medium  of  the  universe  at  about  ten  billion  years  ago. 

Studies  of  stellar  properties  are  important  not  only  for  understanding  the 
characteristics  of  the  stars  themselves  but  also  to  provide  a  firm  basis  for 
the  measurements  on  which  the  conceptions  of  the  structure  and  origin  of 
the  entire  universe  depend.  For  example,  nearly  all  determinations  of  large 
astronomical  distances  depend  on  the  comparison  of  the  apparent  brightness 
of  a  nearby  object  with  that  of  an  identical  object  in  a  distant  cluster  or 
galaxy.  Thus  cepheid  variables48  were  used  by  E.  P.  Hubble  to  fix  the  dis- 
tance of  the  nearby  Andromeda  galaxy,  and  the  galaxies  themselves  were 
used  to  estimate  the  distances  of  clusters  of  galaxies  at  the  extreme  range 
of  telescopic  penetration  into  space.  However,  the  discovery  of  different 
stellar  populations  with  major  differences  in  age  and  chemical  composition 
raised  many  doubts  about  the  identity  in  absolute  magnitude  of  a  star  in 
our  own  neighborhood  with  that  of  a  star  in  a  nearby  galaxy,  which  might 
or  might  not  have  similar  age  or  chemical  composition. 

One  of  the  uncertainties  in  these  extrapolations  toward  a  picture  of  the 
universe  has  been  the  effect  of  light  absorption  by  dust  clouds  along  the 
path  between  star  and  observer.  This  is  especially  troublesome,  since  the 
shorter  wavelengths  of  the  spectrum  are  absorbed  more  strongly  than  the 
longer,  producing  a  reddening  effect.  During  the  year  H.  C.  Arp  reexamined 
this  problem  in  color-magnitude  studies  of  globular  clusters  of  stars.  He 
found  that  the  correction  for  absorption  should  be  appreciably  larger  than 
had  been  allowed  formerly.  A  substantial  revision  downward  of  previously 
determined  globular  cluster  ages  therefore  must  be  made.  This  also  elimi- 
nates a  discrepancy  existing  between  age  determined  from  position  on  the 
color-magnitude  diagram  and  age  determined  from  models  of  cosmological 
expansion.  They  now  become  consistent. 

Extrapolations  to  distant  galaxies  are  also  handicapped  because  most 
are  too  distant  to  permit  observation  of  enough  stars  for  the  construction  of 
a  color-magnitude  diagram.  However,  it  is  possible  to  analyze  the  integrated 

48  Stars  whose  light  emission  varies  in  a  definite  pattern  over  a  relatively  short  period  but 
longer  than  24  hours. 


62  CARNEGIE     INSTITUTION     OF      WASHINGTON 

light  received  from  a  galaxy  and  from  it  obtain  information  about  the 
distribution  in  temperature,  magnitude,  and  chemical  composition  of  the 
component  stars.  W.  A.  Baum  has  studied  some  nearby  galaxies  of  different 
types  by  photoelectric  scanning  methods  in  order  to  obtain  information 
about  the  evolution  of  galaxies.  His  evidence  indicates  that  some  definitely 
are  composed  of  true  Population  II  (older)  stars  whereas  others  (large 
ellipticals)  have  mainly  Population  I  stars.  Such  observations  are  important 
in  the  construction  and  interpretation  of  cosmological  models.  Present 
interpretation  of  the  observable  universe  conceives  it  as  having  a  radius  of 
five  billion  or  so  light  years,49  expanding  at  its  limits  of  observation  at 
nearly  half  the  speed  of  light.  This  interpretation  depends  on  assumptions 
made  about  the  magnitude-redshift  relation,  that  is  the  reddening  of  the 
observed  spectrum  caused  by  recession  of  the  distant  galaxies  in  relation  to 
the  solar  system.  Distant  galaxies,  of  course,  are  seen  at  an  earlier  age 
than  nearby  ones — billions  of  years  of  difference  for  the  most  distant.  Since 
individual  stars  undergo  large  changes  in  luminosity  and  temperature  with 
age,  the  observable  integrated  light  of  a  galaxy  also  changes  with  time. 
How,  for  example,  has  the  extremely  distant  galaxy  3C295  (now  redesig- 
nated 1410+5224),  mentioned  in  Year  Book  59,  changed  in  the  five  billion 
years  since  the  observed  light  that  fell  on  the  Palomar  photographic  plate 
left  the  galaxy?  Answers  to  questions  like  these  will  be  obtained  only  from 
studies  such  as  those  undertaken  by  Baum  and  other  Staff  Members  of  the 
Observatories  on  stellar  properties  and  evolution. 


The  Department  of  Embryology 

Although  the  Department  of  Embryology  was  not  estab- 
lished until  1914,  when  it  was  organized  by  Franklin  P.  Mall, 
1905        even  its  subject  was  not  ignored  among  the  activities  of  the 
Institution  in  1905.  In  that  year  L.  B.  Mendel  of  Yale 
University  was  given  a  grant  for  "  Study  of  physiology  of 
growth,  especially  in  its  chemical  processes/ '  Professor  Mendel's  grant  was 
renewed  in  each  of  several  years  thereafter.  He  reported  that  he  was 
studying  the  " chemical  composition  of  the  developing  animal  body  and 
the  equipment  of  this  organism  for  its  nutrition,  upon  which  growth 
essentially  depends.  Data  are  being  collected  at  first  hand  regarding  the 
composition  of  various  embryonic  tissues  at  different  stages  of  embryonic 
growth.  For  the  nervous  system  a  correlation  between  morphological  and 

49  Light  year — the  distance  traveled  by  light  in  a  vacuum  during  one  year;  about  5.88  X  1012 
miles. 


REPORT    OF   THE    PRESIDENT  63 

chemical  development  is  already  apparent.  The  chemistry  of  embryonic 
muscle  is  also  already  under  investigation.  "The  purin  content  of  the  liver 
and  muscles  at  various  embryonic  stages  has  been  determined.  ...  It  is 
hoped  ...  to  ascertain  whether  the  purin  metabolism  of  the  young  is 
essentially  different  from  that  of  the  adult."50  A  grant  to  L.  E.  Griffin  was 
also  made  in  the  same  year  "to  secure  material  for  a  study  of  the  embry- 
ology, histology,  and  physiology  of  the  Nautilus."  Studies  supported  at  the 
Marine  Biological  Laboratory,  Woods  Hole,  Massachusetts,  included  one 
on  the  "segmentation  of  certain  fertilized  eggs";  on  "regenerative  processes 
and  structures";  and  on  "muscle-fibers  of  the  fish  heart." 

These  studies,  however,  were  not  in  any  sense  an  organized  group.  Nor 
did  they  command  a  major  interest  on  the  part  of  the  Institution's  admin- 
istration, as  was  shown  by  the  termination  of  this  type  of  grant  at  the  end 
of  1908.  It  remained  for  Dr.  Mall  to  set  in  1914  the  lines  on  which  the 
Department  continued  so  long,  an  examination  of  the  morphology  and 
histology  of  the  human  embryo  and  the  embryonic  physiology  of  primates. 
Even  at  the  beginning  of  the  Department,  however,  other  organisms  were 
studied,  as  illustrated  by  the  1914  study  of  E.  L.  and  E.  R.  Clark  on  the 
movements  of  the  lymph  heart  in  living  chick  embryos,  and  their  report  in 
that  year  "that  the  muscle  of  the  lymph  heart  is  derived  from  the 
myotomes."51 

The  year  1961-1962  was  marked  by  the  setting  of  an  im- 
portant milestone  in  the  history  of  the  Department.  After 
1961  1962       several  years  of  preparation  the  new  Department  of  Embry- 
ology building,  adjacent  to  the  Homewood  Campus  of  the 
Johns  Hopkins  University  in  Baltimore,   Maryland,   was 
completed.  This  building,  specially  designed  for  embryological  research, 
should  free  the  staff  of  the  Department  from  the  inconveniences  that 
attended  work  in  their  former  cramped  quarters  at  the  Johns  Hopkins 
Medical  School  near  the  center  of  the  city.  The  Department  started  to 
move  on  August  1,  1961,  and  was  able  to  assume  full  operation  by  early 
November,  in  spite  of  a  long  delay  in  equipping  the  building  because  of  an 
electricians'  strike.  The  new  building  appears  to  have  met  with  staff  ap- 
proval. Director  J.  D.  Ebert  describes  it  as  having  "an  unusual  combination 
of  fine  qualities,  pleasing  to  both  aesthetic  and  practical  senses." 

The  Department  in  1961-1962  is  described  by  Dr.  Ebert  in  the  introduc- 
tion of  his  report  of  this  year  as  one  holding  to  its  "traditional  organization 
of  a  group  of  independent  investigators  whose  interests  range  widely  from 
biochemistry  and  microbiology  to  anatomy  and  physiology,  with  sub- 
stantial overlapping  in  experience  and  approach.     ...  in  developmental 

60  Year  Book  4,  pp.  259-260. 

51  Year  Book  13,  p.  112.  A  myotome  is  a  muscle  mass  in  a  developing  animal. 


64  CARNEGIE     INSTITUTION      OF      WASHINGTON 

biology  today  it  appears  to  favor  the  generation  and  interchange  of 
ideas.  .  .    ." 

The  multifaceted  program  Dr.  Ebert  describes  included  an  interesting 
study  of  the  physiologic  aspect  of  frog-embryo  growth  from  the  stage  of  the 
fertilized  egg  onward  by  D.  D.  Brown  and  J.  D.  Caston,  the  nature  of  the 
testicular  antigen  in  induced  aspermatogenesis52  by  G.  L.  Carlson  and  D. 
W.  Bishop,  the  role  of  deoxyribonuclease  II  during  the  metamorphosis  of 
the  tadpole  by  J.  R.  Coleman,  a  comprehensive  study  of  the  developing 
human  eye  by  R.  O'Rahilly,  and  still  others. 

Of  particular  interest  among  these  was  the  Brown-Caston  study  of  the 
embryonic  development  of  the  frog  Rana  pipiens.  They  found  that  the 
early  embryos  contain  a  measurable  but  small  population  of  ribosomes  in 
their  cells.  The  early  ribosomal  content  changes  little  until  a  stage  near  the 
end  of  morphogenesis,53  when  there  is  a  very  rapid  appearance  of  more 
particles.  This  coincides  with  the  time  when  the  embryo  has  been  shown  to 
require  magnesium  ions  from  outside.  In  addition,  the  iron  storage  molecule, 
ferritin,  was  definitely  identified  in  the  egg.  Also,  although  ribosomal 
synthesis  was  shown  to  begin  after  much  of  morphogenesis  is  completed, 
high-molecular-weight  RNA,  with  a  base  composition  identical  to  ribosomal- 
ion  RNA,  was  found  to  be  present  in  all  stages  of  the  embryo. 

Perhaps  the  most  striking  progress  to  be  reported  from  the  Department 
during  the  year  came  in  the  studies  of  I.  R.  Konigsberg,  who  joined  the 
Institution  as  a  Staff  Member  on  July  1,  1961.  They  will  be  described  in 
some  detail  as  an  illustration  of  the  methods  and  approach  of  the  present- 
day  Department. 

From  its  beginning  the  Department  of  Embryology  has  numbered,  among 
its  Staff  Members,  investigators  dedicated  to  the  study  of  the  development, 
structure,  chemistry,  and  physiology  of  muscles.  W.  H.  and  M.  R.  Lewis 
(Department  of  Embryology,  1915-1940)  pioneered  in  analyzing  the  origin 
of  muscle  fibrils  in  tissue  culture;  and  Arpad  Csapo  (1949-1954)  was  among 
the  first  students  of  muscle  chemistry  to  characterize  the  contractile  pro- 
teins of  the  uterus  and  to  examine  their  regulation  under  different  physio- 
logical conditions.  More  recently  J.  D.  Ebert  and  R.  L.  DeHaan  and  their 
associates  have  focused  attention  on  the  biochemistry  of  developing  con- 
tractile proteins  and  on  morphogenetic  movements  and  relations  of 
contractile  and  conductile  cells  in  the  heart.  D.  W.  Bishop  has  contributed 
importantly  to  our  understanding  of  mechanisms  in  primitive  contractile 
systems  like  sperm  tails. 

To  this  roster  the  Department  now  adds  Konigsberg's  name.  During  the 
year  he  made  substantial  progress  in  a  hitherto  refractory  subject,  the 

62  Destruction  of  the  power  to  produce  sperm. 

63  The  emergence  of  the  specific  structure  of  an  animal  during  embryonic  development. 


REPORT    OF   THE    PRESIDENT  65 

investigation  of  the  cytodifferentiation  of  embryonic  skeletal  muscle  cells  in 
dispersed  cell  culture.  His  system  of  culture  is  designed  to  offer  greater 
opportunity  for  rigorous  control  of  both  the  quantitative  aspects  of  the 
cellular  population  and  the  extracellular  environment  than  can  be  achieved 
either  in  vivo  or  in  organ  culture.  Many  years'  experience  by  numerous 
previous  investigators  suggested  that  such  culturing  techniques  could  be 
expected  to  promote  the  loss  of  differentiative  character  and  would  not 
favor  a  progressive  increase  in  the  effects  of  cell  specialization  on  mor- 
phology. No  generally  satisfactory  explanation  for  this  previously  observed 
incompatibility  has  ever  been  given.  However,  Konigsberg's  results  with 
monolayer  cultures  of  embryonic  skeletal  muscle  cells  are  in  striking  dis- 
agreement with  expectations  from  earlier  experience. 

Monolayer  cultures  prepared  from  suspensions  of  11-  to  12-day  chick 
embryonic  leg  muscle  pass  through  three  recognizable  phases.  The  period 
immediately  following  plating  of  the  cells  is  marked  by  rapid  proliferation 
with  a  mean  generation  time  of  24  hours.  During  this  period  cultures  consist 
exclusively  of  mononucleated  cells  and  have  the  general  appearance  of 
cultures  of  "fibroblast-like"54  cells  such  as  might  be  derived  from  a  great 
variety  of  tissues.  The  transition  from  the  first  to  the  second  phase  occurs  in 
a  matter  of  hours  and  is  characterized  by  the  formation  of  long  multinuclear 
"ribbonlike"  cells.  Formation  of  these  multinuclear  cells  coincides  with  the 
attainment  of  cell  confluence  in  the  culture.  The  effect  of  cell  density  is 
further  suggested  by  experiments  in  which  the  inoculum  size  was  varied. 
The  smaller  the  inoculum,  the  greater  the  time  of  transition  from  phase  one 
("fibroblast-like"  cell)  to  phase  two  (multinuclear  "ribbon"),  and  vice 
versa.  The  abrupt  appearance  of  multinucleated  myotubes55  in  this  second 
phase  is  paralleled  by  an  equally  abrupt  break  downward  in  the  rate  of 
proliferation.  Again,  the  time  required  for  this  development  can  be  shifted 
by  varying  the  inoculum  size. 

Differentiation  beyond  the  stage  of  the  mononucleated  myoblast56  occurs 
in  culture  after  cells  have  ceased  rapid  multiplication.  This  observation  is 
consistent  with  Konigsberg's  earlier  findings,  as  well  as  with  those  from 
several  laboratories,  that  myotube  nuclei  are  postmitotic57  and  that  they 
form  by  cellular  fusion.  The  third  phase  of  muscle  differentiation  in  culture 
is  characterized  by  the  progressive  development  of  the  cross-striated 
myofibrillar  pattern  and  the  initiation  of  the  spontaneous  contraction  char- 
acteristic of  muscular  tissue. 

All  Konigsberg's  studies  before  the  past  year  had  been  restricted  to  mono- 

64  Fibroblasts — elongated  mononuclear  cells  which  develop  into  and  are  also  part  of  connective 
tissue. 

65  Aggregated-cell  constituent  of  muscle. 

66  Unassociated  single  "premuscle"  cell. 

67  Mitosis  is  cell  division. 


66  CARNEGIE     INSTITUTION      OF      WASHINGTON 

layer  cultures  established  with  inocula  of  1  million  to  2.5  million  cells  each. 
Such  cultures  reach  confluence  between  the  second  and  fourth  day  of 
culture,  depending  on  the  size  of  the  inoculum.  To  probe  for  the  lower  limit 
of  inoculum  size  that  would  still  permit  differentiation  to  occur  he  turned  to 
the  single-cell  plating  technique  developed  by  T.  T.  Puck  and  his  associates. 
In  this  procedure  small  numbers  of  cells  are  dispersed  over  a  relatively  large 
area.  During  appropriate  periods  of  incubation  the  individual  cells  give  rise 
to  discrete  colonies  visible  to  the  naked  eye.  The  technique  has  been  applied 
most  successfully  to  permanently  established  cell  strains. 

Using  freshly  isolated  embryonic  muscle  cells  Konigsberg  observed  a 
plating  efficiency  of  approximately  10  per  cent.  In  plates  cultivated  for  10 
to  13  days  approximately  1  in  10  colonies  exhibited  unmistakable  signs  of 
skeletal  muscle  cell  differentiation.  The  proportion  of  differentiated  cells 
ranged  from  colonies  containing  several  elongated  myotubes  in  colonies  of 
predominantly  mononucleated  cells  to  colonies  in  which  virtually  every 
nucleus  was  in  syncytial58  association.  Under  polarized  light  or  bright-field 
illumination  after  staining,  the  myotubes  showed  the  presence  of  longitu- 
dinal fibrils,  which  frequently  exhibited  the  pattern  of  cross  striation  typical 
of  mature  skeletal  muscle  cells.  It  is  apparent  that  some  myoblasts,  at  least, 
through  a  sequence  of  rapid  multiplications,  can  produce  a  large  number  of 
progeny  that  retain  the  capacity  for  differentiation. 

Two  major  questions  emerged  from  these  observations.  First,  what  is  the 
significance  of  the  finding  that  only  1  in  10  colonies  eventually  differentiates? 
Second,  what  is  the  stimulus  initiating  myotube  formation?  Konigsberg  is 
attacking  the  second  problem  by  examining  the  relationship  of  cell  density 
to  myotube  formation.  Two  general  mechanisms  by  which  cell  density 
might  affect  myotube  formation  were  considered.  Since  myotube  formation 
is  a  result  of  cell  fusion,  high  cell  density  might  ensure  that  a  sufficient 
number  of  effective  cell-to-cell  collisions  occur.  Another,  and  equally  likely, 
possibility  is  that  a  high  cell  density  may  be  either  supplementing  the 
culture  medium  with  cell  products  or  removing  some  components. 

Konigsberg  designed  experiments  to  test  that  possibility.  His  first  tests 
showed  that  the  medium  is  altered  by  the  metabolic  activity  of  cells 
cultured  in  it.  In  cultures  grown  on  a  medium  preconditioned  by  the  pres- 
ence of  other  cells,  myotube  formation  commenced  as  much  as  24  hours 
earlier  than  initial  cultures  of  equal  numbers  of  cells  from  the  same  cell 
suspension  but  cultured  in  fresh  medium.  Furthermore,  the  cells  in  condi- 
tioned medium  attached  to  the  glass  more  firmly,  presenting  a  strikingly 
different  appearance  from  the  control  cultures. 

These  results  are  impressive  in  themselves,  and  indeed  they  represent 
something  of  a  technical  breakthrough  in  the  difficult  task  of  cell  culture. 

58  Referring  to  a  multinucleated  aggregate  of  imperfectly  separated  cells,  or  a  multinuclear  cell. 


REPORT   OF   THE    PRESIDENT  67 

But  as  so  often  in  science  they  are  probably  more  important  for  the  questions 
they  raise  than  for  the  results  they  give.  Already  they  have  pointed  the  way 
to  a  number  of  additional  experiments  to  probe  the  relation  between  condi- 
tioned media  and  cell  differentiation.  But  in  the  hint  given  of  a  hitherto 
unsuspected  closeness  of  relation  between  cell  and  environment  we  touch  a 
problem  of  wide  application  and  perhaps  vast  significance  in  understanding 
all  higher  forms  of  life. 


Although  a  report  describing  work  like  Konigsberg's  can  give  something 
of  the  sense  of  high  adventure  experienced  by  scientists  within  the  Institu- 
tion and  elsewhere,  there  are  dimensions  to  the  scientific  life  of  today  that 
must  always  escape  any  progress  report.  Most  of  those  who  work  within  the 
Institution  share  a  deep  conviction  about  the  humanity  of  their  calling  and 
about  the  community  of  fellowship  that  not  only  is  vital  to  the  progress  of 
their  work  but  also  is  a  deeply  felt  reward  in  itself.  Happily  these  convic- 
tions occasionally  shine  through  more  esoteric  daily  concerns.  They  are 
notable  this  year  in  the  comments  of  James  Ebert  and  Merle  Tuve,  each  on  a 
point  of  his  philosophy. 

Ebert  has  written  particularly  of  his  own  deep  attachment  to  the  essential 
unity  of  living  science.  He  quotes  Frank  R.  LilhVs  memorable  words  that 
"  Scientific  discovery  is  a  truly  epigenetic  process  in  which  the  germs  of 
thought  develop  in  the  total  environment  of  knowledge."  The  life  of  the 
laboratory,  where  one  must  be  quick  to  acknowledge  what  has  gone  before, 
alert  to  the  current  actions  of  others  having  similar  interests,  and  mindful  of 
the  needs  for  others  to  know,  can  be  a  social  experience  almost  beyond 
comparison.  Dr.  Ebert  notes  his  pleasure  at  having  visiting  investigators 
from  other  institutions:  "They  do  contribute  vitally  to  the  Department .  .  . 
but  of  far  greater  moment  is  the  question  whether  such  a  visit  adds  measur- 
ably to  the  man's  ability  as  an  investigator  and  teacher  when  he  returns  to 
his  home  laboratory.  Has  he  found  new  direction  or  meaning  for  his  re- 
search? Has  the  opportunity  for  reflection.  .  .  led  to  a  searching  reexamina- 
tion of  his  program?"  With  pride,  the  Institution  can  record  that  its 
Departments  provided  literally  hundreds  of  such  opportunities  during  the 
year. 

Tuve's  comments  touch  upon  the  aesthetic  experience  of  being  a  scientist, 
and  upon  what  is  perhaps  one  of  the  deepest  motivations  in  "exact"  science. 
Contrasting  it  with  the  disorder  and  transience  he  sees  in  the  life  of  men  in 
the  mass,  he  expresses  his  admiration  at  "the  beautiful  regularity  and 
systematic  relatedness.  .  .  in  every  aspect  of  the  natural  phenomena.  .  .  from 
distant  stars  to  living  bacteria."  He  considers  this  a  cause  of  the  sense  of 
very  deep  satisfaction  in  scientific  studies.  Through  science,  man,  bit  by  bit, 


68  CARNEGIE     INSTITUTION      OF      WASHINGTON 

is  adding  to  his  stature  and  to  "his  awe  of  the  stupendous  and  beautifully 
intricate  universe  in  which  he  finds  himself."  Tuve  considers  it  a  "great  good 
fortune"  for  scientists  to  be  able  to  devote  their  energy  and  talents  to 
illuminating  "the  intricate  and  orderly  patterns  of  the  physical  world  around 
us."  To  him  this  is  "a  princely  gift  of  our  time  and  circumstances." 

Such  motivations  lead  to  a  dedication  which  is  the  wonder  of  all  who  have 
not  experienced  their  attractions.  It  is  a  dedication  measured  only  in  part  by 
a  voluntary  70-hour  week,  by  long  nights  on  a  mountaintop  in  below-freezing 
weather,  by  a  hundred  frustrations  with  equipment  design,  or  by  a  willing- 
ness to  work  at  the  modest  salaries  that  fundamental  science  is  able  to 
provide.  We  can  hope  that  Andrew  Carnegie,  after  sixty  years,  might  be 
approving  of  both  the  dedication  and  the  insights  of  these  men  and  their 
predecessors  as  they  have  striven  "to  secure,  if  possible,  for  the  United 
States  of  America  leadership  in  the  domain  of  discovery.  .  .  of  new  forces."59 

Losses  .  .  . 

I  must  report  with  great  sorrow  the  loss  of  a  devoted  member  of  the 
Board  of  Trustees,  the  Honorable  Robert  Woods  Bliss,  and  of  a  highly 
valued  Staff  Member  of  the  Mount  Wilson  and  Palomar  Observatories, 
Don  0.  Hendrix. 

Robert  Woods  Bliss,  a  Trustee  of  the  Institution  for  twenty-six  years, 
died  in  Washington,  D.  C,  on  April  19,  1962.  Elected  a  Trustee  in  1936,  he 
became  a  member  of  the  Executive  Committee  the  following  year.  He  was 
Secretary  of  the  Board  of  Trustees  from  1953  until  his  death.  He  also  served 
continuously  from  1939  on  the  Committee  on  Archaeology,  from  1939  to 
1945  on  the  Auditing  Committee,  and  from  1950  to  1953  and  1958  to  1961 
on  the  Nominating  Committee. 

Before  his  association  with  the  Institution  he  had  already  had  a  distin- 
guished career  of  33  years  in  the  diplomatic  corps  of  the  United  States, 
where  he  held  many  important  posts.  He  was  especially  concerned  with 
efforts  to  bring  about  world  security  through  arms  control  and  international 
organization.  In  1908  he  was  United  States  delegate  to  the  International 
Conference  to  Consider  Measures  for  the  Revision  of  Arms  and  Ammunition 
Regulations  in  Brussels.  As  counselor  to  our  embassy  in  Paris  from  1916  to 
1919  he  assisted  in  preparations  for  the  Versailles  Peace  Conference  and  in 
its  work.  Again,  in  1921,  he  was  a  member  of  the  United  States  delegation 
to  the  Washington  Conference  on  the  Limitation  of  Armaments.  His 
beautiful  estate,  Dumbarton  Oaks,  in  Washington,  was  the  scene  of  the 
conference  that  laid  plans  for  the  United  Nations. 

69  Andrew  Carnegie,  Trust  Deed  Creating  a  Trust  for  the  benefit  of  the  Carnegie  Institution 
of  Washington,  D.  C,  January  28,  1902. 


REPORT    OF   THE    PRESIDENT  69 

Just  before  his  retirement  in  1933  he  had  served  for  six  years  as  am- 
bassador to  Argentina;  and  during  World  War  II  he  was  called  back  from 
his  technical  retirement  to  serve  as  special  consultant  and  then  special 
assistant  to  the  Secretary  of  State. 

Mr.  Bliss  will  be  remembered  by  the  Washington  community  for  his 
many  philanthropic  and  cultural  contributions.  The  Institution  will 
remember  his  dedication  to  its  welfare  and  his  gentle  but  always  penetrating 
counsel  on  every  problem. 

Another  loss  that  is  especially  felt  is  that  of  Don  O.  Hendrix  of  the  Mount 
Wilson  and  Palomar  Observatories,  who  died  on  December  26,  1961,  at  the 
age  of  57.  Joining  the  staff  of  the  Observatories  in  1913,  Hendrix  became 
Superintendent  of  its  optical  shop  in  1947,  where  he  carried  out  such  im- 
portant projects  as  the  optical  design  for  the  48-inch  schmidt  telescope  and 
the  final  figuring  of  the  200-inch  mirror  after  it  had  been  moved  to  Palomar. 
His  extraordinary  skill  was  largely  responsible  for  the  high  efficiency  of  the 
present  equipment  of  the  Observatories. 


With  keen  regret  I  also  record  the  loss  to  the  Institution  of  four  retiring 
members  of  the  staff.  Dr.  Berwind  P.  Kaufmann,  Director  of  the  Depart- 
ment of  Genetics,  Dr.  Robert  K.  Burns,  Staff  Member  of  the  Department  of 
Embryology,  Mrs.  Ruth  L.  McCollum,  Assistant  to  the  President,  and 
Wilbur  A.  Pestell,  Administrative  Assistant  at  the  Department  of  Plant 
Biology,  all  retired  on  June  30,  1962. 

Dr.  Kaufmann  came  to  the  Department  of  Genetics  in  1937  from  the 
University  of  Alabama,  where  he  had  served  for  ten  years  as  professor  and 
department  head.  Since  that  time  his  professional  interests  have  touched  on 
many  facets  of  the  broad  field  of  cytogenetics,  with  emphasis  on  the  varying 
patterns  of  chromosome  structure  that  influence  gene  action.  These  interests 
stemmed  from  experience  in  the  area  of  descriptive  cytology,  gained  in  the 
early  1920's,  when  chromosomes  were  generally  regarded  as  uniformly 
staining  rod-shaped  structures  with  no  discernibly  precise  pattern  of  internal 
organization.  By  developing  and  applying  ingenious  techniques,  Dr. 
Kaufmann  demonstrated  that  chromosomes  contain  paired,  helically  dis- 
posed strands  at  all  phases  of  somatic  and  meiotic  mitoses. 

Upon  joining  the  Institution's  staff,  Dr.  Kaufmann  undertook  an  analysis 
of  the  types  and  frequencies  of  chromosomal  rearrangements  induced  by 
ionizing  radiations,  using  the  giant  chromosomes  in  the  salivary  glands  of 
Drosophila  for  diagnostic  purposes.  His  discovery  and  evaluation  of  the 
effects  of  near-infrared  radiation  on  the  frequencies  of  X-ray-induced  re- 
arrangements was  an  outstanding  accomplishment  of  that  period. 

In  1960  Dr.  Kaufmann  succeeded  Dr.  M.  Demerec,  first  as  Acting 
Director  and  in  1961  as  Director  of  the  Department.  During  his  twenty-five 


70  CARNEGIE     INSTITUTION     OF     WASHINGTON 

years  at  Cold  Spring  Harbor  he  maintained  a  strong  interest  in  science 
education  and  in  the  training  of  young  biologists.  He  has  now  returned  to  a 
university  environment,  having  been  appointed  Professor  of  Zoology  and 
Senior  Research  Scientist  at  the  University  of  Michigan,  where  his  sincerity, 
dedication,  and  technical  skill  will  be  inspiring  to  those  who  have  the  good 
fortune  to  work  with  him. 

Dr.  Burns  joined  the  Department  of  Embryology  in  1940  from  the 
University  of  Rochester,  where  he  had  been  a  member  of  the  Department  of 
Anatomy,  of  which  Dr.  George  W.  Corner  was  the  head  before  his  own  move 
to  the  Institution.  When  he  went  to  Baltimore,  Dr.  Burns  rejoined  Dr. 
Corner  and  another  long-time  Rochester  colleague,  B.  H.  Willier,  who  had 
assumed  the  direction  of  the  Johns  Hopkins  Department  of  Biology.  Burns, 
who  held  the  title  of  Honorary  Professor  of  Biology  at  the  University, 
served  as  an  important  link  between  the  two  departments,  pointing  the 
way  to  the  close  association  that  exists  today. 

Dr.  Burns  has  devoted  his  entire  career  to  studying  the  mechanisms  of  sex 
differentiation.  A  student  of  Ross  G.  Harrison,  he  began  by  demonstrating 
sex  reversal  in  amphibians,  using  the  technique  of  embryonic  parabiosis.  His 
was  the  first  convincing  laboratory  research  following  up  Frank  R.  Lillie's 
analysis  of  the  freemartin.60  Later  he  turned  his  attention  to  mammals,  and 
again  produced  the  first  convincing  evidence  of  sex  reversal  by  the  use  of 
purified  sex  hormones  in  his  analysis  of  the  effects  of  estradiol  on  the  pro- 
spective male  opossum. 

Dr.  Burns  has  returned  to  Bridgewater  College,  where  he  received  his 
first  degree.  He  is  teaching  embryology  and  continuing  his  research  on  sex 
differentiation. 

A  loss  most  keenly  felt  by  the  President  and  the  Office  of  Administration 
was  the  retirement  of  Mrs.  Ruth  McCollum,  Administrative  Assistant  to 
the  President.  Mrs.  McCollum  joined  the  Institution  staff  in  the  administra- 
tive office  of  the  Department  of  Terrestrial  Magnetism  in  1942,  where  she 
gave  distinguished  assistance  during  the  difficult  period  of  the  war.  In  1946 
she  transferred  to  the  Bursar's  office  in  the  Office  of  Administration,  where 
she  served  for  thirteen  years,  first  as  secretary  to  the  Bursar  and  then  as 
Accountant.  During  the  latter  part  of  this  period  Mrs.  McCollum  con- 
tributed part  of  her  time  and  skill  to  general  responsibilities  of  the  Office  of 
Administration.  Early  in  1959  she  became  Administrative  Assistant  to  the 
President.  Her  management  of  arrangements  for  the  Annual  Meeting  of  the 
Board  of  Trustees  was  always  a  model  of  organization  and  good  taste.  Her 
artistic  talent  appeared  in  many  ways  in  her  work,  much  to  the  Institution's 
advantage,  as  in  the  annual  departmental  exhibits.  No  problem  was  too 
difficult  to  tax  her  good  humor,  and  long  hours  only  increased  her  devotion 

60  A  modified  female  of  bovine  heterosexual  twins. 


REPORT   OF   THE   PRESIDENT  71 

to  the  Institution.  Her  many  talents  and  fine  spirit  are  much  missed  by  all 
who  worked  with  her. 

Wilbur  A.  Pestell,  Administrative  Assistant  at  the  Department  of  Plant 
Biology,  also  retired  on  June  30, 1962.  He  was  actively  associated  with  the 
Institution  for  42  years,  a  period  of  dedicated  service  seldom  equaled  by  past 
employees.  He  worked  first  in  the  Division  of  Publications  in  Washington, 
subsequently  at  the  Desert  Botanical  Laboratory  near  Tucson,  then  at  the 
Coastal  Laboratory  at  Carmel,  California,  and  finally  as  Secretary  in  the 
Department  of  Plant  Biology  at  Stanford.  His  faithful  work  cleared  routine 
tasks  from  the  way  of  many  others  whose  scientific  results  have  been  re- 
ported in  these  Year  Books. 


.  .  .  and  Changes  .  .  . 

The  year  1962,  in  addition  to  signalizing  the  sixtieth  anniversary  of  the 
Institution,  also  marked  a  significant  change  in  its  internal  organization. 
Upon  the  retirement  of  Dr.  Berwind  P.  Kaufmann,  the  fourth  Director  of 
the  Department  of  Genetics,  on  June  30,  1962,  the  status  of  genetics  re- 
search within  the  Institution  was  altered.  As  of  July  1,  the  Department  of 
Genetics  became  the  Genetics  Research  Unit,  with  Alfred  D.  Hershey  as 
Director.  The  work  of  the  Unit  will  center  on  the  research  of  Hershey, 
Barbara  McClintock,  and  their  associates  at  Cold  Spring  Harbor.  In 
September  1962,  Helen  Gay,  another  Staff  Member  of  the  Unit,  transferred 
her  work  to  Ann  Arbor,  Michigan,  where  she  will  continue  her  association 
with  Dr.  Kaufmann. 

The  Department  of  Genetics  was  formed  in  1921  from  a  merger  of  the 
former  Department  for  Experimental  Evolution  and  the  Eugenics  Records 
Office.  The  Department  for  Experimental  Evolution,  which  was  formed  in 
1906,  had  been  preceded  by  the  Station  for  Experimental  Evolution, 
established  at  the  present  site  in  Cold  Spring  Harbor,  New  York,  during 
1904.  The  dominant  traits  of  the  Department  of  Genetics  were  clearly  those 
it  inherited  from  the  Department  for  Experimental  Evolution.  For  fifty- 
eight  years  the  research  groups  that  successively  made  Cold  Spring  Harbor 
their  scientific  home  maintained  a  research  tradition  which  in  many  ways 
has  been  the  story  of  genetics  progress  in  the  United  States.  Originally  con- 
ceived by  its  first  Director,  C.  B.  Davenport,  and  inspired  by  the  preceding 
work  of  Hugo  de  Vries  in  the  Netherlands,  the  Cold  Spring  Harbor  labora- 
tory has  had  an  uncanny  record  of  association  with  and  stimulation  of  the 
main  currents  in  genetic  thought  during  the  more  than  half  century  of  its 
existence.  Even  twenty  years  ago  Milislav  Demerec,  on  the  eve  of  his  be- 
coming the  third  Director  of  the  Department,  could  say  that  the  "backing 


72  CARNEGIE     INSTITUTION     OF     WASHINGTON 

given  to  genetical  research  by  the  Institution  undoubtedly  accounts  to  a 
large  degree  for  the  fact  that  the  United  States  now  occupies  the  leading 
position  in  this  branch  of  science."61 

The  Station's  first  work  followed  Dr.  Davenport's  lead.  He  had  a  con- 
suming ambition  to  prove  experimentally  the  broad  application  of  Mendel's 
law  as  rediscovered  in  1900  by  de  Vries  in  the  Netherlands,  Correns  in 
Germany,  and  von  Tschermak-Seysenegg  in  Austria.  Davenport's  early 
work  on  poultry,  birds,  and  mammals  did  actually  furnish  classic  experi- 
mental confirmation  of  the  broad  application  of  Mendelian  inheritance. 

Davenport's  student  and  later  colleague  G.  H.  Shull  provided  one  of  the 
most  unusual  chapters  in  the  Laboratory's  history  by  laying  the  theoretical 
foundations  of  hybrid  corn  cultivation,  described  by  Mangelsdorf  as  "the 
most  far-reaching  development  in  applied  biology  of  this  quarter  century."62 
Shull's  recognition  and  exploitation  of  heterosis  (hybrid  vigor),  which  he 
named,  gave  the  basic  principle  "which  underlies  almost  the  entire  hybrid 
corn  enterprise."63  More  recently  the  same  plant  in  the  hands  of  Barbara 
McClintock  has  been  a  highly  successful  medium  for  discovering  the  muta- 
tional behavior  of  genes.  Both  may  certainly  be  counted  among  the  most 
significant  achievements  in  genetics. 

Illustrative  of  the  range  of  interests  to  be  found  in  the  work  of  the 
Laboratory  through  its  fifty-eight  years  of  history  are  the  pioneering  experi- 
mental studies  of  C.  C.  Little  on  the  inheritance  of  tumors,  followed  later  by 
E.  C.  MacDowell's  and  J.  S.  Potter's  studies  of  mouse  leukemia;  the 
foundation  of  cytogenetics  by  John  Belling,  followed  by  the  productive 
cytogenetic  research  undertaken  on  Datura  (of  the  potato  family)  by  A.  F. 
Blakeslee,  the  second  Director  of  the  Department,  and  his  colleagues,  and 
succeeded  more  recently  by  B.  P.  Kaufmann's  cytogenetic  studies  on 
Drosophila;  the  painstaking  studies  of  Milislav  Demerec,  mapping  the  gene 
loci  of  Escherichia  coli  and  Salmonella;  and  the  work  of  A.  D.  Hershey, 
exploring  the  molecular  structure  of  the  bacterial  phage  chromosome. 

Through  the  years  the  Department  has  been  no  less  favored  by  geneticists 
who  have  been  associated  with  it  on  a  part-time  basis.  Among  the  Research 
Associates  and  Guest  Investigators  who  were  connected  with  the  Depart- 
ment at  one  time  or  another  in  its  history  were:  W.  E.  Castle,  E.  B.  Wilson, 
C.  B.  Bridges,  H.  E.  Crampton,  E.  B.  Badcock,  L.  C.  Dunn,  Th.  Dobzhan- 
sky,  M.  Delbrlick,  A.  Hollaender,  D.  G.  Catcheside,  M.  Westergaard,  C. 
Stern,  and  S.  Brenner. 

The  Institution  will  continue  support  of  genetics  research,  although  at  a 

61  Carnegie  Institution  of  Washington  Year  Book  4U  P-  170. 

62  Paul  C.  Mangelsdorf,  "Hybrid  corn:  its  genetic  basis  and  its  significance  in  human  affairs," 
in  Genetics  in  the  Twentieth  Century,  edited  by  L.  C.  Dunn,  The  Macmillan  Company,  New  York, 
1951,  p.  555. 

63  Ibid.,  p.  653. 


REPORT   OF   THE   PRESIDENT  73 

reduced  scale  by  comparison  with  the  Department's  peak  staff.  The  Genetics 
Research  Unit  will  remain  at  Cold  Spring  Harbor.  It  is  hoped  that  it  will  be 
joined  by  an  interuniversity-sponsored  research  group  investigating  quanti- 
tative biology,  the  formation  of  which  was  being  explored  at  the  year's  end. 

.  .  .  and  Gains 

Two  new  members  were  elected  to  the  Board  of  Trustees  of  the  Institution 
on  May  11,  1962:  William  Walden  Rubey  and  Carl  Joyce  Gilbert. 

Dr.  Rubey  is  one  of  the  country's  most  distinguished  geologists.  From 
1920  until  1960  he  was  associated  with  the  United  States  Geological  Survey, 
where  his  work  received  signal  recognition  in  the  Award  of  Excellence  of  the 
Department  of  the  Interior  in  1943  and  the  Distinguished  Service  award  in 
1950.  His  contributions  have  added  significantly  to  scientific  understanding 
in  several  fields  of  geology,  notably  in  knowledge  of  the  original  formation  of 
the  oceans,  the  transport  of  particles  and  sediments  by  running  water,  and 
the  mechanics  of  very  large  overthrust  faults.  He  is  a  graduate  of  the 
University  of  Missouri  and  holds  honorary  doctoral  degrees  both  from  that 
University  and  from  Yale.  At  present  he  is  serving  as  a  member  of  the 
National  Science  Board  (National  Science  Foundation)  and  of  the  board  of 
directors  of  the  American  Association  for  the  Advancement  of  Science. 

Mr.  Gilbert  is  Chairman  of  the  Board,  Gillette  Company,  Boston.  He  is  a 
graduate  of  the  University  of  Virginia  and  the  Harvard  Law  School.  He  is  a 
member  of  the  board  of  directors  of  several  corporations,  including  the 
Raytheon  Manufacturing  Company,  the  Fiduciary  Trust  Company,  and  the 
Pepperell  Manufacturing  Company.  Devoted  to  public  service  as  well  as  to 
business,  Mr.  Gilbert  is  a  member  of  the  board  of  managers  and  past  presi- 
dent of  the  Boston  Dispensary,  vice-chairman  of  the  Massachusetts  Port 
Authority,  trustee  of  the  New  England  Center  Hospital,  member  of  the 
administrative  board  of  the  New  England  Medical  Center,  vice-president 
of  the  New  England  Council,  and  trustee  and  member  of  the  executive 
committee  of  Tufts  College.  Before  he  became  chairman  of  the  board  of  the 
Gillette  Company  in  1958  he  had  served  as  its  president. 


It  is  always  a  special  pleasure  to  record  the  honors  that  have  come  to 
members  of  the  Institution. 

Presentation  of  the  Kettering  award  for  1961  to  Dr.  Vannevar  Bush, 
retired  President  of  the  Institution,  was  made  at  a  conference  in  Washing- 
ton, D.  C,  of  the  Patent,  Trademark,  and  Copyright  Foundation  of  George 
Washington  University  for  outstanding  work  in  the  field  of  patents,  trade- 
marks, and  related  areas. 


74  CARNEGIE     INSTITUTION     OF     WASHINGTON 

At  the  Mount  Wilson  and  Palomar  Observatories,  Ira  S.  Bowen,  the 
Director,  was  elected  a  member  of  the  Royal  Society  of  Sciences  of  Uppsala, 
Sweden.  The  Newcomb  Cleveland  prize  of  $1000  was  awarded  to  Halton  C. 
Arp,  Staff  Member,  on  December  29,  1961,  by  the  American  Association 
for  the  Advancement  of  Science  for  "a  noteworthy  paper,  representing  an 
outstanding  contribution  in  science.' '  Robert  P.  Kraft,  Staff  Member,  re- 
ceived the  Helen  B.  Warner  prize  of  the  American  Astronomical  Society  for 
outstanding  research  by  a  young  member  of  the  Society.  Guido  Munch  and 
Allan  R.  Sandage,  Staff  Members,  were  elected  fellows  of  the  American 
Academy  of  Arts  and  Sciences.  Fritz  Zwicky,  Staff  Member,  was  elected  a 
member  of  the  International  Academy  of  Astronautics.  This  organization, 
which  is  only  a  year  old,  is  the  first  international  academy  of  scientists  and 
engineers  who  have  made  contributions  to  space  technology.  It  is  limited  to 
165  active  members  in  the  life  sciences,  basic  sciences,  and  engineering. 

At  the  Geophysical  Laboratory,  Philip  H.  Abelson,  the  Director,  received 
on  April  6,  1962,  the  Washington  State  University  Regents'  Distinguished 
Alumnus  award  for  the  academic  year  1961-1962. 

Scott  E.  Forbush,  Staff  Member  of  the  Department  of  Terrestrial 
Magnetism,  was  elected  to  membership  in  the  National  Academy  of 
Sciences,  April  24,  1962,  and  on  June  14,  1962,  he  received  an  honorary 
doctor  of  science  degree  from  Case  Institute  of  Technology,  Cleveland, 
Ohio,  for  his  contributions  to  our  understanding  of  cosmic-ray  phenomena. 

At  the  Department  of  Plant  Biology,  Jens  Clausen,  retired  Staff  Member, 
was  made  a  Knight  of  the  Order  of  Dannebrog  by  the  King  of  Denmark  in 
recognition  of  his  contributions  to  botany  and  genetics.  The  Danish  consul 
presented  the  decoration  to  Dr.  Clausen  on  October  13,  1961,  in  San 
Francisco. 

M.  Demerec,  retired  Director  of  the  Department  of  Genetics,  was  awarded 
the  Kimber  Genetics  medal  of  the  National  Academy  of  Sciences  on  April 
24,  1962,  "in  recognition  of  his  many  contributions  to  the  understanding  of 
the  genetics  of  various  plants,  Drosophila,  bacteria,  and  viruses,  and 
especially  for  his  leadership  in  the  investigation  of  unstable  genes,  the 
mutation  process,  genetics  of  micro-organisms  and  the  genetic  fine  structure 
of  the  gene." 

J.  E.  S.  Thompson,  retired  Staff  Member  of  the  Department  of  Archae- 
ology, received  the  honorary  degree  of  doctor  of  humane  letters  and  the 
Drexel  Medal  for  Archaeology  from  the  University  of  Pennsylvania  in 
February  1962. 


Scientists  and  Scholars,  1902  - 1962 

In  essence,  the  whole  quality  of  the  Institution,  and  its  history,  lie  with  those  who 
have  been  associated  with  it  over  the  years.  Following  are  the  names  of  the  senior 
scientific  staff  members  of  all  departments  of  the  Institution  over  the  last  fifteen 
years  (or  over  the  last  fifteen  years  of  the  existence  of  terminated  departments). 
Below  them  in  each  department  are  listed  the  names  of  eminent  and  representative 
scientists  and  scholars  who  have  been  members  of  the  staff  or  otherwise  affiliated 
with  the  Institution  since  it  was  founded  in  1902.  A  list  is  also  given  of  all  Fellows  of 
the  Carnegie  Institution  of  Washington  since  the  beginning  of  its  Fellowship  Program 
in  1947,  and  a  list  of  grantees  and  others  affiliated  with  the  Institution  but  not  with 
any  particular  department. 


75 


76 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


DEPARTMENT   OF  PLANT   BIOLOGY 

Desert  Laboratory,  opened  in  1903,  became  headquarters  of  Department  of  Botanical  Research 
in  1905;  name  changed  to  Laboratory  for  Plant  Physiology  in  1923;  reorganized  in  1928  as  Division 
of  Plant  Biology,  including  ecology;  name  changed  to  Department  of  Plant  Biology  in  1951. 


Directors 

Daniel  T.  MacDougal,  1906-1927 

Herman  A.   Spoehr,    1927-September   1930,   September   1931-1947    {Chairman))   1947-1950 

{Chairman  Emeritus) 

C.  Stacy  French,  1947— 


Staff  Members 


Jeanette  S.  Brown,  1958 — 
Jens  C.  Clausen,  1931-1956 
David  C.  Fork,  1961 — 
Paul  Grun,  1949-1954 
William  M.  Hiesey,  1926— 
David  C.  Keck,  1928-1951 


Donald  W.  Kupke,  1955-1956 
Harold  W.  Milner,  1927— 
Malcolm  A.  Nobs,  1939-1941,  1951- 
James  H.  C.  Smith,  1925-1961 
Harold  H.  Strain,  1927-1962 
Ellen  C.  Weaver,  1961-1962 


Violet  (Koski)  Young,  1949-1953 


John  Belling,  1921-1933 
William  A.  Cannon,  1903-1924 
Frederic  E.  Clements,  1917-1941 
Waldo  S.  Glock,  1931-1938 
Harvey  M.  Hall,  1918-1932 


Garrett  J.  Hardin,  1942-1946 
Burton  E.  Livingston,  1906-1909 
Francis  E.  Lloyd,  1906 
Winston  M.  Manning,  1941-1946 
Forrest  Shreve,  1908-1945 


Godfrey  G.  Sykes,  1906-1929 


Other  Scientists  and  Scholars  Associated  with  the  Department 


Leroy  R.  Abrams,  1932 

(Stanford  University) 
Ernest  Anderson,  Research  Associate 

1932-1936  (University  of  Arizona) 
William  A.  Arnold,  Research  Associate 

1956-1961 

(Oak  Ridge  National  Laboratory) 
Eric  Ashby,  1930 

(Clare  College,  Cambridge  University) 
Daniel  I.  Axelrod,  1937,  1939,  1944, 1950,  1959 

(University  of  California) 
Ernest  B.  Babcock,  Research  Associate 

1926-1945  (University  of  California) 
Irving  W.  Bailey,  Research  Associate 

1928-1930,  1932-1939 

(Harvard  University) 
Charles  E.  Bessey,  1914 

(University  of  Nebraska) 


Nathaniel  L.  Britton,  Research  Associate 

1902,  1912-1916,  1918-1922 

(New  York  Botanical  Garden) 
Ursula  Brodfiihrer,  1956 

(University  of  Munich) 
Douglas  H.  Campbell,  1911 

(Stanford  University) 
Ralph  W.  Chaney,  Research  Associate 

1923-1956 

(University  of  California,  Berkeley) 
William  S.  Cooper,  1919-1925 

(University  of  Minnesota) 
Frederick  V.  Coville,  1902-1905 

(U.  S.  Department  of  Agriculture; 

later,  U.  S.  National  Museum) 
Pierre  Dansereau,  1949 

(University  of  Montreal;  later, 

New  York  Botanical  Garden) 


REPORT     OF    THE     PRESIDENT 


77 


John  P.  Decker,  1957 

(U.  S.  Forest  Service) 
Lee  R.  Dice,  Research  Associate 

1929-1930,  1932-1934-1938 

(University  of  Michigan) 
Erling  Dorf,  1930,  1936,  1942 

(Princeton  University) 
A.  E.  Douglass,  Research  Associate 

1924-1938  (University  of  Arizona) 
Newton  B.  Drury,  Research  Associate 

1937-1942 

(California  State  Parks  Commission) 
Benjamin  M.  Duggar,  Research  Associate 

1920-1921  (Missouri  Botanical  Garden; 

later,  University  of  Wisconsin) 
Friedrich  Ehrendorfer,  1951-1952 

(University  of  Vienna) 
Robert  Emerson,  Research  Associate 

1937-1941 

(California  Institute  of  Technology) 
G.  E.  Erdtmann,  1930 

(University  of  Stockholm) 
William  G.  Farlow,  1905 

(Harvard  University) 
Edward  E.  Free,  Research  Associate,  1920 

(U.  S.  Department  of  Agriculture) 
Martin  Gibbs,  1962  (Cornell  University) 
John  W.  E.  Glattfeld,  Research  Associate 

1920-1921  (University  of  Chicago) 
Richard  H.  Goodwin,  1950 

(Connecticut  College) 
Verne  E.  Grant,  1949-1950 

(Rancho  Santa  Ana  Botanic  Garden) 
Helen  M.  Habermann,  1959 

(Goucher  College) 
Per  Halldal,  1955-1957  (University  of  Oslo) 
Francis  T.  Haxo,  1957 

(Scripps  Institution  of  Oceanography) 
Robert  Hill,  1952  (Cambridge  University) 
A.  Stanley  Holt,  1959 

(National  Research  Council  of  Canada) 
Ellsworth  Huntington,  Research  Associate  in 

Geology,  1903-1904,  1910-1912,  1915-1917, 

1922-1923  (Yale  University) 
Donald  A.  Johansen,  1931-1932 

(private  research) 
Ivan  M.  Johnston,  1942  (Harvard  University) 
Erik  G.  J0rgensen,  1959 

(Royal  Danish  School  of  Pharmacy) 
Robert  W.  Krauss,  1951-1955 

(University  of  Maryland) 
Elias  Landolt,  1953-1955 

(Swiss  Federal  Institute  of  Technology) 


Charlton  M.  Lewis,  Research  Associate 

1938-1941  (Patent  Agent, 

Barkelow  and  Lewis,  Pasadena) 
Harlan  Lewis,  1954-1955 

(University  of  California,  Los  Angeles) 
Esmond  R.  Long,  1914-1915 

(University  of  Chicago;  later,  Henry  Phipps 

Institute,  University  of  Pennsylvania) 
John  M.  Macfarlane,  1902 

(University  of  Pennsylvania) 
Axel  Madsen,  1962  (Royal  Veterinary 

and  Agricultural  College,  Copenhagen) 
Herbert  L.  Mason,  1925 

(University  of  California) 
Max  Milner,  1957  (UN  Children's  Fund, 

Food  Conservation  Division) 
George  T.  Moore,  1914 

(Missouri  Botanical  Garden,  St.  Louis) 
Vladimir  Moravek,  Research  Associate,  1926 

(University  of  Brno,  Czechoslovakia) 
Jack  E.  Myers,  1950-1951,  1959 

(University  of  Texas) 
Hedda  Nordenshiold,  1949 

(Royal  Agricultural  College,  Uppsala) 
Axel  Nygren,  1950 

(Royal  Agricultural  College,  Uppsala) 
Winthrop  J.  V.  Osterhout,  Research  Associate 

1922-1924  (Harvard  University;  later, 

Rockefeller  Institute  for  Medical  Research) 
James  B.  Overton,  Research  Associate 

1903,  1926-1927  (University  of  Wisconsin) 
George  J.  Peirce,  1910-1912 

(Stanford  University) 
Gifford  Pinchot,  1902 

(U.  S.  Department  of  Agriculture;  later, 

Yale  University  and  Governor  of 

Pennsylvania) 
Thomas  R.  Pray,  1960-1961 

(University  of  Southern  California) 
Joseph  N.  Rose,  Research  Associate 

1908,  1910-1923  (U.  S.  National  Museum) 
Gilbert  M.  Smith,  Research  Associate 

1926-1927  (Stanford  University) 
Roger  Y.  Stanier,  1959 

(University  of  California,  Berkeley) 
G.  Ledyard  Stebbins,  Jr.,  1934-1936,  1945 

(University  of  California,  Davis) 
Bernard  Strehler,  1955  (National  Heart 

Institute,  Baltimore  City  Hospital) 
Walter  T.  Swingle,  1904 

(U.  S.  Department  of  Agriculture) 
Hiroshi  Tamiya,  1952-1953  (Tokugawa 

Institute  for  Biological  Research) 


78 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Edwin  W.  Tisdale,  1959  (University  of  Idaho) 
Sam  F.  Trelease,  1914  (Columbia  University) 
Vladimir  tJlehla,  Research  Associate,  1924 

(University  of  Brno,  Czechoslovakia) 
Cornelius  B.  Van  Niel,  1931-1932 

(Stanford  University) 
Chakrauarti  S.  Venkatesh,  1955-1956 

(Forest  Research  Institute,  India) 
Wolf  Vishniac,  1957  (Yale  University; 

later,  University  of  Rochester) 
Diter  von  Wettstein,  1959 

(University  of  Copenhagen) 


Heinrich  Walter,  1929 

(University  of  Stuttgart) 
John  E.  Weaver,  Research  Associate 

1922-1930  (University  of  Nebraska) 
George  R.  Wieland,  Research  Associate 

1903-1934,  1941  (Yale  University) 
Ira  L.  Wiggins,  Research  Associate 

1932-1933,  1936  (Stanford  University) 
Paul  C.  Wilbur,  1926-1927  (Food  Machinery 

and  Chemical  Corporation,  San  Jose) 
S.  W.  Williston,  1904  (University  of  Chicago) 
Frederick  T.  Wolf,  1960 

(Vanderbilt  University) 


MOUNT   WILSON  AND   PALOMAR   OBSERVATORIES 

Mount  Wilson  Observatory  organized  in  1904;  unified  operation  with  the  Palomar  Observatory 
of  the  California  Institute  of  Technology  began  in  1948. 


Directors 

George  E.  Hale,  1904-1923;  1923-1936  (Honorary) 

Walter  S.  Adams,  1924-1945 

Ira  S.  Bowen,  1946— 


Staff  Members 


Halton  C.  Arp,  1957— 
Walter  Baade,  1931-1958 
Harold  D.  Babcock,  1909-1948 
Horace  W.  Babcock,  1946— 
William  A.  Baum,  1950— 
Arthur  D.  Code,  1956-1958 
Armin  J.  Deutsch,  1951 — 
Olin  Eggen,  1961 — 
Jesse  L.  Greenstein,  1948 — 
Robert  F.  Howard,  1961— 
Fred  Hoyle,  1957-1962 
Edwin  P.  Hubble,  1919-1953 
Milton  L.  Humason,  1917-1957 
Alfred  H.  Joy,  1915-1948 
Robert  B.  King,  1938-1948 
Robert  P.  Kraft,  1960— 
Paul  W.  Merrill,  1919-1952 

J.  A.  Anderson,  1916-1943 
Theodore  Dunham,  Jr.,  1930-1947 
Arthur  S.  King,  1908-1943 

F.  G.  Pease,  1904-1938 

G.  W.  Ritchey,  1905-1919 
Charles  E.  St.  John,  1908-1930 


Rudolph  L.  Minkowski,  1937-1960 
Guido  Munch,  1951— 
Seth  B.  Nicholson,  1915-1957 
J.  Beverley  Oke,  1958— 
Donald  E.  Osterbrock,  1953-1958 
Edison  Pettit,  1920-1955 
Alexander  Pogo,  1950-1959 
Robert  S.  Richardson,  1931-1958 
Allan  R.  Sandage,  1952— 
Roscoe  F.  Sanford,  1918-1949 
Maarten  Schmidt,  1959 — 
Otto  Struve,  1962— 
Henrietta  H.  Swope,  1952 — 
Albert  G.  Wilson,  1948-1953 
Olin  C.  Wilson,  1931 — 
Ralph  E.  Wilson,  1938-1951 
Fritz  Zwicky,  1925— 

Frederick  H.  Seares,  1909-1940 
Harlow  Shapley,  1914-1921 
Sinclair  Smith,  1923-1938 
Gustaf  Stromberg,  1917-1946 
Adrian  van  Maanan,  1912-1946 


REPORT     OF    THE     PRESIDENT 


79 


Other  Scientists  and  Scholars  Associated  with  the  Department 


Charles  G.  Abbot,  1909-1948 

(Smithsonian  Institution) 
Giorgio  Abetti,  1909,  1930 

(Observatorio  di  Arcetri) 
Lawrence  Aller,  1946-1961 

(University  of  Indiana;  later,  University  of 

Michigan  and  University  of  California, 

Los  Angeles) 
Edward  E.  Barnard,  1904-1905,  1912 

(Yerkes  Observatory) 
W.  Becker,  1962  (University  of  Basel) 
Dirk  Brouwer,  Research  Associate 

1940-1944  (Yale  University) 
John  A.  Carroll,  1924  (Cambridge  University) 
William  de  Sitter,  1932 

(Observatory  of  Leiden) 
Albert  Einstein,  1933  (Preussische 

Akademie  der  Wissenschaft,  Berlin) 
E.  Freundlich,  1926 

(Astrophysical  Observatory,  Potsdam) 
Henry  G.  Gale,  Research  Associate 

1909-1911  (University  of  Chicago) 
Leo  Goldberg,  1940 

(Harvard  College  Observatory) 
Guillermo  Haro,  1958 

(Tonantzintla  Observatory) 
George  Herbig,  1948,  1950,  1954 

(Lick  Observatory) 
Ejnar  Hertzsprung,  1912 

(Potsdam  Observatory) 
Erik  Holmberg,  1940-1941,  1947,  1951,  1955 

(Lund  Observatory) 
Sir  James  Hop  wood  Jeans,  Research  Associate 

1922-1947  (Royal  Society  of  London) 
Jacobus  C.  Kapteyn,  Research  Associate 

1908-1922  (University  of  Groningen) 
Philip  C.  Keenan,  1953-1962 

(Perkins  Observatory) 
Gerard  P.  Kuiper,  1942,  1950,  1954 

(Yerkes  Observatory;  later, 

University  of  Arizona) 
Robert  B.  Leighton,  1951-1962 

(California  Institute  of  Technology) 
Abbe  Le  Maitre,  1933  (University  of  Louvain) 
A.  O.  Leuschner,  Research  Associate 

1906-1907,  1924  (University  of  California) 
Bertil  Lindblad,  1920-1921,  1950 

(Stockholm  Observatory) 
Knut  Lundmark,  1922-1923,  1930,  1933,  1938 

(University  of  Uppsala) 
W.  J.  Luyten,  1951,  1959 

(University  of  Minnesota) 


Robert  R.  McMath,  1950-1960 

(McMath-Hulbert  Observatory) 
N.  U.  Mayall,  1951  (Lick  Observatory;  later, 

Kitt  Peak  National  Observatory) 
A.  A.  Michelson,  Research  Associate 

1903-1904,  1919-1931 

(University  of  Chicago) 
Dayton  C.  Miller,  1921 

(Case  School  of  Applied  Science) 
S.  A.  Mitchell,  Research  Associate 

1924-1927, 1934-1944 

(University  of  Virginia) 
W.  W.  Morgan,  1957-1962 

(Yerkes  Observatory) 
Earnest  F.  Nichols,  Research  Associate 

1908-1909  (Dartmouth  College;  later, 

Yale  University  and  Massachusetts 

Institute  of  Technology) 
Y.  Ohman,  1934  (University  of  Uppsala) 
Jan  H.  Oort,  Research  Associate 

1924,  1939,  1952,  1958-1959,  1961 

(Leiden  Observatory) 
P.  Th.  Oosterhoff,  1934,  1960 

(Leiden  Observatory) 
L.  Perek,  1959  (Astronomical  Institute  of 

Czechoslovak  Academy  of  Sciences) 
L.  Plaut,  1956-1959 

(Kapteyn  Astronomical  Laboratory) 
Frank  E.  Ross,  1903-1909,  1927-1939 

(Yerkes  Observatory) 
S.  Rosseland,  192&-1927 

(International  Research  Fellow;  later, 

University  of  Oslo) 
Henry  N.  Russell,  Research  Associate 

1903-1905,  1921-1947 

(Princeton  University) 
Edwin  E.  Salpeter,  Research  Associate,  1959 

(Cornell  University) 
Jan  Schilt,  1925-1926  (International  Research 

Fellow;  later,  Yale  University  and 

Rutherfurd  Observatory,  Columbia 

University) 
Martin  Schwarzschild,  1946-1954 

(Princeton  University) 
C.  D.  Shane,  1929-1930 

(University  of  California) 
Frederick  Slocum,  1933-1934 

(Van  Vleck  Observatory) 
Lyman  Spitzer,  Jr.,  1937-1940,  1948-1959 

(Princeton  University) 
Joel  Stebbins,  Research  Associate 

1930,  1932-1948  (University  of  Wisconsin) 


80 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Carl  Stormer,  Research  Associate,  1912-1915 

(University  of  Christiania) 
Bengt  Stromgren,  1950,  1960 

(Institute  for  Advanced  Study) 
Pol  Swings,  Research  Associate 

1944-1946,  1958-1959  (University  of  Liege) 
A.  D.  Thackeray,  1935-1936  (Commonwealth 

Fellow;  later,  Radcliffe  Observatory) 
Albrecht  Unsold,  1929,  1957,  1961 

(University  of  Kiel) 


H.  C.  Van  de  Hulst,  1954 

(Leiden  Observatory) 
Albert  E.  Whitford,  1933-1957 

(University  of  Wisconsin;  later, 

Lick  Observatory) 
Rupert  Wildt,  1935-1936  (National  Research 

Fellow;  later,  Yale  University) 
R.  v.  d.  R.  Woolley,  Research  Associate 

1929-1931,  1958-1959,  1961 

(Royal  Greenwich  Observatory) 


DEPARTMENT   OF  TERRESTRIAL   MAGNETISM 

Organized  as  the  Department  of  International  Research  in  Terrestrial  Magnetism  on  April  1,  1904. 
Name  changed  to  Department  of  Terrestrial  Magnetism  in  1905. 


Directors 

Louis  A.  Bauer,  1904-1929 

John  A.  Fleming,  1929-1934  (Acting);  1935-1946 

Merle  A.  Tuve,  1946— 


Staff  Members 


Philip  H.  Abelson,  1939-1953 
L.  T.  Aldrich,  1950— 
Lloyd  V.  Berkner,  1933-1951 
Ellis  T.  Bolton,  1951— 
Roy  J.  Britten,  1951 — 
Bernard  F.  Burke,  1953— 
Dean  B.  Cowie,  1944— 
John  W.  Firor,  Jr.,  1953-1961 
Scott  E.  Forbush,  1927— 
W.  Kent  Ford,  Jr.,  1957— 
Oliver  H.  Gish,  1922-1948 
John  W.  Graham,  1951-1958 
Stanley  R.  Hart,  1961 — 
Norman  P.  Heydenburg,  1935- 
Ellis  A.  Johnson,  1935-1956 


Brian  J.  McCarthy,  1960— 
W.  C.  Parkinson,  1913-1950 
Richard  B.  Roberts,  1937 — 
William  J.  Rooney,  1924-1949 
T.  Jefferson  Smith,  1962— 
John  S.  Steinhart,  1961 — 
Howard  E.  Tatel,  1947-1957 
Georges  M.  Temmer,  1953 — 
George  R.  Tilton,  1951-1956 
Oscar  W.  Torreson,  1923-1952 
Ernest  H.  Vestine,  1938-1957 
George  R.  Wait,  1920-1951 
Harry  W.  Wells,  1932— 
George  W.  Wetherill,  1954-1960 


J.  P.  Ault,  1905-1929 
S.  J.  Barnett,  1917-1926 
E.  H.  Bramhall,  1941-1944 
G.  Breit,  1924-1929 
O.  Dahl,  1926-1936 


F.  T.  Davies,  1929-1939 
H.  M.  W.  Edmonds,  1906-1930 
H.  W.  Fisk,  1905-1932 
Lawrence  R.  Hafstad,  1928-1946 
James  A.  Van  Allen,  1939-1941 


REPORT     OF     THE     PRESIDENT 


81 


Other  Scientists  and  Scholars  Associated  with  the  Department 


Samuel  J.  Mcintosh  Allen,  Research  Associate 

1924-1928  (University  of  Cincinnati) 
E.  Amaldi,  1936-1937 

(Royal  University  of  Rome) 
J.  Bartels,  Research  Associate,  1930-1940 

(Fortsliche  Hochschule,  Eberswalde, 

Germany;  later,  Geophysikalisches  Institut, 

Gottingen,  Germany) 
Carl  Barus,  Research  Associate 

1902,  1904-1923,  1926  (Brown  University) 
Jesse  W.  Beams,  1934-1935 

(University  of  Virginia) 
J.  C.  Beattie,  Research  Associate,  1908-1911 

(South  Africa  College,  Cape  Town) 
Ralph  D.  Bennett,  Research  Associate 

1933-1940  (Massachusetts  Institute  of 

Technology;  later,  Vallecitos  Atomic 

Laboratory,  Pleasanton,  California) 
Hans  A.  Bethe,  1936-1941 

(Cornell  University) 
Henry  G.  Booker,  Research  Associate 

1938-1940  (Cornell  University) 
Edward  L.  Bowles,  Research  Associate 

1939-1945  (Massachusetts  Institute  of 

Technology) 
Joseph  C.  Boyce,  Research  Associate, 

1939-1950  (Massachusetts  Institute  of 

Technology;  later,  New  York  University 

and  Illinois  Institute  of  Technology) 
Robert  B.  Brode,  Research  Associate 

1939-1941  (University  of  California, 

Berkeley) 
Richard  E.  Byrd,  1931-1932 

(U.  S.  Navy,  Arctic  explorer) 
Sydney  Chapman,  Research  Associate 

1934-1940,  1951-1953  (Trinity  College, 

Cambridge;  later,  Imperial  College, 

London;  High  Altitude  Observatory, 

Boulder,  Colorado,  and  Geophysical 

Institute,  College,  Alaska) 
Georges  N.  Cohen,  Research  Associate 

1956-1959  (Institut  Pasteur,  Paris) 
Arthur  H.  Compton,  Research  Associate 

1931-1945  (University  of  Chicago;  later, 

Washington  University) 
Karl  T.  Compton,  Research  Associate 

1928-1934  (Princeton  University;  later, 

Massachusetts  Institute  of  Technology) 
T.  G.  Cowling,  1950-1951 

(Princeton  L^niversity) 


Hugh  H.  Darby,  Research  Associate 

1948-1950  (consultant  in  biochemistry, 

Mt.  Airy,  Maryland) 
N.  Ernest  Dorsey,  Research  Associate 

1912-1913  (National  Bureau  of  Standards) 
George  Gamow,  Research  Associate,  1935-1944 

(George  Washington  University;  later, 

University  of  Colorado) 
Enrique  Gaviola,  Research  Associate 

1928-1929  (Comision  de  Astrofisica  y 

Radioastronomia,  Universidad  de  Buenos 

Aires) 
Ross  Gunn,  Research  Associate,  1938-1944 

(Naval  Research  Laboratory;  later, 

U.  S.  Weather  Bureau  and  American 

University) 
Anton  L.  Hales,  Research  Associate,  1960 

(University  of  the  Witwatersrand ;  later, 

Graduate  Research  Center,  Inc.,  Dallas, 

Texas) 
John  S.  Hall,  1954— 

(Lowell  Observatory,  Flagstaff,  Arizona) 
Raymond  G.  Herb,  1935 

(University  of  Wisconsin) 
Victor  F.  Hess,  Research  Associate,  1940-1946 

(Fordham  University) 
Thomas  H.  Johnson,  Research  Associate 

1933-1946  (Bartol  Research  Foundation; 

later,  Brookhaven  National  Laboratory, 

Atomic  Energy  Commission,  and  Raytheon 

Manufacturing  Company) 
Arthur  E.  Kennelly,  Research  Associate, 

1924-1935  (Harvard  University  and 

Massachusetts  Institute  of  Technology) 
Serge  A.  Korff,  Research  Associate,  1936-1945 

(New  York  University) 
D.  la  Cour,  1931-1932  (Danish 

Meteorological  Survey,  Copenhagen) 
H.  A.  Lorentz,  Research  Associate,  1920 

(University  of  Leiden) 
Frank  T.  McClure,  Research  Associate 

1955-1960  (Applied  Physics  Laboratory, 

Johns  Hopkins  University) 
J.  D.  McGee,  Research  Associate,  1959-1960 

(Imperial  College) 
Kenneth  R.  McQuillen,  Research  Associate 

1951-1960  (University  of  Cambridge) 
Robert  A.  Millikan,  Research  Associate 

1921-1945  (California  Institute  of 

Technology) 


82 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


B.  Y.  Mills,  1953-1954 

(Radiophysics  Laboratory,  Commonwealth 

Industrial  and  Research  Organization, 

Sydney,  Australia) 
S.  K.  Mitra,  1936-1937 

(University  of  Calcutta) 
T.  Nagata,  1950-1951 

(Geophysical  Institute,  Tokyo) 
J.  L.  Pawsey,  1957-1958  (Commonwealth 

Industrial  and  Research  Organization, 

Australia) 
Greenleaf  W.  Pickard,  Research  Associate 

1927-1935  (consultant  in  electrical 

engineering,  Newton  Centre,  Massachusetts) 
Wilson  M.  Powell,  Research  Associate 

1942-1943  (Lawrence  Radiation 

Laboratory,  University  of  California) 
A.  T.  Price,  1952 

(The  Royal  Technical  College,  Glasgow) 
Norman  Ramsey,  1938-1939 

(Harvard  University) 
J.  A.  Ratcliffe,  1950-1951 

(Cavendish  Laboratory,  Cambridge) 
Bruno  Rossi,  1932-1933 

(Massachusetts  Institute  of  Technology) 
Sir  Arthur  Rucker,  1904-1915 

(Royal  College  of  Science, 

South  Kensington,  London) 


M.  N.  Saha,  1936-1937 

(Allahabad  University,  India) 
Marcel  Schein,  1939-1944  (University  of 

Chicago;  later,  University  of  California) 

A.  Schmidt,  1905-1907 
(Potsdam  Magnetic  Observatory) 

B.  F.  J.  Schonland,  1952 
(University  of  the  Witwatersrand) 

Frederick  Slocum,  Research  Associate,  1920 

(Brown  University;  later,  Wesleyan 

University) 
F.  Graham  Smith,  1952-1954 

(Cavendish  Laboratory,  Cambridge) 
J.  C.  Street,  Research  Associate,  1933-1934 

(Harvard  University) 
H.  U.  Sverdrup,  Research  Associate,  1926-1939 

(Geophysical  Institute,  Bergen) 
W.  F.  G.  Swann,  1916-1920 

(Bartol  Research  Foundation) 
John  T.  Tate,  Research  Associate,  1941-1945 

(University  of  Minnesota) 
Edward  Teller,  1936-1937  (University  of 

Chicago;  later,  Lawrence  Radiation 

Laboratory,  University  of  California) 
Manuel  S.  Vallarta,  Research  Associate 

1940-1941,  1948-1950 

(National  University  of  Mexico) 
John  von  Neumann,  1948-1949,  1955-1957 

(Institute  for  Advanced  Study,  Princeton) 


DEPARTMENT   OF   GENETICS 

Station  for  Experimental  Evolution  opened  in  1904;  name  changed  to  Department  of  Experimental 
Evolution  in  1906;  combined  with  Eugenics  Record  Office  in  1921  to  form  Department  of  Genetics. 

Directors 

Charles  B.  Davenport,  1904-1934 

Albert  F.  Blakeslee,  1935-November  30,  1941 

Milislav  Demerec,  December  1,  1941-1942  (Acting);  1943- June  30,  1960 

Berwind  P.  Kaufmann,  July  1,  1960- January  31, 1962  (Acting);  February  1,  1962- June  30, 1962 


Staff  Members 

Ernst  Caspari,  1947-1949  Margaret  R.  McDonald,  1946- 

Helen  Gay,  1960-1962  E.  C.  MacDowell,  1914-1952 

Alfred  D.  Hershey,  1950—  George  Streisinger,  1956-1960 

Barbara  McClintock,  1942—  Bruce  Wallace,  1947-1949 

Evelyn  M.  Witkin,  1950-1955 


Arthur  M.  Banta,  1909-1930 
Robert  W.  Bates,  1931-1941 


Barbara  S.  Burks,  1936-1941 
John  Belling,  1920-1929 


REPORT     OF    THE     PRESIDENT 


88 


Ugo  Fano,  1940-1946 
Ross  A.  Gortner,  1909-1914 
J.  Arthur  Harris,  1907-1924 
H.  H.  Laughlin,  1918-1940 
S.  E.  Luria,  1945-1946 
Frank  E.  Lutz,  1904-1909 


Charles  W.  Metz,  1914-1930 
Oscar  Riddle,  1912-1945 
Sophie  Satina,  1924-1942 
George  H.  Shull,  1904-1915 
Morris  Steggerda,  1930-1944 
H.  E.  Warmke,  1938-1945 


Benjamin  B.  Wells,  1941-1942 


Other  Scientists  and  Scholars  Associated  with  the  Department 


Edgar  Anderson,  1941,  1945 

(Washington  University;  later, 

Missouri  Botanical  Garden) 
Ernest  Ball,  1942-1943 

(University  of  North  Carolina) 
Hans  Bauer,   1936  (Kaiser- Wilhelm  Institut 

fur  Biologie,  Berlin-Dahlem;  later, 

Max-Planck  Institut  fur  Meeresbiologie, 

Wilhelmshaven,  Germany) 
George  W.  Beadle,  1935 

(California  Institute  of  Technology; 

later,  University  of  Chicago) 
John  J.  Biesele,  Research  Associate,  1944-1946 

(University  of  Texas) 
Dietrich  Bodenstein,  1944 

(University  of  Virginia) 
Sydney  Brenner,  1954 

(University  of  the  Witwatersrand;  later, 

Cavendish  Laboratory) 
Vernon  Bryson,  Research  Associate,  1942-1943 

(Rutgers  University) 
John  T.  Buchholz,  1921-1941  (University  of 

Arkansas;  later,  University  of  Illinois) 
Sir  Macfarlane  Burnet,  1950 

(The  Walter  and  Eliza  Hall  Institute  of 

Medical  Research,  Melbourne) 
J.  Gordon  Carlson,  1937-1940 

(University  of  Alabama;  later, 

University  of  Tennessee) 
J.  Lincoln  Cartledge,  1921-1924 

(University  of  West  Virginia) 
William  E.  Castle,  Research  Associate 

1904-1943  (Harvard  University;  later, 

University  of  California) 
David  G.  Catcheside,  Research  Associate 

1957-1959  (University  of  Birmingham, 

England) 
Donald  R.  Charles,  1929-1930 

(Sarah  Lawrence  College;  later, 

University  of  Rochester) 
Albert  Claude,  1946 

(Rockefeller  Institute  for  Medical  Research; 

later,  The  Free  University  of  Brussels) 


Marie  E.  Conklin,  1937-1941 

(Adelphi  College) 
J.  N.  Couch,  1925-1926 
M  (University  of  North  Carolina) 

ax  Delbnick,  1937 

(California  Institute  of  Technology) 
Hugo  de  Vries,  Research  Associate,  1904-1918 

(University  of  Amsterdam) 
Th.  Dobzhansky,  Research  Associate 

1936-1949  (California  Institute  of 

Technology  and  Columbia  University; 

later,  Rockefeller  Institute  for  Medical 

Research) 
L.  C.  Dunn,  1929  (Columbia  University) 
Boris  Ephrussi,  1937  (Institut  de  Biologie 

Physico-Chimique,  Paris;  later,  Centre 

National  de  la  Recherche  Scientifique, 

Gif-sur-Yvette) 
Harold  D.  Fish,  Research  Associate,  1919-1924 

(Denison  University;  later,  University  of 

Pittsburgh) 
Pierre  Fredericq,  1958  (University  of  Liege) 
Gabriel  Gasid,  1945-1946  (University  of  Chile) 
Norman  H.  Giles,  Jr.,  1941 

(Yale  University) 
Joseph  S.  Gots,  1954,  1955,  1957 

(University  of  Pennsylvania) 
John  W.  Gowen,  1915,  1940 

(Iowa  State  College) 
Ludwig  von  Graf,  1906  (University  of  Graz) 
C.  C.  Guthrie,  1909  (University  of  Pittsburgh) 
Ike  Gustafsson,  1937-1938  (Institute  for 

Genetic  Research,  Svalof,  Sweden;  later, 

Forest  Research  Institute  of  Sweden, 

Stockholm) 
Olli  Halkka,  1959  (University  of  Helsinki) 
Alexander  Hollaender,  Research  Associate 

1942-1944  (National  Institutes  of  Health; 

later,  Oak  Ridge  National  Laboratory) 
Sally  Hughes-Schrader,  1934 

(Sarah  Lawrence  College;  later, 

Duke  University) 


84 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


C.  Leonard  Huskins,  1935,  1936 

(McGill  University;  later, 

University  of  Wisconsin) 
Fritz  Kaudewitz,  1954  (Max-Planck  Institut 

fur  Virusforschung,  Tubingen) 
Tage  Kemp,  1932  (University  of  Copenhagen) 
P.  C.  Koller,  1938  (University  of  Edinburgh; 

later,   Chester  Beatty  Research  Institute, 

London) 
Jaroslav  Krizenecky,  1928-1929 

(Zootechnical  Research  Institute,  Brno, 

Czechoslovakia) 
Victor  K.  LaMer,  1916-1917 

(Columbia  University) 
Raymond  Latarjet,  1945-1946 

(Institut  Pasteur,  Paris) 
Albert  Levan,  1951  (University  of  Lund) 
Cyrus  Levinthal,  1951 

(University  of  Michigan;  later, 

Massachusetts  Institute  of  Technology) 
C.  C.  Little,  Research  Associate,  1919-1925 

(Roscoe  B.  Jackson  Memorial  Laboratory, 

Bar  Harbor,  Maine) 
Edward  L.  Mark,  Research  Associate 

1904-1910  (Harvard  University) 
Horace  N.  Marvin,  1941-1942 

(University  of  Arkansas) 
William  J.  Moenkhaus,  1904-1906 

(Indiana  University) 
G.  Montalenti,  1951  (University  of  Naples; 

later,  University  of  Rome) 
H.  J.  Muller,  1921  (Indiana  University) 
Robert  K.  Nabours,  Research  Associate 

1929-1930  (Kansas  State  College) 
James  V.  Neel,  1940  (Dartmouth  College; 

later,  University  of  Michigan) 
Howard  B.  Newcombe,  Research  Associate 

1938,  1945-1947  (Atomic  Energy 

Commission  of  Canada,  Ltd.) 
Theophilus  S.  Painter,  1923 

(University  of  Texas) 
Raymond  Pearl,  Research  Associate 

1904-1906  (University  of  Michigan;  later, 

Johns  Hopkins  University) 
Marcus  M.  Rhoades,  1941  (Columbia 

University;  later,  Indiana  University) 
Maurice  N.  Richter,  1930-1952 

(Columbia  University;  later, 

New  York  University  Medical  Center) 
Franz  Schrader,  1934  (Columbia  University; 

later,  Duke  University) 
Edmund  W.  Sinnott,  1938  (Columbia 

University;  later,  Yale  University) 
B.  M.  Slizynski,  1936-1937  (University  of 

Cracow;  later,  University  of  Edinburgh) 


Evelyn  E.  B.  Smith,  Research  Associate,  1958 

(University  of  Glasgow) 
Laurence  H.  Snyder,  1922 

(University  of  Hawaii) 
Arnold  H.  Sparrow,  1940,  1941 

(Brookhaven  National  Laboratory) 
Warren  P.  Spencer,  1935  (College  of  Wooster) 
S.  G.  Stephens,  Research  Associate,  1945-1947 

(North  Carolina  State  College) 
Curt  Stern,  1933,  1938,  1944,  1946 

(University  of  Rochester;  later, 

University  of  California,  Berkeley) 
A.  Tavcar,  1951  (University  of  Zabreb) 
Howard  J.  Teas,  1942-1943 

(Nuclear  Center,  Mayagiiez,  Puerto  Rico) 
Rene  Thomas,  1957-1958 

(University  of  Brussels) 
N.  TimofeerT-Ressovsky,  1932 

(Kaiser- Wilhelm  Institut  fur  Hirnforschung; 

later,   Academy  of  Sciences,   Novosibirsk, 

U.S.S.R.) 
Jun-ichi  Tomizawa,  Research  Associate 

1957-1959  (National  Institute  of  Health, 

Tokyo) 
William  L.  Tower,  Research  Associate 

1904-1917  (University  of  Chicago) 
J.  van  Overbeek,  1940,  1941 

(California  Institute  of  Technology;  later, 

Agricultural  Laboratory,  Shell 

Development  Company,  Modesto, 

California) 
C.  H.  Waddington,  1938  (Cambridge 

University;  later,  University  of  Edinburgh) 
Mogens  Westergaard,  Research  Associate 

1957-1959  (Universitetets  Genetiske 

Institut,  Copenhagen) 
Fritz  von  Wettstein,  1938 

(Kaiser- Wilhelm  Institut  fur  Biologie, 

Berlin-Dahlem;  later,  University  of  Vienna) 
M.  J.  D.  White,  1947,  1950-1952 

(University  of  London  and  University  of 

Texas;  later,  Commonwealth  Scientific  and 

Industrial  Research  Organization, 

Canberra,  Australia) 
P.  W.  Whiting,  1933-1935 

(University  of  Pennsylvania) 
Maurice  Whittinghill,  1938 

(Bennington  College;  later, 

University  of  North  Carolina) 
Edmund  B.  Wilson,  Research  Associate 

1904-1909,  1936-1938 

(Columbia  University) 
Charles  Yanofsky,  1956  (Western  Reserve 

University;  later,  Stanford  LTniversity) 


REPORT     OF     THE     PRESIDENT 


85 


The  following  persons  carried  on  genetic 
studies  with  Carnegie  Institution  support: 

Calvin  B.  Bridges,  Research  Associate 
1916-1938  (Columbia  University;  later, 
California  Institute  of  Technology) 

Thomas  Hunt  Morgan,  Research  Associate 


1916-1945  (Columbia  University;  later, 
California  Institute  of  Technology) 

Jack  Schultz,  Research  Associate,  1929-1941 
(California  Institute  of  Technology;  later, 
Institute  for  Cancer  Research,  Philadelphia) 

A.  H.  Sturtevant,  Research  Associate 
1916-1931  (Columbia  University;  later, 
California  Institute  of  Technology) 


GEOPHYSICAL  LABORATORY 


Organized  in  1906.  Opened  in  1907. 


Directors 


Arthur  L.  Day,  1909-1936 
Leason  H.  Adams,  1936-1937  (Acting);  1938- July  31,  1952 
George  W.  Morey,  August  1,  1952-August  31,  1953  (Acting) 
Philip  H.  Abelson,  September  1,  1953 — 


Staff  Members 


Norman  L.  Bowen,  1910-1937,  1947-1953 

Francis  R.  Boyd,  Jr.,  1953— 

John  S.  Burlew,  1936-1952 

Felix  Chayes,  1947— 

Sydney  P.  Clark,  Jr.,  1957-1962 

Gordon  L.  Davis,  1941 — 

Gabrielle  Donnay,  1955 — 

Joseph  L.  England,  1926 — ■ 

Hans  P.  Eugster,  1952-1958 

Roy  W.  Goranson,  1926-1951 

Hugh  J.  Greenwood,  1960 — 

Joseph  W.  Greig,  1922-1960 

Thomas  C.  Hoering,  1959 — 

Earl  Ingerson,  1935-1947 

Eugene  T.  Allen,  1907-1933 
Olaf  Andersen,  1912-1918 
Tom.  F.  W.  Barth,  1929-1940 
F.  Russell  von  Bichowsky,  1916-1919 
A.  F.  Buddington,  1919-1920 
J.  K.  Clement,  1904-1907 
Pentti  Eskola,  1921-1922 
Clarence  N.  Fenner,  1910-1937 
Michael  Fleischer,  1936-1938 
Ralph  E.  Gibson,  1924-1946 
Sterling  B.  Hendricks,  1926 
James  H.  Hibben,  1928-1939 
John  C.  Hostetter,  1912-1919 


Frank  C.  Kracek,  1923-1956 
Gunnar  Kullerud,  1954 — 
William  S.  MacKenzie,  1951-1957 
Patrick  L.  Parker,  1961 — 
Charles  S.  Piggot,  1925-1947 
Eugene  Posnjak,  1913-1947 
Howard  S.  Roberts,  1917-1947 
J.  Frank  Schairer,  1927— 
George  R.  Tilton,  1956— 
George  Tunell,  1925-1947 
O.  Frank  Tuttle,  1947-1953 
William  D.  Urry,  1938-1949 
Hatten  S.  Yoder,  Jr.,  1948— 
Emanuel  G.  Zies,  1913-1949 

John  Johnston,  1908-1916 
Esper  S.  Larsen,  Jr.,  1907-1909 
Robert  H.  Lombard,  1915-1927 
Herbert  E.  Merwin,  1909-1959 
Elbert  F.  Osborn,  1938-1945 
George  A.  Rankin,  1907-1916 
Earnest  S.  Shepherd,  1904-1946 
Robert  B.  Sosman,  1908-1928 
Henry  S.  Washington,  1912-1934 
Walter  P.  White,  1904-1935 
Erskine  D.  Williamson,  1914-1923 
Fred  E.  Wright,  1906-1944 
Ralph  W.  G.  Wyckoff,  1919-1927 


86 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Other  Scientists  and  Scholars  Associated  with  the  Department 


Frank  D.  Adams,  1903-1911 

(McGill  University) 
Samuel  K.  Allison,  1925-1926 

(University  of  Chicago) 
George  F.  Becker,  1903-1909 

(U.  S.  Geological  Survey) 
J.  C.  Branner,  1906-1907 

(Arkansas  Geological  Survey) 
L.  E.  J.  Brouwer,  1931-1932 

(Royal  Dutch  Petroleum) 
Thomas  C.  Chamberlin,  Research  Associate 

1903-1927  (University  of  Chicago) 
Hessel  de  Vries,  Research  Associate,  1958-1959 

(University  of  Groningen,  the  Netherlands) 
J.  D.  H.  Donnay,  1953— 

(Johns  Hopkins  University) 
William  H.  Emmons,  1903 

(University  of  Chicago) 
Henry  Faul,  1956-1957 

(U.  S.  Geological  Survey) 
Grove  Karl  Gilbert,  1904,  1906-1907 

(U.  S.  Geological  Survey) 
Harry  H.  Hess,  Research  Associate,  1940-1942 

(Princeton  University) 
Joseph  P.  Iddings,  1905 

(University  of  Chicago) 
Emilie  Jager,  1958-1959 

(University  of  Bern,  Switzerland) 
Willard  F.  Libby,  Research  Associate 

1954-1959  (University  of  California  at 

Los  Angeles) 


George  D.  Louderback,  1903-1906 

(University  of  Nevada) 
Gordon  J.  F.  MacDonald,  1955— 

(Institute  of  Geophysics  and  Planetary 

Physics,  University  of  California  at  Los 

Angeles) 
Forest  Ray  Moulton,  Research  Associate 

1903-1922  (University  of  Chicago) 
Paul  Niggli,  1913-1914  (Zurich) 
C.  C.  Patterson,  1958 

(California  Institute  of  Technology) 
Frank  A.  Perret,  Research  Associate 

1938-1943  (Volcanological  Museum, 

St.  Pierre,  Martinique) 
Hans  Ramberg,  Research  Associate,  1955-1958 

(University  of  Chicago) 
Paul  Ramdohr,  Research  Associate,  1960 — 

(University  of  Heidelberg) 
C.  S.  Slichter,  1903,  1906 

(University  of  Wisconsin) 
David  B.  Stewart,  1954-1956 

(U.  S.  Geological  Survey) 
C.  E.  Tilley,  Research  Associate,  1955 — 

(Cambridge  University) 
Johan  August  Udden,  1925,  1928 

(University  of  Texas) 
C.  R.  Van  Hise,  1902-1903 

(University  of  Wisconsin) 
C.  E.  Van  Orstrand,  1904,  1906-1910 

(U.  S.  Geological  Survey) 


Organized  1914. 


DEPARTMENT   OF   EMBRYOLOGY 

Directors 

Franklin  P.  Mall,  1914-1917 

George  L.  Streeter,  1918-1940 

George  W.  Corner,  1941-1955 

James  D.  Ebert,  1956— 


Staff  Members 


George  W.  Bartelmez,  1949-1960 
David  W.  Bishop,  1952— 
Bent  G.  Boving,  1951 — 
Donald  D.  Brown,  1962— 


Robert  K.  Burns,  1940-1962 

Arpad  Csapo,  1951-1955 

Robert  L.  DeHaan,  1956 — 

James  F.  Didusch,  1913-1940,  1945-1955 


REPORT     OF     THE     PRESIDENT 


87 


Louis  B.  Flexner,  1940-1951 
Osborne  0.  Heard,  1913-1956 
Chester  H.  Heuser,  1921-1950 
I.  R.  Konigsberg,  1961 — 
Elizabeth  M.  Ramsey,  1949— 

Herbert  M.  Evans,  1913-1915 
Carl  G.  Hartman,  1925-1941 
Margaret  R.  Lewis,  1915-1946 


Mary  E.  Rawles,  1957— 
Samuel  R.  M.  Reynolds,  1941-1955 
Royal  F.  Ruth,  1956-1961 
David  B.  Tyler,  1947-1950 
Walter  S.  Wilde,  1944-1947 

Warren  H.  Lewis,  1914-1940 
Charles  W.  Metz,  1930-1940 
Adolph  H.  Schultz,  1916-1925 


Other  Scientists  and  Scholars  Associated  with  the  Department 


William  E.  Adams,  1957-1958 

(University  of  Otago,  New  Zealand) 
Ines  de  Allende,  1941-1943 

(University  of  Cordoba) 
Howard  D.  Andervont,  1923-1926 

(Johns  Hopkins  School  of  Hygiene  and 

Public  Health;  later,  National  Cancer 

Institute) 
T.  S.  and  B.  F.  Argyris,  1961-1962 

(Syracuse  University) 
Alexander  Barry,  1947 

(University  of  Michigan) 
T.  H.  Bast,  1929-1930 

(University  of  Wisconsin) 
J.  D.  Boyd,  1934-1935 

(Cambridge  University) 
E.  A.  Boyden,  1939-1940  (University  of 

Minnesota;  later,  University  of  Washington) 
Washington  Buno,  1945-1946 

(University  of  Montevideo) 
Gerald  L.  Carlson,  1960-1962 

(Massachusetts  Institute  of  Technology) 
Eliot  R.  Clark  and  Eleanor  L.  Clark, 

1907-1914  (Johns  Hopkins  School  of 

Medicine;  later,  University  of 

Pennsylvania) 
George  W.  Corner,  Jr.,  1943 — 

(Johns  Hopkins  School  of  Medicine) 

E.  V.  Cowdry,  1913-1916 

(Johns  Hopkins  School  of  Medicine; 
later,  Washington  University) 
Maria  Victoria  de  la  Cruz,  1949-1950 
(Institute  of  Cardiology,  Mexico  City) 

F.  Cuajunco,  1927-1928 
(University  of  the  Philippines) 

Harold  Cummins,  1927-1928 

(Tulane  University) 
Vera  Danchakoff,  1924 

(Columbia  University) 
Carl  Lawrence  Davis,  1920 

(University  of  Maryland) 
Vincent  J.  De  Feo,  1955-1957 

(University  of  Illinois) 


Anatole  S.  Dekaban,  1959 — 

(National  Institutes  of  Health) 
Charles  A.  Doan,  1923-1924 

(Johns  Hopkins  School  of  Medicine; 

later,  Ohio  State  University) 
Jules  Duesberg,  Research  Associate,  1915-1918 

(University  of  Liege) 
Robert  K.  Enders,  1930-1932 

(Swarthmore  College) 
Thomas  R.  Forbes,  1937-1938 

(Johns  Hopkins  School  of  Medicine; 

later,  Yale  University) 
Fritz  Fuchs,  1950-1951 

(University  of  Copenhagen) 
Ernest  D.  Gardner,  1955 

(Wayne  State  University) 
E.  M.  K.  Geiling,  1935-1936 

(Johns  Hopkins  School  of  Medicine; 

later,  University  of  Chicago  and 

U.  S.  Food  and  Drug  Administration) 
Isidore  Gersh,  1934-1935 

(Johns  Hopkins  School  of  Medicine; 

later,  University  of  Chicago) 
G.  O.  Gey,  1924-1930 

(Johns  Hopkins  School  of  Medicine) 
Joseph  Gillman,  Research  Associate 

1941-1942,  1946-1948 

(University  of  the  Witwatersrand) 
G.  Gitlin,  1950 

(Hebrew  University,  Jerusalem) 
Timothy  Glover,  1961 

(University  of  Liverpool) 
Charles  M.  Goss,  1948  (Louisiana  State 

University  School  of  Medicine) 
Donald  J.  Gray,  1955  (Stanford  University) 
Gilbert  S.  Greenwald,  1954-1956 

(University  of  Washington;  later, 

University  of  Kansas) 
Paul  W.  Gregory,  1928-1929 

(Harvard  University;  later, 

University  of  California,  Davis) 
E.  Grodzinski,  1928-1929 

(University  of  Cracow) 


88 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Alan  F.  Guttmacher,  1921-1922 

(Johns  Hopkins  School  of  Medicine;  later, 

Mt.  Sinai  Hospital,  New  York  City) 
Manfred  S.  Guttmacher,  1921-1922 

(Johns  Hopkins  School  of  Medicine; 

later,  private  practice  in  psychiatry, 

Baltimore) 
John  W.  S.  Harris,  1961 

(London  Hospital  Medical  College) 
Arthur  T.  Hertig,  Research  Associate 

1933-1956  (Harvard  Medical  School) 
Marion  Hines,  1925-1947 

(University  of  Chicago  and  Johns  Hopkins 

University;  later,  Emory  University) 
A.  St.  G.  Huggett,  1952-1953 

(St.  Mary's  Hospital  Medical  School, 

London) 
Irwin  H.  Kaiser,  1946-1947 

(University  of  Minnesota) 
Seymour  Katsh,  1955-1958 

(University  of  Colorado) 
Franz  Keibel,  Research  Associate,  1914-1918 

(Anatomical  Institute,  Strassburg) 
Benjamin  F.  Kingsbury,  1917-1918 

(Cornell  University) 
Abraham  Kulangara,  1959-1961 

(University  of  California,  Los  Angeles; 

later,  All  India  Institute  of  Medical 

Sciences,  New  Delhi) 
Orthello  R.  Langworthy,  1924-1930 

(Johns  Hopkins  University) 
Hans  Laufer,  1957-1959 

(Johns  Hopkins  University) 
John  McKenzie,  1959 

(University  of  Aberdeen,  Scotland) 
Joseph  E.  Markee,  1935-1936 

(Stanford  University;  later, 

Duke  University) 
Arthur  Meyer,  1917-1918 

(Stanford  University) 
Tom  Mori,  1960-1961  (Tohoku  University) 
Harland  W.  Mossman,  1934-1935 

(University  of  Wisconsin) 
William  B.  Muchmore,  1959 

(University  of  Rochester) 
Jacques  Mulnard,  1957 

(University  of  Brussels) 
Alton  M.  Mun,  1959-1961 

(Washington  State  College;  later, 

University  of  Maine) 
G.  Muratori,  1934-1935 

(University  of  Padua,  Italy) 
Roberto  Narbaitz,  1959 

(University  of  Buenos  Aires) 


Catherine  Neill,  1952-1953  (London;  later, 

Johns  Hopkins  Medical  School) 
Martin  Nordmann,  1929-1930 

(University  of  Tubingen) 
Ronan  O'Rahilly,  1961-1962 

(St.  Louis  University) 
F.  Orts  Llorca,  1959  (University  of  Madrid) 
John  Papaconstantinou,  1958-1960 

(Johns  Hopkins  School  of  Medicine; 

later,  University  of  Connecticut) 
W.  M.  Paul,  1954  (University  of  Toronto) 
Donald  F.  Poulson,  1936-1937 

(Yale  University) 
Curt  P.  Richter,  1927-1930 

(Johns  Hopkins  School  of  Medicine) 
Eduardo  de  Robertis,  1941-1942 

(University  of  Buenos  Aires) 
John  Rock,  1938-1945 

(Harvard  Medical  School) 
Edward  Roosen-Runge,  1957 

(University  of  Washington) 
Florence  Sabin,  1914-1924 

(Johns  Hopkins  University;  later, 

Rockefeller  Institute) 
Jorgen  U.  Schlegel,  1948-1949 

(University  of  Copenhagen) 
Harold  D.  Senior,  1917-1918 

(New  York  University) 
Ronald  Singer,  1951-1952 

(University  of  Cape  Town,  South  Africa; 

later,  University  of  Chicago) 
William  L.  Straus,  Jr.,  1925-1930 

(Johns  Hopkins  School  of  Medicine) 
Fritz  Strauss,  1950  (University  of  Bern) 
Somers  H.  Sturgis,  1942-1943 

(Massachusetts  General  Hospital) 
Francis  H.  Swett,  1924-1927 

(Johns  Hopkins  School  of  Medicine; 

later,  Duke  University) 
Pierre  Tardent,  1959-1960 

(Zoological  Station,  Naples) 
M.  H.  Toosy,  1948-1949 

(Lahore  Medical  School) 
Theodore  W.  Torrey,  1952 

(Indiana  University) 
U.  U.  Uotila,  1939-1940 

(Harvard  Medical  School) 
W.  J.  van  Doorenmaalen,  1958-1959 

(Municipal  University,  Amsterdam) 
R.  Walmsley,  1935-1936 

(University  of  Edinburgh) 
Lewis  H.  Weed,  Research  Associate 

1914-1919,  1921-1935 

(Johns  Hopkins  University) 


REPORT     OF    THE     PRESIDENT 


89 


Karl  M.  Wilson,  1913-1924 

(Johns  Hopkins  School  of  Medicine; 

later,  University  of  Rochester) 
Milton  C.  Winternitz,  1914-1916 

(Johns  Hopkins  School  of  Medicine; 

later,  Yale  University) 


George  Wislocki,  1916-1931 

(Johns  Hopkins  School  of  Medicine; 

later,  Harvard  University  ) 
Emil  Witschi,  1941-1942 

(University  of  Iowa) 


DEPARTMENT   OF   HISTORICAL   RESEARCH 

Organized  as  a  "bureau"  in  1903;  became  a  "department"  in  1905;  terminated  as  a  department 
and  incorporated  as  the  Section  of  United  States  History  in  a  new  Division  of  Historical  Research, 
1930. 

Directors 

Andrew  C.  McLaughlin,  1903-1905 

John  F.  Jameson,  1905-1928 

None,  1928-1930 


Staff  Members 


Edmund  C.  Burnett,  1907-1932 
Frances  G.  Davenport,  1905-1927 
Elizabeth  Donnan,  1911-1919 
Shirley  Farr,  1921-1922 
Mary  F.  Griffin,  1922-1925 


Waldo  G.  Leland,  1903-1945 
Marguerite  M.  McKee,  1925-1929 
David  W.  Parker,  1909-1913,  1925-1928 
Charles  O.  Paullin,  1912-1936 
Leo  F.  Stock,  1910-1945 


Other  Scientists  and  Scholars  Associated  with  the  Department 


Charles  Francis  Adams,  1902 

(Massachusetts  Historical  Society) 
Ephraim  D.  Adams,  1904 

(Stanford  University) 
William  H.  Allison,  1906-1911  (Bryn  Mawr 

College;  later,  Colgate  University) 
Charles  M.  Andrews,  1904-1918 

(Yale  University) 
James  C.  Ballagh,  1907-1908 

(Johns  Hopkins  University;  later, 

University  of  Pennsylvania) 
Adolf  F.  A.  Bandelier,  1911-1914 

(Columbia  University) 
Eugene  C.  Barker,  1906 

(University  of  Texas) 
John  S.  Bassett,  1921-1928  (Smith  College) 
Herbert  C.  F.  Bell,  1916-1923 

(Bowdoin  College;  later,  Wesleyan 

University) 
Samuel  F.  Bemis,  1923-1925 

(George  Washington  University;  later, 

Yale  University) 
Herbert  E.  Bolton,  1907-1913 

(Stanford  University;  later, 

University  of  California) 


Harold  Martin  Bowman,  1907-1908 

(Boston  University  School  of  Law) 
Julian  P.  Bretz,  1906 

(later,  Cornell  University) 
Helen  T.  Catterall,  1918-1933  (Boston  Bar) 
Isaac  Joslin  Cox,  1906-1908 

(University  of  Cincinnati;  later, 

Northwestern  University) 
Walter  F.  Dodd,  1908  (Library  of  Congress; 

later,  Yale  University) 
Max  Farrand,  1912-1913  (Yale  University; 

later,  Henry  E.  Huntington  Library  and 

Art  Gallery) 
Albert  B.  Faust,  1912-1916 

(Cornell  University) 
William  S.  Ferguson,  1906-1908 

(University  of  California;  later, 

Harvard  University) 
CarlR.  Fish,  1908-1911 

(University  of  Wisconsin) 
Worthington  C.  Ford,  1903-1906 

(Library  of  Congress;  later, 

Massachusetts  Historical  Society) 
Dixon  R.  Fox,  1919-1920  (Columbia 

University;  later,  Union  College) 


90 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Frank  A.  Golder,  1914-1922 

(State  College  of  Washington;  later, 

Stanford  University) 
Evarts  B.  Greene,  1918  (University  of 

Illinois;  later,  Columbia  University) 
Charles  W.  Hackett,  1918-1929 

(University  of  Texas) 
Charles  H.  Haskins,  1905-1908 

(Harvard  University) 
Roscoe  R.  Hill,  1910-1917 

(later,  Nicaraguan  High  Commission; 

National  Archives) 
Frank  H.  Hodder,  1912  (University  of  Kansas) 
William  Wirt  Howe,  1904  (Board  of  Trustees, 

Carnegie  Institution  of  Washington) 
William  I.  Hull,  1914  (Swarthmore  College) 
Herman  G.  James,  1923  (University  of 

South  Dakota;  Ohio  University) 
Marcus  W.  Jernegan,  1907-1926 

(University  of  Chicago) 
Louise  P.  Kellogg,  1922 

(Wisconsin  State  Historical  Society) 
Benjamin  B.  Kendrick,  1910  (Women's 

College  of  University  of  North  Carolina) 
Marion  D.  Learned,  1908-1912 

(University  of  Pennsylvania) 
Orin  G.  Libby,  1912 

(University  of  North  Dakota) 
George  W.  Littlehales,  1915 

(George  Washington  University;  formerly 

with  Carnegie  Institution  of  Washington's 

Department  of  Terrestrial  Magnetism) 
Alfred  T.  Mahan,  1914-1915 

(U.  S.  Navy,  retired) 
William  R.  Manning,  1908-1910 

(George  Washington  University;  later, 

Department  of  State) 


John  J.  Meng,  1936-1954 

(Catholic  University  of  America;  later, 

Hunter  College) 
Herbert  L.  Osgood,  1912-1918 

(Columbia  University) 
Edwin  W.  Pahlow,  1926-1927 

(Ohio  State  University) 
Frederic  L.  Paxson,  1910-1914 

(University  of  Wisconsin;  later, 

University  of  California) 
Francis  S.  Philbrick,  1914-1915 

(University  of  Pennsylvania) 
Jesse  S.  Reeves,  1912-1913 

(University  of  Michigan) 
James  A.  Robertson,  1909-1917,  1931-1932 

(Stetson  University;  later,  Archives  of 

Maryland) 
Robert  W.  Rogers,  1924-1927 

(Drew  Theological  Seminary) 
Joseph  Schafer,  1918-1919 

(University  of  Oregon) 
George  W.  Scott,  1903-1905  (Library  of 

Congress;  Columbia  University) 
William  R.  Shepherd,  1905-1908 

(Columbia  University) 
William  A.  Slade,  1904-1905 

(Library  of  Congress) 
Frederick  J.  Turner,  1916-1917 

(Harvard  University) 
Arnold  J.  F.  van  Lear,  1919-1926 

(New  York  State  Library;  later, 

New  York  State  Education  Department) 
Claude  H.  Van  Tyne,  1904-1908 

(University  of  Michigan) 
Ray  H.  Whitbeck,  1914-1915 

(University  of  Wisconsin) 
Irene  A.  Wright,  1925-1928  (Library  of 

Congress;  later,  National  Archives) 


DIVISION   OF   HISTORICAL  RESEARCH 

Established  1930,  superseding  the  Department  of  Historical  Research,  which  became  a  section  of 
United  States  History  in  the  Division.  The  other  two  sections  were  the  Section  of  Aboriginal  American 
History,  which  continued  the  archaeological  work  already  begun  by  Sylvanus  G.  Morley  in  Central 
America  and  by  E.  H.  Morris  in  southwestern  United  States,  and  the  Section  of  the  History  of 
Science.  Became  the  Department  of  Archaeology,  1951. 


Alfred  V.  Kidder,  Chairman,  1930-1950 
Harry  E.  D.  Pollock,  Director,  1951-1958 


REPORT     OF    THE     PRESIDENT 


91 


Staff  Members 


Eleanor  B.  Adams,  1934-1949 
Robert  S.  Chamberlain,  1937-1947 
Sylvanus  G.  Morley,  1914-1948 
Earl  H.  Morris,  1925-1955 
Alexander  Pogo,  1929-1950 
Tatiana  A.  Proskouriakoff,  1939 — 
Ralph  L.  Roys,  1930-1953 
Karl  Ruppert,  1925-1956 
George  A.  L.  Sarton,  1918-1949 


France  V.  Scholes,  1931-1946 
Anna  0.  Shepard,  1933— 
Edwin  M.  Shook,  1933-1958 
A.  Ledyard  Smith,  1929-1958 
Robert  E.  Smith,  1931-1960 
Gustav  Stromsvik,  1926-1957 
Sol  Tax,  1938-1947 
J.  Eric  S.  Thompson,  1935-1959 
Alfonso  Villa  Rojas,  1932-1947 


Manuel  J.  Andrade,  1932,  1936-1940  J.  Ignacio  Rubio  Man6,  1936-1942 

Abraham  M.  Halpern,  1941-1942  Oliver  G.  Ricketson,  Jr.,  1920-1940 

Henry  B.  Roberts,  1926-1939 


Other  Scientists  and  Scholars  Associated  with  the  Division 


Sophie  D.  Aberle,  Research  Associate 

1933-1940  (United  Pueblo  Agency, 

Albuquerque,  New  Mexico;  later, 

Chief  Nutrition,  Bernalillo  County 

Indian  Hospital) 
Robert  M.  Adams,  Jr.,  1951-1952 

(Oriental  Institute,  University  of  Chicago) 
Monroe  Amsden,  1923-1924,  1927 

(southwestern  archaeologist) 
E.  Wyllys  Andrews, 

1939-1940,  1941-1942,  1947-1948 

(Tulane  University) 
Herman  Beyer,  1937  (Tulane  University) 
Franz  Blom,  1924-1925  (Tulane  University) 
Stephen  F.  de  Borhegyi,  1949 

(University  of  Oklahoma;  later, 

Milwaukee  Public  Museum) 
George  W.  Brainerd,  1939-1942,  1948-1949 

(University  of  California,  Los  Angeles) 
Kirk  Bryan,  1945  (Harvard  University) 
W.  R.  Bullard,  1951-1953 

(Peabody  Museum,  Harvard  University) 
Alfonso  Caso  y  Andrade,  Research  Associate 

1936-1939  (Instituto  Nacional  Indigenista, 

Mexico) 
Kenneth  M.  Chapman,  1935 

(University  of  New  Mexico) 
Jean  Chariot,  1926-1931  (painter  and  teacher) 
Ann  Chowning,  1954-1955 

(Bryn  Mawr  College;  later, 

University  of  Pennsylvania) 
I.  Bernard  Cohen,  1938-1941 

(Harvard  University) 
Fay-Cooper  Cole,  1931 

(University  of  Chicago) 


G.  W.  Collins,  1936-1937 

(U.  S.  Department  of  Agriculture) 
Frank  H.  Connell,  1931-1932 

(Dartmouth  College) 
Luther  S.  Cressman,  Research  Associate 

1936-1942  (University  of  Oregon) 
John  H.  Denison,  Jr.,  1937-1938 

(Big  Horn,  Wyoming) 
Rollins  A.  Emerson,  1934-1935 

(Cornell  University) 
F.  W.  Gaige,  1930-1931 

(University  of  Michigan) 
Rutherford  J.  Gettens,  1955 

(Freer  Gallery  of  Art,  Washington,  D.  C.) 
John  P.  Gillin,  1941-1943,  1945-1946 

(Duke  University;  later,  University  of 

Pittsburgh) 
Antonio  Goubaud,  1944-1945 

(Instituto  de  Antropologia  e  Historia, 

Guatemala  City) 
Carl  E.  Guthe,  Research  Associate,  1921-1922 

(New  York  State  Museum,  Albany) 
Lewis  U.  Hanke,  1935-1939 

(Library  of  Congress;  later,  University  of 

Texas  and  Columbia  University) 
Mark  R.  Harrington,  Research  Associate 

1930-1936  (Southwest  Museum, 

Los  Angeles) 
William  A.  Heidel,  1928-1939 

(Wesleyan  University) 
Edgar  B.  Howard,  1934-1942 

(University  of  Pennsylvania) 
William  T.  Howard,  Jr.,  1924 

(Johns  Hopkins  University) 
Jesse  D.  Jennings,  1936-1937 

(University  of  Utah) 


92 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


J.  H.  Kempton,  1934-1938 

(U.  S.  Department  of  Agriculture) 
J.  Steward  Lincoln,  1940-1941 

(Guatemala  City) 
John  M.  Longyear,  III,  1937-1939, 

1941-1942,  1945-1947,  1949-1950 

(Colgate  University) 
Samuel  K.  Lothrop,  1922-1933 

(Peabody  Museum,  Harvard  University) 
Cyrus  L.  Lundell,  Research  Associate 

1933-1941  (University  of  Michigan) 
Maud  Worcester  Makemson,  1943 

(Vassar  College) 
Norman  A.  McQuown,  1937-1949 

(University  of  Chicago) 
Paul  S.  Martin,  1926-1928 

(Chicago  Natural  History  Museum) 
Ann  Axtell  Morris,  1926-1931 

(Boulder,  Colorado) 
Lila  M.  O'Neale,  1935-1936 

(University  of  California,  Berkeley) 
Arthur  S.  Pearse,  1928-1936 

(Duke  University) 
Wilson  Popenoe,  1935-1936 

(United  Fruit  Company;  later,  Escuela 

Agricola  Panamericana,  Tegucigalpa, 

Honduras) 
Robert  Redfield,  Research  Associate 

1930-1949  (University  of  Chicago) 
Ruth  Reeves,  Research  Associate,  1934-1935 

(New  York  City,  New  York) 
Juan  de  Dios  Rosales,  1944-1946 

(Instituto  Indigenista  de  Guatemala) 
George  M.  Saunders,  1930-1932 

(Harvard  University) 


Adolph  H.  Schultz,  Research  Associate 

1916-1925,  1937-1938  (Johns  Hopkins 

University;  later,  University  of  Zurich) 
George  C.  Shattuck,  1929-1939 

(Boston  City  Hospital;  later, 

Massachusetts  General  Hospital) 
Joseph  L.  Smith,  1941 

(Boston  Museum  of  Fine  Arts) 
Philip  E.  Smith,  1953-1954 

(University  of  Toronto) 
R.  Stadelman,  1936-1938 

(U.  S.  Department  of  Agriculture) 
L.  C.  Stuart,  1932-1933 

(University  of  Michigan) 
John  Teeple,  1928-1931 

(consulting  chemist,  New  York  City) 
Antonio  Tejeda  F.,  1938-1939,  1944-1947 

(Museo  Nacional  de  Arqueologia  y 

Etnologia,  Guatemala  City) 
Donald  E.  Thompson,  1954-1955 

(University  of  Wisconsin) 
Aubrey  S.  Trik,  1935-1938  (University 

Museum,  University  of  Pennsylvania) 
Melvin  Tumin,  1942-1944 

(University  of  North  Carolina) 
George  C.  Vaillant,  1925-1940 

(American  Museum  of  Natural  History; 

later,  University  Museum,  University  of 

Pennsylvania) 
Robert  Wauchope,  1933-1936 

(Tulane  University) 
Howell  Williams,  1949-1950 

(University  of  California,  Berkeley) 
Clark  Wissler,  Research  Associate,  1924-1933 

(American  Museum  of  Natural  History; 

later,  Yale  University) 


DEPARTMENT  OF  ECONOMICS  AND  SOCIOLOGY 

Organized  1904;  terminated  1916. 

Board  Members 

Carroll  D.  Wright,  Director,  1904-1909  (Clark  College) 

Henry  W.  Farnam,  Chairman,  1909-1916  (Yale  University) 

Kenyon  L.  Butterfield,  Agriculture  and  Forestry,  1904-1915  (Rhode  Island  College  of 

Agriculture  and  Mechanic  Arts) 

Victor  S.  Clark,  Manufactures,  1906-1916  (Census  Bureau) 

John  R.  Commons,  The  Labor  Movement,  1909-1915  (University  of  Wisconsin) 

Davis  R.  Dewey,  Money  and  Banking,  1904-1914  (Institute  of  Technology,  Boston) 

Henry  B.  Gardner,  Federal  and  State  Finance,  1904-1914  (Brown  University) 

J.  W.  Jenks,  Industrial  Organization,  1904-1914  (Cornell  University) 


REPORT     OF     THE     PRESIDENT 


98 


Emory  R.  Johnson,  Domestic  and  Foreign  Commerce,  1904-1915  (University  of  Pennsylvania) 

B.  H.  Meyer,  Transportation,  1904-1916  (University  of  Wisconsin) 

S.  N.  D.  North,  Manufactures,  1904  (Census  Bureau) 

Edward  W.  Parker,  Mining,  1904-1915  (U.  S.  Geological  Survey) 

W.  Z.  Ripley,  Transportation,  1904  (Newton  Centre,  Massachusetts) 

Alfred  Holt  Stone,  The  Negro  in  Slavery  and  Freedom,  1906-1914  (Dunleith,  Mississippi) 

Walter  F.  Willcox,  Population  and  Immigration,  1904-1914  (Cornell  University) 


Other  Scientists  and  Scholars  Associated  with  the  Department 


Edith  Abbott,  1905-1910 

(University  of  Chicago) 
Henry  C.  Adams,  1904 

(University  of  Michigan) 
Charles  H.  Ambler,  1910 

(Randolph-Macon  College;  later, 

(University  of  West  Virginia) 
John  B.  Andrews,  1913-1915 

(numerous  activities  in  labor  economics) 
Oliver  Edwin  Baker,  1912 

(U.  S.  Department  of  Agriculture) 
Emily  Greene  Balch,  1904-1907 

(Wellesley  College) 
F.  Spencer  Baldwin,  1908-1909 

(Boston  University) 
J.  Lynn  Barnard,  1905-1908 

(Philadelphia  School  of  Pedagogy) 
Alvard  Longley  Bishop,  1907-1908 

(Yale  University) 
Frank  W.  Blackmar,  1904-1914 

(University  of  Kansas) 
Ernest  Ludlow  Bogart,  1904-1912 

(Oberlin  College;  later,  Princeton 

University  and  University  of  Illinois) 
Beverley  Waugh  Bond,  1908-1909 

(Purdue  University;  later, 

University  of  Cincinnati) 
William  K.  Boyd,  1910-1913 

(Duke  University) 
James  E.  Boyle,  1905-1908 

(University  of  North  Dakota) 
Solon  J.  Buck,  1906-1913 

(University  of  Indiana;  later, 

Archivist  of  the  United  States) 
Thomas  N.  Carver,  1904-1912 

(Harvard  University) 
Robert  E.  Chaddock,  1909 

(Columbia  University) 
John  B.  Clark,  1902  (Columbia  University) 
Frederick  A.  Cleveland,  1905-1913 

(New  York  University;  later,  in  charge  of 

President  Taft's  Commission  on  Economy 

and  Efficiency) 


Thomas  Conway,  Jr.,  1904-1913 

(University  of  Pennsylvania) 
Mary  Roberts  Coolidge,  1907-1910 

(Mills  College) 
John  Lee  Coulter,  1908-1912 

(University  of  Minnesota;  later, 

U.  S.  Tariff  Commission) 
James  Walter  Crook,  1908-1911 

(Amherst  College) 
Ira  Brown  Cross,  1909-1913 

(University  of  California) 
Stuart  Daggett,  1904-1913 

(University  of  California) 
Edgar  M.  Dawson,  1908-1913 

(Princeton  University;  later, 

Hunter  College) 
Clive  Day,  1907-1909  (Yale  University) 
David  T.  Day,  1908-1911 

(U.  S.  Geological  Survey;  later, 

U.  S.  Bureau  of  Mines) 
Robert  N.  Denham,  Jr.,  1908 

(University  of  Michigan;  later, 

National  Labor  Relations  Board) 
Carroll  W.  Doten,  1906-1908 

(Massachusetts  Institute  of  Technology) 
W.  E.  B.  Dubois,  1908  (Atlanta  University) 
Edwin  C.  Eckel,  1904-1908 

(U.  S.  Geological  Survey) 
Richard  T.  Ely,  1904-1909 

(University  of  Wisconsin) 
Fred  Rogers  Fairchild,  1904-1909 

(Yale  University) 
Henry  Pratt  Fairchild,  1908-1909 

(Bowdoin  College;  later, 

New  York  University) 
John  I.  Falconer,  1912-1913 

(Ohio  State  University) 
Albert  B.  Faust,  1907-1910 

(Cornell  University) 
Walter  L.  Fleming,  1908-1911 

(Louisiana  State  University  and 

Vanderbilt  University) 


94 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Albert  A.  Giesecke,  1904-1910 

(University  of  Pennsylvania;  later, 

University  of  Cuzco,  Peru) 
Eugene  A.  Gilmore,  1910 

(University  of  Wisconsin;  later, 

State  University  of  Iowa) 
E.  A.  Goldenweiser,  1904-1908 

(various  economic  posts  in  U.  S. 

Government) 
L.  C.  Graton,  1908-1913 

(U.  S.  Geological  Survey;  later, 

Harvard  University) 
Elmer  C.  Griffith,  1908-1911 

(Kalamazoo  College) 
George  Gorham  Groat,  1904-1908 

(Ohio  Wesleyan  University;  later, 

University  of  Vermont) 
James  Edward  Hagerty,  1905-1909 

(Ohio  State  University) 
Robert  M.  Haig,  1910-1913 

(Columbia  University) 
Matthew  Brown  Hammond,  1904-1909 

(Ohio  State  University) 
Glover D.  Hancock,  1908-1911 

(Amherst  College;  later, 

Washington  and  Lee  University) 
Lewis  Henry  Haney,  1906-1910 

(New  York  University) 
Hugh  Sisson  Hanna,  1905-1908 

(U.  S.  Bureau  of  Labor  Statistics) 
Adelaide  R.  Hasse,  1905-1917 

(New  York  Public  Library;  later, 

Brookings  Institution) 
Frank  I.  Herriott,  1905-1911 

(Drake  University) 
Benjamin  H.  Hibbard,  1908-1914 

(Iowa  State  College;  later, 

University  of  Wisconsin) 
Henry  E.  Hoagland,  1911-1913 

(Ohio  State  University) 
Roy  Jay  Holden,  1908-1915 

(Virginia  Polytechnic  Institute) 
Jacob  H.  Hollander,  1904-1907 

(Johns  Hopkins  University) 
Solomon  S.  Huebner,  1904-1911 

(University  of  Pennsylvania) 
Walter  Renton  Ingalls,  1904-1908 

(construction  engineer,  New  York  City) 
Theodore  H.  Jack,  1909-1911 

(Emory  University;  later, 

Randolph-Macon  College) 
Edward  D.  Jones,  1908-1914 

(University  of  Michigan) 
T.  J.  Jones,  1908  (Hampton  Institute; 

later,  Phelps  Stokes  Fund) 


Clyde  L.  King,  1911-1913 

(University  of  Pennsylvania) 
Julius  Klein,  1909 

(U.  S.  Department  of  Commerce) 
Francis  Baker  Laney,  1904-1915 

(U.  S.  National  Museum;  later, 

U.  S.  Geological  Survey) 
John  Lapp,  1908  (Cornell  University; 

later,  Marquette  University) 
Laurence  M.  Larson,  1905-1909 

(University  of  Illinois) 
C.  K.  Leith,  1904-1915 

(University  of  Wisconsin) 
Isaac  P.  Lippincott,  1909 

(Washington  University) 
Oliver  C.  Lockhart,  1908-1913 

(Ohio  State  University) 
Isaac  A.  Loos,  1905-1909 

(State  University  of  Iowa) 
Gerald  Francis  Loughlin,  1915 

(U.  S.  Geological  Survey) 
David  A.  McCabe,  1912 

(Princeton  University) 
Charles  McCarthy,  1904 

(U.  S.  Commission  on  Industrial  Relations; 

later,  U.  S.  Food  Administration) 
James  Farley  McClelland,  1904-1905 

(Columbia  School  of  Mines;  later, 

Yale  University) 
George  McCutchen,  1908-1912 

(University  of  South  Carolina) 
S.  J.  McLean,  1906-1910 

(University  of  Toronto) 
F.  L.  McVey,  1908-1911 

(University  of  North  Dakota;  later, 

University  of  Kentucky) 
E.  T.  Miller,  1905-1915  (University  of  Texas) 
H.  A.  Millis,  1909-1912 

(Stanford  University;  later, 

University  of  Chicago) 
Wesley  C.  Mitchell,  1904-1908 

(University  of  California;  later, 

New  School  for  Social  Research) 
Blaine  F.  Moore,  1908-1909 

(U.  S.  Commission  on  Industrial  Relations; 

later,  University  of  Kansas) 
Charles  E.  Munroe,  1904-1910 

(George  Washington  University) 
Henry  R.  Mussey,  1904  (Wellesley  College) 
W.  T.  Nardin,  1905  (Pet  Milk  Company) 
Selig  Perlman,  1911-1915 

(University  of  Wisconsin) 
Warren  Milton  Persons,  1908 

(Colorado  College;  later,  Harvard 

University) 


REPORT     OF    THE     PRESIDENT 


95 


John  B.  Phillips,  1908-1909 

(University  of  Colorado;  later, 

University  of  Indiana) 
Ulrich  B.  Phillips,  1904-1910 

(Tulane  University;  later,  University  of 

Michigan  and  Yale  University) 
Charles  F.  Pidgin,  1908 

(Massachusetts  Bureau  of  Statistics  of 

Labor) 
CarlC.  Plehn,  1904-1911 

(University  of  California) 
Fred  Wilbur  Powell,  1909-1913 

(Brookings  Institution) 
Joseph  Hyde  Pratt,  1904-1910 

(University  of  North  Carolina) 
E.  P.  Puckett,  1908-1913  (Central  College) 
Charles  Lee  Raper,  1905-1909 

(University  of  North  Carolina;  later, 

Syracuse  University) 
William  A.  Rawles,  1904-1911 

(University  of  Indiana) 
Heinrich  Ries,  1904-1909 

(Cornell  University) 
Thomas  James  Riley,  1908-1909 

(University  of  Missouri;  later, 

Washington  University) 
Clyde  Orval  Ruggles,  1908-1911 

(Ohio  State  University;  later, 

Harvard  University) 
Aaron  M.  Sakolski,  1906 

(New  York  University) 
David  J.  Saposs,  1911-1915 

(various  government  posts  in  labor 

economics) 
William  O.  Scroggs,  1905-1911 

(Louisiana  State  University) 
A.  E.  Sheldon,  1904-1905 

(Nebraska  Historical  Society) 
St.  George  L.  Sioussat,  1904-1913 

(University  of  the  South;  later, 

University  of  Pennsylvania) 
J.  Russell  Smith,  1904-1908 

(University  of  Pennsylvania;  later, 

Columbia  University) 
Yates  Snowden,  1910-1911 

(University  of  South  Carolina) 


Don  C.  Sowers,  1910-1913 

(University  of  Colorado) 
Robert  James  Sprague,  1908-1909 

(University  of  Maine;  later, 

Rollins  College) 
Harry  Harkness  Stoek,  1904-1908 

(editor,  Mining  and  Minerals;  later, 

University  of  Illinois) 
Edgar  M.  Sydenstricker,  1908-1915 

(U.  S.  Public  Health  Service) 
Henry  C.  Taylor,  1908-1915 

(University  of  Wisconsin;  later, 

Farm  Foundation) 
D.  Y.  Thomas,  1907-1908 

(University  of  Florida;  later, 

University  of  Arkansas) 
William  H.  Tolman,  1908 

(Pawtucket,  Rhode  Island) 
Walter  Sheldon  Tower,  1905-1908 

(Bethlehem  Steel  Corporation;  later, 

Iron  and  Steel  Institute) 
Robert  James  Usher,  1905 

(Howard-Tilton  Memorial  Library, 

Tulane  University) 
Francis  Walker,  1909-1910 

(Federal  Trade  Commission) 
Royal  Brunson  Way,  190&-1908 

(Northwestern  University;  later, 

Beloit  College) 
Nathan  Austin  Weston,  1905-1910 

(University  of  Illinois) 
Horace  L.  Wilgus,  1905-1909 

(University  of  Michigan) 
C.  C.  Williamson,  1905-1908 

(New  York  Public  Library;  later, 

Columbia  University) 
Calvin  Dill  Wilson,  1908-1912 

(clergyman  and  author) 
Edwin  E.  Witte,  1911-1912 

(University  of  Wisconsin) 
R.  R.  Wright,  Jr.,  1908-1910 

(Georgia  State  Industrial  College) 
Allyn  A.  Young,  1905-1910 

(Stanford  University;  later,  Cornell 

University  and  Harvard  University) 
Frederic  G.  Young,  1905-1913 

(University  of  Oregon) 


96 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


DEPARTMENT   OF   MARINE   BIOLOGY 

Established  in  1804.  Name  changed  to  Tortugas  Laboratory  in  1923.  Activities  terminated  in  1939. 

Directors 

Alfred  G.  Mayer,  1904-1922 

William  Harding  Longley,  1923-1927  (Administrative  Officer);  1928-1937  (Executive  Officer) 

David  Hilt  Tennent,  1938-1939  (Executive  Officer) 

Staff  Members 

Paul  S.  Conger,  1924-1929,  1937-1938  Albert  Mann,  1919-1933 

John  W.  Mills,  1906-1939 

Other  Scientists  and  Scholars  Associated  with  the  Department 


Percy  L.  Bailey,  Jr.,  1937 

(College  of  the  City  of  New  York) 
Stanley  C.  Ball,  1913-1914,  1917 

(Massachusetts  Agricultural  College; 

later,  Bishop  Museum,  Honolulu,  and 

Peabody  Museum,  Yale  University) 
Paul  Bartsch, 

1912-1917,  1919,  1921-1927,  1930-1932 

(U.  S.  National  Museum;  later, 

George  Washington  University) 
Norman  J.  Berrill,  1937  (McGill  University) 
Lawrence  R.  Blinks,  1925-1928 

(Rockefeller  Institute;  later,  Stanford 

University  and  Hopkins  Marine  Station) 
H.  Boschma,  1924  (Rijksuniversiteit,  Leiden) 
Howard  H.  M.  Bowman,  1915-1916 

(University  of  Pennsylvania;  later, 

University  of  Toledo) 
Alan  A.  Boyden,  1931,  1933,  1935 

(Rutgers  University) 
Charles  M.  Breder,  Jr.,  1928 

(New  York  Aquarium  and 

American  Museum  of  Natural  History) 
Floyd  J.  Brinley,  1936-1937 

(North  Dakota  Agricultural  College; 

later,  University  of  Toledo) 
William  K.  Brooks,  1905-1907,  1909 

(Johns  Hopkins  University) 
Dugald  E.  S.  Brown,  1934 

(New  York  University  Medical  School; 

later,  University  of  Michigan) 
Walter  E.  Bullington,  1929-1930,  1934 

(Randolph-Macon  College) 
Martin  Burkenroad,  1928  (Tulane 

University;  later,  Marine  Biological  Station, 

National  Museum  of  Panama) 


Lewis  R.  Cary,  Research  Associate 

1910-1918,  1920,  1929-1933,  1935 

(Princeton  University) 
Edward  L.  Chambers,  1936 

(Princeton  University;  later, 

University  of  Miami  School  of  Medicine, 

Coral  Gables) 
Robert  Chambers,  1936 

(Washington  Square  College,  New  York 

University;  later,  Marine  Biological 

Laboratory,  Woods  Hole) 
Frank  M.  Chapman,  1907-1909 

(American  Museum  of  Natural  History) 
Hubert  Lyman  Clark,  Research  Associate 

1912-1917,  1929-1930  (Museum  of 

Comparative  Zoology,  Harvard  University) 
Leonard  B.  Clark,  1936-1937 

(Union  College) 
Frank  W.  Clarke,  1919 

(U.  S.  Geological  Survey) 
Leon  J.  Cole,  1906-1914  (Yale  University; 

later,  University  of  Wisconsin) 
John  Colman,  1930  (Cambridge  University) 
Edwin  G.  Conklin,  1905,  1907,  1909,  1915 

(Princeton  University) 
Benjamin  R.  Coonfield,  1937 

(Brooklyn  College) 
Rheinart  P.  Cowles,  1905-1909,  1914 

(Johns  Hopkins  University) 
Paul  R.  Cutright,  1936  (Beaver  College) 
Ulric  Dahlgren,  1906,  1908,  1911-1922 

(Princeton  University) 
Reginald  A.  Daly,  1919  (Harvard  University) 
John  H.  Davis,  Jr.,  1936-1937 

(Southwestern  College;  later, 

University  of  Florida) 


REPORT     OF     THE     PRESIDENT 


97 


May  W.  de  Laubenfels,  1926-1927,  1931,  1935 

(Pasadena  Junior  College;  later, 

Oregon  State  College) 
George  S.  de  Renyi,  1933 

(University  of  Pennsylvania) 
Richard  B.  Dole,  1913 

(U.  S.  Geological  Survey) 
Henry  H.  Donaldson,  1916 

(Wistar  Institute  of  Anatomy) 
William  L.  Doyle,  1933-1934 

(Johns  Hopkins  University;  later, 

University  of  Chicago) 
George  Harold  Drew,  1911-1913 

(Christ's  College,  Cambridge  University) 
Gilman  A.  Drew,  1912 

(Marine  Biological  Laboratory, 

Woods  Hole) 
Charles  H.  Edmondson,  1906-1907 

(Iowa  Wesleyan;  later,  University  of 

Hawaii  and  Bishop  Museum,  Honolulu) 
Richard  M.  Field,  1919  (Museum  of 

Comparative  Zoology,  Harvard  University ; 

later,  Princeton  University) 
A.  Haldane  Gee,  1929 

(Scripps  Institution  of  Oceanography; 

later,  Foster  D.  Snell,  Inc.,  New  York) 
John  H.  Gerould,  1915,  1921-1922 

(Dartmouth  College) 
Isidore  I.  Gersh,  1934 

(Johns  Hopkins  University  Medical  School; 

later,  University  of  Chicago  School  of 

Medicine) 
Abraham  J.  Goldforb,  1912-1913,  1915-1916 

(College  of  the  City  of  New  York) 
Hubert  B.  Goodrich,  1934 

(Wesleyan  University) 
Myron  Gordon,  1927,  1932 

(Cornell  University;  later, 

American  Museum  of  Natural  History  and 

New  York  Zoological  Society) 
James  N.  Gowanlock,  1929 

(Dalhousie  University) 
Caswell  Grave, 

1924,  1926-1929,  1932,  1934-1935 

(Washington  University) 
George  M.  Gray,  1912 

(Marine  Biological  Laboratory, 

Woods  Hole) 
Eugene  W.  Gudger,  1908,  1912-1915 

(North  Carolina  College  for  Women;  later, 

American  Museum  of  Natural  History) 
George  T.  Hargitt,  1905 

(Northwestern  University;  later, 

Syracuse  University) 


John  E.  Harris,  1933-1934,  1936 

(Cambridge  University) 
J.  A.  Harrison,  1936  (University  of  London) 
Robert  Hartmeyer,  1907 

(Berlin  Zoological  Museum) 
E.  Newton  Harvey,  Research  Associate 

1909-1925,  1929  (Princeton  University) 
Shinkishi  Hatai,  1916-1917 

(Wistar  Institute  of  Anatomy) 
Frederick  R.  Hayes,  1931 

(Institute  of  Oceanography, 

Dalhousie  University,  Halifax) 
Edwin  R.  Helwig,  1932 

(University  of  Pennsylvania;  later, 

University  of  Colorado) 
Walter  N.  Hess,  1930,  1937 

(Hamilton  College) 
Davenport  Hooker,  1905,  1907-1909,  1914 

(Yale  University) 
Dwight  L.  Hopkins,  1928-1930 

(Duke  University;  later, 

Mundelein  College,  Chicago) 
Robert  Tracy  Jackson,  1912 

(Museum  of  Comparative  Zoology, 

Harvard  University) 
Merkel  H.  Jacobs,  1911 

(University  of  Pennsylvania) 
Norris  Jones,  1936-1937 

(Swarthmore  College) 
Harvey  E.  Jordan,  1907,  1909,  1912-1914 

(University  of  Virginia) 
E.  Jorgensen,  1910  (University  of  Bergen) 
Carl  Kellner,  1905-1907,  1909 

(Yale  University) 
Milton  J.  Kopac,  1932-1934,  1936 

(University  of  California;  later, 

New  York  University) 
Beverly  W.  Kunkel,  1930  (Lafayette  College) 
Karl  S.  Lashley,  1913-1915 

(Johns  Hopkins  University;  later, 

Harvard  University  and  Yerkes 

Laboratories  of  Primate  Biology) 
Marius  Le  Compte,  1936 

(Royal  Museum  of  Natural  History, 

Brussels) 
James  L.  Leitch,  1931,  1933,  1935 

(University  of  California;  later, 

Armstrong  College) 
Ivey  F.  Lewis,  1927  (University  of  Virginia) 
Frank  R.  Lillie,  1935-1936 

(University  of  Chicago) 
Edwin  Linton,  1906-1909 

(Washington  and  Jefferson  College;  later, 

University  of  Pennsylvania) 


98 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Charles  B.  Lipman,  1920,  1922-1923 

(University  of  California) 
Balduin  Lucke,  1936-1937 

(University  of  Pennsylvania  Medical 

School) 
Jesse  F.  McClendon, 

1908-1910,  1916-1917,  1919 

(University  of  Missouri;  later, 

University  of  Minnesota  and 

Einstein  Medical  Center,  Philadelphia) 
Oliver  McCoy,  1927-1928 

(Johns  Hopkins  University;  later, 

University  of  Rochester  and 

China  Medical  Board  of  New  York) 
Harold  W.  Manter,  1929-1931,  1933 

(University  of  Nebraska) 
Gordon  Marsh,  1929,  1934-1937 

(University  of  Iowa) 
James  C.  Martin,  1933 

(University  of  California) 
Cloyd  Heck  Marvin,  1932 

(George  Washington  University) 
Samuel  0.  Mast,  1910  (Goucher  College; 

later,  Johns  Hopkins  University) 
George  Matthai,  1915  (Emmanuel  College, 

Cambridge  University) 
Grace  Medes,  1915  (Bryn  Mawr  College; 

later,  Lankenau  Hospital  Research  Center, 

Philadelphia) 
Seth  E.  Meek,  1909 

(Field  'Museum  of  Natural  History, 

Chicago) 
Charles  W.  Merriam,  1932 

(University  of  California;  later, 

Cornell  University  and  U.  S.  Geological 

Survey) 
Harry  M.  Miller,  Jr.,  1924-1926,  1928 

(Washington  University;  later, 

Rockefeller  Foundation,  Paris) 
Sergius  Morgulis,  1923-1924 

(Creighton  University) 
Charles  E.  Moritz,  1935 

(University  of  California;  later,  Redlands 

College  and  Philip  Morris  and  Company) 
Theodor  Mortensen,  1916 

(University  of  Copenhagen) 
Paul  A.  Nicoll,  1932,  1934-1935,  1937 

(Washington  University;  later, 

Indiana  School  of  Medicine) 
Raymond  C.  Osburn,  1908,  1914 

(New  York  Aquarium;  later, 

Ohio  State  University) 
Fernandus  Payne,  1932,  1937 

(University  of  Indiana) 


Arthur  S.  Pearse,  1927,  1930 

(Duke  University) 
Henry  F.  Perkins,  1903-1905 

(University  of  Vermont) 
Alexander  Hamilton  Phillips,  1915 

(Princeton  University) 
Robert  F.  Pitts,  1935  ' 

(New  York  University;  later, 

Cornell  University) 
Harold  H.  Plough,  1935-1937 

(Amherst  College) 
Frank  M.  Potts,  1913-1915,  1920,  1922 

(Cambridge  University) 
Philip  B.  A.  Powers,  1932,  1936 

(University  of  Pennsylvania) 
Henry  S.  Pratt,  1909-1910,  1924 

(Haverford  College) 
Jacob  E.  Reighard,  1905,  1907,  1909 

(University  of  Michigan) 
Edwin  E.  Reinke,  Research  Associate 

1911-1915  (Vanderbilt  University) 
Oscar  W.  Richards,  1933,  1935 

(Yale  University;  later,  American  Optical 

Company,  Southbridge,  Massachusetts) 
Gordon  A.  Riley,  1937  (Yale  University; 

later,  Bingham  Oceanographic  Laboratory, 

Yale  University) 
Asa  A.  Schaeffer,  Research  Associate 

1919,  1921-1927,  1929 

(University  of  Tennessee;  later, 

Temple  University) 
Waldo  L.  Schmitt,  1924,  1929-1931 

(U.  S.  National  Museum) 
William  A.  Setchell,  1920,  1922-1923 

(University  of  California) 
Eugene  W.  Shaw,  1915 

(U.  S.  Geological  Survey) 
Clarence  R.  Shoemaker,  1925 

(U.  S.  National  Museum) 
Charles  F.  Silvester,  1915 

(Princeton  University;  later, 

Captain,  U.  S.  Army) 
H.  G.  Smith,  1933  (University  of  Bristol) 
Frederick  C.  Steward,  1932-1934,  1936 

(University  of  London;  later, 

Cornell  University) 
Charles  R.  Stockard,  1907-1910 

(Cornell  University  Medical  College) 
Raymond  G.  Stone,  1930-1931,  1934 

(University  of  Kansas  City) 
Frank  A.  Stromsten,  1907-1910 

(University  of  Iowa) 
Geoffrey  Tandy,  1930,  1932 

(British  Museum  of  Natural  History) 
Vance  Tartar,  1937  (Yale  University) 


REPORT     OF    THE     PRESIDENT 


99 


Shiro  Tashiro,  1914-1915 

(University  of  Chicago;  later, 

University  of  Cincinnati) 
Charles  V.  Taylor,  1924-1925 

(University  of  California;  later, 

Stanford  University) 
William  R.  Taylor,  1924-1925 

(University  of  Pennsylvania) 
David  M.  Tennent,  1936  (Yale  University; 

later,  Merck  Institute  for  Therapeutic 

Research  and  Hess  and  Clark  Division  of 

Richardson-Merrell,  Inc.) 
Harry  Beal  Torrey,  Research  Associate 

1926-1927  (Cornell  University  Medical 

School;  later,  Stanford  University) 
Aaron  L.  Treadwell, 

1904,  1909-1910,  1913-1916,  1918, 

1920-1921  (Vassar  College) 
Joseph  M.  Valentine,  1925 

(Yale  University;  later, 

Alabama  Museum  of  Natural  History) 
Gilbert  Van  Ingen,  1915 

(Princeton  University) 
T.  Wayland  Vaughan,  Research  Associate 

1908-1917,  1919,  1922-1923 

(U.  S.  Geological  Survey;  later, 

Scripps  Institution  of  Oceanography) 
J.  Paul  Visscher,  1929-1930 

(Western  Reserve  University) 


W.  Seward  Wallace,  1908 

(University  of  Nevada) 
John  C.  Waller,  1915 

(King's  College,  Cambridge  University) 
William  B.  Wartman,  1928 

(University  of  Pennsylvania  Medical 

School;  later,  Northwestern  University) 
John  B.  Watson,  1907,  1909-1915 

(University  of  Chicago;  later, 

William  Esty  and  Company,  New  York) 
John  W.  Wells,  1931  (Cornell  University) 
Roger  C.  Wells,  1919 

(U.  S.  Geological  Survey) 
E.  I.  Werber,  1915  (Yale  University) 
Douglas  M.  Whitaker,  1925 

(Stanford  University) 
J.  L.  Williams,  1931 

(University  of  California) 
Benjamin  H.  Willier,  1935 

(University  of  Rochester;  later, 

Johns  Hopkins  University) 
Henry  V.  Wilson,  1924 

(University  of  North  Carolina) 
J.  M.  Wilson,  1932-1933 

(Medical  College  of  South  Carolina) 
C.  M.  Yonge,  1933  (University  of  Bristol) 
Charles  Zeleny,  1906-1909 

(University  of  Indiana;  later, 

University  of  Illinois) 


DEPARTMENT   OF   MERIDIAN   ASTROMETRY 

1906-1936 

COMMITTEE   ON   MERIDIAN   ASTROMETRY 

1936-1938 

DUDLEY   OBSERVATORY 

Albany,  New  York 

Directors 

Lewis  Boss,  1905-October  5,  1912 
Benjamin  Boss,  1912-1936;  Chairman,  Committee  on  Meridian  Astrometry,  1936-1938 


Staff  Members 

Sebastian  Albrecht,  1913-1937  Harry  Raymond,  1905-1940 

Heroy  Jenkins,  1909-1937  Arthur  J.  Roy,  1903-1936 

William  B.  Varnum,  1903-1936 


100 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


NUTRITION   LABORATORY 

Organized  in  1907,  opened  in  1908.  Activities  terminated  January  1,  1946. 

Directors 

Francis  G.  Benedict,  1907-1937 
Thorne  M.  Carpenter,  1938-1942  (Acting);  1943-1945 


V.  Ooropatchinsky,  1923-1946 


Staff  Members 

Robert  C.  Lee,  1929-1944 


Harold  L.  Higgins,  1908-1915  Walter  R.  Miles,  1914-1922 

H.  Monmouth  Smith,  1913-1920 


Other  Scientists  and  Scholars  Associated  with  the  Department 


Henry  P.  Armsby,  1919-1920 

(Pennsylvania  State  College) 
James  E.  Ash,  1915 

(Harvard  University  Medical  School; 

later,  Army  Medical  Museum) 
Cornelia  Golay  Benedict, 

1918-1920,  1923,  1925-1926,  1929 
Edward  H.  Bensley,  1935 

(Montreal  General  Hospital) 
C.  C.  Benson,  1912,  1928-1929 

(University  of  Toronto) 
Alice  F.  Blood,  1917-1918 

(Simmons  College) 
Samuel  Brody,  1927  (University  of  Missouri) 
Ernest  W.  Brown,  1911 

(U.  S.  Navy  Medical  Corps) 
John  M.  Bruhn,  1932-1934 

(Yale  Anthropoid  Experiment  Station, 

Orange  Park,  Florida;  later, 

University  of  Alabama  School  of  Medicine) 
M.  Lucien  Bull,  1914  (Institut  Marey,  Paris) 
Walter  G.  Cady,  1912-1913 

(Wesleyan  University) 
E.  P.  Cathcart,  Research  Associate 

1912-1914  (University  of  Glasgow) 
Elizabeth  E.  Crofts,  1924 

(Mount  Holyoke  College) 
G.  H.  de  Paula  Souza,  1920  (Sao  Paulo,  Brazil) 
David  B.  Dill,  1935  (Harvard  University) 
Raymond  Dodge,  1912-1913 

(Wesleyan  University;  later, 

Yale  University) 
Eugene  F.  Du  Bois, 

1915,  1921,  1925-1927,  1930 

(Russell  Sage  Institute  of  Pathology;  later, 

Cornell  University  Medical  College) 


David  L.  Edsall,  1912 

(Washington  University  Medical  School; 

later,  Harvard  University) 
H.  T.  Edwards,  1935  (Harvard  University) 
W.  Falta,  1909 

(First  Medical  Clinic,  Vienna) 
Gertrude  A.  Farr,  1925-1929 

(University  of  New  Hampshire) 
John  M.  Fuller,  1925-1926 

(New  Hampshire  Agricultural  Experiment 

Station) 
James  L.  Gamble,  1913 

(Harvard  University  Medical  School) 
H.  S.  D.  Garven,  1927-1932 

(Moukden  Medical  College,  Manchuria) 
Florence  Gustafson,  1925-1927 

(Wellesley  College) 
Tom  S.  Hamilton,  1925  (University  of 

Illinois  Agricultural  Experiment  Station) 
C.  S.  Hicks,  1927-1930 

(University  of  Adelaide,  South  Australia) 
Fred  A.  Hitchcock,  1932 

(Ohio  State  University) 
John  Homans,  1910-1912 

(Harvard  University  Medical  School) 
Roy  G.  Hoskins,  1933 

(Harvard  University  Medical  School; 

later,  Tufts  College) 
Elliott  P.  Joslin,  1909-1925,  1930,  1941-1943 

(New  England  Deaconess  Hospital,  Boston; 

later,  Harvard  University  Medical  School) 
Howard  T.  Karsner,  1914-1915 

(Harvard  University  Medical  School;  later, 

Bureau  of  Medicine  and  Surgery, 

Navy  Department) 


REPORT     OF    THE     PRESIDENT 


101 


Leslie  G.  Kilborn,  1927-1932 

(West  China  Union  University;  later, 

University  of  Hong  Kong) 
Zing  Yang  Kuo,  1938  (Hangchow,  China) 
Walter  Landauer,  1931 

(Storrs  Agricultural  Experiment  Station) 
Milton  0.  Lee,  1935-1936 

(Harvard  University  Medical  School) 
Helge  Lundholm,  1929 

(McLean  Hospital,  Waverley, 

Massachusetts;  later,  Duke  University) 
Grace  MacLeod,  1922-1927 

(Teachers  College,  Columbia  University) 
Eleanor  D.  Mason,  1927-1933 

(Women's  Christian  College,  Madras) 
James  H.  Means,  1913-1915 

(Massachusetts  General  Hospital,  Boston; 

later,  Massachusetts  Institute  of 

Technology) 
Mary  Henderson  Meyer,  1931 

(Massachusetts  Home,  Boston) 
Carey  D.  Miller,  1928-1935 

(University  of  Hawaii  Experiment  Station) 
Sergius  Morgulis,  1913  (Creigh ton  University; 

later,  University  of  Nebraska  College  of 

Medicine) 
John  R.  Murlin,  1909 

(Cornell  University  Medical  College;  later, 

University  of  Rochester  College  of  Medicine) 
Hans  Murschhauser,  Research  Associate 

1914  (Diisseldorf,  Germany) 
Julius  Nitzulescu,  1928 

(Faculty  of  Medicine,  Jassy,  Roumania) 
Francis  W.  Peabody,  1915 

(Peter  Bent  Brigham  Hospital,  Boston) 
Josef  M.  Petrik,  1929 

(Masaryk  University,  Brno, 

Czechoslovakia) 
Joseph  H.  Pratt,  1911-1913 

(New  England  Medical  Center,  Boston) 

E.  G.  Ritzman,  Research  Associate,  1933-1939 
(University  of  New  Hampshire) 

F.  W.  Rolph,  1918  (University  of  Toronto) 
Howard  F.  Root,  1921-1927,  1930, 

1933-1934,  1936,  1939-1943 

(New  England  Deaconess  Hospital,  Boston) 


Paul  Roth,  1911-1914,  1917-1921,  1923 

(Battle  Creek  Sanitarium,  Michigan) 
George  C.  Shattuck,  1929-1930 

(Harvard  University  Medical  School) 
Henry  C.  Sherman,  1934-1936 

(Columbia  University) 
Hazeltine  L.  Stedman-Parmenter,  1925-1927 

(Mount  Holyoke  College) 
Nils  Stenstrom,  1920  (Stockholm,  Sweden) 

F.  Strieck,  1928 

(University  of  Wiirzburg,  Germany) 
Fritz  B.  Talbot,  1911-1922,  1924 

(Harvard  University  Medical  School) 
Carl  Tigerstedt,  Research  Associate,  1913-1914 

(University  of  Helsingfors) 
Harry  C.  Trimble,  1939-1940 

(Harvard  University  Medical  School) 
Abby  H.  Turner,  1924,  1927-1929 

(Mount  Holyoke  College) 
E.  C.  van  Leersum,  1920 

(Institute  for  Human  Nutrition, 

Amsterdam) 
H.  S.  Halcro  Wardlaw,  1931  (Australia) 
Laurence  G.  Wesson,  1938 

(Veader  Leonard  Laboratory  of 

Experimental  Therapeutics,  Baltimore; 

later,  Massachusetts  Institute  of 

Technology) 
Paul  Dudley  White,  1937 

(Massachusetts  General  Hospital,  Boston) 
Priscilla  White,  1936,  1939 

(New  England  Deaconess  Hospital,  Boston) 
John  C.  Whitehorn,  1929 

(McLean  Hospital,  Waverley, 

Massachusetts;  later,  Johns  Hopkins 

Hospital) 
Francis  H.  Williams,  1912 

(Boston  City  Hospital) 

G.  D.  Williams,  1926-1927 
(Washington  University  Medical  School) 

Stanley  D.  Wilson,  1930-1935 

(Yenching  University,  Peiping) 
Robert  M.  Yerkes,  1932-1934 

(Yale  Anthropoid  Experiment  Station, 

Orange  Park,  Florida) 


FELLOWS   OF   THE   CARNEGIE   INSTITUTION 
OF   WASHINGTON 


Office  of  Administration 
Horace  B.  Barlow,  1961  (King's  College,  Cambridge  University) 


102 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Department  of  Plant  Biology 


Herbert  G.  Baker,  1948-1949 

(University  of  California) 
Shao-lin  Chen,  1949-1950 

(Red  Star  Yeast  Company) 
Edwin  A.  Davis,  1949-1950 

(U.  S.  Department  of  Agriculture) 
L.  N.  M.  Duysens,  1952-1953 

(University  of  Leiden) 
Fulton  J.  F.  Fisher,  1956-1957 

(University  of  Melbourne) 
Joop  C.  Goedheer,  1957-1958 

(University  of  Utrecht) 
Bessel  Kok,  1951-1952 

(Research  Institute  for  Advanced  Studies, 

Baltimore) 


Paul  H.  Latimer,  1956-1957 

(Auburn  University) 
Josef  E.  Loeffler,  1954-1955 

(Shell  Development  Company) 
Fergus  D.  H.  Macdowall,  1947-1949 

(Canadian  National  Research  Council) 
Guy  C.  McLeod,  1959-1960 

(SIAS  Institute,  Brooks  Hospital, 

Brookline,  Massachusetts) 
Ruth  Sager,  1961  (Columbia  University) 
Jerome  A.  Schiff,  1962  (Brandeis  University) 
Kazuo  Shibata,  1956 

(Tokugawa  Institute  for  Biological 

Research) 
Hemming  I.  Virgin,  1954 

(University  of  Gothenburg) 


Mount  Wilson  and  Palomar  Observatories 


M.  K.  Vainu  Bappu,  1951-1952 

(Astrophysical  Observatory,  Kodaikanal, 

India) 
Geoffrey  R.  Burbidge,  1955-1957 

(University  of  California,  La  Jolla) 
William  A.  Buscombe,  1950-1952 

(Mount  Stromlo  Observatory, 

Australian  National  University, 

Canberra,  Australia) 
Edward  R.  Dyer,  Jr.,  1949-1950 

(National  Academy  of  Sciences) 
Roger  F.  Griffin,  1960-1961 

(St.  John's  College,  Cambridge  University) 
Colin  S.  Gum,  1959-1960 

(Radiophysics  Laborato^,  Commonwealth 

Scientific  and  Industrial  Research 

Organization,  Sydney,  Australia) 
Karl  G.  Henize,  1955-1957 

(Dearborn  Observatory,  Northwestern 

University) 


Leo  Houziaux,  1960-1962 

(Institut  d'Astrophysique, 

University  of  Liege) 
Thomas  A.  Matthews,  1956-1958 

(California  Institute  of  Technology) 
Charles  Robert  O'Dell,  1962— 

(Mount  Wilson  and  Palomar  Observatories) 
George  W.  Preston,  III,  1959-1961 

(Lick  Observatory,  Mount  Hamilton) 
Alexander  W.  Rodgers,  1959-1960 

(Mount  Stromlo  Observatory,  Australian 

National  University,  Canberra,  Australia) 
John  B.  Rogerson,  Jr.,  1954-1956 

(Princeton  University  Observatory) 
Stewart  L.  Sharpless,  1952-1953 

(U.  S.  Naval  Observatory) 
Carlos  M.  Varsavsky,  1959 

(Comision  de  Astrofisica  y  Radioastronomia, 

Buenos  Aires) 
Merle  F.  Walker,  1952-1954 

(Lick  Observatory,  Mount  Hamilton) 


Department  of  Terrestrial  Magnetism 


Arthur  I.  Aronson,  1959-1960 

(Purdue  University) 
Toshi  Asada,  1960-1962 

(Geophysical  Institute,  Tokyo) 
Manuel  N.  Bass,  1958-1959 

(Northwestern  University) 
Prabhat  K.  Bhattacharya,  1948-1950 

(California  Institute  of  Technology) 
Louis  Brown,  1961 —  (University  of  Basel) 


Mateo  Casaverde,  1948 

(Instituto  Geofisico  del  Peru) 
William  Compston,  1958 

(Australian  National  University) 
E.  H.  Creaser,  1955-1956 

(University  of  Cambridge) 
J.  D.  Duerksen,  1959-1960 

(National  Institute  for  Medical  Research, 

London) 


REPORT     OF    THE     PRESIDENT 


103 


William  C.  Erickson,  1956-1957 

(Leiden  Observatory) 
Gonzalo  Fernandez,  1948-1949 

(Instituto  Geofisico  del  Peru) 
George  B.  Field,  1953 

(Princeton  University  Observatory) 
J.  W.  Findlay,  1952 

(National  Radio  Astronomy  Observatory, 

Green  Bank) 
Kenneth  L.  Franklin,  1954-1956 

(Hayden  Planetarium) 
John  W.  Graham,  1947-1949 

(Woods  Hole  Oceanographic  Institution) 
Ronald  Green,  1961-1962 

(University  of  Tasmania,  Hobart) 
Richard  Hall,  1962 —  (Indiana  University) 
Pembroke  Jones  Hart,  1952-1954 

(National  Science  Foundation) 
H.  Lawrence  Heifer,  1953-1957 

(University  of  Rochester) 
Ellis  S.  Kempner,  1958 

(National  Institutes  of  Health) 
John  J.  Leahy,  1956-1957 

(City  of  Hope  Hospital,  California) 
Howard  M.  Lenhoff,  1958 

(Howard  Hughes  Medical  Institute,  Miami) 
Soren  Lovtrup,  1951-1952 

(Carlsberg  Laboratories,  Copenhagen) 
John  E.  Midgley,  1960-1962 

(Oxford  University) 
Thomas  Murphy,  1947-1948 

(National  University  of  Ireland) 


Jatinder  Nath  Nanda,  1949-1951 

(Indian  Naval  Physical  Laboratory, 

New  Delhi) 
Leif  Owren,  1953-1954 

(Geophysical  Institute,  College,  Alaska) 
W.  D.  Parkinson,  1947-1948 

(Bureau  of  Mineral  Resources, 

Melbourne,  Australia) 
Gerald  C.  Phillips,  1950-1952  (Rice  Institute) 
George  F.  Pieper,  1956-1957 

(Applied  Physics  Laboratory, 

Johns  Hopkins  University) 
Hector  Rojas,  1961-1962 

(Pan  American  College  Observatory, 

Edinburgh,  Texas) 
Hermann  Rudin,  1962 —  (University  of  Basel) 
Jorma  J.  Ruhimas,  1959-1960 

(University  of  Helsinki) 
George  C.  Sponsler,  1950 

(U.  S.  Department  of  the  Navy) 
M.  Sugiura,  1955 

(Geophysical  Institute,  College,  Alaska) 
Harold  Weaver,  1956-1957 

(Lick  Observatory,  Mount  Hamilton) 
James  A.  Weinman,  1958-1960 

(University  of  Wisconsin) 
Dexter  Whitehead,  1947-1948 

(University  of  Virginia) 
Francis  Waverly  Wood,  1949-1951 

(Bureau  of  Mineral  Resources, 

Melbourne,  Australia) 


Department  of  Genetics 


Guiseppe  Bertani,  1948-1949 

(Karolinska  Institutet,  Stockholm) 
Katherine  S.  Brehme  (Warren),  1939-1941 

(National  Institutes  of  Health) 
Hugh  J.  Cairns,  1960-1961 

(Australian  National  University,  Canberra) 
H.  Clark  Dalton,  1948-1950 

(Washington  Square  College, 

New  York  University) 
Berthe  Delaporte,  1948-1949 

(Ecole  Pratique  des  Hautes  Etudes,  Paris) 
A.  H.  Doermann,  1947-1949 

(Vanderbilt  University) 
Kazuo  Hashimoto,  1957-1958 

(Keio  University  School  of  Medicine, 

Tokyo) 
Etta  Kiifer  (Boothroyd),  1956-1957 

(McGill  University) 


Joseph  D.  Mandell,  1955-1957 

(Palo  Alto  Medical  Research  Foundation) 
Hermann  Moser,  1953-1956 

(Frances  Delafield  Hospital, 

New  York  City) 
Frank  H.  Mukai,  1959 

(Biological  Laboratory, 

Long  Island  Biological  Association) 
Kenneth  Paigen,  1950-1952 

(Roswell  Park  Memorial  Institute,  Buffalo) 
Catherine  Roesel,  1950-1951 

(University  of  Georgia  School  of  Medicine) 
Janine  Sechaud,  1960  (University  of  Oregon) 
Atif  Sengiin,  1957 

(University  of  Istanbul,  Turkey) 
Robert  C.  von  Borstel,  1952-1953 

(Oak  Ridge  National  Laboratory) 


104 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Geophysical  Laboratory 


*v- 


Ralph  Arnold,  1956-1959 

(Princeton  University;  later, 

Saskatchewan  Research  Council, 

University  of  Saskatchewan) 
D.  Kenneth  Bailey,  1962— 

(Trinity  College,  Dublin,  Ireland) 
Hubert  L.  Barnes,  1956-1959 

(Columbia  University;  later, 

Pennsylvania  State  University) 
Robin  Brett,  1961 —  (Department  of 

Geological  Sciences,  Harvard  University) 
Charles  W.  Burnham,  1961 — 

(Massachusetts  Institute  of  Technology) 
Peter  R.  Buseck,  1961 — 

(Department  of  Geology,  Columbia 

University) 
G.  A.  Chinner,  1958-1960 

(University  of  Cambridge) 
John  de  Neufville,  1961-1962 

(Yale  University;  later, 

Harvard  University) 
Bruce  R.  Doe,  1960-1962 

(California  Institute  of  Technology;  later, 

U.  S.  Geological  Survey) 
W.  Gary  Ernst,  1955-1958 

(Johns  Hopkins  University;  later, 

University  of  California,  Los  Angeles) 
Jeff  J.  Fawcett,  1961 — 

(University  of  Manchester) 

B.  Halferdahl,  1954-1958 

(Johns  Hopkins  University;  later, 

Research  Council  of  Alberta, 

Edmonton,  Alberta,  Canada) 
Kai  Hytonen,  1959-1961 

(University  of  Helsinki;  later, 

Geological  Survey  of  Finland,  Otaniemi) 
Mackenzie  L.  Keith,  1947-1950 

(Pennsylvania  State  University) 
Donald  H.  Lindsley,  1960-1962 

(Johns  Hopkins  University) 
Giinter  Moh,  1962  (Heidelberg  University) 
Nobuo  Morimoto,  1957-1959,  1962 

(Mineralogical  Institute, 

University  of  Tokyo) 
Kaarlo  J.  Neuvonen,  1948-1950 

(Geological  Survey  of  Finland;  later, 

University  of  Turku,  Finland) 


Louis  Otto  Nicolaysen,  1951-1954 

(Massachusetts  Institute  of  Technology; 

later,  Bernard  Price  Institute  of 

Geophysical  Research,  Johannesburg, 

South  Africa) 
Philip  M.  Orville,  1957-1958 

(Yale  University;  later,  Cornell  University) 
Edwin  W.  Roedder,  1947-1948 

(Columbia  University;  later, 

U.  S.  Geological  Survey) 
Eugene  H.  Roseboom,  1956-1959 

(Harvard  University;  later, 

U.  S.  Geological  Survey) 
Bruno  Sabels,  1962  (University  of  Nevada) 
Th.  G.  Sahama,  1947-1949 

(University  of  Helsinki) 
Werner  F.  Schreyer,  1958-1959,  1962— 

(University  of  Kiel) 
James  R.  Smith,  1954-1957 

(Princeton  University;  later, 

Saskatchewan  Research  Council, 

University  of  Saskatchewan) 
Joseph  Victor  Smith,  1951-1954 

(Cavendish  Laboratory,  University  of 

Cambridge;  later,  Pennsylvania  State 

University  and  University  of  Chicago) 
Yoshio  Suzuki,  1960-1962 

(Hakkaido  University,  Japan;  later, 

Geological  Survey  of  Japan) 
Per-Fredrick  Troften,  1960 

(Norwegian  Geological  Survey;  later, 

Geof  ysisk  Malmleting,  Trondheim,  Norway) 
Allan  C.  Turnock,  1958-1960 

(University  of  Manitoba;  later, 

Department  of  Mines  and  Technical 

Surveys,  Ottawa) 
J.  R.  Vallentyne,  1956-1957 

(Queen's  University,  Ontario;  later, 

Cornell  University) 
Bruce  Velde,  1962— 

(Montana  State  University) 
David  R.  Wones,  1957-1959 

(Massachusetts  Institute  of  Technology; 

later,  U.  S.  Geological  Survey) 
Kenzo  Yagi,  1950-1951,  1960-1961 

(Tohoko  University,  Japan) 
Richard  A.  Yund,  1959-1961 

(University  of  Illinois;  later, 

Brown  University) 


REPORT     OF    THE     PRESIDENT 


105 


Department  of  Embryology 


Michael  Abercrombie,  1962 

(University  College,  London) 
Vittorio  Danesino,  1953-1954 

(University  of  Naples) 
L.  E.  DeLanney,  1957  (Wabash  College) 
Christine  Gilbert,  1950-1951 

(University  of  the  Witwatersrand) 
Perry  W.  Gilbert,  1949-1950 

(Cornell  University) 
E.  Clark  Gillespie,  1948 

(Johns  Hopkins  University;  later, 

University  of  Arkansas) 
Richard  J.  Goss,  1960-1961 

(Brown  University) 
Jerome  S.  Harris,  1948-1949 

(Johns  Hopkins  University;  later, 

private  practice  in  obstetrics  in  Denver) 
Beni  Horvath,  1952-1953 

(Columbia  University;  later, 

National  Institutes  of  Health) 
Yoshihiro  Kato,  1959-1961 

(Tokyo  University;  later, 

University  of  Nagoya) 


Efstathios  J.  Kokrikos,  1953-1954 

(Red  Cross  Hospital,  Athens) 
Ben  C.  Moffett,  Jr.,  1954 

(University  of  Alabama;  later, 

Armed  Forces  Institute  of  Pathology) 
Brenda  Schofield,  1953-1954 

(Oxford  University) 
E.  Carl  Sensenig,  1945 — 

(University  of  Alabama) 
Peter  H.  S.  Silver,  1961-1962 

(Middlesex  Hospital  Medical  School, 

London) 
Malcolm  S.  Steinberg,  1956-1958 

(Johns  Hopkins  University) 
Ikuo  Takeuchi,  1959-1961 

(Princeton  University;  later, 

University  of  Osaka) 
L.  J.  Wells,  1948  (University  of  Minnesota) 
Douglas  R.  Wilkie,  1955 

(University  of  London) 
Ian  B.  Wilson,  1961-1962 

(University  College  of  North  Wales) 
Fred  H.  Wilt,  1958-1960  (Purdue  University) 


Department  of  Archaeology 


Robert  H.  Barlow,  1949-1950 

(Mexico  City  College) 
Heinrich  Berlin,  1952-1955 

(Instituto  de  Antropologia  e  Historia  de 

Guatemala) 


Joseph  A.  Hester,  Jr.,  1952-1954 

(University  of  California,  Los  Angeles) 

William  T.  Sanders,  1954-1955 
(Pennsylvania  State  University) 

Raymond  H.  Thompson,  1950-1952 
(University  of  Arizona) 


GRANTEES  AND   OTHERS  AFFILIATED    WITH   THE 

CARNEGIE  INSTITUTION   BUT   NOT   WITH 

PARTICULAR   DEPARTMENTS 


Chemistry 


Solomon  F.  Acree,  1904-1913 

(Johns  Hopkins  University;  later, 

National  Bureau  of  Standards) 
Charles  Baskerville,  1903-1905 

(College  of  the  City  of  New  York) 
Gregory  P.  Baxter,  Research  Associate 

1904-1914,  1924  (Harvard  University) 
Gustavus  E.  Behr,  1906  (Llarvard  University) 
Amos  P.  Brown,  1904-1908 

(University  of  Pennsylvania) 


Paul  B.  Davis,  Research  Associate,  1916-1917 
(Johns  Hopkins  University;  later, 
Davison  Chemical  Corporation,  Baltimore) 

Louis  M.  Dennis,  1903  (Cornell  University) 

Howard  W.  Doughty,  1904 

(Johns  Hopkins  University;  later, 
Amherst  College) 

George  S.  Forbes,  1906  (Harvard  University) 

Joseph  C.  W.  Frazer,  1916-1918 
(Johns  Hopkins  University) 


106 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Moses  Gomberg,  1904-1905 

(University  of  Michigan) 
Harry  C.  Jones,  1903-1916 

(Johns  Hopkins  University) 
George  B.  Kistiakowsky,  Research  Associate 

1942  (Harvard  University) 
Philip  A.  Leigh  ton,  Research  Associate 

1934-1935  (Stanford  University) 
Harmon  N.  Morse,  1902-1918 

(Johns  Hopkins  University) 
Arthur  A.  Noyes,  Research  Associate 

1903-1930  (California  Institute  of 

Technology) 
I.  I.  Rabi,  Research  Associate,  1934-1935 

(Columbia  University) 
Ira  Remsen,  1902,  1913,  1917 

(Johns  Hopkins  University) 


Theodore  W.  Richards,  Research  Associate 

1902-1928  (Harvard  University) 
Edgar  Fahs  Smith,  Research  Associate 

1902,  1909,  1916-1918,  1920-1922 

(University  of  Pennsylvania) 
Julius  Stieglitz,  1909  (University  of  Chicago) 
Wilfred  N.  Stull,  1903  (Harvard  University) 
James  B.  Sumner,  Research  Associate  in 

Biochemistry,  1931-1932 

(Cornell  University) 
John  Bishop  Tingle,  1903-1905 

(Johns  Hopkins  University) 
Harold  C.  Urey,  Research  Associate,  1934-1935 

(Columbia  University;  later, 

University  of  Chicago) 
Hobart  H.  Willard,  1910  (Harvard  University) 
Edgar  B.  Wilson,  Research  Associate 

1936-1937  (Harvard  University) 


Physics 


Joseph  S.  Ames,  1904-1905 

(Johns  Hopkins  University) 
Carl  D.  Anderson,  1942-1943 

(California  Institute  of  Technology) 
G.  F.  Barker,  1904  (Washington,  D.  C.) 
Samuel  J.  Barnett,  Research  Associate 

1904-1905  (Stanford  University;  later, 

University  of  California,  Los  Angeles,  and 

California  Institute  of  Technology) 
Ralph  D.  Bennett,  Research  Associate 

1932-1933  (Massachusetts  Institute 

of  Technology;  later,  Naval  Ordnance 

Laboratory) 
Charles  F.  Burgess,  1904-1908 

(University  of  Wisconsin) 
William  Campbell,  1904-1905 

(Columbia  University) 
Henry  S.  Carhart,  1904-1905 

(University  of  Michigan) 
Clement  D.  Child,  1903-1904 

(Colgate  University) 
William  W.  Coblentz,  1903-1908,  1911 

(National  Bureau  of  Standards) 
Henry  Crew,  1902-1904 

(Northwestern  University) 
Paul  S.  Epstein,  1937-1939 

(California  Institute  of  Technology) 
J.  A.  Folse,  1926 

(Rosenwald  Industrial  Museum,  Chicago) 
William  S.  Franklin,  1906 

(Lehigh  University;  later, 

Massachusetts  Institute  of  Technology) 


L.  A.  Freudenberger,  1906  (Delaware  College) 
Robert  H.  Goddard,  1929-1930 

(Clark  University) 
John  F.  Hay  ford,  Research  Associate 

1911-1913,  1915-1917,  1919-1925 

(Northwestern  University) 
Henry  M.  Howe, 

1906-1911,  1913-1914,  1916-1920 

(Columbia  University) 
H.  Victor  Neher,  1943 

(California  Institute  of  Technology) 
Edward  L.  Nichols,  Research  Associate 

1905-1906,  1908-1918,  1920-1925 

(Cornell  University) 
Francis  E.  Nipher,  1914 

(Washington  University) 
Gennady  W.  Potapenko,  1937-1939 

(California  Institute  of  Technology) 
Allen  G.  Shenstone,  Research  Associate 

1931-1933  (Princeton  University) 
William  W.  Strong,  1908-1911 

(Johns  Hopkins  University;  later, 

Scientific  Instrument  and  Electrical 

Machine  Company) 
Horace  S.  Uhler,  1905  (Johns  Hopkins 

University;  later,  Yale  University) 
John  B.  Whitehead,  1903-1905 

(Johns  Hopkins  University) 
Robert  W.  Wood,  1902-1904 

(Johns  Hopkins  University) 
Albert  F.  Zahm,  1905 

(Catholic  University  of  America;  later, 

Library  of  Congress) 


REPORT     OF    THE     PRESIDENT 


107 


Mathematics 


Arthur  B.  Coble,  1903-1904 

(University  of  Missouri;  later, 

University  of  Illinois) 
Floyd  F.  Decker,  1910  (Syracuse  University) 
Leonard  E.  Dickson,  Research  Associate 

1904,  1912,  1919,  1922,  1927-1928 

(University  of  Chicago) 
George  W.  Hill,  1905-1907 

(West  Nyack,  New  York) 
John  Holland,  Research  Associate,  1960-1961 

(University  of  Michigan) 
Derrick  N.  Lehmer,  Research  Associate 

1904-1909,  1911-1912,  1925-1928, 

1931-1932,  1936  (University  of  California) 
Arthur  C.  Lunn,  1909  (University  of  Chicago) 
William  D.  MacMillan,  1909 

(University  of  Chicago) 


Eliakim  H.  Moore,  1902 

(University  of  Wisconsin;  later, 

University  of  Chicago) 
Frank  Morley,  Research  Associate 

1902,  1908,  1910-1918,  1920-1921,  1923, 

1926,  1928,  1930-1931,  1933-1936 

(Johns  Hopkins  University) 
James  B.  Shaw,  1907 

(James  Millikin  University;  later, 

University  of  Illinois) 
Henry  W.  Stager,  1911  (Fresno,  California) 
Ormond  Stone,  1902  (Leander  McCormick 

Observatory,  Charlottesville,  Virginia) 
Ernest  J.  Wilczynski,  Research  Associate 

1903-1905  (University  of  California;  later, 

University  of  Chicago) 


Engineering 


William  H.  Burr,  1902 
(Columbia  University) 

William  F.  Durand,  1903-1906 
(Cornell  University;  later, 
Stanford  University) 

George  Gibbs,  1902 

(Baldwin  Locomotive  Works,  Philadelphia; 
later,  consulting  engineer,  Pennsylvania 
Railroad) 

William  F.  M.  Goss,  1904-1908 
(University  of  Illinois) 


George  S.  Morison,  1902 

(civil  engineer,  New  York  City) 
Harold  Pender,  1902-1903 

(Syracuse  University;  later, 

University  of  Pennsylvania) 
Charles  P.  Steinmetz,  1902 

(General  Electric  Company) 
Robert  H.  Thurston,  1902 

(Cornell  University) 
Leonard  Waldo,  1903  (consulting  engineer  in 

metallurgy  and  electronics, 

Plainfield,  New  Jersey) 


Geography,  Geology,  and  Geophysics 


Cleveland  Abbe,  1902  (U.  S.  Weather  Bureau; 

later,  Johns  Hopkins  University) 
Adalbert  E.  Benfield,  Research  Associate 

1940-1941  (Williams  College;  later, 

Harvard  University) 
Tor  Bergeron,  1951-1957 

(University  of  California;  later, 

Meteorological  Institute,  Uppsala,  Sweden) 
J.  Bjerknes,  1951-1957 

(University  of  California) 
V.  Bjerknes,  Research  Associate  in 

Meteorology,  1906-1948 

(University  of  Oslo) 
Eliot  Blackwelder,  1903-1904 

(University  of  Wisconsin;  later, 

Stanford  University) 


Robert  C.  Bundgaard,  1951-1957 

(U.  S.  Air  Force) 
Ian  Campbell,  Research  Associate,  1933-1939 

(California  Institute  of  Technology) 
Rollin  T.  Chamberlin,  1908 

(University  of  Chicago) 
George  Davidson,  1906-1907 

(University  of  California) 
William  Morris  Davis,  1902,  1925-1926 

(Harvard  University) 
C.  L.  Godske,  1951-1957 

(University  of  Bergen) 
Frank  T.  Gucker,  Jr.,  Research  Associate 

1940-1950  (Northwestern  University; 

later,  Indiana  University) 
Norman  E.  A.  Hinds,  1931,  1933-1935 

(University  of  California,  Berkeley) 


108 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


William  H.  Hobbs,  Research  Associate, 

(University  of  Michigan) 
John  H.  Maxson,  Research  Associate 

1932-1939  (California  Institute  of 

Technology;  later,  Anderson-Pritchard 

Oil  Corporation,  Denver) 
Walter  H.  Newhouse,  Research  Associate 

1939-1945  (Massachusetts  Institute  of 

Technology;  later, 

University  of  Chicago) 


1930      Sverre  Petterssen,  1951-1957 

(University  of  Chicago) 
J.  W.  Sandstrom,  1906-1908 

(University  of  Stockholm) 
Alexander  Silverman,  Research  Associate 

1939-1942  (University  of  Pittsburgh) 
H.  Solberg,  1951-1957  (University  of  Oslo) 
William  Van  Roy  en,  Research  Associate,  1934 

(University  of  Nebraska;  later, 

Brooklyn  College) 


Seismology 


Oscar  S.  Adams,  1925-1926 

(U.  S.  Coast  and  Geodetic  Survey) 
F.  B.  Bassett,  1923  (U.  S.  Navy  Department) 
George  L.  Bean,  1928-1931 

(U.  S.  Coast  and  Geodetic  Survey) 
Hugo  Benioff,  1932-1936 

(California  Institute  of  Technology) 
William  Bowie,  1925-1926 

(U.  S.  Coast  and  Geodetic  Survey) 
Perry  Byerly,  1925-1926 

(University  of  California,  Berkeley) 
Charles  Lewis  Gazin,  1931-1932 

(U.  S.  Geological  Survey;  later, 

U.  S.  National  Museum) 
Herbert  E.  Gregory,  1925-1926 

(Yale  University;  later, 

Bishop  Museum,  Honolulu) 
Beno  Gutenberg,  1930-1931,  1933-1935 

(California  Institute  of  Technology) 
William  Stephen  Webster  Kew,  1922-1923 

(U.  S.  Geological  Survey;  later, 

Standard  Oil  Company  of  California) 
Andrew  C.  Lawson,  1906-1907 

(University  of  California,  Berkeley) 
James  B.  Macelwane,  S.J.,  1924-1925 

(St.  Louis  University) 


Levi  F.  Noble,  1922-1923 

(U.  S.  Geological  Survey) 
Harry  Fielding  Reid,  1906-1907 

(Johns  Hopkins  University) 
Charles  F.  Richter,  1927-1928,  1932-1937 

(California  Institute  of  Technology) 
Arnold  Romberg,  1921-1923 

(University  of  Hawaii;  later, 

University  of  Texas) 
Maple  D.  Shappell,  1930-1934 

(California  Institute  of  Technology) 
Frederick  P.  Vickery,  1922-1923 

(University  of  Southern  California, 

Los  Angeles;  later, 

Sacramento  Junior  College) 
Frank  Wenner,  1922-1923 

(National  Bureau  of  Standards) 
Walter  T.  Whitney,  1913,  1917,  1922-1923 

(California  Institute  of  Technology;  later, 

Pomona  College) 
Bailey  Willis,  Research  Associate 

1903-1907,  1912,  1930,  1934 

(Stanford  University) 
Harry  O.  Wood,  Research  Associate 

1920-1931,  1936-1940 

(California  Institute  of  Technology) 


Physiological  Chemistry 


John  J.  Abel,  1903-1905 

(Johns  Hopkins  University) 
Wilder  D.  Bancroft, 

1902,  1904-1906,  1908-1910 

(Cornell  University) 
Russell  H.  Chittenden,  1904-1907 

(Yale  University) 
Walter  H.  Eddy,  Research  Associate 

1927-1933  (Columbia  University) 


Lafayette  B.  Mendel,  1905-1906,  1927-1930 

(Yale  University) 
Thomas  B.  Osborne,  Research  Associate 

1904-1927  (Connecticut  Agricultural 

Experiment  Station) 
Hubert  B.  Vickery,  Research  Associate 

1922-1937  (Connecticut  Agricultural 

Experiment  Station) 
Robert  R.  Williams,  Research  Associate 

1927-1933  (Bell  Telephone  Laboratories) 


REPORT     OF     THE     PRESIDENT 


109 


Psychology 


John  W.  Baird,  1903-1904 
(Cornell  University;  later, 
Clark  University) 

James  Mark  Baldwin,  1902 
(Princeton  University) 

Clarence  B.  Farrar,  1904-1906 

(Shepperd  and  Enoch  Pratt  Hospital, 

Baltimore;  later, 

Toronto  Psychiatric  Hospital) 


Shephard  I.  Franz,  1903-1911, 

1913,  1915-1917  (St.  Elizabeth's  Hospital, 

Washington,  D.  C.) 
S.  Stanley  Hall,  1903-1904  (Clark  University) 
Peter  Milner,  Research  Associate,  1960-1961 

(McGill  University) 
James  P.  Porter,  1907 

(Clark  University;  later,  Ohio  University) 
Henry  A.  Ruger,  Research  Associate 

1927-1929  (Columbia  University) 


Physiology 


Wilbur  O.  Atwater,  1903-1905 

(Wesleyan  University) 
Henry  P.  Bowditch,  1902 

(Harvard  Medical  School) 
Simon  Flexner,  1902-1903 

(University  of  Pennsylvania;  later, 

Rockefeller  Institute  for  Medical  Research) 
Alexander  Forbes,  1906  (Harvard  University) 
Robert  H.  Gault,  Research  Associate 

1927-1929  (Northwestern  University) 
Charles  C.  Guthrie,  1908 

(Washington  University;  later, 

University  of  Pittsburgh) 
Frank  A.  Hartman,  Research  Associate 

1931-1934  (Ohio  State  University) 
William  H.  Howell,  Research  Associate 

1902, 1933-1934  (Johns  Hopkins  University) 


Leo  Loeb,  1903-1905,  1907-1909 

(University  of  Pennsylvania;  later, 

Washington  University) 
S.  Weir  Mitchell,  1902 

(Philadelphia,  Pennsylvania) 
Aubrey  T.  Mussen,  Research  Associate 

1929-1931  (Johns  Hopkins  University) 
Hideyo  Noguchi,  1903-1908 

(University  of  Pennsylvania) 
Earle  B.  Phelps,  Research  Associate,  1931-1933 

(Columbia  University) 
Edward  T.  Reichert,  1904,  1908-1914 

(University  of  Pennsylvania) 
George  Oscar  Russell,  Research  Associate 

(Ohio  State  University;  later, 

Gallaudet  College) 


Zoology 


Anton  J.  Carlson,  1903-1904 

(Stanford  University;  later, 

University  of  Chicago) 
A.  B.  Clawson,  1907  (University  of  Michigan) 
Henry  E.  Crampton,  Research  Associate 

1902,  1904,  1906,  1908,  1916,  1919, 

1923-1925,  1927-1928,  1930-1933,  1935, 

1939  (Columbia  University) 
Joseph  A.  Cushman,  1912,  1919,  1939-1941 

(Cushman  Laboratory  for  Foraminiferal 

Research,  Sharon,  Massachusetts) 
Bashford  Dean,  1906  (Columbia  University) 
Carl  H.  Eigenmann,  1903-1904 

(Indiana  University) 
Ross  G.  Harrison,  Research  Associate 

1944-1948  (Yale  University) 
Leland  O.  Howard,  1903-1904 

(U.  S.  Department  of  Agriculture) 


Herbert  S.  Jennings,  1902-1905 

(University  of  Michigan;  later, 

Johns  Hopkins  University) 
Charles  A.  Kofoid,  Research  Associate 

1921-1925  (University  of  California) 
Ralph  S.  Lillie,  1904  (University  of  Nebraska; 

later,  University  of  Chicago) 
Joseph  A.  Long,  1911  (Harvard  University) 
Clarence  E.  McClung,  1903-1905 

(University  of  Kansas;  later, 

University  of  Penns}7lvania) 
Hansford  MacCurdy,  1907 

(Harvard  University) 
Edward  L.  Mark,  1906-1910 

(Harvard  University) 
C.  Hart  Merriam,  1902 

(U.  S.  Biological  Survey) 
Albert  P.  Morse,  1903-1905 

(Wellesley,  Massachusetts) 


110 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Henry  Fairfield  Osborn,  1902 

(Columbia  University) 
William  Patten,  1904-1905 

(Dartmouth  College) 
Raymond  Pearl,  1904-1906 

(University  of  Michigan;  later, 

Johns  Hopkins  University  Medical  School) 
John  C.  Phillips,  1911 

(Harvard  University) 


Porter  E.  Sargent,  1904 
(Harvard  University;  later, 
Sargent  School  Service) 

Nettie  M.  Stevens,  1904-1905 
(Bryn  Mawr  College) 

Edmund  B.  Wilson,  1903 
(Columbia  University) 

Naohide  Yatsu,  1905-1906 
(Columbia  University) 


Paleontology 


Ira  S.  Allison,  1939-1944 

(Oregon  State  College) 
Earl  H.  Bell,  Research  Associate,  1934-1935 

(University  of  Nebraska;  later, 

U.  S.  Department  of  Agriculture) 
John  P.  Buwalda,  Research  Associate 

1925-1938  (California  Institute  of 

Technology) 
Frank  M.  Carpenter,  Research  Associate 

1931-1932  (Harvard  University) 
Ermine  C.  Case,  Research  Associate 

1903-1905,  1908-1909,  1911-1912, 

1914-1919,  1921-1922 

(University  of  Michigan) 
Carlton  Condit,  1938,  1944 

(University  of  California;  later, 

Illinois  State  Museum,  Springfield) 
Lyman  H.  Daugherty,  1941 

(San  Jose  State  College) 
Hellmut  De  Terra,  Research  Associate 

1934-1939  (Yale  University;  later, 

Columbia  University) 
A.  L.  Du  Toit,  Research  Associate,  1923 

(Pretoria,  South  Africa) 
Eustace  L.  Furlong, 

1921-1924,  1928,  1931-1933,  1938-1942 

(University  of  California) 
William  K.  Gregory,  1938 

(American  Museum  of  Natural  History; 

later,  Columbia  University) 
Oliver  P.  Hay,  Research  Associate 

1902-1907,  1911-1927 

(American  Museum  of  Natural  History) 
Norman  E.  A.  Hinds,  1936 

(University  of  California) 
Edgar  B.  Howard,  Research  Associate 

1934-1942  (University  of  Pennsylvania) 
Hildegarde  Howard, 

1932,  1938,  1939,  1942,  1946,  1949 

(Los  Angeles  County  Museum) 
Hsen  Hsu  Hu,  1940 

(Fan  Memorial  Institute  of  Biology, 

Peiping) 


Remington  Kellogg,  1925-1942 

(U.  S.  National  Museum) 
Robert  Smith  La  Motte,  1935-1936 

(U.  S.  Forest  Service;  later, 

University  of  California) 
Harry  D.  MacGinitie,  1933,  1937,  1941,  1953 

(Humboldt  State  College) 
Edwin  D.  McKee,  Research  Associate 

1936-1942  (U.  S.  National  Park  Service; 

later,  U.  S.  Geological  Survey) 
Earl  L.  Packard, 

1926,  1928,  1931-1932,  1938-1939, 

1941-1943  (Oregon  State  College) 
Llewellyn  I.  Price,  Research  Associate 

1939-1940  (Harvard  University) 
Malcolm  J.  Rogers,  Research  Associate,  1937 

(San  Diego  Museum) 
Paul  B.  Sears,  1936,  1938 

(University  of  Oklahoma;  later, 

Yale  University  and  Wake  Forest  College, 

Winston-Salem,  North  Carolina) 
Chester  Stock,  Research  Associate,  1925-1943 

(California  Institute  of  Technology) 
Alexander  A.  Stoyanow,  Research  Associate 

1928  (University  of  Arizona;  later, 

University  of  California) 
G.  H.  R.  von  Koenigswald,  Research  Associate 

1936-1938,  1947  (Bandung,  Java) 
David  White,  Research  Associate,  1925-1932 

(National  Academy  of  Sciences) 
Henry  S.  Williams,  1902  (Yale  University) 
Howell  Williams,  1943-1944 

(University  of  California) 
Samuel  W.  Williston,  1903 

(University  of  Chicago) 
Robert  W.  Wilson, 

1933-1934,  1936-1937,  1940,  1942,  1949 

(California  Institute  of  Technology;  later, 

University  of  Colorado  and 

University  of  Kansas) 
Wendell  P.  Woodring,  1925,  1928,  1932 

(California  Institute  of  Technology;  later, 

U.  S.  Geological  Survey) 


REPORT     OF     THE     PRESIDENT 


111 


Archaeology  and  Anthropology 


Marion  E.  Blake,  Research  Associate 

1937-1938,  1940-1941,  1945 

(American  Academy,  Rome) 
Franz  Boas,  1902 

(American  Museum  of  Natural  History; 

later,  Columbia  University) 
William  T.  Brigham,  1906-1912 

(Bernice  Pauahi  Bishop 

Museum,  Honolulu) 
George  A.  Dorsey,  1902 

(Field  Museum  of  Natural  History, 

Chicago) 
Arthur  L.  Frothingham,  1913 

(Princeton  University) 
William  H.  Holmes,  1902-1904 

(U.  S.  National  Museum) 
Walter  W.  Hyde,  Research  Associate 

1919-1920  (University  of  Pennsylvania) 
Waldemar  Jochelson, 

1923-1924,  1926,  1929-1930,  1934-1935 

(American  Museum  of  Natural  History) 
Allan  C.  Johnson,  1910-1911 

(Princeton  University) 


W.  Max  Muller,  1904-1907,  1910-1911 

(University  of  Pennsylvania) 
Raphael  Pumpelly,  1903-1906 

(Newport,  Rhode  Island) 
Hubert  Schmidt,  1903-1904 

(Museum  fur  Volkerkunde,  Berlin) 
George  W.  Scott,  1904-1905,  1911-1914 

(Law  Librarian  of  Congress  and 

Supreme  Court) 
Thomas  D.  Seymour,  1903 

(American  School  of  Classical  Studies, 

Athens) 
Esther  Boise  Van  Deman,  Research  Associate 

1906-1925   (American  School  of  Classical 

Studies,  Rome;  later, 

University  of  Michigan) 
William  Hayes  Ward,  Research  Associate 

1903-1908  (editor,  The  Independent, 

New  York) 
Andrew  F.  West,  1905-1911 

(Princeton  University) 
James  R.  Wheeler,  1905-1912 

(American  School  of  Classical  Studies, 

Athens;  later,  Columbia  University) 


Bibliography 


Cyrus  Adler,  1902  (Smithsonian  Institution) 
J.  McKeen  Cattell,  1902-1904 

(Columbia  University;  later, 

editor,  American  Men  of  Science) 
Wilberforce  Eames,  1906-1908 

(librarian,  Bibliographical  Society  of 

America) 


Fielding  H.  Harrison,  Research  Associate 

1903-1927  (Army  Medical  Museum;  later, 

Surgeon  General's  Office) 
Herbert  Putnam,  1902-1907 

(Library  of  Congress) 
J.  David  Thompson,  1903-1907 

(Library  of  Congress) 


Literature,  Linguistics,  and  Philology 


Manuel  J.  Andrade,  Research  Associate 

1933-1940  (University  of  Chicago) 
John  Pawley  Bate,  1910-1917 

(Inns  of  Court,  London) 
Henry  Bergen,  Research  Associate 

1912-1927,  1933  (Brooklyn,  New  York) 
J.  Leslie  Brierly,  1910-1911 

(Lincoln's  Inn,  and  Trinity  College,  Oxford) 
Morgan  Callaway,  1913 

(Johns  Hopkins  University) 
William  Churchill,  Research  Associate 

1911,  1915-1919,  1921  (Committee  on 

Public  Information,  Washington,  D.  C.) 
Lane  Cooper,  1916  (College  of  St.  James) 
Albert  G.  de  Lapradelle,  1910-1916 

(University  of  Paris) 


Charles  G.  Fenwick,  1910-1916 

(Carnegie  Endowment  for  International 

Peace;  later,  Bryn  Mawr  College) 
Ewald  Flugel,  1904-1908 

(Stanford  University) 
George  D.  Gregory,  1916 

(Carnegie  Endowment  for  International 

Peace) 
John  W.  Hebel,  1917  (Cornell  University) 
George  Hempl,  1904-1905,  1909 

(University  of  Michigan;  later, 

Stanford  University) 
Charles  W.  Hodell,  Research  Associate 

1907-1908  (Goucher  College) 
Thomas  Erskine  Holland,  1910-1911 

(LTniversity  of  Oxford) 


112 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Arthur  G.  Kennedy,  1916-1918,  1920,  1923 

(Stanford  University) 
Henry  C.  Lancaster,  Research  Associate,  1912 

(Johns  Hopkins  University) 
Elias  A.  Lowe,  Research  Associate 

1910-1929,  1936-1940,  1947 

(Institute  for  Advanced  Study, 

Princeton  University) 
John  D.  Maguire,  1909-1913 

(Catholic  University  of  America) 
Ernest  Nys,  1917  (University  of  Brussels) 
Charles  G.  Osgood,  1915 

(Princeton  University) 
James  Brown  Scott,  Research  Associate 

1910-1917  (Department  of  State;  later, 

Carnegie  Endowment  for  International 

Peace) 


E.  W.  Scripture,  1903-1906  (Yale  University) 
H.  Oskar  Sommer,  Research  Associate 

1906-1907,  1909-1912  ("Astolat," 

Camberley,  Surrey,  England) 
Benjamin  F.  Stelter,  1916 

(University  of  Southern  California;  later, 

Occidental  College) 
John  S.  P.  Tatlock,  Research  Associate 

1916-1918, 1920, 1923  (Stanford  University; 

later,  Harvard  University  and  University  of 

California) 
Ludwig  von  Bar,  1910-1916 

(University  of  Gottingen) 
John  Westlake,  1910-1912 

(University  of  Cambridge) 
Herbert  Francis  Wright,  1917 

(Catholic  University  of  America;  later, 

Georgetown  University) 


Political  Science 


N.  Andrew  N.  Cleven,  Research  Associate 

1930-1931  (Duke  University) 
Isaac  J.  Cox,  Research  Associate,  1925-1927 

(Northwestern  University) 
Herman  G.  James,  Research  Associate 

1922-1923  (University  of  Texas;  later, 

President,  Ohio  University) 


Percy  A.  Martin,  Research  Associate, 
1926-1928  (Stanford  University) 

William  W.  Pierson,  Jr.,  Research  Associate 
1927-1928  (University  of  North  Carolina) 

Leo  S.  Rowe,  1904,  1906-1910,  1917 
(University  of  Pennsylvania) 

Graham  H.  Stuart,  1924  (Stanford  University) 


Reports  of  Departments 
and  Special  Studies 


Mount  Wilson  and  Palomar  Observatories 

Geophysical  Laboratory 

Department  of  Terrestrial  Magnetism 

Committee  on  Image  Tubes  for  Telescopes 

Department  of  Plant  Biology 

Department  of  Embryology 

Department  of  Genetics 


Mount  Wilson  and  Palomar 

Observatories 


Operated  by  Carnegie  Institution  of  Washington 
and  California  Institute  of  Technology 

Pasadena,  California 


Ira  S.  Bowen 
Director 

Horace  W.  Babcock 
Assistant  Director 


OBSERVATORY  COMMITTEE 

Ira  S.  Bowen, 
Chairman 

Carl  D.  Anderson 

Horace  W.  Babcock 

Jesse  L.  Greenstein 

Robert  B.  Leighton 

Allan  R.  Sandage 


Contents 


Introduction 5 

Observing  Conditions 6 

Solar  Observations 6 

Solar  magnetic  fields 7 

Forbidden  nitrogen  lines  in  the 

solar  spectrum 7 

Planets 7 

Comets 8 

Stellar  Spectroscopy  and  Photometry     .  8 
Chemical  composition  of  stellar 

atmospheres 8 

Line  blanketing    .......  11 

Color-magnitude  and  chemical- 
composition  relationships       .      .      .12 

Color-spectral-type  relationships    .      .  13 

Photometry  of  stellar  clusters  and 

associations 14 

Photometry  of  double  stars  .  .  .15 
Photometry  of  variable  stars  .  .  .16 
Photometry  of  the  Giclas  proper 

motion  catalogue   .      .      .      .      .      .  17 

Subdwarfs 17 

White  dwarfs .18 

Faint  blue  stars 18 

Balmer  lines  in  early-type  stars      .      .  18 

RR  Lyrae  variables 19 

Supernovae 19 

U  Geminorum  stars  (dwarf  novae)       .  20 

Old  novae 20 

Shell  stars 20 

Mass  loss  from  stars  with  extended 

atmospheres 21 


Segregation  of  elements  in  magnetic 

stars 21 

Radial  velocities  of  magnetic  stars       .  22 

Stellar  polarization 23 

Gaseous  Nebulae  and  Interstellar  gas     .  24 

Galaxies 25 

Structure  and  internal  motions  of  the 

Galaxy 25 

Rotation  and  internal  motions  of 

galaxies 26 

Emission  nebulae  in  galaxies     ...  27 

Variable  stars  in  galaxies     ....  28 

Photometry  and  stellar  content  and 

evolution 30 

Catalogue  of  galaxies  and  clusters 

of  galaxies 32 

Internal  motions  of  clusters  of  galaxies  33 

Redshift-magnitude  relations    ...  33 

Radio  Sources 34 

Theoretical  Studies 35 

Stellar  atmospheres 35 

Star  formation 37 

Stellar  dynamics 37 

Cosmology 37 

Miscellaneous 38 

Instrumentation 39 

Guest  Investigators 39 

Staff  and  Organization 45 

Bibliography 47 


Carnegie  Institution  of  Washington  Year  Book  61,  1961-1962 


INTRODUCTION 

In  1904  George  E.  Hale,  acting  under  the  space  age  astronomy,  was  sponsored  by 

auspices   of   the   National    Academy   of  the  Douglas  Aircraft  Company  and  was 

Sciences,  invited  the  scientific  academies  held  at  the  California  Institute  of  Tech- 

and  the  astronomical  and  physical  socie-  nology  on  August  7,  8,  and  9.   It  was 

ties  of  a  number  of  countries  to  send  attended   by   about    100   engineers   and 

representatives  to  a  meeting  to  be  held  in  astronomers. 

connection  with  the  International  Con-  Because  of  the  interest  in  the  large 

gress  of  Science  at  the  St.  Louis  Exhi-  telescopes  at  Mount  Wilson  and  Palomar 

bition   for   the   purpose   of   establishing  Mountain,  arrangements  were  made  by 

"co-operation    among    individuals    and  the  Observatories  to  provide  transporta- 

institutions  engaged  in  Solar  Research."  tion  from  Los  Angeles  and  entertainment 

This  meeting  resulted  in  the  formation  of  on  the  mountains  for  the  foreign  delegates 

the  International  Union  for  Co-operation  to  the  Assembly  of  the  Union.  Trips  to 

in    Solar    Research,    which    held    later  Mount   Wilson  were   scheduled   on   the 

meetings    at    Oxford    (1905),    Meudon  afternoons  of  August  11  and  25,  and  to 

(1907),  Mount  Wilson  (1910),  and  Bonn  Palomar  on  August  12  and  26.  About  275 

(1913).  The  Mount  Wilson  meeting  was  delegates  took  advantage  of  this  oppor- 

attended  by  about  80  members  of  the  tunity    to    visit    the    facilities    on    the 

Union  and  invited  guests.  mountains. 

After  World  War  I,   the  Union  was  Nearly  all  members  of  the  staff  of  the 

reorganized  on  a  broader  basis  to  include  Observatories  attended  the  Assembly  at 

all  branches  of  astronomy  and  its  name  Berkeley  and  participated  in  the  sessions 

was  changed  to  the  International  Astro-  of  the  various  commissions  of  which  they 

nomical  Union.  Assemblies  of  the  Union  were  members. 

were  held  at  Rome  (1922),  Cambridge,  Throughout  the  history  of  the  Observa- 

England    (1925),    Leiden    (1928),    Cam-  tories  the  major  emphasis  has  been  placed 

bridge,     Massachusetts     (1932),     Paris  on  observations  of  the  sun,  stars,  nebulae, 

(1935),  and  Stockholm   (1938).  After  a  and  galaxies.  From  time  to  time,  however, 

ten-year    intermission    caused    by    the  when  the  Observatories'  equipment  was 

second  World  War,  meetings  occurred  at  suitable,    attention   has   been    given   to 

Zurich    (1948),    Rome    (1952),    Dublin  observations   of   planets   and    satellites. 

(1955),   and  Moscow   (1958).   The  next  For  example,  satellites  X,  XI,  and  XII  of 

General  Assembly  in  1961  was  planned  Jupiter  and  the  very  unusual  asteroids 

for  the  United  States  of  America,  and  it  Icarus  and  Geographos  were  discovered 

was   hoped   that   it   might   be   held   in  at    the    Observatories.    High-dispersion 

Pasadena.  However,  a  survey  of  the  hotel  spectroscopic  studies  of  Venus  and  Mars 

situation  indicated  that  to  accommodate  by  Adams  and  Dunham  provide  the  basis 

locally  the  more  than  1000  members  and  for  the  current  knowledge  of  the  compo- 

guests  who  have  attended  these  meetings  sition    of    their    atmospheres.    Infrared 

in  recent  years  would  be  impossible.  The  observations  by  Nicholson  and  Pettit  of 

1961    General    Assembly   of   the    Inter-  the  lunar  surface  during  an  eclipse  led  to 

national  Astronomical  Union  was  there-  the  concept  of  a  surface  covered  with 

fore  held  in  Berkeley,  between  August  15  dust.     Recently    these     infrared     lunar 

and    24,    the    University    of    California  observations  were  refined  by  Dr.  Shorthill 

acting  as  host  institution.  and   Mr.   Saari   of  the  Boeing  Aircraft 

Several    international    symposia    took  Company,  using  the  60-inch  on  Mount 

place  just  before  or  after  the  Berkeley  Wilson.  In  1958  and  1960  the  200-inch 

meeting.  One  of  these,  on  the  subject  of  was  used  by  Dr.  Sinton  of  the  Lowell 


6 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


Observatory  to  map  the  areas  on  Mars 
that  show  the  absorption  bands  near  3.4  y. 
which  are  attributed  to  organic  molecules. 

The  development  in  the  last  few  years 
of  rockets  capable  of  going  to  the  neigh- 
borhoods of  the  moon  and  the  inner 
planets  has  focused  attention  on  lunar 
and  planetary  problems.  Because  of  the 
much  lower  effort  and  cost  required  for 
ground-based  observations  compared 
with  observations  made  from  rockets,  it 
has  become  important  to  push  these  solar 
system  observations  to  the  limits  made 
possible  with  the  new  photometric  and 
infrared  techniques  developed  in  recent 
years. 

In  the  past  year  G.  Munch,  with  the 
assistance  of  Mr.  Robert  Younkin  of  the 
Jet  Propulsion  Laboratory,  has  used  the 
Cassegrain  spectrum  scanner  to  investi- 
gate the  monochromatic  albedo  and  the 
total  intensity  of  the  absorption  bands  in 
the  spectra  of  the  major  planets;  Munch 


has  also  obtained  high-dispersion  spectra 
of  Jupiter,  Saturn,  and  Neptune,  which 
have  been  studied  in  collaboration  with 
Dr.  Hyron  Spinrad  of  the  same  Labora- 
tory. Lines  of  the  hydrogen  molecule  at 
X6367.80  and  X6435.03  were  found  in  the 
spectrum  of  Saturn,  providing  the  first 
definite  evidence  for  the  presence  of 
hydrogen  in  its  atmosphere.  Dr.  Spinrad, 
analyzing  the  high-dispersion  spectra  of 
Venus  available  in  the  files  of  the  Ob- 
servatories, has  found  evidence  for  large 
changes  in  the  apparent  temperature  of 
the  atmosphere.  Dr.  Bruce  Murray  of  the 
Lunar  Research  Laboratory  at  the  Cali- 
fornia Institute  of  Technology  has  con- 
tinued the  studies  of  the  photoelectric 
colorimetry  of  the  moon  with  the  60-inch 
telescope.  He  has  also  developed  and 
tested  on  Mount  Wilson  a  special  20-inch 
infrared  telescope  which  will  be  used  for 
lunar  studies  at  an  altitude  of  13,000  feet 
on  White  Mountain. 


OBSERVING   CONDITIONS 


After  three  years  in  which  the  rainfall 
on  Mount  Wilson  averaged  less  than  45 
per  cent  of  normal,  the  precipitation  for 
1961-1962  jumped  to  46.14  inches,  or 
within  an  inch  of  the  total  for  the 
preceding    three    years.    This   increased 


rainfall  came  just  in  time  to  avoid  a 
serious  water  shortage  on  the  mountains. 
Solar  observations  were  made  on  311 
days;  the  200-inch  was  in  use  on  287 
nights,  the  100-inch  on  292  nights,  and 
the  60-inch  on  265  nights. 


SOLAR   OBSERVATIONS 


Solar  observations  were  made  by 
Cragg,  Hickox,  and  Utter.  The  numbers 
of  photographs  of  the  various  kinds  taken 
between  July  1,  1961,  and  June  30,  1962, 
were  as  follows : 

Direct  photographs  302 

Ha  spectroheliograms,  18-foot  focus  270 

K2  spectroheliograms,  18-foot  focus  258 

K2  prominences,  18-foot  focus  84 
Number  of  days  on  which 

magnetograms  were  obtained  223 

Effective  September  1,  1961,  a  basic 
change  was  made  in  the  method  of 
reduction  of  the  sunspot  magnetic  data. 
In  the  past,  sunspot  groups  were  num- 
bered and  a  list  of  these  groups  along 


with  average  magnetic  classifications  and 
dates  of  central  meridian  passage  was 
published,  or  otherwise  made  available, 
every  year  or  so.  More  recently,  investi- 
gators interested  in  this  information  have 
wanted  it  available  rapidly,  and  they 
have  frequently  been  interested  in  the 
appearance  of  the  sun  on  some  particular 
day.  To  meet  these  needs,  and  also  to 
save  time  in  the  reduction  process,  a  list 
is  now  prepared  each  month  giving  daily 
positions  and  magnetic  classifications  of 
spot  groups  for  which  there  are  magnetic 
measures.  Copies  of  this  list  are  sent  each 
month  to  interested  investigators.  As  the 
routine  observing  program,  except  for  the 
magnetograms,  has  a  low  priority,  the 


MOUNT      WILSON      AND      PALOMAR      OBSERVATORIES 


sunspot  information  is  not  as  complete  as 
in  previous  years.  Sunspot  groups  are  no 
longer  numbered,  and  no  attempt  is  made 
to  keep  track  of  returns.  The  K2  promi- 
nence patrol  was  also  ended  in  September. 
Because  spot  magnetic  polarities  are 
not  observed  as  often  as  in  the  past,  and 
because  the  method  of  reduction  has  been 
altered,  the  tables  of  sunspot  groups  and 
classification  usually  published  in  the 
Annual  Report  will  be  discontinued. 
Magnetic  classifications  of  spot  groups 
were  made  on  161  days  from  July  1,  1961, 
to  June  30,  1962. 

Solar  Magnetic  Fields 

Howard  has  completed  a  preliminary 
study  of  solar  magnetograph  observations 
made  with  very  small  apertures.  Mag- 
netic traces  with  an  aperture  about  2 
seconds  of  arc  on  a  side  show  root-mean- 
square  fluctuations  of  8.2  ±  4.4  gauss. 
The  autocorrelation  function  derived 
from  these  observations  shows  maxima 
near  16,000  km  and  40,000  km  and  in 
general  resembles  the  autocorrelation 
function  that  Rogerson  derived  from 
intensity  fluctuations  on  calcium  spectro- 
heliograms.  Similar  observations  made 
recording  line-of-sight  velocities  yield 
root-mean-square  fluctuations  of  0.39 
db  0.14  km/sec.  Observations  made 
recording  velocities  with  an  aperture  held 
fixed  show  autocorrelation  curves  that 
are  damped  cosine  curves,  indicating  the 
presence  of  oscillations  in  the  solar 
atmosphere.  The  period  observed  is  296 
seconds.  The  spectrum  line  used  for  all 
these  observations  was  X5250.218,  Fe  I. 
Instrumental  improvements  since  these 
observations  were  completed  in  the 
summer  of  1959  enable  us  now  to  make 
much  better  observations  of  this  type. 


Further  observations  are  planned  for  the 
near  future. 

The  daily  solar  magnetograms,  started 
in  1957,  constitute  a  unique  series  of 
observations  giving  valuable  information 
about  daily  configurations  of  the  solar 
magnetic  fields.  Howard  has  begun  an 
extensive  study  of  these  records,  which 
will  include  classification  of  magnetic 
regions  and  their  correlation  with  optical 
and  radio  phenomena.  The  investigation 
starts  with  the  magnetograms  from 
August  1959,  when  the  new  slant-line 
registration  was  begun.  One  interesting 
result  that  has  appeared  at  this  stage  of 
the  investigation  concerns  the  UM  regions 
first  discovered  by  H.  W.  and  H.  D. 
Babcock.  A  large  number  of  UM  regions 
have  been  identified;  invariably,  at  the 
position  of  the  UM  region  the  calcium 
(K2)  spectroheliogram  for  that  day  shows 
mottlings  somewhat  brighter  than  the 
ordinary  background.  Thus  it  may  be 
possible  to  detect  UM  regions  over  a 
period  of  fifty  years  or  more  using  the 
extensive  solar  plate  collection. 

Forbidden  Nitrogen  Lines  in  the  Solar 
Spectrum 

Starting  from  Vitense's  model  of  the 
solar  atmosphere  and  from  a  recent 
determination  of  the  abundance  of  nitro- 
gen by  Neven,  the  intensities  of  the  [N  I] 
X 10397  and  X 10407  lines  have  been 
computed  by  Houziaux.  From  a  com- 
parison of  these  results  with  the  intensity 
of  the  weak  feature  observed  at  X 10397 
on  several  spectrograms  recently  obtained 
by  Migeotte  at  the  Jungfraujoch  it  can 
be  concluded  that  the  forbidden  doublet 
2D5/2-2Pi/2,  3  /2  is  present  in  the  infrared 
solar  spectrum.  The  2D3/2  —  2Pi  /2,  3/2 
doublet  is  hidden  by  a  strong  line  of 
unknown  origin. 


PLANETS 

An  observing  program  has  been  started  atmospheres.  One  aspect  of  the  program 

by  G.  Munch  to  serve  as  a  basis  for  a  is  concerned  with  the  determination  of 

reexamination  of  problems  related  to  the  the  energy  distributions  in  their  spectra 

structure  and  composition  of  planetary  by   photoelectric   scanning.    The   mono- 


8 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


chromatic  albedos  of  the  planets  and  the 
total  intensity  of  their  absorption  bands 
will  thus  be  derived.  Observations  of  the 
integrated  light  of  the  major  planets  have 
been  obtained  with  the  Cassegrain  scan- 
ner on  the  60-inch  telescope,  from  X3400 
out  to  the  long-wavelength  sensitivity 
limit  of  photomultipliers  with  trialkali 
cathodes.  These  scans  have  in  part  been 
studied  by  Mr.  Robert  Younkin  of  the 
Jet  Propulsion  Laboratory.  Repeated 
tracings  of  Jupiter  obtained  during  four 
different  periods  have  provided  consistent 
results  proving  that  there  is  not  a  steep 
fall  in  the  energy  distribution  shortward 
of  X3900,  as  has  been  reported  in  the  past. 
During  this  preliminary  work,  it  has  been 
found  that  the  amount  of  time  involved 
in  an  exhaustive  study  of  the  tracings  is 
so  great  that  such  study  is  impractical 
without  the  aid  of  automatic  data 
reduction  equipment.  An  arrangement  in 
which  the  output  of  the  scanner  is  fed 
directly  into  a  digital  voltmeter  with 
magnetic  tape  recording  has  therefore 
been  tested.  This  will  be  reduced  with  an 
IBM  7090  computer.  Further  observa- 
tions of  the  variations  in  color  and  spectra 
of  planetary  surfaces  will  be  carried  out 
by  means  of  such  auxiliary  equipment. 

Munch  is  utilizing  the  greatly  increased 
resolving  power  and  speed  of  the  coude 


spectrographs  to  take  spectra  of  the 
planets  under  the  highest  dispersion 
possible  with  the  purpose  of  detecting 
new  spectral  features  and  studying  the 
structure  of  known  ones  over  the  various 
parts  of  the  planetary  disks.  Plates  of 
Jupiter,  Saturn,  and  Uranus  in  the  blue 
and  yellow-red  regions  of  the  spectrum 
have  been  obtained.  Part  of  this  material 
is  being  studied  in  collaboration  with  Dr. 
Hyron  Spinrad  of  the  Jet  Propulsion 
Laboratory.  In  these  plates,  Spinrad 
verified  his  discovery  of  the  anomalous 
inclination  of  the  NH3  lines  in  the 
spectrum  of  Jupiter's  equator.  In  Saturn 
it  was  found  that  the  lines  of  CH4  band 
at  X6190  have  an  inclination  greater  than 
half  that  of  the  Fraunhofer  lines  of  the 
scattered  solar  spectrum,  by  about  10 
per  cent — an  amount  twice  as  large  as  the 
probable  error  of  measurement.  In  the 
same  plates  of  Saturn  two  sharp  lines  of 
X6367.80  and  X6435.03  which,  with  cer- 
tainty, must  be  identified  with  the  S(l) 
and  S(0)  lines,  respectively,  of  the  4-0 
quadruple  rotation- vibration  band  of  H2 
were  discovered.  The  possibility  of  ob- 
serving the  quadruple  spectrum  of  H2  was 
suggested  by  Herzberg  in  1938,  but  this 
is  the  first  observation  of  the  S(l)  and 
S(0)  lines  of  the  4-0  band  in  any  astro- 
nomical object. 


COMETS 

The  bright  comet  Seki-Lines  (1962c)  of  NH2  and  C2.  The  distortion  of  the 

was    observed    by    Greenstein.    It    was  CN  (0,  0)  band  by  resonance  fluorescence 

remarkably  dust-free  after  perihelion,  the  was  quite  different  from  that  of  Comet 

Na  I  lines  were  very  weak,  and  the  visual  Mrkos  (1957d).  Spectra  of  1962c  will  be 

region  of  the  spectrum  consisted  largely  measured  by  Greenstein  and  Arpigny. 


STELLAR  SPECTROSCOPY  AND   PHOTOMETRY 


During  the  report  year,  900  spectro- 
grams were  taken  with  the  200-inch 
telescope,  970  with  the  100-inch,  and  550 
with  the  60-inch. 

Chemical  Composition  of 
Stellar  Atmospheres 

The  program  for  the  study  of  the 
abundances  of  the  elements  in  astronom- 


ical objects  continued  under  the  direction 
of  Greenstein  with  the  support  of  the  Air 
Force  Office  of  Scientific  Research  of 
USAF. 

Spectrophotometric  analyses  of  the 
high-galactic-latitude  supergiants  HD 
161796  (F3  lb)  and  89  Herculis  (F2  la) 
have  been  carried  out  by  Searle,  Sargent, 
and  Jugaku.  Comparisons  were  made  with 


MOUNT      WILSON      AND      PALOMAR      OBSERVATORIES 


the  standard  low-latitude  supergiants 
<p  Cassiopeiae  (FO  la)  and  a  Persei 
(F5  lb).  All  the  elements  studied  are 
found  to  have  the  same  relative  abun- 
dances in  all  these  stars.  Spectroscopic 
absolute  magnitudes  were  derived  for  the 
high-latitude  supergiants  using  <p  Cas  and 
a  Per  as  calibration  stars.  It  is  concluded 
that  both  89  Her  and  HD  161796  could 
have  reached  their  present  heights  above 
the  galactic  plane  in  times  comparable  to 
their  estimated  times  of  evolution  from 
the  main  sequence  if  they  were  expelled 
from  the  plane  at  the  time  of  their  forma- 
tion with  velocities  of  the  order  of  100 
km/sec.  These  results — which  are  in 
disagreement  with  those  of  an  earlier 
study  of  these  same  stars  by  Abt — are 
consistent  with  the  view  that  the  high- 
galactic-latitude  supergiants  are  evolved 
runaway  stars. 

Additional  measurements  since  last 
year  on  3  Centauri  A  were  made  by 
Jugaku  and  Sargent  on  a  Radcliffe 
Observatory  coude  plate  of  the  far  ultra- 
violet for  line  identifications.  Most  of  the 
40  lines  between  X3500  and  X3100  can  be 
identified  with  Fe  II,  Mn  II,  and  Ni  II, 
although  a  few  fairly  strong  lines  remain 
unidentified.  The  Be  II  doublet  at  X3130 
is  absent.  A  plate  of  the  visual  region 
obtained  in  April  1962  shows  that  the 
longward  shift  of  X6678  of  He  I,  which 
was  interpreted  as  an  isotope  shift,  is 
still  present. 

The  abundance  analysis  of  k  Cancri, 
B8p,  an  Mn  star,  by  Jugaku  and  Sargent 
is  progressing.  Equivalent  widths  of  250 
lines  in  the  photographic  region  have  been 
measured.  The  ratio  P  III/P  II  gives  a 
value  of  0ion  =  0.39,  which  is  typical  of  a 
normal  star  of  spectral  type  about  B7. 
The  B  —  V  color  also  agrees  with  such  a 
temperature.  A  study  of  the  hydrogen 
lines  gives  log  Pe  =  1.92.  Using  these 
values  of  6  and  log  Pe,  the  preliminary 
abundance  results  are  P/Si  ^  1  (as  in  3 
Cen  A — this  means  that  P  is  overabun- 
dant by  a  factor  of  100).  The  identifica- 
tion of  a  line  at  X3984  with  Hg  II  by 
Bidelman  leads  to  an  overabundance  of 


about  30,000  for  Hg.  Helium  is  deficient 
by  a  factor  of  about  10  (factors  of  6  were 
found  for  3  Cen  A  and  a  Sculptoris).  Be 
is  overabundant  by  a  factor  of  100 
relative  to  the  sun. 

The  study  of  the  infrared  O  I  lines  in 
the  spectra  of  20  Ap  stars  has  been 
completed  by  Searle  and  Sargent.  Oxygen 
is  found  to  be  deficient  with  respect  to 
hydrogen  by  factors  ranging  from  8  to 
more  than  100  in  all  Ap  stars  of  the 
Si-Eu-Cr,  Eu-Cr,  Eu-Cr-Sr,  and  Sr 
classes,  whereas  in  the  Mn  stars  the 
oxygen  abundance  is  normal.  Assuming 
that  they  originated  with  normal  compo- 
sition, the  oxygen-deficient  Ap  stars 
demand  that  O  must  have  been  trans- 
muted into  one  or  more  of  the  cosmically 
abundant  elements  by  an  as  yet  unspeci- 
fied process. 

The  infrared  N  I  lines  fall  at  the  limit 
of  plate  sensitivity  and  for  this  reason 
have  been  studied  in  only  four  bright  Ap 
stars.  Two  of  these,  a2  Canum  Venati- 
corum  and  /3  Coronae  Borealis,  are 
deficient  in  O;  the  remaining  two, 
ip  Herculis  and  n  Leporis,  are  Mn  stars 
with  normal  oxygen  abundance.  In  none 
of  these  stars  are  the  infrared  N  I  lines 
detectable,  although  they  are  clearly 
present  in  the  spectra  of  standard  stars 
from  B5  to  F5.  It  appears  that  nitrogen 
is  deficient  in  all  four  Ap  stars  by  factors 
estimated  to  be  10  or  more. 

Infrared  spectrograms,  at  20  A/mm, 
have  been  obtained  by  Searle  and  Sargent 
of  four  bright  metallic-line  stars  and  four 
standard  stars  to  study  the  behavior  of 
the  O  I  lines.  The  metallic-line  stars 
selected  fall  in  the  two-color  and  color- 
magnitude  diagram  among  the  extreme 
oxygen-deficient  Ap  stars,  but  unlike  the 
Ap  stars  their  oxygen  abundance  is 
normal. 

Spectrograms  of  30  Ap  stars  and  6 
standard  stars  with  types  between  B5 
and  A4  have  been  obtained  at  10  A/mm 
by  Searle  and  Sargent.  The  pressure-  and 
temperature-insensitive  ratios  of  Mg  II 
(X4481)/Si  II  (X4128,  X4130)  and  C  II 
(X4267)/He  I  (X4471)  are  being  studied 


10 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


and  Balmer  line  profiles  obtained. 

For  the  Ap  stars  of  earliest  type,  the 
X4200,  Si  and  Mn  classes,  there  is  an 
excellent  correlation  between  the  central 
depth  of  Hy  and  the  U  —  B  color,  iden- 
tical with  the  correlation  obtained  for 
standard  stars.  There  is  no  systematic 
difference  between  the  Balmer  line  profiles 
of  the  Ap  stars  and  the  standard  stars, 
and  the  location  of  an  Ap  star  in  the 
two-color  diagram  is  a  good  indicator  of 
the  atmospheric  temperature  of  the  star. 

Among  the  hotter  Ap  stars  there  is  no 
evidence  for  Mg  abundance  anomalies, 
and  Si  is  overabundant  (by  about  X  10) 
only  in  stars  that  show  the  high-excitation 
Si  II  lines  at  X4200.  Certain  sharp-line 
late  B-type  stars  (e.g.,  21  Aquilae  and 
HD  207840)  which  have  been  called  Si 
class  Ap  stars  have  Si  lines  of  normal 
strength  for  their  colors,  and  normal 
strength  of  He,  C,  O,  and  Mg.  It  is 
probable  that  they  are  "peculiar"  only  in 
that  they  have  unusually  sharp  lines. 

The  C  II  (X4267)  strength  is  normal 
for  the  color  in  Ap  stars  of  the  Mn  class 
and  in  the  so-called  Si  stars  which  do  not 
show  X4200.  In  these  stars,  the  lines  of 
He  I  are  normal  or  only  slightly  weaker 
than  in  normal  stars  of  the  same  color. 

In  the  stars  that  have  definite  Si  over- 
abundance (the  X4200  stars),  the  C  II  line 
is  weak  (by  a  factor  of  about  5  in  the 
equivalent  width)  and  the  He  I  lines  are 
very  weak  (by  a  factor  of  about  10  in  the 
equivalent  width).  It  seems  hard  to 
escape  the  conclusion  that  the  X4200  stars 
are  very  deficient  in  helium. 

Spectrograms  of  27  Ap  stars  and  4 
standards  at  10  A/mm  in  the  photo- 
graphic ultraviolet  have  been  obtained  by 
Searle,  Sargent,  and  Jugaku  to  study  the 
behavior  of  the  Be  II  doublet  at  X3130. 
In  the  cooler  Ap  stars,  the  Be  II  lines,  if 
present,  are  seriously  blended,  but  in  the 
hotter  Ap  stars  they  are  free  from  serious 
blending.  No  lines  of  Be  II  are  to  be  seen 
in  nine  X4200  stars  observed.  Of  10  Mn 
stars,  6  show  no  trace  of  Be  II,  but  the 
remaining  4  have  very  strong  lines.  In  the 
Be-strong    Mn    stars    (112    Herculis,    k 


Cancri,  /x  Leporis,  and  v  Herculis),  Be  is 
estimated  to  be  overabundant  by  a  factor 
of  about  100.  All  the  stars  that  show 
strong  lines  of  Be  II  also  have  lines  of 
Ga  II  in  their  spectra.  However,  ir1 
Bootis,  in  whose  spectrum  Ga  II  lines  are 
present,  shows  no  trace  of  Be  II. 

Scans  of  the  continuous  spectrum  of 
about  20  Ap  stars  have  been  made  by 
Jugaku  and  Sargent  with  the  Cassegrain 
scanner.  They  will  be  examined  in 
conjunction  with  hydrogen-line  profiles 
obtained  from  coude  plates  of  the  same 
stars  to  see  whether  there  is  definite 
evidence  that  the  atmospheres  of  mag- 
netic stars  differ  from  those  of  normal 
stars.  Preliminary  results  show  that  stars 
like  a  Andromedae  (B8p  Mn) ,  which  have 
anomalously  blue  B  —  V  and  U  —  B 
colors,  have  complete  continuous  energy 
distributions  identical  to  those  of  main- 
sequence  stars  as  hot  as  B5. 

Greens tein,  Parker,  Wallerstein  (Uni- 
versity of  California  at  Berkeley),  Heifer 
(University  of  Rochester),  and  Aller 
(University  of  Michigan)  have  collabo- 
rated in  an  extensive  analysis  of  three  red 
giants  with  extremely  weak  lines:  HD 
122563,  165195,  and  221170.  In  last  year's 
report,  HD  165195  and  221170  were 
mentioned  as  weak-lined  G  dwarfs. 
Further  analysis  and  the  earlier  incom- 
pleted work  of  Greenstein  and  Aller  on 
HD  122563  have  shown  that  these  stars 
are,  in  fact,  giants  with  colors  and  spectra 
like  the  stars  in  globular  clusters.  In 
many  ways,  these  stars  are  extraordinary; 
their  colors  are  quite  red  (B  —  V  ^  -f- 1.0) , 
yet  at  first  glance  they  could  be  mistaken 
for  F  subdwarfs.  The  results  are  tempera- 
tures near  4100°K,  log  Pe  =  -2.5, 
metal/hydrogen  ratios  500  times  lower 
than  in  the  sun  (like  the  most  extreme 
subdwarf).  One  problem  is  in  the  opacity, 
which  seems  to  be  largely  Rayleigh 
scattering,  although  the  expected  colori- 
metric  effects  are  not  found.  In  addition, 
the  ratio  of  iron-group  metals  to  heavy 
elements  like  Sr,  Zr,  Ba,  Ce,  and  Eu  is 
abnormal,  when  compared  with  the  sun, 
in  that  the  heavy  elements  are  deficient 


MOUNT      WILSON      AND      PALOMAR     OBSERVATORIES  11 

by  an  additional  factor  of  50  in  HD  does  not  differ  from  that  of  the  sun  by 
122563.  This  is  the  first  known  example  more  than  a  factor  of  2. 
of  large  changes  within  the  abundances  of  Gunn  is  carrying  forward  a  program  of 
the  metals.  It  indicates  that  the  stars  studying  ' 'strong-line"  versus  "weak- 
condensed  at  an  early  stage  in  the  life  of  line"  field  F  stars  from  spectrograms  of 
our  Galaxy,  perhaps  107  to  109  years  (at  20  A/mm  dispersion  taken  with  the 
the  very  latest),  and  probably  107  to  10 8  60-inch.  It  will  be  determined  whether 
years  after  the  beginning  of  element  the  weak-line  group  can  be  explained  only 
synthesis.  Another  unusual  effect  is  that  in  terms  of  lowered  metal  abundance 
Eu  behaves  like  the  other  heavy  elements;  (relative  to  hydrogen)  or  whether  differ- 
although  Eu  in  the  sun  and  earth  was  ences  in  mean  turbulence  and  mean 
synthesized  by  the  r  process  of  neutron  degree  of  ionization  can  explain  the 
capture,  in  HD  122563  it  was  made  by  existence  of  the  group, 
the  s  process.  There  are  other  elemental 

deficiencies,  e.g.,  V  and  Mn,  of  a  previ-  Lme  Blanketing 

ously  recorded  type,  and  also  evidence  Sandage    and    Smith    completed    an 

that  the  heavier  elements  were  synthe-  observational   study   of   the   differential 

sized  in  a  very  metal-poor  environment,  blanketing  effect  of  weakening  the  Fraun- 

Gunn  and  Kraft  have  completed  a  hofer  lines  in  stellar  spectra.  A  four-color, 
study  based  mainly  on  200-inch  coude  broad-band  photometric  system  was  de- 
spectrograms  of  the  hydrogen-to-metal  vised  using  an  RCA  7263  photomultiplier 
ratio  in  F-type  stars  of  NGC  752,  a  cell  with  an  S20  trialkali  photocathode. 
galactic  cluster  of  age  1  X  109  years.  It  The  system  is  close  to  the  standard  U,  B, 
has  been  suggested  from  earlier  studies  of  V  but  adds  a  fourth  color,  R,  at  an 
small-scale  spectrograms  that  NGC  752  effective  wavelength  of  6800  A.  Observa- 
may  have  a  lowered  metal  abundance  tions  were  made  with  the  Palomar  20-inch 
relative  to  the  sun.  If  this  were  true,  it  telescope  of  64  standard  stars  whose  U, 
would  mean  that  as  little  as  10 9  years  ago  B,  V  values  are  well  known  and  of  32 
different  regions  of  the  Galaxy  were  subdwarfs  of  intermediate  to  large  ultra- 
forming  stars  of  different  metal  content  violet  excess.  Three  results  came  from  the 
at  the  same  time,  though  the  galactic  study.  (1)  It  is  possible  to  transform  the 
orbit  of  NGC  752  is  nearly  the  same  as  natural  photometric  system  of  the  S20 
that  of  the  sun.  photocathode   to   the    U,   B,    V  system 

In  the  present  analysis,  careful  atten-  (usually  observed  with  an  S4  cathode), 
tion  was  paid  to  deriving  accurate  with  a  systematic  accuracy  of  0™02  in 
ionization  temperatures.  With  the  aid  of  B  —  V  and  0™05  in  U  —  B.  The  data 
H7  profiles  based  on  model  atmospheres  show  a  nonlinearity  in  the  (u  —  b) 
(Searle  and  Oke,  1962)  and  known  natural  =f(U  —  B)  transformation  curve 
abundances  (Parker,  Greenstein,  Heifer,  of  amplitude  0™05  which  is  undoubtedly 
and  Wallerstein,  1961),  the  scale  of  Tion  due  to  the  different  ultraviolet  response 
for  Hyades  F-type  stars  was  first  estab-  of  the  1P21  and  RCA  7263  multiplier, 
lished  from  curves  of  growth.  From  the  Therefore,  precise  transformations  of  S20 
models  and  H7  line  strengths,  ionization  data  to  the  U  —  B  system  must  be  done 
temperatures  for  NGC  752  stars  were  with  a  nonlinear  equation.  (2)  The  effect 
estimated  relative  to  Hyades  stars.  This  of  differential  line  blanketing  on  the 
method  of  determining  Tion  replaces  the  positions  of  stars  in  the  U  —  B  versus 
customary,  and  somewhat  unsatisfactory,  V  —  R  diagram  is  clearly  seen.  The  sub- 
procedure  of  estimating  Tion  from  TeKC.  dwarfs,  as  expected,  stand  high  by 
Using  these  temperatures,  it  is  concluded  d(U  —  B)  =  0™2.  (3)  An  extension  of  the 
from  a  study  of  curves  of  growth  that  the  blanketing  theory  discussed  in  Year  Books 
metal  abundance  of  stars  in  NGC  752  59  and  60  was  made  to  include  the  R 


12 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


point,  and  it  was  shown  that  the  theory 
can  with  great  accuracy  correct  the  sub- 
dwarf  positions  in  the  U  —  B  versus 
V  —  R  and  in  the  B  —  V  versus  V  —  R 
diagrams  to  the  position  of  stars  of  high 
metal  abundance.  Therefore,  most  and 
perhaps  all  of  the  previously  observed 
peculiarity  in  the  energy  distribution  of 
subdwarfs  over  the  spectral  range  X3300 
to  X6809  can  apparently  be  explained  as 
due  to  the  effects  of  weak  Fraunhofer 
lines  on  broad-band  photometric  meas- 
urements. Unpublished  data  of  Sears  and 
Whitford  at  the  Lick  Observatory  suggest 
that  this  is  also  true  all  the  way  to  the 
infrared  point  at  X  =  10,000  A  on  their 
six-color  system. 

Spectroscopic  scans  are  being  obtained 
by  Oke  for  a  selection  of  very  metal- 
deficient  stars.  Because  of  the  weakness 
of  the  lines,  the  absolute  energy  distribu- 
tion in  the  spectrum  can  be  accurately 
obtained  over  a  large  wavelength  interval. 
This  facilitates  comparisons  with  model 
atmosphere  fluxes.  In  addition,  model 
atmospheres  for  these  stars  can  be  com- 
puted with  higher  accuracy  than  for 
corresponding  metal-normal  stars.  The 
hydrogen-line  profiles  will  also  be  studied. 
Scans  of  the  extremely  metal-deficient 
red  giant  HD  122563  have  been  obtained 
between  X3400  and  X8000.  Assuming  that 
the  continuous  opacity  is  due  to  the 
negative  hydrogen  ion,  comparison  with 
model  atmospheres  gives  an  effective 
temperature  of  about  4200°K,  in  good 
agreement  with  the  value  determined  by 
Greenstein,  Wallerstein,  and  Parker. 

Color-Magnitude  and 
Chemical-Composition  Relationships 

Eggen  and  Sandage  reexamined  the 
problem  of  the  position  of  the  main 
sequence  as  a  function  of  chemical  com- 
position in  the  M v,  B  —  V  and  the  M boi, 
log  Te  diagrams.  A  sample  of  stars  was 
chosen  (1)  which  were  known  to  be 
dwarfs  from  spectroscopic  luminosity 
criteria,  (2)  which  had  trigonometric 
parallaxes  larger  than  0"034,  and  (3)  for 
which  photometry  on  the  U,  B,  V  or  Uc, 


B,  V  was  available.  From  this  extensive 
material,  Eggen  and  Sandage  confirmed 
their  previous  result  (1959)  that  the 
displacement  of  a  star  below  the  Hyades 
main  sequence  in  the  Mv,  B  —  V  diagram 
is  directly  proportional  to  the  ultraviolet 
excess.  Assuming  that  the  ultraviolet 
excess  is  due  to  Fraunhofer-line  weaken- 
ing caused  by  low  metal  abundance,  and 
applying  the  blanketing  theory  reported 
in  previous  years,  it  was  shown  that 
differential  blanketing  corrections  move 
all  stars,  independently  of  the  size  of  their 
ultraviolet  excess,  onto  the  Hyades  main 
sequence  with  good  accuracy.  In  par- 
ticular, 16  extreme  subdwarfs  with  well 
determined  absolute  magnitudes  and  with 
excess  values  averaging  8 (U  —  B)  =  0^21 
were  observed  to  be  1™05  ±  0™04  fainter 
than  the  Hyades  before  blanketing  cor- 
rections were  applied  but  are  moved  onto 
the  Hyades  main  sequence  to  within 
+0™03  ±  0^05  after  the  corrections  were 
made.  These  results  provide  the  observa- 
tional justification  for  the  main-sequence 
fitting  procedure  to  find  distances  of  star 
clusters  where  it  has  always  been  assumed 
that  the  main-sequence  positions  are 
identical  in  the  M  v,  B  —  V  diagram  for 
clusters  of  different  chemical  composition. 
These  reported  results  show  that  the 
assumption  is  correct,  but  only  if  differ- 
ential blanketing  corrections  are  applied 
before  the  modulus  fit  is  made. 

It  was  further  shown  that  the  large 
scatter  in  the  main  sequence,  between 
B  -  V  =  0m4  and  B  -  V  =  0m8  of 
trigonometric  parallax  stars,  is  due  not 
only  to  errors  in  the  parallaxes  but 
primarily  to  the  line-blanketing  effect  on 
the  colors.  The  distribution  of  ultraviolet 
excess  shows  that  large  variations  in 
chemical  composition  exist  among  the 
parallax  stars  closer  than  29  parsecs 
(ir  >  0"035).  The  main-sequence  scatter 
is  markedly  reduced  when  blanketing 
corrections  are  applied  to  the  observed 
B  —  V  colors. 

The  fact  that  subdwarfs  are  moved 
onto  the  Hyades  line  after  applying 
blanketing    corrections    shows    that    a 


MOUNT      WILSON      AND      PALOMAR      OBSERVATORIES 


13 


separate  subdwarf  sequence  does  not 
exist  in  the  Mhoi,  log  Te  diagram,  despite 
the  low  metal  abundance  of  these  stars. 
This  remarkable  result  permits  the  deter- 
mination of  the  helium  abundance  in  the 
interior  of  subdwarfs  from  the  theory  of 
the  stellar  interior.  Decreasing  the  metal 
abundance  will,  in  general,  move  the  star 
below  the  main  sequence.  A  correspond- 
ing decrease  in  the  helium  abundance 
works  in  the  opposite  direction.  The 
effects  can  be  predicted  qualitatively  by 
homology  arguments,  but  computed  stel- 
lar models  are  needed  for  accurate 
abundance  determinations.  Sandage  used 
the  models  of  DeMarque  to  show  that 
the  hydrogen  (X) ,  helium  ( Y) ,  and  metal 
(Z)  abundances  by  weight  of  the  extreme 
subdwarfs  with  well  determined  absolute 
magnitudes  are  X  «  0.95,  Y  «  0.05, 
Z  ^  0.001,  compared  with  adopted  solar 
values  of  X  =  0.65,  Y  =  0.31,  Z  =  0.04. 
Subdwarfs  are  the  oldest  stars  we  know. 
Therefore,  this  result  suggests  that  the 
primeval  abundance  of  both  helium  and 
the  heavy  elements  was  very  low,  a  result 
in  agreement  with  theories  of  element 
enrichment  of  the  interstellar  medium 
with  time.  The  results  are  not  final. 
Interior  models  with  more  closely  spaced 
abundance  differences  are  needed.  Ikco 
Iben  of  the  California  Institute  Physics 
Department  spent  two  months  at  Los 
Alamos  computing  better  interior  opacity 
values  from  an  IBM  7090  computing 
program  developed  by  A.  N.  Cox  and 
Robert  Brownlee.  Iben's  resulting  models 
will  be  used  when  they  are  completed  for 
a  second  solution  to  the  problem. 

From  a  photometric  study  of  three 
separate  samples  of  main-sequence  stars, 
Eggen  has  found  that,  judged  by  the 
distribution  of  ultraviolet  excesses  with 
respect  to  the  Hyades  stars,  two-thirds  of 
the  stars  in  the  solar  neighborhood  have 
a  higher  ratio  of  metals  to  hydrogen  than 
the  sun.  The  distribution  of  ultraviolet 
excesses  suggests  that,  if  enrichment  of 
the  interstellar  medium  has  been  uniform 
with  time,  the  rate  of  star  formation 
between  5  X  109  (formation  of  the  sun) 


and  5  X  108  (formation  of  the  Hyades) 
years  ago  was  nearly  uniform. 

Color-Spectral-Type  Relationships 

Two  years  ago  a  report  was  published 
by  Wilson  in  which  it  was  indicated  that, 
among  main-sequence  field  stars  later 
than  type  G5,  a  considerable  spread  of 
color  for  a  given  type,  or  of  type  for  a 
given  color,  was  present.  The  data  used 
for  this  purpose  were  the  old  Mount 
Wilson  spectral  types  and  the  photo- 
electric colors  measured  by  Eggen.  Sub- 
sequent spectrograms  of  some  of  these 
stars  revealed  that  the  Mount  Wilson 
spectral  types  were  unreliable.  As  a 
result,  new  spectrograms  of  10  A/mm 
dispersion  have  been  obtained  by  Wilson 
for  more  than  100  of  these  stars,  and  new 
types,  based  on  the  Yerkes  system,  have 
been  derived.  When  the  revised  types  are 
plotted  against  Eggen's  colors,  a  con- 
siderable spread,  amounting  to  0.2  mag 
at  some  types,  is  again  found,  although 
many  stars  no  longer  occupy  the  same 
locations  in  this  diagram  as  in  the  former 
one. 

In  addition,  many  members  of  the 
Hyades  cluster  were  also  observed  spec- 
troscopically  (although  at  smaller  disper- 
sion), and  a  similar  diagram  was  con- 
structed for  them.  When  these  two 
diagrams  are  compared,  the  correlation 
between  spectral  type  and  color  appears 
to  be  tighter  for  the  Hyades  members 
than  for  the  field  stars.  In  the  light  of 
current  knowledge,  the  simplest  explana- 
tion of  this  result  is  to  suppose  that,  as 
regards  chemical  composition,  the  Hyades 
stars  represent  a  more  uniform  group 
than  the  field  stars.  This  conclusion  is  not 
especially  surprising. 

In  recent  Year  Books  Wilson  has 
reported  the  discovery  of  a  definite 
relationship  between  the  width  of  the 
reversals  of  the  H  and  K  lines  and  the 
absolute  magnitude  of  the  star.  However, 
the  total  emission  intensity  of  the  H  and 
K  reversals  seems  to  have  no  obvious 
correlation  with  other  features  of  the 
spectrum.  This  raises  the  question:  why 


14  CARNEGIE     INSTITUTION     OF     WASHINGTON 

is  it  that  among  the  late  main-sequence  members  of  double  and  multiple  systems 

field  stars  there  are  stars  which  appear  to  have  H  and  K  reversals  of  similar 

spectroscopically  identical  in  every  re-  strengths  when  due  allowance  is  made  for 

spect  except  that  one  has  strong  central  differences   in    spectral    type.    The   few 

H  and  K  reversals  and  the  other  little  or  exceptions  to  this  rule  seem  to  be  re- 

none?  In  the  previously  published  work  stricted  to  members  of  systems  that  are 

mentioned  above,  it  appeared  that  there  themselves     short-period     spectroscopic 

was  a  strong  tendency  for  the  redder  binaries. 

stars  of  a  given  type  to  have  stronger  Clearly,  Wilson's  work  summarized 
reversals  than  their  bluer  counterparts,  above  is  still  in  a  preliminary  stage,  and 
With  the  revised  types,  however,  little  if  definite  conclusions  should  be  avoided  at 
any  of  this  tendency  remains,  and  thus  present.  Nevertheless,  enough  has  been 
what  appeared  to  be  a  promising  clue  has  accomplished  to  justify  pursuing  it  fur- 
proved  to  be  illusory.  ther   in   attempting   to   understand   its 

As  an  outgrowth  of  the  present  research  significance  and,  perhaps,  eventually  in 
more  definite  light  is  being  shed  on  this  making  use  of  it  as  a  tool  for  the  further- 
problem.   That  H  and  K  reversals  are  ance  of  other  aims, 
unusually  frequent  in  the  Hyades  stars 

has  long  been  known  and  is  fully  con-  Photometry  of  Stellar  Clusters  and 

firmed  by  the  present  work;  indeed,  if  the  Associations 

discussion  is  restricted  to  types  G5-K0,  Sandage    and    Smith    completed    the 

inclusive,  the  fraction  of  Hyades  members  study  of  the  color-magnitude  diagram  of 

in  this  range  which  has  strong  reversals  NGC  6712,  a  globular  cluster  of  relatively 

seems  to  approach  100  per  cent.  The  same  high  metal  abundance  situated  in  the 

thing  is  true  of  main-sequence  members  Scutum  Cloud.  A  photoelectric  sequence 

of  the  Praesepe  cluster.  Even  in  the  Coma  was  determined  to  V  =  17™5,  B  =  18™5 

cluster,  which  is  relatively  poor  in  known  with   the   200-inch,   and   short-exposure 

main-sequence  members,   the  frequency  plates    were    measured    for    the    color- 

of  strong  reversals  appears  to  be  greatly  magnitude  diagram.  It  suffices  to  remark 

in   excess   of   that   for  the  local   main-  that  the  earlier  conclusion  reached  from 

sequence  objects  of  the  same  range  of  the  study  of  NGC  6536,  that  the  absolute 

spectral  types,  where  the  frequency  is  less  magnitudes  of  the  brightest  giant  stars  in 

than  20  per  cent.  globular  clusters  are  a  function  of  their 

Thus,  on  the  available  evidence,  stars  chemical  composition,  is  confirmed, 

formed  in  clusters  have  a  much  higher  Plates  of  the  globular  cluster  NGC  6712 

probability  of  possessing  strong  H  and  K  taken  by  Sandage  were  photometered  for 

reversals  than  the  noncluster  field  stars  RR  Lyrae  stars  and  reduced  by  Norton 

of  the  same  spectral  types.  This  property  and  Lynden-Bell. 

may  then,  perhaps,  be  thought  of  as  a  Sandage   continued   the   photoelectric 

genetic  factor  that  can  supply  information  measurement  of  faint  stars  in  M  15  and 

of  some  kind  as  yet  unspecified  about  the  M   92.    Complete   color-magnitude   dia- 

circumstances    under    which    the    stars  grams  for  these  clusters  were  prepared, 

were  formed.  Katem  measured  a  special  series  of  plates 

A  natural  extension  of  the  investigation  from  V  =   l?^  to  V  =   22™0  in  both 

along  the  line  of  genetic  thinking  is  to  clusters    to    obtain    the    main-sequence 

look    at    the    reversals    in    double    and  positions  using  the  photoelectric  stars  as 

multiple    main-sequence    systems.    This  standards.  The  results  disagree  by  about 

has   been   done  for   a   number  of   such  0^05  in  the  color  of  the  main-sequence 

objects,   and   the   results   very   strongly  termination    point    with    the    definitive 

support  the  genetic  viewpoint.  That  is,  photometry   of   M    13   reported   several 

there  is  a  very  decided  tendency  for  the  years  ago,  and  a  special  program  of  photo- 


MOUNT      WILSON     AND      PALOMAR     OBSERVATORIES  15 

electric  intercomparison  of  the  five  clus-  globular  clusters  so  far  analyzed,  the 
ters  M  3,  M  5,  M  13,  M  15,  and  M  92  derived  ages  drop  to  between  9  and 
was  begun.  This  study  must  be  completed  14  X  109  years.  The  mean  absolute  mag- 
before  the  M  15  and  M  92  results  will  be  nitude  of  the  RR  Lyrae  stars  should  be 
published.  adopted  as  Mv   =    +0.3,  MB   =    +0.5 

A  photometric  investigation  of  the  mag. 
clusters  and  surrounding  associations  of  Investigation  of  clusters  of  intermedi- 
h  and  x  Persei  using  combined  photo-  ate  age  (109  years)  and  low  metal  content 
electric  and  photographic  techniques  has  (3^2  to  }/%  of  normal)  has  been  completed 
been  made  on  the  U,  B,  V  system  to  by  Arp.  Work  continues  in  the  clusters 
V  =  17.0  by  Wildey.  A  nonunique  epoch  near  the  Galactic  nucleus.  A  very  clear 
of  star  formation  is  suggested  both  by  an  giant  branch  in  the  color-magnitude 
apparent  fine  structure  in  the  bright  end  diagram  of  NGC  6838  has  been  obtained, 
of  the  C-M  diagram  and  by  the  presence  The  brighter  sequences  appear  somewhat 
of  main-sequence  stars  to  the  photometric  similar  to  those  in  47  Tucanae  and  NGC 
limit  even  though  apparently  contracting  6356.  The  photometric  fitting  of  the  main 
stars  are  found  at  brighter  magnitudes,  sequence  and  derivation  of  accurate 
Nuclear  and  gravitational  time  scales  are  reddening  are  now  in  progress, 
in  agreement.  The  data,  when  compared  Eggen  has  isolated  the  "Pleiades 
with  the  theoretical  evolutionary  tracks  Group,"  which,  together  with  the  Hyades 
of  Hayashi  and  Cameron,  suggest  that  and  Sirius  Groups,  accounts  for  nearly 
helium  burning  takes  place  on  the  blue  25  per  cent  of  the  A-type  stars  in  the  solar 
side  of  the  Hertzsprung  gap.  The  differ-  neighborhood.  Also,  two  groups  of  high- 
ences  in  apparent  evolutionary  tracks  velocity  stars,  in  addition  to  the  Groom- 
between  the  Galaxy,  the  Large  Magel-  bridge  1830  Group,  have  been  found.  One 
lanic  Cloud,  and  the  Small  Magellanic  of  these,  Kapteyn's  Star  Group,  contains 
Cloud  are  reconfirmed.  The  two-color  at  least  two  RR  Lyrae  variables,  SU 
diagram  of  the  bluest  stars  is  of  a  gray-  Draconis  (0?66)  and  ST  Leonis  (0d48). 
body  character.  Sp  versus  (B  —  V)  is  The  derived  median,  visual  absolute 
mono  tonic  for  all  stars.  magnitude    is     +0^8    for    both    stars. 

The  publication  of  the  analysis  of  the  Another  possible  member  of  this  group  is 
color-magnitude  diagrams  of  M  5,  which  ST  Comae  Berenices  (0™60).  The  sub- 
indicated  ages  of  20  X  109  years,  empha-  dwarfs  HD  106038,  CC835,  and  CC486, 
sized  the  serious  discrepancy  between  as  well  as  the  horizontal  branch  A-type 
these  ages  and  those  yielded  by  the  star  HD  139961,  are  also  members.  The 
measurements  of  cosmological  expansion,  mean  ultraviolet  excess  of  the  group  is 
In  connection  with  another  investigation,  -f-0?21,  and  the  variables  all  have 
Arp  has  reexamined  the  assumptions  in  AaS  =  6.  No  RR  Lyrae  variables  have 
regard  to  space  reddening  made  in  the  been  identified  as  belonging  to  the  third 
analysis  of  the  color-magnitude  diagrams  high-velocity  group,  which  contains  the 
of  globular  clusters.  He  found  the  following  subdwarfs  HD  74000,  Ross  626,  Ross 
results:  (1)  The  cosecant  reddening  law  451,  Ross  453,  HD  108177,  and  -35°- 
applies  to  high-latitude  globular  clusters  14849. 
despite  the  fact  that  photometric  analysis 

of  stars  within  500  parsecs  shows  on  the  Photometry  of  Double  Stars 

average  0.06  mag  less   reddening   than  Eggen  began  a  program  of  U,  B,   V 

that  given  by  the  law.  The  implication  photometry  of  the  components  of  visual 

that    reddening    in    the    Galaxy    is    not  double   stars  with   the   200 -,    100-,   and 

concentrated  entirely  within  a  thin  sheet  60-inch    reflectors.    The    components    of 

needs  to  be  examined  further.  (2)  If  the  about  100  pairs  have  so  far  been  observed, 

cosecant  reddening  law  is  used  for  all  the  including  the  following  of  special  interest: 


16 


CARNEGIE     INSTITUTION      OF     WASHINGTON 


(1)  Red  dwarf-white  dwarf  pairs.  Ob- 
servations of  the  components  of  six  wide, 
proper-motion  pairs  discovered  by  Giclas, 
Slaughter,  and  Burnham  at  the  Lowell 
Observatory,  G39-27/28,  G 14-57/58, 
G87-28/29,  G102-39/40,  Gil  1-71/72,  and 
G61-16/17,  have  confirmed  the  discovery 
suspicions  that  one  of  the  components  in 
each  pair  is  a  white  dwarf.  An  additional 
pair,  G24-9/10,  has  also  been  found  to 
contain  a  late-type  white  dwarf.  (2) 
Intrinsic  variables.  The  results  for  the 
main-sequence  companions  of  the  (3 
Cephei  variables,  0  Cephei  and  a  Scorpii, 
indicate  that  the  luminosities  of  these 
variables  may  be  about  a  magnitude 
fainter  than  usually  supposed.  Photom- 
etry of  the  solar-type,  main-sequence 
companions  of  the  Mira  variables,  R 
Cassiopeiae  and  RU  Cygni,  indicates  a 
luminosity  near  —  1  for  both  variables.  A 
weak-lined  G-type  subdwarf,  10  seconds 
from  the  RR  Lyrae  variable  AP  Serpentis 
(P  =  0d25),  shows  an  ultraviolet  excess 
of  4-0^12  and  gives  an  absolute  median 
luminosity  near  +2m  for  the  variables  if 
the  stars  form  a  physical  system.  (3)  A 
wide  range  of  eclipsing  stars  with  visual 
companions  is  included  in  the  program  to 
help  in  standardizing  mass-luminosity 
and  radius-luminosity  relations.  Also,  ob- 
servations of  the  physical  companions  to 
several  bright  W  Ursae  Majoris  systems 
indicate  that  the  ultraviolet  excesses  are 
the  same  for  the  variable  and  the  non- 
variable  components.  The  system  of  VW 
Cephei  shares  a  large  space  motion  with 
HD  199476,  although  the  stars  are 
separated  in  the  sky  by  about  1  degree. 
W  Ursae  Majoris,  BD  +55°1351,  has  a 
faint  companion  (VE  =  12m35,  B  —  V 
=  -f-1^70),  a  degree  away,  with  which  it 
forms  ADS  7497.  Although  the  two  com- 
ponents are  not  physically  connected  they 
share  a  common  proper  motion.  The 
radial  velocity  of  the  BD  star  is  not  avail- 
able. From  the  companions  of  these  two 
variables,  as  well  as  that  of  AM  Leonis 
(ADS  8024),  the  variables  are  found  to 
lie  0™75  above  the  main  sequence.  A  faint 
star  (VE   =   13m84;  B  -   V  =    +0m65; 


U  —  B  =  +0^11)  about  10  seconds  from 
UY  Ursae  Majoris  may  be  physically 
connected  to  that  variable.  (4)  Structure 
in  the  mass-luminosity  relationship  has 
been  found  for  stars  with  different  metal 
contents  as  judged  by  their  ultraviolet 
colors  compared  with  Hyades  stars.  The 
colors  of  all  binaries  for  which  orbital 
elements  are  available  are  being  obtained 
by  Eggen  with  the  20-inch  and  60-inch 
reflectors  for  further  study  of  this 
structure. 

Photometry  of  Variable  Stars 

Eggen  has  undertaken  three-color 
photometry  of  all  northern  cepheids  not 
already  observed  photoelectrically.  One 
preliminary  result  from  this  program  is 
that  Baade's  faint  cepheids  in  Cygnus  are 
reddened  by  about  1™0.  Also,  all  known 
contact  binaries  are  being  observed  for 
color.  Two  variables  classified  in  the 
literature  as  contact  binaries  have  been 
found  to  violate  the  period-color  relation 
previously  established  by  Eggen.  Exten- 
sive observations  of  one  of  these,  SZ 
Lyncis,  shows  it  to  be  a  short-period,  RR 
Lyrae  variable  (period  near  0d13)  and  not 
a  contact  binary.  A  previous  conclusion 
that  T  Tauri  variables  and  contact 
binaries  do  not  coexist  in  space  is  appar- 
ently violated  by  the  recent  discovery  by 
Gotz  of  V449  Orionis  (W  Ursae  Majoris, 
P  =  0d44)  and  V441  Orionis  (T  Tauri 
Variable)  which  are  separated  by  less  than 
30  minutes.  Observations  on  five  nights 
in  February  with  the  100-inch  indicate 
that  V449  Ori  is  not  a  contact  binary.  The 
star  showed  very  little  variation  in  visual 
magnitude,  VE  =  15^10  to  15m28,  or 
(B  -  V)  color,  +  lm05  to  -flm26,  but  the 
ultraviolet  color  showed  erratic  variations 
of  more  than  a  magnitude.  V441  Ori 
(VE  =  14m72,  B  -  V  =  lm74,  U  -  B  = 
-p-l^ll)  showed  no  variation.  Other 
variables  observed  include  Nova  Orionis 
(1667  and  1894)  and  X  Leonis.  The  nova 
showed  no  variation  in  the  blue  and  visual 
(VE  =  14m14,  B  -  V  =  +0m48)  on  five 
nights,  but  there  are  erratic  variations 
in  the  ultraviolet.  A  bright  maximum  of 


MOUNT      WILSON     AND      PALOMAR      OBSERVATORIES 


17 


X  Leo  was  observed  from  February  2  to  5, 
1962. 

Photometry  of  the  Giclas  Proper 
Motion  Catalogue 

A  routine  program  of  photoelectric 
observation  of  stars  of  high  proper  motion 
taken  from  the  Giclas  Lowell  Observatory 
Proper  Motion  Catalogue  was  started  in 
September  by  Kowal  under  the  super- 
vision of  Sandage.  The  pilot  program  for 
the  discovery  of  new  subdwarfs,  reported 
last  year,  was  so  successful  that  this 
regular  discovery  program  was  begun. 
Good  progress  was  made  during  the 
report  year  with  more  than  700  stars  ob- 
served and  reduced  in  the  three  colors 
U,  B,  V.  Both  the  60-inch  and  20-inch 
reflectors  were  used  to  make  the  observa- 
tions. Over  100  new  subdwarfs  have  been 
found.  At  least  30  of  them  have  the 
extreme  line-weakened  characteristics  of 
the  globular-cluster  main-sequence  stars. 
Sandage  continued  his  routine  observa- 
tional program  of  determining  radial 
velocities  of  the  stars  in  Kowal's  lists  that 
have  ultraviolet  excess  values  greater 
than  8(U  -  B)  =  0^14.  This  is  a  standby 
program  with  the  200-inch  coude,  which 
is  used  only  when  sky  conditions  prevent 
more  critical  photoelectric  work  at  the 
prime  focus.  But  poor  seeing  conditions 
and  partly  cloudy  weather  were  prevalent 
enough  during  the  winter  season  so  that 
spectra  for  35  program  stars  were  ob- 
tained in  this  way  with  the  18-inch  camera 
giving  a  dispersion  of  18  A/mm.  Again, 
the  results  show  that  stars  with  high 
ultraviolet  excess  values  invariably  have 
large  space  velocities  relative  to  the  sun. 

The  photometric  discovery  program 
will  continue  for  another  year,  by  which 
time  it  is  hoped  that  more  than  2000  stars 
of  the  Giclas  Catalogue  will  have  been 
observed. 

Subdwarfs 

Greenstein  has  completed  a  velocity 
program  on  some  150  F-K  subdwarfs  and 
halo  B  and  A  stars.  The  number  of 
spectroscopic  binaries  found  is  very  low. 
Nevertheless,  spectroscopic  examination 


of  the  high-velocity  stars  revealed  many 
types  of  peculiarities.  After  several  at- 
tempts, an  apparently  reliable  scheme  of 
visual  classification  at  18  A/mm  was 
developed  with  the  following  results: 
extreme  weak-line  subdwarfs  of  F-K 
types,  30  per  cent;  moderately  weak-line 
subdwarfs,  18  per  cent;  slightly  weak-line, 
7  per  cent;  subgiants  or  giants,  10  per 
cent;  horizontal-branch  A-G  stars,  4  per 
cent;  ionized  lines  enhanced,  15  per  cent; 
CH  enhanced  relatively  to  metals,  16  per 
cent.  Some  of  the  so-called  subdwarfs  are 
clearly  above  the  main  sequence  spectro- 
scopically,  and  the  10  per  cent  figure  for 
giants  or  subgiants  means  that  many  very 
high  space  motions  are  included.  The 
radial  velocities  have  internal  probable 
errors  from  0.5  to  1.3  km/sec. 

Jones  has  discussed  a  collection  of  200 
plates  of  late-type  subdwarfs  taken  by 
Greenstein  and  Sandage  with  the  coude" 
spectrograph  of  the  Hale  telescope  at  18 
A/mm.  An  attempt  has  been  made  to 
set  up  a  three-dimensional  classification 
scheme  which  estimates  the  spectral  type 
and  luminosity  of  the  star  independent  of 
any  weakening  of  the  spectral  lines.  The 
spectral  type  is  based  on  four  ratios  of 
line  pairs  of  differing  excitation  potential 
and  the  strength  of  the  hydrogen  lines; 
the  luminosity  is  estimated  from  three 
ratios  of  ionized  to  neutral  lines  in  a 
manner  closely  following  the  Mount 
Wilson  Catalogue.  The  strength  of  six  of 
the  1.5-volt  lines  of  Fe  has  also  been 
estimated,  and  the  values  have  been 
combined  to  form  an  index  of  the  line 
strength.  The  principal  conclusions  are  as 
follows:  spectral  types  on  this  system  are 
well  correlated  with  the  B  —  V  colors 
corrected  for  blanketing  as  far  as  the 
main  sequence  is  concerned,  but  stars 
above  the  main  sequence  appear  to  obey 
another  relation.  Stars  whose  lines  are 
among  the  strongest  for  their  assigned 
spectral  type  have  types  on  this  system 
that  correlate  very  well  with  the  MK  and 
Mount  Wilson  systems.  This  correlation 
was  used  to  convert  the  new  types  from 
an  arbitrary  numerical  scale  to  the  well 


18 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


known  A-F-G-K  scale.  Line  weakening 
occurs  more  frequently  for  stars  earlier 
than  G5,  where  it  may  amount  to  0.7 
logarithmically.  Later  than  G8,  it  rarely 
exceeds  0.1.  Owing  to  the  small  amount  of 
material,  only  one  curve  was  derived  to 
reduce  the  luminosity  characteristic  to 
absolute  magnitude  based  on  trigono- 
metric parallaxes,  but  the  absolute  magni- 
tudes so  derived  agree  very  well  with  the 
Mount  Wilson  Catalogue.  The  Hertz- 
sprung-Russell  diagram  shows  a  marked 
main  sequence  with  a  large  number  of 
weak-line  stars  which  appear  to  define  a 
"turnoff"  at  about  G2.  There  are  also 
several  subgiants,  some  with  markedly 
weak  lines. 

White  Dwarfs 

The  discovery  or  spectroscopic  con- 
firmation of  white  dwarfs  was  continued 
by  Greenstein,  who  finds  Feige  22,  24, 
Giclas  21-16,  24-10,  28-13,  29-38,  61-17, 
67-23,  93-53,  LDS  235B  (helium-rich), 
-37°10500,  Wolf  457  (essentially  con- 
tinuous) to  be  of  this  type.  A  program  of 
observation  of  white-dwarf  members  of 
double  stars  has  been  started  by  Green- 
stein in  collaboration  with  the  photo- 
metric work  of  Eggen.  One  X4670  white 
dwarf,  i.e.,  molecular  carbon-rich,  is  a 
member  of  a  wide  binary. 

Faint  Blue  Stars 

A  project  was  started  by  Zwicky  with 
the  aid  of  a  grant  from  the  National 
Science  Foundation  to  obtain  spectra  of 
special  types  of  blue  stars  for  the  purpose 
of  determining  their  radial  velocities  and 
spectroscopic  characteristics.  The  data 
thus  obtained  will  be  used  to  investigate 
the  statistical  distribution  and  proper 
motions  of  these  objects. 

The  spectrographic  observations  were 
made  by  Berger  with  the  4-inch  camera 
on  the  60-inch  Cassegrain  spectrograph. 
Most  of  the  stars  observed  are  in  the  list 
of  subluminous  hot  stars  discovered  by 
Feige.  A  few  additional  stars  were  sup- 
plied by  Haro  and  by  the  list  of  blue  stars 
published  by  Cowley.  One  to  four  spectra 


of  17  stars  have  been  obtained  in  the 
magnitude  range  between  7.5  and  11.3. 
Preliminary  measures  of  part  of  the 
plates  do  not  indicate  a  large  velocity  for 
these  blue  stars,  \V\   <  50  km/sec. 

A  search  for  very  faint  blue  stars 
(15  <  m  <  19)  near  the  north  galactic 
pole  has  been  started  by  Berger  with  the 
48-inch  schmidt  using  three-color  photog- 
raphy. These  observations  will  be  com- 
bined with  the  observations  of  8500  faint 
blue  stars  near  the  south  galactic  pole 
made  by  Haro  and  Luyten  for  an  in- 
vestigation of  the  statistical  distribution 
of  the  halo  population.  These  plates  will 
also  be  compared  with  the  National 
Geographic  Society-Palomar  Observatory 
Sky  Survey  plates  taken  in  the  early 
1950's  for  the  detection  of  faint  blue  stars 
with  large  proper  motions. 

B aimer  Lines  in  Early-Type  Stars 

Spectra  of  57  B  and  Be  stars  have  been 
obtained  by  Houziaux  in  the  region 
X3900  to  X3550  in  order  to  study  the 
confluence  of  the  Balmer  lines.  It  has  been 
found  that  this  emission-free  region  pro- 
vides a  good  observational  criterion  for 
the  determination  of  the  gravity  of  early- 
type  emission-line  stars.  The  ratios  of  the 
intensities  between  the  Balmer  lines, 
corrected  for  atmospheric  absorption,  to 
the  intensity  at  X3862  are  plotted  versus 
the  principal  quantum  number.  Stars  of 
the  same  spectral  type  and  of  different 
luminosities  are  clearly  separated.  These 
observational  data  are  now  compared 
with  the  results  of  flux  computation  for 
high  lines  at  41  wavelengths  performed 
for  40  model  atmospheres  in  the  tempera- 
ture range  9510°K  to  29,000°K  and  for 
log  g  =  1(1)5. 

Changes  have  been  observed  by 
Houziaux  in  the  infrared  spectrum  of 
Pleione.  The  strong  O  I  absorptions  at 
X7771  and  X8446  have  disappeared.  From 
a  spectrophotometric  study  of  the  in- 
frared region,  it  has  been  shown  that  the 
O/H  ratio  during  the  shell  episode  was 
0.6  X10-4,  a  value  similar  to  the  one 
observed  in  other  B-type  stars. 


MOUNT      WILSON      AND      PALOMAR     OBSERVATORIES 


19 


RR  Lyrae  Variables 

The  photoelectric  spectrum  scanner  has 
been  used  by  Oke  in  a  continued  program 
to  measure  absolute  energy  distributions 
in  the  spectra  of  RR  Lyrae  variable  stars. 
The  stars  being  studied  are  RR  Lyrae, 
SU  Draconis,  X  Arietis,  SW  Androm- 
edae,  and  VZ  Cancri.  Photographic 
spectra  with  a  dispersion  of  9  or  18  A/mm 
are  also  being  obtained.  The  absolute 
energy  distributions  are  compared  with 
fluxes  computed  from  model  atmospheres 
to  obtain  values  of  the  effective  tempera- 
ture at  each  phase.  These  profiles  can  also 
be  used,  along  with  the  photoelectric 
scans,  to  determine  space  reddening  with 
high  accuracy.  After  correction  for  red- 
dening, RR  Lyr,  SU  Dra,  and  X  Ari  all 
appear  to  have  the  same  temperature 
range,  6000°K  to  7500°K.  The  tempera- 
ture curves,  as  a  function  of  phase,  for 
these  three  stars  are  similar  but  not  iden- 
tical. The  radial-velocity  curves  for  SU 
Dra  and  RR  Lyr  demonstrate  conclusively 
that  differential  radial  motions  exist 
throughout  the  atmospheres  at  all  phases. 
Since  the  continuous  opacity  changes  with 
phase,  different  mass  layers  are  observed 
with  different  velocities  at  various  phases. 
Consequently,  the  observed  radial-ve- 
locity curve  does  not  represent  the  motion 
of  the  star's  photosphere,  and  integration 
of  the  velocity  curves  does  not  give  the 
radius-displacement  curve.  Therefore,  the 
Wesselink-Baade  method  cannot  give 
correct  radii  for  RR  Lyrae  stars.  It  is 
found,  however,  that  a  modification  of  the 
method  can  be  used  successfully  to 
determine  the  absolute  radius.  For  SU 
Dra  the  mean  radius  is  5.2Ro.  Using  the 
temperatures  obtained  from  the  scans, 
this  leads  to  a  mean  absolute  visual 
magnitude  of  +0.8  ±  0.4.  The  error  can 
be  reduced  if  only  differences  of  absolute 
magnitudes  are  required. 

The  scanner  is  being  used  by  Oke  to 
measure  absolute  energy  distributions  of 
suspected  horizontal  branch  stars.  These 
measures  will  be  used  in  conjunction 
with  temperatures  obtained  by  fitting  H7 


to  theoretical  profiles  to  obtain  the  red- 
dening. The  value  of  the  effective  gravity 
determined  from  the  scans  indicates 
whether  or  not  a  star  can  be  a  horizontal 
branch  object.  A  comparison  will  be  made 
of  effective  gravities  of  RR  Lyrae  stars 
and  nonvariable  horizontal  branch  stars 
to  study  the  effects  of  the  dynamics  of 
pulsation  on  the  atmosphere.  One  star, 
HD  161817,  is  confirmed  to  be  a  hori- 
zontal branch  star  similar  to  RR  Lyrae 
at  maximum  light. 

Supernovae 

The  search  for  supernovae  has  con- 
tinued under  the  direction  of  Zwicky  and 
with  the  support  of  the  National  Science 
Foundation.  Between  July  1,  1961,  and 
May  31,  1962,  a  total  of  15  supernovae 
was  discovered  at  Palomar,  all  of  them  on 
plates  taken  with  the  48-inch  schmidt 
telescope.  Of  this  number,  4  were  dis- 
covered by  Humason,  2  by  Kearns,  6  by 
Zwicky,  and  1  each  by  H.  S.  Gates, 
Rudnicki,  and  Berger.  Of  these,  2  were  in 
the  Coma  cluster,  as  identified  through 
determination  of  their  symbolic  velocities 
of  recession. 

The  supernova  in  NGC  4303  developed 
into  a  type  II  spectrum  after  a  peculiar 
early  behavior.  Greenstein  found,  near 
maximum,  large  negative  displacements 
of  the  emission  lines  which  were  accom- 
panied by  very  greatly  displaced  absorp- 
tion edges.  Within  a  month,  the  emission 
lines  became  sharper  and  returned  to  zero 
velocity.  Apparently  both  absorption  and 
emission  were  formed  at  the  leading  edge 
of  an  opaque  expanding  shell.  It  is  unclear 
whether  the  apparent  deceleration  is  real 
or  caused  by  the  appearance  of  the  far 
(receding)  side  of  the  star,  but  there  is  no 
doubt  that  the  velocity  spread  was 
greatly  reduced. 

Observations  by  Greenstein  showed 
that  the  supernova  in  NGC  1058  had 
many  extremely  sharp  lines  in  November 
1961,  some  accompanied  by  absorption 
edges.  The  spectrum  resembled  that  of  a 
type  II  object  with  low  velocity  dis- 
persion. H,  He  I,  and  C  III  were  present. 


20  CARNEGIE     INSTITUTION     OF     WASHINGTON 

During  a  secondary  light  maximum  in  spectrograph  (180  A/mm).  Nova  Persei 

December  a  spectrum  was  obtained  at  (1901)  has  a  definitely  composite  spec- 

18  A/mm  providing  excellent  line  profiles  trum  (sdBe  +  K),  and  Nova  Lacertae 

of  the  shallow,  broad  lines.  In  January  (1910)  shows  some  peculiar,  as  yet  not 

1962  the  spectrum  was  like  that  75  days  understood,  variations  in  the  velocity  of 

earlier  except  that  the  lines  had  become  He  II  (X4686)  in  emission. 

sharper  again.  This  object  was  extremely  Nova  Sagittae  (1913,  1946)  proves  to 

complex  in  light  and  spectral  variations,  be    a    spectroscopic    binary    with    the 

and  it  might  be  taken  to  be  a  distinct  shortest   known    period:   81}^   minutes. 

subclass  of  type  II.  This  could  be  detected  only  by  trailing 

Additional  spectra  of  several  of  the  the  star  over  a  long  slit  without  repetition. 

brighter  supernovae  were  obtained  by  An  "S  wave"  was  found  in  the  hydrogen 

Zwicky.  From  a  study  of  the  spectra  and  emission  lines  superimposed  on  the  ab- 

the  light  curves,  Zwicky  believes  that  it  sorption  lines  of  a  white  dwarf  already 

may  be  necessary  to  postulate  several  new  found      by      Greenstein.       Meanwhile, 

types  of  supernovae  in  addition  to  types  Krzeminski  at  Lick  discovered  that  Nova 

I  and  II.  The  supernova  in  NGC  1058  Sge  is  an  eclipsing  binary,  as  well. 

may  be  a  representative  of  a  type  inter-  Greenstein  and  Kraft,  together  with 

mediate    between    ordinary    novae    and  Jon  Mathews  of  the  California  Institute 

supernovae.  Physics    Department,     collaborated    in 

pointing    out    the    possible    importance 

U  Geminorum  Stars  (Dwarf  Novae)  of  Nova  Sge  as  a  test  for  the  part  of 

An  extensive  study  of  several  U  Gem  Einstein's  general  theory  that  predicts 

variables  at  minimum  light  was  continued  the    existence    of    gravitational    waves, 

by  Kraft  with  the  prime-focus  spectro-  However,  since  the  mass  ratio  is  unknown 

graph  of  the  200-inch.  A  spectroscopic  in  Nova  Sge,  it  cannot  be  determined 

binary  orbit  for  the  emission-line  com-  whether    the    star    emits    a    significant 

ponent  of  Z  Camelopardalis  was  obtained  amount  of  gravitational  energy.  If  5tli/ 

withP  =  6.5  hours;  SU  Ursa  Ma j oris  was  y&2  ~  5,  for  example,  the  emission  of 

found  to  vary  in  radial  velocity,  but  a  gravitational  energy  is  found  to  be  30 

period    has    not   yet    been    determined,  times  that  of  the  luminosity,   and  the 

Forty-four  per  cent  of  the  U  Gem  stars  system  would  collapse  in  only  20  million 

that  can  be  reached  from  Palomar  have  years.  The  eclipse  period  would  be  out  of 

been  studied  so  far;  all  have  proved  to  be  phase  by  a  minute  in  15  years — an  easily 

binaries  with  P  <  9  hours.  A  study  of  the  detectable  quantity, 

motions  leads  to    <Mv>    ~   +9.5  at  Another  object  that  may  be  similar  to 

minimum.  Nova  Sge  is  the  suspected  U  Gem  variable 

An  hypothesis  was  advanced  suggesting  EX  Hydrae,  for  which  both  Krzeminski 

U  Gem  stars  are  descendants  of  W  Ursa  and  Kraft  find  P  =»  99  minutes.  These 

Majoris  binaries.  The  two  kinds  of  vari-  two,  together  with  Herbig's  similar  object 

ables  have  comparable  space  distributions  VV  Puppis,  for  which  P  =  100  minutes, 

and  velocities.   Plausible  arguments  on  might   all   be   emitters   of  gravitational 

mass  transfer  between  the  components  of  radiation, 
a  typical  W  UMa  system  show  that  it 

might  well  become  a  U  Gem  star  after  ohell  stars 

107  to  108  years.  Further  plates  of  89  Herculis  have  been 

obtained  by  Sargent  to  continue  work  on 

Old  Novae  ^he  circumstellar  envelope.  During  1961, 

A  search  for  binary  stars  of  short  period  the  shortward  displaced  absorption  com- 

among  old  novae  has  been  started  by  ponents  at  the  Balmer  lines  weakened 

Kraft    using    the    200-inch    prime-focus  considerably  in  an  interval  of  less  than  60 


MOUNT      WILSON      AND      PALOMAR      OBSERVATORIES 


21 


days.  The  other  peculiar  features — the 
emission  in  the  redward  wing  of  Ha,  the 
emission  at  the  intercombination  lines 
of  the  neutral  metals,  and  the  shortward 
displaced  absorption  lines  at  H  and  K  and 
the  D  lines — did  not  change.  A  prelimi- 
nary curve  of  growth  for  the  circumstellar 
H  lines  shows  that  the  turbulent  velocity 
is  very  large — greater  than  20  km/sec. 

Mass  Loss  from  Stars  with  Extended 
Atmospheres 

In  connection  with  his  continuing  study 
of  mass  loss  from  late-type  giants, 
Deutsch  has  observed  the  differential 
motions  occurring  in  the  atmospheres  of 
some  M-type  supergiants.  At  4.5  A/mm, 
spectra  of  \x  Cephei  (M2  la)  reveal  the 
contributions  of  at  least  five  atmospheric 
layers  with  distinguishable  radial  veloc- 
ities in  the  range  —29  to  +29  km/sec. 
The  radial  velocities  of  individual  lines 
correlate  with  equivalent  width  and 
excitation  potential.  Asynchronous  veloc- 
ity variations  can  be  clearly  seen  in  three 
of  the  atmospheric  layers. 

For  the  further  elucidation  of  the  mass- 
loss  process,  Deutsch  has  under  con- 
tinuing observation  a  number  of  late- 
type  giants  which  are  spectroscopic 
binaries.  Several  of  these  exhibit  circum- 
stellar lines  at  the  D  line  as  well  as  at 
H  and  K. 

A  reconsideration  of  the  pronounced 
line  weakening  found  in  early-type  Mira 
variables  has  led  Deutsch  to  the  con- 
clusion that  these  stars,  like  others  be- 
longing to  the  halo  population,  prob- 
ably have  metal  deficiencies  of  the  order 
of  10~2  as  compared  with  the  sun  or 
other  normal  stars.  The  chief  source  of 
the  opacity  in  these  very  cool  atmospheres 
remains  unidentified,  however,  and  until 
it  is  known  the  degree  of  metal  deficiency 
will  remain  uncertain. 

Segregation  of  Elements  in  Magnetic  Stars 

The  problem  of  the  anomalous  abun- 
dance of  elements  in  the  magnetic  stars 
of  type  A  has  been  investigated  by  H.  W. 


Babcock.  These  stars  generally  show  an 
abnormally  high  abundance  of  several 
elements  such  as  Cr,  Sr,  Mn,  Si,  Eu,  and 
other  rare  earths;  further,  in  the  out- 
standing spectrum  variables,  Eu  and 
Cr  are  observed  to  undergo  large  varia- 
tions in  antiphase.  Whatever  the  basic 
mechanism  of  this  variation,  the  process 
by  which  the  particular  elements  are 
segregated  and  the  manner  in  which  this 
segregation  is  maintained  in  the  face  of 
gaseous  diffusion  are  matters  of  consider- 
able interest. 

In  1947  it  was  pointed  out  that  many 
of  the  anomalous  elements  in  the  mag- 
netic stars  belong  to  the  iron  group  or 
the  rare-earth  group,  and  that  their 
atoms  generally  have  large  magnetic 
susceptibility  owing  to  the  occurrence  of 
partly  filled  internal  electron  shells.  Each 
such  atom  has  a  magnetic  moment,  fi, 
the  effective  value  of  which,  measured  in 
Bohr  magnetons,  is  expressed  by  the 
product  gM,  in  which  g  is  the  Lande 
factor  and  M  is  the  magnetic  quantum 
number.  In  a  magnetic  field  H,  having  a 
gradient  VH,  the  atom  will  experience  a 
force  gMvH  in  the  direction  of  stronger 
or  weaker  field,  depending  upon  whether 
its  alignment  is  parallel  or  antiparallel  to 
the  field.  The  idea  that,  as  a  result  of 
this  force,  a  selective  paramagnetic 
migration  of  atoms  might  occur  in  a 
stellar  atmosphere  was  set  aside  because 
in  thermal  equilibrium  the  magnetic  sub- 
levels  will  be  very  nearly  equally  popu- 
lated. Therefore  the  net  magnetic  moment 
would  be  vanishingly  small. 

The  phenomenon  of  "optical  pumping,' ' 
recently  investigated  by  microwave  phys- 
icists, now  offers  a  nonthermal  process  by 
which  atoms  in  a  magnetic  field  can  be 
polarized;  i.e.,  they  can  be  given  a  prefer- 
ential orientation.  It  has  been  found  that 
irradiation  of  atoms  by  polarized  light,  in 
a  magnetic  field,  can  drastically  alter  the 
relative  population  of  the  magnetic  sub- 
levels.  This  results  in  a  net  paramagnetic 
susceptibility  and  in  a  migration  toward 
stronger  or  weaker  regions  of  the  field 
provided  that  the  magnetic  gradient  is 


22 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


sufficient.  It  is  also  known,  as  a  result  of 
recent  laboratory  investigations,  that 
polarized  atoms  have  a  rather  remarkable 
resistance  to  disorientation  by  collisions 
with  other  atoms.  This  disorientation 
resistance,  according  to  Princeton  investi- 
gators, is  particularly  marked  for  spheri- 
cally symmetric  atoms — those  in  S  states 
— as  compared  with  those  in  states  having 
orbital  anisotropy. 

The  possible  application  of  the  fore- 
going facts  relating  to  optical  pumping 
and  disorientation  resistance  has  been 
considered  with  respect  to  the  abundance 
anomalies  of  the  magnetic  stars.  For  the 
elements  of  the  periodic  table,  the  mag- 
netic moment  of  the  ground  state  of  the 
neutral  atom,  as  well  as  for  the  first  two 
stages  of  ionization,  has  been  computed. 
Elements  whose  ground  states  are  not  $ 
states  have  been  rejected  by  reason  of 
insufficient  disorientation  resistance.  Then 
elements  of  very  low  astrophysical  abun- 
dance have  been  deleted,  and,  finally,  in 
the  accompanying  tabulation,  all  re- 
maining atoms  having  a  magnetic  mo- 
ment greater  than  2  Bohr  magnetons 
have  been  listed. 


Atomic 

Number 

Element 

±gM 

7 

NI 

3,1 

8 

Oil 

3,1 

15 

PI 

3,1 

16 

SII 

3,1 

24 

CrI 

6,4,2 

CrII 

5,3,1 

25 

Mnl 

5,3,1 

Mnll 

6,4,2 

Mnlll 

5,3,1 

42 

Mo  I 

6,4,2 

Moll 

5,3,1 

63 

Eii  I 

7,  5,  3,  1 

EuII 

8,  6,  4,  2 

It  is  seen  that  a  few  atoms  best  fitting 
the  conditions  for  paramagnetic  migra- 
tion in  a  magnetic-field  gradient  are  Cr, 
Mn,  Mo,  and  Eu.  With  the  exception  of 
Mo  (which  has  no  prominent  lines  in  the 
commonly  observed  spectral  region), 
these  elements  are  known  to  show  striking 
abundance    anomalies   in    the   magnetic 


stars.  Indeed,  Cr  and  Eu  are  the  most  out- 
standing peculiar  elements  in  the  mag- 
netic spectrum  variables.  This  result 
lends  support  to  the  initial  suggestion 
that  migration  or  segregation  of  para- 
magnetic elements  actually  occurs  in 
magnetic  stars,  even  though  this  idea 
seems  a  priori  quite  unlikely  because  of 
the  requirement  of  a  large  magnetic 
gradient  if  the  paramagnetic  force  is  to 
overcome  backward  diffusion.  If  the  selec- 
tive migration  occurs  horizontally,  an  in- 
crease in  concentration  of  one  order  of 
magnitude  over  a  distance  of  10 ll  cm  is  a 
minimum  need.  The  diffusion  equation, 
relating  the  concentration  gradient  to  the 
selective  force  on  a  particular  kind  of 
atom,  then  shows  that  a  magnetic 
gradient  of  at  least  10~3  gauss/cm  is 
required.  This  is  about  10 3  times  the 
gradient  over  a  large  sunspot.  If  para- 
magnetic concentration  of  elements  actu- 
ally occurs — and  no  other  theory  has  been 
offered  to  maintain  the  segregation  of 
particular  elements — this  will  have  a 
decided  bearing  on  possible  models  of 
magnetic  stars. 

Radial  Velocities  of  Magnetic  Stars 

As  a  by-product  of  the  study  of  stellar 
magnetic  fields,  accurate  radial  velocities 
have  been  derived  by  H.  W.  Babcock  for 
several  A-type  stars,  most  having  very 
sharp  lines.  In  nearly  all  cases  the  spectro- 
grams, made  with  the  200-inch  telescope, 
have  a  dispersion  of  4.5  A/mm.  Velocities 
depend  on  upward  of  20  lines  measured  on 
each  plate  by  Miss  Burd.  For  some  36 
stars  brighter  than  magnitude  6.5,  the 
new  radial  velocities  substantially  aug- 
ment the  data  of  the  Yale  Bright  Star 
Catalogue.  Of  these  36  stars,  25  are  found 
to  have  variable  velocity.  For  most  of 
them  the  range  is  a  few  kilometers  per 
second.  For  two  of  these  stars,  HD  15144 
(HR  710)  and  HD  187474  (HR  7552), 
which  are  evidently  spectrosopic  binaries 
with  periods  of  2.9978  days  and  700  days, 
respectively,  Miss  Burd  has  computed 
orbital  elements. 


MOUNT      WILSON     AND      PALOMAR      OBSERVATORIES  23 

Stellar  Polarization  gathering   power   provides   an    excellent 

signal-to-noise  ratio. 

The  first  form  in  which  stellar  polariza-  The  new  polarimeter  is  of  an  uncon- 
tion  was  detected  was  the  circular  and  ventional  design,  developed  with  the 
elliptical  type  due  to  the  longitudinal  intent  of  overcoming  the  effects  of  stellar 
Zeeman  effect  on  line  profiles.  It  is  this  scintillation  and  of  the  nonuniformity  of 
that  permits  the  line-of-sight  component  the  cathode  of  multiplier  phototubes, 
of  the  star's  magnetic  field  to  be  meas-  These  effects  have  placed  a  serious  limita- 
ured.  But,  if  the  direction  of  the  stellar  tion  on  precision  of  measurement  with 
field  is  essentially  perpendicular  to  the  existing  stellar  polarimeters.  Scintillation, 
line  of  sight,  the  transverse  Zeeman  effect  closely  related  to  astronomical  seeing,  is 
can  be  expected.  If  the  absorption  lines  an  intensity  fluctuation  in  the  low- 
are  numerous  and  strong,  there  can  occur  frequency  range  ( <  500  cps) .  To  over- 
a  resultant  plane  polarization  of  the  light  come  it,  the  polarization  vector  of  the 
which  has  been  called  polarization  by  light  can  be  resolved  into  two  orthogonal 
magnetic  intensification.  Owing  to  an  components  that  are  chopped  at  a 
imbalance  in  equivalent  width  between  frequency  considerably  greater  than  the 
the  "perpendicular"  and  the  "parallel"  scintillation  frequency  and  admitted  al- 
components  of  the  saturated,  Zeeman-  ternately  to  a  single,  stationary  photo- 
broadened  profiles,  an  excess  of  polariza-  tube.  The  instrument  employs  a  slowly 
tion  in  a  plane  parallel  to  the  magnetic  rotating  electrooptic  crystal  ( ADP  plate) , 
field  will  result.  The  integrated  effect  in  a  excited  by  a  3500- volt  square  wave  at 
broad  region  of  the  spectrum  of  a  star  2000  cps.  The  crystal  becomes  birefrin- 
having  a  field  of  a  few  kilogauss  may  be  gent,  with  a  phase  shift  alternating  be- 
of  the  order  of  1  per  cent.  Plane  polariza-  tween  +90°  and  —90°  for  blue  light, 
tion  due  to  magnetic  intensification  This  is  followed  by  a  fixed  circular 
should  be  observable  not  only  in  sharp-  analyzer.  The  plane-polarized  components 
line  stars  but  also  in  those  with  lines  of  star  light  parallel  to  the  two  axes  of 
broadened  by  axial  rotation.  This  kind  of  the  crystal  are  alternately  transmitted  as 
polarization  should  be  distinguishable  these  axes  alternate  between  optically 
from  the  well  known  interstellar  polariza-  "fast"  and  "slow"  at  the  applied  fre- 
tion,  due  to  dust  grains  in  space,  because  quency.  The  output  of  the  photomulti- 
that  is  constant,  whereas  intrinsic  stellar  plier  is  amplified,  demodulated,  and 
polarization  may  be  expected  to  show  filtered  by  an  amplifier  of  the  so-called 
variations  in  intensity  and  in  position  "lock-in"  type,  which  has  already  proved 
angle  as  the  star  rotates  or  as  its  magnetic  indispensable  in  other  astronomical  in- 
field changes.  Observation  and  analysis  struments  required  to  measure  a  weak 
of  intrinsic  plane  polarization  should  give  signal  in  the  presence  of  noise, 
valuable  supplementary  data  on  which  to  The  ADP  crystal  is  rotated  at  a  rate  of 
base  models  of  magnetic  stars.  about  1  turn  in  5  minutes.  As  a  result,  the 

A  sensitive  polarimeter  for  the  investi-  demodulated  signal  produces  a  sine  wave 

gation  of  such  effects  has  been  under  on  a  strip-chart  recorder.  The  amplitude 

development  during  the  year  by  H.  W.  and  phase  of  the  sine  wave  are  readily 

Babcock,  after  initial  tests  with  a  simple  related  to  the  percentage  polarization  and 

rotating  analyzer  in  front  of  a  photom-  position  of  the  electric  vector  in  the  light 

eter  indicated  promise  for  this  approach,  of  the  source.  Calibration  is  accomplished 

The  instrument  is  designed  to  work  at  the  by  inserting  a  depolarizer  followed  by  a 

prime  focus   of  the   200-inch   telescope,  tilted  glass  plate  designed  to  introduce 

which  is  ideal  for  the  purpose  because  either  1  per  cent  or  4  per  cent  polarization, 

there    is    only    one    reflection     (nearly  Tests  of  the  polarimeter  during  various 

normal)    and    because    the    large    light-  stages  of  development  have  shown  that 


24 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


the  2-kc/sec  modulation  frequency  and 
the  very  narrow  band  width  of  the  lock-in 
amplifier  result  in  a  satisfactory  signal-to- 
noise  ratio,  and  that  a  precision  of  a  few 
hundredths  of  1  per  cent  can  be  obtained 
in  measures  of  the  polarization  of  stars 
brighter  than  about  seventh  magnitude. 
Observations  of  fainter  sources  are  gener- 
ally limited  by  shot  noise  (randomness  of 
arrival  of  the  incoming  photons)  rather 
than  by  scintillation.  It  is  satisfying  to 
find  that  no  detectable  plane  polarization 
is  introduced  by  reflection  from  the  200- 
inch  mirror.  An  upper  limit  at  present  is 
0.1  per  cent. 


Observations  of  some  80  stars  have  now 
been  obtained.  Practically  all  show  at 
least  a  small  degree  of  polarization,  and 
it  is  evident  that  considerable  care  will 
have  to  be  exercised  to  maintain  standard 
instrumental  conditions  and  consistent 
calibration  in  the  determination  of  small 
variations  due  to  intrinsic  stellar  causes. 
Preliminary  results  for  the  percentage 
polarization  of  a  few  stars  are:  WY 
Geminorum  1.75,  9  Geminorum  3.26,  R 
Leonis  2.47,  XX  Ophiuchi  5.15,  e  Ursae 
Majoris  0.06,  a2  Canum  Venaticorum 
0.04-0.11,  AD  Leonis  0.54-4.16,  HD 
153882  0.13-0.34,  and  HD  154445  4.0. 


GASEOUS   NEBULAE  AND  INTERSTELLAR  GAS 


A  program  has  been  initiated  by  O'Dell 
to  test  the  feasibility  of  using  the  doublet 
ratios  of  [Ar  IV],  XX4711,  4740,  and  [S  II], 
XX6717,  6731,  in  a  manner  analogous  to 
[O  II],  XX3726,  3729,  for  the  determina- 
tion of  the  electron  density  and  tempera- 
tures in  gaseous  nebulae.  The  former 
ratios  should  be  excellent  criteria  for 
nebulae  of  densities  above  104  electrons/ 
cm3,  where  (O  II)  becomes  insensitive. 
It  is  hoped  that  forthcoming  theoretical 
calculations  will  aid  in  placing  the 
calibration  on  a  reliable  basis.  Measures 
are  being  made  on  a  number  of  planetary 
nebulae  covering  a  large  range  in  densities 
with  both  photographic  and  photo- 
electric spectrographs  on  the  60-inch  and 
100-inch  telescopes. 

Although  the  spectra  of  the  planetary 
nebulae  have  been  the  subject  of  numer- 
ous photographic  studies,  substantial 
errors  exist  in  the  relative  intensities  of 
the  emission  lines,  due  to  the  inherently 
small  photometric  range  of  the  photo- 
graphic plate.  With  this  difficulty  in 
mind,  O'Dell  has  begun  a  systematic 
survey  of  the  bright  planetary  nebulae 
with  the  photoelectric  scanning  spectro- 
graph. Particular  attention  is  being  paid 
to  emission  lines  that  should  be  indicative 
of  the  conditions  in  these  nebulae.  Since 
the  effect  of  interstellar  extinction  can 
become     large     for     even     moderately 


reddened  nebulae,  the  interstellar  extinc- 
tion is  being  determined  from  observa- 
tions of  Paschen  and  Balmer  series  lines 
of  hydrogen  arising  from  the  same  upper 
level.  Although  the  detailed  recombina- 
tion theory  of  Burgess  predicts  a  small 
variation  in  the  ratio  of  these  Paschen 
and  Balmer  lines  with  electron  tempera- 
tures, the  interstellar  extinction  correc- 
tion should  be  much  more  accurate  than 
the  result  by  any  other  current  technique. 
As  part  of  this  program,  the  relative 
energy  distribution  of  the  continuum  from 
the  central  stars  is  also  being  determined. 

The  prints  of  the  National  Geographic 
Society-Palomar  Observatory  Sky  Survey 
plates  were  used  by  Struve  to  prepare  a 
list  of  74  interesting  nebulae  located  in 
and  near  obscuring  clouds  in  the  Milky 
Way.  Struve  finds  that  the  great  nebula 
near  Antares  extends  to  an  angular  dis- 
tance of  several  degrees  from  the  star. 
The  reddish  glow  appears  to  illuminate 
the  southern  edge  of  the  long,  opaque  lane 
which  extends  from  22  Scorpii  toward 
the  east. 

Several  faint  red  and  blue  nebulosities 
were  found  in  the  vicinity  of  p  Ophiuchi 
and  CoD  -24°12684.  Several  illuminat- 
ing stars  are  reduced  in  brightness  by  3 
mag,  but  no  luminous  nebulosity  could 
be  assigned  to  a  star  whose  light  is 
dimmed  by  more  than  3  mag. 


MOUNT      WILSON     AND      PALOMAR      OBSERVATORIES 


25 


The  nature  of  the  emission  nebulosity 
near  a  Scorpii  confirms  the  previous  result 
that  this  object  does  not  coincide  with  the 
reflection  nebula  belonging  to  the  same 
star. 

There  is  a  pronounced  tendency  for 
luminous  nebulosities  (probably  of  the 
reflection  type)  to  be  more  frequent  in 
the  Taurus  complex  of  dense  clouds  than 
in  the  Ophiuchus-Scorpius  complex.  The 
quantity  discussed  is  the  number  of 
nebulosities  per  square  degree  of  dark 
cloud.  Possibly  this  is  connected  with  the 
fact  that  the  average  absorption  of  the 
clouds  is  smaller  in  Taurus  than  in  the 
Ophiuchus-Scorpius  complex. 

Parker  has  continued  his  study  of  S22, 
S147,  NGC  6888,  IC  443,  and  the  Cygnus 
Loop,  which  are  possible  supernovae 
remnants.  The  observing  program,  which 
has  been  completed,  includes  spectro- 
grams and  spectral  scans  of  16  separate 
filaments  in  these  objects.  On  the  pro- 
gram of  reduction,  relative  intensities  of 
all  lines,  corrected  under  various  assump- 
tions for  reddening,  have  been  obtained. 
For  the  condition  of  collisional  excitation 
and  ionization,  relative  intensities  have 
been  computed  for  the  emission  lines  in- 
volved for  several  values  of  the  electron 
pressure  and  electron  temperature.  Com- 
parison of  the  computed  and  observed 
ratios  will  enable  statements  to  be  made 
about  the  temperature,  density,  and 
abundances  in  the  filaments.  Monochro- 
matic net  fluxes  are  also  being  measured 


for  the  objects,  both  to  investigate  the 
mass  of  the  visible  filaments  and  to  inves- 
tigate the  amount  of  free-free  radio 
emission  that  could  be  expected.  The 
dynamics  of  the  five  objects  has  also 
been  examined. 

G.  Munch  and  Wilson  have  prepared 
a  reply  to  the  criticism  raised  by  K. 
Wurm  against  the  model  of  the  Orion 
nebula  proposed  earlier  by  them.  The 
lack  of  agreement  between  the  radial 
velocities  of  the  He  I  nebular  absorption 
lines  and  the  emission  lines  at  the  same 
position,  the  essence  of  Wurm's  criticism, 
is  explained  in  terms  of  the  density  fluc- 
tuations existing  in  the  nebula.  Wilson 
and  Munch  have  therefore  reviewed  the 
variety  of  observational  data  related  to 
such  density  fluctuations  and,  within  this 
framework,  have  discussed  the  radial- 
velocity  data  provided  by  their  early 
observations.  On  the  whole,  they  find  that 
the  model  does  not  need  as  drastic  a  re- 
vision as  Wurm  proposes. 

G.  Munch  and  Dr.  A.  Unsold  observed 
that  the  star  a  Ophiuchi,  at  a  distance  of 
25  parsecs,  shows  a  K-line  component 
undoubtedly  of  interstellar  origin.  The 
observations  of  other  near-by  stars  in  the 
same  area  of  the  sky  lead  them  to  infer 
that  the  interstellar  cloud  in  front  of  a 
Oph  has  linear  dimensions  no  larger  than 
1  parsec.  The  study  of  this  near-by  com- 
plex of  interstellar  matter,  which  may 
extend  right  to  the  sun,  is  being  continued 
by  observing  additional  near-by  stars. 


GALAXIES 


Structure  and  Internal  Motions  of  the 
Galaxy 

A  determination  of  the  solar  motion 
and  the  parameter  A  of  differential 
Galactic  rotation  from  cepheid  variables 
is  being  finished  by  Schmidt  and  Kraft. 
The  total  solar  motion  is  found  to  be  16  or 
17  km/sec,  rather  less  than  the  value  21 
km/sec  adopted  by  Blaauw  and  Morgan. 
The  value  of  A  is  found  to  be  14  to  15 
km/sec  kpc.  No  reliable  determination 
of  the  curvature  of  the  angular  velocity 


curve  can  be  made.  A  small  negative  K 
term  is  found,  but  the  observations  are 
equally  well  represented  by  a  constant 
error  of  about  —3  km/sec  in  the  radial 
velocities. 

The  (U,  V,  TF)-velocity  vectors  for 
221  well  observed  dwarf  stars  have  been 
used  by  Eggen,  Lynden-Bell,  and  San- 
dage  to  compute  the  eccentricities  and 
angular  momenta  of  the  galactic  orbits  in 
a  model  galaxy.  It  is  shown  that  the 
eccentricity  and  the  observed  ultraviolet 


26 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


excess  are  strongly  correlated.  The  stars 
with  the  largest  excess  (i.e.,  the  lowest 
metal  abundance)  are  invariably  moving 
in  highly  elliptical  orbits,  whereas  stars 
with  little  or  no  excess  move  in  nearly 
circular  orbits.  Correlations  also  exist 
between  the  ultraviolet  excess  and  the 
TT-velocity.  Also,  the  excess  and  the 
angular  momentum  are  correlated;  stars 
with  large  ultraviolet  excesses  have  small 
angular  momenta.  These  correlations 
have  been  discussed  in  terms  of  the 
dynamics  of  a  collapsing  galaxy.  The  data 
require  that  the  oldest  stars  were  formed 
out  of  gas  falling  toward  the  galactic 
center  in  the  radial  direction  and  collaps- 
ing from  the  halo  onto  the  plane.  The 
collapse  was  very  rapid,  and  only  a  small 
number  X  10 8  years  was  required  for  the 
gas  to  attain  orbits  in  equilibrium  (i.e., 
gravitational  attraction  balanced  by  cen- 
trifugal acceleration).  The  scale  of  the 
collapse  was  tentatively  estimated  to  be 
at  least  10  in  the  radial  direction  and  25 
in  the  Z  direction.  The  initial  contraction 
must  have  begun  near  the  time  of  forma- 
tion of  the  first  stars,  some  10 10  years  ago. 

In  connection  with  the  study  of  the 
collapsing  galaxy,  mentioned  above, 
Eggen  has  prepared  for  publication  a 
catalogue  of  some  700  stars  whose  space 
motions  with  respect  to  the  sun  are  almost 
certainly  greater  than  100  km/sec.  In 
addition  to  astrometric  and  photometric 
data,  the  catalogue  contains  the  values  of 
the  "modular  velocities"  from  which  new 
space  motion  vectors  can  be  computed 
for  any  future  changes  in  the  available 
radial  velocity,  proper  motion,  or  lumi- 
nosity estimates  of  these  objects. 

The  globular  cluster  system  in  our  own 
Galaxy  has  been  reanalyzed  by  Arp.  The 
globular  clusters  in  the  Galaxy  are  shown 
to  be  90  per  cent  discovered.  The  best 
available  moduli  with  the  new  RR  Lyrae 
zero  point  (Mv=  +0.3,  MB  =  +0.5  mag) 
give  a  distance  to  the  center  of  the  Galaxy 
of  Ri  =  9.9  kpc  d=  0.5  kpc  (minimum 
error  for  an  error  of  0.1  mag  in  RR 
Lyrae  absolute  magnitude) . 

The  analysis  of  the  stars  in  Baade's 


field  near  the  Galactic  center  has  been 
completed  by  Arp.  There  is  a  definite 
giant  branch  which  clearly  emerges  in  the 
color-magnitude  diagram  and  represents 
the  population  in  the  nucleus  or  nuclear 
bulge.  This  giant  branch  is  not  made  up 
of  globular  clusterlike  giants.  More 
detailed  consideration  of  this  diagram  will 
be  made  shortly. 

The  material  gathered  by  Guido  and 
Luis  Munch  for  the  determination  of 
motions  and  distances  of  faint  OB  stars 
in  directions  near  that  of  the  Galactic 
center  has  been  studied  and  reduced. 
All  together,  they  determined  U,  B,  V 
colors  and  spectral  types  for  35  stars  and 
radial  velocities  for  the  24  objects  of 
higher  luminosity  from  90  plates  obtained 
in  a  variety  of  dispersions  with  the  X 
spectrograph  and  the  coude  spectro- 
graphs. This  material  is  now  being  pre- 
pared for  publication  and  will  be  dis- 
cussed with  the  earlier  results  obtained 
by  them,  as  well  as  with  the  data  for  the 
interstellar  lines  in  these  objects  obtained 
during  the  past  few  years. 

Rotation  and  Internal  Motions  of  Galaxies 

The  rather  extensive  observations  of  the 
internal  motions  in  the  elliptical  galaxy 
NGC  3115  obtained  by  Minkowski  and 
Oort  in  1958  and  extended  by  Minkowski 
during  the  following  observing  season 
were  discussed  by  Oort. The  principal  aims 
were  to  find  the  distribution  of  mass  and 
to  obtain  data  on  the  distribution  of 
random  motions  in  the  galaxy.  It  appears 
that,  whereas  up  to  a  distance  of  about  60 
seconds  from  the  center  along  the  major 
axis  the  distribution  of  mass  seems  to  be 
roughly  the  same  as  that  of  the  light,  the 
mass  density  in  the  outermost  parts 
observed  decreases  more  slowly  than  the 
light  density.  The  ratio  of  mass  density  to 
light  density,  M/L,  expressed  in  terms  of 
the  mass  and  light  of  the  sun  as  units,  is 
about  15  in  the  inner  parts,  and  rises  to 
values  of  100  in  the  outer  shells.  For  the 
entire  galaxy,  M/L  was  found  to  be 
about  60.  Such  a  high  value  has  never  yet 
been  found  from  observations  of  rotation, 


MOUNT      WILSON     AND      PALOMAR      OBSERVATORIES 


27 


presumably  because  they  have  never 
reached  such  large  distances.  The  result 
is  important  because  of  its  bearing  on  the 
problem  of  stability  of  groups  and 
clusters  of  galaxies. 

The  preliminary  reduction  of  the  Ha 
plates  taken  by  Brandt  with  the  Mount 
Wilson  B  spectrograph  for  the  purpose  of 
redetermining  the  rotation  curve  of  M  33 
is  complete.  The  rotation  curve  deter- 
mined optically  extends  to  30  minutes 
from  the  center  of  M  33,  but  it  can  be 
extended  60  minutes  from  the  center 
with  the  aid  of  the  radio  observations. 
The  shape  of  the  rotation  curve  is  very 
similar  to  that  of  M  31 ;  hence,  the  relative 
mass  and  density  distributions  should  be 
similar.  For  an  assumed  distance  of  630 
kpc,  the  approximate  mass  becomes 
2.3  X  1010  Mo. 

Emission  Nebulae  in  Galaxies 

Work  done  by  Baade  on  the  precise 
position  of  all  emission  nebulae  in  M  31 
is  being  prepared  for  publication  by 
Arp.  An  effort  will  be  made  to  give  illus- 
trative data  on  the  connection  between 
these  emission  nebulae  and  the  spiral 
structure  in  M  31. 

Another  attempt  was  made  by  Schmidt 
to  determine  the  helium  abundance  of  the 
interstellar  gas  in  H  II  regions  at  various 
positions  in  the  Andromeda  galaxy.  The 
observations  were  planned  to  furnish 
absolute  values  of  the  helium-to-hydrogen 
abundance  ratio  and  the  diameters  of 
both  the  hydrogen  emission  region  and 
the  helium  emission  region.  The  first 
objective  was  attained  by  plates  taken 
in  the  Orion  nebula,  where  Mathis  has 
obtained  an  absolute  calibration.  Expo- 
sures on  the  H  II  regions  in  Andromeda, 
of  which  one  was  a  multinight  exposure, 
were  carefully  guided  with  the  spectro- 
graph slit  over  the  H  II  region  and  a 
near-by  star.  The  star  spectrum  allows 
evaluation  of  effects  of  seeing,  guiding, 
and  scattering  in  the  photographic  plate, 
and  the  diameters  of  the  emission  regions 
are  corrected  for  these  effects.  Once 
these  diameters  are  determined  for  both 


hydrogen  and  helium  emission,  the 
abundance  ratio  can  be  determined 
independent  of  a  possible  variation  with 
abundance  ratio  of  the  far  ultraviolet 
continua  of  the  exciting  stars.  The  obser- 
vational problem  is  a  marginal  one  for  the 
200-inch  prime  focus  spectrograph,  and 
the  photographic  photometry  on  the 
plates  is  difficult.  Results  thus  far  give  a 
helium-to-hydrogen  number  ratio  of  0.08 
at  89  minutes  from  the  center  of  M  31, 
0.14  at  70  minutes,  and  0.17  at  25 
minutes.  Each  determination  is  uncertain 
by  a  factor  between  V/2  and  2,  so  that  the 
variation  as  well  as  the  difference  from 
Orion  nebulae  (0.13)  is  hardly  significant. 
The  implication  of  an  essentially  constant 
helium  abundance  would  be  far  reaching, 
as  was  briefly  indicated  in  the  Annual 
Report  for  last  year. 

The  observation  of  the  emission  lines  of 
[O  II]  in  the  nuclear  region  of  M  31  was 
continued  by  G.  Munch.  An  area  approxi- 
mately rectangular  with  dimensions  100 
X  300  seconds  centered  at  the  nucleus 
has  now  been  covered  with  slits  at  various 
position  angles.  The  emission  lines  at 
greater  distances  from  the  nucleus  be- 
come so  faint  that  it  is  impractical  to 
attempt  to  follow  them  with  existing 
equipment.  The  material  so  far  collected, 
therefore,  will  be  discussed  shortly  and 
published. 

Munch's  lack  of  success  in  detecting 
emission  lines  in  the  patches  which  were 
observed  by  Baade  in  the  disk  of  NGC 
4594,  and  which  Lindblad  considered  to 
be  H  II  regions,  was  reported  last  year. 
Confirming  the  tentative  explanation 
then  given,  it  has  been  found  that  an  Ha 
exposure  through  an  interference  filter, 
with  the  /:3.67  focal  ratio  of  the  200- 
inch,  does  not  show  such  features  with 
stronger  contrast  than  broad-band  photo- 
graphs do.  The  patches  undoubtedly, 
then,  do  not  have  an  emission  spectrum, 
and  their  true  nature  raises  an  important 
problem.  Further  work  on  these  objects 
is  being  planned  with  a  photoelectric 
scanner. 

A    systematic    search    for    planetary 


28 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


nebulae  in  other  galaxies  in  the  local 
system  is  being  executed  by  O'Dell,  by 
means  of  photographic  plate  plus  filter 
combinations  that  give  sensitivities 
around  the  Ha,  Nl  +  N2,  and  X5400 
regions,  since  the  nebulae  should  appear 
in  the  strong  emission  regions  and  not  in 
the  relatively  line-free  region  at  X5400. 
Thus  far,  four  suspected  nebulae  have 
been  found  in  the  Leo  I  system. 

A  program  of  direct  photography  of 
nearly  resolved,  Irr,  Sc,  and  late  Sb 
galaxies  whose  redshifts  are  less  than 
2000  km/sec  was  begun  by  Sandage  using 
a  4  by  4  inch  Ha  filter  of  80  A  half -width. 
The  purpose  is  to  isolate  the  H  II  regions 
and  to  measure  their  apparent  diameter 
as  distance  indicators.  Absolute  calibra- 
tion of  the  linear  diameter  will  be  made 
using  galaxies  in  the  local  group,  such  as 
the  Large  and  Small  Magellanic  Clouds, 
M  33,  IC  1613,  and  NGC  6822,  whose 
distances  are  known  from  the  cepheid 
variables.  Preliminary  calibration  during 
the  report  year  gave  the  linear  diameter 
of  the  largest  H  II  region  as  245  parsecs 
and  the  mean  diameter  of  the  first  five 
largest  as  175  parsecs.  Plates  were  ob- 
tained of  NGC  2403,  M  101,  NGC  925, 
NGC  2903,  and  M  51  in  a  first  trial  of  the 
problem,  and  it  will  be  feasible  to  measure 
diameters  in  galaxies  with  redshifts  as 
large  as  2000  km/sec.  The  ultimate  aim 
of  the  problem  is  to  improve  the  values  of 
the  Hubble  constant  (H).  Preliminary 
results  using  H  II  sizes  determined  in 
NGC  925  and  NGC  4321  show  that  H 
lies  in  the  neighborhood  of  75  to  100 
km/sec  106  psc,  in  agreement  with  Sersic's 
earlier  study  of  the  H  II  region  problem 
using  blue  plates  of  galaxies  taken  with 
the  200-inch,  and  in  agreement  with  the 
value  of  75  km/sec  106  psc  which  has 
been  generally  used  during  the  past 
several  years. 

Widened  spectra  of  NGC  1068  and 
NGC  4151  have  been  obtained  by 
Sargent  at  a  dispersion  of  86  A/mm. 
Exposures  of  stars  whose  continuous 
energy  distribution  have  been  obtained 
with  the  Cassegrain  scanner  have  been 


made  on  the  same  plates.  Line  profiles 
have  been  measured  for  NGC  1068.  In 
NGC  4151  the  broad  wings  at  the  Balmer 
lines,  which  were  reported  by  Seyfert, 
extend  to  ±5000  km/sec.  There  are  no 
absorption  lines  in  NGC  4151  that  can 
be  attributed  to  stars.  Two  strong  absorp- 
tion features  which  occur  at  X3885  and 
X3732  and  which  are  probably  to  be 
identified  with  He  I  are  displaced  by 
about  —200  km/sec  relative  to  the 
emission  line  centers.  NGC  4151  is  seen 
almost  face-on,  and  this  indicates  that 
material  is  being  expelled  from  the 
nucleus  along  the  direction  of  the  axis  of 
rotation. 

Variable  Stars  in  Galaxies 

Sandage  completed  the  photoelectric 
measurements  of  selected  stars  of  Baade's 
variable  star  sequences  in  IC  1613  in 
order  to  put  Baade's  extensive  photo- 
metric material  of  the  cepheids  in  this 
galaxy  on  the  Progson  scale.  The  photo- 
electric sequence  extends  from  B  =  12m 
to  B  =  22™0,  which  is  not  yet  faint 
enough  to  correct  the  cepheid  photom- 
etry at  minimum  light  but  is  faint 
enough  to  correct  all  Baade's  data  at 
maximum  light.  This  has  been  done,  with 
the  result  that  the  slope  of  the  period- 
luminosity  relation  for  IC  1613  is  identical 
with  that  found  by  Arp  for  the  Small 
Magellanic  Cloud.  The  relation  is 

J5max  =  22.61  -  2.25  log  P 

which  gives  an  apparent  modulus  of 
(m  —  M)B  =  24™33,  using  Kraft's  zero- 
point  calibration  of  the  cepheids  in 
galactic  clusters. 

Extensive  U,  B,  V  photometry  of  field 
stars  in  the  direction  of  IC  1613  indicates 
a  reddening  due  to  our  Galaxy  of  E(B  — 
V)  =  0?03,  which  gives  (m  -  M)0  = 
24m2  for  the  true  modulus  of  IC  1613. 
Sandage  expects  to  extend  the  photo- 
electric sequence  to  B  =  23ni5  next 
season.  All  Baade's  material  will  then  be 
published  on  this  photometric  scale. 

Extensive  material  on  three  variable- 
star   fields   in   M   31,    accumulated   by 


MOUNT      WILSON     AND      PALOMAR     OBSERVATORIES  29 

Baade  during  the  past  ten  years,  is  being  to  be  nonvariable.  For  22  variables  in  the 

analyzed    by    Miss    Swope.    There    are  same  area,  periods  could  be  derived  from 

about  120  variables  in  a  field  15  minutes  the  brightness  estimates.   Among  these 

of  arc  from  the  nucleus,  330  variables  in  22,  16  belong  to  Bailey's  type  (a  +  b). 

a  field  45  minutes  from  the  center,  and  The  mean  period  of  these  variable  stars  is 

more   than   50   that   are   in   a   field   96  P  =  0"?617,  almost  the  same  mean  period 

minutes  south  preceding  the  nucleus.  This  as  Miss  Swope  found  in  the  Sculptor-type 

last  field  is  the  only  one  which  is  suitable  system  in  Draco.  For  6  variable  stars, 

for  precision  photometry  and  in  which  van  Agt  determined  periods  shorter  than 

there  is  a  photoelectric  sequence  (Arp,  0?45,  but,  since  the  scatter  in  the  esti- 

Year  Book   58).  In  this  field  there  are  20  mated  brightness  relative  to  the  bright- 

cepheids   with    both    photographic    and  ness-amplitude  is  larger  for  variable  stars 

photovisual  light  curves  and  also  a  pre-  with  smaller  amplitude,  these  periods  are 

liminary  color-magnitude  diagram.  Miss  established   with   less   certainty.    For   4 

Swope  finds  that  the  apparent  distance  variable  stars  in  the  group  of  28  it  has  not 

modulus  of  M  31  is  B  =  24™75  or  V  =  yet  been  possible  to  determine  a  period. 

24^60,    assuming   a   zero   point   of    the  One  bright  variable  star  at  the  northern 

cepheids  from  Arp  and  Kraft's  period  border  of  the  system  was  also  estimated 

luminosity    curves.    There    is    probably  by  way  of  the  eyepiece  method.  From  a 

about  0™15  general  reddening  due  to  our  limited  number  of  plates,  the  period  of 

Galaxy,  which  gives  a  distance  modulus  this  star  was  determined  to  be  2?697.  A 

corrected  for  absorption  of  24™  15  in  both  rough   estimate   of   the   maximum    and 

B  and  V  for  M  31.  minimum  B  magnitude  was  max  =  18.2, 

The    slope    of    the    period-luminosity  min  =  18.7.  Scale  transfers  for  blue  plates 

curve  is  essentially   the   same    as    that  only  are  available  for  the  system  in  Ursa 

found  by  Arp  for  the  Small  Magellanic  Minor.  From  a  scale  transfer  of  Baum's 

Cloud.  sequence  in  M  13  a  provisional  magnitude 

Baade  also  obtained  a  series  of  plates  of  sequence    in    the    Sculptor-type    system 

the  Sculptor-type  system  in  Ursa  Minor,  could  be  established.  The  median  magni- 

These  plates  are  being  assessed  by  Dr.  tude  of  8  variable  stars  the  periods  of 

S.    L.    Th.    J.    van   Agt   of   the   Leiden  which   are    known    and    for   which    the 

Observatory.  In  this  system  91  variable  brightness  estimates  were  transferred  into 

stars  are  known.  They  were  discovered  on  magnitudes    is    20™04.    Because    of    the 

plate  pairs,  each  of  which  covered  only  scatter    in    the    light    curves    of    the    8 

part  of  the  system.  The  outlines  of  the  variables,   all  within   6  minutes  of  the 

Sculptor-type   system   as   inferred   from  center  of  the  system,  this  mean  value  of 

the  positions  of  the  variable  stars  are  the   median   magnitudes   is   still   rather 

roughly  elliptical  and  indicate  dimensions  uncertain. 

on  the  sky  of  60  by  22  minutes.  About  50  All  stars  visible  on  a  plate  with  an 

per  cent  of  the  variable  stars  discovered  103aO  emulsion  and  exposed  with  a  WG2 

in  this  system  are  concentrated  on  the  filter  in  front  of  the  plate  were  measured 

plates    that    cover    the    center    of    this  in   the   iris   photometer   of   the   Leiden 

Sculptor-type   system.    The   number   of  Observatory.  In  addition,  the  same  stars 

plates  is  large  enough  to  determine  the  were  measured  on  a  plate  with   103aD 

periods  of  the  cluster-type  variable  stars  emulsion  and  exposed  with  a  GG11  filter, 

in  the  central  area  only.  Estimates  of  the  This    plate    was    exposed    immediately 

magnitudes  for  28  stars  in  the  central  after  the  first  one  mentioned.  No  scale 

area  were  obtained  by  way  of  inspection  transfers  for  either  of  these  two  plates 

with  an  eyepiece  and  comparison  with  exist.  The  iris  scale  readings  were  used, 

a  sequence  of  close-by  comparison  stars,  therefore,   to   construct   a   pseudo-color- 

Two  of  the  investigated  stars  turned  out  magnitude  diagram  for  the  central  part 


30 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


of  the  Sculptor-type  system.  All  stars 
used  in  this  diagram  are  located  within 
305  seconds  of  the  center  of  the  plates 
measured.  These  plates  cover  the  center 
of  the  Sculptor-type  system,  and,  since 
the  center  of  the  system  and  the  plate 
centers  are  very  close  together,  the 
pseudo-color-magnitude  diagram  repre- 
sents the  general  trend  in  the  central  part 
of  the  system.  The  pseudo-color-magni- 
tude diagram  shows  a  distinct  giant 
branch  and  a  clearly  developed  horizontal 
branch.  In  the  region  where  the  horizontal 
branch  meets  the  upgoing  giant  branch, 
the  number  of  stars  is  considerably  lower 
than  in  the  part  of  the  horizontal  branch 
immediately  before  that  region.  To  the 
blue  end  of  the  horizontal  branch  only 
one  star  was  found.  The  ratio  of  the 
number  of  stars  at  the  blue  end  and  at  the 
red  end  of  the  horizontal  branch  is  about 
1 :  50.  The  position  of  the  gap  of  the  short- 
period  cluster-type  variable  stars  is  such 
that  only  the  one  blue  star  mentioned  is 
found  to  the  blue  side  of  the  gap. 

The  faint  part  of  the  giant  branch  sets 
in  at  about  1%  m&g  below  the  horizontal 
branch.  At  the  point  of  bifurcation  there 
is  no  clear  indication  of  a  doubling  of  the 
giant  branch.  The  plates  used  for 
the  construction  of  the  pseudo-color- 
magnitude  diagram  were  exposed  under 
conditions  of  poor  seeing.  Much  of  the 
scatter  in  the  branches  of  the  diagram  is 
due  to  the  low  quality  of  the  two  plates. 
The  scatter  outside  the  branches  is  low; 
therefore,  we  may  conclude  that  the 
number  of  faint  field  stars  entering  the 
diagram  is  limited. 

Photometry  and  Stellar  Content  and 
Evolution 

The  distribution  of  stars  in  a  number  of 
local  group  dwarf  galaxies  was  studied 
by  Hodge.  These  include  the  Fornax, 
Sculptor,  Leo  I,  Leo  II,  Ursa  Minor, 
Draco,  NGC  147,  NGC  185,  and  NGC 
205  systems.  In  all,  the  projected  density 
was  found  to  obey  Hubble 's  law  in  the 
central  regions  but  to  fall  off  more  steeply 


in  the  outer  regions.  A  cutoff  is  found 
which  corresponds  in  location  to  the 
expected  tidal  cutoff.  Photoelectric  sur- 
face luminosity  distributions  in  two 
colors  were  obtained  for  other  dwarf 
galaxies,  specifically  the  irregular  systems 
NGC  6822,  IC  1613,  Sextans,  and  WLM. 

Photoelectrically  calibrated  measures 
of  the  distribution  of  luminosity  and 
color  have  been  made  by  Hodge  for  29 
galaxies  primarily  of  the  SO  type.  The 
results,  not  yet  completely  reduced,  show 
a  clear  difference  in  physical  properties 
between  Sandage's  subgroup  in  this  class 
as  given  in  The  Hubble  Atlas. 

Two  near-by  dwarf  galaxies  were  ob- 
served photoelectrically  by  Baum  during 
the  report  year.  The  integrated  light  of 
these  very  faint  tenuous  systems  must 
be  measured  by  making  a  series  of  slow 
scans  across  them  with  a  photoelectric 
photometer.  For  this  purpose  the  prime- 
focus  photometer  at  the  200-inch  has 
been  equipped  with  a  scanning  motor. 

The  purpose  in  observing  these  dwarfs 
is  to  obtain  their  color  indices  and 
absolute  magnitudes  so  as  to  fit  them 
into  a  color-luminosity  diagram  for  ellip- 
tical galaxies.  This  diagram  divides 
itself  into  two  color-index  groups,  the 
large  elliptical  galaxies  having  color  in- 
dices 0.2  to  0.3  mag  redder  than  the  dwarf 
systems.  All  evidence  indicates  that  the 
dwarfs  are  truly  Population  II,  whereas 
eight-color  observations  show  that  large 
ellipticals  must  be  mainly  old  Population 
I.  The  newly  observed  dwarfs  add  two 
points  where  they  are  most  needed  in  the 
difficult  part  of  the  color-luminosity 
diagram.  One  of  these,  NGC  6822,  is 
clearly  in  the  true  dwarf  class,  whereas 
the  other,  NGC  185,  is  in  the  transition 
region  between  classes. 

Any  interpretation  of  the  redshift- 
magnitude  relation  in  terms  of  cosmologi- 
cal  models  depends  on  the  way  in  which 
evolutionary  effects  are  taken  into  ac- 
count. Since  individual  stars  undergo 
large  changes  in  luminosity  and  tempera- 
ture as  they  age,  the  integrated  light  of  a 
galaxy  will  tend  to  change  with  time. 


MOUNT      WILSON     AND      PALOMAR      OBSERVATORIES  31 

Owing  to  the  light-travel  time,  distant  center  of  an  arm,  weakest  near  the  inner 

galaxies  are  seen  at  an  earlier  age  than  edge  of  an  arm,  and  intermediate  near  the 

near-by  ones.  It  is  therefore  necessary  to  outer  edge.  The  wavelength  dependence 

know  something  about  the  stellar  content  is  such  that  asymmetry  in  the  dust  lanes 

of  galaxies  of  the  kind  used  for  redshift-  does  not  by  itself  appear  to  be  an  ade- 

magnitude  work.  quate  explanation.  Evidently  the  relative 

Photoelectric  observations  on  the  eight-  number  of  early-type  stars  tapers  off  less 
color  system  were  used  in  1958  for  com-  sharply  at  the  outer  edge  than  at  the 
puting  a  population  model  for  large  inner  edge.  Such  would  be  the  situation 
elliptical  galaxies.  At  4830  A  the  model  if  the  rotating  disk  of  the  galaxy  slips 
could  be  described  as  25  per  cent  Popula-  through  the  arms,  dragging  them  slightly 
tion  II  plus  75  per  cent  old  Population  I.  and  coiling  them  up. 
During  the  current  year,  a  first  attempt  The  evolution  of  the  integrated  proper- 
was  made  by  Baum  to  extend  this  eight-  ties  of  clusters  of  stars  has  been  computed 
color  population  analysis  to  sample  by  Arp.  The  total  colors  and  magnitudes 
regions  in  spiral  galaxies.  Experimental  of  groups  of  stars  of  different  ages  have 
results  were  obtained  by  scanning  across  been  derived.  The  results  indicate:  (1) 
selected  regions  of  M  74  (NGC  628),  The  E  and  SO  galaxies  contain  about  30 
which  is  a  face-on  Sc  spiral  of  unusually  per  cent  more  K7  dwarfs  (or  equivalent) 
good  symmetry.  As  expected,  the  disk  by  number  than  galactic  or  globular 
population  underlying  the  inner  spiral  cluster  populations.  This  conclusion  sup- 
arms  has  a  spectral  energy  distribution  ports  evidence  from  M/L  ratios  and 
roughly  similar  to  that  of  an  elliptical  computations  by  Spinrad  on  the  com- 
galaxy.  The  outer  regions  of  the  disk,  posite  spectra.  (2)  The  total  increase  in 
however,  are  evidently  bluer.  The  arms  magnitude  from  a  "young"  E  or  SO 
themselves,  although  photographically  galaxy  to  brightness  at  10 10  years  is 
impressive,  contribute  astonishingly  little,  about  4  magnitudes.  The  brightness-age 
perhaps  10  per  cent  photo  visually,  to  the  curve  also  enables  the  evolutionary 
total  light  of  M  74.  The  width  of  an  magnitude  correction  for  large  redshifted 
individual  arm  is  found  to  increase  nebulae  to  be  read  off.  This  completely 
systematically  with  wavelength,  the  in-  empirical  method  yields  the  same  K 
frared  width  being  about  1.5  times  the  correction  as  was  computed  by  Sandage 
ultraviolet  width.  for  the  latest  cosmological  solution  from 

The  most  interesting  feature  of  Baum's  the    redshift    data.    (3)    By    classifying 

scans  is  a  slight  asymmetry  in  the  color  galaxies  in  the  two  nonevolving  param- 

distribution   across   a   spiral   arm.    This  eters  of  mass   and   angular   momentum 

provides  a  clue  to  the  solution  of  an  old  it  is  shown  that  evolution  probably  does 

dilemma.  Since  its  birth,  a  typical  spiral  not  take  place  from  one  type  of  galaxy 

galaxy  like  M  74  has  had  time  for  more  into  another.  It  is  suggested  that  only 

than  50  revolutions  of  its  disk,  but  its  relatively  slowly  rotating  masses  contract 

arms  are  coiled  to  the  extent  of  less  than  into  E0  giants,  and  higher  rotation  in  the 

two  visible  turns.  If  the  arms  rotate  with  more-flattened    galaxy    types    sets    in- 

the  disk,  the  outer  ends  of  the  arms  must  creasingly   smaller  mass   limits.    Groups 

share  most  of  the  rotation.  If,  on  the  other  and  clusters  of  galaxies  are  considered, 

hand,  the  outer  ends  of  the  arms  are  tied  and  it  is  shown  that  a  high  density  (in  a 

to  intergalactic  magnetic  fields,  the  disk  cluster  of  galaxies)  is  strictly  correlated 

must  slip  through  the  arms,  which  are  with   E   and   SO   membership,    and   low 

zones  of  new-star  formation,  with  only  a  average  density  with  high  membership,  of 

small  drag.  The  clue  is  this:  Relative  to  spirals.  This  gives  a  physical  explanation 

other  colors  the  strengths  of  violet  and  for  Hubble's  often-made  statement  that 

ultraviolet  are  found  to  be  greatest  at  the  spirals  tend  to  be  field  nebulae.  The  above 


32 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


observations  on  cluster  composition  are 
interpreted  as  confirmation  of  the  hy- 
pothesis about  formation. 

Recomputation  by  Arp  of  Seare's  early 
work  using  modern  magnitude  scales 
yields  a  surface  brightness  of  the  Galaxy 
in  the  solar  neighborhood  of  8b  =  23.8 
mag/sq  sec  (perpendicular  to  plane  with 
absorption  layer  allowed  for).  The  com- 


parable surface  brightness  occurs  11  kpc 
from  the  center  of  M  31,  assuming  cose- 
cant reddening  models  for  both  galaxies. 
It  is  also  shown  that  with  such  reddening 
models  the  conclusion  of  Kron  and  Mayall 
is  incorrect  and  that  the  M  31  globular 
clusters  are,  in  fact,  the  same  intrinsic 
color  as  in  our  own  Galaxy.  The  following 
comparison  has  been  made: 


Mass,  Mq 

Radius  (to  solar-brightness  isophote),  kpc 

Number  of  clusters 

Per  cent  of  gas 

M31 

34  X  1010 

11 

300-400 

0.7 

Milky  Way 

7  X  1010 

10 

130 

4-6 

All  these  characteristics  indicate  that 
our  own  Galaxy  is  more  like  an  Sc  than 
M  31  is,  and  of  course  the  Milky  Way  is 
more  like  an  Sb  than  M  33  is,  but  it 
seems  difficult  to  classify  our  own  system 
more  quantitatively  at  present. 

Work  is  being  continued  by  Oke  on  the 
measurement  of  absolute  energy  distribu- 
tions in  the  spectra  of  the  central  regions 
of  galaxies.  Measurements  are  now  com- 
plete between  X3400  and  X6000  for  about 
20  galaxies.  Nearly  all  observations  have 
been  made  on  giant  elliptical  systems. 
The  two  brightest  galaxies  and  two 
fainter  ones  in  the  Coma  cluster  have 
been  observed.  The  data  are  being  used  to 
compute  K  corrections  for  distant  galaxies 
and  for  studying  the  stellar  content  of 
ellipticals. 

Catalogue  of  Galaxies  and  of  Clusters 
of  Galaxies 

Volume  I  of  the  Catalogue  of  Galaxies 
and  of  Clusters  of  Galaxies  by  Zwicky, 
Herzog,  and  Wild  was  published  in 
October  1961.  In  the  meantime,  work  on 
volume  II  has  progressed  to  the  point  that 
the  data  on  all  clusters  (about  2500) 
have  been  analyzed  and  prepared  for 
publication,  and  the  work  on  the  galaxies 
involved  is  expected  to  be  finished  in 
October  1962.  Volume  II  covers  the  area 
from  Milky  Way  to  Milky  Way  between 
the  declinations  +15°  and  +35°.  Volume 
III,  which  covers  the  south  galactic  cap 
north    of    declination     —3°,    has    been 


started.  This  project  was  supported  in 
part  by  the  National  Science  Foundation. 

The  data  of  the  catalogue  have  been 
reduced  in  a  dozen  different  ways  and 
have  been  used  to  test  for  interstellar  and 
intergalactic  absorption.  The  fact  has  also 
been  confirmed  that  clustering  among 
galaxies  is  universal  and  is  statistically 
of  the  same  nature  at  all  distances  up  to 
redshifts  corresponding  to  symbolic  veloc- 
ities of  recession  of  the  order  of  100,000 
km/sec.  It  has  been  further  confirmed 
that  no  clusters  of  clusters  of  galaxies 
exist. 

The  relative  areas  of  the  sky  that  are 
covered  by  open,  medium  compact,  and 
compact  clusters  of  galaxies  have  been 
investigated  by  Zwicky  and  Rudnicki. 
The  largest  clusters  are  all  of  the  same 
linear  size,  independent  of  type  and  dis- 
tance. 

The  so-called  "cluster  cell,"  every  one 
of  which  contains  the  equivalent  of  one 
large  cluster  of  galaxies,  on  the  average 
was  found  to  have  a  diameter  of  45 
million  parsecs,  assuming  a  redshift 
constant  of  100  km/sec  per  million 
parsecs. 

In  the  catalogue  of  galaxies  by  Zwicky, 
Herzog,  and  Wild  the  peripheral  contours 
used  for  the  delineation  of  clusters  of 
galaxies  are  the  isopheths  or  equal- 
population  contours  along  which  the 
numbers  of  galaxies  per  square  degree  are 
equal  to  about  twice  the  number  of 
galaxies  per  square  degree  in  the  adjacent 


MOUNT      WILSON     AND      PALOMAR     OBSERVATORIES 


33 


fields.  The  clusters  thus  drawn,  not  in- 
cluding the  near-by  Virgo  cluster,  how- 
ever, cover  13  per  cent  of  the  3024  square 
degrees  of  the  sky  included  in  volume  I 
of  the  catalogue. 

In  the  course  of  the  work  on  volume  II 
of  the  catalogue  it  was  found  that  the 
field  (of  36  square  degrees)  centered  at 
R.A.  llh17m  and  decl.  -r-35°30'  (1950)  is 
the  richest  field  of  galaxies  and  of  clusters 
of  galaxies  observed  thus  far,  containing 
about  150,000  galaxies  and  113  clusters 
of  galaxies  as  counted  on  a  limiting  103aE 
plate  (+  red  filter)  obtained  with  the  48- 
inch  schmidt  telescope. 

Internal  Motions  of  Clusters  of  Galaxies 

About  60  spectra  of  galaxies  in  the 
cluster  CI  0123-0138  have  now  been 
photographed  by  Zwicky,  and  the  follow- 
ing results  have  been  derived:  (1)  The 
average  symbolic  velocity  of  recession  is 
Vs  =  5321  km/sec.  (2)  The  dispersion 
in  velocities  (radial)  is  AVS  =  406  km/sec. 
(3)  The  dispersion  in  Vs  is  essentially  con- 
stant from  the  center  of  the  cluster  to  the 
periphery,  a  fact  indicating  that  the 
cluster  is  stationary  and  is  neither  ex- 
panding nor  contracting.  (4)  The  disper- 
sion in  Vs  is  greater  for  the  fainter 
galaxies  than  for  the  brighter  ones, 
although  the  average  values  of  Vs  are  the 
same. 

The  distribution  of  the  galaxies  within 
the  cluster  indicates  an  elliptical  shape 
of  the  cluster.  There  is  no  indication  of 
any  rotation,  however,  from  the  analysis 
of  the  radial-velocity  data. 

Additional  spectra  have  been  obtained 
of  member  galaxies  of  clusters  in  Cancer, 
Hydra    I,    and    the    Coma    cluster    for 


determining    velocity     dispersions    and 
mass-luminosity  ratios. 

Redshift-M agnitude  Relations 

In  principle,  there  are  several  observa- 
tional tests  for  distinguishing  between  an 
exploding  universe  and  a  steady-state 
universe.  With  methods  available  today, 
the  best  test  is  the  relation  between  the 
redshifts  and  the  distances  of  large 
clusters  of  galaxies.  More  exactly,  the 
observable  parameters  are  the  redshifts 
and  the  apparent  bolometric  magnitudes 
of  representative  cluster  members. 

Photoelectric  observations  collected  by 
Baum  since  1955  have  been  mentioned 
in  previous  Year  Books.  The  absolute 
amounts  of  energy  received  from  various 
galaxies  are  measured  photoelectrically 
in  eight  colors  ranging  from  ultraviolet 
to  infrared.  Effective  wavelengths  and 
bandwidths  of  the  eight  colors  are: 

Color  Effective  X,  A    Bandwidth,  A 

Ultraviolet  3730  500 

Violet  4335  740 

Blue  5065  430 

Green  5525  470 

Red  6705  850 

Infrared  I  7525  600 

Infrared  J  8520  800 

Infrared  K  9875  1100 

When  the  resulting  spectral-energy  dis- 
tributions of  galaxies  are  compared,  the 
displacements  between  them  yield  both 
their  redshift  and  their  relative  bolo- 
metric magnitudes.  In  this  way,  the  red- 
shift-magnitude  relation  has  been  ex- 
plored to  a  much  greater  distance  than 
before.  Final  photoelectric  values  for  the 
three  observed  clusters  of  largest  red- 
shift  are  as  follows: 


Cluster 
0024  +  1654 
1448+2617 
1410+5224 

Redshift 
AX/X 

0.29 
0.36 
0.44 

Symbolic  Velocity 
cAX/X,  km/sec 

84,000 
108,000 
132,000 

Probable  Error 
in  Redshift,  % 

2.5 
4.0 

7.1 

Cluster  1410  +  5224  is  the  cluster  found 
by  Minkowski  at  the  position  of  Cam- 
bridge radio  source  3C295.  An  optical 
emission  line  at  X5448,  presumed  to  be 


O  II  3727,  provides  a  good  check  on  the 
redshift  above. 

When  these  photoelectric  redshifts  were 
reported  earlier,  the  corresponding  bolo- 


34 


CAKNEGIE     INSTITUTION     OF     WASHINGTON 


metric  magnitudes  could  not  be  specified 
with  the  fullest  attainable  certainty,  and 
the  apparent  shape  of  the  redshift- 
magnitude  relation  had  to  be  taken  as 
tentative.  Observations  by  Baum  during 
the  report  year  have  been  devoted  to 
resolving  this  difficulty.  The  uncertainty 
arose,  not  because  of  limited  precision 
in  the  photometry,  but  because  auxiliary 
data  were  needed  for  intercomparing  the 
magnitudes  in  one  cluster  with  those  in 
another  in  the  best  possible  way.  During 


the  report  year,  photoelectric  and  photo- 
graphic observations  have  been  collected 
for  constructing  the  needed  luminosity 
functions  of  seven  key  clusters  of  galaxies 
distributed  in  redshift  from  AX/X  =  0.02 
to  AX/X  =  0.44.  The  new  photoelectric 
sequences  include  57  additional  galaxies, 
many  of  them  relatively  faint.  As  before, 
most  of  the  observations  were  made  with 
the  pulse-counting  photometer  at  the 
200-inch  prime  focus. 


RADIO  SOURCES 


Schmidt  has  engaged  in  a  spectroscopic 
investigation  of  galaxies  believed  to  be 
connected  with  radio  sources.  This  pro- 
gram is  planned  in  close  cooperation  with 
Dr.  Thomas  Matthews,  who  made  most 
of  the  identifications.  Galaxies  connected 
with  radio  sources  3C33,  88,  98,  171,  198, 
219,  234,  317,  433,  445,  from  the  third 
Cambridge  Catalogue,  and  Coma  A  were 
investigated.  All  spectra  show  emission 
lines,  the  number  ranging  from  1  to  13. 
The  magnitudes  are  in  the  range  15  to  20; 
the  redshifts  (AX/X)  vary  from  0.03  to 
0.24.  The  absolute  magnitudes  of  all 
galaxies  are  close  to  —20.  The  largest 
number  of  emission  lines  is  seen  in  3C234, 
the  spectrum  of  which  resembles  that  of  a 
planetary  nebula  of  high  excitation.  In  a 
case  like  this,  the  spectrum  constitutes  a 
strong  confirmation  of  the  identification. 
For  3C88,  where  the  spectrum  shows 
weak  X3727  emission  only,  the  spectrum 
has  hardly  any  confirmatory  value. 
Spectra  taken  of  some  three  or  four  other 
galaxies  identified  with  radio  sources 
show  no  emission  features.  At  least  one  of 
these  is  now  known  to  be  a  misidentifica- 
tion.  It  may  be  estimated  conservatively 
that  70  per  cent  of  the  galaxies  fainter 
than  15  mag  that  are  identified  with 
bright  radio  sources  show  emission  lines 
in  the  spectrum.  A  spectrum  of  the 
galaxy  identified  with  the  distant  radio 
source  3C295,  observed  and  discussed 
earlier  by   Minkowski,   shows   that   the 


X3727  emission  relative  to  the  continuum 
is  rather  weak  in  comparison  with  that 
observed  in  some  other  radio  sources. 
Spectroscopic  work  on  some  stellar  ob- 
jects believed  to  be  connected  with  radio 
sources  is  continuing. 

Greenstein  obtained  spectra  of  the 
radio  source  3C442,  which  is  a  double 
elliptical,  with  only  X3727  emission.  A 
common  feature  of  the  radio  galaxies  in 
Herculis  A,  3C278,  and  3C442  is  a  low 
gradient  of  surface  brightness.  The  sys- 
tems seem  to  be  ellipticals,  but  their 
brightness  distribution  optically  is  more 
like  that  of  an  Sc  pole-on  system. 

Further  spectra  of  the  radio  star  3C48 
reveal  no  changes,  and  one  in  the  visual 
region  showed  two  more  unidentifiable 
emission  lines.  He  II,  X5411,  was  absent. 

Extending  the  work  reported  last  year, 
Dr.  Thomas  Matthews  and  Sandage 
identified  two  additional  radio  stars 
similar  to  3C48.  This  brings  the  total  of 
such  identifications  to  three.  The  objects 
are  3C48,  3C196,  and  3C286.  The  radio 
positions  of  all  three  sources  were  deter- 
mined before  optical  identification  was 
made,  and  the  only  object  within  the 
error  rectangle  of  the  radio  position  is 
stellarlike  in  each  of  the  three  cases.  The 
agreement  of  the  radio  and  optical 
positions  is  remarkable,  being  within  4 
seconds  of  arc  in  all  objects.  Photoelectric 
photometry  of  the  stars  shows  that  each 
has  very  unusual  colors.  The  photometry 


MOUNT      WILSON     AND      PALOMAR      OBSERVATORIES 


35 


gives  V  =  16»20,  B  -  V  =  0M0, 
U  -  B  =  -0m59  for  3C48;  V  =  17^79, 
B  -  V  =  0m57,  U  -  B  =  -0^43  for 
3C196;  and  V  =  17m25,  B  -  V  =  0m26, 
[/  _  ^  =  -0?91  for  3C286.  Spectra  of 
3C196  and  3C286,  obtained  by  Schmidt, 
also  show  that  the  stellar  objects  are 
peculiar  and  uniquely  new. 

The  optical  flux  of  3C48  was  found  to 
vary  from  V  =  16^02  to  V  =  16m44  over 
the  13-month  observation  period  since  its 
initial  discovery.  The  time  resolution  of 
Sandage 's  observations  is  not  great,  and 
so  nothing  is  known  about  variations  in 
the  order  of  minutes  or  hours,  but  the 
flux  does  vary  from  night  to  night. 
Special  observations  at  radio  frequencies 
by  Matthews  showed  that  the  radio  flux 
is  constant  to  within  the  probable  error 
of  the  determination  even  though  the 
optical  flux  varies. 

The  optical  U,  B,  V  measures  for  the 
three  sources  were  transformed  to  abso- 
lute flux  units  and  compared  with  the 
radio  data.  For  3C48  and  3C196  the 
power  spectrum  computed  from  the 
theory  of  synchrotron  radiation  was 
shown  to  predict  the  U,  B,  and  V  values 
to  within  a  few  hundredths  of  a  magnitude 
when  all  but  one  of  the  adjustable 
parameters  of  the  theory  are  determined 
from  the  radio  data.  The  remaining 
parameter  is  the  critical  frequency  at  the 
high-energy  cutoff,  and  this  could  be 
adjusted  for  the  excellent  fit.  This  fit 
bridges  a  gap  of  more  than  20  octaves  of 
the  power  spectrum.  But  Matthews  and 
Sandage  were  not  convinced  that  this 
agreement  shows  that  the  optical  radi- 
ation is  necessarily  due  to  synchrotron 
emission  alone,  and  the  question  remains 
open  for  the  future.  If  the  optical  flux 
were  due  to  synchrotron  emission,  the 
energy  of  the  relativistic  electrons  respon- 
sible for  the  radiation  would  be  about 


5  bev  in  a  field  of  1  gauss,  or  50  bev  in  a 
field  of  10~2  gauss. 

Additional  data  at  hand  suggest  that 
future  identifications  of  radio  stars  can 
be  expected,  and  many  of  the  questions 
raised  by  these  unique  objects  will 
undoubtedly  be  better  understood  in  the 
near  future. 

One  of  these  sources,  3C48,  studied  by 
Sandage  has  also  been  observed  photo- 
electrically  by  Baum  on  the  eight-color 
system  used  in  connection  with  the 
redshift-magnitude  program.  This  obser- 
vation permits  the  optical  energy  distri- 
bution to  be  investigated  over  a  broader 
range  extending  from  the  ultraviolet  to 
the  infrared.  The  eight  colors  were  found 
to  fit  a  slope  of  —2.25  db  per  1014  cps. 
In  absolute  terms,  the  observed  flux  of 
3C48  at  5490A  (5.46  X  1014  cps)  on  two 
nights  in  December  1961  was  found  to  be 
1.04  X  10"29  watt/m2  per  cps. 

Over  the  past  few  years,  plates  of  the 
Crab  Nebula  have  been  taken  by  G. 
Munch  with  the  object  of  following 
changes  in  the  large-scale  structure  of  the 
amorphous  mass  emitting  synchrotron 
radiation.  Particular  attention  is  being 
given  to  the  moving  ripples,  which  at 
irregular  intervals  appear  near  the  hypo- 
thetical central  star.  On  February  2,  1962, 
a  diffuse  wisp,  about  4  seconds  long,  at  a 
position  angle  45°  and  at  about  2  seconds 
distance  from  the  nuclear  star,  was 
detected  for  the  first  time.  The  other 
moving  wisp  discovered  by  Baade  and 
discussed  by  Oort  appears  somewhat 
farther  out.  Poor  weather  and  lack  of 
observing  time  prevented  following  the 
development  of  this  ripple,  but  its 
observation  at  a  smaller  distance  from 
the  central  star  points  to  the  likelihood 
that  the  central  star  is  still  very  active 
in  injecting  into  the  nebula  large  numbers 
of  relativistic  particles. 


THEORETICAL  STUDIES 

Stellar  Atmospheres  and  Oke  for  a  range  of  effective  tempera- 

Line  profiles  of  H7  have  been  computed     tures  and  surface  gravities  corresponding 
on  the  Kolb-Griem-Shen  theory  by  Searle     to  those  of  F-type  stars  of  different  lumi- 


36 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


nosity  classes,  and  compared  with  ob- 
served profiles  for  (1)  the  cluster-type 
variables  RR  Lyrae  and  SU  Draconis,  (2) 
two  F-type  subdwarfs,  and  (3)  F-type 
stars  of  normal  metal  abundance.  The 
observed  and  computed  profiles  are  in 
excellent  agreement,  and  the  line  profiles 
can  be  used  as  temperature  indicators 
independent  of  reddening.  A  comparison 
of  the  derived  Hy  temperatures  with 
those  determined  by  fitting  observed 
absolute  energy  distributions  to  fluxes 
computed  from  model  atmospheres  shows 
good  agreement,  provided  allowance  is 
made  for  interstellar  reddening. 

For  the  normal  metal  stars  of  spectral 
type  later  than  F5,  the  continuum  at  H7 
is  depressed  by  line  blanketing  to  such  an 
extent  that  a  temperature  determination 
by  H7  fitting  is  not  possible.  For  late  F- 
and  G-type  stars,  however,  Ha  profiles 
remain  a  practicable  temperature  indica- 
tor, and  computations  of  a  grid  of  Ha 
profiles  have  been  completed.  In  the  late 
G-type  stars,  the  Ha  profile  is  dependent 
not  only  on  temperature  but  also  on 
metal-to-hydrogen  ratio. 

Weidemann  has  analyzed  the  hydro- 
gen-line profiles  and  the  colors  of  normal 
white  dwarfs  of  spectral  type  DA.  The 
Kolb-Griem-Shen  theory  was  used,  and 
allowance  was  made  for  the  dependence 
of  pressure  on  depth.  The  profiles  so  com- 
puted were  essentially  independent  of  the 
H/He  ratio  and  covered  a  wide  range  of 
surface  gravity  and  temperature.  The 
maximum  of  the  intensity  of  the  hydrogen 
lines  is  shifted  about  4000°  hotter  than  in 
main-sequence  stars,  and  the  maximum 
in  the  Balmer  discontinuity  about  2500°. 
The  effects  of  the  lines  on  U,  B,  V  colors 
and  of  the  reemitted  blanketed  radiation 
were  estimated.  The  temperature  scale 
derived  is  close  to  an  earlier  estimate  by 
Greenstein.  Weidemann  finds  that  the 
DA  stars  range  from  18,000°K  to  7300°K 
in  effective  temperature,  log  g  from  7.3  to 
8.5,  and  masses  from  0.25  to  0.S5M o.  The 
problem  of  the  DB  stars,  which  show  only 
He  I  lines,  was  considered  briefly;  it  was 
found  that  a  very  high  He/H  ratio  is 


needed  to  alter  the  opacity  source  from 
pure  H.  Therefore,  only  if  He/H  >  103 
will  the  H  lines  disappear  and  be  replaced 
for  hot  white  dwarfs  by  He  I  lines. 

Wright  has  reported  extraordinarily 
low  excitation  temperatures  for  normal 
A0  stars.  Jugaku  explored  possible  theo- 
retical explanations  for  this  phenomenon, 
which  he  found  also  to  be  present  in 
Hunger's  analysis  of  a  Lyrae,  which  gives 
only  5350°  for  Texc.  Jugaku  has  computed 
the  effects  of  stratification  in  depth  of  line 
formation,  which  corresponds  to  only  300° 
temperature  change,  and  the  model-at- 
mosphere effects  which  predict  Texc  = 
8460°K  for  Teii  =  9500°— again  too  small 
a  temperature  change  to  account  for  the 
low  observed  TeKC.  A  possible  explanation 
is  a  large  deviation  from  local  thermody- 
namic equilibrium,  and  another  is  in- 
crease of  turbulent  velocity  for  lines  of 
low  excitation  potential. 

In  metal-poor  stars  of  low  T,  Rayleigh 
scattering  becomes  important,  as  Traving 
pointed  out  for  the  analysis  of  globular- 
cluster  red  giants  and  as  Greenstein  and 
Wallerstein  find  for  the  metal-poor  field 
red  giants  like  HD  122563.  Jugaku  evalu- 
ated the  contribution  of  metallic  absorp- 
tion continua,  finding  them  to  be  small 
compared  with  H~  and  Rayleigh  scatter- 
ing. There  is  no  direct  observational  sup- 
port for  Rayleigh  scattering,  since  it 
would  produce  a  steeply  wavelength- 
dependent  depression  of  the  blue  and 
ultraviolet  (i.e.,  a  U  —  B  deficiency  in 
metal-poor  stars). 

Nishida  has  studied  the  problems  of  the 
structure  and  evolution  of  the  helium 
stars.  A  series  of  models  for  a  helium  star 
of  1  solar  mass,  both  in  the  helium  burn- 
ing phase  and  in  the  carbon  burning 
phase,  have  been  constructed.  Results 
show  that  the  evolutionary  track  lies 
near  the  location  of  nuclei  of  planetary 
nebulae  in  the  H-R  diagram,  but  that  it 
does  not  move  to  the  right  across  the  main 
sequence  of  the  Population  I  stars.  Cal- 
culations of  the  stellar  models  for  the 
following  problems  are  in  progress  using 
the  IBM  7090:  (1)  Evolution  along  the 


MOUNT      WILSON     AND     PALOMAR     OBSERVATORIES  37 

horizontal  branch  of  the  Population  II  to  be  that  deficient.  Thus,  the  birth 
stars.  (2)  Stellar  models  for  stars  with  luminosity  function  in  the  solar  neighbor- 
very  small  masses  (less  than  O.lMo).  (3)  hood  must  have  been  different  at  earlier 
Evolutionary  tracks  in  the  H-R  diagram  times,  specifically  such  that  in  the  early 
for  stars  that  gravitationally  contract  to  stages  of  the  Galaxy  the  rate  of  formation 
become  white  dwarfs.  of  stars  of  10  solar  masses,  relative  to  that 

G.  Munch  and  Dr.  R.  Kippenhahn  of  of  stars  of  1  solar  mass,  was  about  10  or 

the  Max  Planck  Institut  fur  Astrophysik  20  times  larger  than  it  is  at  present, 
in  Munich,  Germany,  are  studying  the 

spectrum  of  late  F-type  supergiants  to  Stellar  Dynamics 
verify  whether  the  Balmer  lines  can  be  Theory  predicts  that  the  only  exactly 
explained  on  the  basis  of  models  with  a  steady  states  of  an  unrelaxed  stellar  sys- 
unique  temperature  as  function  of  depth  tern  which  shows  star  streaming  directed 
in  the  atmosphere.  Suggestions  have  been  to  and  from  the  center  are  those  of  Ed- 
made  in  the  past  to  explain  a  supposed  dington's  type.  These  have  potentials  of 
increase  in  the  strengths  of  the  Balmer  the  form  \p  =  [£(\)  —  ij(/jl)]/(\  —  ju)> 
lines,  as  the  surface  gravity  of  the  stars  where  A  and  n  are  spheroidal  coordinates 
decreases  at  constant  effective  tempera-  and  $*  and  77  are  arbitrary  functions, 
ture,  in  terms  of  the  temperature  fluctu-  Satisfactory  models  of  this  form  have 
ations  produced  by  the  strong  turbulence  been  discussed  by  Kusmin.  An  attempt 
observed  in  the  line  contours  and  curves  was  made  by  Lynden-Bell  to  discover 
of  growth  for  such  stars.  If  the  absolute  whether  the  light  distribution  of  NGC 
magnitude  effect  is  confirmed,  an  attempt  4594  could  arise  from  such  a  system, 
will  be  made  to  construct  models  non-  Plates  taken  by  Munch  were  reduced  to 
homogeneous  in  temperature  and  to  relate  relative  light  intensities  which  were  com- 
the  temperature  fluctuations  to  the  prop-  pared  with  the  predicted  functional  form 
erties  of  the  turbulent  fields.  of  the  projected  mass  density.  Results 

.  showed  that  NGC  4594  cannot  be  of 

Star  Formation  Eddingtonian  form  unless  the  mass-to- 

Theoretical  work  on  star  formation  on  light  ratio  varies  very  considerably  and 

a  phenomenological  basis  was  continued  systematically  as  a  function  of  distance 

by  Schmidt.  The  main  assumption  made  from  the  center. 

in  these  considerations  is  the  absence  of  Two  further  stellar  dynamical  investi- 
systematic  radial  transport  of  stars  or  gas  gations  were  completed  by  Lynden-Bell. 
in  the  Galaxy.  It  appears  that  the  average  The  first  answers  the  question  how  long 
past  rate  of  formation  of  stars  around  1  it  takes  a  nonsteady  stellar  system  to 
solar  mass  is  less  than  three  times  their  approach  an  unrelaxed  steady  state  and 
present  formation  rate.  The  past  forma-  how  it  does  so  in  the  absence  of  dissipa- 
tion history  of  bright,  rapidly  evolving  tion.  The  time  is  about  10  times  the 
stars  can  be  deduced  indirectly  from  the  period  of  a  typical  star  around  the  system, 
distribution  of  ultraviolet  excesses  in  late  and  the  mechanism  is  Landau  damping, 
G-type  dwarfs.  It  can  be  shown  that  a  well  known  in  plasma  physics.  The  second 
time-independent  birth  luminosity  func-  gives  a  method  of  solving  the  self-gravity 
tion  implies  a  certain  distribution  of  equation  for  flattened  steady-state  sys- 
excesses,  independent  of  the  past  forma-  terns,  and  as  a  result  the  first  exact 
tion  rates.  Specifically,  in  this  case  72  per  theoretical  model  is  produced, 
cent  of  the  late  G-type  dwarfs  would  be 

metal-deficient  by  a  factor  exceeding  2  Cosmology 

relative  to  the  Hyades.  The  actual  dis-  Previous  reports  have  mentioned  that 

tribution  of  excesses  shows  between  40  a  decision  between  the  several  proposed 

and  50  per  cent  of  the  late  G-type  dwarfs  models  of  the  expanding  universe,  such 


38  CARNEGIE     INSTITUTION     OF      WASHINGTON 

as  exploding  world  models  compared  with  steady-state  model,  a  galaxy  with  Z  —  0.4 
the  steady-state  model,  can  be  made  if  at  the  present  epoch  will  experience  an 
the  deceleration  of  the  expansion  can  be  acceleration  rather  than  deceleration,  in- 
measured.  This  is  because  the  deceleration  creasing  its  redshift  at  the  instantaneous 
is  caused  by  the  gravitational  attraction  rate  of  +9.2  km /sec  per  million  years, 
of  matter  on  individual  galaxies.  The  With  present  optical  techniques  there  is 
rate  of  deceleration  measures  the  density  no  hope  of  detecting  such  small  changes 
of  matter  in  space,  which  in  turn  deter-  of  redshift  for  time  intervals  of  less  than 
mines  the  spatial  curvature  and  the  10  million  years.  If  radio  techniques  are 
intrinsic  geometry  of  space  via  the  field  used  with  observation  of  the  21-cm  H  I 
equations  of  general  relativity.  Various  line,  the  detection  of  a  frequency  shift  of 
ways  of  finding  the  deceleration  param-  3  X  10~2  cps  per  year  in  a  signal  of  fre- 
eter  q0,  described  in  Year  Book  59,  depend  quency  2000  Mc/sec  is  required,  which 
on  measurement  of  deviations  from  line-  again  appears  to  be  impossible  by  present 
arity  of  the  redshift-magnitude  relation  methods.  To  solve  the  problem  in  this 
or  a  similar  relation  such  as  that  between  way  will  require,  at  our  present  level  of 
redshift  and  apparent  diameter.  The  technology,  a  precision  redshift  catalogue 
principle  of  these  measurements  is  that  to  be  stored  away  in  a  stable  society  for 
one  looks  back  in  time  as  one  looks  out  10  million  years, 
in  space  and  can  therefore  sample  the 

expansion  rate  of  the  universe  in  past  Miscellaneous 
cosmic  times.   This  indirect  method  is 

observationally    very    difficult,    because  The  problem  of  the  transfer  of  the 

galaxies  with  redshifts  of  the  order  of  radiation  in  the  [O  I]  line  3P2  —  ZP\  at 

AX/X0  =  0.5  must  be  observed  for  signifi-  158.13    cm-1,   between   the   two   lowest 

cant  answers.  Furthermore,  as  was  point-  sublevels  of  the  ground  state,  through 

ed  out  last  year,  uncertain  corrections  the  earth's  ionosphere,  has  been  studied 

for  the  evolution  of  the  stellar  content  of  by  G.  Munch  by  removing  a  number  of 

galaxies  must  be  made  to  account  for  the  restrictive    assumptions    introduced    in 

change  in  absolute  luminosity  of  these  previous   attempts   at   a   solution.    The 

distant  galaxies  in  the  light  travel  time,  solution    found    has    been    numerically 

which  is  of  the  order  of  5  X  109  years.  applied  to  a  model  atmosphere  by  com- 

Sandage   looked   into   the   theoretical  puting  the  specific  intensity  and  flux  of 

possibility  of  detecting  the  deceleration  radiation  as  a  function  of  height.  In  a 

directly,  if  a  series  of  redshift  measure-  paper  submitted  for  publication  to  the 

ments  of  a  given  galaxy  were  made  over  Astrophysical  Journal,  the  possibility  of 

a  suitable  time  interval.  Exact  predictions  measuring  the  flux  emergent  at  the  top 

of  the  change  of  redshift  in  a  given  galaxy  of  the  atmosphere,  which  amounts  to  0.1 

with  time  can  be  made  using  the  various  erg/cm2   sec,   has  been   suggested   as  a 

world  models.  It  turns  out  that  this  test  method    for    the    determination    of    the 

of  world  models  is  a  powerful  one  in  kinetic    temperature    and    the    oxygen 

principle  because  the  sign  of  the  effect  is  concentration. 

different  for  exploding  models  and  the  The   eventual   possibility   of   carrying 

steady-state  model.  However,  the  test  is  out  observations  of  galactic  and  extra- 

beyond  our  present  technical  capabilities,  galactic    objects   in    Lyman-a   radiation 

because  the  effect  is  extremely  small.  For  was  investigated  by  Munch.  It  has  been 

a  galaxy  with  a  present  redshift  of  Z  =  found  that  the  decay  of  Lyman-o:  through 

AX/Xo  =  0.4,  the  change  of  redshift  with  2-photon  emission  and  dust  absorption 

time  is  only  —5.9  km/sec  (decelerating)  makes    it    quite   unlikely    that    galactic 

in  a  million  years  for  the  Euclidean  model,  diffuse  Lyman-o;  radiation  exists,  unless 

Similar  numbers  hold  for  the  hyperbolic  the  immediate  neighborhood  of  the  sun 

and  elliptical  exploding  models.  In  the  is  an  H  II  region,  as  is  suggested  by  the 


MOUNT     WILSON     AND     PALOMAR     OBSERVATORIES  39 

emission  nebulosities  excited  by  the  near-  neutral  gas  and  with  low  dust  content 

by  stars  y  Velorum  and  £  Puppis.  Lyman-  (as  the  coronas  of  M  31  and  the  Galaxy) 

a  emission   from   the   H   II   regions  in  may  possibly  reach  the  solar  system  when 

extragalactic  systems  not  surrounded  by  their  redshift  exceeds  1000  km/sec. 

INSTRUMENTATION 

The  10-inch  ruling  engine  has  been  in  addition  to  Mount  Wilson  and  Palomar. 
operation  with  the  new  system  of  inter-  During  the  last  year,  a  number  of 
ferometric  control.  As  was  described  in  minor  modifications  have  been  made  by 
last  year's  report,  this  system  employs  Oke  to  the  photoelectric  coude*  spectrum 
intermittent  spacing,  with  a  fringe  clamp  scanner  on  the  100-inch  telescope.  The 
and  differential  corrector.  Gears  are  now  seeing  compensation  is  now  found  to  be 
on  hand  for  ruling  at  407,  610,  and  915  quite  satisfactory,  provided  that  the  see- 
grooves  per  millimeter.  With  water  cool-  ing  is  average  or  better  and  the  zenith 
ing  of  the  mercury-198  source  for  the  distance  is  not  more  than  approximately 
interferometer,  the  contrast  of  the  fringes  45°.  A  program  has  been  begun  to  meas- 
is  more  than  ample  for  a  path  difference  ure  line  profiles  and  equivalent  widths  in 
of  10  inches.  The  performance  of  the  the  spectra  of  selected  stars.  Accurate 
control  system  has  been  accurate  and  photoelectric  profiles,  such  as  those  of 
reliable,  so  that  the  average  quality  of  H7  in  A  stars,  can  be  used  to  check  the 
the  gratings  has  been  raised  and  the  photographic  calibration  systems  used  at 
productivity  of  the  machine  has  been  various  observatories, 
increased.  Virtually  theoretical  resolving  To  facilitate  the  reduction  of  very 
power  is  obtained  in  the  higher  orders  of  small-aperture  observations  made  with 
the  gratings,  and  scattered  light  is  held  the  solar  magnetograph,  digitizing  equip- 
to  very  low  levels.  ment  has  been  installed  at  the  150-foot 

Eleven  plane  gratings  of  high  quality  tower  telescope  under  the  supervision  of 

were   produced   by   Roberts   under   the  Howard.   A  shaft  encoder  digitizes  the 

direction  of  H.  W.  Babcock  in  sizes  rang-  shaft  position  of  a  strip-chart  recorder, 

ing  from  3  by  4  to  6  by  10  inches;  most  and  this  information  is  punched  on  paper 

were  5  by  8  inches  with  a  spacing  of  610  tape  at  the  telescope.  These  data  will  be 

grooves  per  millimeter.  fed  to  a  digital  computer  at  a  later  time, 

Gratings  have  been  delivered  to  the  and  the  autocorrelation  analysis  can  pro- 

Kodaikanal,  Sacramento  Peak,  Dominion  ceed  with  no  intermediate  steps  in  the 

Astrophysical,   and   David   Dunlap   Ob-  reduction.   Thus  it  will   be  possible  to 

servatories  and  to  the  National  Bureau  accumulate  a  great  number  of  observa- 

of   Standards.    The    records   show   that  tions  with  small  apertures  and  greatly 

gratings  produced  here  are  now  in  use  at  increase    the    accuracy    of    the    existing 

16  observatories  throughout  the  world  in  autocorrelation  functions. 

GUEST   INVESTIGATORS 

The  following  programs  have  been  photographic  photometry,  are  now  avail- 
carried  out  by  guest  investigators  during  able.  Except  for  scale  factors  that  depend 
the  report  year.  upon  the  richnesses  of  the  clusters,  the 

Dr.  George  O.  Abell  of  the  Department  luminosity  functions  are  all  similar.  That 

of  Astronomy,  University  of  California,  of  the  Coma  cluster  is  representative.  The 

Los  Angeles,  continued  his  study  of  rich  number    of    galaxies    brighter    than    m, 

clusters  of  galaxies.  The  luminosity  func-  N(m),  is  given,  approximately,  by 
tions  for   galaxies  in   six  rich   clusters, 

determined  by  a  method  of  extrafocal  log  N(m)  =  constant  +  s  log  m 


40  CARNEGIE     INSTITUTION     OF     WASHINGTON 

where  s  =  0.78  for  m  <  14.7,  and  lies  in  aid  of  photoelectric  observations  of  the 

the  range  0.23  to  0.29  for  14.7  <  m  <  stronger  emission  lines,  an  attempt  was 

18.3,  depending  upon  what  correction  is  made  to  reduce  the  intensities  of  the  lines 

applied  for  the  nonmember  galaxies  in  as  measured  on  the  spectrograms  to  a 

the  cluster  field.  The  discontinuity  in  s  is  true  relative  scale  for  a  study  of  recom- 

due  to  a  maximum  in  the  bright  end  of  bination  rates  and  ionic  abundances  in 

the  luminosity  function.  If  the  luminosity  the    nebula.    Further    spectrograms    of 

functions   of   the   different   clusters   are  shorter  exposure  and  probably  additional 

fitted  together  at  this  discontinuity,  and  photoelectric    measurements    of    weaker 

the  relative  distance  moduli  so  obtained  lines  will  be  necessary  to  obtain  an  ac- 

are  plotted  against  the  known  redshifts  curate    wavelength-dependent    intensity 

of    the    clusters,    the    root-mean-square  calibration. 

velocity  dispersion  about  a  straight  line  Dr.  Stanley  J.  Czyzak  of  the  Aeronau- 

is  less  than  600  km/sec.  On  the  basis  of  tical  Research  Laboratories  at  the  Wright 

the  clusters  investigated  so  far,  therefore,  Patterson  Air  Force  Base  continued  his 

it  appears  that  the  luminosity  functions  calculations  of  accurate  wave  functions 

of  clusters  can  provide  good  estimates  of  of  various  ions  of  P,  S,  CI,  and  A,  all  of 

their  relative  distances.  which  are  of  astrophysical  interest.  The 

In  the  course  of  this  photometry  of  values  for  25  ions  of  the  3p q  configuration 

galaxies  Abell  made  a  detailed  investiga-  were  completed. 

tion  of  some  properties  of  the  U,  B,  V  It  was  now  possible  to  begin  a  detailed 

system.  He  has  numerically  integrated  25  examination  of  the  screening  constants, 

stellar  energy  distributions  published  by  spin-orbit,  spin-spin,  and  spin-other-orbit 

Code  and  Melbourne  against  the  response  calculation  for  the  transition  probabilities 

functions  of  the  U,  B,  V  cell-filter  com-  for  the  Sp9  configurations.  Spectral  data 

binations,   through   various   air  masses,  of  various  gaseous  nebulae  were  examined 

The   computed   variation   of   extinction  for  forbidden  lines  to  determine  which  of 

with  color  (for  a  given  air  mass)  is  present  the  transitions  had  been  observed.  Also  a 

but  small  for  V,  is  in  approximate  agree-  study    was    made    to    determine    other 

ment  with  the  variation  usually  assumed  transitions  which  are  significant.  In  ad- 

for  B  —  V,  and  is  appreciable  and  non-  dition,  preliminary  calculations  of  colli- 

linear  for  U  —  B.  Observed  values  of  the  sion  cross  sections  were  carried  out  by 

extinction  at  Mount  Wilson  are,  on  the  Dr.  Czyzak  on  ions  with  q   =    1,  since 

average,  in  agreement  with  those  com-  these  calculations  would  be  simpler  than 

puted.  In  addition,  the  computed  extinc-  those  for  q  >  1. 

tions  for  a  given  color  of  star  vary  non-  Dr.  H.  Gollnow  of  the  Mount  Stromlo 

linearly  with  air  mass;  as  a  consequence,  Observatory  of  the  Australian  National 

the  usual  procedure  for  reducing  photo-  University  took  spectra  of  about  20  stars 

electric  observations  can  introduce  errors  with  the  coude  spectrographs  of  the  100- 

in  U  —  B  colors  of  as  much  as  0.1  mag.  inch  and  200-inch  telescopes  in  a  search 

Finally,   accurate   color  equations  were  for  magnetic   stars.    Dispersions   of  4.5 

derived  to  transfer  from  instrumental  to  A/mm  and  10  A/mm  and  a  differential 

U,  B,  V  colors,  and  improved  colors  of  analyzer  in  front  of  the  slit  were  used, 

blackbodies  have  been  obtained.  The  stars  were  selected  between  declina- 

Dr.  Lawrence  Aller  of  the  University  tions    +25°   and    —40°,    so    that    their 

of  Michigan  Observatory  obtained  a  series  observations   can   be   continued   at   the 

of  spectrograms  of  the  planetary  nebula  Mount  Stromlo  Observatory.  About  50 

NGC  7009  with  exposures  ranging  from  per  cent  of  the  stars  show  too  large  ro- 

a  few  hours  to  two  nights.  They  revealed  tational  broadening  for  the  measurement 

a  large  number  of  recombination  lines  of  of  Zeeman  displacements.  Of  the  other 

C,  O,  Ne,  and  other  elements.  With  the  stars,   HD  24712  was  studied  in  some 


MOUNT      WILSON     AND     PALOMAR     OBSERVATORIES  41 

detail  and  a  magnetic  field  varying  be-  was  an  extension  of  an  earlier  investiga- 

tween  +574  and  +997  gauss  was  found,  tion   during   the   previous   year   of   the 

The    observation    of    this    star   will    be  rotational  velocities  of  the  B0-B3  Orion 

continued.  stars.   The  B5-B9  stars  were  found  to 

During    the    report    year,    studies    of  rotate    somewhat   more    slowly   on    the 

velocity  fields  in  the  solar  atmosphere  average  than  the  general  field  stars  of  the 

have  been  continued  by  Dr.  R.  B.  Leigh-  same    type.    The    maximum    rotational 

ton  of  the  California  Institute  of  Tech-  velocities   were   found   to   occur   in   the 

nology  with  the  assistance  of  Robert  W.  B5-B7  spectral  types.  The  observations 

Noyes  and   George  W.   Simon.   Special  also  indicate  that  there  is  a  smaller  per- 

emphasis  was  given  to  oscillatory  motions  centage  of  slow  rotators  among  the  B5-B9 

and  large-scale  currents  discovered  with  group  than  among  the  B0-B3  group, 

the  Mount  Wilson  instruments  in  1960.  Narrow-band  photometric  observations 

The  main  results  may  be  summarized  of  stars  in  the  star  clusters  h  and  %  Persei 

as  follows:  (1)  The  small-scale  velocity  and   M36  have   been   obtained   by   Dr. 

field  (1000-5000  km  linear  dimension)  in  McNamara  with  the  60-inch  telescope, 

the  upper  photosphere  exhibits  a  strong  A    narrow-band    photometric    study    of 

tendency  to  repeat  itself  in  time  with  a  eclipsing  variables  has  also  been  initiated 

5-minute  period.  That  is,  each  local  region  with   the  20-inch  telescope  at  Palomar 

undergoes     a     quasi-sinusoidal     motion  Observatory.  No  results  are  yet  available 

which  may  persist  for  several  cycles.  (2)  on  these  photometric  programs. 

The  period  of  the  above  oscillation  does  Dr.   Walter   E.   Mitchell,   Jr.,   of  the 

not  seem  to  be  strictly  constant  with  Perkins  Observatory  continued  the  solar 

altitude  but  tends  to  decrease  by  10  to  20  observations   with    the    Snow    telescope 

per  cent  as  one  proceeds  from  the  middle  during   the   summer   of    1961   with   the 

photosphere  into  the  lower  chromosphere,  assistance  of  Mr.  John  C.  Muster.  Nu- 

(3)  The  intensity  variations  in  the  lower  merous  improvements  were  made  to  the 
chromosphere,  as  seen  at  the  cores  of  such  telescope  and  spectrograph.  To  reduce  the 
strong  lines  as  Na  X5896,  Mg  X5173,  or  scattered  light  due  to  Rowland  ghosts,  an 
similar  lines,  are  also  observed  to  fluctu-  arrangement  of  mirrors  and  intermediate 
ate  in  time  with  the  period  of  the  velocity  slit  was  designed  to  deliver  the  beam  twice 
oscillations.  This  suggests  that  the  oscilla-  to  the  grating,  i.e.,  to  have  the  spectro- 
tory  motions  are  connected  with  waves  graph  act  as  its  own  monochromator.  A 
which  transport  energy  into  the  chromo-  beam  splitter  consisting  of  a  plane  parallel 
sphere  and  liberate  it  as  heat  or  radiation,  plate  of  fused  quartz  was  mounted  just 

(4)  A  network  of  horizontal  currents,  inside  the  entrance  slit  of  the  spectro- 
grouped  into  a  system  of  large-scale  graph.  The  fraction  of  the  beam  returned 
(5000-30,000  km)  "convective  cells,"  is  by  this  plate  was  used  as  a  monitoring 
observed.  These  cells  appear  to  be  dis-  signal  for  ratio  recording. 

tributed  rather  uniformly  over  the  solar  The  double-pass  system  was  employed 
surface,  and  their  correlation  properties  to  make  photoelectric  tracings  of  the 
suggest  a  strong  tendency  toward  an  following  regions:  Na  Di  and  D2,  Mg  'b\ 
ordered  array  over  distances  up  to  50,000  Ca  II  H  and  K,  H/3,  H7,  H5,  Ca  I  4226, 
or  100,000  km.  and  X3570.  Throughout,  there  is  a  notice- 
Dr.  D.  H.  McNamara  of  North  Ameri-  able  lowering  of  central  intensities  (by 
can  Aviation  investigated  the  rotational  amounts  up  to  10  per  cent  of  the  con- 
velocities  of  B5-B9  stars  in  the  Orion  tinuum)  both  as  compared  with  the 
association.  The  spectra  from  which  the  Utrecht  Photometric  Atlas  of  the  Solar 
rotational  velocities  were  determined  were  Spectrum  and  with  the  Snow  single-pass 
obtained  with  the  16-inch  camera  of  the  observations  with  the  same  grating.  The 
100-inch  coude  spectrograph.  This  study  region  6700-3900  A  was  recorded  in  first 


42 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


order  with  full  resolution  and  with  band 
passes  of  4,  8,  and  16  A. 

Infrared  stellar  photometric  observa- 
tions were  obtained  by  Dr.  Mitchell  and 
Mr.  Philip  E.  Barnhart  with  the  assist- 
ance of  Messrs.  John  C.  Muster,  Ronald 
E.  Roll,  and  John  H.  Hill.  Instrumental 
assistance  was  also  provided  by  Messrs. 
Charles  E.  Gramm,  Anthony  J.  Prasil, 
and  Dr.  William  H.  Haynie  of  the  East- 
man Kodak  Company.  Preliminary  in- 
frared magnitudes  were  measured  for  31 
G,  K,  and  M  giants,  6  red  supergiants, 
and  e  Aurigae  using  an  improved  East- 
man Kodak-Ohio  State  University  in- 
frared stellar  photometer  on  the  60-inch 
and  100-inch  telescopes.  The  photometric 
system  has  the  following  characteristics: 

Magnitude      Effective  Band 

Designation    Wavelength,  n  Pass,  ju 

X  2.2  0.24 

Y  3.7  0.43 

When  the  observed  visual-infrared 
color  indices  of  the  measured  stars  are 
compared  with  the  indices  deduced  the- 
oretically for  the  stars  considered  as 
blackbody  radiators,  the  following  con- 
clusions may  be  drawn:  (1)  Nonvariable 
giants  and  supergiants  lie,  in  general, 
close  to  the  theoretical  relation;  i.e.,  to  a 
first  approximation  these  stars  behave  as 
blackbody  radiators.  (2)  A  few  nonvari- 
able giants  and  supergiants  fall  unexpect- 
edly far  above  or  below  the  theoretical 
relation,  suggesting  large  blanketing  ef- 
fects on  visual  magnitude  or  errors  in 
temperature  assignments.  (3)  When  ob- 
served at  visual  magnitudes  well  below 
maximum,  long-period  variable  stars, 
whose  temperatures  are  derived  from 
their  spectrum  characteristics  at  mean 
maximum  light,  show  an  excessive  red- 
dening compared  with  the  theoretical 
relationship;  that  is,  the  variability 
occurs  almost  entirely  at  wavelengths 
shorter  than  2  p.  (4)  Epsilon  Aurigae 
shows  an  infrared  excess  of  approximately 
1.2  mag,  thus  supporting  the  hypothesis 
that  it  has  a  large  infrared  component. 

Attempts  were  made  by  Dr.  Mitchell 


and  Mr.  Barnhart  to  operate  a  helium- 
cooled  Ge :  Cd  detector  for  the  8-  to  13-/* 
region,  but  no  stellar  signal  was  distin- 
guishable from  photometer  and  telescope 
signals. 

Dr.  Bruce  C.  Murray  of  the  Lunar 
Research  Laboratory  at  the  California 
Institute  has  continued  the  program  of 
photoelectric  colorimetry  of  the  moon 
using  the  spectrum  scanner  at  the  Casse- 
grain  focus  of  the  60-inch.  The  scanning 
technique  initiated  during  1960-1961  has 
been  perfected,  including  the  successful 
implementation  of  "lunar  rate"  for  the 
60-inch,  to  a  point  where  an  accuracy  of 
0.01  to  0.02  mag  has  been  achieved  for  the 
eleven  independent  color  values  obtained 
from  each  object  examined.  Approximate- 
ly fifteen  lunar  areas  of  15  by  15  km  size 
have  been  observed  as  well  as  various 
planetary  and  stellar  objects  for  com- 
parison. The  data  are  in  the  final  phase 
of  reduction  preparatory  to  being  sub- 
mitted for  publication. 

The  testing  and  development  of  a  long- 
wavelength  infrared  photometer  have 
continued  during  the  year,  two  nights 
during  1961  having  been  devoted  to  this 
project  at  the  60-inch.  Recently,  however, 
a  special  20-inch  infrared  telescope  has 
been  designed  and  built.  This  telescope 
with  a  novel  optical  and  photometer 
system  has  been  given  preliminary  trials 
at  Mount  Wilson  but  will  later  be  placed 
in  operation  at  a  13,000-foot  site  on  White 
Mountain. 

Dr.  Robert  L.  Wildey  and  Mr.  Howard 
A.  Pohn  of  the  Lunar  Laboratory  have 
initiated  a  U,  B,  V  photometric  program 
to  investigate  an  apparent  asymmetrical 
phase  lag  in  the  brightness  versus  phase 
curves  of  different  localities  on  the  moon. 
This  phenomenon  is  apparent  in  the  older 
photographic  photometry  of  the  moon;  if 
confirmed  photoelectrically,  it  represents 
a  most  surprising  natural  phenomenon  of 
the  moon. 

Observations  were  continued  by  Dr. 
Daniel  M.  Popper  of  the  University  of 
California  at  Los  Angeles  on  the  program 
of   establishing   absolute   dimensions   of 


MOUNT      WILSON     AND      PALOMAR      OBSERVATORIES  43 

stars  of  various  kinds  from  the  analysis  of  already  taken  by  Dr.  G.  Waller  stein  for  a 
eclipsing  binary  systems.  Relatively  few  joint  investigation  of  the  velocity  curve 
spectrograms  were  obtained  during  the  of  the  M  component  of  the  system. 
year.  Reanalysis  has  been  completed  for  Dr.  H.  Spinrad  of  the  Jet  Propulsion 
three  solar-type  eclipsing  binaries :  VZ  Laboratory  of  the  California  Institute  has 
Hydrae,  WZ  Ophiuchi,  and  UV  Leonis.  analyzed  infrared  spectrograms  of  Venus 
The  new  spectrographic  observations  with  in  the  plate  files.  Rotational  temperatures 
higher  dispersion  lead  to  masses  about  30  have  been  derived  from  the  intensity 
per  cent  smaller  than  those  obtained  pre-  distributions  of  CO  2  rotational  lines  in 
viously  for  two  of  the  systems.  The  the  X7820  band.  The  rotational  tempera- 
revised  values  are  more  in  accord  with  tures  vary  from  214°K  to  445°K.  Total 
the  values  from  visual  binaries  of  the  pressures  have  been  obtained  from  meas- 
same  spectral  types.  The  photometric  urements  of  the  corrected  widths  of  the 
observations  used  in  the  analysis  of  WZ  C02  rotational  lines;  these  pressures 
Oph  are  also  new,  having  been  obtained  correlate  quite  well  with  the  rotational 
with  the  20-inch  at  Palomar.  A  modern  temperatures  in  the  sense  that  the  high 
light  curve  is  badly  needed  for  VZ  Hyd.  pressures  correspond  to  observations  of 

The  following  new  results  are  based  on  high  rotational  temperatures, 

incomplete  observations.  (1)  RR  Arietis  Dr.  Spinrad  has  also  found  that  the 

is  a  sixth-magnitude  K  star  found  to  be  ammonia  and  methane  rotational  lines  in 

eclipsing  by  Archer.  The  velocity  vari-  the  yellow-red  region  of  the  spectrum  of 

ation  appears  to  be  less  than  5  km /sec.  Jupiter  do  not  have  the  expected  incli- 

(2)  Revised  values  of  the  masses  of  the  nation  on  coude  spectra  in  which  the 

K-type  giants  of  RZ  Cancri  are  3.2  and  spectrograph  slit  was  placed  along  the 

0.5.  (Dr.  Popper's  earlier  published  values  planet's  equator.  This  result  is  interpreted 

were  0.4  and  2.6.)  (3)  The  D  lines  of  the  to  mean  that  these  gases  are  probably 

fainter  components,  not  previously  an-  not  rotating  with  the  same  velocity  as  the 

nounced,    have    been    observed    in    the  Jovian  cloud  layer.  Examination  of  100- 

f  olio  wing  eclipsing  systems :  TW  Draconis  inch  and  200-inch  coude"  spectra  indicated 

(difficult),    WW    Draconis,    RR    Lyncis  marked  variations  in  the  relative  inten- 

(metallic-line    star;    observations    com-  si  ties  of  the  Jovian  NH3  lines  near  X6460. 

plete),  XY  Puppis   (difficult),   and  TX  Dr.  Uli  Steinlin  continued  his  observa- 

Ursae  Majoris  (difficult).  tions  with  the  48-inch  schmidt  camera  to 

Dr.  Jorge  Sahade  of  the  La  Plata  obtain  material  for  the  program  on  three- 
Observatory,  Argentina,  continued  his  color  photometry  of  the  Observatory  in 
spectroscopic  observations,  obtaining  Basel,  Switzerland.  Dr.  W.  Becker  from 
plates  of  the  following  objects:  (1)  The  Basel  participated  in  the  observations 
eclipsing  star  V453  Scorpii  to  supplement  from  February  until  April.  The  observing 
material  previously  obtained  at  Bosque  program  was  completed  in  April  with  572 
Alegre  with  lower  dispersion;  (2)  HD  plates  taken  (416  of  them  after  July  1961) 
188439,  an  early-type  object  which  Lynds  in  the  following  fields:  eight  Milky  Way 
had  announced  as  showing  a  photometric  fields:  NGC  1807/17,  M37,  Great  Sagit- 
period  of  about  9  hours;  (3)  HD  207739  tarius  cloud,  Small  Sagittarius  cloud, 
and  AG  Pegasi  to  detect  spectral  changes,  Scutum,  Aquila,  Lacerta,  Cassiopeia;  ten 
if  any,  relative  to  observations  of  previous  fields  in  higher  galactic  latitudes :  Selected 
years;  (4)  the  eclipsing  system  V367  Areas  51,  54,  57,  82,  94,  107,  133,  141, 
Cygni  to  compare  the  spectral  features  158,  and  Hyades. 

with    those    of    other    systems    already  Plates  have  mostly  been  taken  in  R, 

investigated;  (5)  HD  192281  to  supple-  G,  and  U  for  three-color  photometry  of 

ment  observations  made  in   1960;   and  some  clusters  and,  above  all,  of  field  stars 

(6)  17  Leporis  to  be  used  with  material  in  the  Milky  Way  as  well  as  in  higher 


44  CARNEGIE     INSTITUTION     OF      WASHINGTON 

galactic  latitudes.  In  some  fields,  plates  present  both  spectra  are  being  analyzed 
have  also  been  taken  in  B  and  V  to  permit  in  Kiel.  For  comparison  with  HD  161817, 
three-color  photometry  in  the  U,  B,  V  the  more  or  less  normal  stars  5  Delphini 
system  as  well,  and  to  make  possible  a  A7V,  a  Ophiuchi  A5III,  and  111  Herculis 
comparison  of  the  effectiveness  of  the  two  A3V  had  been  selected.  Delta  Del  has 
systems.  The  limiting  magnitude  lies  in  quite  sharp  lines  and  has  since  been 
general  between  18m  and  19m.  The  three-  measured  for  wavelengths  and  identifi- 
color  photometry  should  provide:  (1)  cations.  The  other  two  stars  show  strong 
density  function  and  luminosity  function  rotational  broadening.  Although  a  Oph  is 
in  different  directions  from  the  sun;  (2)  an  MK-type  star  for  A5III,  the  three- 
color-magnitude  diagrams  of  clusters  and  dimensional  Paris  classification  would 
of  clouds  of  stars  within  the  Milky  Way;  place  it  under  A5V.  This  is  probably  due 
(3)  possibly  a  separation  of  disk  and  halo  to  rotational  broadening  of  the  high 
populations  in  higher  galactic  latitudes,  members  of  the  Balmer  series  simulating 

About  1500  stars  in  each  of  the  follow-  the  Stark  broadening  in  main-sequence 

ing  fields  have  already  been  measured:  stars.    This    problem    is    being    further 

Selected  Area  54,  57,  82,  and  107.  The  analyzed,    with    J.    Kaler    (Michigan), 

reduction    of    these    measurements    and  working  at  present  in  Kiel, 
work  in  other  fields  is  under  way  at  the         Spectra  for  investigating  possible  differ- 

Basel  Observatory.  ences    in    composition    connected    with 

Photoelectric  U,  B,  V  standards  for  the  evolution  were  taken  by  Dr.  Unsold  with 

Basel   Observatory  program  have  been  the  32-inch  camera  in  the  photographic 

obtained  with  the  100-inch  by  Dr.  A.  Th.  and    visual    regions.    Five    stars    with 

Purgathofer   of   the   Vienna   University  spectral   types   F5   to   G2V  of   Eggen's 

Observatory.  Observations  of  stars  in  the  y   Leonis   group,   including   5/3   Virginis 

magnitude  range  from  V  =   16m  to  18m  (F8),  supposed  to  be  metal-superrich,  and 

were  obtained  for  most  of  the  Selected  two  later-type  (dK5)  stars  taken  from 

Area  fields.  O.  C.  Wilson's  "red"  and  "violet"  groups 

Dr.  A.  Unsold  of  the  University  of  Kiel,  of  the  main  sequence,  HD  156026  and 

Germany,    in    14    nights    of    observing,  HD  192310,  were  observed, 
obtained  high-dispersion  spectra  of  vari-         Some  visual  test  plates  of  a  Cygni  A2 

ous    groups    of    stars   which    might    be  la  showed  that  the  structure  of  its  Ha 

suitable  for  studying  the  relations  be-  emission   component   has  changed   con- 

tween  chemical  composition  and  evolu-  siderably  since  the  last  visual  plates  were 

tion.  Most  of  the  plates  were  taken  with  taken  in  1957.  The  photometric  analysis 

the  32-inch  camera  of  the  coude  spectro-  is  being  carried  through  by  Dr.  Comper 

graph  of  the  100-inch  telescope,  and  they  in  Kiel. 

cover  the  photographic  and  the  visual         A    considerable    number   of    100-inch 

regions.  coude  plates  of  y  Serpentis  F6IV-V  taken 

HD  161817,  usually  classified  as  sdA2,  in  1957  by  Unsold,  in  1959  by  Traving, 

is  most  probably  a  horizontal  branch  star,  and    in    1960    by    Bonsack    have    been 

Its    huge    space    motion,    according    to  analyzed  in  detail  by  W.  Kegel  in  Kiel. 

Eggen,   is  shared  by  Wilson   10367    =  The  variation  of  turbulence  with  depth 

LPM661,  an  11-mag  F8  main-sequence  turned  out  to  be  an  essential  feature.  The 

subdwarf ,  of  which  at  least  the  photo-  relative  abundances  of  the  metals  are  the 

graphic  region  could  be  obtained  with  the  same   as   in   the   sun,    but,   relative   to 

16-inch  camera.  Both  stars  are  obviously  hydrogen,  all  the  heavier  elements  are 

metal-poor.  Quantitative  comparison  of  reduced  by  a  factor  of  about  1.7.  That,  as 

their  chemical  composition  should  give  well  as  the  space  velocity  and  the  weak 

most  interesting  indications  about  evolu-  ultraviolet  radiation,  indicates  that  y  Ser 

tionary  events  in  the  red  giant  or  super-  is  a  member  of  the  intermediate  Popu- 

giant  region  of  the   H-R  diagram.   At  lation  II. 


MOUNT      WILSON     AND      PALOMAR     OBSERVATORIES  45 

Dr.  George  Wallerstein  of  the  Astro-  Dr.  R.  v.  d.  R.  Woolley  and  Mr.  C.  A. 

nomical  Department  of  the  University  of  Murray  of  the  Royal  Greenwich  Observa- 

California  at  Berkeley  has  been  observing  tory  carried  out  two  programs  with  the 

K  giants  in  order  to  obtain  abundances  of  coude  spectrograph  attached  to  the  100- 

the  elements.  The  observations  of  G8-K2  inch  reflector.  They  exposed  a  number  of 

stars  in  the  general  field  at  6  A/mm  in  the  plates  with  the  32-inch  camera  and  with 

yellow  region  are  now  complete.   Stars  the  72-inch  camera.  Some  of  these  were 

with  ultraviolet  excesses  from  0.20  mag  exposed  to  an  intensity  suitable  for  the 

to  deficiencies  of  0.10  mag  will  be  com-  measurement  of  radial  velocity;  others 

pared  with  the  K0  giants  in  the  Hyades.  were  more  lightly  exposed  so  that  they 

Many  of  the   stars  included   are  high-  would  be  suitable  for  spectrophotometry, 

velocity  stars.  Some  of  the  other  more  The    radial-velocity    plates    have    been 

interesting  stars  on  the  list  are  a  few  that  measured,  and  the  results  will  be  pub- 

Gyldenkern    suspects    to    be    metal-rich  lished;  the  remaining  plates  have  been 

from  his  narrow-band  photometry,  and  examined  with  a  spectrophotometer.  The 

some  "4150  stars"  that  show  strong  CN  spectra    of    r    Ceti,    107    Piscium,    and 

in  the  blue  region.  In  addition,  two  K0  o2  Eridani  obtained  with  these  plates  have 

giants  in  Praesepe  have  been  observed  at  been  investigated  by  Dr.  Pagel  at  the 

15  A/mm  in  the  yellow.  This  work  is  in  Royal  Greenwich  Observatory,  and  the 

cooperation    with    Dr.    Heifer    of    the  results    have    been    worked    up    for    a 

University  of  Rochester.  determination  of  the  abundances  of  ele- 

The  high-velocity  A  star  HD  109995  ments    in    these    stars    by    differential 

has  been  observed  in  order  to  compare  it  curve-of-growth  analysis. 

with  Sirius  and  another  high-velocity  A  Dr.    Woolley   and    Mr.    Murray   also 

star,  7  Sextantis.  A  cursory  examination  carried  out  a  program  of  direct  photog- 

of  one  4  A/mm  and  two  10  A/mm  plates  raphy   at   the   Cassegrain   focus   of   the 

shows  that  the  lines  in  HD  109995  are  60-inch  telescope.  They  took  repeat  plates 

very  much  weaker  than  in  either  of  the  of  a  number  of  cluster  fields  that  had  been 

other  two  stars.  observed  by  van  Maanen,  including  the 

Dr.  Wallerstein  obtained  several  plates  cluster    M    67.    In    all,    15    fields   were 

of  31  Cygni  that  showed  chromospheric  photographed;  the  plates,  together  with 

Ti  II  lines  as  well  as  the  K  line.  These  a  selection  of  van  Maanen's  first  epoch 

plates  will  be  reduced  in  cooperation  with  plates,    are    at    the    Royal    Greenwich 

Dr.    Wright    of    the    Dominion    Astro-  Observatory  awaiting  measurement  for 

physical  Observatory.  proper  motion. 

STAFF    AND    ORGANIZATION 

The   Observatories   suffered   a   severe  the  200-inch  mirror  after  it  had   been 

loss  in  the  sudden  death  on  December  26,  moved  to  Palomar.  While  on  leave  from 

1961,  of  Don  O.  Hendrix,  Superintendent  the  Observatories  he  ground  and  figured 

of  the  Optical  Shop.  Mr.  Hendrix  joined  the  120-inch  mirror  of  the  Lick  Observa- 

the   staff   in    1931    and   became   Super-  tory.  The  high  efficiency  of  the  present 

intendent  of  the  Optical  Shop  in  1947.  He  optical  equipment  of  the  Observatories  is 

developed  extraordinary  skill  in  the  hand  to  a  large  extent  due  to  Hendrix'  skill  and 

figuring   of   large   nonspherical    surfaces  ingenuity. 

required  in  many  modern  optical  designs.  Drs.  Robert  Howard  and  Olin  Eggen 

Among  the  projects  he  carried  out  were  joined  the  staff  of  the  Observatories  in 

the  optics  for  the  48-inch  schmidt  tele-  September  1961.   Dr.   Howard  plans  to 

scope  at  Palomar  and  the  corrector  plates  investigate  solar  magnetic  fields,  and  Dr. 

for    the    15    schmidt    cameras    on    the  Eggen  has  undertaken  an  extensive  photo- 

spectrographs   at   Palomar   and   Mount  metric  program.  Dr.  Otto  Struve  became 

Wilson.  He  also  did  the  final  figuring  of  a  member  of  the  staff  in  March  1962. 


46 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Research  Division 

Staff  Members 
Halton  C.  Arp 

Horace  W.  Babcock,  Assistant  Director 
William  A.  Baum 
Ira  S.  Bowen,  Director 
Armin  J.  Deutsch 
Olin  J.  Eggen 
Jesse  L.  Greenstein 
Robert  F.  Howard 
Robert  P.  Kraft 
Guido  Munch 
J.  Beverley  Oke 
Allan  R.  Sandage 
Maarten  Schmidt 
Otto  Struve 
Olin  C.  Wilson 
Fritz  Zwicky 

Research  Associates 
Jan  H.  Oort 
Kenneth  0.  Wright 

Staff  Members  Engaged  in  Post-Retirement 
Studies 
Harold  D.  Babcock 
Milton  L.  Humason 
Alfred  H.  Joy 
Seth  B.  Nicholson 

Senior  Research  Fellows 

Rudolph  Kippenhahn 
Satoshi  Matsushima 
Minoru  Nishida 
Evry  Schatzman 
Leonard  T.  Searle 
Volker  Weidemann 

Carnegie  Research  Fellows 
Leo  Houziaux 
Charles  R.  O'Dell 

National  Science  Foundation  Fellows 

John  C.  Brandt 
Paul  W.  Hodge 

Research  Fellows 

Jacques  Berger 
John  Hazlehurst 
D.  H.  P.  Jones 
Jun  Jugaku 
Donald  Lynden-Bell 
Luis  Munch 
Konrad  Rudnicki 
Wallace  L.  W.  Sargent 
Henrietta  Swope 


Research  Assistants 
Christine  Arpigny 
Jeanne  Berger 
Frank  J.  Brueckel 
Sylvia  Burd 
Subhash  Chandra 
Jai  H.  Choy 

Mary  F.  Coffeen,  Librarian 
Thomas  A.  Cragg 
Donald  S.  Hayes1 
Emil  Herzog 
Joseph  0.  Hickox2 
Basil  N.  Katem 
Charles  T.  Kowal 
A.  Louise  Lowen 
Joyce  E.  Sheeley 
Merwyn  G.  Utter 

Student  Observers 

James  E.  Gunn 
Manuel  E.  Mendez 
Dimitri  M.  Mihalas 
Robert  H.  Norton 
Robert  A.  R.  Parker 
Lewis  L.  Smith 
Robert  L.  Wildey 

Photographer 

William  C.  Miller 

Instrument  Design  and  Construction 

Lawrence  E.  Blakee,  Senior  Electronic 

Technician 
Eileen  I.  Challacombe,  Draftsman 
Floyd  E.  Day,  Optician 
Kenneth  E.  DeHufT,  Machinist 
Robert  D.  Georgen,  Machinist 
Don  0.  Hendrix,  Superintendent  of  Optical 

Shop3 
Melvin  W.  Johnson,  Optician 
Stuart  L.  Roberts,  Instrument  Maker 
Bruce  Rule,  Project  Engineer 
Marlin  N.  Schuetz,  Electronic  Technician 
Russell  R.  Van  Devender,  Jr.,  Designer  and 

Superintendent  of  Instrument  Shop 
James  S.  White,  Electronic  Technician4 

Maintenance  and  Operation 

Mount  Wilson  Observatory  and  Offices 
Paul  F.  Barnhart,  Truck  Driver 

1  Resigned  March  23,  1962. 

2  Retired  September  30,  1961. 

3  Died  December  26,  1961. 

4  Resigned  December  15,  1961. 


MOUNT      WILSON     AND      PALOMAR      OBSERVATORIES 


47 


Wilma  J.  Berkebile,  Secretary 

Herbert  A.  Cole,  Laborer5 

Hugh  T.  Couch,  Carpenter 

Helen  S.  Czaplicki,  Editorial  Typist 

Stewart  F.  Frederick,  Janitor6 

Eugene  L.  Hancock,  Night  Assistant 

Mark  D.  Henderson,  Gardener 

Margaret  Higgins,  Stewardess 

Anne  McConnell,  Administrative  Assistant 

Leah  M.  Mutschler,  Stenographer  and 

Telephone  Operator 
Bula  H.  Nation,  Stewardess 
Alfred  H.  Olmstead,  Night  Assistant 
Arnold  T.  Ratzlaff,  Night  Assistant 
Glen  Sanger,  Janitor 
John  E.  Shirey,  Laborer 
William  D.  St.  John,  Janitor  and  Relief 

Engineer 
Wilma  G.  Totten,  Stewardess7 
Benjamin  B.  Traxler,  Superintendent 

6  Resigned  March  9,  1962. 

6  Resigned  October  31,  1961. 

7  Resigned  November  15,  1961. 


Palomar  Observatory  and  Robinson  Laboratory 

Audrey  A.  Acrea,  Stewardess 
Fred  Anderson,  Machinist 
Jan  A.  Bruinsma,  Custodian 
Maria  J.  Bruinsma,  Stewardess 
Eleanor  G.  Ellison,  Secretary  and 

Librarian 
Arlis  R.  Grant,  Stewardess8 
Leslie  S.  Grant,  Relief  Night  Assistant  and 

Mechanic 
Byron  S.  Hill,  Superintendent 
Helen  D.  Hollo  way,  Secretary 
Charles  E.  Kearns,  Night  Assistant 
J.  Luz  Lara,  Laborer 
Harley  C.  Marshall,  Office  Manager 
Dwight  M.  Miller,  Mechanic 
George  W.  Pettit,  Janitor9 
Robert  E.  Sears,  Night  Assistant 
William  C.  Van  Hook,  Electrician  and 

Assistant  Superintendent 
Gus  Weber,  Assistant  Mechanic 

8  Resigned  October  13,  1961. 

9  Retired  September  3,  1961. 


BIBLIOGRAPHY 


Abt,  Helmut  A.,  Hamilton  M.  Jeffers,  James 
Gibson,  and  Allan  R.  Sandage,  The  visual 
multiple  system  containing  Beta  Lyrae,  Astro- 
phys.  J.,  185,  429-438,  1962. 

Arp,  Halton  C,  The  globular  cluster  M5,  As- 
trophys. J.,  135,  311-332,  1962. 

Arp,  Halton  C,  The  effect  of  reddening  on  the 
derived  ages  of  globular  clusters  and  the  abso- 
lute magnitudes  of  RR  Lyrae  cepheids,  Astro- 
phys.  J.,  135,  971-975,  1962. 

Arp,  Halton  C,  Intrinsic  variables  and  stellar 
evolution,  Proc.  Symposium  on  Stellar  Evolu- 
tion, La  Plata,  Argentina,  Nov.  7-11,  1960, 
edited  by  J.  Sahade,  Observatorio  Astro- 
nomico,  Secci6n  Publicaciones,  La  Plata, 
Argentina,  1962,  pp.  87-117. 

Arp,    Halton    C,    Stellar   content   of   galaxies, 

Science,  134,  810-819,  1961. 
Arpigny,  Claude,  see  Greenstein,  Jesse  L. 

Baade,  Walter,  and  Henrietta  H.  Swope,  The 
Draco  system,  a  dwarf  galaxy  (abstract), 
Astron.  J.,  66,  278,  1961;  Astron.  J.,  66,  300- 
347,  1961. 

Babcock,  Horace  W.,  The  sun's  magnetic  field 
(abstract),  Science,  134,  1425,  1961. 

Banner,  K.,  W.  A.  Hiltner,  and  Robert  P.  Kraft, 
Colors  and  magnitudes  for  45  cepheids  of  the 
Northern  Milky  Way  (abstract),  Astrophys. 
J.,  134,  1026-1027,  1961;  Astrophys.  J.,  suppl. 
59,  319-355,  1962. 


Baum,  William  A.,  Image  converters — A  new 
instrumental  technique  in  astronomy,  I.C.S.U. 
Rev.,  8,  199-201,  1961. 

Baum,  William  A.,  The  kind  of  universe  implied 
by  recent  photoelectric  redshift  observations 
at  Palomar,  Observatory,  81,  114-115,  1961. 

Baum,  William  A.,  Photoelectric  test  of  world 

models  (abstract),  Science,  184,  1426,  1961. 
Baum,    William   A.,    9a.    Sous-commission    des 

convertisseurs  d'images,  Trans.  Intern.  Astron. 

Union,  10,  143-154,  edited  by  D.  H.  Sadler, 

University  Press,  Cambridge,  1960. 
Baum,    William   A.,    9a.    Sous-commission   des 

convertisseurs  d'images,  Trans.  Intern.  Astron. 

Union,  11  A,  34-47,  edited  by  D.  H.  Sadler, 

Academic  Press,  London,  1962. 

Bowen,  Ira  S.,  John  August  Anderson,  Bio- 
graphical Memoirs,  36,  1-18,  published  for 
National  Academy  of  Sciences  of  the  United 
States  by  Columbia  University  Press,  New 
York,  1962. 

Brandt,  John  C,  Helium  resonance  radiation  in 
the  night  and  day  sky  (abstract),  Astron.  J., 
66,  279,  1961. 

Brandt,  John  C,  The  problem  of  the  Gegen- 
schein,  Astron.  Soc.  Pacific  Leaflet  391,  8  pp., 
January  1961. 

Brandt,  John  C,  Interplanetary  gas,  V,  A  hy- 
drogen cloud  of  terrestrial  origin,  Astrophys. 
J.,  134,  394-400,  1961. 


48 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Brandt,  John  C,  Interplanetary  gas,  VI,  On 
diffuse  extreme  ultraviolet  helium  radiation  in 
the  night  and  day  sky,  Astrophys.  J.,  184,  975- 
980,  1961. 

Brandt,  John  C,  On  the  problem  of  the  distance 
to  the  center  of  the  Galaxy  (abstract),  Publ. 
Astron.  Soc.  Pacific,  78,  324,  1961. 

Brandt,  John  C,  A  note  on  the  scale  of  the 
Galaxy,  Publ.  Astron.  Soc.  Pacific,  74,  142-145, 
1962. 

Brandt,  John  C.,  On  the  interpretation  of  the 
night  sky  Lyman  a  radiation  and  related 
phenomena,  Space  Research  II,  Proc.  Second 
Intern.  Space  Sci.  Symposium,  Florence,  pp. 
624-638,  edited  by  H.  C.  van  de  Hulst,  D.  de 
Jager,  and  A.  F.  Moore,  North-Holland  Pub- 
lishing Co.,  Amsterdam,  1961. 

Burbidge,  E.  M.,  see  Wildey,  R.  L. 

Burbidge,  G.  R.,  see  Wildey,  R.  L. 

Cameron,  A.  G.  W.,  The  formation  of  the  sun 
and  planets,  Icarus,  1,  13-69,  1962. 

Cayrel  de  Strobel,  Guisa,  A  comparison  of  photo- 
electric and  photographic  spectrophotometry, 
Ann.  Astrophys.,  24,  509-515,  1961. 

Cragg,  Thomas,  Rotation  of  Saturn,  Publ.  As- 
tron. Soc.  Pacific,  78,  318-322,  1961. 

Cragg,  Thomas,  Three  new  variable  stars,  Publ. 
Astron.  Soc.  Pacific,  78,  453-455,  1961. 

Deutsch,  Armin  J.,  Non-catastrophic  mass-loss 
from  stars,  Intern.  Astron.  Union  Symposium 
12:  Aerodynamic  phenomena  in  stellar  atmos- 
pheres, part  III,  pp.  238-259,  edited  by  R.  N. 
Thomas,  1961. 

Deutsch,  Armin  J.,  The  prospects  of  extrater- 
restrial astronomy,  Navigation,  8,  12-17,  1961. 

Deutsch,  Armin  J.,  P.  W.  Merrill,  and  P.  C. 
Keenan,  Behavior  of  absorption  features  in 
the  spectra  of  Mira  variables  (abstract), 
Astron.  J.,  66,  282,  1961. 

Deutsch,  Armin  J.,  see  also  Maestre,  Leonard  A. 

Eggen,  Olin  J.,  Stellar  groups,  Proc.  Symposium 
on  Stellar  Evolution,  La  Plata,  Argentina,  Nov. 
7-11,  1960,  edited  by  Jorge  Sahade,  Observa- 
torio  Astronomico,  Secci6n  Publicaciones,  La 
Plata,  Argentina,  1962,  pp.  233-250. 

Eggen,  Olin  J.,  SZ  Lyncis:  A  new  ultrashort- 
period  RR  Lyrae  variable,  Publ.  Astron.  Soc. 
Pacific,  74,  159-161,  1962. 

Fowler,  William  A.,  Jesse  L.  Greenstein,  and 
Fred  Hoyle,  Deuteronomy:  the  synthesis  of 
deuterons  and  the  light  nuclei  during  the  early 
history  of  the  solar  universe,  Am.  J .  Phys.,  29, 
393-403,  1961. 

Fowler,  William  A.,  Jesse  L.  Greenstein,  and 
Fred  Hoyle,  Nucleosynthesis  during  the  early 
history  of  the  solar  system,  Geophys.  J .,  Roy. 
Astron.  Soc,  6,  148-220,  1962. 


Fowler,  William  A.,  Jesse  L.  Greenstein,  and 
Fred  Hoyle,  Nuclear  history  of  the  solar  sys- 
tem (abstract),  Publ.  Astron.  Soc.  Pacific,  78, 
326-327,  1961. 

Gates,  H.  S.,  see  Zwicky,  Fritz. 

Gibson,  James,  see  Abt,  Helmut  A. 

Gomes,  Alercio  M.,  see  Humason,  Milton  L.; 
Zwicky,  Fritz. 

Greenstein,  Jesse  L.,  Stellar  evolution  and  the 
origin  of  the  chemical  elements,  Am.  Scientist, 
49,  449-473,  1961. 

Greenstein,  Jesse  L.,  Radio  sources  containing 
peculiar  ellipticals,  Astrophys.  J.,  135,  679-683, 
1962. 

Greenstein,  Jesse  L.,  and  Claude  Arpigny,  The 
visual  region  of  the  spectrum  of  Comet  Mrkos 
(1957d)  at  high  resolution,  Astrophys.  J.,  135, 
892-905,  1962. 

Greenstein,  Jesse  L.,  and  Fritz  Zwicky,  Super- 
nova 1959  SN  64,  Publ.  Astron.  Soc.  Pacific, 
74,  35-40,  1962. 

Greenstein,  Jesse  L.,  see  also  Fowler,  William  A.; 
Jugaku,  Jun. 

Griffin,  Roger  F.,  The  positions  of  optical  objects 
in  the  fields  of  42  radio  sources,  Monthly 
Notices  Roy.  Astron.  Soc,  in  press. 

Herzog,  Emil,  and  Konrad  Rudnicki,  Bright 
galaxies  with  unknown  redshifts,  Publ.  Astron. 
Soc.  Pacific,  74,  234-237,  1962. 

Hiltner,  W.  A.,  see  Bahner,  K.;  Kraft,  Robert  P. 

Hodge,  Paul  W.,  The  Fornax  dwarf  galaxy,  II, 
The  distribution  of  stars,  Astron.  J.,  66,  249- 
257,  1961. 

Hodge,  Paul  W.,  The  distribution  of  stars  in 
dwarf  elliptical  galaxies  (abstract),  Astron.  J., 
66,  286,  1961. 

Hodge,  Paul  W.,  The  distribution  of  stars  in  the 
Sculptor  dwarf  galaxy,  Astron.  J.,  66,  384-390, 
1961. 

Hodge,  Paul  W.,  Studies  of  the  Large  Magellanic 
Cloud,  VII,  The  open  cluster  NGC  1844, 
Astrophys.  J.,  134,  226-231,  1961. 

Hodge,  Paul  W.,  The  gravitational  stability  of 
the  NGC  7619  group  of  galaxies,  Astrophys. 
J.,  184,  262-264,  1961. 

Hodge,  Paul  W.,  Studies  of  the  Large  Magellanic 
Cloud,  VI,  Properties  of  1057  field  stars  (ab- 
stract), Astrophys.  J.,  184,  666-667,  1961. 

Houziaux,  Leo,  Stark  and  Doppler  broadening 
for  high  Balmer  and  Paschen  lines,  I,  The 
absorption  coefficient,  Ann.  Astrophys.,  24, 
541-549,  1961;  (abstract)  Publ.  Astron.  Soc. 
Pacific,  73,  328-329,  1961. 

Houziaux,  Leo,  Mesure  et  distribution  spectrale 
de  l'energie  recue  des  etoiles,  Scientia,  96,  369- 
375,  1961 


MOUNT      WILSON     AND      PALOMAR     OBSERVATORIES 


49 


Houziaux,  Leo,  Forbidden  nitrogen  I  lines  in  the 
infrared  solar  spectrum,  Z.  Astrophys.,  53,  237- 
239,  1961. 

Houziaux,  Leo,  On  the  infrared  spectrum  of 
Pleione,  Publ.  Astron.  Soc.  Pacific,  74,  250- 
253,  1962. 

Howard,  Robert,  Solar  magnetic  fields,  Astron. 
Soc.  Pacific  Leaflet  896,  8  pp.,  June  1962. 

Hoyle,  Fred,  see  Fowler,  William  A. 

Humason,  Milton  L.,  Photographs  of  planets 
with  the  200-inch  telescope,  The  Solar  System, 
vol.  Ill,  Planets  and  Satellites,  chapter  16,  p. 
572  and  15  plates,  edited  by  G.  P.  Kuiper  and 
B.  M.  Middlehurst,  University  of  Chicago 
Press,  1961. 

Humason,  Milton  L.,  C.  E.  Kearns,  and  Alercio 
M.  Gomes,  The  1961  Palomar  supernova 
search,  Publ.  Astron.  Soc.  Pacific,  74,  215-218, 
1962. 

Humason,  Milton  L.,  see  also  Zwicky,  Fritz. 

Jeffers,  Hamilton  M.,  see  Abt,  Helmut  A. 

Jones,  D.  H.  P.,  The  double  star  ADS  3210, 
01185,  Observatory,  82,  29-31,  1962. 

Joy,  Alfred  H.,  Ralph  Elmer  Wilson,  Biographi- 
cal Memoirs,  vol.  36,  pp.  314-329,  published 
for  National  Academy  of  Sciences  of  the 
United  States  by  Columbia  University  Press, 
1962. 

Joy,  Alfred  H.,  Spectroscopic  absolute  magni- 
tudes: A  review,  J.  Roy.  Astron.  Soc.  Canada, 
56,  67-78,  1962. 

Joy,  Alfred  H.,  Paul  Willard  Merrill,  1887-1961, 
Publ.  Astron.  Soc.  Pacific,  74,  41-43,  1962. 

Joy,  Alfred  H.,  Paul  Willard  Merrill,  Quart.  J. 
Roy.  Astron.  Soc,  3,  45-47,  1962. 

Jugaku,  Jun,  and  Wallace  L.  W.  Sargent,  The 
spectrum  of  a  Sculptoris,  Publ.  Astron.  Soc. 
Pacific,  73,  249-255,  1961. 

Jugaku,  Jun,  Wallace  L.  W.  Sargent,  and  Jesse 
L.  Greenstein,  An  abundance  analysis  of  3 
Centauri  A,  Astrophys.  J.,  134,  783-796,  1961. 

Jugaku,  Jun,  see  also  Sargent,  Wallace  L.  W. 

Kearns,  C.  E.,  see  Humason,  Milton  L. 

Keenan,  P.  C,  see  Deutsch,  Armin  J. 

Kraft,  Robert  P.,  Color  excesses  for  supergiants 
and  classical  cepheids,  V,  The  period-color  and 
period-luminosity  relations:  a  revision,  Astro- 
phys. J.,  134,  616-632,  1961. 

Kraft,  Robert  P.,  Binary  stars  among  cataclys- 
mic variables,  I,  U  Geminorum  stars  (dwarf 
novae),  Astrophys.  J.,  135,  408-423,  1962. 

Kraft,  Robert  P.,  Nova  (WZ)  Sagittae  as  a 
binary  star  (abstract),  Science,  134,  1433,  1961. 

Kraft,  Robert  P.,  Exploding  stars,  Sci.  American, 
206,  54-63,  1961. 


Kraft,  Robert  P.,  and  W.  A.  Hiltner,  Color  ex- 
cesses for  supergiants  and  classical  cepheids, 
VI,  On  the  intrinsic  colors  and  Hess  diagram 
of  late-type  supergiants,  Astrophys.  J.,  134, 
850-860,  1961. 

Kraft,  Robert  P.,  see  also  Bahner,  K. 

Maestre,  Leonard  A.,  and  Armin  J.  Deutsch, 
List  of  absorption  lines  in  two  ultra-sharp  line 
A  stars,  Astrophys.  J.,  134,  562-567,  1961. 

Merrill,  Paul  W.,  Unidentified  lines  in  spectra  of 
sun  and  stars,  Astrophys.  J.,  134,  556-561, 
1961. 

Merrill,  Paul  W.,  see  also  Deutsch,  Armin  J. 

Mihalas,  Dimitri  M.,  Light  curve  of  Humason's 
supernova  in  Virgo  (abstract),  Astron.  J.,  67, 
118-119,  1962;  Publ.  Astron.  Soc.  Pacific,  74, 
116-124,  1962. 

Minkowski,  R.,  NGC  6166  and  the  cluster  Abell 
2199,  Astron.  J.,  66,  558-561,  1961. 

Minkowski,  R.,  Radio  sources,  galaxies,  and 
clusters  of  galaxies,  Proc.  Natl.  Acad.  Sci. 
U.  S.,  in  press. 

Munch,  G.,  and  A.  Unsold,  Interstellar  gas  near 
the  sun,  Astrophys.  J.,  135,  711-714,  1962. 

Nicholson,  Seth  B.,  Award  of  the  Bruce  Gold 
Medal  to  Grote  Reber,  Publ.  Astron.  Soc. 
Pacific,  74,  183-186,  1962. 

Nicholson,  Seth  B.,  and  Oliver  R.  Wulf,  The 
diurnal  variation  of  K  indices  of  geomagnetic 
activity  on  disturbed  days  in  1940-1948,  J. 
Geophys.  Res.,  66,  2399-2404,  1961;  (abstract) 
Science,  134,  1434,  1961. 

Oke,  J.  B.,  An  analysis  of  the  absolute  energy 
distribution  in  the  spectrum  of  8  Cephei,  As- 
trophys. J.,  134,  214-221,  1961. 

Oke,  J.  B.,  see  also  Searle,  Leonard. 

Pettit,  Edison,  Planetary  temperature  measure- 
ments, The  Solar  System,  vol.  Ill,  Planets  and 
Satellites,  chapter  10,  pp.  400-428,  edited  by 
G.  P.  Kuiper  and  B.  M.  Middlehurst,  Univer- 
sity of  Chicago  Press,  1961. 

Preston,  George  W.,  A  coarse  analysis  of  three 
RR  Lyrae  stars,  Astrophys.  J.,  134,  633-650, 
1961. 

Rodgers,  A.  W.,  Photoelectric  spectrophotom- 
etry of  O-type  stars,  Monthly  Notices  Roy. 
Astron.  Soc,  122,  413-419,  1961. 

Rudnicki,  Konrad,  see  Herzog,  Emil. 

Sandage,  Allan,  The  light  travel  time  and  the 
evolutionary  correction  to  magnitudes  of  dis- 
tant galaxies,  Astrophys.  J.,  184,  916-926, 
1961. 

Sandage,  Allan,  Photometric  data  for  the  old 
galactic  cluster  NGC  188,  Astrophys.  J.,  135, 
333-348,  1962. 

Sandage,  Allan,  The  ages  of  M67,  NGC  188,  M3, 
M5,  and  M13,  according  to  Hoyle's  1959 
models,  Astrophys.  J.,  185,  349-365,  1962. 


50 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Sandage,  Allan,  Introductory  report,  Proc.  Sym- 
posium on  Stellar  Evolution,  La  Plata,  Argen- 
tina, Nov.  7-11,  1960,  edited  by  Jorge  Sahade, 
Observatorio  Astron6mico,  Secci6n  Publica- 
ciones,  La  Plata,  Argentina,  1962,  pp.  1-22. 

Sandage,  Allan,  The  ages  of  the  oldest  stars  in 
the  Galaxy  compared  with  the  cosmic  expan- 
sion time,  Proc.  Symposium  on  Stellar  Evolu- 
tion, La  Plata,  Argentina,  Nov.  7-11,  1960, 
edited  by  Jorge  Sahade,  Observatorio  Astro- 
n6mico,  Secci6n  Publicaciones,  La  Plata, 
Argentina,  1962,  pp.  119-135. 

Sandage,  Allan,  Travel  time  for  light  from  distant 
galaxies  related  to  the  Riemannian  curvature 
of  the  universe  (abstract),  Science,  184,  1434, 
1961. 

Sandage,  Allan,  see  also  Abt,  Helmut  A.;  Smith, 
Lewis;  Wildey,  R.  L. 

Sargent,  Wallace  L.  W.,  Circumstellar  envelope 
of  rho  Cassiopeiae,  Astrophys.  J.,  134,  142-160, 
1961. 

Sargent,  Wallace  L.  W.,  and  Jun  Jugaku,  The 
existence  of  He3  in  3  Centauri  A,  Astrophys.  J., 
134,  777-782,  1961. 

Sargent,  Wallace  L.  W.,  see  also  Jugaku,  Jun. 

Schmidt,  Maarten,  Galactic  structure  at  optical 
and  radio  wavelengths,  J.  Roy.  Astron.  Soc. 
Can.,  55,  149-166,  1961. 

Schmidt,  Maarten,  A  preliminary  discussion  of 
helium  abundances,  Proc.  Symposium  on  Stel- 
lar Evolution,  La  Plata,  Argentina,  Nov.  7-11, 
1960,  edited  by  Jorge  Sahade,  Observatorio 
Astron6mico,  Secci6n  Publicaciones,  La  Plata, 
Argentina,  1962,  pp.  61-66. 

Schmidt,  Maarten,  The  evolution  of  the  sun's 
neighborhood,  Proc.  Symposium  on  Stellar 
Evolution,  La  Plata,  Argentina,  Nov.  7-11, 
1960,  edited  by  Jorge  Sahade,  Observatorio 
Astronomico,  Secci6n  Publicaciones,  La  Plata, 
Argentina,  1962,  pp.  67-75. 


Searle,  Leonard,  and  J.  B.  Oke,  Effective  temper- 
atures of  F-type  stars  using  Hy  line  profiles, 
Astrophys.  J.,  185,  790-801,  1962. 

Smith,  Lewis,  and  Allan  Sandage,  Color-magni- 
tude diagram  of  the  strong-line  globular  cluster 
NGC  6712  (abstract),  Astron.  J.,  67,  121,  1962. 

Swope,  Henrietta  H.,  see  Baade,  Walter. 

Tifft,  William  G.,  Multicolor  photometry  of 
bright  galaxies,  I,  Astron.  J.,  66,  390-404, 
1961. 

Traving,  Gerhard,  Atmospheres  of  two  B-type 
stars  in  the  galactic  halo,  Astrophys.  J.,  135, 
439-458,  1962. 

Unsold,  G.,  see  Munch,  G. 

Wildey,  R.  L.,  E.  M.  Burbidge,  A.  R.  Sandage, 
and  G.  R.  Burbidge,  On  the  effect  of  Fraun- 
hofer  lines  on  U,  B,  V  measurements,  Astro- 
phys. J.,  135,  94-108,  1962. 

Wulf,  Oliver  R.,  see  Nicholson,  Seth  B. 

Zwicky,  Fritz,  L'astronomie  et  la  marche  dans 
l'univers,  I  Astronomie,  76,  55-65,  1962. 

Zwicky,  Fritz,  Die  Kerne  von  Galaxien  als  neue 
charakteristische  Aggregate  kosmischer  Ma- 
terie,  Physik.  Blatter,  17,  393-399,  1961. 

Zwicky,  Fritz,  Cosmic  and  terrestrial  tests  for 
the  rest  mass  of  gravitons,  Publ.  Astron.  Soc. 
Pacific,  73,  314-317,  1961. 

Zwicky,  Fritz,  Supernovae  and  chains  of  super- 
novae  as  launchers  of  stars  and  galaxies,  Publ. 
Astron.  Soc.  Pacific,  74,  70-73,  1962. 

Zwicky,  Fritz,  Possible  operations  on  the  moon, 
Space  Flight,  3,  177-179,  1961. 

Zwicky,  Fritz,  M.  L.  Humason,  A.  M.  Gomes, 
and  H.  S.  Gates,  Progress  report  on  super- 
novae,  Publ.  Astron.  Soc.  Pacific,  73,  351-354, 
1961. 

Zwicky,  Fritz,  see  also  Greenstein,  Jesse  L. 


Geophysical  Laboratory 


Washington,  District  of  Columbia 
Philip  H.  Abelson 
Director 


Contents 


Introduction 


53 


Experimental  Petrology 56 

Pyroxenes 56 

The    join    diopside-Ca    Tschermak's 

molecule  at  atmospheric  pressure  .      56 
Phase      relations      in      the      system 
CaMgSi2q6-CaAl2Si06-Si02        at 
low  and  high  pressure    ....      59 
X-ray  data  for  diopsidic  pyroxenes   .      61 
Melting    relations    in    the    system 

diopside-anoi  thite-silica  ...     65 
The  system  MgSi()3-CaMgSi20^    .      .     68 
The  system  diopside-enstatite-silica      .     75 
Preliminary  results  on  melting  rela- 
tions of  synthetic  pyroxenes  on  the 
diopside-hedenbergite  join       ...     82 

Metamorphic  petrology 82 

Metamorphic  reactions  involving  two 

volatile  components 82 

Synthesis  and  stability  of  anthophyllite  .     85 
Quartz-chlorite    assemblages    in    the 
system  MgO-Al203-Si02-H20       .      .     88 
Alkali-rich  igneous  rocks  and  minerals      .     91 
The    system  _  Na20-Al203-Fe203-Si02 

and  its  bearing  on  the  alkaline  rocks  .     91 
Peralkaline     residual     liquids:     some 

petrogenetic  considerations     ...     95 
The  system  nepheline-diopside  ...     96 
A     reconnaissance     of     the     systems 
acmite-diopside    and    acmite-neph- 

eline 98 

Accessory  minerals 100 

Investigations    in    the    system    FeO- 

Fe203-Ti02     .      .      .      .      .      .      .100 

Magnetite-ilmenite  relations  .      .      .   100 
Application  of  experimental  data  to 

natural  minerals      - 105 

Relations  between  ilmenite,  hema- 
tite, magnetite,  and  rutile  .      .      .    106 
Stability     relations     of     dravite,     a 

tourmaline 106 

Mantle  minerals 107 

Effect  of  pressure  on  the  melting  of 
pyrope 109 

Statistical  Petrography 112 

Sanidine  phenocrysts  in  some  peralkaline 

volcanic  rocks 112 

Bulk  analyses  and  norms      .      .      .      .114 
Feldspar  phenocrysts  of  the  analyzed 

rocks 115 

Projection  of  results  into  "petrogeny's 
residua  system" 116 

Variance    relations    in    some    published 
Harker  diagrams 118 

The   treatment   of   FeO   and   Fe203  in 
Harker  diagrams 119 

On  the  relative  scarcity  of  intermediate 
members  in  the  oceanic  basalt-tra- 
chyte association 121 

Granite  in  Port  Clyde  peninsula    .      .      .    123 

Feldspar  in  the  granite  of  the  Port  Clyde 
peninsula 126 

Two-mica  granite  and  hornblende-biotite 
granite 128 


Crystallography 130 

Relationships  between  crystal  structure 

and  crystal  morphology 130 

Lattice  constant  refinement  .  .  .  .132 
The  crystal  structure  of  sillimanite  .  .135 
The  crystal  structure  of  Fe  mica  .  .  .139 
On  the  transitions  of  bornite    .      .      .      .139 

Ore  Minerals 142 

The  Mo-S  system 143 

The  Fe-Ni-S  system 144 

Liquid  immiscibility 144 

Pentlandite  stability  relations    .      .      .146 
Bravoite  stability  relations   .      .      .      .149 

The  Fe-Mo-S  system 150 

The  Cu-Ni-S  system 151 

The  Fe-Ni-As  system 152 

The  Cu-Fe-S  system      ......    154 

Pyrrhotite-pyrite-chalcopyrite  relations   154 
Exsolution  textures  and  rates  in  solid 

solutions  involving  bornite        .      .155 

Exsolution  textures 155 

Rates  of  exsolution 156 

Chalcocite-chalcopyrite  assemblages     .    157 
Heating  experiments  on  natural  born- 

ites  .      .      . 159 

Method  for  mixing  liquids  at  controlled 

temperatures  160 

Pyrrhotite  from  Tern  Piute,  Nevada  .      .161 

Stony  Meteorites 163 

Iron  Meteorites 165 

The  system  Fe-Ni-S 165 

The  system  Fe-Ni-P 166 

Higher  phosphides  in  the  system  Fe-P      .  166 

Geothermal  Calculations 168 

The  Ages  of  Rocks  and  Minerals       .      .      .173 
Geographic  distribution  of  mineral  ages 
in    the    central    portion    of    North 

America 173 

Ages  of  minerals  from  the  Coutchiching 

sediments,  Rainy  Lake,  Ontario      .      .176 
Age  relation  between  the  Karelian  and 
Svecofennian  orogenies  in  Finland   .      .178 

Organic  Geochemistry 179 

Paleobiochemistry 179 

Thermal  stability  of  algae     .      .      .      .179 
Fatty  acids  in  sedimentary  rocks     .      .181 
The  isolation  of  organic  compounds 
from  Precambrian  rocks    ....    184 
The  biogeochemistry  of  the  stable  iso- 
topes of  carbon 187 

The  isotopic  composition  of  the  carbon 

of  fatty  acids 187 

The  stable  isotopes  of  carbon  in  the 
carbonate  and  reduced  carbon  of 
Precambrian  sediments      .      .      .      .190 

Miscellaneous  Administration  .  .  .  .192 
Institute  on  Isotopes  and  Radioactivity  .  192 
Journal  of  Geophysical  Research     .      .      .192 

Lectures 192 

Penologists'  Club 193 

Summary  of  Published  Work       .      .      .      .194 

Bibliography 201 

References  Cited 202 

Personnel 208 


Carnegie  Institution  of  Washington  Year  Book  61,  1961-1962 


INTRODUCTION 

The  Geophysical  Laboratory  continues  fields    of    sillimanite,     corundum,     and 

its  diversified  program  of  studying  the  probably     also     "/?    alumina"     on    the 

earth    through    application    of    physical  liquidus  near  its  composition.   Pressure 

science  and  mathematics.   Superficially,  greatly  increases  the  maximum  amount 

the  areas  of  effort  during  the  report  year  of  alumina  that  the  pyroxene  can  accom- 

appear  to  be  very  similar  to  those  of  the  modate  in  its  structure ;  lime  Tschermak's 

preceding  period.  There  was  activity  in  molecule    (CaAl2Si06)    is    stable    up    to 

experimental  petrology,  statistical  petrol-  1500°C  at  20  kb,  and  there  is  complete 

ogy,  crystallography,  ore  minerals,  mete-  solid   solution   between   this   phase   and 

orites,  geothermal  calculations,  the  ages  diopside.  This  work  clearly  shows  that 

of    rocks    and    minerals,    and    organic  important  changes  in  melting  relations 

geochemistry.  There  were,  however,  sub-  are  produced  by  pressure  and  that  phase 

stantial   shifts   of   emphasis   within   the  diagrams  determined  at  atmospheric  pres- 

program.  For  instance,  much  of  the  work  sure  cannot  be  applied  to  the  production 

this  year  in  experimental  petrology  was  of  magma  at  great  depths  in  the  earth, 

focused  on  the  pyroxenes,   and  greater  Boyd  and  Schairer  present  final  results 

emphasis  was  placed  on  phase  equilibria  on  the  system  MgSi03-CaMgSi206.  Most 

at  higher  pressures.  A  substantial  effort  mafic  rocks  contain  two  pyroxenes,  and 

was  expended  on  studies  of  the  miner-  this  binary  system  is  fundamental  to  an 

alogy  of  meteorites.  New  investigations  understanding    of    the    mineralogy    and 

of  organic  geochemistry  were  initiated,  genesis  of  these  rocks.  It  was  found  that 

including  analysis  of  Precambrian  carbo-  the  solvus  intersects  the  solidus  over  a 

naceous  materials.  composition  interval   of  35   weight  per 

The  Laboratory  is  continuing  its  broad  cent,  so  that  solid  solution  between  the 

program  of  studying  the  phase  relations  Ca-rich  and  Mg-rich  pyroxenes  is  much 

in  basalts  and  their  derivative  rocks.  The  more    restricted    than    was    previously 

alkali-type  basalts  in  particular  have  been  thought.  Evidence  for  a  hitherto  unrecog- 

examined   from  several  viewpoints   this  nized    form    of    Mg-rich    pyroxene    was 

year.  The  fundamental  joins  nepheline-  found    in    runs    at    temperatures    above 

diopside,  acmite-diopside,  and  nepheline-  1385°C. 

acmite,  bearing  on  alkali-type  rocks,  were  Yoder     and     Tilley     continued     with 

worked  out  by  Schairer,  Yagi,  and  Yoder.  Schairer  their  heating  experiments  at  1 

Studies  in  the  system  Na20-Fe203-Al203-  atmosphere  on  natural  basalts  and  nat- 

Si02  have  been  carried  out  by  Schairer  ural  coexisting  pyroxenes,  and  completed 

and  Bailey  in  view  of  its  importance  to  the  preparation  of  their  extensive  mono- 

the  peralkaline  derivative  rocks.  graph,  "The  origin  of  the  basalt  magmas: 

Work  by  Clark,  Schairer,  and  de  An  experimental  study  of  natural  and 
Neufville  on  the  composition  plane  synthetic  rock  systems." 
CaMgSi206-CaAl2Si06-Si02  at  low  and  One  of  the  crucial  systems  in  petrology, 
high  pressures  represents  the  first  exten-  diopside-forsterite-silica,  has  been  re- 
sive  study  of  the  effect  of  pressure,  investigated  in  part  by  Schairer  and 
transmitted  by  an  inert  medium,  on  Yoder.  The  revisions  are  of  exceptional 
melting  and  subsolidus  relations  in  a  import  to  the  general  differentiation  of 
portion  of  a  complex  quaternary  system,  magmas  and  the  nature  of  pyroxenes  that 
Some  of  the  phase  relations  at  20  kb  are  crystallize  in  them.  Many  refinements 
totally  different  from  those  at  atmos-  occur  within  a  very  small  temperature 
pheric  pressure.  Anorthite  melts  incon-  interval,  and  by  close  temperature  con- 
gruent ly  at  this  pressure,  and  there  are  trol    and    calibration    seven    isothermal 

53 


54  CARNEGIE     INSTITUTION      OF      WASHINGTON 

sections  in  5°  intervals  were  mapped  out.  ilmenite- magnetite      intergrowths     may 

Turnock  has  surveyed  melting  relations  form  by  oxidation. 

of    synthetic    pyroxenes    on    the    join         The  most  important  natural  compound 

diopside-hedenbergite,    using    controlled  of    boron   is   tourmaline.    The    stability 

atmospheres  with  fixed  partial  pressures  range   of   one   of  the   end   members   of 

of  oxygen.  tourmaline,   dravite,   has   been   outlined 

In  his  studies  of  metamorphic  reactions  under  hydrous  conditions,  special  care 
Greenwood  has  made  two  contributions,  being  taken  to  retain  the  sodium  and 
one  theoretical  and  the  other  experi-  boron.  The  work  was  carried  out  mainly 
mental.  He  has  derived  equations  for  by  Robbins,  of  the  National  Bureau  of 
equilibrium  boundaries  between  reacting  Standards,  working  during  his  free  time 
phase  assemblages  in  systems  that  contain  at  the  Laboratory,  with  the  help  of  Yoder. 
two  volatile  components  and  are  subject  Boyd  and  England  have  found  that 
to  variation  of  pressure  and  temperature,  pyrope  garnet  melts  incongruently  to 
These  equations  have  the  same  form  as  spinel  and  liquid  in  the  same  pressure 
the  usual  expressions  for  crystal-liquid  range  as  that  in  which  basalts  are  be- 
equilibria  but  do  not  carry  the  restriction  lieved  to  form  in  suboceanic  areas.  Spinel 
that  the  relative  proportions  of  the  two  contains  no  silica,  and  so  the  liquid  that 
volatile  components  are  limited  by  the  forms  by  this  reaction  is  oversaturated. 
proportions  of  the  other  components.  The  Hence,  the  incongruent  melting  of  pyrope 
effect  of  removing  this  restriction  is  to  may  explain  the  development  of  over- 
make  stable  many  reactions  that  would  saturated  basalts  from  mantle  rocks 
normally  be  regarded  as  metastable.  He  believed  to  be  undersaturated.  The 
has  also  been  able  to  delineate  the  upper  classic  reaction  that  has  been  used  io 
and  lower  stability  limits  of  anthophyllite  explain  the  development  of  oversaturated 
and  observe  the  nucleation  kinetics  of  the  lavas  is  the  incongruent  melting  of 
mineral.  enstatite,  discovered  by  Bowen  in  runs 

Many  of  the  principal  mineral  assem-  at  atmospheric  pressure.  However,  Boyd 

blages  found  in  low-grade  metamorphic  and    England    showed    last    year    that 

rocks  include  chlorite  and  quartz.  Previ-  enstati(te  melts  congruently  at  pressures 

ous  synthesis  studies  had  indicated  that  greater  than  those  present  at  moderate 

these  are  not  compatible  minerals.  New  depths  in  the  earth's  crust, 
studies  by  Fawcett  and  Yoder,  involving         In  statistical  petrography,  studies  of 

experiments  of  long  duration,  have  sue-  the  phenocryst-groundmass  relations  in 

cessfully  resolved  this  apparent  conflict  peralkaline  lavas  (Zies  and  Chayes)  and 

with  nature.  of  the  relations  between  two-mica  and 

Lindsley    has    shown    that    synthetic  biotite-hornblende  granites  in  the  Port 

titaniferous    magnetites    in    equilibrium  Clyde   peninsula    (Suzuki   and    Chayes) 

with  ilmenite  contain  little  or  no  ilmenite  have    been    continued,    with    results    of 

in  solid  solution  at  temperatures  below  considerable   interest.    Chayes   has   now 

1000°C;  compositions  of  these  magnetites  fully  substantiated  his  suggestion  of  last 

lie  close  to  the  magnetite-ulvospinel  join,  year  that  in  the  typical  Harker  array  the 

The  titanium  content  of  the  magnetites  variance   of   silica   is   approximately  an 

is  strongly  dependent  on  oxygen  f ugacity  order  greater  than  any  other  variance ;  in 

as   well   as   on   temperature   and   hence  23    of    25    published    arrays    for   which 

cannot    be    used    as    a    simple    geologic  calculations    have    been    completed    the 

thermometer.  Hydrothermal  oxidation  of  variance  of  silica  is  larger  than  the  sum 

magnetite-ulvospinel  solid  solutions  yields  of  all  other  variances.  This  great  excess 

ilmenite-magnetite   intergrowths   textur-  of  silica  variance  is  the  most  important 

ally  similar  to  natural  occurrences,  sup-  single  influence  on  the  correlations  that 

porting  the  hypothesis  that  many  natural  characterize  a  Harker  array. 


GEOPHYSICAL   LABORATORY 


55 


J.  D.  H.  Donnay  (Johns  Hopkins  Uni- 
versity) and  G.  Donnay  have  applied 
their  second  generalization  of  the  law  of 
Bravais  to  the  elucidation  of  the  external 
forms  of  ionic  crystal  structures.  Their 
study  of  the  mineral  barite,  the  morphol- 
ogy of  which  had  heretofore  remained 
unexplained,  leads  to  the  new  concept  of 
"centers  of  charges."  These  turn  out  to 
be  the  equipoints  of  the  bond  assem- 
blages. This  punctualization  of  charges 
is  apparently  the  key  to  the  morphologies 
of  ionic  crystals,  the  interpretation  of 
which  is  a  generalization  of  that  of  the 
NaCl  morphology. 

Burnham  has  developed  a  least-squares 
technique  for  refinement  of  lattice  con- 
stants of  crystals,  which  has  been 
programmed  for  the  IBM  7090  digital 
computer.  The  procedure,  already  em- 
ployed successfully  by  other  staff  mem- 
bers, has  the  following  features.  It  is 
applicable  to  crystals  of  any  symmetry 
and  will  accept  data,  from  cards  or  tape, 
either  as  angle  measurements  for  any 
wavelength  or  in  the  form  of  calculated 
d  values.  Observations  may  be  weighted 
according  to  any  scheme,  and  up  to  nine 
systematic  correction  terms  may  be 
included  with  each  observation. 

Morimoto  has  studied  the  transition 
mechanisms  by  which  the  three  poly- 
morphic forms  of  bornite  are  inter- 
converted. 

Studies  of  many  facets  of  the  ore 
minerals  continue.  Increased  emphasis  is 
being  placed  on  application  of  laboratory 
findings  to  ores.  However,  the  program  is 
still  predominantly  a  laboratory  investi- 
gation of  phase  equilibria.  The  Mo-S 
system  was  studied  by  Morimoto  and 
Kullerud,  who  found  that  the  "Mo2S3" 
phase  is  only  stable  above  605°C.  Very 
important  results  have  been  obtained  on 
the  Fe-Ni-S  system,  in  which  at  high 
temperatures  Kullerud  found  liquid  im- 
miscibility  over  a  large  region  extending 
across  the  sulfur- rich  part  of  the  system. 
In  other  studies  he  also  showed  that 
pentlandite  breaks  down  at  610°C  and 
that  bravoite  is  only  stable  below  137°C. 


In  the  Fe-Mo-S  system  Kullerud  and 
Buseck  found  that  the  minerals  pyrite 
and  molybdenite  are  stable  together 
below  726°C.  The  Cu-Ni-S  system  is 
being  investigated  by  Moh  and  Kullerud, 
who  have  finished  the  600°C  isothermal 
section.  Buseck  studied  the  Fe-Ni-As 
system  and  has  synthesized  the  new 
mineral  oregonite.  The  solid  solutions  in 
the  uni variant  region  containing  pyrite, 
pyrrhotite,  chalcopyrite,  and  vapor  were 
examined  by  von  Gehlen  and  Kullerud. 
They  found  that  at  600°C  application  of 
the  pyrite-pyrrhotite  thermometer  gives 
temperatures  about  50°C  lower  when 
chalcopyrite  is  present  than  when  it  is 
absent.  Brett  studied  exsolution  textures 
from  solid  solutions  involving  bornite 
(digenite-bornite,  chalcocite-bornite,  chal- 
copyrite-bornite) .  His  results  indicate 
that  textural  evidence  alone  does  not 
permit  drawing  of  reliable  conclusions 
about  the  thermal  history  of  the  minerals 
that  form  these  solid  solution  pairs. 

Two  studies  were  directed  at  problems 
involving  meteorites.  Clark,  studying  the 
system  Fe-Ni-P,  has  demonstrated  that 
the  Fe/Ni  ratio  of  the  schreibersite, 
(Fe,Ni)3P,  in  equilibrium  with  both 
kamacite  and  taenite  changes  measurably 
with  temperature.  The  phosphide  is  a 
common  constituent  of  iron  meteorites, 
and  its  composition,  along  with  the 
compositions  of  the  alloy  phases,  will  help 
to  trace  the  history  of  these  extraterres- 
trial bodies.  A  curious,  but  simple, 
relationship  that  has  emerged  from  the 
synthetic  system  is  that  the  ratio 
Ni/(Fe  -f-  Ni)  is  the  same  in  the  schrei- 
bersite as  in  the  taenite  (7  alloy)  with 
which  it  is  in  equilibrium.  Ramdohr  has 
examined  polished  sections  of  more  than 
a  hundred  stony  meteorites.  He  has 
identified  more  than  twenty  minerals  y 
half  of  which  had  not  been  seen  previously 
in  stony  meteorites.  Native  gold  was  seen 
in  specimen.  In  addition,  he  observed 
twelve  new  minerals  that  have  not  been 
fully  identified  but  whose  composition 
can  be  partially  inferred  from  associations 
with     known     substances.     Noteworthy 


56 


CARNEGIE      INSTITUTION      OF      WASHINGTON 


structural  and  textural  relations  were  also 
seen,  including  localized  droplets  of  fused 
troilite  and  iron. 

In  a  continuation  of  the  theoretical 
geothermal  studies  of  the  past  few  years, 
the  effect  on  heat  flow  at  the  surface 
produced  by  very  high  thermal  conduc- 
tivity at  depth  in  the  earth  has  been 
investigated  by  Clark.  He  has  found  that 
under  certain  circumstances  high  conduc- 
tivity at  depth  reduces  the  surface  flux. 
Under  other  conditions  the  opposite 
effect  is  produced.  The  possibility  of 
variable  conductivity  in  the  earth  intro- 
duces an  ambiguity  in  the  interpretation 
of  heat-flow  measurement  in  addition  to 
ambiguities  caused  by  lack  of  precise 
knowledge  of  the  distribution  of  radio- 
activity and  the  initial  temperature.  A 
second  geothermal  investigation  concerns 
the  cooling  of  the  deep  mantle.  It  has 
been  found  that  appreciable  cooling  could 
take  place  if  the  initial  thermal  gradient 
were  sufficiently  steep,  but  it  is  not  yet 
clear  whether  such  a  steep  gradient  is 
tolerable  on  other  grounds. 

Zircon  age  studies  have  been  made  by 
Davis  on  the  ancient  igneous  and  sedi- 
mentary rocks  at  Rainy  Lake,  Ontario. 
The  results  indicate  that  all  the  zircons 
crystallized  about  2750  million  years  ago. 
The  rocks  were  eroded  to  form  sediments, 
which  were  subsequently  metamorphosed 
about  2600  million  years  ago. 

Additional  age  determinations  by  Til- 
ton  and  Kouvo  in  Finland  show  that  the 
Karelian  and  Svecofennian  orogenies 
occurred  at  about  the  same  time,  although 
geological  evidence  suggests  that  the 
Svecofennian  orogeny  is  the  older. 


A  geochronological  map  of  the  United 
States  and  southern  Canada,  based  on 
several  hundred  mineral  age  determina- 
tions, has  been  constructed.  The  Pre- 
cambrian  rocks  occur  in  belts  or  zones, 
with  younger  rocks  on  either  side  of  an 
old  central  belt. 

Abelson  and  Parker  have  isolated 
saturated  fatty  acids  including  stearic, 
palmitic,  and  myristic  from  rocks  as  old 
as  500  million  years.  Very  recent  sedi- 
ments contain  these  same  entities  and 
virtually  no  unsaturated  types.  This 
relationship  differs  sharply  from  that 
noted  in  algae,  which  ostensibly  are  the 
major  source  of  organic  matter  in 
sediments.  Parker  has  isolated  pure  fatty 
acids  from  algae  and  found  that  their 
C13/C12  ratios  differed  from  the  C13/C12 
ratio  of  total  cell.  Different  types  of  fatty 
acids  from  the  same  organism  have  the 
same  C13/C12  ratio. 

Hoering  has  studied  geochemical  evi- 
dence for  the  existence  of  life  in  Pre- 
cambrian  rocks.  The  fractionation  of  the 
stable  isotopes  of  carbon  into  a  C13-rich 
carbonate  fraction  and  a  C13-depleted 
reduced  fraction,  which  is  characteristic 
of  sedimentary  rocks  of  known  biological 
association,  was  found  to  exist  in  some  of 
the  oldest  known  sedimentary  rocks, 
including  the  Bulawayan  limestone,  which 
has  a  minimum  age  of  2.7  billion  years. 
Hoering  also  has  isolated  a  number  of 
organic  compounds  from  these  Precam- 
brian  rocks.  The  chemicals  are  similar  to 
those  that  have  been  found  in  coal.  Both 
these  results  are  consistent  with  the 
existence  of  life  and  photosynthesis  during 
early  Precambrian  times. 


EXPERIMENTAL   PETROLOGY 


Pyroxenes 

The  Join  Diopside-Ca  Tschermak's 

Molecule  at  Atmospheric  Pressure 

John  de  Neufville  and  J.  F.  Schairer 

In  order  to  study  the  extent  of  the 
substitution  of  A1203  in  diopside,  Hytonen 


(Year  Book  60)  prepared  three  series  of 
compositions  on  the  plane  enstatite- 
wollastonite- corundum.  During  the  past 
year  his  work  on  the  diopside-Ca  Tscher- 
mak's  molecule  (CaAl2Si06,  henceforth 
abbreviated  CTs)  series  has  been  extend- 
ed to  the  CTs  composition.  Figure  1  is  a 
temperature-composition   section  at   at- 


GEOPHYSICAL   LABORATORY 


57 


1600°  - 


Diopside 
Ca  Mg  Si206 


30  40  50  60 

Wt.  %CaAI2Si06 


90  CaTschermak's 
Molecule 
Ca  AI2Si06 


Fig.  1.  Temperature  /composition  plot  of  data  obtained  on  the  join  diopside-Ca  Tschermak's 
molecule  (CaAl2Si06)  at  1  atmosphere.  Abbreviations  for  phases  encountered:  DiS8,  diopside  solid 
solution;  Mel8S,  melilite  solid  solution;  Ak89Gehn,  etc.,  specific  melilite  composition  in  terms  of  weight 
per  cent  akermanite  and  gehlenite;  Sp,  spinel;  Fe,  forsterite;  An,  anorthite;  Al,  "(3  alumina"  and/or 
corundum;  L,  liquid. 


mospheric  pressure  along  this  join.  It  has 
been  constructed  using  Hytonen's  unpub- 
lished data  on  eleven  compositions  and 
new  data  on  four  more  aluminous 
compositions.  Hytonen's  X-ray  deter- 
minative work  on  diopsidic  pyroxenes  has 
been  extended  to  more  aluminous 
pyroxenes  and  melilite  solid  solutions  in 
polyphase  assemblages.  The  Di-CTs  join 
intersects  at  a  high  angle  the  "grossu- 
larite"-"pyrope"  join  studied  under  sim- 
ilar conditions  by  Chinner  and  Schairer 
(Year  Book  59).  Thus  it  continues  the 
assault  on  uni variant  lines,  invariant 
points,  and  solidus  volumes  in  the  silica- 
poor  part  of  the  CaO-MgO-Al203-Si02 
quaternary  system.  Quenching  experi- 
ments on  the  Di-CTs  compositions  at 
temperatures  between  the  liquidus  and 
the  solidus  have  given  new  data  on  the 
temperatures  of  three  quaternary  invari- 
ant points  and  on  the  positions  of  critical 
planes  separating  the  seven  solidus  vol- 
umes encountered. 

The  maximum  degree  of  stable  pyrox- 
ene solid  solution  is  estimated  indirectly 
to  be  greater  than  40  mole  per  cent  A1A1 


for  (Ca,Mg)Si.  Sakata  (1957)  observed  a 
continuous  shift  in  pyroxene  lattice 
parameters  in  day-long  runs  at  1200°C  on 
compositions  between  diopside  and 
Di6oCTs4o.  We  find  that  the  same 
extreme  pyroxene  solid  solution  phases 
form  and  persist  indefinitely  at  somewhat 
higher  temperatures  in  the  presence  of  a 
small  amount  of  liquid.  The  pyroxene 
crystallizing  in  Di6oCTs4o  at  1  atmosphere 
must  lie  off  the  Di-CTs  join,  probably  on 
the  diopside-MgAl2Si06  (Mg  Tscher- 
mak's molecule)-CaAl2Si06  plane,  because 
it  coexists  with  two  relatively  magnesium- 
poor  phases,  anorthite  and  akermanite. 
Lattice  parameters  determined  by  Sakata 
(1957)  closely  fit  the  lattice  parameter 
curves  (fig.  4,  p.  63)  at  Di6oCTs40  for 
pyroxenes  crystallized  directly  on  the 
Di-CTs  join  at  20  kb.  Present  evidence 
indicates,  accordingly,  that  the  Di6oCTs40 
pyroxene  crystallized  at  atmospheric 
pressure  has  approximately  the  same 
degree  of  A1A1  for  MgSi  substitution  as  if 
it  lay  on  the  Di-CTs  join  at  that  bulk 
composition. 

In  molar  coordinates  the  Di-CTs  join 


58 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


is  parallel  to  the  akermanite-gehlenite 
series  of  melilites  within  the  tetrahedron 
defined  by  the  four  oxides,  both  series 
involving  the  exchange  of  A1A1  (gehlenite 
and  CaAl2Si06)  for  MgSi  (akermanite  and 
diopside) .  Melilites  are  encountered  in  six 
of  the  seven  solidus  volumes  cut  by  the 
Di-CTs  join,  and  they  range  in  compo- 
sition from  pure  gehlenite  to  pure  or 
nearly  pure  akermanite.  Ervin  and  Os- 
born  (1949)  have  determined  the  d  values 
of  several  X-ray  reflections  as  a  function 
of  composition  in  this  series.  Only  the 
strongest  peak,  (211),  may  be  used  for 
X-ray  determinative  work  in  polyphase 
assemblages.  Although  the  change  in 
d(2ii)  from  akermanite  (2.871  A)  to 
gehlenite  (2.846  A)  is  linear,  it  is  very 
small.  Thus,  an  estimated  possible  meas- 
uring error  of  ±0.02°  in  20  corresponds  to 
about  ±7  weight  per  cent  akermanite. 
Nine  melilite  compositions  determined  by 
this  method  are  shown  in  figure  2. 

The  melilite  composition  changes  as  a 
function  of  bulk  composition  in  the  three- 
phase  assemblages  Melss  +  An  +  Sp, 
Melss  +  An  -f-  Fo,  and  Melss  +  Diss  + 
An.  It  is  invariant  in  the  four-phase 
assemblages  Geh  +  Sp  +  An  +  "alu- 


mina," Ak69Geh3i  +  An  +  Sp  +  Fo,  and 
Ak89Gehu  +  An  +  Diss  +  Fo.  The 
melilite  composition  can  be  calculated  as 
a  function  of  bulk  composition  in  the 
three-phase  regions  where  melilite  co- 
exists with  phases  of  fixed  compositions, 
that  is  in  Melss  +  An  +  Fe  and  in 
Melss  +  An  +  Sp.  It  cannot  be  uniquely 
calculated  in  three-phase  assemblages 
containing  Mel8S  +  An  +  Diss.  To  calcu- 
late the  melilite  composition  in  four-phase 
solidus  assemblages,  the  critical  planes 
bounding  the  tetrahedron  must  be  pre- 
cisely located;  only  for  Geh  +  Sp  +  An 
+  "alumina"  was  it  possible  to  do  this. 
Where  these  compositions  could  be 
calculated  as  a  function  of  bulk  compo- 
sition they  are  shown  on  figure  2  as  solid 
lines.  In  all  other  assemblages  the  melilite 
compositions  have  been  estimated  from 
the  approximate  position  of  critical 
planes,  and  are  dotted. 

At  least  three  invariant  points  have 
been  encountered  in  the  phase-equi- 
librium studies  of  the  Di-CTs  series 
compositions.  All  of  them  lie  outside 
their  respective  four-phase  volumes  and 
are  reaction  points.  The  Geh  +  "alumi- 
na" +  Sp  +  An  +  L  reaction  point  lies 


-2.8700< 


H2.860-T 

e 

>*- 

o 
-2.850     = 


Diopside  '0 

Co  Mg  Si206 


20  30  40  50  60 

Wt.  %  Co  AI2Si06 


70 


80 


90    Co  Tschermok's 
Molecule 
Co  AI2Si06 


Fig.  2.  Plot  of  melilite  compositions  in  solidus  assemblages  versus  bulk  composition  along  join 
diopside-Ca  Tschermak's  molecule  (CaAl2Si06).  Solid  lines,  calculated  theoretical  melilite  compo- 
sitions; dashed  lines,  estimated  theoretical  melilite  compositions;  dots,  compositions  actually  deter- 
mined by  measurement  of  d(2in  of  the  melilites;  bars,  estimated  possible  error  in  determinations 
(±0.02°  20  =  ±7  weight  per  cent  akermanite).  Abbreviations  as  in  figure  1. 


GEOPHYSICAL   LABORATORY 


59 


within  the  Mel68  +  An  +  Sp  volume  at  a 
temperature  of  1360°  =fc  5°C.  This  is  only 
in  fair  agreement  with  De  Vries  and 
Osborn  (1957),  who  measured  a  value  of 
1350°C  for  this  point.  " Alumina"  refers 
to  corundum  and/or  "(3  alumina."  These 
phases  often  occur  together,  although 
"(3  alumina"  is  predominant  near  liquidus 
temperatures,  and  only  corundum  peaks 
are  observed  in  X-ray  patterns  of  solidus 
assemblages. 

The  Sp  +  An  +  Fo  +  Ak69Geh3i  +  L 
invariant  point  lies  within  the  Melss  + 
An  +  Fo  volume,  and  its  temperature  is 
1225°  =fc  5°C.  The  Diss  +  An  +  Fo  + 
AkggGehn  -f-  L  invariant  point  also  lies 
within  the  same  volume.  Its  temperature 
is  not  known  precisely.  It  is  drawn  in 
figure  1  at  1225°C,  which  probably  repre- 
sents a  maximum  value.  Since  these  two 
invariant  points  have  closely  similar 
temperatures  and  compositions,  it  is 
difficult  to  decipher  their  mutual  relation- 
ship. If  they  have  the  same  temperature 
and  different  compositions,  it  is  likely 
that  some  compositions  along  the  Di-CTs 
join  will  pass  through  neither  point  and 
will  crystallize  directly  to  Melss  +  An  + 
Fo  without  forming  any  pyroxene  or 
spinel.  This  possibility  is  depicted  on 
figure  1.  Another,  equally  likely,  possi- 
bility is  that  the  reaction  point  at  which 
pyroxene  is  consumed  occurs  at  a  slightly 
lower  temperature  than  the  reaction 
point  at  which  spinel  is  consumed.  If  this 
is  so,  some  compositions  along  the  join 
may  pass  through  both  points  as  they 
crystallize,  before  winding  up  as  a  mixture 
of  melilite,  anorthite,  and  forsterite. 
Since  compositions  along  the  "grossu- 
larite"-"pyrope"  join  (Chinner  and 
Schairer,  Year  Book  59)  appear  to  raise 
but  not  answer  the  same  questions,  the 
answers  can  be  supplied  only  by  the  study 
of  compositions  lying  off  these  joins  in  the 
Ak-Ak70Geh3o-An-Fo  volume. 

Chinner  and  Schairer  observed  that 
several  compositions  on  the  "grossular- 
ite"-"pyrope"  join  crystallized  an  alumi- 
nous pyroxene  that  reacted  with  liquid  at 
lower  temperatures  to  produce  a  melilite- 


anorthite-forsterite  assemblage.  They 
suggested  that  extensive  fractionation  of 
diopside  crystallized  from  a  basaltic  melt 
as  the  result  of  limestone  syntexis  would 
enrich  the  melt  in  A1203.  Aluminous 
pyroxene  could  then  store  up  CaO  and 
A1203,  which  would  contribute  to  the 
formation  of  melilite  as  pyroxene  redis- 
solved  in  the  magma  during  the  final 
stages  of  crystallization.  This  mechanism 
is  in  complete  qualitative  agreement  with 
the  classic  contamination  sequence  de- 
scribed by  Tilley  and  Harwood  (1931)  at 
Scawt  Hill.  It  also  receives  excellent 
qualitative  confirmation  by  the  relations 
observed  on  the  Di-CTs  join.  The  analo- 
gous reaction  point  in  this  system  is 
DiS8  +  AksgGehn  +  An  +  Fo  +  L,  at 
which  melts  do  indeed  consume  pyroxene 
and  form  melilite  and  other  phases.  This 
is  presumably  the  same  reaction  point 
encountered  by  Chinner  and  Schairer. 
Their  mechanism,  however,  has  at  least 
one  serious  limitation  in  the  application 
to  the  melilite  rocks  of  Scawt  Hill.  The 
pyroxene  at  this  reaction  point  in  the 
synthetic  system  has  been  shown  to 
contain  about  40  mole  per  cent  A1A1  for 
(Ca,Mg)Si.  This  is  more  than  twice  as 
much  AI2O3  as  is  found  in  Scawt  Hill 
aluminous  pyroxenes  (Tilley  and  Har- 
wood, 1931).  Thus  the  analogy  between 
the  synthetic  and  the  natural  pyroxene 
reaction  point  is  less  direct  than  Chinner 
and  Schairer  inferred. 


Phase  Relations  in  the  System 

CaMgSi20&-CaA  hSiOG-Si02 

at  Low  and  High  Pressure 

Sydney  P.  Clark,  Jr.,  J.  F.  Schairer,  and 
John  de  Neufville 

There  is  substantial  indication  that 
basaltic  magmas  are  generated  in  the 
mantle,  perhaps  at  considerable  depths. 
Some  of  this  evidence  is  seismic,  some 
geothermal,  some  geologic.  With  the 
exception  of  the  seismic  activity  associ- 
ated with  Hawaiian  eruptions,  it  is 
indirect   and   perhaps   capable   of   other 


60  CARNEGIE     INSTITUTION     OF      WASHINGTON 

interpretations.  But  enough  evidence  corundum,  forsterite,  the  melilites  aker- 
points  in  the  same  direction  to  make  a  manite  and  gehlenite,  wollastonite,  diop- 
study  of  the  effect  of  pressure  on  melting  side,  enstatite  and  its  polymorphs,  the 
relations  in  systems  of  petrological  inter-  aluminosilicates  andalusite,  kyanite,  sli- 
est worth  while.  Furthermore,  an  upper  limanite,  and  mullite,  the  garnets  pyrope 
limit  to  the  temperature  in  the  mantle  is  and  grossularite,  cordierite,  anorthite, 
set  by  the  liquidus  of  whatever  material  and  the  polymorphs  of  silica.  During  the 
is  down  there,  for  superheated  liquid  must  report  year  a  new  phase,  with  the 
either  move  relatively  rapidly  toward  the  composition  of  lime  Tschermak's  mole- 
surface  or  lose  its  superheat  by  reaction  cule  (CaAl2Si06),  was  synthesized  for  the 
with  surrounding  solid  material.  Effects  first  time  at  a  pressure  of  20  kb. 
of  pressure  on  liquidus  relations  must  be  Because  of  the  importance  of  this 
studied  in  systems  of  moderate  complex-  quaternary  system,  much  previous  work 
ity  before  inferences  about  melting  in  the  has  been  done  to  elucidate  phase  relations 
mantle  can  be  drawn  with  any  confidence,  in  it.  Most  were  investigations  of  lines 
It  has  been  found  that  a  pressure  of  20  kb  and  planes  joining  two  or  three  of  the 
produces  large  effects  on  the  liquidus  that  phases  listed  above.  In  this  way  the 
could  not  have  been  predicted  from  data  tetrahedron  is  crossed  in  many  directions, 
obtained  at  atmospheric  pressure  alone,  and,  given  enough  such  studies,  it  should 

The    quaternary    system    CaO-MgO-  be  possible  to  deduce  with  high  precision 

Al203-Si02,     which     contains     the    join  the  quaternary  equilibrium  relations  at 

CaMgSi206    (diopside)-CaAl2Si06    (lime  atmospheric  pressure. 

Tschermak's  molecule)-Si02,  is  of  great  For  initial  study  in  this  system  at  high 

importance,    because    it    is    sufficiently  pressure  we  selected  the  join  diopside- 

complicated    to    represent    qualitatively  anorthite.  As  Bowen  recognized,  this  is  a 

the  phase  relations  of  the  basic  igneous  simple,    pseudobinary   representation   of 

rocks,    particularly    basalts,    and    rocks  many  basalts  and  diabases.  The  system 

arising  from  their  metamorphism.   The  was  first  shown  not  to  be  truly  binary  by 

main  constituents  commonly  present  in  Osborn    (1942).    We   expected   that   the 

such  rocks  and  absent  from  this  quater-  nonbinary  behavior  would  be  accentuated 

nary  system  are  iron  in  both  its  valence  by  pressure,  and  this  has  proved  to  be  so. 

states,  soda,  water,  and  to  a  lesser  extent  Part  of  the  join  is  quaternary  because  of 

K20,  Ti02,  and  MnO.  In  this  simplified  the  incongruent  melting  of  anorthite  at 

system   it  is   impossible  to   study  such  high  pressures.  We  have,  however,  only 

important  relationships  as  the  effects  of  studied  compositions  lying  in  the  plane 

the  fugacities  of  water  and  oxygen  or  diopside-lime   Tschermak's    molecule-sil- 

changes    in    composition    of    feldspars,  ica,  which    contains    the   join   diopside- 

Experimental  difficulties   occasioned  by  anorthite. 

the  various  possible  oxidation  states  of  For  purposes  of  orientation  it  is  helpful 

transition  elements  and  the  volatility  of  to  consider  the  composition  plane  ensta- 

alkalies  and  water  at  high  temperatures,  tite-wollastonite-corundum    (fig.    3).    All 

however,  are  avoided.  Despite  the  sim-  phases  shown  in  the  figure  lie  precisely  in 

plifi cations,    a    number   of   reactions    of  this     plane;     none     are     projected.     Of 

petrological  importance  take  place  in  this  particular  interest   are   the   intersecting 

system;  because  of  this  and  its  relative  joins  diopside-lime  Tschermak's  molecule 

chemical     tractability     the     system     is  and    grossularite-pyrope.    The   pyroxene 

well  suited  for  a  beginning  to  the  study  join  is  characterized  by  complete  solid 

of  complex  chemical  equilibria   at  high  solution  at  20  kb;  the  garnet  join,  by 

pressures.  complete    solid    solution    above    30    kb. 

Among  the  important  phases  lying  in  Magnesian  Tschermak's  molecule,  shown 

this   system  are  the   oxides  spinel  and  on  the  diagram,  has  never  been  synthe- 


GEOPHYSICAL   LABORATORY 


61 


CORUNDUM 


Mg   TSCHERMAK'S, 
MOLECULE 


Co  TSCHERMAK'S 
MOLECULE 


PYROPE 


ENSTATITE 


DIOPSIDE 


GROSSULARITE 


WOLLASTONITE 


Fig.  3.     Solid   phases   in   the   plane   CaSi03- 
MgSi03-Al203. 


sized,  although  Boyd  and  England  (Year 
Book  59,  p.  49)  have  made  enstatites  with 
at  least  15  mole  per  cent  A1203  in  solid 
solution. 

The  plane  in  figure  3  contains  a  number 
of  phases  with  a  striking  variety  of  crystal 
structures,  all  characterized  by  a  metal- 
to-oxygen  ratio  of  2 :3.  Their  densities  are 
closely  correlated  with  the  structure. 
Densities  of  diopsidic  pyroxenes  men- 
tioned in  the  ensuing  paragraphs  have 
been  calculated  from  X-ray  data  dis- 
cussed below.  Cell  edges  of  the  grossular- 
ite-pyrope  series  of  garnets  were  given 
by  Chinner,  Boyd,  and  England  (Year 
Book  59,  p.  77),  and  the  densities  of  other 
phases  were  taken  from  the  literature, 
using  X-ray  data  whenever  possible. 

Wollastonite  and  the  pyroxenes  are 
chain-type  silicates.  The  lightest,  wollas- 
tonite, has  a  density  slightly  greater  than 
2.9  g/cm3.  The  density  of  enstatite  is 
3.212  g/cm3,  that  of  diopside  is  3.281 
g/cm3,  and  that  of  lime  Tschermak's 
molecule  is  3.437  g/cm3.  The  garnet 
structure  is  composed  of  isolated  silica 
tetrahedra,  connected  by  irregularly  co- 
ordinated cations.  The  density  of  pyrope 
is  3.566  g/cm3;  that  of  grossularite  is 
3.603  g/cm3.  The  densest  structure  is  that 
of  the  closely  packed  oxide  corundum, 
4.02  g/cm3. 

It  has  been  suggested  that  pyroxenes 
might  undergo  transitions  to  the  corun- 


dum structure  at  very  high  pressures,  and 
this  inversion  has  been  reported  in 
MgGe03  (Ringwood  and  Seabrook,  1962). 
It  does  not  seem  to  have  been  remarked 
that  garnets  also  have  the  metal-to- 
oxygen  ratio  appropriate  to  undergo  a 
transition  to  a  corundum  structure.  Such 
an  inversion  may  take  place  deep  in  the 
transition  zone  in  the  mantle. 

Two  other  comparisons  of  density  are 
interesting  to  make.  The  first  is  between 
the  density  of  crystalline  lime  Tscher- 
mak's molecule  (CaAl2Si06)  and  the 
densities  of  its  low-pressure  breakdown 
products  gehlenite  (p  =  3.038  g/cm3), 
anorthite  (p  =  2.765  g/cm3),  and  either 
"|8  alumina"  or  corundum.  The  density  of 
"/3  alumina"  is  not  well  known,  but 
neither  alumina  phase  is  present  in  large 
amounts.  The  mean  density  of  the  break- 
down products  cannot  be  far  from  2.9 
g/cm3.  Lime  Tschermak's  molecule  is  18 
per  cent  denser  than  this.  The  density 
change  between  the  pyroxene  and  garnet 
at  the  intersection  of  the  two  joins  shown 
in  figure  3  is  6  per  cent  (3.368  versus  3.592 
g/cm3). 

X-ray  data  for  diopsidic  pyroxenes.  In 
order  to  set  up  suitable  determinative 
procedures  for  complex  solid  solutions 
such  as  those  shown  by  diopsidic  py- 
roxenes, careful  crystallographic  work 
must  be  done.  The  fine-grained  nature  of 
synthetic  crystals  precludes  single-crystal 
studies,  and  care  must  be  taken  that 
determinative  peaks  on  powder  patterns 
can  be  unambiguously  indexed.  Other- 
wise errors  from  effects  of  preferred 
orientation  may  influence  measurements 
of  unresolved  multiple  reflections. 

In  crystals  of  low  symmetry  it  is  all  but 
impossible  without  the  aid  of  a  high-speed 
computer  to  be  sure  that  all  indexing 
allowed  by  the  space  group  has  been 
compared  with  the  observed  reflections. 
Only  by  being  certain  that  all  possibilities 
have  been  considered  can  one  be  sure  that 
a  reflection  is  not  multiple.  Such  pre- 
cautions have  not  always  been  taken  in 
the  past. 

All  data  processing  was  carried  out  on 


62 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


an  IBM  7090  digital  computer  using 
programs  written  by  Charles  W.  Burn- 
ham.  His  program  for  calculating  unit- 
cell  parameters  by  least  squares  is 
described  elsewhere  in  this  report.  His 
program  for  calculating  d  values  permit- 
ted by  the  space  group  from  the  param- 
eters of  the  unit  cell  was  used  in  indexing 
powder  patterns. 


TABLE  1.     Miller  Indices  and  d  Values  of 

Reflections  Used  in  Calculating  Unit-Cell 

Parameters  of  Diopsidic  Pyroxenes 


d  Value 

hkl 

Diopside 

Lime  Tschermak's 
Molecule 

221 

2.9897 

2.9412 

310 

2.9492 

2.8997 

311 

2.8924 

2.8613 

311 

2.3009 

2.2739 

330 

2.1546 

2.1062 

331 

2.1322 

2.0902 

421 

2.1067 

2.0752 

132 

1.9679 

1 . 9468 

150 

1.7535 

1.7026 

The  starting  point  in  our  investigation 
was  a  carefully  indexed  powder  pattern 
of  pure  diopside.  This  pattern  was  com- 
pared with  patterns  obtained  on  material 
prepared  by  completely  crystallizing 
glasses  on  the  join  diopside-lime  Tscher- 
mak's molecule  at  20  kb.  The  positions  of 
the  peaks  were  found  to  shift  smoothly  as 
a  function  of  composition  from  one  end 
of  the  join  to  the  other.  No  peaks 
appeared  that  could  not  be  traced  into 
their  counterparts  in  the  diopside  pattern ; 


this  plus  optical  examination  provides 
evidence  that  only  one  phase,  a  pyroxene, 
was  present  in  these  runs.  As  a  check,  the 
complete  pattern  for  the  composition  50 
per  cent  diopside,  50  per  cent  lime 
Tschermak's  molecule,  was  calculated. 
No  unexpected  interferences  between 
peaks  were  found.  The  reflections  used 
and  their  d  values  for  diopside  and  lime 
Tschermak's  molecule  are  given  in  table 
1.  These  reflections  were  chosen  because 
they  can  be  indexed  unambiguously  and 
are  sharp  and  strong — an  important 
feature  if  they  are  to  be  used  for  deter- 
minative purposes  in  mixtures  of  phases 
that  do  not  contain  very  much  pyroxene. 

The  first  three  reflections  listed  in 
table  1  fall  at  20  angles  less  than  31°  for 
copper  radiation.  Hence  the  d  values 
cannot  be  determined  with  high  accuracy. 
The  parameters  of  the  unit  cells  were 
calculated  by  least-squares  adjustment 
both  with  and  without  these  peaks.  The 
resulting  parameters  do  not  differ  sig- 
nificantly, but  the  standard  errors  are 
usually  smaller  if  the  low-angle  peaks  are 
rejected. 

The  unit-cell  parameters  of  lime 
Tschermak's  molecule  and  diopside  are 
given  in  table  2,  along  with  parameters 
for  diopside  from  other  observers.  The 
agreement  is  good.  The  change  of 
parameters  along  the  joins  diopside-lime 
Tschermak's  molecule  and  diopside- 
enstatite  is  shown  in  figures  4  and  5.  The 
data  in  figure  5  were  obtained  by  applying 
the  procedures  described  above  to  a  series 
of  glasses  that  had  previously  been 
crystallized  at  1  atmosphere.  Compo- 
sitions containing  more  than  40  per  cent 


TABLE  2.     Unit-Cell  Parameters  of  Lime  Tschermak's  Molecule  and  Diopside 


Lime  Tschermak's 
Molecule 
(present) 


Diopside 
(present) 


Diopside 
(Sakata,  1957) 


Diopside 
(H.  H.  Hess, 
unpublished) 


a,  A 

9.615  ±0.003 

9.745  ±0.001 

9.743 

9.741 

b,  A 

8.661  ±0.002 

8.925  ±0.001 

8.923 

8.924 

c,  A 

5.272  ±0.003 

5.248  ±0.001 

5.251 

5.247 

0,  deg 

73.88    ±0.03 

74.13    ±0.01 

74.07 

74.15 

V,  A3 

421.79    ±0.28 

439.08    ±0.07 

438.98 

438.77 

GEOPHYSICAL   LABORATORY 


63 


o< 


9.60 


8.60 


to 

<D 

■o 


74.25  - 


74.00 


73.75 


ro 


440 
435  [ 
430 
425 


420 


-+- 


■+ 


_r- 


25  50  75 

Wt  %  CaAI2Si06 


100 


Fig.  4.     Unit-cell  parameters  along  the  join  diopside-lime  Tschermak's  molecule. 


64 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


9.75 
o      9.70 


L 


+      -+-- 


-T 


o  <t 


9.00 


1.901- 


5.30 


°<      5.25 


o 


5.20 


74.25 


74.00 


en  73.75 

CD 

|  73.50 

^  73.25 


73.00 

72.75- 

72.50- 


440J- 

ro 

o<        435 
> 


430 


M" 


.+ 


~T 


20 


80 


40  60 

Wt  %   Enstatite 
Fig.  5.     Unit-cell  parameters  along  the  join  diopside-enstatite. 


100 


GEOPHYSICAL  LABORATORY  65 

enstatite  do  not  crystallize  to  a  single  larger  than  those  found  by  other  workers 

phase  under  these  conditions.  to  cast  serious  doubt  on  the  determinative 

Along  the  join  lime  Tschermak's  mole-  curves  given  by  Hytonen  and  Schairer. 
cule-diopside,  the  substitution  is  Al-Al  for  Melting  relations  in  the  system  diopside- 
Mg-Si.  One  would  expect  that  replacing  anorthite-silica.  Liquidus  data  for  this 
an  Mg  atom  with  a  relatively  small  Al  system  at  atmospheric  pressure  are  shown 
would  cause  a  and  b  to  decrease.  Likewise,  in  figure  6.  Dots  indicate  the  compositions 
replacing  an  Si  atom  with  a  relatively  studied  by  the  quenching  method.  Except 
large  Al  atom  in  the  silica  chains  would  for  compositions  near  the  diopside-silica 
cause  a  slight  increase  in  c.  These  are  the  join,  the  figure  has  approximately  the 
observed  effects.  /3  changes  little  in  this  appearance  of  the  simplest  type  of 
series.  The  internal  consistency  of  the  ternary  diagram,  that  is  one  in  which 
data  for  this  parameter,  i.e.,  the  lack  of  only  three  pure  solid  phases  exist  and 
scatter  of  the  points  about  the  curve  in  liquid  miscibility  is  complete.  That  this 
figure  4,  is  remarkable  considering  the  is  only  approximately  true  was  first 
scale  of  the  diagram.  The  parameter  that  shown  by  Osborn  (1942),  who  demon- 
changes  most  is  b,  and  hence  reflections  strated  that  the  join  diopside-anorthite  is 
with  large  k  are  most  satisfactory  for  not  binary  owing  presumably  to  solid 
determinative  purposes  along  this  join.  solution  of  alumina  in  the  pyroxene.  This 

Volumes  in  this  solid  solution  series  result  has  been  confirmed  by  Hytonen 

depart  systematically  from  a  straight  line  and  Schairer  ( Year  Book  60) .  To  obtain 

connecting  the  end  members  in  a  way  more  precise  information  on  the  compo- 

that  implies  that  they  are  nonlinearly  sition   of   the   pyroxene,    careful   X-ray 

related   to   composition.    The   departure  work  was  done  on  a  composition  lying  on 

from  linearity,  although  apparently  real,  the    diopside-anorthite    join    that    was 

is  not  large.  A  straight  line  would  fit  the  equilibrated  with  liquid  at  1260°C  and  on 

data  within  0.5  per  cent.  a  composition  lying  in  the  ternary  plane 

Edges  of  the  unit   cell   change  little  that    was    equilibrated    with    liquid    at 

along  the  diopside-enstatite  join;  the  most  1220°C.   In  both,  the  departure  of  the 

conspicuous  feature  of  figure  5  is  the  large  unit-cell  parameters  from  those  of  pure 

decrease  in  /3  with  increasing  content  of  diopside  was  small ;  it  was  greater  for  the 

enstatite.  The  volume  is  essentially  linear  composition   crystallized    at    the   higher 

with  composition  over  the  limited  range  temperature.  Hytonen  and  Schairer  (Year 

of  the  data.  Book  60,  p.  137)  indicate  that  at  1135°C 

There  is  a  systematic  difference  be-  in  this  system  (a  temperature  well  below 

tween  our  results  and  those  of  Hytonen  the  solidus)  the  pyroxene  contains  about 

and  Schairer  (Year  Book  60,  p.  136).  They  3  per  cent  lime  Tschermak's  molecule, 

based    a    determinative    procedure    for  They   considered  it  probable   that  this 

diopsidic  pyroxenes  on  the  positions  of  amount  of  solid  solution  was  metastable, 

the  (150)  and  (510)  reflections.  We  did  and  our  results  suggest  the  same.  Because 

not  read  (510)  because  of  possible  inter-  of  the  small  shift  in  properties  relative  to 

ferences  with  (422)  and  (332),  but  we  can  experimental  error,  it  is  not  possible  to 

calculate  its  position  from  our  data.  For  determine  the  direction  in  which  these 

both  reflections  our  20  angles  are  about  pyroxenes  differ  from  pure  diopside. 

0.1°    larger    than    those    reported    by  In  all  the  sixteen  compositions  within 

Hytonen  and  Schairer.   By  assuming  a  the  triangle  diopside-anorthite-silica,  the 

value  for  0,  it  is  possible  to  calculate  a  third  solid  phase  first  appeared  on  cooling 

and  b  for  diopside  from  their  data.  Using  at    temperatures    between     1218°    and 

the  extreme  values  of  /3  in  table  2,  it  is  1225°C.    This   implies   that   the   system 

found  that  a   =   9.751  to  9.755  A  and  diopside-anorthite-silica    is    very    nearly 

b  =  8.937  A.  These  values  are  sufficiently  ternary,   and  that   the  stable   pyroxene 


66 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Si02 


TWO_         '» 
LIQUIDS 


10              20              30             40             50              60             70  80             90 

DIOPSIDE  Ca-TSCHERMAK'S 

CaMgSi206  MOLECULE 

Weight  per  cent  CaAI2Sl  °s 

Fig.  6.     Equilibrium   diagram   for  the  system  diopside-lime  Tschermak's  molecule-silica  at   1 
atmosphere. 


must  lie  close  to  the  plane  of  figure  6  at 
1222°C.  The  X-ray  evidence  implies  that 
it  is  essentially  pure  diopside.  The 
piercing  point,  or  ternary  eutectic,  must 
be  close  to  or  at  the  thermal  maximum  on 
the  quaternary  univariant  line  connecting 
two  quaternary  eutectics.  At  one,  wollas- 
tonite,  diopsidic  pyroxene,  anorthite,  and 
a  silica  phase  coexist  with  liquid,  and  at 
the  other  enstatitic  pyroxene,  diopsidic 
pyroxene,  anorthite,  and  a  silica  phase 
coexist  with  liquid.  Determination  of  the 
composition  of  the  latter  eutectic  is  of 
great  geologic  significance,  since  it  repre- 
sents the  goal  of  crystallization  of  a 
simplified  silica-saturated  basalt  at  low 
pressures. 

High-pressure  studies  of  the  liquidus  in 
this  system  have  been  carried  out  in  a 
"single-stage"  type  of  apparatus  similar 
to  that  described  by  Boyd  and  England 
in  Year  Book  60.  Results  at  20  kb  are 
shown  in  figure  7 ;  a  large  number  of  runs 
have  also  been  made  at  30  kb,  but  this 


work  is  not  yet  ready  for  presentation. 
In  all  the  work  described  the  load  pres- 
sure has  been  decreased  by  3  per  cent  to 
allow  for  the  effect  of  friction. 

The  accuracy  with  which  temperature 
can  be  measured  is  much  lower  at  high 
pressures  than  at  atmospheric  pressure. 
At  high  pressures  the  uncertainty  in 
temperature  ranges  from  d=10°C  in 
favorable  cases  to  d=20°C  or  so.  These 
estimates  are  based  on  the  internal 
consistency  and  the  reproducibility  of 
some  of  our  results.  There  is  in  addition 
a  correction  for  the  systematic  effect  of 
pressure  on  the  emf  of  a  thermocouple; 
this  has  been  omitted  because  the  elusive 
problem  of  quantitative  determination  of 
the  correction  remains  to  be  successfully 
attacked.  In  contrast,  at  atmospheric 
pressure  an  accuracy  of  d=2°C  can  be 
achieved  with  care. 

The  eutectic  temperature  in  the  binary 
system  diopside-silica  is  raised  by  slightly 
more  than  200°C  by  a  pressure  of  20  kb. 


GEOPHYSICAL   LABORATORY 


67 


This  is  essentially  the  same  as  the  change 
in  melting  point  of  diopside  itself.  The 
composition  of  the  eutectic  is  not  measur- 
ably affected  by  pressure.  In  this  system, 
as  in  all  the  work  at  20  kb,  quartz  is  the 
silica  phase  stable  on  the  liquidus.  The 
effect  of  pressure  on  the  two-liquid  region 
in  this  system  has  not  been  investigated. 

The  system  anorthite-silica  is  not 
binary  at  20  kb  because  of  the  incongru- 
ent  melting  of  anorthite,  probably  to 
corundum  +  liquid  (Boyd  and  England, 
Year  Book  60,  p.  119).  Between  the  fields 
of  corundum  and  quartz  on  the  liquidus 
there  is  a  field  of  sillimanite.  The  temper- 
ature of  lowest  point  on  the  liquidus, 
between  the  quartz  and  sillimanite  fields, 
is  1540°C,  and  the  composition  is  48 
weight  per  cent  Si02.  At  atmospheric 
pressure  the  binary  eutectic  lies  at  1368°C 
and*59  weight  per  cent  Si02.  (Both  silica 
contents  are  determined  relative  to  lime 
Tschermak's  molecule.) 

Changes  produced  by  pressure  in  the 
system    diopside-anorthite    are    greater 


than  in  the  other  limiting  systems.  Not 
only  does  anorthite  melt  incongruently  at 
high  pressures  but  also  the  amount  of 
alumina  in  the  pyroxene  increases  dra- 
matically. At  compositions  near  anorthite, 
corundum  and  "/?  alumina"  both  appear 
at  high  temperatures,  with  and  without 
other  crystalline  phases.  One  of  these 
alumina  phases  must  be  metastable  on 
the  liquidus,  but  it  is  not  clear  which. 
There  is  some  evidence  that,  although 
corundum  is  stable  at  the  anorthite 
composition,  "(3  alumina"  is  the  stable 
liquidus  phase  at  neighboring  magnesian 
compositions.  It  will  be  difficult  to  work 
out  the  correct  relationship  between  these 
phases  because  of  the  stubbornness  with 
which  they  both  persist  metastably. 

The  nature  of  the  minimum  on  the 
liquidus  in  this  system  has  not  yet  been 
determined.  It  may  be  a  cusp,  resembling 
a  eutectic,  or  it  may  be  a  smooth  trough, 
depending  on  whether  the  minimum  lies 
within  the  pyroxene  field  or  at  its 
boundary.  Figure  7  is  drawn  as  if  this 


Si02 


20  Kilobars 


ANORTHITE 
20CaAI2Si20( 


I585±l5 


DIOPSIDE 
CaMgSi20g 


Ca-TSCHERMAK'S 
MOLECULE 
CaAI2Si06 


Weight  per  cent 
Fig.  7.     Equilibrium  diagram  for  the  system  diopside-lime  Tschermak's  molecule-silica  at  20  kb. 


68  CARNEGIE     INSTITUTION      OF      WASHINGTON 

minimum  were  a  cusp  at  the  boundary  of  kb  is  not  even  qualitatively  similar  to  the 

the  field,  but  future  work  may  indicate  system    at    atmospheric    pressure,    and 

the  need  for  modification  of  this  feature  quantitative   differences   in  melting  be- 

of   the   diagram.    The  temperature   and  havior   occur   at   all   compositions.    The 

composition  of  this  point  are  1480°C  and  most   striking   new   features   caused   by 

71  weight  per  cent  anorthite.  At  atmos-  pressure  are  the  incongruent  melting  of 

pheric  pressure  this  point  lies  at  1274°C  anorthite,  the  appearance  of  sillimanite 

and  43  weight  per  cent  anorthite.  on  the  liquidus,  the  appearance  of  quartz 

The  complex  relations  at  high  pressures  on    the    liquidus    above     1000°C,    and 

found  in  the  systems  diopside-anorthite  extensive  solid  solution  in  the  pyroxene, 

and    anorthite-silica    continue    into    the  None  of  these  effects  occurs  at  atmos- 

triangle  of  figure  7.  The  fields  adjacent  to  pheric  pressure,  and  none  of  them  could 

the  anorthite  composition  have  not  yet  have  been  inferred  without  high-pressure 

been  fully  delineated.  There  must  be  a  experimentation. 

field    of   sillimanite,    one   of   corundum,  There  is  an  interesting  possible  geo- 

probably  one  of  "/3  alumina,"  and,  near  logical     consequence     of    the    shift    in 

the  piercing  point,  one  of  anorthite  itself,  composition  of  the  piercing  point  with 

Although  pyroxene,  anorthite,  and  quartz  pressure.  If  a  small  amount  of  liquid  were 

are  the  solid  phases  present  at  the  piercing  formed  by  fractional  fusion  at  20  kb  in 

point,    the    relationship    there    is    not  this  system,  it  would  have  the  approxi- 

ternary.  There  is  a  melting  interval  of  mate  composition  diopside22-lime  Tscher- 

about  30°C.  This  point  contains  about  10  mak's  molecule 42-quartz36.  If  this  liquid 

weight  per  cent  more  anorthite  than  its  were  then  decompressed  suddenly,  per- 

counterpart  at  atmospheric  pressure,  and  haps  by  rapid  upward  intrusion,  it  would 

its  temperature  is  raised  about  125°C  by  arrive  in  a  superheated  condition  and  the 

20  kb.  This  is  somewhat  less  than  the  composition  of  the  liquid  would  be  well 

increase  in  the  minima  in  the  diopside-  inside  the  anorthite  field  at  low  pressure, 

anorthite  and  anorthite-silica  systems.  The  liquid  would  crystallize  large  quan- 

That  pyroxenes  grown  in  this  system  tities  of  feldspar  before  other  solid  phases 

at  20  kb  do  not  lie  on  the  join  diopside-  appeared,  which  suggests  a  mechanism 

lime  Tschermak's  molecule  is  shown  by  for  the  origin  of  anorthosites.  It  is  to  be 

the  failure  of  compositions,  as  determined  expected    that    in    the    system    albite- 

by  X  rays,  to  bear  the  relations  to  each  diopside-silica    the    piercing    point    will 

other  demanded  by  principles  of  phase  behave  in  a  similar  way  because  of  solid 

equilibria,  and  by  the  fact  that  different  solution  of  jadeite  in  the  pyroxene  and 

parameters  of  the  unit  cell  have  values  the  eventual  disappearance  of  albite  from 

that     would     correspond     to     different  the  liquidus.   An  important  unexplored 

amounts  of  lime  Tschermak's  molecule  in  question  is  the  behavior  of  intermediate 

solid  solution.  Correction  for  enstatite  in  plagioclases ;  it  is  not  yet  known  whether 

solid  solution,   determined  from  /5,   im-  the    mechanism    outlined    can    produce 

proves  the  internal   consistency  of  the  feldspars  of  the  compositions  found  in 

data,  but  the  remaining  discrepancies  are  anorthosites. 
probably  large  enough  to  be  considered 

real.  Presumably  there  is  also  magnesian  The  System  MgSi03-CaMgSi206 

Tschermak's  molecule  (or  corundum)  in  F  R  Boy^  ^  and  j  p  Schairer 
solid  solution  in  the  pyroxene. 

These  results  should  dispel  any  doubts  Mineral   assemblages    containing   two 

that   pressure,   even  in  the   absence   of  pyroxenes     are     of     almost     ubiquitous 

volatile     constituents,     can    profoundly  occurrence  in  mafic  and  ultramafic  igne- 

affect  phase  diagrams.   In  part   of  the  ous  rocks.  The  two  pyroxenes  are  usually 

range  of  compositions,  the  system  at  20  a  calciferous  pyroxene,  augite  or  ferro- 


GEOPHYSICAL   LABORATOEY 


69 


augite,  and  a  lime-poor  hypersthene  or 
pigeonite.  Such  pyroxenes  show  a  wide 
variation  in  Mg/Fe  ratio  together  with  a 
more  limited  variation  in  Ca/(Mg  +  Fe). 
Understanding  of  the  equilibria  between 
such  pyroxene  pairs  is  of  great  petrologic 
interest,  and  the  simplest  system  through 
which  the  problem  can  be  approached  is 
the  join  MgSi03-CaMgSi206. 

Liquidus-solidus  relations  along  this 
join  were  determined  many  years  ago  by 
Bowen  (1914).  Atlas  (1952)  was  the  first 
to  study  the  subsolidus  equilibria,  and  by 
means  of  fluxes  he  located  the  solvus  and 
showed  that  two  pyroxenes  coexist  at  all 
temperatures  below  1350°C.  The  crest  of 
the  solvus  as  determined  by  Atlas  was 
shown  to  be  about  50°  below  the  solidus 
curve ;  within  this  50°  interval  a  complete 
solid  solution  between  MgSi03  and 
CaMgSi206  seemed  to  exist.  Atlas  showed 
that  orthorhombic  MgSi03  was  stable  at 
temperatures  up  to  985°C.  Although 
clinoenstatite  is  commonly  obtained  in 
runs  on  MgSi03  composition  quenched 
from  above  1000°C,  Atlas  argued  that 
protoenstatite  was  the  stable  form  in  this 
range  and  that  clinoenstatite  formed  in 
the  quench.  High-temperature  X-ray 
studies  by  Foster  (1951)  showed  that 
both  orthorhombic  MgSi03  and  clino- 
enstatite could  be  inverted  to  proto- 
enstatite at  temperatures  above  1275°C. 
These  studies  proved  that  protoenstatite 
has  a  stable  field  at  high  temperature. 

Boyd  and  Schairer  (Year  Book  56) 
determined  the  solvus  on  this  join  by 
both  dry  and  hydrothermal  techniques. 
We  found  that  the  solvus  intersected  the 
solidus  over  a  composition  interval  of 
about  15  weight  per  cent,  so  that  a 
complete  solid  solution  between 
CaMgSi206  and  MgSi03  does  not  exist  at 
any  temperature. 

Scatter  of  our  preliminary  results  along 
the  part  of  the  solvus  curve  that  defines 
the  limit  of  solubility  of  CaMgSi206  in 
protoenstatite  led  to  a  further  investiga- 
tion of  this  part  of  the  system.  Results 
obtained  this  year  indicate  that  the 
composition    interval    over    which    the 


solvus  intersects  the  solidus  is  much 
wider  than  was  previously  indicated.  Also 
evidence  was  found  for  an  additional  form 
of  Mg-rich  pyroxene,  stable  above 
1385°C. 

Figure  8  shows  the  liquidus  and 
subsolidus  equilibria  for  the  system 
MgSi03-CaMgSi206.  Quenching  data  for 
an  extensive  series  of  compositions  along 
the  join  locate  the  liquidus  temperatures 
and  the  equilibrium  between  crystals  and 
liquid.  Small  circles  indicate  the  tempera- 
tures as  determined.  Liquidus  tempera- 
tures and  the  temperature  of  appearance 
of  a  Mg-rich  pyroxene  were  determined 
for  some  compositions  by  Bowen  (1914). 
Our  results  are  in  complete  agreement 
with  these  data  within  the  error  of 
measurement.  Bowen's  data  are  given  in 
figure  8  as  triangles. 

Several  of  the  crystal -liquid  fields  on 
the  diopside  side  of  the  equilibrium 
diagram  are  so  small  that  they  cannot  be 
shown  on  the  scale  of  figure  8.  This  part 
of  the  diagram  is  expanded  in  figure  9. 
There  is  a  series  of  solid  solutions  with  a 
temperature  minimum  on  the  melting 
and  freezing  curves.  These  relations  are 
interrupted  by  the  incongruent  melting  of 
pyroxenes  to  forsterite  and  liquid.  The 
composition  of  the  binary  reaction  point 
forsterite  +  diopside  -f-  liquid  was  deter- 
mined as  En23.25  Di76.75  (weight  per  cent). 
The  temperature  is  1389°  ±  2°C.  Bowen 
(1914,  p.  233,  fig.  18)  inferred  these 
relations.  Our  data  also  tie  in  very  closely 
with  those  of  Schairer  and  Yoder  on  the 
system  forsterite-diopside-silica  given 
elsewhere  in  this  report  (pp.  75-82). 

Compositions  on  the  join  MgSi03- 
CaMgSi206  that  are  crystallized  dry  in 
the  two-phase  field  form  cryptoperthitic 
intergrowths  of  protoenstatite  and  diop- 
side. X-ray  methods  are  therefore  neces- 
sary to  fix  the  composition  of  individual 
phases  and  to  locate  the  solvus  bounda- 
ries. Since  protoenstatite  inverts  to 
clinoenstatite  in  the  quench,  the  X-ray 
data  are  given  for  clinoenstatite.  The 
shift  of  several  reflections  with  compo- 
sition   is    sufficiently    large    to    fix    the 


70 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


1600- 
I577±2°J 
1557*2°' 


T 1 1 1 1 1 1 1 1 r 


LIQUID 


1 1 1 1 1 r 


700 


Diss+Fo  +  L 


BINARY  MINIMUM 
I388±2° 


i°  °      A 

±2*D  JO  D  D    / 


Di„+  L 


J L 


J L 


MgSi03  10  20  30  40  50  60 

Weight  per  cent 


J _L 


70 


J L 


1391.5° 


80 


90    CaMgSi206 


o  Quench  data, dry 

a  Quench  data ,  dry ,  Bowen  (1914) 

a  Single  phase  run, dry 

[]  Single  phase  run ,  500  bors  H20 

O  Point  on  solvus  boundary  from  dry  run  or  runs 


0  Point  on  solvus  boundary  determined  by 
homogenizing  pyroxenes, dry 

0  Point  on  solvus  boundary  from  runs  at  500  bars  H20 
0  Single  phase  run, 1000  bars  H20 
i  Two  phase  run .  1000  bars  H20 


Fig.  8.  Liquidus  and  subsolidus  equilibria  in  the  system  MgSi03-CaMgSi206.  Some  liquidus 
points  determined  by  Bowen  (1914)  are  shown  along  with  the  data  obtained  by  the  authors.  The 
value  of  1025°C  given  for  the  inversion  of  rhombic  MgSi03  to  protoenstatite  is  based  largely  on 
extrapolation  of  preliminary  high-pressure  results.  See  figure  9  for  an  expanded  view  of  the  phase 
relations  on  the  diopside  side  of  the  diagram. 


GEOPHYSICAL   LABORATORY 


71 


90- 


LIQUID 


I39!.5C 


90 
80 
70 
60 
1450 
40 
30 
20 
-     10 


I389i2 


+  oo 


BINARY  MINIMUM 


DU+Fo  +  L  DU+L  DU+L 


ss 


'ss 


'ss 


-1400 


90 
80 
70 
60 
350 


50 
MgSi03 


2^6 


70  80  90  CaMgSi?0 

Weight  per  cent 


Fig.  9.     Expanded  view  of  the  phase  relations  in  a  part  of  the  system  MgSi03-CaMgSi206.  See 
figure  8  for  the  entire  diagram. 


composition  of  a  phase  within  ±2  per 
cent.  Figure  10  shows  the  diffractometer 
patterns  of  the  220  peaks  for  a  series  of 
compositions  across  the  solvus.  These 
runs  were  made  dry,  without  flux,  and 
were  held  at  1365°C  for  2  weeks.  In  the 
single-phase  regions,  the  220  reflection 
forms  a  sharp,  single  peak,  but  compo- 
sitions within  the  two-phase  field  show  a 
double  reflection,  indicating  the  presence 
of  two  intimately  intergrown  pyroxenes. 


Figure  11  shows  measurements  of  the 
220  reflection  for  a  series  of  compositions 
from  pure  MgSi03  to  En65Di35.  Silicon 
was  used  as  an  internal  standard  for  these 
measurements.  The  points  for  runs  at 
1365°C  shift  progressively  with  compo- 
sition from  pure  MgSi03  to  a  bulk 
composition  of  En8oDi2o.  For  composi- 
tions richer  in  diopside  than  En8oDi2o,  the 
composition  of  the  enstatite  is  fixed  at 
En78Di22  independent  of  the  bulk  compo- 


72 


CAKNEGIE     INSTITUTION     OF      WASHINGTON 


CLINOPYROXENE    220 
T=  1365° 


Di25En75  Di30En70  Di40En60  0'50En50  Di6c£n40  D'TO^O 


I*  29 


Fig.  10.  Tracings  of  the  220  reflection  from  X-ray  diffractometer  patterns  of  a  series  of  runs 
across  the  solvus  in  the  system  MgSi03-CaMgSi206.  The  220  reflection  is  a  sharp,  single  peak  in  the 
single-phase  regions  of  solid  solution  bordering  the  solvus.  Within  the  solvus  the  220  reflection  splits 
into  a  doublet  indicating  the  presence  of  a  cryptoperthitic  intergrowth  of  a  Ca-rich  diopsidic  pyroxene 
and  a  Mg-rich  clinoenstatite. 


o 
o 

C\J 
<M 


o 

CD 
CM 


28.10 


.04 


.02 


28.00 


.94 


27.92 


i — : — ! — : — i — i — r 


-i — ; — i — r~— t — r~i — i — i — l    I — r 


ill — ! — r~i — r-r 


O  13S5° 
©  1300° 
0  !250° 


eJ?ooiQ 


1365# 


I    l    l    l 


O 0-'^2I 

O 


I      I      I      I      I      I      I     I      I      I      I      I      I      I      I      I      I      I      I      I      I      I      I      I      I      1      I      I      I     I      I      !      I      I      I      I      I      I      I      I 


10 


MgSi03 


20  25  30 

Weight  per  cent 


35  40 

CoMgSi206 >■ 


Fig.  11.  Shift  of  the  220  reflection  in  clinoenstatites  in  the  system  MgSi03-CaMgSi206.  Runs  in 
the  single-phase  field  at  1365°C  fall  on  a  smooth  curve  when  plotted  against  the  bulk  compositions 
of  the  runs.  Runs  in  the  two-phase  field  at  various  temperatures  fall  off  this  curve,  indicating  that 
diopside  is  present  as  well  as  clinoenstatite  and  that  the  composition  of  the  clinoenstatite  is  fixed  at 
constant  temperature,  independent  of  the  bulk  composition  of  the  run. 


GEOPHYSICAL   LABORATORY 


73 


sition;  diopside  is  present  as  a  phase  in 
these  runs,  and  the  Mg-rich  pyroxene  is 
saturated.  Two-phase  runs  at  1300°  and 
1250°C  are  also  shown  in  figure  11,  and 
the  points  on  the  solvus  determined  by 
these  data  are  plotted  in  the  equilibrium 
diagram,  figure  8. 

Reversals  have  been  obtained  at  1365°C 
for  both  sides  of  the  solvus.  At  1365°C 
the  points  on  the  solvus  obtained  by 
unmixing  solid  solutions  are  En78Di22  and 
En35Di65.  Values  obtained  by  homoge- 
nizing pyroxenes  previously  unmixed  at 
lower  temperatures  are  En78.5Di2i.5  and 
En34Di66.  It  has  not  proved  possible  to 
reverse  the  solvus  at  temperatures  below 
1365°C.  At  most  temperatures,  however, 
two  or  more  bulk  compositions  within  the 
two-phase  field  gave  the  same  value  for 
the  solvus.  This  is  a  strong  presumption 
of  equilibrium  though  not  a  proof  of  it. 

The  solvus  curve  defining  the  limit  of 
solubility  of  MgSi03  in  diopside  is  little 
changed  from  our  preliminary  diagram 
(Year  Book  56),  although  additional  data 
have  been  incorporated.  Present  results, 
however,    show   that    the    solubility   of 


CaMgSi206  in  protoenstatite  is  much 
more  restricted  than  was  indicated  by 
our  preliminary  data.  The  intersection  of 
the  solvus  curve  with  the  solidus  on  the 
MgSi03  side  is  at  a  composition  of 
En75Di25  rather  than  En55Di45.  The  error 
in  our  preliminary  results  developed 
chiefly  because  runs  were  too  short.  The 
time  required  to  reach  equilibrium  on  the 
MgSi03  side  of  the  solvus  is  2  to  4  times 
as  long  as  on  the  diopside  side.  Present 
results  on  the  MgSi03  side  are  based  on 
runs  of  2  to  6  weeks. 

A  curious  phenomenon  has  been  found 
in  the  temperature  interval  between  1365° 
and  the  solidus  at  1405°C.  Measurements 
of  the  220  reflections  for  a  series  of 
compositions  from  MgSi03  to  En6oDi40 
crystallized  at  1395°C  are  shown  in 
figure  12.  These  points  show  a  progressive 
shift  of  220  with  composition  out  to  a 
bulk  composition  of  En75Di25.  Points  for 
compositions  richer  in  diopside  fall  off  the 
curve  and  indicate  that  the  solvus  for 
this  temperature  is  at  En76Di24.  This 
point  fits  well  with  the  points  on  the 
solvus  obtained  in  the  range   1250°  to 


i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  |  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i 


27.92 
MgSi03 


i    i    i    t    I    i    i    i    i    I    i    t    i    i    I    i    i    i    i    I    i    i    i    i    I    i    i    i    i    I    i    i    i    i    I    i    t    i    i    I    i    i    i    i 


10  15  20  25 

Weight  per  cent 


30 


35 


40 
CoMgSi206 


Fig.  12.     Shift  of  the  220  reflection  in  Mg-rich  pyroxenes  crystallized  at  1395°C.  The  curve  for 
1365°C  runs  is  reproduced  from  figure  11. 


74 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


1365°C  (fig.  11).  However,  220  measure- 
ments for  pyroxenes  in  the  single- phase 
field  at  1395°C  fall  on  a  curve  different 
from  the  curve  obtained  for  1365°C  runs. 
The  1365°C  curve  is  reproduced  in  figure 
12,  and  the  difference  can  be  seen  to  be 
insignificant  for  pure  MgSi03  but  appre- 
ciable for  bulk  compositions  containing 
diopside. 

Full  X-ray  patterns  of  these  1395°C 
runs  show  that  most  of  them  are  clino- 
enstatite  but  that  some  of  them  are 
orthorhombic  enstatite.  The  position  of  the 
220  reflection  does  not  seem  to  be 
significantly  influenced  by  whether  the 
crystal  form  is  orthorhombic  enstatite  or 
clinoenstatite.  There  can  be  no  doubt 
that  the  orthorhombic  enstatite  in  these 
runs  formed  in  the  quench,  inasmuch  as 
the  orthorhombic  form  has  been  proved 
to  be  unstable  at  temperatures  above 
approximately  1000°C  (see  below).  We 
have  obtained  orthorhombic  enstatite  in 
the  quench  in  a  considerable  number  of 
runs  over  a  range  of  bulk  compositions 
in  this  system  but  always  at  temperatures 
above  1385°C.  We  have,  however,  never 
observed  it  to  form  in  the  quench  in  runs 
on  pure  MgSi03  composition. 

Dry  runs  at  all  temperatures  below 
1365°C  in  the  single-phase  field  have  220 
spacings  that  fall  on  the  1365°C  curve. 
These  runs  are  normal  clinoenstatite.  A 
limited  number  of  runs  made  above 
1395°C  seem  to  fall  on  the  1395°C  curve, 
whereas  runs  at  1385°C  scatter  in  be- 
tween. These  data  suggest  that  there  is 
a  hitherto  unrecognized  form  of  Mg-rich 
pyroxene,  stable  above  1385°C.  This  form 
inverts  in  the  quench  to  a  distorted 
clinoenstatite  or,  sometimes,  to  ortho- 
rhombic enstatite. 

High-temperature  X-ray  studies  are 
needed  to  confirm  or  disprove  this 
suggestion.  It  is  difficult  to  reach  tem- 
peratures above  1350°  to  1400°C  with 
diffractometer  heating  stages,  but  the 
problem  is  being  investigated  in  the 
laboratory  of  J.  V.  Smith,  of  the  Univer- 
sity of  Chicago. 

The  orthorhombic  enstatite  ;=±  proto- 


enstatite  inversion  is  extremely  sluggish 
at  atmospheric  pressure.  Pure  ortho- 
rhombic enstatite,  prepared  hydrother- 
mally,  has  been  heated  for  more  than  2 
months  at  1080°C  without  change.  Partial 
conversion  to  protoenstatite  was  observed 
in  a  dry  run  for  3  months  at  1100°C. 
Addition  of  Na2W04  as  a  flux  lowers  the 
temperature  at  which  orthorhombic  en- 
statite will  invert  to  protoenstatite.  With 
Na2W04,  partial  inversion  of  the  ortho- 
rhombic form  was  observed  at  tempera- 
tures as  low  as  1025°C  However,  we  were 
unable  to  convert  clinoenstatite  or  proto- 
enstatite to  orthorhombic  enstatite  at 
any  temperature  at  atmospheric  pressure, 
even  with  the  use  of  Na2W04  as  a  flux. 

The  presence  of  H20  increases  the 
reaction  rate  somewhat.  Orthorhombic 
enstatite  can  readily  be  prepared  from 
MgSi03  glass  under  hydro  thermal  con- 
ditions, but  not  from  clinoenstatite  or 
protoenstatite.  A  reversal  of  the  transi- 
tion over  a  temperature  interval  of  about 
50°  was  accomplished  in  runs  at  500  bars 
H20  which  were  also  fluxed  with  Na2W04. 
In  evaluating  the  hydro  thermal  data, 
however,  account  must  be  taken  of  the 
effect  of  pressure  on  the  inversion. 

Experiments  made  in  single-stage  appa- 
ratus (Boyd  and  England,  Year  Book  60) 
have  shown  that  the  orthorhombic  ensta- 
tite ^  protoenstatite  inversion  is  very 
sensitive  to  pressure.  Pressure  favors  the 
orthorhombic  form,  and  preliminary  data 
indicate  that  the  slope  of  the  transition 
curve  is  about  75°/kb.  A  reversed  bracket 
on  the  transition  was  obtained  at  1525°C 
and  6.7  ±  0.6  kb.  Extrapolation  of  pre- 
liminary hydrothermal  and  high-pressure 
data  indicates  an  inversion  temperature 
at  atmospheric  pressure  of  about  1025°C, 
in  agreement  with  the  runs  at  atmospheric 
pressure  that  were  fluxed  with  Na2W04. 
This  value  is  in  rough  agreement  with  the 
inversion  temperature  of  985°C  deter- 
mined by  Atlas  (1952)  with  LiF  flux. 

The  solubility  of  diopside  in  ortho- 
rhombic enstatite  was  determined  by 
hydrothermal  runs  at  1000  bars  H20  in 
the  temperature  range  800°  to  1000°C.  It 


GEOPHYSICAL   LABORATORY 


75 


proved  impractical  to  use  X-ray  methods 
on  this  part  of  the  solvus.  The  runs  were 
made  long  enough  so  that  the  presence 
or  absence  of  diopside  could  be  estab- 
lished by  microscopic  examination. 

Attempts  to  locate  the  solvus  in  the 
range  1000°  to  1250°C  on  the  MgSi03 
side  of  the  diagram  by  crystallization  of 
runs  with  Na2W04  flux  were  not  success- 
ful. The  flux  differentially  dissolves  Si02 
and  CaO,  and  so  the  products  of  these 
runs  were  usually  pyroxenes  +  forsterite. 
As  long  as  the  two  pyroxenes  are  on  the 
join  MgSi03-CaMgSi206  their  mutual 
solubility  should  not  be  influenced  by  the 
presence  of  the  forsterite.  However,  the 
220  spacings  of  the  clinoenstatites  in  these 
runs  indicated  that  they  contain  virtually 
no  diopside.  The  results  of  the  fluxed  runs 
are  inconsistent  with  the  dry  data  at 
1250°  to  1400°C  and  inconsistent  with  the 
usual  form  of  a  solvus  curve.  A  check  on 
these  results  was  attempted  by  making  a 
hydro  thermal  run  at  1150°C  and  500  bars 
H20.  In  spite  of  a  quartz  buffer  around 
the  run  it  was  severely  desilicated,  and 
the  products  were  clinoenstatite  +  diop- 
side +  forsterite.  Again  the  220  spacing 
of  the  clinoenstatite  indicated  that  it 
contained  virtually  no  diopside.  Hydro- 
thermal  runs  on  the  CaMgSi206  side  of 
the  solvus  gave  results  in  excellent 
agreement  with  dry  runs,  but  for  the  most 
part  these  runs  were  shorter  and  at  lower 
temperature,  and  desilication  was  not  a 
problem.  Various  explanations  are  possi- 
ble for  the  failure  of  the  fluxed  runs  to 
give  consistent  results.  It  may  be  that  a 
variation  of  the  (Mg  +  Ca)/Si  ratio  in 
the  pyroxene  is  responsible.  Evidence  for 
the  existence  of  such  a  variation  in 
orthorhombic  enstatite  in  high-pressure 
runs  has  been  described  (Boyd  and 
England,  Year  Book  59). 

The  System  Diopside-Enstatite-Silica 
J.  F.  Schairer  and  H.  S.  Yoder,  Jr. 

New  studies  on  the  join  diopside- 
enstatite  (see  pp.  68-75)  indicate  that 
the  solid  solution  of  these  pyroxenes  is 


not  complete  and  that  a  large  solid 
immiscibility  gap  exists  at  the  solidus. 
Additional  relations  on  the  pyroxene 
liquidus  of  the  diopside-enstatite-silica 
system  are  thereby  introduced  that  were 
not  distinguishable  by  Bowen  with  the 
techniques  available  to  him  in  1914. 
Difficulties  arose  in  the  new  studies  of 
diopside-enstatite  near  the  solidus  be- 
cause of  the  complex  changes  within  a 
small  temperature  interval,  and  it  was 
realized  that  some  advantage  was  to  be 
gained  by  studying  the  pyroxene  relations 
in  the  presence  of  the  additional  compo- 
nent silica.  For  these  reasons  a  revision 
of  the  system  diopside-enstatite-silica 
was  undertaken. 

The  revised  liquidus  diagram  of  the 
diopside  (CaMgSi206)-forsterite  (Mg2- 
Si04)  -silica  (Si02)  system  of  which  diop- 
side-enstatite-silica is  a  part  is  shown  in 
figure  13.  The  data  in  the  diopside- 
forsterite-enstatite  (MgSi03)  portion  are 
those  of  Bowen  (1914)  with  a  suggested 
revision  on  the  diopside  solid  solution 
(Diss) -forsterite  (Fo)  boundary  curve.  A 
minimum  relation  is  proposed  instead  of 
the  continuous  drop  of  temperature  to  the 
CaMgSi206-Mg2Si04  join,  which  has  been 
shown  to  be  binary.  The  temperatures  in 
the  region  of  the  proposed  minimum  are 
within  experimental  error,  and  the  exact 
relations  cannot  be  elucidated  with 
present  techniques.  The  two-liquid  region 
is  taken  from  the  work  of  Greig  (1927). 
The  significant  additions  to  Bowen's 
study  are  the  realization  of  a  boundary 
curve  separating  the  fields  of  Diss  and 
Pr8S  (protoenstatite  solid  solution)  and  a 
minimum  on  the  Diss-Tr  (tridymite) 
boundary  curve.  At  point  A  the  reaction 
Fo  +  L  =  Diss  iakes  place  at  a  tempera- 
ture of  1405°  ±  2°C.  The  Fo-Pr8S 
boundary  curve  at  temperatures  higher 
than  point  A  is  the  well  known  reaction 
curve  involving  Fo  +  L  =  Prss.  The 
boundary  curve  Fo-Diss  at  temperatures 
below  point  A  is  also  a  reaction  curve  for 
a  part  of  its  traverse  to  the  minimum  in 
the  CaMgSi206-Mg2Si04-MgSi03  compo- 
sition triangle  and  involves  Fo  +  L  = 


76 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


CaMgSi206 
1391.5" 


1387*2 


I890±20? 
Mg2Si04 


1697*5° 
I7I3±5# 


SiO; 


Weight  per  cent 


Fig.  13.  Revised  liquidus  diagram  at  atmospheric  pressure  of  the  diopside-forsterite-silica 
system.  Data  in  diopside-forsterite-enstatite  portion  are  those  of  Bowen  (1914);  two-liquid  region 
based  on  work  of  Greig  (1927). 


Di88.  At  point  B  the  reaction  Prss  +  L 
=  Diss  +  Tr  proceeds  at  a  temperature 
of  1374°  ±  2°C.  There  are  reasons  (see 
pp.  68—75)  to  believe  that  proto- 
enstatite  is  not  the  correct  crystal 
structure  of  the  solid  solutions  crystal- 
lizing on  the  liquidus  labeled  Prss.  The 
powder  X-ray  diffraction  patterns  of 
MgSi03-rich  pyroxenes  quenched  from 
above  about  1370°C  have  unique  charac- 
teristics that  cannot  be  specifically 
assigned  to  the  now  recognized  forms  of 
MgSi03.  They  may  be  related  to,  but  are 
different  from,  what  Glasser  and  Osborn 
(1960)  referred  to  as  "high  enstatite."  It 
is  to  be  understood  that  until  more 
definitive  data  are  at  hand  the  crystal 


structure  of  the  MgSi03-rich  pyroxene 
crystallizing  in  the  fields  labeled  "Prss" 
is  open  to  question. 

Because  of  the  many  significant 
changes  that  take  place  between  the 
temperatures  1410°  and  1370°C,  isother- 
mal sections  were  studied  in  5°  intervals. 
Most  of  these  are  presented  in  figures  14 
to  20,  on  which  are  plotted  the  bulk 
compositions  of  runs  carried  out.  Runs  on 
other  bulk  compositions,  at  slightly 
higher  and  lower  temperatures,  of  course, 
contribute  to  fixing  the  relationships. 

The  relations  at  1410°C,  shown  in 
figure  14,  represent  a  temperature  imme- 
diately above  the  first  critical  change. 
Only   MgSi03-rich   pyroxenes,   Prss,   are 


GEOPHYSICAL   LABORATORY 


77 


stable.  Tridymite  is  believed  to  be  the 
stable  phase  of  Si02,  but  cristobalite  is 
most  often  obtained  metastably.  Points 
C  and  D  lie  on  the  Fo-PrS8  and  Prss-Tr 
boundary  curves  of  figure  13,  respec- 
tively. The  dashed  crystal-liquid  tie  line 
in  the  field  marked  Prss  +  L  in  figure  14 
and  similar  tie  lines  in  fields  involving 
solid  solutions  in  subsequent  figures  are 
estimated. 

At  1405°C,  figure  15,  the  intersection 
of  the  pyroxene  solvus  and  the  solidus 
takes  place.  The  reaction  is  Fo  +  L  — » 
Diss.  The  composition  of  Diss  is  marked 
by  the  letter  G,  about  Di58Pr42,  and  is  the 
maximum  content  of  MgSi03  that  Di  can 
contain  in  solid  solution.  The  maximum 
amount  of  CaMgSi206  held  in  solid 
solution  by  Pr,  H,  is  also  reached  at  this 
temperature;  it  is  estimated  to  be  about 
24  weight  per  cent  Di.  The  point  E  lies  at 
the  junction  of  the  Fo-PrS8,  Fo-Diss,  and 
Pr8S-Diss  boundary  curves  of  figure   13. 


Point  F  marks  a  position  on  the  PrS8-Tr 
boundary  curve  of  figure  13.  All  bulk 
compositions  in  the  triangle  Mg2Si04-(z- 
H  become  crystalline  at  essentially  this 
temperature. 

Lowering  the  temperature  to  1400°C, 
figure  16,  gives  rise  to  five  new  fields: 
Fo  +  DiSs,  Diss  +  L,  Fo  +  Di8s  +  L, 
PrS8  +  DiS8  +  L,  and  Fo  +  Pr88  +  Di88. 
Points  /,  /,  and  K  lie  respectively  on  the 
boundary  curves  of  Fo-Diss,  Prss-Di88,  and 
Pr88-Tr  of  figure  13. 

In  figure  17  are  given  the  relationships 
found  at  1390°C.  A  second  field  of 
Digs  +  L  has  evolved.  Neither  the  extent 
of  solid  solution  in  the  very  CaMgSi206- 
rich  pyroxenes  nor  the  precise  limits  of 
the  crystal  +  liquid  field  were  deter- 
mined. The  points  M,  N,  and  0  lie 
respectively  on  the  boundary  curves 
Fo-Di88,  Prss-Diss,  and  PrS8-Tr  of  figure  13. 

In  the  temperature  interval  1390°  and 
1385°C   (compare  figs.    17  and   18)    the 


CoMgSi206 


Mg2Si04 


Weight  per  cent 
Fig.  14.     Phase  relations  of  diopside-forsterite-silica  system  at  1410°C. 


78 


CARNEGIE     INSTITUTION     OF      WASHINGTON 

CaMaSi206 


Mg2Si04  MgS,03  Si02 

Weight  per  cent 
Fig.  15.     Phase  relations  of  diopside-forsterite-silica  system  at  1405 °C. 

CaMqSi206 


Mg2Si04 


MgSiOj 


SiO, 


Weight  per  cent 
Fig.  16.     Phase  relations  of  diopside-forsterite-silica  system  at  1400 °C. 


GEOPHYSICAL   LABORATORY 


79 


CoMgSLOe 


Di,.+  L 


Mg2Si04 


MgSiO, 


SiO, 


Weight  per  cent 
Fig.  17.     Phase  relations  of  diopside-forsterite-silica  system  at  1390  °C. 


CoMgSi20€ 


1385° 


SiO, 


Mg2Si04  MgSi03  ~'"2 

Weight  per  cent 
Fig.  18.     Phase  relations  of  diopside-forsterite-silica  system  at  1385 °C. 


80 


CARNEGIE     INSTITUTION     OF     WASHINGTON 

CoMgSi206 


Mg2Si04  MgSi03 

Weight  per  cent 

Fig.  19.     Phase  relations  of  diopside-forsterite-silica  system  at  1375  °C. 

CaMgSi206 


SiO; 


Mg2Si04  MgSi03  Si02 

Weight  per  cent 
Fig.  20.     Phase  relations  of  diopside-forsterite-silica  system  at  1370°C. 


GEOPHYSICAL   LABORATORY 


81 


remaining  liquids  in  the  CaMgSi206- 
MgSi03-Mg2Si04  part  of  the  system 
crystallize;  presumably  the  last  liquid  is 
consumed  at  a  minimum  on  the  Fo-DiS8 
boundary  curve.  In  addition,  all  bulk 
compositions  on  the  join  MgSi03- 
CaMgSi206  become  crystalline.  The  reac- 
tion relationship  of  Fo  +  L  — »  Diss 
terminates  at  various  temperatures  along 
the  Fo  +  DiS8  +  L  curve,  depending  on 
the  bulk  composition.  The  points  P  and 
Q  of  figure  18  are  on  the  Prss-Diss  and 
Prss-Tr  boundary  curves,  respectively,  of 
figure  13.  No  attempt  was  made  to  show 
the  tie  lines  in  the  Diss  +  L  region  because 
of  the  wide  spread  in  possible  orientations. 
Figure  19  portrays  the  relations  at 
1375°C,    indicating    the    nature    of    the 


closure  of  the  lowest  temperature  liquids 
and  the  precursory  conditions  of  the 
Pr8S  -f-  L  — >  Diss  +  Tr  reaction.  The 
points  R  and  S  lie  respectively  on  the 
boundary  curves  Prss-Diss  and  Pr88-Tr  of 
figure  13.  Attention  is  called  to  the  new 
fields  DiS8  +  Tr  +  L  and  Diss  +  Tr  that 
result  from  the  complete  crystallization 
of  compositions  on  the  CaMgSi206-Si02 
join  at  1376°  ±  2°C. 

All  compositions  in  the  system 
CaMgSi206-MgSi03-Si02  are  completely 
crystalline  at  1371°  ±  2°C,  and  the  nature 
of  the  system  is  shown  in  figure  20  for  a 
temperature  of  1370°C.  The  important 
points  X  and  Y  are  approximately 
Di24Pr76  and  Di64Pr36,  respectively.  With 
the  exception  of  polymorphic  transitions 


P 


Q> 

13 
+- 
O 
k_ 

a> 

CL 

E 
a> 


1400 


1300 


1200 


1100 


1000 


900 


Clino-  pyroxene 


»  pyroxene  +  liquid 
«  wollastonite  +  liquid 

•  pyroxene  +  wollastonite 


+  liquid 


Mg5Ca5Si03 


30  50 

Mol   per  cent 


is- 


Fe5Ca5Si03 


Fig.  21.     Preliminary    T-X    section    across    the    join    Mgo.6Cao.6Si03    (diopside)-Fe0.6Ca0.5SiO3 
(hedenbergite).  Total  pressure  =  1  atm.  Partial  oxygen  pressure  is  in  equilibrium  with  iron  +  wiistite. 


82 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


and  exsolution  phenomena  little  change 
in  the  character  of  the  isothermal  sections 
takes  place  with  further  lowering  of  the 
temperature. 

The  application  of  the  revisions  of  the 
diopside-forsterite-silica  system  to  peno- 
logical problems  is  of  exceptional  import. 
No  attempt  will  be  made  here  to  evaluate 
the  new  implications.  Light  is  cast  on  the 
presence  or  absence  of  hypersthene  in 
natural  rocks  (see  Tilley,  1961),  the 
reaction  relations  of  olivine  with  liquid 
to  produce  hypersthene  in  some  cases  and 
augite  in  others,  the  phenocryst-ground- 
mass  relations  of  hypersthene,  pyroxene 
zoning  and  exsolution,  and  many  aspects 
of  fractionation  in  magmas  and  the 
evolution  of  derivative  magmas. 


Preliminary  Results  on  Melting  Relations 

of  Synthetic  Pyroxenes  on  the 

Diopside-Hedenbergite  Join 

A.  C.  Turnock 

A  study  of  melting  relations  of  the 
Mg-Fe-Ca  pyroxenes,  with  compositions 
in  the  three-component  system  MgSi03 
(En)-FeSi03(Fs)-CaSi03(Wo),  has  been 
started  with  compositions  along  the  join 
diopside  (MgCaSi206)-hedenbergite 
(FeCaSi206)  using  a  controlled-atmos- 
phere  quenching  furnace  with  a  total 
pressure  of  1  atmosphere  and  a  partial 
pressure  of  oxygen  that  would  be  in 
equilibrium  with  Fe  -f-  Fei_xO.  The 
oxygen  pressure  was  regulated  by  mixing 
carbon  dioxide  and  carbon  monoxide 
(Darken  and  Gurry,  1945). 

A  diagram  of  the  experimental  results 
is  presented  in  figure  21.  There  is  a 
complete  series  of  monoclinic  pyroxenes 
from  diopside  to  hedenbergite,  but  the 
Fe-Mg  substitution  causes  important 
changes  in  the  stability  of  the  pyroxenes. 
The  Mg-rich  pyroxenes,  as  shown  on  the 
left-hand  side  in  figure  21,  melt  through 
a  temperature  range  given  by  the  solidus 
and  liquidus  curves,  and  this  part  of  the 
diagram  is  essentially  binary.  The  effect 
of  iron  content  in  lowering  the  tempera- 


tures of  the  solidus  and  liquidus  is 
pronounced.  Pyroxenes  richer  in  the 
hedenbergite  molecule  than  about  60  per 
cent,  however,  will  not  melt  but  convert 
to  wollastonite  solid  solution.  These  two 
phases  may  be  polymorphs  across  the 
range  Hed  76  to  Hed  100.  The  wollas- 
tonite solid  solution  persists  metastably 
at  lower  temperatures,  and  in  the  diagram 
the  two  curves  that  define  its  subsolidus 
conversion  to  pyroxene  are  based  on  the 
reverse  reaction,  pyroxene  — ■»  wollastonite 
solid  solution.  Time  studies  of  this 
reaction  satisfactorily  showed  that  the 
transition  interval  was  not  occasioned  by 
incomplete  reaction,  and  the  two  curves 
intersect  the  liquidus  at  positions  that 
satisfy  boundary  points  for  the  field 
"pyroxene  +  wollastonitess  +  liquid." 

Wollastonite  solid  solution  melts 
through  an  interval  of  about  90°C.  In  the 
low-temperature  part  of  the  field  "wollas- 
tonitess +  liquid"  there  is  probably 
another  field,  "wollastonitess  +  liquid  + 
tridymite."  Small  amounts  of  a  silica 
phase  have  been  observed,  but  there  is 
not  yet  enough  information  to  draw  in  a 
field  boundary. 

Metamorphic  Petrology 

Metamorphic  Reactions  Involving 
Two  Volatile  Components 

H.  J.  Greenwood 

Many  metamorphic  reactions  involve 
more  than  one  volatile  or  mobile  compo- 
nent and  are  therefore  influenced  by 
pressure,  temperature,  and  the  compo- 
sition of  the  coexisting  fluid  phase.  The 
equilibrium  relationships  may  be  por- 
trayed in  a  variety  of  ways,  for  example, 
by  plotting  the  chemical  potentials  of  the 
mobile  components  against  one  another 
at  constant  temperature  and  pressure 
(Korzhinskii,  1959;  Zen,  1961).  Alterna- 
tively, the  situation  may  be  represented 
on  an  isobaric  T-x  diagram,  on  which  are 
plotted  the  temperature  and  the  compo- 
sition of  the  coexisting  fluid  phase,  the 
other    components    being    regarded    as 


GEOPHYSICAL   LABORATORY 


83 


nonvolatile  or  immobile.  This  kind  of 
diagram  has  some  advantages  over  the 
chemical  potential,  or  \ii  versus  /*/, 
diagram,  not  the  least  of  which  is  its 
direct  use  of  the  measurable  variables 
temperature,  pressure,  and  composition. 

Equations  have  been  derived  for  the 
equilibrium  boundaries  between  reacting 
phase  assemblages  in  such  systems.  These 
have  the  same  form  as  the  usual  expres- 
sions for  crystal-liquid  equilibria,  but 
they  do  not  carry  the  restriction  that  the 
relative  proportions  of  the  two  volatile 
components  are  limited  by  the  propor- 
tions of  the  other  components.  The  effect 
of  removing  this  restriction  is  to  make 
stable  many  reactions  that  would  nor- 
mally be  regarded  as  metastable. 

The  slope  of  an  equilibrium  boundary 
for  a  reaction  taking  place  at  constant 
pressure  in  the  presence  of  a  one-phase 
binary  fluid  having  zero  enthalpy  of 
mixing  is 

\dx2/p        AS  \x2        xj 

where  x2  is  the  mole  fraction  of  component 
2  in  the  fluid  and  v2  is  the  stoichiometric 
coefficient  of  component  2  in  the  reaction. 
All  reactions  that  take  place  in  such 
systems  can  be  expressed  by  the  general 
relation 

aA  -^bB  +  v\  +  v2 

in  which  a  moles  of  solid  phases  A  react 
to  give  b  moles  of  solid  phases  B  and  v\ 
and  v2  moles  of  the  volatile  components 
1  and  2,  respectively.  The  equation  of  the 
reaction  should  be  written  so  that 

\v\\  +  |  v%  |  =  1     and     vi  +  v2  ^  0 

to  make  the  stoichiometric  coefficients 
equivalent  to  the  mole-fraction  compo- 
sition of  the  gas  given  off  in  the  reaction. 
Inspection  of  these  equations  reveals 
several  points  of  interest  to  metamorphic 
petrology.  If  v2  =  0, 


If  pi  = 


dT\         RT  ( J_ 

,dx2/P        AS  \x2 
>  0 


(2) 


If  Vi  >  0  and  v2  >  0, 

(dT/dx2)P  =  0     (Tmax) 

where  vx  =  xh  v2  =  x2.  (3) 

If  vi  =  —v2,  equal  amounts  of  compo- 
nents 1  and  2  appear  on  opposite  sides  of 
the  reaction,  and  their  entropies  tend  to 
cancel,  making  AS  for  the  reaction  small 
and  (dT/dx2)p  correspondingly  large. 
Accordingly,  as 

A£->0,     (dT/dx2)p-><x>         (4) 
If  -1  <  vx  <  0,  1  >  v2  >  0 

(l^il  <  W), 


+  00    > 


If  1  >  Vi  >  0,  - 1  <  v2  <  0 


(5) 


—  00    < 


(U)p<S(-ir)  (6) 


The  importance  of  these  rather  terse 
statements  to  metamorphic  petrology  can 
best  be  appreciated  by  examining  some 
geologically  interesting  reactions  that 
lend  themselves  to  this  treatment.  Equa- 
tion 1  describes  a  reaction  in  which  only 
component  1  is  given  off  (H20  in  fig.  22). 
As  an  example  of  such  a  reaction  we  might 
take 


Ca2Mg5Si8022(OH)2 

Tremolite 


2CaMgSi206    + 

Diopside 


dT\     =  RT 

,dx2 ) P        AS 
<  0 


(1) 


3MgSi03  +  Si02  +  H20 

Enstatite  Quartz 

Equation  2  describes  a  reaction  in  which 
only  component  2  is  given  off  (C02  in 
fig.  22).  Example  (see  fig.  22): 

MgC03     ->     MgO  +  C02 

Magnesite  Periclase 

Equation  3  describes  a  reaction  in  which 
both  volatile  components  are  given  off, 
such  as 

iCa,Mg5Si8022(OH)2  +  |CaC03  + 

Tremolite  Calcite 


84 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


|Si02 
Quartz 


£CaMgSi206  +  |C02  +  |H20 

Diopside 

(See  fig.  22,  Tmax  at  xCoz  =  0.75.) 
Equation  4  describes  a  reaction  such  as 


Mg(OH)2    +    C02 

Brucite 


MgC03   -f   H20 

Magnesite 


(See  fig.  22,  vertical  boundary.) 

Equation  5  describes  a  reaction  such  as 
|CaMg(C03)2    +    |Si02    +    iH20    -> 


Dolomite 


Quartz 


fCa2Mg5Si8022(OH)2  +  |CaC03  +  |C02 

Tremolite  Calcite 

Equation  6  describes  a  reaction  such  as 
4H4Mg3Si209  +  9CaC03  +  5C02  -> 

Serpentine  Calcite 

Ca2Mg5Si8022(OH)2  + 

Tremolite 

7CaMg(C03)2  +  7H20 

Dolomite 


E 


H20 


025  0.5  0.75 

Mole  fraction  C02 


CO^ 


Fig.  22.  Diagrammatic  sketch  illustrating 
the  six  types  of  crystal-vapor  equilibrium  reac- 
tions in  binary  gas  mixtures.  H20  is  component 
1,  and  CO2  is  component  2,  of  the  equations. 


We  may,  for  the  sake  of  discussion, 
regard  these  reactions  as  models  of 
metamorphic  isograds.  The  most  obvious 
feature  is  that  an  isograd  defined  on  the 


basis  of  a  reaction  that  evolves  one 
volatile  component  may  cross  an  isograd 
defined  on  the  basis  of  a  reaction  that 
evolves  the  other  volatile  component.  In 
addition,  a  plot  like  figure  22  may  be 
regarded  as  a  map  of  an  area  that  has  a 
gradient  in  the  proportions  of  C02  and 
H20  across  it  at  a  large  angle  to  the 
thermal  gradient.  If  such  an  area  could 
be  found  in  the  field,  containing  rocks  of 
suitable  compositions,  it  should  be  pos- 
sible to  demonstrate  the  crossing  of 
isograds.  In  reactions  like  the  formation 
of  diopside  from  tremolite,  calcite,  and 
quartz,  it  is  clearly  of  great  importance 
to  know  something  of  the  composition  of 
the  fluid  phase  in  equilibrium  with  the 
minerals  before  coming  to  any  conclusion 
about  the  temperature  of  metamorphism, 
even  assuming  some  knowledge  of  the 
total  pressure. 

Reactions  like  those  described  by 
equations  4,  5,  and  6  are  perhaps  the  most 
interesting  of  all  when  they  are  regarded 
as  isograds.  Their  steep  slopes  in  T-x 
plots  like  figure  22  show  that  the  progress 
of  many  such  reactions  is  affected  more 
by  the  composition  of  the  coexisting  fluid 
phase  than  by  either  temperature  or 
pressure.  This  observation  leads  directly 
to  the  concept  of  an  isograd  that  is 
essentially  neither  isotherm  nor  isobar 
but  that  provides  a  firm  limit  on  the 
composition  of  the  fluid  with  which  the 
minerals  of  the  rock  could  have  been  in 
equilibrium.  It  cannot  be  too  strongly 
urged,  therefore,  that  when  an  isograd  is 
under  discussion  the  chemical  reaction  be 
precisely  defined. 

Experiments  are  now  under  way  that 
will  fix  the  positions  of  reactions  of  the 
sort  just  discussed  in  the  system  MgO- 
CaO-Si02-H20-C02.  The  apparatus  is 
essentially  the  same  as  was  used  in  an 
earlier  investigation  of  the  system 
NaAlSi206-H20-argon  (Greenwood,  1961) 
in  which  the  solid  phases  are  held  in  open 
capsules  in  a  bomb  containing  a  mixture 
of  C02  and  H20.  Pressure  and  tempera- 
ture are  measured,  and  the  composition 
of  the  gas  is  analyzed  at  the  end  of  each 


GEOPHYSICAL   LABORATORY 


85 


run.  The  stability  of  wollastonite  has 
been  studied  rather  fully,  and  preliminary 
data  are  now  available  on  a  number  of 
other  equilibria.  Figure  23  shows  the 
stability  relations  of  wollastonite  in 
mixtures  of  C02  and  H20  at  1000  and 
2000  bars.  All  the  reactions  shown  repre- 
sent reversals  of  the  equilibrium.  The 
data  are  in  good  agreement  with  those 
of  Harker  and  Tuttle  (1956),  assuming 
that  the  CO 2  and  H20  mix  ideally.  This 
apparent  close  approach  to  ideal  mixing 
is  probably  illusory,  because  it  seems 
likely  that  the  gas  mixture  contains  three 


800 


700 


o 


600 


£  500 


400 


300 


P  =  2000  bars 


Calcite  +  Quartz 


H,0 


0.25  0.50  0.75 

Mole  fraction  CO2 


CO; 


Fig.  23.  Stability  relations  of  calcite,  quartz, 
and  wollastonite  in  mixtures  of  H20  and  C02. 
Circles,  2000  bars;  rectangles,  1000  bars. 


molecular  species  rather  than  two.  Re- 
action between  C02  and  H20  to  produce 
H2CO3  could  easily  produce  the  same 
effect  as  ideal  mixing  of  C02  and  H20  on 
a  solid-gas  equilibrium.  The  accumulation 
of  more  data  on  mineral  equilibria  in  the 
mixtures  will  allow  direct  estimation  of 
the  extent  of  reaction  between  H20  and 
CO 2.  In  addition  to  the  wollastonite 
reaction,  preliminary  runs  indicate  that 
the  reaction  of  talc,  calcite,  and  quartz  to 


give  diopside  occurs  at  a  lower  tempera- 
ture than  the  wollastonite  reaction. 


iMg3Si4O10(OH)2  +  |CaC03  +  JSi02  -» 

Talc  Calcite  Quartz 

3CaMgSi206  +  iH20  +  |C02 


According  to  equation  3  this  reaction 
curve  must  pass  through  a  maximum  in 
temperature  where  #002  =  0.75.  At  a 
total  pressure  of  1000  bars  the  tempera- 
ture of  this  maximum  has  been  deter- 
mined to  be  600°  =fc  25°C,  at  least  25° 
lower  than  the  wollastonite  curve  at  this 
composition,  confirming  the  field  obser- 
vation that  diopside  can  be  formed  at 
lower  temperatures  than  wollastonite. 


Synthesis  and  Stability  of  Anthophyllite 
H.  J.  Greenwood 

Pure  magnesian  anthophyllite,  though 
of  limited  natural  occurrence,  has  been 
the  subject  of  considerable  attention  in 
the  geological  and  geochemical  literature. 
A  significant  part  of  this  interest  may  be 
traced  to  the  classic  paper  by  Bowen  and 
Tuttle  (1949)  on  the  system  MgO-Si02- 
H20,  describing  experiments  in  which 
they  were  unable  to  demonstrate  the 
stability  of  this  mineral.  Since  that  time 
there  have  appeared  a  number  of  papers, 
both  theoretical  and  experimental,  agree- 
ing with  their  conclusion  that  the  mineral 
is  not  stable  in  the  presence  of  excess 
H20.  Anthophyllite  has  been  under  study 
at  this  Laboratory  for  more  than  three 
years.  Last  year  (Greenwood,  Year  Book 
60,  p.  105)  the  existence  of  a  stability 
range  in  the  presence  of  excess  H20  was 
indicated.  Continuation  of  the  study  has 
produced  enough  data  to  allow  detailed 
discussion  of  the  upper  and  lower  stability 
limits  and  nucleation  kinetics  of  the 
mineral.  Fyfe  (1962)  has  recently  given 
an  account  of  some  experiments  with 
unanalyzed  natural  materials,  which  also 
indicate  that  the  mineral  has  a  range  of 
stability  in  the  presence  of  excess  H20, 
but  which  do  not  define  the  reactions  by 


86 


CARNEGIE      INSTITUTION      OF      WASHINGTON 


which  anthophyllite  may  form  from 
lower-temperature  assemblages  including 
talc. 

Synthesis  of  anthophyllite  is  not  easy. 
Most  of  the  hydrothermal  experiments 
that  have  failed  to  produce  the  mineral 
have  failed  because  of  its  extreme 
reluctance  to  nucleate,  even  well  within 
its  own  field  of  stability.  Glasses,  oxide 
mixes,  and  mixtures  of  the  other  minerals 
in  the  system  MgO-Si02-H20  in  various 
proportions  do  not  crystallize  directly  to 
anthophyllite,  even  when  maintained 
within  the  anthophyllite  stability  field 
for  periods  as  long  as  4  months.  The  use 
of  solutions  of  MgCl2  and  of  HC1  did  not 
seem  to  facilitate  the  nucleation.  The  only 
way  in  which  anthophyllite  could  be 
produced  in  the  absence  of  preexisting 
nuclei  was  by  the  metastable  decompo- 
sition of  talc  at  1000  bars  and  830°C  for 
a  period  of  20  hours.  This  procedure  for 
obtaining  starting  materials  with  which 
to  test  the  stability  of  anthophyllite  was 
employed  throughout  most  of  the  investi- 
gation. The  materials  so  obtained  are 
known  to  be  of  the  pure  magnesian 
anthophyllite  composition;  they  have  a 
refractive  index  of  ftz  =  1.615  and  give 
all  the  characteristic  X-ray  reflections  of 
natural  anthophyllite. 

This  metastable  nucleation  of  antho- 
phyllite about  80°C  above  its  upper 
thermal  stability  limit  at  1000  bars  is 
readily  explained  on  the  basis  of  crystal 
structure.  It  seems  unlikely  that  the 
mineral  could  form  in  the  complete 
absence  of  any  nuclei  so  far  above  its 
point  of  thermal  breakdown.  The  require- 
ment that  talc  be  used  as  the  starting 
material  suggests  that  the  talc  is  supply- 
ing nuclei  having  the  anthophyllite 
structure.  Disintegration  of  the  sheets  of 
tetrahedra  in  the  talc  structure  into 
strips  could  provide  the  necessary  struc- 
tural units  having  aanth  =  ctaic  and  canth  A 
frtaic  =  30°.  Rate  studies  support  this 
conclusion.  Talc  held  under  these  condi- 
tions breaks  down  rapidly  (half-life  73^2 
hours)  to  anthophyllite,  protoenstatite, 
and  quartz.  Anthophyllite  increases  faster 


than  the  other  reaction  products  until 
16J/2  hours  have  elapsed,  after  which  time 
it  decreases,  becoming  undetectable  after 
120  hours.  The  final  products  are  the 
phases  stable  at  830°C  and  1000  bars, 
quartz  and  enstatite.  Von  Gehlen  (1962) 
has  shown  that  talc  heated  at  1300°C  at 
atmospheric  pressure  is  transformed  into 
protoenstatite  and  quartz,  with  the 
protoenstatite  oriented  in  the  same  way 
with  respect  to  the  parent  talc  structure 
as  is  postulated  here  for  anthophyllite. 

Rate  studies  on  the  decomposition  of 
an  analyzed  natural  anthophyllite  (A1203, 
1.94;  FeO,  11.12;  CaO,  0.64  weight  per 
cent)  at  1000  bars  have  shown  that  for 
times  up  to  40  days  at  800°C  (50°C  above 
its  stability  limit)  the  amount  of  break- 
down is  barely  perceptible,  although  at 
850°C  it  is  complete  in  5  days.  The 
excellent  crystallinity  and  relatively 
coarse  grain  size  of  the  natural  material, 
perhaps  together  with  its  departure  from 
the  pure  magnesian  end  member,  evi- 
dently make  it  very  slow  to  react,  and 
extrapolation  indicates  that  4  or  5  months 
would  be  required  to  decompose  the 
mineral  near  the  equilibrium  curve. 

Starting  materials  for  the  runs  used  to 
define  the  limits  of  stability  of  the 
amphibole  were  prepared  in  the  manner 
described.  Oxide  mixes  on  each  of  the 
bulk  compositions  MgO-Si02,  7MgO- 
8Si02,  and  3MgO«4Si02  were  separately 
crystallized  well  inside  the  stability  field 
of  talc,  and  then  given  the  heat  treatment 
to  produce  anthophyllite,  protoenstatite, 
and  quartz  from  the  talc,  together  with 
the  other  phases  inherited  from  the 
crystallization  in  the  talc  field.  These 
starting  materials  consisted  of  various 
mixtures  of  forsterite,  enstatite  (proto 
and  clino),  quartz,  cristobalite,  antho- 
phyllite, and  talc.  This  is  an  obvious 
disequilibrium  mixture  on  the  requisite 
bulk  composition  containing  as  nuclei  all 
the  phases  to  which  the  mixture  could 
finally  crystallize  at  equilibrium.  Runs 
are  from  3  to  4  months  in  duration,  at  the 
end  of  which  time  the  reactions  are  from 
about    30    to    100    per    cent    complete. 


GEOPHYSICAL   LABORATORY 


87 


4000- 


3000 


o 

-Q 


Q> 


in 


2000 


1000 


o 


HpO 


0 


MgO  po     En        Si02 


600 


800 


700 

Temperature  ,°C 

Fig.  24.     Stability  relations  of  talc,  anthophyllite,  and  enstatite 


Depending  on  the  bulk  composition,  the 
final  products  consist  of  mixtures  of  any 
pair  of  the  phases  forsterite,  orthoensta- 
tite,  anthophyllite,  talc,  quartz.  Proto- 
enstatite,  clinoenstatite,  and  cristobalite 
disappear. 

The  experimental  results  are  shown  in 
figure  24.  All  runs  shown  represent 
reversals  of  the  reactions  represented  by 
the  equilibrium  curves.  The  arrows  beside 
the  run  symbols  indicate  the  direction 
from    which    the    equilibrium    was    ap- 


proached. Shorter  runs  in  which  no 
reaction  occurred  are  not  considered 
significant  and  are  not  reported.  Both  the 
upper  and  the  lower  stability  limits  of 
anthophyllite  have  been  determined  by 
reversing  the  reactions  at  several  points, 
and  they  are  considered  to  represent 
stable  reversible  equilibria.  The  upper 
stability  limit  of  talc  and  the  lower  limit 
of  enstatite  in  equilibrium  with  H20  are 
indicated  by  the  two  dashed  curves.  Both 
reactions    must    occur    in    this    narrow 


88  CARNEGIE     INSTITUTION      OF      WASHINGTON 

interval,  but  it  has  not  been  possible  to  inary  literature   survey  reveals   only  a 

determine  their  relative  positions.  small   number   of   chemical   analyses   of 

In  summary,  anthophyllite  has  a  range  chlorites    from    low-grade    metamorphic 

of  stability  in  the  presence  of  excess  H20;  rocks.  Turnock  (Year  Book  59)  showed 

and   talc,    anthophyllite,    and    H20   can  that  iron  chlorite  may  exist  in  equilibrium 

coexist    in    stable    equilibrium    over    a  with  quartz  up  to  almost  600°C  at  a  total 

narrow  temperature  interval.  pressure  of  2000  bars.   The  magnesium 

analogues    of    these    chlorites    may    be 

Quartz-Chlorite  Assemblages  in  the  System  represented  in  the  system  MgO-Al203- 

MgO-Al2Os-Si02-H20  Si02-H20,  studied  by  Yoder  (1952)  and 

J.J.  Fawcett  and  H.S.  Yoder,  Jr.  Roy  and  Roy  (1955);  neither  of  these 

studies,    however,    shows    chlorite    and 

Investigation    of    synthetic     systems  quartz  as  a  compatible  mineral  pair  in 

closely  related  to  low-grade  metamorphic  the  temperature  range  450°  to  900°C  at 

rocks  offers  many  opportunities  to  the  15,000  psi   (Yoder)   or   130°  to   1300°C 

experimental    petrologist.    Experimental  between  5000  and  30,000  psi  (Roy  and 

study    of    the    low-grade    metamorphic  Roy). 

rocks  is  handicapped  by  the  slow  rates  of  As  natural  occurrences  indicate  that 

reaction,  which  are  probably  due  to  the  chlorite   and   quartz   may   constitute   a 

absence  of  a  liquid  silicate  phase  so  that  stable  assemblage,  a  series  of  experiments 

diffusion  of  ions  must  take  place  either  in  have  been  performed  in  an  attempt  to 

the  solid  state  or,  more  likely,  through  a  define  a  stability  field  for  Mg  chlorites 

gas    phase.     To    understand     chemical  and  quartz.   Starting  materials  for  the 

reactions  in  low-grade  metamorphic  rocks  experiments  were  glasses  whose  compo- 

and   quantitatively   evaluate   prevailing  sitions,  shown  in  figure  25,  for  the  most 

physical   conditions   it   is   important   to  part  lie  on  the  anhydrous  join  anthophyl- 

study  in  the  laboratory  the  phase  rela-  lite   (7MgO-8Si02)-Mg  gedrite   (5MgO- 

tions    of   the    minerals    and    groups    of  2Al203-6Si02);  the  glasses  were  prepared 

minerals  that  play  a  significant  role.  under  the  supervision  of  Dr.  Schairer  in 

Chlorite  and  quartz  are  two  of  the  most  connection  with  a  study  of  these  amphi- 
common  minerals  present  in  low-grade  boles.  Other  glasses  were  supplied  by 
metamorphic  rocks.  Indeed,  they  charac-  Schairer  and  by  Yoder.  The  compositions 
terize  the  chlorite  zone  of  progressive  plot  between  the  composition  of  the 
metamorphism  and  often  persist  into  the  chlorite  solid  solution  series  and  quartz  as 
biotite  and  even  garnet  zones  (Barrow,  projected  from  H20  onto  the  face 
1893;  Tilley,  1925;  Mason,  1962).  In  MgO-Al203-Si02,  in  the  system  MgO- 
terms  of  the  facies  concept  of  meta-  Al203-Si02-H20.  Determinative  runs  were 
morphism  the  quartz-chlorite  assemblage  made  at  pressures  of  2  and  5  kb  in  cold- 
characterizes  in  particular  the  quartz-  seal  hydrothermal  bombs,  the  duration  of 
albite-muscovite-chlorite  subfacies  of  the  the  runs  varying  from  1  to  6  weeks.  At 
greenschist  facies  (Fyfe,  Turner,  and  2  kb  and  temperatures  of  400°  and  600°C 
Verhoogen,  1958,  p.  218).  The  large  talc  was  found  to  grow  readily,  in  the 
volume  and  wide  distribution  of  these  early  stages  of  the  runs,  with  charges 
rocks  on  the  earth's  surface  suggest  that  containing  less  than  15  per  cent  A1203. 
these  two  minerals  may  exist  together  in  Some  of  the  talc  reacts  only  very  slowly 
equilibrium  over  a  considerable,  though  to  produce  a  more  stable  mineral  assem- 
as  yet  unspecified,  range  of  temperature  blage.  Talc  also  grows  from  compositions 
and  pressure.  containing  more  than  15  per  cent  A1203, 

Hutton  (1940)  suggested  that  the  but  the  smaller  amounts  produced  from 
chlorites  of  low-grade  metamorphic  rocks  these  compositions  are  consumed  corn- 
are  dominantly  iron  rich,  but  a  prelim-  pletely  by  reaction  in  less  than  3  weeks. 


GEOPHYSICAL    LABORATORY 


89 


Si02 

(QUARTZ 
\TRIDYMITE 
\CRISTOBALITE 


TALC 
STEVENSITE. 

ANTHOPHYLLITE 
KUPFFERITE] 

„   ,        ENSTATIT 
CLINOENSTATITE, 
PROTOENSTATITI 


PYROPHYLLITE 


Al- MONTMORILLONITE 
KAOLIMITE 


PERICLASE 
BRUCITE 
MgO 


SERPENTINE 
FORSTERITE 


KYANITE 
ANDALUSITE 
SILLIMANITE 


MULLITE 


SPINEL 


CORUNDUM 
DIASPORE 
BOEHMITE 
GI6BSITE 
A!203 


Weight  per  cent 


Fig.  25.  Composition  of  glasses  (+)  used  in  the  determination  of  the  quartz-Mg  chlorite  stability 
field.  The  projected  composition  of  Al  montmorillonite  should  plot  on  top  of  pyrophyllite  and  not 
kaolinite,  as  shown. 


Preliminary  results  are  illustrated  in 
figure  26.  Quartz  and  chlorite  are  stable 
together  up  to  almost  600°C,  but  the 
projected  shape  of  the  chlorite-quartz 
field  is  a  reflection  of  the  stability  of  the 
chlorite  solid  solution  series.  At  lower 
temperatures  (450°C)  a  wide  range  of 
chlorite  compositions  is  stable;  with 
increasing  temperature  the  range  becomes 
narrower.  The  chlorite  coexisting  with 
quartz  at  the  maximum  temperature  of 
the  quartz-chlorite  field  contains  about 
20  weight  per  cent  A1203.  An  increase  or 
decrease  in  the  A1203  content  of  the 
chlorite  results  in  a  reduction  of  the 
maximum  temperature  of  the  quartz- 
chlorite  stability  field.  Montmorillonite 
crystallized  at  temperatures  below  450°C. 
Talc  formed  metastably  in  the  less 
aluminous  compositions  in  the   quartz- 


chlorite  field,  but  in  the  longer  runs 
(6  weeks)  quartz  and  chlorite  represent  a 
more  stable  assemblage.  Many  runs 
produced  the  7  A  polymorph  of  the 
chlorites  (aluminous  serpentine,  Yoder, 
1952;  septechlorite,  Nelson  and  Roy, 
1958),  but  most  runs  of  3  weeks  or  longer 
produced  the  14  A  chlorite. 

Chlorite  and  quartz  coexist,  together 
with  a  third  phase  (talc  or  cordierite)  on 
either  side  of  the  quartz-chlorite  stability 
field.  These  two  small  fields  are  limited  at 
higher  temperatures  by  the  field  of  talc 
+  chlorite  +  cordierite  and,  for  more 
Si02-rich  compositions  than  those  in  the 
section  of  figure  26,  by  talc  +  cordierite 
+  quartz. 

At  temperatures  in  the  700°  to  850°C 
range  a  careful  search  has  been  made 
for    the    phases    in    the   anthophyllite 


90 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Q. 

E 


700" 


650' 


600' 


550° 


500' 


450* 


400' 


TALC  +  CHLORITE  +  CORDIERITE  +  GAS 
o  o  o 


CHLORITE 


CORDIERITE 
O     + 
QUARTZ 


MONTMORILLONITE  +  QUARTZ +CHL0RITE  +  GAS 


O  o 

MONTMORILLONITE   +  GAS 

o  o 


7MgO-8Si02-H20 
ANTHOPHYLLITE 


6Mg0-AI203-7Si02H20 
MgSi    <^-r    AIAI 


5MgO-2AI2036Si02H20 
Mg-GEDRITE 


Fig.  26.     Phase  relations  along  the  extended  join  anthophyllite  (Mg7Si8022(OH)2)-Mg  gedrite 
(Mg5Al4Si6022(OH)2)  in  the  system  MgO-Al203-Si02-H20. 


(Mg7Si8022(OH)2)-Mg  gedrite  (Mg5Al4- 
Si6022(OH)2)  group.  Amphiboles  of  un- 
known composition  have  been  synthe- 
sized, but  they  have  been  shown  to  be 
unstable.  In  view  of  the  existence  of  a 
stability  field  for  pure  anthophyllite 
(Greenwood,  this  report),  work  on  the 
aluminous  anthophyllites  will  be  con- 
tinued in  the  coming  year. 

To  test  the  results  obtained  with 
synthetic  starting  materials,  naturally 
occurring  minerals  have  been  used  as 
starting  materials  for  some  experiments; 
mixtures  of  talc  +  cordierite  have  been 
converted  to  chlorite  +  quartz  +  minor 
talc  within  the  quartz-chlorite  field  of  the 
synthetic  materials,  and  similarly  mix- 
tures of  quartz  +  chlorite  have  been 
converted  to  cordierite  +  talc  +  quartz 
within  the  stability  field  of  that  assem- 


blage, as  indicated  by  runs  using  glasses. 

An  increase  in  pressure  from  2  to  5  kb 
raises  the  temperature  of  maximum 
stability  of  the  quartz-chlorite  assemblage 
from  about  600°  to  625°C.  It  is  of  interest 
to  note  that  the  quartz-chlorite  reaction 
curve  is  about  100°C  below  the  maximum 
stability  of  clinochlore  at  2000  bars, 
whereas  the  analogous  muscovite-quartz 
curve  is  only  15°C  below  the  upper 
stability  limit  of  muscovite  (Yoder  and 
Eugster,  1955).  The  effect  of  pressure  on 
the  lower  stability  limits  of  the  quartz- 
chlorite  stability  field  has  not  yet  been 
determined. 

All  runs  below  450°C  produced  a 
montmorillonite  phase,  but  there  is,  on 
theoretical  grounds,  a  narrow  tempera- 
ture interval  between  the  montmorillonite 
field  and  the  quartz-chlorite  field  in  which 


GEOPHYSICAL   LABORATORY  91 

the  stable  assemblage  is  quartz  or  talc  +  due  to  complex  reactions  involving  chlo- 

montmorillonite  +  chlorite.  The  precise  rite,  quartz,  or  both. 

phase  relationships  have  not  yet  been 

worked  out  at  lower  temperatures.  Alkali-Rich  Igneous  Rocks 

Nelson  and  Roy  (1958)  determined  the  AND  Minerals 
maximum  stability  of  the  chlorite  solid 

solutions  in  the  absence  of  quartz  to  be  The  System  Na20-Al20s-Fe203-Si02  and 

710°C    at    1000    atm.    The    maximum  Its  Bearing  on  the  Alkaline  Rocks 

stability  of  the  chlorite  +  quartz  assem-  Jm  Fm  Schairer  and  D.  K.  Bailey 
blage  was  found  in  the  present  work  to  be 

675°C  at  2  kb.  Although  the  pressures  The  alkaline  rocks,  by  virtue  of  their 

are  not  equivalent,  the  data  are  in  accord  uncommon  chemistry  and  mineralogy — 

with  the  general  rule  that  heterogeneous  with  essential  amounts  of  feldspathoids 

reactions   must   take   place   within   the  and  alkali  pyriboles  in  a  wide  range  of 

stability    fields    of    the    reactants    and  proportions,     and     often     with     minor 

products.    The    present   data   indirectly  amounts  of  rare  minerals — have  engaged 

suggest  that  the  maximum  stability  of  the  attention  of  petrographers  and  ana- 

the  alumina-poor  and  alumina-rich  chlo-  lysts    to    an    extent    that    belies    their 

rites   is   lower  than  that   suggested   by  quantitative  importance;  consequently,  a 

Nelson  and  Roy.  The  breakdown  products  large  body  of  information  is  available  on 

of  clinochlore  and  amesite  obtained  by  them,  and  the  problem  of  their  origin  has 

Nelson  and  Roy  (1958)  do  not  appear  to  provoked   much   speculation   and   argu- 

represent  equilibrium  assemblages.  Runs  ment.  It  does  not  follow,  however,  that 

of   much    greater   duration    than    those  the    considerable    attention   devoted    to 

carried  out  by  Nelson  and  Roy  on  the  these  rocks  has  been  misdirected,  for  the 

chlorite    breakdown    may    yield    assem-  typical  alkaline  centers,  in  common  with 

blages  different  from  the  forsterite,  talc,  kimberlites  and  carbonatites  (with  which 

and  spinel  obtained  by  those  authors.  alkaline  rocks  are  frequently  associated), 

The    information    on    the    limits    of  are  restricted  to  the  stable  continental 

stability  of  the  quartz-chlorite  mineral  areas  of  the  earth's  crust,  and  the  rocks 

assemblage  may  be  applied  in  a  general  are  probably  the  surface  expression  of 

way  to  the  conditions  of  formation  of  deep-crustal  and  subcrustal  activity  in 

potassium-deficient  low-grade  metamor-  epeirogenic   zones.   The  natural  compo- 

phic  rocks.  As  the  chlorite  minerals  in  the  sitional  ranges  in  the  alkaline  rocks,  and 

rocks  always  contain  iron,  however,  the  the  awesome  range  of  rock  names,  make 

data  can  only  indicate  maximum  tern-  generalizations  about  their  composition 

peratures  in  the  pressure  range  under  difficult,  but  a  large  proportion  of  the 

consideration.    Moreover,    the   effect   of  rocks    are    essentially    assemblages    of 

other  phases  such  as  muscovite,  biotite,  nepheline  (or  related  feldspathoids  such 

feldspar,  or  epidote  on  the  quartz-chlorite  as  sodalite  and  cancrinite) ,  sodic  pyroxene 

stability  field  is  unknown.  (or  amphibole) ,  and  alkali  feldspar,  thus 

It   is    clear   from    these    results    that  falling   into   two   broad   groups:   ijolites 

previous  investigators  had  not  made  runs  (nepheline-pyroxene  rocks)  and  foyaites 

of    sufficient    duration    to    obtain    the  (nepheline-f eldspar-pyroxene  rocks) .  Both 

quartz-chlorite  assemblage.   It  can  now  types  bulk  largely  in  alkaline  complexes, 

be  concluded  that  the  synthetic  studies  are  commonly  associates  of  carbonatite, 

support  the  field  observations  that  Mg-  and  hence  figure  prominently  in  theories 

rich    as    well    as    Fe-rich    chlorites    can  of  the  origin  and  the  differentiation  of 

coexist  with   quartz   over  a   large   P-T  these  rocks.   The  study  of  the  system 

range.  Limitation  of  the  coexistence  of  Na20-Al203-Fe203-Si02  offers  an  oppor- 

quartz  and  chlorite  in  the  natural  rocks  is  tunity    to    observe    the    essential    rock- 


92 


CARNEGIE      INSTITUTION      OF      WASHINGTON 


Fe2Oj 


FIELD    SYMBOLS 


Ab 

Ac  

Beto_ 

Cg 

Cor 

Ds 

Hem 

Mul 

Ne 

Ns 

Os 

0 

Trid 

51  e_ 

6  45_ 


ALBITE 
ACMITE 
0  ALUMINA 
CARNEGIEITE 
CORUNDUM 
SODIUM  DISIUCATE 
HEMATITE 
MULLITE 
NEPHELINE 

SODIUM  METASILICATE 
SODIUM  ORTHOSILICATE 
.QUARTZ 
.TRIDYMITE 
.5No20  Fe203-8Si02 
6No20  4Fe2Os  5S>02 


NUMBERED    POINTS  INDICATE   COMPOSITION: 

1.  ACMITE 

2.  5Na20  Fe203  8Si02 

3.  (No20  4SI02)  COMPN. 

4.  SODIUM  DISIUCATE 

5.  SODIUM  METASILICATE 

6.  ALBITE 

7.  JADEITE 
NEPHELINE 


A_AI2OyFe2O3-(Na20  4Si02) 

B_NEPHELINE  -  SILICA  -ACMITE 

C— NEPHELINE  -SODIUM  OISILICATE-ACMITE 

O—ALBITE-SODIUM  DISILICATE-ACMITE 

E_NEPHELINE-ACMITE-5No2OFe,Os8SiO, 

F_NEPHELINE-ACMITE - (No£0  4Si02) 


Fe,0, 


Fig.  27.     The  system  Na20-Al203-Fe203-Si02  to  show  relations  of  the  compounds  and  joins 
studied.  Three  faces  of  the  tetrahedron  have  been  laid  flat  in  the  plane  of  the  base. 


forming  minerals  nepheline,  acmite,  and 
albite  in  equilibrium  with  liquids  the 
compositions  of  which  are  analogous  to 
those  of  natural  alkaline  rocks. 

Study  of  equilibrium  within  the  quater- 
nary system  Na20-Al203-Fe203-Si02  at  1 
atmosphere  pressure  began  with  an 
examination  of  the  join  jadeite-acmite  in 
1948-1949  (Year  Book  48,  p.  32)  as  part 
of  a  more  general  study  of  the  stability 
relations  of  jadeite.  This  join  is  not 
binary,  and  it  was  found  that  the  primary 
phase  for  compositions  acmite  100-20  was 
hematite  or  hematite-corundum  solid 
solution,  and  that  compositions 
jadeite  100-80  gave  nepheline-albite  solid 
solution  as  the  primary  phase  with 
hematite-corundum  solid  solution  as  the 


second  phase.  This  meant  that  relations 
in  the  join  could  be  described  correctly 
only  in  terms  of  the  quaternary  system, 
and  work  was  started  in  five  joins  in  this 
system  the  following  year. 

Phase-equilibrium  data  for  the  bound- 
ing ternary  system  Na20-Fe203-Si02, 
establishing  the  incongruent  melting  of 
acmite,  have  been  published  by  Bowen, 
Schairer,  and  Willems  (1930)  and  for  the 
system  Na20-Al203-Si02  by  Schairer  and 
Bowen  (1956).  Because  reduction  of 
Fe203  to  FeO  increases  with  temperature, 
only  the  parts  of  the  system  Na20-Al203- 
iron  oxide-Si02  with  low  liquidus  temper- 
atures can  be  treated  as  essentially 
quaternary  and  in  the  system  Na20- 
Al203-Fe203-Si02.  Fortunately  this  low- 


GEOPHYSICAL   LABORATORY  93 

temperature  region  embraces  the  compo-  — is    due    to    the    incongruent    melting 

sitions    of    greatest    geological    interest,  relationship  of  acmite,  the  primary  phase 

approximately  within  the  volume  sodium  volume     of    hematite     thus     extending 

metasilicate-acmite-nepheline-silica.    The  through  this  join. 

position  of  this  volume  within  the  tetra-  It  seemed  most  likely,  from  geometric 

hedron    Na20-Al203-Fe203-Si02    can    be  considerations,  that  the  join  nepheline- 

seen  from  figure  27.  The  positions  of  the  acmite-Na20»4Si02  would  intercept  the 

five  joins  first  studied  can  also  be  seen  univariant   line    nepheline-acmite-albite- 

from  this  figure;  they  are  Na20«4Si02-  liquid,  and  that  in  addition  this  plane 

Al203-Fe203  (which  includes  the  acmite  should  contain  the  ternary  reaction  point 

and    jadeite    compositions),    nepheline-  nepheline-acmite-hematite,  which  is  also 

silica-acmite,  nepheline-sodium  disilicate-  of   particular   interest   in   alkaline   rock 

acmite,    albite-sodium    disilicate-acmite,  problems.  During  the  past  few  months 

and       nepheline-5Na20*Fe203-8Si02-ac-  thirty-three  compositions  have  been  pre- 

mite.  Brief  progress  reports  on  these  joins  pared,  and  work  on  this  join  is  nearing 

were  given  in  Year  Books  49,  PP-  46-47,  completion;  the  preliminary  equilibrium 

SO,  pp.  53-54,  and  SI,  pp.  52-53,  noting  diagram  is  given  here  as  figure  28.  The 

the   existence   of  various   eutectics   and  piercing  points  of  two  univariant  lines, 

piercing  points  without  further  details,  acmite-albite- quartz-liquid    and    nephe- 

Clearly,    however,    the    system    has    a  line-acmite-albite-liquid,  and  the  ternary 

bearing  on  problems  other  than  jadeite  reaction  point  nepheline-acmite-hematite 

stability,  which  was  the  initial  stimulus  have  been  located,   at  temperatures  of 

for  its  study.  One  such  is  its  application  758°,  845°,  and  905°C,  respectively. 

to  studies  of  alkaline  rocks,  and  it  was  We   intend   to   publish   the   complete 

during    a    joint    visit    to    the    alkaline  diagrams    for    the    six    joins    and    the 

rock/carbonatite  complex  of  the  Fen  area  crystallization  flow  diagram  in  the  near 

in  southern  Norway  that  we  decided  to  future,  when  we  hope  that  the  bearing  of 

extend  the  work  on  this  system.  the  results  on  the  crystallization  history 

From  the  data  on  the  first  five  joins  in  of  related  alkaline  rocks  and  other  rocks 

the  system  it  is  possible  to  deduce   a  such  as  peralkaline  granites  and  rhyolites 

crystallization  flow  diagram,   but  there  can  be  discussed.  Some  of  the  petrological 

were    no    observational    data    on    the  implications   are   already  apparent   and 

important     univariant     line     nepheline-  will  be  of  interest  to  geologists  working 

acmite-albite-liquid.    Its    existence    and  on  alkaline  rock  problems ;  for  this  reason 

extent  have  to  be  inferred  from  other  a  few  of  them  are  stated  briefly  here, 

data  and  the  geometry  of  the  system.  The  univariant  line  along  which  the 

Such  a  univariant  line  should  link  the  three   important   rock-forming   minerals 

quaternary  reaction  point  hematite-neph-  nepheline,  acmite,  and  albite  crystallize 

eline-acmite-albite-liquid  and  the  quater-  in  equilibrium  with  liquid  spans  a  rela- 

nary    eutectic    point    nepheline-acmite-  tively  large  composition  and  temperature 

albite-sodium    disilicate-liquid,    between  interval    with    the    composition    of    the 

which  there  is  a  considerable  composition  liquid    moving    toward    a    quaternary 

interval  and  a  temperature  interval  of  eutectic  where  the  fourth  solid   phase, 

200°C  (from  915°  ±  5°C  to  below  727°C).  sodium  disilicate,  begins  to  form.  Liquid 

From  the  point  of  view  of  the  alkaline  compositions  along  this  line  range  from 

rocks    it    was    important    to    have    an  those  analogous  to  ijolite  to  a  foyaitic  or 

intersection   of   this   univariant   line   to  nepheline  syenitic  composition,  approx- 

confirm  these  deductions.  That  such  an  imately  from  liquids  containing  40  to  10 

intersection  had  not  been  found  in  any  of  per  cent  potential  acmite :  the  tempera- 

the    previous    five    joins — notably    the  ture  interval  is  from  915°  ±  5°C  to  below 

composition  plane  nepheline-acmite-silica  727°C.  This  means  that  the  three  min- 


94 


CAKNEGIE     INSTITUTION     OF      WASHINGTON 


NEPHELINE 
Na20-AI203-2Si02 

"  1526*2° 


1285*5° 
60 


Undersaturated  60, 

Oversaturated    'c°'eA 


1114*5 

Na20-4Si02       '" 


JO  867*3°       40  50  60 

Weight    per  cent 


70 


80 


1280*5° 
90  ACMITE 

Na20Fe2034Si02 


Fig.  28.     Preliminary  equilibrium  diagram  of  the  join  nepheline-acmite-(Na20-4Si02).  This  is  a 
portion  of  the  join  nepheline-hematite-(Na20-4Si02). 


erals,  in  various  proportions,  can  exist  in 
equilibrium  with  a  wide  range  of  liquid 
compositions  over  a  long  temperature 
interval.  If  similar  conditions  pertain  to 
magmas  of  analogous  composition  there 
would  be  ample  scope  for  differentiation 
and  separation  of  residual  liquids  by  any 
of  the  normally  invoked  physical  proc- 
esses. Such  a  condition  is  entirely  in 
keeping  with  the  wide  compositional 
ranges  seen  in  alkaline  complexes  and 
even,  in  some  hand  specimens  and 
outcrops,  by  segregation  of  early-formed 
minerals.  Equilibrium  between  nepheline- 
acmite-albite  and  a  liquid  containing 
potential  sodium  disilicate  is  consistent 
with  the  observation  that,  of  the  analyses 
in  Washington's  (1917)  tables  that  give 


sodium  metasilicate  in  the  norm,  about 
half  are  of  unsaturated  alkaline  rocks, 
but  probably  more  important  is  the  fact 
that  the  residual  liquid  is  becoming 
progressively  enriched  in  sodium  disilicate 
with  increasing  crystallization.  In  nature 
the  residual  liquid  would  also  become 
increasingly  enriched  in  volatiles,  and 
production  of  such  residual  fluids  would 
explain  the  almost  invariable  alkali 
metasomatism  of  country  rocks  around 
alkaline  intrusions. 

The  liquid  phase  at  the  ternary  reaction 
point  nepheline-acmite-hematite-  liquid 
has  the  composition  of  a  simplified  ijolite, 
figure  28.  This  point  should  be  a  tempera- 
ture maximum  on  the  corresponding 
univariant  line,  such  that  liquids  with  an 


GEOPHYSICAL   LABORATORY 


95 


initial  composition  on  the  more  siliceous 
side  of  the  composition  plane  move 
toward  the  quaternary  reaction  point 
nepheline-acmite-albite-hematite,  i.e.  have 
a  foyaitic  trend,  whereas  those  on  the 
opposite  side  move  toward  a  quaternary 
eutectic  at  which  5Na20«Fe203*8Si02 
crystallizes  with  the  other  three  phases. 
This  increasingly  basic  liquid  trend,  with 
60  per  cent  or  more  acmite  in  some 
compositions,  may  be  compared  with  the 
ijolite-melteigite  trend  seen  in  natural 
rocks.  This  indication  of  the  possibility 
of  an  ijolitic  liquid's  having  two  possible 
divergent  differentiation  trends  depend- 
ing on  small  compositional  variations  is 
interesting,  but  an  even  more  interesting 
feature  is  that  there  appears  to  be  very 
little  temperature  difference  between  the 
ternary  reaction  point  nepheline-acmite- 
hematite  and  the  quaternary  reaction 
point  nepheline-albite-acmite-hematite . 
This  means  that  only  slight  fluctuations 
in  physical  conditions  or  composition  will 
determine  the  differentiation  trend  of  the 
liquid  toward  either  foyaitic  residual 
liquids  or  mafic  melteigitic  liquids. 

The  critical  consideration  in  regard  to 
all  the  above  suggestions  is  whether  the 
equilibrium  relations  at  1  atmosphere  are 
seriously  changed  under  pressure  and  in 
the  presence  of  volatiles,  and  it  is  planned 
in  the  coming  year  to  investigate  the 
effects  of  water  pressure  on  some  of  the 
critical  joins  in  the  system. 

Peralkaline  Residual  Liquids:  Some 
Petrogenetic  Considerations 

D.  K.  Bailey  and  J.  F.  Schairer 

It  was  noted  in  the  above  discussion 
of  the  anhydrous  system  Na20-Al203- 
Fe203-Si02  that  the  residual  liquids 
resulting  from  crystallization  of  nephe- 
line,  albite,  and  acmite  became  increas- 
ingly enriched  in  sodium  disilicate.  In 
nature  this  enrichment  would  be  expected 
to  be  concomitant  with  enrichment  in 
volatiles,  and  fluids  of  this  type  would  be 
expected  to  react  with  wall  rocks  con- 
taining A1203  and  Si02  to  form  nepheline, 


nepheline-albite,  or  albite,  depending  on 
the  proportions  of  A1203  and  Si02  present. 
For  convenience  of  discussion  the  simplest 
expressions  of  the  reactions  are  given  in 
the  equations 

Na20-2Si02  +  A1203  ->  nepheline     (1) 

Na20.2Si02  +  2Si02  +  A1203  -> 

(nepheline  +  albite)     (2) 

Na20-2Si02  +  4Si02  +  A1203  -> 

albite     (3) 

In  pelitic  wall  rocks  reaction  2  would 
have  the  most  general  application,  the 
silica  and  alumina  balance  corresponding 
to  that  of  the  clay  minerals.  Calcareous 
and  ferruginous  pelites  might  be  expected 
to  push  the  reaction  even  further  in  the 
direction  of  greater  production  of  nephe- 
line. It  is  not  inconceivable,  for  instance, 
that  nephelinization  such  as  that  demon- 
strated at  Bancroft,  Ontario  (Tilley, 
1958),  could  result  from  reaction  of 
sodium  silicate-bearing  fluids  and  impure 
aluminous  members  of  the  limestone 
series;  such  a  reaction  would  require  far 
less  transfer  of  material  than  that  needed 
for  metasomatic  replacement  of  pure 
limestone.  With  higher  Si02-Al203  ratios 
in  the  country  rocks  the  reactions  would 
trend  toward  equation  3,  giving  eventu- 
ally only  feldspathization  of  country 
rocks.  The  generation  of  residual  fluids 
rich  in  alkali  silicate  may  thus  be  seen  to 
offer  a  simple  explanation  of  the  meta- 
somatism around  alkaline  intrusions. 

The  parent  peralkaline  undersaturated 
liquid  giving  rise  to  these  fluids  might 
arise  by  partial  melting  of  alkali  basalt. 
Bowen  (1945,  p.  88)  pointed  out  that 
residual  liquids  from  fractionation  of 
basalt  might  become  enriched  in  sodium 
silicate  by  operation  of  the  "plagioclase 
effect,"  and  he  was  well  aware  that 
reaction  of  such  liquids  with  aluminous 
wall  rocks  would  tend  to  form  nepheline 
and  albite.  (Certainly  such  liquids  would 
account  for  adinole  formation  at  dolerite- 
shale  contacts  and  might  play  a  part  in 
the    formation    of    some    spilites.)    The 


96 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


converse  of  residual  liquid  origin — partial 
melting  of  alkali  basalt — could  yield  the 
same  reactive  undersaturated  liquids  and 
perhaps  give  rise  to  nepheline-syenite 
complexes  with  no  associated  basalt. 

It  is  less  obvious  perhaps  that  over- 
saturated  compositions  could  jdeld  resid- 
ual liquids  capable  of  producing  meta- 
somatic  effects  similar  to  those  derived 
from  unsaturated  liquids.  In  the  system 
Na20-Al203-Fe2Q3-Si02,  quartz,  acmite, 
and  albite  also  crystallize  in  equilibrium 
with  a  liquid  becoming  enriched  in  sodium 
disilicate,  the  quaternary  eutectic  lying 
to  the  silica-poor  side  of  the  join  acmite- 
nepheline-Na20-4Si02  (fig.  28),  which 
means  that  liquid  compositions  near  this 
point  can  be  expressed  in  terms  of  the 
molecules  acmite,  albite,  sodium  disili- 
cate,  and  silica,  the  amount  of  silica  being 
less  than  that  required  for  a  composition 
Na20*4Si02.  If  residual  fluids  of  this 
nature  were  to  react  with  sediments  in 
which  the  Si02/Al203  ratio  was  2  or  less, 
such  as  bauxitic,  calcareous,  or  ferrugi- 
nous pelites,  conditions  intermediate 
between  equations  2  and  3  would  obtain, 
and  it  would  be  possible  to  have  neph- 
elinization  of  certain  rocks  around  a 
peralkaline  granite  intrusion. 

Peralkaline  residual  liquids  might  re- 
sult from  a  variety  of  originally  over- 
saturated  compositions,  for,  as  Tuttle  and 
Bowen  (1958,  pp.  84-87)  have  indicated, 
fractionation  of  liquids  on  the  alkali  side 
of  the  albite-orthoclase-quartz  section  in 
the  system  Na20-K20-Al203-Si02  would 
be  expected  to  yield  residuals  rich  in 
alkali  silicates  and  volatiles.  They  point 
out  that  such  liquids  escaping  from  a 
cooling  granite  "may  affect  granitization 
of  the  adjacent  rocks,  providing  the  rocks 
have  appropriate  composition,"  but  their 
data  indicate  that  the  content  of  alkali 
silicates  in  such  liquids  is  such  that 
reaction  with  aluminous  country  rocks 
could  yield  undersaturated  mineral  as- 
semblages in  the  manner  described  above. 
These  considerations  only  serve  to  under- 
line a  point  that  perhaps  receives  too 
little  emphasis,  namely  that  the  system 


NaAlSi04-KAlSi04-Si02  ("petrogeny's 
residua  system")  is  a  residua  system  only 
for  compositions  that  are  subaluminous, 
in  the  sense  used  by  Shand ;  compositions 
off  this  plane,  on  the  alkaline  side,  would 
be  expected  to  fractionate  to  liquids  rich 
in  alkali  silicates.  It  follows  that  partial 
melting  of  basement  or  buried  sediments, 
with  a  bulk  composition  slightly  less  than 
subaluminous,  should  first  yield  such  a 
liquid  fraction.  Usually  this  liquid,  during 
its  uprise,  would  be  expected  to  react 
with  country  rocks  to  produce  granite, 
either  magmatically  or  metasomatical- 
ly,  or  it  might  produce  a  more  diffuse 
regional  metasomatism;  but  in  an  alumi- 
nous environment  it  should  also  be 
possible  for  undersaturated  assemblages 
to  result. 

The  System  Nepheline-Diopside 
J.  F.  Schairer,  Kenzo  Yagi,1  and  H.  S.  Yoder,  Jr. 

In  view  of  the  significance  of  the 
system  nepheline-diopside  as  a  principal 
join  of  the  petrologically  important 
tetrahedron  diopside-nepheline-forsterite- 
quartz,  further  study  was  initiated  in 
1950-1951  (Year  Book  SO,  p.  54),  and  it 
has  been  continued  intermittently  in 
subsequent  years.  The  system  was  found 
to  be  of  such  complexity  that  over  five 
hundred  runs  have  been  made  in  an  effort 
to  determine  the  stability  regions  of  the 
various  mineral  solid  solutions.  Some 
forty  years  ago  the  system  nepheline- 
diopside  was  investigated  by  Bowen 
(1922)  in  connection  with  the  genesis  of 
alnoitic  rocks  of  the  Province  of  Quebec, 
Canada.  The  subsolidus  relations  were 
not  established  at  the  time  because  of  the 
difficulties  in  the  determination  of  minute 
crystalline  phases  under  the  microscope 
without  the  aid  of  X-ray  techniques. 

As  is  evident  from  an  inspection  of 
figure  29,  the  system  is  not  binary,  and 
should  be  considered  a  join  in  the  quinary 
system  Na20-CaO-MgO-Al203-Si02.  All 
the  crystalline  phases  obtained  are  solid 

1  Tohoku  University. 


GEOPHYSICAL   LABORATORY 


97 


I526±2 


1100 


1000 


900- 


—  -o.«^Nej$+OI  +  Mel  +  Di„+  L 


\    Di„+OI+Mel+Ne„+L " 


0i„  +Mel  +Ne,$+L 


Nes,  +  Mel  +  DiM 


NEPHELINE  10 

N020-AI203-2Si02 


20 


30 


60 


1500 


1400 


1300 


1200 


1100 


1000 


900 


70 


80 


90 


40  50 

Weight  per  cent 
Fig.  29.     Pseudobinary  diagram  of  equilibrium  in  mixtures  of  nepheline  and  diopside 


DIOPSIDE 
CoO-MgO-2Si02 


solutions,  and  their  precise  compositions 
have  not  been  determined ;  however,  they 
may  be  designated  by  the  principal  end 
member  present.  The  phases  are  diopside 
solid  solution  (Diss),  nepheline  solid  solu- 
tion (Ness),  carnegieite  solid  solution 
(Cgss),  melilite  (Mel),  olivine  (01),  and 
liquid  (L) .  In  some  regions  of  temperature 
and  composition  a  specific  solid  solution 
may  have  a  fixed  composition.  This 
conclusion  is  based  on  the  principle  that 
a  solid  solution  of  fixed  composition  will 
appear  or  disappear  during  cooling  at  the 
same  temperature  from  a  range  of 
compositions.  For  example,  in  compo- 
sitions rich  in  the  diopside  component,  it 
is  seen  that  01,  Mel,  and  Ness  appear 
successively  at  specific  temperatures.  It 
may  be  concluded  that  in  the  designated 
ranges  of  bulk  composition  the  solid 
solutions  involved  here  were  of  fixed,  but 


unknown,  composition.  Such  solid  solu- 
tions, believed  to  be  of  fixed  composition, 
are  underlined  in  the  figure. 

It  was  not  possible  to  fix  the  solidus  of 
the  system  with  assurance  because  of  the 
difficulty  of  recognizing  small  amounts  of 
glass  in  the  quench  products  or,  for  some 
compositions,  the  presence  or  absence  of 
small  amounts  of  olivine.  In  addition, 
crystal  growth  was  sluggish  and  equi- 
librium could  not  be  established  with 
certainty.  All  runs  having  more  than  5 
per  cent  crystals  were  examined  with 
powder  X-ray  diffraction  techniques. 

The  new  results  reaffirm  Bowen's 
observations  that  melilite  and  olivine 
separate  from  liquids  whose  total  compo- 
sition can  be  expressed  as  a  mixture  of 
nepheline  and  diopside.  Olivine  appears 
to  react  with  liquid  until  consumed, 
producing  diopside  solid  solution  (see  also 


98 


CAKNEGIE     INSTITUTION     OF      WASHINGTON 


Schairer  and  Yoder,  1960,  on  the  system 
nepheline-diopside-silica)  and  melilite. 
The  intimate  association  of  nepheline, 
clinopyroxene,  and  melilite  in  lavas 
suggests  that  melilite  may  indeed  be  a 
differentiation  product  of  an  alkali  basalt 
magma.  Further  information  is  needed  on 
the  composition  of  the  melilites  that 
crystallize  in  the  join  nepheline-diopside 
and  on  the  crystallization  relations  in  the 
portion  nepheline  -  diopside  -  albite-f  orster- 
ite  of  the  simplified  basalt  tetrahedron 
nepheline  -  diopside  -  f orsterite  -  quartz  of 
Yoder  and  Tilley  (Year  Book  59,  p.  67). 

A  Reconnaissance  of  the  Systems 
Acmite-Diopside  and  Acmite-N epheline 

Kenzo  Yagi 

The  main  constituent  molecules  of  the 
pyroxenes  in  the  alkaline  rocks  are 
diopside,  hedenbergite,  acmite,  and  jade- 
ite.  Jadeite  is  present  only  in  minor 
amounts  in  most  of  these  pyroxenes. 
When  the  compositions  of  these  pyrox- 


enes are  plotted  in  the  triangular  diagram 
diopside-hedenbergite-acmite  (+  jadeite) 
the  points  line  in  a  zone  extending  from 
the  diopside  corner  through  the  center  of 
the  triangle  to  the  acmite  corner.  This 
suggests  complete  solid  solution  between 
diopside  and  acmite  and  extensive  solid 
solution  with  hedenbergite  (see  Yagi, 
1953).  Nepheline  syenites  and  related 
rocks  have  nepheline  and  acmitic  pyrox- 
ene as  the  most  important  constituent 
minerals  in  addition  to  alkali  feldspars. 
To  investigate  the  relations  in  alkaline 
rocks  a  study  of  the  joins  diopside- 
hedenbergite-acmite  and  nepheline-diop- 
side-acmite  is  necessary. 

First  the  join  nepheline-diopside  was 
studied ;  the  results  are  given  elsewhere  in 
this  report  (pp.  96-98).  The  results  on 
the  join  acmite-diopside  are  given  here  as 
figure  30.  There  is  a  complete  series  of 
solid  solutions  between  acmite  and  diop- 
side. Bowen,  Schairer,  and  Willems  (1930) 
showed  the  incongruent  nature  of  the 
melting  of  acmite  at  990°C  to  hematite 


IH<JU 

1                     1                     1 

111111                 ^-^ 

Liquid                                                                      ^-^^^"^ / 

^-— -"■"*'             s 

_— -—                        s 

_o— " S 

1300 

""""^           Pyroxene  +  Liquid                 ^ 

1200 

^^\^     ^tf 

*<* 
**                                                 - 

Hematite  +  Liquid         ^x^^sq 

1100 

-                                   ^r 

^v                     *»» 

jr      Pyroxene  +  Hematite 

^v,^* 

/^                   +  Liquid         ^* 

IOOO< 

Pyroxenes 

390±5° 

900 

- 

800 

1                               !                               1 

i                   i                   i                   i                   i                   i 

1400 
I39l.5« 


1300 


-1200 


-1100 


-1000 


-  900 


-  800 


Acmite  10 

Na20Fe203-4Si02 


20 


30 


40 


50 


60 


70 


80 


90 


Fig.  30. 


Weight  per  cent 
Equilibrium  diagram  for  the  join  acmite-diopside. 


Diopside 
CaO-MgO-2Si02 


GEOPHYSICAL   LABORATORY 


99 


1500 


1400 


1300 


1200 


1100 


1000 
990  ±5° 


900 


Liquid 


Carnegieite  +  Liquid 


Carnegieite  +  Hematite  +  Liquid 

■— O  0  o— —  —  —  — 


Carnegieite  +  Nepheline  +  Liquid 


Hematite  +  Liquid 


Nepheline  +  Hematite  +  Liquid 


Acmite  +  Nepheline 


I526±2° 
500 


1400 


1300 


200 


1100 


1000 


900 


Acmite  |0 

Na20Fe203-4Si02 


20 


30 


40 


50 


60 


70 


80 


Weight  per  cent 
Fig.  31.     Equilibrium  diagram  for  the  join  acmite-nepheline. 


90       Nepheline 
Na20-AI203-2Si02 


and  liquid.  In  the  join  acmite-diopside  all 
compositions  with  40  per  cent  or  more 
acmite  melt  incongruently,  and  the  join 
is  not  binary.  Attention  is  called  to  the 
fact  that  some  of  the  iron  is  always 
present  as  ferrous  iron,  although  most  of 
the  iron  in  these  melts  is  ferric  iron. 
Therefore,  the  system  is  never  truly 
binary  even  in  the  portion  richer  in 
diopside  than  40  per  cent,  and  there  is 
always  a  small  amount  of  liquid  (glass) 
present  in  the  region  labeled  pyroxenes. 
The  results  of  a  study  of  the  join 
acmite-nepheline  are  given  here  in  figure 
31.  A  very  wide  primary  field  of  hematite 
appears  on  the  liquidus  surface  as  a  result 
of  the  incongruent  melting  of  acmite. 
The  system  is  not  binary.  The  phases 
present  are  acmite,  hematite,  carnegieite, 
nepheline,  and  liquid.  There  is  a  narrow 
region  of  coexistence  of  nepheline  and 


carnegieite,  suggesting  that  they  are  solid 
solutions  with  a  narrow  range  of  compo- 
sitions. Hematite  present  in  the  melts 
varies  in  color  from  deep  reddish  brown 
to  pale  brown,  suggesting  differences  in 
composition  perhaps  due  to  the  presence 
of  A1203  in  solid  solution  in  the  hematite. 
The  precise  compositions  of  solid  solu- 
tions have  not  been  determined.  Mixtures 
of  acmite  and  nepheline  begin  to  melt  at 
about  908°C. 

Studies  of  the  join  acmite-diopside- 
nepheline  in  progress  at  Tohoku  Univer- 
sity are  nearly  completed.  They  will  be 
presented  during  the  next  year.  Most  of 
the  studies  on  acmite-diopside  and 
acmite-nepheline  were  made  at  Tohoku 
University,  but  some  of  the  quenching 
experiments  were  run  at  the  Geophysical 
Laboratory  in  December  1960  and  Jan- 
uary 1961. 


100 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Accessory  Minerals 

Investigations  in  the  System 
FeO-FetOrTiO* 

D.  H.  Lindsley 

The  system  FeO-Fe203-Ti02  contains 
several  phases  of  geologic  interest  (fig. 
32).  In  addition  to  the  Ti02  minerals 
rutile,  anatase,  and  brookite,  the  im- 
portant phases  are: 

1.  The  rhombohedral  hematite-ilmen- 
ite  (aFe203-FeTi03)  series,  with  com- 
plete solid  solution  above  ^950°C  (Car- 
michael,  1961),  referred  to  as  the  a  series 
by  Verhoogen  (1962). 

2.  The  cubic  magnetite-ulvospinel 
series,  lying  on  the  binary  Fe304-Fe2Ti04 
join,  with  complete  solid  solution  above 
~600°C  (Vincent,  Wright,  Chevallier, 
and  Mathieu,  1957),  called  /3  series  or  /S 
spinels  for  convenience. 


3.  The  cation-deficient  spinels  that  lie 
on  the  Fe203  and  Ti02  side  of  the 
Fe304-Fe2Ti04  join,  called  7  spinels  by 
analogy  with  maghemite  (7Fe203). 

4.  The  pseudobrookite  series,  between 
Fe2Ti05  (pseudobrookite  proper)  and 
FeTi206,  an  unnamed  end  member  not 
known  to  occur  in  nature.  Complete  solid 
solution  in  this  series  is  found  above 
1150°C  (Akimoto,  Nagata,  and  Katsura, 
1957).  The  term  7  spinel  is  used  only  for 
convenience  in  reference,  and  does  not 
imply  existence  of  the  hypothetical 
7FeTi03  end  member  that  has  been 
postulated  for  these  spinels  (e.g.,  Nicholls, 
1955).  Inasmuch  as  these  spinels  appa- 
rently form  by  oxidation  of  0  spinels,  the 
concept  of  cation  deficiency  is  more  useful 
and  more  valid  than  that  of  7FeTi03 
solid  solution. 

Magnetite-ilmenite  relations.  Magnetite 
grains   containing   lamellae   of   ilmenite 


Ti02 
(Rutile,  Anatase, Brookite) 


FeTi205 


FeTiO 
(Ilmenite), 


FeO 


Fe2Ti05 
Pseudobrookite) 


FegTiO^ 
(Ulvospinel! 


(Wiistite) 


Fe304 
(Magnetite) 

Mol  per  cent 


(Hematite) 
(Maghemite) 


Fig.  32.  Phases  in  the  system  FeO-Fe203-Ti02.  Temperatures  of  complete  solid  solution  (heavy 
lines)  in  the  /3  series,  a  series,  and  pseudobrookite  series  are  approximately  600°,  950°,  and  1150°C, 
respectively.  The  join  magnetite-rutile  (dashed  line)  is  found  at  low  temperatures. 


GEOPHYSICAL   LABORATORY 


101 


oriented  in  the  (111)  planes  of  the  host 
are  found  in  a  variety  of  rocks  and  ores. 
The  ilmenite  lamellae  have  been  widely- 
interpreted  as  due  to  exsolution  from 
original  ilmenite-magnetite  solid  solu- 
tions. From  a  study  of  natural  specimens 
Ramdohr  (1955)  concluded  that  the 
parental  phase  could  contain  up  to  50 
mole  per  cent  ilmenite.  Several  workers 
have  noted,  however,  that  ilmenite- 
magnetite  intergrowths  cannot  be  homog- 
enized by  heating  to  1000°  to  1200°C  if 
bulk  composition  is  maintained;  it  has 
therefore  been  suggested  that  the  original 
phase  contained  ulvospinel  rather  than 
ilmenite  in  solid  solution  and  that 
ilmenite  is  formed  by  oxidation  of 
ulvospinel.  Ramdohr  (1953)  pointed  out 
that  ulvospinel  exsolves  in  the  (100) 
planes  of  magnetite  and  that  ilmenite 
formed  by  oxidation  of  such  ulvospinel 
lamellae  has  oblique  extinction — a  mode 
of  occurrence  very  different  from  that  of 
most  ilmenite-magnetite  intergrowths. 
Several  workers  have  also  suggested  that 
primary  magnetite-ulvospinel  solid  solu- 
tions may  be  oxidized  to  y  spinels,  which 
then  break  down  to  ilmenite-magnetite 
intergrowths.  The  oxidation  hypothesis 
has  gained  added  support  from  recent 
experimental  data  (Webster  and  Bright, 
1961;  R.  Taylor,  unpublished  Ph.D. 
thesis  at  Pennsylvania  State  University) 
and  from  theoretical  considerations  (Ver- 
hoogen,  1962),  which  indicate  that  the 
stable  solubility  of  ilmenite  in  magnetite 
even  at  1200°  to  1300°C  is  much  too  small 
to  explain  observed  amounts  of  ilmenite 
in  natural  ilmenite-magnetite  inter- 
growths.2 Results  of  the  current  investi- 

2  The  extensive  solubility  of  ilmenite  in 
magnetite  reported  by  Schmahl,  Frisch,  and 
Hartgartner  (1960)  at  1000°C  is  questionable 
because  their  experimental  method  could  not 
distinguish  between  mixtures  and  solid  solutions 
of  ilmenite  and  magnetite,  and  no  X-ray  or 
optical  observations  were  reported.  The  limits  of 
solid  solution  shown  in  their  phase  diagram  are 
based  on  Ramdohr's  estimates  from  natural 
occurrences  and  hence  cannot  be  used  as  inde- 
pendent evidence  on  the  extent  of  natural  solid 
solutions. 


gation  support  a  variant  of  the  oxidation 
hypothesis  and  show  that  natural  tex- 
tures as  well  as  assemblages  can  be 
explained  by  that  hypothesis. 

Reagents  used  were  Fisher  " certified' ' 
Fe2C>3  and  Ti02,  and  United  Mineral  and 
Chemical  Corporation  99.999  per  cent  Fe 
sponge.  Before  weighing,  the  Fe  sponge 
was  analyzed  for  02,  and  appropriate 
corrections  were  made  in  the  proportions 
of  Fe  and  Fe203.  Material  was  mixed  by 
grinding  under  acetone  or  toluene  to 
inhibit  further  oxidation.  Single-phase 
starting  materials  for  hydrothermal  ex- 
periments were  synthesized  from  the 
mixes  by  heating  at  1000°C  in  evacuated 
silica  glass  tubes  or  at  1200°C  in  Alundum 
crucibles  in  a  controlled  atmosphere  of 
N2  +  H2.  Homogeneity  was  checked  by 
optical  and  X-ray  examination. 

Stability  relations  were  determined  by 
the  hydrothermal  buffer  technique  of 
Eugster,  using  the  buffers  wlistite-mag- 
netite  ( WM) ,  fayalite-magnetite-quartz 
(FMQ),  nickel-nickel  oxide  (NNO),  and 
magnetite-hematite  (MH)  to  control 
oxygen  fugacity  (/o2).  Oxygen  fugacities 
of  these  buffers  as  functions  of  tempera- 
ture and  total  pressure  can  be  derived 
from  the  expressions  given  in  table  3. 
Alloying  of  Fe  from  the  charge  with  Pt 
containers  at  low  oxygen  fugacities  was  a 
problem  in  early  hydrothermal  buffer 
experiments.  Wrapping  the  charge  in  Ag 
foil  or  using  Ag  instead  of  Pt  containers 
prevents  Fe  loss  but  introduces  additional 
disadvantages.  A.  Muan  (unpublished 
data  presented  in  a  Penologists'  Club 
lecture)  has  shown  that  Ag-rich  Ag-Pd 
alloys  have  melting  points  higher  than 
that  of  pure  Ag,  but  are  still  almost 
immiscible  with  Fe.  The  high  permea- 
bility of  Ag-Pd  alloys  to  hydrogen  makes 
them  ideal  as  charge  containers  for 
buffered  hydrothermal  experiments.  Most 
of  the  data  here  presented  were  obtained 
from  runs  made  in  Ag7oPd3o  (weight  per 
cent)  containers.  Temperatures  were 
measured  to  ±2°C  and  regulated  to 
d=2°C.  At  and  below  800°C  most  runs 
were  made  at  2  kb  total  pressure;  at 


102 


CAKNEGIE     INSTITUTION     OF      WASHINGTON 


TABLE  3.     Calculations  of  Oxygen  Fugacities  of  Buffers  as  Functions  of  Temperature 

and  Total  Pressure 

T  in  °K;/o2  and  Ptot  in  bars.  From  Eugster  and  Wones,  1962. 

T  ,  A      ,      „      ,     „  (Ptot    -    1). 


^vgjoz    —    - 

T 

T  J->   T  ^ 

T 

Buffer 

A 

B 

C 

Wiistite-magnetite  (WM) 
Fayalite-magnetite-quartz  (FMQ) 
Nickel-nickel  oxide  (NNO) 
Magnetite-hematite  (MH) 

32,730 
27,619 
24,709 
24,912 

13.12 
10.55 

8.94 
14.41 

0.083 
0.092 
0.046 
0.019 

higher  temperatures,  lower  pressures 
(usually  1  kb)  were  used  to  protect  the 
pressure  vessels.  All  compositions  are 
given  in  mole  per  cent  unless  otherwise 
noted. 

A  series  of  standard  /3  spinels  ranging 
from  Mtioo  to  Mt20Usp8o  were  synthesized 
hydrotherapy  at  800°C  using  the  WM 
buffer,  each  composition  being  made  in 
duplicate  or  triplicate.  Ten  measure- 
ments of  the  (333,  511)  peak,  using 
internal  standards  of  CaF2  or  Si  with  Fe 
radiation  on  a  Phillips  powder  X-ray 
diffractometer,  were  averaged  for  each 
sample.  Pure  ulvospinel,  which  is  not 
stable  at  the  lowest  /o2  attainable  with 
available  buffers,  was  synthesized  at 
1200°C  in  a  controlled  atmosphere  of  N2 
and  H2.  Because  the  (311)  peaks  of  CaF2 
and  Si  interfere  with  the  (333,  511)  peak 
of  ulvospinel,  quartz  was  used  as  an 
internal  standard.  Comparison  of  results 
obtained  with  samples  of  intermediate 
compositions  using  CaF2,  Si,  and  quartz 
standards  showed  no  detectable  differ- 
ences in  20  values.  The  26  (Fe  KaJ  versus 
composition  data  were  plotted  for  use  as 
a  determinative  curve.  Corresponding 
unit-cell  edges  are  given  in  figure  33,  with 
the  data  of  Akimoto,  Katsura,  and 
Yoshida  (1957)  for  comparison.  Repeated 
measurements  of  both  standards  and 
unknown  specimens  showed  a  repro- 
ducibility of  ±0.01°  26;  this  internal 
consistency  permits  determination  of 
compositions  to  at  least  d=2  mole  per  cent 
despite  any  systematic  errors  that  might 


affect  the  absolute  accuracy  of  the  X-ray 
data.  Unit-cell  edges  are  believed  to  be 
accurate  to  ±0.001  A. 

Data  are  now  available  on  the  compo- 
sitions of  titaniferous  magnetite  in  equi- 
librium with  ilmenite  for  the  buffers 
NNO,  FMQ,  and  WM.  Starting  materials 
were  either  /3  spinels  alone  or  /?  spinels  + 
ilmenite.  The  following  reactions  take 
place  during  buffered  runs : 

Ulvospinel-rich  /3  spinel  +  02 

=  Fe-Ti  spinel  +  ilmS8  (1) 

Magnetite-rich  /3  spinel  +  ilm 

=  Fe-Ti  spinel  +  ilmss  +  02         (2) 

(Ilmss  means  ilmenite  with  Fe203  in  solid 
solution;  for  the  buffers  NNO,  FMQ,  and 
WM  this  Fe203  content  is  less  than  15 
per  cent.  Fe-Ti  spinel  is  used  as  a  general 
term  to  indicate  either  /3  or  7  spinels.) 
For  each  temperature  and  oxygen  fugac- 
ity  the  unit-cell  edges  of  the  Fe-Ti 
spinels  formed  in  reactions  1  and  2  are 
nearly  identical  and  are  always  inter- 
mediate between  those  of  the  initial  /3 
spinels,  suggesting  that  the  Fe-Ti  spinels 
have  approached  the  composition  of  the 
spinel  that  is  in  equilibrium  with  ilmen- 
itess.  However,  the  unit-cell  edges  alone 
cannot  yield  unique  compositions  if  the 
product  spinels  are  7  phases,  which  they 
must  be  if  there  is  solid  solution  of 
ilmenite  in  the  spinel.  Akimoto,  Katsura, 
and  Yoshida  (1957)  have  determined  cell 
edges  of  0  and  7  spinels;  their  data 
permit     determination    of    composition 


GEOPHYSICAL   LABORATORY 


103 


from  the  cell  edge  if  the  Fe/Ti  ratio  of 
the  spinel  is  known.  The  presence  of 
ilmenite  in  a  run,  however,  prevents 
identification  of  the  Fe/Ti  ratio  of  the 
product  spinel  with  the  known  ratio  of 
the  entire  charge.  Chemical  analysis  of 
the  spinel  is  likewise  not  feasible  when 
ilmenite  is  present.  The  amount  of 
ilmenite  in  solid  solution  with  the  Fe-Ti 
spinels  of  reactions  1  and  2  must  therefore 
be  determined  indirectly. 

The  equilibrium  compositions  of  the 
products  of  reactions  1  and  2  are  uniquely 
fixed  at  constant  pressure  by  the  tempera- 
ture and  oxygen  fugacity;  the  relative 
amounts  of  the  products  are  determined 
by  the  bulk  composition  (or  by  the  Fe/Ti 
ratio  in  f0z  buffered  systems)  of  the 
starting  materials.  For  each  fixed  P,  T, 


and  /o2  there  must  exist  a  bulk  compo- 
sition for  which  the  Fe-Ti  spinel  formed 
will  coexist  with  an  infinitesimal  amount 
of  ilmenitess.  That  critical  composition 
can  be  estimated  for  reaction  1  by  making 
a  series  of  runs  in  which  the  Fe2Ti04 
content  of  the  starting  /3  spinel  is  suc- 
cessively lowered  until  no  ilmenite  is 
found  in  the  products.  The  composition 
of  the  spinel  thus  formed  can  be  deter- 
mined from  the  unit-cell  edge  and  by 
chemical  analysis.  No  differences  in  unit- 
cell  edge  were  detected  before  and  after 
runs  at  several  critical  compositions, 
indicating  within  the  accuracy  of  the 
X-ray  data  that  the  product  spinel  is 
essentially  a  binary  p  spinel  of  the 
starting  composition. 

One    composition    indicated    by    the 


8.54 


8.52 


8.50 


o<    8.48 


o 

"S   8.46 


5  8.44 


8.42 


8.40 


0 

Magnetite 
Fe304 


a   This  Investigation 
•    Akimoto  ei  at,  1957 


Mol  per  cent 


100 
Ulvbspinel 
Fe2Ti04 


Fig.  33.  Plot  of  composition  versus  unit-cell  edge  for  magnetite-ulvospinel  solid  solutions  (/3 
spinels).  Vertical  extent  of  data  points  shows  uncertainty  in  cell-edge  determinations;  horizontal 
extent  not  significant.  Compositions  Mtioo  to  Mt2oUsp8o  made  hydrothermally  at  800°C;  pure 
ulvospinel  made  at  1200°C  in  N2  +  H2  mixture.  Data  of  Akimoto,  Katsura,  and  Yoshida  (1957) 
from  samples  made  in  evacuated  silica  glass  tubes  are  shown  for  comparison. 


104 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


X-ray  data  was  checked  as  follows.  At  Fe-Ti  spinels  in  equilibrium  with  ilmenite 

960°C  and  the  /o2  of  the  NNO  buffer,  the  can   be   determined   from   the   unit-cell 

spinel  in  equilibrium  with  ilmenitess  is  edges. 

Mt52Usp48  by  X-ray  determination.  Runs  The  available  data  indicate  negligible 

were  made  on  spinels  of  that  composition  solid  solution  of  ilmenite  in  magnetite, 

at  960°C  using  the  NNO,  FMQ,  and  WM  That  ilmenite-magnetite  intergrowths  (as 

buffers,   and   at   800°C  using  the   WM  well  as  ilmenite-magnetite  assemblages) 

buffer.  Part  of  each  sample  was  removed  can  be  formed  by  reaction  1  is  shown  in 

for  X-ray  and  optical  examination,  and  figure   34,  plate  1.  Pure  ulvospinel  was 


the  remainder,  about  100  mg,  was  dried 
for  16  hours  at  200°C  in  N2  for  chemical 
analysis.  No  ilmenite  was  detected  opti- 
cally or  by  X  ray.  Ferrous  iron  was 
determined  by  the  modified  Pratt  method 
(table  4).  The  determinations  of  FeO  are 


held  at  1000°C  at  the  f0z  of  the  Stellite 
bomb  (roughly  equal  to  that  of  the 
NNO  buffer)  for  3  hours.  The  resulting 
texture — lamellae  of  ilmenitess  in  the 
(111)  planes  of  a  Mt5oUsp5o  solid  solu- 
tion— closely  resembles  natural  textures. 


TABLE  4.     FeO  Contents  and  Unit-Cell  Edges  of  Some  Fe-Ti  Spinels  in  Equilibrium 

with  Ilmenites3 


Unit-Cell 

Weight 

Mole 

Run  No. 

T,  °C 

Buffer 

Edge, 
A 

Per  Cent 
FeO 

Per  Cent 
FeO 

L333 

800 

WM 

8.462 

46.2 

59.3 

L334 

960 

FMQ 

8.462 

45.6 

58.8 

L335 

960 

WM 

8.462 

45.0 

58.3 

L336 

960 

NNO 

8.462 

45.3 

58.5 

Theoretical  FeO  for  Mt£ 

2Usp48 

46.96 

59.68 

minimum  values,  as  any  alloying  of  Fe  Reactions   1  and  2,  carried  out  at  a 

with   the   charge   container,   incomplete  series  of  temperatures  for  each  of  several 

drying  of  the  sample  before  weighing,  or  buffers,  permit  us  to  bracket  the  compo- 

oxidation  of  the  solution  before  titration  sitions  of  /3  spinels  that  are  in  equilibrium 

would  tend  to  reduce  the  value  for  FeO.  with  a  phases.  Data  obtained  using  the 

Evidently  the  spinel  in  equilibrium  with  buffers    NNO,    FMQ,    and    WM    are 

ilmenitess  at  960°C  and  the  foz  of  the  presented  in  figure  35;  in  all  runs  with 

NNO  buffer  deviates  from  a  binary  /3  these  buffers  the  coexisting  a  phases  are 

spinel  by  approximately  1  mole  per  cent  ilmenite  rich.  Exact  compositions  of  the 

FeO,  a  deviation  that  is  not  detected  by  a  phases  are  now  being  determined.  Runs 

the  X-ray  method  used.  The  maximum  are  also  being  made  using  the  buffers  MH 

deviation  of  the  Mt52Usp48  composition  and  MnO-Mn304.  Figure  35  is  simply  a 

from  the  binary  join  is  less  than  0.5  mole  graphical  representation  of  the  compo- 

per  cent  FeO  at  800°C  with  the  WM  sitions   of   /3   spinels   that   coexist   with 


buffer  and  should  be  further  lowered  with 
decreasing  temperature.  The  apparently 
stoichiometric  binary  compositions  indi- 
cated by  X-ray  data  for  other  spinels  in 


ilmenitess  for  given  temperatures  and 
buffers;  it  is  not  a  phase  diagram, 
inasmuch  as  the  compositions  of  the 
ilmenites  are  not  represented.  For  a  given 


equilibrium  with  ilmenite  will  be  checked  temperature  and  buffer  fo2,  a  spinel  of  a 

by  similar  analysis.  It  seems  justified  to  composition  to  the  right  of  the  appropri- 

conclude  that  any  deviation  is  small  and  ate  curve  will  break  down  by  reaction  1 

that    the    approximate    composition    of  to  yield   two  phases:  ilmenitess  plus  a 


GEOPHYSICAL   LABORATORY 


105 


100 


1000 


500 


Reaction  2  Reaction  I 

NNO     ►  < 

FMQ    >  <3 

WM    ►  *m 


NNO  FMQ 


WM 


o 

Magnetite 
Fe304 


20 


40 


60 


80 


Mol  per  cent 


100 
Ulvbspinel 
Fe2Ti04 


Fig.  35.  Composition  of  /3  spinel  in  equilibrium  with  ilmenite88  as  a  function  of  temperature  and 
the  oxygen  fugacities  of  three  buffers.  NNO,  nickel-nickel  oxide  buffer;  FMQ,  fayalite-magnetite- 
quartz  buffer;  WM,  wustite-magnetite  buffer.  Reaction  1,  ulvospinel-rich  /3  spinel  +  02  =  Fe-Ti 
spinel  +  ilmenite88.  Reaction  2,  magnetite-rich  /3  spinel  +  ilmenite  =  Fe-Ti  spinel  +  ilmenite88  +  02. 
Point  A  is  discussed  in  the  text. 


spinel  whose  composition  is  indicated  by 
the  curve.  Any  spinel  to  the  left  of  the 
curve  is  stable  by  itself  but  in  the  presence 
of  ilmenite  will  form  by  reaction  2  the 
spinel  indicated  by  the  curve.  For 
example,  at  1000°C,  the  spinel  Usp55 
(point  A  in  fig.  35)  is  unstable  at  the  /o2 
of  the  NNO  buffer  and  will  oxidize  to 
Usp5i  +  ilmeniteS8.  The  same  spinel  (A) 
with  the  FMQ  buffer  is  stable  down  to 
910°C;  at  successively  lower  temperatures 
spinel  A  is  unstable  and  breaks  down  to 
spinels  successively  richer  in  magnetite, 
plus  ilmenitess.  At  the  lower  oxygen 
fugacities  of  the  WM  buffer,  spinel  A 
would  remain  stable  upon  cooling  until 
the  magnetite-ulvospinel  solvus  is  reached 
(^500°  to  550°C  according  to  the  data  of 
Vincent,  Wright,  Che vallier,  and  Mathieu, 
1957). 

It  is  well  to  recall  that  the  curves  in 


figure  35  are  drawn  not  at  constant 
oxygen  fugacity  but  at  the  fugacity  of 
each  buffer,  which  varies  with  tempera- 
ture (see,  for  example,  Eugster  and  Wones, 
1962).  The  data  are  nevertheless  sufficient 
to  establish  the  principle  that  the 
composition  of  (3  spinel  in  equilibrium 
with  ilmeniteS3  is  strongly  dependent  on 
oxygen  fugacity  as  well  as  on  tempera- 
ture. 

Application  of  experimental  data  to 
natural  minerals.  Available  experimental 
data  in  the  ternary  system  FeO-Fe203- 
Ti02  indicate  that  the  stable  solubility  of 
ilmenite  (=  cation  deficiency)  in  Fe-Ti 
spinels  is  small  at  and  below  magmatic 
temperatures,  although  metastable  cation- 
deficient  Fe-Ti  spinels  are  easily  made  by 
oxidation  of  /?  spinels  in  air  at  400°  to 
550°C.  It  thus  seems  likely  that  natural 
cation-deficient   Fe-Ti   spinels    ("titano- 


106 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


maghemites")  form  metastably  by  oxi- 
dation at  moderate  temperatures,  say 
below  600°C.  (However,  many  natural 
7  spinels  contain  minor  amounts  of  Mg, 
Mn,  Al,  V,  and  Cr,  and  the  experimental 
data  from  the  pure  synthetic  system 
cannot  rule  out  the  possibility  that  the 
presence  of  these  elements  might  stabilize 
the  cation-deficient  structure.)  As  cation- 
deficient  spinels  are  less  dense  than  the 
equivalent  assemblages  /?  spinel  +  a 
phase,  high  pressure  should  inhibit  their 
formation  in  plutonic  rocks. 

The  experimental  data  presented  here 
do  not  disprove  the  theory  that  ilmenite- 
magnetite  intergrowths  form  by  exsolu- 
tion  from  a  primary  ilmenite-magnetite 
solid  solution,  but  they  strongly  support 
the  alternative  hypothesis  that  such 
intergrowths  result  from  the  oxidation  of 
magnetite-ulvospinel  solid  solutions.  An 
intermediate  7  phase  may  form  in  some 
volcanic  and  hypabyssal  rocks,  but  direct 
oxidation  to  magnetite-rich  /3  spinel  + 
ilmeniteS8  seems  likely  in  plutonic  rocks. 
In  rocks  where  oxygen  fugacity  remains 
sufficiently  low  upon  cooling,  little  or  no 
ilmenite  will  form  and  ulvospinel-magne- 
tite  intergrowths  may  result. 

Relations  between  ilmenite,  hematite, 
magnetite,  and  rutile.  Attempts  to  deter- 
mine the  hematite-ilmenite  solvus  hydro- 
thermally  have  been  unsuccessful  because 
no  buffer  is  available  with  an  oxygen 
fugacity  at  which  hematitess  and  ilmen- 
itess  can  coexist.  Initial  compositions  of 
Hem50Ilm5o  are  oxidized  to  hematitess  + 
pseudobrookitess  (or  hematitess  +  rutile) 
by  the  MH  buffer,  and  are  reduced  to 
magnetitess  +  ilmenitess  by  the  NNO 
buffer.  If  both  hematitess  and  ilmeniteS8 
can  coexist  at  equilibrium  for  a  given 
temperature,  the/o2  ranges  at  which  each 
is  stable  must  overlap ;  the  zone  of  overlap 
must  lie  between  the  oxygen  fugacities  of 
the  MH  and  NNO  buffers.  As  the 
compositions  of  the  coexisting  hematiteS8 
and  ilmenitess  move  farther  apart  upon 
cooling,  the  corresponding  f0z  range  of 
mutual  stability  probably  decreases.  It 
is  possible  that  at  low  temperatures  (say 


below  200°  to  400°C)  there  is  no  f0z  at 
which  both  hematitess  and  ilmeniteas  can 
coexist  at  equilibrium.  Under  this  hypoth- 
esis hematite-ilmenite  intergrowths  ex- 
solved  at  low  temperatures  are  meta- 
stable.  They  may  form  because  less 
energy  is  required  for  migration  of  Fe 
and  Ti  in  the  inherited  oxygen  framework 
of  the  original  phase  than  for  complete 
reorganization  into  new  phases  such  as 
magnetite  plus  rutile.  In  this  regard  it  is 
significant  that  many  low-grade  meta- 
morphic  rocks  contain  the  assemblage 
magnetite  +  rutile,  which  is  chemically 
equivalent  to  hematite  +  ilmenite.  The 
reaction 

Fe304  +  Ti02  =  Fe203  +  FeTi03    (3) 

has  a  small  positive  free  energy,  AG(S)  = 
+  1  to  +2  kcal,  over  the  temperature 
range  100°  to  1200°C,  according  to  the 
best  available  data.  A(?(3)  is  probably 
smaller  than  the  uncertainty  involved  in 
its  derivation;  if,  however,  its  sign  is 
correct  it  accounts  for  the  assemblage 
magnetite  +  rutile  in  low-grade  meta- 
morphic  rocks.  At  higher  temperatures 
the  free  energy  of  mixing  of  Fe203 
becomes  sufficient  to  favor  ilmenite- 
hematite  solid  solutions  over  the  magnet- 
ite +  rutile  assemblage. 

Stability  Relations  of  Dravite, 
a  Tourmaline 

C.  R.  Robbins3  and  H.  S.  Yoder,  Jr. 

The  most  abundant  and  geochemically 
most  important  of  the  boron  minerals  are 
the  tourmalines.  They  are  found  in  a 
variety  of  igneous,  metamorphic,  and 
sedimentary  rocks  of  all  ages.  Their 
authigenic  formation  at  low  temperatures 
in  some  limestones  and  sandstones  is  of 
particular  interest. 

Chemically,  tourmalines  are  complex 
borosilicates  of  variable  composition,  the 
variation  resulting  from  the  large  number 
of  substitutions  permitted  by  the  struc- 
ture.   Preliminary    calculations    suggest 

3  U.  S.  National  Bureau  of  Standards. 


GEOPHYSICAL   LABORATORY 


107 


that  tourmalines  may  be  described  as 
isomorphous  mixtures  of  several  end 
members. 

A  number  of  the  tourmalines  have  been 
synthesized  from  rather  complex  systems, 
but  their  stability  relations  were  not 
determined.  The  objective  of  the  present 
study  is  the  determination  of  the  pressure- 
temperature  stability  range  of  a  tour- 
maline of  specific  composition,  the  iron- 
free  end  member  dravite,  NaMg3Al6B3- 
Si6027(OH)4.  This  composition  was  chosen 
because  the  crystal  structure  studies  of 
Hamburger  and  Buerger  (1948)  had 
established  the  ideal  formula.  It  was  also 
of  interest  to  relate  this  composition  to 
the  petrologically  important  system 
MgO-Al203-Si02-H20. 

For  this  work  a  glass  of  the  requisite 
anhydrous  composition  was  carefully 
prepared  in  several  steps  to  avoid  loss  of 


decomposition  products  are  cordierite, 
liquid,  gas,  and  a  crystalline  phase  that 
has  not  yet  been  identified.  Sporadic 
occurrences  of  trace  amounts  of  spinel, 
mullite  (?),  and,  once,  of  sapphirine  have 
been  observed  microscopically  in  the 
dissociation  products  of  both  the  natural 
tourmaline  and  the  glass.  These  dissoci- 
ation products  are  fine  grained  and 
frequently  occur  as  inclusions  in  the 
cordierite  or  glass.  They  may  well  be  the 
result  of  leaching. 

Above  895°C  and  5000  bars  both  glass 
and  natural  tourmaline  form  the  assem- 
blage kornerupine  +  sapphirine  +  liquid 
+  gas.  Previous  synthesis  of  kornerupine 
is  unknown  to  the  writers.  The  phase  is 
well  crystallized,  and  its  X-ray  pattern 
agrees  well  with  that  of  a  natural  mineral 
from  Kazebanza,  Quebec  (U.  S.  N.  M. 
no.  106.774). 


TABLE  5.     Comparison  of  Indices  of  Refraction  and  Unit-Cell  Dimensions  of  Synthetic 

Dravite  and  Dobruva,  Austria,  Tourmaline 

a,  A 

c,  A 

OJ 

€ 

Synthetic  dravite 
Dobruva  tourmaline 

15.93 
15.931 

7.18 
7.197 

1.632 
1.634 

1.610 
1.613 

Na20  and  B203.  In  addition,  a  natural 
tourmaline  from  Dobruva,  Carinthia, 
Austria  (U.  S.  N.  M.  no.  103.791),  was 
selected  for  comparative  studies.  A 
chemical  analysis  of  the  material  by 
H.  B.  Wiik  showed  it  to  be  exceptionally 
close  to  the  ideal  dravite  composition. 
Indices  of  refraction  and  unit-cell  dimen- 
sions of  synthetic  dravite  and  the 
Dobruva  tourmaline  are  given  in  table  5. 

Preliminary  results  of  this  study  at 
various  temperatures,  and  water  pres- 
sures up  to  5000  bars,  are  summarized  in 
figure  36.  The  part  of  the  curve  below 
500  bars  was  calculated  from  the  inte- 
grated Clausius-Clapeyron  equation.  It  is 
evident  that  dravite  is  stable  over  a  wide 
range  of  temperatures  and  pressures. 

At  temperatures  above  865°C  and 
pressures    up    to    2000    bars    the    main 


At  925°C  and  5000  bars  kornerupine  is 
no  longer  stable,  and  the  phases  coexisting 
in  equilibrium  are  sapphirine  +  liquid  + 
gas.  Inclusions  observed  in  the  sapphirine 
appear  to  be  minute  spinel  octahedra  and 
probably  result  from  leaching. 

Although  these  results  are  only  pre- 
liminary, it  would  appear  that  they  will 
have  application  to  natural  occurrences, 
since  the  associations  tourmaline  +  cor- 
dierite, tourmaline  +  kornerupine,  and 
cordierite  +  kornerupine  +  sapphirine 
(Ussing,  1889)  are  known. 

Mantle  Minerals 

F.  R.  Boyd,  Jr.,  and  J.  L.  England 

The  discovery  that  pressures  as  low  as 
5  kb  cause  enstatite  to  melt  congruently 
(Boyd  and  England,  Year  Book  60)  raises 


108 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


5000 


4500 


4000 


3500 


3000 


2500- 


2000 


1500 


1000 


500 


i 1 1 1 — i — i — i — I — i — rr 


i — • — r 


i — ■ — r 


-i 1 r 


ORAVITE 

+ 

GAS 


XXX  X 


*  i  • 


VARIOUS 
CRYSTALS 


+ 
LIQUID 


GAS 


*  B  9      » 


--1 tJ L 


400  500  600  700  800  900  1000  1100  1200  1300  1400 

Temperature  ,°C 
Fig.  36.     Preliminary  P-T  diagram  of  the  system  dravite-water. 


a  problem  in  accounting  for  the  formation 
of  basaltic  magma  that  is  oversaturated 
in  silica.  Basalts  in  the  Pacific  Ocean 
basin  are  sometimes  considerably  over- 
saturated  in  silica.  For  example,  the 
degree  of  oversaturation  of  the  primitive 
shield  basalts  of  Hawaii  ranges  up  to 
about  6  weight  per  cent  Si02  (Powers, 
1955,  p.  81).  Nevertheless,  there  is 
considerable  evidence  to  indicate  that  the 
mantle  rocks  from  which  these  basalts 
were  derived  contain  olivine.  The  prin- 
cipal minerals  in  rocks  in  the  upper 
mantle  are  probably  enstatite,  diopsidic 
pyroxene,  olivine,  and  py rope-rich  garnet. 
It  is  impossible  to  derive  a  liquid 
oversaturated  in  silica  by  partial  fusion 
of  a  mixture  of  pyroxene,  olivine,  and 
garnet  in  the  absence  of  an  incongruent 
melting  reaction.  For  many  years  it  was 
thought  that  the  incongruent  melting  of 
enstatite  found  by  Bowen  and  Andersen 
(1914)  at  atmospheric  pressure  provided 


a  mechanism  for  generating  oversaturated 
liquids  from  olivine-bearing  parent  rocks 
at  depth.  High-pressure  data,  however, 
indicate  that  this  reaction  must  be 
restricted  to  relatively  shallow  depths  in 
the  crust.  Estimates  of  the  depth  of 
formation  of  basaltic  lava  in  the  ocean 
basins  range  from  50  to  100  km  where  the 
pressure  is  in  the  range  15  to  30  kb.  Since 
experimental  data  have  shown  that 
enstatite  melts  congruently  in  this  range, 
an  alternative  explanation  for  the  compo- 
sition of  these  oversaturated  basalts  must 
be  sought. 

A  possible  alternative  is  that  an 
incongruent  melting  reaction  involving 
garnet  is  effective  at  depth  in  the  mantle. 
Boyd  and  England  suggested  in  Year 
Book  60  that  pure  pyrope  must  melt 
incongruently  and  that  this  reaction 
might  also  be  present  in  more  complex 
natural  melts.  The  melting  relations  of 
pyrope   have   been   restudied   with   im- 


GEOPHYSICAL   LABORATORY  109 

proved  techniques,  and  it  has  been  found  boundary  (A  in  fig.  37)  is  essentially  the 

that  pyrope  does  melt  incongruently  over  same  as  that  given  in  our  preliminary 

a  wide  P-T  range.  As  is  described  here-  diagram  in  Year  Book  58  except  that  no 

after  there  are  at  least  three  incongruent  friction  correction  has  been  made  for  the 

melting    reactions    for    pyrope    in    the  present   results.    Experience   has   shown 

pressure  range  25  to  36  kb.  Above  36  kb  that  the  friction  in  single-stage  runs  at 

pyrope  melts  congruently.  high  temperature  is  less  than  was  initially 

The     principal     incongruent    melting  estimated.   The  pressure  on  the  run  is 

reaction  of  pyrope  is  to  spinel  +  liquid,  believed  to  be  within  d=5  per  cent  of  the 

Since  spinel  contains  no  silica,  the  liquid  load    pressure.    The    results    shown    for 

that  forms  in  the  incongruent  interval  temperatures  above  1500°C  are  new  and 

contains  more  silica  than  pyrope  compo-  were  obtained  by  techniques  developed 

sition  does.  In  the  melting  of  pure  pyrope  in  a  study  of  the  melting  curves  of  albite 

the  composition  of  the  liquid  lies  in  the  and  diopside.  Details  of  these  techniques 

three-phase  field  pyrope  +  Al-enstatite  +  have  been  published  recently  (Boyd  and 

quartz.  If  this  incongruent  melting  reac-  England,  1962). 

tion    were    present    in    more    complex  Liquid  of  pyrope  composition  cannot 

natural   systems,    oversaturated   basalts  be  quenched  to  a  glass  over  most  of  the 

could    theoretically    be    generated    by  investigated  P-T  range.  Recognition  of 

fractionation    of    liquid    from    a    partly  the  various  melting  reactions,  therefore, 

melted    garnet    peridotite.    By    analogy  depends  on  textural  differences  in  the 

with  the  melting  relations  for  pure  pyrope  runs.  Fortunately,  pyrope  itself  will  not 

it  would  be  expected  that  the  incongruent  form  in  the  quench.  Runs  quenched  from 

melting  would  be  present  over  a  restricted  the   fields   above   curves   B,   C,   and    D 

depth  range.   Pyrope-rich  garnet  would  crystallize  in  the  quench  to  assemblages 

not  be  stable  in  the  melting  interval  at  consisting  wholly  or  largely  of  aluminous 

lower  pressures  and  would  melt  congru-  enstatite. 

ently  at  higher  pressures.  For  pure  pyrope  At  temperatures  below  curve  E  and  at 
the  depth  range  would  be  from  about  75  pressures  below  curve  A  pyrope  compo- 
to  105  km,  but  it  would  probably  be  sition  crystallizes  to  a  fine-grained  mix- 
shallower  in  a  natural  system.  ture  of  aluminous  enstatite,  sapphirine, 

The  phase  relations  determined  in  the  and  sillimanite.  Enstatite  and  sapphirine 

pyrope  study  are  a  further  demonstration  can  be   distinguished   in  X-ray  diffrac- 

of  the  importance  of  high  pressure  in  tometer  patterns  of  such  runs,  and  they 

modifying    crystal-liquid     equilibria    in  can  be  recognized  under  the  microscope, 

silicate  systems.  In  the  pressure  range  20  Aluminous  enstatite  forms  about  80  per 

to  40  kb  a  liquid  of  pyrope  composition  cent  of  the  products.   The  presence  of 

can    be    crystallized    by    at    least    five  sillimanite  is  known  from  the  study  of 

different  paths,  depending  on  the  pres-  other  compositions  in  the  system  MgO- 

sure.  It  is  not  certain  that  the  incongruent  Al203-Si02  in  this  P-T  range,   but  the 

melting  of  pyrope  is  a  significant  feature  amount  that  forms  in  a  run  on  pyrope 

of  magma  differentiation  at  depth  in  the  composition  is  too  small  to  detect  by 

mantle,  but  there  is  little  question  that  optical  or  X-ray  methods, 

pressure   will   prove    to   have    a   major  At    1615°C   and   24.5   kb   there   is   a 

influence  on  such  differentiation.  pronounced  break  in  the  boundary  of  the 

pyrope    stability    field.    This    break    is 

Effect  of  Pressure  on  the  Melting  of  Pyrope  beljwed  to  be  due  to  the  intersection  of  a 

melting  curve  in  the  breakdown  products 

Phase  relations  for  pyrope  composition  field  with  the  pyrope  subsolidus  boundary. 

in  the  pressure  range   15  to  50  kb  are  Most  likely,   though   not  certainly,   the 

shown    in    figure    37.     The    subsolidus  melting  curve  (E  in  fig.  37)  is  the  solidus 


110 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


curve  for  the  breakdown  products.  It  is 
known  that  aluminous  enstatite  is  a 
stable  phase  in  the  P-T  field  between 
curves  E  and  F,  but  it  is  not  known 
whether  E  marks  the  disappearance  of 
sapphirine  or  sillimanite.  Runs  quenched 
from  pressures  and  temperatures  imme- 
diately above  or  below  curve  E  have 
essentially  identical  X-ray  patterns  and 


are  indistinguishable  under  the  micro- 
scope. 

Curve  F  in  figure  37  is  the  liquidus 
curve  for  aluminous  enstatite.  In  runs 
quenched  from  below  curve  F  the  ensta- 
tite forms  a  fine-grained  mosaic  of 
crystals.  In  runs  quenched  from  above 
curve  F  the  enstatite  crystallizes  as 
coarse  blades  with  undulate  extinction 


i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i  i 


|  i  i  i  i  i  i  i  i  i  | 


i  i  i  i  i  i  i  i 


i  i  i  i  i  i  i 


1900 


1800 


1700 


1600 


O 

o 

a> 

■*- 
O 
i_ 
0> 
CL' 

£ 


1500 


1400 


SPINEL  +  LIQUID 


L,QUID      JU  -  x  / 
/  L---i^rK      x         D 


1300 


1200 


1100 


1000 


ALUMINOUS  ENSTATITE7    □/<  X 
+  ?+LIQUID 


ALUMINOUS  ENSTATITE 

+ 
SAPPHIRINE 

+  i    : 

SILLIMANITE 


PYROPE 
Mg3Al2Si30|2 


D        [DOC 


i  i  i 


J— l—L 


\    i    i    t    l 


I    I    I 


i    i    t    I    I    i    I    I    I    I    I    I    I    t    I 


I    I    I    I    


10 


20  30 

Load  pressurewkilobars 


40 


50 


Fig.  37.  The  stability  field  of  pyrope  garnet.  Synthetic  pyrope  was  used  as  starting  material  for 
all  runs  shown  except  for  the  two  runs  indicated  by  open  circles  on  curve  A.  In  these  two  runs  pyrope 
formed  from  seeded,  crystalline  breakdown  products.  The  dashed  curve,  H,  is  a  nucleation  boundary; 
pyrope  forms  readily  from  glass  or  crystalline  starting  materials  at  pressures  higher  than  curve  Hf 
but  it  will  not  nucleate  in  the  P-T  field  between  curves  A  and  H.  For  a  further  discussion  of  the 
nucleation  problem,  see  Boyd  and  England,  Year  Book  58. 


GEOPHYSICAL   LABORATORY  111 

under    crossed    nicols.    This    texture    is  should  react  to  form  some  other  phase, 

characteristic  of  enstatite  crystals  that  A  mixture  of  20  per  cent  crystalline 

have    formed    in   the    quench    in   high-  sapphirine   +  80  per  cent  pyrope  glass 

pressure  runs  (Boyd  and  England,  Year  and  a  mixture  of  24  per  cent  crystalline 

Book  60).  spinel  +  76  per  cent  crystalline  pyrope 

In  the  P-T  field  bounded  by  curves  F,  were  run  at  temperatures  midway  be- 

C,  and  G  the  quench  crystals  of  enstatite  tween  curves  F  and  G  at  the  pressures 

contain  scattered,  subhedral  grains  of  an  21.5  kb  and  28.7  kb.  The  products  of 

isotropic   phase  with  a  refractive  index  these  runs  looked  identical  with  those  of 

appreciably    higher    than    that    of    the  runs  made  in  this  P-T  field  on  pyrope 

enstatite  in  which  they  are  poikilitically  composition  except  that  the  concentra- 

enclosed.    These    isotropic    grains    have  tion    of    the    primary,    isotropic    phase 

rounded  to  rectangular  shapes  when  seen  poikilitically  enclosed  in  enstatite  quench 

under   the   microscope   and   are   1  to  5  crystals    was    greatly    increased.    X-ray 

microns  in  diameter.  The  texture  of  the  patterns  of  these  runs  showed  that  the 

runs  quenched  from  this  field  is  strikingly  products  obtained  with  both  mixes  were 

similar  to  that  of  the  granules  of  primary  spinel  +  enstatite.  Since  sapphirine  was 

forsterite  embedded  in  quench  enstatite  converted  to  spinel  in  these  experiments, 

found  in  runs  on  MgSi03  composition  the  primary  phase  in  the  field  bounded  by 

cooled  from  a  temperature  within  the  curves  F,  C,  and  G  is  proved  to  be  spinel, 

incongruent  melting  interval  at  atmos-  The  melting  relations  shown  in  figure 

pheric  pressure.   The  isotropic  granules  37    only   partly   define   the   melting   of 

decrease  in  abundance  as  the  temperature  pyrope  composition  in  the  pressure  range 

is  raised  in  the  interval  between  curves  F  15  to  29  kb.  There  must  be  at  least  one 

and  G.  Runs  quenched  from  above  curve  more  curve  than  is  shown  in  the  P-T 

G    contain    only    glass    and/or    quench  range  below  curve  F.   Quenching  diffi- 

crystals  of  enstatite  metastably  rich  in  culties  and  the  small  amounts  of  phases 

A1203.  other  than  aluminous  enstatite  that  are 

The  quantity  of  the  isotropic  primary  present  on  pyrope  composition  prevented 

phase  in  the  field  bounded  by  curves  F,  the   identification   and   location   of   this 

C,  and  G  is  insufficient  to  show  on  an  curve.  Study  of  the  melting  of  a  variety 

X-ray    diffractometer    pattern.    Optical  of   compositions   in   MgO-Al203-Si02   in 

properties  obtained  indicate  that  it  must  this  P-T  range  would  undoubtedly  clarify 

be    either    spinel    or    sapphirine.     The  the  picture,  but  the  tendency  of  these 

refractive  indices  of  spinel  and  sapphirine  compositions  to  crystallize  in  the  quench 

are  similar.  Sapphirine  has  a  low  bire-  remains  a  formidable  problem, 

fringence,    but   it    appears    isotropic    in  The  melting  curves  of  the  breakdown 

grains  only  a  few  microns  in  diameter.  A  products    (E,   F,   and    G)    intersect   the 

test  was  devised,  however,  that  showed  pyrope  stability  field  and  give  it  a  faceted 

the  primary  phase  to  be  spinel.  boundary.     The    principal    incongruent 

At  constant  temperature  and  pressure,  melting  reaction  is  to  spinel  +  liquid,  but 

changing  the  proportions   of  phases  in  at   least   two   other  reactions   in  which 

equilibrium  will  not  change  the  kinds  of  pyrope  melts  to  Al-enstatite  +  liquid  + 

phases    present    or    their    compositions,  other  crystalline  phases  must  be  present 

Hence,  if  the  primary  isotropic  phase  was  in  the  pressure  range  25  to  29  kb.  Above 

spinel,  it  would  be  possible  to  add  spinel  36   kb   the   melting   is   congruent.    The 

to  the  mixture  of  spinel  and  liquid  on  maximum  incongruent  melting  interval 

pyrope  composition  without  changing  the  at  constant  pressure  is  about  90°  at  25  kb 

phase  relations.  If  the  primary  phase  was  and  diminishes  as  the  pressure  is  raised 

spinel   and    sapphirine   was   added,    the  until    the    melting    becomes    congruent, 

sapphirine    would    not    be    stable    and  The    pressure   interval   over   which   the 


112  CARNEGIE     INSTITUTION     OF      WASHINGTON 

melting  is  incongruent  is   11  kb,  corre-  to  5. 5°/kb.  The  only  other  silicate  melting 

sponding  to  a  depth  interval  in  the  upper  curve  thus  far  determined  in  the  pressure 

mantle  of  about  30  km.  range  above  30  kb  is  diopside.  The  slope 

The  average  slope  of  the  pyrope  solidus  of  the  diopside  curve  in  the  range  35  to  50 

curves    (B   and   C)    in   the   incongruent  kb  is  6.9°/kb.  These  slopes  are  substan- 

melting  interval  between  25  and  36  kb  is  tially  less  than  was  earlier  estimated  for 

about  16.5°/kb.  Above  36  kb,  where  the  most    silicates    on    the    basis    of    data 

melting  is  congruent,  the  slope  decreases  obtained  in  a  lower  pressure  range. 


STATISTICAL  PETROGRAPHY 

o     •  7  •      D7              .    .    0  lected  by  H.  S.  Washington,  was  kindly 

banidine  Jrhenocrysts  in  borne  ,         ,JX            ,       ,,     &TT    ~    -*T  ^       i 

Peralkaline  Volcanic  Rocks  rJlleaSed  *?.  us  £  ^   U"   S:   Natlonal 

Museum.  I  he  other  three  specimens  were 

F.Chayes  and  E.G.Zieswith  X-ray  data  by  collected  by  Chayes,  in  company  with 

Professor  S.  Vardbasso  and  Dr.  A.  Atzeni, 

The  petrologist  often  uses  bulk  chem-  of  the  University  of  Cagiiari. 

ical  analysis  as  in  some  sense  a  substitute  The    Paris    de    Besa    trachyte    flow, 

for  modal  analysis,  and  recent  improve-  situated  about  5  km  west  of  the  town  of 

ments  in  modal  analysis  have  prompted  Ales    and    described    briefly   by    Atzeni 

a  revival  of  RosiwaFs  countersuggestion  (1959),  is  exposed  by  a  small  window 

that   chemical   composition   be   inferred  through  the  post-Miocene  basalts  on  the 

from  modes.  There  are  circumstances  in  southeastern  flank  of  Monte  Arci.  It  is  a 

which  the  first  procedure  is  unavoidable,  fine-grained  blue-gray  rock  studded  with 

and  there  are  also  circumstances  in  which  numerous  blocky  phenocrysts  of  glass- 

the  second  seems  very  convenient.  We  clear  sanidine.  As  Atzeni  remarks,  these 

hope  the  work  reported  here,  part  of  a  sometimes   contain   cores   of   oligoclase. 

long-range  and  rather  general  study  of  The   transition  from  oligoclase   core  to 

rhyolites   and   trachytes,   persuades   the  sanidine  mantle  may  be  either  blurred 

reader  that  much  may  also  be  gained  by  and  gradual  or  sharp  and  abrupt.  In  the 

using  analytical  chemistry  and  petrog-  former   case   the   crystals  usually  show 

raphy   as    supplements    to    rather   than  highly    undulant    extinction,    and    the 

substitutes  for  each  other.  "core"  is  likely  to  have  a  jagged  outline 

In  the  current  report  year  we  have  marked  by  many  reentrants ;  in  the  latter 

completed  examination  of  four  peralka-  there  is  no  suggestion  of  strain  or  replace- 

line  specimens  and  the  feldspar  concen-  ment,    and    the    sharply    euhedral    core 

trates    prepared    from    them.    Siliceous  usually    shows    polysynthetic    twinning, 

lavas  of  this  type  provide  an  excellent —  There  are  also  occasional  phenocrysts  of 

perhaps  the  best — opportunity  to  study  acmitic  diopside,   sublenticular  clots  of 

the  relation  between  crystal  composition  tridymite,    and    irregular    inclusions    of 

and    bulk    composition    in    a    natural  other,  possibly  cognate,  volcanic  rocks, 

"system"  closely  resembling  the  experi-  As  in  most  Sardinian  volcanics  so  far 

mentalists'   version  of   "petrogeny's  re-  collected   in   this   project,    the   feldspar 

sidua."  The  specimens  include  a  trachyte  phenocrysts    show    little    indication    of 

from   Paris   de   Besa,    Sardinia,   a   por-  alteration,      although      joint      surfaces 

phyritic  pantellerite  from  Pantelleria,  and  throughout  the  rock  are  usually  stained 

two  comendites  from  the  type  localities  yellowish    brown    and    in    thin    section 

Le  Commende  and  Le  Fontane,  Isola  San  similar    staining    sometimes    occurs    in 

Pietro,   Sardinia.   The  pantellerite,   col-  phenocrysts.  Carlsbad  twins  are  common, 


GEOPHYSICAL   LABORATORY  113 

but  in  our  specimens  no  other  variety  of  trifling  amount.  Despite  much  effort  we 

twinning    has    been    observed    in    the  were  unable  to  isolate  enough  of  either 

sanidine,  which  is  also  free  of  micro-  and  cossyrite  or  acmite  for  analysis.  The  mode 

cryptoperthitic    intergrowth.    Its    optic  of  Washington's  porphyritic  pantellerite 

angle  is  variable  but  always  very  small,  from  Gelkhamar  (U.  S.  N.  M.  no.  PRC 

(Unless  our  material  is  entirely  atypical,  2000)  is  shown  in  column  2  of  table  6. 
Atzeni's  identification  of  the  alkali  feld-         The   comendites   of   Isola   San  Pietro 

spar  of  this  rock  as  microcline  is  erro-  were  discovered  by  Bertolio  and  described 

neous.)  The  mode  of  our  specimen  (no.  in  considerable  detail  by  Johnsen  (1912); 

25B10)  is  shown  in  column  1  of  table  6.  since  Johnsen's  work  no  new  information 


TABLE  6.     Modes  of  the  Analyzed  Specimens 


Phenocrysts 

25B10 

PRC  2000 

39B2 

40B5 

Quart,z 
Sanidine 

0.1 
19.1 

0.8 
10.7 

8.8 
16.8 

6.0 
9.3 

Plagioclase 
Acmite 

0.7 
0.5 

0.2 

Cossyrite 

Opaque 

Tridymite  and  others 

Groundmass 

0.7 
0.4 

78.5 

0.8 

87.5 

74.4 

0.3 
84.3 

25B10,  trachyte,  Paris  de  Besa,  Sardinia. 

PRC  2000,  pantellerite,  Gelkhamar,  Pantelleria. 

39B2,  comendite,  "Commende  type,"  Capo  Sandolo,  Isola  San  Pietro,  Sardinia. 

40B5,  comendite,  "Fontane  type,"  Le  Fontane,  Isola  San  Pietro,  Sardinia. 


The  porphyritic  pantellerite  is  H.  S.  on  these  interesting  rocks  appears  to  have 
Washington's  specimen  from  Gelkhamar,  been  published.  Our  specimen  39B2  is 
Pantelleria,  described  and  analyzed  by  from  a  roadside  exposure  about  500 
him  (Washington,  1914).  A  molar  excess  meters  east  of  the  lighthouse  at  Capo 
of  alkalies  over  R2O3,  signified  in  the  Sandalo  and  about  1  km  west  southwest 
CIPW  system  by  the  appearance  of  Ns  of  Le  Commende,  from  which  the  rock 
(sodium  metasilicate)  in  the  norm,  has  type  takes  its  name.  It  is  a  completely 
long  attracted  attention  to  the  Pantel-  devitrified  blue-gray  glass  containing 
lerian  lavas.  This  specimen  was  one  of  numerous  phenocrysts  of  bipyramidal 
two  recently  reanalyzed ;  for  a  comparison  quartz  and  blocky,  water-clear  sanidine, 
of  the  new  and  old  analyses  see  Zies  the  sanidine  usually  showing  a  pro- 
(1960).  The  feldspar  analysis,  made  at  nounced  schiller.  The  matrix  consists  of 
the  same  time  as  the  bulk  analysis,  spherulitic  masses  of  extremely  fine- 
appears  here  for  the  first  time.  The  grained  quartz,  feldspar,  and  an  acicular 
principal  phenocryst  of  this  specimen  is  green  mineral,  which  may  be  either  acmite 
sanidine,  called  soda  microcline  by  Wash-  or  arfvedsonite.  Specimen  40B5  is  from 
ington.  In  a  careful  examination  of  two  the  quarry  at  Le  Fontane,  about  500 
thin  sections  and  of  many  granular  miles  southwest  of  the  town  of  Carloforte. 
products  obtained  at  various  stages  of  the  It  is  a  bluish  gray  glass,  closely  matching 
sample  preparation  no  second  feldspar  the  description  of  Johnsen's  "Fontane 
was  observed.  Phenocrysts  of  bipyram-  type"  comendite.  Both  quartz  and  feld- 
idal  quartz  are  abundant;  phenocrysts  spar  phenocrysts  are  similar  in  appear- 
other  than  quartz  and  sanidine,  chiefly  ance  to  those  already  described  from  Le 
cossyrite  and  acmite,  are  present  only  in  Commende,  though  much  less  abundant 


114 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


than  in  our  particular  specimens  from 
that  locality.  Prominent  in  40B5  are 
stringers  of  a  dense  black  glass  distributed 
through  the  rock  in  conspicuously  laminar 
fashion.  The  groundmass  is  glass  showing 
little  evidence  of  devitrification.  The 
modes  of  specimens  39B2  and  40B5  are 
shown  in  columns  3  and  4  of  table  6. 

Bulk  analyses  and  norms.  Analyses  and 
CIPW  norms  of  the  four  specimens,  as 
well  as  of  the  material  forming  one  of  the 
dark  stringers  in  40B5,  are  shown  in 
table  7.  It  will  be  noted  that  although  all 
the  rocks  are  peralkaline  Ns  appears  in 
only  one  of  the  four  norms.  This  is,  of 
course,  the  pantellerite  from  Gelkhamar, 
Pantelleria;  the  large  amount  of  norma- 
tive Ns  shown  in  the  original  analysis  of 
this  specimen  was  the  principal  occasion 
for  its  reanalysis.  Ns  is  recorded  in  seven 
of  the  ten  available  peralkaline  norms  of 
the  Pantellerian  lavas  (Washington, 
1914),  the  first  and  still  the  most  extreme 


example  of  molar  excess  of  alkalies  over 
ferric  oxide  and  alumina. 

Ns  is  not  present  in  either  of  our 
comendite  norms,  occurs  in  only  one  of 
the  norms  of  the  seven  comendite 
analyses  given  by  Johnsen  (1912),  and  is 
evidently  both  uncommon  and  quanti- 
tatively insignificant  in  the  type  locality 
of  comendite.  This  contrast  between 
comendite  and  pantellerite  is  perhaps 
particularly  striking  because  in  the  petro- 
graphic  literature  the  names  are  often 
used  interchangeably.  The  differences 
between  our  two  comendites  and  the 
Gelkhamar  pantellerite  are  about  what 
would  be  anticipated  from  inspection  of 
earlier  analyses.  In  connection  with  these 
particular  rocks,  however,  such  an  inspec- 
tion raises  more  problems  than  it  solves. 
Even  if  we  agree  to  ignore  sampling 
difficulties,  which  are  unusually  acute, 
and  the  total  amount  of  information, 
which  is,  as  usual,  rather  small,  Johnsen 


TABLE  7.     Bulk  Analyses  and  Norms 
(Specimens  as  identified  in  table  6;  40B5  inc.  is  fragment  of  a  stringer  in  40B5.) 


Analyses 

Norms 

25B10 

PRC 

2000 

39B2 

40B5 

40B5 
inc. 

25B10 

PRC 

2000 

39B2 

40B5 

Si02 

67.12 

69.81 

75.36 

75.31 

75.01 

Q 

13.22 

28.07 

33.67 

35.33 

A1203 

15.66 

8.59 

11.44 

10.43 

10.53 

Or 

35.91 

26.56 

28.06 

27.50 

Fe203 

2.58 

2.28 

2.30 

3.22 

3.06 

Ab 

41.70 

19.20 

32.41 

27.75 

FeO 

0.69 

5.76 

0.76 

0.80 

1.14 

An 

2.67 

MgO 

0.36 

0.10 

0.13 

0.10 

0.10 

Ac 

6.61 

1.80 

5.31 

CaO 

1.01 

0.42 

0.07 

0.13 

0.10 

Ns 

5.14 

BaO 

0.01 

* 



tr. 

tr. 

Di 

1.04 

1.11 

0.15 

0.35 

Na20 

4.93 

6.46 

4.07 

3.99 

4.05 

Hy 

10.04 

K20 

6.08 

4.49 

4.75 

4.65 

4.68 

En 

0.41 

0.25 

0.09 

H20+ 

0.30 

0.14 

0.75 

0.51 

0.44 

11 

1.26 

0.85 

0.30 

0.39 

H20" 

0.06 

0.05 

0.38 

0.38 

0.27 

Mt 

0.44 

2.22 

2.32 

Ti02 

0.66 

0.45 

0.16 

0.21 

0.21 

Hm 

2.27 

0.14 

Zr02 

0.01 

0.25 

0.10 

0.18 

0.17 

HI 

1.30 

P205 

0.16 

0.13 

0.02 

0.03 

0.03 

Ap 

0.37 

0.30 

0.05 

0.07 

so3 

0.01 

0.06 



tr. 

tr. 

Z 

0.25 

0.15 

0.27 

C1-H20  sol. 



0.03 



0.05 



Rest 

0.41 

0.25 

1.15 

0.89 

C1-H20  msol. 
MnO 

0.02 
0.05 

0.76 
0.28 

0.02 
0.07 

0.05 
0.09 

0.08 
0.09 

99.76 

99.68 

100.35 

100.27 

100.06 

0  for  CI 

99.71 

0.18 

Sum 

99.88 

100.38 

100.13 

99.96 

Sought  but  not  found. 


GEOPHYSICAL   LABORATORY 


115 


points  out  that  his  analyses  of  comendite 
differ  markedly  from  the  earlier,  incom- 
plete analyses  of  Bertolio,  Washington 
points  out  that  his  analyses  of  pantel- 
lerite  differ  markedly  from  the  earlier, 
incomplete  analyses  of  Forstner,  and 
there  is  now  reason  to  suspect  that 
Washington's  Ti02  estimates  were  sys- 
tematically high  (Zies,  1960,  p.  306),  an 
analytical  bias  that  would  necessarily 
generate  overestimates  both  of  the 
amount  of  Ns  and  of  the  frequency  of  its 
occurrence  in  any  set  of  norms.  Like  so 
many  of  the  problems  of  modern  descrip- 
tive petrography,  satisfactory  comparison 
of  these  two  rock  types  will  require  a 
manifold  increase  in  the  number  of  rock 
analyses  with  no  sacrifice,  and  preferably 
with  some  improvement,  in  their  quality. 
Feldspar  phenocrysts  of  the  analyzed 
rocks.  In  all  four  specimens  the  principal 
phenocryst  is  sanidine,  characterized  by 
very  small  optic  angle,  apparent  mono- 
clinic  symmetry  of  X-ray  powder  spectra, 
absence  of  multiple  twins,  and  lack  of 
cryptoperthic  structure  or  anorthoclase 
type  grill.  The  geological  occurrence  is  of 
course  the  classic  one  for  sanidine;  the 
blocky,  sharply  euhedral  habit  and  the 
glassy,  often  transparent  character  of 
the  crystals  are  equally  appropriate. 
Despite  careful  search,  no  other  feldspar 
has  been  identified  in  three  of  the 
specimens;  in  the  Paris  de  Besa  trachyte, 
as  already  noted,  sanidine  crystals  some- 
times contain  cores  of  oligoclase.  The 
X-ray  spectra  of  the  Paris  de  Besa  and 
Gelkhamar  feldspars  show  very  little 
submicroscopic  unmixing,  whereas  those 
from  the  Capo  Sandalo  and  Le  Fontane 
comendites  seem  to  be  almost  entirely 
unmixed,  the  powder  patterns  being 
interpretable  as  mixtures  of  nearly  pure 
Or  and  Ab.  This  unmixing  can  be 
detected  only  by  X  ray.  It  is  therefore 
rather  startling  to  discover  that  in  the 
very  year  in  which  Laue  first  predicted 
that  crystals  ought  to  diffract  X  rays 
Johnsen  (1912,  p.  6)  unhesitatingly 
attributed  the  schiller  of  the  comendite 
sanidine  to   incipient  unmixing,   which, 


carried  to  completion,  would  finally 
transform  an  "unstable  monoclinic"  crys- 
tal into  a  "stable  triclinic"  one.  The  line 
between  science  and  prescience  is  some- 
times very  hard  to  draw ! 

Analyses  and  norms  of  the  best  feldspar 
concentrates  that  could  be  obtained  from 
each  specimen  by  magnetic  and  heavy- 
liquid  separation  are  shown  in  table  8. 


TABLE  8. 

Analyses  and  Norms  of  Four 

. 

AJkali  Feldspars 

i 

(Specimens  as  identified  in  table  6.) 

25B10 

PRC  2000 

39B2 

40B5 

Si02 

65.56 

67.82 

67.30 

67.12 

A1203 

19.68 

18.06 

18.42 

18.32 

Fe203 

0.40 

1.11 

0.59 

0.83 

FeO 

0.05 



0.02 

0.04 

MgO 

* 



0.02 



CaO 

1.14 

_  : 

0.06 



BaO 

0.03 

0.07 





Na20 

6.08 

7.17 

6.95 

5.89 

K20 

6.78 

5.99 

6.60 

7.92 

Ti02 

0.06 

0.02 

0.01 



MnO 





0.02 



H2Of 

0.02 



0.01 



Sum 

99.80 

100.24 
Norms 

100.00 

100.12 

Q 

1.74 

3.44 

1.63 

2.53 

Or 

40.08 

35.41 

39.02 

46.82 

Ab 

51.45 

59.27 

58.01 

49.83 

An 

5.65 

0.31 

Cs 

0.08 

0.19 

Ac 

1.25 

0.69 

En 

0.05 

Mt 

0.09 

0.12 

11 

0.12 

0.02 

Hm 

0.40 

0.69 

0.35 

0.75 

C 

0.22 

0.05 

Rest 

0.02 

0.01 

*  Sought  but  not  found. 

f  Samples  dried  at  110°C  before  analysis. 


Results  of  fragment  counts  made  on 
three  of  the  analyzed  specimens  are 
recorded  in  table  9;  in  the  feldspar  from 
40B5  no  quartz  or  plagioclase  was  found, 
and  the  amount  of  groundmass  adhering 
to  sanidine  grains  was  too  small  to 
estimate  by  this  technique.  In  the  counts, 
incidentally,   the  precision  of  ratios  of 


116 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


TABLE  9.     Fragment  Counts  of  Analyzed 

Feldspar  Concentrates 

(Specimens  as  identified  in  table  6.) 


25B10 

PRC  2000 

39B2 

Alkali  feldspar 

Plagioclase 

Quartz 

Others 

Groundmass 

92.4 
3.3 
3.0 

1.3 

95.4 

3.9 
0.2 
0.4 

98.4 

1.0 
0.7 

Count  length 

1347 

1446 

1827 

minerals  to  each  other  is  about  that 
appropriate  to  the  count  length,  but, 
since  i 'groundmass"  occurs  almost  entire- 
ly as  thin  discontinuous  margins  about 
sanidine  in  all  four  concentrates,  estima- 
tion of  the  amount  of  groundmass  from 
the  number  of  grains  in  which  it  is 
observed  requires  an  "adjustment"  or 
"correction"  factor  that  is  not  much 
better  than  a  shrewd  guess.  The  values 
given  are  probably  overestimates. 

Although  FeO  appears  in  barely  more 
than  trace  amounts  in  the  analyses  of 
table  8,  Fe203  is  present  in  quantities  far 
greater  than  can  reasonably  be  attributed 
to  visible  impurities  or  analytical  error. 
In  all  four  feldspars  a  strong  buff  tint  can 
be  produced  by  heating  for  10  minutes 
or  less  at  ^850°C  in  air,  and  discharged 
by  heating  over  a  Meker  blast  for  a  few 
minutes  at  ^4100°C.  In  our  view  most 
of  the  Fe203  in  these  analyses  must  be 
regarded  as  part  of  the  feldspar,  probably 


proxying  for  A1203,  as  already  suggested 
by  Johnsen  (1912,  p.  6). 

It  will  be  noted  that  quartz  and 
hematite  are  present  in  all  four  and  a 
little  corundum  in  two  of  the  feldspar 
norms.  In  three  of  the  specimens  the 
content  of  normative  quartz  is  roughly 
comparable  to  the  modal  amounts  shown 
in  table  9.  The  agreement  is  far  from 
exact,  but  the  presence  of  measurable 
amounts  of  modal  quartz  sharply  limits 
the  use  of  these  analyses  as  a  basis  for 
speculation  about  the  nature  of  the 
silica-cation  balance  in  alkali  feldspars. 
We  may  point  out,  however,  that  despite 
diligent  search  no  quartz  at  all  was  noted 
in  feldspar  40B5,  yet  the  norm  shows  2.53 
per  cent.  Most  of  the  norms  of  literature 
analyses  of  alkali  feldspar  so  far  examined 
show  normative  quartz  in  comparable  or 
greater  amounts,  amounts  large  enough 
so  that  they  could  scarcely  be  overlooked 
by  the  petrographer  or  fabricated  by  the 
chemist.  Many  also  show  appreciable 
amounts  of  normative  hematite  and 
corundum.  Despite  serious  analytical  and 
sampling  uncertainties  in  available  data 
it  seems  to  us  that  the  possibility  of 
systematic  departure  from  the  assumed 
1:1:6  ratio  of  RO:R203:Si02  in  alkali 
feldspar  deserves  more  than  casual  con- 
sideration. 

Projection  of  results  into  "petrogeny's 
residua  system."  Ternary  coordinates  of 
the  four  rock-feldspar  phenocryst  pairs 


TABLE  10.     Rock  and  Feldspar  Compositions  Projected  into  the  Ternary  System  Q-Or-Ab 
(Numbered  specimens  as  identified  in  table  6;  Commende  and  Fontane  from  Johnsen,  1912.) 


25B10 

PRC  2000           39B2 

40B5 

Johnsen 

Commende 

Fontane 

Rock 

Q 

Or 

Ab 

14.6 
39.5 
45.9 

38.0                35.8 
36.0                34.4 
26.0                29.8 

Feldspar 

39.0 
30.4 
30.6 

33.8 
31.4 
34.9 

40.1 
30.2 
29.8 

Or  (analytical) 
Or  (X-ray) 

44.0 
40 

37.4                39.9 
35                    37 

48.4 
44 

40.8 

48.6 

GEOPHYSICAL   LABORATORY 


117 


Fig.  38.  Data  of  table  10  plotted  in  Q-Or-Ab  diagram.  JC,  rock  anal.  p.  11,  no.  2,  feldspar  anal. 
p.  5,  no.  1;  JF,  rock  anal.  p.  22,  no.  4,  feldspar  anal.  p.  19,  no.  1,  in  Johnsen  (1912).  Other  specimens 
as  identified  in  table  6. 


described  in  this  note  together  with  two 
evidently  similar  pairs  taken  from  John- 
sen  (1912)  are  listed  numerically  in  table 
10  and  shown  graphically  in  figure  38. 
The  last  line  of  the  table  gives  compo- 
sitions of  alkali  feldspar  determined  in 
the  way  described  by  Tuttle  and  Bowen 
(1958,  pp.  11-13)  on  specimens  homoge- 
nized at  850°C  and  room  pressure  for  24 
hours.  As  far  as  could  be  determined  from 
the  X-ray  powder  spectra  the  materials, 
which  were  initially  monoclinic,  were 
completely  homogenized  by  this  treat- 
ment. We  could  find  no  significant 
differences  between  Or  content  estimated 
on  specimens  treated  in  this  way  and  that 
obtained  from  specimens  heated  hydro- 
thermally  for  extended  periods  of  time. 
It  will  be  noted  that  all  four  estimates  of 
Or  by  X  ray  are  lower  than  the  relevant 


Or/ (Or  +  Ab)  ratios  calculated  from  the 
analytical  data.  We  suspect  that  the  high 
Fe203  and  excess  Si02  already  noted  may 
be  responsible  for  this  discrepancy,  since 
the  determinative  curves  are  developed 
for  synthetic  materials  as  free  of  Fe  and 
as  close  to  the  1:1:6  ratio  as  possible. 

In  figure  38  a  line  connects  each 
projected  bulk  composition  with  its 
appropriate  feldspar,  our  data  being 
shown  with  solid  dots  and  lines,  the  two 
pairs  taken  from  Johnsen  by  open  circles 
and  dashed  lines.  Our  Fontane  specimen 
obviously  checks  Johnsen's  very  closely, 
and  the  two  Commende  pairs  are  also  in 
good  agreement.  The  differences  between 
the  slopes  of  these  lines  may  appear  small 
in  ternary  projection  but  are  in  fact  very 
large.  In  compositions  distributed  along 
the    line    connecting    the    Pantellerian 


118  CARNEGIE     INSTITUTION      OF     WASHINGTON 

feldspar  and  its  host  rock  the  slope  of  the  of  rock  analyses,  as  portrayed  graphically 

regression  of  Or  as  a  function  of  Ab  or  Q  in  the  Harker  variation  diagram.4  Be- 

would  be  zero  and  the  slope  of  the  regres-  cause  of  algebraic  restrictions  arising  from 

sionofAb  as  a  function  of  Q  would  be  —  1.  the    cloture    property,    co variances    and 

In  compositions  distributed  along  the  line  correlations     are     not     independent     of 

connecting  our  Fontane  feldspar  with  its  variances,    as    is    normally    assumed    in 

host  rock,  on  the  other  hand,  the  slope  of  statistical  (or  other)  testing.  The  larger 

the  regression  of  Or  as  a  function  of  Ab  the  variance  of  a  particular  variable,  the 

would    be    +1    and    the    slope    of    the  more  strongly  negative  are  its  expected 

regression  of  either  Ab  or  Or  as  a  function  correlations  with  other  variables,  and  as 

of  Q  would  be  —  %.  long  as  no  variance  is  greater  than  the 

The  relations  between  phenocryst  and  sum  of  the  other  variances  all  expected 

host  implied  by  these  two  pairs  are  thus  correlations  are  negative.  The  particular 

very  different.  In  the  first  the  normative  effect  of  variance  or  co  variance  of  most 

groundmass    feldspar    must    be    much  importance  for  an  appreciation  of  the 

richer  in  Or  than  the  phenocryst  feldspar ;  Harker  diagram  may  be  illustrated  by  a 

in  the  second  the  two  must  be  nearly  very  simple  model. 

identical    in    composition.    From    data  Let  us  suppose  an  M-variable  closed 

already  given  we  may  estimate  that  the  array  in  which  the  parent  variances,  07 2, 

normative    Or    content    of    groundmass  of  variables  X,  are  equal  for  2  $  j  ^  M, 

feldspar  in  the  Gelkhamar  pantellerite  is  but   in  which   ci2    9*    aj2.   Further,   we 

61.2  compared  with  37.4  in  the  pheno-  specify  that,  although  ci2  is  potentially 

crysts.  In  the  Capo  Sandolo  comendite,  variable,     the    system    remains    stable 

on  the  other  hand,  the  Or  content  of  the  during  any  particular  sampling  or  set  of 

normative  groundmass  feldspar  is  47.7  as  samplings.   Under  these   conditions   the 

compared  with  39.9  in  the  phenocrysts,  expected  correlation  between  X\  and  X3 

and    in    the    Fontane    comendite    the  in  such  a  sampling  is 

comparable  values  are  50.7  and  48.4,  this  , 

last  pair  probably  differing  by  less  than  pl>  "  °"l/o"^1  ~  M)               W 

the  total  experimental  error.  The  possible  and  that  between  any  pair  of  variables 

importance  of  this  distinction  between  not  including  Xi  is 

pantellerite   and   comendite  is   obvious;  P 

fractionation   of   the    Pantellerian   type  _       ±         ± <Tl  (2) 

would  involve  extensive  end -stage  enrich-  2  —  M  L         (M  —  1 )  o"y2- 

ment  in  potassium,  whereas  fractionation  Nqw      fa  b    definition  nonnegative,  and 

of  the  comenditic  type  would  proceed  ft  can  be  ghown  (        for  instance>  chayeS; 

with  no  notable  shift  in  the  Na/K  ratio  lQm)  ^       <  {M  _  l}      We  are  thug 

It  is  equally  obvious  that  a  discussion  of  concemed  with  variations  in  Plj  and  Pjk 

this  problem  based  on  data  from  three  or  ^              arige    ag    ft    consequence    of 

four  specimens  is  scarce  y  more  than  idle  itti              to               in    the    range 

speculation.  Once  more  the  desirability  01  ^  <-       <(M—Vi 

a  drastic  increase  in  analytical  potential  ^substitution"^    these    values    in 

becomes  apparent.  equations  1  and  2  we  have 

Variance  Relations  in  Some  Published  0  ^  piy  >  —  1 1        ,^ 

Harker  Diagrams  —1/(M  —  2)  ^  pjk  <  +1) 

F.  Chayes  ^s  Gl  m0ves  from  its  lower  to  its  upper 

Much  of  the  interest  in  petrographic  limit>    the    negative    correlation    to    be 

closed  arrays  centers  on  relations  between  4  The  diagram  known  by  Harker's  name  seems 

silica  and  other  essential  oxides  in  suites  to  have  been  invented  by  Iddings  (1892). 


GEOPHYSICAL   LABORATORY 


119 


expected  between  X\  and  Xj  becomes 
progressively  stronger  while  the  initially- 
strong  negative  correlation  between  Xj 
and  Xk  becomes  progressively  weaker;  if 
a i2  >  (M  —  l)aj2  the  expected  value  of 
the  latter  correlation  is  positive.  For  the 
present  we  specify  no  mechanism  by 
which  to  manipulate  the  variances, 
arguing  only  that,  if  variance  relations  of 
the  specified  sort  did  in  fact  occur,  the 
correlations  to  be  expected  in  the  absence 
of  any  other  relation  between  the  vari- 
ables would  be  dictated  by  them. 

That  the  principal  negative  correlations 
of  the  Harker  diagram  seemed  intimately 
related  to  the  variances  in  the  fashion 
suggested  had  already  been  noted 
(Chayes,  1962),  but  in  fact  this  inference 
was  largely  based  on  examinations  of 
graphs.  During  the  report  year  calcula- 
tions were  carried  through  on  twenty-five 
suites  of  analyses  of  volcanic  rocks  that 
had  served  as  the  basis  of  published 
Harker  diagrams.  Satisfactory  description 
of  the  results  calls  for  bibliographical  and 
other  detail  not  appropriate  in  a  report 
of  this  sort.  Certain  of  the  findings  are  so 
extreme,  however,  that  further  work  is 
hardly  likely  to  lead  to  substantial 
modification. 

In  every  suite,  for  instance,  the  vari- 
ance of  silica  was  considerably  larger  than 
that  of  any  other  oxide.  In  twenty-three 
of  the  twenty-five  suites  the  variance  of 
silica  was  larger  than  the  sum  of  all  other 
variances,  both  exceptions  being  suites  of 
oceanic  lavas,  assemblages  to  which  the 
Harker  diagram  is  rarely  applied.  The 
ratio  of  silica  variance  to  the  sum  of  other 
variances  is  so  far  never  larger  than  3.38; 
its  average  value,  1.95,  taken  as  an 
estimate  of  the  ratio  <ri2/(rj2(M  —  1) 
would  lead  to  pu  ~  —0.63  in  equation  1, 
if  we  count  the  variables  in  the  way  pro- 
posed in  last  year's  report.  Although  this 
is  hardly  more  than  a  very  crude  approx- 
imation, the  variance  relations  are  clearly 
such  as  to  require  very  strong  negative 
correlation  between  silica  and  all  other 
oxides  that  contribute  materially  to  the 
total  variance  of  a  Harker  array.  These 


oxides  are,  in  order  of  increasing  average 
variance,  Fe203,  A1203,  FeO,  MgO,  CaO; 
curiously,  and  perhaps  significantly,  this 
is  also  in  order  of  increasingly  strong 
negative  correlation  with  Si02.  Since 
Ti02,  Na20,  and  K20  do  not  contribute 
materially  to  the  total  variance,  the  effect 
of  cloture  on  their  correlations  with  Si02 
should  be  negligible.  Although  Ti02  is 
usually  negatively  correlated  with  Si02, 
it  is  well  known  that  the  correlations  of 
Na20  and  K20  with  Si02  are  nearly 
always  strongly  positive.  Systematically 
strong  positive  correlations  involving 
Si02  thus  emerge  only  where  the  variance 
relations  permit. 

A  considerable  excess  of  silica  variance 
over  the  sum  of  other  variances  seems  to 
be  characteristic  of  the  continental  basalt- 
andesite-dacite-rhyolite  association ;  if  the 
basic  and  acid  parts  of  such  suites  are 
considered  separately,  the  excess  variance 
of  silica  usually  persists  in  the  rhyolite  or 
dacite-rhyolite  and,  to  date,  always 
persists  in  the  basalt-andesite  parts  of  the 
assemblages.  An  apparently  similar  ex- 
cess is  not  uncommon  in  suites  of  the 
oceanic  basalt- trachyte  association  but 
seems  to  be  generated  here  by  the 
grouping  of  analyses  that  do  not  belong 
together,  as  suggested  in  a  later  section 
of  this  report.  At  any  rate,  in  the  basaltic 
portions  of  oceanic  basalt-trachyte  suites 
silica  variance  never  exceeds  the  sum  of 
the  other  variances  and,  indeed,  is  rarely 
the  largest  variance.  (Unfortunately, 
there  are  very  few  oceanic  suites  in  which 
nonbasalts  are  sufficiently  numerous  to 
make  separate  computation  worth  while.) 
A  full  account  of  variance-covariance 
computations  in  volcanic  suites  is  now  in 
preparation. 

The  Treatment  of  FeO  and  Fe2Os  in 
Harker  Diagrams 

F.  Chayes 

In  most  Harker  diagrams  only  one  Fe 
oxide  is  shown;  customarily,  a  new 
variable  is  formed  by  adding  the  adjusted 
weight  of  one  of  the  oxides  to  the  posted 


120 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


amount  of  the  other  in  each  analysis,  viz., 


or 


(Fe203)r  =  Fe203  +  l.llFeO 
(FeO)r  =  FeO  +  0.901Fe2O3 


5- 


Al though  modern  justifications  of  it  are 
rarely  explicit,  the  practice  itself  was 
proposed  by  Iddings  (1892)  in  the  first 
publication  containing  what  we  would 
now  call  Harker  diagrams.  In  this 
pioneering  discussion  Iddings  often  shows 
total  Fe  as  FeO,  but  always  shows  FeO 
and  Fe203  separately  as  well.  (In  the  first 
text  treatment  of  the  subject,  however, 
Harker  [1909]  shows  the  iron  oxides 
separately  in  only  six  of  the  twelve 
diagrams  he  presents.) 

From  close  examination  of  the  "very 
carefully  executed  analyses  of  the  rocks 
from  the  region  of  the  Yellowstone 
Park,"  Iddings  argues  (1892,  p.  153)  that 
"In  this  group  of  rocks  the  reciprocal 
behavior  of  the  ferrous  and  ferric  oxides 
is  one  of  the  most  marked  chemical 
features"  and  concludes  "it  seems  highly 
probable  that  during  the  differentiation 
of  the  magma  all  of  the  iron  existed  in  the 
ferrous  condition  .  .  .  and  that  subse- 
quently it  was  in  part  more  highly 
oxidized,  so  that  the  more  ferric  oxide 
was  produced  the  less  ferrous  remained." 
Before  hastily  concluding  that  this  dictum 
implies  negative  correlation  between  FeO 
and  Fe203,  the  reader  is  urged  to  examine 
figure  39,  a  display  of  the  data  upon 
which  it  is  based.  The  correlation  between 
FeO  and  Fe203  is  actually  positive, 
though  very  weak.  Denoting  Si02  by  x, 
Fe203  by  y,  and  FeO  by  z,  the  Iddings' 


*  4 


ro3 
O 

CVJ 

u.  2 


L 


•  •       • 


•       m 


2        3        4 
FeO.  wt.% 


Fig.  39.  FeO  and  Fe203  in  volcanic  rocks  of 
Yellowstone  Park  (data  from  table  1  of  Iddings, 
1892). 


data  give  ryz  =  +0.246.  The  partial 
correlation  is  very  different,  viz.,  ryz.x  = 
—  0.690.  For  a  fixed  silica  content  there  is 
indeed  some  tendency  for  FeO  and  Fe203 
to  vary  inversely  in  the  sample,  but  the 
tendency  hardly  seems  strong  enough  to 
warrant  either  Iddings'  detailed  specu- 
lations about  a  pooled  Fe  variable  or  the 
tacit  conviction  of  modern  petrographers 
that  no  other  Fe  variable  is  desirable  in 
Harker  diagrams. 

There  is,  nevertheless,  strong  negative 
correlation  between  both  iron  oxides  and 
silica  in  this  earliest  "Harker  array,"  as 
in  so  many  of  its  successors.  In  fact,  for 
the  Iddings  data  rxy  =  0.643,  rxz  = 
-0.834,  and  rx(y+z)  =  -0.925.  This  last 
correlation  is,  I  believe,  the  common 
though  rarely  stated  occasion  for  pooling 
the  Fe  oxides  into  a  single  variable.  We 


Fig.  40.     Histogram  of  product  moment  correlations  between  FeO  and  Fe203  in  the  raw  data  of 
twenty-five  published  Harker  diagrams. 


GEOPHYSICAL   LABORATORY  121 

all  like   points  that  lie   fairly   close   to  authenticity  is   beyond   any   reasonable 

fairly  simple  curves,  and  in  most  Harker  doubt.  (For  a  review  of  the  oceanic  lava 

arrays    the    linear    correlation    between  associations,    see,    for    instance,    Tilley 

silica  and  some  form  of  the  sum  of  the  [1950],  or  Turner  and  Verhoogen  [1951, 

iron  oxides  will  be  stronger  than  that  pp.  124-155].) 

between   silica   and   either   of   the   iron  We  are  equipped  with  a  full  comple- 

oxides  separately.  The  explanation  offered  ment  of  names  for  lavas  intermediate  in 

by  Iddings  is  only  one  of  a  large  class  of  composition  between  basalt  and  trachyte 

hypotheses  compatible  with  this  relation-  — trachybasalt,    trachydolerite,    trachy- 

ship ;  not  all  members  of  this  class  require  andesite,  kohalaite,  mugearite — but  the 

inverse  or  "reciprocal"  variation  of  the  rocks  themselves  seem  not  at  all  common 

iron    oxides,    whether    the    correlation  on  oceanic  islands.  This  apparent  scarcity 

implied  is  total  or  partial.  of   intermediate   members,    noted   by   a 

The  correlation  between  FeO  and  Fe203  number  of  petrologists,  has  been  perhaps 

in  Harker  arrays  is  in  fact  extraordinarily  most  dramatically  presented  by  Barth 

variable.  Figure  40  is  a  histogram  showing  (Barth,  Correns,  and  Eskola,  1939,  p.  65). 

the  distribution  of  ry»  in  the  twenty-five  Citing    Hawaii    as    an    example,    Barth 

arrays  for  which  computations  have  so  remarks  that  in  the  intrapacific  province 

far  been  completed.  Observed  relations  rocks  containing  between  53  and  58  per 

between  FeO  and  Fe203  run  the  gamut  cent    of   silica   seem   to   be    completely 

from    strong    negative     correlation    to  lacking.    Since    Barth    wrote,    however, 

virtually  perfect  positive  correlation.  To  three  Hawaiian  specimens  with  silica  in 

the  extent  that  the  Harker  diagram  is  the  forbidden  range  have  been  reported, 

intended  to  provide  a  condensed  descrip-  two  from  Maui  (Macdonald  and  Powers, 

tion  of  the  data  the  use  of  a  single  Fe  1946,  pp.    119  and   122)   and  one  from 

variable  will  often  be  either  misleading  or  Oahu  (Tilley,  1950,  p.  41).  Although  the 

uninformative.  It  would  be  preferable  to  search   for   them   was   perhaps   in   part 

return  to  the  practice  of  Iddings;  the  stimulated  by  Barth 's  remark,  these  new 

separate  oxides  of  iron  should  be  retained  finds    do    render    his    example    strictly 

as  Harker  variables  whether  or  not  some  incorrect.   And   whether  the   precept  it 

form  of  pooled  Fe  variable  is  also  con-  illustrated  was  ever  correct  will  depend 

structed.  on  whose  analyses  are  to  be  discarded. 

From  elsewhere  in  the  intrapacific  prov- 

On  the  Relative  Scarcity  of  Intermediate  ince  there  are— and  were  at  the  time  of 

Members  in  the  Oceanic  Basalt-Trachyte  Barth's    writing— at    least    seven    other 

Association  analyses  of  lavas  falling  in  the  forbidden 

„  Ch  silica  range,  one  from  Easter  Island  and 

two  each  from  Samoa,  the  Marquesas, 

There  seems  to  be  no  question  that  and  the  Society  group.  But  there  are  also 

basalt   is  by  an  enormous  margin  the  at  least  twenty  analyses  of  Pacific  lavas 

principal  oceanic  lava  and  that  trachyte  in  which  58   <  Si02   <  63,  so  that,  al- 

— often    moderately    feldspathoidal    in  though   lavas   with   silica  in   the   range 

either   norm   or   mode — is   a   poor   but  53-58  are  perhaps  not  so  rare  in  this 

uncontested    second.     The    next    most  region  as  Barth  suggested,  analyses  of 

important  oceanic  lava  is  probably  pho-  them  are  nevertheless  considerably  less 

nolite,  and  where  phonolite  is  abundant,  common  than  those  of  true  trachytes.  In 

as  in  Tahiti  or  Reunion,  for  instance,  other  oceans  this  central  minimum  is  not 

hypabyssal  or  even  plutonic    feldspath-  quite  so  clear  cut;  in  the  Indian  Ocean, 

oidal  rocks  may  occur.    Rhyolite  is  on  for  instance,  available  data  suggest  no 

the  whole  very  uncommon,  though  there  shortage  of  intermediate  silica  values  for 

are    indeed    a    few    occurrences    whose  the  Kerguelen  archipelago  but  at  Reunion 


122 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


and  Mauritius  the  situation  is  about  like 
that  in  the  Pacific.  In  the  Atlantic  a 
"Barth  gap"  seems  to  occur,  though  very 
weakly,  at  St.  Helena,  but  on  Ascension 
and  in  the  Canaries  the  shortage  is  most 
evident  in  the  58-63  per  cent  Si02  range, 
while  there  are  many  analyses  in  the 
range  63-68  per  cent. 

Table  11  shows  the  incidence  of  silica 
values  in  classes  whose  width  is  Y§  of  the 
range  observed  in  each  island  or  island 
group.  It  will  be  noted  that  zeros  occur 
only  in  classes  4  and  5,  and  that  with  two 
exceptions  l's  are  confined  to  classes  3,  4, 
and  5.  The  column  totals  of  the  table 
leave  little  doubt  that,  for  the  array  as  a 
whole,  the  inference  of  some  kind  of 
hiatus — either  a  minimum  or  an  outright 
discontinuity — in  the  parent  distribu- 
tion (s)  is  almost  unavoidable. 


Is  it  possible  that  this  preponderance 
of  trachyte  over  trachyandesite  is  merely 
a  consequence  of  traditional  cabinet- 
specimen  sampling  aimed  at  collecting 
rare  and  unusual  material?  On  the 
assumption  that  the  various  lavas  can  be 
adequately  identified  in  the  field,  it  does 
not  seem  at  all  likely  that  collectors 
would  have  ignored  material  as  rare  and 
interesting  as  trachyandesite.  If,  as  is 
suggested  in  many  of  the  source  papers, 
the  hand-specimen  distinction  between 
trachyte  and  basalt  is  in  fact  difficult  and 
unsatisfactory,  there  is  even  less  likeli- 
hood of  a  serious  sampling  bias.  There 
seems  no  reason  to  doubt  that  in  this 
respect  the  distribution  of  analyses 
reflects,  approximately  at  least,  the  dis- 
tribution of  rocks,  and  that  lavas  inter- 
mediate in  composition  between  basalt 


TABLE  11.     Frequency  of  Silica  Values  in  Sixths  of  Group 

Rar 

iges 

for  Larger 

Groups 

of 

Oceanic  Basalt-Trachyte  Suites 

Class 

Range 

S 

1 

2 

3 

4 

5 

6 

Pacific  Ocean 

Kohala-Hulalai1 

41-62 

9 

6 

2 

0 

1 

3 

21 

Georgian  and  Society  Islands2 

41-63 

10 

10 

2 

1 

3 

8 

34 

Marquesas3 

42-66 

14 

3 

1 

1 

7 

2 

28 

Samoa4 

43-72 

5 

1 

1 

0 

2 

5 

14 

Easter5 

42-74 

4 

4 

1 

2 

2 

1 

14 

42 

24 

7 

4 

15 

19 

111 

Atlantic  Ocean 

Ascension6 

47-73 

2 

4 

1 

2 

4 

3 

16 

St.  Helena7 

43-63 

3 

2 

1 

0 

0 

5 

11 

Canary  Islands8 

Feldspathoidal 

39-64 

9 

10 

3 

4 

18 

7 

51 

Nonfeldspathoidal 

38-71 

12 

15 

3 

1 

5 

7 

43 

Azores9 

37-68 

3 

10 

2 

1 

3 

6 

25 

29 

41 

10 

8 

30 

28 

146 

Indian  Ocean 

Reunion10 

44-64 

4 

6 

2 

1 

0 

3 

16 

Mauritius11 

43-63 

9 

3 

1 

1 

0 

6 

20 

Kerguelen12 

43-69 

8 

9 

3 

3 

2 

2 

27 

21 

18 

6 

5 

2 

11 

63 

All 

92 

83 

23 

Yt 

r 

47 

5£ 

320 

Sources:  1,  Washington  (1923);  2,  Iddings  and  Morley  (1918)  and  Lacroix  (1923,  pp.  279-289); 
3,  Chubb  (1930);  4,  Daly  (1924);  5,  Bandy  (1937);  6,  Daly  (1925);  7,  Daly  (1927);  8,  Fuster,  Ibarrola, 
and  Lobato  (1954);  9,  Berthois  (1953);  10,  Lacroix  (1923,  pp.  227-237);  11,  Walker  and  Nicolaysen 
(1954);  12,  Edwards  (1938). 


GEOPHYSICAL   LABORATORY 


123 


and  trachyte  appear  to  be  less  abundant 
than  trachytes  because  they  are  in  fact 
less  abundant.  One  is  tempted  to  point 
out  imposing  continental  analogies:  the 
lavas  of  eastern  Otago,  East  Africa,  the 
Iki  Islands  of  Japan,  etc.  It  is  important 
to  realize,  however,  that,  although  the 
data  as  a  whole  appear  to  indicate  a 
minimum  in  the  frequency  distribution  of 
silica  in  oceanic  lavas,  from  only  one  island 
group  do  we  have  enough  analyses  to  provide 
a  reliable  test  for  the  significance  of  observed 
departures  from  uniform  density  in  classes 
8,  4,  5,  and  6  of  table  111 

This  is  a  regrettable  and  highly 
unsatisfactory  state  of  affairs,  one  that 
should  be  remedied  at  the  earliest 
opportunity.  With  the  increasing  funds 
available  for  oceanographic  research  it  is 
to  be  hoped  that  the  lavas  of  the  oceanic 
islands  will  soon  receive  from  ocean- 
ographers  the  same  kind  of  attention 
space  scientists  are  devoting  to  meteor- 
ites. 

Granite  in  Port  Clyde  Peninsula 
Y.  Suzuki  and  F.  Chayes 

It  is  commonly  supposed  that  there  is 
a  gradual  transition  from  true  granites  to 
gabbros,  with  parallel  tendencies  toward 
increase  of  plagioclase  over  potash  feld- 
spar, increase  of  An  over  Ab  in  plagio- 
clase, increase  of  color  index,  and  decrease 
of  quartz  content.  Our  area,  however, 
provides  no  support  for  this  classical 
notion  of  a  compositional  continuum. 
Rather,  there  is  a  very  strong  suggestion 
that  the  two  principal  facies,  biotite- 
muscovite  granite  and  biotite-hornblende 
granite,  are  quite  distinct  and  readily 
distinguishable. 

In  the  peninsula  stretching  from  Rock- 
land southwestward  to  Port  Clyde, 
Maine,  there  are  many  excellent  expo- 
sures of  granitic  and  dioritic  rocks.  These 
rock  types  outcrop  within  the  outlined 
areas  of  figure  41,  and  sample  localities 
are  marked.  They  are  intrusive  into 
Paleozoic  sediments.  The  unmarked  area 
is  underlain  by  Paleozoic  rocks  or  glacial 


drift.  Along  the  shoreline  outcrops  are 
abundant,  although  often  deeply  weath- 
ered. The  area  was  once  the  center  of  a 
large  quarrying  industry. 

Following  up  some  earlier  studies  of 
the  quarries  by  Chayes,  Suzuki  spent 
about  a  month  during  the  summer  of  1960 
attempting  to  sample  the  intrusive  com- 
plex systematically  on  a  1-km  grid.  The 
final  distribution  of  specimens  is  shown  in 
figure  41.  Despite  the  enormous  amount 
of  natural  and  artificial  outcrop  in  the 
area,  the  sample  density  varies  greatly 
over  the  grid.  By  conventional  standards, 
however,  we  have  an  unusually  large 
sample  of  the  complex  upon  which  to  base 
our  report. 

Most  of  the  outcrop  area  is  underlain 
by  fine-grained  two-mica  granite.  The 
northeast  part  of  the  complex  consists 
of  coarse-grained  hornblende-biotite  gran- 
ite. Trondhjemite  and  biotite-quartz 
diorites  are  mostly  confined  to  the 
western  border. 

The  average  values  for  the  two  main 
granite  facies  are  shown  in  table  12.  The 
first  column  is  based  on  modal  analyses 
of  thirty-nine  specimens,  the  second  on 
modes  of  twelve  specimens.  By  conven- 
tional variance  analyses  all  differences 
are  highly  significant.  Although  the 
differences  in  the  two  average  modes  are 
in  the  expected  directions,  their  numer- 
ical values  hardly  suggest  the  expected 
continuity.  The  color  index  of  the 
hornblende  facies  is  only  4.1  per  cent 
greater  than  that  of  the  muscovite  facies, 
but  its  quartz  content  is  8.0  per  cent  less. 
Differences  between  the  other  major 
constituents  are,  similarly,  very  much 
larger  than  the  color-index  difference. 
The  real  clue  to  the  situation  seems  to  be 
the  presence  or  absence  of  hornblende — 
sometimes  of  very  little  hornblende. 

Although  plagioclase  is  dominant  over 
potash  feldspar  in  all  the  hornblende- 
bearing  granites,  the  most  calcic  plagio- 
clase so  far  encountered  is  An  35,  and  the 
quartz  content  of  these  rocks  is  usually 
well  within  the  granite  range  (fig.  42). 

Rocks  with  no  potash  feldspar  at  all 


124 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


3  Km. 


Fig.  41.  Sample  localities  in  the  Port  Clyde  peninsula  complex.  Double  circle,  muscovite  granite; 
open  circle,  muscovite-biotite  granite;  cross  circle,  biotite  granite;  solid  circle,  hornblende-biotite 
granite;  triangle,  trondhjemite;  diagonal  cross,  quartz-biotite  diorite. 


may  or  may  not  be  conspicuously  inter- 
mediate between  granite  and  gabbro  in 
quartz  content.  But  rocks  containing  even 
a  little  potash  feldspar  are,  in  this 
respect,  not  transitional  at  all.  Rather, 
they  are  simply  granitic. 

The  plagioclase  of  all  facies  of  the 
complex  exhibits  mild  but  persistent 
zoning,  so  that  reliable  estimation  of  its 
An  content  is  difficult.  The  work  on  this 
problem  has  already  been  described  ( Year 


Book  60,  p.  169).  Here  it  is  only  necessary 
to  point  out  that,  although  in  broad 
outline  the  results  are  compatible  with 
the  proposed  transitional  relation,  there 
are  striking  exceptions,  and  the  total 
range  of  An  content  is  rather  small.  The 
plagioclase  of  the  two-mica  granites  is 
oligoclase,  or  occasionally  andesine.  In 
the  hornblende-biotite  granite  it  is  sodic 
andesine,  rarely  oligoclase. 

Sharp  contacts  between  sizable  masses 


GEOPHYSICAL   LABORATORY 


125 


TABLE  12.     Modal  Compositions  of  Muscovite-Biotite  Granite  and  Hornblende-Biotite 

Granite  of  Port  Clyde  Peninsula,  Maine 


Muscovite-Biotite  Granite 

Hornblende-Biotite  Granite 

Mean 

Standard 
Deviation 

Mean 

Standard 
Deviation 

Quartz 

Potash  feldspar 
Plagioclase 

31.91 
27.79 
29.95 

3.55 
6.64 
6.27 

23.90 
15.17 
46.52 

3.30 

7.78 
5.74 

Muscovite 

Biotite 

Hornblende 

2.52 
6.81 

1.45 
4.23 

12.18 
1.40 

2.98 
1.63 

Color  index 

10.35 

4.57 

14.42 

3.95 

Sample  size 

39 

12 

QUARTZ 


PLAGIOCLASE  POTASH  FELDSPAR 

Fig.  42.     Modal  ternary  ratios  of  Port  Clyde  peninsula  specimens.  Symbols  as  defined  in  figure  41. 


126  CARNEGIE     INSTITUTION      OF      WASHINGTON 

of  the  two  principal  types  of  granite  are  modal    or    mineralogical    properties    of 

nowhere  exposed,   but  we  have  so  far  these  rocks. 

found  no  evidence  for  compositional  The  Ab  content  of  a  grain  of  potash 
gradation  between  the  two.  Although  feldspar  may  receive  contributions  from: 
reliable  identification  cannot  always  be  (1)  plagioclase  not  removed  by  the 
made  in  hand  specimen,  microscopic  separatory  procedure,  (2)  perthitic  inter- 
examination  nearly  always  permits  ready  growths,  (3)  Ab  in  solid  solution.  Source 
assignment  of  an  outcrop  to  one  or  the  1,  contamination,  can  be  held  suitably 
other  of  the  two  main  classes.  Difficulties  low  in  most  of  the  Port  Clyde  peninsula 
arise  only  when  both  muscovite  and  rocks  if  only  small  amounts  of  concen- 
hornblende  are  absent,  and  such  rocks  trate  are  required.  Optical  properties 
are  rare.  measured    on    individual    fragments    or 

The  exact  relation  between  hornblende-  parts    of   fragments    of   perthitic    inter- 

biotite  and  muscovite-biotite  granite  in  growths  are  concerned  only  with  source  3, 

the  Port  Clyde  area  is  still  not  known,  but  in  estimates  of  composition  by  X-ray 

but  it  seems  quite  clear  that  some  type  of  powder  techniques  preliminary  heat  treat- 

geochemical  or  stratigraphic  discontinu-  ment5  converts  contribution  from  source  2 

ity    separates    them.    Further,    there    is  into  contribution  from  source  3.  If  the 

field  evidence — inclusions,  schlieren,  brec-  perthite   appears   to  be   of  replacement 

ciation,    etc. — that   the   relation   of   the  origin,   the   composition  determined  by 

granites    to    the    diorites    and    gabbro-  X    ray    after   heat    treatment   may   be 

diorites  of  the  complex  involves  extensive  mineralogically    interesting    but    petro- 

hybridization.  graphically  uninterpretable.  If,  as  in  the 

It  is  to  be  noted,  too,  that  it  is  the  Port  Clyde  area  rocks,  there  seems  no 

two-mica   granite,   not   the  hornblende-  reason  to  suppose  other  than  an  exsolu- 

biotite  granite,  that  is  usually  involved  in  tion  origin  for  the  perthites,  the  compo- 

this    migmatization.    Indeed,    the    large  sition  determined  by  X  ray  after  homog- 

mass  of  hornblende-biotite  granite  in  the  enization    (and    inversion)    is    properly 

northeastern  part  of  the  outcrop  area  is  regarded  as  an  estimate  of  the  compo- 

separated  by  the  two-mica  granite  from  sition    of    the    alkali    feldspar    before 

the  principal  outcrop  area  of  the  dioritic  exsolution,  and  this  is  a  valuable  datum, 

facies,  which  lies  to  the  southwest.  We  should  also  like  to  know  the  Ab 

Thus,  although  the  notion  of  a  com-  content    of   the    K-feldspar   phase   now 

plete    compositional    continuum    might  visible  in  the  rock.  At  present  the  only 

provide  here,  as  elsewhere,  a  convenient  methods  purporting  to  give  this  informa- 

nomenclature  and  classification  for  the  tion  are  optical,  and  they  are  exceedingly 

various   facies   of  the   complex,   genetic  rough.     The    biggest    index    difference 

inferences  drawn  from  or  based  upon  such  between  pure  orthoclase  and  pure  albite, 

a  concept  would  be  misleading.  for  instance,  is  0.017  (Tuttle,  1952).  With 

measurements  subject  to  an  uncertainty 

of,  say,  5  in  the  fourth  place,  it  is  obvious 

Feldspar  in  the  Granite  of  the  Port  Clyde  that  jf  tne  apparent  difference  between 

Peninsula  ^wo  observations  is  not  zero  it  cannot  be 

Y.  Suzuki  less  than  3  per  cent  Ab.  Whereas  the 

X-ray  procedure  gives  an  average  value 

This  section  describes  variations  in  Ab  for  the  homogenized  specimen,  the  index 

content  of  potash  feldspar  in  the  granites  measurements  give  a  minimum  value  for 

of  the  Port  Clyde  peninsula  and  reports  Or  from  the  maximum  gamma  found  in  a 

attempts    to    determine    whether    such  particular  sample. 

fluctuations  appear  to  be  systematically  s  At  800°C,  1  kb,  1  week,  water  saturated, 

related  to  variations  in  other  measurable  after  grinding. 


GEOPHYSICAL   LABORATORY 


127 


In  table  13  the  average  Or  per  cent  as 
determined  by  X  ray  is  shown  in  column 
2,  and  in  column  3  the  maximum  observed 
gamma  index  is  recorded,  followed  by  an 
estimate  of  the  equivalent  Or  content 
from  Tuttle's  diagram  (1952,  p.  559).  The 
fourth  column  gives  the  angle  from  which 
the  "triclinicity"  of  Goldsmith  and  Laves 
(1954)  is  computed,  and  the  fifth  the 
average  An  content  of  accompanying 
plagioclase,  determined  by  measurement 
of  X-ray  powder  diagrams.  The  An 
content  of  accompanying  plagioclase  is 
clearly  smaller  in  two-mica  granites  than 


in  hornblende  granites.  The  Goldsmith- 
Laves  angle  is  somewhat  larger  in  the 
two-mica  granites,  the  largest  value  in 
the  hornblende-biotite  granites  being  less 
than  the  smallest  in  the  two-mica  granite. 
The  difference  between  minimum  and 
average  Or  per  cent  in  alkali  feldspar  is 
large  enough  to  suggest  that  the  K20 
content  of  this  mineral  varies  from  grain 
to  grain.  In  several  specimens  its  hetero- 
geneity seems  beyond  reasonable  doubt, 
and  it  is  curious  that  three  of  the  five 
hornblende-biotite  granites  fall  in  this 
category.  It  is  also  curious  that,  although 


TABLE  13.     Composition  of  Alkali  Feldspar  and  Accompanying  Plagioclase  in  Granite 

of  the  Port  Clyde  Peninsula 


Average  An 

No. 

Average  Or 

Or  Per  Cent  from 

Angle  between 

Per  Cent  in 

Per  Cent 

Minimum  y 

20  131  and  131 

Associated 

7 

Or% 

Plagioclase 

Two-Mica  Granite 

21 

83 

1.5260 

77 

0.79° 

16 

31 

89 

1.5265 

75 

0.78 

30 

49B 

88 

1.5245 

86 

0.80 

12 

51 

88 

1.5260 

77 

0.78 

18 

58 

83 

1.5255 

80 

0.80 

13 

62 

84 

1 . 5260 

77 

0.78 

19 

65 

90 

1.5275 

69 

0.79 

27 

67 

82 

1.5270 

72 

0.77 

25 

68 

81 

1.5260 

77 

0.78 

16 

80 

86 

1.5265 

75 

0.78 

18 

89 

87 

1.5260 

77 

0.78 

18 

93 

86 

1.5270 

72 

0.78 

20 

109 

88 

1.5260 

77 

0.77 

30 

Hornblende-Biotite  Granite 

25 

90 

1.5275 

69 

0.74 

32 

27 

93 

1 . 5265 

75 

0.71 

29 

28 

91 

1.5280 

66 

0.71 

30 

41 

87 

1.5245 

86 

0.64 

25 

42 

86 

1.5255 

80 

0.70 

29 

Muscovite  Granite 

26 

94 

1.5260 

77 

0.79 

3 

55 

91 

1.5240 

89 

0.80 

5 

Biotite  Granite 

71 

80 

1.5275 

69 

0.76 

33 

44 

92 

1.5250 

83 

0.61 

18 

Two-Mica  Granite  (minor  dike) 

77 

83 

1.5265 

75 

0.78 

30 

128 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


TABLE  14.     Or  in  K  Feldspar  and  An  in  Plagioclase  by  Rock  Types 

(Data  of  table  13) 


No.  Samples 

Average  Or 

Per  Cent  in 

Potash  Feldspar 

Minimum  Or 

Per  Cent  in 

Potash  Feldspar 

Average  An 
Per  Cent  in 
Plagioclase 

All  Data 

Average 
Standard  deviation 

23 

87.04 
3.91 

Two-Mica  Granite 

76.09 
5.61 

21.30 
8.69 

Average 
Standard  deviation 

13 

85.77 
2.89 

76.23 
4.10 

20.03 
6.06 

Hornblende-Biotite  Granite 

Average 
Standard  deviation 

5 

89.40 
2.88 

74.20 
7.59 

29.02 

2.54 

in  the  two-mica  granites  both  the  modal 
content  of  K  feldspar  and  the  minimum 
Or  content  of  K  feldspar  show  strong 
negative  correlation  with  An  in  accom- 
panying plagioclase,  the  average  Or  per 
cent  in  K  feldspar  does  not.  Table  14 
shows,  by  rock  type,  the  average  values 
of  columns  2,  4,  and  6  of  table  13. 

Certain  of  the  classical  "gradations" 
appear  to  be  present  within  the  mica 
granites.  There  is,  for  instance,  a  mark- 
edly inverse  variation  between  the 
amount  of  K  feldspar  in  the  mode  and 
the  average  An  content  of  the  plagioclase 
in  the  rock;  since  the  total  feldspar 
content  is  relatively  stable  it  is  not 
surprising  then  to  find  rather  strong 
positive  correlation  between  An  content 
of  plagioclase  and  plagioclase  content  of 
rock. 

Although  the  Goldsmith-Laves  angle  of 
microcline  increases  almost  linearly  with 
increase  of  An  content  in  plagioclase  of 
the  hornblende-bio tite  granite,  there  is  a 
strong  suggestion  of  an  opposite  trend  in 
the  two-mica  granites ;  at  present  we  have 
no  explanation  to  offer  for  either  of  these 
effects.  Although  average  Or  in  K  feldspar 
does  not  appear  to  be  significantly 
correlated  with  any  other  sample  statistic, 
there  is  a  fairly  strong  inverse  variation 
between  minimum  Or  in  K  feldspar  and 
average  An  content  of  plagioclase. 


Two-Mica  Granite  and  Hornblende-Biotite 
Granite 

Y.  Suzuki 

The  marked  modal  differences  between 
the  two-mica  and  hornblende-biotite 
granites  of  the  Port  Clyde  peninsula 
prompted  a  literature  search  for  quanti- 
tative modal  data  about  other  closely 
associated  granites  of  these  two  types.  As 
might  have  been  expected,  this  search 
was  unsuccessful;  the  only  detailed  com- 
parison between  these  rock  types  possible 
at  present  is  one  that  utilizes  chemical 
analyses,  and  such  a  comparison  is  now 
in  progress. 

To  qualify  for  inclusion,  an  analysis  (1) 
must  be  of  a  rock  called  granite  in  the 
source  publication,  (2)  must  list  deter- 
minations of  the  nine  essential  oxides, 
(3)  must  be  accompanied  in  the  source 
publication  by  a  "qualitative"  mode  or, 
at  least,  a  list  of  essential  minerals.  (This 
list  must  of  course  show  that  it  belongs 
in  one  of  the  two  groups  under  dis- 
cussion.) 

The  need  for  the  third  requirement  is 
obvious.  The  second  was  adopted  largely 
as  a  means  of  eliminating  partial  analyses, 
on  the  perhaps  questionable  assumption 
that  if  a  rock  is  not  sufficiently  interesting 
to  warrant  a  full  analysis  it  may  also  fail 


GEOPHYSICAL   LABORATORY 


129 


TABLE  15.     Averages  of  Muscovite-Biotite  and  Hornblende-Biotite  Granites 


Muscovite-Biotite  Granite 

Hornblende-Biotite  Granite 

Mean 

Standard 

Mean 

Standard 

Deviation 

Deviation 

Si02 

72.20 

2.22 

70.70 

2.67 

A1203 

14.54 

1.28 

14.02 

1.31 

Fe203 

0.69 

0.41 

1.03 

0.50 

FeO 

1.46 

0.88 

2.43 

0.96 

MgO 

0.58 

0.58 

0.69 

0.58 

CaO 

1.82 

0.87 

2.20 

1.07 

Na20 

3.24 

0.70 

3.36 

0.68 

K20 

4.39 

1.40 

4.31 

1.26 

Ti02 

0.25 

0.17 

0.39 

0.22 

Sample  size 

30 

48 

to  rate  a  good  partial  analysis.  In  fact, 
however,  FeO  and  Fe203  turn  out  to  be 
of  major  importance. 

The  desirability  of  the  first  requirement 
will  be  immediately  apparent  only  to 
readers  who  have  attempted  to  make  use 
of  any  of  the  standard  petrographic 
classifications  in  a  study  of  granitic  rocks. 
Although  the  classifications  without  ex- 
ception assign  very  broad  compositional 
limits  to  "granite" — up  to  80  per  cent 
Si02  in  CIPW,  or  as  little  as  5  per  cent 
quartz  in  Johannsen,  for  instance — 
petrologists  have  customarily  used  the 
term  in  a  much  more  restrictive  sense. 
Our  interest  here  is  with  real  rocks  that 
have  actually  been  described  as  granite. 


To  date,  78  analyses  satisfying  all  three 
requirements  have  been  found:  32  from 
North  America,  16  from  Finland,  and  30 
from  Japan.  Means  and  standard  devi- 
ations of  the  nine  essential  oxides  are 
shown  in  table  15.  The  difference  be- 
tween silica  averages  is  suggestive,  where- 
as that  between  the  ferrous  oxide  averages 
is  decisive.  The  range  of  Si02  in  the 
hornblende-biotite  granites  is  64.47-76.68, 
and  that  of  the  muscovite-biotite  granites 
is  67.20-75.86.  There  is  thus  a  suggestion 
that  FeO  and  Si02  may  be  merely 
compensating  for  each  other.  This,  how- 
ever, is  by  no  means  the  whole  story.  In 
subsets  containing,  respectively,  all  anal- 
yses with  (a)  more  than  69  per  cent  Si02, 


TABLE  16.     Effect  of  Various  Silica  Restrictions  on  Average  Compositions  of 
Muscovite-Biotite  (A)  and  Hornblende-Biotite  (B)  Granites 


Restriction 
of  Si02 


More  than  69.00 
Per  Cent 


More  than  72.00 
Per  Cent 


Between  71.00  and 
74.00  Per  Cent 


Per  Cent 

A 

B 

A 

B 

A 

B 

Si02 

72.86 

71.87 

73.51 

73.92 

72.74 

72.64 

A1203 

14.32 

13.72 

13.90 

13.07 

13.91 

13.41 

Fe203 

0.66 

0.93 

0.64 

0.84 

0.72 

0.76 

FeO 

1.26 

2.15 

1.09 

1.62 

1.22 

2.17 

MgO 

0.43 

0.50 

0.37 

0.24 

0.51 

0.37 

CaO 

1.67 

1.83 

1.56 

1.22 

1.57 

1.57 

Na20 

3.14 

3.37 

3.15 

3.35 

3.35 

3.44 

K20 

4.55 

4.47 

4.74 

4.98 

4.78 

4.72 

Ti02 

0.25 

0.32 

0.23 

0.22 

0.31 

0.25 

Sample  size 

26 

35 

20 

15 

15 

15 

130 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


(b)  more  than  72  per  cent  silica,  and  (c) 
between  71  and  74  per  cent  silica,  the 
amount  of  FeO  is  significantly  greater  in 
the  hornblende-biotite  granites.  These 
calculations  are  summarized  in  table  16. 
One-third  of  the  hornblende-biotite 
granites  and  two- thirds  of  the  muscovite 
granites  contain  less  than  2  per  cent  FeO, 


and  no  significant  differences  between 
these  "less  than  2  per  cent  FeO"  sub- 
groups were  found.  For  these  speci- 
mens the  distinction  between  hornblende- 
biotite  and  biotite-muscovite  granite  is 
thus  primarily  physical  rather  than 
chemical.  Work  on  this  problem  is 
continuing. 


CRYSTALLOGRAPHY 


Relationships  between  Crystal  Structure 
and  Crystal  Morphology 

J.  D.  H.  DonnayG  and  G.  Donnay 

The  second  generalization  of  the  law  of 
Bravais,  which  was  reported  under  this 
heading  in  last  year's  report  (Year  Book 
60,  pp.  208-214),  has  now  been  success- 
fully applied  to  the  unraveling  of  a 
particularly  challenging  morphology,  that 
of  the  mineral  barite.  Ever  since  Mallard 
(1879,  p.  318)  and  Friedel  (1904,  p.  339) 
applied  the  classical  law  of  Bravais  to 
barite,  the  morphological  development  of 
this  species  has  remained  an  enigma  to 
the  present  day,  even  though  the  crystal 
structure  has  been  known  for  a  long  time 
(James  and  Wood,  1925).  The  first 
generalization  (1937)  of  the  law  of 
Bravais  is  powerless,  as  was  shown  by 
Hartman  and  Perdok  (1955)  and  by 
Seager  (1959);  the  consideration  of  pseu- 
doperiods  (Hartman,  1961)  was  helpful 
but  not  entirely  satisfactory. 

Barite  has  an  ionic  crystal  structure. 
It  is  well  known  that  in  certain  simple 
ionic  structures  ions  of  equal  charges  but 
of  opposite  signs  play  the  role  of  equiv- 
alent points  when  the  law  of  Bravais  is 
called  upon  to  explain  the  morphology. 
The  punctualization  of  the  ionic  charges 
is  the  basic  postulate  in  this  interpre- 
tation. Friedel  made  use  of  it  in  the 
classical  case  of  NaCl,  the  crystal 
structure  of  which  is  governed  by  a 
face-centered  cubic  lattice,  but  where  the 
morphology  is  controlled  by  the  primitive 

6  The  Johns  Hopkins  University. 


cubic  lattice,  with  half  the  cell  edge,  that 
is  obtained  when  all  the  ions  are  replaced 
by  unit  charges  concentrated  in  the  nodes 
of  this  new  lattice.  The  sign  of  the  charge 
can  be  disregarded  because  the  strength 
of  the  bond  Na+-Cl~  is  equal  to  that  of 
the  bond  Cl~-Na+.  Friedel  explains  the 
morphology  of  calcite  in  the  same  way: 
the  rhombohedral  lattice  whose  nodes 
carry  the  double  charges  (either  positive 
or  negative)  is  the  morphological  lattice 
obtained  by  applying  the  law  of  Bravais 
of  1849;  in  this  case,  not  only  elementary 
ions  (Ca++)  but  complex  ions  (SO4)™  as 
well  are  punctualized.  The  application  of 
the  second  generalization  of  the  law  of 
Bravais  to  the  problem  of  barite  has  led 
us  to  a  new  type  of  punctualization, 
namely  that  of  pairs  of  neighboring  ions 
with  the  same  sign.  The  reasoning  pro- 
ceeds as  follows. 

Let  us  start  with  the  structural  space 
group  of  James  and  Wood  (1925),  Pnma, 
with  axial  ratios  a :  b :  c  =  1.6304:1:1.3136. 
We  first  note  that  the  dominant  general 
form  z  receives  the  symbol  (211)  in  this 
structural  setting,  whereas  it  should  be 
symbolized  (111)  from  the  morphological 
point  of  view;  it  would  then  correctly 
define  a  primitive  morphological  lattice. 
The  conclusion  is  that  all  (hkl)  faces  in 
the  structural  notation  must  obey  the 
criterion  "h  even,"  which  implies  the 
halving  of  the  structural  a  unit  length  in 
the  three-dimensional  bond  assemblage. 
This  agrees  with  a  previous  result  of 
Hartman  and  Perdok  (1955),  that  the 
energy  period  of  the  bond  chain  along  the 
x  axis  is  a/2. 


GEOPHYSICAL   LABORATORY 


131 


The  zone  of  the  (Old)  faces  is  a  simple 
zone  with  (Oil)  dominant,  which  requires 
the  corresponding  reciprocal-lattice  net 
to  be  primitive  from  the  point  of  view  of 
morphology,  that  is  to  say,  of  bonds.  But 
the  structural  net  b*c*  has  its  mesh 
centered,  owing  to  the  n  glide  plane.  We 
must,  therefore,  postulate  additional 
extinction  criteria  that  will  require  both 
k  and  I  to  be  even.  The  dominant  face  will 
have  to  be  symbolized  (022).  To  the 
reciprocal  net  (26*,  2c*)  there  should 
correspond  a  direct  net  (b/2,  c/2)  that 
will  express  the  periodicity  of  the  two- 
dimensional  bond  assemblage  of  the 
projection  of  the  crystal  structure  onto 
the  yz  plane.  Turning  now  to  the  known 
barite  structure  (fig.  1  of  James  and 
Wood,  1925),  we  actually  observe  the 
predicted  net  if  we  replace  by  equivalent 
points  the  pairs  of  neighboring  projected 
ions  with  the  same  sign. 

The  zone  of  the  (hOl)  faces  is  also  a 
simple  zone  with  unit  face  dominant, 
which  must  obey  the  criterion  "h  even" 
(see  above)  and  the  additional  criterion 
"I  even"  in  order  that  (202)  be  the 
dominant  face.  As  a  group,  the  faces 
(hOl)  do  not  occur  frequently  enough  to 
have  their  indices  co-prime:  multiplying 
all  the  indices  by  2  makes  the  faces  in  this 
zone  recede  to  their  correct  ranks  in  the 
list  of  decreasing  frequencies  predicted  by 
the  generalized  law  of  Bravais.  Whereas 
the  structure,  projected  onto  the  zx  plane, 
has  a  primitive  net  with  mesh  ca,  the 
two-dimensional  bond  assemblage  should 
have  a  primitive  mesh  (c/2,  a/2).  This 
predicted  mesh  can  indeed  be  recognized 
in  figure  1  of  James  and  Wood  (1925) 
after  the  ca  projection  is  suitably  ex- 
tended. The  pairs  of  ions  with  the  same 
sign  that  are  to  be  punctualized  are  not  as 
obvious  on  inspection  as  in  the  be 
projection:  there  are  two  kinds  of  pairs 
of  Ba  ions  and  two  kinds  of  pairs  of  S04 
ions,  according  as  the  line  segment  that 
connects  the  two  ions  in  a  pair  slopes  to 
the  right  or  to  the  left. 

The  zone  of  the  (kkO)  faces  is  a  simple 
zone  with  (210)  dominant.  The  condition 


"h  even,"  which  is  imposed  on  all  (hkl) 
faces,  is  also  the  criterion  of  the  structural 
a  glide  plane,  which  requires  the  unit 
length  a  to  be  halved  in  the  xy  projection 
of  the  structure.  This  halving  holds  for 
morphology  too:  this  zone  does  not  yield 
any  information  other  than  the  prediction 
that  both  the  projected  structure  and  the 
corresponding  two-dimensional  bond  as- 
semblage have  the  same  periodicity.  As 
shown  in  figure  1  of  James  and  Wood 
(1925)  no  punctualization  of  ions  or  pairs 
of  ions  can  be  found  to  define  a  mesh 
other  than  (a/2,  b). 

In  addition  to  planar  projections,  we 
must  consider  linear  projections,  on  the 
coordinate  axes.  Such  a  projection  of  the 
crystal  structure  has  a  one-dimensional 
bond  assemblage,  whose  period  may  be 
the  same  or  smaller.  The  relative  im- 
portances (frequencies  of  occurrence)  of 
the  pinacoids  constitute  the  experimental 
data:  we  observe  that  c  is  the  most 
frequent,  and  a  the  least  frequent. 

According  to  the  structural  space  group 
Pnma,  the  linear  projections  onto  the 
coordinate  axes  x,  y,  z  have  periods  a/2, 
b/2,  c/2,  respectively.  This  would  imply 
the  sequence  a(200),  c(002),  6(020)  as  the 
order  of  importance  of  the  pinacoids, 
which  is  contrary  to  facts.  To  express  the 
fact  that  a  is  the  least  frequent  of  the 
pinacoids,  we  must  write  it  a (400),  so 
that  the  predicted  sequence  becomes 
c(002),  6(020),  a(400).  This,  in  turn, 
requires  that  the  one-dimensional  bond 
assemblage  of  the  projection  of  the 
structure  onto  the  x  axis  have  period  a/4. 
This  prediction  can  be  checked  in  figure  1 
of  James  and  Wood  (1925) :  the  projected 
charges  along  the  a  length  look  as  follows : 

+  + +  + +  + 

where  the  first  and  the  last  pair  of  positive 
signs  are  separated  by  translation  a.  If 
all  pairs  of  equal  signs  are  considered 
equivalent  and  replaced  by  points,  the 
length  a  is  divided  by  4. 

We  must  now  check  that  the  other  two 
linear  projections  of  the  structure  are  not 
divided     by    4.     This    is    immediately 


132 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


apparent  for  the  b  axis  (fig.  1;  James  and 
Wood,  1925).  Here,  equal  numbers  of 
plus  signs  and  minus  signs  are  projected 
on  the  same  points  at  y  =  0  and  y  =  %; 
the  structure  is  composed  of  electrically 
neutral  planes;  b  is  halved,  both  for  the 
projected  (linear)  structure  and  for  its 
one-dimensional  assemblage.  The  situ- 
ation is  not  so  clear  for  the  a  axis.  Here 
the  centers  of  the  barium  and  sulfur 
atoms  do  not  lie  exactly  in  the  same  plane 
(parameters  that  should  ideally  be  equal 
to  3^3  are  found  to  be  0.333  and  0.305  by 
James  and  Wood,  who  place  their  origin 
at  a  center  of  symmetry).  Although  the 
structure  cannot  be  said  to  consist  of 
electrically  neutral  planes,  it  nevertheless 
results  from  the  stacking  of  neutral 
layers,  with  period  c/2.  This  period 
controls  the  linear  bond  assemblage  in 
keeping  with  the  morphological  symbol 
(002)  of  the  basal  pinacoid. 

Finally  we  must  justify  the  punctual- 
ization  of  pairs  of  ions.  If  we  replace  by  a 
single  central  charge  the  two  charges  of  a 
pair  of  ions  with  the  same  sign,  we  must 
introduce  a  compensating  quadrupole 
that  consists  of  the  original  two  charges 
and  two  opposite  charges  placed  in  the 
center  of  the  pair,  next  to  the  punc- 
tualized  charge.  Then  we  see  that  the 
bonds  between  successive  equipoints  of 
the  bond  assemblage  are  indeed  rigorously 
equal  in  strength.  Consider,  for  instance, 
along  the  a  length,  a  first  pair  of  Ba  ions, 
followed  by  a  pair  of  S04  ions,  itself 
followed  by  a  second  pair  of  Ba  ions.  The 
interactions  to  be  taken  into  account 
between  two  successive  pairs  of  ions  are 
of  four  kinds:  charge-charge,  charge- 
quadrupole,  quadrupole-charge,  and 
quadrupole-quadrupole.  These  interac- 
tions between  the  first  Ba  pair  and  the 
S04  pair  are  equal,  each  to  each,  by 
symmetry,  to  the  interactions  between 
the  SO 4  pair  and  the  second  Ba  pair.  The 
centers  of  charge  can  thus  be  considered 
equivalent  points  from  the  point  of  view 
of  bonding. 

We  are  indebted  to  Dr.  H.  F.  Hameka, 
Johns  Hopkins  University,  for  suggesting 


to  us  the  consideration  of  the  quadrupole. 
The  above  results  were  given  in  the 
special  issue  of  Kristallografiya  published 
in  honor  of  Professor  N.  V.  Belov. 


Lattice  Constant  Refinement 
Charles  W.  Burnham 

Practically  all  phases  of  experimental 
mineralogy  require  knowledge  of  precise 
crystallographic  lattice  constants.  Such 
values  form  the  basis  of  detailed  three- 
dimensional  crystal  structure  refinements 
as  well  as  studies  of  subsolidus  phase- 
equilibrium  relationships.  To  place  pre- 
cise lattice  constant  determination  on  a 
routine  basis  a  least-squares  technique 
for  lattice  constant  refinement  has  been 
developed  and  programmed  for  the  IBM 
7090  digital  computer. 

The  refinement  procedure  is  complete^ 
general  in  the. following  respects: 

1.  It  is  applicable  to  crystals  of  any 
symmetry. 

2.  It  will  accept  data,  from  cards  or 
tape,  either  as  angle  measurements  for 
any  wavelength  or  in  the  form  of  calcu- 
lated d  values. 

3.  Observations  may  be  suitably 
weighted  according  to  any  scheme. 

4.  Up  to  nine  systematic  correction 
terms  may  be  included  with  each  obser- 
vation. Each  term  consists  of  an  unknown 
refmable  parameter  and  a  coefficient 
whose  form  may  be  any  one  of  five 
different  types.  The  functional  form  of 
each  type  of  coefficient  is  programmed  in 
a  separate  subroutine  to  suit  individual 
experimental  conditions. 

To  allow  for  systematic  errors,  Bragg's 
law  is  modified  to  include  an  error  in  0: 

n\/2d  =  sin  (0  +  Ad)  (1) 

In  practice  n  is  absorbed  by  the  reflection 
indices  and  will  not  appear  in  subsequent 
equations.  Following  the  method  of 
Cohen  (1935),  equation  1  is  squared  and 
expanded  in  a  Taylor  series  retaining 
terms  not  involving  powers  of  the  error, 
Ad* 


GEOPHYSICAL   LABORATORY 


133 


=  sin2  0  +  sin  20  Ad      (2) 


The  term  A0  contains  all  systematic 
errors;  it  can  be  expanded  to  separate  n 
distinct  types  of  errors : 


=  sin2  0  +  2  sin  20  Adk     (3) 


-2dhki- 


/k=l 


The  error  terms,  Adk,  are  of  the  form 
Xkfk(6),  where  Xk  is  an  experimental 
factor  whose  value  is  initially  unknown 
and  is  to  be  refined,  and  7^(0)  is  a  function 
of  0,  which  generally  vanishes  at  6  =  t/2 
(Buerger,  1942;  Klug  and  Alexander, 
1954). 

Cohen's  method  may  be  generalized  by 
introducing  reciprocal  lattice  notation: 


X 

JZdhki- 


X2 

=   ~T  (Thkl'Thkl)  (4) 


Here  rhki  is  the  reciprocal  lattice  vector 
for  the  reflection  hkl.  When  the  dot 
product  is  evaluated  in  terms  of  reciprocal 
lattice  constants  equation  3  is  expanded 
and  rearranged  to  give 

h2a*2  +  k2b*2  +  Z2c*2  +  2/*/ca*6*  cos  7* 


+  2hla*c*  cos  |8*  -  -  2klb*c*  cos  a* 

(5) 


,    f     /MV        4  sin2  6 


k=l 


where 

9k(6)  =  -fk(6)  (4/X2)  sin  26       (6) 

and   e   represents   random   error  in   the 
observation. 

Equation  5  can  be  transformed  to  a 
linear  equation  in  terms  of  the  variations 
of  the  parameters  by  expansion  in  a 
Taylor  series  about  a  set  of  trial  reciprocal 
lattice  constants  and  experimental  un- 
knowns. If  only  the  first  two  terms  of  the 
expansion  are  retained,  the  transforma- 
tion yields 


Qcalc  +   S  ^~  dClj  +   S  ^v~  ^* 


fc=i  dXk 


=   (Jobs  +   6  (7) 

where  Qcaic  represents  the  left  side  of 
equation    5    evaluated    using    the    trial 


parameters,  the  a3-  are  the  reciprocal 
lattice  constants,  and 

Qobs  =  (4  sin2  0Obs)A2 

Since  equation  7  is  linear  in  terms  of  the 
parameter  variations,  8a3-  and  8Xk,  a  set 
of  m  of  these  equations,  one  for  each 
observed  6,  can  be  solved  by  standard 
least-squares  techniques  (Whittaker  and 
Robinson,  1944)  for  the  parameter  vari- 
ations, provided  that  m  ^  n  +  6  (in  the 
triclinic  case).  The  refinement  program 
generates  and  inverts  the  least-squares 
normal  equations  matrix  according  to  a 
method  developed  by  Busing  and  Levy 
(1962). 

When  each  observation  is  weighted  in 
proportion  to  its  reliability,  equation  7  is 
multiplied  by  -\/wi,  and  the  least-squares 
procedure   minimizes    ^Wiu2.    The   two- 

i 

term  Taylor  expansion  is  exact  for 
orthogonal  unit  cells,  hence  the  least- 
squares  parameter  shifts,  when  algebra- 
ically added  to  the  trial  parameters,  will 
yield  a  set  of  lattice  constants  for  which 
the  random  errors  are  minimized.  Two, 
or  perhaps  three,  consecutive  cycles  in 
which  the  corrected  parameters  from  the 
preceding  cycle  make  up  the  new  set  of 
trial  parameters  may  be  required  for 
complete  convergence  in  the  nonorthog- 
onal  crystal  systems. 

Since  the  residual,  e,  represents  the 
difference  between  an  observed  and  a 
calculated  Q,  the  standard  deviation  of  a 
measurement  of  6  must  be  converted  to 
the  equivalent  standard  deviation  of  Q. 
The  program  automatically  computes  the 
proper  least-squares  weight  for  each  Q 
according  to 

VwQ  =  -  -  =  -A — _._  0/>  (8) 


<jq       4o-0  sin  2d 

The  least-squares  standard  error  of  fit, 
corresponding  to  the  standard  error  of  an 
observation  of  Q  of  unit  weight,  is  com- 
puted after  each  cycle  of  refinement 
according  to 


(TO 


S  Wi** 


t'=l 


L  m 


n  J 


(9) 


134 


CARNEGIE      INSTITUTION      OF      WASHINGTON 


where  m  is  the  number  of  observations 
and  n  is  the  total  number  of  varied 
parameters.  The  variance-covariance  ma- 
trix, \Vi\,  of  the  varied  parameters  is 
obtained  from 

\Vi\  =  <ro\B\-i  (10) 

where  \B\  is  the  n  X  n  least-squares 
normal  equations  matrix.  The  standard 
errors  of  the  varied  parameters  are,  of 
course,  the  square  roots  of  the  diagonal 
terms  of  \Vi\. 

Following  each  least-squares  cycle  the 
new  values  of  the  direct  lattice  constants 
and  the  unit-cell  volume  are  evaluated 
using  standard  formulas  (Buerger,  1942). 
The  direct  lattice  constant  variance- 
covariance  matrix,  |  V a  \ ,  is  obtained  from 
the  reciprocal  variance-covariance  matrix 
according  to  (D.  Handwerker,  personal 
communication,  1962): 

\Vd\  =  \D\   \Vr\   \D\T        (11) 

where  \Vr\  is  the  6X6  reciprocal  lattice 
constant  variance-covariance  matrix,  con- 
taining terms  from  \Vi\  plus  appropriate 
zeros  for  nontriclinic  cases,  and 


D\  = 


da    da 
da*d6* 

da*' 


da 
dy* 


dy    dy 
Ja*db* 


dy 


(12) 


The  standard  errors  of  the  direct  lattice 
constants  correspond  to  the  square  roots 
of  the  diagonal  terms  of  \Vd\.  The 
standard  error  of  the  unit-cell  volume  is 
evaluated  in  an  analogous  manner: 


erV  = 


'E\   IF, 


E 


(13) 


where  \E\  is  the  row  vector  containing 
the  partial  derivatives  of  V  with  respect 
to  the  direct  lattice  constants. 

The  printed  results  from  each  refine- 
ment cycle  include  a  list  of  observed  and 
calculated  d  values,  the  residuals  (d0bs  — 


dcaic)  and  (Qobs  —  Qcaic),  and  the  weighted 
residuals  (dohs  —  dcaic)/<rd  and  (Qobs  — 
Ocaic)/^  based  on  the  trial  parameters. 
The  least-squares  results  contain  the 
reciprocal  lattice  constant  and  systematic 
correction  term  experimental  parameter 
shifts  and  standard  errors  in  addition  to 
the  direct  lattice  constant  shifts  and 
standard  errors.  The  asymmetric  part  of 
the  direct  lattice  constant  variance- 
covariance  matrix  is  made  available  for 
subsequent  inclusion  in  interatomic  dis- 
tance and  angle  error  computations. 

To  illustrate  the  results  obtained  with 
this  procedure,  table  17  lists  refined 
lattice  constants  for  kyanite,  Al2Si05 
(triclinic).  A  single  crystal  of  kyanite 
from  Burnsville,  North  Carolina  used  for 
intensity  measurement  for  structure  re- 
finement (Burnham,  1962),  was  used  to 
obtain  precision  Weissenberg  (Buerger, 
1937)  photographs  about  the  a,  b,  and  c 
axes.  Of  the  79  film  measurements 
employed  in  the  least-squares  analysis  23 
were  of  Old  reflections,  20  were  of  hOl 
reflections,  and  36  were  of  hkO  reflections. 
Since  the  precision  Weissenberg  film 
measurement,  /,  is  linearly  related  to  0, 
and  all  measurements  were  considered  to 
have  equal  precision,  all  0Obs  were 
weighted  unity.  Column  1  of  table  17 
lists  the  results  obtained  when  systematic 
correction  terms  were  included  to  com- 
pensate for  film  shrinkage,  specimen 
absorption,  and  camera  eccentricity.  The 
coefficients,  g(6),  of  equation  5  were 
assigned  the  following  forms  (Buerger, 
1942) : 

Film  shrinkage : 

<7(0)8hr  =  ^  (|  -  0)  sin  26 
Absorption : 

4 

A' 
Eccentricity : 

4 

V 


0(0)  abs  =  ^  cos2  6  sin  20 


^(0)  ecc  ==^-sin220 


Separate  film  shrinkage  and  absorption 
corrections  were  applied  to  data  from 


GEOPHYSICAL   LABORATORY 


135 


TABLE  17.     Kyanite  (Al2Si05)  Lattice  Constants 


Seven  Systematic 

Correction 

Terms 


No  Systematic 

Correction 

Terms 


Precession  (Skinner, 

Clark,  and  Appleman, 

1961) 


a>  A 

7.1192  ±0.0005 

7.1197  ±0.0004 

7.121  ±0.002 

6,1 

7.8473  ±0.0004 

7.8479  ±0.0003 

7.846  ±0.002 

c,  A 

5.5724  ±0.0006 

5.5736  ±0.0004 

5.577  ±0.005 

a,  deg 

89.977    ±0.005 

89.969    ±0.006 

89.97    ±0.08 

0,  deg 

101.121    ±0.005 

101.126    ±0.006 

101.15    ±0.08 

7,  deg 

106.006    ±0.003 

106.001    ±0.003 

106.00    ±0.08 

Unit-cell  volume,  A3 

293.16      ±0.06 

293.28      ±0.03 

292.74 

different  films.  One  eccentricity  term  was 
applied  to  all  observations.  Complete 
least-squares  convergence  was  attained 
after  two  iterations. 

Column  2  of  table  17  contains  the 
results  obtained  with  the  same  data  using 
no  systematic  correction  terms.  Column 
3  lists  the  results  obtained  by  Skinner, 
Clark,  and  Appleman  (1961)  with  quartz- 
calibrated  precession  data  from  another 
specimen  of  Burnsville  kyanite. 

It  must  be  emphasized  that  the  least- 
squares  standard  errors  represent  the 
precision  attainable  with  a  specific  set  of 
data.  The  precision  will,  of  course,  vary 
with  the  ratio  of  observations  to  refinable 
variables.  The  values  are,  in  general, 
conservative,  since  they  implicitly  involve 
all  correlation,  or  parameter  interaction, 
effects.  They  should  not  be  construed, 
however,  as  the  accuracy  to  be  expected 
when  several  sets  of  parameters  obtained 
by  the  same  or  different  X-ray  techniques 
on  the  different  samples  are  compared. 

The  Crystal  Structure  of  Sillimanite 
Charles  W.  Burnham 

Details  of  the  crystal  structure  of 
sillimanite  are  essential  to  an  understand- 
ing of  the  crystal  chemical  relationships 
between  the  Al2Si05  polymorphs  (anda- 
lusite,  sillimanite,  kyanite).  Because  of 
the  extreme  similarity  of  their  X-ray 
diffraction  patterns,  a  well  determined 
sillimanite  structure  must,  in  addition, 
form  the  basis  of  detailed  studies  of  the 
structures  of  mullites  of  various  compo- 


sitions. A  three-dimensional  refinement  of 
the  previously  determined  sillimanite 
structure  (Taylor,  1928;  Hey  and  Taylor, 
1931)  was  undertaken  with  single-crystal 
counter  diffractometer  data  measured  on 
a  small  cleavage  fragment  of  clear 
sillimanite  from  LaBelle  County,  Quebec. 
The  unit-cell  dimensions  of  this  specimen 
were  refined  to  the  following  values: 
a  =  7.4856  ±  0.0006,  b  =  7.6738  ± 
0.0003,  c  =  5.7698  ±  0.0008  A. 

Preliminary  least-squares  refinement  of 
the  structure  (Burnham,  1961)  reduced 
the  unweighted  disagreement  factor  R  to 
10.3  per  cent  and  the  weighted  (root- 
mean-square)  R  to  4.8  per  cent.  At  that 
stage  agreement  between  observed  and 
calculated  structure  factors  for  the  sub- 
structure reflections  {I  even)  was  excellent 
(table  18)  whereas  that  for  reflections 
with  I  odd  indicated  almost  complete  lack 
of  complement  structure  convergence. 
Attempted  refinement  of  disordered  mod- 
els and  a  noncentrosymmetric  model 
failed  to  improve  complement  structure 
agreement  but  had  little  adverse  effect  on 
the  substructure  R  value.  This  indicated 
that  the  substructure  reflections  are  very 
insensitive  to  minor  structural  changes 
and  that  substantial  convergence  of  the 
complement  structure  will  be  required 
before  the  details  of  the  structure  can  be 
evaluated. 

During  the  past  year  the  sillimanite 
study  has  been  continued.  Analysis  of  the 
refinement  procedure  demonstrated  that 
convergence  had  not  been  attained  be- 
cause of  strong  mathematical  interactions 


136 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


TABLE  18.     Sillimanite  Disagreement  Factors,  R 


Hey  and 
Taylor  (1931), 

% 


Weighted 
Refinement, 

% 


Unweighted 
Refinement, 

% 


Unweighted  R 


Weighted  R 


z\\Fo\  -  \Fe\r 

'Hw(\Fo\  -  \Fc\) 


ZwFo2 


Even-level  unweighted  R 
Odd-level  unweighted  R 


29.5 


26.2 

23.5 
61.9 


10.3 


4.8 

5.0 
40.1 


5.6 


4.7 

3.6 
16.8 


between  structure  parameters.  Least- 
squares  correlation  coefficients  between 
pairs  of  atomic  coordinates  whose  differ- 
ences determine  the  complement  struc- 
ture varied  from  —0.66  to  —0.81. 

Observed  structure  factors  had  been 
assigned  least-squares  weights  in  inverse 
proportion  to  their  variances  as  deter- 
mined by  counting  statistics.  The  distri- 
bution of  sillimanite  structure  factor 
magnitudes  in  reciprocal  space  is  not 
random;  reflections  on  odd  reciprocal 
lattice  levels  normal  to  the  c  axis  receive 
intensity  contributions  from  the  comple- 
ment  structure   alone.    Because   of   the 


resulting  unfavorable  counting  statistics 
the  average  weight  assigned  to  these 
observations  was  0.07,  compared  with  the 
average  weight  of  0.20  assigned  to  the 
substructure  observations  on  even  recip- 
rocal lattice  levels. 

Underweighting  of  the  critical  class  of 
observations  proved  to  be  the  primary 
cause  of  the  strong  parameter  inter- 
actions. When  all  structure  factors  with 
measurable  values  were  assigned  weight 
1.0  and  those  whose  values  were  below 
the  minimum  observable  value  were 
assigned  weight  0.01,  all  structure  param- 
eters were  effectively  uncoupled.   Com- 


TABLE  19.     Sillimanite  Atom  Coordinates 


Atom, 

Hey  and 

Total 

Final 

Standard 

coordinate 

Taylor  (1931) 

Change 

Error 

Oa: 

X 

0.35 

+0.0099 

0.3599 

0.0006 

y 

0.43 

3/ 

-0.0222 

0.4078 

3/ 

0.0006 

Ob: 

z 

X 

/a 
0.35 

+0.0076 

/a 
0.3576 

0.0006 

y 

0.43 

+0.0052 

0.4352 

0.0006 

z 

Va 

Va 

Oc: 

X 

0.47 

+0.0053 

0.4753 

0.0007 

y 

0.03 

3/ 

-0.0292 

0.0008 

z/ 

0.0007 

Od: 

z 

X 

/a 
0.11 

+0.0148 

/A 

0.1248 

0.0004 

y 

0.22 

+0.0037 

0.2237 

0.0004 

z 

0.5 

+0.0164 

0.5164 

0.0006 

Si: 

X 

0.14 

+0.0135 

0.1535 

0.0003 

y 

0.35 

-0.0096 

0.3404 

0.0003 

z 

% 

Va 

Ali: 

X 

0 

0 

y 

0 

0 

z 

0 

0 

Al2: 

X 

0.14 

+0.0019 

0.1419 

0.0003 

y 

0.35 

-0.0053 

0.3447 

0.0003 

z 

Va 

Va 

GEOPHYSICAL   LABORATORY 


137 


plete  convergence  was  attained  after  nine 
additional  least-squares  cycles  during 
which  all  atomic  coordinates  and  aniso- 
tropic temperature  factors  were  varied. 
The  final  R  values,  listed  in  table  18, 
confirm  Pbnm  as  the  correct  sillimanite 
space  group. 

The  refined  atomic  coordinates  are 
compared  with  those  of  Hey  and  Taylor 
(1931)  in  table  19.  Although  refinement 
produced  significant  coordinate  shifts,  it 
did  not  alter  the  basic  geometrical 
relationships  between  coordination  poly- 
hedra.  Chains  of  slightly  distorted  alumi- 
num octahedra  run  parallel  to  the  c  axis 
and  are  supported  by  double  chains  of 
aluminum  and  silicon  tetrahedra.  Differ- 
ences in  interatomic  distances  (table  20) 


TABLE  20. 

Sillimanite  Interatomic 

Distances* 

Atom 

Multi- 

Distance, 

Standard 

Pair 

plicity 

A 

Error 

Si  tetrahedron 

Si-Oa 

1 

1.629 

0.007 

Si-Oc 

1 

1.564 

0.006 

Si-Od 

2 

1.633 

0.004 

Oa-Od 

2 

2.628 

0.005 

Oa-Oc 

1 

2.608 

0.007 

Oc-Od 

2 

2.627 

0.006 

Od-Od' 

1 

2.696 

0.007 

Al  tetrahedron 

Al2-06 

1 

1.758 

0.005 

Al2-Oc 

1 

1.721 

0.006 

Al2-Od 

2 

1.800 

0.004 

06-0<* 

2 

2.834 

0.005 

Ob-Oc 

1 

2.903 

0.007 

oc-od 

2 

2.843 

0.006 

Od-cv 

1 

3.074 

0.007 

Al  octahedron 

Ah-Od 

2 

1.957 

0.003 

Ali-Oa 

2 

1.919 

0.003 

Ah-O* 

2 

1.861 

0.003 

Oa-Ob 

2 

2.893 

0.001 

Oa-Ob"  (shared)      2 

2.434 

0.006 

Oa-Od 

2 

2.776 

0.005 

(VOd 

2 

2.698 

0.005 

0d'"-0o 

2 

2.705 

0.005 

Od"'-06 

2 

2.703 

0.005 

*  Atoms  designated  with  a  single  prime 
represent  transformation  of  the  unprimed  atom 
in  the  same  coordination  polyhedron  according 
to  x'  —  x,  y'  —  y,  z'  =  }/2  —  z.  Double  primes 
represent  transformation  according  to  x"  =  —  x, 
y"  =  —y,  z"  =  l/2  -\-  z.  Triple  primes  represent 
transformation  to  a  centrosymmetric  equivalent. 


show  that  the  distribution  of  aluminum 
and  silicon  in  the  tetrahedra  is  ordered. 
Figure  43  illustrates  the  bonding  within 
and  between  coordination  polyhedra. 

The  tetrahedral  double  chains  are  of 
particular  crystal  chemical  interest.  Each 
double  chain  may  be  thought  of  as  a 
continuous  series  of  four-membered  rings, 
each  ring  containing  two  silicon  and  two 
aluminum  tetrahedra  in  the  sequence 
Si-Al-Si-Al.  The  Si-Oc-Al2  bond  angle  of 
171.6°  and  the  Si-Od-Al2  bond  angle  of 
114.4°  control  the  basic  configuration  of 
the  ring.  Whereas  the  O-Si-0  tetrahedral 
angles  are  close  to  ideal  (107.4°  to  111.3°), 
the  Si-0  bond  distances  show  that  the 
silicon  atom  is  not  at  the  center  of  its 
tetrahedron  but  is  measurably  displaced 
toward  Oc.  The  aluminum  tetrahedron  is 
more  irregular,  but,  again,  the  cation  is 
closest  to  Oc. 

The  average  Si-0  distance  is  1.615  A, 
and  the  average  tetrahedral  Al-0  distance 
is  1.770  A.  Smith  and  Bailey  (1962)  show 
that  this  average  Si-0  distance  is  close  to 
that  for  other  silicates  in  which  three 
corners  of  each  tetrahedron  are  shared 
with  other  tetrahedra.  They  also  suggest 
that  the  average  Al-0  distance  is  close  to 
the  extrapolated  value  for  layer  silicates. 
If  these  are  to  be  accepted  as  "expected" 
averages  for  sillimanite,  the  anomalous 
positions  of  the  cations  must  be  explained. 

The  anisotropic  temperature  factors 
for  Oc  indicate  a  vibrational  configuration 
corresponding  to  an  oblate  spheroid 
whose  circular  equator  lies  in  the  plane 
normal  to  the  mirror  plane  containing  Si, 
Al2,  and  Oc,  and  is  essentially  parallel  to 
the  bisector  of  the  Si-Oc-Al2  angle.  The 
root-mean-square  amplitude  of  vibration 
in  the  equatorial  section  is  0.12  ±  0.01  A; 
that  normal  to  this  section  and  directed 
toward  Si  and  Al2  is  0.06  ±  0.02  A.  The 
equivalent  isotropic  temperature  factor 
calculated  for  Oc  is  0.86,  as  compared 
with  values  ranging  from  0.35  to  0.50  for 
the  other  three  oxygen  atoms  in  the 
asymmetric  unit,  and  0.36  to  0.42  for  the 
oxygen  atoms  in  andalusite  (Burnham 
and  Buerger,  1961). 


138 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


Fig.  43.     Projection  on  (001)  of  the  refined  structure  of  sillimanite  showing  cation-anion  bonds. 
The  z  coordinate  of  each  atom  is  given  beside  its  designation. 


Since  there  is  a  local  charge  imbalance 
of  —0.25  on  Oc,  the  abnormally  short 
bond  distances  could  be  attributed  to 
excess  Coulomb  attraction  between  the 
cations  and  Oc.  The  indicated  vibrational 
anisotropy  may  consequently  arise  as 
compensation  for  an  electron-density 
distribution  related  to  anomalous  bond 
character  not  considered  in  the  spherical 
scattering  factor  curves  used  in  refine- 
ment. This  explanation,  however,  does 
not  appear  to  be  consistent.  The  Al2-0& 
distance  is  0.04  A  larger  than  the  Al2-Oc 
distance,  yet  0&  also  bears  a  charge 
deficiency  of  —0.25.  The  shortest  Si-0 
bond  in  andalusite  is  directed  to  an 
oxygen  with  a  charge  excess  of  0.2,  and 
the  two  shortest  Al-0  bonds  in  the  five- 
coordinated  aluminum  group  of  andalu- 
site involve  oxygens  with  charge  imbal- 
ances of  —0.4  and  +0.1. 

Alternatively,  the  indicated  thermal 
motion  of  Oc  may  represent  an  average 


electron-density  distribution  of  atoms 
with  normal  vibration  amplitudes  but 
different  time-average  coordinates  in 
different  unit  cells.  Under  this  hypothesis 
the  actual  position  of  Oc  from  ring  to  ring 
is  displaced  from  the  coordinates  listed  in 
table  19  to  positions  within  the  equator 
of  the  oblate  vibrational  spheroid  but  not 
necessarily  on  the  mirror  plane.  An 
approximation  to  the  resulting  effect  on 
bond  distances  can  be  computed  by 
averaging  the  distances  over  the  indicated 
thermal  motion  assuming  the  cation  and 
anion  to  vibrate  independently.  Averag- 
ing increases  the  Si-Oc  distance  to  1.576  A 
and  the  Al2-Oc  distance  to  1.732  A.  Both 
increases  are  significant  relative  to  the 
standard  errors  of  the  bond  distances,  but 
are  not  sufficiently  large  to  normalize  the 
distances. 

Further  studies  of  this  important 
crystal  chemical  problem  are  now  under 
way.  The  positional  variation  hypothesis 


GEOPHYSICAL   LABORATORY 


139 


will  be  tested  by  examining  the  structure 
at  very  low  temperatures.  If  the  oxygen 
atom,  Oc,  is,  in  fact,  statistically  distrib- 
uted, the  apparent  thermal  motion  should 
not  diminish  with  decreased  temperature. 
If,  however,  the  large  temperature  factor 
actually  represents  thermal  vibration,  it 
will  be  measurably  reduced  at  low 
temperatures. 

The  Crystal  Structure  of  Fe  Mica 
N.  Morimoto,  J.  D.  H.  Donnay,7  and  G.  Donnay 

Work  on  synthetic  iron  mica  {Year 
Book  60,  p.  214)  has  been  continued.  The 
crystal  structure  was  determined  by  the 
three-dimensional  least-squares  methods. 
The  computations  were  carried  out  at  the 
National  Bureau  of  Standards,  with  the 
help  of  Dr.  Helen  Ondik,  on  the  IBM 
7090,  using  the  modified  Busing  program. 
The  intensity  data,  without  absorption 
correction,  gave  R  —  0.23,  including  non- 
observed  reflections,  and  R  =  0.13, 
excluding  nonobserved  reflections,  after 
four  cycles  of  least-squares  refinement. 
The  absorption  correction  was  then 
applied  to  the  data  by  means  of  the 
program  of  C.  W.  Burnham.  The  cor- 
rected data  gave  R  =  0.21  or  0.09, 
according  as  the  nonobserved  reflections 
were  or  were  not  included,  after  three 
cycles  of  refinement.  These  last  compu- 
tations were  performed  on  the  IBM  7090 
of  the  Johns  Hopkins  Computing  Center, 
using  the  Trueblood  program  as  modified 
by  Koenig  with  different  isotropic  tem- 
perature factors  for  the  different  atoms. 
The  work  is  still  in  progress. 

On  the  Transitions  of  Bornite 
N.  Morimoto 

The  transition  mechanisms  of  the  three 
polymorphic  forms  of  bornite  (Morimoto 
and  Kullerud,  1961)  were  studied  from 
the  structural  viewpoint. 

The  crystal  structure  of  the  high- 
temperature  form  is  essentially  the  anti- 

7  The  Johns  Hopkins  University. 


fiuorite  structure,  only  slightly  more 
complicated.  The  sulfur  atoms  occupy 
the  nodes  of  the  cubic  face-centered 
lattice  with  a  =  5.50  A,  being  cubically 
close-packed.  Each  sulfur  tetrahedron,  on 
the  average,  contains  %  of  a  metal  atom. 
This  fractional  atom  is  itself  statistically 
distributed  over  twenty-four  equivalent 
sites  inside  the  sulfur  tetrahedron.  Thus, 
in  the  whole  unit  cell,  six  metal  atoms 
are  statistically  distributed  over  24  X  8 
=  192  sites. 

The  cubic  edifice  of  the  metastable  form 
is  a  result  of  twinning  of  a  large  number 
of  small  domains  in  eight  different 
orientations.  Each  such  crystal  has  a 
rhombohedral  cell  with  arh  =  6.70  A  and 
a  =  33°32'. 

The  structure  of  this  rhombohedral 
form  can  be  derived  from  that  of  the 
high-temperature  form  considered  along 
the  body  diagonal  (111)  of  the  cube 
(fig.  44).  All  the  sulfur  atoms  stay  in 
place,  retaining  the  cubic  close  packing. 
Of  the  four  sulfur  tetrahedra  sites,  two 
do  not  change  at  all.  One  becomes  vacant, 
and  the  metal  atom  that  occupied  it  in 
the  high-temperature  form  is  redistrib- 
uted among  the  other  three  sites.  The 
corresponding  three  sulfur  tetrahedra 
now  contain  one  full  atom  apiece.  To 
compensate  for  the  vacant  site,  the  last 
metal  site  is  slightly  displaced.  The 
statistical  distribution  of  %  of  a  metal 
atom  among  twenty-four  possible  sites 
inside  each  sulfur  tetrahedron  changes  to 
the  statistical  distribution  of  one  metal 
atom  among  four  possible  sites. 

Figure  45  shows  the  structural  relations 
between  the  high-temperature  and  the 
metastable  forms,  both  of  which  consist 
of  layers  parallel  to  (lll)rh-  Two  struc- 
tures are  built  on  the  basis  of  the  cubic 
close  packing  of  the  sulfur  atoms.  The 
statistically  distributed  metal  atoms  are 
represented  as  bands. 

The  distance  between  the  Mi  layer  and 
the  sulfur  layer  becomes  shorter  in  the 
metastable  form,  suggesting  the  possi- 
bility that  the  Fe  atoms  concentrate  in 
Mi  layers.  Although  the  structure  of  the 


140 


CAKNEGIE     INSTITUTION      OF      WASHINGTON 


SI 


A 


">- 


Mil 


18.95  A 


SI. 


Mm 


^ 


(a) 


(b) 


Fig.  44.     Derivation  of  the  structure  of  metastable  form  from  that  of  high-temperature  form. 


GEOPHYSICAL   LABORATORY 


141 


1 6.95  A 


A 

B 

i 

A 

■  • . 

B 

i 

! 

C 

1 
i 

B 

C 

i 

1 

A 

I 
i 

1 

C 

A 

if 

* 

B 

1 

i 

1 

A 

B 

1 

C 

! 

I 

B 

C 

1 

i 

A 

1 

i 

C 

A 

-* 1 

s 

0.75  M 


0.75  M 
S 
0.75  M 


0.75  M 
S 
0.75  M 


0.75  M 
S 
0.75  M 


0.75  M 
S 
0.75  M 


0.75  M 
S 
0.75  M 


0.75  M 
S 


1 r 1 

,; \s                            ... 

T 

•••'  :   .      '"''.••    "     "  .  - 

•     '    .  ■,.    "  \  '"'  -      -:  ■  .'.' 

[ i 1 

31 

Mn 


MI 

sn 


Mm 

si 

Mn 


ME 

sn 


Mm 

si 

Mn 


MI 

sn 


Mm 
si 


1 — - 

.J 

V 

, 

3 

K 

*- 

Cu 


Cu 
S 


Cu 
Cu 
S 

Cu 
Cu 


Cu 

Cu 

S 

Cu 

Cu 


Cu 
Fe 
S 


Cu 
Cu 
S 


(a) 


(b) 


(c) 


Fig.  45.     Layer  structures  of  the  three  polymorphic  forms  of  bornite. 


low-temperature  form  was  not  actually 
determined,  it  seems  likely  that  the  metal 
atom,  statistically  distributed  at  the  four 
corners  of  a  tetrahedron  in  the  metastable 
form,  will  occupy  one  of  the  four  sites  in 
the  low-temperature  form.  The  stoichio- 
metric composition  confirmed  for  most 
natural  bornite  suggests  that  the  Fe 
atoms  must  take  some  definite  positions, 
the  Mi  positions,  which  are  closer  to  the 
sulfur  atoms  than  other  metal  positions. 
This  relation  in  the  low-temperature  form 
is  shown  by  lines  in  figure  45. 

The  arrangements  of  the  metal  vacant 
layers  change  their  orientations  according 
to  a  simple  twin  law  in  the  metastable 
and  possibly  in  the  low-temperature 
forms.  Domain  structures  always  take 
place  on  transition  from  the  high- 
temperature  to  metastable  forms.  This 
indicates  that  the  metal  vacant  layers 


cannot  stably  keep  their  orientation  over 
a  long  distance.  The  diffracted  X  rays 
from  each  crystal  ( =  domain  orientation) 
are  not  coherent  with  those  from  other 
crystals.  The  domains  themselves,  how- 
ever, must  be  small,  since  the  twins 
cannot  be  recognized  as  such  by  direct 
methods  of  observation.  The  volumes  of 
the  different  domain  orientations  must  be 
nearly  equal  so  as  to  give  cubic  or 
tetragonal  symmetries. 

The  twinning  found  in  the  metastable 
bornite  is  different  from  usual  twins  in 
that  most  of  the  atoms  (Si,  Sn,  Mn,  and 
Mm)  build  a  continuous  periodic  struc- 
ture throughout  the  whole  edifice,  so  that 
the  twin  relations  apply  only  to  the  Mi 
and  the  vacant  M  positions.  Such  slight 
structural  rearrangements  take  place  that 
the  transition  heat  should  be  very  small 
and  the  transition  unquenchable. 


142  CARNEGIE     INSTITUTION      OF      WASHINGTON 

ORE   MINERALS 

Increased  emphasis  has  recently  been  the  Fe-S  system  at  temperatures  below 

placed  on  applications  of  the  synthetic  200°C. 

systems  to  ores.   Specimens  have  been  The  mineral  bravoite  was  found  to  be 

systematically  collected  in  a  number  of  stable    below     137°C    by    experiments 

mines,    and    their   mineral    assemblages  employing  the  method  of  mixing  aqueous 

have  been  studied  in  polished  sections  and  liquids.  Bravoite  is  a  common  product  of 

by  X  rays.  Employment  on  these  mineral  alteration  in  numerous  ores,  and,  since 

assemblages  of  the  geological  thermom-  its  thermal  stability  is  now  known,  its 

eters  that  we  have  developed  in  recent  presence  may  serve  as  a  valuable  geo- 

years    has    produced    interesting    new  logical  indicator.  These  experiments  also 

information  about  the  formation  of  the  indicate  that  the  a(Ni,Fe)i^J3  phase  is 

ores   but   has   also   brought   forth   new  stable  at  least  down  to  150°C,  and  they 

problems  that  demand  solution.  have  further  produced  data  enabling  us 

Progress  in  these  studies  frequently  to  draw  the  phase  diagram  for  the  system 
depends  on  development  of  new  research  at  150°C.  Differential  thermal  analysis 
methods  and  their  subsequent  systematic  and  high-temperature  X-ray  diffraction 
employment  to  produce  a  steady  flow  of  studies  of  synthetic  as  well  as  natural 
new  data.  The  investigations  started  with  pentlandite  (Fe ,  Ni)  9S8  have  demon- 
the  most  basic  systems  and  have  pro-  strated  that  pentlandite  is  stable  only 
gressed  step  by  step  to  those  sufficiently  below  610°C,  at  which  temperature  it 
complex  to  include  the  most  important  decomposes  to  pyrrhotite  (Fei_xS)  and 
minerals  of  many  common  ore  types.  To  the  Ni3±xS2  phase.  This  mineral  corn- 
date  twelve  binary  and  twelve  ternary  monly  occurs  with  pyrrhotite  in  ultra- 
systems  have  been  studied,  and  rapid  basic  rocks.  The  sulfides  are  believed  to 
progress  is  being  made  in  quaternary  have  segregated  as  liquid  drops  from  the 
systems.  rock  magma.   Earlier  interpretations  of 

Laboratory  experimentation  has  re-  the  ore  assemblages  were  based  on  the 
cently  been  facilitated  by  the  develop-  assumption  that  the  pentlandite-pyrrho- 
ment  of  a  simple  apparatus  for  mixing  tite  pair  is  stable  to  at  least  850°C.  The 
aqueous  solution  in  closed  systems  and  at  new  information  on  pentlandite  stability 
controlled  elevated  temperatures.  The  relations  necessitates  reinterpretation  of 
upper  temperature  limit  of  this  method  field  occurrences  and  has  significant  effect 
in  investigated  systems  slightly  overlaps  on  the  theory  of  formation  of  such  ores, 
the  lower  temperature  limit  for  attain-  Differential  thermal  analysis  experiments 
ment  of  equilibrium  in  corresponding  dry  have  also  demonstrated  that  a  two-liquid 
systems.  The  mineral  assemblages  and  +  vapor  region  extends  across  the  sulfur- 
solid  solution  compositions  obtained  in  rich  part  of  the  ternary  Fe-Ni-S  system 
the  overlapping  temperature  range  after  at  temperatures  above  991°C  for  the  Ni-S 
years  of  heating  of  the  dry  systems  are  side  and  above  1083°C  for  the  Fe-S  side, 
identical  with  those  developed  in  a  few  The  upper  stability  curve  of  linneite 
hours  in  the  aqueous  systems.  Therefore,  (Co3S4)  has  been  determined  up  to  2000 
when  such  identity  can  be  established,  bars.  In  the  presence  of  vapor  this 
this  method  can  be  employed  to  deter-  mineral  decomposes  to  Coi-^S  and  CoS2 
mine  the  phase  relations  in  many  impor-  at  665°C.  This  study,  similar  to  the  study 
tant  systems  at  low  temperatures  and  by  of  poly dy mite,  is  of  interest  because  the 
experiments  lasting  only  a  few  hours  each,  reaction  Co3S4  ^  Coi-^S  +  CoS2  involves 
This  method  has  already  been  applied  to  only  solids. 

parts  of  the  Fe-Ni-S  system  and  is  now  In  the  literature  several  phases  have 

being  used  to  clarify  phase  relations  in  been    reported    to    exist    in    the    Mo-S 


GEOPHYSICAL   LABORATORY 


143 


system.  However,  only  hexagonal  molyb- 
denite (MoS2)  has  been  established  as  a 
mineral  species.  Two  MoS2  forms  were 
made  synthetically,  one  rhombohedral  at 
low  temperatures  and  one  hexagonal  at 
elevated  temperatures.  Apparently  the 
rhombohedral  form  is  metastable.  The 
only  other  phase  obtained  in  the  system 
is  a  monoclinic  compound  of  approxi- 
mately M02S3  composition.  This  phase  is 
not  stable  below  610°C,  where  Mo  and 
MoS2  are  stable  together. 

The  phase  relations  in  the  Cu-Ni-S 
system  have  been  studied  at  600°C.  No 
ternary  compound  occurs,  and  solid 
solutions  extend  only  very  short  distances 
into  the  ternary  system  from  the  binary 
joins. 

The  Fe-Ni-As  system  has  been  studied 
at  800°C.  Two  ternary  phases  occur: 
(Fe,Ni)As3  solid  solution,  which  is  not 
found  in  nature,  and  an  intermediary 
solid  solution,  (Ni,Fe)2As,  corresponding 
to  the  mineral  oregonite.  Extensive  solid 
solutions  exist  between  some  of  the 
phases,  for  instance  between  FeAs  and 
NiAs. 

Studies  of  the  Fe-Mo-S  system  have 
centered  on  the  stability  relations  of  the 
pyrite  (FeS2) -molybdenite  (MoS2)  min- 
eral pair.  These  two  minerals  are  stable 
together  below  726°C.  At  this  tempera- 
ture invariant  conditions  exist  in  the 
system,  and  the  five  phases  pyrite, 
molybdenite,  pyrrhotite,  liquid,  and  va- 
por are  all  stable.  Above  the  invariant 
point  pyrite  is  no  longer  a  stable  phase  in 
the  presence  of  molybdenite,  and  pyrrho- 
tite-molybdenite  becomes  the  stable  min- 
eral pair. 

Investigations  of  the  complicated  sys- 
tem Cu-Fe-S  have  shown  that  at  various 
temperatures  the  pyrrhotite  compositions 
of  the  pyrite-pyrrhotite-chalcopyrite  as- 
semblage are  significantly  different  from 
those  of  the  pyrite-pyrrhotite  assemblage. 
These  results  indicate  that  pyrrhotite 
temperatures  determined  on  ores  con- 
taining chalcopyrite  as  well  as  pyrrhotite 
and  pyrite  are  from  45°  to  60°C  lower 
than  those  that  would  have  been  obtained 


had  chalcopyrite  not  been  present. 

Exsolution  textures  developed  on  cool- 
ing of  synthetic  bornite-type  solid  solu- 
tions have  been  correlated  to  those  found 
in  ores.  This  study  indicates  that  the 
thermal  history  of  an  ore  body  cannot  be 
surmised  from  the  presence  of  exsolution 
lamellae  of  one  mineral  in  another. 
Exsolution  lamellae,  as  shown  in  labora- 
tory experiments,  may  indicate  rapid 
cooling,  which  probably  does  not  take 
place  in  ore  deposits,  or  they  may  origi- 
nate from  a  solid  solution  of  low  concen- 
tration that  cooled  at  a  slow  rate  so  that 
the  degree  of  supersaturation  was  always 
relatively  low. 

Chalcocite-chalcopyrite  assemblages 
are  sometimes  observed  in  ores.  These 
minerals  are  incompatible  at  high  tem- 
peratures but,  owing  to  the  variation  in 
the  metal-to-sulfur  ratio  in  the  chalco- 
pyrite field,  may  form  a  stable  assemblage 
at  very  low  temperatures. 

The  phase  relations  determined  on 
synthetic  systems  have  been  applied  to 
systematically  collected  ore  specimens 
from  many  localities.  Polished-section 
studies  of  minerals  from  the  copper 
deposits  of  the  Keweenaw  peninsula 
revealed  the  presence  of  several  important 
minerals  not  previously  reported  from 
this  district.  These  mineral  associations 
give  valuable  information  about  the  phase 
relations  in  the  ternary  system  Cu-Ni-As. 

Sphalerite-pyrrhotite  and  pyrrhotite- 
pyrite  temperatures  have  been  deter- 
mined from  numerous  samples  from  the 
Brabant  Lake,  Saskatchewan,  ores;  from 
the  Ducktown,  Tennessee,  mines;  from 
the  Elisabeth  Mine,  Vermont;  from  the 
Outukompu  district  in  Finland;  and  from 
Sulitjelma,  Norway. 

The  Mo-S  System 

Ar.  Morimoto  and  G.  Kullerud 

Study  of  the  Mo-S  system  by  quench- 
ing, microscope,  and  X-ray  methods  was 
initiated  primarily  to  elucidate  the  phase 
relations  between   molybdenite    (MoS2), 


144 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


the  most  important  source  of  molyb- 
denum, and  the  other  phases  reported  in 
the  system.  The  information  obtained 
will  serve  as  a  necessary  basis  for  studies 
of  more  complicated  systems  involving 
molybdenum  and  sulfur,  such  as  Mo-Fe-S. 

Among  the  many  reported  phases  in 
this  system,  only  molybdenite,  the  hex- 
agonal form  of  molybdenum  disulfide,  is 
established  as  a  mineral  species.  Recently, 
molybdenum  sesquisulfide  (Mo2S3)  was 
confirmed  as  a  stable  phase  above  1Q00°C, 
coexisting,  depending  on  the  composition, 
with  Mo  or  MoS2  and  sulfur  vapor 
(McCabe,  1955;  Stubbles  and  Richard- 
son, 1960).  A  new  form  of  MoS2  with 
rhombohedral  symmetry  was  synthesized 
at  about  900°C  (Bell  and  Herfert,  1957). 

Rhombohedral  MoS2  has  the  cell 
dimensions  a  =  3.16  ±  0.1  A  and 
c  =  18.37  ±  0.03  A.  The  c  translation  is 
1  J/2  times  as  long  as  that  of  the  hexagonal 
form.  The  crystal  structure,  given  by  Bell 
and  Herfert  and  later  revised  by 
Semiletov  (1962),  has  the  same  kind  of 
layer  structures  as  the  hexagonal  form, 
where  Mo  atoms  are  in  triangle  prisms  of 
S  atoms.  Mo2S3  has  monoclinic  symmetry 
with  a  =  8.6335,  b  =  3.208,  and  c  =  6.092 
A,  and  (3  =  102°43/.  In  this  compound, 
however,  Mo  atoms  are  coordinated  by 
octahedral  arrangements  of  S  atoms 
(Jellinek,  1961). 

Mo2S3  appears  to  be  stable  only  above 
610°  ±  5°C.  When  the  elements  are  used 
as  starting  materials,  Mo2S3  appears 
above  610°  ±  5°C.  Below  this  tempera- 
ture Mo  and  MoS2  are  obtained.  But 
when  Mo  and  MoS2  are  the  starting 
materials,  Mo2S3  is  not  obtained  even 
after  30  days  at  650°C.  On  the  other 
hand,  once  Mo2S3  is  formed  it  does  not 
break  down  even  after  being  heated  for 
1  month  at  600°C.  The  reaction  rates  of 
the  system  are  so  slow  that  equilibrium 
assemblages  are  not  obtained  even  at 
800°C  in  a  reasonable  time.  Above  900°C, 
however,  equilibrium  is  usually  estab- 
lished in  less  than  1  week.  The  exact 
composition  of  the  Mo2S3  phase,  deter- 
mined at  935°C,  was  found  to  be  Mo2.0eS3, 


which  deviates  slightly  from  the  stoichio- 
metric ratio.  Measurements  of  the  posi- 
tions of  reflections  in  X-ray  powder 
patterns  of  "Mo2S3"  grown  in  equilibrium 
with  Mo  and  of  those  of  "Mo2S3"  grown 
in  equilibrium  with  MoS2  give  identical 
results,  indicating  a  very  limited  solid 
solution,  if  any,  in  this  phase  at  935°, 
800°,  and  700°C,  the  temperatures  of 
these  experiments. 

According  to  Semiletov,  the  structural 
differences  between  the  hexagonal  and 
the  rhombohedral  forms  of  MoS2  can  be 
explained  by  assuming  different  stacking 
orders  of  S-Mo-S  layers.  MoS2  synthe- 
sized below  900°C  gives  X-ray  powder 
diffraction  patterns  with  broad  peaks, 
and,  in  general,  the  lower  the  temperature 
of  synthesis  the  broader  are  the  peaks. 
These  poorly  defined  peaks  do  not  fit 
exactly  either  with  those  of  the  hexagonal 
form  or  with  those  of  the  rhombohedral 
form  and  are  on  the  whole  similar  to 
diffraction  effects  commonly  attributed 
to  stacking  faults  in  layered  structures. 
Above  900°C  the  peaks  become  sharp  and 
distinctly  show  the  hexagonal  pattern. 
Natural  MoS2  always  shows  the  hexag- 
onal form,  and,  once  synthesized,  the 
hexagonal  form  of  MoS2  does  not  change 
to  the  rhombohedral  form  or  to  any 
intermediate  form  even  at  low  tempera- 
tures or  after  prolonged  heating.  We 
believe  that  the  rhombohedral  form  is 
metastable  throughout  the  entire  tem- 
perature range. 

Experiments  designed  to  determine 
possible  solid  solution  on  either  side  of 
MoS2  composition  showed  that,  within 
the  limits  of  error  of  our  methods,  M0S2 
is  stoichiometric. 

The  Fe-Ni-S  System 
G.  Kullerud 

Liquid  immiscibility .  Liquid  immisci- 
bility  between  sulfides  and  silicates  has 
been  postulated  as  a  mechanism  for  the 
enrichment  of  many  important  ores 
through  magmatic  segregation  with  the 
sulfides  separated  from  the  silicate  magma 


GEOPHYSICAL   LABORATORY 


145 


by  gravity  settling.  These  sulfides  consist 
mainly  of  mixtures  of  pentlandite 
(Fe,Ni)9S8  and  pyrrhotite  (Fei_xS), 
which  are  sulfur  poor  compared  with 
sulfides  of  other  types  of  deposits.  It  was 
suggested  in  last  year's  report  that  the 
metal-rich  sulfide  drops  not  only  separate 
by  gravity  settling  from  a  silicate 
magmatic  solution  but  may,  even  before 
this  event,  have  formed  through  liquid 
immiscibility  among  the  sulfide  phases. 
This  view  is  supported  by  results  of 
recent  investigations  in  the  ternary 
system  Fe-Ni-S. 

A  region  of  liquid  immiscibility  was 
found  by  Kullerud  and  Yund  (1962)  to 
exist  in  the  Ni-S  system  above  991°C  and 


over  a  composition  range  from  54.5  to 
more  than  97  weight  per  cent  S.  Kullerud 
(Year  Book  60)  reported  the  existence  of 
a  liquid  immiscibility  region  above 
1083°C  and  over  a  composition  range 
from  46.2  to  more  than  95.5  weight  per 
cent  S  in  the  binary  system  Fe-S. 
Additional  differential  thermal  analysis 
experiments  on  ternary  compositions 
have  now  shown  that  the  liquid  immisci- 
bility region  extends  across  the  Fe-Ni-S 
system.  Figure  46  shows  the  results 
obtained  for  various  amounts  of  sulfur  in 
a  section  in  which  the  Fe/Ni  ratio  is 
constant  (61.4  Fe,  38.6  Ni  weight  per 
cent). 

In   all    experiments    with   more   than 


noo 


30       35  X     40 

(Fe.Ni)S 


Weight  per  cent 


Fig.  46.     Phase  relations  in  the  section  from  Fe,Ni  alloy  with  38.6  per  cent  nickel  to  sulfur.  Only 
the  part  containing  more  than  30  per  cent  sulfur  is  shown. 


146 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


41.5  weight  per  cent  S  a  thermal  effect, 
caused  by  the  breakdown  of  nickel- 
bearing  pyrite  (Clark  and  Kullerud, 
Year  Book  58),  was  observed  at  729°C. 
At  this  temperature  invariant  conditions 
exist,  and  the  five  phases,  nickel-bearing 
pyrite,  iron-bearing  vaesite,  hexagonal 
(Fe,Ni)i_a;S  solid  solution,  liquid,  and 
vapor,  are  all  stable.  A  second  heat  effect 
observed  at  862°C  in  all  experiments  with 
41.8  weight  per  cent  or  more  sulfur 
demonstrated  the  disappearance  of  iron- 
bearing  vaesite  from  the  section. 

Above  this  temperature  divariant  con- 
ditions exist.  The  phases  are  now 
(Fe,Ni)i_a;S  solid  solution,  liquid,  and 
vapor.  Since  the  liquid  and  vapor  both 
contain  more  than  99.9  weight  per  cent 
sulfur  the  section  is  now  binary.  Below 
862°C  it  is,  of  course,  not  binary  but 
represents  only  a  projection  onto  a  phase. 

Stoichiometric  (Fe,Ni)S  in  the  section 
is  stable  below  860°C.  Above  this 
temperature  the  solid  solution  becomes 
metal  deficient  even  in  the  presence  of 
excess  (61.4  Fe,  38.6  Ni)  alloy.  The 
melting  relations  are  similar  to  those  of 
the  Fei-zS  and  aNii-^S  solid  solutions. 

The  maximum  melting  point  is  at 
1074°C,  where  the  solidus  and  liquidus 
curves  intersect  at  about  40.5  weight  per 
cent  S.  Mix  crystals  of  this  composition 
are  the  only  ones  that  melt  directly  to  a 
liquid  of  the  same  composition  as  the 
solid.  The  corresponding  maximum  melt- 
ing point  in  the  Fe-S  system  is  at  1192°C 
and  about  38.1  weight  per  cent  S,  and  in 
the  Ni-S  system  at  992°C  and  about  38.2 
weight  per  cent  S.  Thus  in  the  ternary 
system  a  curve  marking  maximum  melt- 
ing of  the  (Fe,Ni)i_xS  solid  solution 
series  is  slightly  concave  toward  the 
sulfur  corner  (see  fig.  47).  In  a  T-X  plot 
this  curve  also  slopes  uniformly  without 
a  maximum  or  a  minimum.  In  all 
experiments  with  49.1  weight  per  cent  or 
more  sulfur  a  heat  effect  was  also  recorded 
at  1028°C.  The  liquidus  curve  on  the 
sulfur  side  of  the  maximum  melting  point 
recorded  for  various  compositions  was 
found  to  reach  1028°C  when  there  was 


38.6 

Weight  per  cent 


Ni 


Fig.  47.  Liquid  immiscibility  in  the  Fe-Ni-S 
system  is  shown  in  the  upper  part  of  the  diagram. 
The  heavy  line  extending  across  the  system  from 
1192°C  on  the  left  to  992°C  on  the  right  indicates 
compositions  and  temperatures  of  maximum 
melting  of  the  (Fe,Ni)i_IS  solid  solution  series. 


about  51.0  weight  per  cent  S.  In  all 
experiments  with  51  to  97  weight  per  cent 
S  a  single  strong  peak  was  recorded  at 
1028°C  in  addition  to  the  heat  effects  at 
726°  and  826°C. 

The  liquid  immiscibility  region  in  this 
section,  therefore,  exists  above  1028°C 
and  over  a  composition  range  extending 
from  about  51  to  more  than  97  weight 
per  cent  S. 

Pentlandite  stability  relations.  This  min- 
eral, our  most  important  source  of  nickel, 
usually  occurs  in  intimate  association 
with  pyrrhotite,  often  in  oriented  inter- 
growths  that  presumably  are  produced  by 
exsolution.  It  is  found  in  basic  rocks  like 
norites  and  may  well  be  derived  from  such 
rocks  by  magmatic  segregation.  Pent- 
landite, (Fe,Ni)9S8,  has  cubic  symmetry. 
The  literature  reports  its  melting  point 
at  about  875°C.  It  is  readily  synthesized 
in  quenching  experiments  in  closed, 
evacuated  silica  tubes  at  temperatures 
above  500°C.  Below  this  temperature 
reaction  rates  are  slow  and  considerable 
time  is  required  to  obtain  a  homogeneous 


GEOPHYSICAL   LABORATORY 


147 


product.  X-ray  diffraction  patterns  of 
these  materials  are  invariably  identical  to 
the  X-ray  diffraction  pattern  of  natural 
pentlandite  regardless  of  the  temperature 
of  synthesis.  On  studying  synthetic 
pentlandite  in  polished  sections  and  by 
using  oil  immersion,  however,  pronounced 
differences  in  textures  were  observed 
between  those  synthesized  at  500°  to 
6Q0°C  and  those  synthesized  at  700°  to 
800°C.  The  lower-temperature  products 
appeared  homogeneous  when  studied  by 
means  of  both  X  rays  and  the  microscope, 
whereas  the  higher-temperature  materials 
displayed  distinct  textures  due  either  to 
inversion  or  to  breakdown  in  the  solid 
state. 

Since  microscopical  studies  alone  could 
not  explain  the  texture  variations,  differ- 
ential thermal  analyses  were  tried;  the 
results  are  given  in  figure  48.  A  few 
milligrams  of  Lake  Toxaway  quartz 
served  as  internal  standard.  The  high-low 


inversion  in  this  material  appears  at 
573°C  both  on  heating  and  on  cooling. 
On  the  left  side  of  figure  48  are  shown  the 
heating  curves  (bottom)  and  cooling 
curves  (top)  recorded  for  synthetic  pent- 
landite of  (Fe,Ni)9S8  composition  in 
which  the  Fe:Ni  ratio  equals  1.  A  very 
strong  thermal  effect  appears  at  610°C  on 
heating  and  at  609°C  on  cooling.  A  second 
strong  peak  was  recorded  at  862°C  on 
heating  and  at  863°C  on  cooling.  The 
temperature  at  which  this  peak  occurs 
coincides  more  or  less  with  the  melting- 
point  temperature  of  about  875°C  given 
for  pentlandite  in  the  literature.  Com- 
parison of  the  two  peaks  shows  that  the 
one  at  610°C  is  at  least  as  strong  as  that 
produced  by  the  melting  process.  There- 
fore, the  lower-temperature  effect  cannot 
readily  be  explained  as  the  result  of  a 
polymorphic  inversion  but  rather  indi- 
cates the  breakdown  of  the  pentlandite 
phase. 


573°C  !  MV   > 

«%7^°r  609  C 

*  6I0°C 
i     i 


863*C     57ioc 
,862PC 


881  °C 


Fig.  48.  Differential  thermal  analysis  curves  of  synthetic  pentlandite,  heating  curve  bottom  left 
and  cooling  curve  top  left;  and  of  natural  pentlandite,  heating  curve  bottom  right  and  cooling  curve 
top  right.  Heating  and  cooling  rate  was  3°C  per  second  in  all  experiments.  The  small  peak  at  573°C 
is  due  to  inversion  in  quartz,  which  was  used  as  internal  standard. 


148 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


DTA  curves  on  pentlandite  from  the 
bottom  of  Frood  Mine,  Sudbury,  are 
shown  on  the  right  side  of  figure  48.  The 
heating  curve  is  on  the  bottom  and  the 
cooling  curve  on  the  top.  Lake  Toxaway 
quartz  was  again  used  as  internal 
standard.  The  heating  and  cooling  rates 
of  3°C/min  and  all  other  experimental 
conditions  were  the  same  as  those  for  the 
synthetic  material. 

The  first  strong  exothermal  peak,  which 
in  the  synthetic  material  appeared  at 
610°C  on  heating,  is  recorded  at  613°C  on 
both  heating  and  cooling  of  the  natural 
pentlandite.  The  second  heat  effect  is 
recorded  at  864°C  on  heating  and  at 
881°C  on  cooling.  The  small  disturbances 
recorded  between  these  peaks  (see  right 
side  of  fig.  48)  may  be  due  to  accessory 
minerals  in  the  natural  sample. 

To  determine  whether  breakdown  ac- 
tually occurs  at  610°C,  X-ray  diffraction 
films  were  made  with  a  high-temperature 
X-ray  camera  first  at  room  temperature, 
then  at  about  600°C,  then  at  about  650°C, 
and  finally  again  at  room  temperature. 
To  avoid  oxidation  the  pentlandite 
specimen  was  kept  in  a  sealed  silica  tube 
constructed  for  this  purpose.  The  first 
exposure  gave  the  pentlandite  pattern 
with  no  other  reflections.  The  second  also 
gave  the  pentlandite  pattern  but  the 
reflections  were  considerably  displaced 
from  their  positions  on  the  film  taken  at 
room  temperature.  The  displacements 
indicate  a  much  larger  unit-cell  size  at 
600°C  than  at  room  temperature.  The 
thermal  expansion  of  pentlandite  appears 
significantly  larger  than  that  reported  for 
any  other  sulfide.  The  exact  thermal 
expansion  is  being  determined.  The  films 
made  at  650°C  contained  none  of  the 
pentlandite  reflections.  Instead  they 
showed  all  the  stronger  reflections  of 
hexagonal  pyrrhotite  and  all  the  reflec- 
tions of  the  high-temperature  Ni3±xS2 
phase  described  by  Kullerud  and  Yund 
(1962).  On  cooling  to  room  temperature 
the  pattern  obtained  was  again  that  of 
pentlandite ;  no  other  reflections  appeared. 

In  the  Fe-Ni-S  system  pentlandite  lies 


on  a  straight  line  from  the  (Fe,Ni)i_xS 
to  the  Ni3±a;S2  solid  solution.  In  figure  49, 
which  shows  the  breakdown  of  pentland- 
ite schematically,  (Fe,Ni)i_xS  is  on  the 
left  side.  The  Ni  content  of  the 
(Fe,Ni)i_xS  mix  crystals  and  the  Fe 
content  of  the  Ni3±xS  phase  at  the 
temperature  of  the  breakdown  have  not 
yet  been  accurately  established.  This 
section  is  not  binary  because  of  the 
variable  metal-to-sulfur  ratios  of  the  end 
members.  Pentlandite  and  pyrrhotite  are 
stable  together  below  61Q°C.  Pentlandite 
and  heazlewoodite  are  stable  together 
below  about  550°C.  Stability  depends  in 
part  on  the  Ni-to-S  ratio:  if  this  ratio  is 
high  the  phase  may  invert  to  Ni3±xS3  at 


E    600 


(Fe,N 


1         1         1 
pyrr 

1           1           1           i           I           I 

hotite  +  Ni3+r  S2 

- 

61012 

- 

pentlandite  +  Ni3»  x  Sg 

pentlandite 

+ 

pyrrhotite 

~550 

pentlandite 
+ 

heazlewoodite 

,)      glO      20       30 

40       50       60       70       80       90   m; 
(Fe,N,)9S8                                   Nl3 

Atomic  per  cent 

Fig.  49.  Schematic  illustration  of  the  break- 
down of  pentlandite  to  pyrrhotite  +  the  Ni3±a:S; 
phase  at  610°  ±  2°C. 


535°C;  if  it  is  low,  inversion  may  take 
place  at  524°C  (Kullerud  and  Yund, 
1962).  Solid  solution  of  iron  in  heazle- 
woodite may  also  affect  the  temperature 
of  inversion  significantly.  Above  the 
temperature  of  inversion  but  below  610°C 
pentlandite  is  stable  with  the  unquench- 
able Ni3±xS2  phase.  Above  610°C  pyrrho- 
tite and  the  Ni3±xS2  phase  are  stable 
together.  The  binary  Ni3±xS2  phase  melts 
incongruently  at  806°C  (Kullerud  and 
Yund,  1962)  to  liquid  +  aNii_sS. 

Very  slight  disturbances  are  noticed  in 
the  heating  and  cooling  curves  in  the  820° 
to  830°C  region.  They  are  too  small  to  be 
caused  by  incongruent  melting  of  Ni3±xS2. 
In  DTA  experiments  on  S3mthetic  mix- 


GEOPHYSICAL   LABORATORY 


149 


tures  of  FeS  and  (Fe,Ni)9S8  in  the  1:1 
weight  per  cent  ratio  the  heat  effects 
were  again  recorded  at  610°  and  862°C. 

It  is  probable  that  a  considerable 
amount  of  pyrrhotite  is  soluble  in  the 
Ni3±xS2  phase  and  that  the  melting 
temperature  of  that  phase  increases  to 
862°C  with  increasing  pyrrhotite  content. 

Many  ores  containing  pentlandite 
formed  originally  much  above  600°C. 
Pentlandite,  therefore,  is  a  phase  that 
must  have  formed  during  the  cooling  of 
the  ore  bodies.  This  new  information  has 
important  bearings  on  the  interpretation 
of  mineral  assemblages  containing  pent- 
landite. 

Bravoite  stability  relations.  Bravoite, 
(Fe,Ni)S2,  is  a  typical  low-temperature 
mineral.  It  is  commonly  found  as  an 
alteration  product  of  pentlandite,  and  it 
often  forms  by  alteration  of  linnaeite.  It 
occurs  as  pore  fillings  and  in  cavities  in 
many  lead-zinc  ores;  it  occurs  in  certain 
sediments;  and  it  has  recently  been 
reported  (Ramdohr  and  Kullerud,  Year 
Book  60)  as  a  secondary  phase  in  certain 
chondritic  meteorites.  In  hydro  thermal- 
type  deposits  bravoite  is  one  of  the 
youngest  minerals  and  often  is  associated 
with  older  minerals  like  pyrite,  chalcopy- 
rite,  millerite,  linnaeite,  and  polydymite. 
In  ores  believed  to  have  formed  through 
magmatic  differentiation  of  sulfide  melts 
bravoite,  one  of  the  youngest  minerals,  is 
associated  with  older  pyrrhotite,  chal- 
copyrite,  pentlandite,  platinum  minerals, 
etc.  Bravoite  in  such  ores  is  formed  by 
alteration  of  older  minerals  through  the 
action  of  water. 

It  was  important  to  investigate  the 
stability  field  of  bravoite  because  of  its 
wide  geological  range  of  occurrence.  This 
was  first  attempted  (Clark  and  Kullerud, 
Year  Book  59)  by  the  dry  method 
involving  the  heating  of  mixtures  of  iron, 
nickel,  and  sulfur  in  silica  tubes.  But 
bravoite  did  not  form  in  these  experi- 
ments, which  owing  to  slow  reaction  rates 
could  not  be  performed  below  200°C. 
Next,  wet  chemical  methods  were  at- 
tempted.  Bravoite  was  precipitated   at 


room  temperature  with  ammonium  poly- 
sulfide  from  aqueous  solutions  containing 
weighed  amounts  of  dissolved  ferrous 
ammonium  sulfate  and  nickel  sulfate. 
These  precipitates,  which  were  exceed- 
ingly fine  grained,  were  next  heated  in 
silica  tubes  with  a  slight  excess  of 
ammonium  polysulfide  at  specified  tem- 
peratures and  for  specified  periods  of 
time.  The  products  of  the  heating 
experiments  were  readily  identified  in 
X-ray  diffraction  patterns  and,  much  less 
readily,  in  polished  sections. 

The  results  of  dozens  of  experiments 
are    shown    in    figure    50.    On    heating, 


400 


±    300 


200- 


100 


8        12       16      20     24     28      32      36     40     44      48 

Time  in  days 

Fig.  50.  The  curve  shows  the  rate  of  break- 
down of  precipitated  bravoite  at  various  tem- 
peratures. It  is  practically  parallel  to  the 
horizontal  axis  after  36  days  and  indicates  that 
bravoite  is  stable  below  about  140°C. 


bravoite  breaks  down  to  pyrite  +  vaesite ; 
the  rate  of  breakdown  is  given  by  the 
curve.  Below  the  curve  bravoite  persists, 
and  above  it,  has  decomposed.  Extrapola- 
tion of  this  curve  to  the  point  where  it 
parallels  the  time  axis  indicates  that 
bravoite  is  stable  below  approximately 
140°C.  This  method  is  very  time- 
consuming,  and  since  the  reactions  cannot 
be  reversed  the  temperature  derived  by 
extrapolation  of  the  rate  curve  may  be 
significantly  too  high.  To  save  time  in  this 
kind  of  experimentation  and  to  assure 
equilibrium  conditions,  a  simple  method 


150 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


was  devised  by  which  the  two  solutions 
could  be  heated  separately  to  the  desired 
temperature  and  then  mixed;  it  is 
described  in  a  separate  section.  The  first 
experiments  were  performed  at  200°C. 
The  solutions  were  heated  separately  to 
this  temperature  and  then  mixed.  The 
immediate  reaction  taking  place  on  mix- 
ing was  strongly  exotherm  and  manifested 
itself  by  raising  the  temperature  in  the 
reaction  vessel  by  about  5°C.  The 
temperature  gradually  decreased  to  200°C, 
and  the  vessel  was  thereafter  kept  at  that 
temperature  for  1  hour.  The  products 
were  pyrite  and  vaesite,  which  by  X  rays 
were  found  to  have  identical  cell  dimen- 
sions and  the  same  compositions  as  pyrite 
and  vaesite  synthesized  together  at 
20Q°C  by  dry  experimentation  over  a 
period  of  550  days.  The  identities  of  the 
products  obtained  by  these  two  methods 
at  one  and  the  same  temperature  are 
encouraging  and  indicate  that  equilib- 
rium data  may  be  obtained  by  the 
mixing-of-solutions  method  at  tempera- 
tures too  low  for  dry  synthesis.  Additional 
experiments  showed  that  bravoite  is 
stable  below  137°  ±  6°C.  The  reversi- 


200 
180- 

160- 

O 

°.    '40 

a> 

B     120 
o 

g.    100 
£     80 


60 
40 


py  +  vs. 


137  ±6 


FeS2+  bv 


bv  +  vs. 


FeS2     10     20     30     40     50     60     70     80     90    NiS2 
Weight  per  cent 

Fig.  51.  Bravoite  is  stable  below  137°  ±  6°C 
as  determined  in  experiments  involving  mixing 
of  liquid  at  elevated  temperatures.  Below  137°C 
bravoite  is  stable  with  FeS2  (pyrite  or  marcasite) 
or  with  vaesite,  depending  on  the  bulk  compo- 
sition. Above  this  temperature  pyrite  and  vaesite 
form  a  stable  mineral  assemblage. 


bility  of  the  reaction  2(Fe,Ni)S2  ^  FeS2 
+  NiS2  was  demonstrated.  In  one  experi- 
ment performed  for  this  purpose  the 
liquids  were  mixed  at  150°C,  where 
pyrite  +  vaesite  form.  Then  the  tempera- 
ture was  lowered  to  130°C  and  main- 
tained for  72  hours.  After  this  period  of 
time,  bravoite  was  detectable  in  X-ray 
diffraction  powder  patterns. 

In  figure  51  the  stability  of  bravoite  is 
shown  in  relation  to  pyrite  and  vaesite. 
Bravoite  and  FeS2  (marcasite  or  pyrite) 
are  stable  together  below  137°C  and  form 
a  common  mineral  assemblage  in  nature. 
Bravoite  and  vaesite  are  also  stable 
together  below  137°C.  This  assemblage 
was  previously  not  known  to  exist  in 
nature,  but  we  have  now  found  it  in 
specimens  from  southeast  Missouri. 

The  Fe-Mo-S  System 
G.  Kullerud  and  Peter  R.  Buseck 

Minerals  in  the  Fe-Mo-S  system  are 
pyrite  (FeS2)  and  pyrrhotite  (Fei_xS) 
along  the  Fe-S  join,  and  molybdenite 
(M0S2),  our  most  important  source  of 
molybdenum,  on  the  Mo-S  join.  In 
addition,  a  phase  of  approximately  Mo2S3 
composition  occurs  in  the  synthetic 
system  but  has  not  been  established  as  a 
mineral. 

The  phase  relations  between  pyrite  and 
molybdenite  are  of  immediate  interest 
because  these  two  minerals  occur  together 
in  the  majority  of  the  ores  mined  for 
molybdenum.  Pyrite  is  stable  to  743°C, 
where  it  melts  incongruently  to  pyrrho- 
tite +  liquid.  Pure  molybdenite  is  stable 
to  about  1350°C. 

Since  pyrite  and  molybdenite  are 
stoichiometric  compounds  as  closely  as 
can  be  determined  by  our  methods,  the 
join  FeS2-MoS2  is  essentially  binary  even 
in  the  presence  of  excess  sulfur.  This 
sulfur  is  added  to  avoid  decomposition  of 
FeS2  at  temperatures  below  its  stability 
limits  through  loss  of  sulfur  to  the  vapor 
phase.  Mixtures  containing  excess  sulfur 
were  heated  at  700°C  and  lower  tempera- 


GEOPHYSICAL   LABORATORY 


151 


tures  for  extended  periods  of  time.  Pyrite 
and  molybdenite  remained  stable  together 
in  all  these  experiments.  Determinations 
of  the  cell  dimensions  by  means  of  X-ray 
difTractometer  methods  of  both  phases 
before  and  after  heating  showed  no 
measurable  change  in  either  pyrite  or 
molybdenite.  In  polished  sections  pure 
synthetic  pyrite  appears  identical  with 
pyrite  heated  together  with  molybdenite, 
and  pure  synthetic  molybdenite  appears 
identical  with  molybdenite  heated  with 
pyrite. 

To  determine  the  solubility  of  pyrite  in 
molybdenite  and  that  of  molybdenite  in 
pyrite,  synthetic  FeS2  and  MoS2  were 
heated  together  at  724°C  for  11  days. 
Subsequent  measurements  of  d3n  of  the 
pyrite  with  Si  internal  standard  gave 
a0  =  5.418  ±  0.002  A,  which  is  identical 
with  the  values  given  by  Swanson, 
Gilfrich,  and  Ugrinic  (1955)  and  Kullerud 
and  Yoder  (1959)  for  pure  FeS2.  Measure- 
ments of  dooe  of  molybdenite  after  being 
heated  with  FeS2  using  Si02  as  internal 
standard  gave  c  =  12.294  A,  which  is 
identical  with  the  value  of  c  =  12.295  A 
given  by  Swanson,  Gilfrich,  and  Ugrinic 
for  pure  MoS2.  These  results  indicate  that 
very  little  if  any  MoS2  is  soluble  in  FeS2 
at  724°C,  and  that  very  little  if  any  FeS2 
is  soluble  in  MoS2  at  the  same  tempera- 
ture. This  conclusion  is  based  on  the 
assumption  that  if  solid  solubility  existed 
in  either  phase  it  would  have  measurable 
effects  on  the  lattice  dimensions  of  the 
host  materials. 

In  DTA  experiments  on  various  FeS2- 
MoS2  mixtures,  all  with  excess  sulfur,  a 
strong  thermal  effect  was  recorded  at 
726°  rt  3°C  both  on  heating  and  on 
cooling.  This  is  the  maximum  tempera- 
ture at  which  pyrite  and  molybdenite  can 
coexist  as  a  mineral  pair  in  the  presence 
of  vapor.  Above  this  temperature  pyrrho- 
tite  and  molybdenite  form  the  stable 
mineral  association.  The  five  phases 
pyrrhotite,  pyrite,  molybdenite,  liquid, 
and  vapor  are  all  stable  at  726°C,  and 
invariant  conditions,  therefore,  exist  in 
the  ternary  system  at  this  temperature. 


This  invariant  point  is  situated  on  the 
FeS2-MoS2  join  at  about  95  weight  per 
cent  FeS2.  This  join  is  binary  below  the 
726°C  invariant  temperature. 

Tpie  Cu-Ni-S  System 

G.  Moh  and  G.  Kullerud 

The  phase  relations  in  this  system  have 
been  studied  at  600°C  in  evacuated, 
sealed  silica  tubes.  The  phases  that  occur 
are  chalcocite  (Cu2S),  digenite  (Cu9S5), 
and  covellite  (CuS)  along  the  copper- 
sulfur  join;  heazlewoodite  (Ni3S2)  and  the 
high-temperature  Ni3±a;S2  phase  as  well  as 
Ni7Se,  millerite  (NiS),  aNii_xS,  polydym- 
ite  (Ni3S4),  and  vaesite  (NiS2)  on  the 
nickel-sulfur  join.  At  600°C  the  only 
stable  binary  phases  are  chalcocite, 
digenite,  Ni3±xS2,  aNix_xS,  and  NiS2. 
There  are  no  ternary  compounds.  The 
limited  solid  solutions  among  the  stable 
phases  and  their  stability  relations  are 
shown  in  figure  52.  At  600°C  complete 
solid  solution  exists  between  digenite  and 
chalcocite,  which  we  will  refer  to  as  the 
chalcocite  solid  solution.  However,  this 
solid  solution  does  not  extend  very  far 
into  the  ternary  system.  Experiments 
with  mixtures  of  members  of  the  chalco- 
cite solid  solution  and  Ni3±xS2,  Nii_xS,  or 
NiS  2  showed  that  the  ternary  solid  solu- 
tion extends  much  less  than  0.5  per  cent 
toward  Ni3±xS2  and  less  than  1  per  cent 
toward  both  NiS  and  NiS2.  NiS2  is  a 
stoichiometric  compound  (Kullerud  and 
Yund,  1962)  that  takes  1.0  per  cent  Cu9S5 
into  solid  solution  at  600°C.  The  aNii_xS 
phase  forms  solid  solution  with  the 
chalcocite  solid  solution.  This  solubility 
is  very  low  in  the  nickel-deficient  part  of 
the  Nii_xS  solid  solution  but  increases  as 
the  nickel  deficiency  decreases  and  is 
about  1.3  per  cent  at  the  point  of  stoichi- 
ometry.  The  Ni3±xS2  phase  that  forms 
the  most  extensive  binary  solid  solution 
of  all  the  compounds  in  this  system  also, 
expectedly,  forms  the  largest  ternary 
field.  It  extends  about  3.5  per  cent  toward 
Cu2S  and  2.5  per  cent  toward  Cu.  The 
solubility  of  sulfur  is  too   small   to  be 


152 


CARNEGIE      INSTITUTION      OF      WASHINGTON 


600  °C 


Cu9S, 
Cu2S 


3±XS2 


Weight  per  cent 


Fig.  52.     Phase  relations  in  the  Cu-Ni-S  system  at  600°C.  All  phases  or  phase  assemblages  coexist 
with  vapor,  and  the  vapor  pressure  is  that  of  the  system. 


measured  in  Cu  or  Ni  or  in  the  Cu-Ni  solid 
solution  series. 

Since  all  experiments  were  performed 
in  rigid  tubes  vapor  is  present  in  equi- 
librium with  all  phases  or  phase  assem- 
blages given  in  figure  52.  The  univariant 
assemblages  are:  chalcocite  s.s.,  vaesite, 
liquid,  and  vapor;  chalcocite  s.s.,  vaesite, 
aNii_a;S  s.s.,  vapor;  chalcocite  s.s., 
aNii-sS  s.s.,  Ni3±a;S2  s.s.,  vapor;  chalco- 
cite s.s.,  Ni3±a;S2  s.s.,  CuNi  alloy,  vapor. 
This  CuNi  alloy  contains  about  68  per 
cent  Ni  as  determined  from  tie-line 
intersections  and  by  X  rays.  Divariant 
regions  are  digenite,  liquid,  vapor;  chal- 
cocite s.s.,  vaesite,  vapor;  chalcocite  s.s., 
aNii_xS  s.s.,  vapor;  chalcocite  s.s.,  Ni3±xS2 
s.s.,  vapor;  chalcocite  s.s.,  CuNi  alloy 
(ranging  in  composition  from  pure  Cu  to 
CuNi  with  about  68  per  cent  Ni),  vapor; 
Ni3±a;S2  s.s.,  NiCu  alloy  (ranging  in 
composition  from  pure  Ni  to  NiCu  with 
about  32  per  cent  Cu),  vapor. 


The  Fe-Ni-As  System 

Peter  R.  Buseck 

Eleven  binary  compounds  occur  in  the 
system  Fe-Ni-As.  Along  the  nickel-arsenic 
join  these  include  Ni5_zAs2  in  both  a 
stable  (jS)  and  a  metastable  (/3r)  form;  a 
low-temperature,  possibly  metastable, 
phase  of  approximate  composition  Ni2As 
(Heyding  and  Calvert,  1957);  NinAs8, 
corresponding  to  the  mineral  maucherite ; 
Nii_zAs,  corresponding  to  the  mineral 
niccolite;  and  NiAs2  in  two  polymorphs 
that  correspond  to  the  minerals  rammels- 
bergite  and  pararammelsbergite.  The 
compounds  Fe2As,  FeAs,  and  FeAs2  are 
stable  along  the  Fe-As  join;  of  these  com- 
pounds only  FeAs  2  has  an  established 
mineral  equivalent,  loellingite.  Complete 
solid  solution  exists  above  912°C  between 
Fe  and  Ni.  Below  this  temperature  the 
solid    solution    is    limited,    with    stable 


GEOPHYSICAL   LABORATORY  153 

a  (bcc)  and  y  (fee)  phases  and  a  meta-  contains    Co.    The    composition   of   the 

stable  a2  (bcc)  phase  occurring.  A  binary  other  ternary  phase,  the  " intermediary 

phase,    Ni3Fe,    is    stable    below    503°C  solid   solution,"   is   not  so   well  known, 

(Hansen  and  Anderko,  1958) ;  awaruite  is  primarily   because    of   the    difficulty    of 

its    mineral    equivalent.    At    least    two  detecting  it  optically.  In  reflected  light  it 

ternary  compounds  exist  in  the  system,  is  white,  has  a  high  reflectivity,  is  weakly 

Of  these,  the   (Fe,Ni)As3  phase  is  not  anisotropic,   and  is  practically  indistin- 

established  as  a  mineral  species,  whereas  guishable  from  synthetic  maucherite  and 

the  other,  having  an  approximate  compo-  j3Ni5_xAs2.  For  this  reason  the  determina- 

sition  (Fe,Ni)2As,  evidently  corresponds  tion  of  its  solid  solution  range  is  based 

to  the  mineral  oregonite.  Knowledge  of  largely    on    X-ray    diffraction    studies. 

the  phase  relations  in  this  system  is  an  Unfortunately,  this  method  is  not  precise 

important   step   in   our   efforts   to   gain  because  a  phase  present  in  only  a  few  per 

understanding  of  the  conditions  prevail-  cent  cannot  be  detected.  However,  the 

ing  during  formation  of  the  "magmatic  (Fe  +  Ni)/As  ratio  is  about  2,  and  its 

segregate"  type  of  ore.  Moreover,  this  is  solid-solution  range  extends  from  at  least 

one  of  the  bounding  systems  of  the  very  Ni/Fe  =  5  to  Ni/Fe  =  1. 

important  quaternary  system  Fe-Ni-As-S,  The  X-ray  pattern  of  the  "intermedi- 

which  includes  many  minerals  of  common  ary  solid  solution"  corresponds  closely  to 

occurrence  both  in  arsenide  and  sulfide-  that  published  by  Ramdohr  and  Schmitt 

type  deposits,  and  thus  provides  a  link  (1959)  for  the  mineral  oregonite,  found 

between  the  two.  with    awaruite    in    Josephine    County, 

The  system  was  studied  at  800°C  in  Oregon.  The  composition  given  by  Ram- 
evacuated,  sealed  silica  glass  tubes  in  dohr  and  Schmitt  for  oregonite,  however, 
which  vapor  was  always  present.  Owing  does  not  lie  within  the  range  of  the 
to  the  very  slow  reaction  rates  of  the  synthetic  solid  solution  field  at  800°C, 
arsenides,  high  temperatures  are  required  but  it  is  known  only  from  X-ray  fluores- 
for  the  attainment  of  equilibrium  within  cence.  At  the  low  temperature  at  which 
a  reasonable  period  of  time.  All  the  the  oregonite  presumably  formed,  the 
results  were  obtained  from  experiments  ''intermediary  solid  solution"  may  extend 
involving  heating  for  at  least  1  month,  to  the  oregonite  composition. 
During  this  period  the  materials  were  A  number  of  the  tie  lines  for  the  system 
reground  one  or  more  times  to  facilitate  Fe-Ni-As  that  have  been  located  are 
the  reactions.  listed  below.  Those  described  by  Rose- 

Because  of  the  sluggish  reaction  rates  boom  (1958)  for  the  higher  arsenides  are 

and  the  appreciable  solid  solution  between  not  included.  Tie  lines  run  from  a2Fe-Ni 

many  of  the  phases  some  of  the  tie  lines  to    Fe2As,    to    the    "intermediary    solid 

have  so  far  been  located  only  approxi-  solution,"  and  to  j8Ni5_xAs2.  Others  ex- 

mately.    Much    effort    was    devoted    to  tend  from  the  "intermediary  solid  solu- 

verifying  the  stability  and  determining  tion"   to  maucherite,  j8Ni5_xAs2,  Fe2As, 

the  composition  of  the   "intermediary"  FeAs,  to  both  the  Fe-rich  and  Ni-rich 

or  (Fe,Ni)2As  solid  solution.  sides  of  the  solvus  for  the  solid  solution 

Both   ternary   phases   form   extensive  series  between  FeAs  and  niccolite,  and 

solid  solution.  The  field  of  stability  of  the  toward  niccolite  solid  solution. 

(Fe,Ni)As3   phase   does    not   extend    to  There  are  several  univariant  regions, 

either  of  the  binary  end  members.   Its  Those   containing   a2Fe-Ni   also   include 

stability  range  has  been  investigated  by  aFe  and  Fe2As,  Fe2As  with  "intermediary 

Roseboom  (1962)  and  Pleass  and  Heyding  solid  solution,"  and  "intermediary  s.s." 

(1962).    The    compound    (Fe,Ni)As3    is  with  jSNis-^As-i.  Other  univariant  regions 

analogous   to   the  mineral   skutterudite,  containing   the    "intermediary   s.s."    are 

though   apparently   skutterudite   always  those  with  Fe2As  and  FeAs,  jSNi5_a;As2 


154 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


and  maucherite,  and  two  phases  of  the 
FeAs-niccolite  solid  solution  series. 

There  are  also  a  number  of  prominent 
divariant  regions  such  as  the  one  between 
/3Ni5_a;As2  and  7Fe-Ni  and  those  between 
the  "intermediary  s.s."  and  Ni-bearing 
FeAs  as  well  as  Fe-bearing  niccolite. 

Grains  whose  compositions  lie  within 
the  two-phase  regions  between  the  "inter- 
mediary s.s."  and  Fe-bearing  niccolite  or 
Ni-bearing  FeAs  commonly  display  a 
distinctive  myrmekitic  texture.  It  is 
extremely  fine  grained  and  visible  only 
under  very  high  magnification;  even 
then  the  phases  present  cannot  be 
optically  identified.  However,  it  appears 
that  in  some  samples  both  host  phases 
contain  these  intergrowths,  whether  as 
the  result  of  decomposition  of  a  solid 
solution  on  quenching  or  as  the  result  of 
a  liquid's  having  been  present  is  not  clear. 

Evidence,  as  yet  inconclusive,  has  been 
found  for  the  existence  of  a  third  ternary 
phase  situated  in  the  field  bounded  by  the 
uni variant  region  containing  c^Fe-Ni  and 
the  "intermediary  s.s."  together  with 
Fe2As  on  the  one  hand  and  /3Ni5_xAs2  on 
the  other. 

At  lower  temperatures  awaruite  (Ni3Fe) 
appears  as  a  phase.  Oregonite,  in  its  only 
known  occurrence,  is  associated  with 
awaruite.  As  the  tie  lines  at  800°C  run 
from  a2Fe-Ni  to  /3Ni5_xAs2,  a  switch  in  tie 
lines  is  necessary  to  establish  oregonite 
and  awaruite  as  a  stable  mineral  pair  at 
low  temperatures. 

The  Cu-Fe-S  System 

Pyrrhotite-Pyrite-Chalcopyrite  Relations 
K.  v.  Gehlen8  and  G.  Kullerud 

The  composition  of  pyrrhotite  when 
deposited  in  equilibrium  with  pyrite  is  a 
useful  indicator  of  the  temperature  con- 
ditions that  existed  when  the  assemblage 
formed,  provided  that  the  pyrrhotite 
maintained  its  original  iron-to-sulfur  ratio 
during  the  subsequent  cooling  process. 

8  University  of  Erlangen-Ntirnberg. 


Most  ores  that  contain  these  two  min- 
erals also  contain  additional  minerals  such 
as  chalcopyrite,  sphalerite,  galena,  and 
often  small  amounts  of  magnetite,  some 
of  which  are  known  to  form  measurable 
solid  solution  with  one  or  both  members 
of  the  pyrrhotite-pyrite  pair.  The  pyrrho- 
tite solid  solution  is  then  no  longer  binary 
but  becomes  ternary  or  even  more 
complex,  and  application  of  the  phase 
relations  in  the  strictly  binary  synthetic 
system  Fe-S  to  such  ores  becomes 
hazardous.  Chalcopyrite  is  perhaps  the 
most  common  mineral  occurring  with  the 
pyrrhotite-pyrite  pair,  and  pyrrhotite  can 
take  significant  amounts  of  copper  into 
solid  solution.  Therefore,  it  was  of 
interest  to  investigate  whether  pyrrhotite 
compositions,  as  determined  by  di02 
measurements,  in  ternary  pyrrhotite- 
pyrite-chalcopyrite  assemblages  at  con- 
trolled temperatures  would  coincide  with 
the  values  given  by  Arnold  for  binary 
pyrrhotite-pyrite  assemblages.  Mixtures 
of  iron,  copper,  and  sulfur  with  bulk 
compositions  inside  the  univariant  pyr- 
rhotite-pyrite-chalcopyrite  field  were 
heated  in  evacuated,  sealed  silica  tubes  at 
various  temperatures.  Measurements  of 
the  dw2  value  of  the  pyrrhotite  synthe- 
sized in  equilibrium  with  pyrite  and 
chalcopyrite  at  600°C  gave  2.0532  ± 
0.0017  A.  Pyrrhotite  synthesized  in  equi- 
librium with  pyrite  in  the  absence  of 
chalcopyrite,  at  the  same  temperature, 
gives  dw2  =  2.0497  =b  0.0007  A  according 
to  Arnold  (1962).  This  value  is  identical, 
within  the  limit  of  error  of  our  measure- 
ments, with  our  results  on  pyrrhotite 
synthesized  in  equilibrium  with  pyrite  at 
600°C. 

The  difference  in  d  values  of  pyrrho- 
tites  synthesized  with  and  without  chal- 
copyrite present  is  significant  to  geo- 
logical thermometry.  Application  of  the 
d-versus-  composition  and  T-versus-com- 
position  curves  by  Arnold  would  only  in 
the  second  case  lead  to  the  correct 
temperature  estimate  of  600°C.  The  d 
value  of  pyrrhotite  coexisting  with  chalco- 
pyrite   and    pyrite    would    indicate    a 


GEOPHYSICAL   LABORATORY  155 

temperature    of    only    about    550°C.    A  chalcocite  and  bornite-chalcopyrite,  sug- 

positive  correction  of  about  50°C  would  gests  that  lamellae  may  be  retained  as  an 

therefore  be  required  at  this  temperature,  exsolution  texture  only  when  the  solid 

The    magnitude     of    this     correction  solution  is  cooled  from  above  the  solvus 

depends  on  the  variation  with  tempera-  in    relatively    short    periods.    Complete 

ture  of  the  copper  content  of  the  pyrrho-  migration  to  grain  boundaries  or  a  mutual 

tite  phase  and  the  composition  of  the  boundary    texture    results    if    the    solid 

chalcopyrite   phase.    That   the   required  solution  is   cooled   over  longer   periods, 

correction    diminishes    with    decreasing  The   common   occurrence   of   exsolution 

temperature  is  probable  but  remains  to  lamellae   in   ore   sulfides   has   led   some 

be  shown.  recent  investigators  (e.g.,  Lyon,  1959)  to 

The  solubility  of  copper  in  pyrrhotite  conclude  that  some  ore  bodies  cooled  from 

exceeds  3  weight  per  cent  at  700°C  (Yund  600°  to  200°C  in  a  matter  of  minutes  or 

and  Kullerud,  Year  Book  S 9),  and  we  have  hours.    Clearly,    masses    of    ore,    some 

now  determined  the  solubility  at  600°C  comprising  millions  of  tons,  cannot  cool 

to  be  about  2  weight  per  cent.  by  conduction  at  so  rapid  a  rate. 

Sphalerite  commonly  coexists  with  the  The   present   study   was   initiated   to 

pyrrhotite-pyrite  mineral  pair;  however,  investigate  this  paradox  and  to  gain  a 

the   solubility   of   ZnS   in   pyrrhotite   is  better  understanding  of  exsolution  tex- 

negligible  even  at  very  high  temperatures  tures. 

(Kullerud,    1953).    For   this   reason   the  A  study  of  the  literature  of  metallurgy 

presence  of  sphalerite  should  not  measur-  and    solid-state    physics    (e.g.,    Geisler, 

ably  affect  the  pyrrhotite-pyrite  solvus.  1951;    Baker,    Brandon,    and    Nutting, 

A    situation    very    similar    to    that    for  1959)    reveals   that   exsolution   lamellae 

sphalerite  exists  for  galena,  which  is  also  need  not  necessarily  be  formed  only  by 

a  common  mineral  in  pyrrhotite-pyrite  rapid  cooling.  Lamellae  are  retained  as 

ores.  the  stable  exsolution  texture  if  the  degree 

On  the  contrary,  the  presence  of  iron  of  supersaturation  is  low,  in  other  words, 

oxides  with  the  pyrrhotite-pyrite  assem-  if  the  solid  solution  is  initially  dilute  or 

blage  may  affect  the  pyrrhotite  compo-  the  cooling  rate  is  slow, 

sition  significantly  since  pyrrhotite  is  very  The  solid  solutions  involving  bornite 

susceptible  to  oxidation  (Kullerud,  Year  (digenite-bornite,  chalcocite-bornite,  and 

Book  56).  Numerous  such  ores  contain  chalcopyrite-bornite)  in  the  system  Cu- 

small  amounts  of  magnetite.  The  effect  Fe-S  were  chosen  for  experimental  study 

of  its  presence  on  the   pyrrhotite  geo-  because  phase  relations  are  fairly  well 

logical  thermometer  may  be  significant,  understood,  because  extensive  solid  solu- 
tion    occurs,     and     because     exsolution 

_,      ,     .      _                   .  _         .     «  , . ,  textures   in   this   system   are   commonly 

Exsolution  1  extures  and  Rates  in  bolid  ■ 

Solutions  Involving  Bornite  Lamellae '  have  now  been  obtained  in 

P.  R.  Brett  runs  cooled  at  rates  as  low  as  3°C  per  day 

for  6  months.  As  a  general  rule,  the  more 

Exsolution  textures.  Very  little  is  known  concentrated  the  initial  solid  solution,  the 

of    the    contribution    of    diffusion    and  less   common  are  lamellae  as  the   final 

exsolution  to  the  formation  of  textures  exsolution    product.    Lenses    or    mutual 

during  the  cooling  of  ores.  There  has  been  boundary  textures  are  the  end  products 

little   systematic   experimental   work   to  of  exsolution  in  the  more  concentrated 

back    up    the    interpretation    of    such  solid    solutions.     Lamellae    were    often 

textures.  observed  in  combination  with  a  mutual 

All  previous  work  on  exsolution  tex-  boundary  texture,  suggesting  that  either 

tures    in    sulfides,    mainly    on    bornite-  some  of  the  lamellae  did  not  coalesce  to 


156  CARNEGIE     INSTITUTION     OF      WASHINGTON 

form  irregular  grains  or  there  was  a  late  rigorous  theory  of  kinetics,  for  a  reaction 

stage  of  formation  of  lamellae.  rate  in  the  solid  state  is  dependent  not 

In  addition  to   lamellae  and   mutual  only  on  such  variables  as  concentration, 

boundary  textures,   "veining"   and   "re-  temperature,  and  pressure  but  also  on  the 

placement"    relations   were   occasionally  rate  of  nucleation,  diffusion,  recrystalliza- 

observed.  The  "veining"  textures  result  tion,  etc.   Moreover,  the  difficulties  in- 

from  the  exsolved  phase  depositing  along  volved  in  determining  the  exact  time  for 

a  continuous  series  of  grain  boundaries  a  reaction  to  proceed  to  a  certain  point 

(fig.     53,    pi.     2).     The     "replacement"  are    considerable.    The    composite    rate 

textures,   in  which   the   exsolved   phase  cannot  be  quantitatively  considered  by 

appears  to  replace  the  host  phase,  were  separate  treatment  of  each  process,  for 

observed  in  chalcopyrite  exsolved  from  the  effect  of  individual  variables  cannot 

bornite  (fig.  54,  pi.  2).  Eutectoid  textures  be   isolated.    Nevertheless,   such   studies 

observed  as  products  of  exsolution  in  the  can  at  least  ascertain  whether  reequili- 

bornite-digenite    and    bornite-chalcocite  bration  in   the   system   studied   can  be 

pairs    are    similar    to    those    commonly  expected    during    the    slow    cooling    of 

present  in  these  minerals  in  ores.  natural  mineral  assemblages.   The  time 

The  results  of  this  study  indicate  that  taken  for  a  reaction  like  exsolution  to 
few  interpretations  from  textural  evi-  proceed  to  a  certain  point  can  be  deter- 
dence  may  be  made  on  the  thermal  mined  only  by  measurement  of  the 
history  of  minerals  that  form  solid  change  in  a  composition-dependent  prop- 
solution  pairs.  Exsolution  lamellae  can  erty  such  as  cell  edge,  hardness,  or 
indicate  extremely  rapid  cooling  (which  magnetic  susceptibility, 
is  not  to  be  expected  in  mineral  deposits)  The  rates  of  exsolution  in  the  solid 
or  cooling  of  a  solid  solution  of  a  relatively  solution  field  bornite-digenite-chalcopy- 
low  initial  concentration,  in  which  the  rite  in  the  system  Cu-Fe-S  were  chosen 
cooling  rate  was  such  that  the  degree  of  for  the  present  study.  This  system  was 
supersaturation  was  never  high.  Veining,  selected  because  phase  relations  are 
pseudo-replacement,  and  mutual  bound-  relatively  well  known  and  because  there 
ary  textures  can  occur  as  products  of  have  been  suggestions  in  the  past  that 
exsolution.  The  utmost  caution  must  be  solvi  in  this  system,  when  determined, 
taken  in  the  interpretation  of  the  textural  would  be  useful  for  geologic  thermometry, 
relations  between  any  minerals  that  may  The  change  in  composition  during 
form  solid  solution  pairs.  exsolution  could  be  ascertained  by  meas- 

Rates  of  exsolution.  It  has  long  been  uring  the  a  cell  edge  of  exsolved  bornite. 

suspected  that  many  ore  minerals  and  This  cell  edge  varies  markedly  with  the 

mineral  assemblages  remain  unchanged  Cu/(Cu  +  Fe)  ratio.  Solid  solutions  of 

during  the  cooling  period,  in  this  way  various  compositions  about  bornite  along 

retaining  the  evidence  required  to  infer  the    bornite-digenite    and    bornite-chal- 

the  conditions  during  ore  deposition.  It  copyrite  joins  were  prepared  at  700°C. 

is  on  the  supposition  that  many  systems  They  were  then  annealed  or  cooled  to 

do  not  equilibrate  with  falling  tempera-  400°,   300°,   200°,   and   50°C  in  various 

ture  that  the  principles  of  geothermom-  times.  If  the  cell  edge  of  the  bornite  was 

etry  are  based.  constant  for  a  given  temperature  below 

An    understanding    of    the    extent    of  the    solvus    regardless    of    its    original 

equilibration  of  mineral  systems  can  be  composition,  equilibrium  was  assumed  to 

obtained  only  by  the  study  of  reaction  have  been  attained. 

rates     at     different     temperatures     and  All    runs    above    50°C    were    held    at 

pressures.  Unfortunately,  data  on  rates  of  temperature  for  2  months  or  more,  and 

solid-state  reactions  such  as  exsolution  equilibrium  was  attained  in  all.  When  the 

cannot    be    considered    in    terms    of    a  solid  solutions  were  cooled  to  50°C  in  3 


GEOPHYSICAL   LABORATORY  157 

months,  equilibrium  was  also  attained;  rapidity  in  nature.  Unfortunately,  these 

the  same  was  observed  when  they  were  systems  are  useless  as  potential  geother- 

cooled  from  600°  to  50°C  in  1  hour.  mometers,   because   the   exsolved   phase 

When  solid  solutions  of  chalcopyrite  in  consistently    migrates    out    of    the    host 

bornite   were   annealed   at    100°C   for  2  mineral,    making    reconstitution    of    the 

weeks,  equilibrium  was  attained,  whereas  original   solid    solution   impossible.    The 

digenite  in  bornite,  held  at  50°C  for  3  minerals  with  greatest  potential  as  geo- 

months,  approached  equilibrium  within  thermometers    are    therefore    the    most 

approximately  3  weight  per  cent.  When  refractory  ones  (such  as  arsenides,  oxides, 

solid  solutions  of  either  chalcopyrite  or  pyrite,  and  sphalerite)  and  the  ones  most 

digenite  in  bornite  were  cooled  from  above  difficult  to  study, 
the    solvus    to    50°C    in    7.5    minutes, 

equilibrium  was  closely  approached,  the  Chalcocite-Chalcopyrite  Assemblages 

variation  in   cell   edge  being   10.944   to  P  R  Brett 
10.950  A  (±0.005  A). 

All  the  runs  in  which  disequilibrium  In  the  course  of  the  investigation  of 

was  most  pronounced  were  those  in  which  bornite-chalcopyrite  exsolution  textures, 

the  original  solid  solution  was  dilute.  In  bornites  with  maximum  sulfur  content 

many    runs    the    dilute    solid    solution  were    prepared    along    the    bornite-chal- 

exhibited  no  exsolution  at  all  after  being  copyrite  join  at  700°C  (see  Year  Book  59, 

cooled  or  annealed  in  spite  of  the  fact  that  figs.  43-45).  Chalcopyrite  exsolved  from 

more    concentrated    solid    solutions    ex-  the  bornite  on  annealing  or  cooling  to 

solved  to  equilibrium.  Doubtless  a  nucle-  50°C ;  another  phase  was  seen  in  amounts 

ation    problem    is    involved;    the    more  insufficient  for  determination  by  X-ray 

concentrated  solids  have  a  greater  degree  diffraction.  By  reason  of  the  geometry  of 

of    supersaturation    at     the     annealing  the  Cu-Fe-S  phase  diagram,  the  appear- 

temperature,  hence  have  a  greater  ten-  ance  of  the  phase,  and  the  composition  of 

dency  to  nucleate.  the  runs,  this  phase  is  probably  chalcocite. 

It  may  be  concluded  from  the  study  The  tie  line  chalcopyrite-chalcocite  is 

that  rates  of  exsolution  (and  indirectly  of  possible  only  because  chalcopyrite  always 

solid  diffusion)  are  rapid  in  this  part  of  contains  less  sulfur  than  is  indicated  by 

the  Cu-Fe-S  system.  Complete  equilibra-  its  stoichiometric  formula,  so  that  chal- 

tion  would  be  expected  in  nature  in  times  cocite,  bornite,  and  chalcopyrite  are  not 

of  the  order  of  1  hour.  exactly  collinear. 

To  check  the  extent  of  equilibration  of  Chalcopyrite  formed  at  700°C  is  more 

natural    bornites    the    a    cell    edges    of  deficient    in    sulfur    than    chalcopyrite 

thirteen    bornites    from    nine    different  formed  at  lower  temperatures  (Yund  and 

localities  and  environments  were  meas-  Kullerud,    Year  Book  59).   Accordingly, 

ured  by  means  of  the  X-ray  diffractom-  chalcopyrite  exsolving  from  a  chalcopy- 

eter.  Except  for  the  anomalous  red  bed  rite- bornite  solid  solution  must  become 

bornites    mentioned    elsewhere    in    this  more  sulfur  rich  as  cooling  proceeds.  The 

report,  all  cell  edges  correspond  to  those  bornite  in  equilibrium  with  the  exsolving 

of  stoichiometric  bornite  (10.950  =fc  0.005  chalcopyrite  must  therefore  become  in- 

A).  This  is  further  evidence  that  equi-  creasingly  poor  in  sulfur  (and  iron)   as 

librium  is  complete  in  nature.  exsolution   proceeds,   and   must   form   a 

The  fact  that  equilibration  can  occur  chalcocite-bornite  solid  solution  that 
in  so  short  a  time  casts  grave  doubts  on  breaks  down  at  low  temperature, 
the  use  of  this  part  of  the  system  for  Chalcocite-chalcopyrite  has  often  been 
geothermometry.  The  majority  of  sulfide  observed  as  a  natural  assemblage,  par- 
systems,  like  the  minerals  in  the  system  ticularly  under  supergene  conditions.  In 
Cu-Fe-S,    also    equilibrate    with    great  view  of  the  collinearity  mentioned  above, 


158 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


it  has  invariably  been  taken  to  be  a 
disequilibrium  assemblage.  The  present 
work  suggests  that  this  need  not  neces- 
sarily be  so. 

Studies  along  the  join  bornite-digenite 
previously  reported  by  Kullerud  were 
continued  down  to  50°C.  The  join 
bornite-digenite  was  found  to  exist  at 
least  to  50°C.  Therefore,  a  pyrite- 
chalcocite  join  is  impossible  below  the 
bornite-digenite  immiscibility  gap.  Above 
this  solvus  bornite,  digenite,  and  chal- 
cocite  form  a  complete  solid  solution  field 
(Yund  and  Kullerud,  Year  Book  59), 
again  prohibiting  a  chalcocite-pyrite  join. 

The  assemblage  pyrite-chalcocite  has 
commonly  been  reported  in  copper-iron 


were  not  formed  simultaneously.  The 
chalcocite  was  probably  formed  at  low 
temperatures  where  reaction  rates  are 
slowest. 

The  persistence  of  pyrite  with  chalco- 
cite in  both  hypogene  and  supergene  ores 
can  only  be  ascribed  to  the  lack  of 
reactivity  of  pyrite.  In  view  of  this 
inertness,  all  sulfide  assemblages  involv- 
ing pyrite  cannot  definitely  be  regarded 
as  equilibrium  assemblages  until  more 
conclusive  evidence  is  available. 

The  relationships  between  composi- 
tions and  the  a  cell  edge  of  bornite  solid 
solutions  were  determined  by  measuring 
the  20  values  of  the  (440)  reflection  on  the 
X-ray  diffractometer.  Provided  that  the 


c 


o| 


10.950' 

1 

"i 1  -         I        " 

-j                   | 

10.930 

- 

10.910 
in  «or> 

• 

i                   l      _ 1 

I         1 

100 

Cu5FeS4 


95 


90 


85 


80 


75 


70 


65 


>-CuFeS2_x 


Moi  per  cent 


Fig.  55.     Variation  in  the  a  cell  edge  of  bornites  with  maximum  sulfur  content  in  the  bornite- 
chalcopyrite  solid  solution  field  at  700°C. 


sulfide  deposits.  Although  much  of  this 
chalcocite  may  be  misidentified  digenite, 
there  are  some  well  authenticated  exam- 
ples of  chalcocite-pyrite.  The  assemblage 
must  be  due  to  the  breakdown  of  a 
chalcocite-digenite-bornite  solid  solution 
that  formed  in  grain  boundary  equilib- 
rium with  pyrite.  A  chalcocite-pyrite 
hypogene  assemblage  should  therefore 
always  be  accompanied  by  bornite  and/or 
digenite,  as  at  Butte. 

Assuming  equilibrium  during  deposi- 
tion, a  pyrite-chalcocite  assemblage  with- 
out accompanying  digenite  or  bornite  is 
thus  an  indication  that  the  two  minerals 


bornite  solid  solution  is  of  maximum 
sulfur  content,  the  relation  between  a  and 
Cu/(Cu  +  Fe)  is  linear.  This  was  demon- 
strated by  Kullerud  in  his  studies  on  the 
bornite-digenite  join  {Year  Book  59).  The 
relationship  has  now  been  verified  for 
bornite  containing  chalcopyrite  in  solid 
solution  (fig.  55).  However,  the  present 
investigation  revealed  that  a  decrease  in 
sulfur  content  of  less  than  0.5  weight  per 
cent  can  increase  the  cell  edge  by  as  much 
as  0.015  A.  The  cell  edge  of  sulfur- 
deficient  bornite  is  dependent  on  (1)  the 
Cu/(Cu  +  Fe)  ratio,  (2)  the  sulfur 
content,    (3)    the   quenching   procedure, 


GEOPHYSICAL   LABORATORY  159 

(4)  the  type  of  grinding  used  in  prepa-  studied  by  means  of  an  electron  micro- 
ration  of  the  diffractometer  mount.  For  scope  (at  the  National  Bureau  of  Stand- 
this  reason  caution  should  be  exercised  in  ards) ,  but  no  submicroscopic  phase  of  the 
applying  data  published  on  bornite  cell  type  suggested  by  Takeuchi  and  Nambu 
edges  to  natural  bornite.  (1956)  was  noted. 

In  general,  bornite  cannot  be  chem- 

Heating  Experiments  on  Natural  Bormtes  ically  anaIyzed  accurately  because  small 

P.  R.  Brett  inclusions    of   other   sulfides    cannot   be 

Bornite   can   take   some   chalcopyrite  eliminated.  A  sample  of  the  "anomalous" 

into   solid   solution,   but   this   is   totally  bornite   was   examined   on   the   electron 

exsolved  on  cooling,  even  if  the  cooling  probe  at  the  U.  S.  Geological  Survey;  this 

time  is  only  some  few  minutes   (Brett,  technique  gave  analyses  with  such  a  large 

this  report).  When  heated  to  tempera-  standard  deviation  that  the  results  could 

tures    below    400°C,    however,    certain  not  be  applied  in  the  present  study, 

natural  bornites  exsolve  up  to  25  volume  The  possibility  that  oxidation  could  be 

per   cent   chalcopyrite    (Wandke,    1926;  responsible  for  the  formation  of  chalco- 

Takeuchi  and  Nambu,   1956;  Prouvost,  pyrite  in  some  heated  bornites  seemed 

1960;  McCauley,  1961).  This  situation  is  worth  examining.  Two  samples  of  Bristol 

anomalous  in  view  of  the  slow  cooling  bornite  were  finely  ground  and  exposed 

that  presumably  occurs  in  nature.  to   the   air  for   5   days   and   4  months, 

Bornite    from    various    localities    was  respectively.    There   was   no    detectable 

heated  in  evacuated  silica  glass  tubes  at  change  in  weight  on  oxidation.  Bornite 

temperatures  ranging  from  75°  to  600°C.  from  Bristol  was  chosen  because  of  its 

Bornites  from  Moonta,  South  Australia;  purity  and  because  it  did  not  exsolve 

Messina,  South  Africa;  Bristol,  Connec-  chalcopyrite  on  heating.  Approximately 

ticut;   and   Magma,   Arizona,   remained  2  volume  per  cent  chalcopyrite  was  found 

unchanged,  but  those  from  Similkaween,  as    irregular    blebs    when    the    oxidized 

B.  C,  Beaverdell,   B.   C,  and  red  bed  bornite  was  heated  at  270°C  for  1  hour, 

copper  localities  in  Utah  exsolved  chalco-  After  1  hour  at  300°C  synthetic  stoichio- 

pyrite  up  to  25  volume  per  cent.  The  metric  bornite  oxidized  in  the  same  way 

Utah  bornites  exsolved  chalcopyrite  after  contained  small  amounts  (less  than  1  per 

only  10  minutes  at  400°C,  but  no  exsolu-  cent)  of  a  very  fine-grained  second  phase, 

tion    occurred    even    after    10    days    of  possibly  chalcopyrite. 

heating  at  temperatures  lower  than  75°C.  The    cell    sizes    of    all    bornites    were 

The   exsolved   phase   was   identified    as  measured  both  before  and  after  heating, 

chalcopyrite  both  optically  and  by  its  with   results   shown   in   table   21.    It   is 

three  principal  X-ray  diffraction  peaks,  apparent  from  table  21  that  the  normal 

Anomalous     unheated     bornite     was  bornites  have  a  close  to  that  of  stoichi- 

TABLE  21.     Cell  Edge  of  Bornite  from  Various  Localities  before  and  after  Heating 


a 

a 

Locality 

before  Heating, 
±0.005  A 

after  Heating, 
±0.005  A 

Magma,  Ariz. 

10.950 

10.950 

Bristol,  Conn. 

10.945 

10.945 

Messina,  South  Africa 

10.956 

10.956 

Red  bed  bornite,*  Utah 

10.932 

10.950 

Red  bed  bornite,*  Utah, 

second 

locality 

10.906 

Oxidized  Bristol,  Conn., 

bornite 

* 

10.958 

10.950 

Synthetic  Cu5FeS4 

10.950 

10.950 

*  Exsolves  chalcopyrite  on  heating. 


160 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


ometric  bornite,  whereas  bornites  that 
exsolve  chalcopyrite  may  have  a  con- 
siderably smaller  cell  edge  before  heating 
and  on  heating  revert  to  a  cell  size  similar 
to  that  of  stoichiometric  bornite.  The 
small  cell  cannot  be  ascribed  to  oxidation, 
because  the  cell  edge  of  oxidized  Bristol 
bornite  before  heating  is  greater  than  that 
of  the  unoxidized  bornite. 

When  heated  at  700°C  and  quenched 
the  Utah  bornite  contained  no  exsolved 
chalcopyrite,  but  the  cell  size  had  reverted 
to  that  of  stoichiometric  bornite.  The  low 
value  of  a  for  the  anomalous  bornite 
evidently  cannot  be  ascribed  to  a  high 
iron  content.  It  has  been  mentioned 
elsewhere  in  this  report  that  the  lower  the 
sulfur  content  of  a  bornite  the  greater  the 
cell  edge.  Possibly  the  anomalous  bornites 
contain  more  sulfur  than  stoichiometric 
bornite.  Heating  at  700°C  would  probably 
expel  the  excess.  At  any  rate,  this  is  a 
reasonable  explanation  of  the  behavior  of 
the  Utah  bornite. 

It  is  suggested  that  at  temperatures 
between  75°  and  400°C  these  sulfur-rich 
bornites  break  down  to  chalcopyrite, 
stoichiometric  bornite,  and  chalcocite. 
The  chalcocite  compensates  for  the 
increased  copper  content  of  the  bornite 
caused  by  the  exsolution  of  chalcopyrite. 
A  phase  resembling  chalcocite  or  digenite 
has  been  observed  with  chalcopyrite  in 
heating  experiments  on  anomalous  born- 
ites by  Greig  (personal  communication, 
1961)  and  Prouvost  (personal  communi- 
cation, 1962). 

Yund  (personal  communication,  1962) 
reports  that  synthetic  sulfur-rich  bornites 
with  Cu/Fe  ratio  equal  to  or  less  than  the 
stoichiometric  ratio  exsolve  chalcopyrite 
when  heated.  This  is  additional  evidence 
that  the  anomalous  bornites  are  anom- 
alous not  because  they  are  iron  rich  or 
contain  oxygen  but  because  they  are  rich 
in  sulfur.  The  conclusion  is  by  no  means 
proved,  however. 

There  is  good  evidence  that  the 
anomalous  bornites  never  attained  the 
temperature  of  approximately  75°C  dur- 
ing  their   formation    or   later,    as   it   is 


possible  to  cause  chalcopyrite  to  exsolve 
from  them  at  this  temperature. 

Method  for  Mixing  Liquids  at 
Controlled  Temperatures 

G.  Kullerud 

In  the  past,  studies  of  mineral  assem- 
blages in  aqueous  solutions  have  ordi- 
narily been  conducted  by  mixing  the 
liquids  at  25°C  then  slowly  heating  the 
mixture  to  a  specified  elevated  tempera- 
ture at  which  it  was  kept  for  desired 
periods  of  time.  In  this  procedure 
precipitation  often  takes  place  as  soon  as 
the  solutions  are  mixed,  and  the  phases 
formed  at  room  temperature  or  during 
the  heating  period  may  persist  metastably 
for  considerable  lengths  of  time.  A 
disadvantage  of  the  method  is  that 
equilibrium  cannot  be  proved  to  have 
existed  in  any  one  experiment.  The 
results  of  many  earlier  studies  on  the 
pyrite-marcasite  relations  in  which  this 
method  was  used  are  probably  examples 
of  the  shortcomings  of  the  procedure. 

Progress  in  our  studies  of  dry  systems 
is  commonly  hampered  by  slow  reaction 
rates.  At  low  temperatures  these  rates  are 
usually  so  slow  that  dry  experimentation 
is  out  of  the  question  because  of  the  time 
involved  in  obtaining  equilibrium.  Since 
many  minerals  in  nature  form  at  low 
temperatures  and  are  often  stable  only  in 
the  region  not  available  to  dry  synthesis, 
it  is  desirable  that  we  develop  other 
methods  by  which  equilibrium  can  be 
obtained  in  a  short  time. 

Bravoite  is  an  example  of  a  low- 
temperature  mineral  that  we  could  not 
synthesize  by  the  dry  method.  Phase 
equilibrium  in  this  part  of  the  Fe-Ni-S 
system  could  be  obtained  only  down  to 
200°C,  and  this  accomplishment  required 
13  months. 

The  new  method  allows  separate 
heating  to  a  preassigned  temperature  of 
two  or  if  necessary  more  liquids.  One 
liquid  is  sealed  into  a  long  evacuated 
Pyrex  tube;  the  other  is  poured  directly 
into  a  cylindrical  Teflon  container  that 


GEOPHYSICAL   LABORATORY  161 

has  a  pressure-tight  closure.  The  Pyrex  The  tungsten-copper  deposits  at  Tern 

tube  is  then  also  inserted  into  the  Teflon  Piute,  Lincoln  County,  Nevada,  occur  in 

tube,  and  the  pressure  seal  is  closed.  This  a    skarn    aureole    surrounding    a    small 

unit,    which    has    a    thermocouple    well  granodiorite  stock.  One  of  the  mines,  the 

similar  to  that  of  cold  seal  and  Tuttle  Free  Tunnel,  was  studied  in  detail  because 

bombs,   is   next  placed   in  a  horizontal  of   its    rather   complex    and    varied    ore 

preheated    furnace    and    heated    to    the  assemblages.    All   the   metallic   minerals 

temperature  of  the   experiment.    Teflon  occur    in    a    diopside,    andradite    skarn, 

softens  on  heating  but  readily  withstands  which  separates  the  barren  intrusion  from 

the    internal    pressure    to    200°C    under  unmineralized    limestone    and,     locally, 

these  conditions.  When  the  temperature  hornfels. 

of  the  experiment  has  been  reached  the  The  metallic  minerals  occur  as  dissem- 

liquids  are  mixed  by  exerting  pressure  on  inations   or  small  lenses,   veining  being 

the  Teflon  tube,  by  means  of  a  specially  extremely  minor.  With  the  exception  of 

constructed  pair  of  pliers  with  jaws,  to  pyrite,  chalcopyrite,  and  scheelite,  all  of 

the  point  where  the  Pyrex  tube  shatters,  which  occur  throughout  the  aureole,  the 

The    reaction    that    immediately    takes  minerals  are  roughly  thermally  zoned.  In 

place  is  recorded  by  the  thermocouple,  the  "inner,"  formerly  hotter,  portions  of 

which  shows  a  rapid  increase  in  tempera-  the  aureole,  in  approximately  paragenetic 

ture  of  as  much  as  5°C.  Thereafter  the  sequence    are    molybdenite,    pyrrhotite, 

temperature  drops  back  over  the  next  10  magnetite,    and    marcasite.     Near    the 

minutes  or  so  to  that  recorded  before  the  limestone  in  the  "outer,"  formerly  cooler, 

liquids  were  mixed.  The  Teflon  tubes  are  parts,  are  sphalerite,  galena,  galenobis- 

kept  in  the  furnace  at  temperature  from  mutite,    cosalite,    and    native    bismuth. 

1  to  100  hours,  depending  on  the  tern-  Except  for  sphalerite  these  minerals  are 

perature  of  the  experiment  after  mixing  sparse. 

of  the  liquids,  to  produce  well  crystallized  Pyrrhotite  was  one  of  the  first  metallic 

materials.    The    products    are    filtered,  minerals  to  form,  and,  as  such,  it  was 

washed,  and  studied  by  means  of  X-ray  deposited    during    the    earliest,    hottest 

diffraction    patterns     and    in    polished  stages  of  the  mineralization  period.  It  is 

sections.  most  prominent  along  the  granodiorite- 

skarn  contact  but  occurs  in  decreasing 

Pyrrhotite  from  Tem  Piute,  Nevada  amounts  farther  from  the  granodiorite. 

The  pyrrhotite  is  commonly  associated 
with  and  generally  surrounds  eunearal 

The  composition  of  hexagonal  pyrrho-  crystals  of  pyrite,  thereby  indicating  its 

tite  when  in  equilibrium  with  pyrite  has  later  origin.  In  an  attempt  to  determine 

found   considerable  use  as  a  geological  its  temperatures  of  formation  all  available 

thermometer.  Efforts  to  apply  the  ther-  pyrrhotite  was  sampled  and  examined  in 

mometer    to     pyrrhotite-pyrite     assem-  the  laboratory. 

blages  from  the  Tem  Piute  district  Arnold  (1962)  demonstrated  that  the 
encountered  difficulties  because  at  this  di02  spacing  of  hexagonal  pyrrhotite  is  a 
locality  the  pyrrhotite  is  monoclinic.  function  of  its  composition.  The  compo- 
Arnold  and  Reichen  (1962)  suggested  that  sition  is  dependent  on  the  temperature  of 
the  thermometer  may  be  valid  even  for  formation  provided  that  the  pyrrhotite 
such  assemblages  and  that  the  compo-  formed  in  equilibrium  with  pyrite  and 
sition  of  monoclinic  pyrrhotite  can  be  that  it  did  not  reequilibrate  with  de- 
determined  by  the  standard  X-ray  creasing  temperatures.  Most  of  the  Tem 
method  if  the  specimen  is  first  inverted  Piute  pyrrhotite  was  sampled  within  1 
to  the  hexagonal  form  by  heating  in  mm  of  pyrite,  and  all  such  samples  have 
vacuo.  similar  dio2  values.  It  is  therefore  assumed 


162 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


that  these  pyrrhotites  formed  in  equi- 
librium with  pyrite.  Pyrrhotites  that  did 
not  form  close  to  pyrite  have  different 
compositions  from  those  in  contact  with 
it.  This  would  presumably  not  be  so  had 
all  the  pyrrhotite  reequilibrated  as  tem- 
peratures fell  during  cooling. 

In  the  inversion  of  monoclinic  pyrrho- 
tite to  the  hexagonal  form  the  time  and 
temperature  allowed  for  annealing  are 
critical ;  with  too  long  an  annealing  period 
or  too  high  a  temperature  the  pyrrhotite 
reequilibrates,  and  with  too  short  an 
annealing  time  or  too  low  a  temperature 
it  does  not  invert.  To  determine  the 
optimum    time    and    temperature    for 


temperature  of  the  hexagonal-monoclinic 
transition  lies  below  260°C,  but  at  the 
same  time  the  reaction  is  too  slow  for 
quick  annealing  at  temperatures  below 
300°C.  As  the  samples  inverted  rapidly 
but  did  not  reequilibrate  in  0.1  hour  at 
346°C  the  other  Tern  Piute  specimens 
were  annealed  under  these  conditions. 

Several  pyrrhotites  were  sampled  from 
specimens  containing  no  pyrite.  In  speci- 
mens that  contained  appreciable  pyrite 
the  pyrrhotite  was  extracted  with  a 
dentist's  drill  kept  in  contact  with  pyrite 
at  all  times  so  that  no  pyrrhotite  was 
collected  farther  than  1  mm  from  pyrite. 

Table   22   lists   the   results   of   X-ray 


TABLE  22.     Average  20(1O2)  and  Corresponding  d(i02)  of  Tern  Piute  Pyrrhotite 


Sample 

20(102) 

sie* 

^(102) 

Comments 

1 

44.036 

0.0041 

2.0563  ±0.0002 

2 

43.966 

0.0097 

2.0594  ±0.0004 

3 

44.002 

0.0033 

2.0578  ±0.0002 

4 

44.017 

0.0053 

2.0572  ±0.0002 

5 

43.95i 

0.0044 

2.0601  ±0.0002 

6 

43.90o 

0.0109 

2.0623  ±0.0005 

Pyrite  absent 

7 

43.87i 

0.0067 

2.0636  ±0.0003 

Pyrite  absent 

8 

44.01! 

0.0046 

2.0574  ±0.0002 

9 

43.939 

0.0082 

2.0606  ±0.0004 

Pyrite  absent 

10 

43.882 

0.0042 

2.0631  ±0.0002 

Pyrite  absent 

11 

43.95i 

0.0053 

2.0601  ±0.0002 

*  Standard  error  of  eight  or  more  successive  oscillations. 


annealing,  natural  pyrrhotite  was  finely 
ground  under  acetone  and  concentrated 
magnetically;  replicate  runs  prepared 
from  this  material  were  heated  in 
evacuated  silica  glass  tubes. 

Runs  were  annealed  for  0.1  hour  at 
different  temperatures.  At  700°C  and 
555°C  the  pyrrhotite-pyrite  reaction  is 
sufficiently  rapid  for  the  samples  to  have 
reequilibrated.  Two  runs  heated  at  346° 
and  one  at  455°C  inverted  but  did  not 
have  time  to  reequilibrate.  They  indicate 
the  same  composition  within  the  limits 
of  error  of  the  method.  A  sample  heated 
at  300°C  did  not  invert  in  0.1  hour  but 
did  in  1}^  months.  Likewise,  one  at 
260°C  did  not  invert  within  \x/i  months 
but    did    within    1    year.    Clearly    the 


measurements  on  a  number  of  Tern  Piute 
pyrrhotites,  all  of  which  were  originally 
monoclinic.  Numbers  3  and  4  are  from 
the  same  sample;  they  provide  almost 
identical  results.  Likewise,  all  the  pyr- 
rhotite samples  that  were  adjacent  to 
pyrite  have  very  similar  di02  values  and, 
had  they  been  hexagonal  when  collected, 
would  indicate  temperatures  between 
455°  and  510°C.  Those  that  were  not  in 
contact  with  pyrite  have  consistently 
larger  dw2  values  and,  had  they  also  been 
originally  hexagonal,  would  correspond  to 
minimum  temperatures  between  390°  and 
450°C.  Although  very  reasonable  for 
contact  metasomatic  deposits  such  as 
Tern  Piute,  these  temperatures  must  be 
regarded  as  tentative.  At  present  it  is  not 


GEOPHYSICAL   LABORATORY  163 

clear  that  the  relations  between  hexagonal  material  yield  sound  estimates  of  tem- 

and  monoclinic  pyrrhotite  are  such  that  peratures  of  formation, 
measurements    on    inverted    monoclinic 

STONY   METEORITES 
P.  Ramdohr9  and  G.  Kullerud 

During  this  past  year  more  than  a  structure    and    containing    Fe-C-S    was 

hundred    stony    meteorites    have    been  observed  in  10  per  cent  of  the  meteorites, 

studied  in  polished  sections  in  addition  to  Daubreelite,  FeCr2S4,  is  also  present  in 

those  described  in  last  year's  report.  The  about    10   per   cent    of   the    specimens, 

following  opaque  and  semiopaque  min-  Sphalerite,  ZnS,  occurs  in  trace  amounts 

erals    have    been    identified:    Minerals  only.  Chalcopyrite,  CuFeS2,  was  observed 

containing  elemental  iron  include  a  iron  in  a  few  meteorites,  and  pyrite,  FeS2,  was 

(kamacite)    with    variable    Ni    content,  identified  only  once. 

Fe-Ni  solid  solutions  (taenite)  with  the  Besides  these  minerals  a  number  of 

structure  of  7  iron,  and  intergrowths  of  new  ones  were  observed  in  small  amounts 

the  a  and  7  phases,  plessite.   Cohenite  and  mostly  in  single  meteorites.  These 

(Fe3C)    occurs    only    in    a    few    stony  phases  are  referred  to  by  the  letters  A 

meteorites  and  in  small  amounts.  Schrei-  through  L.  For  most  of  them  the  compo- 

bersite   (Fe3P)   is  widely  distributed  in  sitions  are  partly  or  completely  unknown 

small  amounts.  A  new  mineral,  which  by  although   their   major   constituents   can 

synthesis  was  found  to  have  the  compo-  often    be    deduced    from    the    mineral 

sition  (Ni ,  Fe)  2S  and  which  we  refer  to  as  assemblages  with  which  they  are  associ- 

the  Henderson  phase,  was  observed  in  ated.  Mineral  A  is  strongly  anisotropic 

three  meteorites.  Graphite  (C)  occurs  in  and  has  a  dark  yellow-green  color.   It 

about  one- tenth  of  the  specimens.  Native  almost    invariably   occurs    as   lenses    or 

copper  (Cu)  is  commonly  observed,  but  lamellae  in  daubreelite  and  only  rarely  is 

in    trace    amounts.    Native    gold    was  found  independent  of  troilite.  Its  optical 

observed  in  only  one  specimen.  Troilite  properties  indicate  that  it  has  a  pseudo- 

(FeS)  is  present  in  all  specimens  examined  hexagonal  orthorhombic  symmetry,  and 

and   is   frequently   the   most   abundant  it  may  be  a  transformation  product  of 

opaque  mineral.  Chalcopyrrhotite,  (Fe ,  daubreelite.    Mineral    B    occurs    inter- 

Cu,Ni,Zn)S,   a  cubic  high-temperature  layered  with  mineral  A  and  appears  to 

solid   solution,   was   observed   in   about  have  formed  from  it,  not  directly  from 

one-third    of    the    specimens.    Valleriite  daubreelite,  with  which  it  is  also  closely 

occurs   as   a   disintegration   product   of  associated.     This    mineral    may    be    a 

chalcopyrrhotite   and   as   an   exsolution  terrestrial   alteration   product,   although 

product     of     pentlandite.     Pentlandite,  the  neighboring  minerals,  some  of  which 

(Fe,Ni)9S8,  is  present  in  about  one-fourth  are  very  susceptible  to  weathering,  show 

of  the  meteorites  examined.  Oldhamite,  no  sign  of  alteration. 

(Ca,Fe,Mn)S,  is  limited  to  meteorites  Mineral    C    is    olive-brown,    weakly 

that  are  highly  reduced  or  that  have  a  reflecting,  and  apparently  isotropic.  It  is 

high  sulfur  content.  A  new  (Fe,Mg,Mn,  commonly,   but   not   always,   associated 

Ca)S  phase  similar  to  oldhamite  but  with  with  daubreelite.  Mineral  D  is  colorless 

much  higher  reflectivity  is  rather  com-  and    transparent    with    high    refractive 

mon.  Alabandite,  MnS,  was  not  observed,  index.  It  replaces  ilmenite  and  chromite 

A  new  mineral  with  a  hexagonal  layer  and  is  always  associated  with  chromite. 

Mineral  E  is  dark  brown  and  occurs  with 

9  University  of  Heidelberg.  troilite.  It  is  relatively  soft,  is  isotropic, 


164 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


and  shows  traces  of  internal  reflections. 
It  contains  exsolution  bodies  of  troilite. 
Mineral  F  is  white  and  resembles  certain 
terrestrial  arsenides  and  sulfarsenides 
such  as  loellingite  and  arsenopyrite.  This 
mineral  is  isotropic  and  has  good  cleavage 
parallel  to  (111).  It  probably  contains 
arsenic.  Mineral  G,  a  light  blue  mineral, 
occurs  in  association  with  troilite.  It  is 
relatively  soft,  is  isotropic,  and  shows 
traces  of  internal  reflections.  It  contains 
troilite  exsolution  bodies  and  is  pre- 
sumably a  sulfide,  but  is  optically  differ- 
ent from  any  known  sulfide.  Mineral  H 
has  a  yellow-gray  color  and  almost 
metallic  characteristics.  It  is  anisotropic 
with  greater  reflectivity  than  chalco- 
pyrite  but  with  paler  colors.  Its  prop- 
erties differ  from  those  of  any  previously 
described  mineral.  Mineral  I  is  colorless; 
it  is  of  the  spinel  type  and  often  contains 
ilmenite  exsolution  lamellae.  It  is  trans- 
parent, its  refractive  index  is  in  the  range 
1.8-1.9,  and  its  hardness  is  greater  than 
that  of  olivine  but  lower  than  that  of  pure 
Mg-Al  spinel.  This  mineral  was  found  by 
synthesis  to  have  the  composition 
Mg2Ti04.  Mineral  K  has  a  very  dark  gray 
color  and  is  definitely  isotropic.  It  is 
sometimes  partly  rimmed  by  troilite,  and 
its  optical  properties  indicate  that  it  is  a 
sulfide.  It  may  possibly  be  a  member  of 
the  (Fe,Cu,Zn)S  mineral  group.  Mineral 
L  is  strongly  pleochroic  and  is  commonly 
intergrown  with  mineral  A.  This  inter- 
growth  indicates  that  mineral  L  may  be 
hexagonal.  Its  hardness  is  moderate  and 
similar  to  that  of  troilite. 

In  contrast  to  the  numerous  sulfides 
observed  in  stony  meteorites  the  common 
oxides  are  limited  to  chromite,  magnetite, 
and  ilmenite. 

With  few  exceptions  the  stony  meteor- 
ites are  more  uniform  in  their  silicate 
mineralogy  than  terrestrial  rocks.  The 
major  silicates  are  olivine  and  pyroxene 
(usually  orthopyroxene) ,  with  minor 
amounts  of  plagioclase.  Glass  is  fre- 
quently present  in  small  amounts ;  quartz 
and  tridymite  are  rare.  The  individual 
meteorites    commonly    display    complex 


mineralogical  relationships.  A  chondrule 
or  even  a  zone  within  a  chondrule  may 
represent  local  equilibrium.  Sometimes 
reactive  gases  apparently  brought  about 
changes  in  mineralogy.  Evidently  an 
increasing  degree  of  reduction  was  accom- 
panied by  increasing  temperature,  result- 
ing in  changes  of  both  mineralogical 
composition  and  grain  size  and  in 
elimination  of  brecciation.  Complex  ge- 
netic histories  are  displayed  in  many 
meteorites  through  spontaneous  melting 
processes  resulting  in  droplets  of  nickel- 
iron,  sulfides,  and  probably  glass.  In 
many  specimens  the  mineralogy  indicates 
that  sulfur  has  been  introduced  in  some 
form.  In  such  meteorites  FeS  is  observed 
to  replace  Fe,  sometimes  even  taenite. 
Since  nickel  is  less  reactive  than  iron  with 
respect  to  sulfur,  the  introduction  of 
sulfur  leads  to  relative  enrichment  of 
nickel  with  a  consequent  decrease  in  the 
amount  of  kamacite.  Continued  intro- 
duction of  sulfur  leads  to  complete  dis- 
appearance of  the  metal  phase  and  often 
produces  pentlandite.  Further  complica- 
tions include  the  presence  of  reactive 
hydrocarbons  in  many  meteorites. 

Noteworthy  structural  and  textural 
phenomena  are:  fusion  on  "dislocations," 
manifested  by  the  occurrence  of  droplets 
of  fused  troilite  and  iron  in  varying 
amounts  in  veinlets ;  spontaneous  melting, 
resulting  in  patches  of  glass  with  fused 
droplets  of  sulfide  or  iron  in  the  interior 
of  many  meteorites,  not  related  to  the 
fusion  crust,  to  heat  developed  on  impact, 
or  to  brecciation  or  sintering  and, 
therefore,  distinctly  different  from  fusion 
on  "dislocations";  exsolution,  which  is 
common,  as  for  instance  ilmenite  from 
chromite,  chalcopyrite  from  troilite,  mag- 
netite from  olivine ;  mechanical  distortion 
and  recrystallization,  evident  in  very 
many  meteorites;  terrestrial  weathering 
effects,  observed  in  numerous  stony 
meteorites,  the  products  of  which,  such 
as  magnetite,  can  sometimes  be  mistaken 
for  primary  components. 

The  new  minerals  discovered  in  stony 
meteorites  occur  in  amounts  too  small  to 


GEOPHYSICAL   LABORATORY  165 

permit  standard  chemical  analysis  and  tialities  in  this  field.  The  new  phases  can 

almost    invariably    also    in    too    small  also  be  synthesized  if  the  major  constitu- 

amounts  to  allow  investigation  by  X-ray  ents  can  be  surmised  from  the  mineral 

powder  diffraction  methods.  The  electron  paragenesis,    and    our    efforts    in    this 

probe,  however,  shows  promising  poten-  direction  are  increasingly  successful. 


IRON   METEORITES 

S.  P.  Clark,  Jr. 

During  the  past  several  years,  work  has 

been  done  in  the  systems  Fe-Ni-S  and  The  System  Fe-Ni-S 
Fe-Ni-P    for    the    primary    purpose    of 

finding    the     compositions    of    troilite,  As  was  stated  in  last  year's  report  ( Year 

(Fe,Ni)S,  or  schreibersite,  (Fe,Ni)3P,  in  Book  60,  p.  184),  the  amount  of  nickel  in 

equilibrium  with  both  kamacite  (a  alloy)  the    troilite    in    equilibrium    with    both 

and  taenite  (7  alloy)  at  various  tempera-  kamacite  and  taenite  is  small.  Since  the 

tures  and  low  pressures.  Knowledge  of  7    structure    in    the    alloy    cannot    be 

these    compositions    as    a    function    of  quenched  in  the  range  of  temperatures 

temperature  provides  an  indication  of  the  over    which    investigations    have    been 

temperature  of  formation  of  iron  meteor-  made,    the   composition   of   the   troilite 

ites  which  supplements  the  one  provided  cannot    be    closely   determined    by   the 

by  the  Fe:Ni  ratios  of  the  two  alloy  standard  methods  of  phase  equilibria.  If 

phases.  Disagreement  between  these  ther-  the  charge  contains  the  metastable  a2 

mometers    might    indicate    a    lack    of  phase,  which  forms  from  the  7  alloy  on 

equilibrium  among  the  phases  present,  quenching,  an  upper  limit  to  the  possible 

This,  coupled  with  textural  observations,  nickel  content  of  the  troilite  is  found  by 

might  give  important  information  about  projecting  the  line  connecting  the  bulk 

the  cooling  histories  of  the  meteorites,  composition    of    the    charge    with    the 

Alternatively,  such  a  discrepancy  might  composition  of  the  a  alloy  to  the  FeS-NiS 

result  because  the  iron  meteorites  were  join.  The  composition  of  the  a  alloy  is 

formed  at  high  pressure,  which  would  be  taken   from   the   known   system   Fe-Ni, 

interesting  to  know.  assuming  it  to  be  unaffected  by  sulfur. 

Besides  the  work  on  schreibersite,  This  is  plausible  because  sulfur  is  prac- 
X-ray  studies  of  the  higher  phosphides  of  tically  insoluble  in  nickel  and  raises  the 
iron,  Fe2P  and  FeP,  have  been  made,  a-7  transition  in  iron  by  only  3°C. 
They  were  stimulated  by  the  discovery  The  sensitivity  of  the  method  depends 
by  Chao,  Adler,  Dwornik,  and  Littler  on  the  limiting  amount  of  the  a2  phase 
(1962)  that  metallic  spherules  in  some  that  can  be  detected.  The  distinction 
tektites  consist  of  kamacite  plus  a  second  between  a  and  a2  is  based  on  the  sharp- 
phase,  which  they  tentatively  identified  ness  of  the  X-ray  reflections  in  the  back- 
as  a  phosphide.  Under  the  microscope  reflection  region.  For  the  a  phase,  the 
this  other  phase  is  highly  anisotropic,  KQ1  and  K«2  reflections  of  (220)  are  easily 
which  is  inconsistent  with  the  optical  and  sharply  resolved;  for  a2  they  are 
properties  of  ordinary  schreibersite.  The  blurred  into  a  single  diffuse  reflection, 
studies  of  the  higher  phosphides  were  Since  the  cell  dimensions  of  the  two 
made  in  order  to  help  in  the  identification  phases  are  nearly  the  same,  it  is  difficult 
of  this  phase.  to  detect  even  moderate  amounts  of  one 


166 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


in  the  presence  of  the  other.  The  X-ray 
reflections  are  superimposed. 

Runs  that  limit  the  possible  compo- 
sitions of  troilite  in  equilibrium  with 
kamacite  and  taenite  at  800°  and  700°C 
and  low  pressure  are  shown  in  table  23. 
The  run  at  800°C  limits  the  NiS  content 
of  the  troilite  at  that  temperature  to 
0-0.3  weight  per  cent.  Similarly  at  700°C 
the  amount  of  NiS  present  in  the  troilite 


error  of  less  than  0.01  the  ratio  Ni/(Fe  + 
Ni)  is  the  same  in  the  schreibersite  as  in 
the  7  alloy  with  which  it  is  in  equilibrium. 
This  is  true  for  ratios  up  to  at  least  0.25 
and  at  temperatures  of  both  700°  and 
800°C.  It  is  somewhat  surprising  in  view 
of  the  strong  segregation  of  nickel  in  the 
metal  phase  in  alloy-sulfide  and  alloy- 
silicate  systems  of  iron  and  nickel.  A 
possible  interpretation  is  that  the  metal- 


TABLE  23. 


Data  Fixing  the  Compositions  of  Troilite  in  Equilibrium  with 
Kamacite  and  Taenite 


T,  °C 

Bulk  Composition  of  Run, 
Fe                        Ni 

wt.  % 

S 

Phases 
Present 

Duration  of 
Run,  days 

800 
700 
700 

68.6 
68.5 
68.2 

0.4 
0.5 
1.0 

31.0 
31.0 
30.8 

«2 

ct{-\-aJ) 
a2 

184 
226 
204 

must  lie  between  0  and  0.5  weight  per 
cent.  These  quantities  of  nickel  are  so  low 
that  troilite  does  not  appear  to  be  suitable 
for  use  as  a  thermometer.  The  accuracy 
with  which  the  desired  equilibrium  com- 
positions can  be  determined  in  the 
laboratory  is  too  low  for  this  purpose, 
although  the  picture  could  possibly  be 
changed  by  use  of  the  electron  probe 
microanalyzer. 

The  System  Fe-Ni-P 

The  nickel  content  of  the  schreibersite 
in  equilibrium  with  kamacite  and  taenite 
is  much  larger  than  that  of  the  com- 
parable troilite,  and  it  changes  demon- 
strably with  temperature.  At  800°C  the 
Ni/(Fe  +  Ni)  ratio  of  this  schreibersite 
is  between  0.065  and  0.10,  and  at  700°C 
it  lies  between  0.125  and  0.15.  Further 
work  is  required  to  fix  these  compositions 
more  closely.  Present  results  are  con- 
sistent with  the  assumption  that  the 
phase  diagram  of  the  Fe-Ni  system  is 
significantly  affected  by  phosphorus.  This 
is  known  to  be  so  for  iron;  the  system 
Fe-P  is  of  the  "Y-loop"  type. 

An  interesting  result  of  these  investi- 
gations is  that  within  an  experimental 


phosphorus  bond  is  nearly  metallic  in 
schreibersite.  That  a  metallic  form  of 
phosphorus  can  be  made  at  high  pressures 
is  at  least  consistent  with  metallic 
behavior  of  the  phosphorus  atom  in  the 
schreibersite  lattice. 

Higher  Phosphides  in  the  System  Fe-P 

In  addition  to  the  work  on  schreiber- 
site, some  of  the  properties  of  Fe2P  and 
FeP  have  been  investigated.  The  powder 
diffraction  patterns  of  these  phases  have 
been  completely  indexed  out  to  the 
minimum  d  values  observed  with  Fe  Ka 
radiation,  and  the  optical  properties  of 
the  solid  phases  formed  by  quenching 
liquids  in  this  system  and  those  formed 
by  growth  in  the  solid  state  have  been 
observed. 

Fe2P  grown  in  equilibrium  with  FeP  at 
1000°C  has  unit  cell  parameters  measur- 
ably smaller  than  those  of  Fe2P  equili- 
brated with  Fe3P  at  the  same  tempera- 
ture. This  indicates  that  at  high  tem- 
perature Fe2P  departs  from  stoichiometry. 
The  possibility  of  lack  of  stoichiometry  in 
FeP  has  not  yet  been  investigated. 

The  fact  that  liquids  in  this  system  are 
known  to  be  easily  supercooled  suggested 


GEOPHYSICAL   LABORATORY 


167 


that  metastable  solid  phases  might  be 
formed  on  quenching  liquids.  Charges  of 
composition  89.5  weight  per  cent  Fe,  10.5 
weight  per  cent  P,  and  73.3  weight  per 
cent  Fe,  26.7  weight  per  cent  P,  were 
fused  at  1070°  and  1300°C,  respectively, 
and  quenched  by  dropping  into  ice  water. 
These  compositions  are  close  to  the 
eutectics  between  Fe  and  Fe3P,  and  Fe2P 
and  FeP,  respectively.  The  temperatures 
are  a  few  tens  of  degrees  above  the 
eutectic  temperatures  of  1050°  and 
1262°C  (Hansen  and  Anderko,  1958). 

These  runs  yielded  the  phases  to  be 
expected  if  equilibrium  had  been  reached, 
as  shown  by  the  X-ray  patterns  of  the 
charges.  The  first  produced  Fe3P  and 
metal,  and  the  second  Fe2P  and  FeP.  The 
optical  properties  of  the  quenched  charges, 
however,  are  strikingly  different  from 
those  of  the  same  phases  when  grown  by 
combination  of  the  elements  at  subsolidus 
temperatures.  Both  Fe3P  and  FeP  were 
highly  anisotropic  under  the  microscope 
and  exhibited  properties  corresponding  to 
the  description  of  the  unknown  phase  in 
the  metallic  spherules  in  tektites  de- 
scribed by  Chao,  Adler,  Dwornik,  and 
Littler  (1962).  These  properties  are  in 
sharp  contrast  to  the  properties  of  these 
phases  when  grown  at  lower  tempera- 
tures. The  difference  is  possibly  due  to 
strains  in  the  lattice.  It  does  not  appear 
to  affect  the  X-ray  properties. 

On  the  basis  of  its  X-ray  properties,  the 
unknown  phase  in  the  tektites  can  be 
identified  as  an  iron-rich  schreibersite. 
The  d  values  of  the  reflections  observed 
by  Chao,  Adler,  Dwornik,  and  Littler 
(1962)  are  compared  with  those  of  Fe3P 
in  table  24.  The  spacing  of  the  (411) 
reflection  indicates  a  Ni/(Fe  +  Ni)  ratio 
of  about  0.05  according  to  the  determina- 
tive curve  given  in  Year  Book  60,  page 
184.  This  estimate  may  be  somewhat  high 
because  of  absorption. 

The  agreement  between  the  d  values  in 
columns  1  and  3  of  table  24  is  excellent. 
The  identification  of  the  last  reflection  in 
the  table  as  (402)  must  be  considered 
somewhat  uncertain,  however,  since  this 


TABLE  24.     Comparison  of  the  X-Ray 

Properties  of  the  Unknown  Phosphide 

and  Fe3P 


Unknown  Phase 
(Chao  et  al.,  1962) 


Fe3P 


d 

/ 

d 

hkl 

2.19 

w 

2.1984 

321 

2.14 

vw 

2.1455 

330 

2.11 

vw 

2.1090 

112 

1.973 

vw 

1.9787 

411 

1.600 

WW 

(1.5936) 

(402) 

is  one  of  the  weaker  reflections  in  the 
schreibersite  pattern.  It  is  surprising  that 
Chao  et  al.  should  observe  this  reflection 
and  not  stronger  ones  such  as  (510)  or 
(132).  The  other  reflections  listed  in  the 
table  are  among  the  strongest  ones  in  the 
schreibersite  pattern.  None  of  the  lines  of 
Fe2P  or  FeP  has  d  values  close  to  1.600  A. 
These  results  indicate  that  little  new 
information  about  the  origin  of  tektites 
can  be  inferred  from  the  presence  of 
optically  anisotropic  schreibersite.  It  is 
clear  from  the  glassy  nature  of  these 
bodies  and  the  spherical  shape  of  the 
metallic  particles  that  they  have  been 
melted  and  then  relatively  rapidly 
quenched.  These  seem  to  be  the  condi- 
tions necessary  to  produce  the  observed 
phosphide.  Of  greater  interest  is  the  new 
evidence  that  the  unknown  phase  is 
indeed  schreibersite.  Tektites  are  com- 
monly thought  to  be  the  result  of  the 
"splash"  produced  by  the  impact  of  an 
iron  meteorite.  The  principal  disagree- 
ment about  their  origin  centers  around 
whether  the  impact  occurred  on  the  earth 
or  the  moon.  In  either  event  the  metallic 
spherules  are  presumably  part  of  the 
meteorite  itself,  and  as  such  they  may  be 
virtually  identical  in  all  tektites  produced 
by  a  given  impact.  Hence  the  content  of 
minor  elements  like  sulfur,  phosphorus, 
or  carbon  in  the  spherules  should  help  in 
determining  whether  or  not  there  was  a 
multiplicity  of  falls  in  regions  of  complex 
strewn  fields  such  as  southeast  Asia  and 
Australia. 


168 


CARNEGIE     INSTITUTION      OF      WASHINGTON 

GEOTHERMAL   CALCULATIONS 

S.  P.  Clark,  Jr. 


The  past  year  has  witnessed  a  striking 
reawakening  of  interest  in  terrestrial  heat 
flow  stimulated  in  part  by  geothermal 
investigations  in  the  Pacific  Ocean  basin. 
This  work,  which  in  recent  years  has  been 
carried  on  mainly  by  R.  P.  von  Herzen 
at  the  University  of  California  at  La 
Jolla,  has  shown  much  fine-scale  irregu- 
larity, which  must  in  part  at  least  be  real. 
For  example,  the  extremely  high  heat 
flows  on  the  East  Pacific  Rise  now  appear 
to  be  confined  to  two  relatively  narrow 
zones  trending  parallel  to  the  crest  of  the 
Rise.  The  sharpness  of  these  features  is 
suggestive  of  volcanic  origin;  in  any  event 
their  cause  must  lie  at  very  shallow 
depths.  The  interesting  question  whether 
similar  features  exist  in  continental 
regions  cannot  be  answered  with  present 
data.  A  number  of  proposals  for  drilling 
holes  for  geothermal  purposes  have  been 
submitted  to  the  National  Science  Foun- 
dation, and  it  is  to  be  hoped  that  the 
observational  basis  of  this  subject  can  be 
greatly  broadened  in  the  next  few  years. 

Theoretical  investigations  of  subjects 
related  to  earth  temperatures,  such  as 
those  described  in  Year  Books  59  and  60, 
have  been  continued.  This  type  of  work 
forms  an  essential  background  for  the 
interpretation  of  geothermal  results.  The 
studies  have  been  facilitated  by  the 
replacement  of  the  IBM  704  digital 
computer  at  the  National  Bureau  of 
Standards  by  the  more  powerful  IBM 
7090,  decreasing  the  expense  and  labor 
involved  in  treating  the  rather  cumber- 
some problems  that  have  been  considered. 
Most  investigative  effort  has  been  de- 
voted to  the  effect  on  surface  heat  flow 
of  very  high  thermal  conductivity  at 
depth  and  to  the  cooling  of  a  uniform, 
nonradioactive  earth. 

In  some  of  the  cases  considered  below, 
radioactive  generation  of  heat  is  involved. 
It  is  assumed  that  40  per  cent  of  the 
present  heat  production  is  by  uranium, 


40  per  cent  by  thorium,  and  20  per  cent 
by  potassium — proportions  similar  to 
those  commonly  observed  in  terrestrial 
rocks.  Account  is  taken  of  radioactive 
decay  by  fitting  the  decay  curve  of  such 
an  assemblage  of  radioactive  isotopes  with 
a  single  decay  constant.  This  approxi- 
mation, which  is  amply  accurate  for 
present  purposes,  leads  to  a  fourfold 
reduction  in  machine  time. 

The  effect  of  high  thermal  conductivity 
at  depth  has  been  investigated  as 
described  in  Year  Book  59  (p.  144).  We 
consider  a  sphere  composed  of  an  outer 
shell  with  finite  and  constant  thermal 
conductivity  surrounding  a  central  region 
with  infinite  conductivity.  This  gives  a 
rough  upper  limit  to  the  effect  of  such 
processes  as  radiative  transfer,  which  lead 
to  high  conductivities  at  high  tempera- 
tures and  hence  imply  high  conductivity 
at  great  depths.  The  model  has  obvious 
imperfections:  the  thickness  of  the  outer 
shell  must  be  set  arbitrarily,  the  thermal 
gradient  must  vanish  in  the  central 
region,  and  a  discontinuous  change  in 
properties  is  introduced  at  a  level  where 
the  properties  of  the  real  earth  are  likely 
to  be  continuous.  But  this  approach  has 
the  great  advantage  that  it  leads  to  linear 
equations,  and  the  radiogenic  heat  and 
heat  flow  can  be  clearly  and  uniquely 
separated  from  the  thermal  effects  of 
initial  heat. 

The  first  theoretical  investigation  of 
heat  flow  involved  calculations  of  the 
thermal  flux  from  an  earth  of  constant 
properties  with  radioactive  elements  dis- 
tributed uniformly  throughout  a  surface 
layer  of  variable  thickness  (Clark,  1961). 
It  was  found  that  the  flux,  when  regarded 
as  a  function  of  thickness  of  the  radio- 
active layer,  passed  through  a  broad 
maximum  at  moderate  thicknesses  and 
decreased  when  the  thickness  exceeded 
500  km.  MacDonald  (1961)  later  pub- 
lished a  similar  calculation  for  a  non- 


GEOPHYSICAL   LABORATORY 


169 


linear  model  in  which  radiative  transfer 
was  taken  into  account.  His  results  show 
a  monotonic  rise  in  heat  flow  with  thick- 
ness of  the  radioactive  layer.  Since  there 
is  no  satisfactory  way  of  separating 
radiogenic  flux  from  that  due  to  initial 
heat  in  MacDonald's  problem,  and  since 
he  took  very  high  initial  temperatures 
(1880°C  at  a  depth  of  100  km),  it  seems 
worth  while  to  examine  further  the  case 
of  an  earth  with  a  perfectly  conducting 
center.  This  examination  should  reveal 
the  reasons  for  the  qualitative  differences 
between  the  uniform  and  nonlinear 
models. 

In  the  first  problem  considered  it  was 
assumed  that  radioactivity  was  uniformly 
distributed  through  the  outermost  500 
km  of  the  earth.  The  initial  temperature 
was  taken  to  be  zero,  and  the  thickness 
of  the  outer  shell  of  finite  conductivity 
was  allowed  to  range  from  200  to  500  km. 
Results  are  shown  in  figure  56;  as  the 
thickness  of  the  outer  shell  is  increased, 
the  curve  levels  off  and  asymptotically 
approaches  a  value  of  about  1.1.  The 
pronounced  minimum  in  the  curve  of 
figure  56  is  perhaps  surprising  at  first.  It 


200  400 

Thickness  of  shell, km 


600 


Fig.  56.  Ratio  of  surface  heat  flow  to  present 
heat  production  as  a  function  of  thickness  of 
outer  shell  of  finite  conductivity.  Radioactivity 
uniformly  distributed  through  outermost  500  km. 


results  from  conduction  of  heat  toward 
the  earth's  center  as  well  as  toward  the 
surface.  Downward  conduction  is  most 
efficient  if  the  shell  is  200  to  300  km  thick. 
The  temperature  reaches  a  fairly  pro- 
nounced maximum  at  shallow  depths 
(fig.    57),   but   this  maximum   does   not 


2000 


1500- 


o 

o 


1000- 


Q. 

E 


500- 


800 


1000 


I2C0 


1400 


Depth, km 

Fig.  57.     Temperatures  for  outer  shells  200,  500,  and  6371  km  thick.  Initial  temperature  zero 
Radioactivity  uniformly  distributed  through  outermost  500  km. 


170 


CARNEGIE      INSTITUTION      OF      WASHINGTON 


exist  if  plausible  nonzero  initial  tempera- 
tures are  adopted.  In  that  case  the 
concentration  of  radioactive  heat  pro- 
duction near  the  surface  tends  to  lessen 
the  amount  of  cooling  at  shallow  depths; 
the  consequences  are  discussed  below. 

In  another  set  of  calculations,  the 
effect  of  a  central  region  of  infinite  con- 
ductivity on  the  cooling  of  a  nonradio- 
active earth  was  investigated.  The  initial 
temperature  was  assumed  to  be  of  the 
form  To  +  mx,  where  x  is  depth  and  To 
and  m  are  constants.  The  thermal 
conductivity  does  not  enter  this  problem 
explicitly,  and  it  is  convenient  to  consider 
the  thermal  gradient  at  the  surface  rather 
than  heat  flow  as  a  function  of  the  thick- 
ness of  the  outer  shell.  Results  are  shown 
in  figure  58,  where  the  various  curves  are 
labeled  with  appropriate  values  of  T0 
and  m. 


E 
O 


to   4  - 


D 


i         r 


j L 


0  200  400  600  800 

Thickness  of  shell,  km 

Fig.  58.  Thermal  gradient  at  the  surface  as  a 
function  of  the  thickness  of  the  outer  shell  of 
finite  conductivity.  No  heat  production.  Initial 
temperature  To  +  mx.  Numbers  beside  curves 
give  values  of  To  and  m. 


From  figure  58  we  see  that  the  thermal 
gradient  at  the  surface,  and  hence  the 
heat  flow,  is  sensitive  to  the  thickness  of 
the  outer  shell  if  T0  is  large  and  m  is  small. 
The  gradient  is  large  for  small  thicknesses 


of  the  shell  and  decreases  markedly  as 
the  thickness  is  increased.  For  small  T0 
and  large  m  there  is  a  very  much  weaker 
effect  in  the  opposite  sense. 

In  the  foregoing  examples  the  problem 
is  put  somewhat  differently  from  the  way 
it  appears  in  the  published  work  cited 
above.  Here  the  independent  variable  has 
been  the  thickness  of  the  outer  shell,  a 
parameter  most  closely  related  to  the 
assumed  effectiveness  of  processes  such  as 
radiative  transfer  which  lead  to  high 
conductivity  at  high  temperatures.  Figure 
59,  however,  shows  heat  flow  as  a  function 
of  thickness  of  the  radioactive  layer  for 
several  thicknesses  of  the  outer  shell.  T0 
was  taken  to  be  1800°C,  and  m  0.8°C/km; 
these  constants  lead  to  initial  tempera- 
tures close  to  those  tabulated  by  Mac- 
Donald. 

These  results  indicate  that  MacDonald's 
findings  of  an  increase  in  heat  flow  with 
increasing  thickness  of  the  radioactive 
layer  is  due  to  nonlinearity  in  his  earth 
model.  It  appears  to  result  from  the 
reduced  cooling  caused  by  shallow  radio- 
activity. Thickening  the  radioactive  layer 
maintains  near-surface  temperatures  at 
higher  values,  causing  high  thermal 
conductivity  because  of  radiative  trans- 
fer. This  produces  an  effect  analogous  to 
reducing  the  thickness  of  the  outer  shell 
of  finite  conductivity.  Radioactive  heat- 
ing in  effect  promotes  the  escape  of  initial 
heat. 

In  the  linear  case  the  T0  term  contrib- 
utes between  30  and  40  per  cent  of  the 
total  flux,  the  lowest  proportion  corre- 
sponding to  the  thickest  outer  shells.  If 
this  term  were  cut  in  half,  which  is  a 
plausible  adjustment,  the  extreme  ratios 
of  heat  flow  to  heat  production  shown  in 
figure  59  would  be  reduced  to  about  0.6 
and  1.1.  MacDonald's  estimates  of  the 
contribution  of  initial  heat  to  the  flux  at 
the  surface  led  to  values  less  than  25  per 
cent,  which  seem  too  low  by  contrast  with 
the  present  results. 

The  results  given  above  extend  the 
previous  study  of  the  effect  of  depth  of 
burial  of  radioactivity  on  surface  heat 


GEOPHYSICAL   LABORATORY 


171 


o  1.4- 


■D 
O 


o  1.2- 


1        1         1        1        1        1 

^<^0 

\^ 

1        1         1         1         1         1 

0  200  400  600 

Thickness  of  radioactive  layer, km 

Fig.  59.  Ratio  of  heat  flow  to  heat  produc- 
tion as  a  function  of  the  thickness  of  the  radio- 
active layer.  Initial  temperature  1800  +  0.8z. 
Numbers  beside  the  curves  give  thicknesses  of 
the  outer  shell  of  finite  conductivity  in  kilo- 
meters. 


flow  to  cases  involving  further  variable 
parameters.  The  effect  of  high  thermal 
conductivity  at  depth  proves  to  be 
greater  than  that  of  changing  the  thick- 
ness of  a  surface  layer  of  radioactivity, 
especially  if  the  initial  temperature  is 
high.  These  results  point  up  our  inability 
to  find  a  connection  between  radioactive 
heat  production  and  heat  flow  at  the 
surface  without  precise  hypotheses  about 
thermal  properties  and  initial  tempera- 
tures in  the  mantle. 

A  second  major  field  of  investigation 
has  been  the  cooling  of  a  uniform, 
nonradioactive  earth.  Interest  in  this 
problem  arises  mainly  from  its  geomag- 
netic implications,  discussed  below.  Initial 
temperatures  in  the  earth  are  assumed  to 
be  of  the  form  m(Rn  —  rn),  where  R  is 
the  radius  of  the  earth,  r  is  the  radial 
coordinate,  and  m  and  n  are  constants. 
Initial  temperatures  for  n  ranging  from 
1  to  4  and  for  a  central  temperature  of 
500°C  are  shown  in  figure  60,  and  the 
amount  of  cooling  in  5  X  109  years  is 
shown  in  figure  61.  For  the  higher  values 
of  n,  the  cooling  is  greatest  near  the 
surface  and  becomes  small  at  great 
depths.  For  n  =  1,  however,  the  cooling 
is  nearly  independent  of  depth,  and  even 


increases  slightly  toward  the  center. 
These  results  agree  with  the  earlier 
conclusions  (Year  Book  59,  p.  146)  that 
the  cooling  of  the  deep  layers  cannot 
exceed  100°  or  200°C  on  this  model.  These 
data  show  that  cooling  cannot  be  large  at 
depth  unless  initially  the  thermal  gradient 
was  relatively  steep. 

The  geomagnetic  importance  of  this 
problem  arises  from  the  fact  that  forceful 
arguments  can  be  made  in  support  of  the 
notion  that  the  earth's  magnetic  field 
results  from  fluid  motions  in  the  outer 
core.  The  simplest  way  to  produce  such 
motions  proves  to  be  thermal  convection. 
The  temperature  at  the  boundary  of  the 
inner  core  is  probably  fixed  by  latent  heat 
of  crystallization,  and  it  becomes  neces- 
sary to  find  conditions  under  which  the 
temperature  at  the  outer  boundary  of  the 
core  remains  steady  or  decreases  slightly 
with  time.  If  this  condition  is  not  met,  an 
adiabatic  gradient  will  not  be  maintained 
and  thermal  convection  will  cease. 

Two  processes  tend  to  heat  the  lower 
mantle :  conduction  of  heat  from  the  core 
down  the  adiabatic  gradient,  which  is 
presumed  to  exist;  and  residual  radio- 
activity in  the  mantle  itself.  If  conditions 
can  be  found  under  which  cooling  more 
than  offsets  these  sources  of  heating,  an 
obstacle  in  the  path  of  dynamo  theories 
of  the  magnetic  field  will  be  removed. 

The  usual  way  of  estimating  initial 
temperatures  in  the  mantle  is  to  assume 
that  they  correspond  to  some  melting 
curve.  Empirically  the  most  satisfactory 
such  curve  is  given  by  the  Simon  equa- 
tion, which  predicts  a  very  small  thermal 
gradient  in  the  lower  mantle.  But  this 
prediction  does  not  take  account  of  the 
transition  zone  between  400  and  1000  km. 
Evidence  that  phase  changes  are  respon- 
sible for  this  region  is  accumulating,  and, 
if  they  are,  the  melting  curve  should 
steepen  in  this  range  of  depth.  Further 
work  will  be  required  to  show  whether 
steep  gradients  can  persist  throughout 
the  lower  mantle  and  prevent  any  rise  in 
temperature  near  the  boundary  of  the 
core. 


172 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


5000 


500 


1000      1500      2000      2500      3000 

Depth  ,  km 


Fig.  60.     Four  cases  of  initial  temperature  considered.  Reading  from  bottom  to  top,  curves  are 
for  n  =  1,  2,  3,  and  4,  respectively. 


300- 


o 

o 

>> 
CD 

m 

c 

o> 

c. 

"o 
o 
o 


200- 


100- 


1 

1 

i 

i                    i 

1 

\N  =  4 

^^^^ 

s^           N  =  3       " 

&■"                                   1 

i 

N  =  2 

0        500       1000      1500      2000      2500      3000 

Depth,  km 

Fig.  61.     Cooling  of  nonradioactive  earth.  Initial  temperatures  shown  in  figure  60. 


GEOPHYSICAL   LABORATORY  173 

THE  AGES  OF  ROCKS  AND  MINERALS 

G.  R.  Tilton,  G.  L.  Davis,  S.  R.  Hart,10  and  L.  T.  Aldrich10 

Some  drill  and  bore 
The  solid  earth,  and  from  the  strata  there 
Extract  a  register,  by  which  we  learn 
That  he  who  made  it,  and  revealed  its  date 
To  Moses,  was  mistaken  in  its  age. 

Cowper,  The  Task 

Knowledge  of  the  earth's  crust  has  been  given  showing  the  extent  of  the  observa- 
extended  into  the  past  by  interpreting  tions  as  of  1957.  So  much  more  informa- 
the  isotopic  ages  of  minerals.  Some  age  tion  is  now  available  that  it  seems 
measurements  give  new  dimension  to  appropriate  to  discuss  the  problem  again, 
existing  geological  concepts;  others  may  Our  interest  has  been  particularly 
allow  a  choice  between  conflicting  ideas  stimulated  by  additional  age  investiga- 
or  provide  a  basis  for  new  consideration  tions  in  the  southwestern  United  States. 
unhampered  by  preexisting  conceptions.  The  results  summarized  in  table  25  and 
An  example  of  the  extension  of  knowledge  figure  62  indicate  that  crystallization  of 
is  afforded  by  an  outline  map  of  the  igneous  rocks  occurred  approximately 
central  part  of  the  North  American  1300  to  1500  m.y.  ago  in  an  area  extending 
continent.  A  regular  pattern  of  ages  has  from  southeast  Missouri  to  eastern  New 
been  developed  from  the  age  measure-  Mexico.  The  earlier  investigations  of 
ments  by  a  number  of  laboratories  in  the  Aldrich,  Wetherill,  Davis,  and  Tilton 
United  States  and  Canada.  No  similar  (1958)  and  Giletti  and  Damon  (1961) 
regularity  is  apparent  on  other  conti-  have  reported  similar  ages  in  western 
nents.  Geophysical  implications  of  these  Arizona  and  Colorado, 
results  await  further  study.  The  relation  When  these  ages  are  compared  with 
of  measured  ages  to  a  geologic  problem  others  from  central  North  America  it  is 
at  Rainy  Lake  on  the  Minnesota-Ontario  seen  that  older  ages  occur  to  the  north 
boundary  is  being  studied.  Zircon  meas-  and  west  of  the  1300-1500  m.y.  rocks; 
urements  are  being  made  in  an  attempt  younger  ages,  to  the  south  and  east.  The 
to  find  ages  predating  a  2600-million-  distribution  of  ages  is  shown  in  figure  63, 
year-old  metamorphic  event.  In  Finland  based  on  a  survey  of  ages  in  the  literature, 
the  approximate  contemporaneity  of  the  Some  of  the  localities  have  been  more 
Karelian  and  Svecofennian  orogenies  has  thoroughly  studied  than  others.  In  favor- 
been  established,  although  geologic  evi-  able  places  ages  from  zircon,  mica,  and 
dence  had  earlier  been  interpreted  to  show  feldspar  are  in  agreement ;  in  less  favor- 
that  the  Svecofennian  belt  is  the  older.  able  ones,  only  K-Ar  or  Rb-Sr  ages  have 

been  measured  for  a  single  mineral.  The 

Geographic  Distribution  of  Mineral  Ages  Paleozoic  ages  (200-450  m.y.)  from  the 

in  the  Central  Portion  of  North  America  Appalachian  chain  that  serve  to  define 

the   <0.5-m.y.  zone  have  been  omitted 

One  important  application  of  the  dating  for  simplification;  likewise  the  post- 
of  rocks  is  to  ascertain  the  ages  and  their  precambrian  ages  from  the  Rocky  Moun- 
geographic  distribution  in  the  crystalline  tain  area  m  the  western  United  States 
basement  rocks  of  the  continent  of  North  have  not  been  shown. 
America.  Information  about  this  problem  Figure  63  shows  that  the  age  measure- 
has  been  accumulating  from  many  labora-  ments  jn  central  North  America  can  be 
tories  at  an  ever-increasing  rate  over  the 
past  decade.  In  Year  Book  57  a  map  was  10  Department  of  Terrestrial  Magnetism. 


174 


CARNEGIE     INSTITUTION      OF     WASHINGTON 


No. 


TABLE  25.     Mineral  Ages  from  the  Central  and  Southwestern  United  States 


Locality 


Mineral  and 
Rock* 


Age,  million  years 


Sr87         Ar40       Pb206      Pb207      Pb207      Pb208 


Rb87        K40        U238        U235       Pb206      Th232 


St.  Francis  Mts. 
M-5  Fredericktown,  Mo. 

M-16  Granite,  Mo. 

Decaturville  Uplift 
M-20  Decaturville,  Mo. 

Arbuckle  Mts. 
M-23  Tishomingo,  Okla. 

Sandia  Mts. 
A-26  Albuquerque,  N.  M. 


Muscovite  (P)  1430       1405 
Zircon  (G) 

Microcline  (G)  1300 

Biotite  (G)  1320       1280 


970       1120       1425       1230 


Muscovite  (P)  1445  1290 
1350  1250 
1340       1300 


Zircon  (G) 
Biotite  (G) 


Zircon  (G) 
Biotite  (G) 


970   1080   1320   1200 
1120   1250   1475   1290 


P,  pegmatite;  G,  granite. 


V.CHITA   MTS "X  .FRB,ER,CK  ...HSjSt^  «TS 


Fig.  62.     Locations  of  samples  from  the  central  and  southwestern  United  States. 


GEOPHYSICAL   LABORATORY 


175 


Fig.  63.  The  distribution  of  ages  in  crystalline  rocks  from  the  central  part  of  North  America. 
Circles  represent  ages  that  are  within  the  limits  specified  on  the  map  for  a  particular  zone.  Crosses 
are  ages  that  are  not  within  the  limits. 


grouped  by  age  and  geographic  location 
in  such  a  manner  that  few  exceptions  are 
found.  In  some  cases,  such  as  the  300-350 
m.y.  ages  for  biotite  from  Precambrian 
gneisses  in  the  western  part  of  the 
Appalachian  belt,  the  exceptions  have 
obvious  explanations.  These  ages  reflect 
Appalachian  metamorphism.  Others,  such 
as  the  HOO-m.y.-old  granite  at  Pikes 
Peak,  Colorado,  seem  to  represent  iso- 
lated intrusions  of  younger  bodies  of  rock. 
The  area  comprising  much  of  the  states 
of  Minnesota  and  Wisconsin,  the  northern 
peninsula  of  Michigan,  and  part  of 
Ontario  is  a  "mixed  age  zone"  in  which 
ages  similar  to  those  in  each  of  the 
surrounding  zones  can  be  found.  In 
general,  the  present  results  confirm  and 
extend  the  pattern  of  age  distribution 
given  in  Year  Book  57. 

There  is  as  yet  no  evidence  that  the 
regularities  in  the  occurrence  of  age  zones 


found  for  the  central  part  of  North 
America  will  be  found  on  other  conti- 
nents. On  the  contrary,  such  data  as 
exist  for  Europe,  Africa,  and  Australia, 
although  perhaps  less  extensive  than 
those  for  North  America,  indicate  rather 
complex  patterns  of  age  distribution. 

The  geophysical  significance  of  the  age 
distribution  in  figure  63  is  a  matter  for 
further  study.  Taken  at  face  value  the 
pattern  suggests  that  the  continent  has 
increased  in  extent  over  geologic  time, 
but  the  results  do  not  constitute  proof  of 
this.  For  a  land  mass  to  grow  at  the 
expense  of  an  ocean  basin  it  is  necessary 
to  form  a  crust  some  30  to  40  km  thick. 
At  present  it  is  not  certain  that  the  age 
distribution  in  North  America  applies  to 
such  a  thick  layer  of  rock.  This  factor, 
together  with  the  lack  of  similar  regu- 
larity in  age  distribution  on  other 
continents,  indicates  a  need  for  caution  in 


176  CARNEGIE      INSTITUTION      OF      WASHINGTON 

interpreting   the   results.    Whatever  the  than  any  age  found  previously.  That  the 

interpretation,   the   distribution  pattern  zircon    is    old    was    indicated    by    the 

shows  an  impressive  regularity.  Pb207-Pb206  age  of  2760  m.y.  The  search 

,  __.        .    „         .     _,       .. .  _  .  for  very  old  rocks,  as  well  as  the  need  to 

Ages  of  Minerals  from  the  Coutchiching  establisn    the    mechanisms    of    loss    of 

Sediments,  Rainy  Lake,  Ontario  daughter   elements,   was   stimulated   by 

In    the    vicinity    of    Rainy    Lake,    at  these  results,  and  rocks  of  the  area  were 

International  Falls  and  Fort  Frances  on  collected. 

the    Minnesota-Ontario    border,    A.    C.  New    age    measurements    have    been 

Lawson  mapped   a   series   of  metamor-  made  in  an  effort  to  determine  the  time 

phosed  sedimentary  strata  that  he  named  intervals  involved  in  the  formation  of 

Coutchiching,    lying   below   a   series    of  this  rock  sequence.   The  initial  studies 

metamorphosed    volcanic    rocks     (Kee-  have  been  on  the  mineral  zircon,  because 

watin) .  Circular  bodies  of  granite-gneiss  past  work  has  shown  that  zircon  ages  are 

are  enclosed  by  the  Coutchiching.  There  but  little  affected,  if  at  all,  by  the  forces 

is  lack  of  agreement  among  geologists  attendant  upon  regional  metamorphism. 

whether    they    are    intrusive    granites,  Consequently,  zircons  preserve  a  record 

mantled   gneiss   domes,   or   paragneisses  of  an  initial  crystallization.  They  survive 

derived    by    intense    metamorphism    of  detrital  cycles  because  of  their  physical 

Coutchiching  sediments.  properties,  thus  providing  some  clues  to 

In  Year  Book  59  the  ages  measured  for  the  source  of  the  sediments.  The  ages  of 

a  single  zircon  sample  from  the  Coutchi-  micas,  feldspars,  and  amphibole  minerals 

ching    metasediments    at    Rainy    Lake,  are  much  more  sensitive  to  the  effects  of 

Ontario,  were  reported.  The  age  pattern  geological  cycles. 

was  very  discordant,  so  much  so  that  The  ages  measured  are  given  in  table 

when  examined  from  the  viewpoint  of  26.  The  first  sample  is  the  one  to  which 

continuous  loss  of  lead  by  solid  diffusion  reference  has  already  been  made.  An  even 

(Tilton,  1960,  and  Year  Book  59)  the  very  more  discordant  pattern  was  found  for 

old  age  of  3800  m.y.  was  derived — older  sample  CC  35,  implying  an  impossibly 


TABLE  26.     Zircon  Ages,  Ontario 


No.  Rock  and  Location 


u, 

ppm 

Th, 
ppm 

Age,  million  years 

pb206 

TJ238 

pb207 

TJ235 

pb207 
pb206 

Pb208 

rpj1232 

648 

858 

1150 

1840 

2750 

1250 

874 

877 

1870 

2280 

2670 

1470 

823 

613 

2340 

2540 

2700 

2040 

1134 

n.d. 

2280 

2460 

2620 



264 

n.d. 

2450 

2600 

2730 

— 

121 

n.d. 

1960 

2300 

2630 

— 

1460 

3380 

520 

1080 

2500 

290 

263 

n.d. 

2140 

2450 

2730 

— 

261 

n.d. 

1540 

2020 

2550 

— 

RL  109     Coutchiching,  Rainy  Lake,  Ont. 

(impure) 
CC  21       Coutchiching?,  Rainy  Lake 

(or  granite  gneiss) 
CC  22       Coutchiching?,  Rainy  Lake 

(or  granite  gneiss) 
CC  26*     Coutchiching,  Rainy  Lake 
CC  29       Gneiss,  Rainy  Lake 
CC  20       Keewatin,  Rainy  Lake 

(metavolcanic) 
CC  35       Granite,  Bad  Vermilion  Lake 

(impure) 
CC  33       Granite,  Bad  Vermilion  Lake 
CC  43       Granite,  Saganaga  Lake,  Minn. 


*  Biotite  from  CC  26:  Rb-Sr  age,  2510  m.y. 


GEOPHYSICAL   LABORATORY 


177 


old  age  when  corrected  for  loss  of  lead  by 
continuous  diffusion.  Careful  study  of 
this  sample,  as  well  as  the  earlier  one, 
revealed  the  presence  of  an  impurity  in 
both  (15  per  cent  in  CC  35,  5  per  cent  in 
RL  109).  The  impurity  did  not  yield  an 
X-ray  pattern,  and  positive  identification 
has  not  yet  been  made.  The  ages  of  these 
two  samples  result  from  the  analysis  of  a 
mixed  system,  not  comparable  with  that 
of  the  other  zircons.  That  this  can  be  so 
is  evident  from  figure  64,  representing  all 
the  zircon  data  on  a  concordia  diagram. 


The  preliminary  conclusions  that  can 
be  drawn  from  the  results  of  the  zircon 
analyses  are: 

1.  The  Pb207-Pb206  values  for  all  the 
Rainy  Lake  zircons  lie  within  the  range 
2620-2750  m.y.,  a  very  narrow  range  in 
view  of  the  geological  complexity  of  the 
area.  The  ages  are  only  a  little  greater 
than  the  mica  ages  from  the  area. 

2.  The  source  rocks  for  the  zircons  in 
the  sediments  crystallized  earlier  than 
2600  m.y.  ago,  and  possibly  earlier  than 
2750  m.y. 


0.6- 


CC29 


OCIear.  somples 
©Impure  somples 


Pb207 
U235 

Fig.  64.     Parent-daughter  ratios  for  zircons  from  Rainy  Lake,  Ontario,  compared  with  the  curve 
calculated  for  loss  of  lead  by  continuous  diffusion  for  2750  m.y. 


The  least-squares  line  of  best  fit  to  the 
points,  excluding  the  two  questionable 
samples,  coincides  with  the  essentially 
linear  part  of  the  continuous  diffusion  loss 
curve  for  zircons  crystallizing  2750  m.y. 
ago.  The  impure  samples  lie  off  this  line. 
Another  sample  of  the  Bad  Vermilion 
Lake  zircon,  CC  33,  obtained  from  a 
different  part  of  the  granite,  gave  a 
pattern  conforming  to  that  of  the  rest  of 
the  zircons. 


3. 


A  single  Keewatin  sample  is  not 
significantly  different  in  age  from  the 
Coutchiching  zircons. 

4.  The  discordant  ages  of  the  pure 
samples  can  be  explained  by  loss  of  lead 
by  continuous  solid  diffusion. 

The  results  can  be  explained  in  two 
ways.  The  zircons  may  have  crystallized 
at  a  single  time  about  2700  m.y.  ago  in 
the  source  for  all  the  Coutchiching 
samples,  or  else  older  rocks  or  sediments 


178 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


were  so  strongly  metamorphosed  that  the 
zircon  ' 'clocks"  were  completely  reset 
2600-2700  m.y.  ago. 

Age  Relation  between  the  Karelian  and 
Svecofennian  Orogenies  in  Finland11 

Two  orogenic  belts  with  distinctly 
different  trends  have  long  been  recognized 
in  the  Precambrian  rocks  of  Finland.  The 
Karelian  belt  extends  with  a  north- 
northwest  trend  from  Lake  Ladoga  in 
southeastern  Finland  to  Finnish  Lapland, 
whereas  the  Svecofennian  belt  extends 
across  southern  Finland  in  a  general 
east- west  direction.  Many  geologists  be- 
lieve that  the  Svecofennian  belt  is  older 
than  the  Karelian.  The  evidence  in 
support  of  this  view  has  recently  been 
summarized  by  Eskola  (1961).  The 
principal  observations  are  that  the  trend 
of  the  Karelian  belt  seems  to  interrupt 
that  of  the  Svecofennian  belt  and  that 
granites  and  granite-gneisses  similar  to 
those  found  in  the  Svecofennian  belt 
occur  as  blocks  in  the  basement  complex 
of  the  Karelian  belt.  The  basement 
complex  in  the  Svecofennian  belt  is 
nowhere  recognized  with  certainty  by 
geologists. 

Kouvo  (1958)  reported  several  mica 
and  zircon  ages  for  the  intrusive  rocks  in 
both  belts  and  found  ages  of  1750  to  1850 
m.y.  in  each,  in  agreement  with  the  few 
measurements  of  earlier  workers.  Some 
geologists  have  accepted  the  viewpoint 
that  intrustion  of  rocks  occurred  simul- 
taneously in  the  two  belts;  others,  notably 
Eskola  (1961),  have  advanced  another 
interpretation.  Eskola  suggested  that  the 
influence  of  metamorphism  on  the  "older" 
Svecofennian  belt  at  the  time  of  intrusion 
of  rocks  of  the  "younger"  Karelian  belt 
was  sufficiently  strong  to  erase  the 
existing  age  record  in  the  different 
minerals.  This  will  be  called  a  "rejuvena- 
tion hypothesis." 

Wetherill,  Kouvo,  Tilton,  and  Gast 
(1962)  found  a  Pb207-Pb206  age  of  2240 

11  In  collaboration  with  Olavi  Kouvo,  Geo- 
logical Surve}-  of  Finland,  Otaniemi. 


m.y.  for  zircon  from  a  Svecofennian 
schist  near  Tampere,  100  miles  northwest 
of  Helsinki,  showing  that  complete  era- 
sure of  ages  had  not  taken  place  1800 
m.y.  ago  in  the  Svecofennian  belt.  Since 
no  age  determinations  had  been  made  on 
zircon  from  the  intrusive  rocks  in  this 
area,  the  question  of  rejuvenation 
throughout  much  of  the  Svecofennian 
belt  was  not  resolved  by  this  result.  The 
possibility  existed  that  neither  the  sedi- 
ments nor  the  intrusives  were  completely 
regenerated  1800-1900  m.y.  ago  in  the 
Tampere  area,  but  were  in  the  other 
areas  studied.  This  postulate  has  been 
shown  to  be  most  unlikely  by  work  in  the 
past  year. 

At  Tampere  a  Svecofennian  granodio- 
rite  intrudes  graywacke  and  phyllitic 
schists.  The  body  is  approximately  20  km 
in  diameter  with  numerous  dikes  and 
stringers  cutting  the  surrounding  sedi- 
ments. Many  large  outcrops  of  granodio- 
rite  occur  in  the  area,  so  that  it  is  possible 
that  this  body  is  part  of  a  considerably 
larger  mass.  Zircon  age  determinations 
have  been  made  on  three  samples:  a 
specimen  of  granodiorite  taken  about  1 
km  from  the  observed  contact  with  the 
schists;  schist  A,  collected  about  2  km 
from  the  contact  with  the  granodiorite; 
and  schist  B,  collected  5  km  from  the 
contact.  Zircons  from  these  specimens 
differ  in  appearance  in  that  rounding  is 
more  frequent  in  the  samples  from  the 
schist  than  from  the  granodiorite.  Obser- 
vations on  200  crystals  from  each  sample 
indicated  that  75  to  85  per  cent  were 
rounded  in  the  schist  samples,  only  5  per 
cent  in  the  granodiorite.  The  granodiorite 
contained  several  per  cent  of  crystals  with 
length-to- breadth  ratios  of  3  to  5;  such 
elongated  crystals  were  not  observed  in 
the  schists.  The  zircons  from  the  grano- 
diorite and  schist  are  dissimilar  in  form 
and  appear  to  represent  two  distinctly 
different  populations.  The  age  of  the 
granodiorite  zircon  should  not  be  appre- 
ciably influenced  by  zircon  from  the 
schist. 

The  age  results  are  given  in  table  27. 


GEOPHYSICAL   LABORATORY 


179 


TABLE  27.     Ages  for  Zircon  from  a  Svecofennian  Intrusive  and  the  Neighboring  Schists 


Concentration 
U 

i,  ppm 
Th 

Age,  : 

million 

years 

Rock 

pb206 

TJ238 

pb207 

TJ235 

pb207 
pb206 

pb208 

Granodiorite 
Schist  A 
Schist  B 

524 
476 
465 

105 
220 
214 

1710 
1850 
1790 

1810 
2030 
2000 

1920 
2220 
2230 

1900 
1580 
1720 

The  zircon  from  the  granodiorite  has 
nearly  concordant  age  values  compatible 
with  a  time  of  crystallization  about  1900 
m.y.  ago.  This  is  in  agreement  with  the 
results  of  Kouvo  on  other  Svecofennian 
intrusive  rocks.  On  the  other  hand,  the 
Pb207-Pb206  age  values  for  the  zircons  from 
the  schists  are  distinctly  older,  showing 
that  the  rejuvenation  hypothesis  does  not 
apply  in  this  area.  These  ages  are  strong 
evidence  that  intrusion  of  igneous  rocks 
occurred  about  1900  m.y.  ago  in  both  the 
Svecofennian  and  the  Karelian  belts  and 
that  the  two  orogenies  are  therefore 
approximately  contemporaneous. 

The  data  are  also  pertinent  to  the 
problem  of  discordant  lead  ages  for 
zircons.  The  temperature  conditions  un- 
der which  zircons  lose  lead  are  not  well 
understood  and  are  based  on  studies  of 
zircons  that  have  undergone  regional 
metamorphism.  Here  the  magnitudes  of 
time  and  temperature  are  poorly  known. 
The    present    observations    show    that 


intrusion  of  sizable  masses  of  rock  can 
occur  without  completely  erasing  the  age 
record  in  zircon  in  the  immediately 
surrounding  rock.  Knowledge  of  the  true 
age  of  the  zircons  would  permit  a  more 
restrictive  statement  to  be  made  about 
the  amount  of  lead  loss.  At  present  these 
data  do  not  uniquely  determine  the  age 
or  ages  of  the  zircons  from  the  schists. 
The  zircons  may  have  been  derived  from 
a  source  somewhat  older  than  shown  by 
the  Pb207-Pb206  age  values,  perhaps  2300 
m.y.  old.  Alternatively,  if  some  loss  of 
lead  from  the  zircons  did  occur  at  the 
time  of  granodiorite  intrusion  1900  m.y. 
ago,  or  if  sources  of  more  than  one  age 
contributed  zircon  to  the  schists,  some  or 
all  of  the  zircons  might  be  considerably 
older  than  2300  m.y.  Wetherill,  Kouvo, 
Tilton,  and  Gast  (1962)  found  2700-m.y.- 
old  rocks  in  the  pre-Karelian  basement 
complex;  rocks  of  this  age  might  have 
contributed  zircons  to  the  schists  at 
Tampere. 


ORGANIC   GEOCHEMISTRY 


Paleobiochemistry 

The  fatty  acids  are  major  components 
of  all  living  matter  and  are  among  the 
more  thermally  stable  organic  substances. 
In  principle,  they  can  survive  at  low 
temperatures  for  billions  of  years  and 
thus  might  be  found  in  sedimentary  rocks 
that  have  been  deposited  since  the  origin 
of  organisms  employing  fatty  acids.  Fatty 
acids  have  been  postulated  to  be  the 
major  material  from  which  some  petro- 
leum hydrocarbons  are  formed. 

We  have  extracted  and  identified  fatty 


acids  from  recent  and  ancient  rocks.  In 
an  attempt  to  provide  a  background  for 
interpreting  our  observations  we  have 
also  conducted  laboratory  studies  of  the 
stability  of  the  crude  components  of 
living  matter  at  elevated  temperatures. 
These  will  be  described  first. 

Thermal  Stability  of  Algae 
P.  H.  Abelson 

When  organic  detritus  is  deposited  in 
anaerobic  sediments  a  number  of  mecha- 
nisms act  to  alter  or  destroy  it,  including 
biological  activity  and  chemical  changes 


180 


CARNEGIE      INSTITUTION      OF      WASHINGTON 


resulting  from  interaction  among  the 
components  and  degradation  due  to  the 
intrinsic  instability  of  organic  matter. 
Unusual  circumstances  may  shield  the 
detritus  from  most  of  these  factors  ex- 
cept intrinsic  chemical  instability,  which 
thus  sets  an  upper  limit  to  the  long- 
term  survival  of  components.  Thermal 
stability  can  be  estimated  by  laboratory 
experiments  on  pure  compounds  at 
elevated  temperatures  coupled  with  ap- 
plication of  the  Arrhenius  equation  to 
extrapolate  to  ambient  temperatures.  For 
saturated  fatty  acids  this  procedure 
yields  decomposition  times  of  many 
billions  of  years.  In  nature,  however,  most 
of  the  organic  matter  is  degraded  more 
rapidly,  and  chemical  interactions  surely 
play  an  important  role.  To  investigate 
them,  work  was  started  last  year  on 
thermal  stability  of  the  components  of 
algae.  These  studies  have  been  extended 
to  include  an  examination  of  the  changes 


C-16  Sat. 


in  major  biochemical  components  and  a 
more  detailed  look  at  the  fate  of  fatty 
acids. 

Chlorella  pyrenoidosa  was  incubated 
both  wet  and  dry  in  the  absence  of  oxygen 
at  temperatures  of  190°  and  142°C.  The 
product  was  split  into  major  fractions  by 
the  conventional  procedure  employed  for 
fresh  tissue.  This  has  obvious  drawbacks 
since  after  degradation  the  fractions  no 

TABLE  28.     Thermal  Degradation  of  Algae 


190°C 


Control 


20 

hours 


12 
days 


142°C 

40 
days 


Cold  TCA  (H20- 

soluble  fraction) 

6 

9 

5 

9 

Lipides 

32 

25 

20 

24 

Hot  TCA 

(nucleic  acid) 

9 

2 

1 

2 

Protein 

48 

33 

10 

31 

Residue 

5 

31 

64 

34 

18  Days 

190° 

C 

C-I8(-2H) 


C-16  Sat 


C-I8(-2H) 


3  Days  190°  C 


Fig.  65.     Gas-liquid  chromatograms  of  methyl  esters  of  fatty  acids  extracted  from  Chlorella.  The 
algae  were  exposed  to  heat  for  varying  periods  of  time.  The  bottom  curve  is  for  a  control  specimen. 


GEOPHYSICAL   LABORATORY  181 

longer  behave  exactly  like   those  from  As  we  shall  see,  these  thermal  tests  agree 

fresh  material.  However,  we  can  make  only  in  part  with  what  has  been  observed 

useful  comparisons,  obtain  a  feeling  for  in  nature,  and  it  appears  that  additional 

what  is  happening,  and  gauge  the  role  of  mechanisms  operate  in  the  sediments  to 

chemical  interactions  and  degradation  in  destroy  unsaturated  fatty  acids, 
altering  the  organic  sediments.  Results  of 

such  experiments  are  displayed  in  table  Fatty  Acids  in  Sedimentary  Rocks 

28.  It  may  be  noted  that  with  prolonged  p  H  AMson  and  p  L  Parker 
incubation  a  major  amount  of  the  organic 

matter  is  converted  to  a  black  insoluble  Fatty  acids  have  been  extracted  from 

residue  similar  to  the  kerogen  of  sedi-  rocks  ranging  in  age  from  recent  to  500 

mentary  rocks.  Even  with  short  exposures  m.y.  old.  The  most  abundant  species  seen 

significant  changes  occur,  including  the  were  the  saturated  acids,  stearic  (C-18), 

virtual    disappearance    of    nucleic    acid,  palmitic    (C-16),    and    myristic    (C-14). 

Increase  in  the  cold  trichloroacetic  acid  Large  qualitative  and  quantitative  differ- 

extract  (water-soluble  fraction)  probably  ences  between  the  contents  of  source  algal 

arises  from  breakdown  products  of  other  detritus  and  the  residual   carbonaceous 

fractions.    Detailed   examination   of  the  material    of    the    sediments    have    been 

protein  and  lipide  fractions  provides  more  noted.   Most  striking  is  the  absence  of 

detail  on  what  has  occurred.  Some  of  the  unsaturated  fatty  acids  in  even  recent 

more  unstable  amino  acids  disappear  as  samples. 

expected,  but  even  the  more  stable  ones  Many  samples  have  been  examined,  of 

like  alanine  vanish  at  a  faster  rate.  Thus,  which  the  following  are  typical:  (1)  recent 

when  incubated  in  dilute  solution,  pure  mud  from  Gulf  Coast  off  Port  Aransas, 

alanine  has  a  half -life  of  2  X  10 7  sec  at  Texas,   collected  by  P.   L.   Parker;   (2) 

190°C.  When  it  is  incubated  as  part  of  recent  mud  from  San  Nicolas  Basin  off 

protein  of  algae  the  half-life  diminishes  to  Southern  California  collected  by  K.  0. 

106  sec  at  190°C.  Emery;  (3)  core  from  Pedernales,  Vene- 

The  fate  of  some  of  the  individual  fatty  zuela,  5000  years  old,  furnished  by  John 

acids  was  also  examined.   The  striking  M.   Hunt;   (4)   Green  River  shale  from 

feature  was  the  relatively  rapid  rate  of  Mahogany    ledge    about    40    m.y.    old, 

disappearance  of  the  more  highly  unsat-  collected  near  Rifle,  Colorado,  by  P.  H. 

urated  substances.  This  is  illustrated  in  Abelson;    (5)    sample  from   Alun  shale, 

figure  65,  which  displays  chromatograms  Sweden,    approximately    500    m.y.    old, 

of  methyl  esters  of  fatty  acids  extracted  furnished  by  Gosta  Salomonsson  of  the 

from  heated  and  from  control  Chlorella.  Swedish  Shale  Oil  Company. 

The    chromatogram    of    the    control  Samples  were  ground  in  a  ball  mill  if 

specimen  reveals  a  substantial  content  of  necessary  to  attain  small  particle  size, 

an  unsaturated  C-18  fatty  acid  containing  treated  with  aqueous  HC1,  filtered,  dried, 

three  double  bonds  (-6H).  On  heating  at  and  extracted  in  a  Soxhlet  extractor.  The 

190°C  for  3  days  this  compound  largely  crude  product  consisted  mainly  of  highly 

disappears,  and  it  has  practically  van-  colored  tarry  materials  amounting  to  1 

ished  after  18  days.  The  C-18  compound  to  10  per  cent  of  the  weight  of  organic 

with  two  double  bonds  is  less  sensitive,  matter  in  the  original  sample.  In  turn  the 

but  it  also  disappears  after  the  longer  desired  fatty  acids  constituted  as  little  as 

incubation.    The   saturated    and   mono-  0.1   per  cent  of  the   crude  extract.   To 

unsaturated    acids   were   about   equally  obtain  reasonably  resolved  peaks  from 

resistant,  the  saturated  apparently  being  gas-liquid  chromatography,  partial  purifi- 

the    more    enduring.    Under    favorable  cation  was  essential.  This  included  chem- 

thermal  conditions  both  these  classes  of  ical  refining  and  solvent  extraction.  The 

compounds  could  last  millions  of  years,  saturated  fatty  acids  can  withstand  rather 


182 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Fig.  66.     Chromatogram  of  esters  of  fatty  acids  extracted  from  surface  mud  collected  at  Port 
Aransas,  Texas. 


Fig.  67.     Chromatograms  of  esters  of  fatty  acids  extracted  from  a  grab  sample  from  San  Nicolas 
Basin  (lower  curve)  and  a  core  from  Pedernales,  Venezuela. 


drastic  oxidizing  and  reducing  treatment, 
during  which  the  tars  are  destroyed.  The 
methyl  esters  of  fatty  acids  are  much 
more  soluble  in  petroleum  ether  than  the 
tars  are.  Various  combinations  of  these 
treatments  were  employed,  and  their 
effectiveness  was  monitored  by  radio- 
active tracers.  In  the  exploratory  stages 
special  care  was  taken  to  preserve 
unsaturated  acids.  Later  it  became  clear 
that  saturated  fatty  acids  were  present 
in  substantial  quantities  in  old  rocks,  and 


procedures   were   modified   accordingly. 

Cooper  (1962),  who  has  recently  re- 
ported on  the  occurrence  of  fatty  acids  in 
rocks  and  petroleum  reservoir  waters, 
employed  urea  adduction  in  his  studies. 
This  procedure  concentrates  fatty  acids 
with  respect  to  unsaturates  and  tars  but 
with  the  small  quantities  of  saturated 
acids  available  tends  to  lead  to  relatively 
large  losses. 

In  figures  66  to  69  are  shown  chromat- 
ograms of  fatty  acids  obtained  from  rocks 


GEOPHYSICAL   LABORATORY 


183 


C-16 


Fig.  68.     Chromatograms  of  esters  from  Green  River  shale.  The  upper  curve  was  obtained  from 
material  that  had  a  special  treatment  with  concentrated  HI. 


C-16 


Fig.  69.     Chromatogram  of  esters  of  fatty  acids  extracted  from  Alun  shale.  The  crude  acids  were 
treated  with  alkaline  permanganate  to  partly  free  the  fatty  acids  of  tars. 


184 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


of  a  variety  of  ages  ranging  from  a  few 
years  to  500  m.y.  The  chromatograms  are 
not  strictly  comparable  because  of  differ- 
ences in  chemical  processing.  Neverthe- 
less, there  is  considerable  similarity  in 
these  traces.  All  indicate  the  presence  of 
C-14,  C-16,  and  C-18  saturated  fatty 
acids. 

In  the  recent  sediments  palmitic  acid 
(C-16)  was  the  major  constituent.  This  is 
in  keeping  with  its  ubiquitous  occurrence 
and  large  abundance  in  present-day 
organisms.  Oleic  acid  (C-18  —  2H),  which 
is  a  major  constituent  of  algae  and  which 
readily  survives  thermal  tests,  is  not 
present.  In  the  older  rocks  stearic  acid 
(C-18)  was  relatively  more  important.  We 
believe  that  this  may  or  may  not  imply 
differences  in  the  utilization  of  fatty  acids 
at  an  earlier  period.  There  are  many 
mechanisms  that  could  lead  to  relative 
losses  of  one  or  another  of  the  constitu- 
ents. 

The  amounts  of  fatty  acids  extractable 
from  old  rocks  are  relatively  small, 
2  X  10-4  to  10~5  gram  per  gram  of 
organic  matter.  The  same  is  also  true  of 
recent  sediments.  Since  the  original 
detritus  might  have  contained  10  to  20 
per  cent  fatty  acids,  a  rather  dramatic 
change  has  occurred.  Only  1  part  in  1000 
of  these  acids  apparently  remains  in  an 
extractable  form  after  a  short  period  of 
exposure  to  the  anaerobic  environment. 

To  investigate  this  low  yield,  tracer 
experiments  employing  C14-tagged  pal- 
mitic acid  were  carried  out.  These 
experiments  checked  the  efficiency  of 
procedures  once  acids  were  freed  from  the 
matrix;  they  could  not  shed  light  on  the 
efficiency  of  the  original  extraction.  They 
showed  that  a  labeled  substance  could  be 
added  to  a  mud  and  be  recovered  in  a 
crude  extract,  and  that  purifications 
could  be  carried  through  without  undue 
loss. 

The  disappearance  of  most  of  the  fatty 
acid  or  its  tight  binding  to  the  matrix 
when  detritus  is  converted  to  kerogen 
thus  remains  an  important  but  unsolved 
problem  of  organic  geochemistry. 


Another  significant  aspect  is  that  the 
fatty  acid  content  of  organic  matter  does 
not  change  much  with  time  from  the 
present  to  the  500-m.y.-old  specimen. 
This  gives  hope  that  even  older  occur- 
rences may  be  found. 

The  Isolation  of  Organic  Compounds  from 
Precambrian  Rocks 

T.  C.  Hoering 

The  ultimate  fate  of  most  organic 
materials  is  to  be  oxidized  to  carbon 
dioxide.  Some  organic  substances  escape 
this  fate  by  being  buried  in  sediments.  A 
few  remain  as  compounds  similar  to  those 
of  the  original  living  cells.  Thus  amino 
acids,  carbohydrates,  fatty  acids,  and 
pigments  have  been  found  in  sedimentary 
rocks.  However,  the  majority  of  the 
organic  compounds  in  rocks  have  been 
converted  to  an  insoluble  substance 
known  as  kerogen. 

At  25°C  organic  substances  are  un- 
stable with  respect  to  decomposition  into 
methane,  carbon  dioxide,  and  graphite. 
At  this  temperature  reactions  leading 
toward  these  products  require  times  as 
long  as  billions  of  years.  The  so-called 
■  'graphite"  of  Precambrian  sedimentary 
rocks  may  contain  intermediate  molecules 
in  the  chemical  pathways  of  the  decom- 
position of  kerogen. 

It  is  the  purpose  of  this  work  to  con- 
sider the  chemical  nature  of  the  carbon 
of  Precambrian  sedimentary  rocks  and  to 
see  whether  any  recognizable  organic 
compounds  can  be  isolated  from  it.  Any 
such  organic  compounds  need  not  bear 
much  resemblance  to  the  chemical  com- 
ponents of  living  cells,  but  as  the  nature 
and  transformations  of  kerogen  are 
gradually  understood  they  may  give  some 
insight  into  the  existence  and  the  nature 
of  Precambrian  life. 

The  reduced  carbon  of  Precambrian 
rocks  is  reminiscent  of  high-rank  anthra- 
cite coal,  and  therefore  some  of  the 
techniques  for  the  elucidation  of  coal 
structure  were  employed.  The  reactions 
used  included  (a)  oxidation  and  recovery 


GEOPHYSICAL   LABORATORY 


185 


of  aromatic  and  aliphatic  acids,  (6) 
thermal  pyrolysis  and  isolation  of  ali- 
phatic and  olefmic  hydrocarbons,  (c)  re- 
duction with  anhydrous  hydrogen  iodide 
and  identification  of  saturated  hydro- 
carbons, (d)  solvent  extraction  followed 
by  spectroscopy  of  the  extracts. 

For  experimental  simplicity,  much  of 
the  work  was  done  on  massive  graphite 
of  Precambrian  age.  The  samples  include 
the  following: 

1.  Michigami  coal  from  the  Iron  River 
formation  of  northern  Michigan.  It  has 
been  described  by  Tyler,  Barghoorn,  and 
Barret  (1957).  Samples  were  collected  by 
E.  S.  Barghoorn  and  P.  H.  Abelson. 

2.  Anthroxolite  from  Sudbury,  On- 
tario, Canada  (Thompson,  1956).  The 
samples  were  collected  by  P.  H.  Abelson. 

3.  Graphitic  material  from  the  Soudan 
iron  mine,  Oliver  Mining  Company, 
Soudan,  Minnesota.  Samples  were  col- 
lected by  F.  L.  Klinger. 

4.  Carbon  leader  from  the  Main  Reef 
series,  Transvaal,  South  Africa.  Sample 
was  donated  by  P.  Ramdohr. 

Some  work  was  done  also  on  the  finely 
dispersed  carbon  of  the  Gunflint  chert, 
the  Bulawayan  limestone,  and  the  Trans- 
vaal dolomite.  These  rocks  are  described 
in  another  section  of  the  writer's  report. 

Oxidation  of  coal  by  alkaline  potassium 
permanganate  is  a  well  known  reaction. 
The  products  are  a  mixture  of  benzene 
polycarboxylic  acids  (Holly  and  Mont- 
gomery, 1956).  Figure  70  is  a  drawing  of 
a  paper  chromatogram  of  the  aromatic 
acids  isolated  from  the  oxidation  of  the 
carbonaceous  material  from  the  Soudan 
iron  mine.  The  acids  on  the  chromato- 
gram appeared  as  dark  blue  and  fluores- 
cent spots  when  viewed  under  ultraviolet 
light.  The  ultraviolet  adsorption  spec- 
trum of  an  aromatic  acid  from  one  of  the 
spots  of  the  chromatogram  is  shown  in 
figure  71;  it  is  typical  of  this  class  of 
compounds.  Through  a  comparison  of  the 
rate  of  migration  of  known  substances  on 
paper  chromatograms,  the  presence  of 
benzenetricarboxylic,  benzenetetracar- 
boxylic,      and      benzenepentacarboxylic 


-< —   Ammonia-Ethonol 


Start 


© 


® 


A 

\) 

o 

£ 

CD 

U. 

o 

O 

o 

:3 
CD 

o 

T 

<35 

CD 


O    o    O 


o         % 


Fig.  70.  Paper  chromatogram  of  the  aromatic 
acids  from  the  oxidation  of  Michigami  coal.  A 
mixture  of  1  part  of  coal  with  1.6  parts  of  KOH 
was  renuxed  with  excess  KMn04  for  24  hours. 
The  solution  was  acidified,  treated  with  SO  2,  and 
evaporated  to  dryness.  The  solids  were  extracted 
with  diethyl  ether.  The  extract  was  separated  by 
two-dimensional  paper  chromatography  accord- 
ing to  the  procedure  of  Germain  (1959).  The 
separated  acids  gave  a  deep  blue  color  or  a  bright 
fluorescence  when  viewed  under  ultraviolet  light. 
A  comparison  of  Rf  values  and  colors  under 
ultraviolet  light,  with  known  acids,  indicated  the 
presence  of  benzene  polycarboxylic  acids. 


220     230     240     250     260     270     280     290 
Millimicrons 

Fig.  71.  The  ultraviolet  adsorption  spectrum 
of  an  aromatic  acid  from  the  oxidation  of 
Michigami  coal.  The  spot  numbered  2  in  the 
paper  chromatogram  shown  in  figure  70  was 
eluted  with  dilute  sodium  hydroxide,  and  the 
ultraviolet  adsorption  spectrum  was  taken. 


186 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


acids  was  indicated.  Benzenehexacar- 
boxylic  acid  (mellitic  acid)  was  identified 
in  all  samples,  but  as  this  compound  can 
be  made  from  purely  inorganic  graphite 
its  presence  is  of  little  significance  to  this 
research.  The  oxidation  products  were 
also  examined  for  low-molecular- weight 
aliphatic  acids,  but  only  acetic  acid  was 
identified. 

The  pyrolysis  of  coals  is  a  well  studied 
process.  Figure  72  is  a  tracing  from  the 
gas-liquid  chromatography  separation  of 


6°C/min 
Time >- 


Fig.  72.  Gas-liquid  chromatogram  of  the 
hydrocarbons  from  the  pyrolysis  of  Michigami 
coal.  Samples  of  graphite  were  pyrolyzed  in  a 
vacuum,  and  the  gases  were  pumped  off  for 
chemical  analysis.  The  temperature  was  raised 
gradually.  The  gases  given  off  below  250°C  were 
due  to  adsorbed  air.  At  300°C  hydrocarbon  gases 
began  to  be  evolved,  and  above  600°C  molecular 
hydrogen  was  observed.  The  gases  were  trans- 
ferred to  a  temperature-programmed  gas-liquid 
chromatograph  and  separated  with  a  6-foot 
silicone  rubber  packed  column.  This  figure  shows 
a  typical  chromatogram  with  gases  from  methane 
through  pentane  being  observed. 


the  hydrocarbons  derived  from  the  heat- 
ing of  Michigami  coal  in  a  vacuum.  All 
the  hydrocarbons  from  methane  through 
pentane  have  been  identified. 

Thermal    pyrolysis    of    coal    is    very 
destructive  to  any  structure  and  does  not 


give  much  insight  into  the  nature  of  the 
organic  substances  present.  The  chemical 
reduction  and  liquefaction  of  coal  may 
be  more  informative.  Figure  73  shows  a 
typical  mass  spectrometric  analysis  of  the 
saturated  hydrocarbons  obtained  from 
the  action  of  anhydrous  hydrogen  iodide 
on  Michigami  coal.  A  mixture  of  hydro- 
carbons from  methane  through  hexane  is 
indicated.  The  orders  of  magnitude  of  the 
yields  of  hydrocarbons  liberated  by 
pyrolysis  and  reduction  range  from  10  to 
100  parts  per  million  of  starting  rock. 

The  exhaustive  extraction  of  coal  by 
basic  solvents  such  as  pyridine  has  long 
been  a  means  of  isolating  organic  sub- 
stances. Infrared  adsorption  spectra  of 
organic  molecules  are  very  specific  for  the 
types  of  chemical  bonds  contained  in 
them.  Figure  74  is  an  infrared  adsorption 
spectrum  of  the  organic  substances 
extracted  from  the  carbonaceous  ma- 
terial of  the  Transvaal  dolomite  by 
pyridine.  The  presence  of  methylene 
groups  (-CH2-)  is  the  most  conspicuous 
feature  of  this  spectrum. 

The  chance  of  contamination  is  an 
ever-present  danger  in  the  search  for 
trace  amounts  of  organic  substances  in 
material  that  has  had  such  a  long  history 
as  the  rocks  studied  in  this  work.  Con- 
tamination in  the  chemical  reagents  and 
water  used  for  the  work  can  be  tested  by 
running  the  appropriate  blanks.  Airborne 
dust  or  pollen  is  another  source  of 
contamination.  By  using  a  number  of 
different  procedures  and  by  looking  for  a 
number  of  different  organic  substances, 
we  can  hope  to  decide  whether  laboratory 
contamination  is  a  problem.  Natural 
contamination  of  the  rock  during  the  long 
period  from  its  deposition  in  the  Pre- 
cambrian  to  the  present  is  much  more 
difficult  to  evaluate. 

The  results  obtained  so  far  support  the 
premise  that  the  so-called  "graphite"  of 
Precambrian  rocks  was  originally  kerogen. 
If  so,  it  is  of  interest  to  ask  whether  the 
organic  compounds  that  formed  this 
kerogen  were  the  product  of  living  cells 
or  whether  they   could  represent  abio- 


GEOPHYSICAL   LABORATORY 


87 


Mass 


Fig.  73.  The  mass  spectrum  of  the  hydrocarbons  from  the  treatment  of  Precambrian  "graphite" 
with  anhydrous  hydrogen  iodide.  Ten  grams  of  Michigami  coal  was  placed  in  a  bomb,  and  50  grams 
of  anhydrous  hydrogen  iodide  was  distilled  in.  The  bomb  was  heated  to  180°C  for  16  hours.  Substances 
volatile  at  100°C  were  distilled  off  the  reaction  mixture,  and  the  iodine  and  hydrogen  iodide  were 
removed.  The  gases  were  separated  into  fractions  by  gas-liquid  chromatography,  and  various  frac- 
tions were  admitted  into  the  mass  spectrometer  for  analysis.  This  figure  shows  the  mass  spectrum  of 
a  mixture  of  hydrocarbons  obtained  in  this  manner  from  Michigami  coal.  A  mixture  of  hydrocarbons 
from  methane  through  pentane  is  shown. 


e  io 

Wavelength  in  microns 


16 


Fig.  74.  Infrared  adsorption  spectrum  of  pyridine  extract  from  Precambrian  Transvaal  dolomite. 
The  "graphite"  from  the  Transvaal  dolomite  was  exhaustively  extracted  in  a  Soxhlet  apparatus  with 
pyridine.  The  pyridine  was  evaporated,  and  the  resulting  oil  was  pressed  into  a  KBr  pellet.  The 
sharp  adsorptions  at  3.4-3.5,  6.8-6.9,  and  14.0  microns  are  characteristic  of  isolated  methylene 
(-CH2-)  groups.  The  broad  adsorptions  at  5.5-5.6  microns  are  suggestive  of  substituted  aromatic 
hydrocarbons. 


logically  produced  organic  compounds 
from  a  period  that  preceded  terrestrial 
life.  A  number  of  the  rocks  studied  have 
textures  that  are  generally  described  as 
due  to  colonial  algae.  The  carbon  isotope 
studies  reported  by  the  writer  here 
indicate  that  photosynthesis  was  occur- 
ring during  the  time  of  their  formation. 
Thus  the  evidence  is  in  favor  of  the 
existence  of  biological  activity  very  early 
in  the  Precambrian  era. 


The  Biogeochemistry  of  the  Stable 
Isotopes  of  Carbon 

The  Isotonic  Composition  of  the  Carbon  of 
Fatty  Acids 

P.  L.  Parker 

Nier  and  Gulbransen  (1939)  first 
measured  variations  in  the  C13/C12  ratios 
of  naturally  occurring  carbon.  They  noted 
that    plant    and    animal     carbon    was 


188  CARNEGIE     INSTITUTION     OF      WASHINGTON 

slightly  depleted  in  C13  compared  with  recovered.  A  small  amount  of  the  total 

the  inorganic  carbon  of  limestone.  Craig  sample    was    analyzed    to    locate    and 

(1953)  in  a  detailed  survey  of  variations  identify  the  fatty  acids  present.  Then  a 

in  the  relative  abundance  of  the  carbon  large    sample    was    injected    into    the 

isotopes    confirmed    and    expanded    this  instrument,  and  the  pure  ester  of  each 

observation.   In  both  these  studies  the  fatty  acid  was  collected  as  it  streamed  out 

whole  plant  or  animal  was  combusted  to  of  the  detector.  Small  glass  tubes  passing 

carbon  dioxide,  and  so  the  measured  ratio  through  a  paper  cup  of  dry  ice  served  as 

represents    an    average    of    the    many  collectors.  To  obtain  2  or  3  mg  of  ester  it 

different  chemical  compounds  present  in  was  necessary  to  repeat  the  collection  two 

living    matter.    Living    matter    can    be  or  three  times.  The  glass  tube  containing 

broken    down    into    different    types    of  the  sample  was  placed  directly  in  the 

chemical    compounds    and    the    C13/C12  combustion   line,    and   the    sample   was 

ratios    of    these    compounds    compared,  burned  to  C02.  This  C02  was  used  for  the 

Abelson  and  Hoering  (1961)  carried  the  mass  analysis.  The  results  of  the  mass 

study  of  the  C13/C12  variations  to  the  analyses  are  expressed  in  terms  of  <5C13, 

molecular  level  for  several  amino  acids  the  parts  per  thousand  difference  between 

isolated  from  a  variety  of  photosynthetic  the  C13/C12  ratio  of  the  sample  and  a 

organisms.     Measurements     of     C13/C12  reference  material, 
ratios  of  fatty  acid  molecules  relative  to 
the    C13/C12   ratios    of    organisms    from 

which  the  acids  were  isolated  are  de-  dCu  =  ^"'wi/nu^'''  reference  X  1000 

scribed  in  the  following.  ^    '^   ^f^nce 

The  total  lipide  was  Soxhlet- extracted  A  negative  5C13  value  indicates  that  the 
from  the  samples  with  methanol  and  sample  contains  less  C13  than  the  stand- 
chloroform.  The  extract  was  taken  to  ard;  a  positive  value,  that  it  contains 
dryness  on  a  steam  bath  under  a  stream  more. 

of  dry  nitrogen.  The  residue  was  saponi-  In  view  of  the  complex  physical  manip- 
fied  for  2  hours  with  a  5  per  cent  solution  ulations  and  chemical  procedure  it  was 
of  potassium  hydroxide  in  methanol  and  necessary  to  run  a  number  of  control 
acidified  with  sulfuric  acid.  A  few  milli-  experiments  to  ensure  that  the  isotope 
liters  of  water  were  added,  and  the  fatty  fractionation  measured  was  not  thereby 
acids  were  extracted  from  the  mixture  brought  about.  The  esterification  reaction 
with  chloroform  which  was  then  taken  to  was  shown  not  to  fractionate  isotopes  by 
dryness.  The  water-free  residue  was  a  comparison  of  stearic  acid  with  methyl 
esterified  by  the  boron  trifluoride  method  stearate  made  from  the  stearic  acid.  If 
(Metcalfe  and  Schmitz,  1961).  This  final  the  acid  is  taken  as  0.0  per  mil  the  ester 
solution  was  a  complex  mixture  of  the  is  —0.5.  The  isotope  effect  in  the  gas 
methyl  esters  of  several  fatty  acids  as  chromatography  was  measured  by  corn- 
well  as  any  material  that  happened  to  paring  the  ester  before  and  after  chro- 
follow  the  chemical  procedure.  Final  matography.  If  the  ester  before  chro- 
purifi cation  and  separation  of  the  mixture  matography  is  taken  as  0.0  the  ester  after 
of  esters  into  specific  esters  was  brought  chromatography  and  100  per  cent  collec- 
about  by  high-temperature  gas  chro-  tion  is  —0.4;  after  only  50  per  cent 
matography.  collection  the  ester  is  +4.8  (100  per  cent 

The  chromatographic  analysis  was  collection  was  used  throughout  this  work) . 
performed  with  an  8-foot  copper  column  Isotope  fractionation  due  to  the  pro- 
packed  with  diethylene  glycol  succinate  cedure  is  less  than  1.0  per  mil.  On  the 
(LAC  3R  728)  on  acid-washed  chromo-  basis  of  repeated  runs  with  the  same 
sorb-P.  A  thermal  conductivity  detector  starting  material  the  overall  error  is 
was  used  so  that  the  samples  could  be  estimated  to  be  1.0  per  mil. 


GEOPHYSICAL   LABORATORY 


189 


By  means  of  these  techniques  the 
isotopic  compositions  of  the  fatty  acids 
of  two  algae,  a  marine  grass,  and  a 
plankton  tow  were  measured.  The  results 
are  given  in  table  29.  Without  exception 
the  fatty  acids  were  found  to  be  signifi- 
cantly depleted  in  C13  as  compared  with 
the  whole  cell.  Variations  between  differ- 
ent fatty  acids  from  the  same  organism 
are  too  close  to  experimental  error  to  be 
considered  significant. 

Chlorella  was  grown  in  the  laboratory 
in  a  solution  of  the  type  described  by 
Sorokin  and  Krauss  (1958),  5  per  cent 


TABLE  29.     5C13  of  Fatty  Acids 


Versus 

Versus 

Inorganic 

Total 

Sample 

Carbon  as 

Cell  as 

Reference 

Reference 

Chlorella  pyrenoidosa 

(inorganic  carbon 

was  tank  CO2) 

Total  cells 

-18 

0.0 

Palmitic 

-22 

-4 

Palmitoleic 

-22 

-4 

Stearic  plus  oleic 

-23 

-5 

Ulva  sp.  (inorganic  carbon 

was  sea-water  carbon) 

Total  cells 

-17 

0.0 

Palmitic 

-24 

-7.0 

Stearic 

-24 

-6.8 

Oleic 

-24 

-7 

Linoleic 

-24 

-7.5 

Linolenic 

-25 

-7.6 

Thalassia  sp.  (inorganic 

carbon  taken  as  sea- 

water  carbon) 

Total  cells 

-6.5 

0.0 

Palmitic 

-19 

-13 

Palmitoleic 

-21 

-15 

Stearic 

-21 

-14 

Oleic 

-21 

-14 

Linoleic 

-21 

-14 

Linolenic 

-17 

-11 

Plankton,  mostly  euphau- 

siids   (inorganic  car- 

bon   taken    as    sea- 

water  carbon) 

Total  cells 

-20 

0.0 

Myristic 

-24 

-4.4 

Palmitic  plus 

palmitoleic 

-26 

-6.2 

Stearic  plus  oleic  and 

linoleic 

-25 

-5 

CO2,  95  per  cent  air,  agitation,  and 
constant  illumination.  Table  29  shows 
that  the  fatty  acids  of  Chlorella  are  about 
4  per  mil  depleted  in  C13  as  compared 
with  the  total  cells.  The  Chlorella  used  in 
the  present  work  was  grown  in  the  same 
way  and  had  the  same  carbon  isotope 
ratio  as  the  Chlorella  used  by  Abelson  and 
Hoering  2  years  before.  According  to 
Abelson  and  Hoering  the  total  amino 
acids  of  Chlorella  are  enriched  in  C13  by 
3  per  mil  relative  to  the  total  cells.  Thus 
the  depletion  in  C13  of  the  fatty  acids  is 
balanced  by  the  enrichment  in  C13  of  the 
amino  acids.  The  isotope  variations  are 
in  the  right  direction  and  of  the  magni- 
tude to  yield  a  material  balance. 

The  other  three  samples,  from  the 
ocean,  were  collected  by  the  Marine 
Laboratory  of  the  University  of  Miami. 
Again,  the  fatty  acids  of  these  three 
samples  are  depleted  in  C13  relative  to  the 
whole  organism.  Ulva  is  a  marine  alga 
that  grows  attached  to  rocks  along  the 
coast.  Thalassia  is  a  marine  "grass"  that 
grows  in  great  abundance  in  the  shallow 
bays  of  the  Gulf  and  Atlantic  coast.  The 
euphausiids  are  small  animals  that  live 
in  the  open  sea. 

The  isotope  fractionation  in  the  forma- 
tion of  the  fatty  acids  is  in  the  same 
direction  for  all  three  of  the  photosyn- 
thetic  organisms.  If  the  <5C13  of  the  feed 
C02  is  taken  as  0.0,  the  5C13  values  of  the 
fatty  acids  for  all  the  plants  fall  between 
—  17  and  —25,  suggesting  that  the 
biochemical  reactions  involving  isotope 
fractionation  in  going  from  C02  to  fatty 
acids  are  similar  for  all  the  plants  studied. 

Petroleums  are  depleted  in  C13  relative 
to  modern  organisms  (Silverman  and 
Epstein,  1958).  Petroleum  derived  in 
large  part  from  fatty  acids,  which  have 
been  shown  to  be  generally  depleted  in 
C13  relative  to  whole  organisms,  might 
reflect  this  depletion  in  C13.  This  is 
probably  too  simple  a  picture,  because 
the  organic  molecules  enriched  in  C13 
must  also  be  accounted  for.  Nevertheless, 
knowledge  of  the  isotopic  composition  of 
specific  types  of  molecules  from  living 


190 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


organisms  may  give  some  clues  about  the 
ultimate  fate  of  the  many  different  types 
of  organic  molecules  trapped  in  the 
sediments. 


The  Stable  Isotopes  of  Carbon  in  the 

Carbonate  and  Reduced  Carbon  of 

Precambrian  Sediments 

T.  C.  Hoering 

When  inorganic  carbon  is  fixed  by  plant 
photosynthesis  there  is  an  isotope  effect 
and  living  cells  have  a  lower  concentra- 
tion of  the  heavy  isotope  of  carbon  than 
the  carbon  dioxide  or  bicarbonate  ion  of 
their  environment.  The  analyses  of  the 
C13/C12  ratio  of  a  large  number  of 
carbonates  and  reduced  fossil  carbons 
have  been  published.  Ages  of  these 
samples  range  from  recent  back  to  the 
Cambrian.  In  general,  the  C13/C12  ratio 
of  carbonates  is  1.02  to  1.03  times  that  of 
the  associated  reduced  carbon,  undoubt- 
edly because  of  the  isotope  fractionation 
during  photosynthesis. 

Wickman  (1941)  and  Rankama  (1948) 
have  proposed  taking  the  isotopic  compo- 
sition of  the  graphite  in  very  old  rocks  as 
an  indication  of  biological  or  nonbio- 
logical  origin.  Their  reasoning  was  criti- 
cized by  Craig  (1954)  for  a  number  of 
reasons,  including  their  rather  arbitrary 
grouping  of  carbon  isotope  ratios  into 
biological  and  nonbiological. 

The  purpose  of  the  present  work  was 
to  measure  the  isotopic  composition  of  the 
carbon  in  coexisting  carbonates  and 
reduced  carbons  in  some  of  the  very 
oldest  rocks  of  the  Precambrian.  The 
existence  of  isotope  fractionation  between 
the  oxidized  and  reduced  forms  of  carbon 
in  a  rock  that  has  had  a  mild  thermal 
history,  especially  if  the  magnitude  of  the 
fractionation  is  similar  to  that  found  in 
rocks  of  known  biological  association, 
suggests  that  photosynthesis  was  occur- 
ring during  the  time  of  deposition  of 
the  rock.  In  Precambrian  rocks,  in 
which  fossil  evidence  is  meager  or  non- 


existent, such  geo chemical  studies  give 
especially  important  evidence  on  the 
record  of  early  terrestrial  life. 

The  following  criteria  were  set  up  for 
the  selection  of  samples:  (a)  they  are 
sedimentary  rocks  that  have  suffered  as 
low  a  degree  of  metamorphism  as  possible ; 
(6)  they  have  presumptively  remnant 
algal  structures;  (c)  their  minimum  age 
can  be  estimated  from  the  isotopic  ages 
of  neighboring  igneous  rocks.  On  this 
basis,  the  following  rocks  were  used: 

1.  The  Gunfhnt  chert  from  near  Port 
Arthur,  Ontario.  This  rock,  carefully 
described  by  Tyler  and  Barghoorn  (1954), 
has  a  minimum  age  of  1.7  b.y.  Structures 
contained  in  it  have  been  described  as 
filamentous  blue-green  algae.  The  samples 
were  collected  by  E.  S.  Barghoorn. 

2.  The  algal  limestone  from  the  Belt 
series  of  Glacier  Park,  Montana.  They 
contain  structures  described  as  colonial 
algae  and  are  documented  by  Fenton  and 
Fenton  (1937).  General  aspects  of  the 
geology  of  the  region  are  described  by 
Ross  (1954).  A  minimum  age  of  1.2  b.y. 
is  suggested  (Tilton  and  Davis,  1959). 
The  samples  were  collected  by  P.  H. 
Abelson. 

3.  Domed  algal  growths  of  the  Dolo- 
mite series,  from  near  Schmidt's  Drift, 
Union  of  South  Africa.  The  structures 
have  been  described  by  Young  and 
Mendelsohn  (1948),  and  isotopic  ages 
measured  by  Nicolaysen  (1958)  have  set 
a  minimum  age  of  2.0  b.y. 

4.  The  Bulawayan  limestone  of  the 
Zwankendaba  series,  from  Bulawayo, 
Southern  Rhodesia.  The  algal  stromato- 
lites in  these  rocks  have  been  described 
by  McGregor  (1940),  and  the  rocks  have 
a  minimum  age  of  2.7  b.y.  (Holmes, 
1954).  These  are  taken  by  many  geolo- 
gists to  be  among  the  oldest  known 
sediments.  Samples  were  collected  by 
I.  Goldberg. 

5.  The  Randville  dolomite  of  the  Iron 
River  formation  from  near  Crystal  Falls, 
Michigan.  The  rocks  have  been  described 
by  James  (1958).  They  have  a  minimum 
age  of  1.5  b.y.   The  samples  were   col- 


GEOPHYSICAL   LABORATORY 


191 


lected  by  H.  James  and  P.  H.  Abelson. 
Thin  sections  have  been  cut  from  the 
carbonate  rocks  for  microscopic  examina- 
tion.   Typically   they   consist   of   partly 


6C13  = 


\C12/x      \C12/ 


standard 


X  1000 


\S^     /^     )  standard 

recrystallized    calcite   or   dolomite   with  There  is  clearly  a  large  difference  in  the 

black  specks  of  dispersed  carbon  in  them,  carbon  isotope  ratio  between  the  oxidized 

In  some,  the  black  particles  have  been  and  the  reduced  forms  of  carbon.  The 

concentrated  along  grain  boundaries  of  dCu  of  the  reduced  carbon  tends  to  be 

the  recrystallized  material.  slightly  more  negative  than  is  reported 

The    carbon   dioxide   for   the   isotope  for  coals  of  more  recent  ages,  possibly 

analysis  of  the   carbonate  fraction  was  because  of  isotope  fractionation  during 

generated  by  treating  with  concentrated  the  transformation  of  organic  material 

phosphoric    acid.    The    reduced    carbon  while  stored  in  the  sediments, 
fraction  was  isolated  from  the  carbonate         A  hypothesis  is  that  the  carbonate  and 

fraction  by  treating  with   hydrochloric  the  carbon  are  related  to  each  other  by 

and    hydrofluoric    acid.    The    resulting  some  inorganic  process.  The  reduction  of 

insoluble  residue  contained  the  reduced  carbon  dioxide  by  magmatic  gases  to  give 

carbon,     pyrite,     and     insoluble     metal  graphite    or   the   interaction   of    carbon 


TABLE  30.     Isotopic  Composition  of  the  Carbon  in  Precambrian  Rocks 


Sample 


5C13  Carbonate     5C13  Reduced         Difference 


Gunflint  chert 

-4.0 

-31.7 

27.7 

Algal  limestone,  Glacier  Park 

1.5 

-23.5 

25.0 

Algal  domes,  Dolomite  series,  S.  Africa 

0.8 

-29.9 

29.1 

Bulawayan  limestone 

-0.7 

-29.1 

28.4 

Randville  dolomite 

2.4 

-18.8 

21.4 

fluorides.    It    was    combusted    to    give  dioxide   and   methane   to   give   graphite 

impure    carbon   dioxide.    This    gas    was  would  involve  high  temperatures.  These 

unsuitable  for  isotope  analysis  and  was  processes  would  yield  isotope  fractiona- 

purified    by    gas-solid    chromatography  tion,  the  heavy  isotope  concentrating  in 

with  a  heated  column  of  silica  gel  and  the  carbon  dioxide  and  the  light  isotope 

with  helium  as  the  sweep  gas.  After  the  in  the  graphite  phase.  The  rocks  used  in 

eluted  carbon  dioxide  passed  through  a  the   present   study,    however,    show   no 

conductivity   cell  and  its  response  was  evidence    of    exposure     to     such    high 

measured,  it  was  frozen  from  the  helium  temperatures.    It   would   have   to   be   a 

stream   by   passing   it   through   a   trap  coincidence  that  the  distribution  of  the 

cooled  with  liquid  nitrogen.  This  purifi-  isotopes  is  so  similar  to  that  found  in 

cation  requires  only  about  5  minutes  and  rocks  of  known  biological  origin, 
yields  very  pure  carbon  dioxide,  suitable         The  results  of  this  work  are  consistent 

for  the  mass  spectrometer.  with  a  model  of  the  existence  of  photo- 

The    results    of    this    experiment    are  synthesis  and  biological  activity  in  the 

shown  in  table  30  and  are  expressed  in  oldest  rocks  of  the  Precambrian  era.  The 

parts    per    thousand    difference    in    the  experiments  described  in  another  part  of 

C13/C12  ratio  of  the  sample  and  a  standard  the   report   on  the   isolation   of  organic 

material,  NBS  Isotope  Reference  Sample  compounds  from  the  carbon  of  Precam- 

20.  brian  rocks  give  support  to  the  model. 


192 


CARNEGIE      INSTITUTION      OF      WASHINGTON 

MISCELLANEOUS   ADMINISTRATION 


Institute  on  Isotopes  and  Radioactivity 

A  week-long  institute,  or  special  course, 
"Isotopes  and  Radioactivity,"  designed 
to  acquaint  secondary  school  science 
teachers  of  the  Washington  area  with  the 
role  of  radioactive  isotopes  in  science  and 
civil  defense,  was  held  from  October  30 
to  November  3  at  the  Administration 
Building  of  the  Carnegie  Institution  of 
Washington.  It  attracted  much  favorable 
attention  from  the  press,  radio,  and 
television,  and  drew  enthusiastic  praise 
and  thanks  from  the  participants. 

Conceived  by  Philip  H.  Abelson,  Pres- 
ident of  the  Washington  Academy  of 
Sciences,  the  Institute  was  sponsored  by 
the  Academy  and  the  Joint  Board  on 
Science  Education.  At  the  request  of  Dr. 
Abelson,  the  morning-lecture  and  after- 
noon-laboratory curriculum  was  organ- 
ized by  Ralph  T.  Overman,  Chairman  of 
the  Training  Division  of  the  Oak  Ridge 
Institute  of  Nuclear  Studies.  About  140 
teachers  from  parochial,  private,  and 
public  schools  were  released  from  their 
classrooms  to  take  this  intensive  course, 
one  or  two  from  each  school.  Their  classes 
were  met  by  scientists  and  engineers  who 
had  volunteered  through  the  Joint  Board 
to  substitute  for  them. 

The  Institute  is  discussed  in  more 
detail  in  an  article,  by  Frank  L.  Campbell, 
which  appeared  in  the  December  1961 
issue  of  the  Journal  of  the  Washington 
Academy  of  Sciences. 

Journal  of  Geophysical  Research 

The  Journal  of  Geophysical  Research  is 
published  monthly  by  the  American 
Geophysical  Union  with  P.  H.  Abelson 
(Geophysical  Laboratory)  and  J.  A. 
Peoples,  Jr.  (University  of  Kansas),  as 
coeditors.  About  half  of  the  editorial 
work,  including  manuscripts  on  upper 
atmosphere  and  space,  as  well  as  some  of 
the  papers  involving  geochemistry,  are 
handled  at  this  Laboratory.  The  Journal 


is    regarded    by    many    as    the    world's 
leading  geophysical  publication. 

Though  publishing  about  5400  pages  a 
year,  the  Journal  has  one  of  the  fastest 
publication  times  among  scientific  jour- 
nals. This  accomplishment  is  due  to  the 
effective  efforts  of  Dr.  and  Mrs.  Peoples 
at  Kansas,  and  the  cooperation  of  Mrs. 
Lucile  Stryker  and  Miss  Mary  Jane  Miles 
of  Carnegie  Institution,  and  Mr.  A.  D. 
Singer  and  Miss  Marjorie  E.  Imlay  of  the 
Geophysical  Laboratory. 

Lectures 

During  the  report  year  staff  members 
and  fellows  were  invited  to  present 
lectures  as  follows : 

As  the  recipient  of  the  Regents'  Dis- 
tinguished Alumnus  Award  for  1961- 
1962,  P.  H.  Abelson  addressed  a  gathering 
at  the  Washington  State  University  on 
April  5,  1962.  At  the  American  Associ- 
ation for  the  Advancement  of  Science 
meetings  in  Denver  he  participated  in  the 
Extraterrestrial  Biochemistry  and  Bi- 
ology Symposium  and  the  Symposium  on 
Geochemical  Evolution — the  First  Five 
Billion  Years.  He  delivered  the  Retiring 
President's  Address  before  the  Washing- 
ton Academy  of  Sciences  and  the  Sigma 
Xi  Lecture  at  the  Institute  of  Biosciences, 
Florida  State  University;  and  he  partici- 
pated in  the  Panel  Discussion  on  the 
Chemical  Origin  of  Life  before  the 
Chemical  Society  of  Washington.  Dr. 
Abelson  also  lectured  to  the  Department 
of  Botany,  University  of  Missouri;  the 
Applied  Physics  Laboratory,  Johns  Hop- 
kins University;  the  Medical  School  at 
Georgetown  University;  the  Institute  for 
Space  Studies,  New  York  City;  Research 
Associates  at  the  National  Institutes  of 
Health;  and  the  National  Academy  of 
Sciences  at  its  annual  meeting  in  Wash- 
ington, D.  C. 

F.  R.  Boyd  lectured  at  the  Department 
of  Geology,  Pennsylvania  State  Uni- 
versity. 


GEOPHYSICAL    LABORATORY 


193 


C.  W.  Burnham  gave  two  talks  to  the 
Geology  Department  at  the  University  of 
Minnesota  and  addressed  the  Washington 
Crystal  Colloquium  at  the  National 
Bureau  of  Standards. 

S.  P.  Clark,  Jr.,  lectured  at  the  College 
of  Mineral  Industries,  Pennsylvania  State 
University;  the  Department  of  Geology, 
University  of  Minnesota;  the  Institute  of 
Geophysics,  University  of  California  at 
Los  Angeles;  and  the  National  Academy 
of  Sciences  Summer  Study  Session  on 
Nuclear  Processes  in  Geology,  Woods 
Hole,  Massachusetts. 

G.  Donnay  gave  a  lecture  on  color 
symmetry  groups  at  the  Mineralogical 
Institute  of  the  University  of  Tokyo, 
Japan. 

H.  J.  Greenwood  delivered  two  lectures 
at  the  Department  of  Geology,  California 
Institute  of  Technology. 

T.  C.  Hoering  addressed  the  Research 
and  Development  Laboratory  of  the  Gulf 
Oil  Company,  Pittsburgh;  the  Depart- 
ment of  Botany,  University  of  Maryland ; 
and  the  National  Academy  of  Sciences 
Summer  Study  Session  on  Nuclear  Proc- 
esses in  Geology,  Woods  Hole,  Massa- 
chusetts. He  also  participated  in  the 
Symposium  on  the  Biogeochemistry  of 
the  Isotopes  of  Sulfur  at  Yale  University. 

G.  Kullerud  lectured  at  the  National 
Research  Council,  Ottawa,  and  the 
Departments  of  Geology  at  Lehigh 
University  and  McGill  University.  He 
also  gave  a  series  of  five  talks  at  the 
Department  of  Geology,  Queen's  Univer- 
sity, Kingston,  Ontario,  and  two  lectures 
at  the  Department  of  Geology,  University 
of  Western  Ontario. 

N.  Morimoto  lectured  at  the  Depart- 
ments of  Geology  at  the  University  of 
California,  Berkeley,  and  the  University 
of  California,  Los  Angeles. 

P.  L.  Parker  addressed  the  Department 
of  Zoology,  Cornell  University,  and  the 
Institute  of  Marine  Science,  University 
of  Texas. 

H.  S.  Yoder,  Jr.,  gave  three  lectures  at 
Clemson  College  and  one  at  the  Lamont 
Geological  Observatory  of  Columbia  Uni- 


versity. During  a  visit  to  Japan,  sup- 
ported in  part  by  the  National  Science 
Foundation,  he  gave  lectures  at  the 
International  Symposium  on  Volcanology 
held  in  Tokyo  and  the  Departments  of 
Geology  of  Hokkaido,  Tohoku,  and 
Kyoto  Universities.  He  also  spoke  on 
high-pressure  techniques  at  symposia  in 
Kyoto  and  Osaka  sponsored  jointly  by 
the  Department  of  Geology  of  Kyoto 
University  and  the  Matsushita  Electric 
Industrial  Company. 

Penologists'  Club 

Six  meetings  of  the  Penologists'  Club 
were  held  at  the  Laboratory  this  year. 
The  following  papers  were  presented: 

"The  system  Fe-Zn-S;  a  preliminary  report 
after  five  years,"  by  Paul  Barton  and  Pete 
Toulmin  (U.  S.  Geological  Survey). 

"The  petrology  of  the  Rainier  underground 
tests,"  by  D.  E.  Rawson  (Lawrence  Radi- 
ation Laboratory). 

"New  observations  on  the  opaque  minerals 
of  stony  meteorites:  Facts  without  hypoth- 
eses," by  Paul  Ramdohr  (University  of 
Heidelberg  and  Geophysical  Laboratory). 

"Penological  applications  of  the  electron 
probe,"  by  S.  0.  Agrell  (Cambridge  Uni- 
versity) . 

"Field  and  laboratory  observations  pertain- 
ing to  the  origin  of  granite  pegmatites,"  by 
R.  H.  Jahns  (Pennsylvania  State  University). 

"Some  applications  of  sedimentary  petrol- 
ogy to  layered  intrusions,"  by  E.  Dale 
Jackson  (U.  S.  Geological  Survey). 


The  Summary  of  Published  Work 
below  briefly  describes  the  papers  pub- 
lished in  scientific  journals  during  the 
report  year.  In  addition,  the  following 
papers  are  now  prepared  for  publication: 
P.  H.  Abelson,  "Geochemistry  of  amino 
acids";  P.  H.  Abelson,  "Paleobiochem- 
istry";  R.  G.  Arnold,  R.  G.  Coleman,  and 
V.  C.  Fryklund,  "Temperature  of  crystal- 
lization of  pyrrhotite  and  sphalerite  from 
the  Highland-Surprise  Mine,  Coeur 
d'Alene  District,  Idaho";  F.  Chayes, 
"Numerical  correlation  and  petrographic 
variation";  G.  A.  Chinner  and  J.  F. 
Schairer,       "The       join       CasALSisO^- 


194 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


Mg3Al2Si30i2  and  its  bearing  on  the 
system  CaO-MgO-Al203-Si02  at  atmos- 
pheric pressure";  L.  A.  Clark,  "X-ray 
method  for  rapid  determination  of  sulfur 
and  cobalt  in  loellingite" ;  S.  P.  Clark,  Jr., 
"Temperatures  in  the  continental  crust"; 
B.  R.  Doe,  "Relationships  of  lead  isotopes 
among  granites,  pegmatites,  and  sulfide 
ores  near  Balmat,  New  York";  H.  J. 
Greenwood  and  H.  L.  Barnes,  "Binary 
mixtures  of  volatile  components";  G. 
Kullerud,  "Sulfide  research";  G.  W. 
Morey,  "The  action  of  water  on  calcite, 
magnesite,  and  dolomite";  N.  Morimoto, 
"On  the  transition  of  bornite";  N. 
Morimoto    and     G.     Kullerud,     "Poly- 


morphism in  digenite";  J.  V.  Smith  and 
W.  Schreyer,  "Location  of  argon  and 
water  in  cordierite";  G.  R.  Tilton,  G.  W. 
Wetherill,  and  G.  L.  Davis,  "Mineral  ages 
from  the  Wichita  and  Arbuckle  Moun- 
tains, Oklahoma,  and  the  St.  Francis 
Mountains,  Missouri";  A.  C.  Turnock 
and  H.  P.  Eugster,  "Fe-Al  oxides:  Phase 
relationships  below  1000°C";  D.  R. 
Wones,  "Phase  equilibria  of  'ferriannite/ 
KFe3+2Fe+3Si3O10(OH)2";  H.  S.  Yoder, 
Jr.,  and  C.  E.  Tilley,  "Origin  of  basalt 
magmas:  An  experimental  study  of 
natural  and  synthetic  rock  systems"; 
R.  A.  Yund,  "The  system  Ni-As-S:  Phase 
relations  and  mineralogical  significance." 


SUMMARY  OF  PUBLISHED   WORK 


(1352)  Heat  flow  in  the  Austrian  Alps.  S.  P. 
Clark,  Jr.     Geophys.  J.,  6,  54-63,  1961. 

Data  on  underground  temperature  obtained 
during  the  construction  of  the  Arlberg  and 
Tauern  tunnels  in  Austria  have  been  combined 
with  measurements  of  the  thermal  conduc- 
tivity of  42  samples  of  rock  from  near  the 
tunnels  to  calculate  the  terrestrial  heat  flow. 
The  value  in  the  Arlberg  is  found  to  be 
(1.9  +  0.2)  X  10-6  cal/cm2  sec;  that  in  the 
Tauern,  (1.8  ±  0.2)  X  10~6  cal/cm2  sec.  The 
new  results  are  in  good  agreement  with  the 
value  1.9  X  10~6  cal/cm2  sec  found  earlier  in 
the  Loetschberg  tunnel  in  Switzerland,  and 
indicate  that  relatively  high  geothermal  fluxes 
extend  into  the  eastern  Alps.  The  high  flux 
can  be  attributed  to  radioactive  heat  gener- 
ation in  a  thickened  crust. 

(1353)  Ponctualisation  des  charges  dans  les 
structures  cristallines  du  type  ionique. 
J.  D.  H.  Donnay  and  G.  Donnay. 
Compt.  Rend.,  253,  291-292,  1961. 

Pairs  of  neighboring  ions  with  the  same  sign 
are  replaced  by  points,  in  which  is  concen- 
trated the  total  charge  of  the  two  ions.  Such 
points,  regardless  of  their  sign,  are  equivalent 
as  far  as  morphology  is  concerned.  This 
punctualization  is  performed  on  various  pro- 
jections of  the  crystal  structure  of  barite 
(planar,  onto  coordinate  planes;  linear,  onto 
coordinate  axes). 


(1354)  A  structural  explanation  of  the  poly- 
morphism and  transitions  of  MgSiOg. 
W.  L.  Brown,  N.  Morimoto,  and  J.  V. 
Smith.     /.  Geol,  69,  609-616,  1961. 

Differences  between  the  polymorphs  of 
MgSiOa  consist  essentially  of  different  ways  of 
stacking  slabs  of  Si03  chains,  and  transitions 
between  the  polymorphs  may  be  effected  by 
movements  of  chains  by  two-thirds  of  the 
2-axis  spacing,  together  with  associated  dis- 
placements of  Mg  atoms  by  one-third  of  c. 
The  transitions  from  proto-  to  rhombic 
enstatite  and  from  proto-  to  clinoenstatite 
involve  the  same  percentage  of  displaced 
atoms,  but,  because  the  displaced  atoms  are 
distributed  more  uniformly  in  the  second 
transition,  it  is  thought  that  a  nucleus  of 
clinoenstatite  will  propagate  more  easily  than 
one  of  rhombic  enstatite.  This  suggestion  is 
consistent  with  the  rapid  metastable  forma- 
tion of  clinoenstatite  at  low  temperatures  and 
with  the  sluggish  formation  of  rhombic 
enstatite  (often  very  disordered)  from  proto- 
enstatite.  Shearing  stress  should  favor  the 
formation  of  clinoenstatite  in  conformity  with 
the  experiments  of  Turner  et  al.,  and  thus  it 
may  be  a  ''stress  mineral"  in  the  sense  of 
Harker.  Highly  complex  schemes  for  arranging 
the  Si03  chains  are  possible,  and,  as  an 
example,  three  possible  sequences  are  proposed 
for  the  enstatite  with  a  36  A  a  axis  described 
by  Bystrom. 


GEOPHYSICAL   LABORATORY 


195 


(1355)  Compositions  and  structural  states  of 
anhydrous  Mg-cordierites :  A  re-investi- 
gation of  the  central  part  of  the  system 
MgO-Al203-Si02.  W.  Schreyer  and  J. 
F.  Schairer.  J.  Petrol,  2,  324-406, 
1961. 

The  central  portion  of  the  system  MgO- 
Al203-Si02  has  been  studied  with  the  aim  of 
determining  the  range  of  solid  solution  as  well 
as  the  stability  limits  of  the  various  structural 
states  of  the  ternary  compound  cordierite.  The 
previously  suggested  limited  solid  solution 
between  cordierite  of  the  composition  2MgO- 
2A1203  •  5Si02  (2:2: 5)  and  Si02  is  now  believed 
to  exist  only  metastably.  Between  800°  and 
1300°C  the  composition  of  cordierite  was 
found  to  be  invariably  2MgO-2Al203-5Si02. 
Above  1300°C,  however,  there  is  evidence  for 
the  existence  of  limited  solid  solution  in 
cordierite  (2:2:5)  toward  a  theoretical  com- 
pound "Mg-beryl"  (3:1:6).  The  existence  of 
cordierite  solid  solution  at  liquidus  tempera- 
tures has  an  important  bearing  on  the  melting 
relations  of  many  compositions  within  the 
system.  Because  of  this  solid  solution  the 
courses  of  crystallization  of  melts  consisting  of 
normative  cordierite  (2:2:5)  and  small 
amounts  of  MgSi03,  for  example,  have  to 
follow  parts  of  the  boundary  curve  between 
the  cordierite  and  spinel  fields  with  these  two 
phases  coprecipitating  over  a  limited  range  of 
temperatures.  The  dividing  line  between 
compositions  that  complete  their  crystalliza- 
tion at  the  ternary  eutectic  forsterite  +  proto- 
enstatite  -f  cordierite  +  liquid,  1364°  ±  3°C, 
and  those  that  complete  their  crystallization 
at  the  ternary  eutectic  protoenstatite  +  cor- 
dierite +  tridymite  +  liquid,  1355°  ±  3°C, 
was  formerly  considered  to  be  the  join 
MgSi03-cordierite  (2:2:5).  Because  of  solid 
solution  in  cordierite  coexisting  with  liquid 
this  dividing  line  is  displaced  slightly  in  the 
direction  toward  more  siliceous  bulk  compo- 
sitions. Furthermore,  the  temperature  maxi- 
mum along  the  boundary  curve  cordierite  + 
protoenstatite  +  liquid  cannot  lie  at  the 
intersection  of  this  boundary  curve  with  the 
join  MgSi03-2:2:5,  but  must  lie  with  the  tie 
line  MgSi03-cordieriteS8.  The  position  of  this 
temperature  maximum  thus  moves  closer  to 
the  ternary  eutectic  protoenstatite  -f-  cor- 
dierite +  tridymite  +  liquid.  Temperatures 
and  compositions  of  some  of  the  invariant 
points  in  the  system  have  been  redetermined. 

On  the  basis  of  Miyashiro's  distortion  index, 
the  structural  states  of  the  cordierites  synthe- 


sized are  subdivided  into  high-cordierite, 
intermediate-state  cordierite,  and  "low"- 
cordierite.  High-cordierite  was  obtained  in  all 
compositions  at  any  temperature  as  the  first 
form  of  cordierite  to  crystallize.  With  con- 
tinued heating  at  appropriate  temperatures, 
this  metastable  high-cordierite  was  found  to 
go  over  gradually  through  intermediate-state 
cordierite  to  the  stable  form  "low"-cordierite. 
The  rate  of  this  transition  varies  with  bulk 
composition  and  generally  increases  with 
temperature.  In  contrast  to  this  metastable 
behavior  are  the  stable  relations  among  the 
polymorphs,  which  were  found  to  be  a  func- 
tion of  temperature  as  well  as  total  bulk 
composition  of  the  cordierite-bearing  mixtures. 
In  bulk  compositions  with  low  Al203/Si02 
ratios  {%  1:5)  high-cordierite  was  not  found  to 
be  a  stable  phase  at  any  temperature;  in  bulk 
compositions  with  intermediate  Al203/Si02 
ratios  high-cordierite  is  stable  only  in  the 
presence  of  much  liquid;  in  those  with  high 
Al203/Si02  ratios  ( >  1 : 2 : 5)  a  stable  transition 
from  "low"-cordierite  to  high-cordierite  takes 
place  at  subsolidus  temperatures.  This  rela- 
tionship is  considered  indirect  evidence  that 
Al/Si  ordering  is  the  principal  cause  of  the 
transition  from  high-  to  "low"-cordierite. 

Owing  to  solid  solution  the  transition  from 
"low"-  to  high-cordierite  in  the  presence  of 
liquid,  for  certain  bulk  compositions  with 
intermediate  Al203/Si02  ratios,  takes  place  in 
a  manner  that  cannot  be  described  by  a 
varying  distortion  index.  For  this  reason  a 
new  variable,  the  intensity  index,  defined  as 
*  =  /(5ii+42i)//(i3i),  is  introduced,  which  is  zero 
for  high-cordierite  solid  solutions  and  1.15  to 
1.35  for  "low"-cordierite. 

The  sensitive  dependence  of  the  structural 
behavior  of  cordierite  on  its  chemical  environ- 
ment excludes  the  possibility  of  using  this 
property  as  a  geologic  thermometer  to  a  very 
large  extent.  Experimental  investigations  on 
cordierite-bearing  synthetic  "haplobuchites," 
as  well  as  on  a  fused  shale  from  the  Bokaro 
coalfield  in  India,  revealed  that  high-cordierite 
is  not  a  stable  phase  for  these  bulk  compo- 
sitions at  any  temperature.  Natural  cordierites 
with  structural  states  close  to,  or  identical 
with,  high-cordierite,  which  have  been  found 
in  rocks  formed  at  high  temperatures  (buch- 
ites,  etc.),  are  believed  to  be  metastable 
products  of  crystallization.  They  are  preserved 
because  the  duration  of  heating  was  not 
sufficient  to  produce  the  stable  low-tempera- 
ture  form.   Petrographic   and   X-ray  studies 


196 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


show  that  there  is  a  close  relationship  between 
the  distortion  index  and  the  degree  of  perfec- 
tion of  the  crystal  form  of  cordierites  in  these 
rocks.  On  the  basis  of  these  results  it  seems 
possible  to  use  the  structural  state  of  cor- 
dierites, at  least  qualitatively,  as  a  geologic 
timer  for  the  crystallization  history  of  the 
enclosing  rock. 

(1356)  A  redetermination  of  equilibrium  rela- 
tions between  kyanite  and  sillimanite. 
S.  P.  Clark,  Jr.  Am.  J.  Sci.,  259, 
641-650,  1961. 

The  equilibrium  curve  between  kyanite  and 
sillimanite  has  been  established  by  quenching 
experiments  at  temperatures  between  1000° 
and  1500°C  and  pressures  between  17  and  24 
kb.  The  curve  is  given  by  the  expression 
P  =  4.1  +  13.2  X  10-3  T,  where  the  pressure, 
P,  is  in  kilobars  and  the  temperature,  T,  is  in 
degrees  Centigrade.  There  is  some  evidence 
that  the  phase  boundary  may  depart  from 
linearity  at  low  temperatures,  but  no  quanti- 
tative estimate  of  the  amount  of  curvature 
can  be  obtained  from  present  data. 

If  kyanite  forms  stably  in  nature,  pressures 
of  nearly  10  kb  are  required.  This  is  equivalent 
to  the  weight  of  about  30  km  of  overburden. 
Such  great  depths  of  burial  are  not  required  if 
pressure  is  contained  by  the  strength  as  well 
as  by  the  weight  of  the  overlying  rock.  It  is 
suggested  that  "tectonic  overpressures"  of  a 
kilobar  or  more  may  exist  in  rocks  undergoing 
deformation. 

(1357)  Metastable  solid  solutions  with  quartz- 
type  structures  on  the  join  Si02- 
MgAl204.  W.  Schreyer  and  J.  F. 
Schairer.     Z.  Krist.,  116,  60-82,   1961. 

Various  members  of  a  series  of  metastable 
solid  solutions  with  a  quartz-type  structure 
and  with  compositions  between  Si02  and 
MgAl204  have  been  synthesized  from  glass. 
Increasing  amounts  of  Mg+2  and  Al+3  in  the 
quartz  structure  cause  a  gradual  contraction 
parallel  to,  and  a  gradual  expansion  perpen- 
dicular to,  the  c  axis.  Siliceous  members  of  the 
series  are  optically  positive,  and  less  siliceous 
members  negative;  for  a  member  with  about 
73  weight  per  cent  Si02  the  birefringence  is 
zero.  The  refractive  indices  of  the  solid 
solutions  increase  with  decreasing  Si02  con- 
tent. Members  with  less  than  about  92  weight 
per  cent  Si02  exhibit  high-quartz  structures 
even   at   room   temperature,    whereas   more 


siliceous  members  go  through  an  inversion  to 
a  low-quartz  structure  when  quenched  to 
room  temperature.  The  temperature  of  this 
inversion  is  lower  than  that  of  pure  quartz 
(Si02)  as  a  result  of  the  presence  of  Mg+2  and 
A1+3  in  the  structure. 

(1358)  Phase  relations  in  the  system  Ni-As. 
R.  A.  Yund.  Econ.  Geol,  56,  1273- 
1296,  1961. 

Phase  relations  in  the  system  Ni-As  were 
determined  in  rigid  silica  glass  tubes,  in 
collapsible  gold  tubes,  and  by  differential 
thermal  analyses.  The  system  includes  the 
well  established  minerals  maucherite  (NinAss) 
niccolite  (Nii±IAs),  and  the  NiAs2  polymorphs 
rammelsbergite  and  pararammelsbergite. 

A  phase  with  the  composition  of  Ni3As 
(dienerite)  could  not  be  synthesized,  and  if 
this  phase  exists  it  must  be  stable  only  below 
200°C.  Ni5_xAs2  is  stable  to  approximately 
993°C  and  has  a  large  variation  in  its  Ni/As 
ratio.  Maucherite,  which  is  essentially  re- 
stricted to  NinAss  composition,  melts  incon- 
gruently  at  830°  ±  5°C  to  niccolite  plus  a 
liquid.  The  existence  of  a  metastable  form  of 
NinAss  appears  to  be  likely. 

Niccolite,  which  is  stable  to  962°  ±  3°C, 
also  has  a  large  variation  in  its  Ni/As  ratio. 
The  niccolite  solvus  between  NiAs  and  NiAs2 
is  not  useful  as  a  geothermometer,  however, 
since  it  is  nearly  vertical  in  the  temperature 
range  of  geologic  interest.  The  pararammels- 
bergite-rammelsbergite  inversion  was  found  to 
occur  at  590°C  under  the  vapor  pressure  of 
the  assemblage  when  pure  NiAs2  is  in  equi- 
librium with  niccolite.  The  inversion  tempera- 
ture is  raised  22°C/1000  bars,  giving  a  AH  of 
0.57  kcal/mole  at  590°C.  When  pure  NiAs2  is 
in  equilibrium  with  metallic  arsenic  instead  of 
niccolite,  the  inversion  temperature  is  approx- 
imately 8°C  higher.  Investigation  of  the 
inversion  temperatures  of  natural  specimens 
of  rammelsbergite  and  pararammelsbergite 
shows  that  solid  solution  of  elements  such  as 
Fe,  Co,  and  S  may  lower  the  inversion  by 
more  than  100°C. 


(1359)  Molar  volumes  and  thermal  expansions 
of  andalusite,  kyanite,  and  sillimanite. 
B.  J.  Skinner,  S.  P.  Clark,  Jr.,  and  D.  E. 
Appleman.  Am.  J.  Sci.,  259,  651-668, 
1961. 

Precise  measurements  of  unit-cell  param- 
eters of  four  andalusites,  four  sillimanites,  and 


GEOPHYSICAL   LABORATORY 


197 


five  kyanites  from  different  localities  lead  to 
the  following  molar  volumes  at  25°C:  anda- 
lusite,  51.550  +  0.011  cm3/mole;  sillimanite, 
49.918  ±  0.015  cm3/mole;  kyanite,  44.116  + 
0.021  cm3/mole. 

Unit-cell  parameters  at  high  temperatures 
were  measured  with  a  heating  stage  on  an 
X-ray  difTractometer.  From  these  data  the 
molar  volumes  and  thermal  expansions  of  all 
three  minerals  were  obtained  between  25°  and 
1050°C. 

(1360)  Woodring  Conference  on  Major  Bio- 
logic Innovations  and  the  Geologic 
Record.  P.  E.  Cloud,  Jr.,  and  P.  H. 
Abelson.  Proc.  Natl.  Acad.  Sci.  U.  S., 
47,  1705-1712,  1961. 

The  Woodring  Conference  was  held  at  Big 
Meadows  Lodge,  Skyline  Drive,  Virginia, 
June  14-16,  1961.  It  was  attended  by  twenty- 
three  biologists  and  geologists.  The  conference 
was  a  multidisciplinary  approach  to  major 
biological  innovations  in  the  context  of  the 
geologic  record,  and  with  emphasis  on  the 
nature,  manifestations,  and  timing  of  events 
leading  to  the  first  Metazoa.  This  report 
describes  the  proceedings  of  the  meeting  and 
includes  an  excellent  bibliography. 

(1361)  The  system  NaAlSi206-H20-argon :  To- 
tal pressure  and  water  pressure  in 
metamorphism.  H.  J.  Greenwood.  J. 
Geophys.  Res.,  66,  3623-3946,  1961. 

Phase  equilibrium  in  metamorphic  rocks  is 
affected  by  temperature,  pressure,  the  pro- 
portions of  nonvolatile  components,  and  the 
chemical  potentials  of  the  reacting  volatile 
components.  Theory  interrelating  these  vari- 
ables has  been  tested  by  studying  the  reaction 
analcite  — >  albite  -f-  nepheline  +  water  in  the 
presence  of  mixtures  of  water  and  argon.  New 
data  on  the  system  Ar-H20  permit  calculation 
of  the  composition  of  the  water-argon  mixture, 
which  should  equilibrate  with  the  phase 
assemblage  analcite  +  albite  +  nepheline. 
Experimental  determination  of  this  compo- 
sition as  a  function  of  pressure  at  constant 
temperature  is  in  good  agreement  with  the 
theory. 

(1362)  Stability  relations  of  glaucophane.  W. 
G.  Ernst.  Am.  J.  Sci.,  259,  735-765, 
1961. 

Stability  relations  have  been  determined  for 
glaucophane   [oNa2Mg3Al2Si8022(OH)2]  +  ex- 


cess vapor  and  for  quartz  +  glaucophane  + 
vapor  by  means  of  conventional  hydrothermal 
techniques.  The  high-temperature  stability 
limit  of  this  amphibole  ranges  from  850°C  at 
175  bars  vapor  (  =  total)  pressure  to  868°C  at 
2000  bars  Pvapor-  Neither  differential  stress  nor 
high  pressures  are  necessary  for  the  formation 
of  glaucophane.  The  presence  of  excess  silica 
lowers  its  high-temperature  stability  limit 
only  3°  to  6°C. 

Unusually  large  enthalpy  values  for  the 
reactions  glaucophane  — *  forsterite  +  ensta- 
tite  +  albite  -f-  vapor  and  quartz  +  glauco- 
phane — >  enstatite  +  albite  +  vapor  (330  + 
60  and  320  +  60  kcal/mole,  respectively)  can 
be  explained  only  in  part  by  the  change  in 
coordination  of  aluminum  from  6  in  glauco- 
phane to  4  in  albite.  The  entropy  of  glauco- 
phane at  864°C  and  1000  bars  vapor  pressure 
is  150  +  50  cal/deg/mole. 

Optical  properties  of  synthetic  glaucophane 
agree  well  with  data  for  natural  specimens. 
Unit-cell  dimensions  of  the  synthetic  material 
are  slightly  larger  than  those  of  natural 
glaucophanes. 

The  experimental  investigation  indicates 
that  glaucophane  is  stable  over  a  wide  range 
of  physical  conditions,  given  appropriate 
chemical  conditions.  Bulk  compositions  rich  in 
soda  and  magnesia  and  poor  in  lime  relative  to 
alumina  should  favor  production  of  glauco- 
phane. The  rare  occurrence  of  such  chemical 
environments  severely  restricts  the  crystal- 
lization of  glaucophane  in  nature. 


(1363)     Annual    report    of    the    Director    for 
1960-1961. 


(1364)  Age  measurements  on  rocks  from  the 
Finnish  Precambrian.  G.  W.  Wetherill, 
O.  Kouvo,  G.  R.  Tilton,  and  P.  W.  Gast. 
J.  Geol,  70,  74-88,  1962. 

New  mineral  age  measurements  are  reported 
from  several  subdivisions  of  the  Finnish 
Precambrian.  Samples  of  zircon,  feldspar,  and 
muscovite  collected  from  the  gneissose  pre- 
Karelian  basement  area  in  eastern  Finland 
indicate  an  age  of  about  2700  m.y.  for  these 
rocks.  In  contrast,  biotite  ages  from  the  same 
rocks  agree  at  1800  m.y.,  presumably  repre- 
senting the  effect  of  the  orogeny  at  this  time. 

Measurements  on  samples  of  mantled  gneiss 
domes  within  the  Karelian  belt  give  feldspar 
and  zircon  ages  supporting  the  correlation  of 
these  rocks  with  the  pre-Karelian  basement 


198 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


to  the  east,  and  again  the  biotite  ages  represent 
the  time  of  the  1800-m.y.  orogeny.  These 
results  are  closely  analogous  to  data  previ- 
ously reported  for  mantled  gneiss  domes  near 
Baltimore,  Maryland. 

Additional  measurements  on  the  younger 
Precambrian  rocks  of  Finland  confirm  earlier 
data  indicating  an  age  of  around  1800  m.y. 
for  plutonic  rocks  associated  with  both  the 
Svecofennian  and  Karelian  orogenic  belts. 

(1365)  Polymorphism  in  bornite.  N.  Mori- 
moto  and  G.  Kullerud.  Am.  Mineralo- 
gist, 46,  1270-1282,  1961. 

Synthetic  Cu5FeS4  and  natural  bornite  were 
observed  in  three  crystalline  modifications: 
(1)  a  high-temperature  form,o  face-centered 
cubic,  with  a  =  5.50  ±  0.01  A,  Z  =  1,  and 
probably  antifluorite  structure;  (2)  a  meta- 
stable  form,  cubic,  FdZm  or  F43m,  with 
a  =  10.94  ±  0.02  A,  Z  =  8;  (3)  a  low- 
temperature  form,  primitive  tetragonal,  space 
group  P!2iC,  pseudo-i42d,  with  a  =  10.94  ± 
0.02,  c  =  21.88  ±  0.04  A,  Z  =  16.  The  high- 
temperature  form  is  nonquenchable  and  is 
stable  only  above  228°  ±  5°C  (for  synthetic 
materials).  The  metastable  form  appears  on 
rapid  cooling  from  temperatures  above  that 
of  the  polymorphic  inversion;  it  changes  to 
the  low-temperature  form  slowly  at  room 
temperature.  The  low-temperature  and  the 
metastable  forms  are  closely  related  in  crystal 
structure,  as  shown  by  their  similar  intensity 
distributions  in  X-ray  patterns.  Twinning  of 
the  tetragonal  form  about  a  threefold  twin 
axis  [221 J  accounts  for  other  previously 
reported  "modifications." 

(1366)  Arsenopyrite  crystal-chemical  relations. 
N.  Morimoto  and  L.  A.  Clark.  Am. 
Mineralogist,  46,  1448-1469,  1961. 

The  composition  of  naturally  occurring 
arsenopyrite  varies  from  about  FeAso.gSi.i  to 
FeAsi.iSo.9,  as  indicated  by  the  more  credible 
published  chemical  analyses  and  one  new 
analysis.  Analytical  errors  probably  account 
for  any  apparent  deviations  of  the  Fe/(As+S) 
ratio  from  1:2. 

Five  arsenopyrites  of  different  compositions 
were  studied  by  single-crystal  X-ray  methods. 
The  changes  caused  by  increasing  arsenic 
content  are  (1)  the  triclinic  symmetry 
approaches  monoclinic  and  (2)  metrically  the 
cell  approaches  the  orthorhombic.  These 
pseudosymmetries  give  rise  to  two  types  of 


twinning.  Although  refinements  of  the  arseno- 
pyrite crystal  structure  by  means  of  (hOl)  and 
(hkO)  data  were  hampered  by  twinning,  the 
atomic  coordinates  obtained  in  this  investi- 
gation confirm  those  of  Buerger.  The  inter- 
atomic distances  Fe-As,  Fe-S,  and  As-S  are 
2.35,  2.25,  and  2.33  A,  respectively. 

Indexed  X-ray  powder  data  are  given.  The 
metrically  monoclinic  cell  constants  for  six 
analyzed  arsenopyrites  relate  linearly  to 
arsenic  content  and  inversely  to  sulfur  content. 
Provided  the  combined  minor  element  content 
is  below  1  per  cent,  the  curve  d1Si  =  1.6106  -+- 
0.00098:r,  where  x  is  the  arsenopyrite  arsenic 
content  in  atomic  per  cent,  enables  rapid 
determination  of  arsenopyrite  compositions  to 
within  1  atomic  per  cent. 

(1367)  Stability  relations  of  the  ferruginous 
biotite,  annite.  H.  P.  Eugster  and 
D.  R.  Wones.  J.  Petrol,  3,  82-125, 
1962. 

Annite,  KFe3AlSi3Oio(OH)2,  a  member  of 
the  iron  biotites  and  the  ferrous  analogue  of 
phlogopite,  has  been  synthesized  and  its  phase 
relations  have  been  determined  as  functions 
of  temperature,  fugacity  of  oxygen  (/o2),  and 
total  pressure  (Ptotai  ~  Ph2o  +  Pn2).  A 
method  for  controlling  /02  at  high  total 
pressures  is  described,  and  data  for  the 
"oxygen  buffers"  used  are  given.  Buffers  range 
from  quartz  -f-  iron  -f  fayalite  assemblages 
(low  /o2)  to  magnetite-hematite  assemblages 
(high  /o2).  Optical  properties  and  unit-cell 
dimensions  of  synthetic  annites  depend  on  the 
conditions  of  synthesis. 

By  recalculating  published  analyses  of 
natural  iron-rich  biotites  it  can  be  shown  that 
a  constant  hydrogen  content  cannot  be 
assumed  for  such  biotites.  Oxidation  may  have 
occurred  by  drying  at  115°C.  Octahedral 
occupancy  therefore  cannot  be  calculated  from 
such  data. 

Phase  relations  of  annite  are  presented  in 
2070  and  1035  bar  sections.  Depending  on 
foz-T  values,  annite  was  found  to  decompose 
to  one  of  the  following  assemblages:  hematite 
-f-  sanidine,  magnetite  +  sanidine,  fayalite  -f- 
leucite  +  kalsilite,  iron  +  sanidine.  All  de- 
compositions are  dehydration  and  redox 
reactions  and  are  sensitive  to  changes  in  /H2o 
and/o2  (or/H2o  and/H2).  At  2070  bars  total 
pressure  annite  +  magnetite  +  sanidine  can 
coexist  between  425°  and  825°C,  depending  on 
the  magnitude  of /o2. 

In  the  presence  of  quartz  the  stability  field 


GEOPHYSICAL   LABORATORY 


199 


of  annite  is  more  restricted.  Phase  equilibria 
in  the  system  KAlSi04-Si02-Fe-02-H2  have 
been  summarized  schematically. 

Wherever  possible,  thermodynamic  extrap- 
olations are  made  to  test  the  internal 
consistency  of  the  data.  Enthalpies  of  forma- 
tion are  calculated  for  both  annite  and 
phlogopite.  Ranges  of  /o2  values  in  nature  as 
well  as  mechanisms  for  changes  in  /o2  are 
investigated.  It  is  useful  to  distinguish  be- 
tween assemblages  that  are  internally  buffered 
with  respect  to  /o2  changes  and  those  that  are 
not  buffered.  The  applications  of  individual 
reactions  involving  annite  to  specific  geologic 
problems  are  discussed  with  respect  to 
igneous,  metamorphic,  and  sedimentary  rocks. 

(1368)  The  Ni-S  system  and  related  minerals. 
G.  Kullerud  and  R.  A.  Yund.  J. 
Petrol,  3,  126-175,  1962. 

The  system  Ni-S  has  been  studied  sys- 
tematically from  200°  to  1030°C  by  means  of 
evacuated,  sealed  silica  glass  tube  experiments 
and  differential  thermal  analyses.  Compounds 
in  the  system  are  Ni3S2  (and  a  high-tempera- 
ture, nonquenchable  Ni3±xS2  phase),  Ni7Se, 
Nix_xS,  Ni3S4,  and  NiS2.  The  geologic  occur- 
rence of  the  minerals  heazlewoodite  (Ni3S2), 
millerite  (/sNii_*S),  polydymite  (Ni3S4),  and 
vaesite  (NiS2)  can  now  be  described  in  terms 
of  the  stability  ranges  of  their  synthetic 
equivalents. 

Hexagonal  heazlewoodite,  which  is  stoichio- 
metric within  the  limit  of  error  of  the  experi- 
ments, inverts  on  heating  to  a  tetragonal  or 
pseudotetragonal  phase  at  556°C.  This  high- 
temperature  phase  (Ni3±a;S2)  has  a  wide  field 
of  stability,  from  23.5  to  30.5  weight  per  cent 
sulfur  at  600°C,  and  melts  incongruently  at 
806°  ±  3°C.  The  /3Ni7S6  phase  inverts  to 
aNi7S6  at  397°C  when  in  equilibrium  with 
Ni3S2  and  at  400°C  when  in  equilibrium  with 
aNiS.  Crystals  of  aNi7S6  break  down  to 
Ni3_xS2  +  «NiS  at  573°  ±  3°C.  The  low- 
temperature  form  of  Nii_xS,  corresponding  to 
the  mineral  millerite,  is  rhombohedral,  and  the 
high-temperature  form  has  the  hexagonal 
NiAs  structure.  Stoichiometric  NiS  inverts  at 
379°  ±  3°C,  whereas  Nii_xS  with  the  maxi- 
mum nickel  deficiency  inverts  at  282°  ±  5°C. 
The  Nii_xS-NiS2  solvus  was  determined  to 
985°  ±  3°C,  the  eutectic  temperature  of  these 
phases.  Stoichiometric  NiS  is  stable  at  600°C 
but  breaks  down  to  Ni3_xS2  and  aNii_xS 
below    797°C,    whereas    aNi^S    with    38.2 


weight  per  cent  sulfur  melts  congruently  at 
992°  ±  3°C.  Vaesite  does  not  vary  measurably 
from  stoichiometric  NiS2  composition  and 
melts  congruently  at  1007°  ±  5°C.  Polydym- 
ite breaks  down  to  aNii_xS  +  vaesite  at 
356°  +  3°C.  Differential  thermal  analyses 
showed  the  existence  of  a  two-liquid  field  in 
the  sulfur- rich  part  of  the  system  above  991°C 
and  over  a  wide  compositional  range. 

(1369)  Equilibrium  relations  between  pyrrho- 
tite  and  pyrite  from  325°  to  743°C. 
R.  G.  Arnold.  Econ.  Geol,  57,  72-90, 
1962. 

The  pyrrhotite  solvus  that  represents  the 
compositions  of  pyrrhotite  coexisting  in  equi- 
librium with  pyrite  was  determined  in  the 
temperature  range  325°  to  743°C  by  experi- 
ments conducted  in  sealed,  evacuated,  silica 
glass  capsules  and  at  pressures  equal  to  the 
pressure  of  the  vapor  in  equilibrium  with  the 
condensed  phases.  Experiments  conducted  in 
sealed,  collapsible  gold  tubes  demonstrate  that 
confining  pressures  of  2000  bars  do  not 
measurably  affect  the  position  of  the  solvus 
below  670°C. 

The  compositions  of  synthetic  hexagonal 
pyrrhotite  were  measured  within  ±0.13 
atomic  per  cent  Fe  with  the  aid  of  an  X-ray 
determinative  curve  that  relates  d(102)  to 
composition. 

X-ray  powder  data  and  a  general  descrip- 
tion are  given  for  an  unidentified  lamellar  iron 
sulfide  phase  occurring  in  rapidly  quenched 
iron-deficient  pyrrhotite. 

Temperatures  of  crystallization  of  ten 
natural  pyrrhotite-pyrite  assemblages  are 
estimated  by  means  of  the  pyrrhotite  solvus. 
The  temperature  of  crystallization  of  sphal- 
erite coexisting  with  pyrrhotite  and  pyrite  in 
four  of  these  samples  was  also  measured.  With 
very  few  exceptions  the  estimates  obtained 
from  the  two  methods  agree  well  within  the 
experimental  error. 

(1370)  Measurement  of  the  metal  content  of 
naturally  occurring,  metal-deficient, 
hexagonal  pyrrhotite  by  an  X-ray 
spacing  method.  R.  G.  Arnold  and 
L.  E.  Reichen.  Am.  Mineralogist,  47, 
105-111,  1962. 

It  is  shown  on  the  basis  of  fourteen  chem- 
ically analyzed  pyrrhotites  that  the  metal 
content  of  metal-deficient  natural  pyrrhotites 
may  be  measured  to  ±0.25  atomic  per  cent 


200 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


by  means  of  an  experimentally  derived  X-ray 
determinative  curve,  provided  that  the  com- 
bined concentration  of  nickel,  cobalt,  and 
copper  in  solid  solution  is  less  than  about  0.6 
per  cent  by  weight. 

(1371)  Metastable  osumilite-  and  petalite-type 
phases  in  the  system  MgO-Al203-Si02. 
W.  Schreyer  and  J.  F.  Schairer.  Am. 
Mineralogist,  47,  90-104,  1962. 

Two  new  compounds  have  been  synthesized 
metastably  in  the  system  MgO-Al203-Si02. 
One  has  a  structure  similar  to  that  of  osumilite 
and  other  related  phases,  such  as  the  synthetic 
compound  Na20-5MgO-  12Si02.  It  has  a 
composition  along  the  line  Si02-MgAl204, 
probably  close  or  equal  to  MgO-Al203-4Si02 
as  deduced  from  the  phase  assemblages.  On 
the  other  hand,  the  measured  mean  index  of 
refraction  (1.535),  according  to  the  Gladstone- 
Dale  relationship,  suggests  a  composition 
containing  less  Si02,  such  as  4Mg0-4Al203- 
7Si02.  The  other  compound,  whose  composi- 
tion is  unknown,  yields  a  powder  X-ray 
diffraction  pattern  similar  to  those  of  petalite, 
Li20-Al203-8Si02,  and  lithium  disilicate, 
Li20-2Si02.  The  two  metastable  phases  form 
during  devitrification  of  glass  of  certain  bulk 
compositions  at  relatively  low  subsolidus 
temperatures.  Upon  further  heating  they  are 
gradually  replaced  by  assemblages  that  are 
more  stable  for  these  bulk  compositions  and 
include  cordierite  and  a  silica  modification 
with  or  without  protoenstatite. 

(1372)  A  titaniferous  basalt  from  the  Island  of 
Pantelleria.  E.  G.  Zies.  J.  Petrol,  8, 
177-180,  1962. 

A  new  analysis  of  a  highly  titaniferous 
basalt  from  the  Island  of  Pantelleria,  first 
described  by  H.  S.  Washington,  is  presented. 
The  new  values  for  both  Ti02  and  A1203  differ 
appreciably  from  Washington's  and  produce 
marked  changes  in  the  calculation  of  the 
CIPW  norm.  The  analytical  procedures  by 
which  the  new  values  were  obtained  are  given 
in  outline. 

(1373)  Centers  of  charges  inferred  from  barite 
morphology.  J.  D.  H.  Donnay  and 
G.  Donnay.  Soviet  Phys.  Cryst.,  6, 
679-684,  1962. 

A  comparison  between  the  crystal  structure 
of  barite  and  its  morphological  development 
leads  to  the  concept  of  centers  of  charges.  The 


centers  of  charges  act  as  equivalent  points  in 
the  bond  assemblage  that  controls  the 
morphology. 


(1374)  Mineral  ages  from  the  Appalachian 
province  in  North  Carolina  and  Ten- 
nessee. G.  L.  Davis,  G.  R.  Tilton,  and 
G.  W.  Wetherill.  J.  Geophys.  Res.,  67, 
1987-1996,  1962. 

Age  measurements  are  given  for  nine  zircons 
and  nine  micas  from  the  Appalachian  orogenic 
zone  in  western  North  Carolina  and  eastern 
Tennessee.  These  measurements  provide  fur- 
ther evidence  for  the  existence  of  crystalline 
rocks  as  old  as  1000  m.y.  in  the  area.  A  still 
older  age  of  1300  m.y.  is  found  for  zircons  from 
two  gneissic  rocks;  these  older  zircons  are 
probably  detrital.  All  the  zircons  have 
discordant  ages.  The  discordances  are  com- 
patible with  loss  of  lead  by  continuous 
diffusion  or  episodic  loss  as  a  result  of  Paleo- 
zoic metamorphism.  Possible  difficulties  in 
ascribing  the  discordances  solely  to  episodic 
loss  during  Paleozoic  metamorphism  are 
pointed  out.  The  problem  of  loss  of  lead 
during  fusion  of  zircon  has  been  studied; 
losses  are  shown  to  be  negligible. 

( 1 376)  Skutterudites  (Co, Ni, Fe)  As3_ x '  Compo- 
sition and  cell  dimensions.  E.  H. 
Roseboom,  Jr.  Am.  Mineralogist,  47, 
310-327,  1962. 

Skutterudites  were  synthesized  by  heating 
mixtures  of  Co,  Ni,  Fe,  and  As  at  600°  to 
800°C  in  sealed,  evacuated  silica  tubes.  The 
resulting  phases  were  identified  by  powder 
X-ray  diffraction  methods  and  by  ore 
microscopy. 

Analyses  of  natural  skutterudites  by 
numerous  workers  have  indicated  nonstoichio- 
metric  compositions  with  a  deficiency  of  As. 
It  has  been  suggested  that  the  As  deficiency 
may  be  due  to  the  presence  of  other  phases  as 
impurities.  In  the  present  study,  synthetic 
cobalt  skutterudite  was  found  to  have  a  small 
but  real  As  deficiency,  even  in  the  presence  of 
crystalline  As,  but  this  deficiency  is  too  small 
to  account  for  the  large  deficiencies  indicated 
in  many  analyses  of  natural  skutterudites. 

Natural  skutterudites  are  known  to  vary 
widely  in  their  Co,  Ni,  and  Fe  content,  but 
pure  Ni  and  Fe  members  are  unknown.  The 
same  is  true  for  synthetic  skutterudites.  The 
limits  of  solid  solution  vary  little  with  tern- 


GEOPHYSICAL   LABORATORY 


201 


perature  between  600°  and  800°C,  and  most 
natural  skutterudites  fall  within  the  limits  of 
solid  solution  observed  for  the  synthetic 
phases. 

The  cell  edges  of  twenty-six  synthetic 
skutterudites  with  nickel  content  equal  to  or 
greater  than  iron  content  are  related  to 
composition  by  the  function  a  =  0.1240X  — 
0.0246F  +  8.2060,  where  a  is  the  cell  edge  in 
k,  X  is  the  mole  ratio  Ni/(Co  -f-  Ni  +  Fe), 
and  Y  is  the  mole  ratio  Fe/(Co  +  Ni  +  Fe). 
The  function  describes  the  measurements  to  a 
standard  deviation  of  0.00086  A.  The  cell 
edges  of  thirteen  analyzed  natural  skutteru- 
dites of  other  workers  show  fair  agreement 
with  the  synthetic  ones,  and  are  described  by 
the  above  function  to  a  standard  deviation  of 

o 

0.0097  A.  The  deviations  of  the  measured  cell 
edges  of  natural  skutterudites  from  cell  edges 
computed  using  the  function  are  not  demon- 
strably due  to  differences  either  in  (As  +  S)/ 
(Co  -j-  Ni  +  Fe)  ratios  or  in  total  sulfur 
content. 

(1377)  Erzmikroskopische  Untersuchungen  an 
Magnetiten  der  Exhalationen  im  Valley 
of  the  10,000  Smokes.  P.  Ramdohr. 
Neues  Jahrb.  Mineral.,  Monatsh.,  49-59, 
1962. 

These  fumaroles  produced  locally  large 
quantities  of  loosely  coherent  crystals  of 
magnetite.  Analyses  by  E.  G.  Zies  revealed 
the  presence  of  substantial  amounts  of  Zn,  Cu, 
Pb,  Mn,  Ni,  Co,  Mo,  and  Sn.  Various  writers 
have  thought  of  these  metals  as  existing  in  an 
anomalous  form  of  mixed  crystals  in  the 
magnetite  in  spite  of  the  fact  that  Zies  gave 
evidence  and  expressed  the  opinion  that  most 
of  the  metallic  constituents  were  present  as 
sulfides.  Actually  the  magnetite  contains  the 
following  sulfides:  FeS,  FeS2,  CuFeS2,  chalco- 


pyrrhotite,  bornite,  Cu2S,  Cu,  ZnS,  MoS2, 
FeAsS.  Besides  that  there  is  zincite;  only  a 
part  of  Zn  and  Mn  are  in  magnetite  itself. 
Paragenetically,  that  assemblage  is  of  interest 
for  ore  deposits  of  exhalative  origin  in  general. 

(1382)  Phase  equilibria  in  silicate  systems  at 
high  pressures  and  temperatures.  F.  R. 
Boyd,  Jr.  In  Modern  Very  High  Pres- 
sure Techniques,  edited  by  R.  H. 
Wentorf,  Jr.,  Butterworths,  Washing- 
ton, D.  C,  pp.  151-162,  1962. 

Studies  of  mineral  equilibria  at  high  pres- 
sures yield  data  for  estimating  the  conditions 
of  formation  of  igneous  and  metamorphic 
rocks.  These  data  also  provide  a  basis  for 
speculation  about  the  mineralogy  of  rocks  in 
the  earth's  mantle.  In  the  pressure  range  up 
to  50  kb,  experiments  are  most  easily  and 
accurately  made  with  single-stage  apparatus. 
Pressures  up  to  100  kb  can  be  obtained  with 
two-stage  apparatus  in  which  the  piston  is 
supported  by  a  KBr  cell  compressed  to  about 
20  kb. 

Most  silicates  whose  atomic  structures  are 
relatively  open  networks  invert  or  break  down 
to  denser  phases  in  the  pressure  range  10  to 
30  kb.  The  quartz-coesite  inversion  is  a 
chemically  simple  example,  and  the  P-T  curve 
for  this  reaction  in  the  temperature  range  700° 
to  1700°C  is  given.  Minerals  with  more  closely 
packed  atomic  structures  are  stable  to  much 
higher  pressures,  but  some  inversions  in  these 
minerals  have  been  discovered.  The  inversion 
of  the  olivine  Fe2Si04  to  a  spinel  form  is 
briefly  discussed.  Few  measurements  of  the 
effect  of  pressure  on  the  melting  relations  of 
silicates  have  thus  far  been  made,  although 
such  data  will  have  important  geologic  appli- 
cations. A  preliminary  melting  curve  for 
diopside  to  35  kb  is  given. 


BIBLIOGRAPHY 


Abelson,  P.  H.,  see  Cloud,  P.  E.,  Jr. 

Appleman,  D.  E.,  see  Skinner,  B.  J. 

Arnold,  R.  G.,  Equilibrium  relations  between 
pyrrhotite  and  pyrite  from  325°  to  743°C, 
Econ.  GeoL,  57,  72-90,  1962. 

Arnold,  R.  G.,  and  L.  E.  Reiehen,  Measurement 
of  the  metal  content  of  naturally  occurring, 
metal-deficient,  hexagonal  pyrrhotite  by  an 
X-ray  spacing  method,  Am.  Mineralogist,  47, 
105-111,  1962. 


Boyd,  F.  R.,  Jr.,  Phase  equilibria  in  silicate 
systems  at  high  pressures  and  temperatures, 
in  Modern  Very  High  Pressure  Techniques, 
edited  by  R.  H.  Wentorf,  Jr.,  Butterworths, 
Washington,  D.  C,  pp.  151-162,  1962. 

Brown,  W.  L.,  N.  Morimoto,  and  J.  V.  Smith, 
A  structural  explanation  of  the  polymorphism 
and  transitions  of  MgSi03,  J.  GeoL,  69,  609- 
616,  1961. 

Clark,  L.  A.,  see  Morimoto,  N. 


202 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Clark,  S.  P.,  Jr.,  Heat  flow  in  the  Austrian  Alps, 
Geophys.  J.,  6,  54-63,  1961. 

Clark,  S.  P.,  Jr.,  A  redetermination  of  equilibri- 
um relations  between  kyanite  and  sillimanite, 
Am.  J.  Sci.,  259,  641-650,  1961. 

Clark,  S.  P.,  Jr.,  see  also  Skinner,  B.  J. 

Cloud,  P.  E.,  Jr.,  and  P.  H.  Abelson,  Woodring 
Conference  on  Major  Biologic  Innovations  and 
the  Geologic  Record,  Proc.  Natl.  Acad.  Sci. 
U.  8.,  47,  1705-1712,  1961. 

Davis,  G.  L.,  G.  R.  Tilton,  and  G.  W.  Wetherill, 
Mineral  ages  from  the  Appalachian  province 
in  North  Carolina  and  Tennessee,  J.  Geophys. 
Res.,  67,  1987-1996,  1962. 

Donnay,  G.,  see  Donnay,  J.  D.  H. 

Donnay,  J.  D.  H.,  and  G.  Donnay,  Ponctualisa- 
tion  des  charges  dans  les  structures  cristallines 
du  type  ionique,  Compt.  Rend.,  253,  291-292, 
1961. 

Donnay,  J.  D.  H.,  and  G.  Donnay,  Centers  of 
charges  inferred  from  barite  morphology, 
Soviet  Phys.-Cryst.,  6,  679-684,  1962. 

Ernst,  W.  G.,  Stability  relations  of  glaucophane, 
Am.  J.  Sci.,  259,  735-765,  1961. 

Eugster,  H.  P.,  and  D.  R.  Wones,  Stability 
relations  of  the  ferruginous  biotite,  annite, 
J.  Petrol,  3,  82-125,  1962. 

Gast,  P.  W.,  see  Wetherill,  G.  W. 

Greenwood,  H.  J.,  The  system  NaAlSi206-H20- 
argon:  Total  pressure  and  water  pressure  in 
metamorphism,  J.  Geophys.  Res.,  66,  3923- 
3946,  1961. 

Kouvo,  O.,  see  Wetherill,  G.  W. 

Kullerud,  G.,  and  R.  A.  Yund,  The  Ni-S  system 
and  related  minerals,  J.  Petrol.,  3,  126-175, 
1962. 

Kullerud,  G.,  see  also  Morimoto,  N. 

Morimoto,  N.,  and  L.  A.  Clark,  Arsenopyrite 
crystal-chemical  relations,  Am.  Mineralogist, 
46,  1448-1469,  1961. 

Morimoto,  N.,  and  G.  Kullerud,  Polymorphism 


in  bornite,  Am.  Mineralogist,  46}  1270-1282, 
1961. 

Morimoto,  N.,  see  also  Brown,  W.  L. 

Ramdohr,  P.,  Erzmikroskopische  Untersuchung- 

en  an  Magnetiten  der  Exhalationen  im  Valley 

of  the  10,000  Smokes,  Neues  Jahrb.  Mineral., 

Montash.,  49-59,  1962. 
Reichen,  L.  E.,  see  Arnold,  R.  G. 
Roseboom,  E.  H.,  Jr.,  Skutterudites  (Co,Ni,Fe)- 

As3~:r:  Composition  and  cell  dimensions,  Am. 

Mineralogist,  47,  310-327,  1962. 
Schairer,  J.  F.,  see  Schreyer,  W. 

Schreyer,  W.,  and  J.  F.  Schairer,  Compositions 
and  structural  states  of  anhydrous  Mg- 
cordierites:  A  re-investigation  of  the  central 
part  of  the  system  MgO-Al203-Si02,  J.  Petrol., 
2,  324-406,  1961. 

Schreyer,  W.,  and  J.  F.  Schairer,  Metastable 
solid  solutions  with  quartz-type  structures  on 
the  join  Si02-MgAl204,  Z.  Krist.,  116,  60-82, 
1961. 

Schreyer,  W.,  and  J.  F.  Schairer,  Metastable 
osumilite-  and  petalite-type  phases  in  the 
system  MgO-Al203-Si02,  Am.  Mineralogist,  47, 
90-104,  1962. 

Skinner,  B.  J.,  S.  P.  Clark,  Jr.,  and  D.  E. 
Appleman,  Molar  volumes  and  thermal  expan- 
sions of  andalusite,  kyanite,  and  sillimanite, 
Am.  J.  Sci.,  259,  651-668,  1961. 

Smith,  J.  V.,  see  Brown,  W.  L. 

Tilton,  G.  R.,  see  Davis,  G.  L.;  Wetherill,  G.  W. 

Wetherill,  G.  W.,  O.  Kouvo,  G.  R.  Tilton,  and 
P.  W.  Gast,  Age  measurements  on  rocks  from 
the  Finnish  Precambrian,  J.  Geol.,  70,  74-88, 
1962. 

Wetherill,  G.  W.,  see  also  Davis,  G.  L. 

Wones,  D.  R.,  see  Eugster,  H.  P. 

Yund,  R.  A.,  Phase  relations  in  the  system  Ni-As, 
Econ.  Geol,  56,  1273-1296,  1961. 

Yund,  R.  A.,  see  also  Kullerud,  G. 

Zies,  E.  G.,  A  titaniferous  basalt  from  the  Island 
of  Pantelleria,  J.  Petrol,  3,  177-180,  1962. 


REFERENCES   CITED 


Abelson,  P.  H.,  and  T.  C.  Hoering,  Carbon- 
isotope  fractionation  in  formation  of  amino 
acids  by  photosynthetic  organisms,  Proc.  Natl 
Acad.  Sci.  U.  S.,  47,  623-632,  1961. 

Akimoto,  S.,  T.  Katsura,  and  M.  Yoshida, 
Magnetic  properties  of  TiFe204-Fe304  system 
and  their  change  with  oxidation,  J.  Geomagnet. 
Geoelec.  (Kyoto),  9,  165-178,  1957. 

Akimoto,  S.,  T.  Nagata,  and  T.  Katsura,  The 
Ti-Fe206-Ti2Fe05  solid  solution  series,  Nature, 
179,  37-38,  1957. 


Aldrich,  L.  T.,  G.  W.  Wetherill,  G.  L.  Davis,  and 
G.  R.  Tilton,  Radioactive  ages  of  micas  from 
granitic  rocks  by  Rb-Sr  and  K-A  methods, 
Trans.  Am.  Geophys.  Union,  39,  1124-1134, 
1958. 

Arnold,  R.  .G.,  Equilibrium  relations  between 
pyrrhotite  and  pyrite  from  325°  to  743°C, 
Econ.  Geol,  57,  72-90,  1962. 

Arnold,  R.  G.,  and  L.  E.  Reichen,  Measurement 
of  the  metal  content  of  naturally  occurring, 
metal-deficient,   hexagonal  pyrrhotite  by  an 


GEOPHYSICAL   LABORATORY 


203 


X-ray  spacing  method,  Am.  Mineralogist,  47, 

105-111,  1962. 
Atlas,  L.,  The  polymorphism  of  MgSi03  and 

solid  state  equilibria  in  the  system  MgSi03- 

CaMgSi206,  J.  Geol,  60,  125-147,  1952. 
Atzeni,  A.,  Sopra  una  particolare  colata  post- 

miocenica  delFedificio  vulcanico  del  M.  Arci, 

Rend.    Seminar.    Fac.    Sci.    Univ.    Cagliari, 

29{Y2),  1-15,  1959. 
Baker,  R.  G.,  D.  G.  Brandon,  and  J.  Nutting, 

The  growth   of  precipitates,   Phil.   Mag.,   4j 

1339-1345,  1959. 
Bandy,  M.  C.,  Geology  and  petrology  of  Easter 

Island,  Bull.  Geol.  Soc.  Am.,  48,  1589-1610, 

1937. 
Barrow,  G.,  On  an  intrusion  of  muscovite  biotite 

gneiss  in  the  southeast  Highlands  of  Scotland, 

Quart.  J.  Geol.  Soc.  London,  49,  330-358,  1893. 

Barth,  T.  F.  W.,  C.  W.  Correns,  and  P.  Eskola, 

Die  Entstehung  der  Gesteine,  Springer- Verlag, 

Berlin,  1939. 
Bell,  R.  E.,  and  R.  E.  Herfert,  Preparation  and 

characterization  of  a  new  crystalline  form  of 

molybdenum  disulfide,  J.  Am.  Chem.  Soc,  79, 

3351-3355,  1957. 
Berthois,  L.,  Contribution  a  F  etude  lithologique 

de  l'Archipelo  des  Acores,  Comun.  Serv.  Geol. 

Port.,  84,  1953. 
Bowen,   N.   L.,   The  ternary   system   diopside- 

forsterite-silica,  Am.  J.  Sci.,  88,  207-264,  1914. 

Bowen,  N.  L.,  Genetic  features  of  alnoitic  rocks 
at  Isle  Cadieux,  Quebec,  Am.  J.  Sci.,  3,  1-34, 
1922. 

Bowen,  N.  L.,  Phase  equilibria  bearing  on  the 

origin  and  differentiation  of  the  alkaline  rocks, 

Am.  J.  Sci.,  248A,  75-89,  1945. 
Bowen,   N.   L.,   and  0.   Andersen,   The  binary 

system  MgO-Si02,  Am.  J.  Sci.,  37,  487-500, 

1914. 

Bowen,  N.  L.,  J.  F.  Schairer,  and  H.  W.  V. 

Willems,  The  ternary  system  Na2Si03-Fe203- 

Si02,  Am.  J.  Sci.,  20,  405-455,  1930. 
Bowen,  N.  L.,  and  O.  F.  Tuttle,  The  system 

MgO-Si02-H20,  Bull.  Geol.  Soc.  Am.,  60,  439- 

460,  1949. 
Boyd,  F.  R.,  and  J.  L.  England,  Effect  of  pressure 

on  the  melting  of  diopside,  CaMgSi2Oc  and 

albite,  NaAlSi308,  in  the  range  up  to  50  kb,  J. 

Geophys.  Res.,  67,  in  press,  1962. 
Braun,     F.,     Morphologische,     genetische    und 

paragenetische  Trachtstudien  an  Baryt,  Neues 

Jahrb.  Mineral.,  Geol.,  Beilage  Bd.  A,  65,  173- 

219,  1952. 

Buerger,  M.  J.,  The  precision  determination  of 
the  linear  and  angular  lattice  constants  of 
single  crystals,  Z.  Krist,  A,  97,  433-468,  1937. 

Buerger,  M.  J.,  X-Ray  Crystallography,  John 
Wiley  &  Sons,  New  York,  1942. 


Burnham,  C.  W.,  The  structures  and  crystal- 
chemistry  of  the  aluminum-silicate  minerals, 
Ph.D.  thesis,  Massachusetts  Institute  of 
Technology,  1961. 

Burnham,  C.  W.,  Refinement  of  the  crystal 
structure  of  kyanite,  Z.  Krist.,  in  press,  1962. 

Burnham,  C.  W.,  and  M.  J.  Buerger,  Refinement 
of  the  crystal  structure  of  andalusite,  Z.  Krist., 
115,  269-290,  1961. 

Busing,  W.  R.,  and  H.  A.  Levy,  A  procedure  for 
inverting  large  symmetric  matrices,  Comm.  of 
the  Assoc,  for  Computing  Machinery,  in  press, 
1962. 

Carmichael,  C.  M.,  The  magnetic  properties  of 
ilmenite-hematite  crystals,  Proc.  Roy.  Soc. 
London,  A,  263,  508-530,  1961. 

Chao,  E.  C.  T.,  I.  Adler,  E.  J.  Dwornik,  and  J. 
Littler,  Metallic  spherules  in  tektites  from 
Isabela,  Philippine  Islands,  Science,  135,  97- 
98,  1962. 

Chayes,  F.,  On  correlation  between  variables  of 
constant  sum,  J .  Geophys.  Res.,  65,  4185-4193, 
1960. 

Chayes,  F.,  Numerical  correlation  and  petro- 
graphic  variation,  /.  Geol.,  60,  440-452,  1962. 

Chubb,  L.  J.,  Geology  of  the  Marquesas  Islands, 
Bernice  P.  Bishop  Museum,  Honolulu,  Bull., 
68,  1930. 

Clark,  S.  P.,  Jr.,  Heat  flow  from  a  differentiated 
earth,  J.  Geophys.  Res.,  66,  1231-1234,  1961. 

Cohen,  M.  U.,  Precision  lattice  constants  from 

X-ray  powder  photographs,  Rev.  Sci.  Instr.,  6, 

68-74,  1935. 
Cooper,  J.  E.,  Fatty  acids  in  recent  and  ancient 

sediments    and    petroleum    reservoir    waters, 

Nature,  193,  744-746,  1962. 

Craig,  H.,  The  geochemistry  of  the  stable  carbon 
isotopes,  Geochim.  Cosmochim.  Acta,  3,  53-92, 
1953. 

Craig,  H.,  Geochemical  implications  of  the 
isotopic  composition  of  carbon  in  ancient  rocks, 
Geochim.  Cosmochim.  Acta,  6,  186-196,  1954. 

Daly,  R.  A.,  The  geology  of  American  Samoa, 
Carnegie  Inst.  Wash.  Publ.  840,  pp.  95-142, 
1924. 

Daly,  R.  A.,  The  geology  of  Ascension  Island, 
Proc.  Am.  Acad.  Arts  Sci.,  60,  1-80,  1925. 

Daly,  R.  A.,  The  geology  of  Saint  Helena  Island, 
Proc.  Am.  Acad.  Arts  Sci.,  62,  31-92,   1927. 

Darken,  L.  S.,  and  R.  W.  Gurry,  The  system 
iron-oxygen,  I,  The  wtistite  field  and  related 
equilibria,  J.  Am.  Chem.  Soc,  67,  1398-1412, 
1945. 

De  Vries,  R.  C,  and  E.  F.  Osborn,  Phase  equi- 
libria in  high  alumina  part  of  the  system 
CaO-MgO-Al203-Si02,  J.  Am.  Ceram.  Soc,  40, 
6-15,  1957. 


204 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Edwards,  A.  B.,  Tertiary  lavas  from  the  Ker- 

guelen  Archipelago,  Rept.  B.A.N.Z.  Antarctic 

Res.  Expedition,  ser.  A,  2,  pt.  5,  72-99,  1938. 
Ervin,   G.,  and  E.  F.  Osborn,  X-ray  data  on 

synthetic    melilites,     Am.    Mineralogist,    34, 

712-722,  1949. 
Eskola,  P.,  Regeneration  of  rocks  in  the  Archean, 

Acta  Geol.  Acad.  Sci.  Hung.,  7,  109-113,  1961. 
Eugster,    H.   P.,   and   D.   R.   Wones,   Stability 

relations  of  the  ferruginous  biotite,  annite,  J. 

Petrol,  3,  82-125,  1962. 

Fenton,  C.  L.,  and  M.  A.  Fenton,  Belt  series  of 
the  North:  stratigraphy,  sedimentation,  and 
paleontology,  Bull.  Geol.  Soc.  Am.,  48,  1873- 
1970,  1937. 

Foster,  W.  R.,  High-temperature  X-ray  diffrac- 
tion study  of  the  polymorphism  of  MgSiOs, 
J.  Am.  Ceram.  Soc,  34,  255-259,  1951. 

Friedel,    G.,    Groupements   cristallins,    Thomas, 

Saint-Etienne,  1904. 
Fuster,  J.  M.,  E.  Ibarrola,  and  M.  P.  Lobato, 

Analisis  quimicos  de  rocas  espanolas  publi- 

cados    hasta    1952,    Inst.    "Lucas    Mallada" 

(Madrid),  Mon.  14,  1954. 

Fyfe,  W.  S.,  On  the  relative  stability  of  talc, 
anthophyllite,  and  enstatite,  Am.  J.  Sci.,  260, 
460-466,  1962. 

Fyfe,  W.  S.,  F.  J.  Turner,  and  J.  Verhoogen, 
Metamorphic  reactions  and  metamorphic 
facies,  Geol.  Soc.  Am.  Mem.  73,  1958. 

v.  Gehlen,  K.,  Orientierte  Umwandlung  von  Talk 
in  Protoenstatit  (7-MgSi03)  in  gebrannten 
Presskorpern  von  Speckstein,  Ber.  Deut. 
Keram.  Ges.,  39,  155-161,  1962. 

Geisler,  A.  H.,  Precipitation  from  solid  solution 
of  metals,  in  Phase  Transformations  in  Solids, 
edited  by  R.  Smoluchowski,  John  Wiley  & 
Sons,  New  York,  1951. 

Germain,  J.  E.,  Analyse  chromatographique  des 
melanges  d'acides  benzenepolycarboxyliques, 
Bull.  Soc.  Chim.  France,  115,  1959. 

Giletti,  B.  J.,  and  P.  E.  Damon,  Rubidium- 
strontium  ages  of  some  basement  rocks  from 
Arizona  and  northwestern  Mexico,  Bull.  Geol. 
Soc.  Am.,  72,  639-644,  1961. 

Glasser,  F.  P.,  and  E.  F.  Osborn,  The  ternary 
system  MgO-MnO-Si02,  J.  Am.  Ceram.  Soc, 
43,  132-140,  1960. 

Goldich,  S.  S.,  A.  O.  Nier,  H.  Badsgaard,  J.  H. 
Hoffman,  and  H.  W.  Krueger,  The  Precam- 
brian  geology  and  geochronology  of  Minnesota, 
Minn.  Geol.  Survey  Bull.,  41,  193  pp.,  1961. 

Goldsmith,  J.  R.,  and  F.  Laves,  The  microcline- 
sanidine  stability  relations,  Geochim.  Cosmo- 
chim.  Acta,  5,  1-19,  1954. 

Greenwood,  H.  J.,  The  system  NaAlSi206-H20- 
argon:  Total  pressure  and  water  pressure  in 
metamorphism,  J.  Geophys.  Res.,  66,  3923- 
3946,  1961. 


Greig,  J.  W.,  Immiscibility  in  silicate  melts,  Am. 

J.  Sci.,  13,  1-44,  133-154,  1927. 
Hamburger,   G.   E.,   and   M.   J.   Buerger,   The 

structure  of  tourmaline,  Am.  Mineralogist,  33, 

532-540,  1948. 

Hansen,  M.,  and  K.  Anderko,  Constitution  of 
Binary  Alloys,  McGraw-Hill  Book  Co.,  New 
York,  1305  pp.,  1958. 

Harker,  A.,  Natural  History  of  the  Igneous  Rocks, 
The  Macmillan  Co.,  New  York,  1909. 

Harker,  R.  I.,  and  O.  F.  Tuttle,  Experimental 
data  on  the  Pco2-T  curve  for  the  reaction 
calcite  +  quartz  ^±  wollastonite  +  carbon 
dioxide,  Am.  J.  Sci.,  254,  239-256,  1956. 

Hartman,  P.,  Proposition  pour  Interpretation 
de  la  loi  de  Bravais-Friedel  generalisee  (loi  de 
Donnay-Harker),  Compt.  Rend.,  252,  3288, 
1961. 

Hartman,  P.,  and  W.  G.  Perdok,  On  the  relation 
between  structure  and  morphology,  III,  Acta 
Cryst.,  8,  525-529,  1955. 

Hey,  J.  S.,  and  W.  H.  Taylor,  Coordination  of 
aluminum  in  the  aluminosilicates,  Z.  Krist.,  80, 
428-441,  1931. 

Heyding,  R.  D.,  and  L.  D.  Calvert,  Arsenides  of 
the  transition  metals,  II,  The  nickel  arsenides, 
Can.  J.  Chem.,  35,  1205-1215,  1957. 

Holly,  E.  D.,  and  R.  S.  Montgomery,  Structures 
of  the  water-soluble  acids  obtained  by  the 
alkali-oxygen  oxidation  of  bituminous  coal,  I, 
Chromatographic  fractionation,  Fuel,  35,  49- 
55,  1956. 

Holmes,  A.,  The  oldest  dated  minerals  of  the 
Rhodesian  Shield,  Nature,  173,  612-614,  1954. 

Hutton,  C.  O.,  Metamorphism  in  the  Lake 
Wakatipu  region,  New  Zealand  Dept.  Sci.  Ind. 
Res.,  Geol.  Mem.  5,  1940. 

Iddings,  J.  P.,  On  the  crystallization  of  igneous 
rocks,  Bull.  Phil.  Soc.  Wash.,  11,  67-202,  1892. 

Iddings,  J.  P.,  and  E.  W.  Morley,  A  contribution 
to  the  petrography  of  the  South  Sea  Islands, 
Proc  Natl.  Acad.  Sci.  U.  S.,  4,  110-117,  1918. 

James,  H.  L.,  Stratigraphy  of  Pre-Keweenawan 
rocks  in  parts  of  northern  Michigan,  U.  S. 
Geol.  Survey  Profess.  Paper  314-C,  1958. 

James,  R.  W.,  and  W.  A.  Wood,  The  crystal 
structures  of  barytes,  celestine,  and  anglesite, 
Proc.  Roy.  Soc.  London,  A,  109,  598-620,  1925. 

Jellinek,  F.,  Structure  of  molybdenum  sesqui- 
sulfide,  Nature,  192,  1065-1066,  1961. 

Johnsen,  A.,  Die  Gesteine  der  Inseln  S.  Pietro 

und    S.    Antioco,    Sardinien,    Abhandl.    Kgl. 

Preuss.  Akad.  Wiss.  Berlin,  Phys.-Math.  Kl., 

Anhang.  Abh.  II,  1-82,  1912. 
Klug,    H.    P.,    and    L.    E.    Alexander,    X-Ray 

Diffraction  Procedures,  John  Wiley  &  Sons, 

New  York,  1954. 


GEOPHYSICAL   LABORATORY 


205 


Korzhinskii,  D.  S.,  Physicochemical  Basis  of  the 

Analysis    of    the    Paragenesis    of    Minerals, 

Consultants  Bureau,  Inc.,  New  York,  1959. 
Kouvo,  Olavi,  Radioactive  age  of  some  Finnish 

Precambrian    minerals,     Bull.    Comm.     Geol. 

Finlande,  182,  70  pp.,  1958. 
Kullerud,  G.,  The  FeS-ZnS  system,  A  geological 

thermometer,  Norsk  Geol.  Tidsskr.,  32,  61-174, 

1953. 
Kullerud,  G.,  and  H.  S.  Yoder,  Pyrite  stability 

relations  in  the  Fe-S  system,  Econ.  Geol.,  54, 

533-572,  1959. 
Kullerud,  G.,  and  R.  A.  Yund,  The  Ni-S  system 

and  related  minerals,  J.  Petrol.,  3,   126-175, 

1962. 
Lacroix,  A.,  Mine'ralogie  de  Madagascar,  vol.  3, 

A.  Challamel,  Paris,  1923. 

Lowden,    J.    A.,    Age    determinations    by    the 

Geological    Survey    of    Canada,    Can.    Dept. 

Mines   Tech.  Surv.,    Geol.  Surv.  Can.,   Paper 

61-17,  127  pp.,  1961. 
Lyon,  R.  J.  P.,  Time  aspects  of  geothermometry, 

Mining  Eng.,  11,  1145-1151,  1959. 

McCabe,  C.  L.,  Sulfur  pressure  measurements  of 
molybdenum  sesquisulfide  in  equilibrium  with 
molybdenum,  J.  Metals,  8,  61-63,  1955. 

McCauley,  J.  F.,  Uranium  in  Pennsylvania, 
Penn.  Topographic  and  Geol.  Surv.  Bull.,  M43, 
671  pp.,  1961. 

Macdonald,  G.  A.,  and  H.  A.  Powers,  Contribu- 
tion to  the  petrography  of  Haleakala  volcano, 
Hawaii,  Bull.  Geol.  Soc.  Am.,  57,  115-124, 
1946. 

MacDonald,  G.  J.  F.,  Surface  heat  flow  from  a 
differentiated  earth,  J.  Geophys.  Res.,  66, 
2489-2493,  1961. 

McGregor,  A.  M.,  A  Precambrian  algal  limestone 
in  Southern  Rhodesia,  Trans.  Proc.  Geol.  Soc. 
S.  Africa,  43,  9-16,  1940. 

Mallard,  E.,  Traite"  de  cristallographie,  vol.  1, 
Dunod,  Paris,  1879. 

Mason,  B.,  Metamorphism  in  the  southern  Alps 
of  New  Zealand,  Bull.  Am.  Museum  Nat.  Hist., 
123,  211-248,  1962. 

Metcalfe,  L.  D.,  and  A.  A.  Schmitz,  The  rapid 
preparation  of  fatty  acid  esters  for  gas 
chromatographic  analysis,  Anal.  Chem.,  33, 
363-364,  1961. 

Morimoto,  N.,  and  G.  Kullerud,  Polymorphism 
in  bornite,  Am.  Mineralogist,  46,  1270-1282, 
1961. 

Nelson,  B.  W.,  and  R.  Roy,  Synthesis  of  the 
chlorites  and  their  structural  and  chemical 
constitution,  Am.  Mineralogist,  43,  707-725, 
1958. 

Nicholls,  G.  D.,  The  mineralogy  of  rock  mag- 
netism, Advances  in  Phys.,  4,  113-190,  1955. 

Nicolaysen,  L.  O.,  New  measurements  relating  to 


the  absolute  age  of  the  Transvaal  system  and 
of  the  Bushveld  igneous  complex,  Trans.  Proc. 
Geol.  Soc.  S.  Africa,  61,  137-163,  1958. 

Nier,  A.  O.,  and  E.  A.  Gulbransen,  Variations  in 
the  relative  abundance  of  the  carbon  isotopes, 
J.  Am.  Chem.  Soc,  61,  697-698,  1939. 

Osborn,  E.  F.,  The  system  CaSi03-diopside- 
anorthite,  Am.  J.  Sci.,  240,  751-788,  1942. 

Pleass,  C.  M.,  and  R.  D.  Heyding,  Arsenides  of 
the  transition  metals,  VI,  Electrical  and 
magnetic  properties  of  the  triarsenides,  Can. 
J.  Chem.,  40,  590-600,  1962. 

Powers,  H.  A.,  Composition  and  origin  of 
basaltic  magma  of  the  Hawaiian  Islands, 
Geochim.  Cosmochim.  Acta,  7,  77-107,  1955. 

Provoust,  J.,  Remarques  sur  la  composition  et  le 
pouvoir  reflecteur  de  la  bornite  (Cu5FeS4), 
Compt.  Rend.  Congr.  Soc.  Savantes  Paris  De'pt., 
Sect.  Sci.,  Dijon,  1959,  365-372,  1960. 

Ramdohr,  P.,  Ulvospinel  and  its  significance  in 
titaniferous  iron  ores,  Econ.  Geol.,  48,  677-688, 
1953. 

Ramdohr,  P.,  Die  Erzmineralien  und  ihre  Ver- 
wachsungen,  Akademie-Verlag,  Berlin,  1955. 

Ramdohr,  P.,  and  M.  Schmitt,  Oregonit,  ein 
neues  Nickel-Eisenarsenid  mit  metallartigen 
Eigenschaften,  Neues  Jahrb.  Mineral.,  Mo~ 
natsh.,  239-247,  1959. 

Rankama,  K.,  New  evidence  on  the  origin  of 
Precambrian  carbon,  Bull.  Geol.  Soc.  Am.,  59, 
389-416,  1948. 

Ringwood,  A.  E.,  and  M.  Seabrook,  High- 
pressure  transition  of  MgGeOs  from  pyroxene 
to  corundum  structure,  /.  Geophys.  Res.,  67, 
1690-1691,  1962. 

Roseboom,  E.  H.,  Jr.,  Phase  relationships  in  the 
arsenic-rich  portion  of  the  system  Co-Ni-Fe- 
As,  Ph.D.  thesis,  Harvard  University,  65  pp., 
1958. 

Roseboom,  E.  H.,  Jr.,  Skutterudites  (Co,Ni,Fe)- 
Asz-X:  Composition  and  cell  dimensions,  Am. 
Mineralogist,  47,  310-327,  1962. 

Ross,  C.  P.,  The  geology  of  Glacier  Park  and 
vicinity  in  Montana,  Science,  119,  658-659, 
1954. 

Roy,  D.  M.,  and  R.  Roy,  Synthesis  and  stability 
of  minerals  in  the  system  MgO-Al203-Si02- 
H20,  Am.  Mineralogist,  40,  147-178,  1955. 

Sakata,  Y.,  Unit-cell  dimensions  of  synthetic 
aluminian  diopsides,  Japan  J.  Geol.  Geography, 
Trans.,  28,  161-168,  1957. 

Schairer,  J.  F.,  and  N.  L.  Bo  wen,  The  system 
Na20-Al203-Si02,  Am.  J.  Sci.,  254,  129-195, 
1956. 

Schairer,  J.  F.,  and  H.  S.  Yoder,  Jr.,  The  nature 
of  residual  liquids  from  crystallization,  with 
data  on  the  system  nepheline-diopside-silica, 
Am.  J.  Sci.,  258 A,  273-283,  1960. 


206 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Schmahl,  N.  G.,  B.  Frisoh,  and  E.  Hargartner, 
Zur  Kenntnis  der  Phasenverhaltnisse  im 
System  Fe-Ti-0  bei  1000°C,  Z.  anorg.  u.  allgem. 
Chem.,  305,  40-54,  1960. 

Seager,  A.  F.,  The  morphology  of  the  baryte 
group,  Mineral.  Mag.,  32,  63-86,  1959. 

Semiletov,  S.  A.,  The  crystalline  structure  of 
rhombohedral  MoS2,  Soviet  Phys.-Cryst.,  6, 
428-431,  1962. 

Silverman,  S.  R.,  and  S.  Epstein,  Carbon  isotopic 
compositions  of  petroleum  and  other  sedi- 
mentary organic  materials,  Bull.  Am.  Assoc. 
Petrol.  Geologists,  42,  998-1012,  1958. 

Skinner,  B.  J.,  S.  P.  Clark,  Jr.,  and  D.  E. 
Appleman,  Molar  volumes  and  thermal  expan- 
sions of  andalusite,  kyanite,  and  sillimanite, 
Am.  J.  Sci.,  259,  651-668,  1961. 

Smith,  J.  V.,  and  S.  W.  Bailey,  Second  review  of 
Al-0  and  Si-0  tetrahedral  distances,  Acta 
Cryst.,  15,  in  press,  1962. 

Sorokin,  C,  and  R.  W.  Krauss,  The  effects  of 
light  intensity  on  the  growth  rates  of  green 
algae,  Plant  Physiol.,  33,  109-113,  1958. 

Stubbles,  J.  R.,  and  F.  D.  Richardson,  Equi- 
librium in  the  system  molybdenum  +  sulfur 
+  hydrogen,  Trans.  Faraday  Soc,  56,  1460- 
1466,  1960. 

Swanson,  H.  E.,  N.  T.  Gilfrich,  and  G.  M. 
Ugrinic,  Standard  X-ray  diffraction  powder 
patterns,  Natl.  Bur.  Standards  U.  S.  Circ.  539, 
v.  5,  1955. 

Takeuchi,  T.,  and  M.  Nambu,  On  the  genesis  of 
chalcopyrite  lattice  in  bornite,  Sci.  Rept. 
Tohoku  Univ.,  A,  8,  31-44,  1956. 

Taylor,  W.  H.,  The  structure  of  sillimanite  and 
mullite,  Z.  Krist.,  68,  503-521,  1928. 

Thompson,  J.  E.,  Geology  of  the  Sudbury  Basin, 
Ann.  Rept.  Ontario  Dept.  Mines,  65,  1-56, 
1956. 

Tilley,  C.  E.,  Metamorphic  zones  in  the  southern 
Highlands  of  Scotland,  Quart.  J.  Geol.  Soc. 
London,  81,  100-112,  1925. 

Tilley,  C.  E.,  Some  aspects  of  magmatic  evolu- 
tion, Quart.  J.  Geol.  Soc.  London,  106,  37-61, 

1950. 
Tilley,  C.  E.,  Problems  of  alkali  rock  genesis, 

Quart.  J.  Geol.  Soc.  London,  113,  323-360,  1958. 
Tilley,  C.  E.,  The  occurrence  of  hypersthene  in 

Hawaiian  basalts,   Geol.  Mag.,  98,   257-260, 

1961. 
Tilley,  C.  E.,  and  H.  F.  Harwood,  The  dolerite- 

chalk   contact   of   Scawt   Hill,    Co.    Antrim, 

Mineral.  Mag.,  22,  439-468,  1931. 
Tilton,  G.  R.,  Volume  diffusion  as  a  mechanism 

for  discordant  lead  ages,  J.  Geophys.  Res.,  65, 

2933-2945,  1960. 
Tilton,  G.  R.,  and  G.  L.  Davis,  Geochronology, 

in  Researches  in  Geochemistry,  edited  by  P.  H. 


Abelson,    John    Wiley   &   Sons,    New   York, 
pp.  190-216,  1959. 
Turner,  F.  J.,  and  J.  Verhoogen,  Igneous  and 
Metamorphic   Petrology,    McGraw-Hill    Book 
Co.,  New  York,  1951. 

Tuttle,  O.  F.,  Optical  studies  on  alkali  feldspars, 
Am.  J.  Sci.,  Bowen  Vol.,  553-567,  1952. 

Tuttle,  O.  F.,  and  N.  L.  Bowen,  Origin  of  granite 
in  the  light  of  experimental  studies  in  the 
system  NaAlSi308-KAlSi308-Si02-H20,  Geol. 
Soc.  Am.  Mem.  74,  153  pp.,  1958. 

Tyler,  S.  A.,  and  E.  S.  Barghoorn,  Occurrence  of 
structurally  preserved  plants  in  Precambrian 
rocks  of  the  Canadian  Shield,  Science,  119, 
606-608,  1954. 

Tyler,  S.  A.,  E.  S.  Barghoorn,  and  L.  P.  Barrett, 
Anthracitic  coal  from  Precambrian  upper 
Huronian  black  shale  of  the  Iron  River 
district,  northern  Michigan,  Bull.  Geol.  Soc. 
Am.,  68,  1293-1304,  1957. 

Ussing,  N.  V.,  Untersuchungen  der  Mineralien 
von  Fiskernas  in  Gronland,  Z.  Krist.,  15, 
596-615,  1889. 

Verhoogen,  J.,  Oxidation  of  iron-titanium  oxides 
in  igneous  rocks,  J.  Geol,  70,  168-181,  1962. 

Vincent,  E.  A.,  J.  B.  Wright,  R.  Chevallier,  and 

5.  Mathieu,  Heating  experiments  on  some 
natural  titaniferous  magnetites,  Mineral.  Mag., 
31,  624-655,  1957. 

Walker,  F.,  and  L.  O.  Nicolaysen,  The  petrology 
of  Mauritius,  Colonial  Geol.  Mineral  Resources 
(Gt.  Brit.),  4,  3-43,  1954. 

Wandke,  A.,  Molecular  migration  and  mineral 
transformation,  Econ.  Geol.,  21,  161-171,  1926. 

Washington,  H.  S.,  The  volcanoes  and  rocks  of 

Pantelleria,  J.  Geol.,  21,  653-713;  22,  16-27, 

1914. 
Washington,  H.  S.,  Chemical  analyses  of  igneous 

rocks,   U.  S.  Geol.  Surv.  Profess.  Paper,  99, 

1917. 
Washington,  H.  S.,  Petrology  of  the  Hawaiian 

Islands,  parts  I  and  II,  Am.  J.  Sci.,  5,  465-502; 

6,  100-126,  1923. 

Webster,  A.  H.,  and  N.  F.  H.  Bright,  The  system 
iron-titanium-oxygen  at  1200°C  and  oxygen 
partial  pressures  between  1  atm  and  2  X  10-14 
atm,  J.  Am.  Ceram.  Soc,  44>  110-116,  1961. 

Wetherill,  G.  W.,  O.  Kouvo,  G.  R.  Tilton,  and 
P.  W.  Gast,  Age  measurements  on  rocks  from 
the  Finnish  Precambrian,  /.  Geol.,  70,  74-88, 
1962. 

Whittaker,  Sir  Edmund,  and  G.  Robinson,  The 
Calculus  of  Observations,  Blackie  and  Son, 
London,  4th  ed.,  pp.  209-259,  1944. 

Wickman,  F.  E.,  On  a  new  possibility  of  calcu- 
lating the  total  amount  of  coal  and  bitumen, 
Geol.  Foren.  Stockholm  Forh.,  63,  419-421, 
1941. 


GEOPHYSICAL   LABORATORY 


207 


Yagi,  K.,  Petrochemical  studies  on  the  alkalic 

rocks  of  the  Morotu  District,  Sakhalin,  Bull. 

Geol.  Soc.  Am.,  64,  769-810,  1953. 
Yoder,    H.   S.,   Jr.,   The   MgO-Al203-Si02-H20 

system  and  related  metamorphic  facies,  Am. 

J.  Set.,  Bowen  Vol.,  569-627,  1952. 

Yoder,  H.  S.,  Jr.,  and  H.  P.  Eugster,  Synthetic 
and  natural  muscovites,  Geochim.  Cosmochim. 
Acta,  8,  225-280,  1955. 


Young,  R.  B.,  and  E.  Mendelsohn,  Domed  alga 
growths  in  the  Dolomite  series  of  South  Africa 
Trans.  Proc.  Geol.  Soc.  S.  Africa,  51,  53-62 
1948. 

Zen,  E.,  The  zeolite  facies:  An  interpretation, 
Am.  J.  Set.,  259,  401-409,  1961. 

Zies,  E.  G.,  Chemical  analyses  of  two  pantel- 
lerites,  J.  Petrol.,  1,  304-308,  1960. 


208 


CARNEGIE     INSTITUTION      OF      WASHINGTON 

PERSONNEL 


Scientific  Staff 

Director:  P.  H.  Abelson. 

Retired  Associate:  E.  G.  Zies,  Chemist. 

Staff  Associates:  G.  J.  F.  MacDonald,  W.  F. 
Schreyer.1 

Physical  Chemists:  F.  R.  Boyd,  H.  J.  Green- 
wood, T.  C.  Hoering,  J.  F.  Schairer,  G.  R. 
Tilton. 

Penologists:  F.  Chayes,  H.  S.  Yoder,  Jr. 

Geochemists:  G.  L.  Davis,  G.  Kullerud. 

Organic  Geochemist:  P.  L.  Parker.2 

Geophysicist:  S.  P.  Clark,  Jr.3 

Physicist:  J.  L.  England. 

Crystallographer:  G.  Donnay. 

Guest  Investigators:  H.  Baadsgaard,  University 
of  Alberta;  B.  R.  Doe,  U.  S.  Geological 
Survey;  J.  D.  H.  Donnay,  Johns  Hopkins 
University;  H.  Faul,  U.  S.  Geological 
Survey;  B.  M.  French,  Johns  Hopkins 
University;  K.  v.  Gehlen,  University  of 
Erlangen-Ntirnberg;  P.  Ramdohr,  Univer- 
sity of  Heidelberg;  B.  E.  Sabels,  Desert 
Research  Institute  of  the  University  of 
Nevada;  C.  E.  Tilley,  Cambridge  Uni- 
versity. 

Fellows:  D.  K.  Bailey,  Trinity  College, 
Dublin;1  P.  R.  Brett,  Harvard  University; 
C.  W.  Burnham,  Massachusetts  Institute  of 
Technology;2  P.  R.  Buseck,  Columbia 
University;2  J.  de  Neufville,  Yale  Univer- 
sity; B.  R.  Doe,  California  Institute  of 
Technology;4  J.  J.  Fawcett,  University  of 
Manchester;2  K.  Hytonen,  University  of 
Helsinki;5  D.  H.  Lindsley,  Johns  Hopkins 
University;6  G.  Moh,  University  of  Heidel- 
berg;1 N.  Morimoto,  Tokyo  University;1 
B.  E.  Sabels,  Desert  Research  Institute  of 
the  University  of  Nevada;7  W.  F.  Schreyer, 
University  of  Munich;8  Y.  Suzuki,  Hok- 
kaido University;  A.  C.  Turnock,  Johns 
Hopkins  University;9  R.  A.  Yund,  Univer- 
sity of  Illinois.10 

Operating  and  Maintenance  Staff 

Executive  Officer:  A.  D.  Singer. 

Accountant:  E.  T.  Orozco. 

Editor  and  Librarian:  Miss  D.  M.  Thomas. 


Stenographer:  Miss  M.  E.  Imlay. 

Typist:  Mrs.  N.  0.  Doe.11 

Stockroom  Assistant:  M.  L.  Kirby. 

Mechanic's  Helper:  C.  Baylor.12 

Janitors:  M.  Ferguson,  E.  L.  Jackson,13  C.  E. 
Wilkerson.14 

Chief  Mechanician:  F.  A.  Rowe. 

Instrument  Makers:  C.  A.  Batten,15  A.  M. 
Fyfe,16  L.  C.  Garver,17  J.  F.  Kocmaneck, 
W.  H.  Lyons,18  0.  R.  McClunin,  G.  E. 
Speicher,  W.  H.  Suerth.19 

Mechanic  and  Carpenter:  E.  J.  Shipley. 

Electrician:  E.  C.  Huffaker. 

Machinist:  J.  R.  Thomas. 

Building  Engineer:  R.  L.  Butler. 


Appointment  from  January  1,  1962. 
2 Appointment  from  September  1,  1961. 
3Resigned  June  30,   1962,  to  accept  post  of 
Sidney  J.  Weinberg  Professor  of  Geophysics  at 
Yale  University. 

4Appointment  terminated  July  31,  1961,  to 
accept  position  with  U.  S.  Geological  Survey. 

6 Appointment  terminated  July  31,  1961,  to 
accept  position  with  Geological  Survey  of  Fin- 
land. 

6 Appointment  terminated  June  30,  1962,  to 
accept  position  on  staff  of  Geophysical  Labora- 
tory. 

7Appointment  from  May  1,  1962. 
8Appointment  terminated  December  31,  1961, 
to  accept  position  at  University  of  Kiel. 

9Appointment  terminated  June  30,  1962,  to 
accept  position  with  Department  of  Mines  and 
Technical  Surveys,  Ottawa. 

"Appointment  terminated  August  31,  1961,  to 
accept  position  at  Brown  University. 
"Appointment  from  November  1,  1961. 
12Retired  June  30,  1962. 
"Appointment  from  May  28,  1962. 
14 Appointment  from  April  2  through  May  31, 
1962. 

"Appointment  from  June  1,  1962. 
"Appointment    from    February    20    through 
April  30,  1962. 

"Appointment  from  June  8,  1962. 

"Resigned  June  15,  1962. 

"Appointment  terminated  February  28,  1962. 


Plate  1 


Geophysical  Laboratory 


Fig.  34.  Ilmenite-' 'magnetite"  intergrowth  made  by  holding  a  charge  of  pure  ulvospinel 
at  1000°C  and  the  /02  of  the  bomb  walls  for  3  hours.  The  darker  gray  "magnetite''  host 
is  about  MtsoUspsoJ  the  light  gray  ilmenitess  lamellae  lie  in  the  (111)  planes  of  the  host.  X2200. 
Photograph  by  Professor  Paul  Ramdohr. 


Plate  2 


Geophysical  Laboratory 


cp 


Fig.  53.  Chalcopyrite  (cp)  exsolved  from  bornite  (bn).  Exsolved 
chalcopyrite  is  in  the  form  of  a  vein  and  as  tiny  lamellae.  Specimen 
annealed  at  300°C  for  43^  months.    X  800   (oil). 


Fig.  54.  Chalcopyrite  (cp)  exsolved  from  bornite  (bn).  Note 
pseudoreplacement  texture.  Specimen  cooled  from  700°  to  50°C 
in  22  hours.   X  1600  (oil).   (Retouched.) 


Department 
of  Terrestrial  Magnetism 


Washington,  District  of  Columbia 
Merle  A.  Tuve, 
Director 


Contents 


Introduction 211 

Experimental  Geophysics 214 

Radio  astronomy 214 

Radio  hydrogen  observations 214 

South  American  cooperation 219 

Continuum  observations 219 

The  earth's  crust 221 

Seismic  studies 221 

Radioactive  ages  of  rocks 234 

Theoretical  and  Statistical  Geophysics 239 

Morphology  and  temporal  variations  of  the  intensity  of  charged  particles  in  the 

Van  Allen  trapped-radiation  belt 239 

Geomagnetic  equatorial  ring  current  measures  and  latitude  of  auroral  currents       .  240 

Conductivity  anomaly  program  for  Peru 241 

Cosmic-ray  program 241 

Laboratory  Physics 242 

Nuclear  physics 242 

Polarized  ion  source 242 

Summary  of  cooperative  program  with  Florida  State  University 243 

Biophysics 244 

Introduction 244 

Incorporation  of  RNA  bases  into  the  metabolic  pool  and  RNA  of  E.  coli  .      .      .  244 

RNA  composition 253 

Kinetic  studies  of  RNA  composition 260 

Purification  of  D-RNA 276 

Kinetics  of  labeling  of  turnip  yellow  mosaic  virus 276 

Control  mechanisms 279 

Cell-free  synthesis 282 

Doublet  code 282 

Cooperation 285 

Conclusions 285 

Image  Tubes  for  Large  Telescopes 288 

Bibliography 288 

Major  publications 290 

Personnel 291 


Carnegie  Institution  of  Washington  Year  Book  61,  1961-1962 


Frontispiece 


Department  of  Terrestrial  Magnetism 


Purification  of  "messenger"  nucleic  acid.  Single-stranded  DNA,  the  genetic  material  of  the  cell,  is 
immobilized  in  a  bed  of  agar.  RNA  is  passed  through  the  bed,  where  the  RNA  molecules  ("mes- 
sengers") containing  nucleotide  sequences  complementary  to  those  in  the  DNA  are  specifically 
hybridized  by  means  of  hydrogen  bonding.  The  hydrogen  bonds  are  then  "melted,"  and  the  purified 
messenger  is  collected  for  further  study. 


INTRODUCTION 

Two  beautiful  examples  of  laboratory  end  result  of  standard  separation  pro- 
work  at  the  Department,  each  one  cedures,  has  meant  that  the  "messenger 
exhibiting  in  conspicuous  simplicity  the  RNA"  which  carries  to  the  nascent 
patterns  governed  by  the  laws  of  nature,  protein  the  DNA  information  or  code  (by 
moved,  during  the  year,  from  the  idea  being  a  complementary  sequence  of 
stage  into  the  factual  stage  of  accom-  nucleotides)  has  not  been  available  for 
plishment.  study  as  a  separate  item.   " Messenger 

One  example  was  the  use  of  a  spin-  RNA"  to  date  has  chiefly  been  useful  as 

polarized    beam    of    deuterons    in    our  a  concept.  Toward  the  close  of  the  report 

4-Mev    Van    de    Graaff    generator    for  year  a  new  and  simple  technique  was 

studies  of  light-element  nuclear  reactions,  developed    here    for   the    adsorption    of 

Appropriately,  in  view  of  the  Swiss  em-  highly    specific    fractions    of    RNA    on 

phasis,  and  our  own,  on  nuclear  physics  single-stranded    DNA    which    has    long 

as  a  scientific  discipline  (not  technology),  sequences  of  biological  coding  identical 

this  was  a  joint  activity  of  our  Depart-  (complementary)   to  the  coding  of  this 

ment  with  Professor  P.  Huber  of  Basel,  specific  RNA.  The  DNA  is  entrapped  in 

Switzerland,    and    his    colleagues.    Two  agar  gel  in  the  single-strand  form,  and 

years  ago,  his  group  there  developed  the  the  total  RNA  mixture  is  slowly  filtered 

first  successful  spin-polarized  hydrogen-  through  the  gel.  The  specifically  adsorbed 

ion  source,  and  our  high-voltage  gener-  RNA  is  a  small  fraction  of  the  total  RNA, 

ator,  now  25  years  old,  is  the  only  one  perhaps    1   or   2   per   cent,   and   it   can 

in  the  world  with  the  obsolescent  charac-  subsequently  be  released  from  the  DNA 

teristic  of  large  bulk,  so  that  the  19-foot  by   suitable   elution,   thus   yielding   the 

ball,  which  is  the  high-voltage  electrode,  coded  "messenger"  RNA  as  a  separated 

is  able  to  accommodate  the  large  6-kilo-  constituent.  Our  experiments  have  dem- 

watt  analyzer  that  provides  the  polarized  onstrated    that    DNA   and    RNA   from 

beam.  This  spin-polarized  beam  provides  closely  related  bacteria  show  appropri- 

a  direct  and  simple  approach  to  many  ately  related  degrees  of  specific  adsorp- 

familiar  questions  of  spin  and  parity  in  tion,  while  unrelated  bacteria  exhibit  no 

nuclear  reactions.  such  adsorption  between  the  DNA  of  one 

The  second  example  evolved  abruptly  and  the  RNA  of  the  other. 

in  the  course  of  studies  of  nucleic  acids  Another  research  result  of  considerable 

by    our    biophysics    group,    who    were  consequence  which  was   confirmed   and 

examining  the  early  products  of  biosyn-  convincingly  illustrated  during  the  year 

thesis   and    the    handing-on    of   genetic  relates  to  the  study  of  the  earth's  crust 

information  by  "biological  coding"  from  by    sound    waves    from    explosions.    At 

the  DNA  to  RNA  and  protein.  The  idea  increasing  distances  from  an  explosion  the 

has  been  familiar  for  several  years  that  velocity  of  the  waves  which  first  arrive 

the  double-stranded  helix  of  DNA  (as  in  appears  to  increase,  and  this  effect  has 

chromosomes,  etc.)  must  somehow  sepa-  been  widely  interpreted  as  showing  the 

rate  into  single  strands  and  then  transmit  existence  of  various  "layers"  of  rock  with 

its  genetic  information  (e.g.,  in  the  form  intermediate    velocities    interposed    be- 

of  sequences  of  the  nucleotides  of  adenine,  tween  the  low- velocity  rocks   near  the 

guanine,  cytosine,  and  thymine)  to  RNA  surface  and  the  upper  boundary  of  the 

and  thence  to  proteins,  with  their  many  earth's  mantle  at  a  depth  of  40  or  50 

highly     specific     enzymatic     properties,  kilometers  under  most  land  areas.   De- 

Nevertheless,  the  complex  mixture  of  old  tailed  study  of  the  results  of  an  intensive 

and  new  RNA  molecules,  which  is  the  program  of  sea  explosions  in  the  Gulf  of 

211 


212  CARNEGIE     INSTITUTION      OF      WASHINGTON 

Maine   during    1961,    arranged    by   our  dating  of  ancient  rocks.  A  program  of 

Department  with  the  Navy,  the  Coast  studies   of   deep   electrical   conductivity 

Guard,  and  a  group  of  scientific  collab-  (80  km)  by  magnetic  variations  in  Peru 

orators    (especially    the    University    of  and    Bolivia   is    also    now   becoming   a 

Wisconsin),  led  to  the  emphatic  recog-  reality. 

nition,  again,  of  a  mathematical  result  In  a  quite  different  area  and  a  highly 
known  for  several  decades.  Changes  in  specialized  technical  field,  the  Depart- 
the  rate  of  increase  of  velocity  with  depth  ment  is  guiding  another  cooperative 
give  rise  to  " cusps"  (multiple-valued  program,  largely  based  to  date  on  our 
regions)  on  the  travel- time  curve,  and  close  contacts  with  the  electronic  tube 
these  cusps  are  both  frequent  and  super-  industry,  which  endeavors  to  put  into 
posed  on  one  another  when  the  travel-  the  service  of  optical  astronomy  the  best 
time  curve  is  plotted  for  any  ordinary  efforts  of  electronic  technicians  relating 
sample  of  the  real  earth.  Not  only  this,  to  the  intensification  of  faint  optical 
but,  as  a  result  of  it,  when  efforts  are  images.  This  image  tube  work  has  now 
made  to  determine  the  velocity  structure  reached  a  stage  where  a  real  gain  in 
of  regions  in  the  crust  below  16  or  18  research  use  ("figure  of  merit")  of  6  to  15 
kilometers'  depth,  the  observed  travel-  over  the  best  photographic  plates  is  just 
time  curves  can  be  reproduced  by  any  one  beginning  to  be  realized  by  means  of 
of  a  considerable  variety  of  assumed  convenient  (and  permanent)  sealed-off 
structures  of  velocity  as  a  function  of  image  tubes.  "Cooperation"  thus  has 
depth,  owing  to  the  importance  of  the  many  very  different  aspects  in  the  current 
rates  of  change  of  velocity,  superposed  on  work  of  the  Department,  but  in  each  case 
the  actual  values  of  velocity  at  stated  it  represents  a  situation  where  there  is 
depths.  This  unequivocal  demonstration,  special  value  in  our  freedom  of  initiative 
again,  of  the  very  real  limitations  imposed  and  a  recognition  of  the  infectious 
on  our  ability  to  learn  about  crustal  characteristic  of  personal  enthusiasm, 
"layers"  by  the  customary  methods  of  The  image  tube  work  is  now  primarily 
explosion  seismology  came  at  a  time  of  supported  by  the  National  Science  Foun- 
special  pertinence,  in  view  of  the  greatly  dation,  and  we  find  that  our  relationships 
increased  activities  brought  about  by  the  with  the  NSF  form  a  conspicuously 
Geneva  proposals  for  detection  of  bomb  satisfying  aspect  of  our  entire  Depart- 
explosions  in  relation  to  arms  control.  ment's  cooperative  ventures,  including 
The  cooperative  expedition  to  Maine  the  current  emphasis  on  Latin  American 
has  been  only  one  aspect  of  a  general  projects  involving  radio  astronomy 
procedure  of  intensive  cooperation  with  studies  of  the  southern  sky  and  geo- 
selected  individual  colleagues  which  is  physical  studies  of  the  Andes.  NSF  funds 
followed  in  most  of  our  work.  Other  are  now  applied  to  defray  the  Latin 
examples  this  year  relate  to  studies  of  American  expenditures  incurred  in  con- 
local  earthquakes  in  the  Andes  jointly  nection  with  our  collaboration  and  also 
with  three  groups  (now  four)  in  different  to  provide  some  modest  stipends  for  our 
western  Latin  American  countries,  and  to  South  American  colleagues,  especially 
initial  preparations  here,  jointly  with  students.  The  various  items  are  discussed 
other  colleagues  from  eastern  South  under  the  individual  sections. 
America,  for  radio  astronomy  studies  of  Except  for  human  problems,  rooted  in 
the  southern  sky,  with  cooperative  instru-  conflicting  idea-structures  and  in  man's 
mentation  we  are  constructing  for  use  in  traditional  will  to  dominate  over  other 
Argentina  and  Brazil.  Similar  emphasis  men — problems  of  which  the  research 
on  colleague-to-colleague  arrangements  man  is  painfully  and  rather  helplessly 
have  been  effective  in  our  biophysics  aware — the  world  we  observe  around  us 
program  and  in  our  work  on  the  isotope  is  an  intensely  beautiful  and  interesting 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM  213 

place.  It  is  true  that  powerful  men  in  biophysicist,  but  still  in  his  traditional 
different  nations,  both  as  individuals  and  role  as  a  "natural  philosopher"  who 
in  groups,  may  express  unalterable  con-  studies  dynamic  and  experimental  rela- 
victions  for  a  time,  and  oppose  each  other  tionships  (as  contrasted  with  "natural 
even  with  violence  concerning  various  history"  and  classification),  has  found 
relationships  in  the  society  of  human  here  also  a  deeply  satisfying  exhibit  of 
beings.  But  these  social  differences  have  the  laws  of  nature.  An  additional  element 
a  way  of  melting  into  one  another,  and  is  undoubtedly  and  ineffably  present 
even  reversing,  in  the  course  of  roughly  when  he  moves  toward  studies  of  such 
one  lifetime,  so  that  the  visible  patterns  factors  as  memory,  idea,  or  self-will  in  a 
of  human  group  relationships,  to  the  eye  living  organism.  There  are  clear  and 
of  the  physicist,  seem  largely  transient,  obvious  limits  to  the  fields  of  inquiry 
mostly  lacking  in  beauty,  and  not  visibly  amenable  to  mathematics  and  experi- 
governed  by  perceptible  regularities  or  ment,  and  each  individual  human  aware- 
natural  laws.  Contrast  this  immediate  ness  is  a  demonstration  of  these  limits, 
and  unavoidable  confusion  of  his  daily  The  course  of  history  and  the  usual 
environment  with  the  beautiful  regularity  current  distress  of  political  adjustments 
and  systematic  relatedness  he  observes  in  everywhere  may  be  another  demonstra- 
every  aspect  of  the  natural  phenomena  tion  of  the  same  limits,  but  the  great  good 
he  studies,  from  distant  stars  to  living  fortune  of  being  enabled  to  devote  our 
bacteria,  and  you  can  sense  his  deep  energies  and  talents,  modest  as  they  may 
satisfaction  in  scientific  studies.  Every  be,  to  further  illuminating  the  intricate 
question  he  frames  and  every  reaction  he  and  orderly  patterns  of  the  physical 
observes  reveals  in  striking  measure  the  world  around  us,  including  the  material 
immanence  of  natural  law,  a  universally  interactions  and  patterns  in  living  cells 
patterned  relatedness  of  the  kind  that  and  creatures,  is  the  "princely  gift"  of 
men  have  always  recognized  as  tran-  our  time  and  circumstance, 
scendent.  Yet  these  relationships  are  This  search  for  understanding  and  for 
more  and  more  seen  and  understood,  bit  heightened  awareness  is  an  ancient  use  of 
by  bit  adding  to  man's  stature  and  his  leisure  in  a  society,  usually  limited  to  a 
awe  of  the  stupendous  and  beautifully  perceptive  few,  and  we  can  all  rejoice  that 
intricate  universe  in  which  he  finds  it  is  honored  so  widely  in  our  own  time, 
himself.  This  is  the  continuing  miracle  of  The  Carnegie  Institution  and  even  this 
human  awareness,  as  it  is  observed  and  Department  have  indeed  witnessed  a 
experienced  in  the  natural  sciences.  Based  most  remarkable  half  century  of  expan- 
on  the  demonstrations  of  repeatable  sion  in  the  history  of  the  human  search 
experiments  and  on  the  definitions  and  for  knowledge  and  awareness.  Recog- 
formal  logic  of  mathematics  the  satisfying  nition  of  a  problem  is  at  least  a  necessary 
activities  of  the  research  scientist,  espe-  step  toward  its  active  solution,  and  so 
cially  in  the  "exact"  sciences,  comprise  perhaps  we  may  hope  that  some  of  the 
and  exhibit  a  kind  of  dedicated  vocation  patterns  of  sustained  and  critical  effort 
long  familiar  in  other  fields  of  activity.  which  are  the  tradition  in  the  natural 
The  stars  are  as  remote  as  they  ever  sciences  may  ultimately  prove  of  value  in 
were,  yet  we  now  are  fully  confident  of  resolving  some  of  the  conflicts  and 
our  identification  in  them  of  the  familiar  irregularities  in  social  relationships.  Even 
atoms  that  comprise  the  earth  and  even  if  these  human  problems  continue  to  defy 
our  own  bodies.  The  biologists,  as  solution,  today's  towering  structure  of 
geneticists,  have  exposed  to  view  during  scientific  knowledge,  as  a  human  expres- 
the  last  few  decades  the  beautifully  sion  of  the  laws  of  nature,  will  long  stand 
regular  and  intricate  mechanisms  of  as  a  monument  to  the  efforts  and  satis- 
inheritance.    The    physicist,    now    as    a  factions  of  our  own  epoch. 


214 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


EXPERIMENTAL   GEOPHYSICS 


RADIO  ASTRONOMY 
B.  F.  Burke  and  M.  A.  Tuve 

Radio  Hydrogen  Observations 

The  galactic  center.  During  the  past 
year  much  of  the  observing  time  on  the 
60-foot  dish  and  multichannel  21-cm 
receiver  has  been  spent  on  a  program  to 
study  the  motions  of  the  interstellar 
hydrogen  line  in  the  inner  parts  of  the 
galaxy.  Previous  observations  at  Leiden 
and  Sydney  have  given  ample  evidence 
that  the  motions  close  to  the  galactic 
center  are  complex,  with  clear  indication 
that  the  hydrogen  gas  is  not  only  rotating 
about  the  center  of  mass  of  the  galaxy 
but  is  also  expanding.  Since  our  observing 
station  at  Derwood,  Maryland,  is  at  a 
more  southern  latitude  than  Dwingeloo 
(the  Dutch  station)  we  were  able  not  only 
to  confirm  the  observations  of  the  Dutch 
but  also  to  extend  the  observations 
nearly  20°  farther  south  along  the  galactic 
plane.  The  Sydney  observations  were 
made  with  a  small  dish,  using  a  receiver 
of  relatively  wide  bandwidth,  which  also 
showed  the  existence  of  the  expansion  but 
with  little  detail  on  its  structure. 

Initially,  the  program  has  been  to 
obtain  hydrogen-line  profiles  along  the 
galactic  plane  at  intervals  of  1°  in  longi- 
tude from  I1  =  301°  to  I1  =  355°,  and  to 
take  cross  sections  every  2°  in  longitude 
running  from  2°  above  the  plane  to  2° 
below  the  plane  in  0.5°  increments  in 
latitude.  As  the  new  conversion  tables  for 
(I11,  b11)  were  not  available,  all  measure- 
ments were  made  in  the  (I1,  b1)  system, 
using  latitude  b1  =  — 1.5°  as  the  approx- 
imate galactic  equator  on  which  the  grid 
of  observation  points  was  centered. 

Within  about  4°  of  the  galactic  center, 
the  Dutch  observed  high-velocity  wings 
on  the  hydrogen-emission  profiles,  ex- 
tending nearly  300  km/sec  both  to  the 
red  and  to  the  blue.  These  they  have 
interpreted,  in  part,  as  a  double  structure: 


an  inner  disk  of  hydrogen,  about  300 
parsecs  in  radius,  rotating  rapidly  (but 
not  expanding);  and  an  outer  ring, 
extending  from  R  =  500  to  590  pc,  like- 
wise in  rapid  rotation. 

We  have  confirmed  the  existence  of 
such  high- velocity  wings,  which  show 
quite  clearly  on  scans  taken  near  the 
center.  Scans  taken  at  latitudes  a  few 
degrees  above  and  below  the  plane  are 
used  to  give  a  "cold  sky"  reference,  to 
check  on  the  zero  base  line  of  the  receiver. 
Our  receiver  can  only  cover  a  band  about 
200  km/sec  wide,  and  hence  it  is  necessary 
to  take  several  scans,  centered  on 
different  frequencies,  to  obtain  a  complete 


^x  =  324 
bz  =  -l.5 


A 


H, 


.£I  =  322 
bx  =  -l.5 


j*?      V^f 


H 


j*1  =  320 
bT  =  -1.5 


H. 


O        40       80      120      160     200 

Scale  (km  /sec) 


Fig.  1.  Sample  records  taken  along  the  plane 
of  the  Milky  Way  in  the  vicinity  of  the  galactic 
center.  Each  curve  is  a  superposition  of  four  to 
six  individual  tracings  from  the  54-channel 
spectrograph. 


DEPARTMENT   OF   TERRESTRIAL   MAGNETISM 


215 


^  J  =316 

b1  =  -1.0 


fX 


/W^ 


v\ 


**v.». 


H, 


^  X  =  3I4 
bx  =  -l.5 


-^  =  308 
b1  =  -1.5 


0       40      80     120     160    200 

Scale  (km/sec) 


Figure  2.  Sample  records  taken  along  the 
plane  of  the  Milky  Way  in  the  vicinity  of  the 
galactic  center.  Each  curve  is  a  superposition 
of  four  to  six  individual  tracings  from  the  54- 
channel  spectrograph. 


line  profile.  Detailed  reduction  and 
interpretation  of  our  curves  are  not  yet 
complete,  but  preliminary  reductions 
show  good  agreement  with  Leiden. 

Most  of  our  analysis  so  far  has  been 
concentrated  on  features  lying  farther 
from  the  galactic  center  and  in  particular 
on  the  southern  side.  Typical  sets  of 
records  are  shown  in  figures  1  and  2, 
which  are  representative  of  the  profiles 
obtained  in  the  southern  sector  of  the 
survey.  The  three  profiles  in  figure  1, 
centered  on  the  "blue"  side  of  the  main 
emission  peak,  show  the  striking  feature 
known  as  the  "3.5-kpc  expanding  arm" 
from  which  Rougoor  and  Oort  demon- 
strated that  noncircular  motions  exist  in 
the  inner  parts  of  the  galaxy. 

Figure  2  shows  representative  profiles 


of  more  southern  longitudes,  invisible 
from  Leiden.  The  profiles  are  not  as 
simple,  since  not  one  but  several  peaks 
can  be  clearly  seen.  Figure  3  shows  the 
apparent  peak  velocities  for  all  features, 
and  also  the  velocity,  with  respect  to  the 
local  standard  of  rest,  as  a  function  of 
galactic  longitude.  If  all  the  gas  were  in 
circular  motion,  the  radial  velocity  should 
go  to  zero  at  I1  —  327.8°,  and  the  actual 
value  of  this  intercept  is  a  direct  measure- 
ment of  expansion  velocity.  The  "3.5-kpc 
expanding  arm"  is  represented  by  the  set 
of  points  labeled  A.  The  variation  in 
velocity  with  longitude  is  remarkably 
linear  for  this  feature  as  far  south  as 
ll  =  320°.  The  apparent  expansion 
velocity   derived   from   our   data   is   52 


E 


10 

1 

1 

1           1           1           1 

• 

20 

- 

• 

• 

C 

• 

30 

— 

• 
• 

40 

- 

• 

50 

• 

B                       •     .     .. 

•          • 
•  • 

60 

70 

• 

• 
• 

A 

• 

•     • 

•                      • 

80 
90 

o 

0 

• 

• 

•  • 

• 

100 

110 
120 
130 

- 

1 

I 

•  • 

• 

... 

• 

• 
• 

• 
I          l          I          1 

328        324         320         316         312         308        304 

Peak  velocities  near  galactic  center 


300   € 


Fig.  3.  Variation  of  peak  velocities  observed 
on  the  southern  side  of  the  galactic  center  in 
the  plane. 


216 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


km/sec,  in  very  good  agreement  with  the 
53  km/sec  found  by  Rougoor  and  Oort. 
From  the  slope  of  the  curve  (assuming 
uniform  rotation  and  expansion)  the 
difference  in  angular  velocity,  co(R)  — 
co(Ro),  can  be  derived,  where  the  distance 
of  the  sun  from  the  galactic  center  is  RQ. 
If  we  adopt  R0  =  8.2  kpc,  which  for  con- 
venience has  been  used  in  the  following 
discussion  as  a  scale  factor  for  all  galactic 
distances,  u(R)  —  o)(R0)  =  35.6  km/sec/ 
kpc.  Unless  R  is  known,  the  corresponding 
circular  velocity  at  R  cannot  be  deter- 
mined; if,  following  Leiden,  we  place  this 
feature  at  3.5  kpc,  and  assume  dc(Ro)  = 
216  km/sec,  then  6C(R)  =  216  km/sec 
also.  If  R  is  smaller  than  3.5  kpc,  6C{R) 
is  also  reduced.  For  any  reasonable  guess 
at  R,  6c(R)  does  not  fall  far  from  the 
Leiden  circular  velocity. 

At  approximately  ll  =  319°,  the  regu- 
larity of  feature  A  disappears.  The 
feature  divides  into  two  peaks,  which  can 
be  clearly  traced  as  far  as  I1  =  302°  but 
with  much  more  scatter  to  the  points. 
The  slope  of  the  highest-velocity  feature 
becomes  slightly  steeper,  as  far  as 
I1  =  315°,  when  it  flattens  out  and  main- 
tains nearly  constant  radial  velocity  as 
far  as  the  peak  can  be  traced.  The 
lower-velocity  part,  which  may  well  be  a 


different  feature  altogether,  exhibits  near- 
ly constant  radial  velocity  over  its  entire 
range. 

The  two  lower-velocity  features, 
labeled  B  and  C,  exhibit  similar  charac- 
teristics. The  slope  of  C  near  the  longitude 
of  the  galactic  center  is  very  nearly  the 
same  as  for  A,  although  the  intercept  at 
the  longitude  of  the  center  is  about  +35 
km/sec,  implying  that,  if  this  is  an 
expanding  feature,  it  is  on  the  far  side  of 
the  galactic  center.  The  feature  B  has  a 
less  negative  slope,  implying  co(R)  — 
u)(Ro)  =  28  km/sec /kpc.  From  the  appar- 
ent velocity  at  the  galactic  center,  an 
expansion  velocity  of  18  km/sec  is 
obtained.  Both  these  values  are  con- 
sistent with  this  feature's  lying  farther 
from  the  center  than  the  "3.5-kpc  arm." 

A  different  presentation  of  the  data  is 
given  in  figure  4,  which  shows  the 
intensity  of  hydrogen  emission  as  a 
function  of  longitude  and  velocity  for 
latitude  bl  =  —1.5°,  the  approximate 
galactic  plane.  The  most  intense  hydro- 
gen emission,  at  low  velocities,  is  not 
shown,  since  it  refers  primarily  to  local 
structures.  The  dotted  lines  give  the  run 
of  the  peaks  as  shown  in  figure  3.  The 
ridge  of  the  "3.5-kpc  expanding  arm"  can 
be  clearly  seen,  and  also  what  at  first 


e 

-2£ 


Fig.  4.     Brightness  of  hydrogen  emission  as  a  function  of  velocity  and  longitude,  along  galactic 
plane  (61  =  -1.5°).  Solid  lines  show  the  peak  velocities  from  figure  3. 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 

0 


217 


«x=330 


320 


Fig.  5.     Same  relief  map  as  in  figure  4  but  with  solid  lines  showing  various  possible  models  of 
circular,  uniformly  rotating,  and  expanding  arms. 


sight  looks  like  its  extension  to  longitude 
304°,  where  the  contours  suggest  that  we 
are  looking  tangentially  to  the  arm.  If  we 
are  looking  tangentially  at  this  longitude, 
the  feature  is  3.8  kpc  from  the  center. 
Closer  examination  shows  that  the  two 
parts  do  not  join  very  smoothly  across 
the  gap  around  I1  =  316°.  Indeed,  there 
is  a  faint  extension  between  I1  =  310°  and 
315°,  which  may  well  be  the  remnants  of 
the  "3.5-kpc  expanding  arm."  This  small 
discrepancy,  though  not  conclusive,  led  us 
to  other  tests,  represented  in  figure  5.  On 
the  same  contour  map  are  superimposed 
the  expected  velocities  for  uniformly 
rotating,  expanding,  circular  arms.  One 
curve  has  been  chosen  to  fit  the  straight 
part  of  feature  A,  and  the  other  to 
approximate  the  "turn  around"  at  ll  = 
304°.  It  can  easily  be  seen  that  either 
fitting  fails  for  the  other  half. 

The  "3.5-kpc  expanding  arm,"  there- 
fore, is  not  a  uniform  circular  feature,  and 
in  fact  may  not  be  at  3.5  kpc.  The  feature 
certainly  extends  as  far  as  I1  =  316°  and 
possibly  to  I1  =  310°.  These  longitudes 
correspond  to  lower  limits  of  1.7  kpc  and 
2.6  kpc,  respectively,  for  the  center-to- 
arm  distance.  It  appears  that  the  problem 
of  the  space  distribution  of  the  expanding 


gas  in  the  central  region  of  the  galaxy  is 
still  far  from  solved.  Work  is  proceeding 
on  the  reduction  of  the  remainder  of  the 
observations,  in  the  expectation  that  the 
angular  extent  in  latitude  will  provide 
further  clues  to  the  nature  of  these  gas 
complexes. 

Extragalactic  observations.  Among  the 
members  of  our  local  galactic  group  the 
Andromeda  nebula,  M  31,  and  the 
prominent  galaxy  in  Triangulum,  M  33, 
are  particularly  interesting  since  they  can 
be  partly  resolved  by  instruments  like 
our  60-foot  dish.  Both  these  galaxies  have 
been  studied  by  the  slightly  larger  Leiden 
antenna,  but  it  was  thought  that  a  repeat 
survey  would  be  of  some  value.  A  more 
compelling  incentive  was  to  prepare  for 
our  projected  observing  program  at 
Green  Bank  (West  Virginia)  on  the 
300-foot  telescope  now  in  construction 
there.  During  the  spring  and  early  sum- 
mer an  intensive  program  has  revealed 
the  necessity  of  understanding  the  factors 
involving  base-line  stability,  since  the 
antenna  temperatures  observed  on  these 
two  objects  are  never  greater  than  4°K. 
To  improve  the  statistical  fluctuation 
level,  the  receiving  bandwidth  was 
broadened  by  averaging  the  outputs  of 


218 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


four  adjacent  channels,  and  the  inte- 
grating time  was  lengthened  to  5  minutes. 
Ordinarily,  a  series  of  three  observations 
was  made  at  each  point  on  the  galaxy  and 
observations  at  standard  reference  points 
in  the  sky  were  interleaved  to  provide  a 
check  on  the  stability  of  the  instrument 
base  line.  Each  averaging  has  an  rms  of 
about  0.2°K,  and  with  such  sensitivity 
small  perturbations  in  the  base  line  were 
frequently  observed.  It  proved  to  be 
extremely  tedious  to  track  these  down, 
since  a  complete  set  of  three  averages 
takes  about  15  minutes,  and  initially  it  is 
not  known  whether  the  fluctuations  are 
caused  by  external  interference  (L-band 
radars,  for  example),  self-generated  inter- 
ference ("birdies"  in  oscillators  and 
amplifiers),  mechanical  difficulties  such 
as  loose  connections,  or  temperature 
effects  in  the  front  end  of  the  receiver. 
We  have  experienced  trouble  from  all 
these  causes,  but  temporary  summer 
temperature  problems  give  the  most 
difficulty,  and  we  feel  confident  that 
observations  of  the  necessary  accuracy 
can  now  be  performed  by  our  receiver. 
There  are  still  slow  drifts  during  the  day, 
but  by  careful  checking  at  reference  spots 
they  can  be  allowed  for. 

A  survey  on  M  31  has  been  completed; 
when  allowance  is  made  for  our  poorer 
angular  resolution  and  our  better  reso- 
lution in  frequency  (Doppler  velocity), 
we  are  in  substantial  agreement  with  the 
Leiden  survey  reported  in  B.A.N.  482.  A 
similar  survey  on  M  33  is  still  in  progress. 

Hydrogen-line  equipment  and  opera- 
tions. The  60-foot  radio  telescope  at 
Derwood  was  operated  throughout  the 
year,  only  minor  repairs  being  required. 
Observations  were  made  exclusively  at  21 
cm  with  the  54-channel  spectrograph. 
Tests  on  the  pointing  accuracy  have  been 
repeated,  by  both  radio  and  optical 
methods.  The  linear  scale  correction  to 
the  declination  indicators  was  determined 
by  photographing  stars  at  various  decli- 
nations with  the  6  inch  Dallmeier  camera 
mounted  on  the  dish  and  was  checked  by 
radio  observations  of  the  brighter  radio 


sources.  By  applying  the  new  declination 
correction  the  dish  can  be  pointed  with 
an  accuracy  of  better  than  0.02°. 

The  54-channel  spectrograph  performed 
well  throughout  the  year.  The  channel 
stability  has  been  greatly  improved  by 
converting  all  phase-sensitive  detectors 
from  hot-tube  diodes  to  matched  silicon 
diodes.  Daily  adjustments  of  all  diode 
balances  are  no  longer  required  for 
ordinary  galactic  observations. 

A  new  amplifier,  using  WE  437A's,  was 
installed  after  the  crystal  mixer,  with 
resultant  improvement  in  noise  figure. 
Direct  measurements  of  the  excess  noise 
were  made  by  measuring  the  increase  in 
noise  when  a  resistor  was  substituted  for 
the  antenna  feed.  In  the  six  months  since 
the  preamplifier  was  installed,  the  double- 
sideband  excess  noise  has  increased 
slightly,  from  440°K  at  the  outset  to 
550°K  in  late  spring,  with  no  changes  in 
mixer  crystal  or  437 A  tubes. 

A  great  deal  of  effort  has  been  expended 
in  understanding  causes  of  base-line 
variation  from  hour  to  hour.  When  one 
looks  at  the  regions  of  sky  known  to  have 
very  little  hydrogen  radiation,  the  base 
line  is  usually  displaced  from  the  meter 
zero  and  has  a  slope.  Changes  in  antenna 
tuning  and  in  the  matching  transformer 
between  antenna  and  mixer  affect  the 
zero  displacement,  but  finding  of  the 
cause  of  the  slope  (and  of  an  adjustment 
to  remove  it)  proved  to  be  an  elusive 
problem.  It  appears  that  the  only  adjust- 
ment that  affects  the  slope  significantly 
is  the  cascade  input  tuning.  Apparently 
the  noise  generated  in  the  cascade  input 
circuit  at  27  Mc/s  is  converted  up  to 
1420  Mc/s,  is  reflected  by  the  antenna,  is 
converted  down  again,  and  returns  to  the 
cascade,  but  in  frequency-dependent 
quantities.  Removing  the  zero  slope  by 
adjusting  the  cascade  input  tuning  affects 
the  zero  displacement  as  well,  but  this 
can  be  corrected  by  other  adjustments. 

In  addition  to  the  existing  54  channels, 
a  single  extra-narrow  filter,  only  2  kc/s  in 
bandwidth,  was  added.  Narrow  absorp- 
tion features  suspected  of  being  too  sharp 


DEPARTMENT   OF   TERRESTRIAL    MAGNETISM  219 

to  be  resolved  with  our  present  band-  venture  we  had  two  visitors  from  Brazil 

width  have  been  investigated  by  means  join  in  our  work  here  at  Derwood:  Dr. 

of  this  filter.  A.  H.  G.  Vieira,  Escola  Politecnica,  Sao 

Narrow-band  observations.  During  their  Paulo,  and  Dr.  G.  Schwachheim,  Centro 

visit  with  us,  Drs.  Vieira  and  Schwach-  Brasileiro  de  Pesquisas  Fisicas,  Rio  de 

heim  installed  the  above-mentioned  nar-  Janeiro.    During  their  four- month   visit 

row-band    (2-kc/s)    filter    as    an    extra  they  participated  in  hydrogen-line  obser- 

channel  on  the  54- channel  receiver  and  vations    and    in    planning    for    a    new 

with    it    observed    the    two    narrowest  interferometer  experiment.  Dr.  R.  A.  R. 

absorption  features  in  the  direction  of  Palmeira,  also  of  the  CBPF,  who  was 

Cygnus  X  and  the  galactic  center.   In  here  five  months  last  year,  is  a  partner  in 

both  cases  the  absorption  profile  seemed  the  Brazil  activities  as  well, 
slightly  deeper,  but  it  appears  that  the 

difference    between    these    narrow-band  Continuum  Observations 

observations  and  those  made  with  our         Precise  position  array.  In  Year  Book  60 

normal  bandwidth  (about  10  kc/s  to  half  a  list  of  derived  right  ascensions  of  radio 

power)  is  in  general  negligible.  sources  was  given.  The  small  area  of  the 

array  has  been  a  handicap,  and  only  the 

South  American  Cooperation  brighter  sources  could  be  measured  with 

The  Carnegie  radio  astronomy  station,  the  desirable  precision.  Hercules  A,  which 
to  be  located  between  Buenos  Aires  and  is  among  the  ten  brightest  radio  sources, 
La  Plata  in  Argentina,  is  well  under  way.  has  had  a  somewhat  questionable  optical 
The  major  instrument,  a  30-meter-  identification.  Originally  it  was  suspected 
diameter  equatorially  mounted  parabolic  that  the  radio  source  was  associated  with 
antenna,  will  be  built  during  the  coming  a  peculiar  galaxy,  of  about  the  same 
year.  Several  large  parts,  including  the  apparent  magnitude  as  Cygnus  A,  and, 
declination  and  polar  axis  assemblies  and  like  Cygnus  A,  showing  strong  emission 
drive  assemblies,  are  now  being  fabricated  lines.  Roberts,  Bolton,  and  Harris  at  Cal 
in  Baltimore  and  will  be  shipped  this  fall.  Tech  derived  an  improved  radio  position 
A  prototype  aluminum  rib  has  been  built  that  appeared  to  rule  out  that  possibility, 
to  test  fabrication  procedures,  and  suffi-  and  they  suggested  instead  that  an  even 
cient  aluminum  stock  and  other  materials  fainter  galaxy  of  about  19m  was  associated 
for  the  entire  dish  are  now  in  storage  at  with  the  radio  source.  A  position  derived 
Derwood,  also  to  be  shipped  this  fall.  A  at  the  Cavendish  Laboratory  by  Elsmore 
preliminary  site  in  Argentina  has  been  and  others  specified  still  another  spot  in 
agreed  upon,  and  construction  can  start  the  sky,  and  Dewhirst  noted  that  there 
when  final  on-the-spot  evaluation  has  was  also  a  peculiar  galaxy,  with  optical 
been  made.  The  Instituto  Nacional  de  brightness  of  about  18m,  at  this  position. 
Radio  Astronomfa  has  been  created  by  These  last  two  possible  identifications  are 
the  Consejo  Nacional  de  Investigaciones  separated  in  the  sky  by  little  more  than 
Cientificas  y  Tecnicas  and  by  the  Re-  a  minute  of  arc,  and  only  by  the  most 
search  Council  of  the  State  of  Buenos  careful  measurements  could  it  be  hoped 
Aires  to  provide  an  Argentine  organiza-  to  distinguish  between  them, 
tion  to  participate  in  this  joint  venture,  Figure  6  is  reproduced  from  Dewhirst 's 
which  will  involve  cooperation  between  discussion,  galaxies  a,  b,  and  c  being  the 
the  Carnegie  Institution  of  Washington,  three  successive  identifications.  Although 
the  University  of  Buenos  Aires,  and  the  our  "PPA"  (the  400-Mc/s  arrays  de- 
University  of  La  Plata,  with  invitations  scribed  in  previous  reports)  measures  only 
to  colleagues  in  other  institutions  to  right  ascension  at  present,  it  was  con- 
participate,  sidered  worth  while  to  make  a  new  series 

To  extend  the  base  of  our  cooperative  of  right-ascension  measurements  on  Her- 


220 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


1 

1           '    1 

1  cC 

T 

4- 



•         • 

•  _ 

1 

i 

Ob 

• 

• 

i 

o°  ♦ 

• 

• 

1 

i        I 

_L 

+5°05' 


c 
o 

o 

c 

75 
Q 


+  5°00' 


55* 


5CT 


45s 


40' 


I6h48m 


Right  Ascension 


Fig.  6.  Field  in  vicinity  of  Hercules  A.  Solid 
circles  are  stars.  Open  circles  show  possible 
galaxies  associated  with  radio  source.  Cross  near 
galaxy  c  is  Cal  Tech  position.  Cross  near  galaxy 
6  is  Dewhirst  position.  DTM  right  ascension 
shown  as  vertical  line  with  ±1  second  error. 


cules  A  to  try  to  discriminate  between  the 
two  possibilities.  These  measurements 
were  made  during  the  winter  and  early 
spring  of  1962  with  the  collaboration  of 
Dr.  J.  W.  Hollinger  of  George  Washing- 
ton University. 

To  determine  the  collimation  error  of 
the  antenna,  17  transit  measurements  of 
the  Cygnus  A  source  were  made  during 
the  period  January  6-30,  1962.  Compari- 
son of  the  measured  radio  right  ascension 
with  the  right  ascension  of  the  optical 
source  gave  a  collimation  error  of  + 1  .'8, 
which  was  applied  to  the  Hercules  A 
observations.  Earlier  measurements  indi- 
cated that  level  and  azimuth  errors  of 
less  than  5  seconds  of  arc  were  to  be 
expected,  and  therefore  the  data  were 
not  corrected  for  these  sources  of  error. 

A  total  of  30  transit  measurements  of 
the  Hercules  A  source  were  made  during 
the  period  February  3  to  May  7,  1962. 


The  transit  times  as  observed  are  shown 
in  figure  7.  These  results  lead  to  a  right 
ascension  (1950.0)  of  16h48m428  for  the 
Hercules  A  source.  This  value  can  be 
compared  with  the  right  ascension  deter- 
mined from  a  much  smaller  number  of 
transits  in  last  year's  report  a (1950)  = 
16h48m43s.  The  new  right  ascension  is 
indicated  in  figure  6,  with  the  estimated 
error  of  db  Is  shown.  The  error  shown  was 
derived  solely  from  the  statistics  of  the 
observations  and  does  not  include  syste- 
matic effects  such  as  changes  in  collima- 
tion correction  after  January  30.  The 
source  c  is  closer  to  the  observed  right 
ascension,  but  source  b  is  sufficiently 
close  to  be  a  possibility.  Source  a, 
however,  appears  to  be  clearly  ruled  out 
by  our  observations. 

Angular  size  interferometer.  The  need 
for  precise  positions  of  radio  sources  has 
continued  to  occupy  an  important  place 
in  our  planning,  but  it  is  becoming  evident 
that  measurements  will  also  be  needed  of 
source  angular  sizes.  To  get  both  suffi- 
cient angular  resolution  and  enough 
antenna  collecting  area,  it  is  necessary  to 
build  rather  large  antenna  structures. 
During  the  past  year  we  have  studied  the 
possibility  of  using  cylindrical  parabo- 
loids and  large  paraboloidal  dishes  as  an 
interferometer  designed  to  resolve  sources 
only  a  few  seconds  of  arc  in  size.  Sur- 
prisingly enough,  a  dish  of  30-meter 
diameter,  closely  following  our  La  Plata 
design  but  only  mounted  as  a  transit 
instrument,  compares  favorably  in  ex- 
pense with  a  cylindrical  paraboloid  of 
comparable  area  and  has  the  very 
important  simplicity  of  a  single  antenna 


I6h49m00s 
Tronsit  time 
(sidereal) 


05s         10s         15s        20s        25s 

Hercules  A  transits  1962 


Fig.  7.     Observed  transit  times  of  radio  source 
Hercules  A. 


DEPARTMENT   OF   TERRESTRIAL    MAGNETISM  221 

"feed."  It  was  finally  decided,  in  view  of  more  and  more  critical  evaluation  of  the 

the  several  practical  advantages  of  dishes  results  obtained  in  explosion  seismology 

over    cylindrical    paraboloids,    to    start  and  of  the  reality  of  the  physical  struc- 

construction  of  a  large  base-line  inter-  tures  deduced  from  the  observations. 

ferometer  having  two  or  three  such  dishes  The    questions    may    then    fairly    be 

as  elements.  The  first  dish  is  now  being  asked:   What   real   physical   entities   or 

constructed  at  Derwood;  it  will  be  used  structures  in  the  earth  can  we  hope  to 

with  our  present  60-foot  dish  as  a  two-  determine    from    explosion    seismology? 

element  interferometer  to  evaluate  the  What  criteria  are  sufficient  to  establish 

potentialities  of  the  system.  these    results    as    definite    rather    than 

merely    plausible?    In    the    subsequent 

THE  EARTH'S  CRUST  paragraphs,  some  modest  successes  from 

«             c  the  Maine  experiments  will  be  presented, 

Seismic  Studies  ,       ,,         .,,  ^          ,           ,     .?        e ,, 

together  with  some  demonstrations  of  the 

/.  S.  Steinhart,  R.  Green,1  T.  Asada,*  A.  very  great  difficulties  and  uncertainties 

Rodriguez  B.,«  L.  T.  Aldrich,  and  M.  A.  Tuve  encountered  with  gome  pr0Cedures  that 

For  more  than  50  years  the  structure  have  been  used  by  us  and  by  others  in 

of  the  crust  and  upper  mantle  has  been  the  past, 

represented,    chiefly    for    mathematical  . 

convenience,  as  one  or  more  horizontal  The  Mame  Seismic  Experiment 

layers  of  constant  wave  velocity  bounded  In  July  1961  an  intensive  study  of  the 

by  discontinuities  in  velocity.  During  the  earth's  crust  in  Maine  was  made  from  61 

last  10  years  or  more  this  simple  picture  explosions    detonated    in    the    Gulf    of 

has    appeared    inadequate,    for    several  Maine.    This    was    a    truly    cooperative 

reasons:    (1)    Efforts   to   find   the   near-  project,    in   which    the    Department    of 

vertical   reflections   from   the   supposed  Terrestrial    Magnetism    was    joined    by 

discontinuities   have   been  unsuccessful,  colleagues  from  the  University  of  Wis- 

(2)  Laboratory  measurements  of  seismic  consin,  Princeton,  Penn  State,  University 
velocities  in  various  rock  types  at  high  of  Michigan,  and  Woods  Hole  Oceano- 
pressures  and  temperatures,  coupled  with  graphic  Institution.  Some  other  groups, 
observed  geological  complexities,  make  it  largely  under  government  research  con- 
extremely  unlikely  that  the  upper  20  km  tracts,  were  able  to  utilize  the  explosions 
of  the  crust  is  a  constant-velocity  layer,  for  their  own  special  objectives.  This  kind 

(3)  Seismograms  obtained  in  the  field  are  of  cooperative  experiment  enabled  us  to 
more  complex  than  would  be  expected  obtain  far  more  detailed  information  from 
for  a  simply  layered  crust  of  one  or  two  the  explosions  than  ever  before  (more 
layers.  (4)  The  concept  of  horizontal  than  900  seismograms  were  obtained  from 
layers  may  be  approximately  true,  but  18  stations).  Such  a  large  amount  of 
geographical  variation  in  physical  prop-  detailed  information  was  obtained  that 
erties  has  been  clearly  demonstrated  on  a  analysis  of  the  results  is  far  from  corn- 
large  scale  and  fairly  small  lateral  plete,  although  certain  general  results  are 
changes  over  the  horizontal  dimensions  now  clear. 

of  an  experimental  region  may  introduce  Travel  times  and  crustal  structure.  The 

large  errors  into  the  calculated  result,  usual  representation  of  the  results  of  a 

These  objections  and  others  detailed  in  seismic  profile  studying  the  earth's  crust 

reports  of  previous  years  have  led  us  to  is  in  terms  of  a  plot  of  the  times  of  arrival 

of  wave  groups  as  a  function  of  range. 

1  Carnegie  Institution  Fellow;  from  University  Figure    8    shows    such    a    plot,    for    one 

o     asmama  station's  records,  of  all  the  shots  on  the 

2  Carnegie  Institution  Fellow;  from  University  _.    ,„„,..  .               r>,      T            i      ,•        n 
of  Tokyo.  Gulf  of  Maine  profile.  In  evaluating  the 

3  University  of  San  Agustin,  Arequipa,  Peru,  crustal    velocity    structure   from   such    a 


222 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


14 

6 

7           8 

12 

# 

• 

depth 
KM 

* 

km/sec 

10 

o 

• 

• 

1 

10 
20 

8 

6 

• 

>• 

o 

o 

• 

o 
o 

30 
40 

4 

•  1     *     <b 

"    o     W             o 

MODEL 

"O         2 

c 
o 
o 

$ 

A   -2 

• 
•o 

»             o    °* 

• 

• 

•  •    • 

?:  ? 

• 

o 

•  • 

•  O 

• 

• 

• 

•     • 

• 

• 
• 

•  • 

• 

•  o.    o 

• 
• 

• 

o 

• 

• 

An 

iplitudes 

o    !g 

• 
• 

o°. 

o        •     •» 

•«.  : 

o 

• 

o 

•    • 

» 

3 

» 

0 

a 
o 

-6 
-8 

-10 
-12 

o     1.0  to  3.0 

•  3.0  to  6.0 
o     6.0  to  10.0 

•  10  to  20 

MA 

INE  1961 
/ICTOR 

c 

•  •       ■ 

♦  .      o0     • 

♦  :  o- 

• 

+ 
+ 

o 

o           o 

o    •           • 

O 

•                   • 

0 

• 
• 

35  to  60 
nrwre  than  60 

\ 

•       • 

a 

• 
• 

o                 • 

• 

• 

0 

50 


«X) 


150 


200 

Kilometers 


250 


300 


350 


400 


Fig.  8.     Plot  of  the  times  of  arrival  of  wave  groups,  for  records  of  one  station,  of  all  the  shots  on 
the  Gulf  of  Maine  profile. 


plot  it  is  necessary  to  find  a  velocity 
structure  whose  travel  times  agree  with 
the  observations.  To  obtain  the  tradi- 
tional solution  in  terms  of  flat,  homoge- 
neous layers,  it  is  only  necessary  to  fit 
straight-line  segments  to  the  travel  times 
and  compute  the  depths.  For  Maine,  a 
two-layered  crust  with  velocities  of  6.05 
and  6.8  km/sec  would  fit  quite  well, 
giving  an  overall  crustal  thickness  of 
about  34  km.  But  we  are  led  to  inquire 
whether  this  is  the  only  solution  appro- 
priate to  the  observations,  especially  in 
view  of  the  objections  raised  above  to 
homogeneous  layers.  Tuve,  Tatel,  and 
Hart  showed,  for  Mary  land- Virginia,  a 
whole  range  of  possibilities  that  would 
satisfy  the  observations. 


Figure  9  presents  a  model  in  which  the 
homogeneous  layers  have  been  replaced 
by  layers  having  internal  velocity  gradi- 
ents, and  the  discontinuities  in  velocity 
have  been  replaced  by  rapid  gradients. 
There  is  no  a  priori  reason  for  assuming 
this  particular  model  to  be  more  or  less 
physically  real  than  the  homogeneous 
layer  model,  except  that  the  increase  in 
velocity  in  the  upper  part  of  the  crust 
approaches  more  nearly  to  some  measure- 
ments given  below.  The  important  change 
is  that  there  is  no  longer  any  velocity 
discontinuity  whatever,  although  there 
are  discontinuous  changes  in  gradient.  It 
will  be  noted  that  this  model  provides  a 
good  fit  to  the  travel-time  data  and  that 
the  curvature  in  the  travel  times  implied 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


223 


by  the  velocity  gradients  is  quite  small 
in  the  region  of  first  arrivals.  This  matter 
of  measurement  of  velocity  gradients 
from  curvature  in  the  travel  times  is 
discussed  in  more  detail  below. 

Figure  10  shows  a  further  change  in 
the  model  in  which  anything  that  could 
be  called  an  intermediate  layer  has 
disappeared.  Again  it  is  noteworthy  that 
the  travel  times  fit  the  observations  well 
and  that  the  curvature  of  the  travel-time 
curve  in  the  first-arrival  region  is  small. 

Finally,  in  figure  11a  model  is  shown 
with  a  continuous  velocity  depth  function 
that  has  nothing  "layerlike"  about  it. 
This  model  was  obtained  from  an  objec- 
tive statistical  program  for  the  IBM  7090 
computer,     in    which    the    first-arrival 


observations  are  fed  in  and  the  machine 
fits  the  best  model  to  them.  There  is  a 
restriction  on  the  models  produced  by 
this  program  to  single-valued  travel-time 
curves  (that  is,  no  reversed  cusps)  which 
makes  it  far  from  ideal  for  producing  a 
"best-fit"  model  in  any  general  sense.  The 
valuable  lesson  to  be  learned  here  is  that 
the  first-arrival  observations  may  be 
satisfied  by  a  model  like  that  in  figure  1 1 
which  is  very  different  from  our  custom- 
ary ideas  of  crustal  structure.  This  is 
somewhat  disappointing,  because  the 
most  definite  travel  times  that  can  be 
measured  are  those  of  the  first  arrival  of 
energy.  The  relative  reliability  of  meas- 
urement of  first-arrival  time  has  caused 
much  weight  to  be  placed  on  this  infor- 


200 

Kilometers 


400 


Fig.  9.     A  model  in  which  the  homogeneous  layers  have  been  replaced  by  layers  having  internal 
velocity  gradients,  and  the  discontinuities  in  velocity  have  been  replaced  by  rapid  gradients. 


224 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


200 

Kilometers 


400 


Fig.  10.  A  further  change  in  the  model  in  which  anything  that  could  be  called  an  intermediate 
layer  has  disappeared.  Note  that  the  travel  times  fit  the  observations  well,  and  that  the  curvature 
of  the  travel-time  curve  in  the  first-arrival  region  is  small. 


mation  in  the  past.  Figures  8  through  11 
show  clearly  that  first  arrivals  do  not 
furnish  enough  information  to  define  a 
velocity  depth  function. 

The  employment  of  arrival  times  of 
later  phases  is  necessary  for  a  more 
detailed  conclusion,  but  there  is  a  danger 
here.  Choice  of  the  specific  later  arrivals 
to  be  used  is  unavoidably  conditioned  by 
their  local  amplitudes,  because  the  only 
phases  that  can  be  noticed  are  those 
whose  amplitudes  at  this  point  are  sub- 
stantially above  the  general  confused 
seismic  signal  that  persists  for  some 
seconds  after  the  first  arrival.  Thus,  while 
a  complete  travel-time  curve  would  give 
an  unambiguous  velocity  depth  profile, 


some  parts  of  it  cannot  be  identified 
because  of  small  amplitudes  of  the 
arrivals  or  because  of  interference  be- 
tween coincident  arrivals.  Given  these 
restrictions,  we  may  conclude  that  travel- 
time  curves,  as  they  are  usually  thought 
of,  are  not  enough.  We  must  look  at 
apparent  velocities  (which  are  really  the 
first  derivative  of  the  travel-time  curve, 
but  can  be  measured  separately)  and 
amplitudes. 

One  positive  result  does  become  clear 
from  the  above.  The  layered  model  in 
Maine  gives  a  crustal  thickness  of  about 
34  km.  The  two  models  in  figures  9  and  10 
give  a  crustal  thickness  of  36  and  38  km. 
For  the  continuous  model  (fig.  11),  the  M 


DEPARTMENT    OF    TERRESTRIAL    MAGNETISM 


225 


is  more  difficult  to  define,  but  taking  the 
definition  from  Carnegie  Publication  622, 
the  crustal  thickness  is  about  39  or  40  km. 
We  have  then  placed  bounds  on  the  crust- 
al thickness  for  all  reasonable  models. 
The  depth  to  the  Mohorovicic  discon- 
tinuity in  Maine  is  36  d=  3  km  for  any 
reasonable  velocity  structure.  If  we 
consider  the  uncertainty  in  this  calculated 
depth  implied  by  the  scatter  of  points, 
and  assume  that  the  layered  model  is 
exactly  correct,  we  find,  from  the  statis- 
tical procedures  outlined  in  Carnegie 
Publication  622,  that  the  scatter  of  points 
about  the  fitted  model  means  at  least  ±  1 
km  uncertainty  in  the  depth.  If  all  we 
wish  to  know  is  the  crustal  thickness,  we 
know  it  to  ±3  km,  and,  with  the  present 


data,  we  cannot  expect  to  know  it  better 
than  ±1  km,  even  if  we  know  the  form 
of  the  velocity  depth  function.  But  to 
understand  the  mechanics  of  the  earth 
we  wish  to  know  the  velocity  depth 
structure,  and  by  specialized  methods  of 
observation,  such  as  were  used  in  Maine, 
we  may  be  able  to  decrease  the  scatter  of 
the  travel-time  points  through  averaging. 
The  use  of  apparent  velocities.  Over  the 
years  the  results  of  many  seismic  refrac- 
tion profiles  have  been  used  to  find  the 
thickness  of  the  crust  in  terms  of  " vir- 
tual" depths  (Tatel  and  Tuve)  rather 
than  obtaining  the  velocity  depth  func- 
tion for  the  crust  and  hence  its  true 
thickness.  The  virtual  depths  indicate 
distinct  structural  difference  between  one 


M 
12 
10 
8 

i 

( 

7            8 

# 

• 

depth 
KM 

\                ' 

km/se 

f            1 

: 

O 

• 

• 

1 

10 

°0 

( 

)  % 

• 

30 

6 

®o 

0 

40 

A 

2 

0 

-2 

3  i  •    ^o 

O 

o 

o  O 

MO 

DEL 

■o 

c 
o 
o 

m 

•o 

i                o     <* 

• 

•      • 

t: 

• 

• 
• 
8 

• 
o 

•       • 

• 

o 

• 

• 

• 

•     • 

• 

• 
• 

•      • 

• 

.     o.    o 

• 
• 

• 
■ 

0 

• 

ID 

<] 

1 

An 

plitudes 

■a         P               * 

0°. 

o         *      •  * 

e 

• 
o 

» 

0 

0 

•      • 

» 

o 

©     1.0  to  30 

•   • 

V^*      o  o     * 

o 

-6 
-8 

•  30  to  6.0 
o     60  to  100 

•  10  to  20 

MA 

INF  1961 

>J  o- 

•SS^ 

o 

0     .              . 

o 

•                    • 
o 

•     35  to  60 

VICTOR 

• 
• 
o                     • 

-10 
-12 

# 

Txxe  thon  60 

o 

50 


100 


150 


200 

Kilometers 


250 


300 


350 


400 


Fig.  11.    A  model  with  a  continuous  velocity  depth  function  that  has  nothing  "layerlike"  about  it. 
This  model  was  obtained  from  an  objective  statistical  program  for  the  IBM  7090  computer. 


226 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


area  and  another,  and  this  in  itself  is 
valuable  geophysical  information.  On  the 
other  hand,  most  questions  about  the 
detailed  structure  of  the  crust  are 
difficult  to  answer  with  any  certainty. 
The  important  question  considered  above 
is  whether  the  crust  is  horizontally 
layered  or  whether,  as  one  would  expect, 
a  better  accord  with  reality  can  be 
afforded  by  considering  the  rocks  at 
increasing  depths  in  the  crust  to  be  so 
affected  by  increasing  temperature  and 
pressure  with  depth  that  a  continuous, 
but  nonuniform,  velocity  gradient  exists 
from  the  surface  downward. 

To  demonstrate  the  presence  of  velocity 
gradients  in  the  crust,  however,  it  is 
necessary  to  show  that  the  time-distance 
curve  has  not  only  slope  but  also  curva- 
ture, and  furthermore  it  is  necessary  to 
show  that  this  curvature  (of  the  order  of 
1/500  km-1)  is  significant.  In  other  words, 
for  a  linear  velocity  gradient  it  is  neces- 
sary to  demonstrate  the  significant  differ- 
ence between  the  curves  given  by 

t  =  A/V  (1) 

and 

t  =  2/ a  arcsinh  aA/2V0 
=  2/a{aA/2V0  - 

(gA/2F0)3/6+  •••]      (2) 
where 

t  —  the  travel  time. 
A  =  the  shot-point  distance. 
V  =  the  average  velocity  for  the 
constant-velocity  layer 
model. 
V0  =  the  surface  velocity  for  the  model. 
a  =  the  velocity  gradient. 

For  small  values  (0.01-0.04  km/sec/km) 
of  gradient,  which  are  possible  in  the 
crust,  the  difference  in  arrival  times  given 
by  (1)  and  (2)  is  usually  less  than  0.1  sec. 
Unfortunately,  the  recognition  of  the 
first  arrival  on  a  seismic  record  is  not 
unequivocal.  Tatel  and  Tuve  demon- 
strated with  model  experiments  and 
Gamburtsev  et  al.  showed  that  the  time 
of  the  ''preferred  pick"  for  the  apparent 
first  arrival  is  dependent  on  the  gain 
setting  of  the  amplifier  of  the  recording 


system.  The  laboratory  and  field  work 
showed  that  it  is  only  when  the  signal-to- 
noise  ratio  is  high  that  a  consistent  pick 
of  a  P  event  can  be  made. 

This  emphasis  on  curvature  in  the 
travel-time  curve  led  us  to  an  examina- 
tion, in  terms  of  information  theory,  of 
the  accuracy  with  which  arrivals  can  be 
determined  from  a  seismic  record.  Con- 
sider a  seismic  record  that  is  the  result  of 
the  transmission  and  reception  of  infor- 
mation through  a  ground  geophone 
system,  and  that  has  a  frequency  band- 
width of  W  c/s  and  a  signal-to-noise  ratio 
of  P/N.  The  maximum  rate  of  trans- 
mission of  information  through  such  a 
system  in  terms  of  bits  of  information  per 
second  is  given  (by  Shannon  and  Weaver) 
as 

-      !        lod    i    -i-  ^)( 


-  /o  log(1  +  0 


An  "arrival"  cannot  be  identified  with 
the  uncertainty  in  arrival  time  less  than 
the  time  taken  to  transmit  one  bit  of 
information  through  the  channel.  To 
evaluate  (3),  a  knowledge  of  the  spectral 
distribution  of  noise  over  the  frequency 
band  of  the  recorder  is  required. 

If,  to  obtain  an  indication  of  the 
magnitude  of  the  expected  uncertainty, 
the  assumption  is  made  that  the  noise  is 
random  over  this  band  of  bandwidth  40 
c/s,  and  for  a  poor  site  the  signal-to-noise 
power  ratio  is  2,  the  channel  capacity  is 
44  bits/sec.  Thus  the  minimum  uncer- 
tainty in  the  pick  of  an  identifiable 
arrival  is  22.8  milliseconds.  For  the  case 
of  an  "ideal  site"  where  the  signal-to- 
noise  ratio  S/N  is  100,  the  channel 
capacity4  is  less  than  185  and  the  un- 
certainty in  the  arrival  time  of  an  event 
is  reduced  to  6  milliseconds.  In  practice 
the  optimal  value  will  never  be  attained, 
because  the  sufficient  and  necessary 
information    required    to    identify    the 

4  For  large  S/N  ratios  the  channel  capacity  is 
C  =  W  log  (2TreS/N  +  1)(1  +  e),  where  e  is 
arbitrarily  small.  Thus  the  asymptotic  approx- 
imation in  equation  3  will  yield  too  large  a  figure 
for  channel  capacity. 


DEPARTMENT    OF    TERRESTRIAL    MAGNETISM 


227 


arrival  of  an  event  has  never  been  stated 
in  precise  language. 

With  the  above  discussion  of  the 
uncertainty  in  identifying  an  arrival  as  a 
guide,  the  expected  accuracy  of  the 
determination  of  the  velocity  V  of  a  wave 
front  over  the  interval  between  two 
points  a  distance  I  apart  will  be  evaluated. 

V  =  l/t 
AV  =  (V2/l)M  +  (V/l)M         (4) 

Inserting  numerical  values  of  0.02  sec  for 
At,  2  km  for  I,  5  per  cent  for  Al/l,  and 
6  km/sec  for  V  leads  to  errors  of  1  km/sec 
in  velocity.  It  might  appear  that  im- 
proved accuracy  can  be  obtained  by 
increasing  /,  but  this  may  introduce  other 
problems  due  to  local  geological  varia- 
tions. For  example,  whenever  the  second 
derivative  of  the  velocity  depth  function 
becomes  positive,  a  triplication  occurs  in 
the  travel-time  curve,  leading  to  the  loss 
of  coherence  of  the  wave  form  over 
distances  greater  than  a  few  kilometers. 
In  such  circumstances  an  observer  has  no 
assurance  that  he  is  basing  his  velocity 
determination  on  the  successive  arrivals 


of  the  same  phase  (Green  and  Steinhart) . 
However,  using  radial  seismometer  arrays 
2  km  long,  the  same  event  can  usually  be 
traced  with  confidence  across  the  whole 
array,  and  the  apparent  velocities  can  be 
determined  with  the  accuracy  stated 
above. 

The  measurement  of  apparent  velocity 
has  been  carried  out  for  nine  stations 
having  shot-point  ranges  of  40  to  140  km 
for  Maine  (fig.  12).  This  range  was 
selected  because  the  time-distance  curve 
is  apparently  fitted  by  a  straight-line 
showing  of  a  velocity  of  6.05  km/sec  over 
this  range  and  also  the  first  arrivals  here 
are  rays  that  have  not  been  refracted 
from  below  the  Moho  as  head  waves. 

For  every  station  a  plot  of  apparent 
velocity  as  a  function  of  shot-point 
distance  has  been  made  and  the  slope  of 
the  straight  line  of  best  fit  (least  squares) , 
together  with  the  estimated  standard 
deviation,  S\,  has  been  determined.  The 
statistics  show  that  for  every  station  the 
best  estimate  of  slope  is  positive  (i.e., 
apparent  velocity  increasing  with  dis- 
tance),   and    zero   slope    is    statistically 


i?.u 

• 

O 

« 

• 

<D 

# 

-iC    8.0 

- 

> 

• 
• 

e 

• 

.*: 

>N 

• 
•  • 

• 
• 

•      •    • 

• 

o 

• 

•   • 

• 

*•         -^•""■""" — 

o   _^ 

• 

•    • 

•  • 

~^^>          • 

a>    7.0 

— 

• 

• 

• 
• 

• 

• 

> 

• 

*    -~*^~^ 

• 

• 

•  • 

•  ^^^-""^"^ 

•  • 

• 

• 

•     «                    • 

•  • 

^^*^*^    • 

■ 

c 

• 

*         • 

<L) 

+ 

•  • 

it- 

• 

•     • 

CD 

v^ •     • 

•   • 

•     A 

Q. 

_«•  ••^-- — 

• 

a.  6.0 
< 

•         • 
• 

• 
• 

i 

• 
• 

• 

• 
• 

• 

• 

• 

1 

•  * 
• 

• 

50  100 

Range,  km 


150 


Fig.  12.  Measurement  of  apparent  velocity  carried  out  for  nine  stations  having  shot-point 
ranges  of  40  to  140  km  for  Maine.  This  range  was  selected  because  the  time-distance  curve  is 
apparently  fitted  by  a  straight-line  showing  of  a  velocity  of  6.05  km /sec  over  this  range  and  also  the 
first  arrivals  here  are  ravs  that  have  not  been  refracted  from  below  the  Moho  as  head  waves. 


228  CARNEGIE     INSTITUTION      OF      WASHINGTON 

unacceptable  to  four  stations.  However,  amplitude     are     inexorably    associated, 

a    slope    of   0.01    km/sec/km    range    is  measurements   tend   statistically   to   be 

acceptable  to  all  stations  with  a  confi-  confined    to    signals    enhanced    by    the 

dence  of  95  per  cent.  This  would  suggest  various  slight  increases  in  the  velocity 

velocity  gradients  in  the  crust.   Before  gradient,    those    from    zones    having    a 

discussing  a  smooth  velocity  gradient  in  decrease  in  velocity  gradient  being  ig- 

the  crust,  it  is  appropriate  to  examine  the  nored,  and  the  records  will  display  chiefly 

suitability  of  a  number  of  discrete  layers  the  features  associated  with  the  larger 

there.   This  is  simple,   because  discrete  amplitudes.  Hence  the  velocity  gradients 

layers  in  the  crust  would  be  expected  to  as  measured  by  apparent  velocities  over 

appear  as  distinct  steps  in  the  data  for  short  spreads   (over  which  coherence  is 

apparent    velocity    as    a    function    of  maintained)  will  tend  to  be  too  high, 

distance.  Figure  12  shows  that,  irrespec-  In  figure  13  the  temperature  is  plotted 

tive  of  what  may  be  the  true  velocity  as  the  abscissa  and  pressure  in  kilobars  as 

depth  function,  discrete  layers  of  constant  the  ordinate,  and  on  this  field  are  drawn 

velocity  are  distinctly  not  indicated.  curves  of  equal  velocity  for  a  rock  of 

If,  as  a  first-order  approximation,  the  granitic  composition.  Superposed  on  this 

apparent   velocity   is   considered   to   in-  temperature-pressure    field    is    shown    a 

crease  linearly  with  distance  at  less  than  heavy  curve  indicating  an  extreme  veloc- 

X   =   0.01  km/sec/km  range,  the  corre-  ity  depth  gradient  of  0.05  km/sec/km 

sponding  increase  in  velocity  with  depth  depth.   It  can  be  seen  that  at  depths 

is  not  linear  but  is  given  implicitly  by  greater  than  3  km  this  gradient  cannot 

be   maintained   for   a   rock   of   granitic 

%v  —  (Fp/X7r)[(F0/Fp)(arccoshFp/Fo)  composition  without  a  temperature  in- 

—  (arccos  VQ/VP)]                    (5)  version.    But   it   is   necessary   that   the 

temperature  increase  with  depth,  because 

where  Zp  is  the  depth,  V p  is  the  corre-  the  heat  flow  is  outward.  Consequently, 

sponding  velocity,  and  VQ  is  the  surface  to  maintain  the  gradient  and  to  avoid  the 

velocity.  Equation  5  shows  that  initially  inversion,  we  must  admit  a  change  in 

the  velocity   Vp  increases  very  rapidly  composition    toward    more    basic    rock 

with  depth  and  that  with  increasing  depth  types.  This  is  in  contrast  to  the  situation 

the  rate  of  increase  becomes  gradual.  where  the  gradient  is  small  ( <0.01  sec-1), 

Before  discussing  a  velocity  gradient  in  and  can  be  maintained  down  to  depths  of 
the  crust  in  terms  of  petrological  compo-  over  20  km.  In  this  case,  however,  if  the 
sition  and  likely  temperatures,  it  should  temperature  at  about  20  km  appears  to 
be  pointed  out  that  the  superimposition  exceed  300°  the  gradient  will  further 
of  minor  fluctuations  upon  a  general  decrease,  and  ultimately  a  velocity  re- 
velocity  gradient  can  be  both  important  versal  may  result.  In  other  words,  it  is 
and  observationally  confusing.  difficult  to  conceive  of  a  granite  layer 

Every  positive  increase  in  the  velocity  thicker  than  20  km  that  does  not  have  a 
gradient  will  lead  to  the  recording  of  velocity  reversal,  especially  when  the  high 
(1)  large  amplitudes,  (2)  triplication  in  heat  production  rates  in  granites  are 
the  time-distance  curve  (by  a  cusp),  and  considered.  The  situation  is  markedly 
(3)  a  rapid  increase  in  the  apparent  different  if  compositional  changes  in  the 
velocities,  whereas  every  decrease  in  the  crust  are  assumed.  Birch  has  suggested 
velocity  gradient  leads  to  the  recording  of  that  velocity  depth  gradients  can  be 
(1)  vanishingly  small  amplitudes,  (2)  a  altered  by  increasing  the  amount  of 
single  arrival,  or,  in  an  extreme  case,  a  gabbroic  material  present.  In  figure  14,  a 
discontinuity  in  the  time-distance  curve,  set  of  curves  for  a  gabbro  (Hughes  and 
and  (3)  an  effectively  constant  value  for  Maurette)  is  given.  The  falloff  in  velocity 
the  apparent  velocity.  Because  the  re-  with  temperature  is  even  more  remark- 
corded  time  of  arrival  of  a  phase  and  its  able  at  temperatures  above  300°C  than 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 

Temperature  -  °C 


229 


0 


1000 


2000 


3000 


E 
„$_> 

en 
i 

CD 


&   4000 


CO 
CD 


5000 


6000 


7000 


8000 


200 


300 


Fig.  13.  Temperature-pressure  field  showing  curves  of  equal  velocity  for  a  rock  of  granitic  com- 
position. Heavy  curve  indicates  an  extreme  velocity  depth  gradient  of  0.05  km/sec/km  depth, 
assuming  no  compositional  change. 


230 


CARNEGIE     INSTITUTION      OF      WASHINGTON 

Temperature  -  °C 


100 


200 


300 


400 


1000 


2000 


3000 


E 

CD 
±£ 
I 

<1> 


4000 


5000 


6000 


7000 


8000 


GABBRO 


Fig.  14.  Temperature-pressure  field  showing  curves  for  a  rock  of  gabbroic  composition.  The  fall- 
off  in  velocity  with  temperature  is  even  more  remarkable  at  temperatures  above  300  °C  than  for 
granite;  consequently,  if  no  low-velocity  layer  is  present  in  the  crust,  the  temperature  is  below  300 °C. 


for    granite;    consequently,    if    no    low-  that  roughly  the  upper  3  km  of  the  crust 

velocity  layer  is  present  in  the  crust,  the  is  granitic  and  below  this  the  percentage 

temperature  is  below  300°C.  of  gabbro  increases  so  as  to  maintain  a 

The  situation  in  Maine  appears  to  be  generally  steady  gradient  with  but  minor 


DEPARTMENT    OF    TERRESTRIAL    MAGNETISM 


231 


fluctuations.  At  a  depth  of  20  km  the 
material  is  almost  entirely  gabbroic.  At 
depths  greater  than  20  km  there  can  be 
very  little  increase  in  velocity  with  depth 
(see  fig.  14).  Indeed,  it  is  quite  possible 
that  there  is  actually  a  slight  velocity 
decrease  for  some  distance  below  20  km. 
Furthermore,  such  a  decrease  in  the 
velocity  gradient  below  20  km  is  in 
agreement  with  the  travel- time  curve. 

The  velocity  depth  function  as  deduced 
from  the  observations  is  in  very  close 
agreement  with  the  curve  given  by 
equation  5,  which  shows  a  rapid  increase 
in  velocity  with  depth  close  to  the  surface 
and  a  decreasing  rate  at  greater  depths. 

It  is  appropriate  to  mention  that  the 
proposed  model  crust  has  appreciably  less 
granitic  material  than  is  often  assigned 
to  a  continental  crust  (Clark,  Year  Book 
59). 

South  American  Local  Earthquake  Network 

Operations  have  been  continued  during 
the  year  in  Peru,  Bolivia,  and  northern 
Chile  by  our  academic  colleagues  there, 
using  the  array  of  about  19  relatively 
temporary  stations  where  our  simple 
vertical  seismographs  have  been  installed 
in  the  Andes.  Difficulties  with  precision 
timekeeping  and  with  local  seismic  noise 
have  limited  the  value  of  some  of  these 
efforts,  but  a  profusion  of  interesting  data 
exists  in  the  voluminous  records  obtained. 
Some  of  the  records  have  been  interpreted 
by  Fr.  G.  Saa  at  Antofagasta,  Chile,  and 
by  our  staff  members  with  Professor  A. 
Rodriguez  B.  at  Arequipa,  Peru.  The  data 
obtained  in  Bolivia  under  the  guidance 
of  Dr.  R.  Salgueiro  and  Fr.  R.  Cabre  have 
not  been  intensively  studied.  The  U.  S. 
National  Science  Foundation  (grant  G- 
14593)  has  contributed  generously  to  the 
expenses  in  South  America  of  this 
cooperative  project,  which  was  initiated 
and  equipped  with  the  help  of  special 
Carnegie  grants. 

Studies  of  attenuation.  The  anomalous 
high  attenuation  first  reported  in  Year 
Book  57  has  been  examined  again  in  the 
light  of  records  from  the  South  American 


earthquake  network.  These  studies  have 
been  progressing  slowly,  because  of 
operating  difficulties  and  equipment 
troubles  at  the  various  sites,  but  it  is 
clear  that  the  local  earthquakes  show  the 
same  marked  attenuation  of  seismic 
energy  as  the  1957  explosions. 

Recent  studies  further  indicate  that 
the  high  attenuation  occurs  in  the  upper 
tens  of  kilometers  of  the  crust  and  that 
attenuation  is  most  extreme  for  higher 
frequencies.  As  our  equipment  is  im- 
proved and  additional  stations  begin  to 
operate  on  a  continuing  basis  it  will  be 
possible  to  work  more  quantitatively  with 
this  problem. 

The  San  Agustin  Fault  System.  The 
discovery  of  a  near  vertical  fault  zone  300 
or  more  kilometers  deep,  parallel  to  the 
margin  of  the  Pacific  in  southern  Peru, 
was  first  reported  in  Year  Book  60.  More 
complete  and  detailed  information  was 
gathered  in  1961  to  confirm  this  finding 
and  delineate  the  details  of  the  fault  zone. 
The  system  was  named  the  San  Agustm 
Fault  System  in  recognition  of  the 
University  in  Arequipa,  Peru,  where  the 
work  was  done.  A  paper  by  Rodriguez, 
Steinhart,  and  Asada,  giving  these  results, 
is  in  press.  Further  investigations  of  the 
vertical  and  horizontal  dimensions  of  the 
fault  zone  are  in  progress. 

Crustal  structure  in  southern  Peru.  From 
the  small,  four-station  earthquake  net- 
work used  to  study  the  San  Agustin  Fault 
System,  a  new  method  of  studying 
crustal  structure  has  been  evolved.  It 
depends  on  the  frequent  occurrence  of 
earthquakes  in  appropriate  positions  and 
so  is  not  suitable  for  universal  applica- 
tion. Nevertheless,  it  offers  a  new  look  at 
crustal  structure  and  may  be  of  wide 
applicability  in  the  Andes  and  perhaps  in 
other  active  earthquake  areas. 

If  the  observed  travel  time  for  some 
station  is  T,  the  distance  to  the  earth- 
quake hypocenter  is  kT,  where  k  has  the 
dimensions  of  a  velocity  and  can  be 
identified  physically  as  the  average 
velocity  from  hypocenter  to  station.  Let 
us  assume  that  the  earth  is  a  constant- 


232 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


velocity  half-space.  Then  if  we  have 
observations  of  the  travel  time  at  four 
stations  we  can  set  up  four  simultaneous 
equations  to  solve  for  the  three  position 
coordinates  of  the  hypocenter  and  the 
value  of  k.  Most  graphical  methods  of 
locating  earthquakes  are,  in  fact,  exactly 
this  procedure  done  graphically. 

If  our  naive  assumption  that  the  earth 
is  a  constant-velocity  half-space  were 
correct,  we  could  plot  the  depth  of  focus 
against  the  calculated  value  of  k  for  the 
various  earthquakes  and  obtain  a  straight 
line,  since  k  would  be  the  same  inde- 
pendent of  depth.  But  this  assumption  is 
known  to  be  not  true,  for  velocity  is 
known  to  vary  with  depth.  The  values  of 
the  depth  of  focus,  H,  and  k  obtained  by 
this  simple  procedure  may  then  be  called 
apparent  depth  and  apparent  k. 

It  can  be  shown  analytically  that  the 
geographical  coordinates  of  the  epicenter 
directly    above    the    quake    are    very 


E 

.a: 

parent  depth 
8            8 

^A 

A    * 

A 

UV 

A 

Q. 

< 

Region 
i 

A 
l 

i 

i                 i 

i 

5.5     60  7.0  8.0  9.0 

Apparent  k  km /sec 


10.0 


11.0 


Fig.  16.  Values  of  apparent  depth  plotted 
against  apparent  constant  velocity  for  region 
A  encircled  in  figure  15. 


insensitive  to  the  simple  assumption 
above.  That  is,  we  will  get  the  correct 
epicenter  but  the  wrong  values  of  depth 
and  k.  Now,  if  we  restrict  ourselves  to 
consideration  of  earthquakes  occurring  in 
some  small  epicentral  region  and  apply 
our  simple  procedure  we  will  be  dealing 
with  earthquakes  that  have,  in  fact, 
occurred  beneath  that  region,  and  for 
each  earthquake  we  will  have  values  for 
apparent  depth  and  k. 

Figure  15  shows  the  Arequipa,  Peru, 
network  with  two  regions  encircled  and 
the  earthquake  epicenters  shown.  Figures 
16  and  17  show  the  values  of  apparent 
depth  plotted  against  apparent  k  for 
these  two  regions.  The  observed  points 
are  not  in  a  straight  vertical  line,  as  they 
would  be  for  constant  velocity  (k),  and 


Fig.  15.  Arequipa,  Peru,  seismic  network. 
Regions  A  and  B  (encircled)  designate  where 
numerous  local  earthquakes  have  occurred,  and 
the  earthquake  epicenters  are  shown. 


7.0  8.0  9.0 

Apparent  k  km /sec 

Fig.  17.  Values  of  apparent  depth  plotted 
against  apparent  constant  velocity  for  region  B 
encircled  in  figure  15. 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


233 


0 

6             8            10           12            14            16 

|31 

•32 

K 

km/sec 

(  r 

])- 

20 

j 

i — ; 

u 

i         i 

J 

) 

40 

|     | 

l 
i 

! 

60 

[ 

$ 

30 

1 
\ 

i 

132 

i 

--.31           ! 

(d) 

E 

_x: 

>-—, 

parent  depth 
8             8 

A    A        i    Ai^ 

Q. 

i        \ 

< 

Region 

i 

A 

1                               1 

\     \ 

«           \ 

5.5     6.0  7.0  8.0  9.0  100 

Apparent  k  km /sec 


no 


8  10  12  14  16 


74 


A 


Region  A 


(e) 

X*       *    *>A 

\N?V\a  A 

\\  \ 

\\  \ 

I ■  \  ■■■■■  ■  * 


5.5     6.0 


7.0  8.0  9.0  10.0 

Apparent  k  km /sec 


11.0 


6             £ 

S 

10            1 

2            14           16 

0 

~TI 

K 

km/sec 

(    , 

:)- 

v  c 

20 

1 

40 

60 

41) 

72 

E 


50 


a. 

CD 


Q.I00 
Cl 
< 


I50 


72 


(  f  ) 

**  *a\\ 

A    A     ^  XAA      a 

%    \A  A 

\  \ 

Region  A  S.       \ 

\    \ 


55      6.0  7.0  8.0  9.0  100 

Apparent  k  km /sec 


I  ID 


Fig.  18.     Possible  models  for  velocity  versus  depth,  and  the  corresponding  theoretical  curves 
superimposed  on  the  observations  from  region  A  of  figure  15. 


234 


CARNEGIE     INSTITUTION      OF     WASHINGTON 


the  amount  by  which  they  deviate  is  a 
reflection  of  the  velocity  structure  of  the 
crust  and  upper  mantle.  Hence,  figures 
like  16  and  17  contain  quantitative 
information  about  crustal  structure,  and 
we  should  find  a  way  to  extract  it. 

At  this  point  we  must  digress  to  point 
out  one  experimental  problem  that  exists 
with  these  data.  Because  of  small  irregu- 
larities in  the  speed  of  the  recording 
drums  and  various  local  difficulties  in 
receiving  time  signals,  it  was  not  possible 
to  determine  absolute  time  accurately 
enough  to  use  the  travel  time  of  the  first 
arrival,  the  compressional  wave,  directly. 
Recourse  was  then  had  to  the  measure- 
ment of  the  time  difference  between 
arrival  of  the  compressional  and  shear 
waves  at  each  site.  This  PtoS  time  is  also 
proportional  to  the  distance  traveled, 
provided  that  the  compressional  and 
shear  waves  travel  essentially  the  same 
paths.  We  can  proceed  with  the  analysis 
as  before  with  this  additional  assumption, 
which  is  equivalent  to  requiring  the  ratio 
of  these  two  velocities  to  be  a  constant. 
As  will  be  seen  below,  it  is  not  entirely 
justified. 

Having  the  observational  data  of 
figures  16  and  17  we  may  then  propose  a 
typical  crustal  structure  and  compute  the 
true  S-P  travel  times  for  that  specific 
structure.  Treating  these  times  as  though 
they  were  observational  data,  we  may 
apply  the  same  procedure  as  was  applied 
to  the  actual  earthquake  data  and 
produce  a  theoretical  curve  of  apparent  k 
versus  apparent  depth.  If  the  model  we 
assumed  is  correct,  the  theoretical  curve 
should  pass  through  the  observed  points. 
This  theoretical  analysis  was  programmed 
for  the  IBM  7090  and  a  number  of  models 
were  calculated.  In  figure  18,  a,  b,  and  c 
show  the  models,  and  d,  e,  and  /  show  the 
theoretical  curves  superimposed  on  the 
observations  from  region  A.  It  will  be 
noted  that  none  of  the  theoretical  curves 
fit  the  observations  especially  well,  de- 
spite the  wide  range  of  models  investi- 
gated. 

The  only  conclusion  that  seems  reason- 


able is  that  for  the  lower  part  of  the 
crust  the  compressional  and  shear  waves 
do  not  travel  the  same  paths.  This,  in 
itself,  may  be  a  demonstration  that 
Poisson's  ratio  and  some  of  the  elastic 
constants  change  with  depth  in  a  way 
different  from  that  usually  assumed. 
Recent  work  has  been  directed  toward 
improving  the  timing  of  the  instruments. 
When  the  compressional-  and  shear-wave 
travel  times  can  be  treated  separately, 
the  above  procedure  can  be  followed  for 
each  and  the  crustal  structure  can  be 
examined  in  more  detail. 

Radioactive  Ages  of  Rocks 

L.  T.  Aldrich,  S.  R.  Hart,  I.  Hayase,5  G.  L.  Davis,* 
G.  R.   Tilton,6  B.  R.  Doe,1  and  H.  Baadsgaard* 

The  work  of  the  group  this  year  strikes 
a  balance  between  application  and  meth- 
odology. In  the  course  of  international 
cooperation,  preliminary  investigations 
have  been  carried  out  on  rocks  from 
Japan,  Brazil,  and  Thailand.  Though 
furnishing  much  helpful  information,  the 
results  support  the  general  finding  that 
ages  from  areas  of  complex  geological 
history  have  complex  interpretations  and 
that  careful  work  by  various  methods  on 
several  minerals  is  necessary  for  real 
understanding. 

Studies  of  minerals  low  in  potassium 
and  of  the  effects  of  contact  meta- 
morphism  on  mineral  ages  have  been 
continued  as  a  means  for  understanding 
the  complex  age  patterns  frequently 
encountered.  The  results  of  these  studies 
are  summarized  below. 

Kyoto  University-Carnegie  exchange. 
During  this  report  year  the  Department 
activated  an  exchange  program  with  the 
Geological  and  Mineralogical  Institute  of 
the  University  of  Kyoto  which  had  been 

5  Guest  Investigator;  from  University  of 
Kyoto. 

6  Geophysical  Laboratory,  Carnegie  Institu- 
tion of  Washington. 

7  U.  S.  Geological  Survey. 

8  Guest  Investigator;  from  University  of 
Alberta. 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


235 


Fig.  19.     Locations  of  rock  samples  from  Japan  on  which  radioactive  age  determinations  were 
made  by  Professor  I.  Hayase. 


in  the  discussion  stages  for  the  last  two 
years.  As  a  result,  Dr.  I.  Hayase  came  to 
DTM  as  a  Carnegie  Guest  Investigator. 
While  at  the  Department  he  became 
familiar  with  the  techniques  of  measuring 
mineral  ages  by  analyzing  several  samples 
collected  in  Japan.  Modern  geochrono- 
logical  methods  are  of  major  importance 
in  Japan,  where  a  combination  of  over- 
lapping regional  metamorphic  and  vol- 
canic phenomena  and  numerous  faults, 
both  vertical  and  thrust,  make  the 
geological  structure  difficult  to  under- 
stand at  best.  In  addition,  there  are 
numerous  isolated  igneous  bodies  whose 
age  relationships  can  be  found  only  by 
age  measurements.  The  geological  history 
is  so  complex  that  too  simple  an  interpre- 
tation of  these  measurements  could  result 
from  using  a  single  mineral  and  a  single 
decay  system. 

Figure  19  shows  the  locations  from 
which  the  samples  were  collected.  Table  1 
gives  the  ages  measured  by  Dr.  Hayase 


while  at  the  Department.  Several  obser- 
vations may  be  made  about  the  data. 
First,  there  are  no  contradictions  between 
the  ages  and  known  geological  informa- 
tion about  them.  Second,  discordancies  of 
Rb-Sr  and  K-Ar  ages  are  common  enough 
to  indicate  the  complexity  of  the  geo- 
logical history,  already  suggested.  Third, 
the  samples  at  Miyazaki  show  a  difference 
between  biotite  and  muscovite  ages  very 
similar  to  differences  found  on  older 
pegmatites  in  other  parts  of  the  world. 
Finally,  the  measurements  have  only 
increased  the  need  for  further  analyses 
on  different  minerals  by  various  decay 
systems  before  they  themselves  can  be 
completely  understood. 

The  second  part  of  the  exchange 
between  the  two  departments  is  now  in 
progress.  A  Carnegie  staff  member,  L.  T. 
Aldrich,  is  now  in  Kyoto  as  a  visiting 
professor  in  Dr.  Hayase 's  Institute, 
advising  in  the  establishment  of  a  com- 
plete laboratory  for  the  measurement  of 


236 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


TABLE  1.     Ages  Measured  on  Micas  Collected  in  Japan 


Sample     T 
Tyr           Location 

Rock 

Mineral 

Ages, 
of 

millions 
years 

Geological  Information 

Rb-Sr 

K-Ar 

1 

Ishikawayama 

Pegmatite 

Biotite 

115 

80 

? 

2 

Miyazaki- 

Pegmatite 

Muscovite 

100 

70 

? 

Mikawa 

Biotite 

40 

40 

3 

Unoyama 

Pegmatite 

Muscovite 

80 

60 

Post  Permo-Carboniferous 

4 

Miyazu 

Granite 

Biotite 

65 

45 

Post  Permo-Carboniferous 

5 

Kyoto 

Granite 

Biotite 

100 

90 

Post  Permo-Carboniferous 

6 

Nara 

Pegmatite 

Muscovite 

70 

Post  Permo-Carboniferous 

7 

Ningyo  Pass 

Granite 

Biotite 

130 

50 

Pre-Miocene 

8 

Tokushima 

Schist 

Muscovite 

265  ±  50 

Paleozoic? 

mineral  ages.  To  expedite  this  phase  of 
the  work  the  Department  constructed 
and  shipped  to  Japan  mass-spectrometric 
equipment  which  is  now  in  operation  in 
Kyoto.  This  apparatus  will  serve  two 
purposes:  to  enable  the  measurement  of 
ages  to  begin  at  once;  and  to  serve  as  a 
model  for  similar  equipment  to  be  built 
in  Japan.  The  laboratory  in  Kyoto  has 
been  and  will  be  a  meeting  place  for  those 
in  Japan  interested  in  geochronology,  and 
other  laboratories  now  about  to  make 
similar  measurements  will  also  be  visited. 
The  Department  looks  forward  to  many 
years  of  collaboration  with  Japanese 
scientists  in  this  field,  primarily  because 
of  the  personal  contacts  possible  only  on 
an  extended  visit  such  as  this. 

Thailand.  Another  group  of  Asian 
samples  measured  this  year  resulted  from 
the  collaboration  with  Dr.  Saman  Bura- 
vas  of  the  Royal  Department  of  Mines  of 
Thailand.  Two  samples  of  particular 
interest  to  that  Department,  which  were 
also  of  some  general  interest  because  of 
their  correlation  with  Paleozoic  sedi- 
mentary formations,  were  separated,  and 


the  appropriate  minerals  were  analyzed. 
The  results  are  given  in  table  2.  The 
Phuket  granite  was  thought  on  very 
loose  geological  grounds  to  be  Cretaceous, 
and  the  age  measurements  confirm  the 
assignment.  The  Tak  granite  is  known  to 
have  metamorphosed  Permian  limestone 
in  the  area  from  which  it  was  collected, 
and  this  fact  places  a  lower  limit  on  the 
age  of  the  formation.  The  granite  had 
been  assigned  a  post-Permian,  pre-Tri- 
assic  age,  and  the  data  agree  with  that  of 
the  Holmes J  "B"  time  scale  for  that 
boundary.  It  is  seen,  even  for  these  young 
samples,  that  the  ages  of  the  different 
minerals  of  the  same  rock  exhibit  con- 
siderable discrepancy. 

Brazil.  Further  international  ties  were 
developed  this  year  through  the  visit  to 
the  Department  of  Dr.  Lenz  Cesar  from 
the  Universidad  do  Ceara,  Brazil.  While 
becoming  acquainted  with  the  techniques 
of  age  determination,  Dr.  Cesar  measured 
several  mineral  ages  on  a  sample  of 
granite  from  near  Sera  da  Moeda, 
Quadrilatero  Ferrifero,  Brazil.  On  the 
basis  of  K-Ar  ages  of  biotite,  this  granite 


TABLE 

2.     Data  on  Minerals  from  Granites  from  Thailand 

Sample 

Ages,  millions  of  years 
Lat.-Long.                     Mineral 

K-Ar                       Rb-Sr 

Tak  granite 
Phuket  (tin)  granite 

17°5'-99°5'                   Muscovite                    60  ±    5 

Feldspar                          .... 
8°10'-98°20'                 Biotite                        215  ±  10 

85  db    5 
150  ±  25 
225  ±  10 

DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


237 


was  assigned  by  Herz,  Hurley,  Pinson, 
and  Fairbairn9  to  a  group  of  "late" 
granites  having  intrusive  ages  of  about 
500  m.y.  Table  3  shows  the  mineral  ages 
determined  here  by  several  methods  for 
this  granite.  It  is  clear  that  this  granite  is 
at  least  1050  m.y.  old;  the  younger  ages 


TABLE  3.     Mineral  Ages  from  Granite  from 
Moeda  Road,  Quadrilatero  Ferrifero,  Brazil 


Mineral 

Ages,  millions  of  years 
K-Ar                      Rb-Sr 

Biotite 

Feldspar 

640                           630 
470                        1050 

undoubtedly  represent  the  effect  of  a 
later  metamorphic  event.  The  actual 
occurrence  of  500-m.y.  intrusive  rocks  in 
this  area  as  given  by  Herz  et  al.  must  be 
considered  unproved  until  other  ages 
than  the  biotite  K-Ar  ones  become 
available. 

Excess  Argon  in  Pyroxenes 

In  Year  Book  60  we  reported  on  the 
promising  results  obtained  in  the  use  of 
amphiboles  and  pyroxenes  for  K-Ar 
dating.  Subsequent  work  supports  the 
value  of  amphiboles  such  as  hornblende. 
However,  several  pyroxenes  have  been 
found  that  contain  large  excesses  of 
radiogenic  argon,  thus  severely  limiting 
their  general  usefulness  for  K-Ar  dating. 

Damon  and  Kulp10  demonstrated  the 
existence  of  excess  or  initial  radiogenic 
argon  and  helium  in  the  minerals  beryl, 
cordierite,  and  tourmaline.  These  min- 
erals have  a  ring  structure  with  large 
channels,  which  appears  to  account  for 
the  incorporation  of  the  initial  radiogenic 

9  N.  Herz,  P.  M.  Hurley,  W.  H.  Pinson,  and 
H.  W.  Fairbairn,  Age  measurements  from  part 
of  the  Brazilian  Shield,  Bull.  Geol.  Soc.  Am.,  72, 
1111-1115,  1961. 

10  P.  E.  Damon  and  J.  L.  Kulp,  Excess  helium 
and  argon  in  beryl  and  other  minerals,  Am. 
Mineralogist,  48,  433-459,  1958. 


gas.  In  the  same  paper  Damon  and  Kulp 
postulated  that  amphiboles  might  also 
contain  excess  gas  due  to  the  partial 
vacancy  in  the  alkali-cation  position. 
Hart,11  investigating  amphiboles  and 
pyroxenes  for  use  for  K-Ar  dating,  found 
no  evidence  for  excess  argon  in  horn- 
blendes. At  that  time  pyroxenes  were 
considered,  a  priori,  to  be  unlikely  hosts 
for  excess  argon,  as  they  do  not  have 
structural  vacancies  or  holes  of  the  size 
found  in  the  amphiboles  and  ring  silicates. 
Recent  data,  shown  in  table  4,  clearly 
demonstrate  the  presence  of  excess  radio- 
genic argon  in  pyroxenes.  These  purified 
mineral  samples  were  generously  made 
available  by  R.  T.  Dodd  of  Princeton 
University.  The  Bear  Mountain  locality 
is  well  dated  at  about  1000-1 100  m.y.  by 
concordant  U-Pb  ages  on  zircons  from 


TABLE   4.     K-Ar   Ages   of   Hornblendes   and 

Pyroxenes    from    Gneisses    at    Bear   Mountain, 

New  York 


Ages,  mil- 

^r40* 

lions  of 

Mineral 

K,  % 

in  10~5  cc 

STP/g 

years 
K-Ar 

Hornblende 

1.40 

7.39 

1,000 

Hornblende 

1.30 

5.96 

900 

Hypersthene 

0.102 

0.937 

1,500 

Diopside 

0.00863 

0.969 

10,400 

basement  gneisses  and  intrusive  granites 
(Tilton,  Wetherill,  Davis,  and  Bass12). 
The  two  hornblende  ages  are  somewhat 
lower  but  in  essential  agreement  with  the 
zircon  ages.  Both  pyroxene  ages  are 
clearly  too  high.  The  only  reasonable 
explanation  for  these  ages  is  incorporation 
of  radiogenic   argon  into   the   pyroxene 

11  S.  R.  Hart,  The  use  of  hornblendes  and 
pyroxenes  for  K-Ar  dating,  ./.  Geophys.  Res.,  66, 
2995-3001,  1961. 

12  G.  R.  Tilton,  G.  W.  Wetherill,  G.  L.  Davis, 
and  M.  N.  Bass,  1000-million -year-old  minerals 
from  the  eastern  United  States  and  Canada, 
J.  Geophys.  Res.,  65,  4173-4179,  1960. 


238 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


during  either  initial  crystallization  or  a 
later  recrystallization.  As  the  structural 
sites  in  pyroxene  are  not  large  enough  to 
admit  argon,  it  is  suggested  that  the 
excess  argon  is  held  in  crystal  imperfec- 
tions such  as  dislocations  and  defects.  If 
this  explanation  is  correct,  excess  gases 
might  be  expected  in  many  other  min- 
erals, as  most  natural  minerals  contain 
abundant  defect  structures.  As  yet  there 
have  been  no  recognized  occurrences  of 
excess  argon  in  micas,  feldspars,  or 
hornblende.  This  should  be  considered  a 
possibility  in  very  young  samples,  espe- 
cially those  from  deep-seated  environ- 
ments where  appreciable  argon  pressures 
can  exist. 

Study  of  Mineral  Ages  in  a  Contact 
Metamorphic  Zone 

Work  has  been  continuing  on  the 
contact  zone  study  described  in  Year  Book 
60.  Additional  samples  have  been  col- 
lected at  distances  from  2400  to  22,000 
feet.  As  shown  in  figure  20,  the  samples 
at    the    greatest    distances    have    been 


virtually  unaffected  by  the  contact  heat- 
ing. A  complete  transition  has  been 
traced  from  severely  affected  ages  adja- 
cent to  the  contact  to  unaffected  ages  at 
22,000  feet.  A  study  of  the  mineral 
assemblages  in  this  zone  shows  only  one 
change  that  can  be  related  to  the  contact 
heating — a  change  in  the  structure  of  the 
K-feldspar  from  monoclinic  near  the 
contact  to  triclinic  at  distances  greater 
than  1000  feet.  It  occurs  in  the  same 
interval  where  the  biotite  ages  are 
changing  most  rapidly.  Interestingly,  the 
feldspars  retain  appreciable  argon  in  this 
zone  where  the  structure  has  changed. 
This  fact  probably  reflects  the  rather 
minor  nature  of  the  disordering  that 
accompanies  the  feldspar  transition  from 
triclinic  to  monoclinic.  On  the  basis  of 
laboratory  studies,  this  transition  is 
believed  to  occur  at  about  500°C.  It  is 
difficult  to  reconcile  this  temperature 
with  that  obtained  from  consideration  of 
any  reasonable  model  of  heat  flow  from 
this  intrusive,  which  suggests  tempera- 
tures at  least  100°C  lower. 

The  observed  biotite  age  curves  can  be 


c    o 

S  </> 
o    c 

Cl    O 

Q.  1= 

<    — 

E 


1400- 


1200 


d>     o     l00° 
U>    CD 
O     ^ 


800 
600 
400 

200 

Age  of 
intrusive - 
0 


Biotite  K-Ar 


10  100  1000  10,000  30,000 

Distance  from  contact  (feet,  log  scale) 

Fig.  20.  Variation  of  ages  of  minerals  as  a  function  of  distance  from  an  intrusive  contact.  Note 
the  variability  of  response  to  the  intrusive  event  between  minerals  and  between  the  methods  of 
measuring  the  age. 


DEPARTMENT    OF    TERRESTRIAL    MAGNETISM 


239 


closely  fitted  by  a  model  in  which  volume 
diffusion  of  argon  and  strontium  is 
assumed  to  be  dependent  on  a  tempera- 
ture-time function  obtained  from  simple 
heat-flow  calculations.  The  model  shows 
that  the  temperature  dependence  of  the 


diffusion  constants  of  argon  and  stron- 
tium is  essentially  the  same  and  that 
the  difference  in  the  K-Ar  and  Rb-Sr  age 
curves  is  due  to  a  constant  difference  of 
about  a  factor  of  4  in  the  diffusion 
constants  of  argon  and  strontium. 


THEORETICAL  AND   STATISTICAL   GEOPHYSICS 

S.  E.  Forbush 


Morphology  and  Temporal  Variations  of 

the  Intensity  of  Charged  Particles  in  the 

Van  Allen  Trapped- Radiation  Belt13 

The  results  of  a  preliminary  investiga- 
tion of  the  temporal  and  spatial  variations 
of  intensity  in  the  Van  Allen  trapped- 
radiation  belt  were  described  in  Year  Book 
60.  This  investigation  was  continued, 
utilizing  all  the  data  available  for  transits 
of  Explorer  VII  through  the  radiation 
belt  from  October  1959  through  Decem- 
ber 1960. 

The  purpose  of  the  investigation  was  to 
determine  how  the  temporal  changes  of 
intensity  and  morphology  in  the  trapped- 
radiation  belt  are  related  to  geomagnetic 
variations.  An  objective  of  particular 
interest  was  to  determine  whether  tem- 
poral variations  of  intensity  in  the  belt 
might  be  responsible  for  variations  in  the 
geomagnetic  field  of  the  equatorial  ring 
current  responsible  for  geomagnetic 
storm-time  changes.  Trapped  charged 
particles  spiral  around  geomagnetic  lines 
of  force  and  oscillate  between  mirror 
points  in  the  two  hemispheres.  In  addi- 
tion, such  particles  undergo  a  longitudinal 
drift  around  the  earth:  westward  for 
positively  and  eastward  for  negatively 
charged  particles.  This  longitudinal  drift 
current  seems  the  likely  source  of  the 
geomagnetic  storm-time  changes  that 
have  previously  been  ascribed  to  an 
equatorial    ring    current.     The    present 

13  An  investigation  carried  out  jointly  with  D. 
Venkatesan  and  G.  Pizzella  of  the  Department 
of  Physics  and  Astronomy  of  the  State  Univer- 
sity of  Iowa. 


investigation  is  based  on  counting  rates 
from  a  Geiger  counter  which  are  most 
likely  due  to  electrons  with  energies  above 
about  1.5  Mev. 

From  observations  the  intensity  in  the 
belt  is  determined  as  a  function  of 
position  and  time.  An  investigation  of 
temporal  variations  of  intensity  as  a 
function  of  three  positional  coordinates 
would  be  practically  impossible.  Fortu- 
nately, it  has  been  shown14  that,  for  the 
time-equilibrium  state,  a  parameter  L 
may  be  defined  as  a  function  of  the 
integral  invariant  /  and  scalar  magnetic 
field  B  such  that,  everywhere  on  the  shell 
defined  by  the  motion  of  a  trapped 
particle  in  the  earth's  actual  field,  L 
closely  approximates  the  equatorial  radius 
of  the  shell.  This  means  that,  along  a 
given  shell,  L  =  constant,  the  intensity 
depends  on  the  scalar  magnetic  field  B 
and  the  distribution  of  mirror  points, 
which  in  turn  depends  on  the  distribution 
of  pitch  angles  at  the  equator.  L  is 
measured  from  the  earth's  center  in  units 
of  the  earth's  radius. 

The  orbit  of  Explorer  VII  was  such 
that,  for  a  given  L,  B  varied  systemat- 
ically with  period  of  about  100  days  and 
opposite  phase  in  the  northern  and 
southern  hemispheres.  Utilizing  the  ob- 
served counting  rates,  for  a  given  value 
of  L,  for  consecutive  passes  in  the  two 
hemispheres,  and  the  corresponding  val- 
ues of  B,  it  was  possible  to  determine 
empirically  the  dependence  of  intensity 
on  B.  This  dependence  was  then  used  to 

14  C.  E.  Mcllwain,  J.  Geophys.  Res.,  66, 
3681-3691,  1961. 


240 


CARNEGIE      INSTITUTION      OF      WASHINGTON 


1  'correct"  the  observed  intensities  to  a 
constant  B  for  each  of  several  values  of 
L  between  2.5  and  4.7.  The  corrected 
intensities  were  then  used  to  investigate 
temporal  variations  and  morphology  to 
determine  how  changes  in  them  were 
related  to  changes  in  the  equatorial 
geomagnetic  field  of  the  ring  current  U, 
which  were  derived  from  geomagnetic 
data  from  four  equatorial  magnetic 
observatories  about  6  hours  apart  in 
longitude. 

The  results  of  the  investigation  showed 
that,  for  a  given  L,  the  correlation 
between  the  "corrected"  counting  rates 
from  consecutive  passes  (50  minutes 
apart)  in  the  northern  and  southern 
hemispheres  was  decidedly  higher  than 
that  for  the  observed  counting  rates.  This 
indicated  not  only  the  necessity  for  but 
also  the  validity  of  the  correction  for  B. 

The  value  of  L  at  which  the  maximum 
intensity  occurred  (Xmax)  showed  a 
marked  tendency  to  decrease  with  in- 
creasing U;  that  is,  the  maximum  inten- 
sity in  the  belt  tended  to  occur  closer  to 
the  earth  during  magnetic  storms.  On  the 
average,  L,  in  units  of  the  earth's  radius 
measured  from  the  earth's  center,  de- 
creased from  4.0  to  3.0  for  an  increase  of 
about  100  gammas  in  the  southward 
geomagnetic  field  of  the  equatorial  ring 
current. 

For  all  the  transits  through  the  belt 
over  Australia  the  values  of  B  were  at 
least  as  great  as  those  that  (for  corre- 
sponding values  of  L)  occur  at  sea  level 
south  of  Africa.  Thus  all  the  trapped 
particles  measured  over  Australia  have 
mirror  points  that  would  not  be  above 
sea  level  over  the  region  south  of  Africa 
where  they  would  consequently  be  com- 
pletely absorbed.  This  means  that  the 
intensity  observed  over  Australia  is 
maintained  by  replenishment  of  the 
particles  absorbed  in  the  region  south  of 
Africa.  Moreover,  the  replenishment  must 
take  place  within  the  time  taken  for 
electrons  to  drift  longitudinally  eastward 
from  the  longitude  of  south  Africa  to  that 
of  Australia,  that  is,  within  a  period  of 
the  order  of  an  hour  or  less.  This  replen- 


ishment may  be  due  to  some  mechanism 
causing  a  migration  of  mirror  points  to 
lower  values  of  B. 

For  values  of  L  >  Lmax  the  intensity 
shows  a  weak  tendency  to  decrease  during 
magnetic  storms,  whereas  for  L  <  Lmax 
the  opposite  tendency  occurs.  No  evi- 
dence was  found  to  indicate  that  any 
important  contribution  to  the  ring  cur- 
rent could  come  from  the  longitudinal 
drift  of  the  measured  electron  intensity 
(above  1.5  Mev).  Thus  it  seems  likely 
that  the  major  contribution  to  the  ring 
current  must  arise  either  from  electrons 
of  energy  less  than  1.5  Mev  (the  threshold 
value  for  the  counters  used)  or  from 
protons  of  energy  less  than  18  Mev,  the 
threshold  value  of  the  counter  for  protons. 

For  values  of  L  ^  4.1  earth  radii,  vari- 
ations of  a  factor  of  10  or  more  in  counting 
rate  occurred  on  the  average  every  15  or 
20  days.  For  L  ^  2.5  these  shorter-period 
fluctuations  were  practically  absent,  al- 
though there  were  at  L  =  2.5  two  or 
three  increases  of  intensity  by  a  factor  of 
about  100  which  appeared  to  decay 
slowly  over  a  period  of  two  or  three 
months.  These  longer-period  variations 
diminished  rapidly  with  decreasing  L 
until  at  L  =  1.5  they  were  barely 
discernible. 

Geomagnetic  Equatorial  Ring  Current 

Measures  and  Latitude  of  Auroral 

Currents 

For  the  investigation  described  in  the 
preceding  section  3-hour  values  of  the 
geomagnetic  equatorial  ring-current  field 
(ERC  or  U)  were  derived  for  the  period 
from  October  1959  through  December 
1960.  In  addition,  3-hour  ERC  values 
were  maintained  as  near  to  date  as  receipt 
of  the  necessary  observatory  magnetic 
data  permits,  which  now  is  through 
December  1961.  These  values  were  de- 
rived to  facilitate  cooperation  in  the 
interpretation  of  magnetic  measurements, 
which  ultimately  will  locate  the  equa- 
torial ring  current,  now  being  obtained 
by  others  from  satellite-borne  magnetom- 
eters. The  ERC  values  were  derived  after 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM  241 

the  method  of  Kertz,15  using  data  from  type  is  to  be  constructed  and  tested.  If  it 
magnetic  observatories  about  equally  proves  satisfactory,  timers  will  be  pro- 
spaced  in  longitude  and  nearest  the  vided  for  use  with  Mercer  battery- 
equator,  operated  chronometers.  Thus,  reliable 
The  strong  auroral  zone  current  system  time  marks  would  be  available  at  the 
and  its  southward  shift  during  magnetic  field  stations — a  matter  of  importance, 
storms  are  likely  to  be  closely  associated  since  the  interpretation  of  any  anomalies 
with  particles  coming  from  the  outer  Van  may  depend  to  considerable  degree  upon 
Allen  belt.  To  permit  closer  investigation  the  relative  phases  of  the  observed  com- 
of  this  connection  the  latitude  of  the  ponents  of  the  night-time  sudden  corn- 
auroral  currents  has  been  derived  for  a  mencements  which  will  provide  the 
number  of  magnetic  storms  using  data  uniform  external  inducting  field.  More- 
from  several  magnetic  observatories  in  over,  the  timers  will  provide  automatic 
Scandinavia,  which  is  about  the  only  control  of  daily  calibrations, 
longitude  sector  where  observatories  near  In  view  of  troubles  experienced  with 
the  same  longitude  are  close  enough  the  Z  system  of  the  Askania  variograph 
together  to  make  such  a  determination  during  the  International  Geophysical 
feasible.  For  these  determinations  correc-  Year  the  equations  for  the  system  were 
tions  were  made  for  the  ERC  field  at  the  derived  and  practical  experiments  out- 
auroral  zone  by  means  of  the  ERC  values  lined  for  determining  the  cause  for  any 
that  had  been  derived.  These  values  will  anomalous  drift  in  Z  base  lines  after 
be  applied  in  studies  involving  measure-  temperature  compensation  in  Peru.  For 
ments  in  the  outer  Van  Allen  belt  and  some  of  the  variometers  previously  used 
also  in  connection  with  an  investigation  in  Peru  during  the  International  Geo- 
of  storm- time  variations  in  cosmic-ray  physical  Year  the  analysis  showed  that 
intensity  begun  in  collaboration  with  most  of  the  troubles  were  probably  due 
Professor  Alfven  of  the  Royal  Institute  to  a  mechanical  unbalance  in  the  Z 
of  Technology,  Stockholm,  Sweden.  system.  Tests  were  outlined  for  deter- 
mining all  the  important  parameters 
Conductivity  Anomaly  Program  for  Peru  required  to  specify  the  state  of  the  system. 

An  observational  program  was  outlined 
to  provide  magnetic  data  for  determining  Cosmic- Ray  Program 

whether  crustal  conductivity  anomalies 

(Bartels-Schmucker  anomalies)  may  exist         Observations    and    reductions    of   data. 

in  Peru.  The  program  will  be  carried  out  Cosmic- ray    ionization    chambers    were 

by  the  Instituto  Geofisico  del  Peru  with  operated  throughout  the  report  year  at 

about  eight  Askania  variographs  loaned  Huancayo,  Peru,  and  at  Fredericksburg, 

by  the  U.  S.  Coast  and  Geodetic  Survey.  Virginia.  Scalings  and  reduction  of  rec- 

The  existence  or  absence  of  such  anom-  ords  have  been  maintained  current  for 

alies  provides  one  means  of  exploring  the  Fredericksburg  and  Huancayo. 
earth's  crust  at  depths  of  the  order  of         Cooperation  in  operation  of  cosmic-ray 

100  km.  meters.  Grateful  appreciation  is  expressed 

A  preliminary  model  to  provide  time  to  the  U.  S.  Coast  and  Geodetic  Survey 

marks  at  10-minute  intervals  on  the  film  and  the  staff  of  its  magnetic  observatory 

was  constructed  and  tested.  One  proto-  at  Fredericksburg,  Virginia,  for  efficient 

operation  of  the  meters  during  the  past 

15  Walter   Kertz,    Ein   neues   Mass   fur   die  report  year,  and  to  the  Government  of 

Feldstarke    des    erdmagnetischen    aquatorialen  peru  anc[   the   Director  and  staff  of  the 

R^fstroms    Abhandl.  Akad.   Wiss    Gottingen,  Instituto  Geofisico  del  Peru  for  making 
Math.-thysik.  At.,  Beitrage  zum  inter nationalen  .  ° 

geophysikalischen    Johr,    Heft    2,    Gottingen  cosmic-ray  records  from  Huancayo  avail- 

Vandenhoeck  8,  Ruprecht,  1958.  able. 


242 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


LABORATORY  PHYSICS 


NUCLEAR  PHYSICS 

L.  Brown,16  N.  P.  Heydenburg,  H.  Rudin,17  and 
G.  M.  Temmer 

Polarized  Ion  Source 

The  desirability  of  having  available 
polarized  beams  of  protons  and  deuterons 
for  the  study  of  nuclear  interactions  was 
discussed  in  some  detail  in  Year  Book  57. 
It  was  pointed  out  at  that  time  that  the 
Department's  pressurized  Van  de  Graaff 
generator,  with  its  large  high-voltage 
terminal  equipped  with  6  kw  of  a-c  power, 
was  ideally  suited  for  the  installation  of  a 
complex  atomic  beam  apparatus  for  the 
production  of  polarized  particles.  It  was 
announced  in  last  year's  report  that 
arrangements  had  been  made  for  a 
cooperative  endeavor  between  the  De- 
partment and  the  Basel  group  for  the 
installation  of  their  atomic  beam  appa- 
ratus in  the  DTM  accelerator.  This  was 
to  be  a  copy  of  their  first  model,  which 
successfully  demonstrated  the  production 
of  polarized  deuterons  by  the  atomic 
beam  method  for  the  first  time. 

The  Basel  polarized  ion  source  was 
completed  by  the  summer  of  1961  and 
was  shipped  to  our  laboratory  in  August. 
The  installation  of  the  source  in  the  DTM 
accelerator  was  begun  in  September  and 
was  essentially  completed  and  ready  for 
testing  by  the  end  of  January  1962.  The 
Basel  polarized  source  has  been  described 
elsewhere.18  Basically  it  consists  of  (1)  a 
discharge  tube  for  the  production  of 
atomic  hydrogen,  (2)  a  set  of  diaphragms 
and  fast  pumps  for  defining  the  atomic 
beam,  (3)  a  quadrupole  magnet  for  the 
selection  and  focusing  of  atoms  having 
the  desired  orientation  in  space,  (4)  an 
ionizer  for  the  atomic  beam,  and  (5)  an 

16  Carnegie  Institution  Fellow. 

17  Carnegie  Institution  Fellow;  from  Univer- 
sity of  Basel. 

18  H.  Rudin,  H.  R.  Striebel,  E.  Baumgartner, 
L.  Brown,  and  P.  Huber,  Helv.  Phys.  Acta,  34, 
58,  1961. 


arrangement  for  preaccelerating  and 
focusing  the  ionized  atoms.  After  the 
source  had  been  assembled  in  the  acceler- 
ator terminal,  a  system  was  constructed 
to  control  from  outside  the  steel  pressure 
tank  the  various  vacuum  pumps,  vacuum 
valves,  and  electronic  equipment  associ- 
ated with  the  source  inside  the  high- 
voltage  terminal.  A  cooling  arrangement 
was  also  necessary  for  cooling  the 
diffusion  pumps,  baffles,  and  ionizer  in 
the  high- voltage  terminal.  An  external 
heat  exchanger,  and  a  circulating  system 
for  Freon  11,  were  installed  for  this 
purpose. 

After  a  number  of  unpleasant  failures 
of  the  cooling  system,  a  deuteron  beam 
of  about  10~9  ampere,  after  acceleration, 
was  obtained.  Although  somewhat  weaker 
than  the  beam  obtained  during  tests  at 
Basel,  it  was  sufficient  to  proceed  with  a 
test  for  polarization.  The  polarization  was 
determined  from  the  angular  asymmetry 
of  the  disintegration  protons  in  the 
reaction  He3(d,  p)He4,  measured  some- 
what above  the  broad  resonance  occurring 
at  470  kev.  A  small  reaction  chamber  was 
constructed  for  this  test  consisting  of  a 
gas  cell  for  the  He3  and  two  Csl  particle 
detectors  placed  at  20°  and  90°  with 
respect  to  the  direction  of  the  incident 
beam.  The  polarization  for  a  deuteron 
with  spin  1  cannot  be  characterized  by 
vector  polarization  alone,  as  it  can  for  a 
proton  with  spin  3^2,  but  the  polarization 
tensor  must  be  considered  as  well.  For  the 
above  reaction,  the  ratio  of  the  cross 
sections  at  20°  and  90°  is  given  by 

o-(20°)       1  -  MPZJ&  cos2  20°  -  1) 


A  = 


(7(90°)  "  1  -  %P„{Z  cos2  90°  -  1) 


where  Pzz  is  a  component  of  the  polariza- 
tion tensor  and  is  predicted  to  have  the 
value  Pzz  =  —  3^  for  our  polarized  beam. 
The  experimental  value  of  A  was  found 
to  be  1.216  ±  0.048.  Using  this  equation, 
Pzz  was  found  to  be  -0.301  ±  0.063,  in 
good  agreement  with  the  expected  value. 


DEPARTMENT   OF   TERRESTRIAL    MAGNETISM  243 

We  plan  to  utilize  the  polarized  deuteron  23.0  Mev  level  4+,  and  the  25.0  Mev 

beam  in  the  study  of  a  number  of  nuclear  level  2+. 

reactions,    among    them    the    following:  Angular  distributions  and  an  excitation 

d(d,  p)t,  He4(d,  d)He4,  and  Li6(d,  a)a.  curve  were  observed  for  the  Li6(p,  a)He3 

reaction    (Han    and    Heydenburg).    No 

Summary  of  Cooperative  Program  resonance  structure  is  seen  in  the  excita- 

with  Florida  State  University  *"»  curve  for  Proton  bombarding  ener- 
gies  from   3    to    12    Mev.    A   complete 

The  arrangement  for  the  participation  analysis  of  the  angular  distribution  curves 
by  Heydenburg  and  Temmer  in  the  has  not  yet  been  made,  but  strong 
nuclear  research  program  at  Florida  forward  and  backward  peaking  of  the  He3 
State  University  has  continued  during  the  cross  section  suggests  that  the  reaction 
past  year,  Heydenburg  spending  part  mechanism  involves  both  a  deuteron 
time  at  DTM  on  the  polarized  ion  source  pickup  by  the  incident  proton  and  a 
project.  The  experimental  work  on  the  direct  interaction  of  the  proton  with  a 
Li7(p,  a) a  reaction  (Han  and  Heyden-  triton  in  the  nucleus.  This  reaction  has 
burg) ,  reported  in  Year  Book  60,  has  been  special  interest  for  cluster-model  con- 
completed,  and  the  angular  distributions  siderations,  since  Li6  is  one  of  the  best- 
have  been  analyzed  in  terms  of  the  understood  nuclei  from  that  point  of 
Legendre  polynomials.  In  addition  to  the  view. 

resonances   at   3.0    Mev   and    5.6   Mev  The  Be9(p,  a)Li6  reaction  (Blieden  and 

previously  reported,  there  is  evidence  for  Temmer)  is  a  striking  example  of  a  direct 

another  at  6.5  Mev  and  a  weak  one  at  interaction    process.    The    yield    curve 

9.0  Mev.   These  four  resonances  corre-  varies  smoothly  with  proton  bombarding 

spond  to  energy  levels  in  the  compound  energy,  showing  no  resonance  structure, 

nucleus  Be8  at  19.9,  22.1,  23.0,  and  25.0  The  angular  distributions  have  identical 

Mev.  shapes  between  6  and  11.5  Mev  proton 

The  Li6(<i,   a)  a  reaction  can  be  em-  energy  for  alphas  to  the  ground  state  of 

ployed  to  study  the  excitation  region  of  Li6,  and  between  9  and   11.5  Mev  for 

Be8  above  22.3  Mev.  The  Be8  levels  at  alphas    to    the    first-excited    state.    The 

22.1  and  23.0  Mev  cause  a  sharp  rise  in  curves  are  being  analyzed  in  terms  of 

the   yield    of   alphas   and   an   apparent  Satchler's  distorted  wave  analysis,  and 

resonance  peak  at  a  bombarding  energy  preliminary  results  show  good  agreement, 

of  0.4  Mev,  as  was  shown  in  earlier  work  In  contrast  to  the  above  reaction  the 

at  DTM.  We  have  extended  this  early  F19(p,  a)016  reaction  (Warsh  and  Tem- 

work  to  bombarding  energies  up  to  12  mer)  shows  a  rather  anomalous  behavior, 

Mev.  The  Be8  level  at  25.0  Mev  causes  a  having  angular  distribution  curves  that 

prominent  resonance  peak  in  the  excita-  cannot    be    fitted    with    the    Satchler 

tion  curve  at  a  deuteron  energy  of  3.85  analysis.    It    appears    that    the    heavy- 

Mev  (as  compared  with  the  very  weak  particle  stripping  process  may  also  have 

resonance  in  the  Li7(p,  a) a  reaction  occur-  to  be  considered  in  analyzing  these  data, 

ring  at  the  same  excitation  in  Be8).  From  as  indicated  by  the  large  alpha  yields  in 

this  evidence  and  other  considerations  it  the  backward  direction, 

is    believed    that    these    four    relatively  Two  new  energy  levels  in  Ne20  were 

narrow  states  in  Be8  can  be  described  in  found  from  a  study  of  the  alpha  groups 

terms  of  Li6  +  d  configurations  rather  from  the  Na23(p,  o:)Ne20  reaction;  they 

than  a  +  cl  configurations.  M.  Nomoto  are  particularly  interesting  in  that  they 

has  been  able  to  make  spin  and  parity  lend  further  support  to  the  surprisingly 

assignments    for    these    levels    from    an  regular  rotational  band  structure  of  Ne20. 

analysis  of  our  data.  These  are  as  follows :  The  excitation  curves  and  angular  dis- 

19.9  Mev  level  2+,  22.1  Mev  level  2-f,  tributions    for    this    reaction    and    the 


244 


CAENEGIE     INSTITUTION      OF      WASHINGTON 


A\27(p,  a)Mg24  reaction  are  very  complex. 

A  yield  curve  at  a  90°  angle  has  been 
obtained  for  the  reaction  N15(p,  o;)C12 
(Roy,  Adams,  and  Temmer)  showing 
considerable  resonance  structure.  This 
reaction  has  special  interest  in  connection 
with  the  level  structure  of  O16,  which  has 
closed  proton  and  neutron  shells  and  for 
which  several  detailed  theoretical  pre- 
dictions have  been  made.  Angular  distri- 
butions will  be  measured  for  this  reaction. 

Temmer  has  considered  the  possibility 
of  a  so-called  resonant  transfer  process, 
involving  no  energy  loss,  of  one  or  more 
nucleons  between  the  target  and  projec- 
tile in  a  nuclear  encounter.  This  is  based 
on  an  analogy  to  a  recently  discovered 
atomic  process  demonstrating  the  repeat- 
ed resonant  transfer  of  an  electron  be- 
tween two  identical  nuclei.  A  nuclear 
example  would  be  the  transfer  of  a  proton 
from  O16  to  N15.  Experiments  attempting 
to  demonstrate  the  existence  of  this 
effect,  which  contributes  to  elastic  scat- 
tering, are  contemplated. 

BIOPHYSICS 

E.  T.  Bolton,19  R.  J.  Britten,  D.  B.  Cowie,  B.  J. 
McCarthy,  J.  E.  Midgley,20  and  R.  B.  Roberts 

Introduction 

The  report  of  the  Biophysics  Section  in 
Year  Book  56  was  the  first  in  which  we 
mentioned  the  particles  (now  called 
ribosomes)  found  in  cells  and  their 
possible  role  in  the  synthesis  of  nucleic 
acids  and  proteins.  Before  that  time  our 
attention  had  been  directed  toward  the 
biosynthesis  of  amino  acids  and  nucleo- 
tides and  later  toward  the  mechanisms  by 
which  those  small  molecules  were  concen- 
trated and  held  in  the  cells.  Five  years 
ago  our  knowledge  of  the  particles  was 
barely  adequate  to  appreciate  that  the 
synthesis  of  ribosomes  and  their  role  in 
protein  synthesis  was  a  promising  field 
for  investigation. 

19  Visiting  investigator  at  universities  in 
Canberra  and  Adelaide,  Australia,  and  Osaka, 
Japan,  July-December,  1961. 

20  Carnegie  Institution  Fellow. 


After  five  years  of  concentration  on 
these  problems  the  broad  outlines  and 
many  of  the  details  have  been  worked 
out.  Year  Book  57  reported  the  first 
separation  of  ribosome  precursors  by 
chromatography  and  by  sedimentation. 
These  techniques  were  markedly  im- 
proved during  the  years  to  provide  the 
detailed  flow  diagrams  in  Year  Book  60. 
Year  Book  58  reported  the  identification 
of  newly  synthesized  protein  still  attached 
to  ribosomes  which  pointed  to  their  role 
in  protein  synthesis.  We  consider  our- 
selves privileged  to  have  been  able  to 
participate  in  the  unusually  rapid  devel- 
opments that  have  occurred  during  these 
five  years. 

At  the  present  moment  it  is  possible  to 
account  quantitatively  for  the  flow  of 
material  as  it  passes  from  the  external 
medium  through  pools  of  low-molecular- 
weight  compounds  through  two  sequen- 
tial macromolecular  precursors  and  finally 
into  the  stable  end  products.  In  addition, 
it  appears  that  one  of  these  precursors, 
while  attached  to  a  ribosome,  can  act  as 
the  template  for  protein  synthesis.  A 
more  complete  discussion  of  our  present 
knowledge  and  the  questions  remaining 
to  be  answered  will  be  found  under 
"Conclusions,"  at  the  end  of  this  section. 
The  details  of  the  experimental  work  are 
given  in  the  body  of  the  report. 

Incorporation  of  RNA  Bases  into  the 
Metabolic  Pool  and  RNA  of  E.  coli 

Studies  of  the  incorporation  of  C14- 
uracil  reported  last  year  showed  direct 
entry  of  radioactivity  into  the  RNA 
without  delay  by  the  large  pool  of  nucleo- 
tides. The  scheme  shown  in  diagram  1 
was  proposed  to  describe  the  kinetics  of 
uracil  incorporation.  P  represents  a  very 
small  pool  or  sequence  of  reaction  steps 
leading  from  uracil  to  a  chemical  form 
suitable  for  incorporation  into  RNA.  S 
represents  a  large  pool  of  compounds  that 
can  exchange  with  some  uracil  compound 
in  P.  The  rate  of  exchange  between  S  and 
P  is  not  fast,  and  equilibrium  between 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


245 


Endogenous 
synthesis 


Exogenous 
uracil 


d 

+>P 


v 


^Incorporation 

c         into  RNA 


5 
Diagram  1 

the  specific  radioactivity  of  P  and  S 
requires  several  minutes,  at  least.  P  then 
effectively  forms  a  bypass,  around  the 
large  pool,  for  the  entry  of  exogenous 
uracil  into  RNA. 

Since  an  alternative  explanation  of  the 
undelayed  entry  of  radioactivity  into 
RNA  has  been  proposed  by  Gros  et  al.  it 
seemed  worth  while  to  examine  the 
kinetics  of  incorporation  of  the  three 
other  RNA  bases.  The  additional  infor- 
mation reported  here,  particularly  the 
fact  that  the  time  constants  and  bypass 
fractions  vary  widely  among  the  four 
bases,  clarifies  the  situation.  The  existence 


of  a  bypass  mechanism  is  amply  demon- 
strated. 

Incorporation  of  cytosine.  The  results 
of  an  experiment  in  which  10-6  M 
cytosine-2-C-14  was  supplied  to  exponen- 
tially growing  cells  are  shown  in  figure  21. 
It  is  clear  that  the  kinetics  of  cytosine 
incorporation  are  qualitatively  similar  to 
the  kinetics  of  uracil  incorporation  re- 
ported last  year.  There  is  an  initial  rapid 
incorporation  into  RNA,  and  this  rate  is 
maintained  while  exogenous  cytosine 
remains.  At  the  end  of  this  first  phase  the 
rate  of  incorporation  into  RNA  abruptly 
falls  by  a  large  factor.  During  the  second 
phase  (after  the  exogenous  cytosine  is 
exhausted)  the  radioactivity  of  the  pool 
falls  slowly.  The  semilog  plot  (fig.  22) 
shows  that  the  radioactivity  of  the  pool 
decreases  in  an  approximately  expo- 
nential fashion.  The  time  constant  (decay 
to  1/e)  is  about  21  minutes — more  than 
twice  that  observed  for  uracil. 

The  results  of  an  experiment  at  a 
higher  concentration  of  cytosine  (10-4  M) 
are  shown  in  figure  23.  Here  again  two 
phases  are  observed.  Initially  C14  from 
cytosine  enters  the  RNA  at  less  than  half 


<u  15,000  - 


O  IQ000  - 


o 
o 
o 

'"O 

o 
DC 


5000  - 


Sample  time -minutes 

Fig.  21.  Incorporation  of  106  M  C14-cytosine  by  E.  coli  ML  30  growing  at  37°C  with  a  generation 
time  of  51  minutes.  Cell  density  0.5  mg  (wet)  per  ml.  Open  circles  represent  radioactivity  of  total 
cell  samples  collected  by  membrane  filtration.  Solid  circles  represent  RNA  radioactivity,  samples 
collected  by  membrane  filtration  after  treatment  with  5  per  cent  TCA. 


246 


CARNEGIE     INSTITUTION      OF     WASHINGTON 


to 


8000  - 


§  4000 

o 

o 


2000  - 


o 
o 
o 

^Z    1000 
o 

£ 


-% 

I0"6M  Cytosine 

i             i             I 

1          1          1 

1 

10    15   20   25   30   35 

Sample  time -minutes 


Fig.  22.  Decay  of  the  radioactivity  of  the 
C14-cytosine  labeled  pool.  Data  obtained  from 
experiment  of  figure  21  by  subtracting  the  RNA 
radioactivity  from  the  whole  cell  radioactivity. 


the  final  rate.  Only  after  a  relatively  long 
period  does  it  achieve  its  final  rate. 

The  curves  shown  in  figures  21,  22,  and 
23  are  precisely  those  to  be  expected  on 
the  basis  of  the  schematic  diagram  if  it  is 
assumed  that  the  time  constant  is  a 
measure  of  the  quantity  of  compounds  in 
the  pool  $  in  relationship  to  the  flow  b. 

The  fraction  of  the  total  flow  bypassing 


the  pool  S  may  be  estimated  from  the 
fraction  of  the  radioactivity  (fig.  21)  that 
has  entered  the  RNA  at  the  time  the 
exogenous  cytosine  is  exhausted.  This 
appears  to  be  about  45  per  cent.  The 
bypass  fraction  of  the  flow  may  also  be 
estimated  from  the  ratio  of  the  initial  to 
ultimate  slopes  in  figure  23.  The  result  is 
consistent  with  the  experiment  at  low 
concentration.  Although  the  shape  of  the 
curve  in  figure  23  suggests  the  presence 
of  components  in  the  pool  of  intermediate 
time  constant,  their  effect  is  not  apparent 
in  figure  22.  If  such  additional  compo- 
nents are  present,  the  time  constant 
quoted  is  a  weighted  average  of  many  and 
the  accuracy  of  the  estimate  of  the  bypass 
is  reduced.  The  calculation  of  the  bypass 
flow  and  time  constants  is  discussed  in 
more  detail  below. 

The  time  constant  estimated  here  is 
probably  influenced  by  the  appearance  of 
C14  from  cytosine  in  uracil  compounds  in 
the  pool  and  RNA.  Table  5  shows  the 
results  of  several  measurements  of  the 
ratio  of  the  amount  of  radioactivity  in 


10,000 


c 

-  200 

F 

"v 

8000 

-   100 

<n 

r 

~5 

-      0 

o 

o 

;^ 

6000 

> 

<  > 

n 

o 

"O 

4000 

< 

7T 

en 

2000 


10'*  M  Cytosine 


20  40  60 

Cell  mass,  sample  times  indicated  (minutes) 


90 


Fig.  23.  Incorporation  of  10-4  M  C14-cytosine  by  E.  coli.  Initial  cell  density  1/3  mg  (wet)  per  ml. 
RNA  radioactivity  (TCA-precipitable)  plotted  against  cell  mass  with  sample  times  indicated.  The 
upper  curve  represents  the  data  at  early  times  with  both  scales  expanded  by  a  factor  of  10.  Thus  the 
slopes  on  the  two  curves  may  be  directly  compared. 


DEPARTMENT   OF   TERRESTRIAL    MAGNETISM 


247 


TABLE  5.     Inter-conversion  of  Cytosine  and  Uracil  Compounds 


Labeled  Supplement 


Competitor 


Ratio*  of  Radioactivity  of 
RNA  Uridylic  to  RNA  Cytidylic 


C14-cytosine  10~4  M 
C14-cytosine  10"4  M 
C14-cytosine  KT6  M 
C14-cytosine  5  X  KT5 
C14-cytosine  5  X  10~5 
C14-uracil  5  X  10"6  M 
C14-uracil  5  X  10"5  M 


M 

M 


C12-uracil  5  X  10"5  M 
C12-cytosine  5  X  10~5  M 


1.4 

1.6f 

1.7 

1.6 

0.6 

1.0 

1.0 


*  After  the  radioactivity  was  completely  incorporated  into  RNA  the  cells  were  treated  with  5  per 
cent  TCA,  washed,  and  hydrolyzed  with  alkali,  and  the  nucleotide  residues  were  separated  by  electro- 
phoresis. 

f  In  this  experiment  samples  were  taken  at  10-minute  intervals.  By  10  minutes  the  ratio  had  already 
reached  1.3,  and  at  20  minutes  it  had  reached  essentially  its  final  value. 


the  uridylic  acid  to  that  in  the  cytidylic 
acid  of  the  RNA.  In  view  of  the  very 
considerable  interconversion  between  cy- 
tosine and  uracil  compounds  it  is  sur- 
prising that  the  effective  time  constants 
of  pool  S,  measured  with  the  C14-uracil 
and  C14- cytosine,  are  so  different.  By 
comparing  lines  4  and  5  with  lines  6  and 
7  of  table  5,  however,  it  appears  that 
C12-uracil  has  an  effect  on  the  conversion 
of  C14-cytosine  whereas  C12-cytosine  does 
not  affect  the  conversion  of  C14-uracil. 
This  lack  of  symmetry  in  the  competition 
experiments  and  the  relatively  greater 
conversion    of    cytosine    compounds    to 


uracil  compounds  indicates  the  com- 
plexity of  the  interconversion  processes. 
Guanine.  Figure  24  shows  the  results 
of  an  experiment  in  which  10~6  M  guanine 
was  supplied  to  exponentially  growing 
cells.  Here  again  there  is  a  rapid  incorpo- 
ration into  RNA  during  the  first  phase 
when  guanine  is  present  externally. 
During  the  second  phase,  after  the 
external  guanine  has  been  exhausted,  the 
radioactivity  of  the  pool  is  relatively 
slowly  transferred  to  RNA.  No  logarith- 
mic plot  of  the  decay  of  the  guanine  pool 
is  presented,  since  the  small  amount  of 
radioactivity  in  the  pool  and  the  scatter 


8000  - 


6000  - 


a 

o 
o 

£  4000 

o 
o 
o 
■o 
o 


2000 


- 

<-Totol  cell 

n  n°      -   nK,      -       "     ° 

n 

" 

« 

♦ 

/                                                          *     * 

• 

- 

f     /*              ^  RNA 

i                    i 

I0"6  M  Guanine 

20 


25 


0  5  10  15 

Sample  time  -minutes 
Fig.  24.     Incorporation  of  10" fi  M  C14-guanine  by  E.  coli.  Cell  density  0.7  mg  (wet)  per  ml. 


248 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


Cell  mass,  sample  times  indicated  (minutes) 

5  10  15  20 


<u  8000  - 


§  6000 

o 

o 


o  4000 
g 

""O 

o 


o:  2000  - 


t               ^ 

i    ■ 

Upper  curve  ( • ) 

Guanine  I0"5M                               > 

/^ 

y 

/ 

X               <* 

X              / 

x        y 

X         y 

_ 

X        s 

X       s 

X     y 

X      s* 

X   y 

X   S 

x  y 
X  y 

Lower  curve, 

J?" 

expanded  ( ° )      * 

I 

>*/                             i 

i 

600 


-  400 


-  200 


Time  in  minutes 

Fig.  25.  Incorporation  of  10-5  M  C14-guanine  by  E.  coli.  Cell  density  initially  0.3  mg  (wet)  per 
ml.  RNA  radioactivity  plotted  against  cell  mass  with  sample  times  indicated.  Solid  circles  refer  to 
upper  and  left  scales.  Open  circles  refer  to  lower  and  right  scales,  both  of  which  are  expanded  by  a 
factor  of  10  so  that  the  initial  slope  may  be  directly  compared  with  the  final  slope. 


in  the  points  lead  to  great  inaccuracy. 
Various  experiments  have  given  mean 
time  constants  between  3  and  6  minutes 
for  the  decay  of  the  pool  radioactivity 
after  the  external  guanine  is  exhausted. 

Figure  25  shows  the  results  of  an 
experiment  in  which  a  higher  concentra- 
tion of  guanine  (10~5  M)  was  supplied. 
Here  a  comparatively  small  amount  of 
curvature  is  observed,  and  a  straight  line 
through  the  points  taken  after  5  minutes 
extrapolates  to  about  1  minute. 

The  results  with  guanine  are  qualita- 
tively similar  to  those  with  uracil  and 
cytosine.  Two  phases  in  the  incorporation 
curves  are  observed  at  both  high  and  low 
concentrations.  The  quantitative  aspects, 
however,  are  quite  different.  The  flow 
through  the  pool  S  is  small,  and  the  time 
constant  is  not  long  (about  3  minutes). 
The  pool  of  guanine  nucleotides  may  be 
calculated  from  its  time  constant  and 
flow.  The  total  pool  of  guanine  com- 
pounds may  also  be  estimated  directly 
from    the    extrapolated    time    in    the 


experiment  of  figure  25  to  be  sufficient  to 
supply  the  guanine  required  for  1  min- 
ute's growth  of  the  cellular  nucleic  acid. 
This  estimate  is  valid  if  there  is  little 
exchange  between  pool  guanine  com- 
pounds and  external  guanine  and  if  the 
conversion  to  adenine  compounds  is  not 
too  large.  This  is  equivalent  to  7  mM  per 
gram  dry  cells. 

C14-guanine  does,  in  fact,  appear  to 
only  a  slight  extent  in  the  adenylic  acid 
of  RNA.  In  three  experiments  C14- guanine 
at  concentrations  of  2  X  10~6  M,  10" 5  M, 
and  5  X  10-5  M  was  allowed  to  be 
entirely  incorporated  into  RNA.  The 
ratio  of  the  radioactivity  of  the  adenylic 
acid  of  the  RNA  to  that  of  the  guanylic 
acid  ranged  between  0.1  and  0.2. 

Figure  26  shows  the  results  of  a  "chase" 
experiment  in  which  C14- guanine  (3.6  X 
10-7  M)  was  initially  supplied  the  cells 
and  was  followed  15  seconds  later  by 
10~5  M  C12-guanine.  This  result  differs 
from  that  obtained  with  uracil  (McCarthy, 
1962),  in  which  the  specific  activity  of  the 


DEPARTMENT    OF    TERRESTRIAL    MAGNETISM 


249 


Time  in  minu'es  -(upper  curve) 

5  10  i5  20 


3000 


2   looo 


Add  !0"5  M 


C    Guanine 


Guanine  3.6  xlO"7M,  chased 


20         30        40        50        60        70        80        90        100 

Time  in  seconds  -  (lower  curve) 


Fig.  26.  Guanine  chase  experiment.  Exponen- 
tially growing  E.  coli.  Cell  density  1.8  mg  (wet) 
per  ml.  Initially  C14-guanine  (3.6  X  10~ 7  M) 
added.   Fifteen   seconds   later   C12-guanine 


was 


was  added  to  bring  the  concentration  to  10~5  M . 
Lower  curve  represents  the  same  data  as  the 
upper  curve  but  with  time  scale  (alone)  ex- 
panded by  a  factor  of  15. 


uracil  passing  through  the  bypass  ap- 
peared to  be  diluted  almost  instantly. 
With  guanine  there  appears  to  be  a  delay 
of  20  to  30  seconds  before  the  tracer 
passing  through  the  bypass  is  completely 
diluted.  There  is  an  instantaneous  change 
in  slope  to  about  half  that  reached  during 
the  initial  15  seconds.  This  implies  the 


existence,  in  the  pool  of  guanine  com- 
pounds, of  a  small  component  with  a 
short  time  constant.  Such  a  complexity 
is  also  suggested  by  the  shape  of  the  RNA 
curve  in  figure  24,  just  after  the  external 
guanine  has  been  exhausted  at  3  minutes, 
and  by  the  uncertainty  in  the  determina- 
tion of  the  decay  time  constant  for  the 
pool  of  guanine  compounds. 

Adenine.  Figures  27  and  28  show  the 
incorporation  of  C14-adenine  at  10~7  and 
10-5    M.    Here    again    the    qualitative 


3000 


2000 


1000 


o 
o 
■o 
o 
en 


Total  cell 

0/n  o-o Q o ° ° o °- 


30 


Time  in  minutes 


Fig.  27.  Incorporation  of  10~7  M  C14-adenine 
by  growing  E.  coli.  Cell  concentration  1/3  mg 
(wet)  per  ml.  Open  circles  represent  total  cell 
radioactivity.  Solid  circles  represent  TCA- 
precipitable  (RNA)  radioactivity. 


in 


100  - 


or 

CD 

O 

o 

, — » 

>* 

i 

v. 

— 

XJ 

■o 

CD 

h 

O 

O 

v.. 

v_ 

O 

o> 

Q.  \ 

O 

to    50 

CJ 

O 

t_ 

E 

CD 

o 

i_ 

o 

CD 

e 

T3 

< 

Adenine  I0"5  M 

•        »              * 

^^*~*     ■                                          • 

/      — Control 

s£ 

^  I0"5M  C12  Adenine 

added  at  -10  minutes 

2r* 

i                           i                              i 

0  10  20  30  40 

Cell  mass,  sample  times  indicated  (minutes) 

Fig.  28.     Incorporation  of  10~5  M  C14-adenine  by  growing  E.  coli.  Solid  circles,  control.  Open 
circles  10~5  M.  C12-adenine  added  10  minutes  before  carrier-free  C14-adenine. 


250 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


10  15 

Time  in  minutes 


25 


Fig.  29.     Semilogarithmic  plot  of  radioactivity  of  pool  after  external  adenine  was  exhausted. 
Data  from  figure  27. 


features  are  similar  to  those  of  the  other 
three  bases.  Figure  29  (derived  from  the 
experiment  of  figure  27)  shows  a  semi- 
logarithmic  plot  of  the  radioactivity  of 
the  pool  after  the  external  adenine  was 
exhausted.  It  is  immediately  apparent 
that  the  decay  of  the  pool  radioactivity 
cannot  be  represented  as  a  single  expo- 
nential. 

It  might  be  suggested  that  the  shape  of 
the  curve  in  figure  28  is  influenced  by  a 
change  in  the  pool  size.  The  open  circles 
in  figure  28  represent  the  results  of  an 
experiment  in  which  10~5  M  C12-adenine 
was  added  10  minutes  before  the  tracer. 
There  is  no  indication  of  any  difference 
between  the  two  curves  except  that  to  be 
expected  from  the  utilization  of  a  certain 
fraction  of  the  carrier  adenine  before  the 
tracer  was  added.  There  was,  therefore, 
no  measurable  expansion  of  the  pool  of 
adenine  compounds  even  at  this  relatively 
high  concentration  of  external  adenine. 

The  appearance  of  radioactivity  from 
C14-adenine  in  the  guanylic  acid  residues 
of  RNA  was  measured  at  two  concen- 
trations (10~7  M  and  10~5  M).  At  both, 
the  radioactivity  of  the  adenylic  acid 
residues  was  about  three  times  that  of  the 
guanylic  acid  residues. 


Calculation  of  the  bypass  flows  and  time 
constants.  To  adequately  compare  the 
experimental  results  with  the  predictions 
of  the  schematic  diagram,  the  bypass 
flows  and  time  constants  of  the  pool  S 
have  been  calculated  for  each  of  the  four 
bases  from  experiments  at  both  high  and 
low  concentration. 

An  experiment  is  considered  to  be  at  a 
low  concentration  if  the  external  supply 
of  labeled  base  is  exhausted  before  the 
specific  activity  of  the  pool  S  has  become 
comparable  to  the  specific  activity  of  the 
tracer.  In  other  words,  the  external  supply 
is  exhausted  before  the  rate  of  incorpora- 
tion of  radioactivity  has  risen  significantly 
above  the  initial  rate. 

If  the  external  supply  lasts  well  beyond 
the  time  when  the  final  rate  of  incorpora- 
tion into  RNA  has  been  achieved  the 
experiment  is  at  a  high  concentration,  and 
little  further  change  occurs  as  the 
concentration  is  increased.  Experiments 
at  intermediate  concentrations,  where 
neither  of  these  conditions  is  met,  are 
more  difficult  to  interpret. 

For  experiments  at  low  concentration 
the  bypass  flow  can  be  estimated  directly 
from  the  fraction  of  the  total  radioactivity 
entering  the  cell  that  enters  the  RNA. 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


251 


Thus  (c  —  b)/c  listed  in  the  first  row  in- 
table  6  was  calculated  from  the  ratio  of 
the  slopes  of  the  RNA  incorporation 
curve  (TCA-precipitable  radioactivity) 
and  the  total  curve.  Where  experiments 
were  available  at  both  10~6  and  10-7  M 
the  results  agreed. 

For  experiments  at  high  concentrations 
the  bypass  flow  was  calculated  from  the 
ratio  of  the  initial  slope  of  the  RNA 
incorporation  curve  to  its  slope  after  the 
final  rate  had  been  achieved  (i.e.,  after  S 
had  reached  its  maximum  specific  radio- 
activity). Since  there  was  significant  cell 


and  uracil)  very  little,  if  any,  expansion 
occurs.  The  agreement  between  the  data 
in  row  3  and  row  4,  table  6,  supports  this 
conclusion. 

Row  4,  table  6,  lists  the  time  constant 
for  the  increase  in  the  rate  of  entry  of 
radioactivity  into  RNA  as  it  rises  from 
its  initial  rate  to  its  final  rate.  For  this 
purpose  the  linear  part  of  the  curve  (e.g., 
fig.  23)  at  late  times  is  extrapolated  until 
it  strikes  the  time  axis,  giving  the 
effective  delay  time  T' .  To  a  sufficiently 
close  approximation  the  desired  time 
constant  is  given  by  T  =  R2T'/(R2—  Ri), 


TABLE  6.     Bypass  Flows  and  Time  Constants 


Base  Supplied 


Uracil* 


Cytosine  Guanine 


Adenine 


Fraction  of  flow  in  bypass 

Low-concentration  experiments  f 
High-concentration  experiments  J 

Pool  time  constant  (minutes) 
Low-concentration  experiments! 
High-concentration  experiments  || 


0.40 

0.45 

0.74 

0.4 

0.37 

0.37 

0.68 

0.46 

10 

21 

2-6 

2-12 

11 

24 

3.1 

4 

*  Data  from  Year  Book  60. 

t  From  ratio  of  RNA  incorporation  rate  to  total  cell  incorporation  rate. 

%  From  ratio  of  initial  RNA  incorporation  rate  to  final  RNA  incorporation  rate. 

§  Time  constant  of  exponential  decay  of  pool  radioactivity. 

||  Extrapolated  delay  time  (corrected). 


growth,    all    the    experiments    at    high  where  Ri  is  the  initial  slope  and  R2  the 

concentration  have  been  plotted  against  final  slope. 

the  increase  in  cell  mass.  Straight  lines  on         Discussion.    The    evidence    presented 

such  a  plot  correspond  to  constant  rates  here  and  in  last  year's  report  shows  that 

of  incorporation  per  cell.  By  marking  the  the  incorporation  of  the  four  RNA  bases 

sample  time  on  the  abscissa  it  becomes  may  be  represented  by  diagram  1.  The 

possible  to  estimate  the  time  constant  relative  bypass  flow  (c  —  b)/c,  the  size  of 

directly  by  extrapolation.  the  pool  S,  and  the  time  constant  of  the 

Row  3  (table  6)  lists  the  time  constant  pool  (proportional  to  S/b)  vary  widely 

of  the  pool  S  estimated  on  semilogarith-  among  the  four  bases.   The  nucleotide 

mic  plots  of  the  pool  radioactivity  as  a  pool  appears  to  contain  more  than  one 

function  of  time  from  experiments  at  low  component,  and  the  time  course  of  the 

concentrations.  The  time  constant  esti-  decay  of  the  radioactivity  of  the  pool  is 

mated  in  this  way  is  a  measure  of  the  not    always    represented    by    a    single 

ratio  of  the  flow  through  the  pool  to  the  exponential.  Further,  uracil  compounds 

size  of  the  pool  if  the  pool  is  constant  in  and    cytosine    compounds    are    rapidly 

size.  There  is  no  evidence  that  supple-  inter  converted.  Studies  of  cytosine  (table 

mentation  with  RNA  bases  expands  the  5)   and  uracil  show  that  some  of  this 

nucleotide  pools.  In  cases  that  have  been  conversion    occurs    before    entering    the 

tested  by  "preload"  experiments  (adenine  pool  S. 


252 


CARNEGIE      INSTITUTION      OF      WASHINGTON 


The  schematic  diagram  would  obvi- 
ously grow  in  complexity  if  these  features 
were  explicitly  indicated.  It  is  clear  that 
a  very  large  number  of  consequential 
steps  are  ignored  or  briefly  symbolized. 
As  a  result,  the  question  must  be  raised 
whether  the  central  feature  of  the 
diagram — the  existence  of  a  bypass 
around  the  large  pool — is  indeed  sup- 
ported by  the  evidence. 

In  the  first  place  it  is  clear  that  the 
evidence  rules  out  models  of  the  type 

External  tracer  <±  S  — >  RNA 

where  the  flow  into  RNA  is  precisely  that 
required  for  growth.  Such  models  are  not 
consistent  with  an  undelayed  entry  into 
RNA  of  a  given  fraction  of  the  tracer  that 
enters  the  cell  (e.g.,  fig.  21),  nor  do  they 
give  any  explanation  of  the  second  phase 
rise  in  rate  in  experiments  at  high 
concentration  (e.g.,  fig.  23). 

The  only  alternative  that  has  been 
proposed  that  will  explain  the  qualitative 
features  may  be  represented  by  diagram 
2. 


T 


s 

^ 

tr— 1 

R 

a 

r 

Diagram  2 


S  represents  the  nucleotide  pool.  R 
represents  stable  RNA,  and  the  flow  to  R 
is  exactly  that  required  for  growth.  T 
represents  an  unstable  (but  TCA-precipi- 
table)  RNA.  It  has  been  proposed  by 
Jacob  and  Monod  that  RNA  acting  as 
template  for  protein  synthesis  might  have 
the  property  of  rapid  synthesis  and 
breakdown.  If  the  flow  /?  due  to  the 
turnover  of  T  is  very  large,  T  and  S  will 
effectively  have  the  same  specific  radio- 
activity at  any  time.  Thus  a  certain 
fraction  (determined  by  the  relative  size 
of  T  and  S)  of  the  radioactivity  entering 
the   cell   will   appear   without   delay   in 


TCA-precipitable  RNA.  With  the  proper 
choice  of  S  and  T,  calculations  from  this 
model  give  the  experimentally  observed 
curves  for  any  particular  base  (e.g., 
guanine).  For  a  different  base  (e.g., 
uracil),  it  is  necessary  to  change  the  size 
of  both  S  and  T  to  fit  the  observations. 
The  sum  of  S  and  T  determines  the  time 
constant  listed  in  table  6,  and  the  ratio 
(T/(T  +  S)  is  the  "bypass  flow."  The 
size  of  S  may  be  chosen  for  each  base 
(in  the  absence  of  direct  measurements 
of  the  nucleotide  pool  size);  the  relative 
quantities  of  T  for  the  different  nucleo- 
tides are  known  from  the  nucleotide 
composition  of  the  early-labeled  RNA 
reported  below.  Further,  the  time  con- 
stant for  the  rise  in  radioactivity  of  T, 
according  to  this  model,  is  the  same  as 
that  of  the  pool  S — a  particularly  useful 
argument  as  it  is  not  affected  by  the 
inter  conversion  of  the  nucleotides.  Since 
all  other  RNA  labeling  will  be  delayed 
(according  to  this  scheme)  by  the  time 
constant  determined  by  S  -j-  T,  all  the 
observed  early-labeled  RNA  must  be 
considered  to  be  T.  With  uracil  as  tracer 
the  time  constant  of  the  early-labeled 
RNA  fraction  is  about  2  3^2  minutes  and 
the  time  constant  listed  in  table  6  for  S 
is  10  minutes. 

Thus,  at  present  there  appears  to  be  no 
alternative  to  the  existence  of  a  bypass 
around  the  nucleotide  pools.  The  exist- 
ence of  the  bypass  does  not,  of  course, 
rule  out  the  presence,  in  addition,  of  a 
fraction  of  RNA  which  turns  over  by 
degradation  to  low-molecular-weight  frag- 
ments. Equally,  the  existence  of  a  rapidly 
labeled  fraction  of  RNA  does  not  prove 
the  existence  of  turnover  by  degradation. 
As  described  below,  it  appears  that  about 
one-third  of  the  early- labeled  RNA 
fraction  does  turn  over  but  that  the  rate 
of  synthesis  and  degradation  is  not  as 
rapid  as  that  required  for  T  in  schematic 
diagram  2. 

In  one  sense  it  is  not  surprising  that 
the  nucleotide  pools  are  bypassed.  Their 
principal  function  may  not  lie  in  their 
role  as  nucleic  acid  precursors.  ATP  and 


DEPARTMENT    OF    TERRESTRIAL    MAGNETISM 


253 


GTP  presumably  function  in  the  energy 
transport  system,  and  the  other  com- 
pounds presumably  play  a  role  in  a  great 
variety  of  reactions. 

In  last  year's  discussion  of  the  relation- 
ship of  the  bypass  to  the  mechanism  of 
pool  formation  it  was  pointed  out  that 
the  carrier  model  for  amino  acid  pools  was 
consistent  with  the  observations  of  uracil 
incorporation.  Since  the  incorporation  of 
the  other  three  bases  is  in  essential 
respects  similar  to  that  of  uracil,  the 
carrier  model  gains  further  support. 

RNA  Composition 

The  RNA  of  bacteria  is  remarkably 
invariable  in  nucleotide  composition;  the 
DNA  nucleotide  composition  may  vary 
widely  from  species  to  species.  There  may 
exist  a  small  fraction  of  the  bacterial 
RNA  having  a  nucleotide  composition 
like  that  of  the  DNA,  uracil  substituting 
for  thymine.  In  no  case  as  yet,  however, 
has  the  composition  of  such  fractions  been 
reported  to  be  identical  with  that  of  the 
DNA.  In  the  present  work,  five  species  of 
bacteria  and  one  of  yeast  have  been 
examined  for  any  consistent  correlation 
between  the  RNA  and  the  DNA  nucleo- 
tide composition  in  one  or  more  of  the 
RNA  fractions  that  could  be  isolated. 
The  technique  of  isotope  dilution  was 
used  to  determine  the  nucleotide  compo- 
sition of  these  RNA  fractions  with  the 
greatest  possible  accuracy,  so  that  even 
fairly  small  differences  could  be  detected 
among  the  fractions  of  a  given  species. 

The  bacteria  and  yeast  were  grown  in 
the  presence  of  P32  orthophosphate  for 
several  hours  during  exponential  growth. 
They  were  then  harvested  and  were 
washed  three  times.  The  pellet  was 
resuspended,  and  the  cells  were  broken  in 
the  French  pressure  cell.  The  cell  extract 
was  centrifuged  at  105,0Q0#  for  2  minutes 
to  remove  cell  walls  and  unbroken  cells. 
The  supernatant  was  further  centrifuged 
at  105,0000  for  45  minutes  to  pellet  the 
70S  ribosomes.  The  pellet  was  then 
washed    with    buffer,    resuspended,    and 


repelleted  by  a  further  centrifugation  for 
45  minutes.  In  this  way  a  purified  sample 
of  70S  ribosomes  was  prepared.  E.  coli 
ML  30  unlabeled  70S  ribosomes  were  also 
prepared  from  one  batch  of  cells  by  the 
same  method.  Using  the  sucrose  density- 
gradient  sedimentation  method,  30S  and 
50S  ribosomes  derived  from  the  70S 
particles  were  purified  (fig.  30). 


q_  5X10 


Fraction  number 

Fig.  30.  Sucrose  density-gradient  sedimenta- 
tion pattern  of  P32-labeled  SOS  and  50S  ribo- 
somes of  P.  vulgaris.  Sucrose  concentration  5  per 
cent  to  20  per  cent  in  TCM/100  buffer.  Centrif- 
ugation at  37,000  rpm  for  160  minutes,  4°C. 


S-RNA  was  purified  by  further  centri- 
fugation of  the  bacterial  extract  from 
which  the  70S  particles  had  been  removed 
(240  minutes  at  105, 000$).  After  precipi- 
tation by  3  volumes  of  cold  95  per  cent 
ethanol,  the  S-RNA  was  dissolved  in 
TCM  buffer  and  was  adsorbed  on  DEAE. 
S-RNA  eluted  at  0.5  M  NaCl,  and  any 
degraded  ribosomal  RNA  not  pelleted  by 
centrifugation  eluted  at  0.8-1.0  M  NaCl 
(%  31). 

Unfractionated  cell  RNA  was  obtained 
by  precipitating  labeled  cells  in  cold  5  per 
cent  (w/v)  TCA  and  filtering  off  the 
material  on  Millipore  filters. 

The  composition  of  the  single  batch  of 
E.  coli  ML  30  70S  RNA  used  as  unlabeled 
carrier  in  all  subsequent  determinations 
was  measured  by  alkaline  hydrolysis  of  a 
sample,  column  chromatography,  and 
summation  of  the  ultraviolet  absorb- 
encies.  In  the  digests,  approximately  98 


254 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


(VJ 

to 
Q_ 


8X10*  - 


J-       lO5  L 
I 

P    6XI04 

to 

c    4XI04 

O 
O 

2XI04 


S-RNA 

A 

t  \ 
i     \ 

1           \ 

^s^ 

—                                        i 

\                 RNA 

\    *  ■ 

-*Hh— *--- < f-4— •— f*"        1             1 

1          1          1 

1.0 


10 


15 


20       25        30       35       40       45 


Fraction  number 


0.8 

c 

o 

-t— 

n 

0.6 

^ 

c 

Q) 

O 

0.4 

C 

8 

0.2 

o 

Fig.  31.  Elution  by  NaCl  from  a  DEAE-cellulose  column  of  phenol-treated  supernatant  of  re- 
labeled B.  subtilis  cell  extract,  obtained  after  centrifugation  at  105,000*7  for  240  minutes.  Linear 
gradient  of  NaCl  (0.2  M  to  1.0  M)  in  TCM  buffer. 


per  cent  of  the  material  hydrolyzed  was 
recovered  from  the  column.  These  meas- 
urements were  checked  against  the  result 
obtained  by  the  summation  of  the  P32 
counts/minute  contained  in  each  nucleo- 
tide after  hydrolysis  and  column  chro- 
matography of  a  labeled  sample  of  E.  coli 
70S  RNA,  prepared  in  the  same  way. 

The  possibility  of  the  fractionation  of 
the  standard  RNA  by  the  phenol  pro- 
cedure was  also  checked  by  comparison 
of  the  composition  determined  from 
phenol-extracted  70S  RNA  and  from 
TCA-precipitated  P32-labeled  70S  ribo- 
somes  of  E.  coli.  No  significant  differences 
could  be  detected. 

Table  7  indicates  the  nucleotide  compo- 
sition of  the  E.  coli  ML  30  70S  RNA  as 


determined  by  two  methods.  The  results 
are  the  mean  of  several  determinations 
by  each  method. 

Composition  of  the  major  fractions.  The 
nucleotide  compositions  of  the  unfrac- 
tionated  cell  RNA  precipitable  by  cold 
5  per  cent  TCA,  the  70S,  50S,  and  30S 
ribosomes,  and  the  S-RNA  in  the  five 
bacterial  species  are  given  in  tables  8  to 

12.  In  comparison,  the  nucleotide  compo- 
sition of  the  80S,  60S,  and  40S  ribosomes 
of  yeast  and  the  S-RNA  is  given  in  table 

13.  The  slight  differences  observed  in  the 
30S  and  the  50S  ribosomes'  nucleotide 
composition  in  a  given  species  are  repro- 
ducible to  better  than  1  per  cent.  As,  in 
several  of  the  determinations,  the  compo- 
sitions of  the  RNA  in  the  50S  and  30S 


TABLE  7.     Determinations  of  the  Nucleotide  Composition  of  Escherichia  coli  70S  RNA 


By  Summation  of  P32 

Counts 

By  Summation  of  UV 

Nucleotide 

in  Nucleotides, 

Absorbencies  of  Nucleotides 

mole  % 

at  pH  2,  mole  % 

C 

21.7 

22.0 

A 

25.2 

25.1 

G 

32.8 

32.4 

U 

20.3 

20.5 

Several  determinations  by  each  of  the  two  methods  were  carried  out.  The  mean  nucleotide  com- 
position used  in  experiments  was:  C  21.9  mole  per  cent,  A  25.1  mole  per  cent,  G  32.6  mole  per  cent, 
U  20.4  mole  per  cent. 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


255 


TABLE  8.     Compositions  of  RNA  Fractions  of  Pseudomonas  aeruginosa  ATCC  9027 
(DNA  composition  A  =  T  =  13  mole  %,  G  =  C  =  32  mole  %) 


Nucleotide 

Total    RNA 

70S 

50S 

30S 

S-RNA 

C 

22.2 

21.7 

21.2 

21.6 

28.3 

A 

25.7 

25.7 

26.3 

25.1 

20.8 

G 

31.3 

31.6 

31.2 

32.8 

33.8 

U 

20.8 

21.0 

21.3 

20.5 

17.1 

Purine 
Pyrimidine 

1.33 

1.35 

1.35 

1.36 

1.20 

G  +  C 

A  +  U 

1.15 

1.14 

1.10 

1.19 

1.64 

TABLE  9.     Composition  of  RNA  Fractions  of  Aerobacter  aerogenes  ATCC  211 
(DNA  composition  A  =  T  =  22  mole  %,  G  =  C  =  28  mole  %) 


Nucleotide 

Total  RNA 

70S 

50S 

30S 

S-RNA 

C 

22.6 

21.9 

22.0 

22.4 

29.2 

A 

25.0 

25.5 

25.6 

25.3 

19.7 

G 

31.7 

31.5 

31.2 

30.8 

32.5 

U 

20.7 

21.1 

21.2 

21.5 

18.8 

Purine 
Pyrimidine 

1.32 

1.33 

1.32 

1.27 

1.10 

G  +  C 

A  +  U 

1.19 

1.15 

1.14 

1.15 

1.60 

TABLE  10.     Composition  of  RNA  Fractions  of  Escherichia  coli  ML  30 
(DNA  composition  A  =  T  =  24  mole  %,  G  =  C  =  26  mole  %) 


Nucleotide 

Total  RNA 

70S 

50S 

30S 

S-RNA 

C 

22.1 

21.9 

21.5 

22.7 

29.5 

A 

25.2 

25.1 

25.4 

24.8 

19.7 

G 

32.5 

32.6 

33.5 

31.0 

33.8 

U 

20.2 

20.4 

19.6 

21.5 

17.0 

Purine 
Pyrimidine 

1.37 

1.36 

1.44 

1.26 

1.17 

G  +  C 

A  +  U 

1.20 

1.20 

1.22 

1.16 

1.71 

particles  differ  in  individual  nucleotides 
by  as  much  as  10  to  15  per  cent  in  a  single 
species,  the  differences  are  probably  real. 
Neither  the  unfractionated  cell  RNA  nor 
the  70S  RNA  nor  the  S-RNA  was  found 
to  have  a  definite  correlation  with  the 
DNA  for  any  species.  In  fact,  the 
compositions  of  these  fractions  in  the  five 


bacterial  species  are  all  invariable  within 
the  limits  of  the  experimental  error  of 
determination.  Yeast  has  a  ribosomal 
RNA  and  total  cell  RNA  nucleotide 
composition  basically  unlike  that  of 
bacteria.  If  there  is  in  these  fractions  an 
RNA  with  a  composition  like  that  of  the 
DNA,  the  accuracy  of  measurement  by 


256 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


TABLE  11.     Composition  of  RNA  Fractions  of  Bacillus  subtilis  ATCC  6051 
(DNA  Composition  A  =  T  =  29  mole  %,  G  =  C  =  21  mole  %) 


Nucleotide 

Total  RNA 

70S 

50S 

30S 

S-RNA 

C 

22.1 

22.3 

22.5 

22.3 

28.3 

A 

25.5 

25.9 

26.5 

26.5 

20.2 

G 

31.4 

31.0 

32.0 

29.6 

33.9 

U 

21.0 

20.8 

19.3 

21.6 

17.6 

Purine 
Pyrimidine 

1.32 

1.32 

1.39 

1.28 

1.17 

G  +  C 

A  +  U 

1.17 

1.15 

1.20 

1.08 

1.65 

TABLE  12.     Composition  of  RNA  Fractions  of  Proteus  vulgaris  ATCC  4669 
(DNA  composition  A  =  T  =  31  mole  %,  G  =  C  =  19  mole  %) 


Nucleotide 

Total  RNA 

70S 

50S 

30S 

S-RNA 

C 

22.6 

21.7 

21.3 

23.0 

29.3 

A 

24.6 

26.2 

26.5 

24.7 

19.1 

G 

32.0 

31.4 

31.4 

31.9 

33.3 

U 

20.8 

20.7 

20.8 

20.4 

18.3 

Purine 
Pyrimidine 

1.30 

1.35 

1.37 

1.30 

1.11 

G  +  C 
A  +U 

1.21 

1.13 

1.11 

1.22 

1.67 

TABLE  13.     Composition  of  RNA  Fractions  of  Saccharomyces  cerevisiae 
(DNA  composition  A  =  T  =  32  mole  %,  G  =  C  =  18  mole  %) 


Nucleotide 

Total  RNA 

80S 

60S 

40S 

S-RNA 

C 

19.4 

19.2 

19.0 

19.1 

26.3 

A 

26.8 

27.2 

27.9 

25.2 

19.2 

G 

28.3 

28.2 

28.4 

28.4 

34.3 

U 

25.5 

25.4 

24.7 

27.3 

20.2 

Purine 
Pyrimidine 

1.23 

1.24 

1.29 

1.15 

1.15 

G  +  C 
A  +U 

0.91 

0.90 

0.90 

0.91 

1.55 

the  isotope  dilution  technique  cannot 
permit  it  to  be  more  than  10  per  cent  of 
the  RNA. 

The  composition  of  the  14S  RNA  frac- 
tion. It  has  been  established  that  the  first 
detectable  labeled  polynucleotide  ma- 
terial formed  during  the  incorporation  of 
P32  or  C14-uracil  into  bacterial  RNA  has 


different  sedimentational  and  chromato- 
graphic properties  from  the  RNA  detect- 
able by  ultraviolet  absorption.  It  has  also 
been  found  that  most  of  the  C14-uracil 
that  is  incorporated  into  this  fraction  is 
eventually  incorporated  into  the  RNA  of 
the  ribosomes.  This  fraction  is  termed  the 
"eosome"    or    14S    component.    As   this 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


257 


icr 


-   T  2.0  - 


c 

e 

in 
T3 

CO 

ro 
Q_ 

S  5XI03 
o 

g. 
o 

Q. 


C 

o 


c 

CD 
O 

c 
o 
o 

o 
o 


20  40 

Fraction  number 

Fig.  32.     Elution  by  NaCl  from  a  DEAE-cellulose  column  of  a  cell  extract  from  an  E.  coli  culture 
labeled  for  3  minutes  by  P32.  Linear  NaCl  gradient  (0.2  M  to  1.0  M)  in  TCM  buffer. 


material  accounts  for  effectively  all  the 
P32-labeled  RNA  present  in  short  periods 
of  isotope  incorporation,  its  nucleotide 
composition  should  be  similar  to  that  of 
unfractionated  cells  at  these  times. 

The  five  species  of  bacteria  used  in  the 
bulk  RNA  studies  above  were  exposed  to 
short  periods  of  P32-orthophosphate  in- 
corporation during  exponential  growth. 
The  cells  were  then  squirted  into  10  per 
cent  cold  TCA  and  filtered  on  Millipore 
filters.  Many  washes  of  TCA  were  given 
to  remove  most  of  the  5 '-nucleotides  on 
the  filter.  From  an  aliquot  of  cells  that 
had  been  poured  onto  crushed  ice  rather 
than  into  TCA,  14S  RNA  was  then 
isolated.  Analyses  of  the  pulse-labeled 
RNA  in  the  five  species  are  given  in 
table  14. 


The  extracts  from  the  cells  poured  onto 
crushed  ice  were  adsorbed  on  DEAE  and 
eluted  by  a  linear  NaCl  gradient  of  0.2  M 
to  1.0  M  NaCl  in  TCM  buffer.  Figure  32 
shows  a  typical  elution  pattern.  It  can  be 
seen  that  only  one  labeled  component, 
not  tracking  with  any  of  the  ultraviolet- 
absorbing  material,  elutes  at  0.6  M  NaCl. 
This  material  was  pooled,  TCA-precipi- 
tated,  and  collected  by  filtration.  Analysis 
of  the  filters  gave  the  compositions  listed 
in  table  15.  In  each  of  the  species  exam- 
ined, the  base  composition  of  the  14S  or 
"eosome"  obtained  in  this  way  is  identical 
within  experimental  error  to  that  of  the 
total  cell-labeled  RNA  at  this  time. 

A  culture  of  B.  subtilis  was  given  a 
3-minute  labeling  period  with  P32  during 
exponential  growth.  The  nucleotide  com- 


TABLE  14.     Composition  of  Labeled  RNA  Formed  during  Short  Exposure 
of  Bacteria  to  P32  Orthophosphate 


Species 


Time  of  Labeling  with 
Isotope,  minutes 


Labeled  RNA  Composition, 


G  +C 


""^j  iiiv 

^    /u 

A  +  U 

c 

A 

G 

U 

25.4 

21.1 

31.9 

21.6 

1.34 

23.4 

24.8 

30.3 

21.5 

1.16 

22.9 

25.0 

29.5 

22.6 

1.10 

23.3 

25.6 

27.7 

23.4 

1.04 

22.2 

26.7 

27.0 

24.1 

0.97 

Ps.  aeruginosa 

A.  aerogenes 
E.  coli 

B.  subtilis 
P.  vulgaris 


258 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


TABLE  15. 


Composition  of  the  14S  (Eosome)  RNA  Component  of  Bacteria 
Purified  by  DEAE  Chromatography 


Species 

Time  of  Labeling  with 
Isotope,  minutes 

14S  RNA 

Composition 

G  +  C 

C 

A 

G 

U 

A  +  U 

Ps.  aeruginosa 
E.  coli 
B.  subtilis 
P.  vulgaris 

4 
2 
2 
4 

25.6 
22.7 
22.5 
21.9 

20.8 
25.1 
25.3 
27.0 

31.7 
29.1 
28.0 

27.6 

21.9 
23.1 
24.2 
23.5 

1.31 
1.07 
1.02 
0.98 

position  of  the  total  cell-labeled  RNA 
was  measured,  and  a  sample  of  the  cell 
juice  was  treated  with  phenol;  after 
alcohol  precipitation,  and  dissolving  the 
RNA  in  TCM  buffer,  it  was  then  ad- 
sorbed on  a  methylated  serum  albumin- 
coated  kieselguhr  column.  The  RNA  was 
eluted  by  a  linear  gradient  of  NaCl  from 
0.4  to  1.1  M  in  0.04  M  phosphate  buffer, 
pH  6.7.  The  elution  pattern  is  shown  in 
figure  33. 

The  labeled  RNA  does  not  track 
exactly  with  the  16S  and  23S  RNA 
produced  from  the  bulk  of  the  RNA 
components  of  the  cell.  There  are  three 
radioactive  peaks,  but  analysis  of  each 


showed  that  there  was  no  difference  in 
base  compositions  of  any  one  peak  from 
the  composition  of  the  material  eluted  at 
0.6  M  NaCl  from  DEAE  or  from  the 
total  cell-labeled  RNA  at  this  time.  It  is 
evident  that  under  these  conditions  no 
further  fractionation  of  the  newly  formed 
RNA  labeled  with  P32  has  been  achieved. 
The  14S  component  of  E.  coli  labeled 
for  3  minutes  by  P32  was  isolated  by 
sucrose  density-gradient  centrifugation  in 
the  swinging  bucket.  After  centrifugation 
at  37,000  rpm  for  160  minutes  a  peak 
sedimenting  at  about  14S  was  clearly 
resolved  by  its  radioactivity  (fig.  34). 
This    peak    was    collected    and    TCA- 


600 


S.  RNA 

& 


Fractions  pooled  and 

analysed 
I      II         n 


600 


400 


200 


t 

i 

J 

J, 

a) 

"5 

c 


t/> 
c 

3 
O 

o 

CM 

CL 


Fraction  number 

Fig.  33.  Elution  by  NaCl  from  a  methylated  serum  albumin-coated  kieselguhr  column  of  a  cell 
extract  from  an  E.  coli  culture  labeled  for  3  minutes  by  P32.  The  cell  extract  was  treated  with  phenol 
to  remove  protein  from  the  ribosomes  before  adsorption  on  the  column.  Linear  gradient  of  NaCl 
(0.4  M  to  1.1  M)  in  0.04  M  potassium  phosphate  buffer,  pU  6.7. 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


259 


Bottom 


6  8         10         12 

Fraction  number 


Top 


Fig.  34.  Sucrose  density-gradient  sedimentation  of  a  cell  extract  from  P.  vulgaris  labeled  for  3 
minutes  by  P32.  Sucrose  concentration  5  to  20  per  cent  in  TCM/100  buffer.  Centrifugation  at  37,000 
rpm  for  160  minutes,  4°C. 


precipitated.  Its  analysis  showed  that  it 
was  identical  to  the  total  cell-labeled 
RNA  at  this  time,  and  to  the  material 
eluted  from  DEAE  at  0.6  M  NaCl.  This 
would  indicate  that  the  eosome  or  14S 
RNA  can  be  isolated  as  a  discrete  object 
without  measurable  change  in  nucleotide 
composition  and  that  column  chromatog- 
raphy, either  by  the  Mandell  and  Hershey 
column  of  phenol- treated  RNA,  or  by 
DEAE  of  untreated  cell  extracts,  does 
not  result  in  the  isolation  of  newly  formed 
RNA  with  a  nucleotide  composition  any 
different  from  that  obtained  by  TCA 
precipitation  of  unfractionated  labeled 
cells. 

The  analyses  of  the  bulk  RNA  com- 
ponents in  the  five  species  of  bacteria 
used  indicate  no  obvious  relationship  in 
the  nucleotide  composition  of  the  various 
purified  RNA  fractions  to  the  DNA.  The 
composition  of  the  RNA  comprising  most 
of  this  material  in  the  cells,  the  70S 
ribosomes,  is  remarkably  constant  from 
species  to  species.  The  S-RNA  also 
appears  to  be  constant  in  composition 
(fig.  35). 

Subfractionation  of  the  70S  component 


of  bacteria  into  50S  and  30S  or  of  yeast 
80S  into  60S  and  40S  has  brought  to  light 
some  differences  in  nucleotide  compo- 
sition of  the  two  fractions.   In  general, 


0.9       1.1        1.3       1.5       1.7 
G+C/A+T  of  DNA 

Fig.  35.  Comparison  of  G  +  C/A  +  U 
values  for  S-RNA  and  70S  RNA,  from  bacteria 
with  DNA  G  +  C/A  +  T  values  ranging  from 
0.6  to  1.78. 


purine  contents  are  higher,  pyrimidines 
lower,  in  the  larger  (50S  or  60S)  than  in 
the  smaller  (SOS  or  40S)  ribosomal  sub- 
units.   There  is,  however,  no  uniformly 


260 


CARNEGIE     INSTITUTION      OF     WASHINGTON 


consistent  relationship  in  composition 
between  the  DNA,  and  either  30S  or  50S, 
in  the  bacterial  species. 

The  14S  fraction  has  been  found  to  be 
very  different  in  nucleotide  composition 
from  the  normal  total  TCA-precipitable 
RNA  in  the  cell  or  from  the  ribosomes. 
Figures  36  and  37  indicate  a  possible 
relationship  between  the  DNA  compo- 
sition of  the  bacteria  and  the  14S  RNA 


and  70S  RNA  nucleotide  compositions  in 
each  of  the  five  species.  It  can  be  seen 
(fig.  36)  that,  if  the  14S  fraction  were  in 
fact  composed  of  two  entities  of  RNA 
with  different  nucleotide  compositions 
corresponding  to  either  the  DNA  or  the 
ribosomal  RNA,  then  in  each  case  the 
14S  RNA  would  be  made  up  of  approxi- 
mately 33  per  cent  DNA-like  and  67  per 
cent  ribosomal  RNA-like  material. 


Gradients 


?DNA 


Newly 
formed  RNA- 

J- Ribosomal 
1  RNA(70S)    " 


0.7 


0.9        I.I         1.3        1.5        1.7 

G  +  C/A+Tof  DNA 


21 


Fig.  36.  Comparison  of  G  +  C/A  +  U  (T) 
values  for  70S  ribosomal  RNA,  14S  RNA,  and 
DNA  from  bacteria  with  DNA  G  +  C/A  +  T 
values  ranging  from  0.6  to  1.78. 


30 
25 
20r 


DNA 

x , 

I4S 

y^o           >. 

^s* 

•^, — ■ — "i 

70S 

Ps.  aeruginosa  A.  aerogenes 


E.coli 


Key 

* G 

* — C 

o a 

-U(T) 


B.  subtil  is 


P  vulgaris 


Fig.  37.  Graphical  representation  of  nucleo- 
tide composition  of  DNA,  14S  RNA,  and  70S 
ribosomal  RNA  in  each  of  the  five  bacterial 
species  used.  Ordinate,  nucleotide  composition 
(mole  per  cent).  On  the  left  ordinate  of  each 
graph,  DNA  base  composition;  on  the  right 
ordinate,  70S  RNA  base  composition.  14S  RNA 
composition  is  best  fit  between  these  two  com- 
positions. 


Kinetic  Studies  of  RNA  Composition 

Published  studies  of  rapidly  labeled 
RNA  contain  two  conflicting  sets  of 
observations.  On  the  one  hand  it  is 
evident  from  kinetic  studies  of  the  flow 
of  P32-orthophosphate  or  C14-uracil  in  and 
out  of  this  fraction  that  it  can  be  con- 
sidered predominantly  as  a  precursor  to 
the  ribosomal  RNA.  On  the  other  hand, 
analyses  of  the  pulse-labeled  total  RNA 
and  the  purified  14S  fraction  indicate  an 
apparent  nucleotide  composition  inter- 
mediate between  that  of  ribosomal  RNA 
and  the  bacterial  DNA.  Consequently, 
depending  on  the  type  of  observations 
made,  the  rapidly  labeled  fraction  has 
been  described  as  mostly  ribosome  pre- 
cursor or  mostly  the  "messenger"  or 
informational  RNA  postulated  to  be 
necessary  for  the  genetically  directed 
synthesis  of  specific  proteins  in  ribosomes. 

It  therefore  seemed  possible  that  the 
14S  RNA  fraction  contained  molecules  of 
different  compositions  and  functions.  The 
present  study  was  undertaken  to  correlate 
kinetic  and  composition  measurements 
and  to  determine  the  rate  of  synthesis  of 
D-RNA.  The  term  DNA-like  RNA  or 
D-RNA  has  been  used  when  the  poly- 
nucleotide composition  is  observed;  tem- 
plate RNA  designates  RNA  observed  to 
act  as  template;  and  messenger  RNA 
describes  RNA  shown  to  have  all  the 
properties  postulated  by  Jacob  and 
Monod. 

The  nucleotide  composition  of  labeled 
RNA  formed  after  various  times  of 
exposure  to  P32  has  been  measured  for 
each  of  five  species  of  bacteria.  In  each 


DEPARTMENT   OF   TERRESTRIAL    MAGNETISM  261 

case  P32  was  added  to  the  bacteria  material  can  seriously  contaminate  the 
growing  exponentially  in  the  low-phos-  nucleotide  peaks  in  the  analysis.  There- 
phorus-tris  medium.  Growth  and  incorpo-  fore,  very  early  samples  were  chilled 
ration  were  terminated  in  samples  taken  rapidly  by  addition  to  crushed,  frozen 
at  intervals  by  adding  them  to  an  equal  medium  and  washed  in  cold  tris  mag- 
volume  of  10  per  cent  TCA.  Since  the  nesium  buffer.  The  frozen  cells  were 
method  of  nucleotide  analysis  itself  disrupted  in  the  French  pressure  cell,  and 
represented  an  adequate  purification  of  the  effluent  was  added  directly  to  phenol 
the  2',3'-nucleotides,  no  attempt  was  at  37°C.  The  RNA  was  then  purified  by 
made  to  remove  other  macromolecules  means  of  a  second  phenol  extraction  and 
from  the  RNA.  Likewise,  DNA  contami-  two  alcohol  precipitations.  The  final 
nation  is  of  no  account,  since  DNA  is  alcohol  precipitate  was  taken  up  in  buffer 
unaffected  by  the  alkaline  hydrolysis  and  made  to  5  per  cent  with  TCA. 
conditions.  On  the  other  hand,  labeled  Filtration  and  washing  on  the  filter  com- 
5 '-nucleotides  from  the  TCA-soluble  pool  pleted  the  purification.  In  this  way 
are  serious  contaminants.  For  all  but  very  samples  of  labeled  RNA  could  be  ana- 
brief  exposures  to  P32  it  was  found  that  lyzed  from  cells  given  P32  for  periods  as 
5 '-nucleotide  contamination  could  be  short  as  10  seconds, 
eliminated  by  filtration  of  the  TCA  Tables  16  to  20  contain  the  results  of 
precipitate  through  membrane  filters  these  nucleotide  analyses  of  RNA  after 
followed  by  multiple  washes  with  5  per  P32  exposures  of  10  seconds  to  several 
cent  TCA.  After  washing,  the  filters  were  hours.  They  also  compare  the  G  +  C/ 
dried  and  kept  frozen  until  hydrolyzed.  A  +  T  ratio  of  the  DNA  (which  varies 
In  this  way  nucleotide  compositions  of  from  0.61  to  1.78  among  the  five  species) 
the  pulse- labeled  part  of  the  total  RNA  with  the  G  +  C/A  +  U  ratio  of  the  newly 
could  be  measured  without  the  possibility  formed  RNA.  Assuming  that  the  newly 
of  fractionation  or  selective  degradation  formed  RNA  is  a  mixture  of  two  types, 
during  purification  steps.  comparison  of  the  G  +  C/A  +  U  ratio 
For  samples  taken  after  exposures  to  with  that  for  the  bacterial  DNA  and  the 
P32  for  a  period  shorter  than  1  per  cent  of  ribosomal  RNA  (G  +  C/A  +  U  =  1.15) 
the  generation  time  this  technique  proved  or  the  total  RNA  (G  +  C/A  +  U  =  1.20) 
insufficient.  In  this  time  range  the  specific  gives  a  measure  of  the  amount  of  labeled 
radioactivity  of  the  5 '-nucleotides  may  be  RNA  in  the  sample  having  a  nucleotide 
a  thousand  times  that  of  the  2',3'-nucleo-  composition  resembling  DNA.  Such  a 
tides,    and    small    amounts    of    soluble  comparison  is,  of  course,  most  useful  in 


TABLE  16.     Nucleotide  Composition  of  Newly  Formed  RNA  in  Proteus  vulgaris,  mole  % 


Time  of  Labeling 

m  * 

C 

A 

G 

U(T) 

G  +  C 

with  Isotope, 

minutes        ' 

A  +  U 

0.2£ 

0.0019 

23.5 

26.4 

26.6 

23.5 

1.00 

1 

0.0076 

23.5 

27.3 

26.4 

22.9 

1.00 

2 

0.915 

23.5 

27.1 

26.7 

22.7 

1.01 

4 

0.030 

22.7 

26.7 

28.1 

22.5 

1.03 

8 

0.061 

22.9 

26.3 

28.7 

22.1 

1.07 

16 

0.106 

23.0 

25.6 

30.6 

20.8 

1.16 

20 

0.152 

22.5 

24.9 

31.6 

21.0 

1.18 

40 

0.304 

22.4 

24.8 

31.5 

21.3 

1.17 

300 

2.72 

22.6 

24.6 

32.0 

20.8 

1.21 

DNA 

19.0 

31.0 

19.0 

31.0 

0.61 

*  Fraction  of  eth  time  (see  text), 


262 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


TABLE  17.     Nucleotide  Composition  of  Newly  Formed  RNA  in  Bacillus  subtilis,  mole  % 


Time  of  Labeling           T 
with  Isotope,  minutes         e 

C 

A 

G 

U(T) 

G  +  C 

A  +  U 

0.17                0.0023 

23.5 

25.5 

27.3 

23.7 

1.03 

0.5                  0.0069 

24.0 

25.1 

27.9 

23.0 

1.08 

1                      0.014 

22.8 

25.5 

27.2 

24.5 

1.00 

2                      0.028 

23.3 

25.6 

27.7 

23.4 

1.04 

4                      0.055 

23.7 

26.2 

28.0 

22.1 

1.07 

8                      0.100 

22.5 

26.2 

29.3 

22.0 

1.08 

14                      0.193 

22.1 

25.5 

31.5 

20.9 

1.15 

28                     0.386 

21.6 

26.0 

31.7 

20.7 

1.14 

360                     4.96 

22.1 

25.5 

31.4 

21.0 

1.15 

DNA 

21.0 

29.0 

21.0 

29.0 

0.72 

TABLE  18.     Nucleotide  Composition  of  Newly  Formed  RNA  in  Escherichia  coli,  mole  % 


Time  of  Labeling 

Te 

C 

A 

G 

U(T) 

G  +  C 

with  Isotope, 

minutes 

A  +  U 

0.17 

0.0023 

22.8 

25.1 

28.8 

23.3 

1.11 

0.5 

0.0069 

23.2 

25.3 

29.8 

21.7 

1.13 

1 

0.014 

23.4 

24.8 

30.1 

21.7 

1.14 

2 

0.028 

22.9 

25.0 

29.5 

22.6 

1.10 

5 

0.069 

22.4 

26.0 

29.9 

21.7 

1.10 

10 

0.138 

22.0 

25.1 

30.9 

22.0 

1.12 

20 

0.276 

20.9 

25.6 

32.5 

21.0 

1.14 

30 

0.414 

21.4 

25.8 

33.3 

19.5 

1.20 

50 

0.690 

21.6 

25.5 

32.8 

20.1 

1.18 

360 

4.96 

21.9 

25.1 

32.6 

20.4 

1.08 

DNA 

26.0 

24.0 

26.0 

24.0 

species  in  which  the  compositions  of  the 
DNA  and  the  ribosomal  RNA  are  very 
different,  i.e.,  P.  vulgaris,  B.  subtilis,  and 
Ps.  aeruginosa.  Even  in  species  in  which 
these  differences  are  small,  i.e.,  E.  coli  and 
Aerobacter  aerogenes,  the  very  early- 
labeled  RNA  and  the  total  RNA  are  not 
identical. 

Two  features  are  immediately  clear 
from  examination  of  the  data  in  tables 
16  to  20.  In  each  bacterial  species,  the 
composition  of  the  RNA  at  the  earliest 
times  is  that  which  would  result  from  a 
mixture  of  30-40  per  cent  having  a 
DNA-like  composition  and  60-70  per  cent 
ribosomal.  Moreover,  the  change  in  base 
composition  or  G  +  C/A  +  U  ratio  is  not 
an  especially  rapid  one.  In  general  there 
is  little  difference  among  the  first  three  or 


four  analyses  representing  times  from 
about  0.3  per  cent  to  about  4  per  cent  of 
a  generation  time.  If  the  nucleotide 
composition  represents  the  weighted  mean 
between  different  amounts  of  two  differ- 
ent types  of  RNA  molecules,  the  relative 
amounts  of  radioactivity  in  the  two 
different  molecules  present  do  not  change 
during  the  first  5  per  cent  or  so  of  the 
generation  time.  There  is  no  indication 
of  an  extremely  rapidly  labeled  compo- 
nent. Analyses  made  during  the  period 
between  5  and  20  per  cent  of  a  generation 
time  show  a  change  in  composition  toward 
that  of  ribosomal  RNA.  At  times  greater 
than  20  per  cent  of  a  generation  time,  the 
nucleotide  composition  of  the  labeled 
fraction  is  indistinguishable  from  that  of 
the  total  RNA. 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM  263 

Fractionation    of    pulse-labeled    RNA.  fuged  at  40,000  rpm  for  120  minutes.  The 

The  early-labeled  RNA  can  be  separated  magnesium  concentration  resulting  large- 

into  two  fractions  of  different  composition  ly  from  the  bound  ribosomal  ions  can  be 

by    differential    dissociation    from    ribo-  estimated  at  2-4  X  10-4  M. 

somes.  The  pulse-labeled  cells  of  the  four  The  top  half  of  the  supernatant  was 

bacterial  species  shown  in  table  21  were  removed,  TCA-precipitated,  and  filtered, 

prepared  by  exposure  for  3  minutes  to  The  pellet  was  also  resuspended,  precipi- 

P32.  After  chilling  and  washing,  the  cells  tated,  and  filtered.  Table  21  contains  the 

were    broken    in   tris    buffer    containing  results  of  the  nucleotide  analyses  of  the 

10~2  M   MgCl2.   After   removal   of   cell  various  fractions  together  with  the  frac- 

debris  by  a  short  centrifugation  (2  min-  tion  of  labeled  macromolecules  recovered, 

utes  at  105,000$)  most  of  the  ribosomal  For  E.  coli  the  results  are  the  mean  of  six 

material  and  about  90  per  cent  of  the  different  experiments;  the  other  results 

labeled  RNA  were  pelleted  by  means  of  a  are  the  mean  of  two  experiments. 

45-minute  centrifugation  at  40,000  rpm.  The    separation    technique     is    most 

The  ribosome  pellet  (2-5  mg)  and  the  successful  for  E.  coli  material.  Very  little 

interior  of  the  centrifuge  tube  were  rinsed  of  the  pulse-labeled  RNA  was  degraded 

with  distilled  water  at  2°C  to  remove  all  during  the  procedure,  and  the  fractions 

traces  of  the  buffer.  The  pellet  was  then  obtained    had    nucleotide    compositions 

resuspended  in  distilled  water  and  centri-  very  close  to  those  of  pure  ribosomal 


TABLE  19.     Nucleotide  Composition  of  Newly  Formed  RNA  in  Aerobacter  aerogenes,  mole  % 


Time  of  Labeling 

Te 

C 

A 

G 

U(T) 

G  +  C 

with  Isotope, 

minutes 

A  +  U 

1 

0.012 

23.5 

23.4 

32.0* 

21.1 

1.25 

2 

0.023 

24.1 

24.5 

30.2 

21.2 

1.19 

4 

0.046 

23.4 

24.8 

30.3 

21.5 

1.17 

8 

0.092 

22.8 

24.9 

30.7 

21.6 

1.17 

14 

0.161 

22.1 

25.4 

31.4 

21.1 

1.15 

20 

0.230 

21.9 

25.7 

31.2 

21.2 

1.16 

40 

0.460 

21.9 

25.7 

31.5 

20.9 

1.13 

360 

4.14 

22.0 

25.6 

31.7 

20.7 

1.16 

DNA 

28.0 

22.0 

28.0 

22.0 

1.27 

*  The  measurement  of  G  in  the  sample  labeled  for  1  minute  is  inaccurate,  owing  to  contamination 
by  orthophosphate. 


TABLE  20.     Nucleotide  Composition  of  Newly  Formed  RNA  in  Pseudomonas  aeruginosa,  mole  % 

Time  of  Labeling  T  p  G  +C 

with  Isotope,  minutes       ie                   ^  A  u  u  w  A  +  U 

0.25                0.0015  26.2  21.4  31.9  20.5  1.39 

2                     0.012  25.3  21.6  32.3  20.8  1.36 

4                      0.025  25.4  21.1  31.9  21.6  1.34 

8                     0.050  25.8  20.8  31.6  21.8  1.35 

14                     0.086  26.1  21.8  31.0  21.1  1.33 

20                      0.126  24.7  22.0  31.7  21.6  1.29 

40                     0.252  23.9  23.5  32.0  20.6  1.27 

60                     0.388  22.6  25.5  31.2  20.7  1.16 

360                     0.207  22.2  25.7  31.3  20.8  1.15 

DNA  32.0  18.0  32.0  18.0  1.78 


264 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


TABLE  21.     Fractionation  of  3-Minute  P32  Pulse-Labeled  RNA  by  Water  Treatment 


G  +  C 

Ps.  aeruginosa 
1.75 

A. 

aerogenes 
1.27 

E. 

coli 

P. 

vulgaris 

A  +  U 

1 

08 

0.61 

inDNA 

SN 

Ppt. 

SN 

Ppt. 

SN 

Ppt. 

SN 

Ppt. 

C 

28.4 

22.0 



21.9 

25.2 

21.8 



22.1 

A 

20.8 

25.5 



25.6 

24.4 

25.0 



25.5 

G 

31.2 

31.6 



31.6 

26.9 

32.6 



31.8 

U 

19.6 

20.9 



20.9 

23.5 

20.6 



20.6 

G  +  C 

A  +  U 

1.47 

1.15 

1.14 

1.09 

1.19 

1.17 

Per  cent  P32 

52 

40 

5 

46 

30 

70 

12 

34 

Per  cent  P32  RNA 

recovered 

92 

51 

100 

46 

RNA  and  of  E.  coli  DNA.  Distortions  of 
the  real  nucleotide  composition  by  a 
combination  of  unequal  labeling  of  pool 
nucleotides  and  nonrandomness  in  DNA 
sequences  are  not  apparent  in  the 
measured  composition  of  the  D-RNA. 
Apparently  the  "water-shock"  treatment 
causes  the  DNA-like  fraction  to  become 
dissociated  from  the  ribosomes,  leaving 
behind  the  labeled  RNA  which  resembles 
ribosomal  RNA.  Examination  of  water- 
treated  70S  ribosomes  in  the  analytical 
ultracentrifuge  shows  that  the  ribosomes 
have  been  dissociated  into  two  fractions 
of  approximately  60S  and  10S,  unlike  the 
usual  dissociation  into  50S  and  30S 
ribosomes.  The  supernatant  fraction  from 
both  E.  coli  and  Ps.  aeruginosa  appears  to 
be  quite  similar  to  the  DNA  in  compo- 
sition. 

Unfortunately,  the  treatment  results  in 
the  degradation  of  a  large  proportion  of 
the  pulse-labeled  RNA  in  both  P.  vulgaris 
and  A.  aerogenes  presumably  due  to 
RNase  liberation.  Even  there,  however, 
it  is  evident  that  the  fraction  remaining 
with  the  ribosomes  is  purely  ribosomal  in 
base  composition.  The  degradation,  there- 
fore, appears  to  be  selective,  the  DNA- 
like  fraction  being  preferentially  de- 
stroyed. Again  the  rapidly  labeled  frac- 
tions appear  to  contain  two  types  of 
molecule.  Sucrose  density-gradient  sedi- 
mentation  of  both   water-shock   super- 


natant and  pelleted  labeled  RNA  in  the 
presence  of  10~4  M  Mg  showed  similar 
broad  peaks  of  radioactivity  with  a 
14-16S  maximum. 

Kinetic  studies  of  RNA  synthesis  with 
P32.  The  delay  in  incorporation  of  P32  into 
RNA  brought  about  by  the  large  pool  of 
TCA-soluble  RNA  precursors  prevents 
direct   correlation   of   base   composition 


20,000  F 


10 


20        30       40 

Minutes 


Fig.  38.  The  incorporation  of  P32  as  ortho- 
phosphate  into  exponentially  growing  Proteus 
vulgaris. 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


265 


changes  (tables  16-20)  with  studies  of  the 
flow  of  C14-uracil  into  ribosomes.  To  make 
such  a  comparison,  four  samples  of  P. 
vulgaris  extracts  were  prepared  after 
various  exposures  to  P32.  Sedimentation 
analysis  on  sucrose  gradients  allowed 
measurement  of  the  fraction  of  labeled 
RNA  present  in  the  various  species  of 
ribosomes  and  precursors.  In  addition, 
the  rate  of  uptake  of  P32  into  the  total 


RNA  gave  a  measure  of  the  TCA-soluble 
pool. 

Figure  38  shows  the  incorporation 
curve  of  P32  into  P.  vulgaris.  The  data 
were  obtained  from  samples  of  whole  cells 
and  TCA-extracted  cells  filtered  through 
membrane  filters.  In  addition,  some 
samples  at  late  times  were  extracted  with 
hot  ethanol  to  make  corrections  for  the 
lipide  phosphorus.  The  size  of  the  phos- 


Q. 
t/> 
O 

Q. 

Q> 
T3 

"-♦— 
O 

o 


S 

•«— 

o 
p 


o 


aoi 


0001 


ooooi 


TIME-T 
(Defined  by  growth  Q  =  Q0  eT  ) 

Fig.  39.  Log-log  plot  of  the  time  course  of  the  labeling  of  the  RNA  of  Proteus  vulgaris.  Labeled 
fraction  of  nucleotide  phosphorus  for  the  total  cell  and  total  nucleic  acid,  data  from  figure  38;  for  the 
total  nucleic  acid,  14S  eosome  fraction,  and  total  ribosomal  material,  data  from  figure  40.  The  lines 
drawn  were  calculated  from  the  equations  for  the  case  where  the  phosphorus  precursor  pool  is  16 
per  cent  of  the  total  nucleotide  phosphorus  and  the  14S  fraction  is  2.7  per  cent  of  the  total  RNA 
(see  text). 


266 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


phorus  pool  was  estimated  by  the  method 
of  kinetic  analysis  described  in  Year  Book 
60.  Times  were  converted  to  r,  where 
Q  =  QoeT  gives  the  growth  of  the  cells  or 
any  component.  Normalization  of  the 
data  so  that  the  labeled  fraction  of 
nucleotide  phosphorus  fitted  1  —  e~T  gave 
the  top  curve  in  figure  39.  Similar  treat- 


ment of  the  data  for  uptake  of  P32  into 
total  RNA  gave  the  second  curve,  from 
which  the  size  of  the  phosphorus  pool  can 
be  obtained.  The  phosphorus  pool  is,  of 
course,  complex,  containing  mono-,  di-, 
and  triphosphates  of  the  various  nucleo- 
tides, and  inorganic  phosphate.  Moreover, 
there  are  a  great  number  of  reactions 


10,000  - 


5000  - 


C 


-  3000 


-  2000 


1000 


C 

O 
O 


O 


10,000  - 


5000  - 


c 
rj 
o 
o 

CO 

-  12,000  **> 
CL 


-  8,000 


4,000 


Fraction  number 

Fig.  40.  Sedimentation  analysis  of  four  total  cell  extracts  from  Proteus  vulgaris  labeled  for  three 
generations  with  C14-uracil  and  given  (a)  a  4-minute  exposure  to  P3204s,  (6)  8  minutes,  (c)  16  minutes, 
(d)  32  minutes.  Extracts  prepared  from  about  0.5  mg  dry  weight  of  cells  in  tris-HCl,  0.01  M}  pH  7.4, 
MgCl2  10~4  M.  Centrifugation  150  minutes  at  37,000  rpm,  4°C. 


DEPARTMENT   OF   TERRESTRIAL    MAGNETISM 


267 


among  the  various  components,  and  a 
part  of  the  phosphorus  is  destined  for 
lipide  synthesis.  Nevertheless,  it  behaves 
for  kinetic  purposes  like  a  single  delay- 
pool.  The  size  of  the  pool  thus  measured 
is  the  ratio  of  the  phosphorus  in  these 
small  molecules  to  the  total  phosphorus 
in  nucleic  acid  and  the  precursor  mole- 
cules. The  above  procedure  gave  0.16  for 
the  size  of  the  phosphorus  precursor  pool 
in  P.  vulgaris.  (Total  phosphorus  in 
nucleic  acid  and  precursors  =  1.)  The 
curve  drawn  through  the  points  in  figure 
39  was  calculated  using  this  value. 

Four  samples  of  P.  vulgaris  extracts 
were  prepared  after  growth  for  two 
generations  in  C14-uracil  and  4,  8,  16,  and 
32  minutes'  exposure  to  P32.  The  extracts 
were  fractionated  by  sedimentation,  and 
the  fractions  were  assayed  for  TCA- 
precipitable  C14  and  P32  (fig.  40).  Such 
pulse-labeling  experiments  are  the  inverse 
of  those  already  described  in  detail  in 
which  P32  was  used  as  the  steady  label 
and  C14  as  the  pulse  label.  The  labeled 
fraction  of  the  nucleotide  phosphorus 
was  computed  for  the  total  RNA,  the 
rapidly  labeled  14S  fraction,  and  the  bulk 
30S  and  50S  ribosomes,  and  was  plotted 
in  figure  39.  The  points  for  the  total  RNA 
fit  the  total  incorporation  curve  already 
described.  The  other  two  sets  of  points 
fit  well  to  theoretical  curves  calculated 
for  diagram  3. 


0.16 

0.027 

0.813 

TCA- soluble       Rapidly  labeled      Ribosomes 
pool  I4S  eosome  fraction 

Diagram  3 


The  quantity  of  the  rapidly  labeled 
fraction  has  been  assumed  to  be  equal  to 
2.7  per  cent,  the  best  estimate  from 
experiments  with  E.  coli.  It  should  be 
pointed  out,  however,  that  the  curves  are 
relatively  insensitive  to  the  magnitude  of 
this  number  except  at  late  times. 

It  is  apparent  then  that  the  kinetics  of 


P32  labeling  of  the  rapidly  labeled  fraction 
are  consistent  with  a  precursor  product 
relationship  between  the  14S  fraction  and 
ribosomes.  These  kinetic  studies  with  a 
different  organism  and  a  different  labeling 
scheme  are  in  agreement  with  the  E.  coli 
studies  already  published,  after  the 
inclusion  of  the  16  per  cent  TCA-soluble 
pool. 

Correlation  of  nucleotide  compositions 
with  kinetic  data.  The  sedimentation 
analyses  of  figure  40  give  a  measure  of 
the  changing  distribution  of  P32  among 
the  various  RNA-containing  components 
of  the  cell.  It  is  therefore  possible  to 
reexamine  the  measurements  of  the  base 
composition  of  the  total  RNA  in  terms  of 
the  relative  amounts  of  ribosomes  and 
14S  component  present.  A  juxtaposition 
of  the  data  of  table  16  and  figure  40  is 
shown  in  figure  41. 

The  time  course  of  the  G  +  C/A  +  U 
ratio  is  plotted  on  a  scale  running  from 
1.00  (the  average  of  the  earliest  deter- 
minations) to  1.22,  the  value  for  total 
RNA  of  P.  vulgaris  after  long  periods  of 
labeling  (table  16).  Since  the  value  for 
DNA  or  pure  D-RNA  on  this  scale  is  0.61, 
the  origin,  representing  the  zero  time 
composition,  would  correspond  to  36  per 
cent  D-RNA  and  64  per  cent  R-RNA. 
The  fraction  of  the  total  P32-labeled 
RNA  present  in  ribosomes,  i.e.,  all  counts 
sedimenting  at  30S  or  greater,  was 
computed  for  each  time  point  in  figure  40. 
These  fractions  are  plotted  in  figure  41 
on  a  scale  running  from  zero  to  0.8,  since 
the  remaining  20  per  cent  of  the  RNA  is 
S-RNA.  Finally,  the  expected  function 
for  the  change  of  the  fraction  of  the  label 
present  in  ribosomes  was  calculated  from 
the  theoretical  curves  already  plotted  in 
figure  40. 

At  early  times,  while  all  the  P32  radio- 
activity is  present  in  the  14S  fraction,  the 
nucleotide  composition  remains  essen- 
tially constant.  Changes  in  composition 
toward  that  of  the  total  RNA  begin  at 
the  time  when  label  first  enters  ribosomes. 
Thus  the  change  in  G  +  C/A  +  U  can  be 
accounted  for  by  the  increasing  relative 


268 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


Toto!_ 

rna" 


.20 


<r 

c 
=> 

■4- 

o 

•4- 


.10 


36%  D.RNA 
64%  R.  RNA 


i ! ! i   i  Li 


■   i  t  1 1 1» 


i i i   i  i  i  1 1 


0.8 


a> 

E 
o 

to 

o 

o.6  ir 


cr 

T3 

0.4     0, 
o 

Q. 


02  S 
o 
2 


1.0 


TIME-T 


10 


too 


Minutes 


Fig.  41.  A  juxtaposition  of  the  data  of  table  16  and  figure  40.  The  nucleotide  analyses  (vertical 
markers)  are  plotted  against  time  in  terms  of  G  +  C/A  +  U  ratios.  The  left-hand  scale  runs  from 
1.00,  the  average  of  the  earliest  determinations,  to  1.22,  the  value  for  total  RNA.  The  fraction  of  the 
total  P32  label  present  in  ribosomes  (30S  or  greater)  has  been  plotted  on  a  scale  from  zero  to  0.8. 
The  data  are  those  shown  in  figure  40.  The  curve  drawn  through  the  points  is  a  theoretical  one  for 
the  fraction  of  P32-labeled  RNA  in  ribosomes  as  a  function  of  time  calculated  as  described  in  the  text. 


proportion  of  labeled  ribosomes  of  pure 
RNA  composition.  The  data  imply  a 
rapidly  labeled  RNA  fraction  of  2-3  per 
cent  of  the  total  RNA,  with  a  time 
constant  of  2-3  minutes,  one-third  of 
which  is  D-RNA.  Since  the  composition 
of  the  14S  fraction  is  constant,  the  life- 
time of  the  two  RNA  molecules  present 
must  be  very  similar. 

Composition  of  the  isolated  14S  fraction. 
It  is  clear  from  table  16  and  figure  41 


that  there  is  no  measurable  change  in  the 
G  +  C/A  +  U  ratio.  More  direct  evi- 
dence for  the  constancy  of  the  composi- 
tion of  the  14S  fraction  is  shown  in  table 
22.  At  longer  times,  when  the  total 
composition  is  beginning  to  change  (table 
16)  and  the  14S  fraction  no  longer 
contains  all  the  radioactivity,  very  similar 
results  were  obtained,  indicating  propor- 
tions of  D-RNA  and  R-RNA  of  1 :2.  The 
agreement  between  these  late-time  values 


TABLE  22.     Nucleotide  Composition  of  Purified  14S  Fraction  of  Proteus  vulgaris,  mole  % 


Time  of  Labeling 
with  Isotope,  minutes 

Te 

C 

A 

G 

U(T) 

G  +  C 

A  +  U 

2 
5 

0.0148 
0.0370 

23.5 
22.7 

27.1 
26.7 

27.1 
27.6 

23.3 
23.0 

1.02 
1.01 

DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


269 


of  the  purified  14S  fraction  and  the  early- 
time  total  compositions  indicates  that 
unequal  5 '-nucleotide  specific  radioactiv- 
ities did  not  seriously  influence  the  early 
determinations.  For  P.  vulgaris,  base 
compositions  of  this  fraction  prove  to  be 
constant  throughout  the  range  from  15 
seconds  to  5  minutes. 

Resolution  of  pulse-labeled  RNA  by 
chromatography.  The  technique  of  chro- 
matography on  columns  of  methylated 
beef  serum  albumin-adsorbed  on  kiesel- 
guhr  has  been  used  to  study  the  pulse 
labeling  of  various  RNA  fractions.  The 
fractionation  of  DNA  by  this  column  has 
been  shown  by  Mandell  and  Hershey  to 
depend  on  the  molecular  weight  of  the 
nucleic  acid.  Furthermore,  preparations 
of  total  bacterial  nucleic  acid  prepared  by 
the  phenol  method  are  resolved  into 
S-RNA,  DNA,  and  two  peaks  of  ribo- 
somal  RNA  resulting  from  the  two  sizes 
of  molecules  (16S  and  23S)  found  after 
phenol  treatment.  This  technique  has 
been  employed  in  the  fractionation  of  the 
rapidly  labeled  RNA  and  in  following  the 
uptake  of  C14-uracil  and  P32-orthophos- 
phate  into  the  various  nucleic  acid 
fractions. 


RNA  labeled  after  brief  exposure  to  the 
isotopes  appears  in  three  peaks  other  than 
S-RNA  and  DNA,  none  of  which  is 
exactly  coincident  with  the  two  peaks  of 
ribosomal  RNA  (figs.  42  and  44).  This 
distribution  does  not,  however,  reflect  a 
separation  into  RNA's  of  different  nucle- 
otide composition.  The  relative  propor- 
tions of  label  among  the  three  peaks  is 
entirely  reproducible  provided  that  the 
conditions  of  phenol  extraction  are  such 
that  they  preserve  the  normal  proportions 
of  the  two  sizes  of  ribosomal  RNA.  If  the 
RNA  is  extracted  in  the  presence  of  10-4 
M  magnesium  some  of  the  23S  RNA  may 
be  converted  to  16S  RNA.  Such  treat- 
ment also  changes  the  distribution  of  the 
early-labeled  RNA  among  the  three 
peaks.  Figure  42  shows  such  an  elution 
diagram  of  P32  pulse- labeled  RNA  of  P. 
vulgaris.  There  is  still  no  fractionation 
into  materials  of  different  nucleotide 
composition  (table  23).  A  similar  result  is 
obtained  for  B.  subtilis  (table  24).  Thus, 
whereas  it  appears  that  the  three  peaks 
of  radioactivity  may  be  aggregates  of  the 
pulse-labeled  RNA  with  the  16S  and  23S 
material,  or  with  itself,  partition  of  the 
D-RNA  among  them  is  closely  similar. 


E 
O 

CO 
C\J 


ci 


0Q5 


S.  RNA 


DNA 


n 

♦    ♦ 


Ribosomal  RNA  fi 

*  n 


O.D.260m^ 


^, 


10 


400 


200 


15 

c 


if) 

Z5 

o 

o 

CO 
rO 

Q_ 


20        30       40       50       60       70 

Fraction  number 


Fig.  42.  Elution  diagram  from  a  column  of  methylated  beef  albumin  on  kieselguhr  of  a  sample 
of  RNA  from  Proteus  vulgaris  labeled  for  90  seconds  with  P32.  RNA  extracted  from  cells  broken  in 
tris-HCl  0.01  M,  pH  7.4,  containing  10~4  M  MgCl2.  Elution  with  300  ml  of  0.02  M  phosphate  buffer, 
pH  6.7,  containing  a  linear  gradient  of  sodium  chloride  from  0.4  to  1.1  M  NaCl. 


270 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


TABLE  23.     Nucleotide  Compositions  of  Two 
Fractions  of  Proteus  vulgaris  90-Second  P32  Pulse- 
Labeled  RNA  Resolved  on  a  Methylated  Beef 
Albumin  Kieselguhr  Column 


TABLE  24.     Nucleotide   Composition  of  Two 
Fractions  of  Bacillus  subtilis  3-Minute  P32  Pulse- 
Labeled  RNA  Resolved  on  a  Methylated  Beef 
Albumin  Kieselguhr  Column 


C 

A 

G 

U(T) 

C 

A 

G 

U(T) 

Total  P^- 

Total  P32- 

labeled  RNA  23.5 

26.9 

26.7 

22.9 

labeledRNA  23.4 

25.4 

27.5 

23.7 

Peak  I 

23.5 

25.7 

27.2 

23.6 

Peak  I 

23.7 

25.0 

27.2 

24.1 

Peak  II 

24.7 

25.5 

27.0 

22.8 

Peak  II 

23.3 

25.5 

27.6 

23.6 

The  appearance  of  essentially  all  the 
early-labeled  RNA  in  this  region  makes 
it  very  convenient  to  compare  the  kinetics 
of  labeling  of  this  mixture  of  D-RNA  and 
R-RNA  with  those  of  the  S-RNA  and 
DNA.  Two  series  of  samples  of  E.  coli 


RNA  were  chromatographed  in  this  way, 
one  prepared  from  cells  that  had  been 
pulse-labeled  with  C14-uracil  and  the 
other  from  P32  pulse-labeled  cells. 

Figure    43    shows    two    of    the    eight 
analyses  made  of  RNA  prepared  from 


3000  - 


2000  - 


1000 


3 

c 


c 

3 
O 

o 

ro 
0_ 


(a) 

S-RNA 

DNA 
1 

Ribosomal  RNA 

r  v     i    '  a 

Ai 

"  ^^"Vo^ 

i  A  t  \  \ 

it,    <^ 

2000  - 


1000  - 


-    400 


-    200 


15 
c 

£ 


c 

o 
o 

o 


-   2000 


-   1000 


30  60 

Fraction  number 

Fig.  43.  Elution  diagram  from  columns  of  methylated  beef  albumin  on  kieselguhr  of  two 
samples  of  RNA  from  E.  coli  grown  for  three  generations  in  P32C>4~  and  given  (a)  1  minute, 
(b)  334  minutes.  Elution  as  in  figure  42. 


DEPARTMENT    OF    TERRESTRIAL    MAGNETISM 


271 


cells  labeled  with  P32  for  three  genera- 
tions, and  for  periods  up  to  15  minutes, 
with  C14-uracil.  Figure  43(a)  (1-minute 
C14-uracil  incorporation)  shows  the  char- 
acteristic three  peaks  of  pulse-labeled 
material  in  the  high-molecular-weight 
region  not  coincident  with  the  two  peaks 
of  P32-labeled  RNA.  The  specific  radio- 
activity, here  given  by  the  ratio  of  C14 
counts/minute  to  P32  counts/minute,  is 
about  three  times  higher  in  the  high- 
molecular-weight  region,  taken  as  a 
whole,  than  in  the  S-RNA.  By  a  labeling 
time  of  334  minutes,  the  time  of  the 
second  analysis  shown  in  figure  43(6),  the 
difference  in  specific  radioactivity  is  not 
as  noticeable.  At  even  later  times,  figure 
47(6),  most  of  the  C14  radioactivity 
appears  under  the  two  main  peaks  of 
P32-labeled  RNA. 

Similar  elution  diagrams  were  obtained 
for  the  other  five  points  after  40  seconds, 
2  minutes,  4 3/2  minutes,  7  minutes,  and 
10  minutes.  The  specific  radioactivities  of 
S-RNA  and  the  mixture  of  D-RNA  and 
R-RNA  were  obtained  by  summation  of 
the  counts  per  minute  throughout  the 
whole  region.  After  the  appropriate  cor- 
rection had  been  made  the  results  were 


plotted  in  figure  44.  The  entry  of  uracil 
into  S-RNA  is  subject  to  a  delay  of  just 
over  1  minute,  not  shown  in  the  labeling 
of     high-molecular-weight     RNA.     The 


o 


£ 

CO 

~c 

3 
O 

o 

CM 

ro 
0_ 


o 

o 
o 

T3 
O 


O 

<D 

Cl 

if) 


4.0 

- 

3.0 

R.  RNA  - 
+D.RNA 

*/ 

- 

2.0 

/o        / 

^  S.  RNA 

1.0 

UV    ! 

1         1 

1           1 



2        4         6 
Minutes 


10 


Fig.  44.  Plot  of  the  specific  radioactivities 
of  S-RNA  (solid  circles)  and  the  mixture  of  D- 
RNA  and  R-RNA  (open  circles)  as  ratios  of  C14 
cpm  to  P32  cpm  against  time.  Data  from  figure 
43  and  five  other  analyses. 


-  60,000 


-  40,000 


20 


40  60 

Fraction  number 


1 

3 

c 


c 

3 
O 

o 


CM 

20000   rQ_ 


Fig.  45.     Elution  diagram  from  a  column  of  methylated  beef  albumin  on  kieselguhr  of  a  sample 
of  E.  coli  RNA  from  cells  labeled  for  three  generations  with  C14-uracil  and  for  5  minutes  with  P32. 


272 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


labeling  of  DNA  is  delayed  to  a  similar 
extent. 

The  relative  delay  in  S-RNA  synthesis 
is  not  a  special  feature  of  uracil  incorpo- 
ration. When  the  identical  experiment 
was  performed  using  P32  as  the  pulse  label 
and  C14-uracil  as  the  steady  label  very 
similar  results  were  obtained.  Figure  45 
shows  an  elution  diagram  of  RNA  labeled 
with  P32  for  5  minutes.  The  lower  specific 
radioactivity,  i.e.,  ratio  of  P32  cpm  to  C14 
cpm,  and  the  separation  of  P32  counts 
from  the  steady  C14  label,  are  immedi- 
ately evident.  Similar  analyses  were  made 
after  1,  2,  3,  7,  and  10  minutes.  The 
specific  radioactivities  of  S-RNA  and  the 
mixture  of  D-RNA  and  R-RNA  are 
plotted    in    figure    46.    In    spite    of   the 


o 
o 

CVJ 

rO 
Q_ 


30.0 


c 

'£ 
\ 
y> 

o 
o 

^20.0 

;> 

"■•— 
o 
o 
o 

2       10.0 

o 
*♦— 
'o 

<D 
CL 


- 

_ 

R.  RNA           / 
♦D.  RNA       / 

v-^x/ 

2  4  6  8 

Minutes 


10 


Fig.  46.  The  specific  radioactivities  of  S- 
RNA  and  the  mixture  of  D-RNA  and  R-RNA 
as  a  function  of  time  plotted  as  ratios  of  P32 
cpm  to  C14  cpm.  Data  from  figure  45  and  five 
other  analyses. 


curvature  brought  about  by  the  large  pool 
of  acid-soluble  P32  it  is  possible  to  see  a 
delay  in  the  entry  of  P32  into  S-RNA  and 
DNA  relative  to  that  of  R-RNA  and 
D-RNA  of  about  2  minutes. 

Thus  the  labeling  of  both  the  pyrimi- 
dine  bases  and  the  phosphorus  atoms  of 
newly  synthesized  S-RNA  and  DNA  is 


delayed  relative  to  other  RNA  by  a  pool 
of  nucleotide  material  equivalent  to  1  or 
2  minutes'  worth  of  RNA.  The  conclusion 
will  be  drawn,  after  the  presentation  of 
further  results,  that  this  delay  is  a 
consequence  of  the  degradation  of  a 
fraction  of  the  high-molecular-weight 
RNA,  identified  with  D-RNA. 

The  effect  of  chloramphenicol  on  the 
synthesis  of  D-RNA.  It  has  been  known 
for  a  number  of  years  that  the  overall 
nucleotide  composition  of  RNA  synthe- 
sized in  the  presence  of  chloramphenicol 
is  similar  to  that  of  normal  bacterial 
RNA.  Fractionation  of  E.  coli  chlor- 
amphenicol RNA  on  columns  of  DEAE- 
cellulose  gives  two  components,  one 
having  the  nucleotide  composition  of 
S-RNA  and  the  other  approximately  that 
of  ribosomal  RNA.  After  10  minutes' 
exposure  to  C14-uracil  and  chloramphen- 
icol, analysis  of  the  extracts  on  the 
sucrose  gradients  shows  all  the  radio- 
activity in  either  S-RNA  or  the  14S 
precursor  peak.  It  therefore  appeared 
that  chloramphenicol  brought  about  an 
accumulation  of  the  14S  material  here 
under  investigation.  Further  experiments 
were  directed  at  determining  the  relative 
quantities  and  behavior  of  the  D-RNA 
and  R-RNA  moieties. 

A  detailed  analysis  was  made  of  the 
rate  of  change  of  the  nucleotide  compo- 
sition of  the  RNA  in  the  presence  of 
chloramphenicol.  Chloramphenicol  at  200 
mg/1  was  added  to  exponentially  growing 
cultures  of  E.  coli  and  P.  vulgaris,  and 
was  followed  5  minutes  later  by  P32. 
Samples  were  taken  for  nucleotide  analy- 
sis as  described  previously ;  the  results  are 
shown  in  table  25.  They  should  be  com- 
pared with  those  for  uninhibited  cultures 
shown  in  tables  16  and  18. 

The  compositions  at  the  earliest  times 
compare  well  with  those  for  control 
cultures,  and  they  indicate  the  synthesis 
of  about  one-third  D-RNA  and  two- 
thirds  R-RNA.  Although  the  composition 
changes  toward  that  for  total  RNA  in  the 
early  stages  as  in  the  control,  the  nucleo- 
tide  composition   at   late   times   differs 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


273 


TABLE  25.     Composition  of  Newly  Formed  RNA  during  Incubation  with  (200  mg/ml) 

Chloramphenicol 


G  +  C 

Time,  minutes 

C 

A 

G 

U(T) 

A  +  U 

Escherichia  coli 

2 

22.5 

25.0 

29.8 

22.7 

1.05 

4 

22.9 

25.6 

29.4 

22.1 

1.06 

7 

22.7 

25.3 

29.4 

22.6 

1.09 

20 

21.7 

25.0 

31.4 

21.9 

1.13 

40 

22.4 

25.0 

31.2 

21.4 

1.15 

60 

22.8 

25.3 

31.9 

21.0 

1.16 

Total  RNA 

22.1 

25.2 

Proteus  vulgaris 

32.5 

20.2 

1.20 

5 

22.6 

26.0 

28.0 

23.4 

1.02 

10 

21.5 

25.4 

31.2 

21.9 

1.11 

20 

21.9 

25.2 

30.7 

22.2 

1.11 

40 

21.6 

26.0 

30.8 

21.6 

1.10 

60 

21.4 

26.2 

30.9 

21.5 

1.10 

Total  RNA 

22.6 

24.6 

32.0 

20.8 

1.20 

significantly  from  that  of  total  RNA.  In 
E.  coli  the  difference  in  G  +  C/A  +  U 
between  1.15  or  1.16  and  1.20  is  hardly 
significant  (although  it  also  appears  in 
the  data  of  Horowitz  et  al.),  but  in  P. 
vulgaris  we  can  be  confident  of  the  reality 
of  the  difference  between  1.10  and  1.20. 
If  these  apparent  compositions  are  taken 
at  their  face  value  the  RNA  formed  after 
1  hour's  exposure  to  chloramphenicol 
would  consist  of  some  10-20  per  cent 
D-RNA.  Thus  these  observations  suggest 
that  D-RNA  and  R-RNA  are  synthesized 
in  the  normal  proportions  in  chloram- 
phenicol but  that  the  breakdown  of 
D-RNA  is  greatly  reduced. 

Further  studies  of  the  nature  of  RNA 
synthesis  in  chloramphenicol  were  made 
by  means  of  chromatographic  analysis  of 
the  RNA.  Chloramphenicol  at  200  mg/1 
was  added  to  exponentially  growing  E. 
coli  prelabeled  for  three  generations  with 
P32,  and  was  followed  5  minutes  later  by 
C14-uracil.  Control  cells  received  no 
chloramphenicol.  Samples  taken  at  1,  2, 
4,  and  8  minutes  were  washed  and  broken 
in  the  usual  way,  and  the  purified  RNA 
was  analyzed  on  columns  of  methylated 
beef  albumin  as  described  above. 

The  two  analyses  made  after  8-minute 


exposures  are  shown  in  figure  47.  By  this 
time  RNA  from  uninhibited  cells  shows  a 
high  degree  of  coincidence  between  the 
C14  and  P32  counts.  On  the  other  hand, 
the  RNA  from  the  inhibited  culture  has 
a  higher  proportion  of  the  newly  made 
RNA  in  the  region  of  16S  RNA.  More- 
over, the  remaining  C14  radioactivity 
appears  in  a  peak  much  broader  than  that 
given  by  the  23S  P32-labeled  RNA.  In 
fact,  the  specific  radioactivity  in  this 
region  is  lowest  at  the  peak,  suggesting 
the  existence  of  two  overlapping  compo- 
nents of  C14  radioactivity. 

The  analyses  of  chloramphenicol  RNA 
at  the  earliest  times,  1  and  2  minutes 
(not  shown) ,  are  quite  similar  to  those  of 
the  control  except  for  the  absence  of  the 
third  peak  of  C14  radioactivity;  see  figure 
43(a).  At  4  minutes  a  degree  of  coinci- 
dence of  the  peaks  of  radioactivity  lower 
than  that  in  the  control  is  already 
apparent. 

Of  greater  interest  is  the  plot  of  the 
specific  radioactivities  of  S-RNA  and 
D-RNA  plus  R-RNA  shown  in  figure  48. 
It  appears  that  the  entry  of  C14-uracil 
into  S-RNA  takes  place  without  signifi- 
cant delay.  A  similar  effect  was  apparent 
in  the  labeling  of  DNA.  The  data  from 


274 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


T 
i 

i 

C 


c 

O 

o 

ro 
CL 


3000 


2000 


1000 


Chloramphenicol 
200  mg /I 


3000  - 


2000  - 


KXX)  - 


3000 


2000 


1000 


-  3000 


3 

c 


in 
c: 

O 
O 

O 


-  2000 


-  1000 


40  60 

Fraction  number 

Fig.  47.  Elution  diagrams  from  columns  of  methylated  beef  albumin  on  kieselguhr  of  two  samples 
of  RNA  from  E.  coli  cells  labeled  for  three  generations  with  P32  and  8  minutes  with  C14-uracil:  (a)  in 
the  presence  of  200  mg/1  of  chloramphenicol;  (b)  control. 


the  control  cultures  gave  the  same 
1-minute  delay  as  shown  in  figure  44.  As 
the  delay  is  abolished  by  chloramphenicol 
it  does  not  appear  to  be  the  result  of  a 
special  small-molecule  precursor  pool.  On 
the  other  hand,  the  lack  of  a  delay  may 
be  correlated  with  the  stability  of  D-RNA 
in  chloramphenicol-inhibited  cultures. 

Discussion.  A  possible  consequence  of 
the  messenger  RNA  hypothesis  of  Jacob 
and  Monod  is  that  these  RNA  molecules 
may  survive  long  enough  to  specify  the 
synthesis  of  only  one  protein  molecule. 
Since  these  molecules  would  presumably 


be  included  in  the  D-RNA  fraction, 
knowledge  of  its  rate  of  synthesis  is  of 
major  importance  in  considering  the 
mechanism  of  protein  synthesis. 

There  is,  however,  no  evidence  for  an 
extremely  rapid  rate  of  turnover  of  the 
D-RNA  fraction  such  that  the  rate  of 
incorporation  into  these  molecules  is 
many  times  that  into  the  stable  RNA. 
In  fact,  nucleotide  compositions  measured 
at  very  early  times  are  consistent  with  a 
flow  of  material  into  D-RNA  equal  to 
half  that  into  stable  ribosomal  RNA. 
There  is,   of  course,   as  in  all  isotope- 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


275 


4  6  8 

Minutes 


10 


Fig.  48.  The  specific  radioactivities  of  S- 
RNA  and  the  mixture  of  D-RNA  and  R-RNA 
as  a  function  of  time  from  chloramphenicol- 
inhibited  cultures  plotted  as  ratios  of  C14  cpm 
to  P32  cpm.  Data  from  figure  47(a)  and  three 
other  analyses. 


labeling  experiments,  the  possibility  that 
the  rate  of  uptake  of  label  does  not 
measure  the  true  rate  of  synthesis.  Thus, 
in  the  present  experiment,  although  it 
appears  that  the  rate  of  RNA  synthesis 
is  50  per  cent  higher  than  the  flow  into 
stable  RNA,  we  may  argue  that  D-RNA 
is  being  synthesized  and  degraded  much 
more  rapidly,  in  equilibrium  with  a 
chemically  or  physically  isolated  pool.  A 
similar  argument  must  be  applied  to  the 
incorporation  of  C14-uracil  and  other 
bases.  Since  this  would  imply  rigid 
separation  between  two  or  more  pools  of 
nucleotides  and  inaccessibility  to  labeling 
by  either  C14-uracil  or  P32  this  process 
has  not  been  considered  further  in  the 
present  discussion. 

In  all  the  five  organisms  studied  the 
composition  of  the  newly  synthesized 
RNA  is  intermediate  between  that  of 
D-RNA  and  R-RNA.  Fractionation  by 
water  treatment  of  the  14S  component  of 
Ps.  aeruginosa  and  of  E.  coli  into  two 


RNA  fractions  provides  a  strong  indica- 
tion that  it  consists  of  a  mixture  of  two 
types  of  molecule,  one  with  a  base 
composition  like  that  of  the  DNA  of  the 
cell  (D-RNA)  and  the  other  pure  R- 
RNA.  Although  the  fractionation  was 
not  as  successful  with  RNA's  of  other 
bacteria,  results  are  consistent  with  the 
existence  of  two  molecules  with  different 
compositions. 

Detailed  analyses  of  changes  in  nucleo- 
tide composition  of  the  newly  formed 
RNA  indicate  that  the  two  types  of 
molecule  have  very  similar  kinetics  of 
incorporation.  The  apparent  composition 
of  the  14S  RNA  fraction  remains  constant, 
and  the  changes  in  overall  nucleotide 
composition  appear  to  result  from  the 
degradation  of  the  D-RNA  moiety  and 
the  conversion  of  R-RNA  to  stable 
ribosomal  material.  Thus  the  compo- 
sitional changes  can  be  correlated  with 
the  appearance  of  radioactivity  in  stable 
ribosomal  material  which  dilutes  a  14S 
component  of  constant  composition.  In 
effect,  the  lifetimes  of  D-RNA  and 
R-RNA  molecules  in  the  14S  fraction  are 
closely  similar,  one  being  removed  mainly 
by  degradation  and  the  other  by  the 
addition  of  protein  and  conversion  to 
ribosomes. 

The  relative  amounts  of  D-RNA  and 
R-RNA  synthesized  in  each  of  the  five 
organisms  studied  appear  to  be  the  same. 
This  ratio  may  have  stoichiometric 
significance  or  it  may  merely  be  a  function 
of  the  conditions  of  growth,  since  other 
authors  report  considerably  higher  frac- 
tions of  D-RNA  in  pulse-labeled  RNA. 
Bacteria  undergoing  a  downward  transi- 
tion in  growth  rate  and  nongrowing 
yeasts  produce  higher  proportions  of 
D-RNA.  The  G  +  C/A  +  U  ratios  for 
the  DNA-like  materials  reported  in  either 
of  these  two  conditions  are,  however, 
consistent  with  an  equal  mixture  of 
D-RNA  and  R-RNA. 

Chromatography  of  pulse-labeled  RNA 
on  methylated  beef  albumin  columns  does 
separate  the  label  from  the  bulk  RNA  but 
does   not   give   any  separation  between 


276 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


D-RNA  and  R-RNA.  Each  of  the  three 
peaks  obtained  appears  to  consist  of 
mixtures  of  the  two  components.  The 
relative  quantity  appears  to  depend  on 
conditions  of  extraction. 

The  separation  between  S-RNA  and 
the  other  RNA  has  proved  useful,  how- 
ever, for  kinetic  studies.  Studies  with 
C14-uracil  and  P32  showed  a  delay  in 
incorporation  of  label  into  S-RNA  relative 
to  the  other  RNA  of  about  1  to  2  minutes. 
Rather  than  being  the  result  of  a  pre- 
cursor pool  of  nucleotides,  this  delay 
seems  to  be  a  reflection  of  the  turnover 
of  D-RNA.  The  apparent  difference 
between  the  delays  experienced  in  C14- 
uracil  and  P32  labeling  may  be  attrib- 
utable to  the  special  features  of  uracil 
incorporation.  Thus  some  of  the  C14- 
uracil  could  enter  S-RNA  directly  so  that 
the  initial  rate  of  entry  is  not  zero  and 
the  delay  resulting  from  the  utilization  of 
nucleotides  derived  from  D-RNA  break- 
down is  apparently  reduced. 

A  unified  picture  of  the  flow  of  label 
into  RNA  would  be  the  following.  The 
initial  flow  is  accounted  for  by  a  one-third 
entry  into  D-RNA  and  the  remaining 
two-thirds  into  R-RNA.  The  former  is 
degraded  and  the  nucleotide  material 
used  to  some  extent  for  S-RNA  and  DNA 
synthesis.  Of  the  nucleotides  originally 
entering  D-RNA  some  would  serve  to 
make  S-RNA  (20  per  cent  of  the  total 
RNA)  and  DNA  (about  15  per  cent  of 
the  cell  nucleic  acid)  and  possibly  R- 
RNA.  If  D-RNA  were  degraded  to 
nucleoside  5 '-diphosphates  or  monophos- 
phates the  conversion  of  D-RNA  to  DNA 
could  proceed  by  the  mechanism  de- 
scribed by  Cohen  et  al.  Thus,  rather  than 
being  an  obligatory  precursor  of  R-RNA 
the  D-RNA  is  the  precursor  of  S-RNA 
and  DNA  and  only  to  a  more  limited 
extent  the  precursor  of  R-RNA.  This 
picture  would  fit  all  the  kinetic  experi- 
ments, and  it  finds  support  from  the 
studies  of  chloramphenicol  inhibition. 

The  presence  of  chloramphenicol  in  the 
growth  medium  during  the  incorporation 
of    C14-uracil    into    RNA    removes    the 


kinetic  delay  of  the  entry  of  label  into 
S-RNA.  Thus  more  C14  enters  S-RNA 
and  DNA  directly.  It  is  unlikely  that  the 
removal  of  the  delay  by  chloramphenicol 
can  be  adequately  explained  by  changes 
in  a  small-molecule  precursor  pool  feeding 
S-RNA.  This  effect  has  been  shown  to  be 
associated  with  an  accumulation  of 
D-RNA.  Although  the  rate  of  synthesis 
of  D-RNA  remains  the  same,  its  degra- 
dation is  markedly  reduced.  The  con- 
version of  "chloramphenicol  RNA"  to 
soluble  material  when  the  antibiotic  is 
removed  may  reflect  the  renewal  of  the 
degradation  process,  although  it  is  not 
clear  whether  both  the  D-RNA  and 
R-RNA  fractions  are  lost. 

Purification  of  D-RNA 

When  the  report  year  was  drawing  to  a 
close  it  was  found  that  single-stranded 
DNA  could  be  immobilized  in  agar,  and 
complementary  RNA  could  be  hybridized 
with  this  DNA.  In  applying  this  principle, 
which  was  discovered  through  a  lead 
provided  by  Bautz  and  Hall  of  the 
University  of  Illinois,  the  D-RNA  of 
several  phages  and  bacteria  has  been 
purified  and  chemically  characterized. 
Kinetic  studies  of  the  synthesis  of 
D-RNA  were  carried  out  with  Proteus 
vulgaris.  It  was  found  that  D-RNA  has  a 
half-life  of  about  2  minutes  and  comprises 
1  per  cent  of  the  total  cellular  RNA. 
Assuming  D-RNA  to  be  template  RNA, 
it  would  therefore  function  catalytically 
to  take  part  in  the  synthesis  of  perhaps 
fifty  polypeptide  strands. 

Kinetics  of  Labeling  of  Turnip  Yellow 
Mosaic  Virus 

Turnip  yellow  mosaic  virus  (TYMV) 
is  a  spherical  plant  virus  containing  36 
per  cent  RNA  and  64  per  cent  protein. 
Preparations  of  the  cell  sap  of  Chinese 
cabbage  plants  infected  with  this  virus 
contain  small  amounts  of  noninfectious 
viruslike  particles.  The  protein  comple- 
ment of  these  particles  is  serologically 
indistinguishable  from  that  of  the  infec- 


DEPARTMENT    OF    TERRESTRIAL    MAGNETISM  277 

tious  unit,  but  the  RNA  complements  are  were  washed  and  placed  in  1  ml  of  P32- 

smaller.  The  RNA  content  appears  to  be  labeled  orthophosphate.  In  about  an  hour, 

"quantitized,"  and  a  series  of  particles  when  the  fluid  was  completely  absorbed, 

containing  about  24,  12,  and  6  per  cent  a  large  volume  of  water  was  added  and 

RNA    can    be    separated.    In    addition,  the  plants  were  maintained  in  the  green- 

another  particle  is  found  that  contains  no  house.  At  intervals  after  administration 

RNA.  The  existence  of  such  a  series  of  of  the  tracer,  disks  were  taken  from  each 

particles  in  the  cell  sap  of  infected  plants  half-leaf  with  a  cork  borer.  Then  with  a 

implies  that  the  particles  may  be  func-  smaller  cork  borer  a  sample  was  obtained 

tionally    related,     perhaps    serving    as  from   the    center   of   the    pile    of   disks 

sequential   stages   along   a   pathway   of  previously  taken.  The  smaller  disks  were 

biosynthesis  leading  to  the  infectious  unit  extracted    with    hot    ethanol,    and    the 

itself.  This  hypothesis  was  tested  by  Dr.  extract  was  analyzed  for  phosphorus  and 

R.  E.  F.  Matthews,  who  discovered  the  radioactivity.     The    larger    disks    were 

RNA-containing  viruslike  particles  at  the  ground  to  a  pulp,  the  pulp  was  clarified 

Plant  Diseases  Division  of  the  Depart-  in  the  centrifuge,  and  the  resulting  cell 

ment   of   Scientific   and    Industrial   Re-  sap  was  treated  with  an  antiserum  specific 

search,  Auckland,  New  Zealand.  Such  a  for  viral  protein  and  nucleoprotein.  The 

hypothesis  was  in  fact  consistent  with  his  specific    precipitates    were    digested    in 

results.   Since  the  experiments  did  not  alkali ;  the  resulting  viral  nucleotides  were 

demonstrate  precursor-product  relation-  separated  chromatographically,  and  their 

ships,    an    attempt    was    made    in    Dr.  specific  radioactivities  were  determined. 

Matthews'  laboratory  to  improve  upon  The  partition  of  phosphorus  and  the 

the  design  of  the  tracer  experiments  in  incorporation    of    radioactivity    in    the 

the  hope  that  they  would  yield  a  definitive  infected  plant  are  shown  in  figure  49.  The 

answer     to     the     problem     of     TYMV  phosphorus  content  expressed  in  milli- 

assembly.  grams  per  leaf  appeared  to  remain  about 

Initially,  "chase"  experiments  of  the  constant  in  the  ethanol  extract,  decreased 
type  used  in  our  studies  on  bacterial  to  half  in  the  nonviral  alcohol-insoluble 
nucleoprotein  synthesis  were  tried,  by  fraction,  and  rose  linearly  fourfold  in  the 
which  means  it  was  hoped  to  follow  the  virus.  The  specific  radioactivities  in  these 
flow  of  radioactive  precursor  through  the  fractions  rose  rapidly  at  first,  then 
synthetic  sequence.  After  a  number  of  appeared  to  level  off.  In  this  particular 
unsuccessful  trials  with  P32  labeling  of  the  experiment  the  specific  activity  of  the 
virus  it  was  found  that  chase  experiments  alcohol-soluble  fraction  reached  the  max- 
were  impractical  because  the  pool  of  low-  imum  value  at  5  hours.  In  other  experi- 
molecular- weight  compounds  in  Chinese  ments,  however,  labeling  of  this  fraction 
cabbage  was  very  large  and  practically  reached  saturation  in  an  hour  or  so. 
unalterable  in  size,  either  as  a  result  of  The  fixation  of  radioactivity  in  the 
extreme  phosphorus  starvation  or  by  alcohol  in  soluble  macromolecular  corn- 
forcing  phosphate  into  the  host  plant,  ponents  of  the  infected  plant  is  a  slower 
Therefore,  the  incorporation  of  radio-  process,  and  the  specific  radioactivity 
activity  was  followed  over  a  long  enough  does  not  reach  that  of  the  alcohol-soluble 
period  of  virus  development  to  permit  low-molecular-weight  fraction  during  the 
comparisons  of  the  early  rates  of  incorpo-  course  of  the  experiment.  After  40-60 
ration  with  those  occurring  later.  hours    there    is    little,    if    any,    further 

In  a  typical  experiment  twelve  young  increase  in  specific  activity  of  the  viral  or 

Chinese  cabbage  plants  that  had  been  host  nucleic  acid.  Nevertheless,  the  virus 

inoculated  with  TYMV  12  days  earlier  continues  to  be  synthesized  at  the  same 

were    trimmed    to    two    well    expanded  rate  as  before.  This  might  be  taken  to 

systemically  infected  leaves.   The  roots  mean  that  nucleic  acid  was  turning  over, 


278 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


1.8 


1.4 


o 

CD 


CL 


h/s 


-— °" 


-^°" 


— -o 


20  60  100  140  180 


0> 

E 

C 


C 

O 
(J 


8000 


6000  - 


4000  - 


2000 


Hours 

Fig.  49.  TYMV  1.  Phosphorus  content  and 
specific  radioactivities  in  TYMV- infected 
Chinese  cabbage  leaves.  Upper:  open  circles, 
total;  triangles,  alcohol  soluble;  solid  circles, 
nonvirus  alcohol  insoluble;  crosses,  virus.  Lower: 
triangles,  alcohol  soluble;  solid  circles,  nonvirus 
alcohol  insoluble;  crosses,  virus. 


but  the  result  can  equally  well  be 
explained  by  assuming  that  there  is  a 
small  pool  of  phosphorus  precursor  which 
"bypasses"  a  large  pool.  In  this  way 
phosphorus  of  relatively  high  specific 
radioactivity  could  enter  the  virus  at 
early  times  whereas  later  phosphorus  of 
low  specific  activity  would  be  derived 
from  the  large  pool,  thus  causing  a 
leveling-off  in  the  specific-activity  curve. 
S35-labeled  sulfate  was  used  to  follow 
the  synthesis  of  virus  protein.  In  an 
experiment,  24  previously  infected  plants 
were  arranged  in  sets  of  3  and  provided 
with  S35  through  their  roots.  At  intervals 


after  the  administration  of  the  tracer  a 
set  of  plants  was  harvested  and  the  leaf 
laminae  were  ground  to  a  pulp.  The  viral 
protein  was  pelleted  in  the  ultracentri- 
fuge.  The  pellet  was  then  fractionated  in 
a  cesium  chloride  density  gradient  in  the 
swinging-bucket  ultracentrifuge.  The  top 
component  (T)  is  the  empty  protein 
shell.  Next  in  order  are  the  minor  compo- 
nents (Boo,  Bo),  the  infectious  virus  (Bi), 
and  a  noninfectious  particle  (B2)  that  is 
chemically  indistinguishable  from  the 
virus.  The  specific-activity  data  for  the 
empty  protein  shell  and  the  virus  are 
shown  in  figure  50.  Another  experiment 


Hours 

Fig.  50.     TYMV  2.  S35  labeling  of  the  empty 
protein  shell  T,  and  of  TYMV  (Bx). 


was  performed  similarly  except  that  non- 
radioactive plants  in  the  same  stage  of 
growth  and  infection  as  the  radioactive 
ones  were  added  to  the  radioactive  plants 
during  preparation  of  the  virus  to  provide 
enough  of  the  minor  components  for 
analysis.  The  results  are  shown  in  figure 
51.  ' 

The  shapes  of  the  curves  for  S35 
incorporation  resemble  those  for  P32 
incorporation  in  the  nucleoproteins  B0o, 
B0,  Bi,  B2,  but  the  S35  curve  for  the  empty 
protein  shell  is  markedly  different.  The 
specific  radioactivity  of  this  shell  rises 
very  rapidly  to  more  than  twice  that  of 
the  nucleoproteins,  then  decreases  slowly. 
The  amino  acid  compositions  of  T  and 
Bx  are  known  to  be  very  similar.  Never- 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


279 


90        120       150      180 

Hours 


210 


Fig.  51.  TYMV  3.  S35  labeling  of  the  empty 
protein  shell  T,  the  minor  nucleoproteins  Boo, 
Bo,  B2,  and  TYMV  (Bi). 


theless  the  S35  of  T  is  in  the  amino  acids 
cysteine  and  methionine,  and  the  radio- 
activity cannot  be  released  from  the  shell 
by  treatments  with  mercaptoethanol, 
urea,  or  sodium  sulfate.  The  high  specific 
radioactivity  could  be  explained  if  T  is 
turning  over  rapidly  and  draws  on  sulfur- 
containing  compounds  having  a  high 
specific  radioactivity  at  early  times.  T  is 
continually  synthesized  during  the  exper- 
iment and  is  one-fifth  to  one-fourth  of  Bi 
in  amount.  Hence,  the  decrease  in  its 
specific  activity  is  strange.  It  could  be 
explained  if  the  empty  protein  shell  were 
breaking  down  at  the  same  time  that  new 
shells  were  being  made  from  sulfur 
compounds  whose  specific  activity  was 
now  low  compared  with  that  at  early 
times,  perhaps  as  a  result  of  a  bypass 
mechanism  as  suggested  for  phosphorus 
incorporation.  This  suggestion  would  also 
account  for  the  shape  of  the  nucleo- 
protein  curves.  From  the  data  of  figures 
50  and  51  it  appears  highly  unlikely  that 
T  is  a  precursor  of  Bi.  In  addition,  the 
minor  components,  which  are  1  to  3  per 


cent  of  the  nucleoprotein,  could  not  be 
precursors  between  T  and  Bi  because 
their  specific  activities  are  too  low  at 
early  times. 

The  hope  of  definitely  establishing  the 
mechanism  of  how  TYMV  is  assembled 
has  not  been  realized.  Nevertheless,  these 
experiments  make  clear  the  fact  that  a 
much  deeper  knowledge  of  plant  physi- 
ology— especially  knowledge  of  the  kinds, 
amounts,  and  kinetic  behavior  of  low- 
molecular-weight  precursors — is  required 
before  any  satisfying  model  of  TYMV 
synthesis  can  be  proposed.  It  is  also  clear 
that  ' 'chase"  experiments,  which  have 
proved  a  powerful  tool  for  biosynthesis 
studies  in  microorganisms,  are  practically 
impossible  in  higher  plants. 

Control  Mechanisms 

In  the  bacterial  cell,  protein  synthesis 
and  RNA  synthesis  are  closely  inter- 
locked. DNA  synthesis,  in  contrast,  is 
relatively  independent.  Thus,  in  the  15 
T-A-U-  mutant  the  lack  of  thymine  (T) 
prevents  DNA  synthesis  without  any 
immediate  effect  on  protein  or  RNA 
synthesis.  Conversely,  the  lack  of  arginine 
(A)  and  uracil  (U)  causes  no  immediate 
change  in  the  rate  of  DNA  synthesis.  The 
lack  of  either  U  or  A  brings  both  protein 
and  RNA  synthesis  to  a  halt.  The 
requirement  of  RNA  synthesis  for  the 
continued  formation  of  protein  is  a 
natural  consequence  of  the  short  lifetime 
of  protein-forming  templates.  However, 
there  is  no  obvious  reason  why  amino 
acids  are  needed  for  RNA  synthesis. 

Protein  synthesis  is  not  essential  for 
RNA  synthesis,  as  RNA  continues  to  be 
synthesized  in  the  presence  of  concentra- 
tions of  chloramphenicol  that  block  the 
incorporation  of  amino  acids  into  protein. 
Such  results  have  been  interpreted  as 
showing  a  "catalytic  role"  of  amino  acids 
in  RNA  synthesis  or  a  "derepression"  of 
RNA  synthesis  either  by  amino  acids  or 
by  their  activated  forms. 

These  interpretations  do  not  seem 
entirely  satisfying.  The  action  of  chlor- 


280 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


amphenicol  is  extremely  rapid  and  much 
faster  than  would  be  expected  on  this 
basis.  Also,  chloramphenicol  can  remove 
the  inhibition  of  RNA  synthesis  caused 
by  an  amino  acid  analog,  5-methyltryp- 
tophan.  Puromycin  allows  continued 
RNA  synthesis  in  the  absence  of  protein 
synthesis,  but  it  is  not  as  effective  as 
chloramphenicol  in  eliminating  the  need 
for  amino  acids.  Such  investigations  are 
being  continued  in  the  hope  of  gaining  a 
better  understanding  of  cellular  control 
mechanisms. 

Effects  of  virus  infection.  Last  year  it 
was  shown  that  a  study  of  the  time  course 
of  enzyme  synthesis  after  the  addition  or 
removal  of  an  inducer  provided  valuable 
clues  to  the  induction  mechanism  and  to 
regulatory  processes  in  the  synthesis  of 
protein.  The  addition  of  inducer  immedi- 
ately accelerated  the  synthesis  of  /?- 
galactosidase,  and  after  2.5  to  3.0  minutes 
a  steady  rate  many  times  greater  than 
that  of  the  uninduced  cells  was  observed. 
It  was  concluded  that  the  brief  period  of 
accelerating  synthesis  was  the  time 
required  to  produce  the  new  enzyme- 
forming  units  (EFU)  necessary  for  the 
induced  rate  of  synthesis  of  enzyme. 
Similarly,  the  removal  of  inducer  caused 
an  immediate  deceleration  in  the  rate  of 
synthesis,  and  within  3  minutes  the 
system  returned  to  the  rate  observed  for 
uninduced  cells.  Such  a  result  would  be 
expected  if  the  induced  enzyme-forming 
units  were  unstable  and  decayed  with  a 
time  constant  of  2.5  to  3  minutes. 

This  year  other  means  of  markedly 
altering  the  rate  of  protein  synthesis  were 
investigated.  In  collaboration  with  Dr. 
Maury  Miranda  at  the  Instituto  de 
Biofisica,  Universidade  do  Brazil,  Rio  de 
Janeiro,  Brazil,  studies  were  carried  out 
to  determine  the  capacity  of  E.  coli  cells 
to  synthesize  host  protein  after  phage 
infection. 

Wild-type  T4  bacteriophage  rapidly 
lyses  K12X  E.  coli,  but  rll  mutants  of  it 
are  unable  to  cause  lysis  of  this  organism 
even  though  infection  occurs.  Further- 
more,  Benzer  has  shown  that  the  rll 


region  of  the  T4  DNA  can  be  divided  into 
the  A  and  B  cistrons,  the  two  subunits 
having  different  mutational  and  func- 
tional characteristics.  Two  rll  mutant 
phages  were  selected  for  our  studies:  164, 
a  mutant  of  the  A  region;  and  196,  a 
mutation  of  the  B  cistron.  K12X  mixedly 
infected  with  both  these  rll  mutants 
causes  efficient  lysis  of  this  host  and  the 
rapid  production  of  mature  phage.  Single 
infection  with  either  the  A  or  B  rll 
mutants  does  not  result  in  lysis,  although 
the  bacteria  are  no  longer  capable  of 
cellular  replication.  Phage  development 
within  the  host  cell  could  not  be  demon- 
strated, and  shaking  infected  cells  with 
chloroform  liberated  no  infective  particles. 

Infection  with  rll  A  phage  followed  at 
a  later  time  by  a  second  infection  with 
the  complementary  rll  B  phage  rendered 
the  K12X  cells  capable  of  subsequent  lysis 
and  phage  production.  After  an  elapsed 
time  of  several  hours  between  infections 
phage  production  still  resulted.  Evidently 
the  initial  infection  provides  the  host  cells 
with  certain  latent  characteristics  that 
can  be  stored  and  maintained.  This 
system  seemed  an  ideal  one  for  the 
investigation  of  the  effect  of  phage 
infection  upon  the  capacity  of  the  cell  to 
synthesize  host  and  viral  proteins. 

The  data  in  figure  52  illustrate  the 
effect  of  rll  phage  upon  the  induced 
synthesis    of    ^-galactosidase.     In    this 


20  30 

Time  in  minutes 


Fig.  52.  Time  course  of  induction  of  /3-galac- 
tosidase  in  K12X  E.  coli  after  infection  with  rll 
mutants  of  T4  phage.  (A  =  mutant  164;  B  = 
mutant  196.) 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


281 


experiment  an  exponentially  growing 
culture  of  K12X  cells  was  induced  for 
/3-galactosidase  synthesis  with  3.5  X  10~4 
M  methyl-thio-iS-D-galactoside  (TMG). 
Nine  minutes  later  rll  A  phage  particles 
(4  phages  per  cell)  were  added  to  the 
culture.  Figure  52  indicates  that  the 
0-galactosidase  synthesis  continues  for 
several  minutes  after  phage  infection. 
Enzyme  production  then  slows  down  or 
stops  and  later  resumes.  Plate  count 
analysis  revealed  that  99  per  cent  of  the 
cells  were  infected  with  the  rll  A  phage. 
Growth  of  the  culture  after  infection 
could  not  be  detected. 

In  figure  53  it  is  shown  that  these 
enzyme  kinetics  are  quite  similar  to  those 


/S-galactosidase 


until  they  decay  or  become  inactive.  It  is 
interesting  that  each  of  these  very 
different  treatments  of  the  culture  per- 
mits the  resumption  of  the  induced 
synthesis  of  enzyme.  Identical  results  are 
obtained  if  the  culture  is  infected  with 
rll  B  mutants  instead  of  the  rll  A  phage. 

The  temporary  inhibition  of  the  syn- 
thesis is  not  caused  by  low-molecular- 
weight  contaminant  material  contained  in 
the  added  phage  lysate.  Dialysis  of  the 
phage  suspension  does  not  reduce  the 
effectiveness  of  the  infection.  More 
significantly,  nondialyzed  preparations 
added  to  the  culture  in  which  no  trypto- 
phan was  present  caused  no  effect. 
Tryptophan  is  required  for  the  adsorption 
of  these  phages  to  the  host. 

The  synthesis  of  /3-galactosidase  can  be 
abruptly  stopped  after  the  initial  infec- 
tion with  rll  A  phage  if  the  comple- 
mentary rll  B  mutants  are  added  to  the 
culture  later.  These  kinetics  are  shown  in 
figure   54.   Preinduced   K12X   cells  were 


Fig.  53. 
induction. 


10         15        20       25 

Time  in  minutes 
Effect  of  glucose  on  0-galactosidase 


O) 

=5 
C 


c 
o 

> 

c 


o 
o 

CD 


C 

UJ 


observed  when  glucose  (10~2  M)  was 
added  to  K12  cells  after  the  induced 
synthesis  of  enzyme  had  been  initiated. 
The  infection  with  rll  A  phage  (or  the 
addition  of  glucose)  appears  to  stop  the 
synthesis  of  active  enzyme-forming  units. 
Those  already  present  at  the  time  of 
infection  continue  to  synthesize  enzyme 


Phage-A(orB) 


/3-galactosidase 


10 


20        25 


30 


Time  in  minutes 


Fig.  54.  Kinetics  of  /3-galactosidase  synthesis 
after  an  initial  infection  with  an  rll  mutant,  A 
(164  phage),  and  subsequent  infection  (36 
minutes  later)  with  rll  mutant  B  (196  bacterio- 
phage). 


282 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


first  infected  with  rll  A  phage  and  18 
minutes  later  with  rll  B  phage.  The 
immediate  cessation  of  synthesis  of  /3- 
galactosidase  is  observed  as  normal  phage 
development  begins.  Presumably,  not 
only  are  no  new  enzyme-forming  units 
synthesized  but,  more  significantly,  those 
previously  existing  and  functioning  after 
the  initial  infection  are  inactivated  within 
seconds. 

The  rapidity  with  which  phage  can  stop 
the  synthesis  of  /3-galactosidase  can  be 
seen  from  the  data  of  figure  55.  Wild-type 


sidase-forming  units.  It  is  also  difficult  to 
imagine  why  the  complete  phage  genome 
is  required  for  permanent  suppression  of 
host  protein  synthesis. 

The  information  in  the  injected  DNA 
of  one  of  the  rll  mutants  may  remain 
dormant  for  long  periods  but  can  be 
expressed  at  the  moment  of  entry  of  the 
second  rll  mutant  DNA.  Our  current 
research  is  directed  toward  determining 
how  and  where  such  information  is  main- 
tained and  the  mechanisms  by  which  it 
is  finally  expressed. 


10 


20        30        40        50 

Time  in  minutes 


Fig.  55.  The  kinetics  of  /3-galactosidase  in- 
duction in  K12X  E.  coli  after  infection  with  wild- 
type   T4  bacteriophage. 


T4  phage  were  added  to  a  preinduced 
culture  of  K12X.  Within  seconds  all 
synthesis  of  /3-galactosidase  ceased. 

It  is  extremely  puzzling  that  the 
injection  of  viral  DNA  into  the  host  cell 
should  cause  instantaneous  alterations  in 
the  capacity  of  the  cell  to  synthesize  host 
protein.  Ribosomes  are  presumably  the 
sites  for  protein  synthesis,  and  it  is 
extremely  difficult  to  imagine  a  direct 
action  of  injected  DNA  on  all  /3-galacto- 


Cell-Free  Synthesis 

In  spite  of  the  successful  use  of  cell-free 
systems  by  other  laboratories,  this  tech- 
nique has  had  little  application  here.  One 
particularly  important  use  of  such  sys- 
tems is  to  provide  an  assay  for  templates 
capable  of  synthesizing  a  biologically 
active  protein.  In  Year  Book  60  two 
apparently  successful  experiments  of  this 
type  were  reported.  Broken  cells,  in- 
capable of  synthesizing  /3-galactosidase, 
synthesized  a  small  quantity  of  the 
enzyme  when  incubated  with  purified 
RNA  extracted  from  induced  cells.  Nu- 
merous attempts  were  made  to  repeat  and 
extend  these  results,  but  no  RNA 
preparation  other  than  the  initial  one 
gave  positive  results. 

Doublet  Code 

In  1954  Gamow  formulated  as  a  coding 
problem  the  role  of  nucleic  acid  in 
specifying  the  order  of  amino  acids  in 
protein.  A  year  ago  an  experimental 
attack  on  the  coding  problem  became 
possible  when  Nirenberg  found  that 
synthetic  polynucleotides  could  serve  as 
templates  for  protein  synthesis.  This  year 
the  "code  letters"  for  19  of  the  20  amino 
acids  were  experimentally  determined. 
Almost  invariably  the  experimental  find- 
ings have  been  interpreted  in  terms  of  a 
"three-letter"  or  triplet  code. 

As  the  experiments  progressed  and  the 
code  letters  for  the  amino  acids  were 
recognized,  a  serious  difficulty  developed 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


283 


TABLE  26.     Calculation  of  Hypothetical  Template 


Proportions  in 
E.  coli  Proteins* 

Code          Common  U 

Proportion  of  Bases  Expected  in 
U                 G                 A 

Template 
C 

Ala 

14.7 

UCG 

14.7 

14.7 

14.7 

Arg 

7.0 

UCG 

7.0 

7.0 

7.0 

Asp 

14.4 

UAG 

14.4 

14.4 

14.4 

Cys 

0.9 

UUG 

0.9 

0.9 

0.9 

Glu 

15.6 

UAG 

15.6 

15.6 

15.6 

Gly 

12.0 

UGG 

12.0 

12.0 

His 

2.9 

UAC 

2.9 

2.9 

2.9 

lieu 

7.5 

UUA 

7.5 

7.5 

7.5 

Leu 

12.4 

UUC 

12.4 

12.4 

12.4 

Lys 

8.5 

UAA 

8.5 

8.5 

Met 

4.3 

UAG 

4.3 

4.3 

4.3 

Phe 

4.8 

uuu 

4.8 

4.8 

Pro 

5.8 

ucc 

5.8 

5.8 

Ser 

6.5 

UUC 

6.5 

6.5 

6.5 

Thr 

7.8 

UAC 

7.8 

7.8 

7.8 

Try 



UGG 



Tyr 

3.9 

UUA 

3.9 

3.9 

3.9 

Val 

10.0 

UUG 

10.0 

10.0 
50.8 

10.0 
90.9 

73.4 

139.0 

62.9 

Composition  of  template  (including  common 
Composition  of  template  (excluding  common 
Composition  of  E.  coli  50S  ribosomal  RNAf 

U) 
U) 

45.1 
18.3 
19.6 

21.8 
32.7 
33.5 

17.6 
26.4 
25.4 

15.1 
22.6 
21.5 

*  Data  of  Sueoka. 

f  Data  of  Midgley. 

in  the  triplet  interpretation.  If  the 
composition  of  the  product  protein  is 
known,  it  is  simple  to  calculate  the 
nucleotide  composition  of  the  template 
according  to  an  assumed  code.  Table  26 
shows  such  a  calculation  of  the  templates 
needed  to  direct  the  incorporation  of 
amino  acids  into  the  proteins  of  E.  coli. 
When  the  uridylic  acid  (U)  common  to 
all  the  triplet  code  words  is  included,  the 
templates  would  include  45  per  cent  U. 
No  RNA  of  this  nature  is  found  in  E.  coli. 
In  fact,  the  template  material  of  E.  coli 
may  be  DNA-like  or  ribosome-like  in 
composition,  but  clearly  not  of  high  U 
content.  The  same  difficulty  holds  with 
other  RNA's  that  can  act  as  templates. 
The  RNA  of  tobacco  mosaic  virus,  which 
acts  as  a  template  in  cell-free  systems, 
does  not  have  the  high  U  content  required 
by  the  present  triplet  code. 

This  failing  has  been  rationalized  on 
the    theory    that    the    code    is    highly 


degenerate  and  contains  many  other  still 
undiscovered  code  words  lacking  U. 

There  is  the  further  difficulty  that  a 
UC  polymer  containing  60  per  cent 
nonsense  words  (according  to  the  triplet 
interpretation)  is  very  effective  as  a 
template.  It  is  increasingly  difficult  to 
believe  that  enough  code  words  low  in  U 
have  remained  undiscovered  to  compen- 
sate for  the  high  U  content  of  the  words 
readily  discovered. 

If,  however,  the  common  U  of  the  code 
words  is  discarded  and  the  code  is 
considered  to  be  a  doublet,  these  major 
failings  disappear.  The  predicted  tem- 
plates for  a  number  of  bacteria  (table  27) 
are  strikingly  close  in  their  composition 
to  the  ribosomal  RNA. 

The  same  calculation  applied  to  the 
protein  of  six  plant  viruses  gives  predicted 
templates  somewhat  like  the  RNA  of  the 
corresponding  virus  (table  28).  Exact 
agreement  would  not  be  expected,  because 


284 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


TABLE  27.     Comparison 

of  RNA's 

GC  Content 

Type  of  RNA 

Mole 

per  cent 

Organism 

of  DNA, 

% 

U 

G 

A 

C 

B.  subtilis 

42 

Template* 

18.5 

32.1 

27.5 

21.8 

50Sf 

19.3 

32.0 

26.5 

22.5 

Newly  synthesized  f 

23.7 

27.3 

25.5 

23.5 

E.  coli 

50 

Template 

18.3 

32.7 

26.4 

22.6 

50S 

19.6 

33.5 

25.4 

21.5 

Newly  synthesized 

22.6 

29.5 

25.0 

22.9 

A.  aerogenes 

57 

Template 

18.1 

33.4 

25.2 

23.2 

50S 

21.2 

31.2 

25.6 

22.0 

Newly  synthesized 

21.5 

30.3 

24.8 

23.4 

Ps.  aeruginosa 

65 

Template 

17.4 

33.9 

24.5 

24.1 

50S 

21.3 

31.2 

26.3 

21.2 

Newly  synthesized 

20.5 

31.9 

21.4 

26.2 

*  Hypothetical  template  calculated  according  to  doublet  code  from  amino  acid  analyses  of  Sueoka. 
t  Observed  RNA  compositions,  Midgley. 


TABLE  28.     RNA  of  Viruses  Compared  with  Template  Calculated  by  Doublet  Code 


Measured  RNA  Composition 

TBSV 

PV 

CV 

TYMV 

SBMV 

TMV 

u 

25 

25 

30 

22 

25 

27 

G 

28 

24 

26 

17 

26 

26 

A 

28 

29 

26 

23 

26 

29 

C 

21 

22 

19 

38 

23 

19 

Calculated  Template 

TBSV 

PV 

CV 

TYMV 

SBMV 

TMV 

U 

21.8 

20.4 

27.0 

20.8 

21.2 

22.8 

G 

30.2 

27.5 

25.2 

21.4 

28.2 

28.4 

A 

21.4 

25.3 

19.0 

25.8 

23.4 

21.4 

C 

26.6 

26.8 

28.9 

31.9 

27.3 

27.4 

TBSV,  tomato  bushy  stunt  virus. 
PV,  polio  virus. 
CV,  cucumber  virus. 


TYMV,  turnip  yellow  mosaic  virus. 
SBMV,  southern  bean  mosaic  virus. 
TMV,  tobacco  mosaic  virus. 


only  a  part  of  the  product  of  the  template     are  rilled  and  that  all  the  ambiguities  lie 


appears  in  the  virus  protein. 

The  doublet  code  obtained  by  discard- 
ing the  common  U  of  the  triplet  code  is 
shown  in  table  29.  The  observed  degen- 
eracies were  omitted.  They  can  be 
attributed  to  errors  in  distinguishing 
leucine  from  isoleucine  and  valine,  since 
these  three  amino  acids  are  well  known  to 
be  confused  by  the  entry  mechanisms  in 
living  cells.  The  order  was  chosen  to  fit 
amino  acid  replacement  data. 

It  is  striking  that  all  the  possible  spaces 


on  the  purine  pairs. 

In  addition  to  correcting  the  failings  of 


TABLE  29.     Doublet  Code 

G 

A 

C 

U 

G 

gly 

try 

glu 
gluN 

ala 

val 

A 

asp 

aspN 

lys 
met 

thr 

ileu 

C 

arg 

his 

pro 

leu 

U 

cys 

tyr 

ser 

phe 

DEPARTMENT    OF   TERRESTRIAL    MAGNETISM  285 

the  triplet  code,  this  code  has  several  brain.   Visitors  who  carried  out  experi- 

other  features.   It  is  equivalent  to  the  mental  work  in  our  laboratory  include 

triplet  code  in  predicting  the  incorpora-  Dr.  McQuillen,  University  of  Cambridge, 

tion    of    any    amino    acid    relative    to  England;  Dr.  Hotta  and  Dr.  Van  Holde, 

phenylalanine,  since  the  ratio  XUU/UUU  University  of  Illinois;  Dr.  Sager,  Colum- 

=  XU/UU.  It  is  nearly  equivalent  again  bia  University;  Dr.  Brown,  Department 

in  correlating  amino  acid  replacements,  of  Embryology;  and  Dr.   Hendler,   Na- 

because  the  amino  acid  data  assign  U  to  tional  Institutes  of  Health.  In  addition, 

a  common  position  in  the  triplet  code,  we  have  carried  out  collaborative  experi- 

The  doublet  code  eliminates  the  unneces-  ments  with  Dr.  Nirenberg  of  the  National 

sary  U  and  with  it  a  large  set  of  possible  Institutes  of  Health, 
mutations  that  have  not  been  observed. 

The  doublet  code  is  not  usually  con-  r      1     . 
sidered  seriously  because  of  the  aimculty 

in  imagining  how  4  letters  could  provide  Figure    56    summarizes    our    present 

more  than  16  different  words.  Perhaps  conception   of  the   flow   of  material   in 

this  view  is  too  abstract,  and  the  proper  nucleic   acid  synthesis.   To  arrive   at   a 

question  is  how  many  of  the  possible  diagram  of  such  complexity  it  has  been 

configurations  of  any  doublet  are  suffi-  necessary  to  combine  evidence  from  a 

ciently  stable.  The  difficulty  can  still  be  variety  of  experimental  approaches.  The 

avoided  by  a  mixed  code  in  which  a  purine  boxes  enclosed  in  dashed  lines  show  how 

pair  indicates  the  beginning  of  a  3-letter  the    problem   has   been   separated   into 

word.    Alternatively,    the    purine    pairs  analyzable  parts,  and  the  titles  indicate 

could  carry  additional  information.  the  classes  of  evidence  used  to  establish 

At  present  no  certain  assignment  of  the  the  existence  of  various  elements  shown, 
coding  ratio  can  be  made.  The  pure  Low-molecular- weight  (acid-soluble)  corn- 
doublet  requires  a  mechanism  for  trans-  pounds  are  shown  by  clear  boxes,  and 
mitting  and  preserving  some  as  yet  macromolecular  (acid-precipitable)  by 
unrecognized  property  of  purine  bases;  crosshatched  boxes.  The  areas  of  the 
the  triplet  disagrees  with  well  known  elements  shown  on  the  diagram  corre- 
data.  Further  experiments  should  elimi-  spond  to  the  relative  quantity  of  each 
nate  one  or  the  other  or  both  in  the  near  present  in  a  steadily  growing  cell, 
future.  There  are  four  major  regions  on  the 

diagram.  Region  1  concerns  the  pool  of 

Cooperation  low-molecular- weight    RNA    precursors. 

The  Biophysics  Section  has  benefited  Although,  of  course,  many  chemical  steps 
greatly  from  collaboration  both  by  visits  and  a  large  number  of  pool  compounds 
of  our  staff  to  other  laboratories  and  by  are  involved,  one  dominant  feature 
visitors  who  carried  out  experiments  here,  becomes  clear  from  an  examination  of  the 
Bolton  spent  six  months  working  with  incorporation  of  labeled  nucleic  acid 
Dr.  R.  E.  F.  Matthews  in  New  Zealand  bases  into  the  pool  and  the  total  RNA. 
and  visiting  numerous  laboratories  in  A  major  fraction  of  the  compounds 
Australia  and  Japan.  Cowie  spent  three  entering  the  cell  are  rapidly  incorporated 
months  working  with  Dr.  Miranda  at  the  into  RNA,  bypassing  the  relatively  large 
Biophysics  Institute,  Rio  de  Janeiro,  nucleoside-phosphate  pools.  Labeled  corn- 
after  visiting  several  leading  South  Amer-  pounds  entering  these  pools  do,  however, 
ican  laboratories.  ultimately     become     incorporated     into 

We  have  continued  our  close  relation-  RNA. 

ship  with  Drs.  L.  B.  and  J.  B.  Flexner  and  Region  2  concerns  the  steps  in  ribosome 

follow   with   the   greatest   interest   their  synthesis,    or    assembly,     carrying    the 

experiments  on  protein  synthesis  in  the  process   only   up   to   the   30S   and    50S 


286 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


FLOW   DIAGRAM -NUCLEIC  ACID  SYNTHESIS  IN  E.  COL/ 

Areas  proportional  to  quantities 


Bypass  mechanism 

\      Interna! 
synthesis 


External 


base 


External 


orthophosphate 


MONO- 


NUCLEOSIDE -PHOSPHATE  . 
POOLS  |j 

J 

From  overall  tracer 
incorporation  kinetics 


\®. 


From  time  course 

of  P32  nucleotide 

composition 

17^ 1 


From  tracer  kinetics  using 
methylated  albumin  column 


From  tracer  kinetics  using  sedimentation  analysis 
and  DEAE  column  analysis 


Fig.  56.  Flow  diagram  for  nucleic  acid  synthesis  in  E.  coli.  The  areas  of  the  boxes  represent  the 
relative  quantities  of  the  different  elements  present  in  exponentially  growing  cells.  Open  boxes  rep- 
resent low-molecular-weight  (acid-soluble)  compounds,  and  the  crosshatched  boxes  represent  macro- 
molecular  fractions.  The  dashed  lines  show  how  the  pattern  has  been  separated  into  analyzable  parts, 
and  the  titles  indicate  the  classes  of  evidence  used. 


subunits  of  the  larger  (70S,  85S,  and 
100S)  ribosomes.  Studies  of  the  kinetics 
of  labeling  of  fractions  of  RNA  resolved 
by  sedimentation  analysis  and  DEAE 
column  chromatography  show  that  there 
is  a  set  of  sequential  steps  starting  with 
the  relatively  small  (14S)  eosome  and 
progressing  to  the  completed  ribosome. 

It  appears,  however,  that  there  are  two 
distinct  classes  of  RNA  of  about  14S, 
which  are  the  first  to  be  labeled  and  are 
not  resolvable  from  each  other  by  sedi- 
mentation analysis  or  column  chro- 
matography. This  is  shown  by  the  fact 
that  the  apparent  nucleotide  composition 
of  this  fraction  (pulse  P32  labeling)  does 
not  correspond  to  that  of  ribosomal  RNA. 
Further  special  procedures  permit  its 
partial  resolution  into  two  subfractions 


having,  respectively,  a  composition  sim- 
ilar to  that  of  DNA  (substituting  uracil 
for  thymine)  and  a  composition  similar 
to  that  of  ribosomal  RNA.  These  sub- 
fractions  have  been  termed  D-RNA  and 
R-RNA,  and  the  nucleotide  composition 
of  the  14S  fraction  in  a  number  of 
bacteria  indicates  that  their  relative 
quantities  are  in  the  ratio  of  1  to  2.  The 
D-RNA  has  been  placed  on  the  diagram 
in  box  3. 

Since  the  nucleotide  composition  of  the 
D-RNA  differs  from  that  of  any  major 
RNA  fraction,  and  it  is  rapidly  labeled, 
it  must  at  least  in  part  be  degraded  to 
low-molecular-weight  fragments.  The  fate 
of  these  degradation  products  is  some- 
what uncertain.  However,  the  entry  of 
labeled  compounds  into  the  DNA  and 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


287 


S-RNA  is  delayed,  and  very  likely  these 
nucleic  acids  are  in  part  synthesized  from 
the  degradation  products.  This  evidence 
combined  with  various  other  arguments 
has  led  to  the  pattern  of  flows  shown  in 
box  4. 

From  the  studies  summarized  in  this 
diagram  the  quantity  of  newly  synthe- 
sized RNA  and  its  distribution  between 
the  D-RNA  and  the  R-RNA  components 
can  be  estimated.  Two  per  cent  of  the 
total  RNA  is  in  the  newly  formed 
R-RNA,  and  1  per  cent  is  D-RNA.  Both 
components  are  labeled  with  the  same 
time  constant  of  2J/£  minutes. 

A  number  of  lines  of  evidence  indicate 
that  this  newly  formed   RNA  fraction 


length  of  600  nucleotides.  According  to 
either  model  the  rate  of  amino  acid 
incorporation  is  rapid,  2-10  seconds  to 
complete  a  peptide  strand,  or  10-30 
milliseconds  per  amino  acid  if  they  are 
added  sequentially. 

The  rate  of  RNA  synthesis  can  also  be 
related  to  the  sites  for  its  synthesis.  If 
these  are  assumed  to  be  the  cells'  DNA 
there  are  15,000  sites  of  600  nucleotide 
lengths  per  nucleus.  The  rate  of  D-RNA 
synthesis  corresponds  to  the  production 
of  2  copies  of  the  entire  complement  of 
DNA  during  each  generation.  Thus  the 
rate  per  site  need  only  be  1  copy  per 
30  minutes. 

A  much  faster  synthesis  of  templates 


TABLE  30.     Calculated  Rate  of  Synthesis  per  Template 


Template  Material 

Ed 

+  ER 

Ed 

Coding  ratio 

2 

3 

2 

3 

Templates  per  cell 

5000 

5000 

1666 

1666 

Length  of  peptide 

300 

200 

300 

200 

Time  for  peptide  synthesis,  seconds 

10 

7 

3 

2 

Milliseconds  per  peptide  bond 

33 

33 

10 

10 

Average  peptides  per  template 

15 

22 

45 

67 

Average  template  is  assumed  to  contain  600  nucleotides. 


carries  the  template  for  protein  synthesis,  must  occur  during  the  induction  of  an 
Its  lifetime  is  the  same  as  that  of  the  enzyme.  An  increase  in  the  rate  of 
enzyme-forming  unit  for  /3-galactosidase,     /5-galactosidase  synthesis  can  be  observed 


and  Nirenberg  has  found  that  this 
fraction  is  the  most  active  in  stimulating 
protein  synthesis  in  his  cell-free  system. 
There  is  no  clear  evidence,  however, 
whether  it  is  the  D-RNA  fraction  (as  was 


within  30  seconds  after  the  inducer  is 
added.  As  an  induced  enzyme  may 
account  for  2  to  5  per  cent  of  the  cells' 
protein  when  fully  induced  we  may 
assume  that  2  per  cent  of  the  templates 


postulated  by  Jacob  and  Monod)  or  the     are  required  for  that  particular  enzyme. 


R-RNA   fraction   or   both   that   act   as 
templates. 

In  table  30  we  have  calculated  the  rate 
of  synthesis  per  template.  Since  it  is  not 
certain  which  fraction  provides  the  tem- 


If  all  these  templates  are  made  at  one  of 
the  15,000  sites  the  rate  per  site  must  be 
600  copies  per  generation  or  1  copy  per  6 
seconds.  The  factor  of  300  from  the 
minimum  rate  to  the  maximum  corre- 


plate  or  whether  the  coding  ratio  is  2  or  3,  sponds  to  the  ratio  from  the  uninduced  to 

the  calculation  has  been  carried  out  for  the  induced  level  of  enzyme, 

two  cases.  For  purposes  of  calculation  the  It  seems  reasonable  that  the  templates 

template  has  been  assumed  to  have  a  for  many  of  the   cells'   active   enzymes 


288 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


would  be  synthesized  at  an  intermediate 
rate  between  these  extremes.  If  1000  sites 
(1/15  of  the  total)  were  active  and 
produced  an  average  of  60  copies  per 
generation  this  rate  would  be  sufficient 
to  account  for  the  production  of  the 
R-RNA. 

The  questions  as  to  which  fraction  of 
the  RNA  serves  as  template  and  how  the 
activity  as  a  template  is  terminated  still 
require  experimental  answers.  At  present 
we  can  only  speculate  and  try  to  find  the 
most  reasonable  hypothesis  consistent 
with  the  known  facts. 

One  other  fact  is  very  pertinent  in 
these  speculations.  The  proteins  of  a 
series  of  bacteria  having  widely  different 
DNA's  and  correspondingly  different 
D-RNA's  have  very  nearly  the  same 
amino  acid  contents.  Thus  if  the  D-RNA 
acts  as  template  there  must  be  a  highly 
degenerate  code  to  allow  quite  different 


templates  to  produce  very  similar  prod- 
ucts. The  code  as  now  worked  out  in 
cell-free  systems  has  not  shown  the 
needed  multiplicity  of  symbols. 

The  R-RNA  fraction,  on  the  other 
hand,  is  constant  in  composition,  so  that 
a  single  nondegenerate  code  would  suffice 
if  the  R-RNA  were  the  template  for  the 
greater  part  of  the  cells'  protein.  The 
doublet  code  does  provide  a  correlation 
between  the  amino  acid  composition  of 
the  protein  and  the  nucleotide  compo- 
sition of  the  R-RNA. 

If  it  turns  out  that  the  R-RNA  fraction 
can  act  as  template,  the  termination  of 
template  activity  could  be  ascribed  to  the 
covering  of  the  template  during  the 
course  of  ribosome  synthesis.  More 
definite  answers,  based  on  experiments 
instead  of  speculation,  should  be  available 
in  the  near  future. 


IMAGE  TUBES  FOR  LARGE  TELESCOPES 


The  activities  of  our  Department 
include  a  vigorous  participation  in  the 
development  of  photoelectric  image  tubes 
to  extend  the  sensitivity  and  range  of 
medium-sized  and  large  telescopes,  based 
on  the  fact  that  modern  photoelectric 
surfaces  are  more  than  a  hundred  times 


more  sensitive  than  the  best  photographic 
emulsions  used  in  astronomy.  The  com- 
plete record  of  these  activities,  including 
the  full-time  work  of  Dr.  W.  K.  Ford,  Jr., 
of  the  DTM  staff,  and  others,  is  found  in 
the  report  of  the  Committee  on  Image 
Tubes  for  Telescopes,  which  follows. 


BIBLIOGRAPHY 


Adams,  H.  S.,  J.  D.  Fox,  N.  P.  Heydenburg,  and 
G.  M.  Temmer,  Angular  distributions  of  4.43- 
Mev  gamma  radiation  from  Cn(p,p'y)C12, 
Phys.  Rev.,  124,  1899,  1961. 

Aldrich,  L.  T.,  see  Meyer,  R.  P. 

Asada,  T.,  Conferencias  sobre  sismologia  mo- 
derna,  Arequipa,  Peru  (report),  Instituto 
Geoffsico  de  la  Universidad  de  San  Agustin, 
104  pp.,  1961. 

Asada,  T.,  see  also  Rodriguez  B.,  A.,  and  Stein- 
hart,  J.  S. 


Baum,  W.  A.,  J.  S.  Hall,  L.  L.  Marton,  and  M.  A. 
Tuve,  Committee  on  Image  Tubes  for  Tele- 
scopes, Carnegie  Inst.  Wash.  Year  Book  60, 
341-342,  1961. 

Blieden,  H.  R.,  see  Warsh,  K.  L. 

Boezi,  J.  A.,  and  D.  B.  Cowie,  Kinetic  studies  of 
/3-galactosidase  induction,  Biophys,  J.,  1, 
639-647,  1961. 

Bonini,  W.  E.,  see  Meyer,  R.  P.,  and  Steinhart, 
J.  S. 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM 


289 


Britten,  R.  J.,  Hydrolysis  of  RNA  by  lead 
acetate,  Compt.  Rend.  Trav.  Lab.  Carlsberg,  82, 
371-380,  1962. 

Britten,  R.  J.,  and  B.  J.  McCarthy,  Analysis  of 
the  kinetics  of  tracer  incorporation  in  growing 
cells,  Biophys.  J.,  2,  49-55,  1962. 

Britten,  R.  J.,  B.  J.  McCarthy,  and  R.  B. 
Roberts,  The  synthesis  of  ribosomal  protein 
and  the  assembly  of  ribosomes,  Biophys.  J., 

2,  83-93,  1962. 

Britten,  R.  J.,  see  also  McCarthy,  B.  J.,  and 
Roberts,  R.  B. 

Burke,  B.  F.,  Radio  frequency  radiometry  of 
the  planets,  in  The  Atmospheres  of  Mars  and 
Venus,  National  Academy  of  Sciences- 
National  Research  Council  Publ.  944,  pp. 
76-79,  1962. 

Burke,  B.  F.,  Radio  observations  of  Jupiter,  I, 
in  Planets  and  Satellites.  The  Solar  System,  vol. 

3,  pp.  473-499,  edited  by  G.  P.  Kuiper  and 
Barbara  M.  Middlehurst,  University  of 
Chicago  Press,  1961. 

Burke,  B.  F.,  and  J.  W.  Firor,  405-Mc  observa- 
tions of  the  galactic  center  (abstract),  Astron. 
J.,  66,  39,  1961. 

Cowie,  D.  B.,  see  Boezi,  J.  A.,  and  Halvorson, 
H.  O. 

Davis,  G.  L.,  G.  R.  Tilton,  and  G.  W.  Wetherill, 
Mineral  ages  from  the  Appalachian  province 
in  North  Carolina  and  Tennessee,  J.  Geophys. 
Res.,  67,  1987-1996,  1962. 

Fahlquist,  D.  A.,  see  Meyer,  R.  P.,  and  Steinhart, 
J.  S. 

Firor,  J.  W.,  and  H.  Zirin,  Observations  of  five 
ionization  stages  of  iron  in  the  solar  corona, 
Astrophys.  J.,  185,  122-137,  1962. 

Firor,  J.  W.,  see  Burke,  B.  F.,  and  Kundu,  M.  R. 

Forbush,  S.  E.,  G.  Pizzella,  and  D.  Venkatesan, 
The  morphology  and  temporal  variations  of 
the  Van  Allen  radiation  belt,  October  1959- 
December  1960  (report),  Department  of 
Physics  and  Astronomy,  State  University  of 
Iowa,  Iowa  ChVy,  Iowa,  39  pp.,  illustrated, 
1962. 

Forbush,  S.  E.,  D.  Venkatesan,  and  C.  E. 
Mcllwain,  Intensity  variations  in  outer  Van 
Allen  radiation  belt,  J.  Geophys.  Res.,  66, 
2275-2287,  1961. 

Forbush,  S.  E.,  see  Venkatesan,  D. 

Fox,  J.  D.,  see  Adams,  H.  S. 

Frick,  G.,  see  Tuve,  M.  A. 

Gast,  P.  W.,  see  Wetherill,  G.  W. 

Green,  R.,  and  J.  S.  Steinhart,  Maine  seismic 
experiment:  the  apparent  velocity  of  the  P- 
phase   across   short   spreads    (abstract),   Am. 


Geophys.  Union  Program,  43rd  Ann.  Meeting, 
Washington,  D.  C,  p.  33,  April  25-28,  1962. 

Hall,  J.  S.,  see  Baum,  W.  A. 

Halvorson,  H.  O.,  and  D.  B.  Cowie,  Metabolic 
pools  of  amino  acids  and  protein  synthesis  in 
yeast,  in  the  Symposium  on  Membrane  Trans- 
port and  Metabolism,  pp.  479-487,  Prague, 
Publishing  House  of  Czechoslovak  Academy  of 
Sciences,  1961. 

Hart,  S.  R.,  The  use  of  hornblendes  and  pyroxenes 
for  K-Ar  dating,  /.  Geophys.  Res.,  66,  2995- 
3001,  1961. 

Heydenburg,  N.  P.,  see  Adams,  H.  S. 
Howell,  B.  F.,  Jr.,  see  Meyer,  R.  P.,  and  Stein- 
hart, J.  S. 

Kouvo,  O.,  see  Wetherill,  G.  W. 

Kundu,  M.  R.,  and  J.  W.  Firor,  Interferometric 
studies  of  type  IV  solar  bursts  of  continuum 
radiation  on  340  and  87  Mc/s,  Astrophys.  J., 
184,  389-393,  1961. 

Lenhoff,  H.  M.,  Digestion  of  protein  in  Hydra  as 
studied  using  radioautography  and  fractiona- 
tion by  differential  solubilities,  Exptl.  Cell. 
Res.,    23,     335-353,  1961. 

McCarthy,  B.  J.,  The  effects  of  magnesium 
starvation  on  the  ribosome  content  of  Esche- 
richia coli,  Biochim.  Biophys.  Acta,  55,  880- 
888,  1962. 

McCarthy,  B.  J.,  and  R.  J.  Britten,  The  in- 
corporation of  C14-uracil  into  the  metabolic 
pool  and  RNA,  Biophys.  J.,  2,  35-47,  1962. 

McCarthy,  B.  J.,  R.  J.  Britten,  and  R.  B. 
Roberts,  The  synthesis  of  ribosomal  RNA, 
Biophys.  J.,  2,  57-82,  1962. 

McCarthy,  B.  J.,  see  also  Britten,  R.  J. 

McClure,  F.  T.,  see  Roberts,  R.  B. 

Mcllwain,  C.  E.,  see  Forbush,  S.  E. 

Marton,  L.  L.,  see  Baum,  W.  A. 

Meyer,  R.  P.,  J.  S.  Steinhart,  B.  F.  Howell,  Jr., 
W.  E.  Bonini,  D.  A.  Fahlquist,  and  L.  T- 
Aldrich,  Cooperative  Maine  experiment  in 
crustal  seismology:  methods  and  application 
of  fixed  linear  recording  arrays  to  crustal 
measurements  (abstract),  Am.  Geophys.  Union 
Program,  43rd  Ann.  Meeting,  Washington, 
D.  C.,  p.  32,  April  25-28,  1962. 

Meyer,  R.  P.,  see  also  Steinhart,  J.  S. 

Pizzella,  G.,  see  also  Forbush,  S.  E.,  and  Ven- 
katesan, D. 

Roberts,  R.  B.,  Alternative  codes  and  templates, 
Proc.  Natl.  Acad.  Sci.  U.  S.,  48,  897-900,  1962. 

Roberts,  R.  B.,  R.  J.  Britten,  and  F.  T.  McClure, 
A  model  for  the  mechanism  of  enzyme  induc- 
tion, Biophys.  J.,  1,  649-656,  1961. 


290 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


Roberts,  R.  B.,  see  also  Britten,  R.  J.,  and 
McCarthy,  B.  J. 

Rodriguez  B.,  A.,  T.  Asada,  and  M.  A.  Tuve,  A 
note  on  seismic  wave  attenuation  in  the 
Andes  (abstract),  Am.  Geophys.  Union  Pro- 
gram, 43rd  Ann.  Meeting,  Washington,  D.  C, 
p.  35,  April  25-28,  1962. 

Rodriguez  B.,  A.,  see  also  Steinhart,  J.  S.,  and 
Tuve,  M.  A. 

Salgueiro  P.,  D.  R.,  see  Tuve,  M.  A. 

Steinhart,  J.  S.,  and  R.  P.  Meyer,  Minimum 
statistical  uncertainty  of  the  seismic  refrac- 
tion profile,  Geophysics,  26,  574-587,  1961. 

Steinhart,  J.  S.,  R.  P.  Meyer,  B.  F.  Howell,  Jr., 
D.  A.  Fahlquist,  W.  E.  Bonini,  and  T.  Asada, 
Maine  seismic  experiment:  general  crustal 
results  (abstract),  Am.  Geophys.  Union  Pro- 
gram, 4^rd  Ann.  Meeting,  Washington,  D.  C, 
p.  32,  April  25-28,  1962. 

Steinhart,  J.  S.,  A.  Rodriguez  B.,  and  T.  Asada, 
Near  earthquakes  and  crustal  structure  in 
southern  Peru  (abstract),  Geol.  Soc.  Am., 
Program  58th  Ann.  Meeting,  University  of 
Southern  California,  Los  Angeles,  California, 
p.  67,  April  16-18,  1962. 

Steinhart,  J.  S.,  see  also  Green,  R.,  and  Meyer, 
R.  P. 

Temmer,  G.  M.,  On  the  possibility  of  resonant 
transfer  processes  in  nuclear  reactions,  Physics 
Letters,  1,  10-12,  1962. 

Temmer,  G.  M.,  see  also  Adams,  H.  S.,  and 
Warsh,  K.  L. 

Tilton,  G.  R.,  see  Davis,  G.  L.,  and  Wetherill, 
G.  W. 

Tuve,  M.  A.,  A.  Rodriguez  B.,  D.  R.  Salgueiro 
P.,  and  G.  Frick,  Local  earthquake  special 
network  in  the  Andes  (abstract),  Am.  Geophys. 
Union  Program,  43rd  Ann.  Meeting,  Washing- 
ton, D.  C,  p.  35,  April  25-28,  1962. 

Tuve,  M.  A.,  see  also  Baum,  W.  A.,  and 
Rodriguez  B.,  A. 


Venkatesan,  D.,  G.  Pizzella,  and  S.  E.  Forbush, 
Time  variations  in  the  outer  radiation  zone, 
October  1959  through  December  1960  (ab- 
stract), Am.  Geophys.  Union  Program,  43rd 
Ann.  Meeting,  Washington,  D.  C,  p.  90,  April 
25-28,  1962. 

Venkatesan,  D.,  see  also  Forbush,  S.  E. 

Warsh,  K.  L.,  H.  R.  Blieden,  and  G.  M.  Temmer, 
Excitation  curve  and  angular  distributions  of 
the  F19(p,  a0)O16  reaction  (abstract),  Bull.  Am. 
Phys.  Soc,  [2]  7,  300,  1962. 

Wetherill,  G.  W.,  O.  Kouvo,  G.  R.  Tilton,  and 
P.  W.  Gast,  Age  measurements  on  rocks  from 
the  Finnish  Precambrian,  J.  Geol.,  70,  74-88, 
1962. 

Wetherill,  G.  W.,  see  also  Davis,  G.  L. 

Zirin,  H.,  see  Firor,  J.  W. 


Major  Publications 

Cosmic-Ray  Results.  Huancayo,  Peru,  January 
1956-December  1959;  Cheltenham,  Maryland, 
January  1956-October  4,  1956,  Fredericks- 
burg, Virginia,  October  5,  1956-December 
1959;  Christchurch,  New  Zealand,  January 
1956- June  1959;  Godhavn,  Greenland,  Jan- 
uary 1954- July  1959;  Universidad  Nacional 
de  Mexico,  July  1957-December  1958.  By 
L.  Beach  and  S.  E.  Forbush.  Carnegie  Inst. 
Wash.  Publ.  175,  vol.  21.  Quarto,  v  +  226  pp., 
221  tables.  1961. 

Equatorial  Electrojet  in  Peru.  By  S.  E.  Forbush 
and  Mateo  Casaverde.  Carnegie  Inst.  Wash. 
Publ.  620.  Quarto,  v  +  135  pp.,  98  tables. 
1961. 

Explosion  Studies  of  Continental  Structure.  By 
J.  S.  Steinhart  and  R.  P.  Meyer,  with  contri- 
butions by  W.  E.  Bonini,  T.  Jefferson  Smith, 
and  G.  P.  Woollard,  Carnegie  Inst.  Wash. 
Publ.  622.  Quarto,  xiii  +  409  pp.,  142  figs.,  2 
plates  in  back  cover  pocket.  1961. 


DEPARTMENT   OF   TERRESTRIAL   MAGNETISM 


291 


PERSONNEL 

Director 
M.  A.  Tuve 


Staff  Members 


L.  T.  Aldrich1 
E.  T.  Bolton2 
R.  J.  Britten 
B.  F.  Burke 
D.  B.  Cowie 
J.  W.  Firor3 
S.  E.  Forbush4 
W.  K.  Ford,  Jr. 


S.  R.  Hart5 
N.  P.  Heydenburg6 
B.  J.  McCarthy 
R.  B.  Roberts 
T.  J.  Smith7 
J.  S.  Steinhart8 
G.  M.  Temmer6 
H.  W.  Wells9 


Section  Chairmen 


Biophysics:  R.  B.  Roberts 


Radio  Astronomy:  B.  F.  Burke  (from  Novem- 
ber 1,  1961) 
Earth's  Crust:  L.  T.  Aldrich1  Theoretical  Geophysics:  S.  E.  Forbush4 

Nuclear  Physics:  N.  P.  Heydenburg6 


Fellows  and  Associates 


T.  Asada,  Geophysical  Institute,  Tokyo 
University,  Tokyo,  Japan  (through  Febru- 
ary 1962). 

L.  Brown,  University  of  Basel,  Basel,  Switzer- 
land (from  September  1961). 

G.  Frick,  D.I.C.A.,  Universidad  de  Chile, 
Antofagasta,  Chile. 

R.  Green,  University  of  Tasmania,  Hobart, 
Tasmania  (September  1961-May  1962). 

R.  C.  Hall,  University  of  Indiana  (June  1962). 


S.  R.  Hart,  Massachusetts  Institute  of  Tech- 
nology (through  August  1961). 

J.  E.  Midgley,  Oxford  University,  England. 

A.  Rodriguez  B.,  Instituto  Geofisico,  Univer- 
sidad de  San  Agustin,  Arequipa,  Peru. 

H.  R.  Rojas,  Observatoire  de  Paris,  Meudon, 
Paris,  France  (through  June  15,  1962). 

H.  Rudin,  University  of  Basel,  Basel,  Switzer- 
land (from  February  1962). 

R.  Salgueiro,  Instituto  Tecnol6gico  Boli- 
viano, La  Paz,  Bolivia. 


1  Guest  investigator  at  Geological  and  Mineralogical  Institute,   University  of  Kyoto,  Kyoto, 
Japan  (from  February  1962). 

2  Visiting  investigator  at  universities  in  Canberra  and  Adelaide,  Australia,  and  Osaka,  Japan, 
July-December  1961. 

3  Through  September  15,  1961. 

4  Guest  investigator  at  Department  of  Electronics,  Royal  Institute  of  Technology,  Stockholm, 
Sweden,  through  September  1961. 

5  From  September  1,  1961. 

6  Professor  at  Florida  State  University,  Tallahassee,  Florida. 

7  From  June  1,  1962. 

8  From  July  1,  1961. 

9  On  leave  of  absence  to  serve  as  State  Department  Scientific  Attache  stationed  in  Rio  de  Janeiro, 
Brazil,  through  April  1962. 


292 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


Collaborators  and  Visiting  Investigators 


H.  Alvarez,  University  of  Michigan  (July- 
September  1961). 

H.  Baadsgaard,  University  of  Alberta,  Ed- 
monton, Canada  (April- June  1962). 

R.  Cabre,  S.  J.,  Observatorio  San  Calixto,  La 
Paz,  Bolivia. 

Mateo  Casaverde,  Instituto  Geofisico  del 
Peru,  Lima,  Peru. 

H.  L.  Cesar,  Instituto  de  Pesquisas  Radio- 
activas,  Belo  Horizonte,  Brazil  (December 
1961-January  1962). 

B.  R.  Doe,  U.  S.  Geological  Survey  (part 
time,  December  1961-April  1962). 

L.  N.  Edmunds,  Jr.,  Princeton  University 
(June  1962). 

L.  B.  and  J.  B.  Flexner,  University  of  Penn- 
sylvania (part  time). 

A.  A.  Giesecke,  Jr.,  Instituto  Geofisico  del 
Peru,  Lima,  Peru. 

I.  Hayase,  University  of  Kyoto,  Japan 
(through  November  1961). 

R.  W.  Hendler,  National  Institutes  of  Health 
(part  time). 

J.  W.  Hollinger,  George  Washington  Univer- 
sity (December  1961-May  1962). 

Y.  Hotta,  University  of  Illinois  (July  1961). 

A.  Kamitsuki,  University  of  Kyoto,  Kyoto, 
Japan  (January  1962). 


T.  Krishnan,  CSIRO,  Sydney,  Australia 
(September  1961). 

F.  T.  McClure,  Applied  Physics  Laboratory, 
Johns  Hopkins  University  (part  time). 

K.  McQuillen,  University  of  Cambridge, 
Cambridge,  England  (July  1961). 

M.  Nirenberg,  National  Institutes  of  Health 
(part  time). 

H.  Rudin,  University  of  Basel,  Basel,  Switzer- 
land. 

G.  Saa,  S.J.,  Seismological  Observatory, 
Antofagasta,  Chile. 

R.  Sager,  Columbia  University  (December 
1961). 

P.  Salas  del  Carpio,  Instituto  Geofisico, 
Universidad  de  San  Agustin,  Arequipa, 
Peru. 

W.  W.  Salisbury,  Microwave  Power  Labora- 
tory, Garland,  Texas  (July  1961). 

U.  Schmucker,  Scripps  Institution  of  Oceanog- 
raphy (February  1962). 

G.  Schwachheim,  Centro  Brasileiro  de  Pes- 
quisas Fisicas,  Rio  de  Janeiro,  Brazil 
(January-April  1962). 

K.  E.  Van  Holde,  University  of  Illinois 
(September-October  1962). 

A.  H.  G.  Vieira,  Escola  Politecnica,  Sao 
Paulo,  Brazil  (January-March  1962). 


Research  Assistants 


J.  B.  Doak 
E.  T.  Ecklund 


W.  E.  Scott 


P.  A.  Johnson 
C.  A.  Little,  Jr. 


Laboratory  Assistants 


Mrs.  L.  Beach 
S.  J.  Buynitzky 
G.  R.  Poe 


Miss  P.  Roddy  (through  June  15,  1962) 
Mrs.  A.  Shirven 
Miss  E.  Stern 


Office 


Chief,  Fiscal  Section:  Miss  H.  E.  Russell 

Office  Manager:  W.  N.  Dove 

Librarian:  Mrs.  L.  J.  Pro  thro  (part  time) 

Accounting  Assistant: 


Secretary:  Mrs.  C.  Ator  (part  time) 
Stenographer:  Mrs.  D.  B.  Dillin 
Typist:  Mrs.  M.  T.  Sheahan  (part  time) 
Miss  G.  J.  Johnston 


DEPARTMENT    OF   TERRESTRIAL    MAGNETISM  293 

Shop 

Chief  of  Section:  W.  F.  Steiner  Machinist-Instrument  Maker:  D.  E.  Mossor 

Senior    Instrument    Makers:    B.    J.  Haase  (from  November  1,  1961) 

(retired  June  30,  1962),  J.  G.  Lorz  Machinist:  F.  J.  Caherty 

Instrument  Maker:  M.  Seemann 


Buildings  and  Grounds 

Carpenter,  Buildings  and  Maintenance  Fore-      Assistant  Caretakers:  S.  Gawrys,  S.  Swant- 

man:  L.  J.  Haber  kowski 

Caretaker:  E.  Quade 


Part-Time  and  Temporary  Employees 

Wayne  Ator  David  James 

John  Brauer  John  Randall 

Manuel  Buchwald  James  Roddy 

Thomas  Garrahan  John  Roddy 

Barry  Goss  Kai  Schwarz 


Special  Project  Appointee 
L.  W.  Fredrick  (Image  Tubes) 


Committee  on  Image  Tubes 

for  Telescopes 

Cooperative  Project  of  Mount  Wilson  and  Palomar  Observatories 

Department  of  Terrestrial  Magnetism,  Lowell  Observatory 

National  Bureau  of  Standards,  and  United  States  Naval  Observatory 


W.  A.  Baum 

Mount  Wilson  and  Palomar  Observatories 

John  S.  Hall 

Director,  Lowell  Observatory 
Flagstaff,  Arizona 

L.  L.  Marton 

National  Bureau  of  Standards 

M.  A.  Tuve  (Chairman) 

Department  of  Terrestrial  Magnetism 


Carnegie  Institution  of  Washington  Year  Book  61,  1961-1962 


Frontispiece 


Image  Tubes 


The  DTM  spectrograph  on  the  Morgan  24-inch  reflector,  Lowell  Observatory,  Flagstaff, 
Arizona.  A  cascaded  image  intensifier  system  is  mounted  on  the  left  side  of  the  spectrograph. 
This  unit  consists  of  the  spectrograph  camera  lens  (just  below  the  mounting  flange),  a  cascade 
tube  and  focusing  magnet  (just  above  the  flange),  the  relay  lens  system  and  plateholder  (shown 
here  with  an  eyepiece  for  viewing  the  phosphor  screen). 


INTRODUCTION 

Improved      samples      of     magnetically  tested.  Each  of  these  tubes  provided  a 

focused   image   tubes   were   tested   and  gain  of  roughly  5  over  direct,  unaided 

evaluated  by  the  Carnegie  Image  Tube  plates. 

Committee  during  the  report  year.  We  While  observations  at  the  telescope 
have  found  that  these  tubes  tend  to  have  demonstrated  the  reliability  and  effective- 
better  operating  characteristics  than  we  ness  of  these  devices,  laboratory  tests 
had  hoped  for  three  years  ago.  were  helpful  in  distinguishing  the  relative 
Our  tests  of  these  tubes  have  been  made  merits  of  the  various  tubes.  In  particular, 
primarily  with  the  DTM  spectrograph  on  resolution  and  screen  quality  were  evalu- 
the  Morgan  24-inch  reflector  at  Lowell  ated  with  an  excellent  test  target  designed 
Observatory.  In  October  1961  and  April  by  Dr.  Baum. 

1962  a  blue-sensitive  cascaded  tube  made  On  the  basis  of  these  spectrographic 

by  ITT  Laboratories  was  used  to  obtain  and    laboratory    tests,    the    Committee 

good-quality  spectra,  but  exposures  were  believes  that  both  the  mica- window  and 

limited    to     20    minutes    by    spurious  the  cascaded  tubes  will  have  wide  appli- 

emission.  In  March  1962  a  mica-window  cation   in   astronomy   because    of   their 

tube  made  by  RCA  proved  to  be  highly  advantages    over    conventional    photog- 

successful.  It  gave  very  good  definition  raphy.   To  help  meet  the  high  cost  of 

over  35  mm  of  spectrum.  In  April  and  completing    the    development    of    these 

early  May  an  RCA  cascaded  tube  with  tubes  the  Committee  applied  for  and  has 

unusually  high  sensitivity  was  tested  and  been   awarded   a   new   grant   from   the 

compared    with    the    tubes    previously  National  Science  Foundation. 

APPLICATION   OF  IMAGE   TUBES 

Image  intensiflers  are  of  importance  observing  conditions  as  well  as  in  the 

because  of  the  exceedingly  low  light  levels  laboratory.    Good    photocathodes,    once 

with    which    astronomers    must    work,  they  are  made,  are  well  protected  and 

Image  intensiflers  make  use  of  the  high  have  an  indefinitely  long  life.  Many  of 

quantum  efficiency  of  the  photoelectric  the  electrostatic  tubes  evaluated  by  this 

process    to    allow    information    to    be  Committee  three  or  four  years  ago  are 

recorded  more  rapidly  on  conventional  still  quite  sensitive  and  are  still  being  used 

photographic    emulsions    or,    in    other  on  some  problems  such  as  double  star 

applications,  to  allow  a  better  measure-  photography,  where  the  smallness  of  the 

ment  of  light  intensity  to  be  made.  field    in    good    focus    is    not    a    serious 

The  advantages  of  the  photoelectric  handicap, 
type  of  image  converter  have  been  The  first  application  that  comes  to 
demonstrated  primarily  by  Professor  mind,  perhaps,  for  an  image  intensifier  is 
Lallemand  and  his  colleagues.  In  his  the  photographing  of  fields  of  faint  stars. 
device,  however,  the  operation  of  the  tube  Here  the  brightness  of  the  night  sky 
occupies  an  undue  amount  of  the  astron-  limits  the  exposure  that  can  be  made  with 
omer's  time.  This  Committee  indeed  was  a  fast  optical  system.  To  see  fainter  stars 
organized  to  explore  possible  ways  of  a  longer-focal-length  telescope  is  required, 
providing  a  simple  and  reliable  alternative  but  with  present  instruments  this  in- 
to the  Lallemand  tube.  volves  going  to  focal  ratios  too  large  for 

The  image  tubes  with  which  we  are  the   rather   low    quantum   efficiency    of 

now  concerned  are  completely  sealed  off  photographic  plates, 

and  are  reliable  in  operation  under  actual  With  image  intensiflers,  such  as  the 

297 


298  CARNEGIE     INSTITUTION     OF     WASHINGTON 

cascaded  tubes  made  by  RCA  and  ITT,  the  signal-to-noise  ratio  is  improved, 
long-focal-length  telescopes  too  slow  for  Consequently,  fainter  stars  can  be  de- 
photography  can  be  used  to  record  faint  tected. 

images  against  the  background  of  the  Perhaps   the    most    important    appli- 

night   sky.   The   long- focus  instruments  cation  of  image  intensifiers  in  astronomy 

provide  greater  magnification,  and  the  will  be  in  obtaining  stellar  spectra.  These 

total  number  of  photons  from  1  square  devices,  when  perfected,  should  make  it 

minute  of  the  night  sky  are  spread  over  a  possible    to    obtain    with    telescopes    of 

larger  area.  When  exposures  are  made  to  moderate   aperture   spectra   that   previ- 

a  given  optical  density  of  the  night- sky  ously  have  been  difficult  to  obtain  with 

background  with  the  aid  of  an  image  even  the  largest  instruments.  Moreover, 

intensifier,    a   longer   time    is    therefore  image  tubes  are  efficient  in  the  infrared 

permitted,  there  are  more  photo  events  region  of  the  spectrum,  where  emulsions 

in  the  still  concentrated  stellar  image,  and  are  relatively  insensitive. 


TESTS  OF  SAMPLE  TUBES 

The  image  intensifiers  tested  were  cathode  is  not  fully  utilized  in  these  tubes, 
experimental  tubes  which  are  still  being  Nevertheless,  they  are  of  value  for 
improved.  Some  of  the  defects  in  these  photographing  infrared  fields  (provided 
first  samples  have  been  corrected  in  more  that  moderately  fast  F  ratios  are  avail- 
recent  tubes.  The  remaining  shortcomings  able),  for  high-resolution  spectroscopy 
can,  we  believe,  be  eliminated  with  where  bright  sources  are  involved,  and 
continued  effort.  finally    for    specialized    work    such    as 

Infrared  converter  tubes.  The  infrared  photographing  the  infrared  coronal  lines 

converter  has  an  SI  photocathode  with  of  the  sun. 

useful  sensitivity  from  0.7  to  1.4  microns.  An   infrared   converter,   FW132-1221, 

One  such  tube  is  the  FW132  image  con-  made  by  ITT  Laboratories,  was  tested  on 

verter   made   by    ITT    Laboratories    (a  the  Morgan  telescope  in  April  and  more 

picture  of  this  tube  appeared  in  last  year's  extensively  in  October  1961.  The  tube 

report).  This  compact  tube  is  character-  had  a  good  photocathode  with  a  sensi- 

ized  by  a  high-quality  SI  photosurface,  a  tivity  of  4.5  juA/lumen  to  2870°K  light 

series    of    accelerating    electrodes,    and,  with  a  no.  2540  filter.  The  photocathode 

finally,  a  phosphor  screen  capable  of  high  was  cooled  with  dry  ice  to  reduce  thermi- 

resolution.   Under   optimum   conditions,  onic  emission,  and,  at  15  kv,  20-minute 

50    line    pairs    per    millimeter    can    be  exposures    on    IIa-0    plates    showed   no 

resolved  with  these  tubes.  A  conventional  spurious  background.  The  relay  lens  was 

camera  with  a  fast  relay  lens  system  is  a  pair  of  //1.5  Zeiss  Bio  tars  mounted 

used    for    photographing    the    phosphor  front-to-front. 

screen.  These  tubes  provide  advantages  With  a   12-inch  camera  lens  on  the 

over  infrared  photographic  plates  because  DTM  spectrograph  (2-inch  collimator) , 

of  their  improved  resolution,  their  sim-  spectra  at  a  dispersion  of  45  angstroms/ 

plicity  in  operation  compared  with  hyper-  mm  at  1.0  micron  were  obtained  at  15  kv. 

sensitization    processes,    and    their    effi-  Typical  exposure  times  at  1  micron  on 

ciency     compared     with     the     infrared  IIa-0  plates  were  20  minutes  for  Jupiter 

photographic   sensitivity.    Because   con-  and  20  minutes  for  0  Andromedae  (2.4 

ventional  optics  are  used  for  photograph-  magv,  MO  III).  The  resolution  of  the 

ing  the  phosphor  screen,  and  since  there  system  was  better  than  40  line  pairs  per 

is  no  internal  multiplication  of  electrons,  millimeter, 

the   quantum   efficiency   of   the   photo-  The  gain  of  this  system  relative  to 


COMMITTEE   ON   IMAGE   TUBES   FOR   TELESCOPES 


299 


infrared  photographic  emulsions  is  diffi- 
cult to  judge,  owing  to  variations  in 
sensitivity  of  the  emulsions  with  hyper- 
sensitization.  We  estimate  that  roughly 
a  gain  of  30  in  exposure  time  is  achieved 
over  typical  I-Z  hypersensitized  emul- 
sions, accompanied  by  a  slight  improve- 
ment in  resolution  and  granularity. 

Slightly  greater  gains  could  be  achieved 
if  the  tube  could  be  operated  at  higher 
voltages.  ITT  Laboratories  has  now 
developed  an  improved  version  of  this 
tube,  known  as  the  FW167,  which  is 
twice  as  long  as  the  FW132.  This  tube 
has  not  yet  been  tested. 

Cascaded  image  tubes.  In  a  cascaded 
image  tube,  photoelectrons  from  the  first 
cathode  are  multiplied  by  a  phosphor- 
photocathode  sandwich;  the  "secondary" 
photoelectrons  thus  produced  are  imaged 
on  a  phosphor  screen.  The  screen  is  then 
photographed  with  a  relay  lens  system. 
The  multiplication  process  is  an  efficient 
one,  and  resolution  is  limited  principally 
by  the  granularity  of  the  phosphor. 

Several  cascaded  tubes  made  by  ITT 
Laboratories  have  been  tested  at  DTM, 
and  the  best  of  them,  FW152-37,  was  used 
with  the  spectrograph  on  the  Morgan 
telescope  in  October  1961  and  April  1962. 
The  resolution  of  this  tube  was  close  to 
25  line  pairs  per  millimeter.  The  cathode 
was  40  mm  in  diameter,  but  the  Zeiss 
Biotar  relay  system  was  effective  in 
imaging  only  a  25-mm  field.  The  tube 
was  focused  with  a  permanent- magnet 
system  and  gave  magnification  close  to 
unity. 

At  an  operating  voltage  of  18  kv, 
exposures  of  20  minutes  could  be  made 
without  the  background  due  to  spurious 
emission  becoming  excessive.  There  was 
no  artificial  cooling  in  these  tests.  The 
Sll  photocathode  had  a  sensitivity  of 
about  40  juA/lumen,  and  the  multipli- 
cation across  the  sandwich  was  approxi- 
mately 12.  Baked  IIa-0  plates  were  used 
to  photograph  the  phosphor  screen. 

Spectra  were  taken  with  a  12- inch 
camera  lens  which  gave  a  dispersion  of 
22   A/mm  at  the  photocathode.   These 


spectra  were  compared  with  direct  spectra 
taken  with  a  7-inch  lens  and  were  judged 
to  be  similar  in  quality.  The  speed  gain 
of  the  image  tube  system  in  this  compari- 
son ranged  from  3  to  5,  depending  on  the 
quality  of  the  seeing. 

Work  in  progress  at  ITT  Laboratories 
is  directed  now  toward  improving  the 
background  in  this  tube  type  to  permit 
operation  at  higher  voltages.  Improve- 
ments have  been  made  in  the  control  of 
phosphor  deposition,  and  more  recent 
tubes  have  somewhat  better  resolution. 
Additional  tubes  are  scheduled  for  early 
delivery. 

RCA  has  also  developed  for  us  a  two- 
stage  cascade  intensifier,  the  C70  056. 
This  tube  is  similar  in  size  to  the  ITT 
FW152  but  has  the  voltage  per  stage 
divided  between  a  series  of  four  accele- 
rating electrodes.  This  results  generally  in 
less  spurious  background.  The  cathode  is 
S20  (multialkali)  and  is  40  mm  in 
diameter.  The  tubes  we  have  received 
have  had  P20  screens. 

The  RCA  tube  we  have  tested  (C70 
056-1)  had  140  /x A/lumen  photosensitivity 
and  a  large  electron  gain  across  the 
sandwich.  It  was  operated  in  a  perma- 
nent-magnet system  at  an  overall  voltage 
of  16.5  kv.  At  this  voltage  the  background 
due  to  ion  scintillations  limited  exposures 
to  2  hours.  On  axis  visual  resolution  of 
the  system  in  these  tests  was  slightly 
better  than  20  line  pairs  per  millimeter. 

Test  exposures  were  made  with  this 
tube  with  the  12-inch  camera  on  the 
DTM  spectrograph.  The  speed  gain  over 
direct  spectra  obtained  with  the  7-inch 
lens  was  about  5.  The  actual  information 
gain  was  probably  somewhat  less,  owing 
to  the  limited  resolution  of  the  intensifier 
system.  Improved  phosphors  and  less 
background  due  to  ion  scintillations 
(which  will  permit  operation  at  higher 
voltages)  will  increase  considerably  the 
gain  that  can  be  obtained.  Delivery  of 
several  more  RCA  cascaded  tubes  of 
slightly  different  types  is  in  progress. 

Mica-window  tubes.  A  single-stage 
image  converter  with  a  phosphor  screen 


300  CARNEGIE     INSTITUTION      OF      WASHINGTON 

deposited  on  a  thin  end  window  provides  developed  in  D76  for  5  minutes  at  68°F. 

an  efficient  method  of  image  intensifica-  Test  exposures  were  made  at  10,  12J^, 

tion.  By  pressing  a  photographic  emulsion  143^,  and  17  kv.  At  the  higher  voltages, 

into    contact    with    the    window,    good  scintillations  were  visible  when  the  phos- 

definition  can  be  maintained,  and  most  of  phor  screen  was  viewed  with  a  micro- 

the    light    produced    in    the    screen    by  scope.  They  are  believed  to  be  due  to 

impinging  electrons  reaches  the  emulsion,  residual  gas  atoms  becoming  positively 

A  magnetically  focused  mica-window  ionized,  bombarding  the  photocathode, 
tube  (C70  026-6)  made  for  the  Committee  and  releasing  bunches  of  electrons.  The 
by  RCA,  Lancaster,  was  tested  by  Ford  scintillations  due  to  these  bunches  of 
and  Fredrick  with  the  DTM  spectrograph  electrons  could  be  recorded  at  14J^  kv  on 
on  the  Morgan  telescope.  This  tube  was  the  Plus-X  emulsion,  and  this  limited  the 
similar  in  size  to  the  ITT  FW1 17  pictured  duration  of  the  exposures.  It  was  deter- 
in  the  Annual  Report  of  the  Committee  mined  that  123/2  kv  approached  an 
on  Image  Tubes  in  Year  Book  60.  The  optimum  operating  voltage  from  the 
mica  window  was  40  mm  long  by  2  mm  point  of  view  of  gain  and  background.  At 
wide  and  approximately  8  microns  thick,  this  voltage  1-hour  exposures  could  be 
The  tube  had  an  S20  photocathode  (76  made.  The  magnetic  field  required  to 
AtA/lumen)  and  a  Pll  phosphor  screen,  focus  the  tubes  was  provided  by  an  array 

Preliminary  tests  had  previously  been  of    cylindrical    rod    magnets,    and    the 

made   with   ITT   magnetic   tubes   with  magnetic  field  was  varied  in  strength  by 

infrared    photocathodes.    Unfortunately  adding  or  removing  rods.  No  attempt  was 

the  preliminary  tubes  were  characterized  made  to  minimize  the  "S"  distortion  of 

by  excessive  field  emission,  optical  dis-  the  image,  and  the  magnification  of  the 

tortion,    and    slumping    photocathodes.  system  was  slightly  greater  than  unity. 

The  tube  used  in  these  RCA  tests  was  The  first  exposures  were  made  with  the 

the  first  one  fabricated  for  the  Committee  7-inch   spectrograph   camera,   and   were 

that  had  acceptably  low  background  and  compared  with  direct  plates  made  with 

a  reasonable  sensitivity.  Under  optimum  the  same  camera.  It  is  estimated  that  for 

conditions,  45  or  50  line  pairs  per  milli-  wavelengths  in  the  range  4000-5500  A  the 

meter  could  be  resolved  visually  on  the  gain  in  exposure  time  for  equal  densities 

phosphor  screen.  With  Plus-X  film,  25  over  103a-G  plates  was  20  to  25  times, 

line  pairs  could  be  resolved  in  contact  accompanied,   of   course,   by   a   loss   in 

exposure.  resolution.  This  gain  in  exposure  time  for 

Exposures  were  made  with  a  film  equal  density  is  estimated  for  similar  slit 
transport  mechanism  built  in  the  DTM  widths,  widening,  and  seeing  conditions, 
shop.  With  this  mechanism,  pieces  of  A  second  series  of  exposures  were  made 
35-mm  film,  approximately  35  mm  long,  with  a  12-inch  spectrograph  camera  lens, 
are  loaded  onto  a  cylindrical  mandrel,  and  giving  a  dispersion  of  22  A/mm  at  the 
the  mandrel  is  pressed  into  mechanical  photocathode.  These  exposures  were  corn- 
contact  with  the  thin  window.  With  the  pared  directly  with  direct  photographs  at 
apparatus  in  its  present  form,  six  man-  39  A/mm  made  with  the  7-inch  spectro- 
drels  are  available  for  preloading,  and  a  graph  camera.  It  was  found  that,  with 
single  exposure  is  made  on  a  piece  of  film,  the  mica-window  tube  and  the  Plus-X 
A  box  with  six  partitions  and  individual  film,  the  spectra  obtained  with  the  12- 
dark  slides  is  available  for  storing  the  inch  camera  were  comparable  in  quality 
mandrels  with  film.  Less  than  1  minute  to  those  obtained  directly  on  103a-G 
is  required  to  change  mandrels  between  plates  with  the  7-inch  lens, 
exposures.  In  the  tests  at  Flagstaff,  all  The  information  gain  in  this  system  is 
exposures   were   made   on   Plus-X    film  estimated  to  approach  5  at  123^  kv.  The 


COMMITTEE    ON   IMAGE   TUBES   FOR   TELESCOPES 


301 


tube  used  had  a  75-juA/lumen  photo- 
cathode,  which  is  somewhat  less  than 
average  for  the  S20  photosurface. 

The  Carnegie  Committee  has  two  more 
tubes  of  this  type  on  order  from  RCA. 
The  prospects  look  good  that  they  will  be 


useful  for  astronomical  observations.  In 
addition,  four  tubes  of  similar  type 
having  infrared  photocathodes  are  on 
order  at  ITT.  One  sample  tube  having  a 
very  good  cathode  has  been  received  and 
will  be  tested  soon. 


ACKNOWLEDGMENTS 


Many  of  the  tests  described  in  this  industrial  laboratories,  under  the  direc- 

report  were  conducted  by  Dr.  Ford  of  the  tion  of  the  Committee,  is  supported  by  a 

Department   of   Terrestrial   Magnetism,  generous  grant  from  the  National  Science 

The    development    of    special    tubes    in  Foundation. 


Department  of  Plant  Biology 


Stanford,  California 
C.  Stacy  French 
Director 


Contents 


Introduction 305 

Personnel 310 

Experimental  Taxonomy  Investigations ...      .      .      ...  311 

New  vistas  in  experimental  taxonomy .      .      .      .      ..  311 

Transplant  station  activities 312 

Physiology  of  climatic  races 313 

Growth  studies  in  controlled  environments 317 

A  control  system  for  carbon  dioxide  concentration  in  plant  growth  chambers     .  319 

Ecotypic  differences  in  response  to  light  intensity  in  Solidago  virgaurea    ....  320 

The  aseptic  culture  of  excised  tissues  of  Mimulus 323 

Studies  in  Poa  hybridization .  325 

The  North  American  field  pansy,  Viola  rafinesquii 333 

Biochemical  Investigations 334 

Factors  affecting  oxygen  evolution  from  Swiss  chard  chloroplasts 334 

A  Teflon-covered  electrode  assembly 343 

Relations  between  the  two  photochemical  reactions  of  photosynthesis     ....  345 

Enhancement  and  photostimulated  oxygen  consumption  in  Porphyridium     .      .      .  350 

Physical  separation  of  pigment  complexes  from  Euglena 352 

Electron  paramagnetic  resonance  studies  on  Chlamydomonas  reinhardi     ....  353 

An  electrically  isolated  transparent  liquid  pump 365 

Speeches        365 

Bibliography 366 


Carnegie  Institution  of  Washington  Year  Book  61,  1961-1962 


INTRODUCTION 

For  the  last  few  years  there  has  been  by  a  pigment  with  a  longer-wavelength 
increasing  interest  in  the  functional  absorption  band.  This  intermediate  reac- 
nature  of  the  two-pigment  system  now  tion,  which  may  turn  out  to  involve 
considered  to  be  of  basic  importance  in  plastoquinone,  or  perhaps  the  copper 
the  process  of  photosynthesis.  This  year  protein  plastocyanin,  appears  to  be 
the  subject  has  become  the  focal  point  of  driven  to  the  reduced  state  by  chloro- 
investigations  in  several  laboratories,  phyll  b  and  to  the  oxidized  state  by  the 
At  the  Department  of  Plant  Biology  a  far-red  pigment.  The  effectiveness  of 
particularly  striking  illustration  of  a  different  wavelengths  in  driving  the 
multipigment  system  was  studied  in  reaction  in  one  direction  or  the  other  is 
isolated  leaf  chloroplasts.  Dr.  David  C.  complicated,  therefore,  by  the  over- 
Fork  found  that  chlorophyll  b  was  more  lapping  of  the  absorption  bands  of  the 
effective  than  chlorophyll  a  for  the  evolu-  two  pigments  which  have  opposite  effects, 
tion  of  oxygen  by  isolated  spinach  Up  to  the  present  time  the  identifica- 
chloroplasts  suspended  in  a  medium  free  tion  of  a  specific  pigment  as  responsible 
of  added  oxidants.  This  endogenous  pro-  for  a  measurable  photochemical  event 
duction  of  traces  of  oxygen  by  illuminated  connected  with  the  photosynthesis  proc- 
chloroplasts  has  been  known  for  about  ess  has  depended  mainly  on  matching 
eighty  years  but  only  recently  was  dis-  the  wavelengths  of  peaks  in  the  action 
covered  to  be  due  more  to  the  action  of  spectrum  to  the  wavelengths  of  absorp- 
chlorophyll  b  than  to  that  of  chloro-  tion  maxima  of  known  pigments.  How- 
phyll  a.  ever,  the  action  spectrum  for  a  reaction 

In  1937  Hill  found  that  oxidants  like  being  driven  in  opposite  directions  by  two 
ferricyanide  added  to  chloroplast  sus-  pigments  may  be  so  complex  as  to  make 
pensions  would  tremendously  increase  the  identification  of  the  functional  pig- 
their  capacity  to  evolve  oxygen.  This  ments  through  their  characteristic  ab- 
process,  in  contrast  to  the  endogenous  sorption  spectra  impossible, 
reaction,  is  driven  by  chlorophyll  a  and  It  is  becoming  evident  that  the  diverse 
chlorophyll  b  about  equally.  Since  Hill's  photochemical  functions  of  different  pig- 
discovery  very  little  attention  has  been  ments  as  well  as  the  shapes  of  their 
given  to  the  slower  and  short-lasting  absorption  curves  must  be  better  under- 
endogenous  process.  stood  in  order  to  account  for  the  various 

When  chloroplasts  without  added  rea-  action  spectra  measured  for  processes 
gents  are  first  illuminated  the  rate  of  related  to  photosynthesis.  Advances  in 
oxygen  evolution  is  reasonably  high,  but  the  study  of  action  spectra  require 
it  drops  in  a  few  seconds  to  a  very  low  further  information  about  both  the  light- 
value.  Presumably  some  material  in  the  absorbing  and  the  photochemical  proper- 
chloroplast  structure  is  used  up  by  the  ties  of  the  various  functional  plant  pig- 
light  reaction.     The  ability  to  give  off  ments. 

oxygen  is  regenerated  by  storage  in  the  The  lack  of  sufficient  information  about 

dark  for  some  minutes.  Dr.  Fork  found  the  absorption  spectra  of  the  different 

furthermore  that  the  recovery  process  is  pigments  and  the  specific  reactions  driven 

greatly  hastened  by  far-red  light.     The  by  them  has  become  emphasized  through 

participation    of    two    pigments    in    the  the  attempt  to  find  a  satisfactory  explana- 

endogenous     evolution     of     oxygen     by  tion  for   two   additional   action  spectra 

chloroplasts  is  therefore  particularly  clear,  determined  by  Dr.  Fork  for  other  light 

Chlorophyll    b   drives    off   oxygen   from  effects    in    leaf    chloroplasts.    The    four 

some  intermediate  which  is  regenerated  effects  giving  very  different  action  spectra 

305 


306  CARNEGIE     INSTITUTION     OF      WASHINGTON 

with  the  following  characteristic  peaks  in  action  spectra  to  identify  the  fractional 
the  red  region  were:  the  evolution  of  absorption  of  each  pigment  at  various 
oxygen  (650  m/x,  680  m/x  shoulder) ;  the  wavelengths.  By  simple  absorption  spec- 
regeneration  of  the  ability  to  do  so  (730  troscopy,  and  even  with  derivative  spec- 
m/x);  the  uptake  of  oxygen,  apparent  troscopic  measurements,  it  has  not  been 
when  the  oxygen-evolving  system  is  possible  to  determine  the  shape  of  the 
poisoned  (690  m/x) ;  and  the  evolution  of  complete  absorption  curves  of  the  individ- 
oxygen  from  chloroplasts  with  added  ual  pigments.  Perhaps  in  the  future  more 
ferricyanide  (678  m/x,  650  mix  shoulder),  precise  measurements  and  deeper  under- 

Of  these  peaks  only  two  can  be  identi-  standing  of  action  spectra  for  the  various 
fied  with  a  reasonable  degree  of  certainty,  photochemical  effects  of  different  pig- 
The  650-mxx  action  peak  is  attributed  to  ments  may  be  used  to  extend  the  now 
chlorophyll  b,  and  that  at  678  mix  to  very  incomplete  knowledge  of  the  absorp- 
chlorophyll  a.  However,  it  is  not  yet  tion  spectra  of  the  functional  plant  pig- 
certain  whether  both  the  670-m/x  and  the  ments  in  their  natural  state. 
683-m/x  absorbing  forms  of  chlorophyll  a  The  need  for  more  specific  information 
contribute  to  the  678-m/x  action  maxi-  about  the  shapes  of  the  complete  absorp- 
mum  for  the  evolution  of  oxygen  when  tion  spectra  of  the  individual  pigments  is 
ferricyanide  is  added.  Furthermore,  this  illustrated  by  the  action  spectrum  for 
action  spectrum,  for  this  reaction,  which  oxygen  uptake  by  poisoned  chloroplasts. 
is  also  typical  of  green  plant  photo-  Is  the  690-m/x  peak  to  be  attributed  to  a 
synthesis,  shows  structure  characteristic  particularly  active  form  of  chlorophyll  a 
of  two  pigments.  The  maximum  is  at  678  with  its  maximum  at  that  wavelength? 
mix,  but  a  shoulder  at  650  mix  shows  that  Or  is  the  position  of  this  maximum  de- 
chlorophyll  b  as  well  as  chlorophyll  a  termined  by  the  wavelength  of  minimum 
is  involved.  Since  two  chlorophylls,  a  and  overlap  between  two  other  forms  of 
b,  are  believed  to  drive  separate  steps  of  chlorophyll  having  opposing  functions? 
the  overall  reaction,  the  relative  heights  Furthermore,  what  is  the  absorbing 
of  the  maxima  in  action  spectra  should  entity  giving  the  maximum  recovery  of 
be  capable  of  experimental  variation,  the  oxygen-evolving  capacity  of  chloro- 
This  could  in  principle  be  accomplished  plasts  at  730  m/x?  Does  this  wavelength 
by  varying  the  amounts  of  the  inter-  correspond  to  a  pigment  like  phyto- 
mediate  substances  linking  the  two  chrome,  or  could  730  m/x  perhaps  be  the 
photochemical  reactions  or  by  the  partial  wavelength  at  which  the  ratio  of  the 
inhibition  of  the  reaction  rates  of  certain  absorption  by  two  pigments  most  strongly 
nonphotochemical  steps  in  the  process,  favors  one  of  them?  Such  questions  are 

A  further  hope  for  the  future  is  that  it  not  easy  to  answer  with  certainty,  yet 
may  be  possible  to  add  certain  inter-  they  are  of  the  greatest  significance  in 
mediates  to  disintegrated  chloroplasts  in  understanding  the  functional  relations  of 
such  a  way  that  the  action  of  the  separate  the  pigments  responsible  for  photo- 
photochemical    steps    may    be    clearly  synthesis. 

distinguished.  So  far  the  attempts  to  do  To  study  further  the  relation  between 

this  in  other  laboratories  have  resulted  in  the  two  light  reactions  in  the  red  alga 

rather  complex  action  spectra  indicating  Porphyridium,   which   was   reported    on 

that  the  separation  of  the  effects  of  the  last  year,  measurements  were  made  with 

individual  pigments  has  not  been  com-  two    light    beams    of    different    colors 

plete.  separated  in  time.  Red  light  absorbed  by 

The  importance  of  the  complete  separa-  chlorophyll  a  and  green  light  absorbed  by 

tion  of  one  effect  from  another  lies  not  phycoerythrin  were  given  in  flashes  of  a 

only  in  elucidating  the  chemical  nature  of  few  seconds'  duration.  A  green  flash  was 

the  individual  steps  but  also  in  using  the  found  more  effective  in  oxygen  produc- 


DEPARTMENT    OF    PLANT   BIOLOGY  307 

tion  if  it  was  immediately  preceded  by  a  of  finding  a  way  around  this  difficulty, 

red  flash.  The  material  produced  by  red  Dr.    J.    S.    Brown    has    studied    how 

light  which  enhances  the  succeeding  green  greatly    the    conditions    of   growth    and 

flash  had  a  half -life  of  about  18  seconds,  measurement  influence  enhancement  and 

Within  the  time  intervals  studied,  how-  photostimulated   oxygen  uptake  in  the 

ever,  the  yield  of  oxygen  from  a  red  flash  red  alga  Porphyridium.  The  initial  values 

was  slightly  decreased  rather  than  en-  in  a  series  of  measurements  are  strongly 

hanced  by  a  previous  exposure  to  green  affected  by  the  light  used  for  culturing 

light.  Such  studies  can  certainly  be  used  the    algae.     During    the    first    5    hours 

to  evaluate  the  lifetimes  of  intermediate  photosynthesis    and    enhancement    may 

products  disappearing  in  side  reactions  vary  inversely;  the  photostimulation  of 

and    may   give  information  on  relative  oxygen  uptake  declines.  After  5  hours  all 

quantities  of  the  several  intermediates,  three  processes  reach  constant  values. 

Flash  experiments  may  have  only  a  Dr.  Brown  also  found  that  the  710-m/x 

limited   value   in   telling   which    photo-  absorbing    compound    of   aged    Euglena 

chemical  reaction  comes  before  another,  cultures  leaks  out  of  the  chloroplasts  into 

A  cyclic  process  has  no  "first  step"  from  the    cytoplasm.  The    685-m^t   absorbing 

the    point    of   view    of    the    permanent  form  of  chlorophyll,  however,  remains  in 

machinery.  The  concept  of  a  first  step  is  the   chloroplasts.   The   670-m/z   form   of 

reasonable    with     respect    to    material  chlorophyll   a  accompanies  the   710-m/z 

flowing  through  the  system,  such  as  a  pigment.  In  collaboration  with  Dr.  J.  H. 

labeled  compound.  For  the  cycling  part  of  C.  Smith  she  found  the  710-nux  pigment 

the  system,  however,  the  idea  of  a  first  to  be  pheophorbide  a  or  a  closely  related 

step   has   significance   only   in   deciding  compound.    It   arises   from   the   695-m/u 

which  intermediate  piles  up  in  a  ready  form  of  chlorophyll  a. 

state  during  a  dark  period.  An  investigation  of  the  production  of 

Last  year  we  used  a  particular  model  free  electrons  induced  by  light  in  the 

scheme  to  compute  hypothetical  time-  green    alga    Chlamydomonas    started    in 

course  curves  for  the  relative  rates  of  Zurich  by  Drs.  Harry  and  Ellen  Weaver 

oxygen  exchange  of  illuminated  photo-  was    continued    at   the    Department   in 

synthetic    systems.    This    scheme    gave  collaboration    with    Varian    Associates, 

reasonably  realistic  results.   The  model  The  measurements  were  made   by  the 

differs  from  others,  currently  popular,  in  electron  paramagnetic  resonance  method, 

predicting   that    the    time-course    curve  Exposure  to  light  produced  a  broad  slowly 

for  gas  exchange  immediately  following  a  disappearing  signal  and  an  intense  narrow 

light   exposure   should   have   a   second-  rapidly  disappearing  signal.  The  action 

order  shape  which  should  vary  with  the  spectrum  for  the  narrow  signal  approxi- 

color  of  the  light  used,  whereas  the  other  mated  the  absorption  spectrum  of  chloro- 

models  predict  a  first-order  decay  curve  phyll  a.  These  observations  agree  with 

whose  shape  should  be  independent  of  the  recent  work  in  other  laboratories.  Earlier 

light  color.  results  on  action  spectra  had  been  dis- 

The  difficulty  in  settling  this  question  torted   by  using  preparations  with  too 

experimentally  lies  in  the  fact  that  con-  great  absorption. 

comitant  with  oxygen  evolution  there  is  For  the  narrow  rapidly  decaying  signal 

also  the  production  of  variable  amounts  the  g  value  was  2.0025,  which  is  close  to 

of   easily   oxidizable   material.    The   in-  that  for  a  pure  conduction  electron.  The 

creased  rate  of  oxygen  uptake  by  this  half-width  was  8.3  gauss.  For  the  slowly 

material  after  a  light  period  complicates  decaying  signal  the  g  value  was  2.0046 

the  study  of  the  shape  of  the  decay  curve  and  the  half -width  was  20  gauss.  This 

for  the  lingering  oxygen  evolution.  Some  signal    appears    to    be    due    to    plasto- 

experiments  are  in  progress  in  the  hope  quinone. 


308 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


The  Experimental  Taxonomy  group  is 
attaining  increased  experimental  pre- 
cision in  different  aspects  of  the  study  of 
plant  relationships.  Several  races  and 
species  of  Mimulus  are  being  used  for 
comparative  growth  studies  at  the  altitu- 
dinal  field  stations,  for  cytogenetic  analy- 
sis, for  controlled  growth  chamber  ex- 
periments, and  for  quantitative  com- 
parisons of  photosynthetic  rates  under 
varied  conditions.  All  these  studies  lead 
to  an  improved  understanding  of  evolu- 
tionary mechanisms  and  of  natural 
selection  in  different  environments. 

One  of  the  most  striking  observations 
in  the  transplant  work  is  the  enhanced 
vigor  in  first-generation  hybrids  between 
climatic  races  of  the  same  species.  The 
vigor  shown  by  such  a  hybrid  depends  on 
the  environment  in  which  it  is  grown 
and  on  the  relation  of  that  environment 
to  the  native  climates  of  its  parents.  The 
approximate  degree  of  genetic  difference 
needed  between  two  climatic  races  to 
yield  first-generation  progeny  capable  of 
surviving  over  a  wider  range  of  climates 
than  either  parent  is  now  being  sys- 
tematically studied  in  Mimulus.  This 
basic  question  has  not  been  adequately 
examined  experimentally  before.  Recipro- 
cal crosses  have  been  made  between  each 
pair  of  eight  key  races  native  to  eight 
different  altitudes.  The  hybrids  and 
parental  races  are  being  established  for 
observation  this  year  as  cloned  trans- 
plants at  the  three  altitudinal  transplant 
stations. 

Milner  and  Hiesey  have  completed  an 
examination  of  the  photosynthetic  re- 
sponse of  six  races  of  Mimulus  cardinalis 
originating  from  different  climates.  The 
measurements  covered  a  temperature 
range  from  0°C  to  the  high  temperature 
at  which  apparent  photosynthesis  ceases, 
about  50°C. 

The  light  intensity  required  to  saturate 
photosynthesis  varies  from  3700  to  5900 
footcandles  for  the  different  races  at  40°C, 
and  from  300  to  600  footcandles  at  0°C. 

All  six  races  showed  their  maximum 
photosynthetic  rate  at  30°,  and  differences 


between  races  were  small  from  20°  to  40°. 
Racial  differences  became  apparent  at  the 
low  and  high  ends  of  the  temperature 
range.  At  0°  the  photosynthetic  rates  of 
the  six  races  vary  from  11  to  20  per  cent 
of  the  maximum  rates.  At  high  tempera- 
ture, apparent  photosynthesis  becomes 
zero  at  46°  for  one  race,  47°  for  three 
races,  and  49°  for  the  other  two. 

The  races  also  differ  in  their  ability  to 
maintain  a  high  rate  of  photosynthesis 
for  a  long  time.  Measurements  of  con- 
tinuous photosynthesis  for  12  hours 
under  optimum  light  intensity  and  tem- 
perature were  made.  At  the  end  of  this 
time  the  rates  for  the  different  races  were 
from  92.5  to  74  per  cent  of  the  maximum 
rates. 

Much  greater  variability  was  found  for 
the  rates  of  carbon  dioxide  production 
in  the  dark  than  for  rates  of  its  photo- 
synthetic uptake  in  the  light.  All  the 
photosynthetic  rates  referred  to  above 
are  rates  of  apparent  photosynthesis,  un- 
corrected for  the  rate  of  carbon  dioxide 
evolution  in  the  dark.  Because  of  the 
intrinsic  interest  in  the  latter  value, 
however,  it  was  measured  many  times 
during  the  course  of  the  work. 

A  single  measurement  of  the  rate  of 
dark  carbon  dioxide  evolution  is  found  to 
be  of  doubtful  value,  and  making  enough 
measurements  to  provide  a  trustworthy 
mean  is  very  time  consuming.  For  these 
reasons,  and  also  because  the  rate  of 
dark  carbon  dioxide  evolution  reflects 
the  operation  of  other  processes  besides 
respiration,  we  question  the  validity  of 
'  'correcting"  photosynthetic  rates  of  Mim- 
ulus leaves  for  it. 

Studies  on  the  growth  and  develop- 
ment of  Mimulus  in  controlled  growth 
chambers  by  Hiesey  and  Milner  are  being 
coordinated  with  the  quantitative 
measurements  of  their  photosynthetic 
rates.  At  present  the  first  objective — de- 
veloping an  effective  working  laboratory 
for  this  type  of  comparison — has  been 
achieved.  The  marked  effects  of  the  inter- 
action of  temperature,  light  intensity, 
and  carbon  dioxide  concentration  on  the 


DEPARTMENT    OF   PLANT   BIOLOGY  309 

growth  and  development  of  Mimulus  suited  to  a  wide  variety  of  parallel  ex- 
clones  are  being  studied.  Clones  of  dis-  perimental  studies  and  are  of  particular 
tinct  climatic  races  show  differential  value  for  cross  comparison  in  two  dis- 
patterns  of  response  that  undoubtedly  tinct  and  complementary  climatic  regions 
are  linked  with  differences  in  internal  of  the  world, 
physiology.  Findings  of  considerable  interest  have 

From  the  light-saturation  and  tempera-  been  made  by  Drs.  Bjorkman,  Holmgren, 

ture    curves    derived    from    short-term  and    Nobs    on    two    races    of    Solidago 

quantitative   measurements,    predictions  virgaurea,  one  a  shade  race  from  Sweden 

can  be  made  about  expectations  from  at  56°N  and  the  other  an  alpine  race 

growth  cabinet  experiments  lasting  several  from  Norway  at  69°N.  The  two  races 

weeks  to  months.  Some  such  predictions  have      very     different     photosynthetic 

have   been   supported   by   experimental  responses   to   different   light   intensities, 

results,  but  others  have  been  reversed,  The  northern  alpine  race  has  a  much 

indicating  the  influence  of  other  factors  higher  requirement  for  light  saturation 

that  remain  to  be  examined.  Enhance-  than  the  southern,  and  its  chloroplasts 

ment  of  growth  has  been  observed  by  remain  normal  when  exposed  to  a  light 

enriching  the  carbon  dioxide  in  the  atmos-  intensity   of   approximately   3500   foot- 

phere,  the  degree  of  enhancement  differ-  candles  in  contrast  to  the  southern  race 

ing  with  races.  whose   chloroplasts  disintegrate   at   this 

An  important  step  was  made  this  year  moderately    high    light    intensity.    The 

by    Mrs.    Ruth    Elliott    in    establishing  leaves  of  the  southern  shade  race  become 

tissue  cultures  from  Mimulus  clones  that  twice  as  large  when  grown  in  light  of  700 

are  also  being  used  for  related  studies,  foot  candles  as  when  grown  in  light  of 

The  significance  of  the  work  with  cul-  3500  foot  candles.   The  direction  of  the 

tures  of  parts  of  plants  is  that  quantita-  corresponding  leaf  modifications  in  the 

tive  growth  and  photosynthesis  measure-  northern  race  is  just  reversed.  The  rates 

ments  can  be  made  of  tissue  from  races  of  photosynthesis  measured  under  con- 

with  contrasting  environmental  require-  trolled  conditions  on  the  modified  leaves 

ments.    It   should   thus   be   possible   to  of  both  races  are  also  changed,  but  the 

examine  the  physiological  requirements  significant  racial  differences  remain, 
of  tissue  from  the  various  organs  of  plants         Dr.  Clausen  has  been  reviewing  some 

as  well  as  for  the  entire  plant  to  localize  of  the  earlier  data  on  Poa,  completing 

the  site  of  the  physiological  differences  records  on  the  fertilities  and  performance 

that  determine  survival  or  extinction  of  of  interspecific  hybrids.  The  older  data 

the  races.  are   being   supplemented   by  some   new 

We  are  collaborating  with   Dr.   Axel  observations  from  current  field  plantings 

Nygren,    Dr.    Olle   Bjorkman,   and   Dr.  made  at  Stanford.  Some  of  the  tentative 

Paul  Holmgren  of  the  Institute  of  Plant  conclusions    about    the    fertilities    and 

Systematics   and   Genetics  at   Uppsala,  characteristics     of     hybrid     progenies, 

with  whom  Dr.  Nobs  has  been  working  arrived  at  during  earlier  years  on  the 

during  most  of  the  year.  The  Swedish  basis  of  less  complete  information,  are 

group  works  mostly  on  races  and  species  being  reexamined.  The  unpredictable  out- 

of  Solidago,  a  goldenrod  genus  common  come  of  interspecific  crossings  between 

to     North     America     and     to     Europe,  species  and  races  of  the  highly  polyploid 

Although  Mimulus  and  Solidago  belong  and  predominantly  apomictic  species  of 

to  different  plant  families,  both  are  well  Poa  becomes  increasingly  evident. 


310 


CARNEGIE     INSTITUTION      OF      WASHINGTON 

PERSONNEL 


Biochemical  Investigations 

Staff:  C.  Stacy  French,  Director,  Jeanette  S. 

Brown,  David  C.  Fork,  James  H.  C.  Smith, 

Emeritus 
Visiting  Investigators:  Paul  Latimer,  Ellen  C. 

Weaver 
Technical  Assistants:  Robert  A.  Clair,  Harriet 

M.  Fulk 

Experimental  Taxonomy 

Staff:  Jens  C.  Clausen,  Emeritus,  William  M. 

Hiesey,    Harold  W.   Milner,   Malcolm  A. 

Nobs 
Visiting  Investigators:  Thomas  R.  Pray,  Henry 

J.  Thompson 
Summer    Research     Assistants:     Steven     N. 

Gilborn,  Andrew  N.  Lenz 
Technical  Assistants:  Ruth  F.  Elliott,  Frank 

Nicholson 
Clerical  Assistant:  Marylee  H.  Eldredge 
Gardeners:    Joseph    S.    Chang,    Emmett    R. 

Clagg,  Wesley  B.  Justice 

Department  Secretary 
Wilbur  A.  Pestell 

A  dministrative  A  ssistant 
Wiley  Knight,  Jr. 

Mechanic 
Richard  W.  Hart 


Custodian 


Jan  Kowalik 


Wilbur  A.  Pestell  retired  on  June  30, 
1962,  after  42  years  with  the  Institution. 
During  this  period  he  was  in  the  Division 
of  Publications  in  Washington;  secretary 
in  the  Desert  Laboratory  at  Tucson, 
Arizona,  and  at  the  Coastal  Laboratory 
at  Carmel,  California;  and  since  1929 
secretary  of  the  Department  of  Plant 
Biology  at  Stanford,  California. 

Dr.  Jens  C.  Clausen  was  made  a 
Knight  of  the  Order  of  Dannebrog  by 
the  King  of  Denmark  in  October  1961.  In 


July- August  1961  Dr.  Clausen  presented 
a  series  of  five  lectures  on  evolution  at  the 
Summer  Institute  for  College  Teachers 
of  Botany,  Washington  State  University, 
Pullman,  Washington. 

Drs.  James  H.  C.  Smith  and  C.  S. 
French  spent  the  summer  of  1961  visiting 
European  laboratories  of  plant  physiology 
and  biochemistry  that  are  concerned  with 
the  study  of  photosynthesis,  and  attended 
international  congresses  on  the  subjects 
of  biochemistry,  biophysics,  and  plant 
physiology. 

Dr.  David  C.  Fork  spent  the  month  of 
March  in  the  Photosynthesis  Laboratory 
of  Professor  A.  Moyse  at  Gif-sur-Yvette 
near  Paris.  There  he  collaborated  with 
Mr.  Y.  de  Kouchkovsky,  who  had  also 
been  working  on  the  endogenous  evolu- 
tion of  oxygen  from  isolated  chloroplasts. 
Since  then  Dr.  Fork  has  been  at  the  De- 
partment of  Physical  Chemistry  of  the 
Philips  University  in  Marburg  to  work 
with  Dr.  H.  T.  Witt.  He  has  been 
applying  Dr.  Witt's  methods  for  measur- 
ing the  changes  in  the  absorption  spectra 
of  cellular  components,  induced  by  light, 
to  algae  and  chloroplasts,  whose  time 
course  of  oxygen  evolution  he  has  pre- 
viously analyzed.  This  work  is  expected 
to  clarify  further  the  relations  between 
the  pigment  systems  and  several  of  the 
intermediate  compounds  involved  in  the 
process  of  oxygen  evolution. 

Dr.  Malcolm  A.  Nobs  worked  at  the 
Institute  of  Plant  Systematics  and  Genet- 
ics at  the  Royal  Agricultural  College  of 
Sweden  from  October  1961  to  June  1962 
in  collaboration  with  Dr.  Axel  Nygren, 
Dr.  Olle  E.  Bjorkman,  and  Dr.  K.  Paul 
Holmgren.  The  program  of  our  Ex- 
perimental Taxonomy  group  on  the 
comparative  physiology  of  climatic  races 
and  that  of  the  Uppsala  group  are  closely 
related.  Dr.  Nobs'  time  in  Sweden  was 
devoted  to  comparative  studies  of  physio- 
logical characteristics  of  latitudinal 
species  and  races  of  Solidago,  native  both 
to  North  America  and  to  Europe.  The 


DEPARTMENT    OF   PLANT   BIOLOGY  311 

newly  constructed  phytotron  at  Uppsala  altitudinal  and  latitudinal  races  of  Mimu- 

is  being  used  for  this  purpose.  The  find-  lus  are  of  much  mutual  interest  because 

ings  of  the  Uppsala  group  in  Solidago  the  two  cover  distinct  but  complemen- 

and  those  of  our  group  in  California  on  tary  regions  of  the  world. 


EXPERIMENTAL  TAXONOMY  INVESTIGATIONS 

,y       v  of  genetics,  developmental  morphology, 

^  ^  physiology,  and  biochemistry.  The  mech- 

EXPERIMENTAL   TAXONOMY  .  x    j-    j    •  •  x 

anisms  are  studied  in  successive  steps, 

William  MHiesey  Harold  W.  Milner,  starting   with    populations    of   plants    in 

and  Malcolm  A.  Nobs  ,,     .  ,.        ,     ,  .,    ,  j  ,.         , 

their  native  habitats  and  proceeding  to 

Current  developments  in  techniques,  more  sharply  focused  stages  down  to  the 

and    new    information    from    long-term  cellular  level. 

experiments  begun  years  ago,  extend  the  The  experimental  steps.  The  importance 
horizon  of  experimental  taxonomy  so  of  studies  on  the  growth  and  development 
markedly  that  a  reevaluation  of  what  responses  of  plants  from  diverse  natural 
this  field  now  encompasses  appears  to  be  habitats  when  grown  at  the  altitudinal 
timely.  The  basic  objectives  formulated  field  stations  is  well  established  from 
during  earlier  years  remain  the  same;  but  earlier  publications,  and  such  studies  may 
the  leverage  afforded  by  more  precise  be  considered  to  be  the  first  step.  The 
means  of  investigation  opens  exciting  necessity  for  cytological  and  genetic  in- 
fresh  avenues  for  the  experimental  study  vestigations  has  likewise  been  demon- 
of  plant  relationships.  strated     through     numerous     examples; 

The  entire  spectrum  of  expression  of  they  are  the  second  essential  step.  The 
higher  plant  species,  ranging  from  the  inclusion  of  experiments  in  controlled 
multitudinous  aspects  they  present  to  the  environments  and  of  quantitative  physio- 
observer  in  the  wild  to  the  detailed  logical  measurements,  as  outlined  in 
analysis  of  the  functioning  of  their  many  Year  Book  58,  pages  344-346,  constitutes 
component  parts  down  to  the  cellular  a  third  vital  step.  The  fourth  step — 
level,  is  now  within  reach  of  systematic  study  of  detached  tissues  of  higher  plants 
experimental  study.  In  prospect  is  a  grown  under  aseptic  conditions — has 
truly  integrated  plant  science  whereby  been  started  during  the  current  year, 
contributions  from  the  various  specialized  All  four  steps  are  directly  linked 
fields,  including  taxonomy,  ecology,  cytol-  through  a  common  basic  unit  of  study, 
ogy,  genetics,  physiology,  developmen-  the  cloned  individual,  which  assures 
tal  morphology,  and  biochemistry,  can  be  genetic  identity.  Clones  from  the  vir- 
incorporated  in  a  panoramic  view  of  tually  unlimited  diversity  of  climatic 
plant  relationships  and  evolution  not  races  of  the  same  or  of  different  species 
only  scientifically  satisfying  but  also  that  occur  in  the  wild,  or  of  controlled 
aesthetically  inspiring.  hybrid  combinations  between  them,  are 

A  restatement  of  objectives.  The  enlarged  the  experimental  materials  required  for 

perspectives  in  experimental  taxonomy  supplying    the    information    needed    to 

call  for  a  revised  statement  of  its  objec-  piece  together  the  overall  picture, 
tives.  The  idealized  goal  toward  which  Methods  and   materials.    Each   of   the 

we  are  working  is  a  fully  integrated  under-  experimental  steps  needs  to  be  examined 

standing  of  the  chain  of  mechanisms  that  in  considerable  detail  to  bring  to  light 

underlie  the  end  products  of  plant  evolu-  the    innumerable    bits    of    information 

tion  in  terms  ranging  from  those  used  in  essential    for    assembling    the    compre- 

purely  descriptive  classification  to  those  hensive  picture  of  relationships  and  evolu- 


312 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


tionary  mechanisms  that  we  seek.  Data 
need  to  be  reexamined  from  many  points 
of  view  before  their  place  and  significance 
in  this  picture  can  be  evaluated.  The 
importance  of  selecting  materials  fulfilling 
the  exacting  requirements  for  such  studies 
has  been  emphasized  in  previous  year 
books  (Year  Books  IS,  pp.  103-104;  53 
157-158),  together  with  reasons  why  our 
current  work  is  centered  principally  on 
latitudinal  races  of  the  Mimulus  cardi- 
nalis-lewisii  complex. 

The  possibilities  of  utilizing  other 
plant  groups  are  also  being  kept  in  mind. 
Especially  promising  from  the  point  of 
view  of  comparing  the  physiological 
characteristics  of  ecological  races  from 
contrasting  latitudes  are  forms  of  the 
goldenrod,  especially  the  Solidago  multi- 
radiata-virgaurea  complex  of  North 
America  and  Europe.  Dr.  Nobs  has  spent 
the  greater  part  of  the  current  year 
working  on  this  group  of  plants  at  the 
Institute  of  Plant  Systematics  and  Genet- 
ics at  Uppsala,  Sweden,  where,  in  co- 
operation with  Dr.  Axel  Nygren,  Dr. 
Olle  E.  Bjorkman,  and  Dr.  K.  Paul 
Holmgren,  he  has  been  studying  rates  of 
photosynthesis  of  races  of  Solidago  from 
Scandinavia  from  56°  to  69°N  latitude, 
and  from  California  at  38°  to  39°N. 

Transplant  Station  Activities 

Malcolm   A.  Nobs,  Jens  Clausen,   William  M. 
Hiesey,  and  Frank  Nicholson 

The  systematic  testing  of  various  com- 
binations of  Fi  hybrids  between  different 
altitudinal  and  latitudinal  races  of  the 
Mimulus  cardinalis-lewisii  complex  men- 
tioned in  last  year's  report  ( Year  Book  60, 
pp.  381-382)  has  been  carried  to  the 
garden  planting  stage  this  year.  Cloned 
individuals  both  of  parental  and  of  hybrid 
plants  have  been  established  at  Stanford, 
Mather,  and  Timberline.  In  most  com- 
binations, reciprocal  Fi  hybrids  are  in- 
cluded in  the  tests  as  well  as  progeny 
resulting  from  selling  the  parents. 

The  preliminary  evidence  indicates 
that    Fi    hybrids    derived  from  crosses 


between  the  same  race  of  M .  lewisii  and 
different  races  of  M .  cardinalis  from  low 
and  middle  altitudes  have  markedly 
different  capacities  to  survive  at  the  three 
altitudinal  stations.  From  the  new  plant- 
ings it  should  be  possible  to  establish 
approximately  the  degree  of  genetic 
differentiation  needed  in  parental  forms 
to  confer  a  given  amount  of  tolerance 
to  progeny  for  survival  in  contrasting 
climates. 

Cloned  propagules  of  the  same  parental 
and  hybrid  plants  used  in  these  altitudinal 
tests  have  been  sent  to  Dr.  Robert  K. 
Vickery,  Department  of  Botany,  Uni- 
versity of  Utah,  Salt  Lake  City.  He  and 
his  associates  are  studying  the  flower 
pigments  by  paper  chromatography  in 
an  effort  to  determine  the  mode  of  in- 
heritance of  the  various  chemical  con- 
stituents that  govern  flower  color  varia- 
tions in  this  group  of  plants.  Flower 
color,  as  described  in  Year  Book  57, 
pages  270-271,  is  one  of  the  morphologi- 
cal characters  that  appear  to  be  linked 
with  capacity  for  survival  at  the  trans- 
plant stations  in  segregating  F2  progeny 
resulting  from  crosses  between  coastal  M. 
cardinalis  and  alpine  M.  lewisii. 

Seeds  and  living  plants  of  a  race  of 
Solidago  multiradiata  from  Umiat,  Alaska, 
were  supplied  to  us  by  Dr.  John  Koranda 
of  the  Alaska  Agricultural  Experiment 
Station  at  Palmer,  Alaska.  This  race  will 
be  studied  as  transplants  at  our  altitu- 
dinal stations.  Plants  of  California  forms 
of  Solidago  have,  in  turn,  been  sent  to 
Dr.  Koranda  for  study  in  Alaska. 
Information  about  the  responses  of 
latitudinal  ecotypes  transplanted  be- 
tween California  and  Alaska  is  almost 
wholly  lacking.  These  exploratory  plant- 
ings may  serve  to  chart  the  way  for  more 
comprehensive  efforts  later. 

Some  new  plantings  of  parental  and 
hybrid  derivatives  of  Poa  have  been 
made  at  Stanford  to  clarify  questions 
about  the  relative  fertilities  and  degrees 
of  apomixis  in  interspecific  combinations 
made  in  earlier  years.  At  the  altitudinal 
stations    some    observations    are    being 


DEPARTMENT    OF    PLANT   BIOLOGY 


313 


continued  on  cloned  transplants  of  key 
parental  and  apomictic  hybrid  strains 
established  as  early  as  1946.  The  new 
data  supplement  the  now  extensive  in- 
formation obtained  from  the  Poa  studies 
in  previous  years. 

During  the  latter  part  of  the  summer  of 
1961  Dr.  Clausen  carried  on  observational 
vegetational  studies  in  the  Harvey 
Monroe  Hall  Natural  Area.  They  were 
aimed  primarily  at  determining  vegeta- 
tional patterns  associated  with  differences 
in  microclimates,  soils,  and  terrain  in 
this  typically  high  Sierran  region.  Floral 
components  both  of  circumboreal  and  of 
more  southern  origin  meet  in  this  area. 
Observational  studies  of  this  kind  help  to 
orient  experimental  work  not  only  with 
alpine  but  also  with  lowland  plants. 

Visitors  at  the  transplant  stations 
during  the  current  year  included  Dr.  John 
Koranda  of  the  Alaska  Agricultural 
Experiment  Station  at  Palmer  and  staff 
members  and  graduate  students  from 
the  University  of  California  at  Davis 
headed  by  Dr.  Charles  M.  Rick.  A  class  of 
graduate  and  undergraduate  students  in 
botany  from  Stanford  under  the  guidance 
of  Dr.  John  Thomas  and  Mrs.  Roxana  S. 
Ferris  of  the  Department  of  Systematic 
Biology  conducted  field  work  at  Mather 
during  the  spring  of  1962.  Dr.  Alexander 
Sokoloff  of  the  Department  of  Genetics 
of  the  University  of  California,  Berkeley, 
collected  Drosophila  material  at  Mather 
for  Dr.  Th.  Dobzhansky,  who  is  con- 
tinuing his  studies  with  species  and 
races  from  along  the  Sierran  Transect  at 
the  Rockefeller  Foundation. 

Physiology  of  Climatic  Races 

Harold  W .  Milner,  William  M.  Hiesey, 
and  Malcolm  A.  Nobs 

Different  climatic  races  of  the  Mimulus 
cardinalis-lewisii  complex  show  different 
responses  in  their  rates  of  photosynthesis 
to  changes  in  light  intensity,  temperature, 
and  duration  of  continuous  photosyn- 
thesis. Following  the  lines  described  in 
earlier  year  books,  we  have  now  com- 


pleted the  examination  of  six  races  of  M. 
cardinalis  over  a  temperature  range  of 
0°  to  50°.  The  results  are  being  prepared 
for  publication  and  are  summarized  here. 

Light  intensity.  The  photosynthetic 
rate  of  each  plant  was  determined  at  high 
light  intensity  and  then  at  a  series  of 
intensity  levels  each  80  per  cent  of  the 
preceding  intensity.  The  true  light 
saturation  level  lies  between  the  last 
measured  intensity  at  which  the  maxi- 
mum photosynthetic  rate  was  main- 
tained and  the  next  lower  intensity  where 
a  decrease  in  rate  was  found.  This  pro- 
cedure was  followed  at  each  5-degree 
temperature  interval  from  0°  to  40°. 
At  each  temperature  the  light  intensity 
levels  between  which  saturation  occurs 
were  plotted.  A  smooth  curve  was  then 
drawn  to  fall  within  the  appropriate 
light  levels  at  each  temperature.  Above 
40°  the  photosynthetic  rate  is  not  stable 
over  the  length  of  time  required  to  make 
satisfactory  light  saturation  measure- 
ments. 

Light  saturation  curves  were  deter- 
mined for  three  clones  each  of  the  Los 
Trancos,  San  Antonio  Peak,  Jackson- 
ville, Priest's  Grade,  and  Yosemite  races 
and  for  five  clones  of  the  Baja  California 
race.  Within  each  race  the  clones  showed 
a  very  similar  response  of  light  saturation 
to     temperature. 

Mean  values  for  the  clones  of  each 
race  are  shown  in  figure  1  to  represent  the 
light  saturation  requirement  of  the  races. 
The  Los  Trancos  and  Baja  California 
races  have  the  lowest  light  saturation, 
from  500  footcandles  at  0°  to  3700  foot- 
candles  at  40°.  At  these  temperatures  the 
range  in  footcandles  for  the  other  races  is : 
San  Antonio  Peak,  600  to  4300;  Jackson- 
ville, 400  to  4800;  Priest's  Grade,  300  to 
5700;  and  Yosemite,  the  highest,  600  to 
5900.  The  saturating  light  intensity  at  0° 
relative  to  that  at  40°  ranges  from  5  per 
cent  for  the  Priest's  Grade  to  14  per  cent 
for  the  San  Antonio  Peak  race. 

The  effect  of  high  and  low  light  inten- 
sity at  different  temperatures  over  a 
period  of  weeks  was  tested  on  plants  in 


314 


CARNEGIE     INSTITUTION      OF     WASHINGTON 


6000 


Temperature  ,°C 

Fig.  1.  Saturating  light  intensity  versus 
temperature.  The  curves  are  identified  by  the 
initial  letters  of  the  race  names:  Yosemite, 
Priest's  Grade,  Jacksonville,  San  Antonio  Peak, 
Baja  California,  and  Los  Trancos.  Reading  down 
at  0°  the  curves  are  Y,  S,  B  and  L  (equal),  J, 
and  P. 


controlled  growth  cabinets.  The  results 
with  clones  of  the  Los  Trancos  and 
Priest's  Grade  races  are  described  in  the 
following  section  of  this  report. 

Temperature.  The  rate  of  light  saturated 
photosynthesis  was  measured  at  5- degree 
intervals  from  0°  to  40°,  then  at  1-degree 
intervals  from  40°  to  the  temperature  at 
which  apparent  photosynthesis  became 
zero,  the  temperature  compensation 
point.  These  measurements  were  made  on 
the  same  plants  that  were  used  for  the 
light  saturation  measurements  described 
above.  The  six  races  have  their  maximum 
rate  of  photosynthesis  at  30°.  This  is 
true  for  19  of  the  20  clones.  The  exception 
is  a  Los  Trancos  clone  that  has  a  slightly 
higher  rate  at  35°  than  at  30°. 

Between  20°  and  40°,  where  photo- 
synthesis is  about  75  per  cent  of  its  maxi- 
mum rate,  there  are  small  differences 
between  the  curves  of  the  six  races.  At 
the  low  and  high  ends  of  the  temperature 
range  the  races  show  considerable  differ- 
ences in  their  ability  to  photosynthesize 
effectively.  Figure  2  shows  these  differ- 


ed 

o 

>*_ 

X 

o 

E 


<L> 


c 

(/> 
o 

o 

SZ 
Q. 


c 

O 
Q. 
Q. 

< 


10         12  "  42        44       46       48 

Temperature,  °C 


Fig.  2.     Rate  of  light-saturated  photosynthesis  versus  temperature.  Identification  of  the  curves 
same  as  in  figure  1. 


DEPARTMENT    OF   PLANT   BIOLOGY 


315 


ences  on  an  expanded  scale  by  omitting 
the  top  half  of  the  rate  curves  and  the 
rates  between  12°  and  42°.  The  photo-syn- 
thetic rate  at  0°  varied  from  7  to  27  per 
cent  of  the  maximum  rate  among  the  20 
clones.  The  mean  values  for  the  clones  of 
each  race  were  about  20  per  cent  for  the 
Los  Trancos  race  and  between  11  and  14 
per  cent  for  the  other  five  races.  At  low 
temperatures  the  Los  Trancos  race  has 
the  highest  photosynthetic  rates,  Priest's 
Grade  and  Jacksonville  the  lowest.  Above 
40°  the  Baja  California  race  has  the 
highest  rates,  and  San  Antonio  Peak 
the  lowest.  The  temperature  compensa- 
tion point  (mean  of  the  clones)  is  46° 
for  the  San  Antonio  Peak  race;  47°  for 
Priest's  Grade,  Jacksonville,  and  Yosem- 
ite;  49°  for  Los  Trancos  and  Baja 
California. 

Continuous  photosynthesis.  Each  plant 
was  allowed  to  photosynthesize  con- 
tinuously for  12  hours  under  saturating 
light  intensity  and  at  constant  tempera- 
ture. The  photosynthetic  rate  was  meas- 
ured every  5  minutes.  All  the  plots  of 
rate  against  time  in  these  experiments 
have  the  same  general  shape.  The  rate 
rises  to  its  maximum,  then  decreases 
during  the  remainder  of  the  12  hours. 
The  curves  differ  in  the  time  taken  for 
the  rate  to  reach  its  maximum  and  in  the 
extent  of  the  subsequent  decrease  in  rate. 
Each  plot  shows  some  random  fluctua- 
tions of  several  per  cent  in  rate  during  the 
time  after  the  maximum  is  passed.  There 
is  wide  variation  in  the  amount  of  such 
fuzziness  observed  with  different  clones. 
This  effect  tends  to  conceal  the  true  trend 
of  the  curves. 

Comparison  and  evaluation  of  the 
curves  were  facilitated  by  applying  a 
smoothing  procedure  to  them.  At  each  5- 
minute  point  on  the  time  scale  we  plotted 
the  mean  of  the  five  rates  determined  in 
the  interval  from  10  minutes  before  to 
10  minutes  after  the  designated  point. 
This  largely  eliminated  the  sawtooth 
appearance  of  the  original  curve  and  made 
the  trend  readily  visible.  Comparison  of 
different  curves  was  further  improved  by 


plotting  each  of  the  above  mean  rates  as 
percentages  of  the  maximum  rate. 

Continuous  light  saturated  photosyn- 
thesis was  measured  for  12  hours  at 
several  temperatures  on  the  three  Los 
Trancos  and  three  Yosemite  clones. 
Within  the  probable  reproducibility  of 
the  curves,  no  difference  in  shape  was 
observed  at  20°,  25°,  30°,  and  35°.  There- 
fore, the  clones  of  the  other  races  were  all 
run  at  30°,  the  temperature  giving  the 
maximum  photosynthetic  rate. 

The  body  of  data  now  available  indi- 
cates that  the  rate  versus  time  curve 
during  12  hours'  continuous  photosyn- 
thesis is  not  a  critical  measurement  for 
detecting  small  differences  between  the 
performance  of  various  Mimulus  clones. 
Duplicate  curves  for  the  same  clone  may 
differ  as  much  as  those  made  on  different 
clones  of  the  same  race. 

Mean  curves  were  calculated  for  the 
clones  within  each  race.  These  curves, 
illustrating  the  performance  of  the  six 
races,  are  shown  in  figure  3.  To  make  the 
small  differences  between  the  curves 
visible,  only  the  top  third  of  the  vertical 
scale  is  shown  in  the  figure.  There  is  a 
marked  difference,  0.5  to  3.5  hours,  in 
the  time  taken  by  different  races  to  reach 
their  maximum  photosynthetic  rate.  The 
least  time  is  taken  by  the  Priest's  Grade 
race,  followed  by  the  San  Antonio  Peak 
and  Jacksonville  races  in  the  first  hour. 
The  Yosemite  race  reaches  its  maximum 
rate  in  about  1.5  hours.  The  Baja  Cali- 
fornia and  Los  Trancos  races  require  3.0 
and  3.5  hours,  respectively,  to  reach  their 
maxima. 

After  reaching  its  maximum,  the  photo- 
synthetic rate  of  each  race  declines  in  an 
irregular  fashion  during  the  rest  of  the  12 
hours.  The  least  loss  in  rate  is  shown  by 
the  Los  Trancos  race,  which  at  the  end  of 
12  hours  photosynthesizes  at  92.5  per 
cent  of  the  maximum  rate.  There  is 
probably  no  real  difference  between  the 
curves  for  the  Priest's  Grade  and  the 
Jacksonville  races.  They  follow  nearly  the 
same  course  during  the  12  hours,  ending 
at  87  and  85  per  cent  of  the  maximum 


316 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Fig.  3.     Course  of  photosynthetic  rate  with  time.  Identification  of  the  curves  same  as  in  two 
preceding  figures.  Reading  down  at  zero  time  the  curves  are  Y,  P,  S,  B,  J,  and  L. 


rate.  Except  for  taking  longer  to  reach  its 
maximum,  the  curve  for  the  Baja 
California  race  is  hardly  distinguishable 
from  the  Priest's  Grade  and  Jacksonville 
curves.  Although  it  ends  at  82  per  cent 
of  maximum  rate,  the  curve  for  the  San 
Antonio  Peak  race  is  decidedly  below  the 
curves  of  the  races  just  mentioned  during 
most  of  the  12  hours.  The  Yosemite  race 
shows  the  sharpest  decline  in  rate,  its 
curve  ending  at  74  per  cent  of  the  maxi- 
mum rate. 

Dark  carbon  dioxide  evolution.  At  the 
start  of  our  work  we  used  the  classical 
approach,  alternating  measurements  of 
photosynthetic  rate  with  measurements 
of  dark  C02  evolution  and  correcting  the 
apparent  rate  of  photosynthesis  for  the 
dark  C02  evolution  rate.  We  soon 
abandoned  this  procedure,  for  three 
reasons.  First,  it  is  the  net  rate  of  C02 
uptake  that  is  of  greatest  interest  to  us. 
Second,  after  a  dark  period  it  takes  too 
long  to  reestablish  a  steady  photosyn- 
thetic rate,  particularly  at  low  tempera- 
ture or  low  light  intensity.  Third,  the 


reproducibility  of  measurements  of  dark 
CO  2  evolution  is  unsatisfactory. 

Because  of  the  intrinsic  interest  in  dark 
C02  evolution,  however,  we  measured  it 
as  time  permitted  on  different  plants 
under  different  conditions.  Enough  data 
have  now  been  accumulated  to  permit  a 
critical  evaluation  of  the  dark  C02 
evolution  by  Mimulus  plants.  The  general 
unreliability  of  these  measurements  is 
illustrated  by  two  examples. 

Using  12  members  of  one  clone  of  the 
Yosemite  race,  113  rates  of  dark  C02 
evolution  were  determined  at  different 
times  and  temperatures.  Each  measure- 
ment was  made  immediately  after  a 
period  of  light  saturated  photosynthesis. 
The  rates  of  dark  C02  evolution  were 
plotted  against  the  immediately  pre- 
ceding photosynthetic  rates.  Two  plots 
were  made,  one  in  which  both  rates  were 
expressed  as  milligrams  of  C02  per 
square  decimeter  per  hour  and  the  other 
in  which  milligrams  of  C02  liberated  in 
the  dark  was  expressed  as  percentage  of 
milligrams  of  C02  absorbed  in  the  light. 


DEPARTMENT    OF    PLANT   BIOLOGY 


317 


The  distribution  of  the  points  in  both 
plots  showed  no  correlation  between  the 
rate  of  dark  C02  evolution  and  the 
immediately  preceding  photosynthetic 
rate. 

The  113  measurements  were  at  5- 
degree  steps  from  0°  to  40°.  The  rates  at 
each  temperature  were  averaged,  and  a 
line  was  drawn  through  the  means.  The 
rate  of  dark  C02  evolution  followed  an 
irregularly  rising  course  from  0°  to  40°. 
There  was  a  wide  scatter  in  the  values  at 
each  temperature.  The  35  values  at  20° 
showed  a  sevenfold  variation  from  the 
lowest  to  the  highest.  By  elimination  of 
the  5  highest  values,  the  scatter  of  the 
remaining  30  was  reduced  to  2.2-fold. 
The  wide  scatter  at  other  temperatures 
was  also  caused  by  a  few  values  much 
higher  than  the  range  of  the  majority. 
The  cause  of  occasional  very  high  values 
of  dark  C02  evolution  is  unknown. 

An  equally  confused  picture  was  ob- 
tained by  applying  the  same  treatment  of 
data  to  41  measurements  of  dark  C02 
evolution  made  on  one  clone  of  the  Los 
Trancos  race. 

These  apparent  irregularities  in  the 
rate  of  dark  C02  evolution  were  confirmed 
by  a  few  experiments  designed  to  test  the 
reliability  of  such  measurements.  The 
generally  accepted  trends  of  dark  C02 
evolution  were  confirmed  qualitatively.  It 
is  greatest  just  after  a  period  of  vigorous 
photosynthesis.  It  decreases  as  the  plant 
remains  dark,  and  reaches  a  roughly  uni- 
form level  overnight.  The  rate  is  not 
measurably  affected  by  C02  concentra- 
tions between  75  and  500  ppm  in  the  air 
surrounding  the  plant.  But,  even  with  a 
plant  that  had  been  dark  overnight, 
duplicate  measurements  of  dark  C02 
evolution  differed  by  as  much  as  25  per 
cent.  When  the  plant  had  not  been  in  the 
dark  for  some  time,  duplicate  rate 
measurements  showed  up  to  twofold 
differences. 

A  single  measurement  of  the  rate  of 
dark  C02  evolution  is  of  doubtful  value, 
and  making  enough  measurements  to 
provide  a  trustworthy  mean  is  very  time 


consuming.  For  these  reasons,  and  also 
because  the  rate  of  dark  C02  evolution 
reflects  the  operation  of  other  processes 
besides  respiration,  we  question  the 
validity  of  "correcting"  photosynthetic 
rates  of  Mimulus  leaves  for  the  rate  of 
dark  C02  evolution. 

Growth  Studies  in  Controlled 
Environments 

Willia?n  M.  Hiesey,  Harold  W.  Milner, 
and  Malcolm  A.  Nobs 

The  facilities  for  growing  plants  under 
controlled  conditions  have  been  improved 
during  the  year.  Four  small  cabinets  as 
described  in  earlier  reports  (Year  Books 
54,  p.  350;  59,  p.  319)  have  been  com- 
pleted. Recent  modifications  in  the 
cabinets  provide  for  different  light  in- 
tensities and  also  for  alternating  day  and 
night  temperatures.  The  power  supply 
was  increased  to  permit  full-time  opera- 
tion of  four  units  utilizing  artificial  light 
and  an  additional  two  units  utilizing 
natural  daylight  in  the  greenhouse.  Dr. 
French  and  Mr.  Clair  have  improved  the 
performance  of  the  C02  controller  men- 
tioned in  Year  Book  58,  p.  352.  Our 
operational  facilities  for  the  study  of 
cloned  plant  materials  in  controlled 
environments  are  beginning  to  approach 
an  optimum  level  for  the  present  size  of 
our  staff.  Studies  utilizing  these  facilities 
are  being  closely  coordinated  with  quanti- 
tative measurements  on  rates  of  apparent 
photosynthesis  and  dark  C02  evolution. 

Some  interaction  effects  between  light 
intensity,  temperature,  and  carbon  dioxide 
concentration  on  the  growth  of  Mimulus 
races.  Among  the  questions  that  arise 
immediately  in  the  comparative  study  of 
ecological  races  are  those  basic  to  formu- 
lating effective  experiments  designed  to 
reveal  the  characteristics  of  the  climatic 
races.  An  example  of  some  of  the  inter- 
acting effects  of  three  external  variables — 
temperature,  light  intensity,  and  C02 
concentration — on  two  clones  of  Mimidus 
cardinalis,  one  from  Los  Trancos  along 
the  cool  coastal  area  of  central  California, 


318 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


the  other  from  Priest's  Grade  at  800  feet 
elevation  in  the  warm  interior  foothill 
region  of  the  Sierra  Nevada,  will  serve  to 
illustrate  the  method. 

The  marked  differences  in  light  satura- 
tion values  of  the  two  races  as  a  function 
of  temperature  are  shown  by  curves  L 
and  P  of  figure  1.  The  curves  of  figure  1 
are  based  on  quantitative  measurements 
made  in  a  short  time.  In  experiments  in 
cabinets  in  which  both  light  intensity  and 
temperature  were  held  at  different  values 
the  growth  in  the  clones  was  compared 
over  a  3 5- day  period,  starting  with  small 
rooted  cuttings  and  ending  with  plants  in 
the  flowering  or  near-flowering  stages. 
Some  of  the  results  of  such  an  experiment 
are  illustrated  in  figure  4,  plate  1. 

At  the  low  light  intensity  of  approxi- 


intensity  the  total  growth  of  the  Los 
Trancos  clone  is  much  less  than  that  of 
the  clone  from  Priest's  Grade.  Increasing 
the  light  intensity  to  4300  footcandles  at 
30°C  increases  the  growth  of  both  clones 
expressed  in  total  dry  weight,  but  the 
amount  of  increase  is  relatively  much 
greater  on  the  Los  Trancos  clone,  as  is 
shown  in  table  1. 

The  cooler  temperature  of  20°C 
markedly  increases  growth  of  both  races 
as  compared  with  30°C,  both  at  low  and 
at  high  light  intensities,  but  the  enhance- 
ment is  greater  on  the  coastal  Los  Trancos 
clone  than  on  the  inland  Priest's  Grade 
clone.  An  unpredicted  result  is  an  actual 
reduction  in  dry  weight  at  the  high  light 
intensity  on  the  clone  from  Priest's 
Grade  when  grown  at  20°C,  as  shown  in 


TABLE  1.     Differential  Effects  of  Light  Intensity  and  Temperature  on  Growth  of  Two  Mimulus 

Clones 

See  text  and  figure  4. 


20c 


30c 


Clone 


Low  Light   High  Light   %  Difference  Low  Light   High  Light   %  Difference 
Intensity       Intensity         at  High         Intensity       Intensity         at  High 
(2300  fc)        (4200  fc)        Intensity        (2300  fc)        (4200  fc)        Intensity 

6546-5,  Los  Trancos 

(coastal  race)  1.26  ±  0.07*  1.33  ±  0.11  0.0      0.37  ±  0.08     1.07  ±  0.22       +66.3 

7210-1,  Priest's 

Grade  (interior 

race)  1.67  ±0.10     1.26  ±0.09       -27.1       1.10  ±0.14     1.73  ±0.21       +36.4 

*  Mean  dry-weight  increase  in  grams  per  propagule  over  a  35-day  period;  10  propagules  of  each 
clone  used  in  each  experiment. 


mately  2300  footcandles  (fig.  4,  lower 
row)  the  clone  from  Los  Trancos,  6546-5, 
is  light-saturated  both  at  20°  and  at  30° 
C,  but  the  clone  from  Priest's  Grade, 
7210-1,  is  barely  light-saturated  at  20°C 
and  not  nearly  saturated  at  30°C.  At  the 
high  light  intensity  of  4200  footcandles 
(fig.  4,  upper  row)  both  clones  are  fully 
light-saturated  at  both  temperatures. 
The  differential  growth  responses  of  the 
two  clones  are  evident  in  the  figure, 
especially  at  30°C,  where  under  low  light 


table  1.  In  contrast,  the  dry- weight  yield 
on  the  Los  Trancos  clone  was  the  same  at 
both  light  intensities  at  this  temperature, 
verifying  a  prediction  based  on  the  fact 
that  under  both  intensities  this  clone  was 
light-saturated. 

When  the  same  two  clones  are  sub- 
jected to  high  day  temperature  (30°C) 
and  cool  nights  (15°C)  at  both  low  and 
high  light  intensities,  a  marked  increase 
in  total  growth  per  unit  time  results  in 
both  clones,  the  increase  being  the  great- 


Plate  1 


Department  of  Plant  Biology 


© 

o 


7 


\ 


to  i «  *  i  . . 

illii 


m  1  ■  M 


u 

o 

o 

CM 


<jE>  us     3  i 

•  SI  •  ■  ■  I 


o 

JS 

p4 

-1-3 

3 

c3 

O 

o 

O 

+3 

CO 

C3 

-t-= 

■4J 

•+3 

+3 

H 

0) 

b£ 

cu 

"C 

t/i 

CD 

A 

rC 

hfl 

+3 

o 

£ 

o 

£ 

O 

O 

© 

T— 1 

oi 

0) 

S-. 

o3 

co 

1? 

53 

o 

Cj 

bfi 

^ 

-^ 

<K 

5*H 

^CD 

>> 

CD 

^ 

<3 

-t-3 

T3 

J3 

CQ 

o 

^ 

CO 

o 

CD 

-G 

So 

<M 

T-H 

Cj 

r^ 

Xi 

o 

.2 

S-l 

"f-l 

ft 

go 

.2 

S3 

.2 

+3 

'■+3 

"t/2 

c3 

fl 

(3 

0) 

•+3 

.9 

1 
£5 

■+-> 
^ 

t— i 

bJD 

^H 

o 

O 

£ 

© 

-t-3 

CO 

CM 

T3 

fH 

£ 

<S 

o 

f-i 

CO 

CD 

cd 

r 

£ 

+= 

o 

c3 

q 

xi 

+3 

£ 

s3 

i> 

o 

-t-3 

CD 

o 

co 

-1-3 

O 

S-3 

o 

+3 

>> 

-1-3 

GO 

o> 

CO 

a 

S3 

^ 

CD 

+3 

"3 

.s 

CO 

-t-3 

^ 

rG 

^ 

bC 

o 

4 

O 

o 

O 

CO 

£ 

^ 

+3 

X 

!H 

CD 

o 

#> 

CO 

°CD 

o 

O 

co 

CD 

a 

fH 

o 

o 

£ 

co 

o 

CD 

m 

tf 

a 

Q 

Q 

"* 

G 

fci) 

CD 

s 

+3 

a 

o 

Plate  2 


Department  of  Plant  Biology 


Fig.  5.  Upper:  Clones  of  Mimulus  cardinalis  grown  in  an  atmosphere  containing  a  high 
(1250  ppm)  and  a  normal  (300  ppm)  C02  concentration.  Temperature  held  constant  at 
25°C  and  the  light  intensity  at  2000  fc  during  12-hour  days.  The  scales  are  10  cm  high. 

Lower:  The  plant  cabinets  and  the  C02  controller. 


DEPARTMENT   OF   PLANT   BIOLOGY 


319 


est  in  the  Los  Trancos  clone  at  the  low 
light  intensity.  These  spectacular  in- 
creases are  thought  of  as  being  associated 
with  a  reduced  rate  of  respiration  during 
the  colder  night  period  that  conserves 
stored  food  reserves  for  increased  growth. 
The  increase  in  both  clones  was  much  less 
marked  at  the  high  light  intensity,  an 
observation  that  challenges  further  study. 
The  importance  of  C02  concentration 
as  a  variable  in  influencing  the  rate  of 
photosynthesis  of  Mimulus  was  men- 
tioned in  Year  Book  59,  pages  316-317. 
Some  effects  on  relative  growth  are 
illustrated  in  figure  5,  discussed  in  the 
section  below.  It  is  still  too  early  to  formu- 
late theoretical  interpretations  about 
internal  mechanisms  or  the  ecological 
significance  of  these  observations,  but  as 
further  information  accumulates  about 
these  and  the  other  races  of  Mimulus  and 
their  hybrid  derivatives  a  sound  basis  for 
doing  so  may  be  expected. 


A  Control  system  for  Carbon  Dioxide 

Concentration  in  Plant  Growth 

Chambers 

C.  S.  French,  R.  W.  Clair,  and 
W.  M.  Hiesey 

For  several  years  we  have  experi- 
mented with  various  types  of  controllers 
for  maintaining  desired  concentration 
levels  of  CO 2  in  the  atmosphere  of  plant 
growth  chambers. 

The  device  now  in  use  is  shown  sche- 
matically in  figure  6.  It  appears  to  work 
reasonably  well,  and  two  successful  runs 
of  several  weeks'  duration  have  been  com- 
pleted with  different  climatic  races  of 
Mimulus.  A  description  of  the  model  now 
operating  and  a  discussion  of  the  models 
previously  tested  are  being  prepared  for 
publication. 

The  responses  of  two  races  of  Mimulus 
to  high  and  normal  C02  concentration  in 
a   15-day  experiment  are  illustrated  in 


Needle  valve 


m 


Absorber|_ 
i     -i   Pump    n 
JL-,  (normally    I 


Plant 
chamber 


d  c 


wnmnml 


Absorber 
(KOH) 


f) 


i 1      pui 

_J   co2  LJbb 

J      |Analyzer[ 


Sample 
pump 

sE 


t$ 


Solenoid  n    , 

(normally  R,edu,cer, 

/  and  relief 

open )  -i  r 


Needle 

valve 

drive 

~~l — 


CO2  solenoid 

and  absorber 

control 


CO2 
supply 


Mechanical 

=  Gas 
Electrical 


Fig.  6.  Diagram  to  show  the  operating  principles  of  the  CO2  controller.  The  infrared  analyzer 
continuously  monitors  the  CO  2  concentration  in  the  plant  chamber.  When  the  concentration  deviates 
from  the  desired  level  a  motor  adjusts  the  rate  of  CO 2  input  by  a  needle  valve.  When  the  apparatus 
is  operating  below  the  normal  atmospheric  level  some  of  the  gas  is  pumped  through  a  CO2  absorber 
whenever  the  concentration  is  too  high;  when  it  is  set  above  the  atmospheric  level  the  concentration 
is  reduced  by  taking  in  outside  air. 


320 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


TABLE  2.     Dry- Weight  Increase  of  Mimulus  Clones  of  Two  CO 2  Concentrations 
Temperature,  25°C;  light  intensity,  2000  footcandles;  initial  mean  dry  weights,  0.14  and  0.17  gram  for 
clones  6546-1  and  7210-1,  respectively.     Mean  increases  in  grams  per  propagule  during  15  days. 


Clone 


CO 2  Concentration,  ppm 


1250 


300 


Increase 
at  1250  ppm,     % 


6546-5,  Los  Trancos  (coastal) 
7210-1  Priest's  Grade  (foothill) 


1.24  ±  0.05 
1.45  db  0.08 


0.65  ±  0.04 
0.94  db  0.07 


90.7 
54.3 


figure  5.  In  10  replicated  clones  C02 
enrichment  resulted  in  an  average  dry- 
weight  increase  of  90.7  per  cent  in  a 
coastal  race  (clone  6546-5)  and  of  54.3 
per  cent  in  a  Sierran  foothill  race  (clone 
7210-1),  as  shown  in  table  2.  Whether  the 
difference  between  the  races  is  of  ecologi- 
cal significance  requires  further  study. 
What  is  immediately  apparent  is  that 
CO  2  concentration  is  a  variable  that  needs 
to  be  carefully  controlled  in  growth 
experiments. 

Ecotypic  Differences  in  Response  to 

Light  Intensity  in  Solidago 

virgaurea 

Olle  E.  Bjorkman,  K.  Paul  Holmgren, 
and  Malcolm  A.  Nobs 

The  light  intensity,  varying  widely 
between  different  habitats,  is  often  a 
highly  important  factor  in  determining 
the  local  distribution  of  different  races 
and  species  of  plants.  At  the  Institute  of 
Plant  Systematics  and  Genetics  at  Upp- 
sala, Sweden,  a  study  is  under  way  on  the 
effect  that  different  light  intensities  dur- 
ing development  have  upon  the  photo- 
synthetic  process.  This  program  was 
designed  to  supply  basic  information  in  a 
cooperative  study  of  the  physiology  and 
genetics  of  ecotype  differentiation  which 
is  being  conducted  by  the  Institute  of 
Plant  Systematics  and  Genetics  and  the 
Department  of  Plant  Biology  of  Carnegie 
Institution. 

The  genus  Solidago  was  selected  for 
these  Swedish  investigations.  In  the 
northern  half  of  the  eastern  hemisphere 


this  genus  is  primarily  represented  by  one 
species,  Solidago  virgaurea  L.,  which  is 
found  nearly  continuously  from  North 
Africa  to  the  Arctic  Ocean  and  from  the 
Atlantic  coast  to  the  Pacific.  The  species 
is  particularly  well  suited  for  these 
studies,  for  it  contains  a  great  number  of 
distinct  ecological  races  within  its  wide 
distribution,  occupying  habitats  ranging 
from  dense  forests  to  open  heaths  and 
meadows. 

The  studies  were  focused  on  two  con- 
trasting races  that  differ  widely  in  the 
intensity  of  the  light  received.  The 
Hallands  Vadero  race  grows  in  the 
subdued  light  of  the  oak  forests  on  the 
island  of  Hallands  Vadero,  just  off  the 
coast  of  southern  Sweden.  The  Beskades 
race  grows  in  a  habitat  with  high  light 
intensity,  an  alpine  heath  at  600  meters 
elevation  in  Finnmark  in  the  Beskades 
region  of  northern  Norway. 

Two  identical  series  of  each  race  were 
vegetatively  propagated.  In  the  Insti- 
tute's phytotron  one  series  was  cultivated 
under  a  low  light  intensity  of  3  X  104 
erg  cm"2  sec"1  (400-700  m/i),  about  700 
footcandles;  the  other  was  grown  under 
higher  light  intensity,  15  X  104  erg 
cm-2  sec"1  (400-700  mM),  about  3500 
footcandles.  Other  conditions  were  identi- 
cal: a  photoperiod  of  16  hours,  a  tempera- 
ture regime  with  a  20°C  day  and  a  10°C 
night,  and  constant  air  humidity  at  70 
per  cent. 

From  4  to  8  weeks  was  allowed  for  the 
establishment  and  development  of  the 
plants.  After  this  period  the  leaf  morphol- 
ogy and  anatomy,  the  amount  of  chloro- 


DEPARTMENT    OF    PLANT    BIOLOGY 


321 


plast  pigments  in  the  leaves,  the  light 
dependence  of  photosynthesis,  and  the 
temperature  dependence  of  photosyn- 
thesis were  compared  in  the  plants  from 
the  two  treatments. 

The  mean  values  for  the  structural 
modifications  in  three  clones  of  each  race 
are  summarized  in  table  3.  The  anatomi- 
cal modifications  were  studied  in  thin 
sections  of  living  leaves.  The  gross  leaf 
shape,  proportions,  and  venation  pattern 
are  unaltered  for  each  clone  in  either 
treatment.  The  effect  of  the  different  light 
intensities  on  leaf  area,  however,  is 
strikingly  different  between  the  two 
races.  In  Hallands  Vadero  the  leaf  area 
produced   by  the   weak  light   plants   is 


races  is  evident  in  the  nature  of  the 
chloroplasts  in  the  upper  layers  of  the 
palisade  parenchyma  in  the  plants  grown 
under  high  light  intensity.  In  the 
Hallands  Vadero  race  grown  under  high 
light  conditions,  the  chloroplasts  in  all 
plants  were  irregular,  fragmented,  and 
often  disintegrated  in  the  two  upper 
layers  of  the  palisade  parenchyma.  These 
abnormalities  in  the  chloroplasts  are  even 
present  in  the  first  layer  of  palisade  cells 
in  young  leaves  that  have  not  fully 
expanded.  This  suggests  that  in  this  race 
the  chloroplasts  are  photolabile.  All  plants 
from  the  Beskades  race  had  normal 
chloroplasts  in  the  mesophyll  tissues. 
An  analysis  of  the  plastid  pigments  from 


TABLE  3.     Comparison  of  Structural  Modifications  Induced  in  Leaves  of  Two  Races  of 
Solidago  virgaurea  Cultivated  under  Two  Light  Intensities 


Hallands  Vadero 


Beskades 


Light 

104  erg  cm-2  sec" 

(400  -  700  m/x) 


Leaf 

Area, 

cm2 


Leaf 
Thick- 
ness, 


No. 

Palisade 

Layers 


Depth  of 
Spongy 
Paren- 
chyma, 


Leaf 

Area, 

cm2 


Leaf 
Thick- 
ness, 


No. 

Palisade 
Layers 


Depth  of 
Spongy 
Paren- 
chyma, 


3  weak 
15  strong 


39.0 
20.5 


210 
290 


80 
120 


3.0 
6.5 


190 
260 


80 
110 


nearly  twice  that  produced  by  the  same 
plants  in  the  strong  light  treatment.  In 
Beskades,  on  the  other  hand,  the  leaf 
area  in  the  weak  light  plants  is  only  about 
half  that  of  those  grown  under  the  higher 
light  intensity.  No  differences  were  found 
in  the  proportions  and  size  of  the  epider- 
mal cells  either  between  the  races  or 
between  the  treatments.  A  parallel  modi- 
fication is  found  in  the  leaf  thickness,  the 
depth  of  the  spongy  parenchyma,  and  the 
number  of  cell  layers  in  the  palisade 
parenchyma.  In  both  races  the  greater 
leaf  thickness  in  the  strong  light  plants 
is  the  result  of  an  increase  in  the  number 
of  palisade  layers  coupled  with  an  in- 
crease in  the  thickness  of  the  spongy 
parenchyma. 

A  striking  difference  between  the  two 


the  Hallands  Vadero  race  shows  that  the 
plants  grown  in  the  higher  light  intensity 
have  about  half  as  much  chlorophyll  on 
a  fresh-weight  basis  as  the  corresponding 
plants  grown  in  low  light  intensity.  In 
the  Beskades  race,  plants  from  both 
strong  and  weak  light  treatments  have 
about  equal  amounts  of  chlorophyll, 
which  also  indicates  that  in  this  race  the 
chloroplasts  are  photostable  under  high 
light  intensities. 

These  differences  between  the  races  are 
connected  with  differences  in  the  light 
and  temperature  dependence  of  photo- 
synthesis. Figure  7  illustrates  typical 
light  saturation  responses  of  these  two 
races  grown  under  different  light  intensi- 
ties. The  solid  lines  indicate  plants  grown 
under  low,  and  the  broken  lines  those 


322 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


HALLANDS  VADERO 


BESKADES 


</> 


t/> 

i 

o> 

i- 

SI 

sz 

c 

tVJ 

E 

o 

o 

o 

CJ 

_c 

O 

Q. 

O 

+- 

o> 

E 

v_ 

o 

Q. 

Q. 

< 

25 


20 


15 


10 


0 
3 

~     2 


Q_ 


/      V 

' 

. 

f  / 

/ 
1 
1 
1 

/  > 

/    / 
/  / 

' 

/  / 
II 

It 

II 

II 

II 

\ 

-\ 

' 

\     V 

\  \ 



J* 

'""" 

/ 
/ 
/ 
/ 
/ 

/  _s 
IS 

i/ 

ft 
ft 

ft 

1 
1 

// 

V 

=^ 

\ 

\ 

\ 

V 

V 

•^^ 

10 


15 


20 


25 
2 


30-KT  0 


10 


15 


20 


Light  intensity,  I, erg  cm'^  sec 


25         30-10 
2    _-l 


Light  intensity ,  I,  erg  cm*^  sec 


Fig.  7.  Light  curves  of  photosynthesis  in  clone  plants  of  the  two  ecotypes,  cultivated  at  one  low 
light  intensity,  3  X  104  erg  cm-2  sec-1  (solid  lines),  and  one  high  light  intensity,  15  X  104  erg  cm-2 
sec-1  (broken  lines). 


under  high,  intensity.  The  leaf  tempera- 
ture was  held  constant  at  22°C.  In  the 
Hallands  Vadero  race  the  plants  grown 
under  low  light  intensity  consistently  had 
a  higher  photosynthetic  rate  than  those 
grown  in  the  higher  light,  which  was 
particularly  evident  at  low  light  intensi- 
ties. In  the  Beskades  race  the  plants 
cultivated  under  the  high  light  have  the 
higher  photosynthetic  rate.  In  both 
treatments  light  saturation  for  photo- 
synthesis occurs  at  a  higher  light  in- 
tensity in  the  Beskades  race  than  in  the 
Hallands  Vadero  race,  and  in  both  races 
light  saturation  occurs  at  higher  light 
intensities  in  the  plants  grown  at  high 
light  conditions. 

The  temperature  dependence  of  photo- 
synthesis in  these  two  races,  shown  in 
figure  8,  was  determined  at  a  constant 
light  intensity  of  15  X  104  erg  cm-2 
sec"1    (400-700   m/x),   about   3500   foot- 


candles.  In  both  races  the  plants  culti- 
vated under  weak  light  have  the  higher 
temperature  optimum. 

The  racial  differences  between  these 
two  ecotypes  in  their  response  to  the  light 
conditions  during  growth  are  consistent 
and  reflect  the  ecological  characteristics  of 
the  natural  environments  from  which 
they  were  obtained.  The  photosynthesis 
of  the  Hallands  Vadero  race  from  the 
shade  of  the  oak  forests  is  highly  efficient 
under  low  light  intensities,  while  the 
photolability  of  the  chloroplasts  in  the 
leaf  tissues  places  it  at  a  disadvantage 
under  high  light  conditions.  The  Beskades 
race  from  the  exposed  alpine  heath,  on 
the  other  hand,  has  photostable  chloro- 
plasts and  requires  high  light  intensities 
for  maximum  photosynthesis. 

Studies  on  another  Scandinavian  popu- 
lation from  an  exposed  alpine  habitat  at 
Tarfala,  Sweden,  show  a  pattern  essenti- 


DEPARTMENT    OF    PLANT    BIOLOGY 


323 


o 


HALLANDS    VADERO 


BESKADES 


100 
90 
80 


4 
/ 

Sy 

^"^l 

\ 

\ 
\ 

\ 

1 

\ 

10  15  20  25  30 

Leaf  temperature, °C 


/"  A" 

/ 
/ 

i' 

\ 

\ 

\ 
\ 

10  15  20  25 

Leaf  temperature, °C 


30 


Fig.  8.  Temperature  curves  of  photosynthesis  measured  at  a  light  intensity  of  15  X  104  erg 
cm-2  sec-1  in  clone  plants  of  the  two  ecotypes,  cultivated  at  one  low  light  intensity,  3  X  104  erg 
cm-2  sec-1  (solid  lines),  and  one  high  light  intensity,  15  X  104  erg  cm-2  sec-1  (broken  lines). 


ally  identical  to  that  found  in  Beskades. 
Additional  studies,  still  incomplete,  on 
other  races  of  Solidago  virgaurea  and 
closely  related  species  from  California 
indicate  that  the  adaptations  found  in 
response  to  different  light  intensities  are 
generally  valid. 

The  Aseptic  Culture  of  Excised 
Tissues  of  Mimulus 

Ruth  F.  Elliott,1  Frank  Nicholson, 
and  William  M.  Hiesey 

Recent  advances  in  the  culture  of  plant 
tissues  under  aseptic  conditions  prompted 
us  to  explore  the  application  of  these 
techniques  to  clones  of  Mimulus.  The 
encouraging  results  to  date  indicate  the 
possibility  of  making  studies  at  the 
cellular  level  on  the  same  clones  of  higher 
plants  that  are  being  used  in  transplant, 
cytogenetic,  and  physiological  investiga- 
tions. 

The  work  of  the  current  year  has  been 
devoted  mainly  to  developing  methods  for 
isolating  and  growing  callus  tissues  from 
the  cambium  of  stem  inter  nodes.  The 
practicality  of  growth  cultures  on  agar 
slants  and  in  liquid  media  has  been 
demonstrated. 

Culture  media.  The  culture  media  for 
these    experiments    are    based    on    the 


1  Now  at  the  Botany  Department,  University 
of  Auckland,  New  Zealand. 


mineral  solution  in  which  Dr.  W.  M. 
Laetsch  and  Dr.  Winslow  R.  Briggs  of 
Stanford  grow  sporelings  of  Marsilea. 
Their  basic  solution  was  chosen  because 
it  is  essentially  the  same  one  used  in 
other  tissue  culture  work,  and  with 
appropriate  additives  it  was  known  to 
support  growth  of  a  green  callus  tissue 
isolated  by  Dr.  Laetsch  from  Euonymus. 

Laetsch 's  solution  was  based  on  Knop's 
solution  and  Berthelot's  trace  element 
solution  as  modified  by  Gautheret. 
Molybdenum  was  added  according  to 
Ball  and  Street.  Calcium  and  magnesium 
were  added  in  half  the  amount  specified 
for  Knop's  solution,  and  NaFeEDTA 
instead  of  ferric  sulfate  was  the  iron 
source.  The  last  two  modifications,  sug- 
gested by  Sheat  and  co-workers  for 
White's  medium,  eliminated  the  precipi- 
tate produced  in  Gautheret 's  mineral 
solution  on  autoclaving  and  reduced  the 
drop  in  pH  that  normally  occurs  during 
autoclaving.  The  pH  was  initially  ad- 
justed to  6.0  by  the  addition  of  0.1  N 
NaOH. 

To  the  mineral  solution  was  added  3 
per  cent  sucrose  and  10  or  15  per  cent 
coconut  milk,  a  fluid  that  may  reactivate 
quiescent  cells  and  cause  their  renewed 
growth. 

Supplements  added  to  this  basal  medi- 
um were  of  three  types:  (1)  amino  acids 
or  casein  hydrolysate  or  peptone;  (2) 
yeast  extract;  and  (3)  growth  regulators 


324  CARNEGIE     INSTITUTION      OF      WASHINGTON 

of  the  auxin  type,  including  indoleacetic  the  relative  humidity  was  maintained  at 

acid  or  naphthalene  acetic  acid,  kinetin,  approximately   70  per  cent  to  prevent 

and  2,4-D.  excessive  drying.  The  best  results  were 

Methods  for  preparing  the  explants.  obtained  when  inocula  were  taken  from 
Internode  material  prepared  by  a  modi-  young,  vigorous  plants  started  from 
fication  of  the  method  described  by  cuttings  and  grown  in  one  of  the  con- 
Blakely  and  Steward2  was  successfully  trolled  cabinets.  Explants  from  green- 
established.  Internodes  from  actively  house-grown  plants  were  also  successful 
growing  plants  which  were  fully  firm  were  when  taken  from  actively  growing  young 
cut  from   the   stem   with   a   razor   and  stems. 

dropped  in  95  per  cent  ethyl  alcohol  for  a  Early    results.    The    first    attempt    at 

few  seconds  to  wet  the  surface.  They  were  establishing  stem  tissue  cultures  of  Mimu- 

then  placed  in  1  per  cent  hypochlorite  lus  cardinalis  and  M .  lewisii  from  green- 

for  10  minutes  and  washed  twice  in  sterile  house   plants  was  unsuccessful.   In  the 

distilled  water.  After  this  washing,  the  second  trial  explants  were  taken  from 

*  'epidermis"  was  stripped  aseptically,  and  young  M.  cardinalis  plants  grown  in  a 

the  explant  was  placed  in  the  culture  controlled  cabinet  at  20°C  under  a  light 

medium.  In  the  internodes  used,  second-  intensity   of   approximately    2000   foot- 

ary  growth  of  the  stem  had  started,  but  candles.    The    tissues    were    placed    in 

the  vascular  tissue  still  formed  only  a  twenty-one   modifications   of   the   basic 

narrow  zone.  In  preparing  the  explants,  medium  using  the  supplements  mentioned 

a  longitudinal  incision,  as  far  in  as  the  above,  and  successful  growth  was  attained 

woody  layer,  was  made  in  the  stem,  and  on  six  of  the  modifications.   The  most 

the  outer  layer  was  peeled  off.  This  re-  rapid  growth  was  made  on  a  medium 

moves  all  the  phloem  and  cortex,  as  the  containing  10~7  ju  per  liter  indoleacetic 

tissue  splits  at  the  cambium.  Thus  the  acid  as  a  supplement,  and  this  medium 

explants  consisted  of  a  core  of  pith  sur-  was  used  for  subsequent  work.  Successful 

rounded  by  a  narrow  zone  of  xylem  and  cultures  were  later  established  for  eight 

traces  of  cambial  tissue.  races  of  M.  cardinalis,  five  of  M .  lewisii, 

Apical  tissues  also  were  established  and  one  of  M .  verbenaceus.  Of  the  three 
from  actively  growing  tips  having  ap-  species,  M.  lewisii  was  the  most  difficult 
proximately  two  short  undeveloped  inter-  to  establish.  Apical  tissues  of  M .  lewisii 
nodes.  The  attached  leaves  were  removed  were  found  to  grow  successfully,  first 
before  the  apices  were  sterilized.  Attempts  developing  green  shoots,  then  roots, 
to  culture  rhizome  tissues  from  plants  Origin  and  structure  of  isolated  tissues. 
grown  in  soil  were  unsuccessful  because  of  Some  preliminary  transverse  sections  and 
difficulties  in  sterilization.  Root  tissues  smears  of  explants  suitable  for  micro- 
were  later  successfully  established,  how-  scopical  examination  were  made.  One  of 
ever,  from  subcultures  by  transfer  from  the  cultures  was  callus  tissue  that  had 
established  internode  tissues  that  had  grown  from  the  cambium  of  stem  inter- 
developed  roots.  nodes.  The  callus  tissue  develops  either 

Conditions  for  growth.  In  most  tissue  on  the  surface  or  at  one  end  (the  morpho- 
culture  work  reported  in  the  literature,  logical  base)  of  the  explant,  but  some- 
light  intensities  of  about  100  footcandles  times  there  is  a  proliferation  of  pith  cells, 
were  used,  and  this  procedure  was  Externally  the  new  tissue  appears  as 
followed  in  the  present  preliminary  small  groups  of  cells  which  eventually 
studies.  The  temperature  was  main-  form  ball-like  outgrowths  covering  the 
tained  between  20°  and  25°C.  Agar  slants  whole  surface  of  the  explant.  Microscop- 
were  placed  in  small  cabinets  in  which  ically  the  callus  is  at  first  more  or  less 

*  American  Journal  of  Botany,  48,  351-357,  serially  organized,   like   young  xylem  or 

1961.  phloem  tissue,  but  with  increased  activity 


DEPARTMENT    OF    PLANT    BIOLOGY  325 

the  cells  become  more  irregularly  organ-  general  were  little  known,  having  been 

ized.  The  larger  "balls"  of  tissue  consist  studied  primarily  in  a  descriptive  way. 

of  large,  loosely  organized  parenchyma-  The  biological  relationships  between  the 

like  cells,  interspersed  with  some  recogniz-  species   of  the  Poa  genus  were   largely 

able  phloem-type  cells,  and  a  central  core  unknown.    The    Department    of    Plant 

of  thickened,  scalariform  tracheids.  Biology  was  in  a  strategic  position  to 

Current  studies.   The  present  work  is  conduct  an  investigation  on  the  biological 

directed    toward    obtaining    cultures    of  relationships  of  a  genus  of  this  kind  be- 

discrete    tissues    of    selected    clones    of  cause   of  its   background   experience   in 

contrasting    altitudinal    and    latitudinal  other  groups  of  plants  and  its  unique 

races  of  Mimulus.  Work  with  these  cul-  transplant  facilities, 

tures  will  be  coordinated  with  the  trans-  The  bluegrasses  constitute  one  of  the 

plant  and  physiological  studies.  One  of  biologically    most    complex    genera    of 

the  first  considerations  is  to  explore  the  higher  plants.  In  the  course  of  the  investi- 

rate  of  growth  in  liquid  as  compared  with  gations  it  was  found  that  the  apomictic 

solid  media  as  a  means  of  speeding  up  processes  are  in  an  intricate  balance  with 

studies    on    relative    growth    rates    for  the   still  existing,   although  suppressed, 

comparative  experimental  purposes.  sexual  processes.  Crossings  between  apo- 

Mimulus  roots  in  rotated  liquid  cul-  mictic    species    often    restore    sexuality, 

tures  were  found  to  grow  much  faster  than  resulting  in  a  phase  of  partial  recombina- 

similar    explants    of    the    same    clones  tion      between      parental      characters, 

established  on  agar.  Both  the  liquid  and  Through  selection  the  apomictic  processes 

solid   media   are   useful,    the   liquid   for  can  be  restored  in  later  generations,  and 

comparative  growth  studies  and  the  solid  in  such  progenies  there  are  many  different 

for  maintaining  stock  cultures.  expressions  of  balance  between  apomixis 

The  necessity  of  adding  indoleacetic  and    sexuality.     Wild    forms    tend    to 

acid   to   the   medium   for   growing   root  stabilize  at  either  a  high  degree  of  sexual 

tissues  of  M.  cardinalis  has  been  demon-  or  a  high  degree  of  asexual  reproduction, 

st rated  in  tests  using  four  clones.  Studies  Apomictic  plants  store  a  great  deal  of 

on    the    effects    of    controlled    external  potential    variability,    but    spontaneous 

environmental  variables  on  the  relative  crossings  occasionally  occur  that  release 

growth  of  tissues  of  selected  clones  are  and  recombine  this  variability, 

planned.  In  the  bluegrasses  apomixis  combined 

with  high  polyploidy  makes  it  possible 

Studies  in  Poa  Hybridization  to  add  the  heredities  of  highly  distinct 

wild  species,  even  though  the  chromosome 

Jens  Clausen,  William  M  Hiesey,  numbers  and  the  morphological  characters 

ana  Malcolm  A.  Nobs  -,.»(>         .  1  ■,     .     ,-,                       T                t 

may  differ  widely  in  the  parents.  In  sexual 

The  Poa  program  was  initiated  in  1943  species  with  fewer  sets  of  chromosomes 

as  a  study  of  a  genus  that  has  developed  this  is  usually  not  possible.  It  was  found 

mechanisms  for  asexual  (apomictic)  pro-  that,  through  selection  in  later  genera- 

duction  of  seed  in  many  of  its  species.  The  tions  of  apomicts  crossed  with  apomicts, 

approximately  600  species  of  this  large  the   hybrid   derivatives   can  become   as 

genus  contribute  prominently  to  the  plant  vigorous  and  as  apomictic  as  wild  species 

cover  of  the  earth  from  the  warm  tern-  of  Poa,  and  such  derivatives  can  repro- 

perate   to   the  arctic  zones  of  both  the  duce  themselves  as  seed  clones  that  can 

northern     and     southern     hemispheres,  readily  be  tested  in  diverse  climates.  The 

Many  of  these  bluegrass  species  are  of  convenient   handling   of   seed    apomicts 

economic  interest  to  man.  made  it  possible  to  conduct  transplant 

When  our  Poa  program  was  started  the  experiments  with  Poa  on  a  worldwide 

reproductive  mechanisms  of  apomixis  in  scale. 


326  CARNEGIE     INSTITUTION     OF      WASHINGTON 

The  success  of  the  Poa  crossings  is  part  of  the  year  was  spent  by  Clausen 
probably  related  to  the  fact  that  species  in  evaluating  and  completing  the  ac- 
from  remote  and  contrasting  habitats  cumulated  records  of  a  series  of  crossing 
were  crossed,  producing  hybrids  that  com-  experiments  that  were  performed  in  1951 
bined  heredities  that  complemented  each  {Year  Book  51,  pp.  111-117).  The  pur- 
other  and  that  had  not  previously  been  pose  of  the  1951  hybridization  was  to 
subjected  to  hybridization  followed  by  include  species  belonging  to  other  sec- 
natural  selection.  These  experiments  dem-  tions  of  the  genus  than  those  used  in  the 
onstrated  that  it  is  sometimes  possible  to  1943-1946  experiments  and  to  broaden 
develop  new  hybrid  apomicts  able  to  the  basis  for  the  conclusions, 
compete  with  natural  apomicts  in  their  In  the  earlier  series  of  crossing  experi- 
native  habitats  even  though  the  natives  ments  52  hybrid  combinations  were 
have  presumably  been  subjected  to  attempted,  37  of  which  actually  pro- 
crossing  and  natural  selection  through  duced  hybrids.  In  the  1951  crossings,  27 
geologic  ages.  In  addition,  a  few  of  the  different  hybridizations  were  attempted 
new  hybrids  were  found  to  have  a  wide  and  16  new  combinations  were  obtained, 
range  of  tolerance  to  contrasting  climates,  In  some  of  these  crossings  hybrids  were 
so  that  they  also  could  compete  with  obtained  in  high  frequencies;  in  others, 
locally  established  bio  types  in  other  parts  none  at  all.  Our  combined  crossing  data 
of  the  world.  from   the   Poas  included   representative 

In  apomictic  high-poly ploid  plants  the  races  from  only  15  species,  or  2.5  per  cent 

inheritance  is  governed  not  so  much  by  of  the  approximately  600  species  within 

individual  genes  as  by  the   component  the    genus.    Fortunately,    however,    the 

sets  or  partial  sets  of  chromosomes,  the  species  studied  represent  about  one-third 

genomes.  It  is  now  clear  that  in  crossing  of  the  sections  of  the  genus, 
apomictic   plants  fertilization  does   not         It  is  now  possible  to  present  a  clearer 

occur  at  random.  Strong  selection  appears  summary  of  the  second  series  of  hybridi- 

to   occur   among   the   masses   of  pollen  zations  than  appeared  in  the  preliminary 

deposited  on  a  single  stigma  and  among  account  in  Year  Book  52,  table  2,  page 

the  ovules  to  be  fertilized,  which  may  be  113.  The  present  tables  4  and  5  record 

either  sexual  or  asexual,  reduced  or  un-  the  revised  data  and  list  the  fertilities  and 

reduced    with    respect    to    chromosome  later-generation  progenies.  Fifty-five  of 

number.  There  appears  to  be  something  the  second-    and  third-generation  prog- 

beyond  a  mere  chance  interplay  between  enies  were  planted  at  Stanford  during 

the  particular  genetic-physiologic  make-  the  spring  of  1962. 

up  of  the  two  parents  that  determines  the         It  is  obvious  from  the  data  that  not  all 

kind  of  fertilization  that  will  take  place,  the  species  of  Poa  can  be  intercrossed  and 

In  apomictic  plants  the  heredities  are  equally  clear  that  it  is  impossible  to  pre- 
genic  as  in  other  kinds  of  plants,  but  the  diet  which  will  be  the  successful  inter- 
hybrids  do  not  segregate  according  to  breeders.  Whether  or  not  two  species  will 
Mendelian  ratios.  The  fertilization  proc-  combine  depends  on  how  their  genomes 
ess  depends  upon  what  can  properly  fit  together  genetically  and  physiologi- 
be  added  to  the  already  complex  heredity  cally. 

of  the  maternal  parent.  Study  of  the  Poa         The    tables    indicate    widely    varying 

genus   therefore   suggests   new   concepts  degrees  of  success  in  combining  the  hered- 

about    evolutionary    processes    in    plant  ities  of  distinct  species  of  Poa.  From  the 

groups    that    combine    polyploidy   with  11   unsuccessful  hybridizations  in  table 

apomixis  in  establishing  what  may  be  5   we   progress   to   the   least   successful 

regarded  as  superspecies.  hybrids  in  table  4,  such  as  the  rare,  barely 

Evaluation    of   progenies    of   the    1951  viable  hybrid  between  P.  scabrella,  Lu- 

crossings.  In  working  on  the  monograph  cerne,  and  P.  arachnifera,  and  to  others 

reporting  the  Poa  experiments,  a  major  that  are  viable  but  completely  sterile, 


DEPARTMENT   OF   PLANT   BIOLOGY 


327 


TABLE  4.     Results  of  Interspecific  Crossings  in  Poa 


Combination 


Fi  Fer- 
No.      Nq        Per     *^T         No. 

Seed-     Fi,s     Cent    R      y      F.Prog- 

lings  r  i  s  &  enies 

/o 


Remarks 


Poa  caespitosa  hybrids 
pratensis  X  caespitosa 
compressa  X  caespitosa 
caespitosa  X  arachnifera 

Poa  arachnifera  hybrids 
scabrella  (2n  =  94)  X 

arachnifera 
Previous  scabrella-arachnifera 
crossings 
scabrella,  Los  Posas 

(2n  =  84),  X  arachnifera 
scabrella,  Watsonville 

(2n  =  82),  X  arachnifera 
scabrella,  Lucerne 

(2n  =  63),  X  arachnifera 
nervosa  9   (2n  =  81)  X 
arachnifera  d71 

Poa  scabrella,  Paso  Robles, 

2n  =  84,  hybrids 
scabrella  X  pratensis, 

Groveland  (2n  =  56) 
scabrella  X  pratensis, 

Mono  Lake  (2n  =  50) 

scabrella  X  pratensis,  Las 
Vegas,  N.  Mex.  (2ra  =  65-74) 

Poa  ampla,  Albion  (2n  =  63)  X 
P.  compressa,  Crescent 
Mills  (2n  =  42) 
ampla  X  compressa 
Poa  arida  X  ampla 

arida  (2n  =  63)  X  ampla, 
Albion  (2n  =  56) 

Reciprocal 
arida  X  ampla,  Wenatchee 

Reciprocal 
Quadruple  hybrids 

arida-ampla  X  ampla-alpigena 


Reciprocal,  line  4683-1 
arida-ampla  X  ampla-pratensis 

Reciprocal 
scabrella-pratensis  X 
ampla-alpigena 


Reciprocal,  line  4683-1 
ampla-alpigena  (line  4273-9) 
X  ampla-pratensis, 
Heise-Newport 
Reciprocal 


120         2         1.7       0 
330         5         1.5     10-40 
111         6         5.4     15-40 


210  52  29.4  0.2-9 

967  0  0 

1060  0  0 

1145  1  0.09  0 

57  55  96.5  0 

1050  6  0.6  0 

810  52  6.4  0-2 

630  30  4.8  6-14 


2       300  Fi  seeds;  1  F2  plant 
1       96%  of  F2  die 
4       21  F3  populations,  both 
parents  sexual 


9       Initial  sterility  high 


0       Fi  exceedingly  weak 

0      43,000  Fi  "seeds"  sown;  no 
germination 


0       Fi  weak,  sterile 

4       Fi  vigorous,  high  sterility; 
2n  =  93,  93 

8      Fi  vigorous;  2rc  =  89,  95, 
115,  116 


503 

1 

0.2 

20 

1 

Variable  F2 

630 

85 

53 

10-50 

78 

Fi  vigorous;  few  apomicts 
in  F2;  2n  =  90  to  95 

510 

0 

0 

324 

33 

52 

3-15 

0 

Fi  fairly  sterile  (2n  =  90  to 

95) 

210 

0 

0 

510 

22 

4.3 

7-15 

2 

Fi  vigorous,  fertile;  F2 
germination  good  but 
76%  died  first  year 

870 

5 

0.6 

64 

1 

5  aberrants  from  the  line 

890 

32 

3.6 

4-20 

5 

Reduced  fertility  in  Fi's 

126 

9 

7 

1-25 

0 

450 

88 

48 

1-90 

6 

Presence  of  ampla-alpigena 
pollen  increases  apparent 
sexuality 

540 

0 

0 

480         9         1.9       1-60 
79         0         0 


328 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


TABLE  5.     Unsuccessful  Hybridizations  of  Interspecific  Crossings  in  Poa 


Combination 


No. 

Seedlings 


No. 

Fi's  Seen 


Per  Cent 
Fertility 


ampla,  Spokane,  X  arachnifera 

Reciprocal 
ampla,  Albion,  X  arachnifera 

Reciprocal 
ampla  X  caespitosa 

Reciprocal 
caespitosa  X  scabrella  (2n  =  94) 

Reciprocal 
arachnifera  X  pratensis 
arachnifera  X  ampla-pratensis 

Reciprocal 
howellii  X  scabrella 

Reciprocal 
howellii  X  pratensis 
howellii  X  douglasii 

Reciprocal 
douglasii  X  arachnifera 
kelloggii  X  ampla,  Albion 


600 

0 

0 

24 

0 

0 

840 

0 

0 

30 

0 

0 

621 

0 

0 

621 

0 

0 

1128 

0 

0 

43 

0 

0 

30 

0 

0 

30 

0 

0 

30 

0 

0 

958 

0 

0 

90 

0 

0 

1771 

0 

0 

660 

0 

0 

150 

0 

0 

30 

0 

0 

150 

0 

0 

such  as  P.  pratensis  X  P.  caespitosa.  In 
contrast  with  these  are  the  easily  pro- 
duced, highly  vigorous  Fi  hybrids  that 
are  completely  sterile,  as,  for  example,  P. 
nervosa  X  P.  arachnifera  in  table  4. 


section  of  bunchgrass  poas,  such  as  Poa 
ampla,  P.  scabrella,  P.  canbyi,  and  P. 
gracillima,  combine  easily  with  many 
races  of  Poa  pratensis  of  the  circumboreal 
section  of  Pratenses   (Stolonosae).   The 


Further  degrees  of  increasing  fertility  Swedish  investigators  Dr.  Arne  Miintzing 

of   Fi   hybrids   are   represented   by   P.  and  Dr.  Erik  Akerberg  have  found  that 

scabrella,  Paso  Robles,    X   P.  pratensis,  Poa  alpina  of  the  Alpinae  (Subbulbosae) 

Groveland,  in  which  the  Fi  was  weak  and  section  likewise  combines  with  various 

nearly  sterile;  by  Poa  compressa   X   P.  forms  of  Poa  pratensis,  and  that  natural 

caespitosa,  in  which  an  Fi  of  normal  vigor  hybrids  occur.  These  hybridizations  have 

was  moderately  fertile  but  almost  all  F2  in  later  generations  resulted  in  successful 

plants  were  weak  and  died ;  and  by  P.  apomictic  hybrid  strains  that  morpholog- 

scabrella,  Paso  Robles,   X  P.  pratensis,  ically  can  be  classified  as  forms  of  Poa 

Mono  Lake,  that  resulted  in  a  fairly  high  pratensis. 

percentage  of  Fi's  which  exhibited  hybrid  Genetic  relationship  and  taxonomic  sec- 
vigor  and  tolerance  but  nevertheless  had  tions.  Poa  arachnifera  from  the  southern 
a  low  degree  of  fertility,  an  unusual  situa-  North  American  Great  Plains  probably 
tion  in  Poa  scabrella-pratensis  hybrids.  belongs    to    the    predominantly    South 

Among    the    unexpectedly    successful  American  section  Dioicopoa.   In  North 


hybrids  were  Poa  caespitosa  X  P. 
arachnifera,  an  Australian  tussockgrass 
pollinated  by  the  dioecious  Texas  blue- 


American  floras  this  species  has  generally 
been  considered  to  be  a  member  of  the 
section  Pratenses,   but  it  is  not  easily 


grass.  Both  species  are  sexual  and  yielded     crossed  with  P.  pratensis  and  also  cannot 


vigorous  and  moderately  fertile  Fi 
hybrids  from  which  23  F3  progenies  were 
planted  in  the  Stanford  garden  during  the 
spring  of  1962. 

We  have  previously  found  that  the 
members  of  the  western  North  American 


be  crossed  with  either  Poa  ampla  or  P. 
scabrella  or  the  western  bunch  poas  (table 
5). 

A  dioecious,  sexual,  and  high-chromo- 
some form  of  Poa  nervosa  from  south- 
eastern Washington  was  easily  crossed 


DEPARTMENT    OF    PLANT    BIOLOGY 


329 


with  P.  arachnifera,  however,  and  the 
hybrids  were  vigorous  although  com- 
pletely sterile  (table  4),  suggesting  some 
affinity  between  the  two  species.  P. 
nervosa  also  has  commonly  been  classified 
with  the  Pratenses,  but  it  probably  be- 
longs to  the  section  Dioicopoa.  High- 
altitude  forms  of  P.  nervosa  occur  in  the 
Sierra  Nevada,  but  they  are  exclusively 
female.  In  previous  experiments  four 
female  P.  nervosa  plants  from  near  our 
Timberline  station  were  pollinated  by 
two  forms  of  Poa  canbyi  and  by  three 
plants  of  Poa  scabrella,  Las  Posas.  The 
progenies  consisted  of  2429  plants,  all 
maternal  Poa  nervosa,  indicating  that  the 
high  Sierran  form  is  apomictic  and  lacks 
affinity  to  the  species  that  cross  with 
Poa  pratensis. 

Poa  caespitosa  of  an  Australian-New 
Zealand  group  of  species  also  does  not 
cross  with  P.  ampla  or  P.  scabrella 
(table  5)  but  crosses  readily  with  P. 
arachnifera  and  is  partly  interfertile  with 
it. 

From  the  crossing  evidence  it  therefore 
appears  that  genetically  the  species  tested 
arrange  themselves  in  two  major  com- 
plexes: the  Poa  pratensis-alpina-ampla- 
scabrella  complex  and  another  consisting 
of  P.  arachnifera,  P.  nervosa,  and  P. 
caespitosa. 

Although  in  our  hybridization  experi- 
ments P.  arachnifera  X  P.  pratensis, 
Mather,  did  not  succeed,  there  is  a  con- 
nection between  these  two  groups  of  Poa. 
Dr.  Marion  E.  Brown  of  the  University  of 


Missouri,  Columbia,  was  able  from  the 
cross  of  Poa  arachnifera  9  X  P.  pratensis, 
Troy  (Turkey),  to  develop  a  vigorous 
apomictic  hybrid  line  that  superficially 
resembles  a  large  form  of  P.  pratensis, 
suggesting  that  Poa  arachnifera  may  also 
have  entered  into  the  parentage  of  Poa 
pratensis. 

The  present  and  previous  evidence  from 
Poa  compressa  of  the  section  Tichopoa 
suggests  that  that  species  is  not  closely 
allied  to  any  of  the  two  groups  above. 
Neither  is  Poa  howellii,  an  annual  species 
of  the  Ochlopoa  section,  closely  related  to 
any  of  the  previous  groups.  Hybrids  of  P. 
howellii  were  attempted  with  P.  scabrella 
of  the  bunchgrass  section,  P.  pratensis  of 
the  Pratenses,  and  P.  douglasii,  possibly 
of  the  Dioicopoa  section,  but  none  of 
them  succeeded  (table  5). 

Fertilities  in  hybrid  progenies.  Table  4 
also  lists  the  range  of  seed  fertilities 
among  the  Fi  plants  of  these  Poa  cross- 
ings. The  fertilities  vary  greatly  from 
plant  to  plant,  for  example,  between  10 
and  50  per  cent  among  78  Fi  plants  of 
Poa  arida  X  P.  ampla  of  which  F2 
progenies  have  been  grown. 

In  the  hybrid  Poa  scabrella-pratensis  X 
P.  ampla-alpigena,  however,  the  fertilities 
of  the  quadruple  Fi  plants  varied  be- 
tween 1  and  90  per  cent,  and  table  6  lists 
the  frequencies  of  fertilities  among  44 
of  these  Fi  plants.  It  is  immediately 
evident  that  approximately  43  per  cent 
of  the  plants  had  low  fertilities,  of  less 
than   10  per  cent,   but  also  that  2   Fi 


TABLE  6.     Seed  Fertilities  of  Fi  Plants  of  Hybrid  Poa  scabrella-pratensis  X  ampla-alpigena 


Seed  Fertility,  % 


No.  Fi  Plants 


0-10 
10-20 
20-30 
30-40 
40-50 
50-60 
60-70 
70-80 
80-90 


19 
5 
6 
2 
3 
3 
2 
2 
2 


44 


330  CARNEGIE     INSTITUTION      OF     WASHINGTON 

plants  had  a  high  fertility,  of  approxi-  upon  the  abnormal  increase  in  frequencies 

mately  90  per  cent,  the  same  as  in  P.  of  Fi  hybrids  having  high  chromosome 

pratensis,  and  exceeding  the  fertilities  of  numbers  when  Poa  arida  is  pollinated  by 

the  P.  ampla  and  P.  scabrella  parents.  two  kinds  of  Poa  ampla.  For  example, 

Some  of  the  Fi  plants  of  this  quadruple  Poa   arida,    North    Platte,    4262-1    and 

cross  were  sexual,  and  others  were  already  4262-11,  were  open-pollinated  at  Mather, 

apomictic  in  various  degrees.  High  seed  and  the  seeds  were  space-planted  at  the 

fertility   and   the   ability   to   reproduce  U.  S.  Plant  Materials  Center  at  Pullman, 

apomictically  was  recombined  with  low  Washington.  Under  these  circumstances 

fertility  and   sexual  reproduction.   Two  no  ampla  hybrids  were  obtained ;  approxi- 

quadruple  Fi's,  for  example,  6310-10  and  mately  78  per  cent  of  the  progeny  were 

6310-313,  both  had  high  fertilities  rang-  of  the  apomictic  Poa  arida  type,  and  22 

ing  between  65  and  75  per  cent  and  a  per  cent  were  weaker  aberrants,  some  of 

high  percentage  of  germination,  but  their  which  died  early,  as  shown  in  table  7. 

F2's  were  highly  variable  and  weak,  as  One    plant    among    the    progeny    from 

was    also    the    progeny    of   the    equally  4262-11,    however,    was    a    spontaneous 

sexual    6310-8,    a    plant    having   a   low  hybrid,  Poa  arida  X  P.  pratensis. 

fertility  ranging  between  15  and  20  per  When  the  same  two  Poa  arida  plants 

cent.  One  moderately  apomictic  Fi  plant,  were  cage-pollinated  at  Stanford  with  a 

6310-1,  had  a  low  seed  fertility,  of  about  plant  of  Poa  ampla,  Albion,  5156-23,  45 

20  per  cent,  and  another  highly  apomictic  and  61  per  cent  of  their  progeny  were 

plant,  6310-2,  had  the  highest  fertility,  hybrids,   Poa  arida    X    ampla,  and  the 

approximately   90   per   cent.    It   visibly  percentage   of   apomictic   and   aberrant 

combines  the  hereditary  characteristics  of  progeny  decreased  dramatically.  A  similar 

the  four  parental  species  native  to  highly  result    was    obtained    with    Poa    arida, 

diverse    climates    ranging    between    34°  4262-13,    when    cage-pollinated    with   a 

and  68°N  latitude.  morphologically  very  different  plant  of 

Fertilization  mechanisms    As  indicated  Poa  ampla  having  63  chromosomes,  from 

by  the  data  of  table  4,  hybrids  in  some  near  Wenatchee,  Washington.  It  yielded 

crossings  occur  with  abnormal  frequency  52  per  cent  hybrids,  the  apomictic  Poa 

and  far  exceed  the  frequencies  observed  in  arida    plants  being  reduced  to  42.5  per 

crossings  made  during  the  1943  to  1946  cent  of  the  progeny, 

period,   in  which   the   frequency   of   Fi  The  chromosome  numbers  of  the  hy- 

hybrids  ranged  between  0.2  and  4.6  per  brids  reveal  that  something  different  from 

cent  of  the  total  progeny.  Emasculation  normal  fertilization  takes  place,  as  shown 

was  not  attempted  in  any  of  the  experi-  in  the  lower  part  of  table  7.  The  parents 

ments,  because  it  does  not  prevent  the  had  2n    =    63  and  56  chromosomes  or 

development  of  apomictic  embryos.  2n   —   63  and  63  chromosomes,  but  59 

One  of  the  two  sperm  cells  from  the  Fi  plants  had  the  high  chromosome  num- 

pollen  fertilizes  the  central  diploid  nucleus  bers  of  2n  =  86  to  96.  This  suggests  that 

of  the  embryo  sac  and  starts  the  develop-  the  hybrids  arose  from  unreduced  egg 

ment  of  the  endosperm  that  nourishes  cells  having  63  chromosomes  pollinated 

either  apomictic  or  sexual  embryos.  The  by  ampla  sperm  cells  having  a  reduced 

other  sperm  cell  presumably  may  fertilize  number  of  chromosomes.  Morphologically 

a  sexual  ovule  and  produce  a  hybrid,  or,  the  hybrids  reflected  the  preponderant 

as  indicated  later,  may  under  certain  cir-  influence  from  Poa  arida. 

cumstances  be  able  to  fertilize  the  cell  Apparently  the  presence  of  Poa  ampla 

which  would  develop  into  the  apomictic  pollen   changes  the  fertilization  mecha- 

embryo.  nism  of  Poa  arida  even  when  the  pollen  of 

The  evidence  suggesting  this  type  of  arida  is  present  in  great  abundance.  When 

fertilization  is  circumstantial  and  rests  the  Poa  arida  plants  are  exposed  to  a 


DEPARTMENT    OF   PLANT   BIOLOGY 


331 


TABLE  7.     Poa  arida,  North  Platte,  Progenies  from  Different  Kinds  of  Pollination 


P. 

arida 
Parent 

P.  ampla  Parent 

Per  Cent  Progeny 

P.  arida 

Fl. 

Hybrids 

Total 

Apomicts 

Aberrants 

4262-1, 

2n  =  63 
4262-11, 

2n  =  63 
4262-13 

Open-pollinated 

X  ampla,  Albion,  5156-23,  2n  =  56 

Open-pollinated 

X  ampla,  Albion,  5156-23,  2n  =  56 

X  ampla,  Wenatchee,  4175-1,  2n  =  63 

77.8 
43.6 
78.1 
29.6 
42.3 

22.2 

11.0 

20.3 

9.3 

5.8 

0 
45.4 

1.6 
61.1 
51.9 

64 
64 
64 

54 
52 

2n  of  Fi  Hybrids 

Total 

63 

77 

86  to  96 

4262-1 

4262-11 

4262-13 

X  ampla,  Albion,  no.  plants 
X  ampla,  Albion,  no.  plants 
X  ampla,  Wenatchee,  no.  plants 

Total  no.  plants 

1 

1 

1 
1 

23 

19 
17 

59 

24 
20 
17 

61 

mixture  of  pollens  as  in  the  Mather 
garden  containing  a  preponderance  of 
different  kinds  of  Poa  pratensis  in  the 
surrounding  meadow,  the  apomictic  de- 
velopment proceeds  unchanged.  The 
weaker  aberrants  have  Poa  arida  charac- 
teristics, presumably  resulting  from  polli- 
nation by  arida  pollen.  A  single  spon- 
taneous hybrid  P.  arida  X  pratensis 
indicates  the  possibility  of  pollination  by 
Poa  pratensis. 

It  does  not  seem  reasonable  that  Poa 
arida  should  produce  such  high  percent- 
ages of  unreduced  egg  cells  that  function 
only  when  pollen  from  Poa  pratensis  is 
present.  Another  possibility  is  that  the 
Poa  ampla  sperm  cells  can  stray  away 
from  the  normal  egg  cell  and  are  capable 
of  fertilizing  cells  that  normally  would 
develop  into  apomictic  embryos. 

It  is  also  challenging  that  in  the  recip- 
rocal cross,  Poa  ampla  X  arida,  there 
were  no  hybrids  among  720  progeny, 
although  when  P.  ampla  receives  pollen 
of  Poa  pratensis  a  fair  percentage  of 
hybrids  arise.  Poa  arida  pollen  therefore 
appears  to  be  ineffective  both  with  sexual 
and  apomictic  egg  cells  of  Poa  ampla,  in 
contrast  with  the  potency  of  Poa  ampla 
pollen  when  applied  to  Poa  arida. 

A  similar  abnormal  increase  in  the  per- 


centage of  hybrids  was  observed  in  the 
quadruple  cross,  culture  6310,  Poa  sea- 
brella-pratensis,  4711-3,  X  P.  ampla- 
alpigena,  4683-1.  Each  of  the  two 
parents  of  this  hybrid  were  recombined 
F2  plants  of  interspecific  hybrids  that 
had  become  apomictic.  Morphologically 
the  two  parent  plants  can  be  classified  as 
forms  of  Poa  pratensis  having  some 
obvious  inheritance  from  P.  scabrella  and 
P.  ampla,  respectively.  When  the  hybrid 
apomictic  plant  4711-3  is  open-pollinated, 
approximately  62  per  cent  apomicts  and 
38  per  cent  weaker  aberrants  develop. 
When  plant  4683-1  was  caged  with  it, 
however,  the  seed  harvested  on  4711-3 
produced  only  about  34  per  cent  of  the 
apomictic  4711-3  type  and  nearly  48  per 
cent  hybrids. 

This  quadruple  cross  resembles  the  Poa 
arida  X  P.  ampla  cross  in  the  abnormally 
large  number  of  hybrids  produced,  but 
the  chromosomal  situation  is  very  dif- 
ferent in  the  two  crosses,  as  is  indicated 
in  the  lower  part  of  table  8.  The  parents 
of  the  quadruple  cross  have  higher 
chromosome  numbers,  2n  =  68  and  70, 
respectively,  as  compared  with  2n  =  63 
and  56  in  the  Poa  arida-ampla  crossings. 
Considering  the  irregularities  in  the  dis- 
tribution of  the  chromosomes  when  sex 


332 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


TABLE  8.     Hybrids  from  a  Quadruple  Cross 
Poa  scabrella-pratensis,  4711-1  (2n  =  68),  X  P.  ampla-alpigena,  4683-1  (2n  =  70) 


Per  Cent  Progeny 


9  Parent 


Pollinator 


Line  4711-3 


Fi  hybrids 


Total 


Apomicts 

Aberrants 

4711-3 
4711-3 

Open-pollinated                 61.7 
(4  progenies) 
X  4683-1                       33.7 

38.3 

28.5 

0                    360 

47.8                  92 

2n  of  Fi  Hybrids  of  4711-3  X  4683-1 

56  to  58      62  to  66      68  to  74      75  to  78 

79  to  80      ! 

32  to  84      88  to  92 

No.  plants 

1                   6                 11                   6 

1 

4            29 

cells  of  Poa  are  formed,  all  29  Fi  plants 
of  the  quadruple  cross  could  have  been 
produced  from  the  union  of  an  unreduced 
sex  cell  with  another  having  a  reduced 
number  of  chromosomes.  Most  likely, 
however,  the  4  plants  having  2n  =  88  to 
92  chromosomes  were  derived  from  the 
union  of  an  unreduced  sex  cell  with 
another  having  a  reduced  number  of 
chromosomes. 

The  high  frequency  of  hybrids  in  the 
quadruple  cross  Poa  scabrella-pratensis 
X  P.  ampla-alpigena  appears  to  have  a 
different  cause  from  that  in  the  Poa 
arida  X  P.  ampla  cross.  Fertilization  of 
unreduced  embryos  obviously  did  not 
occur  to  any  great  extent  in  the  quad- 
ruple cross.  Competition  in  rate  of  de- 
velopment between  fertilized  hybrid  em- 
bryos and  apomictic  ones  would  seem  to 
be  the  factor  governing  the  shift  in 
frequency  of  hybrids  in  this  case. 

The  reciprocal  cross  Poa  ampla- 
alpigena  X  P.  scabrella-pratensis  did  not 
produce    a    single    hybrid    among    720 


plants,  an  example  that  parallels  the 
situation  in  Poa  ampla  X  P.  arida. 

This  new  evidence  on  fertilization 
mechanisms  and  chromosome  numbers  in 
Poa  brings  into  focus  data  reported  in 
Year  Book  49,  page  106,  on  selections  that 
occur  during  the  pollination  of  Poa 
ampla.  Considerably  more  information  is 
now  available;  it  is  presented  in  table  9. 

When  Poa  ampla,  Albion,  plant  4183-1, 
2n  =  64,  is  pollinated  by  Poa  pratensis, 
Mather,  4253-4,  2n  =  68,  unreduced 
"ovules"  are  predominantly  being  fer- 
tilized ;  but  when  the  same  plant  is  being 
pollinated  by  Poa  pratensis-alpigena, 
Lapland,  2n  =  74,  the  hybrids  are  derived 
exclusively  from  ovules  having  the  re- 
duced number  of  chromosomes.  The 
frequency  of  hybrids  in  these  two  crosses 
was  low,  4.6  and  2.0  per  cent,  respectively, 
indicating  that  the  hybrids  were  derived 
from  unreduced  female  cells.  It  is  possible, 
however,  that  unreduced  apomictic  cells 
were  also  being  fertilized. 

The   data  here   presented   bring  into 


TABLE  9.     Chromosome  Numbers  in  Hybrids  of  Poa  ampla,  Albion,  Plant  4183-1,  2n  =  64 


Number  of  Plants 

Parents 

Total 

2n  = 

63  to  66 

68 

70  to  73      77      80  to  84  88  to  100 

104 

117 

4183-1  X  4253-4,  2n 

=  68, 

pratensis,  Mather 

3 



2           _.           _.             14 

2 

1 

22 

4183-1  X  4050-1,  2n 

=  74, 

alpigena,  Lapland 

2 

1 

3 

-- 

-- 

6 

DEPARTMENT    OF    PLANT   BIOLOGY 


333 


renewed  focus  the  complexities  of  the 
intricate  balances  between  the  fertiliza- 
tion mechanisms  in  the  genus  Poa. 

The  North  American  Field  Pansy, 
Viola  rafinesquii 

Jens  Clausen  and  R.  B.  Channell 

While  Clausen  was  lecturing  at  Vander- 
bilt  University  in  the  spring  of  1961  (Year 
Book  60,  p.  379)  an  opportunity  presented 
itself  to  make  a  field  study  of  the  only 
North  American  member  of  the  other- 
wise Old  World  Melanium  section  of  the 
violets,  which  were  the  subjects  of  his 
experiments  and  a  series  of  papers  be- 
tween 1921  and  1931.  Dr.  Channell  and 
Dr.  Uzi  Nur  of  the  Department  of 
Biology  of  Vanderbilt  University,  Nash- 
ville, Tennessee,  have  been  associated  in 
these  studies,  which  also  include  the 
seasonal  change  in  floral  morphology  and 
in  the  mode  of  fertilization  of  a  species 
whose  geographic  isolation  from  its  Old 
World  relatives  is  puzzling. 

In  terms  of  its  technical  classification 
the  American  field  pansy  has  had  an  un- 
usually confused  history:  from  1808  to 
1958  it  was  known  under  various  names, 
such  as  Viola  tenella  rafinesque,  V.  tri- 
color L.,  V.  aruensis  Murr.,  V.  rafinesquii 
Greene,  V.  kitaibeliana  R.  et  S.  var. 
rafinesquii  (Greene)  Fernald,  and  V. 
bicolor  Pursh  in  Shinners.  Alternately,  it 
has  also  been  considered  a  native  North 
American  species  or  an  introduced 
European  weed. 

As  shown  in  the  research  before  1931 
most  of  the  European  species  of  the 
Melanium  (pansy)  section  are  genetically 
interconnected  by  being  able  to  produce 
partly  fertile  hybrids  regardless  of  their 
chromosome  numbers,  which  vary  from 
species  to  species  between  n  =  7,  8,  10, 
11,  12,  13,  17,  18,  20,  24,  and  30.  The 
Melanium  section  of  the  Viola  genus  is 
fairly  unique  in  such  tolerance  to  hy- 
bridization among  nonapomictic  species. 
The  single  North  American  species  has 
n  =  17,  determined  on  plants  originally 


from  Kansas  (Clausen,  1929,  and 
Gershoy,  1934;  from  Dr.  Ezra  Brainerd's 
stock);  from  Tennessee  (Uzi  Nur,  1961); 
and  from  Texas  (Lloyd  Shinners,  1961). 

At  the  Stanford  laboratory  in  1933 
Clausen  crossed  reciprocally  the  Kansas 
stock  of  the  American  field  pansy  with  the 
best  "combiners"  among  the  European 
species,  namely,  V.  tricolor,  n  =  13,  V. 
arvensis,  n  =  17,  and  V.  kitaibeliana, 
n  =  24.  All  seeds  resulting  from  these 
crossings  were  empty  and  did  not 
germinate,  and  Fi  hybrids  were  not 
obtained.  In  contrast,  these  three  species 
cross  easily  and  produce  highly  segre- 
gating later  generations  (Clausen,  Hered- 
itas,  15,  219-308,  1931).  Being  an 
isolated  experiment  with  a  negative  re- 
sult, the  1933  crossings  were  not  reported 
at  the  time,  but  the  record  was  clear  and 
was  preserved.  The  negative  results  in 
these  crossings  strongly  indicate  that  the 
American  field  pansy  is  genetically  well 
separated  from  its  morphologically  closest 
European  relatives. 

The  North  American  species  is  unique 
within  the  Melanium  section  because  it 
has  two  kinds  of  flowers:  from  March  to 
early  April  the  spring  flowers  are  showy, 
bluish  white  to  blue,  and  not  unlike  the 
flowers  of  the  European  V.  tricolor,  except 
that  in  the  American  species  the  orifice 
of  the  stigma  has  no  lip  to  protect  it 
against  self-pollination.  Later  in  the 
season  the  flowers  of  the  North  American 
pansy  become  inconspicuous  and  so 
different  that  the  plant  appears  like  a 
wholly  different  species:  the  late  spring 
flowers  never  open;  their  petals  are  rudi- 
mentary and  hidden  by  the  sepals;  in 
successively  later  flowers  the  four  anthers 
become  rudimentary  or  their  pollen  sacs 
abort,  but  the  anther  between  the  two 
upper  petals  remains  well  developed;  in 
the  changed  flowers  the  pistil  makes  a 
complex  twist  and  places  its  stigma  just 
below  the  fertile  anther  that  opens 
directly  into  the  orifice.  The  develop- 
mental changes  within  the  flower  are 
gradual  as  the  season  progresses,  al- 
though the  end  result  is  rather  drastic. 


334  CARNEGIE     INSTITUTION     OF      WASHINGTON 

This  late  stage  is  known  as  a  cleistogamic  relatives.    A    search    through    botanical 

one,  and  in  this  species  it  has  generally  literature  has  established  that  its  name 

been  overlooked  by  botanists.  according  to  international  botanical  rules 

None  of  the  Old  World  species  of  the  is   Viola  rafinesquii  Greene.  This  single 

Melanium  section  of  Viola  have  cleisto-  American    species    covers    a    territory 

gamic  flowers,  but  such  flowers  are  known  similar  in  size  to  that  occupied  on  the 

from  species  belonging  to  the  Nomimium  other   side    of   the    Atlantic    Ocean   by 

section,  such  as   Viola  mirabilis  L.,  so  approximately  50  species.  The  American 

named  on  account  of  its  remarkable  shift  species  extends  through  more  than  ten 

in  seasonal  development.  In  V.  mirabilis,  latitudes  from  the  Atlantic  east  coast  to 

however,  it  is  the  two  lower  anthers  that  Colorado  and  occupies  habitats  edaphi- 

persist,    and    the    pistil    bends    forward  cally  as  distinct  as  sand,  limestone,  and 

instead  of  twisting  backward  as  in  Viola  ruderal  sites. 
rafinesquii.  The  Viola  rafinesquii  situation  provides 

These  remarkable  seasonal  adjustments  some  food  for  thought  on  the  evolution 

from    open    to    self-pollination    involve  of  the  pansies.  Did  the  North  American 

systemic  and  coordinated  changes  in  the  pansy  cross  the  Atlantic   Ocean  in  an 

growth  mechanism  of  the  flower.   It  is  early  geologic  age,  or  did  the  Melanium 

significant  to  notice  that  species  of  dis-  section  originate  in  North  America  with  a 

tinct  sections  follow  different  paths  in  couple  of  pioneers  making  the  trip  in  the 

achieving  the  same  end  result,  namely,  opposite  direction,  or  did  the  continents 

regular   alternation   between   open   and  simply  float  apart?  The  posing  of  such  a 

forced  self-pollination,  and  extreme  econ-  question    is    probably    more    significant 

omy  in  pollen  production  in  the  cleisto-  than  a  final  answer.  Even  more  intriguing 

gamic  stage.  is    the    hereditary    growth    mechanism 

On  the  basis   of  both   genetical   and  that  must  regulate  the  seasonal  change 

morphological  facts  it  is  now  well  sub-  from  open  cross-pollinated  to  closed  self- 

stantiated  that  the  North  American  field  pollinated  flowering.   The  data  will  be 

pansy  is  a  native  species  and  that  it  is  presented  in  a  joint  paper  with  Dr.  R.  B. 

evolutionally  distinct  from  its  Old  World  Channell  and  Dr.  Uzi  Nur. 


BIOCHEMICAL  INVESTIGATIONS 

-^  .  _  _  Recently,  however,  de  Kouchkovsky  has 

Factors  Affecting  Oxygen  Evolution       ,    ,.    ,  ,[        i  ^         ,   ..      e    _ 

^  ^  ^  studied  the  endogenous  02  evolution  from 

from  Swiss  Chard  Chloroplasts  •    ■,  .    ■,       •       u r        i     ,     T    , ,  , 

isolated  maize  chloroplasts.  In  the  present 

work  a  study  was  made  of  the  endogenous 

02-evolving    capacity    of    Swiss    chard 

It  has  been  known  for  about  eighty     chloroplasts  and  of  various  factors  affect- 


David  C.  Fork 


years  that  chloroplasts  suspended  in  ing  02  evolution  and  consumption, 
sucrose  solutions  retain  a  limited  capacity  It  is  possible  to  study  the  following 
to  evolve  02  upon  illumination.  Since  different  processes  by  measurements  of 
Hill's  discovery  in  1937  that  substantial  02  exchange  in  chloroplasts:  (1)  02  pro- 
quantities  of  02  could  be  produced  from  duction  by  light  from  an  endogenous 
chloroplasts  supplied  with  an  appropriate  substrate  without  added  reagents ;  (2)  the 
hydrogen  acceptor,  there  has  been  a  regeneration  by  a  dark  process  of  the 
multitude  of  publications  on  the  subject,  substrate  used  for  02  evolution;  (3)  the 
Study  of  the  endogenous  02  evolution  photochemical  regeneration  of  this  ma- 
from  remaining  traces  of  the  natural  Hill  terial;  (4)  the  production  of  02  from 
oxidant    has    been    largely    neglected,  added   substrates   such   as   f erricyanide ; 


DEPARTMENT   OF   PLANT   BIOLOGY  335 

(5)  the  increase  in  02  consumption  caused  The    experiments    described    here    were 

by  light.  conducted  at  room  temperature  (19°  to 

It  appeared  worth  while  to  examine  21°C)  with  unwashed  whole  chloroplasts. 

factors  affecting  02  evolution  of  chloro-  The  time  course  of  oxygen  evolution  at 

plasts  before  and  after  Hill  oxidants  had  650  mp.  Figure  9  shows  curves  for  02 

been  added.  Furthermore,  a  comparison  production  upon  repeated  exposures  of 

of  the  action  spectra  for  02  evolution  by  whole  chloroplasts  to  the  same  intensity 

chloroplasts  before  and  after  addition  of  of    650-m/x    light   under   anaerobic    and 

a  Hill  oxidant  might  be  expected  to  reveal  aerobic  conditions.  The  chloroplasts  were 

important  differences  in  view  of  current  suspended  in  the  "standard"  circulating 

findings  of  separate  functions  for  chloro-  solution  described  above  (no  added  Hill 

phyll  a  and  accessory  pigment  systems,  oxidant).    Under    anaerobic    conditions 

Chloroplast    preparation.    Swiss    chard  (part  A,  fig.  9)  the  curves  for  02  produc- 

(Beta  vulgaris  L.  var.  cicla)  used  in  these  tion  show  a  high  initial  rate  (02  produc- 

experiments  was  grown  in  the  garden  at  tion  spike).  The  high  rate  of  02  production 

Stanford.  Mature  leaves  were  picked  as  decreases  rapidly  in  the  light,  reaching 

needed,  rinsed  with  distilled  water,  and  steady-state  net  02  evolution  after  about 

chilled  before  use.  All  the  steps  in  the  3  minutes.  Exposure  to  the  same  650-mju 

preparation  of  the  chloroplasts  were  per-  beam  3  minutes  after  the  initial  exposure 

formed  in  a  cold  room  at  3  to  4°C  in  dim  gives  rise  to  a  lower  02  spike  than  that 

green   light.    Leaf   blades   free   of   large  observed  initially.  A  dark  interval  of  10 

midribs  were  ground  in  a  solution  con-  minutes,  however,  serves  to  regenerate  an 

taining  0.4  M  sucrose,  0.01  M  NaCl,  and  02-producing  capacity  nearly  equal  to  the 

0.05  M  K2HP04-KH2P04  buffer  (pR  6.9).  original. 

The  slurry  from  the  ground  leaves  was  The  steady-state  rate  is  presumed  to 

strained  through  8  layers  of  cheesecloth  correspond  to  the  equality  between  the 

and  centrifuged  at  200  X  g  for  2  minutes,  rate  of  substrate  utilization  by  light  and 

The    supernatant    was    centrifuged    at  its  re-formation  by  both  a  thermal  and  a 

1000  X  g  for  8  minutes  to  sediment  the  separate  photochemical  process, 

whole    chloroplasts.    For    making    frag-  Figure  9,  part  B,  shows  the  effect  of 

ments,  the  chloroplasts  were  washed  by  exposing  these  chloroplasts  to  the  same 

resuspending  them  in  grinding  medium  intensity  of  650-nuz  light  under  aerobic 

and  centrifuging  again  at  1000  X  g  for  conditions.    Upon   illumination,    the    02 

8  minutes.   Chloroplast  fragments  were  production  spike  is  seen  again.  However, 

obtained  by  osmotic  rupture  of  washed  the  rate  of  02  production  declines  in  the 

whole    chloroplasts    which    were    resus-  light,    reaching    a    steady-state    net    02 

pended  in  0.01   M   NaCl  and  0.05   M  uptake    after    about    3    minutes.    The 

phosphate  buffer  for  10  minutes.  Centri-  chloroplasts  show  an  02  gulp  when  the 

fugation  at  18,800   X  g  for  15  minutes  light  is  turned  off.  The  former  dark  base 

sedimented  these  fragments.  line    is    attained    again    after    about    3 

Measurement   of  oxygen   exchange.    02  minutes  in  the  dark, 

changes  in  chloroplast  preparations  were  The  decline  in  rate  of  02  production 

measured  by  a  platinum  electrode  covered  during  the  exposure  is  much  less  steep 

by  a  thin  film  of  Teflon,  as  described  in  with  aerobic  conditions,  which  suggests 

another  section.  The  Teflon  film  permit-  that  the  re-formation  of  the  substrate  is 

ted  02  to  diffuse  through  it  but  protected  faster  with  02.  The  fall  below  the  dark 

the  electrode  surface  from  the  effect  of  base  line  is  evidence  for  a  photooxidation 

substances  in  the  chloroplast  preparation,  process  running  concurrently  with  the  02 

The  solution  used  to  grind  leaves  for  production.  The  gulp  is  caused  by  the 

isolating  the  chloroplasts  was  employed  photooxidation  process  continuing  longer 

as  the   "standard"   circulating  solution,  than  the  02  evolution  after  the  light  is 


336 


CARNEGIE      INSTITUTION      OF      WASHINGTON 


1      i      i      i      i      |      i      i      i      i      |      i      i 
Dark|  650mpt  KDark-H  650m^  \* 

•      i      ;      i      i 
Dark 

i     i     i     i     i     i     l 
*\  650  mji  kDark 

60 

- 

- 

- 

Anaerobic 

A     - 

40 

- 

I 

- 

2>     20 

1     : 

CD 

c       0 

-WjL_ 

V       : 

o 

sz 
o 

X 
CO 

CM 

O 

O    60 

Dark).  650 m^KDark-H  650m// 1* — 

Dark 

A  650mMl*Dark 

<D 

O 

£C    40 

20 

- 

Aerobic 

1                     B 

\                                       J  \-_ 7*~~~~~' 

0 

N^  — ■  —    f                                   1/ 

- 

I     i     i     i     i     1     i     i     i     i     1     i     i 

i     i     1     i     i 

i      1      i      i      I  '    i      1 

0                              5                             10 

15 

20                            25 

Time 

,  min 

Fig.  9.  O2  exchange  with  illuminated  Swiss  chard  chloroplasts  under  anaerobic  (A)  and  aerobic 
(B)  conditions.  A  dark  interval  of  17  minutes  separated  the  last  exposure  of  pait  A  from  the  first 
exposure  in  part  B.  During  this  time  air  instead  of  purified  nitrogen  (99.995  per  cent  N2)  was  bubbled 
through  the  gas  exchanger.  Intensity  of  the  650-nni  beam  =  803  ergs  cm-2  sec-1. 


turned  off.  This  means  that  an  oxidizable 
product  is  made  by  light,  so  that  the 
"photooxidation"  process  may  be  a 
photochemical  production  of  reducing 
power  rather  than  a  strict  photooxidation. 
This  reducing  power  may  be  the  reduced 
form  of  the  substrate  used  for  02  pro- 
duction. 

The  effects  described  above  for  whole 
chloroplasts  could  also  be  observed  for 
chloroplast  fragments.  Whole  chloro- 
plasts, as  well  as  chloroplast  fragments, 
retained  their  ability  to  evolve  02  for 
many  hours,  sometimes  even  after  being 
left  on  the  electrode  overnight  in  the 
dark. 

The  recovery  of  the  oxygen-evolving 
capacity  in  the  dark.  The  dark  build-up  of 
a  product  which  brings  about  increased 
02  production  was  measured  anaerobi- 
cally  by  exposing  the  chloroplasts  to 
650-m/x  light  for  4  minutes  to  deplete  this 
product.  The  light  was  then  turned  off. 


At  increasing  intervals  afterward,  the 
650-m/x  light  was  turned  on  just  long 
enough  (about  5  seconds)  for  the  spike  of 
02  production  to  reach  its  highest  point 
and  was  then  turned  off.  This  " flash" 
sampled  the  amount  of  substrate  re- 
formed but  without  depleting  it  appreci- 
ably. Figure  10  shows  that  the  02 
production  spike  has  recovered  half  its 
original  height  after  about  1.5  minutes  in 
the  dark  under  anaerobic  conditions. 
Figure  10  shows,  furthermore,  that  the 
build-up  continues  for  a  considerable  time 
in  the  dark,  and  that  the  maximum 
build-up  had  not  been  attained  after  16 
minutes. 

The  action  spectrum  for  oxygen  evolution. 
The  action  spectrum  for  the  production 
of  the  02  spike  was  measured  anaero- 
bically  for  chloroplasts  suspended  in  the 
"standard"  circulating  solution.  For  this 
purpose  a  650-m^  reference  beam  was 
turned  on,  as  described  above,  just  long 


DEPARTMENT    OF   PLANT   BIOLOGY 


337 


Q. 
</> 

c 
o 

t> 
o 


<\J 
O 


CT> 

'a> 

I 


8  10 

Time,min 

Fig.  10.  Relative  heights  of  (^-production  spikes  from  650-m/t  light  flashed  at  various  dark 
intervals  after  a  previous  4-minute  exposure  to  this  650-niju  beam.  The  light  flashes  were  on  just  long 
enough  to  permit  the  spike  to  reach  its  highest  value.  The  height  of  the  O2  production  spike  produced 
by  the  initial  4-minute  exposure  was  62  units.  The  intensity  of  the  650-m^  beam  was  803  ergs  cm-2 
sec-1.  Gas  phase,  N2. 


enough  to  allow  the  maximum  rise  in  the 
02  spike,  about  5  to  10  seconds.  This  was 
repeated  after  1 -minute  dark  intervals 
until  a  constant  response  was  attained.  A 
similar  flash  exposure  of  a  measuring 
wavelength  from  the  monochromator  was 
given,  again  just  long  enough  for  the  peak 
rate  to  be  reached,  and  was  then  turned 
off.  After  a  1-minute  dark  interval  this 
measuring  flash  was  followed  by  expo- 
sures to  flashes  of  the  650-m/x  adapting 
light  and  1-minute  dark  intervals  until  a 
constant  response  was  again  attained, 
whereupon  the  chloroplasts  were  exposed 
to  equal  quanta  of  another  wavelength 
for  the  next  measurement.  The  ratio  of 
the  peak  height  of  the  response  to  the 
measuring  beam  to  the  preceding  650-mxi 
peak  height  was  used  to  plot  the  action 
spectrum.  The  action  spectrum  for  en- 
dogenous 02  evolution  in  figure  11  shows 
peaks  at  about  650  and  480  mix.  These 
peaks  correspond  to  regions  of  maximum 
absorption  by  chlorophyll  b.  A  shoulder 
can  be  seen  around  680  m/x  which  would 
correspond  to  chlorophyll  a  absorption.  A 
check  of  the  02  production  spike  versus 


650-m/x  intensity  showed  that,  at  650  m/x 
and  at  the  intensity  used  for  the  action 
spectrum  measurements,  the  response 
fell  on  the  linear  region  of  the  curve. 

It  is  clear  that  illuminated  chloroplasts 
are  able  to  evolve  02  even  though  no  Hill 
oxidants  are  provided.  The  components 
responsible  for  the  02  evolution  must  be 
bound  to  the  chloroplasts,  since  a  con- 
tinual dialysis  of  water-soluble  substances 
takes  place  into  the  circulating  medium 
which  passes  over  the  chloroplasts. 

This  endogenous  evolution  of  02  appa- 
rently results  largely  from  the  functioning 
of  chlorophyll  b,  since  peaks  at  650  and 
480  mix  in  the  action  spectrum  for  the  02- 
production  spike  were  measured  using 
chloroplasts  without  an  added  Hill  oxi- 
dant. The  shoulder  in  this  action  spectrum 
at  680  mix  indicates  that  chlorophyll  a 
may  be  functioning  to  a  limited  extent 
also. 

The  670-mxx  form  of  chlorophyll  a  is 
believed  to  be  a  part  of  the  accessory 
pigment  system.  Presumably  this  is  the 
form  of  chlorophyll  a  active  in  the  present 
system.    The    wavelength    shift    of    the 


338 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


i — i — | — i — i — i — i — \ — i — i — i — i — | — i — i — t — i — | — r — i — i — i — | — i — i — i — r 


450 


500 


550 


600 


650 


700 


750 


Wavelength,  m^ 


Fig.  11.     Action  spectra  for  various  effects  of  light  on  02  exchange  by  isolated  chloroplasts. 

Endogenous  O2  evolution:  The  maximum  rate  of  O2  evolution  from  briefly  illuminated  chloroplasts 
without  added  Hill  reagents.  Gas  phase,  N2.Tempeiature,  19.5°C.  Measured  for  equal  incident  quanta 
at  each  wavelength  above  580  m/z  and  corrected  to  this  value  below  580  m/z  where  less  energy  was 
available.  Intensity  at  650  vein  =  477  ergs  cm-2  sec-1. 

Regeneration  of  ability  to  evolve  02'-  The  effect  of  3-minute  preillumination  by  various  wave- 
lengths on  the  02  evolution  by  650  nuz. 

O2  uptake:  The  O2  consumption  induced  photochemically  by  various  wavelengths  in  chloroplasts 
whose  02-evolving  ability  was  poisoned  with  2.7  X  10~5  M  DCMU.  Automatic  recording  with 
constant  incident  quanta  per  second.  Gas  phase,  air. 

O2  evolution  with  FeCy :  The  effect  of  different  wavelengths  on  O2  evolution  from  chloroplasts  with 
added  f erricyanide :  the  Hill  reaction.  Automatic  recording  with  constant  incident  quanta  per  second. 
At  675  m/z  the  intensity  was  365  ergs  cm-2  sec"1.  Gas  phase,  N2. 


shoulder  to  G80  rriju  can  be  caused  by  the 
addition  of  the  effect  of  the  long-wave 
side  of  the  650  band  to  the  effect  of  the 
670  band. 

The  proximity  of  the  02-evolving  step 
in  photosynthesis  to  the  accessory  pig- 
ment system  has  been  postulated  recently 
by  Witt  and  co-workers,  by  Losada  et  al., 
and  by  Duysens'  group.  Witt's  scheme, 


derived  from  studies  on  absorption 
changes,  suggests  that  electrons  from 
water  reduce  an  unknown  substance,  X 
(plastoquinone?),  and  that  the  accessory 
pigment  system  is  closely  connected  to 
this  reduction.  The  oxidation  of  this 
reduced  compound  (X-) ,  in  turn,  depends 
on  the  activation  of  chlorophyll  a  where- 
by  electrons,    removed   from   X~,    ulti- 


DEPARTMENT    OF   PLANT    BIOLOGY 


339 


<D 


a> 

c 
o 

JZ 

o 

X 

a> 

CM 
O 


0) 

o 
rr 


7~I 

y. 

i 

i 
l-e 

o 
a 

1 

i      > 
K-730 

1 

i      i 

i 

1      i      i      ,      ,      , 

o    S    S          * 

o   in   a 

^>          .      v.   1            *- 

Dark 

• — ■ — —  650  m/z. - 

Dark 
i      i      i      i      i      i" 

40 
20 

0 

-  c 

E 

D     < 
r>     i 

D 
Q 

E 

3 

o 

1 

650 
m/x 

m//- 

650 
rryx 

40 

-     £      6 

_ 

_  S    8    I  650 

K- 

-Dark- 

— H  650 

h Dark- 

— *■ 

*    f    t  rr\fi 

m^ 

20 

o 
Q 

-5C 

o 

a 

L 

_ 

0 



I 

... 

x-^> 

1 

- 

_! 

i 

• 

1          ! 

i 

i 

I      .      .      .      . 

i 

i 

0 


10 
Time,  min 


15 


Fig.  12.  The  effect  of  730-m/i  light  on  the  650-mju  02-production  spike  in  Swiss  chard  chloroplasts 
suspended  in  the  solution  described  in  the  text.  The  lower  part  of  the  figure  is  a  dark  control.  See 
text  for  details.  Intensity  of  650-mju  beam  =  803  ergs  cm-2  sec-1;  of  the  730-m/x  beam  =  459  ergs 
cm-2  sec-1.  Gas  phase,  N2. 


mately  reduce  TPN  (or  an  artificial 
electron  acceptor  such  as  potassium 
f  erricyanide) . 

Effect  of  light  on  the  recovery  of  oxygen- 
evolving  capacity.  The  02  production 
brought  about  by  the  650-m/x  beam  was 
influenced  by  a  previous  light  exposure. 
To  study  this  effect,  flash  exposures  to 
650-nux  light  were  given  an  anaerobic 
preparation  at  1-minute  intervals  until  a 
constant  response  was  attained  as  shown 
in  figure  12,  upper  left.  A  2-minute 
exposure  to  the  650-m/x  beam  was  then 
given.  This  exposure  was  followed  by  a 
1-minute  dark  interval,  whereupon  a 
far-red  light,  730  m/x,  was  turned  on  for 
3  minutes.  After  3  minutes  of  the  far-red 
exposure  the  650-m/x  beam  was  super- 
imposed on  it.  The  effectiveness  of  730 
m/x  in  increasing  the  650-m/x  02  spike  can 
be  expressed  as  the  height  of  the  650-m/x 
02  spike  after  an  exposure  to  730  m/x 
divided  by  the  height  of  the  650-m/x  02 
spike  before  an  exposure  to  730  m/x  (in 
this  case  1.37).  A  control,  with  an 
equivalent  dark  period  placed  between 
the  650-m/x  exposures,  lower  part  of 
figure  12,  gives  a  ratio  of  0.77. 


A  similar  time-course  curve  for  02 
production  and  enhancement  of  the 
height  of  the  spike  is  observed  when  a 
480-m/i  beam  is  substituted  for  650  mix. 

Figure  13  shows  the  increase  in  the 
650-m/x  O  revolution  spike  brought  about 
by  a  previous  exposure  to  730  m/x  for 
varying  times.  Since  the  build-up  by 
730-mxx  light  is  relatively  slow,  a  simul- 
taneous exposure  to  730-m/x  and  to 
650-m/x  light  does  not  give  noticeable 
enhancement.  The  effect  of  the  intensity 
of  730-m/x  light  on  the  increase  of  the 
650-m/x  02  spike  was  approximately 
linear  up  to  500  ergs  cm-2  sec-1.  At  higher 
intensities  there  is  an  appreciable  amount 
of  oxygen  evolution  sustained  in  730-m/x 
light  and  a  corresponding  drop  in  its 
effectiveness  in  causing  increased  02 
production  from  650  m/x. 

The  action  spectrum  for  the  light-induced 
recovery.  The  action  spectrum  for  the 
effectiveness  of  light  in  regenerating  the 
material  used  up  by  02  production  from 
a  650-m/x  beam  was  plotted  from  other 
data  in  the  same  set  of  measurements 
used  for  the  endogenous  02  production.  A 
point  on  the  ordinate  of  the  curve  in 


340 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


50  100 

Time,  sec 


Fig.  13.  The  effect  of  730-m/x  exposure  time 
on  the  02-production  spike  brought  about  by  a 
650-m/x  light  superimposed  on  730  m/x.  After 
730-mpi  light  had  been  on  for  the  time  indicated, 
a  650-m/x  beam  was  turned  on  long  enough  for 
the  spike  to  reach  its  highest  value.  Both  beams 
were  then  turned  off.  The  ordinate  is  expressed 
as  the  ratio  of  the  height  of  the  650-m/i  O2  spike 
after  a  730-mt*  exposure  (or  dark  period  for  dark 
control)  to  the  height  of  the  650-m/i  O2  spike 
before  a  730-m/i  exposure.  The  intensity  of  the 
730-m/i  beam  was  570  ergs  cm-2  sec-1;  of  the 
650-mit  beam,  803  ergs  cm-2  sec-1.  Temperature, 
20°C.  Gas  phase,  N2. 


figure  11  labeled  "Regeneration  of  ability 
to  evolve  02"  is  proportional  to  the  ratio: 
height  of  the  02-production  spike  from 
the  standard  650-m/x  exposure  given  1 
minute  after  a  previous  exposure  to 
monochromatic  light  divided  by  the 
height  of  the  02-production  spike  from 
the  standard  650-m/x  exposure.  The  red 
peak  in  this  action  spectrum  occurs 
around  730  m/x.  Other  measurements  of 
this  action  spectrum  gave  peaks  at  713, 
720,  725,  and  729  m/x.  Blue-green  light 
can  also  bring  about  a  response  similar  to 
that  of  far-red  light.  A  dark  control  for 
this  action  spectrum,  with  a  dark  period 


substituted  for  a  monochromatic  expo- 
sure, gave  a  ratio  of  increased  650-m/x  02 
production  of  1.03. 

The  far-red  peak  around  730  m/x  in  the 
action  spectrum  suggests  that  02  pro- 
duction might  be  influenced  by  a  phyto- 
chrome  system.  However,  the  phy to- 
chrome  system  seems  not  to  be  involved, 
since  the  time-course  curve  of  02  evolu- 
tion at  480  m/x  (not  absorbed  by  the 
reversible  phytochrome  system)  also 
shows  a  decline  similar  to  that  from 
650-m/x  light  in  the  rate  of  02  evolution 
during  the  exposure. 

At  present,  it  would  seem  reasonable 
to  attribute  the  stimulation  of  02 
production  by  far-red  to  a  light  reaction 
which  causes  the  reoxidation  of  the 
unknown  reduced  component  whose  oxi- 
dized form  is  the  substrate  for  02  evo- 
lution. 

A  low-intensity  photooxidative  process 
and  its  action  spectrum.  Oxygen  produc- 
tion by  these  chloroplasts  was  inhibited 
by  DCMU,  3-(3,4-dichlorophenyl)-l,l- 
dimethylurea,  kindly  supplied  by  Dr. 
H.  J.  Thome  of  E.  I.  du  Pont  de  Nemours 
&  Company,  Wilmington,  Delaware. 
Figure  14  shows  the  effect  of  adding  this 
herbicide  to  chloroplasts.  The  650-m/x 
light  was  again  turned  on  36  seconds 
after  addition  of  the  DCMU.  The  02 
spike  is  smaller  than  before,  and  a  rapid 
drop  in  02  evolution  is  seen  during  the 
650-m/x  exposure.  A  smaller  02  gulp  is 
also  seen  when  the  light  is  turned  off. 
Subsequent  exposures  to  650-m/x  light 
result  in  a  net  uptake  of  02.  The  light- 
dependent  02  uptake  becomes  constant  in 
magnitude  about  5  minutes  after  poison- 
ing, and  the  02  gulp  disappears. 

While  the  light  is  on,  02  is  produced 
from  a  substrate  that  is  rapidly  used  up. 
At  the  same  time  a  respiratory  stimula- 
tion is  induced  by  the  formation  of  a 
product  of  the  light  reaction.  When  the 
light  is  turned  off  the  respiratory  stimu- 
lation persists  until  this  photoproduct  is 
used  up,  thus  giving  the  02  gulp.  DCMU 
poisons  the  02  evolution  but  not  the 
photostimulated  respiration,   which  can 


DEPARTMENT    OF   PLANT   BIOLOGY 


341 


Dark — 4—650 
m/j. 


Add 
DCMU 


Dark 


650 
rr\fj. 


Dark 


650- 
m/i 


-Dark- 


8  10 

Time,  min 


14 


16 


18 


Fig.  14.  Swiss  chard  chloroplasts  exposed  to  650-nux  light  before  and  after  the  addition  of  DCMU. 
The  final  concentration  of  the  DCMU  (added  dissolved  in  0.2  ml  95  per  cent  ethanol)  was  2.7  X  10-5 
M.  The  same  intensity  650-mju  beam  (803  ergs  cm-2  sec-1)  was  used  for  all  exposures.  Gas  phase,  air. 


now  be  measured  without  interference  by 
O2  evolution. 

The  action  spectrum  for  02  uptake  was 
determined  for  this  chloroplast  suspension 
by  automatic  recording  using  equal 
numbers  of  incident  quanta.  Figure  11 
shows  that  the  peak  in  the  red  region  for 
this  effect,  labeled  "02  uptake,"  occurs 
at  690  mju.  02  uptake  as  a  function  of 
intensity  of  690  mju  is  shown  in  figure  15. 
Since  the  curve  is  half  saturated  at  about 
250  ergs  cm~2  sec-1  the  action  spectrum 


200  400  600 

Light   intensity, ergs  cm-2  sec1 

Fig.  15.  Rate  of  02  uptake  as  a  function  of 
690-mju  light  intensity  for  the  chloroplasts 
suspension  treated  with  DCMU.  The  intensity 
marked  by  the  arrow  was  used  for  the  action 
spectrum  recording. 


was  determined  by  using  a  quantum  flux 
of  67.2  ergs  cm-2  sec-1,  as  shown  by  the 
position  of  the  arrow.  Upon  the  addition 
of  trichloroacetic  acid  (TCA)  to  the 
circulating  solution  to  a  final  concentra- 
tion of  1.2  per  cent,  the  chloroplasts 
turned  olive-brown  and  02  uptake  was 
abolished  at  the  light  intensity  used  for 
the  action  spectrum.  A  large  02  uptake 
remained,  however,  when  the  TCA- 
treated  chloroplasts  were  exposed  to 
bright  white  light.  It  may  be  a  result  of 
photooxidation  reactions  similar  to  those 
described  by  Franck  and  French.  Where- 
as the  present  photooxidation  process  is 
measurable  at  low  intensity  (i.e.,  with  a 
quantum  yield  roughly  comparable  to 
that  of  photosynthesis),  it  is  believed  to 
be  like  that  described  as  respiratory 
stimulation  in  Porphyridium  (Year  Book 
60,  pp.  351-357). 

This  uptake  of  02,  in  contrast  to  the 
uptake  observed  by  Mehler  after  ethanol- 
catalase  had  been  added  as  a  "trap"  for 
the  H202,  did  not  depend  on  the  func- 
tioning of  the  O revolving  system.  Green 
fragments  of  red  algal  chloroplasts  which 
no  longer  evolved  02  after  phycobilin 
pigments  had  been  leached  out  likewise 
showed  a  light-dependent  uptake  of  02 
(Year  Book  60,  pp.  369-370). 


342 


CARNEGIE     INSTITUTION     OE      WASHINGTON 


The  effect  of  potassium  ferricyanide  on 
oxygen  production.  The  02  production  of 
chloroplasts  to  which  no  Hill  oxidant  had 
been  supplied  was  compared  with  the 
02  production  after  the  addition  of 
potassium  ferricyanide.  The  result  ob- 
tained is  shown  in  figure  16.  The  left  part 
of  the  figure  shows  02  exchanges  obtained 
when  only  the  "standard"  circulating 
solution  was  used.  The  first  exposure  to 
650  m/x  gave  a  time  course  for  02  pro- 
duction very  similar  to  the  time  course 
described  earlier.  The  ability  of  a  previous 


m/x  no  longer  stimulates  02  production  in 
the  650-m/x  beam.  In  some  experiments 
with  ferricyanide  the  time  course  for  02 
production  exhibited  a  long-term  induc- 
tion effect  with  a  protracted  0  ^produc- 
tion spike.  In  these  cases,  however, 
steady-state  02  production  was  attained 
after  about  3  minutes  in  the  light. 

The  action  spectrum  for  oxygen  evolution 
with  potassium  ferricyanide.  The  action 
spectrum  for  the  evolution  of  02  after  the 
addition  of  ferricyanide  could  be  readily 
obtained,  again  by  means  of  procedures 


1      1      1      1       1      1      1      1 
Dark 

1       1              l       1       1       l 

Dark 

i    i    i   I 

|650m^H^730 

i 

n H  Dark 

650  nyxl 

0> 

v. 

Tl/X          1" 

650  m^l     -i 

i 

CD 

CT>    60 

c 
o 

- 

160 

r~] 

/ 

- 

o 

£     4° 

- 

120 

- 

- 

<M 

o 

20 

O 

CD 

O         r\ 

■'                         J 

U 

80 

40 

0 

I 

1      = 

s   ° 

1 

i       i       i       1 

i       i 

i 

i    i   i 

1 

t        i 

25 


30 


Time ,  min 


Fig.  16.  The  effect  of  potassium  ferricyanide  on  the  time-course  curves  of  O2  evolution  and  on 
the  relative  rate  of  O2  evolution  in  the  steady  state.  Potassium  ferricyanide  was  added  to  the  circu- 
lating solution  during  the  17-minute  dark  interval.  Gas  phase,  N2.  The  intensity  of  the  650-m/i 
beam  used  was  803  ergs  cm-2  sec-1;  of  the  730-m/i  beam,  459  ergs  cm-2  sec-1. 


730-m/x  exposure  to  increase  the  650-m/x 
02  spike  is  also  seen.  During  a  dark 
interval  of  17  minutes  potassium  ferri- 
cyanide was  added  to  the  circulating 
solution  to  a  final  concentration  of 
4.3  X  10~3  M.  A  subsequent  exposure  to 
the  same  650-m/x  beam  produced  a 
steady-state  rate  of  02  evolution  about 
25  times  higher  than  that  obtained 
previously.  Moreover,  the  time  course  of 
02  evolution  at  650  m/x  in  the  presence  of 
ferricyanide  remains  at  a  high  level.  A 
previous  or  concurrent  exposure  to  730 


for  the  automatic  recording  of  action 
spectra  with  equal  incident  quanta.  The 
resulting  action  spectrum  of  figure  11  is 
labeled  "02  evolution  with  FeCy."  Its 
main  peak  was  at  678  m/x,  and  there  was 
a  broad  shoulder  in  the  640-  to  650-m/x 
region.  A  check  of  the  02  production  as  a 
function  of  intensity  of  675-m/x  light 
showed  that  the  action  spectrum  was 
determined  well  within  the  linear  region 
of  the  saturation  curve.  This  action 
spectrum  is  similar  to  a  "normal"  action 
spectrum  for  02  evolution  by  green  plants 


DEPARTMENT   OF   PLANT   BIOLOGY 


343 


like  that  of  Haxo  and  Blinks  for   Ulva 
taeniata. 

A  Teflon-Covered  Electrode 
Assembly 

David  C.  Fork 

The  bare  platinum  electrode  in  use  for 
several  years  was  remodeled  to  separate 
the  experimental  material  from  the 
platinum  by  a  thin  Teflon  membrane. 
With  the  present  system  reagents  like 
ferricyanide    that    would    influence    the 


electrode  behavior  can  now  be  used.  The 
3^-mil  Teflon  is  permeable  to  oxygen  but 
not  to  ions.  Since  it  is  an  electrical 
insulator  the  reference  electrode  is  also 
placed  under  the  Teflon.  This  system  has 
the  added  advantage  of  avoiding  the 
passage  of  current  through  the  sample. 
The  present  assembly,  like  the  one  it 
replaces,  gives  a  relative  measure  of 
differences  of  rate  in  oxygen  exchange 
between  the  sample  in  the  light  and  in 
the  dark.  Ag/Ag2O/(0.5  N  KOH)  under 
a  plastic  film  as  used  by  Clark  and  by 


a 


Cover 


/*=/    Gasket 


Dialysis 
membrane 


\ssmi 


3  Pt 


Teflon 


Ag-Ag20 


Electrode 
base 


Fig.  17.     The  Teflon-covered  Pt  electrode  for  measuring  the  rate  of  O*  production  by  chloroplasts, 


344  CARNEGIE     INSTITUTION     OF      WASHINGTON 

Carritt  and  Kanwisher  was  adopted  for  ments  of  the  rate  of  02  production  by 

a  reference  electrode.  chloroplast  preparations. 

An  expanded  view  of  the  electrode  The  fluid  circulating  system  and  gas 
assembly  and  the  relation  of  its  compo-  exchange  system  for  small  volumes  of 
nent  parts  is  shown  in  figure  17.  The  base  solution  is  shown  in  figure  18.  A  Lucite 
of  clear  Lucite  was  designed  so  that  it  centrifugal  pump  similar  to  that  de- 
could  be  positioned  reproducibly  under  scribed  in  another  report  with  a  hold-up 
the  beam  from  a  monochromator.  The  volume  of  about  3  ml  gave  a  fluid  flow  of 
platinum  electrode,  1  by  15.3  mm,  was  about  410  ml/minute  past  the  electrode, 
set  flush  into  the  top  surface  of  the  Lucite.  The  total  volume  of  circulating  fluid 
A  rectangular  Ag/Ag20  reference  elec-  required  to  fill  this  system  was  50  ml. 
trode,  22.6  by  28.2  mm,  with  the  center  The  gas  exchanger  was  made  of  Pyrex 
cut  out  was  made  from  pure  silver  0.8  mm  tubing  15  cm  long  and  4  cm  in  diameter 
thick,  and  was  also  mounted  flush  with  and  was  about  half  filled  with  sections  of 
the  top  of  the  Lucite  base  and  with  the  small  glass  tubing,  which  served  to  in- 
platinum  electrode.  The  Lucite  was  crease  the  surface  area  for  gas  exchange 
grooved  underneath  the  Ag/Ag20  refer-  and  also  prevented  gas  bubbles  from 
ence  electrode  to  make  a  pool  for  the  entering  the  centrifugal  pump.  If  nitrogen 
KOH.  The  area  of  the  Ag/Ag20  reference  was  substituted  for  air  as  the  gas  phase, 
electrode  was  about  75  times  that  of  the  a  new  stable  dark  base  line  was  estab- 
Pt  electrode.  The  Pt  and  the  Ag/Ag20  lished  after  about  8  minutes, 
electrodes  were  covered  with  a  6.4-ju-thick  The  Ag20  coating  of  the  electrode  was 
sheet  of  Teflon  film  held  down  over  the  formed  originally  by  polarizing  the  Pt 
electrodes  by  a  rubber  ring.  A  thin  film  of  electrode  at  —0.8  volt  with  reference  to 
KOH  over  the  surface  of  the  Pt  and  the  Ag  electrode  and  leaving  it  in  the  air 
Ag/Ag20  electrodes  was  trapped  under  for  a  day.  The  resulting  Ag/Ag20  refer- 
the  Teflon.  The  KOH  pool  and  the  KOH  ence  electrode  was  then  ready  for  use. 
under  the  Teflon  film  were  connected  Since  the  Teflon  membrane  is  permeable 
through  channels  cut  away  at  intervals  to  C02  in  the  atmosphere,  the  electrode, 
around  the  Pt  electrode  as  shown  in  the  when  not  in  use,  was  kept  in  a  desiccator 
insert  of  figure  17.  Leads,  soldered  to  the  over  NaOH  pellets  to  prevent  C02  from 
under  sides  of  the  Pt  and  Ag/Ag20  neutralizing  the  KOH  in  the  electrode, 
electrodes,  were  brought  out  under  the  When  in  use,  the  Pt  electrode  was 
level  of  the  rubber  ring  and  were  sealed  polarized  at  —0.8  volt  by  a  circuit  with 
in  place  with  beeswax.  The  leads  were  1.3 5- volt  Mallory  (RM-42RT)  mercury 
attached  to  binding  posts  on  the  electrode  batteries  which  also  powered  a  balancing 
base.  circuit.   The  voltage  across  a   100-ohm 

Chloroplasts  (or  cells)  were  spread  in  a  resistor  in  series  with  the  electrode  was 

thin  layer  on  the  surface  of  the  Teflon  measured    by    a    Beckman    model     14 

above  the  Pt  electrode  and  then  covered  chopper  amplifier  and  a  Varian  recorder, 

with  a  piece  of  moistened  dialysis  mem-  The  electrode  without  the  cover  showed 

brane  held  in  place  by  a  rubber  gasket  a  light  response  and  caused  the  recorder 

and  a  cover.   Circulating  fluid  entered  pen  to  go  in  the  same  direction  as  that 

through  one  end   of  the  cover,   flowed  caused   by   02   production.    This   signal 

across  the  surface  of  the  dialysis  mem-  apparently  resulted  from  a  light  reaction 

brane,  and  passed  out  the  other  end.  A  at    the    Ag/Ag20    reference    electrode, 

modified  flowing  system  suggested  origi-  When  the  gasket  and  cover  were  in  place, 

nally  by  Professor  Jack  Myers  was  used,  the  Ag/Ag20  electrode  was  shaded  and 

so  that  a  solution  at  constant  02  tension  there  was  no  longer  a  response  to  light 

was  passed  over  the  electrode  at  a  steady  with  the  highest  intensities  used  in  these 

rate.  The  electrode  thus  gave  measure-  experiments. 


DEPARTMENT   OF   PLANT    BIOLOGY 


345 


Flow 


Teflon  covered 
Platinum  electrode 


Tygon 
tubing 


Glass  tubing 
sections 


Flow 


Sync,  motor  pump 
Fig.  18.     The  circulating  and  gas  exchange  system  with  the  Teflon-covered  Pt  electrode. 


Relations  between  the  Two 

Photochemical  Reactions 

of  Photosynthesis 

C.  S.  French 

The  two-pigment  nature  of  photo- 
synthesis is  an  aspect  of  the  process 
particularly  suitable  for  experimental 
investigation  now.  Several  different  types 


of  model  schemes  have  been  proposed  to 
explain  the  various  manifestations  of  the 
two  pigment  systems.  We  therefore  need 
simple  experiments  to  distinguish  between 
various  models  of  this  part  of  the  photo- 
synthetic  process.  Two  types  of  kinetic 
experiments  appear  to  be  helpful  in 
evaluating  the  relative  merits  of  these 
models. 


346  CARNEGIE     INSTITUTION     OF     WASHINGTON 

For  practical   reasons   both   types   of  with  one  sequence  than  with  the  other, 

experiments,   02  evolution  from  paired  Furthermore,    by    giving   variable    dark 

flashes  of  two  colors  and  the  time  course  times   between   the   two   flashes,    inter- 

of  the  decay  in  02  evolution  rate  after  a  mediary  nonphotochemical  reactions  may 

light  period,  are  measured  with  the  same  be    observed    that    are    related    to    the 

preparations.  Although  both  are  in  their  activation  of  certain  pigments  but  not 

early  stages,  the  colored  flash  measure-  of  others. 

ments  have  so  far  given  clearer  results  The  measurements  were  made  with  a 

than  the  studies  on  decay  curves  for  02  Teflon-covered  electrode  similar  to  that 

evolution.    Some    illustrations    of   these  described    by    Fork    elsewhere    in    this 

experiments  are  presented  as  a  report  of  report  except  that  the  electrode  system 

work  in  progress  rather  than  as  a  com-  Au/(0.5  M  KHC03,  0.5  M  KCl)/AgCl 

pleted  study.  was  used  under  the  Teflon.  A  thin  layer 

Experiments  with  two  light  flashes  differ-  of  algae  over  the  Teflon  was  held  down  by 

ent  in  color.3  The  two-pigment  nature  of  a  cellophane  membrane  under  tension, 

the    photosynthetic    process    is    clearly  Above  the  cellophane,  the  algal  growth 

shown   by   the   enhancement   found   by  medium  was  circulated  as  described  by 

Emerson  when  both  pigment  systems  are  Fork. 

illuminated  together.  A  major  question  The  response  time  of  this  electrode 

about  the  enhancement  phenomenon  is  covered  with  Teflon  and  cellophane  was 

whether  one  specific  pigment  system  helps  measured  by  injecting  a  small  amount  of 

the  other  one  or  whether  the  enhancement  deaerated  water  into  a  stirred  bottle  of 

is  mutual.  That  is,  does  the  yield  from  air-equilibrated  water  that  was  continu- 

each  pigment  system  increase  when  the  ously    pumped    through    the    electrode 

other  one  is  also  activated?  Steady-state  assembly.    One-half   the   total   response 

rate    measurements    during    concurrent  was  achieved  in  3  seconds.  Presumably 

illumination  by  both  wavelengths  do  not  the    response    to    illumination    of    cells 

distinguish  between  the  effects  of  the  two  placed  directly  over  the  Teflon-covered 

beams,  whereas  illumination  by  the  two  electrode    would    be    shorter    than    the 

colored  beams  separated  in  time  should  response    of    this    complete    system.    A 

give  results  bearing  on  this  question.  similar  electrode  assembly  with  a  piece  of 

Several  years  ago  Myers  found  that  it  fine  nylon  stocking  to  hold  more  electro- 
was  possible  to  alternate  the  light  beams  lyte  between  the  Teflon  and  the  oxygen 
with  periods  of  a  few  seconds  and  still  electrode  was  also  tested.  This  assembly 
observe  enhancement  when  the  continu-  had  a  half-response  time  of  12  seconds, 
ous  rate  of  02  evolution  was  measured.  The  bare  Pt  electrode  previously  used 
The  present  work  is  an  attempt  to  should  be  the  best  of  all  for  speed  of 
measure  the  02  from  single  flashes  or  response,  but  its  response  speed  has  not 
pairs  of  flashes  rather  than  from  a  con-  yet  been  tested  by  this  means, 
tinuous  series  of  alternate  exposures.  The  light  beams  came  either  from  a 

In  principle,  if  one  pigment  produces  monochromator      with      supplementary 

material  necessary  for  the  action  of  the  stray  light  filters  or  from  tungsten  lamps 

other  pigment  it  should  be  possible  to  tell  with  interference  filters.  The  rest  of  the 

the  order  in  which  the  pigments  must  act.  equipment   has    been   described    in   the 

Thus  if  two  consecutive  flashes  of  different  reports    of    the    last    three    years.    The 

colors,  one  favoring  each  pigment,  are  servomotor    shutters    opened    or    closed 

given,    and    the    total    02   evolution   is  within  0.03  second  as  measured  with  a 

observed,  a  greater  effect  might  result  photocell  and  oscilloscope.  For  this  work 

,  q             |.    .                        .       mx      1     j  the  shutters  were  controlled  by  a  pair  of 

6  borne  preliminary  experiments  with  colored  ,  .                                             J        r 

light  flashes  were  made  in  collaboration  with  Dr.  motor-driven  timers. 

Fork  and  with  Dr.  Brown.  The  time  course  for  changes  in  rate  of 


DEPARTMENT    OF   PLANT    BIOLOGY 


347 


k- 
<D 

cn 

c 
o 

jC 

o 

X 

Q> 

CVJ 

o 

O 

o 


19.3  14.0 


17.1  23.0 
R  G 


19.5     18.5  17.5 
G  G  G 


Time,  min 

Fig.  19.  The  rate  of  O2  exchange  produced  by  3-second  flashes  of  green  light,  570  mju  absorbed 
mainly  by  phycoerythrin,  and  of  red  light,  680  m/x  absorbed  mainly  by  chlorophyll  a,  in  Porphyridium 
at  20°C.  The  peak  heights  are  given  above  each  spike.  A  red  flash  before  green  enhances  the  O2 
production  from  green.  A  green  flash  before  red,  however,  instead  of  enhancing  the  O2  evolution  by 
red,  reduces  it  slightly. 


O2  exchange  from  paired  flashes  is  illus- 
trated in  figure  19,  which  also  shows  the 
peak  rates  produced  when  the  flashes 
were  given  in  various  sequences. 

It  appears  that  material  made  by  red 
light  remains  in  the  cells  long  enough  to 
enhance  a  succeeding  green  flash.  For  the 
reversed  order,  however,  green  before 
red,  or  both  together  for  these  short 
times,  there  was  no  such  enhancement. 
There  is,  in  fact,  a  depression  of  the 
response  to  red  light  following  a  green 
flash.  The  effect  is  small,  but  it  has  been 
found  repeatedly. 

An  attempt  was  made  to  see  whether  a 
longer  exposure  to  green  light  absorbed 
by  phycoerythrin  in  Porphyridium  would 
produce  material  remaining  long  enough 
to  give  enhancement  of  a  red  flash.  The 
peak  height  of  a  10-second  red  flash 
before  the  green  exposure  was  10.8. 
Twenty  seconds  after  the  1.5-minute 
green  exposure  the  response  to  red  was 
10.1,  again  showing  a  small  decline  rather 
than  an  enhancement.  The  same  red  flash 
was  also  given  while  the  cells  were 
exposed  to  green  light,  both  at  the  peak 
of  the  initial  spike  of  the  green-light 
time-course  curve  and  again  after  the 
steady-state  rate  of  photosynthesis  had 
been  reached.  The  responses  to  red  light 


were  27.8  and  25.8,  respectively.  This 
shows  strong  enhancement  of  the  red 
flash  by  continuous  green  but  little 
difference  at  the  two  times,  which  were 
about  13  and  120  seconds  after  the  start 
of  the  green  exposure. 

The  peak  heights  reported  are  a 
measure  of  the  speed  with  which  photo- 
synthesis gets  started,  that  is,  a  decrease 
of  the  induction  period,  rather  than  a 
measure  of  the  steady-state  rate  or  of  the 
total  02  evolved.  A  number  of  records 
were  made  with  a  faster  paper  speed  and 
the  areas  under  the  curves  were  measured 
with  a  planimeter  to  give  the  relative 
total  amounts  of  02  produced  by  the 
pairs  of  flashes.  For  the  total  02  per  pair 
of  flashes,  the  results  were  qualitatively 
similar  but  less  clear  than  the  effects 
shown  by  the  peak  heights.  The  area 
measurements  involve  some  uncertainty 
in  deciding  when  the  final  base  line  is 
reached,  and  the  peak  measurements 
become  hard  to  interpret  except  when  the 
overlap  in  time  is  reduced  by  a  dark 
period  between  the  flashes,  as  in  figure  19. 

Following  the  observations  of  Whit- 
tingham  and  of  others  that  a  dark  period 
between  two  flashes  has  an  influence  on 
the  yield  of  the  second  flash,  we  varied 
the  time  between  the  start  of  the  red 


348 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


0  10  20  30 

Time  after  start  of  685  m// 
exposure, sec 

Fig.  20.  The  maximum  rates  of  O2  evolution 
in  Porphyridium  by  a  green  flash  given  at  various 
times  after  the  start  of  a  weak  red  flash.  Red 
light:  685  m/z,  5  sec;  green  light:  567  m/x,  2  sec. 
The  peak  height  from  red  alone  was  only  2.4 
units;  that  from  green  alone,  13.7  units. 


flash  and  that  of  the  green  flash.  The 
results  are  shown  in  figure  20.  The  first 
point  at  t  =  0  of  figure  20  shows  the  result 
when  both  flashes  are  started  together; 
for  the  second  point  the  green  was  put  on 
2  seconds  after  the  start  of  the  5-second 
red;  and  for  the  third  point,  shown  in 
duplicate,  the  green  was  given  immedi- 
ately after  the  red,  that  is,  5  seconds  later 
than  the  starting  time.  The  dropping 
part  of  the  curve  clearly  shows  the  rate 
of  disappearance  of  the  red  product  that 
makes  the  green  light  more  effective.  An 
approximate  half -life  of  18  seconds  for 
the  red-light  product  is  found  from  the 
data  of  figure  20. 

In  summary,  the  experiments  with  two 
colors  of  flashes  have  shown  that  a 
product  of  chlorophyll  a  activity  remains 


long  enough  to  enhance  a  succeeding 
flash  of  light  absorbed  by  the  accessory 
pigment.  The  product  of  the  accessory 
pigment's  action,  however,  does  not 
remain  long  enough  to  enhance  a  suc- 
ceeding red  flash. 

The  flash  times  used,  1  to  10  seconds, 
were  not  short  enough  to  detect  products 
with  half -lives  less  than  a  few  seconds. 
An  extension  to  shorter  times  of  the 
experiments  with  paired  flashes  of  differ- 
ent colors  may  be  undertaken  in  the 
future. 

Experiments  on  the  time  course  of 
oxygen  exchange  following  a  light  period. 
Another  way  to  get  at  the  interrelation 
between  the  two  pigment  systems  of 
photosynthesis  is  to  investigate  the 
kinetics  of  the  slowly  decaying  02-e volu- 
tion process  after  the  photosynthetic 
light  has  been  turned  off.  Certain  model 
schemes  predict  that  the  curve  relating 
rate  of  02  evolution  with  time  after  a 
light  period  should  be  first  order,  and 
that  its  shape  should  not  be  influenced 
by  the  color  of  the  light  previously  used 
to  drive  the  photosynthetic  reaction. 
However,  the  model  reaction  scheme  we 
used  last  year  predicts  that  the  decay 
curve  should  be  second  order  and  that  its 
shape  should  vary  with  the  ratio  of  the 
light  absorbed  by  the  two  pigment 
systems. 

These  concepts  are  so  simple  and  clear 
that  the  resolution  of  the  question  by 
experiment  might  have  been  an  easy 
matter  except  for  a  complication,  which 
is  itself  wavelength-dependent.  The  diffi- 
culty is  the  production  by  light  of 
material  that  greatly  increases  the  con- 
sumption of  02  for  a  short  time.  This 
"respiratory  stimulation,"  perhaps  better 
called  "increased  oxygen  consumption," 
is  at  its  maximum  when  the  light  is 
turned  off.  It  then  declines  rapidly  during 
the  period  of  interest  for  the  study  of  the 
lingering  02  evolution  following  a  light 
period.  The  magnitude  of  the  respiratory 
stimulation,  which  obscures  the  changes 
in  02-e volution  rate,  is  very  dependent 
on  the  previous  treatment  of  the  cells.  It 


DEPARTMENT    OF   PLANT    BIOLOGY 


349 


may  be  possible  to  design  experiments  in 
such  a  way  as  to  reduce  this  source  of 
confusion.  Fortunately,  various  species  of 
algae  show  the  respiratory  stimulation  to 
very  different  degrees  and  with  very 
different  half -lives. 

At  present  this  approach  to  the  com- 
parison of  different  theories  for  two 
pigment  mechanisms  appears  to  have 
some  potential  value  as  a  test  for  the 
relative  merits  of  contrasting  concepts  of 
the  mechanism  of  photosynthesis.  Appro- 
priate organisms  and  favorable  conditions 
for  their  use  to  this  end  are  being  sought. 
A  few  measurements  of  the  time  course 
of  the  02-e volution  rate  after  light 
exposure  under  different  conditions  will 
be  discussed  here. 

The  process  of  respiratory  stimulation 
by  light  shows  so  clearly  in  Porphyridium 
because  its  magnitude  and  its  half-life  of 
about  1.1  minutes  make  its  presence 
obvious  on  a  time  scale  convenient  for 
measurements  of  the  time  course  of 
photosynthesis.  If  the  half-life  of  the 
intermediate  carrying  the  increased  02 
consumption  had  been  an  order  of 
magnitude  higher  or  lower  it  would  not 
have  been  as  evident  in  the  records. 

A  much  more  rapid  process  of  this 
nature  has  recently  been  observed  in  a 
green  alga,  Scenedesmus.  Here  the  respira- 
tory stimulation  shows  only  as  a  brief  dip 
in  the  otherwise  orderly  decay  of  02 
evolution  after  a  light  exposure.  That 
this  irregularity,  frequently  observed  in 
various  laboratories,  is  due  to  the 
increased  02  consumption  by  light  be- 
comes clear  from  the  following  experi- 
ment. 

The  time  course  of  02  exchange  was 
followed  for  a  20-second  exposure  to  650 
m/x  absorbed  mainly  by  chlorophyll  6. 
This  was  done  both  with  and  without 
continuous  background  light  of  a  wave- 
length absorbed  more  by  chlorophyll  a. 
This  background  light  of  690  m/x  speeds 
up  the  rate  of  decay  of  the  lingering  02 
evolution  so  much  that  the  respiratory 
stimulation  becomes  recognizable  as  such 
by  a  short-time  drop  below  the  previous 


a> 


c 
o 

JZ 

o 

X 

CD 


No  background 


With  690  m/x 
background 

A 


Time  ,  m;n 

Fig.  21.  The  rate  of  O2  exchange  from  a  20- 
second  exposure  at  650  niju  in  Scenedesmus.  A 
background  light  of  690  m/x  increases  the  rate  of 
decay  of  the  lingering  O2  evolution  so  much  that 
the  opposing  light-stimulated  O2  uptake  becomes 
evident. 


base  line.  That  is,  an  actual  net  02  uptake 
was  briefly  observed.  This  effect  is 
illustrated  in  figure  21,  which  makes  it 
apparent  that  the  second  maximum  of  the 
normal  decay  curve  is  not  due  to  an 
actual  increase  in  rate  of  02  evolution. 
A  converse  experiment  was  also  per- 


0> 


c 
o 


if. 


No  background 


With  650  m// 
background 


' ' ' #— ' ' 

0  I  2  3       "      0  I  2 

Time,  min 

Fig.  22.  The  decline  in  rate  of  02  evolution 
in  Scenedesmus  after  a  2-minute  exposure  to 
700  m/i.  With  650-m/x  background  the  decline  is 
much  steeper. 


350  CARNEGIE     INSTITUTION      OF      WASHINGTON 

formed,  but  in  a  somewhat  different  way.  consumption  as  measured  by  the  polaro- 

Measurements  were  made  of  the  time  graphic  technique.  We  express  enhance- 

course  of  the  decay  in  rate  of  02  evolution  ment   as   the   ratio   of   the   rate   of   02 

by  Scenedesmus  after  an  exposure  of  2  evolution  when  two  wavelengths  of  light 

minutes  to  700-m/x  light.  These  curves  are  given  simultaneously  to  the  sum  of 

are  shown  in  figure  22.  At  the  start  of  the  rates  from  the  same  two  light  beams 

each  record  the  700-mju  light  was  turned  separately.  This  avoids  any  assumption 

off.  The  sloping  base  line  is  given  by  the  about    the    mechanism    of    the    effect, 

dotted  lines.  Porphyridium  under  optimal  conditions 

The  decay  of  rate  of  02  evolution  after  yields  much  higher  enhancement  ratios 

the  700-m^  exposure  is  greatly  increased  than  green  algae. 

by  the  presence  of  650-mju  background  The  enhancement  ratio  found  is  ex- 
light  as  seen  from  the  slopes  of  these  tremely  sensitive  to  the  physiological 
decay  curves.  In  the  curve  measured  condition  of  the  alga.  Apparently  all  the 
without  background  light  the  small,  growth  conditions  such  as  temperature, 
rapid,  initial  drop  is  attributed  to  stimu-  nutrition,  and  light  must  be  kept  rigidly 
lated  02  uptake  which  is  finished  long  constant  in  order  to  obtain  reproducible 
before  the  02  -exchange  curve  reaches  its  enhancement  values  with  different  cul- 
base  line.  In  the  second  curve  this  is  tures  of  the  same  species, 
obscured  by  the  very  rapid  decay  induced  In  addition,  the  rate  of  photosynthesis 
by  the  650-niju  background.  at  a  single  wavelength  and  intensity  may 

Some  attempts  have  been  made  at  change  as  the  algae  adapt  to  the  condi- 
fitting  similar  data  with  positive  first-  tions  of  measurement.  Once  this  adapta- 
order  decay  curves,  corrected  for  the  tion  period  of  about  4  hours  has  passed, 
stimulated  02  uptake  by  negative  first-  photosynthesis  and  enhancement  shown 
order  curves  with  smaller  time  constants,  by  a  particular  cell  preparation  will 
So  far  no  adequate  fits  have  been  remain  constant  for  a  day  or  two. 
achieved.  Whether  second-order  curves  Whether  the  rate  of  photosynthesis  rises 
for  the  decay  as  predicted  by  one  of  the  or  falls  during  the  adaptation  period 
models  with  first-order  corrections  will  probably  depends  on  the  initial  physio- 
give  better  fits  remains  to  be  seen.  logical  state  of  the  alga  and  the  color  and 

In  any  event,  it  is  clear  that  the  rate  intensity  of  the  light  used  for  the  meas- 

of  decay  of  02  evolution  is  greatly  speeded  urements.   The  enhancement  ratios  are 

up  by  background  light  acting  on  either  considerably  larger  at  low  than  at  high 

pigment  system.  photosynthetic  rates. 

Another  point  of  interest  is  the  ratio  of 
intensities  of  the  two  light  beams  required 

Enhancement  and  Photostimulated  m  orcjer  to  yield  maximum  enhancement. 

Oxygen  Consumption  in  Porphyridium  We  have  confirmed  with  Porphyridium 

T  „  p  the  earlier  work  of  Myers  and  French 

J.  o.  Brown  .  ,     ^7  .                .           ,        i  .    . 

with  Chlorella  that,  to  obtain  maximum 

Some   clarification   of   the   differences  enhancement  ratios,  the  photosynthetic 

between  the  two  primary  photochemical  rate    from    light    absorbed    mostly    by 

reactions    in    photosynthesis    driven   by  accessory  pigment  should  be  two  to  three 

different  pigments  has  come  from  experi-  times  the  rate  from  light  absorbed  by  the 

ments  with  the  red  alga  Porphyridium  long- wavelength     chlorophyll.     Whether 

cruentum  (Year  Book  60,  p.  351).  We  have  the   same   optimum   ratio   of  photosyn- 

continued  these  experiments  to  see  what  thetic  rates  holds  true  for  all  pairs  of 

role,  if  any,  the  physiology  of  the  alga  wavelengths  that  yield  any  enhancement 

plays  in  the  magnitude  of  photosynthesis,  has  not  been  determined. 

enhancement,  and  photostimulation  of  02  Figure  23  shows  two  action  spectra  for 


DEPARTMENT    OF   PLANT    BIOLOGY 


351 


C 


O 

Q. 

O 

<v 
a 


Background 
/"~695  m^ 

\ 

No  background     \ 
light  ->,  \ 


Enhancement  *N 
due  to  695  m/x  f\ 

background 


'       '       1^1 


/' 


\ 


\^*-  Background 

\     550  m/j. 

V 


No  background 
light 

y 


Enhancement 
due  to  550  n\(i 


_i__I i i i i 


550 


600  650 

Wavelength,  m/z 


700 


750 


Fig.  23.  Action  spectra  for  rate  of  photosynthesis  by  Porphyridium  omentum.  The  upper  curves 
were  continuously  recorded  with  a  constant  intensity  of  background  light  of  550  and  695  m/t,  respec- 
tively. Enhancement  was  calculated  by  first  subtracting  the  rate  due  to  the  background  light  alone 
from  the  upper  curves  and  then  determining  the  difference  between  the  two  action  spectra  at  each 
wavelength.  Small  drifts  with  time  of  the  base  lines  were  also  corrected  in  the  calculated  enhancement. 


02  evolution  from  Porphyridium  which 
had  previously  adapted  to  a  steady-state 
rate  of  photosynthesis  on  the  electrode. 
The  solid  curves  give  the  rate  of  photo- 
synthesis as  the  wavelength  was  slowly 
changed  while  the  intensity  in  incident 
quanta  per  second  was  maintained  con- 
stant. The  upper  curves  represent  en- 
hanced photosynthesis,  since  the  algae 
were  receiving  a  constant  intensity  back- 
ground light  of  550  mju  as  the  red  region 
was  being  traversed  and  were  illuminated 
with  background  light  of  695  m^u  for  the 
green  region  of  the  action  spectrum. 
Although  the  difference  between  these 
two  curves,  with  and  without  background 
light,  has  sometimes  been  assumed  to  be 
an  action  spectrum  for  enhancement,  we 
do  not  believe  it  to  be  a  true  representa- 
tion, for  the  following  reason:  the  spectra 
were  measured  with  a  constant  ratio  of 
light  intensity  between  the  pairs  of 
wavelengths,  whereas  we  know  that  there 
is  an  optimum  ratio  between  the  rates  of 
photosynthesis  which  should  be  kept 
constant  when  comparing  effectiveness  of 
different  wavelength  pairs.  It  should  be 
possible  to  determine  the  approximate 
shape  of  the  action  spectrum  for  enhance- 
ment   with    a    constant    photosynthetic 


ratio  from  a  family  of  curves  as  in  figure 
23,  varying  the  intensity  of  the  back- 
ground light  with  each  curve.  From  these 
data  we  could  compare  enhancement 
from  different  wavelength  pairs  but  with 
the  same  ratio  of  the  rates  of  photo- 
synthesis from  each  wavelength  given 
separately. 

The  wavelength-dependent  light  stim- 
ulation of  02  uptake  by  Porphyridium 
was  described  in  last  year's  report.  It  is 
observed  as  a  large  uptake  of  02  when 
the  light  is  turned  off.  After  particular 
wavelengths  and  intensities,  the  apparent 
increased  rate  of  respiration  may  be  as 
large  as  the  previous  photosynthetic  rate 
and  may  have  a  half-life  of  2  to  3  minutes. 
This  photostimulated  02  uptake  is  ini- 
tially proportional  to  the  light  intensity 
but  saturates  at  very  low  intensities.  The 
maximum  amount  of  photostimulated  02 
uptake  has  been  found  to  vary  with  the 
growth  conditions  of  the  alga.  For 
instance,  initially,  greater  photostimu- 
lated 02  consumption  is  shown  by  cells 
cultured  with  cool-white  fluorescent  lights 
than  by  cells  cultured  with  tungsten 
lamps.  It  is  also  greatest  when  the  cells 
are  first  placed  on  the  electrode,  and  then 
it  decreases  gradually  in  time.   By  the 


J.  S.  Brown 


352  CARNEGIE     INSTITUTION     OF      WASHINGTON 

second  day  very  little  remains — a  fortu-  result  from  the  light  activation  of  each  of 

nate  circumstance  which  makes  feasible  the  two  pigment  systems 

the   continuous  recording  of  an  action  We  believe  that  throughout  most  of 

spectrum   for   02   evolution.    Numerous  the  spectrum  the  pigment  ratios  are  such 

action  spectra  for  the  photostimulation  of  that    intermediates    necessary    for    02 

02  uptake  have  invariably  shown  a  broad  production  are  produced  in  near-optimum 

peak  centered  at  685  m/x.  Since  one  of  the  amounts.  But  at  certain  wavelengths — 

forms  of  chlorophyll  a  observed  in  vivo  notably  around  700  mix  with  green  algae 

also  has  an  absorption  maximum  at  685  and  in  the  whole  red  region  with  red 

m/x,  it  is  tempting  to  postulate  that  this  algae — photosynthesis  is  limited  by  too 

form  may  be  responsible  for  the  formation  little  of  the  intermediates  produced  by 

of  easily  oxidizable  material.  the  accessory  pigments  and  therefore  is 

We    have    examined    the    absorption  enhanced  by  adding  light  absorbed  by 

spectra  of  various  Porphyridium  cultures  this  pigment  system, 
to     determine    whether    the    ratio     of 

phycoerythrin  to  chlorophyll  correlates  Physical  Separation  of  Pigment 
with    the    amount    of    enhancement    or  Complexes  from  Euglena 
photostimulated  02  consumption  shown 
by  these  cells.   Our  results  corroborate 
those  of  Marcia  Brody  that  the  ratio  of  Study  of  the  chlorophyll  complexes  of 
phycoerythrin    to    chlorophyll    can    be  aging  Euglena  has  progressed  during  the 
altered  by  changing  the  quality  of  the  past    year.    When    a    dense    culture    of 
light   with   which   the   cells   are   grown.  Euglena  cells  is  allowed  to  age  in  the  dark 
Different  pigment  ratios  may  account  for  cold  room,  the  chlorophyll  a  absorption 
initial  differences  in  enhancement  and  in  peak  shifts  toward  longer  wavelengths, 
photostimulated    02    uptake.    However,  The  shift  is  attributable  to  a  simultaneous 
this  pigment  ratio  does  not  change  during  increase  of  absorption  at  710  mix  and  a 
the  4  to  5  hours'  measurement  period  and  decrease  of  absorption  at  695  m/x. 
therefore  cannot  account  for  the  changes  Last  year  we  reported  that  the  corn- 
occurring  during  this  time.  ponent  absorbing  at  710  m/x  (Pa710)  was 

Derivative  absorption  spectra  of  the  in  the  chloroplast  and  could  be  separated 

red  band  of  chlorophyll  a  in  Porphyridium  by    differential    centrifugation    from    a 

reveal  the  presence  of  the  usual  two  forms  component  absorbing  at  685  m/x  (C0685). 

of  chlorophyll  a  with  maxima  at  about  Now  we  find  that  when  the  cells  are  aged 

670  and  683  m/x,  but  the  ratio  of  these  for  a  longer  period  of  time,  four  and  a 

forms  does  not  seem  to  vary  with  the  half  weeks,  the  brownish-green  Pfl710  is 

different  growth  conditions.  mostly   in   the   cytoplasm   and   can   be 

Although  the  photostimulation  of  02  separated  to  a  large  extent  by  centrifuga- 

consumption  probably  is  induced  via  a  tion  from  the  green  chloroplast  pigments 

specific  pigment,  C„685,  the  magnitude  which  have  absorption  bands  at  670,  685, 

of  the  response  may  depend  on  the  size  and  695  mix.  Pa710  together  with  a  high 

of  the  pool  of  intermediates.  In  turn  this  proportion  of  absorption  which  we  attrib- 

pool  size  may  be  limited  also  by  the  kind  ute  to  Ca670  remains  in  the  supernatant, 

of  alga  being  observed  and  by  its  physio-  These  two  absorption  bands  are  probably 

logical  state.  due  to  pigments  attached  to  the  same 

All    these    observations    support    the  particle,    since    they   sediment   together 

contention  that  enhanced  O2  evolution  is  when  centrifuged  at  very  high  speed, 

not    the    result    of    the    direct    physical  Ether  extraction  and  ascending  paper 

interaction  of  two  pigments  but  rather  chromatography  of  the  cytoplasmic  frac- 

occurs  because  of  the  optimum  produc-  tion  were   carried   out   in   collaboration 

tion  of  intermediates  which  themselves  with  Dr.  James  H.  C.  Smith.  The  results 


DEPARTMENT   OF   PLANT    BIOLOGY 


353 


strongly  indicate  that  the  pigment  ab- 
sorbing at  710  m/z  is  pheophorbide  a  or  a 
closely  related  compound.  The  question 
still  remains  whether  the  organism  first 
forms  pheophytin  a  in  the  chloroplasts 
and  then  excretes  it  into  the  cytoplasm 
as  a  pheophorbide  complex  or  forms  the 
pheophorbide  directly  from  Ca695. 

Electron  Paramagnetic  Resonance 
Studies  on  Chlamydomonas  reinhardi 

Ellen  C.  Weaver 

It  is  possible  to  make  observations  of 
the  free  electrons  in  living  material  by 
means  of  electron  paramagnetic  resonance 
(EPR)  spectroscopy.  The  method  is 
intrinsically  harmless  to  the  system  being 
observed,  and  the  modern  instruments 
can  detect  electron  levels  of  the  order  of 
10 n  for  a  1-gauss  line  width  in  aqueous 
medium.  It  is  a  well  established  fact  that 
chlorophyll-containing  material  has  a 
higher  level  of  unpaired  electrons  when  it 
is  illuminated  than  when  it  is  in  the  dark, 
suggesting  that  at  least  some  phase  of 
photosynthesis  proceeds  by  single-elec- 
tron transfers.  Although  there  has  been 
general  agreement  on  the  gross  observa- 
tions among  several  groups  using  this 
relatively  new  (since  1956)  technique  on 
photosynthetic  material,  there  has  been 
no  convincing  identification  of  the  sub- 
stances responsible  for  the  signal  or  any 
rigorous  demonstration  that  the  electron 
resonance  has  any  direct  connection  with 
photosynthesis. 

The  original  aim  of  the  present  study 
was  to  determine  whether  or  not  the 
resonance  was  associated  with  chloro- 
phyll. It  was  soon  found  that  the  signal 
was  difficult  to  reproduce,  necessitating 
precise  control  of  all  experimental  vari- 
ables. It  was  logical  to  extend  the  study 
to  include  a  determination  of  the  effect  of 
a  few  variations  in  the  cell's  environment 
on  the  electron  resonance  signal.  This  has 
led  us  to  a  tentative  identification  of  one 
of  the  substances  responsible  for  the 
signals  and  to  a  better  understanding  of 
the  behavior  of  the  phenomenon.   The 


work  to  be  described  was  started  in 
Zurich  and  continued  at  the  Department 
of  Plant  Biology  this  past  year.  EPR 
spectroscopy,  already  in  wide  use  by 
chemists,  physicists,  geologists,  and 
others,  seems  certain  to  become  an 
important  tool  in  the  study  of  photo- 
synthesis. 

Materials  and  methods.  The  instrument 
for  these  studies  was  kindly  made 
available  by  Varian  Associates.  The 
experiments  deal  entirely  with  intact, 
living  cells  of  a  green  fresh- water  alga, 
Chlamydomonas  reinhardi.  Cultures  that 
had  lost  the  typical  wild-type  ability  to 
form  chlorophyll  in  the  dark  were 
obtained  from  the  Cambridge  Collection 
of  Algae  and  Protozoa.  Later  work  was 
therefore  done  with  a  different  wild  type 
(21g2)  obtained  from  Dr.  Ruth  Sager, 
but  there  was  no  obvious  difference  in  the 
pigments  or  behavior  of  these  stocks 
when  they  were  grown  in  the  light. 
Mutants  were  derived  from  the  Cam- 
bridge stocks  by  ultraviolet  irradiation, 
only  one  of  which  (no.  100)  is  pertinent 
to  the  present  discussion.  Cultures  were 
grown  in  liquid  shake  culture,  bubbled 
with  5  per  cent  C02  in  air,  and  illuminated 
by  combined  fluorescent  and  incan- 
descent light.  Temperature  was  not 
accurately  controlled  but  was  maintained 
at  approximately  25°C  with  a  water  bath. 
The  medium  contained  only  mineral 
nutrients,  including  trace  amounts  of 
some  seven  transition  ions.  The  only 
nutritional  element  that  was  varied  was 
manganese,  which  was  2  X  10-6  M  unless 
otherwise  specified. 

A  Varian  model  V4502  electron  reso- 
nance spectrometer  equipped  with  100- 
kc/sec  field  modulation  was  used. 
Samples  were  contained  in  a  flattened 
quartz  cuvette,  1  cm  wide,  with  an 
internal  thickness  of  about  0.25  mm.  The 
sensitive  volume  was  approximately  0.01 
ml.  The  sample  was  illuminated  through 
a  slotted  window  in  the  cavity.  Light 
from  a  500- watt  projection  lamp  passed 
through  a  7.5-cm  water  bath  and  was 
focused  on  the  slots  of  the  cavity.  Light 


354 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


intensity  was  regulated  with  a  Variac. 
Nearly  monochromatic  light  was  pro- 
vided by  means  of  interference  filters, 
with  half-widths  varying  from  10  to  13 
TdfjL,  plus  Corning  glass  cutoff  filters  to 
eliminate  the  harmonics.  The  relative 
intensity  of  light  from  all  the  filters  in 
the  optical  system  was  measured  for  a 
series  of  lamp  voltages  with  a  bolometer. 

Ratios  of  chlorophyll  a  to  chlorophyll  b 
were  obtained,  with  the  help  of  Dr. 
Jeanette  S.  Brown,  by  means  of  the 
derivative  recording  spectrophotometer. 
Percentage  light  transmission  of  the 
samples  in  the  cuvette  for  one  of  the 
electron  resonance  experiments  was  meas- 
ured by  Dr.  French. 

Determinations  of  g  values  were  made 
by  Dr.  John  Maling,  of  the  Biophysics 
Laboratory  at  Stanford  University,  by  a 
frequency  counting  method.  The  g  value 
expresses  the  ratio  of  the  magnetic  field 
and  the  microwave  frequency  at  the 
point  where  the  resonance  is  a  maximum. 
Both  field  and  frequency  must  be  deter- 
mined with  great  accuracy,  since  all  free- 
radical  g  values  lie  close  to  the  value  of 
the  free  " conduction"  electron,  which  is 
2.0023.  In  EPR  spectroscopy  the  g  value 
provides  a  measure  of  the  location  for  a 
given  resonance  in  the  magnetic  field 
much  as  does  the  wavelength  of  an 
absorption  maximum  in  optical  spec- 
troscopy. 

Cells  to  be  used  for  resonance  measure- 
ments were  harvested  from  a  4-  to  5-day- 
old  culture.  They  were  centrifuged  in  the 
cold  for  8  minutes  at  3000  rpm,  resus- 
pended  in  an  aqueous  medium  free  of 
manganese,  and  centrifuged  a  second 
time  at  the  same  speed.  They  were  then 
diluted  as  necessary  and  pipetted  into  the 
cuvette,  which  was  placed  in  the  cavity 
of  the  spectrometer  and  adjusted  to  lie 
precisely  at  the  node  of  the  electric  field 
and  at  the  corresponding  maximum  of 
the  magnetic  field.  Since  it  is  intrinsically 
easier  to  vary  the  magnetic  field  than  the 
microwave  frequency,  the  microwave 
frequency  is  held  stationary  and  the  field 
is    swept.    The    resultant    resonance    is 


usually  recorded  as  the  first  derivative  on 
a  chart.  A  Moseley  X-Y  recorder  proved 
to  be  particularly  convenient  for  this 
purpose,  as  successive  signals  can  be 
superimposed.  The  time  constants  of  rise 
and  decay  rates  were  determined  by 
means  of  a  Sanborn  recorder. 

Results  and  discussion.  Two  distinctly 
different  light-induced  resonances  can  be 
observed:  one  seen  only  when  the  cells 
are  illuminated,  and  another  which  per- 
sists in  the  absence  of  illumination.  The 
former  is  centered  at  g  =  2.0025,  is  8.3 
gauss  wide  from  peak  to  peak,  and 
disappears  with  a  half-time  of  a  fraction 
of  a  second  when  the  light  is  turned  off, 
hence  is  termed  the  R  (rapid-decaying) 
signal.  The  other  is  centered  at  g  = 
2.0046,  is  about  20  gauss  wide,  takes  time 
of  the  order  of  hours  to  disappear  in  the 
absence  of  light,  and  so  is  designated  the 
S  (slow-decaying)  signal.  Figure  24  is  an 
example  of  these  two  signals  as  observed 
in  wild-type  Chlamydomonas. 

Our  first  task  was  to  determine  whether 
or  not  the  R  signal  was  associated  with 
chlorophyll,  and  the  evidence  bearing  on 
this  point  follows.  A  mutant  was  isolated 
which  was  clear  yellow  whether  it  was 
grown  in  light  or  dark.  An  absorption 
spectrum  of  the  extracted  pigment  re- 
vealed no  trace  of  chlorophyll.  There  was 
a  slight  resonance  from  a  dense  suspen- 
sion of  these  cells  but  no  observable 
increment  of  the  signal  with  illumination. 
The  two  Cambridge  cultures,  32 A  and 
32B,  formed  no  chlorophyll  when  they 
were  grown  in  the  dark  and  yielded  the 
same  results:  no  detectable  light-induced 
signal  from  the  dark-grown  (yellow) 
material.  In  addition,  it  was  shown  that 
an  aliquot  of  the  same  culture,  when 
allowed  to  green  for  24  hours  in  the  light, 
produced  a  normal  signal. 

If  the  R  signal  is  ascribable  to  chloro- 
phyll, an  action  spectrum  for  its  ampli- 
tude should  correspond  to  other  action 
spectra  for  chlorophyll-mediated  reac- 
tions. Amplitude  refers  solely  to  the 
vertical  peak-to-peak  distance  of  a  given 
trace  and  should  not  be  confused  with 


DEPARTMENT   OF   PLANT   BIOLOGY 


355 


<D 


■o 


Q_ 

E 
o 

"5 

c 

GO 


Light  signal 


Magnetic  field 

Fig.  24.  EPR  signals  from  Chlamydomonas  reinhardi  illuminated  with  red  light,  light  signal,  and 
in  the  absence  of  illumination,  dark  signal.  The  light  R  signal  is  superimposed  on  the  persistent 
dark  S  signal,  whose  hyperfine  structure  is  evident,  owing  to  the  use  of  a  high  gain  and  a  low  modu- 
lation (2.5  gauss  peak  to  peak). 


signal  intensity,  which  is  a  measure  of 
the  number  of  spins  being  observed  and 
can  be  calculated  with  accuracy  only  by 
precisely  determining  (1)  the  base  line, 
(2)  the  width,  and  (3)  the  shape  of  the 
signal.  The  amplitude  is  a  valid  measure 
only  when  the  other  variables  are  known 
to  be  constant. 

Our  action  spectra,  as  well  as  those 
reported  by  other  groups  working  in  this 
field,  originally  seemed  to  show  that  light 
with  an  average  wavelength  longer  than 
that  absorbed  most  strongly  by  chloro- 
phyll a  (about  680  mju)  was  most  effective 
in  producing  an  R  signal.  Various 
explanations  have  been  advanced  to 
account  for  this,  including  one  that 
ascribes  the  EPR  signal  to  the  ''active 


centers"  of  chlorophyll  that  absorb  most 
strongly  at  700  m/x,  like  that  of  Beinert, 
Hoch,  and  Kok.  However,  we  suspected 
it  was  simply  a  self-absorption  phenome- 
non, and  later  we  were  able  to  show  that, 
if  the  cell  suspension  was  diluted  suffi- 
ciently, 680-m^  light  is  indeed  more 
effective  than  longer  wavelengths  in 
producing  the  narrow  R  signal.  The 
degree  to  which  material  being  observed 
can  be  diluted  is  limited  by  the  signal-to- 
noise  ratio  of  the  resonance.  The  reduced 
precision  was  compensated  for  by  making 
traces  of  the  signal  at  each  wavelength; 
by  using  several  light  levels,  all  well  below 
saturation,  for  each  wavelength;  by 
randomizing  the  order  in  which  measure- 
ments were  made;  by  making  frequent 


356 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


E 

o 


CO 


— i r 

680  mfi 


FT 


/ 


X 


/ 


0) 

3       4 


Q. 

£      3 
o 

D       2 

c 

C7> 


S  signal 


550  m/z 


S  signal 


\  =  700 


+ 


/ 


/ 


/ 

.+ 


/ 


S  signal  — 


0  .1  .2         .3        A         .5         .6        .7 

Light  intensity, rel 

Fig.  25.  Signal  amplitude  (vertical  peak-to- 
peak  distance)  plotted  as  a  function  of  light 
intensity  for  three  different  wavelengths.  The  S 
signal  is  constant,  with  an  amplitude  of  1.1  cm, 
and  forms  a  base  line  for  the  R  signal.  The 
nonlinearity  of  the  curve  for  680  mpt  indicates 
appreciable  internal  absorption. 


checks  of  a  standard  filter  and  voltage 
combination  to  monitor  the  cells'  re- 
sponse (which  was  normally  reproducible 
over  a  period  of  several  hours);  and  by 
making  frequent  checks  of  the  signal  in 
the  absence  of  light. 

With  these  data,  it  is  possible  to  plot  a 
light  intensity  versus  signal  amplitude 
curve  for  each  wavelength  (figure  25)  and 
from  this  to  plot  an  action  spectrum  for 


signal  amplitude  at  a  given  light  intensity 
for  each  wavelength.  Figure  26  is  an 
action  spectrum  from  a  cell  suspension 
which  transmitted  about  half  of  the 
incident  light  at  680  m/x  in  the  cuvette 
used  for  EPR  measurements.  The  maxi- 
mum of  the  peak  is  far  flatter  than  the 
absorption  spectrum  for  chlorophyll,  but 
in  view  of  earlier  findings  it  might  be 
postulated  that  self-absorption  is  still 
playing  a  large  part  in  the  curve  shape 
and  that  the  peak  would  be  sharpened  by 
further  dilution.  The  cell  concentration 
for  figures  25  and  26  was  about  108/ml. 
Its  percentage  light  transmission  in  the 
cuvette  used  for  EPR  measurements  was 
as  follows: 


/elength 

Transmission,  % 

650 

46 

680 

45 

694 

58 

730 

90 

°.    3 

•o    2 

13 


Q. 

E  o 
o 

_  5 

2  4 

CO  3 


n| ( f 1 ! f r 


S  signa 


S  signal-^ 


j '      '      ' '      ' 


650  700 

Wavelength,  m// 

Fig.  26.  (a)  Action  spectrum  for  production 
of  EPR  signal,  derived  from  curves  of  the  type 
illustrated  in  figure  25.  Each  of  the  three  upper 
spectra  represents  the  result  for  a  different  light 
intensity.  These  are  flattened  by  the  curvature 
of  the  plots  shown  in  figure  25.  (b)  Action 
spectrum  for  production  of  the  EPR  signals 
obtained  by  back  extrapolation  of  the  same  data 
to  a  very  low  intensity. 


DEPARTMENT   OF   PLANT   BIOLOGY  357 

The    nonlinearity    of    light    intensity  actual   substance   from   which   the   free 

versus  signal  amplitude  curves  of  figure  25  electrons  are  derived  is  chlorophyll  a  or 

for  strongly  absorbed  wavelengths  indi-  some  other  material  is  not  decided  by  the 

cates  that  self -absorption  is  playing  an  present  experiments, 

appreciable  role.  The  extent  of  this  effect  The    tentative    conclusion    we    have 

can  be  roughly  estimated  by  extrapo-  drawn  from  the  foregoing  results  is  that 

lating  the  steepest  part  of  the  curve  and  the  R  signal  is  ascribable  to  chlorophyll 

plotting  the  points  thus  obtained.  This  and  that  it  arises  as  a  consequence  of  the 

has    been    done    for    figure    266,    with  "primary"   act  of  photosynthesis.   This 

resultant    sharpening    of    the    peak    at  hypothesis  is  greatly  strengthened  by  the 

680  m,u.  finding    in    other    laboratories,    notably 

This  estimation  is  strengthened  by  the  Calvin's,    that    the    R    signal    can    be 

shapes   of  the  signal  amplitude  versus  produced    at    very    low    temperatures 

light  intensity  curves  mentioned  above.  (  — 160°C)  with  no  apparent  increase  in 

The    curves    for    wavelengths    strongly  rise  time  of  the  resonance,  indicating  a 

absorbed  by  chlorophyll  have  no  linear  purely  physical  rather  than  a  chemical 

portion,  whereas  those  for  other  wave-  process.  The  R  signal  (or  one  that  seems 

lengths  do  exhibit  strict  proportionality  to  be  identical  with  it)  has  been  reported 

at  low  light  levels,  as  shown  in  figure  25.  not  only  in  other  algae  and  in  chloro- 

There  is  also  a  corresponding  variation  in  plasts  of  higher  plants  but  also  in  meth- 

steepness  of  the  curves.  anol  extracts  of  chlorophyll,  dried  chloro- 

Another    experiment,    by    the    same  phyll  films,  and  single  crystals, 

experimental   techniques   and   with   the  The  amplitude  of  the  R  signal  can  be 

same   precautions   as   above,    was    per-  widely  varied.   Continuous  or  repeated 

formed     to     determine     the    maximum  illumination  may  bring  about  a  reduction 

obtainable,  or  saturation,  signal  from  each  in  /^-signal  amplitude,  but  eventually  a 

wavelength.     It    was    found    that    the  steady  state  is  reached  in  which  signals 

maximum    signal    amplitude    was    inde-  are  reproducible  over  a  period  of  several 

pendent  of  wavelength,  provided  that  the  hours.  Freshly  spun  cells,  which  have  been 

quantum  energy  was  sufficient  to  excite  grown  under  optimal  conditions  and  then 

an    electron.    Far-red    light,    745    m/x,  suspended  in  a  buffered  medium,  exhibit 

excites  very  nearly  a  maximum  signal;  a  rather  small  signal  even  at  high  white 

760-m/x  light,  even  at  rather  high  intensi-  light    intensities    when    they    are    in    a 

ties,   produced   no   detectable    R   signal  physiologically  healthy  state.  Presumably 

(showing,  incidentally,  that  the  effect  at  electrons  are  flowing  smoothly  in  the  well 

745  m/z  is  not  a  spurious  one  due  to  stray  functioning  system,  and  the  steady-state 

light).    There   was   no   obvious   damage  level  is  rather  low.  If  the  cuvette  is  stored 

from  760-mju  light,  since  the  response  to  for  an  hour  or  two  in  the  dark  and  again 

shorter  wavelengths   was   just   as   good  observed,  it  is  found  that  both  signals 

after  exposure  as  before.   The  shortest  have    disappeared.    Under   illumination, 

wavelength  employed  was  470  mju.  Since  however,   the   R  signal  is  induced  and 

a  dilute  suspension  saturates  with  a  lower  shows    a    very    large    amplitude,    and 

light  level  than  a  concentrated  one,  and  subsequent  tracing  of  the  S  signal  reveals 

the    amount    of    light    obtainable    was  it  to  be  similar  to  that  originally  observed, 

limited,  fairly  dilute  ones  were  used  for  This  "starvation  effect"  is  illustrated  in 

this  work.  The  efficiency  of  signal  produc-  figure   27.   Which   of  several  factors   is 

tion   is   very   low   for   wavelengths   not  responsible  for  it  is  not  known,  but  it  is 

strongly  absorbed   by  chlorophyll.   The  more  pronounced  when  cells  are  grown  in 

action  spectrum  shows  that  light  absorp-  medium  deficient  in  manganese  or  are 

tion  by  chlorophyll  a  leads  to  the  pro-  suspended  in  distilled  water  rather  than 

duction  of  the   R  signal.   Whether  the  buffer.  A  similar  enhancement  of  ^-signal 


358 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


amplitude  can  be  demonstrated  by  treat- 
ing the  system  with  10~5  M  DCMU 
(3-[3,4-dichlorophenyl]-l ,  1-dimethyl 
urea),  shown  in  figure  28.  This  compound 
acts  very  specifically  to  block  the  oxygen- 
evolving  mechanism,  with  no  other 
obvious  toxic  effects.  Action  spectra  made 
with  and  without  DCMU  have  the  same 
maxima  and  minima,  but  the  large  signal 
amplitude  produced  by  its  use  permits  a 
greater  accuracy  of  measurement. 

Another  way  to  block  the  oxygen- 
evolving  mechanism  of  the  cell  is  to  limit 
manganese-ion  concentration  in  the  grow- 
ing medium,  since  many  studies  have 
shown  manganese  to  be  an  essential 
component  of  this  system.  Cells  grown  in 
medium  having  no  manganese  added  (but 
in  which  no  effort  was  made  to  exclude  it, 
so  traces  were  certainly  present)  grew 
slowly,    were    clumped,    but    produced 


ultimately  a  culture  in  which  the  propor- 
tions of  chlorophyll  a  and  b  were  similar 
to  those  in  cells  grown  in  normal  medium, 
with  perhaps  a  trace  more  chlorophyll  b. 
The  rate  of  oxygen  production  from  these 
manganese-deficient  cells  was  greatly 
reduced  in  comparison  with  cells  grown  in 
medium  containing  as  little  as  10~6  M 
Mn++.  However,  they  consistently  gave 
rise  to  an  enhanced  R  signal,  in  contrast 
to  reports  from  at  least  two  other 
laboratories. 

It  seems  reasonable,  then,  to  assume 
that,  when  the  pathway  of  the  electrons 
is  in  any  way  impeded,  the  net  level  of 
unpaired  spins  rises.  Thus  far  we  have 
only  disturbed  the  mechanism  for  oxygen 
evolution,  but  it  is  hoped  that  eventually 
other  pathways  can  be  altered  in  specific 
and  known  ways  and  the  effect  on  the 
electron    resonance    spectrum    can    be 


<D 


E 

D 

~5 
C 
C7> 

CO 


Starvation  effect 


(l 


Same  sample 
after  li  hours 

2 

in  the  dark 


Magnetic  field 

Fig.  27.     The  increase  of  the  R  signal  induced  by  light  after  1}4  hours'  previous  storage  in  the  dark. 


DEPARTMENT    OF   PLANT   BIOLOGY 


359 


CD 


■o 

"5. 

e 

o 

"o 

c 
a> 

C/) 


Sample  in  water 


Sample  in  IO"5M  DCMU 


**^\x 


Magnetic  field 

Fig.  28.  Enhancement  of  R  signal  by  10~5  M  DCMU  (3-[3,4-dichlorophenyl]-l,l-dimethyl  urea). 
Two  aliquots  of  the  same  culture  were  used.  Upper  curves:  S  signal  following  a  light  exposure.  Lower 
curves:  R  -\-  S  signals  during  illumination. 


observed.  Obviously,  the  value  of  such 
studies  will  be  greatly  increased  by 
concurrent  biochemical  and  physiological 
studies. 

Thus  far  we  have  been  concerned 
mainly  with  the  R  signal,  which  is 
probably  produced  by  chlorophyll.  The 
*S  signal,  however,  is  of  great  interest 
because  it  shows  that  the  alga  has  some 
mechanism  that  maintains  a  free  radical 
for  an  hour  or  more  in  the  dark.  More- 
over, the  signal  is  structured,  as  can  be 
seen  in  figure  24,  with  six  hyperfine  peaks 
spaced  about  5  gauss  apart.  This, 
together  with  the  higher  g  value  (2.0046), 
indicates  an  organic  free  radical,  possibly 
a  quinone.  Other  EPR  observations 
include  the  fact  that  the  S  signal  can  be 
induced  by  both  red  and  green  light  and 
is  saturated  by  light  levels  an  order  of 
magnitude  lower  than  those  required  to 
saturate  the  R  signal,  as  shown  in  figure 
29.  Even  stray  light  in  a  semidarkened 


room  can  suffice.  For  these  reasons,  no 
action  spectrum  has  been  made  of  the  S 
signal. 

One  set  of  moment  calculations,  the 
double  integral  of  the  recorded  traces, 
made  on  some  twenty  traces  with  and 
after  light  of  various  wavelengths  and  at 
various  intensities  with  a  fairly  dilute 
suspension,  showed  that  there  were 
roughly  ten  times  as  many  unpaired  spins 
in  the  S  signal  as  in  the  R  signal  alone. 
The  S  signal  is  not  altered  in  appearance 
by  "starvation"  or  DCMU.  However,  it 
was  found  that  the  manganese-deficient 
cultures  had  practically  no  S  signal,  as 
seen  in  figure  30.  This  is  interesting  in 
view  of  the  fact  that  it  has  been  reported 
that  the  purple  bacteria,  which  do  not 
evolve  oxygen,  are  also  lacking  the 
typical  broad,  persistent  resonance,  which 
has  been  reported  in  a  number  of  different 
aerobic  organisms.  The  fact  that  DCMU 
does   not  abolish   it   indicates   that   the 


360 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


a> 
■o 

3 


a. 
E 
o 

o 

c 


Magnetic  field 

Fig.  29.  The  induction  of  the  S  signal  by  light  and  its  decay,  (a)  S  signal  from  freshly  prepared 
cells,  (b)  Same  preparation  after  4  hours  in  total  darkness,  (c)  Same  as  (6),  but  traced  immediately 
after  30-second  exposure  to  low  level  of  illumination  at  530  m/x.  (d)  After  further  exposure  for  30 
seconds  to  694  rn.fi  at  a  higher  level  of  illumination.  (Traces  b,  c,  and  d  were  made  with  a  faster 
scanning  rate  than  a,  and  so  the  signal  appears  narrower.) 


resonance  is  not  due  simply  to  the 
process  of  oxygen  evolution  but  to  some 
substance  vital  to  it.  Experiments  started 
in  this  laboratory  some  years  ago  have 
led  Dr.  Norman  Bishop  to  the  identifi- 
cation of  a  benzoquinone,  Q-255,  or 
plastoquinone,  located  in  chloroplasts, 
which  is  a  necessary  and  apparently 
universal  factor  in  the  oxygen-evolving 
mechanism  of  green  plants.  It  seems  very 
possible,  but  has  not  yet  been  demon- 
strated, that  the  reason  plants  deprived 
of  manganese  do  not  evolve  oxygen  is 
that  they  also  lack  plastoquinone. 

Samples  of  purified  crystalline  plasto- 
quinone have  been  obtained  through  the 
kindness  of  Dr.  F.  L.  Crane  of  Purdue, 


Dr.  Karl  Folkers  of  Merck,  and  Dr. 
Bishop.  Dr.  Maling  has  been  able  to 
obtain  a  well  resolved  spectrum  of  the 
purified  compound  (fig.  31)  as  it  is 
oxidized  from  the  semiquinone  to  the 
quinone.  The  g  value  is  2.0044  dr  0.0001, 
which  comes  close  to  that  calculated  for 
the  S  signal.  The  spacing  of  the  hyperfine 
lines  is  2. 1  gauss  in  the  purified  compound, 
less  than  half  that  observed  for  the  S 
signal.  Whether  the  binding  of  this 
compound  and  its  close  association  with 
the  chlorophyll-lipo-protein  complex 
could  result  in  broadening  by  a  factor  of 
2  is  unknown.  This  discrepancy,  there- 
fore, should  not  be  considered  proof  that 
plastoquinone  could  not  be  responsible 


DEPARTMENT   OF    PLANT    BIOLOGY 


361 


for  signal  S.  Plastoquinone  remains  a 
prime  candidate  on  all  other  grounds,  and 
it  is  hoped  that  more  solid  evidence  can 
be  offered. 

A  number  of  preliminary  observations, 
not  yet  adequately  confirmed,  may  be  of 
some  interest.  Figure  32  presents  three 
traces  made  with  Porphyridium  omentum, 
a  red  alga  studied  in  this  laboratory  ( Year 
Book  60,  pp.  351-362)  with  special 
reference  to  a  two-wavelength  model  of 
photosynthesis.    A    typical    light    signal 


results  from  illumination  with  680  m/z. 
However,  when  the  suspension  is  illumi- 
nated with  light  of  equal  energy,  but  at 
a  peak  wavelength  of  567  m/i,  absorbed 
by  phycoerythrin,  there  is  no  effect.  This 
result  supports  the  conclusion  reached 
from  the  Chlamydomonas  work  that  the 
primary  resonance  is  due  to  chlorophyll, 
even  though,  in  this  case,  phycoerythrin 
is  more  effective  in  photosynthesis.  This 
is  a  most  intriguing  subject  for  further 
study    with    special    reference    to    EPR 


o> 


Cl 

E 
o 

"o 

c 

E? 

CO 


Magnetic  field 

Fig.  30.  The  effect  of  manganese-deficient  medium  on  the  S  signal.  The  top  trace  reveals  a  very 
slight  resonance  in  the  dark.  Successive  exposures  to  equal  but  saturating  intensities  of  two  wave- 
lengths of  light  produced  the  second  and  third  traces.  The  bottom  trace,  made  immediately  thereafter 
in  the  dark,  shows  almost  no  resonance.  Instrumental  conditions  provided  high  sensitivity  and  are 
identical  to  those  used  for  figures  27  through  29.  A  modulation  of  5  gauss,  peak  to  peak,  was  used. 
The  dense  cell  concentrations  in  these  four  figures  are  all  about  3  X  108/ml. 


362 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


<1> 

a> 

3 

"a. 

e 

o 

"5 

c 

in 

CH3>* 

0 

0 

0 

CH3          \  /    \    / 
"(CH2CH  =  C-CH2)9H         ^ 

Magnetic  field 

Fig.  31.     The  structure  and  EPR  spectrum  of  plastoquinone.  The  ten  hyperfine  peaks  are  2.1 
gauss  apart;  the  g  value  is  2.0044  ±  0.0001.  (Courtesy  of  Dr.  J.  E.  Maling.) 


a> 


CD 


Q. 

E 
o 

D 

C 
C7> 

CO 


Magnetic  field 

Fig.  32.  EPR  signals  from  Porphyridium  cruentum.  Trace  (a)  was  made  after  the  cells  had  been 
kept  some  hours  in  the  dark.  Trace  (b)  was  made  with  red  light,  680  ran,  and  trace  (c)  was  made  a 
few  minutes  later  with  light  of  the  same  intensity  but  with  a  wavelength  of  567  m^. 


DEPARTMENT    OF   PLANT    BIOLOGY 


363 


experiments  done  with  two  wavelengths 
of  light. 

There  have  been  many  experiments  on 
rise  and  decay  times,  but  they  have 
proved  to  be  so  variable  that  it  is  not 
possible  to  draw  consistent  conclusions. 
Records  are  obtained  by  setting  the 
magnetic  sweep  on  the  point  of  maximum 
deflection,  then  turning  the  light  on  and 
off.  It  seems  fairly  safe  to  say,  however, 
that  both  rise  and  decay  times  have  two 
time  constants:  a  fast  response,  followed 
by  a  slower  one.  The  exact  values  vary 
with  the  number  of  spins  observed,  i.e., 
with  density  of  cell  suspension  and  with 
intensity  and  wavelength  of  the  exciting 
light — the  greater  the  number  of  spins, 
the    faster    the    response.    In    DCMU- 


treated  cultures,  a  minute  or  more  was 
needed  to  attain  maximum  R  signal 
amplitude,  a  correspondingly  long  decay 
time  also  being  typical.  Since  the  shape 
of  the  curves  has  not  been  shown  to  be 
exponential,  the  term  t\/2  cannot,  strictly 
speaking,  be  used.  By  way  of  giving  an 
order  of  magnitude,  however,  the  time 
needed  for  an  average  signal  to  reach 
one-half  of  its  maximum  amplitude  in 
untreated  preparations  is  about  0.2 
second.  Figure  33  illustrates  typical  rise 
and  decay  curves.  Properly  controlled 
studies  of  the  shapes  and  time  constants 
of  the  build-up  and  disappearance  of  both 
R  and  S  signals  should  yield  valuable 
information  on  the  classes  of  events 
involved  in  their  formation. 


a> 


3 


Q. 

E 

D 

"5 

C 

EP 

CO 


24  36 

Time,  sec 


48 


60 


Fig.  33.  (a)  Rise  and  decay  of  R  signal  from  cells  suspended  in  water.  Time  required  for  signal 
to  reach  half  its  maximum  amplitude  is  approximately  0.2  second.  Decay  takes  place  as  fast  as  the 
recorder  can  move.  (6)  Rise  and  decay  of  R  signal  from  cells  suspended  in  10~5  M  DCMU.  Times  are 
2.2  seconds  for  half  of  the  maximum  rise  (not  shown)  and  2  seconds  for  half  of  the  decay. 


364 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


Rubber 
sea 


t 


Plexiglas  coupling 


J         L_U 


l  i 


T^TTTin 


Exploded  side  view 


Front  view 


Fig.  34.     A  Plexiglas  centrifugal  liquid  pump  giving  electrical  isolation,  a  constant  flow  rate,  and 
a  small  internal  volume. 


DEPARTMENT   OF   PLANT   BIOLOGY  365 

We  have  no  direct  evidence  on  the  solutions  over  the  oxygen  electrode  for 

relationship    between    the    two    signals,  photosynthesis  measurements,  it  seems  to 

Commoner  and  co-workers  suggested  in  have  many  potential  uses. 

1957  that  electrons  from  the  R  signal  The  inert  transparent  Plexiglas  allows 

proceeded  to  the  S  signal,  from  there  to  contamination  to  be  detected  and  permits 

disappear  so  slowly  that  their  passage  continuous  observation  of  the  pumping 

could  not  be  traced.  We  have  not  found  action.  An  1850-rpm  synchronous  motor 

any  evidence  to  contradict  this  hypothe-  gives  quiet,  vibration-free  operation  with 

sis,  and  several  other  studies  of  chloro-  a  constant  flow  rate, 

phyll  (or  chlorophyll-like)  and  quinone  The  construction  is  shown  in  figure  34. 

systems  have  demonstrated  that  single-  A  centrifugal  impellor  with  a  low  hold-up 

electron  transfers  do  take  place  between  volume   was   machined   from    Plexiglas. 

them.  We  have  presented  evidence  from  Four  holes  spaced  90°  apart  were  drilled 

an   intact    photosynthetic    organism    to  from  the  circumference  to  the  center  axis, 

support  the  idea  that  chlorophyll  is  the  intersecting  an  inlet  hole  drilled  along  the 

source   of   one   free   electron,   and   that  axis.  The  impellor  shaft  of  stainless  steel 

plastoquinone  is  the  site  of  another;  we  is   threaded   into   a   stainless-steel   disk 

have   been    able    to    demonstrate    some  mounted    on   the   impellor.    A   long-life 

correlation  of  the  behavior  of  these  two  impellor  seal  of  frictionless  Hycar  was 

signals    with    evolution    of    oxygen    by  obtained    from    Schaar    and    Company, 

photosynthesis.  It  is  hoped  that  further  7300  West  Montrose  Avenue,  Chicago  34, 

work  on  these  and  other  materials  will  Illinois  (model  AR4360). 

yield  results  that  will  help  solve  some  of  Several  modifications  in  size  and  pro- 

the  unknowns  of  photosynthesis.  portions  of  the  basic  design  have  been 

made  with  no  apparent  loss  of  versatility 

or  performance.  Figure  34  does  not  show 

An  Electrically  Isolated  the  method  of  mounting  motor  and  pump 

Transparent  Liquid  Pump  as  a  unit  because  of  the  varied  motor 

designs  available.    The   rotation  of  the 
impellor  must  be  counterclockwise  facing 

The  need  for  a  thermally  and  elec-  the  motor  shaft  so  that  the  motive  force 

trically    isolated    pump,    adaptable    to  acts  to  tighten  the  screw  supporting  the 

diverse  applications  for  pumping  liquids,  impellor    shaft.    Electrical    isolation    is 

led  to  the  design  and  fabrication  of  one  assured  by  a  Plexiglas  coupling  on  the 

constructed  mainly  of  Plexiglas.  Although  motor   shaft.    The   performance   of   one 

it   was    built    primarily   for    circulating  model  is  shown  in  table  10. 

TABLE  10.     Performance  of  a  Pump  Having  a  2J^-Inch-Diameter  Impellor  with  34-Inch  Holes 

Head,  in.                         6                12                18  24                30                36                42 

Flow  rate,  liters /min      6.60            6.10            6.05  5.75            5.40            5.00            4.60 


SPEECHES 

Brown,  J.  S.,  Forms  of  chlorophyll  a  and  photo-  Fork,  D.  C,  Evidence  for  a  2-pigment  mecha- 

synthesis,  AAAS,  Denver,  Colorado,  December  nism     in     Prophyridium     cruentum,     AAAS, 

1961.  Lafayette,  Indiana,  August  1961. 

Brown,  J.  S.,  A  study  of  the  induction  period  in  Fork,  D.  C,  Etude  de  remission  d'oxygene  chez 

photosynthesis  with  different  wavelengths  of  les  chloroplastes  de  Beta  vulgaris,  Laboratoire 

light,  Biophysical  Society,  Washington,  D.  C,  de  Photosynthese  du  CNRS,  Gif-sur-Yvette, 

February  1962.  France,  March  1962. 


R.  W.  Hart 


366 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


French,  C.  S.,  Computer  solutions  for  a  2-pig- 
ment  mechanism  of  photosynthesis  showing 
time  courses,  chromatic  transients,  and 
enhancement,  International  Biophysics  Con- 
gress, Stockholm,  Sweden,  July  1961. 

French,  C.  S.,  A  2-pigment  mechanism  of  photo- 
synthesis, Sixth  Scandinavian  Congress  for 
Plant  Physiology,  Lund,  Sweden,  August  1961  ; 
Botanisches  Institut,  Giessen,  Germany, 
August  1961;  Laboratoire  de  Photosynthese 
du  CNRS,  Gif-sur-Yvette,  France,  August 
1961. 

French,  C.  S.,  Plant  pigments  and  their  func- 
tions in  photosynthesis,  three  lectures, 
Harvard  University,  Cambridge,  Massachu- 
setts, October-November  1961. 

French,  C.  S.,  Photosynthesis  and  respiratory 
stimulation  in  Porphyridium,  Harvard-Bran- 
deis  Joint  Photosynthesis  Seminar,  Cam- 
bridge, Massachusetts,  November  1961. 

French,  C.  S.,  and  D.  C.  Fork,  Two  primary 
photochemical  reactions  in  photosynthesis 
driven  by  different  pigments,  Fifth  Interna- 
tional Congress  of  Biochemistry,  Moscow, 
USSR,  August  1961. 


Milner,  H.  W.,  and  W.  M.  Hiesey,  Photosyn- 
thetic  rate  and  light  saturation  from  0°  to  50° 
in  altitudinal  races  of  Mimulus,  American 
Society  of  Plant  Physiologists,  Lafayette, 
Indiana,  August  1961. 

Milner,  H.  W.,  and  W.  M.  Hiesey,  Temperature 
effects  on  photosynthetic  rates  of  climatic 
races  of  Mimulus,  Western  Society  of  Natural- 
ists, Eugene,  Oregon,  December  1961. 

Smith,  J.  H.  C,  The  photochemistry  of  photo- 
chlorophyll  holochrome,  Photosynthesis  Sem- 
inar, Berkeley,  California,  November  1961. 

Smith,  J.  H.  C,  and  J.  Coomber,  Particle  size  of 
the  protochlorophyll  holochrome,  Interna- 
tional Biophysics  Congress,  Stockholm, 
Sweden,  July  1961. 

Smith,  J.  H.  C,  Chlorophyll  formation  and 
photosynthesis,  Fifth  International  Congress 
of  Biochemistry,  Moscow,  USSR,  August 
1961. 

Weaver,  E.  C,  and  H.  E.  Weaver,  Electron 
resonance  signals  in  Chlamydomonas  reinhardi, 
Biophysical  Society,  Washington,  D.  C, 
February  1962. 


BIBLIOGRAPHY 


Brown,  J.  S.,  and  C.  S.  French,  The  long-wave- 
length forms  of  chlorophyll  a,  Biophys.  J.,  1, 
539-550,  1961. 

Brown,  J.  S.,  A  study  of  the  induction  period  of 
photosynthesis  with  different  wavelengths  of 
light,  Abstracts,  Biophys.  Soc.  Sixth  Annual 
Meeting,  FC  13,  1962. 

Clausen,  Jens,  Stages  in  Evolution  of  Plant 
Species,  second  printing,  Hafner  Publishing 
Company,  New  York,  pp.  vii  +  206,  1962. 

Frei,  Yael  F.,  The  derivative  absorption  spectra 
of  chlorophyll  in  algae  and  leaves  at  low 
temperatures,  Biochim.  Biophys.  Acta,  57, 
82-S7,  1962. 

French,  C.  S.,  Computer  solutions  for  a  two- 
pigment  mechanism  of  photosynthesis  show- 
ing time  courses,  chromatic  transients,  and 
enhancement,  International  Biophysics  Con- 
gress, Stockholm,  1961,  Abstracts  of  Contrib- 
uted Papers,  pp.  17-18. 

French,  C.  S.,  and  D.  C.  Fork,  Computer  solu- 
tions for  photosynthesis  rates  from  a  two- 
pigment  model,  Biophys.  J.,  1,  669-681,  1961. 

French,  C.  S.,  and  D.  C.  Fork,  Two  primary 
photochemical  reactions  in  photosynthesis 
driven    by    different    pigments,    Fifth    Inter- 


national Congress  of  Biochemistry,  Moscow, 
1961,  Symposium,  VI,  Preprint  78,  pp.  1-15. 

Jorgensen,  Erik  G.,  and  E.  Steemann-Nielsen, 
Effect  of  filtrates  from  cultures  of  unicellular 
algae  on  the  growth  of  Staphylococcus  aureus, 
Physiol.  Plantarum,  14,  896-908,  1961. 

Nobs,  Malcolm  A.,  Gradual  speciation  and  gen- 
ecology,  in  Recent  Advances  in  Botany,  Univer- 
sity of  Toronto  Press,  pp.  849-853,  1961. 

Smith,  James  H.  C,  Chlorophyll  formation  and 
photosynthesis,  Fifth  International  Congress 
of  Biochemistry,  Moscow,  1961,  Symposium, 
VI,  Preprint  110,  pp.  1-12. 

Smith,  James  H.  C,  Some  physical  and  chemical 
properties  of  the  protochlorophyll  holochrome, 
Biological  Structure  and  Function,  Proc.  First 
IUB/IUBS  Intern.  Symposium,  II,  325-338, 
1962. 

Smith,  James  H.  C,  and  J.  Coomber,  Particle 
size  of  the  protochlorophyll  holochrome, 
International  Biophysics  Congress,  Stockholm, 
1961,  Abstracts  of  Contributed  Papers,  pp. 
10-21. 

Watson,  Patricia  J.,  and  Jens  Clausen,  Pheno- 
typic  responses  to  contrasting  environments 
in  the  genus  Poa,  Scottish  Plant  Breeding 
Station  RepL,  1961,  64-78. 


Department  of  Embryology 


Baltimore,  Maryland 

James  D.  Ebert 
Director 


Contents 


Introduction 369 

Cellular  Regulatory  Mechanisms 372 

Development  of  machinery  for  protein  synthesis 372 

Cytoplasmic  ribosomes 373 

Identification  of  ferritin  in  the  egg 374 

Correlation  of  death  from  magnesium  starvation  and  appearance  of  cytoplasmic 

ribosomes  in  development 375 

Binding  ribosomes 377 

High-molecular-weight  RNA  in  development 380 

Size  and  localization  of  synthesized  ribosomes  after  stage  23 382 

Assay  of  vertebrate  ribonucleic  acid  for  a  ribonucleic  acid  fraction  specifying 

polypeptide  sequence 384 

Regulation  of  antibody  formation 384 

Acquired  tolerance  to  skin  homografts  in  mice 384 

Immunologically  induced  aspermatogenesis 390 

Nature  of  the  material  inducing  aspermatogenesis  in  the  guinea  pig       ....  392 

Antibodies  as  tools 393 

An  approach  to  an  immunochemical  study  of  neuronal  differentiation   ....  393 

A  new  antigenic  system  in  the  chick  embryo 394 

Composition  of  the  cell  surface 396 

Cell  Interaction  in  Differentiation  and  Morphogenesis 397 

Myogenesis  in  vitro 397 

Regional  localization  of  pre-pacemaker  cells  in  the  chick  embryo 400 

The  cultivation  in  fluid  medium  of  the  labile  chorioallantoic  membrane    ....  404 

Inductive  Tissue  Interactions 406 

Interactions  between  dermis  and  epidermis  from  prospective  feather  and  scale 

regions  after  recombination  on  the  chorioallantoic  membrane 406 

Lens  induction 407 

Humoral  Regulatory  Mechanisms 408 

Growth  promotion  during  regeneration  and  the  control  of  growth 408 

Organ  homogenates  and  growth  promotion 409 

Reproduction  and  sex  differentiation  in  the  opossum 415 

Biochemical  Changes  during  Metamorphosis 416 

Acid  deoxyribonuclease  in  amphibian  metamorphosis 416 

An  amylase  activity  in  Rana  pipiens  serum 417 

The  Embryo  in  Relation  to  Its  Environment 419 

Mechanisms  of  implantation  of  the  ovum 419 

Anatomy  and  physiology  of  the  placenta 423 

Study  of  human  placental  vasculature 425 

A  composite  drawing  of  the  placenta  to  show  its  structure  and  circulation       .      .  426 

Differentiation  and  Morphogenesis  in  the  Human  Embryo 426 

The  collection  of  hyman  embryos 426 

Development  of  the  eye 427 

Early  development  of  the  brain 427 

Staff  Activities 428 

Bibliography 430 

Personnel 432 

Carnegie  Institution  of  Washington  Year  Book  61,  1961-1962 


Frontispiece 


Department  of  Embryology 


T3 

a 

i— H 


o3 
JO 

'o 


o3 
0) 


o 

o3 
Fh 
O 


INTRODUCTION 

The    year    1916    saw    Frank   R.   Lillie,  the  Department  into  one  directed  solely 

then  in  his  forty-sixth  year,  at  the  peak  toward  some  major  problem,  emphasizing 

of  his  career  as  investigator,  teacher,  and  either  one  currently  fashionable  experi- 

science  administrator  par  excellence.  Al-  mental  system,  say  antibody  format  on  or 

though   he   was   known   earlier   for   his  myogenesis,  or  one  of  the  areas  largely 

studies    on    the    physiology    of    cellular  neglected  of  late,  like  neurogenesis;  but 

differentiation  and  cell  lineage  in  annelids  each  time  the  idea  has  been  discarded  in 

and  mollusks,   and  much  later  for  his  favor   of   the    Department's   Iraditional 

definitive  analysis  of  tissue  interactions  in  organization  of  a  group  of  independent 

the  developing  feather  germ,  Lillie 's  most  investigators  whose  interests  range  widely 

lasting  contributions,  and  those  of  great-  from  biochemistry  and  microbiology  to 

est  heuristic  value,  came  in  the  middle  of  anatomy  and  physiology,  with  substan- 

his  scientific  lifetime.  It  was  in  1916  that  tial  overlapping  in  experience  and  ap- 

he  published  in  Science  two  "milestones,"  proach.  Such  an  organization  might  not 

his  first  analysis  of  the  freemartin  and  a  be  desirable  in  all  fields,  but  in  develop- 

definitive  article  summarizing  his  theory  mental  biology  today  it  appears  to  favor 

of   the   interactions   of  egg   and   sperm  the  generation  and  interchange  of  ideas; 

substances  in  fertilization.  its  flexibility  permits  expansion  as  needed 

In  the  fertilization  article,  Lillie  wrote,  for  limited  periods  as  advances  on  a  given 

referring  to  the  history  of  the  fertilization  problem  warrant  it ;  and  it  helps  ensure 

problem,     "Possibly    the    results    seem  the  vitality  and  sense  of  accomplishment 

slight  as  a  record  of  265  years  of  continu-  of  the  Department,  it  being  unlikely  that 

ous  study  of  a  single  biological  problem,  all    members    of    the    group    would    be 

But  we  read  the  history  of  science  very  experiencing    at    the    same    time    the 

superficially  indeed  if  we  fail  to  realize  temporary    lulls    or    setbacks    that    are 

the    constant    interdependence     of    all  always  a  part  of  research.   Someone  is 

scientific  thought.   There  has  probably  always  ready  to  take  the  lead,  setting  the 

been  no  time  in  the  history  of  our  par-  pace    and    providing    an    example    and 

ticular  subject  when  a  greater  amount  of  stimulus  for  the  others, 
work  on  its  problems  would  have  caused         During  the  past  year  it  fell  to  a  newly 

a  much  more  rapid  advance.  Scientific  appointed     staff     member,     Irwin     R. 

discovery  is  a  truly  epigenetic  process  in  Konigsberg,  and  two  Fellows  of  Carnegie 

which  the  germs  of  thought  develop  in  Institution  of  Washington,   Donald   D. 

the    total    environment    of    knowledge.  Brown  and  Douglas  Caston,  to  assume 

Investigation  of  particular  problems  can-  that  role.  Substantial  progress  is  reported 

not  be  accelerated  beyond  well-defined  in  the  long-range  programs  of  established 

limits;  progress  in  each  depends  on  the  staff  members,  especially  those  of  David 

movement  of  the  whole  of  science."  W.  Bishop  and  Robert  DeHaan,  but  none 

Lillie 's  words  might  have  been  written  has  generated  the  excitement  of  the 
yesterday,  so  aptly  do  they  fit  the  current  Department's  new  ventures.  Perhaps 
state  of  research  in  developmental  biol-  faithful  readers  of  these  reports  will  be 
ogy;  and,  since,  to  a  large  degree,  the  pleasantly  surprised,  as  the  writer  has 
activities  of  the  Department  of  Embry-  been,  at  the  year's  achievements,  for 
ology  mirror  and  influence  those  of  the  much  less  might  have  been  expected  in  a 
field  as  a  whole,  they  also  represent  the  year  in  which  at  least  the  first  four 
philosophy  of  the  writer  and  the  Depart-  months  were  disrupted  by  difficulties 
ment.  It  cannot  be  denied  that  from  time  experienced  in  occupying  the  new  build- 
to  time  thought  has  been  given  to  molding  ing.  The  move  from  the  former  location 

369 


370 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


in  the  New  Hunterian  Building  to  115 
West  University  Parkway,  which  began 
on  August  1,  1961,  found  the  staff  occu- 
pying a  building  in  which  construction 
was  halted  by  a  prolonged  electricians' 
strike.  Although  a  few  experiments  were 
conducted  almost  at  once,  it  was  Novem- 
ber before  serious  work  was  possible.  But 
those  difficult  months  are  almost  for- 
gotten as  the  building  has  already  proved 
to  be  a  place  in  which  the  staff  can  think 
and  work  effectively.  William  E.  Haible's 
design  provided  a  handsome  and  efficient 
structure,  with  an  unusual  combination 
of  fine  qualities,  pleasing  to  both  aesthetic 
and  practical  senses.  It  is  not  fitting, 
perhaps,  to  lay  too  much  emphasis  on  the 
building  itself,  for  after  all  it  is  just  the 
setting  of  our  continuing  research  story, 
but  it  is  an  accomplishment  in  which  we 
can  take  pride,  being  a  further,  notable 
example  of  the  rewards  of  cooperative 
effort,  which  the  Institution  has  long 
stressed. 

Apart  from  the  new  building,  the  year 
was  one  of  eventful  change.  As  was 
mentioned  above,  Irwin  II .  Konigsberg 
joined  the  research  staff  on  July  1,  1961. 
A  student  of  B.  H.  Willier,  and  former 
associate  of  Heinz  Herrmann  in  the 
University  of  Colorado's  Laboratory  of 
Chemical  Embryology,  Konigsberg 
brought  with  him,  from  his  most  recent 
position  in  the  National  Institutes  of 
Health's  Gerontology  Branch,  a  store  of 
experience  and  a  continuing  interest  in 
problems  of  muscle  development.  During 
the  year  he  has  made  great  strides  in 
developing  a  system  that  would  permit 
rigorous  control  of  the  cellular  population 
and  the  extracellular  environment  when 
embryonic  skeletal  muscle  cells  were 
grown  in  dispersed  cell  cultures.  He  has 
demonstrated  convincingly  that  cells  in 
monolayer  cultures  prepared  from  sus- 
pensions of  11-  to  12-day  embryonic  leg 
muscle  undergo  the  characteristic  changes 
of  myogenesis:  rapid  proliferation  of 
mononuclear  cells,  the  formation  of 
multinucleated  myotubes  by  the  fusion  of 
these  cells  (a  point  on  which  Konigsberg 


had  earlier  provided  some  of  the  definitive 
evidence),  and  the  progressive  develop- 
ment of  the  cross- striated  pattern  (with 
the  initiation  of  spontaneous  contrac- 
tions). He  has  now  begun  to  probe  more 
deeply  into  the  mechanisms  of  these  steps, 
as  the  body  of  the  report  will  reveal. 

It  was  noted  in  Year  Book  60  (p.  440) 
that  frog  embryos  develop  normally  in 
media  lacking  magnesium  until  stage  21, 
when  they  die.  As  Donald  D.  Brown 
observed,  however,  the  addition  of  mag- 
nesium before  stage  20  allows  normal 
development.  At  the  time  that  report  was 
written,  the  observation  was  an  isolated 
one ;  it  did  not  fit  readily  into  any  overall 
scheme.  As  so  often  happens  in  research, 
however,  this  seemingly  isolated  fact  has 
now  become  an  important  piece  in  a 
much  larger  mosaic  of  findings  now  re- 
ported by  Brown  and  his  co-worker,  a 
newly  appointed  Fellow,  Douglas  Caston. 
Early  embryos  contain  a  measurable  but 
small  population  of  cytoplasmic  ribo- 
somes.  These  have  been  isolated  and 
quantitated  by  an  isotope  dilution  tech- 
nique coupled  with  standard  homoge- 
nization  and  differential  centrifugation 
methods.  There  is  little  change  in 
ribosomal  content  until  near  the  end  of 
morphogenesis  after  stage  18,  when  there 
is  a  striking  rapid  appearance  of  these 
particles.  At  about  this  time  the  embryo 
requires  exogenous  magnesium  ions, 
which  can  be  shown  to  preserve  the 
integrity  of  ribosomes  in  the  aggregated 
form.  Hence  the  death  of  embryos  due  to 
magnesium  deficiency  is  the  consequence 
of  the  failure  of  a  key  step  in  the  develop- 
ment of  the  machinery  for  protein 
synthesis.  Further  details  of  this  intensive 
study  of  ribosomal  development  will  be 
found  in  the  pages  to  follow. 

Gerald  L.  Carlson,  Given  Foundation- 
National  Research  Council  Fellow  in 
Academic  Medicine,  continued  his  study, 
in  collaboration  with  Bishop,  of  the  nature 
of  the  testicular  antigen  in  induced 
aspermatogenesis.  At  the  same  time, 
however,  he  has  made  important  progress 
in  analyzing  the  serum  amylase  activity 


DEPARTMENT  OF  EMBRYOLOGY  371 

of  the  frog,   Rana  pipiens;  amylase  is  Anatomy,   Middlesex   Hospital   Medical 

among  the  enzymes  that  change  rapidly  School,  London,  examined  the  develop- 

at  metamorphosis,  to  which  much  atten-  ment  of  the  eye.  O'Rahilly,  Special  Fellow 

tion  is  now  being  devoted.  of  the  U.  S.  Public  Health  Service,  began 

Another  enzyme  in  this  group  is  acid  a  comprehensive  study  of  the  developing 
deoxyribonuclease  (DNase  II),  which  human  eye,  while  Silver,  National  Insti- 
JohnR.  Coleman  observed  (Year  Book  60,  tutes  of  Health  International  Post- 
pp.  400-404)  to  increase  concomitantly  doctoral  Traveling  Fellow,  working  in 
with  resorption  of  the  tail.  Coleman,  who  part  in  collaboration  with  John  Papa- 
left  at  the  end  of  1961  to  begin  a  period  constantinou  of  the  University  of  Con- 
of  further  training  and  collaboration  with  necticut,  began  an  analysis  of  lens 
Heinz  Herrmann  at  the  University  of  induction,  emphasizing  techniques  of 
Connecticut,  has  extended  these  studies,  experimental  morphology  as  a  back- 
the  findings  of  which  now  make  it  clear  ground  for  biochemical  and  immuno- 
that  DNase  II  begins  to  rise  shortly  chemical  approaches, 
before  extensive  tail  resorption  begins.  In  December  J.  W.  S.  Harris  returned 

Two  visiting  investigators  devoted  to  his  post  at  London  Hospital  Medical 
their  time  to  exploratory  forays  into  the  College,  after  a  productive  stay  during 
application  of  immunochemical  tech-  which  he  completed  a  substantial  part  of 
niques  to  problems  of  development.  As  a  a  study  of  the  vascular  pattern  of  the 
Fellow  of  Carnegie  Institution  of  Wash-  human  uterus  with  placenta  in  situ, 
ington,  Michael  Abercrombie,  recently  In  the  fall  of  1961,  Ian  Wilson,  a  recent 
appointed  Jodrell  Professor  of  Zoology,  graduate  in  zoology  at  University  College 
University  College,  London,  was  con-  of  North  Wales,  and  recipient  of  the 
cerned  with  techniques  that  might  ad-  Thomas  Henry  Huxley  Award  of  the 
vance  knowledge  of  the  composition  of  Zoological  Society  of  London  for  1961, 
the  cell  surface.  Arthur  LaVelle,  of  the  arrived  to  spend  a  year  as  a  Fellow  of 
Department  of  Anatomy,  University  of  Carnegie  Institution  of  Washington  work- 
Illinois  College  of  Medicine,  centered  his  ing  in  consultation  with  Bent  Boving  on 
attention  on  the  antigenic  properties  of  factors  effecting  the  orientation,  spacing, 
substances  extracted  from  brain.  and  siting  of  the  blastocyst  in  the  mouse. 

Another   aspect    of   the    interplay    of  Chinami  Takata,  formerly  associated 

embryology   and   immunology   occupied  with  T.  Yamada  at  Nagoya  University, 

Bertie  F.  Argyris  of  Syracuse  University,  and    most    recently    a    member    of    the 

a  Fellow  of  the    U.   S.    Public    Health  Department   of   Anatomy,   Tokyo   Uni- 

Service  who  enjoyed  a  productive  period  versity,  took  up  a  one-year  appointment 

of  research  concerned  with  mechanisms  of  in  the  Department  in  September  1961  to 

acquired  tolerance  to  skin  homografts  in  work  in  consultation  with  James  D.  Ebert 

mice.  Thomas  S.  Argyris,  a  Fellow  of  the  on  factors  affecting  the  lability  of  the 

National    Science    Foundation,    devoted  chorioallantoic  membrane, 

his    stay    in    the    Department    to    the  In  addition  to  the  visiting  scientists 

development  of  biochemical  techniques  already  named,  nearly  fifty  investigators 

for  the  eventual  identification  and  meas-  from   eighteen   countries   shared   in   the 

urement    of    tissue- specific    growth-pro-  activities    of    the    Department.    Among 

moting  substances.  them,  to  name  only  a  few  with  continuing 

Two     visiting     investigators,     Ronan  programs,   were   George  W.   Bartelmez, 

O'RahhTy  of  the  Department  of  Anatomy,  G.  W.  Corner,  Jr.,  Anatole  S.  Dekaban, 

Wayne   State   University,    recently   ap-  Arentje  Dekker,  Martin  W.  Donner,  W. 

pointed  Director  of  the  Department  of  Richard  Ferguson,  and  Sheila  J.  Moody. 

Anatomy  at  St.   Louis  University,  and  As  always,  the  list  of  visiting  investi- 

Peter    H.     S.     Silver,     Department    of  gators  and  their  activities  is  impressive; 


372  CARNEGIE     INSTITUTION      OF      WASHINGTON 

yet  such  a  listing  tells  only  a  part  of  the  well  trained  students  of  development  at  a 

story.  The  visitors  do  contribute  vitally  time  when  the  demand  far  exceeds  the 

to  the   Department,   directly  and  indi-  supply. 

rectly,  but  of  far  greater  moment  is  the  Before  concluding  this  recital  of  arrivals 

question  whether  a  visit  adds  measurably  and  departures,  one  more  departure  must 

to  a  man's  ability  as  an  investigator  and  be  recorded,  that  of  the  senior  member  of 

teacher  when  he   returns   to   his   home  the   research   staff,    R.    K.    Burns,   who 

laboratory.  Has  he  found  new  direction  retired  on  June  30,   1962.   It  is  hardly 

or  meaning  for  his   research?   Has   the  necessary  to  recount  even  the  highlights 

opportunity  for  reflection,  away  from  the  in  his  distinguished  career;  his  achieve- 

usual  distractions  of  his  own  laboratory  ments,  described  in  these  reports  over  the 

and  classroom,  led  to  a  searching  reexam-  last  twenty-two  years,  and  in  numerous 

ination  of  his  program?  publications,  speak  for  themselves.  It  is  a 

Such  questions  can  be  answered  objec-  pleasure  to  report  that  he  has  accepted 

tively  only  with  difficulty — if  at  all.  But  an  appointment  as  Professor  of  Biology 

the  writer  can  state  his  own  impression  at  his  alma  mater,  Bridge  water  College, 

that    the    Department's    emphasis    on  where    he    will    teach    embryology    and 

opportunities  for  visiting  investigators  is  continue   his   investigations   of   sex   dif- 

contributing  effectively  to  the  number  of  ferentiation. 


CELLULAR  REGULATORY  MECHANISMS 

-^                             ,T  RNA   with    other    cellular    constituents 

Development  of  Machinery  for  ■,     •           ,      ,            £  ■,      n            . 

during  early  stages  of  development. 

Brotein  Synthesis  Materials  and  methods.  Fertilized  eggs 
Striking  changes  occur  in  metabolic  developed  in  a  modification  of  Holt- 
activities  of  eggs  after  fertilization.  The  freter's  salt  solution  containing  NaCl, 
degree  of  change  is  particularly  impressive  KC1,  CaCl2,  and  MgCl2;  the  medium 
in  amphibians  like  the  frog,  Rana  pipiens,  described  by  Holtfreter  contains  NaHC03 
since  thousands  of  mature  eggs  remain  in  but  no  MgCl2.  Staging  experiments  were 
the  ovary  over  the  winter  with  little  performed  using  embryos  from  the  same 
change  in  their  metabolism.  If  the  gravid  female.  At  the  appropriate  time  the  jelly 
female  is  injected  with  pituitary  glands  was  removed  manually  and  the  embryos 
she  ovulates  in  about  two  days,  and  if  were  frozen  at  —  70°C  until  a  complete 
the  eggs  are  then  fertilized  development  series  could  be  processed  together.  Usu- 
begins,  resulting  in  a  swimming  larva  in  ally  50  embryos  were  taken  each  time, 
six  dayk  The  only  exogenous  require-  Embryos  were  homogenized  in  0.01  M 
ments  during  this  period  are  several  tris  HC1  (trishydroxyaminomethane) 
inorganic  ions.  New  nucleic  acids,  pro-  buffer,  pH  7.3,  containing  0.001  M  MgCl2 
teins,  and  carbohydrates  are  synthesized  (TM).  Homogenates  were  divided  into 
from  material  stored  within  the  eggs.  three  fractions.  The  first,  termed  "pellet," 
The  following  experiments  were  begun  is  the  part  of  the  homogenate  sedimented 
by  Donald  Brown  and  Douglas  Caston  to  at  12,000#  in  a  Servall  centrifuge.  This 
study  components  of  protein  synthesis  fraction  includes  lysosomes  and  mito- 
during  embryogenesis  with  the  hope  of  chondria  as  well  as  all  heavier  particles 
elucidating  the  steps  that  regulate  the  like  yolk  platelets,  melanin  granules,  and 
amount  of  protein  formed  in  early  stages,  nuclei.  The  Servall  supernatant  was 
In  particular,  this  report  correlates  the  centrifuged  at  105,000^  for  2  hours  in  a 
amount,  time  of  appearance,  and  compo-  Spinco  model  L  ultracentrifuge,  and  the 
sition    of    ribosomes    and    ribosome-like  resulting    soluble    supernatant    was    the 


DEPARTMENT  OF  EMBRYOLOGY 


373 


second  fraction.  The  Spinco  pellet  was 
rehomogenized  in  a  small  volume  of  TM ; 
this  suspension  was  centrifuged  for  20 
minutes  at  12,000#,  and  the  supernatant 
was  saved  as  the  third  or  "ribosomal" 
fraction. 

RNA,  DNA,  and  protein  were  meas- 
ured routinely  by  the  orcinol,  diphenyla- 
mine  (Burton  modification),  and  Lowry 
methods,  respectively. 

To  obtain  purified  RNA,  material  was 
homogenized  in  a  suspension  containing 
sodium  chloride  and  sodium  lauryl  sulfate 
at  pH  5.0  and  then  extracted  with  phenol. 
After  centrifugation,  nucleic  acids  in  the 
aqueous  phase  were  precipitated  twice 
with  ethanol  and  finally  dissolved  for 
column  chromatography.  These  prepa- 
rations had  very  low  levels  of  DNA.  The 
high-molecular- weight  RNA  was  frac- 
tionated on  a  2  by  2  cm  column  of  Celite 
containing  methylated  bovine  serum 
albumin  according  to  the  method  of 
Mandell  and  Hershey.  The  nucleic  acids 
were  eluted  with  a  linear  gradient  of 
NaCl  increasing  from  0.2  M  to  1.0  M, 
5  ml/tube  for  300  ml.  Elution  patterns 
were  determined  by  reading  the  eluate 
from  each  tube  in  a  spectrophotometer  at 
260  millimicrons.  The  peak  tubes  were 
pooled,  dialyzed  against  distilled  HoO, 
and  concentrated  in  vacuo.  Base  compo- 
sitions were  determined  according  to  the 
method  of  Wyatt. 

Radioactive  ribosomes  were  prepared 
by  intraperitoneal  injection  of  large  R. 
pipiens  tadpoles  with  high  specific  activ- 
ity orotic  acid-C14.  Ribosomes  were 
isolated  by  the  same  technique  described 
above  between  30  and  50  hours  after 
injection. 

Density-gradient  centrifugation  was 
performed  according  to  the  method  of 
Roberts  and  Britten.  Horse  spleen  fer- 
ritin, a  product  of  Pentex  Corporation, 
and  its  antiserum  (rabbit)  were  generous 
gifts  of  Zoltan  Ovary. 

Cytoplasmic  Ribosomes 

When  "ribosomes"  prepared  from  ova- 
rian eggs  were  compared  with  purified 


liver  particles  several  distinct  differences 
were  apparent  at  once.  Sucrose  density- 
gradient  studies  show  that  the  major 
component  of  this  fraction  from  eggs 
moves  more  slowly  in  the  centrifugal 
field  than  its  counterpart  isolated  from 
liver.  More  striking  still  is  the  insensi- 
tivity  of  the  egg  particle  to  magnesium 
ions.  The  ribosomes  from  adult  liver  are 
RNA  protein  subunits,  held  together  by 
magnesium.  Removal  of  magnesium  by 
versene  causes  disaggregation  of  the 
normal  80  to  100  S  particles  from  frog 
liver  or  whole  tadpoles  into  2  to  4  S 
pieces.  However,  only  a  minor  fraction  of 
the  egg  preparation  is  solubilized  by 
versene  treatment,  the  major  part  re- 
maining unchanged.  Figure  1,  plate  1 
(at  the  end  of  the  report),  shows  sedi- 
mentation patterns  taken  with  schlieren 
optics  of  particles  contained  in  the 
"ribosomal"  fractions  of  frog  liver,  whole 
tadpole,  and  body  cavity  eggs.  Note  the 
marked  coloration  (yellow)  associated 
with  the  major  egg  peak.  The  smaller 
slower-moving  peak  in  the  liver  prepara- 
tion is  also  colored,  but  has  the  same  S 
value  as  the  egg  particle,  and  is  presum- 
ably identical.  Subsequent  purification  of 
the  minor  liver  peak  after  treatment  with 
versene  and  recentrifugation  confirms  the 
similarity.  Another  striking  difference 
between  the  major  component  of  the  liver 
"ribosomal"  fraction  and  that  of  the  egg 
was  their  behavior  to  high  concentrations 
of  equivalent  cations.  Ribosomes  are 
further  aggregated  at  concentrations  of 
Mg++,  Ba++,  or  Ca++  exceeding  0.01-0.02 
M  and  can  then  be  sedimented  at  very 
low  centrifugal  speeds.  Figure  2  compares 
the  response  of  these  two  preparations 
(egg  and  liver)  to  incubation  for  45 
minutes  at  0°C  in  different  concentrations 
of  BaCl2.  After  incubation  the  prepara- 
tions were  centrifuged  at  lbOOg,  and  the 
supernatants  were  decanted  and  read  in 
a  spectrophotometer.  Figure  2  plots 
percentage  of  original  optical  density 
remaining  in  solution  and  clearly  demon- 
strates the  marked  difference  between  egg 
and  liver  particles.  Identical  curves  were 


374 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


o 

UJ 

l- 
< 


o 

UJ 
Ql 


o 

I- 

z 
o 
o 


0.001  0.01  0.1 

CONCENTRATION   OF  BaCI, 


1.0 


Fig.  2.  Precipitation  of  "ribosomal"  fraction 
of  ovarian  eggs  and  liver  by  different  molar  con- 
centrations of  BaCI  2. 


obtained  using  Mg++  or  Ca++  instead  of 
Ba++. 

At  this  point  in  the  investigation  the 
problems  were  clearly  outlined.  If  the 
particles  present  in  the  egg  were  not 
ribosomes,  were  they  in  fact  related  in 
some  way  to  ribosomes,  perhaps  func- 
tionally or  as  ribosomal  precursors?  If 
not,  what  were  these  particles,  and,  of 
greater  interest,  what  was  the  concen- 


tration of  typical  cytoplasmic  ribosomes 
in  the  egg?  Subsequent  investigations 
unequivocally  identified  the  major  com- 
ponent of  the  egg  "ribosomal"  fraction  as 
the  iron  storage  protein  complex,  ferritin. 
We  will  next  describe  the  studies  sup- 
porting this  conclusion  and  will  then 
proceed  with  a  description  of  analyses  of 
frog  eggs  and  early  tadpoles  for  ribo- 
somes, ribosomal  RNA,  and  other  bio- 
chemical parameters. 

Identification  of  Ferritin  in  the  Egg 

After  the  preparation  of  the  "riboso- 
mal" fraction  from  mature  ovarian  eggs, 
the  egg  particle  was  further  purified  by 
elution  from  a  diethylaminoethyl  cellu- 
lose column  (DEAE)  with  NaCl.  The 
ferritin  was  still  particulate  and  could  be 
recovered  in  the  pellet  after  further 
centrifugation  at  105,000*7  for  2  hours. 
Table  1  compares  this  preparation  from 
eggs  with  a  commercially  obtained  sample 
of  horse  spleen  ferritin  further  purified  in 
the  same  manner  by  elution  from  a 
DEAE  column.  The  two  substances  had 
identical  spectra  and  distinction  coeffi- 
cients, and  both  had  a  high  iron  content. 
It  cannot  be  stated  definitively  at  this 
time  whether  the  small  amount  of  orcinol 
reacting  material  associated  with  purified 
ferritin  is  RNA. 

As  a  further  proof  the  cross  reaction  of 
egg  ferritin  with  antiserum  to  horse 
spleen  ferritin  was  investigated.  Upon 
combination   of  the   two,   a  precipitate 


TABLE  1.     Comparison  of  Purified  Egg  Particles  from  "Ribosomal"  Fraction  and  Horse 

Spleen  Ferritin  (Pentex) 


Egg 


Horse 


O.D.  260  nux/mg  protein /ml 
mg  iron*/mg  protein 
mg  "RNA"f/mg  protein 
O.C.  ratios,  m/t 

235/260 

250/260 

280/260 

400/260 


8.8 

9.0 

0.13 

0.16 

0.011 

0.0012 

1.47 

1.35 

1.05 

1.05 

0.93 

0.93 

0.21 

0.24 

*  Iron  was  measured  by  the  o-phenanthroline  method. 

f  "RNA"  was  measured  by  the  orcinol  reaction,  which  is  not  specific  for  RNA. 


DEPARTMENT    OF    EMBRYOLOGY 


375 


developed.  As  a  more  sensitive  test  the 
passive  cutaneous  anaphylaxis  test  devel- 
oped by  Ovary  was  performed  using 
intracutaneous  sensitization  to  horse 
spleen  ferritin  antiserum  followed 
by  intravenous  challenge  with  egg 
ferritin  mixed  with  Evans  blue  dye 
5  hours  later.  Figure  3  (pi.  1)  is  a 
photograph  of  the  resulting  large  blue 
areas  surrounding  sites  of  antibody 
injection.  Serum  from  a  control  rabbit 
gave  a  completely  negative  response. 
Furthermore,  if  the  antiserum  was  pre- 
incubated  with  horse  spleen  ferritin  and 
then  used  to  sensitize  the  guinea  pig,  no 
blue  spot  was  formed  after  the  challenge 
with  egg  ferritin.  The  tests  further 
demonstrated  the  expected  finding  that 
the  reaction  of  frog  ferritin  with  horse 
ferritin  antibody  was  considerably  less 
intense  than  the  homologous  control,  but 
no  attempt  was  made  to  quantitate  this 
difference.  Subsequent  experiments  have 
shown  that  the  entire  complement  of  iron 
in  the  egg  and  early  developmental  stages 
is  confined  to  the  ferritin  fraction,  which 
contains  about  1  jug  of  iron  per  embryo 
(8  Mg  of  ferritin) .  Iron  measurements  have 
been  carried  to  Shumway  stage  20  (see 
illustration  of  stages  in  fig.  6),  in  which 
blood  circulation  has  already  begun.  In 
spite  of  the  presumed  initiation  of  hemo- 
globin synthesis  as  early  as  stages  16-18, 
the  bulk  of  iron  remained  bound  in  the 
ferritin  fraction  as  late  as  stage  20. 

Correlation  of  Death  from  Magnesium 

Starvation  and  Appearance  of 
Cytoplasmic  Ribosomes  in  Development 

It  was  noted  in  Year  Book  60  that  the 
defined  media  previously  described  by 
others  for  raising  early  R.  pipiens 
embryos  to  feeding  stages  were  inade- 
quate. The  deficiency  could  be  remedied 
by  the  addition  of  10"4  M  MgCl2.  In  the 
absence  of  magnesium,  embryos  develop 
normally  to  Shumway  stages  21-23, 
after  which  they  become  immobile  but 
remain  alive  for  another  2  to  3  days. 
Development  continues  to  some  extent, 
including  the  enclosure  of  gills  within  the 


opercular  fold,  but  growth  in  length  is 
arrested  completely  after  the  onset  of 
obvious  symptoms.  Death  occurs  between 
48  and  72  hours  after  the  signs  of  de- 
ficiency first  appear.  Furthermore,  during 
this  period  the  progression  of  death 
cannot  be  reversed  by  the  addition  of 
magnesium.  Substitution  of  manganese  is 
completely  ineffective,  even  if  provided 
from  the  beginning  of  development. 
Magnesium  starvation  can  be  speeded  up 
or  retarded  by  raising  or  lowering  the 
temperature  and  is  correlated  directly 
with  the  Shumway  stage  and  the  length 
of  the  embryo.  (Deficient  embryos  are 
always  arrested  at  about  9  mm  in  length.) 
Initial  experiments  demonstrated  a 
striking  correlation  between  the  stage 
when  ribosomes  can  first  be  detected  and 
the  onset  of  magnesium  deficiency.  Only 
a  small  fraction  of  total  RNA  can  be 
ascribed  to  cytoplasmic  ribosomes  until 
between  stages  21  and  23,  when  they 
appear  suddenly  and  accumulate  expo- 
nentially, doubling  in  amount  approxi- 
mately each  day  (at  21°C).  Figures  4-6 
compile  values  of  RNA,  protein,  and 
DNA  in  the  three  fractions  of  homoge- 
nized embryos  throughout  the  period  of 
early  embryogenesis.  Two  control  series 
are  plotted  together  with  one  series  of 
embryos  allowed  to  develop  in  the 
absence  of  magnesium.  There  is  a 
similarity  in  the  different  parameters 
except  for  ribosomal  RNA.  At  the  time 
that  cytoplasmic  ribosomes  begin  to 
appear  in  quantity  the  symptoms  of 
magnesium  deficiency  become  apparent. 
It  can  be  seen  that  there  is  some  variation 
between  series  in  levels  of  RNA,  DNA, 
or  protein  per  embryo,  although  the 
shapes  of  the  curves  are  comparable.  To 
remedy  this  discrepancy  embryos  were 
reared  on  the  salt  mixture  containing 
magnesium  to  stage  22  (118  hours).  Half 
of  the  embryos  were  then  transferred  to 
magnesium-deficient  medium  after  care- 
ful washing  with  distilled  water.  Develop- 
ment was  allowed  to  continue  to  stage  23 
to  24  (189  hours),  at  which  time  the 
embryos  were  homogenized  and  the  three 


376 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


fractions  analyzed  for  RNA,  DNA,  and 
protein. 

Table  2  summarizes  the  values.  Al- 
though total  RNA  is  the  same,  the  con- 
trol ribosomal  fraction  contains  about  6 
times  more  RNA  than  the  same  fraction 
from  magnesium-deficient  embryos.  Su- 
crose density-gradient  studies  were  per- 
formed to  compare  the  two  "ribosomal" 
fractions.  Figure  7  shows  that,  unlike  the 
control,  the  ribosome  fraction  of  the 
magnesium-starved  embryos  was  devoid 
of  80  and  100  S  particles.  It  can  be 
concluded  that  at  least  a  contributing 
cause  of  death  from  magnesium  starva- 
tion is  the  inability  to  make  cytoplasmic 
ribosomes  at  the  very  time  when  they 
normally  accumulate  at  the  most  rapid 
rate.  Since  the  ribonucleoprotein  subunits 


o 

>- 
oc 
m 

2 

LU 

v. 

o» 

=1. 

< 

Z 

«r 


0  J^_^_L_J^*^1ol<<£D-l<r'.J 1 1 L_L 

0  40        80        120       160      200      260      300 

HOURS 

I I I 1      I     I    I ■       '       I 1 

1  10         17       19       21        23  25 

SHUMWAY   STAGES 

Fig.  4.  "Pellet"  (solid  line)  and  "ribosomal" 
(dotted  line)  RNA  at  different  stages  of  develop- 
ment. Two  control  series  (open  circles  and  open 
squares)  are  compared  with  one  magnesium- 
jstarved  series  (closed  circles). 


0.8 
0.7 
0.6 

o 

>- 

cd     0.5  - 

S      • 

^     0.4 

2 

*     0.3 


O 

cr     0.2 

QL 


0.1 


I      l      I      I      »      I      »      »      I      I      I      »      I      I      » 


0  40       80       120      160      200     240     280     320 

HOURS 
■ I I I  I 

1  10         17      19  20    22         24 

SHUMWAY   STAGES 

Fig.  5.  "Pellet"  (solid  line)  and  soluble 
(dotted  line)  protein  in  the  same  control  and 
magnesium-starved  (closed  circles)  embryos. 
Symbols  are  the  same  as  in  figure  4. 


are  held  together  by  Mg++  the  reverse 
reasoning  may  be  equally  applicable ;  that 
is,  binding  of  magnesium  by  increased 
quantities  of  ribosomes  could  effectively 
deplete  the  cells  of  the  activity  of  many 
magnesium-dependent  enzymes.  (Ts'o 
found  3.7  moles  of  magnesium  bound  per 
12  moles  of  bases  in  ribosomal  RNA.) 

Unlike  adult  tissues  the  egg  contains 
most  of  its  protein  in  an  easily  sedimented 
form  which  can  be  called  structural  or 
insoluble.  Just  before  the  dramatic  ap- 
pearance of  ribosomes,  soluble  super- 
natant protein  begins  to  rise  and  the 
insoluble  fraction  concomitantly  drops. 
The  drop  is  due  to  utilization  of  yolk. 
Magnesium-deficient  embryos  normally 
utilize  yolk  and  form  soluble  protein 
before  the  onset  of  symptoms  (cf.  fig.  5). 
However,  when  death  occurs  the  gut  is 
still  filled  with  yolk,  and  table  2  shows 
that  magnesium-deficient  embryos  have 
a  larger  quantity  of  precipitable  ("pellet") 
protein  and  less  soluble  protein  than  the 
control. 


DEPARTMENT    OF    EMBRYOLOGY 


377 


Binding  Ribosomes 

Results  of  staging  experiments  (fig.  4) 
required  careful  analysis  to  determine  the 
quantity  of  ribosomes  present  in  embryos 
from  Shumway  stages  1  to  21.  The 
conventional  homogenization  and  isola- 
tion techniques  failed  repeatedly  to 
demonstrate  more  than  0.1  jug  of  "riboso- 
mal"  RNA  per  embryo,  a  small  propor- 
tion of  the  total  RNA  present  in  the  egg 
and  early  embryos.  Homogenization  with 
isotonic  sucrose,  phosphate  buffers  in 
different  concentrations,  different  con- 
centrations of  Mg++,  and  sodium  de- 
oxycholate  at  several  pH  values  varying 
from  6.5  to  8.0  all  failed  to  yield  any 
significantly  larger  quantity  of  ribosomes. 
Prolonged  centrifugation  (105,000#  for  15 
hours)  failed  to  sediment  any  macro- 
molecule  containing  RNA. 


o 

> 

CO 

5 
UJ 


I L 


160     200 

HOURS 

J '   i    '      ' I L 


0     10        17      19  20    22  23 


25 


•    • 


I 


Mitt 


SHUMWAY 
STAGES 


Fig.  6.     "Pellet"  DNA  from  the  same  three 
series  as  shown  in  figures  4  and  5. 


0.4 


0.3 


E 

o 
<0 


o 
o 


0.2 


0.1 


10  15  20         25 

TUBE    NUMBER 


30 


35 


Fig.  7.  Sucrose  density-gradient  centrifuga- 
tion pattern  of  purified  ribosomes  from  control 
and  magnesium-starved  stage  24  embryos  (30 
embryos  each).  The  gradient  decreases  from  tube 
1  (bottom)  to  tube  36  (top). 


Next  the  possibility  was  examined  that 
ribosomes  were  indeed  present  but  bound 
or  destroyed  in  some  way  during  homoge- 
nization. When  liver  and  egg  were 
homogenized  together  and  the  ribosomes 
were  isolated,  extremely  poor  yields  of 
particles  were  obtained  and  under  certain 
conditions  all  the  liver  ribosomes  were 
lost.  Subsequent  investigations  ruled  out 
the  action  of  a  nuclease,  since  there  was 
no  apparent  solubilization  of  the  particles. 
Instead  it  could  readily  be  demonstrated 
that  large  quantities  of  added  ribosomes 
and  in  fact  yeast  RNA  were  bound  by  the 
low-speed  egg  pellet.  To  date  no  technique 
or  medium  has  freed  these  bound 
particles.  Figure  8  gives  the  results  of  a 
staging  experiment  in  which  homogeniza- 
tion was  performed  in  the  presence  of 
added  purified  tadpole  ribosomes  labeled 
with  orotic  acid-C14.  It  can  be  seen  that 
none  of  the  added  radioactivity  could  be 
recovered  in  the  ribosome  fraction  until 
stages  21-23,  when  both  ribosomes   (as 


378  CARNEGIE     INSTITUTION     OF      WASHINGTON 

TABLE  2.     Comparison  of  Control  and  Magnesium-Deficient  Stage  24  Embryos  (189  hours) 


Control, 
^tg/embryo 


Magnesium-Starved, 
Aig/embryo 


RNA 

Pellet 

Supernatant 

Ribosomes 

Total 


25.1 
23.7 

8.7 

57.5 


32.5 

22.5 

1.5 

56.5 


DNA 


44.6 


31.0 


Protein 
Pellet 
Supernatant 

Total 


550 
229 

779 


675 
158 

833 


measured  by  ribosomal  RNA)  and  counts 
began  to  appear. 

Although  no  specific  medium  increased 
the  recovery  of  added  counts  during  the 
early  stages,  two  simple  modifications  of 
the  homogenization  technique  resulted  in 


500 


0D 

o 
to 
o 

> 

r~ 

■X) 

> 


0         40        80       120       160      200     240     280 

HOURS 

■      '      ' I I I I I 1 I 

0     li     15     18    20  21     23         24    25 

SHUMWAY    STAGES 

Fig.  8.  CPM  (closed  circles)  and  jug  RNA 
(open  circles)  in  the  ribosomal  fraction  at  differ- 
ent developmental  stages.  Purified  radioactive 
tadpole  ribosomes  labeled  with  orotic  acid-2-C14 
were  added  to  the  homogenization  medium.  The 
RNA  values  are  corrected  for  the  added  radio- 
active RNA. 


much  greater  recovery  of  added  radio- 
active ribosomes.  These  two  procedures 
entailed  homogenization  in  a  very  large 
volume  (1:100  w/v)  and  centrifugation 
within  5  minutes  after  homogenizing  the 
embryos.  With  these  modifications  as 
much  as  50  per  cent  of  added  radioactive 
ribosomes  was  recovered.  The  technique 
made  quantitation  of  endogenous  ribo- 
somes by  an  isotope  dilution  experiment 
possible. 

The  principle  of  isotope  dilution  is  as 
follows:  It  is  assumed  that  any  endoge- 
nous ribosomes  are  present  in  the  free 
state  in  the  egg  but  upon  homogenization 
become  bound  to  other  particles  in  the 
egg  (or  embryo) .  When  enough  ribosomes 
are  present  under  normal  homogenizing 
conditions  the  binding  sites  become 
saturated  and  free  ribosomes  become 
detectable.  If  homogenization  is  carried 
out  in  the  presence  of  a  known  amount  of 
radioactive  ribosomes  of  known  specific 
activity  (expressed  as  CPM/^ug  RNA), 
the  ribosomes  will  freely  and  completely 
mix  with  the  endogenous  pool.  When  the 
purified  ribosomal  fraction  is  isolated, 
counted,  and  measured  for  RNA  content 
the  new  specific  activity  can  be  calcu- 
lated. From  this  value  the  endogenous 
ribosomal  "pool"  size  can  be  determined. 
Figure  9  demonstrates  the  validity  of  this 
technique.  Jelly  was  removed  from  ovu- 
lated eggs,  and  different  numbers  were 


DEPARTMENT  OF  EMBRYOLOGY 


379 


homogenized  in  the  presence  of  the  same 
amount  of  radioactive  ribosomes.  The 
105,000#  pellet  was  homogenized  in  0.01 
M  versene  and  recentrifuged  at  105,000$ 
for  1  hour.  This  procedure  removed 
substances  other  than  ribosomal  RNA 
which  react  with  orcinol,  thus  ensuring 
an  accurate  RNA  value  for  the  specific- 
activity  determinations.  In  this  experi- 
ment it  was  found  that  the  specific 
activity  of  the  added  labeled  ribosomes 
was  indeed  diluted  after  reisolation. 
When  the  appropriate  correction  was 
made  a  linear  curve  resulted  giving  a 
value  for  endogenous  ribosomal  RNA  of 
about  1.6  jug/embryo,  at  least  10  times 
previous  values.  With  this  technique 
another    staging    experiment    was    done 


50  100 

NUMBER    OF    EGGS 


Fig.  9.  Ribosomal  RNA  in  ovulated  unfer- 
tilized eggs  corrected  for  binding  by  the  isotope 
dilution  technique. 


(figs.  10,  11).  Ribosomal  RNA  isolated  by 
both  the  conventional  and  isotope  dilu- 
tion techniques  is  plotted.  Although  the 
base-line  level  of  ribosomes  is  raised  there 
still  appears  to  be  little  increase  in 
ribosomes  before  stage  20,  the  great  bulk 
of  the  RNA  being  associated  with  the 
' 'pellet"  fraction.  Increase  in  supernatant 
protein  can  now  be  correlated  directly  in 
time  with  increase  in  ribosomes. 

Using  Kutsky's  technique  of  labeling 
eggs  by  intraperitoneal  injection  of  P32 


40        80       120       160      200      240      280 
HOURS 
ii        I        i      I       I      i       I 


0     It     15     18    20  21     23         25 

SHUMWAY   STAGES 

Fig.  10.  Ribosomal  and  "pellet"  RNA  (open 
circles)  from  embryos  at  different  stages  of  de- 
velopment. Values  for  ribosomal  RNA  are 
plotted  before  (squares)  and  after  (closed  circles) 
correction  by  the  isotope  dilution  technique. 


> 


0          40        80        120        160      200      240      280 
HOURS 
J I I I I I I 


I L 


O     II     15     18     20  21     23         24     25 

SHUMWAY    STAGES 

Fig.  11.  DNA  (open  circles),  pellet  (closed 
circles),  and  soluble  supernatant  protein 
(squares)  values  for  the  staging  experiment 
recorded  in  figure  10. 


380 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


into  the  gravid  female,  it  was  possible  to 
reverse  the  isotope  dilution  experiment 
and  study  the  appearance  of  P32-labeled 
ribosomes  in  the  developing  embryo. 
When  radioactive  embryos  were  homog- 
enized in  the  presence  of  a  large  excess  of 
added  purified  nonradioactive  tadpole 
ribosomes  only  trace  amounts  of  P32  were 
detectable  in  the  ribosomal  fraction  until 
after  Shumway  stage  18.  At  this  time 
there  was  a  rapid  appearance  of  radio- 
activity in  the  ribosomal  fraction.  Radio- 
active unfertilized  eggs  have  the  same 
low  level  of  ribosomal  radioactivity  as 
long  as  6  days  after  P32  injection.  At  21°C 
the  lag  period  preceding  the  appearance 
of  radioactivity  in  the  ribosomal  fraction 
in  a  large  tadpole  is  30  to  40  hours. 


E 
o 

<£> 
CVJ 

O 
O 


1.0 


0.9 


0.8 


0.7 


0.6 


0.5 


0.4 


0.3 


0.2 


0. 


Ovulated  Unfertilized  Egg 


TUBE 


40 
NUMBER 


50 


60 


Fig.  12.  Elution  pattern  of  purified  ovulated 
unfertilized  egg  nucleic  acid  before  (closed 
circles)  and  after  (open  circles)  ribonuclease 
digestion. 


10        20        30        40        50  60 

TUBE     NUMBER 


Fig.    13.     Elution  pattern  of  purified  RNA 
from  stage  21  embryos  (115  hours  at  21°C). 


High-Molecular-Weight  RNA  in 
Development 

In  view  of  the  low  level  of  ribosomes 
in  the  frog's  egg  and  early  stages  as  well 
as  the  unusual  characteristics  of  isolated 
RNA  of  these  eggs  described  by  Finamore 
and  Volkin,  techniques  were  developed  to 
isolate  undegraded  high-molecular-weight 
RNA  from  early  stages.  This  RNA  has 
been  compared  with  RNA  isolated  by 
the  same  technique  from  large  whole 
tadpoles  and  adult  frog  liver.  The  features 
of  the  homogenizing  medium  are  (a)  the 
inclusion  of  sodium  lauryl  sulfate,  without 
which  subsequent  phenol  extraction  is 
completely  ineffective,  and  (6)  acidic  pH 
(5.0)  which  both  augments  phenol  pre- 
cipitation   of    protein    and    keeps    the 


DEPARTMENT    OF    EMBRYOLOGY 


381 


extraction  at  a  pH  where  ribonuclease,  if 
present  in  these  early  stages,  is  inactive. 
The  entire  procedure  is  carried  out  at 
0°C.  Very  little  DNA  is  extracted  by  this 
method. 

The  methylated  serum  albumin  column 
developed  by  Mandell  and  Hershey  for 
DNA  separates  nucleic  acids  according  to 
size.  The  column  has  been  applied  to 
RNA  by  McCarthy,  Sueoka,  and  others 
and  has  been  shown  to  separate  soluble 
RNA  from  ribosomal  RNA.  Original 
ultracentrifuge  studies  of  ribosomal  RNA 
described  two  large  sizes,  16  and  23  S. 
More  recent  studies  imply  that  there  is 
only  one  size,  of  about  30  S,  and  that  the 
smaller  pieces  were  the  result  of  nuclease 
action.  As  techniques  for  isolation  of 
RNA's  have  improved,  "messenger" 
RNA  has  also  been  found  to  be  of  high 
molecular  weight.  Figure  12  shows  a 
NaCl  gradient  elution  pattern  of  ovulated 
egg  RNA  and  demonstrates  the  presence 
of  high-molecular-weight  RNA.  Subse- 
quent stages  are  given  in  figures  13-15.  A 
comparable  fraction  of  RNA  was  always 
eluted  at  0.7  M  NaCl.  Table  3  sum- 
marizes the  similar  if  not  identical  base 
composition  of  this  peak  in  the  various 
preparations  and  further  compares  these 
results  with  the  base  composition  of 
purified  tadpole  ribosomal  RNA.  It 
cannot  as  yet  be  determined  what 
percentage  of  the  high-molecular-weight 
RNA  isolated  by  this  technique  is 
attributable  to  the  small  number  of 
ribosomes  present  in  the  egg. 


£ 

S    .4 
CM 
O 

O 


.2 


I     ■ 


Stage  25 
215  hours 


A 


J 


.•.*" 


0  10  20  30  40  50  60 

TUBE    NUMBER 

Fig.    14.     Elution  pattern   of  purified  RNA 
from  stage  25  embryos  (215  hours  at  21°C). 


a. 

E      -3 


O      .2 
O 


Frog   Liver 

1 

h 

i  \ 

"     1 

!       \ 

1                   V* 

i  \  . 

— ~    J 

i        i 

i         i         i     ••••r 

0  10  20  30  40  50  60 

TUBE    NUMBER 

Fig.    15.     Elution   pattern   of  purified   RNA 
from  adult  frog  liver  (0.5  gram  wet  weight). 


TABLE  3.     Base  Composition  of  High-Molecular-Weight  RNA  Eluted  at  0.7  M  NaCl  from 

Methylated  Serum  Albumin  Column 


Shumway  Stage 


Guanine 


Adenine  Cytosine 


Uracil 


0  (unfertilized) 
11  (gastrula) 
21 
25 

Whole  tadpole 
Liver 
Tadpole  ribosomal  RNA* 


32 

21 

29 

18 

36 

18 

29 

17 

36 

20 

27 

17 

34 

19 

29 

18 

32 

19 

28 

21 

35 

21 

27 

17 

34 

21 

28 

17 

*  RNA  was  isolated  from  105,0000  pellet  by  the  phenol  procedure, 
hydrolyzed  directly  and  not  chromatographed  on  the  column. 


The  ethanol  precipitate  was 


382 


CAKNEGIE     INSTITUTION     OF      WASHINGTON 


Size  and  Localization  of  Synthesized 
Ribosomes  after  Stage  23 

Roberts  et  al.  have  demonstrated  by 
isotope  studies  that  large  bacterial  ribo- 
somes are  constructed  by  aggregation  of 
smaller  ribonucleoprotein  particles.  When 
ribosomes  accumulate  in  the  cytoplasm 
of  embryos  after  stages  21-23  they  are 
already  in  the  form  of  80-100  S  particles. 
This  observation  implies  that  these 
particles  had  been  formed  elsewhere, 
perhaps  in  the  nucleus  as  suggested  by 
several  workers. 

The  question  arose  whether  ribosomes 
appeared  in  all  tissues  of  the  developing 
embryo      simultaneously.      Twenty-five 


absence  of  ribosomal  particles  in  the 
ventral  or  visceral  fraction,  implying  that 
ribosome  appearance  here  lags  behind 
that  in  the  tail,  head,  and  dorsal  parts  of 
the  embryo. 

Discussion.  In  a  developing  system 
where  all  components  during  the  critical 
early  period  of  development  are  derived 
from  stored  materials  it  is  not  surprising 
that  the  egg  should  contain  a  completely 
different  array  of  macromolecules  both 
particulate  and  soluble  from  those  present 
in  a  mature  tissue.  From  one  point  of 
view  the  egg  is  a  remarkably  inert  storage 
depot,  consisting  largely  of  temporary 
items  that  will  be  used  for  the  formation 
of  more  permanent  ones  during  develop- 


TABLE  4.     Regional  RNA  and  DNA  Contents  in  Stage  24  Embryo  (189  hours) 


Dissected  Embryo,*  jug 

Whole 

Dorsal 

Visceral              Tail 

Total 

Mg 

RNA 

Pellet 

12.3 

13.7                  3.2 

29.2 

25.1 

Supernatant 

17.2 

5.3                   6.2 

28.7 

23.7 

Ribosomes 

6.8 

0.03                 2.1 

8.9 

8.7 

DNA 

21.9 

10.7                  4.4 

37.0 

44.6 

Ribosomal  RNA/DNA 

0.31 

0.003               0.48 



0.20 

See  text  for  description  of  three  dissected  regions. 


stage  24  embryos  (189  hours)  were 
divided  into  three  parts:  (a)  tail;  (6) 
contents  of  the  abdominal  cavity,  the 
largest  part  of  which  was  yolk-filled  gut; 
and  (c)  the  remainder  of  the  embryo, 
which  included  the  head  and  the  dorsal 
part.  The  usual  homogenization  and 
fractionation  was  performed,  and  RNA, 
DNA,  and  protein  contents  were  meas- 
ured. Table  4  compares  values  of  dissected 
embryos  with  values  for  whole  embryos 
homogenized  at  the  same  time.  The 
summation  of  the  different  parameters  of 
dissected  embryos  is  in  good  agreement 
with  control  values  for  whole  embryo 
homogenates.  In  the  final  line  ribosomal 
RNA/DNA  ratio  represents  a  measure  of 
ribosomes  per  cell.   There  is  a  striking 


ment.  This  report  has  added  the  iron 
binding  protein  complex  ferritin  to  the 
list  of  constituents  in  the  frog  egg. 

Since  early  development  of  the  frog 
(to  feeding)  does  not  require  net  increase 
in  mass  it  can  be  considered  a  rearrange- 
ment of  substances  already  present.  The 
greatest  part  of  this  change  from  both  a 
chemical  and  morphological  point  of  view 
is  structural.  In  fact,  Brown  and  Caston 
have  shown  in  experiments  not  reported 
here  that  in  very  early  stages  and  in  the 
unfertilized  egg  itself  radioactive  pre- 
cursors introduced  into  the  pool  for 
protein  synthesis  are  largely  incorporated 
into  "pellet"  or  structural  protein.  As 
development  progresses  the  ratio  of 
radioactivity  found  in  "pellet"  protein  to 


DEPARTMENT    OF    EMBRYOLOGY  383 

that  incorporated  into  soluble  protein  With  the  rapid  synthesis  of  these  particles 
shifts  greatly,  so  that  by  stage  20  equal  existing  stores  of  Mg++  are  rapidly 
amounts  are  incorporated  into  each  bound,  imposing  a  magnesium  deficiency 
fraction.  Similar  results  were  found  by  on  the  embryo.  As  would  be  expected 
Hultin  in  his  studies  on  sea  urchin  such  a  deficient  embryo  completely  stops 
development.  Whether  this  classification  growing  but  can  in  fact  continue  a  certain 
of  soluble  and  structural  proteins  based  amount  of  development  even  in  a  mori- 
on centrifugal  properties  is  of  biological  bund  state.  Although  few  ribosomes  can 
significance  remains  an  important  subject  be  demonstrated  in  the  early  embryo,  all 
for  future  investigation.  Staging  experi-  stages  contain  large-molecular-weight 
ments,  illustrated  in  figures  4-6  and  RNA  having  a  base  composition  like  that 
10-11,  lend  further  support  to  this  of  ribosomal  RNA.  However,  large- 
distinction,  since  only  a  slight  net  molecular-weight  RNA  isolated  from 
increase  in  soluble  protein  occurs  before  ovulated,  unfertilized  eggs  labeled  6  days 
stage  20.  Then,  directly  concurrent  with  previously  with  P32  is  completely  devoid 
the  appearance  of  cytoplasmic  ribosomes,  of  radioactivity  (less  than  0.01  per  cent 
the  amount  of  soluble  protein  increases  of  the  fraction  isolated  by  the  phenol 
rapidly.  This  correlation  suggests  to  extraction  technique) . 
Brown  and  Caston  that  "soluble"  protein  Summary.  Early  embryos  contain  a 
may  be  formed  by  the  classical  sequence  measurable  but  small  population  of  cyto- 
of  events  where  final  amino  acid  assem-  plasmic  ribosomes.  These  have  been 
bly  occurs  on  "cytoplasmic"  ribosomes  isolated  and  quantitated  by  an  isotope 
whereas  the  "structural"  or  sedimentible  dilution  technique  coupled  with  standard 
protein  so  necessary  for  early  morpho-  homogenization  and  differential  centrifu- 
genesis  is  a  product  of  another  as  yet  gation  methods.  There  is  little  change  in 
undefined  series  of  events  not  requiring  ribosomal  content  until  near  the  end  of 
the  presence  of  typical  ribosomes.  There  morphogenesis  after  Shumway  stage  18, 
is  very  little  evidence  at  present  regarding  when  there  is  a  striking  rapid  appearance 
the  origin  of  "structural"  protein  such  as  of  these  particles.  Simultaneously,  the 
that  contained  in  mitochondria,  cell  wall,  embryo  requires  exogenous  magnesium 
or  the  nucleus.  The  other  known  cellular  ion,  which  can  be  shown  to  retain  the 
system  for  protein  synthesis  is  the  integrity  of  the  ribosomes  in  the  80-100  S 
similar  if  not  identical  sequence  of  events  aggregated  form.  When  ribosomes  are 
that  occurs  in  the  cell  nucleus,  partly  formed  there  is  a  concomitant  increase  in 
elucidated  by  Allfrey,  Mirsky,  and  their  "soluble  protein"  at  the  expense  of 
collaborators.  The  attractive  possibility  precipitable  protein  (yolk).  Ribosomes 
that  "structural"  protein  is  nuclear  in  appear  in  the  cytoplasm  as  "mature" 
origin  remains  to  be  proved.  Staging  particles  80-100  S  in  size  and  do  not 
experiments  show  that  the  great  bulk  of  accumulate  uniformly  in  all  parts  of  the 
RNA  before  stage  20  is  localized  in  parts  embryo,  the  visceral  part  being  prac- 
of  the  cell  that  are  not  in  communication  tically  devoid  of  ribosomes  at  a  time 
with  cytoplasmic  ribosomes.  Whether  or  when  tail  and  head  parts  have  sizable 
not  this  RNA  is  nuclear,  it  parallels  DNA  amounts. 

appearance  closely.  The  pattern  and  base  ratio  of  high- 

Ribosomal  synthesis  begins  after  much  molecular-weight  RNA  throughout  early 

of  organogenesis   and   morphogenesis   is  development  has  been  studied,  revealing 

completed.  Perhaps  we  can  think  of  these  ''ribosomal"  RNA  in  all  stages  having 

ribosomes    as    being    involved    in    the  comparable  base  composition, 

synthesis  of  proteins  required  for  growth  Iron  is   stored   in  the   egg  bound   to 

of  the  differentiated  system  rather  than  ferritin.    The   protein   moiety   has   been 

for  the  process  of  differentiation  itself,  unequivocally    identified    by    its    com- 


384 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


parison  with  purified  horse  spleen  ferritin 
by  both  chemical  and  immunological 
techniques. 

Assay  of  Vertebrate  Ribonucleic 

Acid  for  a  Ribonucleic  Acid  Fraction 

Specifying  Polypeptide  Sequence 

Since  specific  proteins  are  detected  in 
different  tissues,  the  regulation  of  protein 
synthesis  in  developmental  processes  by 
regulation  of  synthesis  of  specific  RNA 
differentially  coded  from  genetic  material 
has  become  a  popular  hypothesis  (e.g., 
Jacob  and  Monod,  Leslie).  With  the 
description  by  Nirenberg  and  Matthaei 
of  a  cell-free  system  from  Escherichia  coli 
which  catalyzes  the  incorporation  of 
amino  acids  into  polypeptide  chains  in 
sequences  somehow  specified  by  the 
sequence  of  nucleotides  in  ribonucleotide 
polymers,  it  became  possible  to  test 
vertebrate  tissues  for  the  presence  of 
similar  RNA  active  in  specifying  protein 
structure.  The  assay  of  RNA  active  in 
coding  for  specific  proteins  would  in  turn 
permit  a  functional  test  of  the  idea  of 
differential  gene  expression  in  develop- 
ment through  differential  elaboration  of 
coded  RNA. 

In  studies  related  to  those  of  Brown 
and  Caston,  but  conducted  independently 
by  G.  L.  Carlson,  cell-free  preparations 
from  E.  coli  corresponding  to  the  30  S 
(30  X  103<7  supernatant)  fraction  of 
Nirenberg  and  Matthaei  were  treated 
with  DNase  and  preincubated  to  inac- 
tivate the  " template"  RNA  in  extracts. 
RNA  was  prepared  from  chicken  testes 
and  frog  liver  by  several  phenol  extrac- 
tion procedures  and  incubated  together 
with  the  30  S  fraction  and  supplements 
described  by  Nirenberg  and  Matthaei. 
Incorporation  of  C14-arginine  into  the 
acid-insoluble  protein  of  the  30  S  fraction 
was  not  stimulated  by  any  of  the  verte- 
brate RNA  preparations,  hence  these 
negative  experiments  are  not  reported  in 
detail.  Efforts  are  being  continued  toward 
preparation  of  vertebrate  RNA  functional 
in  this  or  a  similar  assay  system. 


Regulation  of  Antibody  Formation 

Acquired  Tolerance  to  Skin  Homografts 
in  Mice 

As  an  embryo  develops  into  an  adult, 
it  acquires  the  capacity  to  recognize  and 
reject  foreign  invaders.  The  rate  of 
development  of  the  immune  system  varies 
with  the  species,  but,  in  general,  it  can 
be  said  that  at  birth,  or  shortly  before  or 
after  birth,  most  animals  become  capable 
of  an  immune  response  and  are  able  to 
distinguish  "self"  from  "nonself."  The 
phenomenon  of  actively  acquired  toler- 
ance was  first  demonstrated  experimen- 
tally by  Billingham,  Brent,  and  Medawar, 
who  injected  fetal  or  newborn  mice  with 
suspensions  of  homologous  cells  including 
lymphoid  cells,  and  found  that  these 
mice,  when  challenged  later,  accepted 
skin  homografts  from  mice  of  the  same 
inbred  strain  as  the  original  donor.  The 
"stem  cell"  concept  of  acquired  tolerance, 
which  holds  that  the  immune  system  of 
the  recipient  is  affected  specifically  at  a 
critical  time  during  its  maturation,  is  now 
being  subjected  to  careful  examination  in 
several  laboratories.  It  appears  likely  that 
a  modification  of  the  original  hypothesis 
will  be  necessary,  for  evidence  is  now 
mounting  to  indicate  that  actively  ac- 
quired tolerance  and  immunologic  paraly- 
sis are  closely  related  phenomena.  The 
following  experiments  conducted  by  B.  F. 
Argyris  are  concerned  with  the  mecha- 
nisms of  acquired  tolerance  to  skin 
homografts  in  mice. 

Newborn  C3H  mice  are  injected  intra- 
venously with  spleen  cells  from  adult 
female  mice  of  the  CBA  strain.  Two 
months  later  the  injected  mice  are  test- 
grafted  with  CBA  skin,  and  all  accept 
homografts  for  at  least  2  months.  At 
times  when  the  tolerant  mice  are  bearing 
such  successful  homografts,  donor  (CBA) 
cells  can  be  found  in  the  lymphoid  tissues 
of  all  tolerant  C3H  mice  (table  5).  To 
detect  the  presence  of  these  cells,  the 
"chimera"  test  as  described  by  Billing- 
ham and  his  associates  is  employed. 
Lymphoid  cells  from  the  C3H  chimeras 


DEPARTMENT  OF  EMBRYOLOGY 


385 


TABLE  5.     Chimera  Analysis  of  Fully,  Partly,  and  Post-Tolerant  C3H  Mice 

Assay  C3H  mice  injected  intraperitoneally  with  lymphoid  cells  of  the  tolerant  animals  and  6  days 
later  test-grafted  with  CBA  strain  skin.     Biopsies  removed  for  histological  analysis  7  days  later. 


Tolerance 


No.  Mice       Chimeras  (presence  of  CBA  strain  cells) 


Full 
Partial 
After  1  graft 
After  2  grafts 


are  injected  into  C3H  assay  mice.  Six 
days  later  these  assay  mice  are  test- 
grafted  with  skin  from  a  mouse  of  strain 
CBA.  Histological  analysis  of  7-day 
biopsies  of  these  skin  grafts  indicates  a 
first-  or  second-set  response,  in  turn 
suggesting  the  respective  absence  or 
presence  of  CBA  strain  cells  in  the  C3H 
tolerant  mice. 

To  study  the  fate  of  skin  homografts 
in  older  tolerant  mice,  a  group  of  109  C3H 
mice  which  bore  successful  CBA  skin 
grafts  for  at  least  2  months  has  been  set 
aside  for  long-term  observations.  As  these 
mice  grow  older,  an  increasing  number  of 
them  show  signs  of  losing  their  tolerance. 
The  loss  of  tolerance  usually  starts  with 
hair  loss  on  the  graft  and  subsequent 
contraction.  No  scab  is  formed,  and  the 
end  point  is  marked  by  a  smooth  scar. 
Graft  rejection  can  be  a  prolonged  process 
(average  duration  62  days;  range  1  day 
to  6  months).  When  the  loss  of  tolerance 
is  rapid,  sudden  scab  formation  takes 
place,  starting  on  the  periphery  of  the 
graft  and  spreading  inward.  At  present 
41  mice  (37  per  cent)  have  rejected  their 
grafts  between  2.5  and  10  months  after 
grafting  and  27  (25  per  cent)  are  in  the 
process  of  rejecting  their  grafts.  This 
means  that  a  total  of  62  per  cent  of  the 
once-tolerant  mice  are  losing  or  have  lost 
their  tolerance.  No  clear  correlation  has 
been  found  between  the  number  of  cells 
injected  at  birth  and  the  time  of  onset  or 
completion  of  graft  rejection,  but  a  more 
detailed  analysis  of  the  quantitative 
aspects  of  this  problem  is  in  progress. 

Fifteen  tolerant  mice,  which  rejected 


their  skin  homografts  between  71  and  290 
days  after  grafting  (average  183  days), 
were  regrafted  on  the  contralateral  side 
with  CBA  strain  skin.  Ten  of  the  mice 
that  rejected  their  first  graft  between  71 
and  290  days  (average  166  days)  rejected 
their  second  graft  in  11  to  32  days 
(average  19  days).  This  suggests  that 
these  mice  have  almost  or  completely 
recovered  their  immune  reactivity  to  the 
cells  to  which  they  were  once  tolerant. 
None  gave  a  second-set  response.  The 
remaining  5  regrafted  mice,  which  re- 
jected their  first  graft  in  149  to  257  days 
(average  216  days),  are  still  bearing  their 
second  graft  at  the  time  of  writing  (2-3 
months  after  grafting) .  We  may  conclude 
therefore  that  the  delayed  loss  of  tolerance 
is  not  always  accompanied  by  a  complete 
recovery  of  the  immune  system  of  the 
host,  differing  in  this  respect  from  the 
immediate  immune  recovery  of  chickens 
after  the  loss  of  tolerance  to  homologous 
red  blood  cells  or  of  mice  after  the  loss  of 
tolerance  to  a  soluble  protein,  described 
by  Mitchison  and  Torres,  respectively. 

Chimera  analysis  of  a  small  number  of 
mice  during  the  process  of  graft  contrac- 
tion ("partly  tolerant"  mice)  and  shortly 
after  the  rejection  of  the  first  skin  graft 
("post-tolerant;  1  graft")  indicates  that 
donor  cells  are  present  even  after  the 
graft  is  rejected  (table  5).  At  later  stages 
after  the  loss  of  the  skin  graft  and  after 
the  second  graft  has  been  rejected  no 
donor  cells  can  be  detected  in  the  post- 
tolerant  mice  (table  5).  Since  the 
sensitivity  of  detection  of  CBA  cells  in 
C3H  assay  mice  is  of  the  order  of  0.5 


386 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


million  CBA  cells  (table  6),  we  can 
conclude  that  fewer  than  0.5  million  CBA 
cells  are  present  in  these  post-tolerant 
mice. 

The  delayed  loss  of  tolerance  and  the 
disappearance  of  CBA  cells  in  the 
tolerant  C3H  mice  suggest  a  reaction 
between  the  host  and  donor  cells  with 
final  victory  for  the  host.  To  determine 
whether  there  is  histological  evidence  for 
such  a  host-versus-graft  reaction,  and,  if 


mesenteric  lymph  nodes  were  examined. 
During  the  early  stages  kidney  and  liver 
were  also  studied.  Routine  histological 
procedures  were  followed,  and  5-micron 
sections  were  stained  with  methyl  green/ 
pyronin  and  hematoxylin/eosin.  So  far 
63  tolerant,  7  partly  tolerant,  16  post- 
tolerant,  72  control  (nongraf  ted) ,  10 
control  (isograf ted) ,  13  control  (homo- 
grafted),  and  10  control  (injected  with 
homologous  cell  suspension)   mice  have 


TABLE  6.     Sensitivity  of  Assay  Test  for  Homologous  Cells  in  the  CBA  +±  C3H  Strain 

Combination  of  Mice 

Assay  mice  injected  intraperitoneally  with  homologous  lymphoid  cells  and  6  days  later  test-grafted 
with  donor-type  skin.     Biopsies  removed  for  histological  analysis. 


Strain  Combination 

No.  Homologous 

CBA^C3H 

C3H->CBA 

Cells  Injected, 
X106 

No.  Mice  - 
Tested 

No.  Mice  with 

lst-set             2nd-set 
response          response 

No.  Mice 
Tested 

No.  Mice  with 

lst-set 
response 

2nd-set 
response 

10 
5 

5 

7 

5 

7 

6 
4 

0 
1 

0 
1 

3 
3 

5 
6 

5 
6 

> 

3 

1 

4 

5 
5 

1 

3 

1 
0.5 

0.25 
0.125 

3 
3 

2 

2 

Horizontal  arrows  denote  lower  level  of  sensitivity. 


so,  when  it  develops  in  the  tolerant  host, 
a  detailed  histological  analysis  of  the 
lymphoid  tissues  of  tolerant  mice  has  been 
undertaken.  The  strain  combination 
selected  (CBA— >C3H)  is  particularly 
advantageous,  because  the  high  incidence 
of  acquired  tolerance  (100  per  cent)  after 
injection  of  standard  doses  of  homologous 
lymphoid  cells  allows  for  the  study  of 
early  stages  in  the  development  of 
acquired  tolerance  before  test  grafting 
becomes  technically  feasible.  Moreover, 
the  absence  of  runt  disease,  believed  to 
be  a  graft-versus-host  reaction,  avoids 
additional  complications.  In  these  studies 
the    spleen,    thymus,    and    axillary   and 


been  examined.  Although  the  study  has 
not  been  completed  the  following  account 
can  be  given  with  reasonable  confidence. 
Although  the  postnatal  development  of 
the  lymphoid  system  has  been  described 
in  detail  for  the  guinea  pig  by  Gyllensten, 
and  in  general  seems  to  resemble  that  of 
the  mouse,  species  and  strain  variations 
in  the  rate  of  maturation  of  the  immune 
system  warrant  a  short  description  of  the 
events  taking  place  in  the  lymphoid 
tissues  of  the  postnatal  C-H  mouse.  At 
birth  the  C3H  spleen  consists  mainly  of 
red  pulp,  which  is  densely  packed  with 
myelopoietic  and  erythropoietic  elements, 
megakaryocytes  and  large  pyroninophilic 


DEPARTMENT    OF   EMBRYOLOGY  387 

cells.  Lymphoid  follicles  are  just  begin-  Generally    after    2    months    the    spleen 

ning  to  form  around  a  central  artery.  The  becomes    less    active,    germinal    centers 

lymph  nodes  are  undifferentiated  small  decrease  in  size,  lymph  nodes  become  less 

nodules    of    reticulum    cells.    The    liver  active,   and  very  few  plasma  cells  are 

shows  a  high  rate  of  hemopoiesis,  and  the  found   in  the   axillary  lymph   nodes   of 

thymus  is  well  developed,  consisting  of  a  control  mice.  Mesenteric  nodes,  on  the 

cortex  of  actively  dividing  thymocytes  other  hand,   continue  to  contain  many 

and    a    less    dense    medulla.    Since    no  plasma  cells. 

unusual  changes  take  place  in  the  thymus  Around  4  months  of  age   (2  months 

except  for  a  gradual  involution  in  older  after  grafting)  a  difference  in  histological 

mice,  we  will  make  no  further  mention  of  picture  between  control  and  experimental 

this  organ.  After  8  to  9  days  myelopoiesis  mice    becomes    apparent.    The    axillary 

disappears  from  the  splenic  red  pulp  and  (regional)  lymph  nodes  of  tolerant  mice 

the    liver;    the    lymph    nodes    begin   to  enlarge,  the  number  and  size  of  germinal 

differentiate;   primary   nodules   form   in  centers  increase,  and  many  plasma  cells 

the  cortex,  and  cords  begin  to  develop  in  can  be  found  in  the  medullary  cords  of 

the  medulla.  Around  12  to  14  days  of  age,  these   nodes.   At   7   months   (the   oldest 

few    germinal    centers    develop    in    the  tolerant  mice  studied  so  far)  this  differ- 

splenic  white  pulp.  Erythropoiesis  disap-  ence  between  control  and  experimental 

pears  from  the  liver  at  this  stage,  and  no  mice  is  still  evident.  The  cortex  of  the 

further  pronounced  changes  are  seen  in  axillary    lymph    node    may    also    show 

the    liver.    Mast    cells    appear    in    the  clusters  of  plasma  cells,  a  phenomenon 

medulla  of  the  axillary  lymph  nodes.  No  rarely  observed  in  normal  axillary  lymph 

difference  is  found  between  the  control  nodes.  As  before,  the  spleens  of  tolerant 

and  experimental  (tolerant)  animals  ex-  and   of  control  mice   present   a  similar 

cept  for  an  occasional  plasma  cell  in  the  picture.   At  this  age  the  red  pulp  has 

cortex  or  medulla  of  the  lymph  nodes  in  decreased  considerably  in  cellularity,  the 

the  tolerant  animals.  These  cells  probably  amount    of    erythropoiesis    is    reduced 

represent  donor  cells,  reacting  against  the  significantly,  many  red  blood  cells  are 

host   antigens.   This  weak  graft-versus-  stored    in   the    red    pulp,    and    a    large 

host  reaction  appears  of  no  consequence  amount  of  pigment  is  present, 

to  the  host.  To  be  certain  that  the  enlargement  of 

No  differences  in  histological  appear-  the  regional  lymph  node  in  tolerant  mice 

ance  can  be  detected  between  lymphoid  is  not  a  nonspecific  response  to  trauma  of 

tissues  from  normal  (control)  and  tolerant  the  skin  grafting  procedure,  the  effect  of 

(experimental)  mice  ranging  in  age  from  isografts    on    regional    lymph    nodes    of 

2  weeks  to  4  months.  During  this  time  control    C3H    mice    was    studied.     No 

the  white  pulp  of  the  spleen  expands,  enlargement  or  plasma  cells  are  found  in 

germinal    centers   increase    in   size,    the  the  regional  lymph  nodes  of  mice  1  or  2 

density  of  the  red  pulp  decreases,  and  the  months  after  the  receipt  of  an  isograft. 

number  of  stored  red  blood  cells  in  the  The  enlargement  of  the  regional  nodes  in 

red   pulp   increases.    At    2   months   the  tolerant  mice  therefore  appears  to  be  a 

spleens  of  both  control  and  experimental  specific    immunologic    response    to    the 

mice    can    be    active,    with    extensive  graft. 

germinal    centers    and    occasional    small  In  the  partly  tolerant  mice,  those  killed 

clusters  of  plasma  cells.  Germinal  centers  at   the   time    of   graft    contraction,    the 

can  also  be  found  in  the  cortex  of  the  picture  is  similar  to  that  in  tolerant  mice, 

lymph   nodes  at  this  time,  and  plasma  No  difference  is  observed  in  the  spleen  or 

cells  are  present  in  the  medullary  cords,  mesenteric  lymph  node,  but  an  immune 

The   mesenteric    node   has   many   more  response  of  the  regional  lymph  node  is 

plasma    cells    than    the    axillary    node,  apparent.  A  few  post-tolerant  mice  have 


388  CARNEGIE     INSTITUTION     OF      WASHINGTON 

been  studied  shortly  after  the  contraction  times  partly  and  post-tolerant  mice  are 

of    their    first    graft.    The    histological  still  chimeras.  After  the  rejection  of  a 

picture  of  these  animals  resembles  that  second     graft     by     post-tolerant     mice, 

of  the  preceding  group.  After  the  rejection  however,  the  immune  response  seems  to 

of   a   second   graft,   however,   the   post-  involve  both  spleen  and  lymph  nodes, 

tolerant  mice  present  an  entirely  different  Accordingly  no  donor  cells  can  be  found 

picture.    As   before,    the   regional    node  by   the   chimera   test   at   this   time.    A 

shows  an  immune  response,  but  now  the  summary  of  these  findings  is  presented  in 

spleen  is  involved  also,  showing  enlarged  table  7. 

germinal  centers,  and  clusters  of  plasma  In  summary  the  results  suggest  that: 

cells  in  both  red  and  white  pulp.  1.  Tolerance  to  skin  homografts  in  the 

Before  analyzing  the  results  it  might  be  CBA— >C3H  strain  combination  of  mice  is 

well  to  point  out  that  others  have  shown,  not  permanent. 

and  these  observations  have  confirmed,  2.  A  host-versus-graft  reaction  against 

that  skin  homografts  elicit  an  immune  the  skin  homograft  precedes  that  against 

response  limited  to  the  regional  lymph  the    homologous    lymphoid    cells.    This 

node,  whereas  the  injection  of  homologous  host-versus-graft  reaction  is  evident  in 


TABLE  7.     Summary  of  Histological  and  Chimera  Analysis  in  Tolerant  Mice  of  Different 

Stages  and  in  Control  Mice 

State  of  Tolerance  Control  Graft 


Partial  Homologous 

Full  (contracting)      1st  Graft    2nd  Graft   None     Iso-     Homo-          Cells 

Chimera  test              +  +                     +                 — 
Immune  response 

Regional  lymph 

node                    +  +                     +                 +—          —           +                 + 

Spleen                     —  —                      —                 _|__          _            _                 _|_ 


cells  is  followed  by  an  immune  response  the    regional    lymph    node    before    any 

on  the  part  of  both  the  spleen  and  lymph  external  signs  of  graft  rejection, 

nodes.   Although  these  studies  are  not  3.  After  rejection  of  the  graft  a  state 

complete,   the   results   suggest   that,   in  of  "restricted"  tolerance  sets  in,  during 

tolerant  mice,  a  host-versus-graft  reaction,  which    donor    lymphoid    cells    are    still 

which  is  histologically  demonstrable,  pre-  present. 

cedes  external  signs  of  graft  rejection.  4.  In  the  final  stage  of  the  delayed  loss 

This  host-versus-graft  reaction  appears  of  tolerance,  the  donor  lymphoid  cells  are 

limited    to    the    regional    lymph    node,  rejected    with    a    concomitant    immune 

Chimera  analysis  of  lymphoid  tissues  at  response  in  host  spleen  and  lymph  nodes, 

this  time  does  show  the  presence  of  donor  To    determine    whether    removal    of 

cells.  The  immunological  reaction  appears  lymphoid  cells  from  a  tolerant  animal 

to  be  directed  primarily  against  the  donor  affects   the   animal's   tolerant   behavior, 

skin   graft   and   not   against   the   donor  newborn  C3H  mice  were  again  injected 

lymph   cells.    During   graft   contraction,  with   5  million  spleen  cells  from  adult 

and     shortly    thereafter,     the    immune  female  CBA  donor  mice.  At  2  months  of 

response  is  still  limited  to  the  regional  age   they  were   test-grafted   with    CBA 

lymph  node.  It  is  of  interest  that  prelim-  strain  skin,  and  all  were  tolerant  for  at 

inary   evidence   suggests   that   at   these  least  2  months.  At  this  time  the  lymphoid 


DEPARTMENT    OF    EMBRYOLOGY 


389 


tissues  (bone  marrow,  lymph  nodes,  and 
spleen)  from  the  tolerant  mice  were 
removed  and  transferred  to  lethally 
irradiated  C3H  and  CBA  mice.  The  bone 
marrow  was  injected  intravenously  and 
the  lymph  node  and  spleen  cells  intra- 
peritoneally.  Three  weeks  later  the  irradi- 
ated recipient  mice  were  test-grafted  with 
C3H,  CBA,  and  C57BL/6  skin.  Two 
months  later  they  were  killed  and  their 
lymphoid  tissues  were  analyzed  for  the 
presence  of  C3H  and  CBA  cells  by  the 
"chimera"  test.  Control  irradiated  mice 
were  injected  with  equivalent  amounts  of 
isologous  lymphoid  cells  and  similarly 
test-grafted.  A  second  control  group  was 
irradiated  but  received  no  further  treat- 
ment. 

Table  8  contains  the  results  of  this 
experiment.  The  acceptance  of  CBA 
strain  skin  grafts  of  the  X -irradiated  CBA 
and  C3H  recipients  of  " tolerant"  tissues 
indicates  that  tolerant  tissues  (here,  those 
of  C3H  origin)  remain  tolerant  in  a 
different  environment.  The  acceptance  of 
C3H  strain  skin  grafts  by  the  irradiated 
recipients  of  tolerant  tissues  suggests  that 
the  original  donor  cells  (those  of  CBA 
origin),  which  were  injected  into  the 
newborn  C3H  mice  as  adult  cells,  have 
become  tolerant  of  the  host  type  (C3H) 
antigens.  Here,  then,  is  an  example  of 
host-versus-graft  as  well  as  graft-versus- 


host  tolerance.  The  persistence  of  this 
graft-versus-host  tolerance  in  the  irradi- 
ated CBA  recipients  suggests,  further- 
more, that  the  graft-versus-host  tolerance 
is  not  the  result  of  an  immunological 
paralysis.  If  the  CBA  donor  cells,  upon 
injection  into  the  newborn  C3H  host,  had 
been  overwhelmed  and  paralyzed  by  the 
large  number  of  host-type  antigens 
present,  they  would  have  recovered  their 
immunological  reactivity  to  C3H  type 
antigens  upon  transfer  to  the  irradiated 
CBA  recipients,  where  the  number  of 
C3H  type  antigens  was  greatly  reduced. 
That  no  such  recovery  of  the  CBA  cells 
occurred,  and  that  they  continued  to 
tolerate  C3H  skin  grafts,  suggest  that  a 
"true"  tolerance  was  induced  in  these 
adult  CBA  donor  cells.  The  rejection  of 
C57BL/6  skin  grafts  by  the  irradiated 
recipients  of  "tolerant"  tissues  indicates 
that  the  acceptance  of  CBA  and  C3H  skin 
grafts  by  these  animals  is  not  due  to  a 
general  immunological  impairment  but 
to  a  specific  inhibition  of  an  immune 
response. 

Upon  chimera  analysis,  it  was  found 
(table  9)  that  5  out  of  7  irradiated  C3H 
recipients  of  tolerant  tissues  harbored 
CBA  cells,  confirming  that  both  the 
original  host  (C3H)  and  donor  (CBA) 
cells  were  transferred  from  the  tolerant 
to  the  irradiated  recipient  mouse.   Chi- 


TABLE  8.     Fate  of  Skin  Grafts  in  Irradiated  CBA  and  C3H  Recipient  Mice  Injected  with 
Lymphoid  Cells  from  Normal  and  Tolerant  Donors 


Type  of 

Lymphoid 

Cells 

No. 
Mice 

Early 
Mortal- 
ity 

Fate  of  Skin  Grafts 

XR 

Recipient 

Donor  CBA 
No.            Rej. 

Donor  C3H 

Donor  C57BL/6 

Strain 

No. 

Rej. 

No. 

Rej. 

Tested 

(MST) 

Tested 

(MST) 

Tested 

(MST) 

C3H 

Tolerant 

12 

3 

9 

1  (12) 

8* 

0 

8 

8  (12±0) 

Isologous 

10 

2 

8 

8  (15±) 

3f 

0 

4 

4  (12±0) 

None 

9 

8 

• — ■ 

— 

— 

CBA 

Tolerant 

10 

1 

9 

0 

9 

0 

9 

8  (12  ±0) 

Isologous 

10 

0 

9 

0 

10J 

10  (15±1) 

9 

9  (12±0) 

None 

5 

4 

— 

— 

— 

MST  =  mean  survival  time. 

*  One  mouse  died  after  rejection  of  the  CBA  graft. 

t  Only  3  out  of  the  8  mice  in  this  group  were  tested  with  isografts. 

t  One  mouse  died  after  rejection  of  the  C3H  graft. 


390 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


TABLE  9.     Chimera  Analysis  of  C3H  and  CBA  X-Irradiated  Recipients  of  "Tolerant" 

Lymphoid  Tissues 


X-Irradiated  Recipients 

C3H 

CBA 

No.  Cells 

No.  Cells 

Individual 

Transferred 

Chimera 

Individual 

Transferred 

Chimera 

Mouse  No. 

in  Analysis, 
X106 

Mouse  No. 

in  Analysis, 
X106 

1 

150 

Yes 

1 

130 

No 

2 

175 

No 

2 

125 

No 

3 

220 

Yes 

3 

150 

No 

4 

220 

Yes 

4 

220 

No 

5 

150 

Yes 

5 

200 

No 

6 

175 

No 

6 

250 

Yes 

7 

190 

Yes 

mera  analysis  of  the  irradiated  CBA 
recipients  of  "tolerant"  tissue,  however, 
failed  in  all  but  one  of  them  to  reveal  the 
presence  of  C3H  cells.  Even  though  the 
assay  sensitivity  of  C3H  cells  in  CBA  mice 
is  lower  than  in  the  reversed  direction  (in 
the  order  of  5  million  cells ;  table  6)  these 
results  are  nevertheless  surprising  since 
the  irradiated  CBA  mice  were  injected 
with  lymphoid  cells  from  tolerant  C3H 
mice.  Whether  there  is  a  preferential 
proliferation  of  CBA  cells  in  these 
irradiated  CBA  mice  is  an  open  question 
for  the  moment. 

Immunologically  Induced 
A  spermatogenesis 

D.  W.  Bishop's  previous  studies  of  the 
immunologically  induced  aspermatogenic 
response  in  guinea  pigs  (see  Year  Book  60, 
pp.  412-416)  focused  attention  on  two 
problems  relating  to  mechanisms  of  the 
reaction  that  are  being  further  explored 
by  Bishop  in  collaboration  with  Maurice 
Lessof,  Guy's  Hospital,  London.  The  first 
concerns  the  significance  and  possible 
causal  relation  between  the  presence  of 
circulating  nonprecipitable  antibodies  and 
the  appearance  of  lesions  in  the  germinal 
epithelium  after  injection  of  testicular 
antigen  combined  with  adjuvant.  Al- 
though such  a  causal  relation  has  gen- 
erally been  regarded  as  relatively  heretical 


by  classical  immunologists  concerned 
with  responses  of  the  delayed  hypersensi- 
tivity type,  the  findings  of  Bishop  and 
co-workers  have  repeatedly  shown  a 
significant  correlation  and  suggest  that 
humoral  factors  may  play  an  important 
role.  Recent  support  for  this  view  has 
been  suggested  by  certain  studies  of 
immune  induced  thyroid  disease  and  skin 
homograft  reactions.  Lessof  is  attempting 
to  determine  antibody  titer  at  all  stages 
following  sensitization,  with  particular 
interest  in  events  when  testicular  injury 
first  becomes  apparent.  For  antibody 
assay  he  is  utilizing  extracted  testicular 
antigen  and  serum  standards  prepared  by 
Bishop  with  the  aid  of  G.  L.  Carlson.  No 
results  have  yet  been  reported  by  Lessof, 
who  initiated  this  phase  of  the  investi- 
gation only  recently. 

The  unusual  susceptibility  of  the 
guinea  pig  germinal  epithelium  to  auto- 
immune lesions  has  been  under  study  by 
Lessof  in  an  effort  to  determine  what 
factors  or  physiological  peculiarities  ren- 
der this  system  responsive,  and  whether 
physiological  alterations  in  the  testis 
might  so  change  the  reactivity  as  to  shed 
light  on  mechanisms  of  the  cellular 
response.  Although  many  difficulties  in 
technique  and  interpretation  beset  such 
an  approach,  Lessof  has  evolved  methods 
involving  unilateral  treatment  of  the 
testis  with  low  temperature  or  with  X 


DEPARTMENT    OF   EMBRYOLOGY  391 

irradiation,  the  effects  of  which  can  be  The  possibility  also  exists  that  the  mother 
distinguished  from  autosensitization,  in  is  accidentally  sensitized  at  the  time  of 
an  attempt  to  modify  the  onset  of  fetal  injection  and  her  antibodies  are 
immune  induced  aspermatogenesis.  Thus  transferred  to  the  fetuses  in  utero  or  to 
far,  cold  shock  has  proved  ineffective,  but  the  newborn  during  lactation.  According 
irradiation  offers  some  promise.  The  to  studies  by  Uhr,  the  former  possibility 
effect  of  pharmacodynamic  substances,  is  remote,  since  he  was  unsuccessful  in 
like  hyaluronidase,  histamine,  and  other  passively  sensitizing  guinea  pig  fetuses  by 
agents  that  increase  permeability,  is  injection  of  the  mother  with  diphtheria 
being  scrutinized.  toxoid  or  ovalbumin.  As  pregnancies 
In  last  year's  report  it  was  noted  that  become  available,  the  significance  of 
studies  on  the  immune  capacity  of  fetal  lactation  as  an  anti-BGG  transfer  mech- 
guinea  pigs  were  contemplated  with  the  anism  in  the  guinea  pig  is  being  investi- 
aim  of  testing  the  effect  of  injection,  in  gated.  Although  fetal  sensitizations  with 
utero,  of  soluble  antigen  and  of  the  testicular  antigen  have  lagged  behind, 
aspermatogenic  factor  when  it  can  be  Bishop  has,  thus  far,  been  unable  to 
obtained  in  a  high  degree  of  purity.  The  confirm  Chutna  and  Haskova's  finding 
operative  procedure  for  this  approach  has  that  the  guinea  pig  can  acquire  tolerance 
now  been  refined  so  that  fetal  loss  is  or  be  desensitized  to  the  aspermatogenic 
minimized.  Intraperitoneal  injection  with  reaction.  The  discrepancy  may  lie  in 
bovine  gamma  globulin  (BGG)  of  45-  to  their  use  of  a  different  and  less  responsive 
66-day-old  fetuses  results  in  positive  (aspermatogenetically)  strain  of  guinea 
serological  reactivity  when  determined  on  pig  or  in  the  employment  of  massive 
sera  collected  at  3  and  8  weeks  post-  dosages  of  testicular  antigen  to  "para- 
partum.  Circulating  factors  are  demon-  lyze"  the  immune  reaction, 
strable  by  the  passive  cutaneous  ana-  Previously  reported  were  experiments 
phylactic  (PCA)  reaction  but  are  negative  by  Bishop  and  Gump  demonstrating  that 
by  the  gel-diffusion  technique,  indicating  newborn  guinea  pigs  injected  intramuscu- 
the  presence  of  nonprecipitable  or  of  larly  with  BGG  and  adjuvant  give  rise  to 
low-titer  precipitable  antibodies  in  serum  circulating  antibodies  within  7  to  9  days, 
of  newborn  animals.  Unfortunately,  the  Two  further  simple  experiments  have 
precise  fetal  origin  of  these  antibodies  is  demonstrated  (1)  the  immunologic  com- 
clouded  somewhat  by  positive  PCA  petence  of  newborn  animals  injected 
reactions  in  maternal  serum  collected  intraperitoneally  with  antigen  alone  and 
after  delivery.  More  precision  in  the  (2)  the  relative  efficacy  of  intramuscular, 
procedure  is  necessary  to  determine  intraperitoneal,  and  intravenous  injection 
whether  the  pregnant  animal  also  gives  of  adults.  The  former  data  are  sum- 
rise  to  specific  antibodies  as  a  result  of  marized  in  table  10;  of  17  neonatal 
uterine  transfer  of  BGG  or  whether  she  animals  injected  intraperitoneally  with  20 
is  passively  sensitized  by  fetal  antibody,  mg  BGG  in  saline  within  24  hours  of 


TABLE  10.     Immunologic  Response  of  Newborn  Guinea  Pigs  to  a  Single  Intraperitoneal 

Injection  of  BGG 


^                           Injection 

Bleeding 

PCA  Reactions* 

Ouchterlony 
Reactionf 

Age     Route       Antigen 

Age     Route     -f-  + 

+++     +  +  +  + 
3               11 

17       18-24  hr      IP      20  mg  BGG 

21  days     IC        3 

Negative 

*  PCA  sensitization:  0.1  ml  antiserum.  PCA  challenge:  10  mg  BGG.  PCA  scoring:  ++   =  11-15 
mm,  +  +  +  =  16-20  mm,  +  +  +  +  =   >20  mm. 
t  Ouchterlony  plates  read  daily  to  16  days. 


392 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


delivery,  all  showed  pronounced  PCA 
reactions  in  sera  collected  at  21  days  of 
age.  Ouchterlony  tests,  on  the  other  hand, 
were  consistently  negative. 

In  a  comparison  of  circulating  antibody 
titers  according  to  the  route  of  admin- 
istration, it  was  found  that  antigen  plus 
adjuvant  injected  intramuscularly  into 
the  guinea  pig  gave  a  higher  and  more 
persistent  titer  than  antigen  admin- 
istered alone  either  intravenously  or 
intraperitoneally  (table  11).  Adjuvant 
potentiates    the    reaction,    and    is    the 


The  results  of  gel-diffusion  tests  of  these 
sera,  also  shown  in  table  11,  roughly 
follow  those  obtained  by  the  PCA  pro- 
cedure. Double  bands  appeared  in  the 
Ouchterlony  plates  in  1  serum  sample 
from  the  intraperitoneal  series  and  in  4 
out  of  5  from  the  intramuscular  series.  In 
a  comparison  of  the  two  test  methods,  the 
individual  data  show  clearly  that  the 
PCA  procedure  is  more  sensitive  and  is 
capable  of  detecting  precipitable  antibody 
in  titers  too  low  to  be  demonstrated  by 
the  gel-diffusion  method.  That  the  appar- 


TABLE  11.     Immunologic  Response  of  Adult  Guinea  Pigs  to  BGG  Administered  Intravenously, 
Intraperitoneally,  and  Intramuscularly  with  Adjuvant* 


Bleeding  after 

PCA  Reactions! 

Ouchterlony  Tests 

No.  Animals 

Last  Injection, 

days 

IV 

IP 

IM 

IV 

IP              IM 

5 

5 

18.4 

20.4 

21.2 

1/5 

3/5              4/5 

5 

9 

22.0 

24.6 

25.2 

1/5 

3/5              4/5 

5 

14 

21.4 

25.6 

30.4 

3/5 

4/5              5/5 

5 

30 

11.6 

25.6 

35.8 

3/5 

2/5              5/5 

5 

42 

12.0 

25.0 

36.6 

0/5 

2/5              5/5 

*  Each  animal  received  100  mg  BGG,  in  5  injections  in  IV  and  IP  series,  and  in  2  injections  in  IM 
series. 


f  Values  represent  averages  (in  mm)  for  5  sera. 


method  of  choice.  In  the  intravenous 
series,  antibody  titer  increased  to  a 
maximum  at  about  10  days  and  then 
receded  abruptly.  The  antibody  titer  of 
the  intraperitoneal  series  increased  for 
about  2  weeks  and  leveled  off  at  twice 
the  intravenous  value,  whereas  the  titer 
of  the  intramuscular  series  continued  to 
increase  for  6  weeks.  In  this  experiment, 
done  in  collaboration  with  Timothy 
Glover,  each  animal  received  100  mg  of 
BGG,  distributed  in  5  injections  over  a 
period  of  9  days  in  the  intravenous  and 
intraperitoneal  series  and  in  2  injections 
with  adjuvant  in  the  intramuscular  series. 
All  animals  were  bled  5,  9,  14,  30,  and  42 
days  after  the  last  injection.  PCA  tests 
were  carried  out  according  to  the  pro- 
cedure developed  by  Ovary,  and  the 
tabulated  results  indicate  the  average 
reaction  diameter  for  5  undiluted  sera. 


ent  sensitivity  of  the  PCA  test  here  lies  in 
its  ability  to  detect  nonprecipitable 
humoral  antibody  is  unlikely. 

Nature  of  the  Material  Inducing 
Aspermatogenesis  in  the  Guinea  Pig 

In  collaboration  with  D.  W.  Bishop, 
Gerald  L.  Carlson  has  continued  to 
investigate  the  chemical  nature  of  the 
fraction  purified  from  guinea  pig  testes 
which  when  injected  together  with  adju- 
vant produces  aspermatogenesis  in  the 
guinea  pig.  As  was  mentioned  in  Year 
Book  60,  pages  412-414,  Freund,  Thomp- 
son, and  Lipton  had  purified  an  asper- 
matogenic  fraction  designated  CPM 
(chlorofo  m-purified  material)  from 
guinea  pig  testes  and  characterized  it  in 
a  preliminary  way.  A  failure  of  chemical 
analysis  to  account,  with  reasonable 
assumptions,  for  more  than  35  to  40  per 


DEPARTMENT  OF  EMBRYOLOGY 


393 


cent  of  the  total  mass  determined  in 
dry-weight  measurements  by  Freund  et 
al.  was  possibly  due  to  incomplete  drying 
of  CPM  samples.  Dry-weight  determina- 
tions on  CPM  samples  used  in  the 
present  studies  (table  12)  indicate  that 
carbohydrate,  measured  as  reducing 
power  for  cupric  ion  and  expressed  as 
galactose,  accounts  for  about  20  per  cent, 
and  polypeptide  (s),  determined  by  either 
the  Lowry  or  Biuret  method  with  bovine 
serum  albumin  as  a  standard,  for  about 
70  per  cent,  of  the  total  mass.  Thus,  only 
about  10  per  cent  of  the  CPM  remains  to 
be  assigned  to  a  general  chemical  classifi- 
cation. 

The  amino  acids  composing  the  poly- 
peptide part  of  CPM  were  examined  by 
methods  giving  greater  resolution  than 
techniques  previously  used.  Samples  of 
CPM  were  oxidized  with  performic  acid 


ionophoresis  are  histidine,  lysine,  argi- 
nine,  glutamic  acid,  aspartic  acid,  cystine, 
or  cysteine,  phenylalanine,  tyrosine,  pro- 
line, threonine,  alanine,  and  serine. 
Tryptophan  is  probably  present  as  judged 
from  absorption  by  CPM  at  278  mn 
remaining  at  pR  13.  One  unidentified 
ninhydrin-reactive  compound,  present  in 
quantity  in  all  CPM  hydroly sates,  is 
under  further  study. 

Antibodies  as  Tools 

An  Approach  to  an  Immunochemical 
Study  of  Neuronal  Differentiation 

Having  only  six  months  to  spend  in  the 
Department,  Arthur  LaVelle  elected  to 
center  his  attention  on  mastering  immu- 
nochemical techniques  and  adapting  them 
for  the  study  of  developing  nerve  cells. 


TABLE  12.     Analysis  of  CPM  Samples 


Type  of  Measurement 


Milligrams  per  Milliliter  of  CPM 


Preparation  A 


Preparation  B 


Dry  weight 

Galactose  equivalent* 

Bovine  serum  albumin  equivalent  (biuret) 

Bovine  serum  albumin  equivalent  (Lowry) 


3.20  ±  0.01 
0.60  db  0.02 
2.22  ±  0.04 
2.25  ±  0.07 


3.10  ±0.05 
0.625  ±  0.01 
1.83  d=0.13 
1.98    d=0.06 


*  Reducing  material  determined  after  hydrolysis  with  0.37  N  H2S04  in  sealed  tube  at  95-100c 
16  hours  and  expressed  as  galactose. 


for 


to  give  stable  derivatives  of  sulfur- 
containing  amino  acids,  and  the  oxidized 
CPM  samples  were  then  hydrolyzed  with 
6  N  HC1  in  sealed  tubes.  The  hydroly- 
sates  of  performic-oxidized  CPM  were 
used  in  two-dimensional  ionophoretic 
separations  which  followed  the  procedures 
of  M.  A.  Naughton  (the  ionophoresis 
study  was  made  in  the  Biophysics 
Department  of  Johns  Hopkins  Medical 
School  with  his  aid) .  With  these  methods 
the  presence  of  eighteen  amino  acids  has 
been  shown  in  CPM  samples,  and  a 
quantitative  analysis  of  the  amino  acids 
is  in  progress.  Amino  acids  in  CPM  that 
have  been  identified  so  far  by  position  in 


The  long-range  purpose  of  these  pre- 
liminary attempts  is  to  endeavor  to 
produce,  as  one  extension  of  his  current 
research  in  neuronal  differentiation,  a 
specific  antiserum  to  neuronal  tissue 
(neuroplasm).  Therefore  LaVelle  began  to 
develop  techniques  for  the  production  of 
antibrain  serum  in  several  different 
strains  of  laboratory  mice. 

Because  of  the  ease  of  dissecting 
cellular  areas  from  frozen-dried  sections 
of  nervous  tissue  (an  advantage  that 
could  be  of  later  value)  it  was  decided 
first  to  assess  the  antibody-producing 
potential  of  frozen-dried  whole  brain. 
Individuals  of  several  series  of  C57  and 


394 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


BDFiJ  mice  (Jackson  Memorial  Labora- 
tory), 4  to  6  months  old,  were  injected 
each  week  for  6  to  8  weeks.  At  each 
injection  each  animal  received  0.8  mg 
(dry  weight)  of  saline-perfused  frozen- 
dried  brain  combined  with  saline  and 
Freund's  adjuvant  (Difco  "complete"). 
This  mixture  was  injected  as  a  0.2-ml 
emulsion  into  multiple  sites  either  intra- 
dermally  or  subcutaneously,  depending 
on  the  series.  At  intervals,  and  at  1  week 
after  the  final  injection,  blood  was  drawn 
from  each  animal  by  cardiac  puncture, 
and  the  separated  serum  was  stored  in 
the  deep  freezer  at  —  40°C  for  future 
testing.  No  physiological  signs  of  experi- 
mental encephalomyelitis  (EAE)  were 
observed  at  any  time.  Also,  thus  far, 
titers  of  the  "antiserum"  (serum  dilu- 
tions) against  saline  suspensions  of 
frozen-dried  brain  have  been  negative. 
Preliminary  tests  of  frozen-dried  brain 
suspensions  against  the  serum  from  the 
experimental  animals  using  micro  agar- 
diffusion  techniques  failed  to  show  any 
precipitation  lines.  As  this  report  was 
being  written,  saline  suspensions  of  fresh 
whole  brain  and  frozen-dried  brain  were 
being  run  against  serum  from  the 
experimental  animals  on  macro-Ouchter- 
lony  plates  in  an  effort  to  obtain  precipi- 
tation lines.  Further  tests  with  the 
experimental  serum  will  be  run,  but  they 
will  be  undertaken  relative  to  experiments 
using  fresh  and  frozen-dried  brain  now 
being  injected  into  other  animals.  Full 
testing  of  this  serum  will  not  be  finished 
for  several  months,  in  conjunction  with 
correlative  experiments  now  in  progress. 

Among  several  reasons  for  the  con- 
sistent negative  results  so  far  attained, 
two,  in  particular,  seem  pertinent.  One  is 
that  frozen- dried  tissue,  aside  from  brain 
itself,  is  in  some  way  altered  so  that  it  is 
a  poor  antibody  stimulus.  LaVelle  sug- 
gests that  a  mechanism  similar  to  the 
"enhancement"  effect  obtained  in  homo- 
graft  experiments  may  play  a  role ;  in  this 
type  of  reaction  frozen-dried  spleen  is 
particularly  effective. 

The  other  explanation  is  suggested  in  a 


report  by  Lee  and  Schneider  that 
appeared  while  these  experiments  were  in 
progress,  stating  that  a  critical  relation 
between  the  constituents  of  the  antigen- 
adjuvant  emulsion  is  necessary  to  produce 
EAE  in  completely  susceptible  mice. 
These  authors  also  report  that  the  efforts 
of  some  others  using  the  Difco  "com- 
plete" adjuvant  in  combination  with 
proteolipides  failed  to  produce  EAE  in 
susceptible  mice. 

We  might  add  that  the  high  content  of 
lipide  in  whole  brain  makes  it  particularly 
difficult  to  prepare  satisfactory  suspen- 
sions for  precipitin  tests.  Apparently  the 
lipide  also  interferes  with  free  diffusion  on 
agar  plates.  As  yet,  LaVelle  has  not 
tested  lipide-extracted  residue  or  the 
extracted  total  lipide  against  the  experi- 
mental serum.  The  proteolipide,  prepared 
from  hamster  brain  according  to  the 
method  of  Folch  and  Lees,  has  not  yet 
been  tested  either.  Experiments  will  be 
run  to  take  into  consideration  these  and 
other  factors  that  need  not  be  detailed 
here. 

Obviously  many  factors  must  be  con- 
sidered before  a  satisfactory  resolution  to 
even  this  modest  beginning  of  the  main 
problem  can  be  obtained. 

A  New  Antigenic  System  in  the 
Chick  Embryo 

The  following  report  of  R.  F.  Ruth's 
activities  during  1960-1961  was  sub- 
mitted too  late  to  be  included  in  Year 
Book  60. 

With  the  assistance  of  F.  J.  Kupres, 
Ruth  confirmed  and  expanded  his  earlier 
finding  of  a  new  antigenic  system  in  the 
chick  embryo.  The  salient  feature  of  this 
system  is  its  association  with  the  periph- 
ery of  the  blastoderm.  It  is  at  the 
periphery  that  the  blastoderm  attaches  to 
the  vitelline  membrane,  and  the  effects 
of  antisera  were  first  observed  by  the 
separation  of  the  blastoderm  from  the 
vitelline  membrane.  Closer  examination 
of  this  phenomenon  indicates  that  the 
peripheral  attachment  is  immunologically 
related  to  the  erythrocyte. 


DEPARTMENT    OF    EMBRYOLOGY  395 

The  detachment  of  the  blastoderm  was  that   detachment   is   due   to   antibodies 

discovered    by    treatment    of    cultured  evoked  specifically  in  rabbits  by  injection 

embryos    with    antichicken    erythrocyte  of  chicken  erythrocytes, 
serum.   More  antierythrocyte  sera  were         This   demonstration   of   the   antigenic 

prepared  in  rabbits.  All  such  sera,  when  relationship  between  the  erythrocyte  and 

used    fresh    or    after    one    freezing    and  the  peripheral  attachment  of  the  blasto- 

thawing,     detached    all    embryos    with  derm  to   the  vitelline  membrane   poses 

which  they  were  in  contact  between  the  several  questions.  When  do  the  antigens 

twentieth    and     sixty-eighth    hours    of  appear  in  ontogeny?  Are  they  genetically 

development.    This    was    true    without  controlled?  Is  the  detachment  due  to  the 

exception   of   more   than   a   dozen   sera  presence  of  the  antigens  on  the  vitelline 

tested  with  more  than  100  embryos.  Some  membrane  or  on  the  peripheral  cells?  Are 

antisera    were    fractionated,     and     the  the   antigens   involved   peculiar   to   the 

globulin-rich  and  globulin-poor  fractions  erythrocyte  and  to  the  peripheral  attach- 

were  used  to  treat  embryos.  The  globulin-  ment? 

rich  fractions  detached  the  embryos;  the         The  erythrocytes  used  to  produce  sera 

globulin-poor   fractions    did    not.    Some  were   obtained   from   groups   of  mature 

antisera  were  absorbed  exhaustively  with  chickens.  Detachment  of  treated  embryos 

chicken  erythrocytes.  Despite  difficulties  occurred    during    the    first    3    days    of 

associated  with  the  aging  of  the  sera  and  incubation.  The  effective  antigens  thus 

lysis   of  many  erythrocytes  during  the  appear  early  in  development  and  persist 

absorptions,  these  sera  were  only  slightly  into  maturity.  This  observation  suggested 

toxic  to  embryos.  They  did  not  detach  that  the  antigens  involved  might  be  of 

embryos.  the  classical  Forssman  type.  Since  guinea 

The  detachment  of  embryos  by  anti-  pigs     cannot     produce     antibodies     to 

erythrocyte  sera  and  globulin-rich  frac-  Forssman  antigens,  antierythrocyte  sera 

tions  of  these  sera,  and  the  inability  of  prepared  in  guinea  pigs  were  used  to  treat 

absorbed  sera  and  globulin-poor  fractions  embryos.    These    antisera    detached    all 

to  detach  embryos,  suggested  that  detach-  embryos,    indicating   that   the   effective 

ment  is  effected  by  globulins  that  can  antibodies  are  not  peculiar  to  the  rabbit 

attach  irreversibly  to  erythrocytes.  There  and   that   the   antigens   are   not   of  the 

remained  the  possibility  that  these  globu-  classical  Forssman  type, 
lins   might  be   directed   toward   plasma         Presumably,  the  antigens  involved  in 

contaminants  present  in  the  erythrocytes  detachment    are    genetically    controlled, 

used  for  injection  and  absorption,  despite  Genetic    control    could    not    have    been 

repeated  washings.  detected  in  the  experiments   described, 

Antisera    were    prepared    in    rabbits  since   they  were   designed   to   eliminate 

against  cell-free  chicken  plasma.  The  sera  intraspecies  variations  in  the  production 

obtained  after  one  series  of  injections  did  of   antibodies   and   in   the   response    of 

not  detach  embryos.  This  finding  indi-  embryos  to  treatment.  The  possibility  of 

cated  that  the  ability  to  detach  was  not  genetic  control  was  tested  in  an  inde- 

due  to  contamination  with  any  of  the  pendent  experiment.  Embryos  known  to 

principal  constituents  of  plasma.  Rabbit  be  of  A!AX  genotype  were  treated  with 

antierythrocyte  sera,  which  did  detach,  chicken  anti- A1  erythrocyte  sera,  chicken 

were  incubated  with  equal  volumes  of  anti- A5  erythrocyte  sera,  and  normal  sera, 

chicken    plasma    without    altering    their  Equal  sets  of  A5A5  embryos  were  treated 

ability    to    detach    as    compared    with  with   the   same   reagents.    Since   the   A 

appropriately  diluted  and  incubated  con-  antigens     are     present     very     early     in 

trols.  Various  absorptions  and  technical  development  the  A  alleles  seem  to  provide 

controls  were  employed,  and  these  results  the  best  possible  genetic  test.  The  results 

also  were  consistent  with  the  hypothesis  of  the  experiment  were  negative,  but  a 


396  CARNEGIE     INSTITUTION     OF     WASHINGTON 

consideration    of    the    dilutions    of    the  occasion.    When    fresh    sera    are    used 

reagents  used  suggests  that  the  conditions  individually   it   is    a    simple    matter   to 

for  the  test  were  not  the  best  possible,  obtain  healthy,  vigorous  embryos  which 

The  embryos  and  reagents  used  in  the  continue     development     while     floating 

test  were  provided  by  Elwood  Briles  of  freely  in  serum.  It  is  not  necessary  to 

the  DeKalb  Agricultural  Research  Asso-  dilute  the  serum  in  order  to  demonstrate 

ciation.  this,   but  dilution  might  raise  the  fre- 

The  possibility  that  detachment  is  due  quency    of    good    development    among 

to  a  combination  of  antibodies  with  the  detached  embryos. 

vitelline  membrane  was  tested  in  several  The  contraction  of  the  freely  floating 
ways.  Finely  divided  vitelline  membrane  blastoderm  substantiates  New's  interpre- 
was  incubated  with  antierythrocyte  sera,  tation  of  the  role  of  the  periphery  and 
Agglutination  could  not  be  detected  the  vitelline  membrane  in  maintaining 
macroscopically  or  microscopically.  Anti-  tension  across  the  blastoderm.  The  f re- 
sera  incubated  with  divided  vitelline  quency  with  which  the  embryo  itself 
membrane  retained  their  ability  to  detach  continued  development  and  growth,  even 
embryos.  These  observations  fail  to  though  it  had  to  push  the  contracted 
support  the  theory  that  the  antibodies  blastoderm  out  of  the  way,  affirms  the 
effect  detachment  by  coating  the  vitelline  basic  independence  of  embryonic  develop- 
membrane.  These  negative  experiments  ment  and  the  spreading  of  the  blastoderm, 
cannot  be  considered  conclusive,  however.  Aside  from  the  birds  only  one  major 
In  fact,  the  only  pertinent  positive  results  group  of  vertebrates,  the  teleosts,  displays 
are  consistent  with  the  idea  that  specific  an  analogous  spread  of  the  blastoderm  by 
coating  of  the  vitelline  membrane  may  means  of  a  peripheral  attachment, 
occur. 

Vitelline  membranes  were  prepared  in  Composition  of  the  Cell  Surface 
the    usual    fashion.    The    embryo    was 

removed  from  each  by  fine  needles  and  a  There  seems  good  reason  to  hope  that 

pipette ;   the   vitelline   membranes   were  immunological  methods  may  be  helpful  in 

incubated    with    normal    sera    or    anti-  advancing  knowledge  of  the  composition 

erythrocyte   sera  for   2  hours  at  37°C,  of  the  cell  surface  and  of  the  nature  of 

washed   repeatedly  with   saline,   reincu-  contact-dependent      reactions      between 

bated  with  saline  at  37°C  overnight,  and  cells. 

re  washed ;  and  fresh  embryos  were  placed  The  technique  suggested  by  Michael 

on  the  membranes.  The  embryos  placed  Abercrombie  is  to  apply  antisera,  made 

on  membranes  treated  with  normal  sera  by  injecting  rabbits  with  whole  cells  or 

attached,  and  those  placed  on  membranes  cell  fractions,  to  tissue  cultures;  and  to 

treated  with  antisera  did  not.  measure  the  effects  of  the  treatment  on 

The  periphery  of  a  detached  blastoderm  some    form    of    cell    behavior    likely    to 

may    retain    some    of    its    specialized,  depend  on  the  nature  of  the  cell  surface, 

flattened  characteristics,  but  the  beauti-  Mutual  inhibition  of  cell  movement  by 

ful  sinuous  extremity  of  the  periphery  is  contact  is  such  a  form  of  behavior.  The 

never  present.  Typically,  the  peripheral  overlapping  of  cells  in  a  culture  on  a 

cells  are  condensed  into  a  roll  of  cells  some  plane  surface,  which  largely  depends  on 

of  which  retain  normal  nuclear  morphol-  this   inhibition,    is   the   most   promising 

ogy.   Such   "rolls"  have  lost  all  resem-  assay  system,  since  it  can  be  precisely 

blance  to  the  normal  periphery  and  are  quantitated  and  should  be  highly  sensi- 

largely     made     up     of     vacuoles.     The  tive.  By  this  means  it  is  hoped  to  estab- 

remainder  of  the  embryo  is  usually  more  lish  degrees  of  cross  reaction  between  cells 

normal  in  appearance,  though  it  may  lack  of  different  types  and  in  different  states 

hemoglobin  or  be  grossly  distorted  on  and  to  follow  up  the  interesting  beginning 


DEPARTMENT    OF   EMBRYOLOGY  397 

made  recently  by  Kite  and  Merchant  in  information   on   the   mechanism   of   the 

determining  the  chemical  nature  of  the  contact  inhibition  involved.  For  instance, 

surface  antigens.  diminution  of  overlapping  when  the  cell 

The  precise  nature  of  the  influence  of  surface   is   masked   by   antibody  would 

antibodies  on  the  overlapping  of  cultured  make  it  difficult  to  suppose  that  contact 

cells  cannot  be  predicted  from  present  inhibition  depends  on  specific  adhesion 

knowledge  and  should  itself  give  useful  between  cells. 

CELL  INTERACTION   IN   DIFFERENTIATION 
AND   MORPHOGENESIS 

,,                          T.  characterized  by  the  formation  of  long, 

Myogenesis  in  Vitro  u.       ,          .£,      ri         u     ^         ,• 

multinuclear,  ribbonlike  cells.  lormation 

An  investigation  of  the  cytodifferenti-  of  multinuclear  cells  coincides  with  the 
ation  of  embryonic  skeletal  muscle  cells  attainment  of  confluency.  The  involve- 
in  dispersed  cell  culture  was  originally  ment  of  cell  density  is  further  suggested 
undertaken  by  I.  R.  Konigsberg  in  an  by  experiments  in  which  the  inoculum 
attempt  to  define  a  system  that  would  size  was  varied.  When  inoculum  size  is 
offer  greater  opportunity  for  rigorous  varied  the  time  of  transition  from  phase 
control  of  both  the  quantitative  aspects  one  (fibroblastlike  cell)  to  phase  two 
of  the  cellular  population  and  the  extra-  (multinuclear  ribbon)  can  be  shifted.  The 
cellular  environment  than  could  be  abrupt  appearance  of  multinucleated 
achieved  either  in  vivo  or  in  organ  myotubes  is  paralleled  by  an  equally 
culture.  The  cumulative  experience  of  abrupt  break  in  the  rate  of  proliferation, 
numerous  investigators  over  many  years  which  again  can  be  shifted  by  varying  the 
suggested  that  such  culturing  techniques  inoculum  size.  Differentiation  beyond  the 
could  be  expected  to  promote  the  loss  of  stage  represented  by  the  mononucleated 
differentiative  character  rather  than  the  myoblast  occurs  in  culture  after  rapid 
progressive  increase  of  the  morphological  cell  multiplication  has  ceased.  This  is 
consequences  of  cell  specialization.  No  consistent  with  findings  from  several 
generally  satisfactory  explanation  for  this  laboratories,  using  such  diverse  tech- 
observed  incompatibility  was  or  is  avail-  niques  as  microspectrophotometry,  inhi- 
able.  The  results  Konigsberg  has  obtained  bition  of  DNA  synthesis,  microcinepho- 
with  monolayer  cultures  of  embryonic  tography,  and  radioautography,  that  the 
skeletal  muscle  cells  disagree  with  the  myotube  nuclei  are  postmitotic  and  the 
general  observation  of  a  loss  of  morpho-  mechanism  of  their  formation  is  cellular 
logical  indices  of  the  differentiated  state,  fusion.     The     third     phase     of    muscle 

The  monolayer  cultures  prepared  from  differentiation  in  culture,  whose  initiation 

suspensions    of     11-    to     12-day    chick  is  difficult  to  time  exactly,  is  characterized 

embryonic  leg  muscle  pass  through  three  by  the  progressive  development  of  the 

recognizable  phases.  The  period  immedi-  cross- striated    myofibrillar    pattern    and 

ately   following   plating   of   the   cells   is  the  initiation  of  spontaneous  contraction 

marked   by   rapid   proliferation   with    a  (pi.  2,  fig.  16). 

mean  generation  time  of  24  hours.  During  All  Konigsberg's  studies  before  the  past 

this  period  cultures  consist  exclusively  of  year  had  been  restricted  to  monolayer 

mononucleated  cells  and  have  the  general  cultures     established     with     inocula     of 

appearance  of  cultures  of  fibroblastlike  2.5  X  106to  106  cells  each.  Such  cultures 

cells  such  as  might  be  derived  from  a  reach  confluency  between  the  second  and 

great  variety  of  tissues.  The  transition  fourth  day  of  culture,  depending  on  the 

from  phase  one  to  phase  two  occurs  very  size  of  the  inoculum.  To  probe  for  the 

rapidly   (in  a  matter  of  hours)   and  is  lower  limit  of  inoculum  size  that  would 


398  CARNEGIE     INSTITUTION     OF      WASHINGTON 

still  permit  differentiation  to  occur,  1.  What  is  the  significance  of  the 
Konigsberg  turned  to  the  single -cell  finding  that  only  1  in  10  colonies  eventu- 
plating  technique  of  Puck  and  his  asso-  ally  differentiates,  and,  as  a  corollary, 
ciates.  In  this  procedure  small  numbers  of  what  are  the  origin  and  state  of  colonies 
cells  are  dispersed  over  a  relatively  large  that  remain  morphologically  indifferent? 
area.  During  appropriate  periods  of  2.  What  is  the  stimulus  initiating 
incubation  the  individual  cells  give  rise  to  myotube  formation? 
macroscopically  visible,  discrete  colonies.  The  conclusion  that  1  in  10  colony- 
The  technique  has  been  applied  most  forming  centers  is  different  from  the 
successfully  to  permanently  established  others  is  difficult  to  escape.  The  difference 
cell  strains.  Using  freshly  isolated  embry-  may  be  related  to  the  inherent  uncer- 
onic  muscle  cells  Konigsberg  observes  a  tainty  of  the  plating  technique  in 
plating  efficiency  (approximately  10  per  absolutely  ruling  out  cellular  multiplicity 
cent)  considerably  better  than  that  of  all  colony- forming  centers.  Alterna- 
reported  for  freshly  isolated  cells  of  older  tively,  it  may  reflect  the  relative  ratio  of 
animals.  In  plates  cultivated  for  10  to  13  myoblasts  to  fibroblasts  that  survive  the 
days  approximately  1  in  10  colonies  steps  of  the  culturing  technique.  Lastly, 
exhibits  the  unmistakable  indices  of  the  difference  may  be  indicative  of  the 
differentiation  of  skeletal  muscle  cells,  attainment  of  some  property  by  the  cells 
The  proportion  of  differentiated  cells  which  may  or  may  not  be  of  the  type  we 
ranges  from  colonies  containing  several  generally  regard  as  developmentally  sig- 
elongated  myotubes  in  colonies  of  pre-  nificant.  We  must  take  cognizance  of  the 
dominantly  mononucleated  cells  to  colo-  fact,  also,  that  the  final  expression  by 
nies  in  which  virtually  every  nucleus  is  in  which  we  infer  a  preexisting  difference 
syncytial  association.  Examination  of  the  may  actually  be  the  product  of  inter- 
myotubes  under  polarized  light  or  under  action  between  such  a  difference  and  the 
bright-field  illumination  after  staining  culture  environment.  We  must  bear  in 
with  phosphotungstic  acid  hematoxylin  mind  the  possibility  that  alteration  of  the 
reveals  the  presence  of  longitudinal  culture  conditions  might  reduce  or  elim- 
fibrils  which  are  frequently  observed  to  inate  apparent  differences.  To  extend  this 
exhibit  the  typical  pattern  of  cross  thought  it  would  seem  incautious  in  the 
striation  of  mature  skeletal  muscle  cells,  extreme  to  invoke,  at  this  time,  some 
The  sequence  of  events  in  plating  fundamental  limitation  of  the  cell  itself  to 
cultures  is  entirely  comparable,  except  explain  the  observations  made  on  cells 
for  the  temporal  factor,  to  Konigsberg's  in  vitro.  Despite  the  fact  that  many  of 
previous  observations  on  monolayer  cul-  the  operations  judged  to  be  "impossible" 
tures.  Initially,  the  widely  dispersed  cells  a  short  time  ago  are  performed  with 
proliferate  without  exhibiting  any  overt  relative  ease  today,  tissue  culture  is  still 
indications  of  differentiation  by  the  far  from  being  a  perfect  tool  for  all 
criteria  established.  The  colonies  in  purposes,  particularly  when  applied  to 
cultures  observed  at  the  sixth  day  of  the  study  of  cellular  differentiation,  where 
incubation  are  completely  devoid  of  the  component  processes  of  the  phenom- 
myotubes  which  are  present  in  cultures  enon  are  so  imperfectly  defined, 
fixed  at  the  tenth  or  thirteenth  day  of  While  various  approaches  to  the  first 
incubation.  It  is  apparent  that  some  cells,  question  raised  are  under  investigation 
at  least,  can,  through  a  sequence  of  rapid  Konigsberg  has  chosen  to  attack  the 
multiplications,  produce  a  large  number  second  problem  by  examining  the  relation 
of  progeny  that  retain  the  capacity  for  of  cell  density  to  myotube  formation, 
differentiation.  The  two  major  questions  Such  a  correlation  can  be  made  in  mono- 
emerging  from  these  observations,  how-  layer  cultures,  where  the  attainment  of 
ever,  are:  confluency  and  myotube  formation  are 


DEPARTMENT    OF    EMBRYOLOGY  399 

coincident.  In  colony  formation,  also,  the  Whatever    changes   have    occurred    still 

cell    density    seems    to    increase    as    a  exert  their  effects  after  filtration  through 

function  of  the  number  of  cells  per  colony,  bacteriological  filters  (Selas,  Millipore)  or 

the  intercellular  distances  diminishing.  In  after  storage  for  at  least  2  to  3  weeks, 

this  situation,  however,  the  relation  is  Obviously  any  number  of  alterations  may 

even  more  problematical.  have    occurred,    depletions    as    well    as 

Two  general  mechanisms  by  which  cell  additions.  To  discriminate  between 
density  might  affect  myotube  formation  changes  in  the  macromolecular  constitu- 
tive been  considered.  Since  myotube  ents  as  opposed  to  alterations  in  the 
formation  is  a  result  of  cell  fusion,  high  small-molecular-weight  components  the 
cell  density  might  ensure  that  a  sufficient  latter  were  replaced  by  dialyzing  con- 
number  of  effective  cell-to-cell  collisions  ditioned  medium  against  three  changes  of 
occur.  The  possibility  seemed  equally  4  volumes  each  of  freshly  prepared 
likely  that  a  high  cell  density  might  be  medium  over  a  period  of  3  days.  Using 
either  supplementing  the  medium  with  monolayer  cultures  on  a  rotating  turn- 
cell  products  or  removing  some  compo-  table  conditioned  medium  dialyzed 
nents.  Preliminary  tests  were  run  by  against  fresh  medium  produced  results 
culturing  two  coverslips  seeded  with  similar  to  those  observed  with  untreated 
different  numbers  of  cells  in  a  single  petri  conditioned  medium  (pi.  3,  figs.  17,  18). 
plate  bathed  with  the  same  medium.  The  The  limitations  of  tests  conducted  on 
medium  was  circulated  by  incubating  the  mass  cultures,  particularly  in  the  present 
petri  plates  on  a  slowly  rotating  tilted  application,  became  apparent.  One  draw- 
turntable.  It  was  found  that  myotube  back,  only  partly  compensated  for  by 
formation  was  initiated  on  the  sparsely  continuous  rotation  of  the  test  cultures, 
seeded  coverslip,  despite  the  lack  of  is  that  the  changes  for  which  Konigsberg 
confluency,  at  the  same  time  the  process  is  testing  are  continuously  produced  by 
was  initiated  in  the  denser  culture  and  the  test  cells  themselves.  Another  is  the 
before  it  occurred  in  petri  plates  carrying  difficulty  encountered  in  quant itating  the 
two  sparsely  seeded  coverslips.  The  response.  To  circumvent  these  objection- 
obvious  next  step  consisted  in  feeding  able  features  he  turned  to  a  plating 
sparsely   seeded   cultures   with   medium  system. 

withdrawn  from  cultures  that  had  grown  Using  inocula  of  50-400  cells  he  has 

to  confluency.  For  these  tests  the  rotating  found  that  the  conditioned  media  fre- 

turntable  was  also  employed,  to  minimize  quently   give    plating   efficiencies   much 

any    possible    effects    of    the    test    cells  higher  than  those   obtained  with  fresh 

themselves   during   the   test   period.    In  medium.  Moreover,  colony  size  increases 

cultures    grown    in    the    preconditioned  at  a  strikingly  more  rapid  rate:  colonies 

medium,  myotube  formation  is  initiated  grown  in  conditioned  medium  for  6  days 

as   much   as    24   hours   earlier   than   in  attain    a     size     normally     observed     in 

cultures  initiated  with  equal  numbers  of  colonies  cultivated  for  13  days  in  fresh 

cells  from  the  same  cell  suspension  but  medium.  This  increased  growth  rate  was 

cultured  in  fresh  medium.  Furthermore,  not  apparent  by  inspection  in  the  mono- 

the  cells  in  conditioned  medium  attach  to  layer  test  series.   Reexamination  of  the 

the    glass    more    firmly,    presenting    a  effectiveness   of   "dialyzed"    conditioned 

strikingly  different  appearance  from  the  medium   using   plating   cultures   reveals 

control  cultures.  that,   actually,   low-molecular-weight   as 

It  is  apparent  that  the  medium  has  well  as  macromolecular  factors  are  in- 
been  altered  in  some  way  by  the  meta-  volved  that  could  not  be  readily  demon- 
bolic  activity  of  the  cells  cultured  in  it,  strated  in  the  monolayer  test.  Colonies 
since  merely  incubating  the  medium  for  grown  in  "dialyzed"  conditioned  medium 
an  equal  period  of  time  is  without  effect,  are  noticeably  smaller  than  those  grown 


400 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


in  untreated  conditioned  medium,  and 
they  exhibit  a  reduced  plating  efficiency. 
They  are,  however,  still  vastly  different 
in  both  respects  from  colonies  grown  in 
fresh  medium. 

The  effects  of  conditioned  medium  on 
differentiation  in  plating  cultures  are 
difficult  to  interpret.  Precocious  forma- 
tion of  myotubes  also  occurs  in  plating 
cultures.  On  day  6,  numbers  of  small 
colonies  composed  almost  entirely  of 
syncytial  elements  are  observed ;  myotube 
formation  has  never  been  observed  at 
this  time  in  colonies  cultivated  in  fresh 
medium.  Their  frequency  is  not  strikingly 
different  from  the  frequency  of  differ- 
entiated colonies  in  fresh  medium.  Vari- 
able results  are  observed,  however,  in 
plates  fixed  at  13  days.  Muscle  colonies 
similar  in  size  to  those  observed  on  day  6 
are  frequently  observed.  (Precocious  myo- 
tube formation  in  colonies  consisting  of 
small  numbers  of  cells,  if  it  involves  all  or 
most  of  the  cells  in  the  colony,  might  not 
increase  in  size  simply  because  all  the 
proliferative  "stem  cells"  had  been  re- 
cruited into  postmitotic  syncytia.)  Occa- 
sionally, these  small  colonies  are  either 
reduced  in  number  or  absent  at  day  13, 
but,  with  some  lots  of  conditioned 
medium,  in  plating  cultures  fixed  at  13 
days  muscle  colonies  fully  as  large  as  or 
larger  than  those  in  cultures  grown  in 
fresh  medium  are  observed.  The  fre- 
quency of  differentiated  colonies  in  such 
cultures  may  actually  exceed  the  usual 
frequency  encountered  with  fresh  medi- 
um. Konigsberg  is  now  investigating  the 
source  of  the  variability  in  supporting 
differentiation  among  various  lots  of 
conditioned  medium,  his  working  hypoth- 
esis being  that  with  his  present  condition- 
ing system  he  is  operating  close  to  some 
threshold  condition. 

The  examination  of  the  second  question 
raised  by  the  plating  experiments,  incom- 
plete as  it  is,  leads  us  back  to  the  first 
one,  namely:  what  is  the  significance  of 
the  finding  that  only  1  in  10  colonies 
eventually  differentiates?  The  studies 
involving  conditioned  medium  suggest  a 


possibility  of  reconciling  the  findings  on 
monolayer  culturing  with  the  former  view 
that  differentiation  is  promoted  by  organ 
culture  in  which  the  organization  of  the 
tissue  is  left  undisturbed.  The  necessary 
condition  may  in  reality  be  a  cell  density 
high  enough  to  correct  for  the  deficiencies 
of  the  tissue  culture  environment.  The  1 
in  10  colonies  that  differentiates  may 
represent  a  population  of  variants  able  to 
express  their  developmental  potential 
despite  the  inadequacies  of  the  environ- 
ment, just  as  the  plating  efficiency  itself 
may  represent  an  overlapping  class  of 
variants.  If  we  may  consider  the  possi- 
bility that  a  cell  type  like  the  myoblast 
may  actually  represent  a  population 
homogeneous  with  respect  to  certain 
metabolic  properties  but  heterogeneous 
with  respect  to  others,  the  effect  of 
conditioned  medium  on  plating  efficiency 
might  be  interpreted  as  providing  an 
environment  satisfying  the  requirements 
for  survival  and  multiplication  of  a 
broader  spectrum  of  the  myoblast  group. 
An  environment  adequate  for  survival 
and  multiplication,  however,  may  not 
satisfy  all  the  requirements  for  differenti- 
ation. The  differences  between  organ 
culture,  monolayer  culture,  and  plating 
culture  may  represent  the  different 
degrees  to  which  conditioning  or  cross 
feeding  may  occur. 

Regional  Localization  of 

Pre-Pacemaker  Cells  in 

the  Chick  Embryo 

Previous  year  books  have  recorded 
progress  in  Robert  L.  DeHaan's  analysis 
of  the  formation  of  the  heart  in  the  early 
chick  embryo.  Approaches  to  two  sets  of 
questions  have  been  considered:  the 
spatial  organization  and  morphogenetic 
movements  of  the  precardiac  mesoderm; 
and  the  formation  and  early  function  of 
the  cardiac  pacemaker  tissue.  In  the  past 
year  these  separate  lines  of  investigation 
have  begun  to  meet  in  the  question  of 
regional  localization  of  pre-pacemaker 
cells. 


DEPARTMENT    OF    EMBRYOLOGY  401 

It  will  be  recalled  that  the  regions  of  quite   different,    histologically   and   bio- 

precardiac  mesoderm  can  be  visualized  on  chemically,  from,  say,  those  that  form 

time-lapse  motion-picture  films  as  dark  the  posterior  wall  of  the  definitive  atria, 

condensed  areas  silhouetted  through  the  Moreover,  it  has  long  been  known  that 

endoderm.    The    preheart   mesoderm    is  the  several  regions  differ  in  their  intrinsic 

composed  primarily  of  discrete  clusters  of  pulsation    rate.     Alexander    Barry    has 

cells,    which   migrate   with    the   folding  shown  that  there  is  a  continuous  antero- 

foregut  endoderm,  using  that  layer  as  a  posterior    gradient    of    inherent    rhyth- 

substratum   for  their  own  independent  micity  in  the  chick  heart,  so  that  any 

movements.     The    migration    of    these  fragment    of    myocardium    beats    more 

clusters  was  tracked  on  such  films  from  slowly  than  those  posterior  to  it,   and 

their  initial  positions  in  the  heart-forming  more  rapidly  than  those  anterior,  along 

region  of  the  embryo  at  stage  5  into  the  the  axis  of  the  heart.  These  differences  in 

primitive  tubular  heart.  For  the  first  few  rate  are  a  function  of  individual  cells 

hours  after  these  clusters  condense  out  of  within  the  various  regions :  thus,   after 

the  background  mesenchyme  as  discrete  disaggregation  with  trypsin,  isolated  cells 

structures,  they  appear  to  migrate  in  a  of  the  atria  beat  faster  than  those  from 

random  fashion  within  the  heart-forming  the  ventricle. 

regions.  Gradually,  the  mesoderm  be-  As  DeHaan  has  argued,  the  "spon- 
comes  arranged  into  a  spatially  organized  taneous"  beat  of  cardiac  tissue  depends 
crescentic  pattern.  The  cells  destined  to  on  stimulation  of  myocardial  cells  by  the 
form  noncardiac  structures,  like  extra-  specialized  pacemaker  cells  of  the  con- 
embryonic  vascular  tissue  or  head  mesen-  ductive  tissue.  Therefore,  the  question 
chyme,  leave  the  heart-forming  regions  arises  whether,  in  the  early  heart-forming 
while  each  cluster  remaining  within  them  regions,  pre-pacemaker  cells  are  already 
takes  up  a  position  bearing  a  definite  and  spatially  organized  in  a  localized  fashion 
constant  relation  to  the  part  of  the  heart  with  the  rest  of  the  preheart  mesoderm 
to  which  that  cluster  will  contribute.  The  and  in  accordance  with  the  rate  gradient 
group  of  clusters  in  the  anteriormost  part  seen  in  the  beating  heart.  Specifically,  are 
of  the  lateral  heart-forming  region,  for  there,  in  the  anterior  parts  of  the  heart- 
example,  migrates  into  the  rostral  part  of  forming  regions,  destined  to  contribute  to 
the  heart  rudiments  that  develop  first,  the  distal  branches  of  the  Purkinje 
and  ultimately  forms  conus  and  cono-  system  in  the  conoventricular  region, 
ventricular  tissue.  The  clusters  in  the  pre-pacemaker  cells  which  exhibit  a  low 
middle  of  each  cardiac  primordium  form  level  of  rhythmicity?  In  contrast,  are 
the  belly  of  the  ventricle  of  the  tubular  there,  in  the  posterior  parts  of  the  heart 
heart;  the  most  posterior  clusters  in  the  primordia,  already  localized  with  the 
heart-forming  regions  enter  the  heart  prospective  atrial  and  sinoatrial  meso- 
rudiments  last,  to  form  atrial  and  sino-  derm,  pre-pacemaker  cells  capable  of 
atrial  tissue.  Thus,  in  addition  to  the  developing  high  levels  of  inherent  rhyth- 
prospective  heart  cells  being  differenti-  micity  suitable  for  the  sinoatrial  node  and 
ated,  as  such,  from  other  mesoderm,  at  atrial  conduction  tissue?  The  experiments 
these  early  stages,  the  various  parts  of  reported  here  were  designed  to  answer 
the  heart  are  also  represented  as  localized  these  questions  by  separating  the  heart- 
parts  of  the  heart-forming  regions.  forming  regions  into  anterior,  middle,  and 

This  localization  implies  more  than  just  posterior    parts    and    allowing    each    to 

a    differential    spatial    distribution.     It  develop  in  isolation, 

suggests  that  cells  in  each  of  the  prospec-  For  these  experiments  chick  embryos 

tive  regions  show  distinct  differentiative  were  explanted  ventral  side  up,  each  on 

capacities,  since,  in  the  later  embryonic  its  own  vitelline  membrane.   With  this 

and  adult  heart,  cells  in  the  conus  are  technique   development   progresses   nor- 


402 


CARNEGIE     INSTITUTION     OF     WASHINGTON 


mally  for  as  long  as  2 3/2-3  days.  For 
microsurgery,  embryos  were  allowed  to 
develop  in  culture  to  the  desired  stage 
and  were  cut  into  fragments,  as  dia- 
grammed in  figure  19.  Fragments  1R  and 
1L  were  calculated  to  include  material 
that  would  form  conus  and  conoventricu- 
lar  tissue ;  2R  and  2L  included  prospective 
ventricle,  and  possibly  some  atrioven- 
tricular tissue;  3R  and  3L  contained  the 


shaped  masses  of  tissue.  At  the  end  of  48 
hours  of  incubation,  a  substantial  in- 
crease in  size  is  seen,  and  each  culture 
takes  on  an  appearance  characteristic  of 
the  original  position  of  the  fragment  and 
age  of  the  donor.  Figure  20,  plate  4,  shows 
three  explants  after  48  hours  of  culture, 
as  whole  mounts  and  in  cross  section.  The 
posterior  fragments  exhibit  distinctly 
better  development  than  the  middle  or 


Stage  5 


Stage  7 


___     Stage  9 

"  "1   .        s. 


Fig.  19.  Diagrams  of  chick  embryos  at  stages  5,  7,  and  9.  The  heavy  broken  lines  represent  cuts 
made  to  separate  the  prospective  heart  regions  into  anterior,  middle,  and  posterior  fragments.  At  all 
stages,  fragments  1R  and  1L  contained  prospective  cono ventricular  mesoderm,  2R  and  2L  included 
pre  ventricular  cells,  and  3R  and  3L  had  sinus  and  atrium.  The  fragments  included  all  three  germ 
layers. 


posterior  clusters,  destined  to  form  atrial 
and  sinus  tissue.  Stages  4,  6,  and  8  (not 
shown  in  fig.  19)  were  cut  in  the  same 
fashion.  Each  fragment  was  incubated  in 
tissue  culture  medium  for  48  hours,  after 
which  it  was  examined  for  spontaneous 
pulsatile  activity,  and  the  rate  of  beating 
was  counted  with  the  aid  of  a  stopwatch. 
In  all,  178  embryos  (20-30  at  each  stage) 
were  operated  upon,  yielding  a  total  of 
1068  cultured  fragments,  which  provided 
the  data  for  the  present  study. 

After  8-10  hours  of  incubation  in 
culture  medium,  explanted  fragments 
tend  to  round  up  into  solid,  irregularly 


anterior  ones,   cardiac  tissue  appearing 
histologically  well  differentiated  in  all. 

Figure  21  summarizes  the  heart-form- 
ing potencies  of  the  cultured  fragments, 
in  terms  of  the  fraction  of  the  cultures 
that  developed  spontaneously  beating 
heart  tissue,  as  a  function  of  the  stage  of 
the  embryonic  donor.  The  anterior  frag- 
ments (1R,  1L),  destined  to  form  conus 
and  cono  ventricular  tissue,  contain  very 
few  preheart  cells  before  stage  8,  and  even 
at  stages  8  and  9  only  about  half  the 
cultures  are  capable  of  forming  con- 
tractile heart  tissue.  The  middle  frag- 
ments (2R,  2L)  also  exhibit  very  little 


DEPARTMENT    OF    EMBRYOLOGY 


403 


100 


90 


UJ 

RO 

£ 

3 

en 

to 

<f) 

70 

UJ 

o 

I- 

60 

U) 

q: 

Ul 

< 

50 

> 

UJ 

T 

U. 

40 

o 

O 

z 

UJ 

o 
a: 

UJ 

z 

1- 
< 

UJ 
CD 

30 
20 

Q. 

10 

-       3RLi 


9  B^^$Mi« 


\ 

•  ew^      3R^^ 

2RL  *  ».___. 


•  • 


I  RL  / 


>»•»».** 


6  7 

STAGE    OF  DONOR    EMBRYO 


Fig.  21.     Proportion  of  vesicles  with  beating  heart  tissue  as  a  function  of  age  of  embryonic  donor. 


pacemaker  activity  at  early  stages.  By 
stage  6,  however,  more  than  80  per  cent 
of  the  cultures  of  these  fragments  form 
beating  hearts,  and  by  stage  7  all  of  them 
do.  The  posterior  fragments  (3R,  3L), 
containing  prospective  atrial  and  sino- 
atrial cells,  are  the  first  to  gain  the 
capacity  to  form  pacemaker  tissue.  At 
the  first  sign  of  notochordal  cells  pushing 
out  in  front  of  Hensen's  node  (stage  4+), 
80  per  cent  of  these  posterior  fragments 
can  develop  beating  heart  masses;  and 
this  fraction  very  quickly  increases  to 
90-100  per  cent.  * 

Not  only  are  the  anterior  fragments 
incapable  of  forming  beating  heart  tissue 
until  stage  7-8,  the  vesicles  that  can  beat, 
as  indicated  in  figure  22  (curve  1  RL),  do 
so,  at  the  low  rate  of  40-50  beats  per 
minute.  The  middle  fragments  (2R,  2L), 
from  the  beginning,  contain  cells  with  a 
higher  intrinsic  rhythmicity,  producing 
vesicles  which  by  stage  7  beat  at  a  rate  of 
60-85  per  minute.  The  posterior  frag- 
ments, in  accordance  with  their  content 
of  prospective  atrial  and  sinoatrial  cells, 
produce  heart  vesicles  even  from  very 
early  stages  with  high  rates  of  spon- 
taneous contraction,  which,  like  those 
from  more  anterior  fragments,  gradually 
increase  in  rate  with  age  of  the  donor, 


leveling  off  at  stages  7  to  9  at  120-130 
beats  per  minute. 

The  well  defined  anteroposterior  rate 
gradient  of  cells  in  the  heart-forming 
regions,  demonstrated  in  figure  22,  agrees 
with  earlier  reports  of  a  similar  gradient 
of  inherent  rhythmicity  in  the  formed 
heart  after  it  begins  beating.  This  finding, 
however,  should  not  be  interpreted  as 
indicating,  necessarily,  a  distribution 
within  the  heart-forming  regions  at  these 
early  stages  of  pre-pacemaker  cells, 
already  determined  to  be  appropriate  for 
conus  or  ventricle,  atrium  or  sinus, 
although  this  is  one  possibility.  An 
equally  plausible  hypothesis  is  that  the 
heart-forming  mesoderm  in  early  stages, 
like  the  primordia  of  brain,  limb,  or  eye, 
represents  an  equipotential  system  in 
which  all  parts  of  the  primordial  material 
are  competent  to  form  any  part  of  the 
adult  organ.  Localized  differences  in 
developmental  potential  would  arise  as 
the  result  of  inductive  or  other  influences 
from  the  environment.  The  regional 
differentiation  of  a  specific  group  of 
precardiac  mesoderm  cells  would  be  a 
function  of  the  milieu  provided  by  the 
particular  region  of  endoderm  and/or 
ectoderm  with  which  those  cells  happened 
to  come  in  contact.  Anterior  fragments 


404 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


1R,  1L  would  produce  slowly  beating 
vesicles,  according  to  this  idea,  because 
the  endoderm  overlying  the  anterior 
heart-forming  regions  induces  precardiac 
cells  in  contact  with  it  to  become  pace- 
maker tissues  with  low  levels  of  inherent 
rhythmicity,  whereas  fragments  3R  or  3L 
develop  a  sinuslike,  rapid  beat,  in  similar 
fashion,  as  a  result  of  influences  on  the 
mesoderm  from  the  surrounding  tissues. 
Intrinsic  activity,  therefore,  would  reside, 
not  in  the  precardiac  mesoderm  itself, 
but  in  a  set  of  reciprocal  interactions 
between  mesoderm  and  endoderm.  Such 
inductive  relations  between  preheart 
mesoderm  and  both  endoderm  and  ecto- 
derm are  well  documented  for  other 
species. 

A  critical  test  to  distinguish  between 
these  two  ideas  is  easily  conceived: 
namely,  the  recombination  of  specific 
areas  of  endoderm  with  mesoderm  from 
different  parts  of  the  heart-forming 
regions   or   from   neutral    (i.e.,    nonpre- 


cardiac)  sites.  Such  a  test  awaits  the 
development  of  practical  techniques  for 
recombining  tissues  from  these  early 
stages. 

The  ability  to  localize  pre-pacemaker 
cells  in  specific  regions  of  the  early 
embryo,  before  they  become  functional, 
suggests  the  possibility  of  studying  the 
electrophysiology  of  such  cells  during 
their  differentiation.  Recording  with 
intracellular  electrodes  during  the  transi- 
tion of  a  cell  from  a  state  of  quiescence  to 
one  of  spontaneous  rhythmic  activity 
might  provide  new  insight  into  the 
mechanism  of  action  of  pacemaker  tissues 
in  general. 

The  Cultivation  in  Fluid  Medium 

of  the  Labile  Chorioallantoic 

Membrane 

In  recent  years,  Ebert  and  his  co- 
workers, notably  DeLanney  and  Mun, 
have  paid  increasing  attention  to  the 
labile  chorioallantoic  membrane.  In  pre- 


I40r 
130 
120 
I  10 
100 

90 

80 

70 

60 

50 


S    40 
<    30 


O    20 

Q. 

<°       10 

0 


Fig.  22. 
embryonic 


! 


i 

2RL.#  |      ••*• 


•  • 


•  •? 


•x 
•   i  • 


I.-' 


••'I 

K 
I 


•x* 


•I  • 

X 


1/T>*. 


IRL  i 

i 
i 


JL 


4  5  6  7  8  9 

STAGE  OF  DONOR  EMBRYO 

Spontaneous  pulsation  rate  of  the  beating  heart  vesicles  as  a  function  of  the  age  of 
donor.  The  vertical  bars  at  each  point  represent  standard  error. 


DEPARTMENT    OF    EMBRYOLOGY 


405 


vious  reports  and  several  other  publica- 
tions evidence  has  been  advanced  that 
the  characteristics  of  the  membrane  can 
be  altered  by  a  variety  of  specific  and 
nonspecific  stimuli.  For  example,  cardiac 
muscle  is  formed  in  the  chorioallantoic 
mesenchyme  after  inoculation  of  a  mix- 
ture of  cardiac  microsomes  and  Rous 
sarcoma  virus.  Less  convincing  but 
nevertheless  suggestive  evidence  is  avail- 
able for  the  differentiation  of  granulocytes 
in  the  endodermal  layer.  And,  finally,  a 
graft  of  spleen  and  to  a  lesser  extent  other 
tissues,  or  the  inoculation  of  suspensions 
of  spleen  cells,  results  in  the  appearance 
in  the  membrane  of  a  variety  of  respond- 
ing cells,  including  giant  cells,  which  wall 
off  the  homologous  cells  in  cystic  masses. 
Thus,  contrary  to  Burnet's  earlier  view, 
such  lesions  on  the  chorioallantois  are,  to 
a  large  degree,  derived  from  the  host.  But 
whether  they  migrate  in  from  the  embryo, 
or  are  mobilized  in  the  membrane  in  situ, 
or  both,  is  not  yet  clear.  In  an  article  to 
appear  shortly  in  Biological  Bulletin, 
Mun,  Tardent,  Errico,  Ebert,  DeLanney, 
and  T.  S.  Argyris  offer  evidence  derived 
from  tritium  thymidine  labeling  studies 
which  demonstrates  some  invasion  of 
splenic  grafts  by  cells  of  the  host. 

It  became  evident  that  to  continue 
these  lines  of  research  effectively  it  would 
be  necessary  to  develop  a  technique  for 
the  cultivation  of  fragments  of  the 
chorioallantois  in  vitro.  Such  an  approach 
should  permit  the  controlled  modification 
of  the  membrane  on  a  larger  scale  than 
had  previously  been  possible  as  well  as 
the  resolution  of  the  question  of  the 
origin  of  cells  reaching  homologous  grafts. 

Moscona  has  reported  that,  when  small 
fragments  of  the  chorioallantoic  mem- 
brane of  8-day  chick  embryos  are 
cultured  on  a  plasma  clot  by  Fell's  watch 
glass  technique,  the  chorionic  epithelium 
stretched  on  small  squares  of  rayon 
acetate  net  show  a  keratogenic  meta- 
plasia. Earlier,  Delson  described  a  method 
for  the  cultivation  of  the  membrane 
supported  by  filter  paper  without  such 
metaplasia. 


During  the  past  year  Chinami  Takata 
and  Nancy  Sype,  in  consultation  with 
J.  D.  Ebert,  initiated  a  study  of  several 
techniques  for  cultivating  the  membrane. 
The  following  preliminary  findings  may 
be  reported  briefly. 

In  the  course  of  the  study  to  date, 
several  hundred  large  fragments  of  the 
chorioallantoic  membrane  were  taken 
from  8-day  chick  embryos.  Each  frag- 
ment, supported  on  filter  paper,  was 
placed  on  the  surface  of  the  medium  to  be 
tested  in  an  embryological  watch  glass 
and  was  cultured  in  a  humid  chamber  for 
5  to  7  days  at  37°C.  The  explants  were 
transferred  to  a  fresh  medium  every 
second  day. 

The  filter  paper  (no.  5243-C,  Arthur 
H.  Thomas  Co.)  for  supporting  the 
membrane  was  cut  into  rectangles  about 
2.3  cm  by  2.0  cm  and  cut  out,  leaving  a 
frame  about  3  mm  wide.  The  frames  were 
treated  with  1  per  cent  hydrochloric  acid 
for  1  hour,  washed  in  three  changes  of 
distilled  water,  extracted  a  second  time 
for  1  hour  in  95  per  cent  alcohol,  rinsed 
in  several  changes  of  glass-distilled  water, 
dried  in  an  oven  (80-90°C),  and  sterilized 
by  autoclaving.  The  experimental  media 
were  made  up  from  sterile  stock  solutions 
in  the  proportions  given  in  Table  13. 
Hanks'  balanced  salt  solution  was  buf- 
fered with  tris  or  bicarbonate  at  pH 
7.5-7.8. 

After  cultivation  the  explants  were 
fixed  in  Bouin's  or  Zenker's  fluid,  stained 
with  Weigert's  iron  hematoxylin  and 
eosin,  and  examined  as  whole  mounts. 
Some  explants  were  sectioned  at  5,  6,  or 
8  microns  and  stained  with  Weigert's  iron 
hematoxylin  and  eosin  or  Biebrich's 
scarlet  for  microscopic  observation. 

The  criteria  for  determining  effective- 
ness of  each  medium  were  degree  of 
maintenance  of  mesenchyme,  ectodermal 
and  endodermal  layers,  blood  cells,  and 
proliferation  of  cells. 

Experimental  group  A.  Series  1,  2, 
and  3.  The  fragments  of  membrane  were 
cultured  in  media  containing  horse  or  calf 
serum,  Hanks'  balanced  salt  solution,  and 


406 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


TABLE  13. 

Maintenance  of  Mesenchyme  Found  in  Fragments  of  Chorioallantois  Cultured  in 

Experimental  Media 

Experimental 
Group 

Experimental 
Series 

Medium 

No. 

Explants 

Studied 

No.  in  Which 

Mesenchyme  Was 

Maintained 

A 

1 

2 
3 

10%  HS  +  75%  H  +  15%  EE 
20%  HS  +  65%  H  +  15%  EE 

20%  CS  +  65%  H  +  15%  EE 

11 
46 
21 

8 
38 
17 

B 

4 
5 

20%  HS  +  80%  H 
40%  HS  +  60%  H 

73 

27 

65 
19 

C 

6 

7 
8 
9 

10%  HS  +  50%  H  +  40%  P 
10%  CS  +  50%  H  +  40%  P 
20%  HS  +  65%  H  -  15%  P 
20%  HS  +  40%  H  +  40%  P 

39 
20 
12 

27 

32 

18 

9 

19 

HS  =  horse  serum. 

CS  =  calf  serum. 

H  =  Hanks'  balanced  salt  solution  containing  3  grams  of  glucose  per  liter. 

P  =  Puck's  medium  N-16,  synthetic  medium  without  protein. 

EE  =  9-day  chick  embryo  extract. 


embryo  extract.  In  series  1,  most  of  the 
blood  cells  in  the  blood  vessels  showed 
necrosis.  The  endodermal  layer  and 
mesenchyme  in  this  group  were  kept  in 
healthy  condition  during  the  cultivation 
(pi.  5,  figs.  23  and  24).  On  the  other  hand, 
the  ectodermal  layer  changed  into  a 
one-cell  layer  without  showing  keratini- 
zation. 

Experimental  group  B.  Series  4  and  5. 
When  the  explants  were  cultured  in  media 
containing  only  horse  serum  and  Hanks' 
balanced  salt  solution,  the  endodermal 
layer,  mesenchyme,  and  the  ectodermal 
epithelium  which  consists  of  a  single  layer 
were  maintained  in  good  state  (pi.  5,  fig. 
25).  Necrosis  of  blood  cells,  however,  was 
observed  in  series  5  containing  40  per  cent 
horse  serum. 


Experimental  group  C.  Series  6,  7,  8, 
and  9.  In  this  group,  Puck's  medium  N-16 
was  used  instead  of  embryo  extract. 
Maintenance  of  the  explants  was  favor- 
able (pi.  5,  fig.  26),  and  mitotic  figures 
were  often  observed.  Keratogenic  meta- 
plasia was  not  observed  in  a  flattened  cell 
layer  of  ectodermal  epithelium.  In  series 
7,  the  blood  cells  were  maintained  in 
rather  good  condition,  although  the  media 
of  series  8  and  9  entirely  failed  to  main- 
tain the  blood  cells  in  the  blood  vessels. 

In  summary,  although  the  ectodermal 
layer  of  the  membrane  changed  to  a 
flattened  cell  layer,  keratogenic  meta- 
plasia was  not  observed;  all  media  tested 
were  effective  for  maintenance  of  mesen- 
chyme and  endodermal  layer. 


INDUCTIVE   TISSUE  INTERACTIONS 


made  between  epidermis  and  dermis  from 

prospective  feathered  and  scaled  regions 

of  chick  embryos,  M.  E.  Rawles  observed 

that  epidermis  from  the  foot  (tarsometa- 

tarsus)   failed  to  self-differentiate  when 

underlain  by  dermis  from  the  middorsum. 

In  a  series  of  experiments  with  chorio-      In  this  combination,  epidermal  differenti- 

allantoic  grafts  reported  in  Year  Book  60,      ation  proceeded  in  a  feather  direction 

pages  424-427,  in  which  exchanges  were      rather  than  in  a  scale  direction.   Thus 


Interactions    between    Dermis    and 
Epidermis  from  Prospective  Feather 
and  Scale  Regions  after  Recombina- 
tion on  the  Chorioallantoic 
Membrane 


DEPARTMENT    OF    EMBRYOLOGY 


407 


there  appeared  to  be  no  specific  time 
before  or  during  the  formation  of  the  scale 
ridges  when  the  capacity  to  form  scales  is 
irrevocably  fixed  or  "determined"  in  the 
epidermis  of  the  foot  region.  Differences 
in  the  responsiveness  of  the  epidermis 
with  increase  in  developmental  age  could 
be  detected,  but  only  by  the  degree  of 
abnormality  displayed  by  the  feathers. 

Inasmuch  as  the  dermis  in  the  above- 
mentioned  combination  always  came  from 
the  prospective  dorsal  feather  tract,  a 
region  that  normally  gives  rise  to  feathers 
only,  the  question  arises  about  what 
course  the  epidermal  reaction  would 
follow  if  underlain  by  mesoderm  (dermis) 
isolated  from  the  relatively  featherless 
areas  lying  between  the  feather  tracts — 
the  apteria.  Results  obtained  so  far  from 
combining  epidermis  of  the  foot  (9-  to 
14- day  embryos)  with  dermis  from  the 
apteria  (6-  to  9-day  embryos)  have, 
interestingly  enough,  shown  the  same 
type  of  epidermal  reaction  obtained 
previously  when  the  dermis  was  of 
feather-tract  origin.  Feathers,  normal  and 
abnormal,  depending  on  the  age  of  the 
epidermis  involved,  appeared  on  the 
graft  surface.  Scales  could  not  be  recog- 
nized. 

The  fact  that  the  apteria  are  not 
always  entirely  featherless  in  the  normal 
chick  indicates  that  the  skin  of  these 
areas  is  potentially  capable  of  giving  rise 
to  down  feathers.  In  other  words,  the 
feather-forming  potency  is  not  restricted 
entirely  to  the  feather  tracts. 

Of  the  few  completely  featherless  areas 
of  the  body  of  a  chick,  the  beak  region 
seemed  to  be  most  suitable  for  Rawles's 
purpose.  Experiments  are  now  in  progress 
in  which  exchanges  are  being  made 
between  epidermis  and  dermis  of  foot  and 
beak  regions.  Results  obtained,  so  far, 
from  combinations  of  epidermis  from  the 
foot  of  11-  to  13-day  embryos  with 
dermis  from  the  prospective  beak  of  5-  to 
7-day  embryos  have  shown,  without 
exception,  beak  formation,  normal  in 
form  and  structure. 

Results    of    the    various    combination 


experiments  mentioned  above  have  given 
striking  confirmation  to  the  principle, 
heretofore  well  established  for  the  chick 
by  other  workers,  that  in  the  formation  of 
epidermal  derivations  the  specificity  re- 
sides in  the  mesoderm.  Of  further 
interest,  however,  is  the  observed  fact 
that  epidermis  of  the  foot  region  at  a 
relatively  late  developmental  stage  (13- 
14  days)  after  the  formation  of  scales  is 
still  labile  enough  to  change  its  direction 
of  differentiation  either  into  a  feather  or 
even  more  remarkably  into  a  highly 
keratinized  smooth  beak. 

Lens  Induction 

Another  classic  example  of  induction  is 
found  in  the  interaction  of  optic  vesicle 
and  ectoderm  in  the  formation  of  the  lens. 
It  has  long  been  thought  that  the  forma- 
tion of  the  lens  offers  a  favorable  target 
for  the  analysis  of  induction  at  the 
biochemical  level.  It  was  for  that  reason, 
in  fact,  that  the  goal  of  one  of  the  more 
rewarding  programs  in  the  Department 
in  recent  years,  that  conducted  by  John 
Papaconstantinou  (Year  Books  58,  pp. 
379-385,  and  59,  pp.  378-380)  was  to 
describe  the  normal  sequence  of  protein 
synthesis  in  the  lens  as  a  basis  for 
combining  techniques  of  biochemistry 
and  experimental  morphology. 

During  the  year  P.  H.  S.  Silver  has 
begun  such  a  study,  partly  in  collabora- 
tion with  Papaconstantinou,  now  at  the 
University  of  Connecticut,  the  long-range 
goal  of  which  is  to  determine,  by  bio- 
chemical and  immunochemical  methods, 
the  nature  of  the  proteins  formed  in 
interactions  of  chick  and  duck  embryonic 
tissues.  The  analysis  has  not  advanced 
sufficiently  to  warrant  extensive  discus- 
sion, but  the  following  brief  remarks  are 
appropriate.  None  of  the  methods  for 
homoplastic  lens  induction,  described  pre- 
viously in  the  literature,  has  proved 
satisfactory,  as  the  resulting  lenses  are 
too  small  or  misshapen  for  biochemical  or 
immunological  study.  For  this  reason  an 
extensive  survey  has  been  carried  out  to 
find    the    most    suitable    experimental 


408  CARNEGIE     INSTITUTION     OF      WASHINGTON 

technique.  Several  different  methods,  in  To  do  this,  the  embryo  with  its  blasto- 

vitro  and  in  ovo,  have  been  tried.  The  derm  is  explanted  in  vitro.  The  blast o- 

combination  of  the  primary  vesicle  with  derm  is  folded  so  that  the  embryo  lies  on 

competent   ectoderm   in   ovo   has   been  its  side.  The  graft  is  fixed  by  adjusting 

abandoned,  for  Silver's  material  shows  the  level  of  the  culture  fluid  so  that  the 

that,  despite  a  recent  statement  to  the  surface  tension  presses  the  graft  in  place, 

contrary  by   McKeehan,   the  retina   of  Once   the   graft   has   stuck    (about    V/i 

both  chick  and  duck  has  the  capacity  hours)   the  blastoderm  is  unfolded  and 

itself  to  produce  well  formed  lens  tissue,  cultured  in  the  normal  way.  The  thicken- 

and  so,  if  the  initial  experiment  is  carried  ing  and  invagination  of  the  graft  epi- 

out  in  ovo,  and  a  lens  results,  there  will  thelium   can  be   observed  in  the  early 

always  be  doubt  about  whether  this  lens  stages  of  lens  formation,  and  the  origin  of 

has   originated   by   induction   from   the  the  lens  from  the  surface  ectoderm  can  be 

covering    ectoderm   or   by   regeneration  established.     The    whole    eye    is    then 

from  the  retina.  transplanted    to    the    head    of    another 

A  definitive  technique  has  now  been  embryo  and  left  for  16  to  17  days.  There 

arrived  at  in  which  the  prospective  lens  is  no  doubt  that  the  head  mesenchyme  is 

epithelium  of  the  host  embryo  is  replaced  the  best  milieu  for  the  growth  of  eye 

by  the  corresponding  tissue  of  the  donor,  tissue. 

HUMORAL  REGULATORY  MECHANISMS 

Growth  Promotion  during  th+e   f*^f*    of    growth    in    each 

Regeneration  and  the  situation  ls  due  +t(\a  diffusible  growth- 

Control  or  Growth  promoting    agent    from    the    damaged 

tissue.     Identification    ol    the    growth- 

The  mechanisms  controlling  growth  promoting  agents  involved  is  a  necessary 
and  differentiation  in  adult  mammalian  prerequisite  to  understanding  the  mech- 
tissues  are  even  less  understood  than  anisms  by  which  growth  is  controlled, 
those  involved  in  controlling  growth  and  The  aim  of  Argyris's  research  program 
differentiation  in  the  embryo.  All  students  during  the  past  10  months  has  been 
of  growth  control  in  adult  organs  agree  primarily  to  become  acquainted  with,  and 
that  that  is,  at  least  in  part,  a  function  of  develop,  a  number  of  biochemical  tech- 
intercellular  chemical  communication,  niques  with  which  he  may  try  to  identify 
The  nature  of  the  chemical  communica-  these  growth-promoting  agents.  Specifi- 
tion  is  unknown.  Regeneration  is  an  cally,  two  biochemical  tools  have  been 
obvious  method  for  forcing  organs  to  explored:  the  isolation  of  cell  particulates, 
reveal  some  of  their  potential  for  control-  and  autoradiography.  The  rationale  for 
ling  growth.  Therefore,  it  is  not  surprising  the  development  of  these  two  techniques 
that  the  mechanisms  involved  in  con-  seems  obvious  and  will  not  be  presented 
trolling  growth  during  wound  healing  and  fully.  By  means  of  cell  particulate 
compensatory  hypertrophy  have  been  isolation  he  will  attempt  to  determine 
intensively  studied.  what  cell  fraction  or  fractions  are  respon- 

T.  S.  Argyris  has  been  studying  two  sible  for  the  observed  growth-promoting 

situations  in  which  damage  leads  to  the  effects.  Once  this  is  accomplished  efforts 

stimulation  of  growth :  the  stimulation  of  will  be  made  to  isolate  and  identify  the 

growth  of  resting  hair  follicles  surround-  specific  growth-promoting  agent  (s)  in  the 

ing  a  wound,  and  the  stimulation  of  cell  active  fraction,  thus  achieving  localiza- 

division  in  the  kidney  cortex  following  tion    as    well    as    identification    of    the 

damage  by  insertion  of  a  needle  into  the  growth-promoting  agent(s).  By  labeling 

contralateral  kidney.  It  is  presumed  that  the  damaged  tissues  with  an  appropriate 


DEPARTMENT    OF    EMBRYOLOGY  409 

isotope  and  making  autoradiograms  one  an  approach  containing  an  element  of 

may  hope  to  follow  the  movement  of  the  uncertainty  inasmuch  as  whole  homoge- 

growth-promoting   substances   from   the  nates  may  contain  substances  that  inhibit 

damaged  tissues  to  the  target  tissues.  or  inactivate  the  growth-promoting  sub- 

Argyris  further  plans  to  explore  the  use  stance  (s). 

of  these  techniques  in  studying  a  third  Effect  of  subcutaneously  injected  homog- 

growth-promoting  interaction,  the  stimu-  enates  of  Ehrlich  ascites  tumor  (EA  T)  on 

lation    of    the    overlying    epidermis    by  the  skin  of  mice.  The  EAT  was  secured 

subcutaneous  transplants  of  tumors.  Ar-  from   a   stock   carried    by   Argyris   and 

gyris  and  Argyris  have  found  previously  Argyris  at  Syracuse  University  for  the 

that    the    subcutaneous    inoculation    of  last  3  years.  The  EAT  grows  as  a  fluid 

Ehrlich  ascites  tumor,  sarcoma  180,  or  tumor  in  the  peritoneal  cavity  of  mice, 

adenocarcinoma  755  results  in  invasion  of  It  is  transplanted  every  8  to  10  days  by 

the  skin  and  the  stimulation,  at  a  distance  removing  the  tumor  with  a  no.  23  needle 

of  at  least  100  microns,  of  the  overlying  mounted   on   a   tuberculin   syringe   and 

epidermis.    Contrarily,   the   resting  hair  injecting    0.1    ml    of    the    white    fluid 

follicles  closer  to  the  tumor  are  never  material  intraperitoneally  into  2  female 

stimulated.  mice.  An  additional  sample  of  the  tumor 

is  diluted  with  a  saline  solution  of  0.2 

Orqan  Homoqenates  and  Growth  Promotion  ^     ,               ° ,        '                        .  „  s   a.^e 

made    in    a    hemacytometer.    All    cells 

Preliminary  experiments  homogenizing  staining  black  because  of  Nigrosin  uptake 

liver,  kidney,  skin,  skin  wounds,  and  the  are  excluded  from  the  count,  presumably 

solid  form  of  the  Ehrlich  ascites  tumor  being  damaged  or  dead. 

(EAT)  in  a  Servall  omnimixer  or  in  a  Since  it  is  not  uncommon  to  find  the 

number    of    different     kinds     of    glass  loss  of  stimulating  activity  of  one  line  of 

homogenizers  driven  by  motor  or  by  hand  tumor  in   a   newly   established   subline, 

led  to  the  conclusion  that  the  use  of  the  0.1  ml  of  the  EAT  was  injected  subcu- 

Ten  Broek  glass  hand  homogenizer  was  taneously  into  5  mice  with  skin  in  the 

the  best  method,  especially  for  the  small  resting  phase  to  see  whether  it  would 

amount  of  material  available  for  homog-  invade  the  skin  and  stimulate  the  over- 

enization.   In  all  the  studies  to  be  de-  lying  epidermis.  Mice  were  killed  21  days 

scribed  the  degree  of  homogenization  was  after  tumor  inoculation — ample  time  for 

checked  by  smearing  an  aliquot  of  the  the  tumor  to  invade  the  skin  and  stimu- 

homogenate  on  a  glass  slide,  fixing  it  in  late  the  epidermis.  Six  hours  before  being 

100  per  cent  methanol  for  1  to  2  minutes,  killed  the  mice  were  injected  subcutane- 

and  staining  with  hematoxylin  for  2  to  5  ously  with  0.01  ml  of  colchicine  (10  mg  25 

minutes.  All  tissues  were  homogenized  in  ml  saline)  per  gram  body  weight.  Biopsy 

0.25  M  buffered  sucrose,  pH  7.3.  Buffer  specimens  of  the  skin  and  tumor  were 

contained  10-2  tris  and  10~3  magnesium  fixed    in    cold     10    per    cent    formalin, 

chloride.  Homogenization  was  done  in  an  dehydrated    in    dioxane,  embedded    in 

ice  bucket,  and  sucrose  homogenizers  and  paraffin,    and    sectioned    at    5    microns, 

other    utensils    were    always    precooled  Slides  were  routinely  stained  with  hema- 

before  use.  toxylin  and  eosin.  All  the  sections  showed 

Since  the  fraction  of  the  homogenate  good  tumor  growth,  invasion  of  the  skin, 

that    may    contain    the    active    growth-  and    stimulation    of   the    overlying   epi- 

promoting  principle  cannot  be  predicted  dermis,    the    last    being    evidenced    by 

it  was  necessary  initially  to  inject  the  thickening  of  the  epidermis  due  not  only 

whole  sucrose  homogenate  and  to  begin  to  an  increase  in  cell  number  but  also  to 

injecting  isolated  cell  fractions  only  when  cell  enlargement, 

the  whole  homogenate  showed  activity —  Effect  of  subcutaneoashj  injected  homog- 


410 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


enates  of  EA  T  on  the  skin  of  mice.  Tumor 
material  for  homogenization  was  provided 
by  injecting  8  to  10  mice  subcutaneously 
with  0.1  ml  of  EAT.  After  7  days  the 
animals  were  killed,  and  the  solid  nodules 
of  tumor  were  removed,  quickly  weighed, 
and  minced  with  scissors.  Instead  of 
adding  the  9  ml  of  sucrose  per  gram  of 
tissue  usually  recommended,  which  was 
found  to  be  effective  for  normal  organs, 
about  3  ml  per  gram  of  tumor  material 
was  added  and  homogenized  in  an  ice 
bucket  with  a  Ten  Broek  glass  homoge- 
nizer.  The  use  of  less  than  expected 
amounts  of  sucrose  is  due  to  the  fact  that 
normal  amounts  produce  homogenates 
that  are  too  dilute  and  make  breaking-up 
of  tumor  cells  too  difficult.  This  difficulty 
might  have  been  anticipated,  since  the 
dilutions  have  been  calculated  per  gram 
of  living  tissue.  In  a  7-day  solid  growing 
EAT  only  the  exterior  is  living,  and  it  is 
hardly  more  than  a  quarter  to  a  half  of 
the  entire  tumor  mass.  A  drop  or  two  of 
the  homogenate  was  smeared  on  a  glass 
slide  and  stained  as  described  above  to 
make  certain  that  disruption  of  the  tumor 
cells  had  occurred. 

To  ensure  that  the  mice  had  skin  in 
the  resting  phase,  the  entire  dorsum  of 
each  mouse  was  plucked  21  days  before 
inoculation.  The  plucking  initiated  hair 
growth  in  the  plucked  area  only,  and  the 
hair  follicles  completed  their  growth  and 
by  21  days  came  to  rest  again.  They 
usually  remained  at  rest  for  4  to  8  weeks. 
It  is  necessary  to  have  skin  in  the  resting 
phase  because  the  epidermis  changes  in 
thickness  at  different  stages  of  the  hair- 
growth  cycle,  making  evaluation  of 
stimulation  by  experimental  factors  diffi- 
cult. 

Eighteen  female  C57  mice  were  anes- 
thetized with  ether,  and  0.1  ml  of 
homogenate  of  EAT  was  injected  sub- 
cutaneously. The  injected  area  was 
marked  with  a  drop  of  eosin.  Nine  mice 
were  killed  each  at  2  and  9  days  after 
injection  of  homogenate.  Six  hours  previ- 
ously the  mice  received  colchicine  as 
described    above.    Biopsy    specimens    of 


skin  were  fixed  in  cold  10  per  cent 
formalin,  routine  histological  procedures 
being  followed. 

Stimulation  of  the  epidermis  due  either 
to  cell  enlargement  or  to  cell  division  has 
not  been  observed.  The  nuclei  of  the 
homogenate  persist  and  are  embedded  in 
a  blue-staining  homogeneous  mass.  The 
homogenate  often  results  in  the  stimu- 
lation of  the  loose  areolar  connective 
tissue  underneath  the  skin,  a  reaction  also 
elicited  by  the  living  tumor  cells. 

Effect  of  intraperitoneal  injections  of 
kidney  homogenates  on  the  kidney. 
Material  for  kidney  homogenate  was 
obtained  by  killing  a  female  C57  mouse 
and  removing  both  kidneys.  After  being 
weighed,  kidneys  were  quickly  minced 
and  placed  in  a  Ten  Broek  homogenizer 
and  homogenized  in  0.25  M  buffered 
sucrose  as  described  above.  Various 
amounts  of  homogenate  or  buffered 
sucrose  were  injected  intraperitoneally 
into  female  mice ;  48  hours  later  the  mice 
were  killed,  and  their  left  kidneys  were 
removed  and  prepared  for  histological 
study  as  previously  described.  Six  hours 
before  being  killed  mice  received  0.1  ml 
of  colchicine  (10  mg/25  ml  saline)  per 
gram  body  weight,  subcutaneously.  The 
choice  of  48  hours  as  the  interval  between 
injection  of  homogenate  and  sacrifice  was 
arbitrary,  based  on  the  suggestion  of 
Swann  that  a  mitotic  "inductive"  stim- 
ulus of  the  kind  that  might  be  produced 
by  the  injection  of  homogenate  requires 
about  48  hours  for  its  expression  because 
of  the  time  necessary  for  shifts  in  intra- 
cellular synthesis  to  occur  leading  to  the 
building  of  mitotic  protein.  Mitotic 
counts  were  done  on  2  nonadjacent 
sections  per  kidney,  counting  10  fields 
per  section,  using  a  44  X  objective  and  a 
10  X  ocular.  At  this  magnification  there 
were  about  1000  cells  per  field.  Thus  the 
mitotic  counts  represented  the  number  of 
mitoses  per  20,000  cells.  The  choice  of 
this  counting  procedure  was  based  on 
considerable  previous  experience  indi- 
cating that  such  counts  are  sufficient  to 
establish  obvious  differences.  As  a  further 


DEPARTMENT    OF    EMBRYOLOGY 


411 


check,  a  number  of  kidneys  were  re- 
counted, counting  10  nonadjacent  sec- 
tions, 10  fields  per  section,  or  a  total  of 
about  100,000  cells.  Mitotic  counts  were 
proportionately  the  same. 

Table  14  presents  the  mitotic  counts 
and  shows  that  the  intraperitoneal  injec- 
tion of  0.2  ml,  0.1  ml,  or  0.05  ml  of  kidney 
homogenate  does  not  increase  or  decrease 
the  mitotic  counts  of  the  kidney  cortex 
as  compared  with  the  injection  of  the 
same  amount  of  buffered  sucrose.  More- 
over, these  counts  are  the  same   as  he 


fore,  homogenates  were  prepared  from 
kidneys  at  5  minutes  and  at  5.5,  24,  or 
48  hours  after  damage.  Kidney  damage 
was  produced  in  anesthetized  female 
mice  by  exteriorizing  both  kidneys  and 
puncturing  each  kidney  4  times  with  a 
sterile  no.  23  needle  mounted  on  a 
tuberculin  syringe.  Two-tenths  milliliter 
of  the  homogenate  was  injected  intra- 
peritoneally  into  each  mouse,  and  the 
mice  were  killed  48  hours  later,  after  the 
subcutaneous  injection  of  colchicine  as 
already  described. 


TABLE  14.     Mitotic  Activity  of  Mouse  Kidney  after  Injection  of  Homogenates  of  Normal 

Mouse  Kidney 


Material  Injected  into 

No.  Mitoses  per 
20,000  Kidney  Cortical  Cells 

Each  of  6  Mice 

Average 
±  SEm 

Range 

0.2    ml  kidney  homogenate 
0.2    ml  buffered  sucrose 
0.1    ml  kidney  homogenate 
0.1    ml  buffered  sucrose 
0.05  ml  kidney  homogenate 
0.05  ml  buffered  sucrose 

0.67  ±  0.36 

2.0  ±0.94 
0.33  db  0.36 

0 

1.1  ±0.59 
1.0    ±  1.0 

0-2 
0-5 
0-2 

0 
0-3 
0-3 

has  previously  obtained  after  the  injec- 
tion of  saline  or  from  kidneys  of  non- 
injected  female  mice. 

The  question  arises  whether  the  nega- 
tive results  may  not  be  due  to  the  fact 
that  the  homogenates  are  prepared  from 
normal  kidneys.  But  it  should  be  recalled 
that  the  original  observation  in  vivo  is 
that  damaged  kidney  results  in  the 
mitotic  stimulation  of  the  undamaged 
contralateral  kidney.  Perhaps  if  the 
homogenates  were  prepared  from  dam- 
aged kidneys  kidney  cell  division  might 
be  stimulated. 

Since  in  the  original  in  vivo  experi- 
ments the  increase  in  mitotic  activity  of 
the  contralateral  kidney  is  observed  48 
hours  after  kidney  damage,  it  is  not 
known  whether  the  stimulating  activity 
from  the  damaged  kidney  is  released  at 
once  or  some  time  after  damage.  There- 


Table  15  shows  that  the  intraperitoneal 
injection  into  a  female  mouse  of  0.2  ml 
of  kidney  homogenate  prepared  from 
kidneys  5  minutes,  5.5  hours,  24  hours, 
or  48  hours  after  damage  does  not  result 
in  any  significant  increase  or  decrease  of 
mitotic  activity  in  the  kidney. 

The  effect  of  homogenates  of  skin  wounds 
on  resting  hair  follicles.  Here  the  basic 
premise  is  that  homogenates  of  skin 
wounds  should  be  able  to  induce  the 
growth  of  resting  hair  follicles  when 
injected  subcutaneously,  since  experi- 
ments in  vivo  have  suggested  that  the 
stimulation  of  hair  growth  is  due  at  least 
in  part  to  a  diffusible  substance  from  the 
wounds.  Two  kinds  of  stimulation  of  hair 
growth  can  be  expected  after  subcu- 
taneous injection  of  homogenates:  (1) 
Local  stimulation  of  growth  of  resting 
hair  follicles  directly  over  the  area   of 


412  CARNEGIE     INSTITUTION      OF      WASHINGTON 

TABLE  15.     Effect  of  0.2  Milliliter  of  Damaged  Kidney  Homogenate  on  Mitotic  Activity 

of  Mouse  Kidney 

No.  Mitoses  per 
N     M"  Time  between  Damage  and  20,000  Kidney  Cortical  Cells 


Killing  of  Mice 


Average  ±  SEm  Range 


4  5      minutes  1.7  ±0.97  0-3 

8  5.5  hours  0.5  ±  0.28  0-2 

6  24      hours  1.5  ±  0.24  1-2 

6  48      hours  0.5  ±  0.24  0-1 


homogenate  injection.  This  would  be  due,  were  included  in  the  final  experiment  for 

presumably,  to  diffusion  of  the  requisite  the  sake  of  completeness, 

growth-promoting    substances    into    the  For  injection  the  mice  were  anesthe- 

skin  from  the  underlying  injected  homog-  tized  with  ether  and  the  homogenate  or 

enate.    (2)   General  stimulation  of  hair  buffered  sucrose  was  injected  subcutane- 

growth  over  the  entire  back  of  the  mice,  ously  by  means  of  a  no.  23  needle  mounted 

this  stimulation  presumably  being  due,  in  a  tuberculin  syringe.  Before  injection 

according  to  Chase,  to  an  effect  of  the  mice   were   clipped.    Clipping   does   not 

homogenate  on  a  systemically  circulating  stimulate  hair  growth,  and  it  makes  the 

inhibitor  which  normally  helps  to  control  detection  of  new  hair  growth,  observed 

hair   growth.    That    a   substance   exists  as  a  blue  area  on  the  skin  surface,  easier, 

which   circulates   systemically   and   can  Mice  so  treated  were  examined  every  day 

affect  hair  growth  is  suggested  by  recent  for  a  month  or  more  for  either  a  specific 

experiments  of  Ebling  and  Johnson.  In  bluing  over  the  injected  site,  indicating 

these  experiments,  therefore,  evidence  of  local  hair  growth  stimulation,  or  bluing 

both  kinds  of  stimulation  was  sought.  of  the  entire  back,  indicating  a  general 

Wound  homogenate  was  obtained  by  stimulation  of  hair  growth, 

making  three  or  four  wounds  in  female  Of  49  female  and   12  male  mice  so 

C57/Black  mice  immediately  after  pluck-  treated  none  showed  any  local  stimulation 

ing   their   backs.    Four    days    later   the  over  the  injected  site,  with  either  0.1,  0.2, 

wounds  and  the  surrounding  skin  were  or  0.5  ml  of  wound  homogenate  or  sucrose 

excised,  placed  in  a  precooled  beaker,  and  (table  16). 

thoroughly  minced  with  scissors.  Similarly,    these    animals    showed    no 

Minced  wounds  were  then  transferred  enhanced  general  stimulation  over  their 

to   a  homogenizer  immersed   in  an  ice  entire  backs  (table  17). 

bucket ;  proper  amounts  of  sucrose  were  Since  it  is  known  that  hair  follicles  that 

added ;  and  the  material  was  homogenized  have  been  resting  for  some  time  and  are 

as  described  above.   Since  skin  wounds  about    to    grow    can    be    prematurely 

were  difficult  to  homogenize,   care  was  brought    into   the    growth   phase    by   a 

exercised  that  enough  of  the  smear  was  variety  of  stimuli  (Chase) ,  it  was  decided 

examined  to  be  sure  that  the  cells  were  to   repeat   the   experiments   using   mice 

broken  up.  whose  hair  follicles  had  been  at  rest  for 

To  be  certain  that  the  mice  receiving  18    days.    Perhaps    the    ability    of    the 

the  wound  homogenate  had  resting  hair  homogenate  to  induce  hair  growth  would 

follicles,  their  backs  were  plucked  21  days  then  be  enough.  Sixteen  female  mice  were 

before  injection.  Large  female  mice  were  injected  as  described  above  with  0.2  ml 

used  because  males  tended  to  fight,  and  of  homogenate  18  days  after  their  hair 

bites  induce  hair  growth,  but  a  few  males  follicles  had  come  to  rest.  Again,  neither 


DEPARTMENT    OF    EMBRYOLOGY  413 

TABLE  16.     Local  Effect  of  Wound  Homogenate  on  the  Stimulation  of  Growth  of  Resting 

Hair  Follicles  in  Mouse  Skin 


Presence  of  Local  Hair 

No.  and  Sex  of  Mice 

Material  Injected 

Growth  Stimulation 

7F 

0 . 1  ml  wound  homogenate 

— 

16  F 

0.2  ml  wound  homogenate 

— 

6F 

0.3  ml  wound  homogenate 

— 

6F 

0.3  ml  wound  homogenate 

— 

8F 

0 . 2  ml  buffered  sucrose 

— 

8F 

0 . 5  ml  buffered  sucrose 

— 

6M 

0.2  ml  wound  homogenate 

— 

6M 

0.2  ml  buffered  sucrose 

— 

TABLE  17.     General  Effect  of  Wound  Homogenate  on  the  Stimulation  of  Growth  of  Resting 

Hair  Follicles 


Time,  in  Days, 

of  Appearance 

of  General  Hair  Growth  on 

No.  and  Sex  of  Mice 

Material  Injected 

Backs  of  Mice 

Average  ±  SEm 

Range 

7F 

0 . 1  ml  wound  homogenate 

42.9 

39-58 

16  M 

0.2  ml  wound  homogenate 

30.7 

16-45 

6F 

0.3  ml  wound  homogenate 

33.3 

24-58 

6F 

0 . 5  ml  wound  homogenate 

43.3 

37-52 

6F 

0.2  ml  buffered  sucrose 

22.5 

13-30 

6F 

0.5  ml  buffered  sucrose 

41.1 

28-52 

6M 

0.2  ml  wound  homogenate 

32.3 

13-36 

6  M 

0 . 2  ml  buffered  sucrose 

27.5 

5-50 

specific    nor   nonspecific    stimulation    of  skin  muscle  removed  in  a  small  area  of 

hair  growth  was  seen.  skin  by  a  special  procedure  developed  by 

The  possibility  exists  that,   although  Argyris  at  Syracuse  University.  As  before, 

the  amount  of  homogenate  is  sufficient  to  these  mice  were  plucked  and  their  hair 

induce    hair    growth,    not    enough    can  follicles  were  allowed  to  grow  and  reenter 

diffuse   through  the   skin  muscle   (pan-  the    resting    phase.     They    were    then 

niculus  carnosus)  to  stimulate  hair  growth  subcutaneously  injected  with  0.2  ml  of 

locally.  That  skin  muscle  may  act  as  a  wound  homogenate  or  buffered  sucrose 

barrier  to  growth-promoting  substances  near  the   area  of  skin  that  lacked   its 

has  been  suggested  by  previous  work  in  muscle. 

which   the    growth-promoting   effect    of         No  stimulation  of  hair  growth  was  ever 

tumors  on  the  mammary  gland  and  skin  observed  with  the  dosage  of  homogenate 

of    mice    has    been    studied.    In    these  used  (table  18). 

experiments  it  had  been  noticed  that  the  Since  there  is  a  measure  of  thorough- 
presence  of  muscle  between  the  target  ness  in  this  study,  it  might  appear  that 
tissue  and  the  tumor  is  always  associated  the  negative  results  strongly  suggest  that 
with  an  absence  of  stimulation  of  the  wounds  do  not  contain  growth-promoting 
target  tissue.  Therefore,  the  experiments  substances.  This  can  hardly  be  true.  We 
were  repeated  using  22  female  and  14  must  keep  in  mind  that  the  wound  tissue 
male  C57/Black  mice  which  had  their  used  is  that  present  4  days  after  injury. 


414  CARNEGIE     INSTITUTION      OF     WASHINGTON 

Perhaps  the  use  of  wounds  from  earlier  or  sodium  acetate- 1-C14  and  tritiated  water 

later  stages  would  reveal  growth-promot-  being  used. 

ing   substances.    Moreover,    the   wound  Sodium   acetate- 1-C14   had   a   specific 

homogenates  were  prepared  from  whole  activity  of  5.90  mc/mM.  It  was  diluted 

wounds,  which  have  a  variety  of  tissue  to  the  desired  concentration  with  sterile 

components,    such    as    epithelium    and  saline.  Eight  female  C57/Black  mice  were 

granulation    tissue,     and    the    growth-  injected  intraperitoneally  with  50  fxc  of 

promoting  activity  may  be  inhibited.  It  the  labeled  acetate.  Mice  were  killed  at 

may  well  be  that  homogenates  of  each  1,  8,  and  24  hours  after  the  injection, 

kind  of  tissue  will  have  to  be  prepared  and  Biopsy  specimens  were  fixed  in  10  per 

tested  separately.  To  this  end  Argyris,  cent   formalin,    dehydrated   in   dioxane, 


TABLE  18.     Local  Effect  of  Wound  Homogenate  on  the  Stimulation  of  Growth  of  Resting 
Hair  Follicles  in  Mouse  Skin  in  Which  the  Skin  Muscle  (panniculus  carnosus)  Had  Been  Removed 

Presence  of  Local  Hair 
No.  and  Sex  of  Mice  Material  Injected  Growth  Stimulation 

1 1  F  0 . 2  ml  wound  homogenate  — 

1 1  F  0 . 2  ml  buffered  sucrose  — 

9  M  0 . 2  ml  wound  homogenate  — 

9  M  0 . 2  ml  buffered  sucrose  — 


under  the  guidance  of  Mary  E.  Rawles,  embedded  in  paraffin,  and  sectioned  at  4 

has  developed  trypsinization  procedures  or    5   microns.    Sections   were   then   re- 

that  permit  him  to  separate  the  wound  hydrated  and  dipped  in  Kodak  NTB-2 

epithelium  from  the  rest  of  the  wound  emulsion    according   to    Quastler.    After 

components.  drying,  the  coated  slides  were  stored  for 

Autoradiography  of  skin,  wounds,  and  10  days,  60  days,  and  6  months  before 

Ehrlich    ascites    tumor.    As    was    stated  being   developed   in   DK-19,    and   fixed 

above,   Argyris  hoped  to  develop  tech-  according  to  the  technique  of  Quastler. 

niques  for  labeling  the  damaged  tissues  The  developed  slides  were  then  counter- 

and  following  the  released  growth-pro-  stained  lightly  with  hematoxylin  before 

moting  substances  to  the  target  tissues,  being  mounted. 

To  accomplish  this,  some  prerequisites  Twelve  C57/Black  females  were  in- 
have  to  be  met  in  addition  to  learning  the  jected  intraperitoneally  with  tritiated 
technique  of  autoradiography:  (1)  an  water  of  low  specific  activity.  The  isotope 
isotope  must  be  found  which  enters  into  was  diluted  with  saline,  and  mice  were 
as  many  metabolic  pools  as  possible,  thus  injected  with  50,  100,  200,  1000,  2000,  or 
labeling  as  many  classes  of  substances  as  5000  /xc  each.  The  mice  were  killed  1,  5, 
possible;  (2)  the  distribution  of  the  and  8  hours  after  injection  of  the  isotope, 
labeled  material  in  the  normal  and  Biopsy  specimens  were  taken,  and  auto- 
damaged  skin  and  kidney  tissue,  and  in  radiograph  sections  were  prepared  from 
the  EAT,  must  be  determined.  them    as    described    above    for    sodium 

These  prerequisite  studies  have  been  acetate- 1-C14. 

carried  out  in  a  preliminary  fashion  for  Skin  with  resting  hair  follicles  shows 

resting  and  growing  skin,  for  wounds,  and  much  radioactivity  in  the  epidermis,  hair 

for    subcutaneously    inoculated    Ehrlich  follicles,  and  sebaceous  glands  within  1 

ascites    tumor,    the   labeled    compounds  hour  after  injection  of  the  labeled  acetate. 


DEPARTMENT   OF   EMBRYOLOGY  415 

Seven  days  after  initiation  of  hair  growth  differentiation  of  the  gonads,  and  in  the 

the  growing  hair  follicles  show  intense  study   of   the   reproductive   life   of   the 

activity,   as  does  the  hyperplastic   epi-  opossum  in  a  state  of  nature,  a  study 

dermis.  The  radioactivity  in  the  dermis  essentially  completed  with  accumulation 

and  subcutis  is  high  in  skin  with  either  of  sufficient  data  on  most  points  to  permit 

resting  or  growing  hair  follicles.  general  conclusions,  which  can  be  briefly 

Four  days   after  wounding,   the   epi-  summarized  as  follows : 

dermis  is  markedly  enlarged  as  the  result  1.  In    northern    Florida    breeding    is 

of  both  cellular  hypertrophy  and  hyper-  sharply  limited  to  a  period  of  2  to  3  weeks 

plasia.     The    hair    follicles    lose     their  in   late   January   and   the   first   half   of 

sebaceous  glands  and  are  converted  into  February,  more  than  80  per  cent  of  all 

hyperplastic    cords    of    cells.    Concomi-  births  occurring  then.  The  first  appear- 

tantly,    the    wound    is    filled    with    the  ance  of  young  is  very  constant  from  year 

hyperplastic  epidermis,  hair  follicle  cords,  to  year  and  apparently  has  no  relation  to 

and  granulation  tissue.  local  climatological  conditions,  suggesting 

The  EAT,  7  days  after  subcutaneous  that  in  all  probability  length  of  daylight 

inoculation,  is  composed  of  an  outer  rim  is  the  determining  factor  in  the  onset  of 

of  rapidly  proliferating  cells  and  a  necrotic  breeding. 

center.    Surrounding    the    tumor    is    a  2.  Young  females  may  produce  their 

tumor-induced    hyperplastic    connective  first  litters  at  an  age  of  6  to  7  months,  and 

tissue   sheath.    Except   for  the   necrotic  at  a  weight  of  less  than  1  kilogram,  some 

tumor  mass,  which  exhibits  little  activity,  2  years  before  they  attain  their  final  size, 

all  these  areas  show  high  radioactivity  Nevertheless,  the  average  litter  closely 

within    1    hour    after    injection    of    the  approximates    the    average    for    mature 

labeled  acetate.  females. 

In  all  the  materials  studied — normal  3.  In  spite  of  the  mild   climate  and 

skin,  wounds,  or  tumor — the  pattern  and  relatively  favorable  conditions  in  Florida, 

intensity  of  radioactivity  are  the  same  at  the  average  number  of  young  opossums 

8  and  24  hours  as  at  1  hour.  per  litter  is  definitely  lower  than  in  any 

The  distribution  of  the  labeled  water  other  part  of  the  United  States  for  which 
at   1,  5,  and  8  hours  after  injection  is  data  are  available.  This  number  is  about 
similar    to    that    of    the    acetate.    The  6.2  based  on  more  than  200  litters, 
activity,    however,    is    not    as    intense,  4.  The  opossum  in  Florida  does  not 
irrespective  of  dosage.  Also,  in  contrast  seem  bound  in  any  way  to  a  definite 
to  the  labeled  acetate,  which  shows  high  breeding  territory  or  a  home  range.  Year 
activity  within  1  hour  after  injection,  the  after  year  trapping  data  reveal  a  corn- 
activity  of  tritiated  water  is  not  high  pletely  new  population  of  adults  in  the 
until  8  hours.  This  gradual  increase  in  same  area.  It  is  extremely  rare  to  recap- 
activity  from  1  to  8  hours  after  injection  ture  an  adult  of  either  sex  after  the  lapse 
is  evident  in  mice  injected  at  all  doses  of  a  year.   This  does  not  preclude  the 
tested,  ranging  from  200  to  5000  micro-  possibility  that,  over  shorter  periods  of 
curies.  time,  an  animal  may  remain  in  a  neigh- 
borhood for  a  while  because  of  plentiful 
Reproduction  and  Sex  food  or  other  favorable  conditions. 
Differentiation  in  the  Opossum  The  study  of  histogenesis  in  the  embry- 
onic   gonads    continues    to    follow    the 

R.  K.  Burns's  studies  on  young  opos-  patterns  of  previous  years.  The  testis  is 

sums  in  Florida  consisted  entirely  in  a  readily  transformed  under  the  influence 

continuation  of  the  experiments  of  recent  of  estradiol  into  an  ovotestis  or  an  almost 

years    on  the    effects    of  administration  typical    ovary.     In    these    experiments 

of   steroid    sex   hormones    on   the   early  attention  has  been  chiefly  concentrated 


416  CARNEGIE     INSTITUTION     OF      WASHINGTON 

on  the  role  of  the  secondary  sex  cords  germ  cells  do  not  long  survive,  and  the 

(cortical    cords)    in    the    survival    and  developing  cortex  is  sterile.  Administra- 

multiplication  of  the  germ  cells.  Large  tion   of   androgens,    in    an   attempt   to 

doses    of    estradiol    typically    result    in  transform    the    embryonic    ovary,    has 

delayed  appearance  of  the  cortical  cords;  continued  to  yield  negative  results. 


BIOCHEMICAL   CHANGES  DURING   METAMORPHOSIS 

Acid  Deoxyribonuclease  in  gross  signs  of  tail  resorption  were  evident 

Amphibian  Metamorphosis  un+txl  +10/°  U  d+ays'  at  w^f  tlmf.  the 

extent  of  resorption  varied  from  disap- 

In  Year  Book  60,  pages  401-402,  John  pearance  of  the  tail  fin  with  no  shortening 

R.  Coleman's  preliminary  findings,  indi-  of  the  tail  to  25  per  cent  reduction  in  tail 

eating  that  Rana  pipiens  tadpole  tails  length.  The  only  tadpole  that  survived 

exhibit  an  increase  in  acid  deoxyribonu-  14    days    of    thyroxine    treatment    had 

clease  (DNase  II)  activity  concomitant  undergone  43  per  cent  tail  resorption, 

with  resorption  of  the  tail  at  metamor-  DNase  II  activities  in  the  tails  of  the 

phosis,  were  described.  Since  this  enzyme  thyroxine-treated  tadpoles  were  not  sig- 

is  one  of  the  hydrolytic  group  of  enzymes  nificantly   different   from   those    of   un- 

found  in  rat  liver  lysosomes  by  de  Duve  treated  tadpoles  except  in  tails  that  had 

and    co-workers    and    is    generally    con-  begun  to  be  resorbed  (fig.  27).  But  when 

sidered  to  serve  more  of  a  catabolic  than  the  DNase  II  activity  was  plotted  as  a 

an   anabolic   role   in   the   cell,    a   more  function    of   the    time   in   thyroxine   it 

rigorous    examination    of    its    changes  appeared  to  have  reached  a  maximal  rate 

during    tail    resorption    was    considered  of  increase  by  8  days  (fig.  28),  before  any 

worth  while.   For  convenience,  and  be-  extensive  resorption  had  occurred.  Tails 

cause  this  project  was  undertaken  in  the  in  the  process  of  resorption  (11-14  days) 

autumn  when   R.   pipiens  tadpoles  are  showed  a  2-  to  3-fold  increase  in  enzyme 

difficult  to  obtain,  the  investigation  was  activity  over  the  average  value  obtained 

conducted    on    R.    catesbeiana   tadpoles,  from  tails  not  in  the  process  of  resorption, 

which  are  commercially  available  (Caro-  These  results  are  in  agreement  with  the 

lina  Biological  Supply  Co.).   The  tech-  earlier    findings    of    Weber    that    the 

niques  for  enzyme  assay  were  essentially  specific  activity  of  cathepsins  begins  to 

as  described  in  last  year's  report.  rise  in  the  Xenopus  tadpole  tail  shortly 

Tadpoles  about  to  undergo  spontaneous  before  metamorphosis.  However,  Weber 

metamorphosis  were  not  available.  Assays  found  a  logarithmic  increase  in  cathepsin 

of  DNase  II  activity  in  all  the  available  activity  throughout  resorption,  which  is 

stages    showed    considerable    variability  not   true   for   DNase    II   activity.    The 

from  tadpole  to  tadpole  with  no  signifi-  changes    in    DNase    II    activity    more 

cant  trend  of  increase  or  decrease  with  closely    parallel    the    changes    in    acid 

advancing   development    (fig.    27,    open  phosphatase  activity  in  Xenopus  shown 

symbols).    To    obtain    tadpoles    in    the  by  Weber  and  Niehus  than  changes  in 

process  of  tail  resorption,  thyroxine  (Na  cathepsin  activity. 

salt,  California  Corporation  for  Biochem-  No  evidence  is  available  about  the 
ical  Research)  was  added  to  the  medium  possible  lysosomal  compartmentalization 
in  a  final  concentration  of  0.1  jug/ml.  of  these  enzymes  in  tadpole  tail  tissues, 
Under  these  conditions,  measurable  but  their  patterns  of  activity  are  con- 
growth  in  both  hind  limb  and  tail  sistent  with  the  theory  that  they  play 
occurred  within  the  first  4  to  6  days.  No  primarily  a  catabolic  role. 


DEPARTMENT  OF  EMBRYOLOGY 


417 


z 

I- 
O 

cc 

o» 

E 


W 


o 


CO 

H 

Z 


1500 

4 

+ 

♦ 

1000 

A* 

• 

• 

• 
A 

A 

500 

•• 

AA             o 

o 
o 

•a 

i 

A 
i 

•             i 

i                wm m 

.3 


.4 


.5 


,6 


LEG    /     TAIL 


Fig.  27.  DNase  II  activity  per  milligram  of  protein  in  isolated  tadpole  tails  at  various  stages  of 
development  as  indicated  by  the  ratio  of  hind-limb  length  to  tail  length.  Shaded  symbols  represent 
tadpoles  kept  in  the  presence  of  0.1  jug  of  thyroxine  per  milliliter  for  various  periods;  open  symbols 
refer  to  nonthyroxine-treated  tadpoles.  Each  point  represents  duplicate  assays  on  a  single  tail. 
Triangles  indicate  determinations  made  before  thyroxine  experiment;  open  circles,  control  determina- 
tions carried  out  in  parallel  with  thyroxine  experiment;  closed  circles,  determinations  on  tails  of 
thyroxine-treated  tadpoles  before  resorption  had  begun;  closed  circles  with  crosses,  same,  but  under- 
going resorption. 


An  Amylase  Activity  in  Rana  pipiens 
Serum 

Frieden  and  co-workers  have  described 
the  appearance  of  an  albumin  in  serum  of 
amphibians  during  metamorphosis  and 
have  suggested  that  the  albumin  appear- 
ing during  metamorphosis  is  a  prepara- 
tion for  the  terrestrial  environment  of  the 
adult  inasmuch  as  an  increased  amount 
of  osmotically  active  material  in  the 
blood  would  aid  in  the  maintenance  of 
blood  volume.  These  earlier  studies  of 
blood  proteins  had  been  made  by  means 


of  paper  electrophoresis,  and  G.  L. 
Carlson  thought  it  useful  to  reexamine 
the  findings  taking  advantage  of  the 
greater  resolution  of  the  starch  gel 
electrophoretic  technique. 

In  electrophoresis  of  adult  Rana  pipiens 
sera  on  starch  gel,  an  unusual  degradation 
of  the  starch  near  the  point  of  applica- 
tion of  sera  demonstrated  the  existence 
of  glucan  hydrolase  (amylase)  activity  in 
the  sera.  Of  the  proteins  resolved  on 
starch  gel  electrophoresis,  amylase  has 
been  studied  most  thoroughly  to  date. 


418 


CAKNEGIE     INSTITUTION     OF      WASHINGTON 


111 

»- 
O 

<r 

CL 

o> 

E 


H 

</> 
O 

z 

Q 
CO 


1500 

0 

/  ° 

1000 

0 

^8 

o  / 
o/ 

0 

500 

> 

0 

O 

o 

1            _l  __ 

i 

1 

1     1  , 

8 


10        12 


14 


DAYS    IN   THYROXIN  (O.lmg/ml) 


Fig.  28.  DNase  II  activity  per  milligram  of  protein  in  isolated  tadpole  tails  after  exposure  of 
tadpoles  to  thyroxine  for  various  periods.  Thyroxine  was  added  to  the  medium  in  a  final  concentration 
of  0.1  Aig/mh  The  medium  was  changed  every  2  to  3  days.  Each  point  represents  duplicate  assays 
on  single  isolated  tails  except  at  0  days,  where  the  point  represents  an  average  of  all  control  data 
shown  in  figure  27.  No  tail  resorption  was  apparent  before  10  days  in  thyroxine,  but  by  11  days  all 
tails  exhibited  some  degree  of  resorption. 


A  quantitative  assay  for  the  amylase 
was  devised  taking  as  a  measure  of 
enzyme  activity  the  appearance  of  com- 
pounds reducing  cupric  ion  after  incuba- 
tion of  aliquots  of  serum  with  potato 
starch  in  pR  7.5  tris  buffer  at  30°C  for  25 
minutes.  Under  these  conditions,  a  linear 
proportionality  between  the  amount  of 
serum  added  and  the  amount  of  reducing 
material  liberated  could  be  shown.  With 
color  developed  from  the  reduced  copper 
by  means  of  Nelson's  arsenomolybdate 
reagent,  it  is  easily  possible  to  assay  the 


amount  of  amylase  activity  in  as  little  as 
1  microliter  of  adult  serum.  The  assay 
conditions  were  chosen  after  determining 
that  the  enzyme  activity  of  adult  serum 
had  a  pH  optimum  between  7.5  and  8.0. 
On  a  volume  basis  it  is  interesting  that 
the  amylase  activity  in  adult  R.  pipiens 
serum  obtained  by  heart  puncture  is 
about  one-tenth  that  of  human  saliva. 

Further  definition  of  the  nature  of  the 
amylase  activity  was  obtained  by  paper 
chromatographic  examination  of  the 
products  of  the  action  of  adult  R.  pipiens 


DEPARTMENT    OF    EMBRYOLOGY 


419 


serum  amylase  on  glycogen  at  various 
times  during  hydrolysis.  When  aliquots  of 
reaction  mixtures  were  chromatographed 
with  an  isopropyl  alcohol-water  (4:1) 
solvent  system  and  reducing  compounds 
were  detected  with  alkaline  silver  nitrate 
reagent,  a  complex  mixture  of  reaction 
products  was  observed  as  expected  of 
endoamylase  (a-amylase)  activity.  The 
chromatographic  pattern  of  reducing 
products  compared  well  with  that  ob- 
tained by  digestion  of  glycogen  to 
approximately  the  same  fraction  of  free 
reducing  groups  with  a  human  salivary 
a-amylase  preparation,  suggesting  strong- 
ly that  the  frog  serum  amylase  consists 
mainly  of  a-amylase (s).  Reducing  ma- 
terial with  an  Rf  of  maltose  in  this  solvent 


system  is  formed  in  quantity  later  than 
the  slower-moving  dextrins,  and  it  there- 
fore seems  highly  unlikely  that  a  /3- 
amylase  (exoamylase)  is  present.  This 
conclusion  is  supported  by  the  observa- 
tion that  chromatograms  of  a  reaction 
mixture  of  maltose  incubated  with  R. 
pipiens  serum  show  that  only  glucose  is 
produced  as  a  reaction  product,  and  this 
reaction  proceeds  at  a  very  slow  rate. 

A  quantitative  analysis  of  the  levels  of 
serum  a-amylase  of  R.  pipiens  during 
metamorphosis  is  now  being  made  in 
order  to  relate  the  serum  amylase  levels 
to  changes  in  /3-amylase  activity  previ- 
ously noted  by  E.  Urbani  in  studies  on 
homogenates  made  from  amphibian  tad- 
poles during  metamorphosis. 


THE   EMBRYO   IN   RELATION   TO   ITS   ENVIRONMENT 


Mechanisms  of  Implantation 
of  the  Ovum 

In  the  rabbit.  As  has  been  reported  in 
Year  Books  58  (pp.  368-370),  59  (pp. 
359-362),  and  60  (pp.  431-432),  Bent  G. 
Boving's  proximate  objective  is  to  ana- 
lyze the  mechanisms  of  implantation  of 
the  rabbit  blastocyst.  He  has  elected  to 
study  a  single  species  rather  than  adopt 
the  traditional  comparative  approach. 
Fragments  of  knowledge  of  the  implanta- 
tion process  suggest  that  there  is  con- 
siderable species  variation;  it  appears 
likely,  therefore,  that  a  better  view  of  the 
implantation  process  as  a  whole  may  be 
obtained  by  following  it  in  one  species. 
Ultimately,  we  should  like  to  know  to 
what  extent  the  major  conclusions  from 
studies  of  the  rabbit  apply  to  other 
species,  particularly  the  macaque  and  the 
human. 

As  Boving  has  shown,  implantation 
embraces  a  series  of  mechanisms  grouped 
by  him  into  three  main  categories: 
muscular,  adhesive,  and  invasive.  The 
last  two  (adhesion  of  trophoblast  and 
uterine  epithelium,  and  penetration  of 
epithelium  by  trophoblast)  have  been 
shown  to  occur  selectively  where  there  is 


a  capillary  at  the  base  of  the  epithelium. 
Both  phenomena  have  been  attributed  to 
a  local  alkalinity  elicited  when  bicarbo- 
nate from  the  blastocyst  passes  through 
the  epithelium  and  there  dissociates  into 
(alkaline)  carbonate  and  carbonic  acid, 
which  with  enzymatic  assistance  dissoci- 
ates into  water  and  carbon  dioxide,  the 
carbon  dioxide  being  removed  by  the 
maternal  circulation. 

During  the  year  covered  by  this  report, 
Boving  has  centered  much  of  his  attention 
on  the  uterine  lumen  and  the  uterine 
epithelium. 

Little  is  said  of  the  uterine  lumen  in 
most  treatises,  and  what  is  said  is  largely 
incorrect,  for  there  are  several  popular, 
but  mistaken,  impressions  about  it.  In 
particular,  it  is  supposed  to  be  occupied 
by  "free"  blastocysts  and  uterine  fluid. 
The  "free"  blastocysts  not  only  are 
confined  within  the  uterus  but  also 
depend  on  it  for  their  motion  and  pre- 
sumably for  some  aspects  of  their 
metabolic  maintenance.  That  is  "free- 
dom." When  the  uterus  of  the  rabbit  is 
opened  at  the  time  of  implantation,  its 
lining  is  covered  with  a  thin  layer  of 
mucus  that  is  barely  moist.  That  is 
"uterine  fluid."  Still,  what  there  is  of  it 


420  CARNEGIE     INSTITUTION      OF      WASHINGTON 

is    possibly    interesting,     if    the    usual  Boving's  estimate  of  the  distribution  of 

assumption  is  correct  that  all  chemical  fluid    injected    into    the   uterine   lumen 

exchange  between  mother  and  unattached  provides  the  background  for  a  study  of 

blastocyst  occurs  through  the  fluid  rather  the  effects  on  cohesion  of  uterine  epi- 

than  directly  in  regions  of  contact.  Boving  thelium    of    various    solutions    in    the 

believes    that,    except    for    the    rather  carbonate  family.  His  working  hypothesis 

viscous    uterine    secretion    and    debris  (that   bicarbonate   from   the   blastocyst 

remaining  adherent  to  the  blastocyst  as  yields  carbonate  in  epivascular  epithelial 

the    gloiolemma,    any    fluid    interposed  cells,  where  it  raises  the  pH  and  so  causes 

between  epithelium  and  blastocyst  would  epithelial  dissociation)  suggested  to  him 

be  squeezed  out  to  unoccupied  segments  that  a  bicarbonate  solution  of  appropriate 

of  the  uterus  as  the  blastocyst  expands  concentration    (0.15    N)    in   the    lumen 

and  pushes  out  the  uterine  wall  around  it  might  produce  the  same   effect.   Slight 

to  form  its  dome.  dissociation  was  noted,  but  the  evidence 

The  amount  of  uterine  fluid  in  a  uterine  was  not  convincing.  Earlier,  Boving  had 

horn  was  estimated  by  Kulangara  (Year  observed  that  general  epithelial  disrup- 

Book  59,  p.  361)  to  be  about  0.1  to  0.5  ml.  tion  was  produced  by  0.05  N  NaOH.  It 

Assuming  a  length  of  200  mm,  the  lumen  seemed    possible,     therefore,     that    the 

accommodating    that    amount    of    fluid  bicarbonate  solution  was  being  handled 

would  have  an  average  cross-section  area  in  a  way  significantly  different   (mixed 

of  0.5  to  2.5  mm.  The  open  spaces  seen  in  with  uterine  fluid  and  buffered?)   from 

histological  sections  are  generally  greater,  bicarbonate    coming   from   a   blastocyst 

the    difference    being    almost    certainly  held  in  apposition  to  uterine  epithelium, 

evidence     of     tissue     shrinkage     during  Hence  it  became  necessary  to  provide  a 

preparation.  Boving  has  developed  a  new  direct  test  of  the  action  of  0.15  N  Na2C03 

way  of  visualizing  the  distortion;  it  is  instead  of  depending  on  it  to  be  produced 

derived  from  a  test  of  the  distribution  of  in  the  epithelium  after  injecting  NaHC03 

a  small  amount  of  fluid  injected  into  the  solution  into  the  uterine  lumen, 

uterine  lumen.  About  0.5  ml  of  0.5  per  As  has  been  reported  earlier,  there  was 

cent  AgN03  solution  was  injected  into  a  little  effect  of  bicarbonate  solution,  even 

segment  of  uterus  about  20  mm  long,  when  charged  with  C02,  and  little  effect 

That  is  to  say,  the  segment  had  added  to  of  Na2C03  charged  with  C02;  but  there 

it  about   10  times  as  much  fluid  as  it  was  a  whole  range  in  severity  of  effects 

would  normally  contain.  Even  with  that  produced  by  alkaline  Na2C03  (pis.  7,  8, 

excess,  the  fluid  stained,  and  presumably  figs.  31,  32,  33).  Except  for  differences  in 

touched,   only   the   tips   of   endometrial  degree,  the  kinds  of  effects  observed  were 

folds   nearest  the   center  of  the  uterus  similar  to  those  seen  in  the  invasion  by 

(pi.  6,  figs.  29,  30).  Since  the  now  open  normal    trophoblast    at    7    days    after 

"glands"  were  not  entered  by  the  fluid,  mating.  Not  only  does  carbonate  dissoci- 

we  may  reason  that  they  were  closed  in  ate    the    epithelium    (and    occasionally 

the  living  state.   That  relation  may  be  stroma)    but    it    generally    spares    the 

reconstructed  if  we  envision  the  black-  underlying  blood  vessels  (pi.  8,  fig.  33). 

stained    tips    of    the    endometrial    folds  (It  is  a  day  or  two  later  that  the  blood 

pulled  together  to   compensate  for  the  channels  normally  lose  the  endothelium, 

shrinkage  that  pulled  them  apart.  The  after    they    have    been    enveloped    by 

remaining  space  may  then  be  considered  trophoblast.)    Thus,   not   only  does   the 

to    be    10    parts    artifact    produced    by  experiment    lend    support    to    Boving's 

injecting  the  AgN03  solution  and  1  part  hypothesis;    it    also    points    out    that 

uterine  fluid,  representing  the  extent  of  maternal  tissues  have  different  degrees  of 

the  uterine  lumen  in  vivo.  The  uterine  susceptibility  to  dissociation, 

lumen,  then,  is  small  in  life.  Previously,    the    idea    of    differential 


DEPARTMENT    OF    EMBRYOLOGY  421 

dissociability   had   been   applied   to   the  observed  previously,  both  histologically 

invasion  of  trophoblast  only  in  the  all-or-  and    in    an    experiment    with    a    plastic 

none  sense:  penetration  was  explained  as  chamber  in  vivo,  mentioned  in  Year  Book 

the  persistence  of  trophoblast  by  reason  59  (p.  362)  as  a  possible  basis  for  the  rise 

of  the  syncytial  structure  of  its  invading  in  the  concentration  of  protein  in  the 

knobs,    which   resist   dissociation   in   an  uterine   fluid   that   Kulangara  found   to 

alkaline    microenvironment    where    the  begin  about  4  days  after  mating.  Could 

epithelial  cells  suffer  loss  of  cohesion.  It  it  be  that  the  blastocyst  "turns  on"  the 

remains  to  be  explained  why  the  cellular  increase  of  uterine  fluid  protein  by  a  pH 

trophoblast    between    knobs    does    not  mechanism?  The  idea  is  difficult  to  test, 

dissociate,    although   it   may   adhere   to  for  comparisons  of  protein  concentration 

epithelium.  Conceivably,  it  may  have  a  of  fluids  in  empty  vs.  gravid  horns  must 

high  degree  of  resistance  to  dissociation  at  get   around   the   possibility  that   higher 

high  pH,  as  does  the  maternal  capillary  values  in  the  gravid  horn  may  derive  from 

endothelium.  That  idea  is  being  tested,  blastocysts,  living  or  dead,  rather  than 

along   with    the    idea   that    trophoblast  from    the    uterus    on    which    they    are 

spread  is  arrested  by  the  uterine  epi-  presumed    to    act.    Further    caution    is 

thelium's   becoming   converted   by   pro-  suggested  by  evidence  of  epithelial  shed- 

gesterone  into  a  syncytium  and  thereby  ding  in  cysts  (pi.  6,  figs.  29  and  30),  where 

becoming   insusceptible   to   dissociation,  it  may  be  assumed  no  blastocyst  reached. 

The    experiment    of    putting    carbonate  Such   cysts,    surprisingly   common   once 

solution  into  rabbit  uteri  at  9  days  after  one  is  alert  to  them,  may  also  be  contem- 

mating  has  been  done,  but  the  manipu-  plated  as  a  source  of  protein  of  epithelial 

lation  was  suspected  of  being  improper,  cell  origin,  but  that  idea  should  not  be 

and  repetition  is  planned.  taken  too   seriously   until   timed   histo- 

The   foregoing   questions   of   dissocia-  logical    studies    confirm    that    they    are 

bility  will  be  approached  by  technical  formed  and  rupture  at  an  appropriate 

means  that  may  be  expected  to  provide  time. 

data  on  the  amounts  of  force  necessary  to  In  the  mouse.  The  factors  effecting  the 
separate  the  tissues  and  to  determine  orientation  and  spacing  of  blastocysts  in 
whether  the  binding  strength  varies  with  the  mouse  uterus  have  never  been  clearly 
pH.  A  Duryee-Bush-Hastings  electric  delineated.  It  is  known  that  the  normal 
micromanipulator  is  being  adapted  as  a  orientation  of  the  blastocyst  and  the  sites 
microtension  measuring  device.  The  asso-  of  implantation  are  not  affected  by 
ciated  equipment  for  cinematographic  reversal  of  the  gravitational  axis  of  the 
recording  is  essentially  complete  and  has  uterus.  Evidence  for  and  against  un- 
performed adequately  in  a  single  pre-  plantation's  being  invariably  antimeso- 
liminary  test.  metrial   because   of   the   influence    of   a 

The     question     whether     bicarbonate  chemotactic    stimulus     (from    epithelial 

forms    carbonate    in    the    epithelium    is  lipides)  has  been  presented  in  the  litera- 

being  approached  not  just  by  the  substi-  ture,  and  the  balance  seems  to  be  against 

tution  methods  described  but  also  ana-  such  a  stimulus.  Spacing  of  blastocysts 

lytically.  Nickel  chloride  has  been  found  along  the  length  of  the  uterus  (whether 

to  precipitate  sodium  carbonate  but  not  even  or  random)  is  thought  to  be  caused 

bicarbonate.  A  perfusion  with  it  has  been  by  peristaltic  movements  of  the  uterine 

done,    but    the    histology,    which    is    to  smooth  muscle. 

depend    on    development    with    sodium  In  earlier  experiments   involving   the 

sulfide,  is  just  beginning.  transplantation  of  melanotic  tumor  tissue 

Finally,  it  should  be  noted  that  the  to  the  pseudopregnant  mouse  uterus,  Ian 

peeling  of  epithelium  (pi.  7,  figs.  31  and  Wilson  found  that  such  grafts  "implant" 

32)  induced  by  carbonate  resembled  that  in  the  endometrium  invariably  antimeso- 


422 


CARNEGIE      INSTITUTION      OF      WASHINGTON 


metrially,  as  do  blastocysts.  Grafts  of 
muscle  tissue  (autografts  of  body  wall) 
also  "implant"  at  the  same  site,  and  it 
has  been  reported  that  even  wax  or  glass 
beads,  when  inserted  into  the  uterus, 
become  antimesometrially  located. 

The  above  observations,  considered 
together,  suggest  that  the  antimesome- 
trial  siting  of  implantation  is  brought 
about  by  a  simple,  nonspecific  mechanism. 
It  is  possible  that  the  uterine  contents, 
whether  blastocysts  or  wax  beads,  are 
forced  into  the  antimesometrial  area 
simply  by  compression  of  the  uterine 
lumen  caused  by  contraction  of  the 
circular  muscles. 

Melanoma  cells  in  suspension,  injected 
into  the  uterus,  segregate  into  clumps, 
often  spaced  out  along  the  uterus  exactly 
as  blastocysts  are.  This  finding  supports 
the  idea  that  spacing,  in  the  mouse,  is  a 
result  of  random  scattering  of  the  uterine 
contents  by  peristaltic  muscular  move- 
ments and  not  a  result  of  specific  inter- 
actions between  the  blastocyst  and  the 
uterus. 

It  was  thought  that  a  direct  test  of  the 
effect  uterine  muscular  movements  exert 
on  the  spacing  and  localization  of 
implantation  sites  might  be  made  by 
injecting  mice,  in  early  pregnancy,  with 
a  drug  that  would  inhibit  the  activity  of 
uterine  smooth  muscle.  Such  a  drug, 
isoxsuprine  [2-  (phenoxypropylamino)  - 1- 
('p'-hydroxyphenyl)-l-propanol-HCl],has 
become  available  recently.  It  is  reported 
to  have  a  relatively  specific  inhibitory 
effect  upon  the  smooth  muscles  of  the 
uterus  (as  opposed  to  the  gut),  although 
it  is  also  a  vasodilator. 

It  was  supposed  that  the  muscle 
activity  that  causes  spacing  of  the  blasto- 
cysts preceded,  and  was  relatively  inde- 
pendent of,  the  activity  that  causes  their 
antimesometrial  propulsion.  To  separate 
these  two  stages  for  experimental  work  it 
was  necessary  to  find  out,  from  a  series  of 
normal  pregnancies,  the  exact  location  of 
the  blastocysts  (and,  incidentally,  their 
orientation)  at  any  given  time  before 
their  implantation. 


Wilson's  preliminary  observations  show 
that,  in  the  period  up  to  about  90  hours 
postcoitum,  blastocysts  become  randomly 
located  throughout  the  uterine  lumen; 
they  may  be  found  in  the  mesometrial  or 
antimesometrial  region,  and  their  embry- 
onic-abembryonic  axis  is  haphazardly 
oriented  (pi.  9,  fig.  34).  From  90  hours 
postcoitum  onward  the  blastocysts  are 
invariably  found  in  antimesometrial 
1  'pockets"  properly  oriented  with  their 
embryonic  mass  toward  the  mesometrium 
(i.e.,  orientation  does  not  occur  until  the 
blastocysts  occupy  their  prospective  im- 
plantation sites).  It  is  clear  that,  at  this 
time,  the  uterine  lumen  is  completely 
occluded  except  for  the  small  antimeso- 
metrial pockets  containing  the  blasto- 
cysts. Although  the  walls  of  the  lumen 
are  pulled  apart  during  fixation  and 
mounting  of  the  sections,  the  silhouettes 
of  the  lateral  walls  are  complementary; 
on  the  free  surface  of  the  epithelial  cells 
(except  those  in  the  ' 'pocket")  there  are 
present,  during  this  period,  small  conical 
protrusions  that  correspond  exactly  with 
small  depressions  on  the  opposite  side  of 
the  lumen  (pi.  9,  fig.  35). 

The  dose  of  isoxsuprine  needed  to 
inhibit  uterine  contractions  is  being 
measured  by  observations  in  vivo  upon 
the  rabbit  and  the  rat,  the  mouse  being 
too  small.  It  is  hoped  to  test  the  effect  of 
the  drug  during  the  two  established 
periods  of  early  pregnancy:  (1)  the  period 
during  which  the  blastocysts  become 
randomly  scattered  in  the  uterus,  and 
(2)  that  in  which  they  become  antimeso- 
metrially located. 

Trial  injections  of  isoxsuprine  into  5 
mice  (over  days  3  and  4  of  pregnancy) 
inhibited,  or  interrupted,  pregnancy  in  4. 
In  all  5  animals  the  blastocysts  had 
reached  their  prospective  implantation 
sites.  In  2  animals  the  blastocysts  were 
judged  to  be  healthy  but  retarded  in 
development  by  about  36  hours  (no  egg 
cylinder  formed,  no  sign  of  any  reaction 
in  the  uterine  mucosa).  In  2  others  it 
appeared  that  the  blastocysts  had  actu- 
ally started  to  implant  when  the  vaso- 


DEPARTMENT  OF  EMBRYOLOGY 


423 


dilator  action  of  the  isoxsuprine  precipi- 
tated rupture  of  subepithelial  capillaries 
and  extravasation  of  blood  into  the 
uterine  lumen.  The  exudate  dislodged  the 
blastocysts  from  their  implantation  sites. 
In  1  animal  pregnancy  was  apparently 
unaffected  by  the  drug.  At  the  time  of 
these  trials,  however,  neither  the  inhibi- 
tory effect  of  the  dosage  used  nor  the 
stage  of  pregnancy  at  which  injections 
were  started  had  been  determined 
critically. 

Anatomy  and  Physiology 
of  the  Placenta 

The  team  of  E.  M.  Ramsey,  G.  W. 
Corner,  Jr.,  and  M.  W.  Donner  sustained 
a  severe  loss  in  the  fall  of  1961  in  the 
untimely  death  of  their  technical  asso- 
ciate Herbert  M.  Stran.  Mr.  Stran's 
understanding  of  biology  and  of  the  point 
of  view  of  scientific  investigation,  as  well 
as  his  training  and  skill  in  mechanics  and 
electronics,  made  him  uniquely  valuable. 
His  death  constitutes  a  scientific  loss  as 
well  as  a  source  of  personal  regret. 

Problems  associated  with  the  move  to 
the  new  laboratory  were  agreeably  few 
and  easy  of  solution,  thanks  to  the  valued 
assistance  of  Arthur  G.  Rever  and 
William  I.  Cleary  and  his  staff. 

Adjustment  to  new  conditions  has  been 
the  rule  on  all  fronts  throughout  the  year. 
During  the  winter  and  spring  the  Depart- 
ment of  Radiology  at  the  Johns  Hopkins 
Hospital  installed  new  and  improved 
equipment  for  cineradioangiography. 
Ramsey  and  her  associates  assisted  by 
providing  monkeys  for  tests  and  calibra- 
tion studies  involving  higher  radiation 
dosage  than  is  permissible  for  use  in 
human  patients. 

In  the  experimental  procedures  them- 
selves unanticipated  areas  of  ignorance 
became  manifest  requiring  reorientation 
of  the  schedule  of  operations.  It  was  found 
necessary,  before  all  else,  to  acquire 
background  information  about  labor  and 
parturition  in  the  rhesus  monkey. 

As  was  noted  in  Year  Book  60  it  had 


been  planned  that  the  first  studies  of 
placental  circulation  to  be  undertaken  in 
the  season  just  ended  would  deal  with  the 
final  stage  of  pregnancy.  Inasmuch  as 
symptomatic  diagnosis  of  impending 
parturition  in  the  monkey  cannot  often 
be  made  long  enough  in  advance  of  birth 
to  permit  adequate  study  of  the  process, 
it  was  proposed  to  induce  labor  at  full 
term  (as  judged  by  the  duration  of  the 
pregnancy)  by  administration  of  hor- 
mones. Unexpectedly,  it  was  found  that 
this  cannot  be  done  in  the  monkey  with 
the  same  ease  as  in  human  patients.  High 
dosage  and  prolonged  administration  of 
the  standard  oxytocic  drug  Syntocin 
failed  to  effect  adequate  cervical  dilation 
and  effacement.  Furthermore,  in  some  of 
the  extended  studies  the  drug  seemed  to 
be  incapable  of  maintaining  such  progress 
as  was  achieved.  As  for  the  activity  of  the 
uterus,  recordings  of  intraamniotic  pres- 
sure during  administration  of  Syntocin 
showed  progression  from  the  late-preg- 
nancy type  of  infrequent,  uncoordinated, 
myometrial  contraction  waves  of  medium 
strength  to  characteristic  rapid,  coordi- 
nated, high-amplitude  labor  waves,  but 
the  latter  type  of  activity  was  inter- 
mittent and  unsustained. 

These  observations  afford  renewed 
evidence  that  the  technique  of  intra- 
uterine pressure  recording  permits  accu- 
rate evaluation  of  clinical  status  and 
prediction  of  clinical  performance;  it  is 
gratifying  to  find  them  in  harmony  with 
previous  observations  on  the  early  and 
middle  stages  of  pregnancy,  yet  they  do 
present  a  serious  dilemma  as  far  as  the 
study  of  placental  circulation  at  term  is 
concerned.  First,  why  is  Syntocin  less 
effective  in  inducing  productive  labor  in 
monkeys  than  in  humans?  And  second, 
how  can  labor  be  diagnosed  and  studied? 
The  probable  explanation  appears  to  be 
that  the  attempts  to  initiate  labor  were 
made  before  physiological  "term"  had 
been  reached,  no  matter  what  the  chrono- 
logical age  of  the  pregnancy.  Almost 
nothing  is  known  about  prelabor  develop- 
ments in  the  monkey.  The  observations 


424  CARNEGIE     INSTITUTION     OF     WASHINGTON 

and  motion  pictures  of  parturition  made  in  early  pregnancy,  the  contour  of  the 

by  Carl  Hartman  more  than  30  years  ago  contraction  wave  indicates  a  high  degree 

remain  the  only  recorded  observations,  of  myometrial  incoordination.  Under  such 

and  they  are  entirely  gross  and  external  conditions    the    uterus    is    incapable    of 

and  deal  with  the  final  stage  of  birth  emptying  itself.  On  the  other  hand,  the 

alone.  The  current  pressure  studies  have  contraction  wave  of  midpregnancy  indi- 

at  least  made  it  possible  to  state  when  cates  relative  myometrial  coordination, 

the  myometrium  is  not  ready  to  respond  but  the  contractions  are  of  insufficient 

promptly  to  oxytocics  and  to  expel  the  amplitude  and  frequency  to  expel  the 

fetus.  Although  negative,  this  information  uterus.  The  mounting  myometrial  activ- 

should  make  it  possible  in  the  future  to  ity  associated  with  preabortion  or  prelabor 

study  the  response  to  hormones  of  the  is  indicated  by  the  return  of  a  complex 

"ready"  myometrium  on  a  basis  of  better  wave  which  gradually  breaks  up  into  true 

understanding  of  the  animal's  physiology  labor    units.    The    simplicity,    the    high 

at   the   end   of   pregnancy.    Since   it   is  amplitude,  and  the  rapidity  of  the  labor 

impractical  to  diagnose  impending  labor  wave  show  that  both  myometrial  coordi- 

by  the  fairly  elaborate  process  of  intra-  nation  and  force  have  attained  the  degree 

uterine  pressure  recording,  intensive  clin-  necessary  to  effect  delivery, 

ical  studies  have  been  made  in  pregnant  3.  In  speculating  on  the  possible  role 

monkeys  approaching  term  for  the  pur-  of  progesterone  in  the  observed  phenom- 

pose  of  developing  criteria  that  can  be  ena,   Csapo's  suggestion  that  the  local 

established  by  vaginal  palpation  of  the  action    of    progesterone    may    enhance 

cervix  in  the  unanesthetized  animal.  muscle  incoordination  is  to  be  considered. 

The  urgency  of  the  above  problem  has  The  monkey's  bidiscoid  placenta,  supply- 
made  it  necessary  to  defer  radioangio-  ing  two  foci  of  progesterone  formation, 
graphic  study  of  venous  drainage  of  the  certainly  contributes  to  uterine  asym- 
placenta,  which  was  the  second  series  of  metry.  In  addition,  the  marked  reduction 
studies  planned  for  the  current  season,  as  in  uterine  circulation  just  before  conver- 
noted  in  Year  Book  60.  sion,     found    by    Reynolds,     tends    to 

In  reviewing  for  publication  the  data  enhance  this  action, 

assembled   in   the    1960   season   on   the  The  observation,  previously  reported 

effect  of  administration  of  progestins  at  (Year  Book  59),  of  the  slight  quieting 

various  stages  of  pregnancy  it  was  found  effect    of   progesterone    administered    in 

possible     to     amplify    the     preliminary  early  pregnancy  can  then  be  explained  on 

statements  made  immediately  after  the  the  basis  that,  because  of  poor  circulation, 

conclusion  of  the  experiments  and,  setting  endogenous  progesterone  is  reaching  the 

the   observations   in   the   framework   of  muscle  in  so  reduced  a  quantity  that  the 

recent    studies    by    other    workers,    to  additional  dosage  counteracts  the  asym- 

propose  the  following  generalizations.  metry.   In  midpregnancy,  when  uterine 

1.  In  terms  of  amplitude  and  rate  of  conversion  has  reestablished  circulation, 
myometrial  contraction,  activity  at  the  and  the  myometrium  already  has  a  full 
beginning  and  end  of  pregnancy  is  more  supply  of  progesterone,  added  quantities 
intense  and  rapid  than  in  midpregnancy.  merely  increase  the  surplus  of  proges- 
In  midpregnancy,  rates  and  intensities  terone  and  can  have  no  additional  effect, 
similar  to  those  prevailing  at  the  begin-  It  may  be  postulated  that  the  reduced 
ning  and  end  foreshadow  labor  or  circulation  of  the  late-pregnant  uterus 
abortion.  permits  the  two  foci  of  placental  produc- 

2.  The  contour  of  the  contraction  wave  tion  of  progesterone  to  reestablish  their 
is  as  important  as  its  amplitude  and  dominance.  Thus  there  is  a  return  of  the 
frequency.  Despite  the  strength  and  complex  wave  pattern.  Then,  as  the 
frequency  of  the  myometrial  contractions  uterus   gradually   escapes   from   proges- 


DEPARTMENT    OF    EMBRYOLOGY 


425 


terone  domination,  coordinated  myome- 
trial  activity  supervenes.  Until  a  clearer 
overall  picture  is  available,  however,  such 
speculations  upon  the  role  of  progesterone 
in  the  control  of  myometrial  activity 
cannot  be  considered  more  than  provoca- 
tive conjectures. 

Study    of  Human  Placental    Vasculature 

During  his  year  in  the  Department, 
J.  W.  S.  Harris  made  good  progress  in 
working  up  the  pregnant  human  uteri  in 
the  Carnegie  Collection.  This  material 
has  been  accumulated  over  a  period  of 
years  with  the  ultimate  objective  of 
preparing  a  study  of  the  vasculature  of 
the  human  placenta  along  lines  parallel 
to  the  study  of  the  placenta  of  the  rhesus 
monkey  previously  carried  out  by 
Ramsey. 

It  was  envisaged  that  the  human  study 
would  fall  into  three  phases : 

1.  Analysis  of  available  material.  The 
material  consists  of  specimens  submitted 
to  the  Department  either  as  unsolicited 
contributions  or  by  obstetrical  colleagues 
throughout  the  country  who  have  coop- 
erated in  preparing  and  injecting  uteri 
specifically  for  the  project.  Upon  receipt, 
the  specimens  from  both  sources  were 
screened  and  if  considered  suitable  were 
placed  in  carefully  adapted  storage 
containers.  Only  a  few  sample  blocks  from 
some  half  dozen  of  the  specimens  were 
sectioned. 

Harris's  first  concern  was  to  cull  and 
evaluate  this  special  material  and  to 
survey  the  collection  as  a  whole  to  deter- 
mine what  older  specimens  might  also  be 
included  in  the  study.  Not  only  has  he 
completed  this  part  of  the  program,  but 
he  has  also  prepared  for  more  general  use 
of  all  those  handling  the  collection  a 
definitive  list  of  all  suitable  placental 
material  on  hand  through  1961.  Speci- 
mens are  seriated  by  age;  the  state  of 
preservation  and  type  of  preparation  have 
been  noted;  and  gaps  in  the  series  that 
should  be  closed  in  future  collecting  have 
been  marked. 


2.  Study  of  the  placental  series  and 
modeling  of  representative  examples.  Prep- 
aration of  the  material  is,  of  course, 
preliminary  to  study.  It  has  been  a  basic 
premise  that  the  preparation  should  be 
carried  out  in  the  same  way  as  the 
preparation  of  the  monkey  material,  so 
that  results  will  be  closely  comparable. 
Refinements  and  additional  lines  of 
investigation  are  by  no  means  excluded 
and  are  particularly  to  be  welcomed  in  so 
far  as  they  may  yield  data  on  points 
remaining  obscure  in  the  monkey  series. 
The  fundamental  pattern  of  serial  sec- 
tioning through  the  placenta  and  uterine 
wall  will  be  maintained,  and  transparent 
sheet  reconstructions  of  key  stages  will 
be  prepared. 

Somewhat  more  than  half  of  the 
technical  work  was  done  before  Harris 
returned  to  London,  and  arrangements 
have  been  made  for  completion  of  the 
remainder. 

Study  of  specimens  was  commenced  as 
soon  as  selections  were  ready,  and 
remodeling  proceeded  both  as  an  element 
in  the  study  serving  to  elucidate  relation- 
ships and  as  a  first  step  in  readying  the 
material  for  eventual  publication.  Nu- 
merous sketch  models  have  been  made, 
and  several  plastic  sheet  reconstructions 
have  been  brought  close  to  finished  form 
— one,  entirely  so. 

3.  Preparation  of  manuscript  for  publi- 
cation. This  step  will  be  a  joint  project  of 
Harris  and  Ramsey.  Its  completion  is 
tentatively  scheduled  to  be  in  time  for 
volume  38  of  the  Contributions  to  Embry- 
ology. 

The  services  of  Mrs.  Ranice  Davis, 
Director  of  the  Department  of  Art  as 
Applied  to  Medicine  of  the  Johns  Hopkins 
University,  have  been  enlisted  for  the 
artistic  rendering  of  the  models  as 
illustrations  for  the  publication.  Mrs. 
Davis,  a  former  student  of  the  late  James 
F.  Didusch,  who  drew  the  illustrations 
for  the  monkey  papers,  and  herself  both 
skilled  and  experienced,  is  well  fitted  to 
prepare  drawings  that  will  be  comparable 
to  Mr.  Didusch's. 


426 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


A  Composite  Drawing  of  the  Placenta 
to  Show  Its  Structure  and  Circulation 

Diagrammatic  representations  of  the 
placenta  and  its  circulation,  as  shown  in 
scientific  papers  and  textbooks,  have  not 
kept  pace  with  advances  in  knowledge 
and  understanding  of  this  organ.  Re- 
visions of  diagrams  based  upon  long- 
outmoded  data  and  theories — and  snow- 
balling revisions  of  revisions — have 
produced  confusion  and  inaccuracy.  A 
fresh  start  and  a  rethinking  of  the  whole 
problem  of  how  to  represent  this  compli- 
cated organ  are  long  overdue.  In  an 
attempt  to  meet  this  need,  figure  36 
(pi.  10)  has  been  drawn  by  Mrs.  Davis,  in 
consultation  with  Ramsey.  The  goal  has 
been  to  embody  current  knowledge  of 
placental  structure  and  current  concepts 
of  fetal  and  maternal  circulation  in  a 
drawing  that  reconciles  realism  and 
interpretation  as  far  as  possible.  The 
usual  diagram,  which  frequently  resem- 
bles a  blueprint  of  hydraulic  channels,  has 
been  replaced  with  a  drawing  in  which 
students  can  immediately  recognize  the 
placenta  as  they  see  it  in  the  delivery 
room  and  under  the  microscope.  At  the 
same  time  it  has  been  considered  desirable 
to  express  clearly  the  course  pursued  by 
maternal  blood  in  traversing  the  inter- 
villous space  and  to  represent  the  forces 
propelling  it.  To  achieve  these  diverse 
purposes  the  drawing  has  been  made  in 
five  separate  panels,  each  telling  a  single 
facet  of  the  total  story,  as  indicated  in 
the  legends.  Taken  together  they  illus- 
trate the  "physiological  concept"  of 
circulation  in  the  maternal  placenta, 
which  has  been  deduced  from  anatomical, 
physiological,    and    radiological    studies 


carried  out  by  Ramsey  and  her  associates 
over  a  period  of  years  (see  Year  Books  1^8 
to  date).  This  concept  may  be  expressed 
as  follows: 

Maternal  blood  enters  the  intervillous 
space  of  the  placenta  from  the  endo- 
metrial arteries  in  fountainlike  jets  or 
"spurts"  produced  by  the  higher  pressure 
in  the  maternal  vessels  as  compared  with 
the  low  pressure  prevailing  in  the 
amorphous  intervillous  pool.  The  head  of 
maternal  arterial  pressure  drives  the 
blood  well  along  toward  the  chorionic 
plate  before  lateral  dispersion  occurs,  thus 
preventing  short-cutting  from  arteries 
into  adjacent  venous  orifices  before  the 
blood  has  circulated  through  the  inter- 
villous space.  Gradually,  however,  the 
head  of  pressure  is  spent,  and  general 
spreading  throughout  the  space  occurs. 
The  villi,  acting  as  baffles,  aid  this 
dispersion  and  promote  slowing  and 
mixing.  Their  own  pulsation  also  effects 
a  mild  stirring.  Finally,  the  blood  drains 
from  the  intervillous  space  into  the 
maternal  uterine  veins,  where  pressure  is 
even  lower  than  in  the  placenta. 

Dr.  John  W.  Crawford  of  Glasgow, 
Scotland,  has  kindly  checked  the  repre- 
sentation of  the  fetal  circulation  as  it 
appears  in  this  drawing.  Dr.  Crawford's 
extensive  investigations  over  the  past 
decade  have  largely  illuminated  this 
aspect  of  placental  circulation.  His  help- 
ful suggestions  are  gratefully  acknowl- 
edged. The  fetal  circulation  is  less  fully 
portrayed  here  than  the  maternal  circuit, 
in  part  because  problems  of  magnification 
preclude  representation  of  the  villous 
capillary  bed,  which  is  the  fundamental 
area  of  metabolic  exchange  on  the  fetal 
side. 


DIFFERENTIATION   AND   MORPHOGENESIS  IN   THE 

HUMAN   EMBRYO 


The  Collection  of  Human  Embryos 

In  the  year  covered  by  this  report, 
Elizabeth     M.     Ramsey     examined    30 


specimens  sent  by  physicians  and  labora- 
tories from  three  states  and  one  foreign 
country.  Of  these  specimens,  22  were 
discarded  as  of  no  research  value,  at  the 


DEPARTMENT  OF  EMBRYOLOGY  427 

end  of  three  months  after  reporting  to  under  way.   Specimens  acquired  during 

the  donor  and  in  the  absence  of  instruc-  recent  years,  particularly  those  stained 

tions  to  the  contrary.   Eight  specimens  by    the     "azan"     method,     have     been 

had  sufficient  research  value  to  justify  especially  valuable  in  this  investigation, 

preservation.  Considerable  attention  has  been  paid  to 

In  the  six  months  before  moving  to  the  the  basement  and  other  membranes  of 

new  laboratory  all  the  gross  specimens  in  the  eye,  and  an  abstract  has  already  been 

the  Collection  of  Human  Embryos  were  published  in  the  Anatomical  Record. 

reviewed  carefully.  All  tissues  of  impor-  During  the  closure  of  the  retinal,  or 

tant  early  stages  were  retained,  whether  so-called  "choroid,"  fissure  of  the  eye,  the 

any    part    of    the    specimen    had    been  generally  accepted  view  that  "fusion  of 

sectioned  previously  or  not.  All  specimens  course   must   occur   before   the   internal 

exhibiting  anomalies  were  retained  unless  limiting  membrane  is  differentiated  along 

the  state  of  preservation  precluded  any  the  margins  of  the  cleft"  (Mann,  1949) 

possible  future  study.   This  was  rarely  has  been  found  to  be  incorrect.  The  optic 

true  even  among  specimens  collected  as  cup   is   covered   by,    and   lined   with,    a 

much  as  45  years  ago.  Normal  material  continuous  basement  membrane  from  its 

in  grade  2  condition  (adequate  for  gross  initial  formation  from  the  neural  tube, 

study    but    not    for    histological)     was  and,   at  the  site  of  obliteration  of  the 

reduced    sharply.    Material    in   grade    1  retinal  fissure,  this  membrane  disappears 

condition  was  reduced  to  an  inventory  of  during  closure.  A  similar  occurrence  has 

about    10    specimens    for    each    10-mm  been  seen  during  the  complicated  mor- 

increment  in  crown-rump  length  through  phogenesis  of  the  membranous  labyrinth, 

150  mm.  Older  specimens  up  to  term  were  and  preliminary  observations  have  also 

kept   to   an   average    of   5    per    10-mm  been  made  on  the  early  development  of 

increment.  the  otocyst  in  the  collection  of  human 

Discarded  specimens  in  good  condition  embryos, 
were  presented  to  the  Departments  of 

Anatomy  of  the  Johns  Hopkins  Medical  Early  Development  of  the  Brain 
School  and  the  University  of  West  Vir- 
ginia, School  of  Medicine.  In  the  large  and  closely  spaced  series 

of  well  preserved  embryos  in  the  Carnegie 

Development  of  the  Eye  Collection  the  stage  at  which  structural 

Ronan  O'Rahilly's  main  objective  has  differentiation    appears    in    the    various 

been  to  expand  a  program  of  work  on  the  centers  of  the  brain  has  been  determined 

development  of  the  eye  by   (a)   under-  by  G.  W.  Bartelmez  and  A.  S.  Dekaban 

taking    an    experimental    embryological  and  an  account  has  been  published  in 

study  of  the  chick  eye,  and  (6)  making  a  Contributions  to  Embryology,  volume  37. 

detailed  analysis  of  the  development  of  Description  of  the  progressive  develop- 

the  human  eye  in  staged  embryos.  ment   of   the   human   brain   from   early 

Some  experience  in  several  experi-  neural  folds  is  of  value  from  the  practical 
mental  techniques  on  the  chick  embryo,  as  well  as  from  the  embryological  view- 
such  as  intracoelomic  and  chorioallantoic  point.  Analysis  of  various  complex  con- 
grafting,  has  been  attained,  and  it  is  genital  malformations  of  the  nervous 
proposed  to  make  use  of  these  methods  system  can  be  understood  only  in  the 
subsequently.  light  of  normal  morphogenesis.  It  is  well 

A  detailed  analysis  of  the  development  known  that  the  nervous  system  is  highly 

of  the  human  eye,  from  its  initial  appear-  susceptible    to    various    noxious    factors 

ance  at  horizon  x  (3  postovulatory  weeks)  during  prenatal  life. 

until  the  beginning  of  the  fetal  period  at  To  follow  the  centers  of  the  brain  from 

horizon  xxiii  (7  postovulatory  weeks),  is  later  to  earlier  stages  it  was  necessary  to 


428  CAKNEGIE     INSTITUTION      OF      WASHINGTON 

establish  a  series  of  landmarks.  The  most  an  abrupt  shift  in  the  axis  of  the  central 

obvious  are  the  boundaries  between  the  nervous    system,    namely    the    "cranial 

major  subdivisions  that  characterize  the  flexure." 

brain  of  all  vertebrates:  namely,  the  In  the  early  stages  of  human  develop- 
forebrain,  midbrain,  and  hindbrain.  These  ment  the  hindbrain  is  the  largest  sub- 
have  been  traced  from  embryos  of  the  division,  as  it  is  in  all  other  vertebrates, 
eighth  to  ninth  weeks,  when  all  the  The  earliest  differentiations  appear  in  this 
fundamental  centers  can  be  recognized,  to  region  and  rapidly  proceed  forward  to  the 
the  neural-fold  stages  of  the  early  fourth  basal  (efferent)  region  of  the  midbrain 
week.  and  thence  to  the  hypothalamus  and  the 

The    Carnegie    Collection   has   photo-  basal   centers   of  the   olfactory   system, 

micrographs  of  the  serial  sections  of  many  Neuroblasts  then  appear  in  other  phylo- 

embryos,  each  section  accurately  oriented  genetically  old  centers  of  the  forebrain, 

by  guide  lines  according  to  the  Born-  the  corpus  striatum,  and  the  epithalamus. 

Lewis  procedure.  It  is  accordingly  pos-  As  was  first  reported  by  Hines,  the  first 

sible  to  reconstruct  any  internal  organ  region  of  the  cerebral  cortex  to  differenti- 

with  the  control  of  the  external  form  of  ate   is   the   hippocampus,   which  is   the 

the     embryo,     controlled     in    turn    by  dominant  center  of  the  forebrain  in  many 

photographs  of  the  intact  specimen.  Such  lower  vertebrates.  With  the  appearance 

controls    are    especially    important    for  of  differentiation  in  the  neopallium  the 

reconstruction    of    the    brain    with    its  associated  thalamic  centers  can  be  recog- 

flexures.  nized.   Retinal  fibers  do  not  reach  the 

By  means  of  graphic  reconstructions  brain  until  the  end  of  the  second  month, 

and  models  the  exact  location  of  histo-  The  cerebellar  cortex  differentiates  later 

logical  differentiation  could  be  determined  than  all  others,   and  its   cells  are  still 

in   each   embryo   of   our   series.    In   the  undergoing  rearrangement  and  matura- 

earliest   stages   the   primordium   of   the  tion  after  birth.  It  is  perhaps  significant 

optic    vesicle    from    which    the    retina  that  it  is  the  most  frequently  observed 

develops,      and     the     characteristically  site  of  neoplasms  in  the  central  nervous 

shaped    subdivisions    of    the    hindbrain  system. 

associated  with  the  fifth  and  seventh  Bartelmez  and  Dekaban  are  currently 
cranial  nerves,  mark  the  segment  of  the  attempting  to  follow  the  primary  pro- 
neural  folds  between  them  as  the  mid-  jection  centers  of  the  cerebral  cortex, 
brain,  which  is  at  this  period,  as  well  as  motor,  visual,  and  auditory,  to  stages  of 
throughout  development,  associated  with  the  third  fetal  month. 


STAFF  ACTIVITIES 

One  of  the  most  perplexing  problems  to  imagine  that  soon  each  biologist  will 

with  which  all  scientists  have  to  deal  is  receive  his  daily  or  weekly  newspaper  of 

the  ever-mounting  tide  of  publications,  biology;  extensive  documentation  will  be 

There  can  be  little  doubt  but  that  the  available  in  centers  established  at  several 

next  decade  will  see  far-reaching  changes  strategic   locations,   and   copies  of  data 

in   the    nature    and    scope    of    scientific  will  be  made  available  to  persons  working 

journals  and  in  other  means  of  dissemi-  on  a  given  problem.  Journals,  in  the  usual 

nation  of  data.  Already  specialized  jour-  sense,    might    then   be    devoted    almost 

nals  are  being  established  to  speed  the  entirely  to  synthesis  and  evaluation, 

flow  of  announcements  of  discoveries  in  Similarly    there    may    have    to    be    a 

brief  form,   and  the  number  of  review  change  in  the  direction  of  conferences  and 

journals  is  being  increased.  It  is  not  hard  symposia,  which  have  multiplied  rapidly 


DEPARTMENT  OF  EMBRYOLOGY  429 

in  recent  years.  We  are  beset  by  vast  School  of  Medicine,  including  Anatomy, 
numbers  of  conferences  with  attendance  Biology,  and  Pathobiology.  Other  teach- 
in  the  range  150  to  300  persons,  too  large  ing  activities  included  lectures  in  experi- 
a  group  for  effective  discussion,  too  small  mental  embryology  given  by  one  staff 
to  make  full  use  of  a  speaker's  talents,  member  at  the  University  of  Miami,  and 
Often  the  intermediate-size  meeting  is  the  participation  of  two  others  in  the 
justified  by  the  statement  that  the  Embryology  Training  Program  of  the 
proceedings  will  be  published,  but  such  Marine  Biological  Laboratory,  Woods 
volumes  are  all  too  often  collections  of  Hole,  Massachusetts, 
ill-assorted  papers.  More  difficult,  but  equally  rewarding, 

The  activities  of  our  own  staff  members  are  lectures  directed  toward  lay  audi- 
and  visiting  investigators  reveal  that  they  ences ;  several  members  of  the  staff 
have  found  no  simple  solution  to  these  lectured  before  such  groups,  which  in- 
problems.  Of  all  the  ventures  in  which  the  eluded  the  Maryland  Academy  of  Sciences 
staff  was  engaged  one  stands  out  as  an  Public  Lecture  Series  and  the  Baltimore 
unusually  promising  approach.  Elizabeth  City-County  High  School  Science  Semi- 
M.  Ramsey  organized  and  conducted  a  nar.  Attention  was  focused  on  congenital 
meeting  of  an  informal  group  of  16  defects  for  general  audiences  in  Provi- 
physicians  engaged  in  laboratory  and  dence  and  Washington,  D.  C.  One  mem- 
clinical  studies  of  uterine  motility  and  the  ber  of  the  group  took  part  in  the  Columbia 
physiology  of  the  placenta.  This  confer-  Broadcasting  System's  College  of  the  Air 
ence  was  a  three  days'  session  of  reports  television  series  The  New  Biology. 
and  consultations,  November  3-5,  1961,  Among  the  principal  lectures  presented 
constituting  the  first  conference  to  be  by  members  of  the  staff,  the  following 
held  in  the  new  laboratory.  The  writer,  should  be  mentioned:  a  series  of  lectures 
who  was  privileged  to  attend  several  of  delivered  under  the  auspices  of  the 
the  sessions,  found  the  frank,  uninhibited  Cleveland  Foundation  at  Western  Re- 
discussions  highly  rewarding.  serve  University,  the  Hoffman-La  Roche 

An  effective  training  program  in  which  Lectures  at  Rutgers  University,  a  Macy 
two  members  of  the  group  took  part  was  Lecture  at  Harvard  Medical  School,  and 
the  Summer  Institute  in  Developmental  a  Sloane  Lecture  in  the  Department  of 
Biology  for  College  Teachers  held  at  Obstetrics  and  Gynecology  of  Columbia 
Brevard,  North  Carolina,  during  August  University.  Other  lectures  were  offered  at 
1961,  under  the  auspices  of  the  National  the  following  research  centers  and  uni- 
Institutes  of  Health.  Other  activities  versities:  University  of  Aberdeen,  Scot- 
directed  principally  toward  teaching  in-  land;  University  of  Alabama  School  of 
eluded  the  following :  One  member  of  the  Medicine ;  University  of  Florida ;  Harvard 
staff  served  as  a  consultant  in  develop-  Medical  School;  Johns  Hopkins  Univer- 
mental  biology  in  the  Baltimore  Public  sity;  University  of  Miami;  National 
Schools,  under  the  BSCS  Teacher  Train-  Institutes  of  Health;  State  University  of 
ing  Program  developed  by  the  American  New  York;  University  of  Washington. 
Institute  of  Biological  Sciences ;  another  Staff  members  and  fellows  took  part  in 
lectured  at  the  Summer  Institute  for  several  international  meetings,  including 
Secondary  School  Teachers,  held  at  the  the  Fifth  International  Embryological 
University  of  Maryland ;  and  still  another  Conference,  held  in  London;  the  First 
presented  a  lecture-demonstration  in  the  Inter- American  Conference  on  Congenital 
Basic  Science  Program  of  the  Washington  Defects,  held  in  Los  Angeles;  the  Con- 
Hospital  Center.  As  in  the  past,  several  ference  on  Specificity  of  Cell  Differenti- 
members  of  the  group  took  limited  part  ation  and  Interaction,  Gatlinburg,  Ten- 
in  formal  courses  in  several  departments  nessee;  the  Thirteenth  Conference  on 
of  the   Johns   Hopkins   University   and  Problems  of  Nephrosis,  Princeton,  New 


430 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


Jersey;  and  the  First  International  Con- 
ference on  the  Biology  of  Skin  Cancer, 
Philadelphia. 

Other  meetings  of  learned  societies  in 
which  members  of  the  group  participated 
were  the  American  Association  for  the 
Advancement  of  Science,  American  Asso- 
ciation of  Anatomists,  American  Institute 
of  Biological  Sciences,  American  Society 
of  Biological  Chemists,  American  Society 
of  Zoologists,  Biophysical  Society,  Fed- 
eration of  American  Societies  for  Experi- 
mental Biology,  Johns  Hopkins  Medical 
Society,  Society  of  American  Bacteriolo- 
gists, Society  for  the  Study  of  Develop- 
ment and  Growth,  Society  of  General 
Physiologists,  and  the  Tissue  Culture 
Association. 

Advisory  and  consultative  services 
included  membership  on  the  editorial 
boards  of  the  American  Zoologist,  the 
Biological  Bulletin,  the  Journal  of  Embry- 
ology and  Experimental  Morphology,  the 
section  on  Human  Developmental  Biol- 
ogy of  Excerpta  Medica,  and  the  board  of 
consulting  editors  of  Developmental  Bi- 
ology. One  member  of  the  staff  was 
appointed  to  the  Divisional  Committee 
for  Biology  and  Medicine  in  the  National 
Science  Foundation,  served  on  the  Cell 
Biology  Study  Section  of  the  National 
Institutes  of  Health,  the  Subcommittee 
on  Congenital  Malformations  of  the  U.  S. 
National  Committee  on  Vital  and  Health 
Statistics,  and  the  Visiting  Committee, 
Department  of  Biology,  Massachusetts 
Institute  of  Technology. 

Among  the  services  rendered  by  several 
members  of  the  staff  were  the  following: 
Vice-President,  AIBS ;  Vice-Chairman, 
Scientific  Council,  Maryland  Academy  of 
Sciences;  President,  Maryland  Section, 
Society    of    Experimental    Biology    and 


Medicine;  Chairman,  Division  of  Devel- 
opmental Biology,  American  Society  of 
Zoologists;  Chairman,  American  Organ- 
izing Committee  for  the  forthcoming 
International  Conference,  International 
Institute  of  Embryology ;  Secretary,  Soci- 
ety of  General  Physiologists;  Secretary, 
Section  F,  AAAS;  Chairman,  Publica- 
tions Committee,  American  Society  of 
Zoologists. 

A  distinction  won  by  one  member  of 
the  group  must  be  singled  out  for  special 
mention:  Virginia  LaFleur,  a  senior  at 
College  of  Notre  Dame  of  Maryland,  who 
has  for  the  past  year  been  carrying  on 
research  in  the  Department  on  the 
graft-versus-host  reaction  in  consultation 
with  J.  D.  Ebert,  was  awarded  first  prize 
for  her  paper  presented  at  the  North- 
eastern Regional  Convention  of  the 
undergraduate  biological  society,  Beta 
Beta  Beta,  held  at  American  University, 
Washington,  D.  C. 

Seminars.  The  roster  of  speakers  at  the 
Seminar  organized  by  the  Department  to 
serve  all  those  working  in  developmental 
biology  in  the  area  included  Robert 
Auerbach,  University  of  Wisconsin;  J.  D. 
Biggers,  University  of  Pennsylvania;  J. 
Chutna,  Prague;  A.  L.  Colwin,  Queens 
College;  A.  J.  Coulombre,  National 
Institutes  of  Health;  I.  Finger,  Haverford 
College;  C.  E.  Ford,  Harwell;  V.  Haskova, 
Prague;  T.  J.  King,  Institute  for  Cancer 
Research;  H.  Kroeger,  Zurich;  J.  W. 
Lash,  University  of  Pennsylvania;  A.  B. 
Pardee,  Princeton  University;  P.  Perl- 
mann,  Stockholm;  R.  B.  Roberts,  Depart- 
ment of  Terrestrial  Magnetism;  J.  Runn- 
strom,  Stockholm;  L.  Saxen,  Helsinki; 
F.  Seidel,  Marburg,  Germany;  P.  Sengel, 
Paris;  A.  K.  Tarkowski,  Warsaw;  T. 
Vainio,  Helsinki. 


BIBLIOGRAPHY 


Bartelmez,  G.  W.,  The  proliferation  of  neural 
crest  from  forebrain  levels  in  the  rat,  Carnegie 
Inst.  Wash.  Publ.  621,  Contrib.  Embryol.,  37, 
1-12,  1962. 


Bartelmez,  G.  W.,  and  A.  S.  Dekaban,  The  early 
development  of  the  human  brain,  Carnegie 
Inst.  Wash.  Publ.  621,  Contrib.  Embryol.,  87, 
13-32,  1962. 


DEPARTMENT  OF  EMBRYOLOGY 


431 


Bishop,  D.  W.,  Biology  of  spermatozoa,  in  Sex 
and  Internal  Secretions,  3rd  ed.,  edited  by 
W.  C.  Young,  Williams  &  Wilkins  Company, 
Baltimore,  Md.,  pp.  707-796,  1961. 

Bishop,  D.  W.,  Sperm  motility,  Physiol.  Rev.,  42, 
1-59,  1962. 

Bishop,  D.  W.,  Reactivation  of  extracted  sperm- 
cell  models  in  relation  to  the  mechanism  of 
motility;  and  Epilogue,  in  Spermatozoan 
Motility,  edited  by  D.  W.  Bishop,  American 
Association  for  the  Advancement  of  Science, 
Publ.  72,  Washington,  D.  C,  pp.  251-268  and 
285-295,  1962. 

Bishop,  D.  W.,  Autoimmune  induced  aspermato- 
genesis (abstract),  Bull.  Johns  Hopkins  Hosp., 
110,  222,  1962. 

Bishop,  D.  W.,  R.  Narbaitz,  and  M.  Lessof, 
Induced  aspermatogenesis  in  adult  guinea  pigs 
injected  with  testicular  antigen  and  adjuvant 
in  neonatal  stages,  Develop.  Biol.,  3,  444-485, 
1961. 

Boving,  B.  G.,  Review  of  "Control  of  Ovulation," 
edited  by  C.  A.  Villee,  J .  Am.  Med.  Assoc, 
178,  530,  1961. 

Boving,  B.  G.,  Review  of  C.  R.  Austin,  "The 
Mammalian  Egg,"  Science,  136,  526,  1962. 

Boving,  B.  G.,  Anatomical  analysis  of  rabbit 
trophoblast  invasion,  Carnegie  Inst.  Wash. 
Publ.  621,  Contrib.  Embryol.,  37,  33-55,  1962. 

Burns,  R.  K.,  The  role  of  hormones  in  the 
differentiation  of  sex,  in  Sex  and  Internal 
Secretions,  3rd  ed.,  edited  by  W.  C.  Young, 
Williams  &  Wilkins  Company,  Baltimore, 
Md.,  pp.  76-158,  1962. 

Coffman,  C.  M.,  see  DeLanney,  L.  E. 

Coleman,  J.  R.,  Deoxyribonucleases  in  embry- 
onic and  adult  Rana  pipiens,  Am.  Zoologist,  1, 
348-349,  1961. 

Coleman,  J.  R.,  Deoxyribonuclease  activities  in 
the  development  of  the  frog,  Rana  pipiens, 
Ph.D.  dissertation,  the  Johns  Hopkins  Uni- 
versity, 1961. 

Cooper,  W.  G.,  and  I.  R.  Konigsberg,  Succinic 
dehydrogenase  activity  of  muscle  cells  grown 
in  vitro,  Exptl.  Cell  Res.,  23,  576-581,  1961. 

DeHaan,  R.  L.,  Differentiation  of  the  atrio- 
ventricular conducting  system  of  the  heart, 
Circulation,  24,  458-470,  1961. 

DeHaan,  R.  L.,  Time-lapse  photographic  analy- 
sis of  migration  of  the  precardiac  mesoderm  in 
the  early  chick  embryo,  Am.  Zoologist,  1, 
444-445,  1961. 

Dekaban,  A.  S.,  see  Bartelmez,  G.  W. 

DeLanney,  L.  E.,  and  J.  D.  Ebert,  with  C.  M. 
Coffman  and  A.  M.  Mun,  On  the  chick  spleen: 
origin;  patterns  of  normal  development  and 
their  experimental  modification,  Carnegie  Inst. 
Wash.  Publ.  621,  Contrib.  Embryol,  37,  57-85, 
1962. 


DeVries,  P.,  and  J.  B.  Saunders,  Development 
of  the  ventricles  and  spiral  outflow  tract  in  the 
human  heart,  Carnegie  Inst.  Wash.  Publ.  621, 
Contrib.  Embryol.,  37,  87-114,  1962. 

Ebert,  J.  D.,  Perspectives  in  the  study  of 
congenital  defects,  The  molecular  basis  of 
malformations,  Bull.  Acad.  Med.  New  Jersey, 
7,  224-236,  1961. 

Ebert,  J.  D.,  Tissue  transplantation,  The  Voice 
of  America  Forum  Lectures,  Biology  Series, 
no.  11,  pp.  1-8,  1961. 

Ebert,  J.  D.,  Antibodies,  viruses,  and  embryos, 
in  First  International  Conference  on  Congenital 
Malformations,  edited  by  M.  Fishbein,  J.  B. 
Lippincott  Co.,  Philadelphia,  pp.  291-299, 
1961. 

Ebert,  J.  D.,  Review  of  "The  Reticuloendothelial 
System  and  Reticuloendothelial  Structure  and 
Function,"  Quart.  Rev.  Biol,  36  (1),  88,  1961. 

Ebert,  J.  D.,  see  also  DeLanney,  L.  E. 

Konigsberg,  I.  R.,  Some  aspects  of  myogenesis 
in  vitro,  Circulation,  24,  447-457,  1961. 

Konigsberg,  I.  R.,  Cellular  differentiation  in 
colonies  derived  from  single  cell  platings  of 
freshly  isolated  chick  embryo  muscle  cells, 
Proc.  Natl.  Acad.  Sci.  U.  S.,  47,  1868-1872, 
1961. 

Konigsberg,  I.  R.,  Review  of  "Animal  Growth 
and  Development,"  Quart.  Rev.  Biol.,  36,  299, 
1961. 

Konigsberg,  I.  R.,  see  also  Cooper,  W.  G. 

Kulangara,  A.  C,  Autoradiography  using  melted 
nuclear  emulsion,  Nature,  182,  437-439,  1961. 

Lessof,  M.,  see  Bishop,  D.  W. 

McKenzie,  J.,  The  development  of  the  sterno- 
mastoid  and  trapezius  muscles,  Carnegie  Inst. 
Wash.  Publ.  621,  Contrib.  Embryol.,  37,  121- 
129,  1962. 

Mun,  A.  M.,  see  DeLanney,  L.  E. 

Narbaitz,  R.,  The  primordial  germ  cells  in  the 
male  human  embryo,  Carnegie  Inst.  Wash. 
Publ.  621,  Contrib.  Embryol,  37,  115-119,  1962. 

Narbaitz,  R.,  see  also  Bishop,  D.  W. 

O'Rahilly,  R.,  The  initial  membranes  of  the  eye 
in  staged  human  embryos,  Anal  Record,  14%, 
263-264,  1962. 

Ramsey,  E.  M.,  Normal  embryogenesis,  Bull. 
Acad.  Med.  New  Jersey,  7,  210-223,  1961. 

Ramsey,  E.  M.,  The  placenta,  in  Traumatic 
Medicine  and  Surgery  for  the  Attorney,  edited 
by  P.  D.  Canter,  Butterworth  and  Company, 
London,  pp.  215-238,  1961. 

Saunders,  J.  B.,  see  DeVries,  P. 

Silver,  P.  H.  S.,  Experiments  on  eye  and  orbit 

in  the  chick  embryo,  Anal  Record,  142,  279, 

1962. 


432 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


PERSONNEL 

Year  Ended  June  30,  1962 
(including  those  whose  services  began  or  ended  during  the  year) 


Research  Staff 

David  W.  Bishop,  General  Physiology 

Bent  G.  Boving,  Physiology 

Robert  K.  Burns,  Experimental  Embry- 
ology 

Robert  L.  DeHaan,  Experimental  Embry- 
ology 

James  D.  Ebert,  Director 

Irwin  R.  Konigsberg,  Experimental  Embry- 
ology 

Elizabeth  M.  Ramsey,  Placentology;  Pa- 
thology 

Mary  E.  Rawles,  Experimental  Embryology 

Assistant  Investigators 

Alton  M.  Mun,  Experimental  Embryology 
Chinami    Takata,    Experimental    Embry- 
ology 

Research  Associates  (Extramural) 

Louis  B.  Flexner,  Philadelphia 
Arthur  T.  Hertig,  Boston 
Chester  H.  Heuser,  Augusta,  Georgia 
Samuel  R.  M.  Reynolds,  Chicago 

Fellows 

Michael  Abercrombie,  Fellow  of  Carnegie 

Institution  of  Washington 
Bertie  F.  Argyris,  Fellow  of  the  U.  S.  Public 

Health  Service 
Thomas  S.  Argyris,  Fellow  of  the  National 

Science  Foundation 
Donald    D.    Brown,    Fellow    of    Carnegie 

Institution  of  Washington 
Gerald  L.   Carlson,   Fellow  of  the  Given 

Foundation-National  Research  Council 
J.    Douglas    Caston,    Fellow    of    Carnegie 

Institution  of  Washington 
Timothy  Glover,  Fellow  of  the  Population 

Council 
John  W.  S.  Harris,  Fellow  of  the  Rocke- 
feller Foundation 
Tom    Mori,    Fellow    of    the    Rockefeller 

Foundation 


Ronan  O'Rahilly,  Special  Fellow  of  the  U.  S. 
Public  Health  Service 

Peter  H.  S.  Silver,  National  Institutes  of 
Health  Postdoctoral  Traveling  Fellow 

Ian  B.  Wilson,  Fellow  of  Carnegie  Institu- 
tion of  Washington 

Visiting  Investigators 

Frank  D.  Allan,  Washington,  D.  C. 

George  W.  Bartelmez,  Missoula,  Montana 

George  W.  Corner,  Jr.,  Baltimore 

Anatole  S.  Dekaban,  Bethesda 

Arentje  Dekker,  Leiden,  the  Netherlands 

Louis  E.  DeLanney,  Crawfordsville,  Indiana 

Martin  W.  Donner,  Baltimore 

W.  Richard  Ferguson,  Baltimore 

Arthur  LaVelle,  Chicago 

Ali  Mehrizi,  Baltimore 

Sheila  J.  Moody,  London  and  New  York 

E.  Carl  Sensenig,  Birmingham,  Alabama 

Students  (in  cooperation  with  the  Johns 
Hopkins  University  and  College  of  Notre 
Dame  of  Maryland) 

Timothy  Bishop  (student  assistant) 

John  R.  Coleman  (graduate,  biology) 

James  Errico  (medicine) 

Charles  B.  Kimmel  (student  assistant) 

Virginia  LaFleur  (undergraduate,  biology) 

John  Rowse  (medicine) 

Gretchen  Schabtach  (graduate,  biology) 

R.  Owen  Sear  (medicine) 

Clerical  and  Technical  Staff 

Leon  Allen,  Custodian 
Mary  N.  Barton,  Librarian 
Franklin  R.  Baytops,  Custodian 
George  Boettinger,  Porter 
William  Bouchat,  Assistant  Recorder 
Barbara  Brown,  Dishwasher 
William  I.  Cleary,  Recorder 
Lloyd  Crane,  Technician 
Lawrence  A.  Dorsey,  Custodian 
William  H.  Duncan,  Technician 
Ernest  W.  Edwards,  Custodian 


DEPARTMENT    OF    EMBRYOLOGY 


433 


Linda  Fuson,  Technician 
Wilma  Gabbay,  Technician 
Wilbur  F.  Garde,  Assistant  Recorder 
Thomas  F.  Garnett,  Technician 
Richard  D.  Grill,  Photographer 
Ernest  Harper,  Chief  Custodian 
Elaine  Kerby,  Stenographer 
Francis  J.  Kupres,  Technician 
Edna  G.  Lichtenstein,  Secretary 
Ellen  P.  Monaghan,  Technician 


John  Pazdernik,  Building  Engineer 
Margaret  J.  Proctor,  Secretary 
Arthur  G.  Rever,  Office  Manager 
James  Roland,  Custodian 
Nancy  J.  Sype,  Technician 
John  L.  Wiser,  Machinist 

Special  Technical  Assistant  pro  tempore 
Joseph  P.  Drane 


PLATES 


Plate  1 


Department  of  Embryology 


Fig.  1.  Schlieren  patterns  of  ribosome  fractions.  Centrifugations  were  done  at  50,740  rpm,  and 
pictures  were  taken  320  seconds  after  reaching  speed.  Preparations  are,  from  left  to  right,  adult 
frog  liver;  ovarian  eggs;  ovarian  eggs  treated  in  versene;  stage  25  embryos;  stage  25  embryos  in 
versene.     Arrow  shows  direction  of  sedimentation. 


Fig.  3.  Passive  cutaneous  anaphylaxis  re- 
action elicited  by  intracutaneous  antiferritin 
serum  in  decreasing  concentrations  from  top 
down  on  right.  Compare  with  injection  of  control 
serum  at  top  left. 


Plate  2 


Department  of  Embryology 


Fig.  16 

A:  Culture  initiated  with  400  cells  and  fixed  on  the  tenth  day  of  incubation.  Fixation: 
Bouin's.  Stained  in  phosphotungstic  acid  hematoxylin.  Arrow  indicates  colony  enlarged 
in  B  and  C.  Actual  size. 

B:  Colony  indicated  in  A  above.  Area  encompassed  by  the  square  outline  is  enlarged  in 
C.  Magnification  20  X . 

C:  Area  of  B.  Magnification  370  X.  Note  the  prominently  cross-striated  segment  of  the 
left  branch  of  central  myotube.  In  the  same  field,  smaller,  less  well  differentiated  mj'otubes 
can  be  seen  as  well  as  mononucleated  cells. 


Plate  3 


Department  of  Embryology 


Fig.  17.  Cultures  initiated  with  400  cells  each.  Fixed  in  Bouin's  on  the  sixth  day  of  culture. 
Stained  in  Ehrlich's  hematoxylin.  0.75  actual  size.  A,  colonies  in  fresh  medium;  B,  colonies  in  con- 
ditioned medium ;  C,  colonies  in  conditioned  medium  which  had  been  dialyzed  against  fresh  medium 
as  described  in  text. 


Fig.  18.  Monolayer  cultures  initiated  with  7.5  X  105  cells  in  5-cm  petri  plate.  Cultures 
fixed  and  stained  after  2  days  of  cultivation  on  slowly  rotating  turntable.  A,  culture  in 
fresh  medium,  X17.5;  B,  as  above,  X50;  C,  culture  in  conditioned  medium,  X17.5; 
D,  as  above,  X50. 


i 


Plate  4 


Department  of  Embryology 


'? 


/ 


Fig.  20.  Vesicles  formed  from  fragments  of 
heart-forming  tissue,  after  48  hours  of  incubation. 
Whole  mounts  and  cross  sections,  X50. 

a,  b  vesicle  2315-10-1L 

c,  d  vesicle  2215-8-2R 

e,  /vesicle  2222-15-3R 


Plate  5 


Department  of  Embryology 


o 

5  8 
x  S 

x 


i 


o 
o 

X 


W  W  o  ^ 

iO  lO  V  O 

,— i  ,_i  A  -t 

+  +  .  + 


w  w 


o  o 

00    IO 


+  +  +  + 


S2  M 

o  o 


2  0° 

W  o 

o  o 

(N  rH 


h  CO  -t  N 

CO  02  CO  CO 

_0)  CD  CD  CD 

'C  "C  "C  'fi 

CD  CD  CD  CD 

CO  co  CO  CO 

r-1  r-  f-1  i-1 

S  £  S  S 

2  ^  ^  q 

T3  t  T3  ^ 

CD  CD  CD  CD 

§  2  S  2 


co  co  co  co 

^5  £*}  ^>  r"~> 

o3  c3  o3  o3 

T3  T3  T3  13 

CO  tD  iO  CD 

(h  S_  Sh  Sh 

o  o  o  o 


T3 

-c 

T3 

T3 

CD 

CD 

CD 

CD 

(-H 

tn 

— 

S-H 

i— ' 

q 

q 

-1-3 

-^ 

-f^ 

-1-3 

3 

— 

q 

S 

o 

o 

o 

a 

+= 

+= 

■+3 

-1-3 

fl 

q 

a 

G 

CD 

CD 

CD 

CD 

£ 

r* 

H 

C2 

q 

C 

s 

q 

b£ 

h/0 

b£ 

b£ 

c3 

c3 

C3 

03 

Sh 

fH 

Sh 

£- 

«+H 

^l-H 

S+H 

5-1-1 

^^ 

< 

< 

<1 

<^ 

o 

o 

o 

o 

c3 

03 

C3 

& 

l4-H 

«4H 

«+H 

tM 

o 

O 

O 

o 

q 

q 

a 

q 

o 

o 

O 

o 

+3 

-4-3 

-t-3 

-^ 

a 

o 

O 

o 

CD 

CD 

CD 

CD 

co 

CO 

CO 

co 

co 

CO 

CO 

CO 

CO 

co 

co 

co 

O 

o 

O 

O 

Fh 

S-i 

■— 

t-t 

o 

o 

O 

O 

CO 

T)H 

id 

CO 

CN 

CM 

CN 

CN 

fci 

bi) 

bb 

bb 

fe 

fe 

£ 

E 

Plate  6 


Department  of  Embryology 


Fig.  29.  Black  staining  by  0.5  per  cent  silver  nitrate  solution  shows  the  distribution  of 
a  small  amount  of  fluid  injected  into  the  uterine  lumen.  It  did  not  enter  glands,  which  pre- 
sumably are  closed  in  the  living  state.  From  the  relationships  shown,  the  original  extent 
of  the  uterine  lumen  may  be  estimated.  Dislodged  cells  are  present  in  the  uterine  lumen, 
and  a  cyst  near  the  center  of  the  figure  also  contains  some.     L-44 1/3-4.     X150. 


Fig.  30.     A  second  view  of  the  specimen  of  figure  29  shows  shed  cells  in  lumen  and  in 
cysts  to  better  advantage.     L-441/4-7.     X150. 


Plate  7 


Department  of  Embryology 


Fig.  31.  Sodium  carbonate-induced  epivascular  dissociation  of  uterine  epithelium  at  the 
tips  of  endometrial  folds  exhibits  several  degrees  of  increasing  severity  progressing  diago- 
nally downward  from  right  to  left.  A  mass  of  cells,  debris,  and  perhaps  uterine  secretion  is 
in  the  uterine  lumen.     L-443-A/3-3.     X400. 


Fig.  32.     Another  region  of  the  same  specimen  as  figure  31  shows  more  severe  alkali- 
induced  epithelium  loss.     L-443-A/4-3.     X400. 


Plate  8 


Department  of  Embryology 


Fig.  33.  A  third  region  of  the  same  specimen  shows  extreme  and  even  complete  loss  of 
epithelium,  some  loss  of  stroma,  and  (in  the  upper  right  corner)  complete  stripping  down 
to  capillary  endothelium.  The  absence  of  hemorrhage  is  surprising.  L-443-A/test  slide-2. 
X400. 


Plate  9 


Department  of  Embryology 


CD     "' 


S  2  X 
.2  M  5 


r   cT  o 
v    j_.    o 


ft  o3  CO 


►—(go 

-5  c 
>o  «<_  _. 

S  'S  J 


c  o 

5  "G 

'B  x 


_ 


c  — 


CO 


t£ 


Plate  10 


Department  of  Embryology 


o 

bJD 


o 
O 


CO 
CO 

bfl 


Department  of  Genetics 


Cold  Spring  Harbor,  New  York 

Berwind  P.  Kaufmann 
Director 


Contents 


Introduction 437 

Educational  programs 438 

Cooperative  activities 439 

Terminal  activities 440 

The  library 441 

Research  programs 441 

Growth  and  Inheritance  in  Bacteriophage 443 

The  molecular  weight  of  T5  DNA 443 

The  DNA  of  phage  lambda 444 

The  nature  of  self-protection 445 

Local  denaturation  by  hydrodynamic  shear 445 

The  reversibility  of  thermal  denaturation 447 

Replicating  DNA  of  phage  T2 448 

Conclusion 448 

Topographical  Relations  between  Elements  of  Control  Systems  in  Maize     ....  448 

Origin  from  aiw~5  of  a  two-element  control  system 450 

Analysis  of  axm~2 452 

The  derivatives  of  bz™-2 458 

Enzymology 461 

Ribonuclease 461 

Cytochrome  c 464 

Deoxy ribonuclease  II 466 

Organization  of  Cellular  Materials 466 

Mutagenicity  of  agents  interacting  with  DNA 467 

Effect  of  "near-infrared"  radiation  on  crossing  over 468 

Chromosomal  changes  during  cleavage  mitoses  in  Drosophila 468 

Cytochemical  studies 469 

Electron-microscope  studies 472 

Bibliography 475 

Personnel 475 


Carnegie  Institution  of  Washington  Year  Book  61,  1961-1962 


INTRODUCTION 

"A   bright  prospect  opens  before  us/'  essential  tools  and  techniques  that  will 

said  Hugo  de  Vries  in  his  address  at  the  enable  him  to  shape  the  work  of  evolution 

dedication   of   the    Station   for    Experi-  as  envisioned  by  Hugo  de  Vries.  In  its 

mental  Evolution  in  Cold  Spring  Harbor  progress   from   a   nascent   to   a   mature 

on  June   11,   1904.  "The  matter  of  the  science,  genetics  has  benefited  immeasur- 

evolution  of  organic  life  on  this  earth  .  .  .  ably  from  the  work  carried  on  in  the 

is  to  be  investigated  to  its  very  core.  .  .  .  Station    for    Experimental    Evolution — 

We  want  to  have  a  share  in  the  work  of  later  to  be  known  as  the  Department  of 

evolution,  since  we  partake  of  the  fruit.  Experimental  Evolution  and  finally  the 

We  want  even  to  shape  the  work,  in  order  Department  of  Genetics.   The  entreaty 

to  get  still  better  fruits."  voiced  by  de  Vries  in  1904  has  not  been 

Speaking  as  the  chief  apostle  of  the  forgotten,  and  his  expectations  have  been 

emerging  science   of  genetics,   de   Vries  realized. 

could  envision  the  future  with  a  presci-  A  survey  of  the  accomplishments  of  the 

ence  that  stemmed  from  his  own  pioneer-  departmental  staff  up  to  the  year  1942 

ing  efforts  at  the  turn  of  the  century  in  was  made  by  Dr.  M.  Demerec  when  he 

the  rediscovery  of  Mendel's  fundamental  assumed  the  office  of  Director  (Year  Book 

work  and  the  formulation  of  the  mutation  41,  PP-  169-172).  The  scientific  produc- 

theory.  Within  the  framework  created  by  tivity  of  the  two  previous  directors,  C.  B. 

this  active  participation,  de  Vries  found  Davenport  and  A.  F.  Blakeslee,  and  of 

himself  in  a  position  to  serve  as  godfather  their  associates  (including  G.  H.  Shull, 

of  the  new  laboratory  and  offer  advice  to  C.  W.  Metz,  John  Belling,  A.  J.  Harris, 

its    staff   about    the    experimental    pro-  Oscar  Riddle,  C.  C.   Little,  and  E.   C. 

cedures    requisite    to    its    success    and  MacDowell)  had  indeed  been  "a  glory  to 

perpetuation.  "Increase  of  knowledge  of  the  laboratory  and  to  the  institution  that 

all  the  peculiarities  which  accompany  the  founded  it."  Among  biologists,  at  least, 

phenomenon  of  mutability  is  the  most  the  name  of  Cold  Spring  Harbor  had 

immediate  requirement,"  he  told  them,  gradually   acquired   the   connotation   of 

"A  broad  foundation  knowledge  of  phe-  pioneering  effort  and  major  discovery, 

nomena    is    the    most    assured    way    to  The  pattern  set  in  the  early  years  has 

success.  .  .  .  During  a  long  series  of  years  been   maintained   during   the   past   two 

I  have  fostered  my  conception  of  sudden  decades,  which  lie  fresh  in  the  memory  of 

mutability  and  cultivated  my  primroses  the  writer.   Investigations  for  the  most 

for  myself  and  for  myself  only.  .  .  .  [Now]  part  have  fallen  within  the  purview  of 

I  have  to  yield  my  much  beloved  child,  cytogenetics  in  its  broadest  implications 

But  I  do  it  gladly  and  without  regret.  It  — an  elucidation  of  the  mechanisms  of 

is  the  interest  of  the  child  itself  which  heredity  in  terms  of  the  structure  and 

commands  me.  It  will  be  better  in  your  behavior    of    chromosomes    and    other 

hands.  .  .  .  Pray  have  good  care  of  it  and  cellular  constituents — utilizing  the  facil- 

educate    it    assiduously,    that    it    may  ities  that  have  become  available  in  recent 

become  one  of  the  most  brilliant  parts  of  years  for  exploring  the  details  of  genetic 

your  work,  a  glory  to  this  laboratory  and  mechanisms  down  to  the  level  of  molecu- 

to  the  institution  that  founded  it,  a  pride  lar   organization.    Microorganisms    have 

to  your  country,  and  a  bliss  for  human-  played  a  significant  part  in  facilitating  the 

ity."  (Year  Book  3,  pp.  39,  48,  49,  1905.)  meticulous  analysis  of  genie  fine  structure 

Fifty-eight  years  later,  having  attained  by  M.  Demerec  and  his  associates,  and 

full  stature  among  the  biological  sciences,  the  skillful  diagnosis  of  mechanisms  of 

genetics    is    providing    man    with    the  bacteriophage    reproduction    by    Alfred 

437 


438  CARNEGIE     INSTITUTION      OF      WASHINGTON 

Hershey  and  his  colleagues.  Chromosomes  ^  7      ±.      7  ^ 

j.  ,  .  I       j.           i                    i             j.-i  educational  Proqrams 

oi  higher  forms  have  served  as  material  u 

for    Barbara    McClintock's    penetrating  The    scope    of    these    activities    was 

investigation  of  the  elements  that  control  detailed  in  Year  Book  60.  More  up-to-date 

gene  action,  and  for  the  detailed  studies,  information    with     respect     to     certain 

made  by  Helen  Gay,  Margaret  McDonald,  categories  is  given  below. 

and    the    writer,    of   linear    and    lateral  Drosophila  project.  Distribution  of  mu- 

patterns  of  organization  in  euchromatic  tant  stocks  of  Drosophila  melanogaster  and 

and  heterochromatic  materials  and  their  copies  of  the  Drosophila  Guide  has  been 

roles    in    facilitating    nucleocytoplasmic  continued  throughout  the  year.  A  further 

exchanges.  impressive   increase   in   the    number   of 

This,  then,  is  a  brief  measure  of  past  distributions  is  reported  by  the  Curator 

accomplishment     and     continuing     en-  of  Stocks,  Mrs.  Buchanan.  As  compared 

deavor.  As  this  report  goes  to  press,  the  with  the  figures  recorded  last  year  for  the 

Department   is   being   terminated   as   a  eight  and  a  half  months  from  October  15, 

separate  administrative  unit  of  the  Insti-  1960,  to  June  30,   1961 — namely,  2704 

tution.  Those  of  its  staff  who  remain  at  Drosophila  cultures  and  about  505  copies 

Cold    Spring    Harbor    in    the    Genetics  of  the  Guide — this  year's  tally  shows  that 

Research  Unit  will  cooperate  with  the  between  July  1,  1961,  and  June  30,  1962, 

new  Laboratory  of  Quantitative  Biology  4484  cultures  were  sent  out  in  addition 

now  being  organized.   It  is  anticipated  to  876  copies  of  the  book.  On  a  prorated 

that  the  accomplishments  of  this  group  basis,  this  represents  an  average  increase 

will  continue  to  sustain  the  international  in  the  neighborhood  of  20  per  cent, 

image  of  Cold  Spring  Harbor  as  a  great  As  usual,  the  shipments  were  made  to 

scientific  center.  almost  every  state  in  the  country,  and  a 

The  termination  of  the  Department  of  few  went  as  far  afield  as  Europe,  India, 

Genetics  as  an  administrative  division  of  and  Australia.   In  this  connection  it  is 

the   Institution  has   been   a   matter   of  interesting   to  note   that  the   Comision 

concern  to  many  geneticists.  But  it  is  Nacional  de  Energia  Nuclear,  of  Mexico, 

obvious  that  in  a  sense  the  Department  completed  during  the  year  the  publication 

as  such  has  fulfilled  its  mission.  Genetics  of  a  Spanish  translation  of  the  Drosophila 

now  stands  as  "the  core  science  of  biol-  Guide,  to  be  distributed  to  students  of 

ogy."  It  serves  in  this  capacity  to  inte-  biology  in  Mexican  schools, 

grate  the  efforts  of  physicists,  chemists,  It  appears  that  this  year  by  far  the 

and  biologists  in  resolution  of  the  basic  largest  proportion  of  the  requests,  perhaps 

problems  of  living  systems  concerned  with  more  than  70  per  cent,  came  directly  from 

heredity,  growth,  and  development.  Its  students;  about  20  per  cent  were  made 

practitioners  are  now  firmly  established  in  by  high  school  teachers,  and  about  9  per 

numerous   elaborately  equipped   labora-  cent  by  college  teachers.  The  correspond- 

tories  in  many  lands.  Nonetheless,  much  ence  files  record  eight  science  fair  awards 

work  remains  to  be  done  by  all  of  us,  to  students  who  enlisted  the  aid  of  the 

under  one  patron  or  another,  including  service  during  the  year,  and  in  addition 

the  continuing  Genetics  Research  Unit  of  one  honorable  mention  in  the  Westing- 

the  Institution.  In  the  words  of  Thomas  house  Science  Talent  Search. 

Jefferson,    "Truth   advances,    and   error  Because  of  the  widespread  demand  for 

recedes  step  by  step  only;  and  to  do  to  this  service,  and  the  steadily  increasing 

our  fellow  men  the  most  good  in  our  costs,  a  decision  was  made  to  institute  on 

power,  we  must  lead  where  we  can,  follow  April  1  a  small  charge,  sufficient  to  cover 

where  we  cannot,  and  still  go  with  them,  the  expense  for  materials  and  postage 

watching  always  the  favorable  moment  involved  in  the  distribution, 

for  helping  them  to  another  step."  Other  educational  services.  As  reported 


DEPARTMENT    OF    GENETICS 


439 


in  Year  Book  60,  the  Department  has 
continued  to  respond,  when  possible,  to 
requests  for  information,  illustrative  ma- 
terial, and  bibliographic  references 
received  from  teachers,  students,  and 
other  members  of  the  public.  The 
resources  of  the  library  have  remained 
available  to  students  and  biologists  in  the 
vicinity. 

Drs.  Gay  and  Kaufmann  again  partici- 
pated in  graduate  courses  in  biology  at 
Adelphi  College.  One  graduate  student  of 
the  College,  Miss  Myrna  Thomas,  sought 
and  gained  permission  to  utilize  our 
facilities  in  undertaking  the  research 
required  for  the  M.S.  degree.  Carried  out 
under  Dr.  Gay's  supervision,  her  project 
has  been  completed  and  the  degree 
granted.  Dr.  Kaufmann  has  been  serving 
as  Director  of  Graduate  Program  Devel- 
opment in  the  Department  of  Biology  at 
Adelphi,  in  an  effort  to  establish  a  Ph.D. 
program.  During  March  1962,  Dr. 
McClintock  delivered  several  lectures  in 
a  course  in  advanced  genetics  at  New 
York  University.  In  February  she  took 
part  in  a  conference  at  Raleigh,  North 
Carolina,  for  the  purpose  of  organizing  a 
program  of  research  in  maize  cytogenetics 
for  four  Latin  American  fellows  who  will 
be  in  residence  at  North  Carolina  State 
College  during  the  coming  academic  year. 
Sponsored  by  the  Inter- American  Maize 
Improvement  Program  of  the  Rockefeller 
Foundation,  this  project  will  be  under  the 
direction  of  Dr.  McClintock  and  Dr. 
William  L.  Brown. 

Cooperative  Activities 

In  the  course  of  the  year  Dr.  Gay 
presented  seminar  talks  at  the  University 
of  Michigan  and  Cornell  University;  Dr. 
Kaufmann  was  a  symposium  speaker  at 
the  A.I.B.S.  meetings  at  Purdue  Uni- 
versity; Dr.  McDonald  attended  the 
meetings  of  the  Federation  of  American 
Societies  for  Experimental  Biology  in 
Atlantic  City  and  a  symposium  on 
proteins  and  nucleic  acids  at  Columbia 
University;  Dr.  Hershey  lectured  at  the 
University  of  Oregon;  and  Dr.  McClin- 


tock delivered  lectures  at  Columbia 
University,  New  York  University,  North 
Carolina  State  College,  and  Yale  Uni- 
versity. 

Under  the  joint  chairmanship  of  Dr. 
McDonald  and  Dr.  H.  E.  Umbarger  of 
the  Biological  Laboratory,  an  interesting 
program  of  seminar  lectures  included  the 
following  invited  speakers:  Dr.  Edward 
A.  Adelberg  of  Yale  University;  Dr.  V.  G. 
Allfrey,  Rockefeller  Institute;  Dr.  Elias 
Balbinder,  Scripps  Institution  of  Ocean- 
ography; Dr.  Maurice  Bessman,  Johns 
Hopkins  University;  Dr.  Liebe  F.  Cava- 
lieri,  Sloan-Kettering  Institute;  Dr. 
Ludwig  E.  Feinendegen,  Brookhaven 
National  Laboratory;  Dr.  Allen  Fox, 
Michigan  State  University;  Dr.  Kathryn 
Fuscaldo,  St.  John's  University;  Dr.  A.  F. 
Graham,  Wistar  Institute;  Dr.  Samson 
R.  Gross,  Duke  University  School  of 
Medicine;  Dr.  Berwind  N.  Kaufmann, 
Johns  Hopkins  Hospital;  Dr.  Victor  R. 
Larsen,  Adelphi  College;  Dr.  Arthur  B. 
Pardee,  Princeton  University;  Dr.  J.  L. 
Sirlin,  University  of  Edinburgh;  Dr. 
Stephen  Taub,  Harvard  University;  Dr. 
H.  Ursprung,  University  of  Zurich  and 
Johns  Hopkins  University;  and  Dr. 
Geoffrey  Zubay,  Brookhaven  National 
Laboratory.  Other  meetings  were  devoted 
to  discussion  of  work  within  the  Depart- 
ment or  at  the  Biological  Laboratory, 
presented  by  Drs.  Arthur  Chovnick, 
Helen  Gay,  Edward  Goldberg,  Christoph 
Jungwirth,  B.  P.  Kaufmann,  Margaret 
McDonald,  Paul  Margolin,  and  H.  E. 
Umbarger. 

As  was  reported  briefly  last  year,  the 
twenty-sixth  annual  Cold  Spring  Harbor 
Symposium  on  Quantitative  Biology, 
sponsored  by  the  Biological  Laboratory, 
was  held  in  the  Department's  Lecture 
Hall  from  June  4  to  June  12,  1961.  This 
symposium  on  regulatory  mechanisms, 
with  its  brilliant  and  timely  contributions 
to  an  understanding  of  gene  action  and 
interaction,  served  to  underline  the 
central  role  played  by  modern  genetics  in 
effecting  a  rapprochement  among  scien- 
tists   of    divers    backgrounds    and    thus 


440  CARNEGIE     INSTITUTION     OF      WASHINGTON 

facilitating  the  "breakthroughs"  that  are  During  a  visit  to  Cold  Spring  Harbor,  Dr. 

the  hallmark  of  present-day  genetics.  Berwind    N.    Kaufmann    extended    his 

Although    the    laboratories    at    Cold  studies   of   the   action   of   deoxyribonu- 

Spring  Harbor  have  always  provided  in  cleases  and  base  analogues  on  chromo- 

many   ways   an   ideal   setting   for   such  somes  of  human  blood  cells  maintained 

meetings  as  the  Symposia  on  Quantitative  in  culture. 

Biology,  an  element  of  physical  comfort  During  the  past  year  Dr.  McClintock 

has  sometimes  been  lacking  in  assembly  has  continued  as  a  research  associate  of 

rooms  during  hot  and  humid  weather.  Columbia    University,    and    Drs.    Gay, 

Therefore  it  is  a  pleasure  to  report  that  Kaufmann,    and    McClintock    as    guest 

an  air-conditioning  system,  installed  in  investigators    of    Brookhaven    National 

the  Lecture  Hall  during  the  winter  and  Laboratory;   Dr.   McClintock  grew  her 

spring  of  1962,  was  completed  in  time  for  maize   crop  at  Brookhaven  during  the 

the  twenty-seventh   Symposium,   which  summer  of  1962.  Gay  has  served  as  an 

was  held  June  7  to  13,  1962.  Fifty-four  associate  editor  of  Biological  Abstracts; 

speakers    took    part    in    this    program,  Kaufmann   as    an   editor    of    Biological 

"Basic    Mechanisms    in    Animal    Virus  Abstracts   and   associate   editor   of    The 

Biology,"  which  was  attended  by  approx-  Nucleus  and  the  International  Journal  of 

imately  200  participants.  Radiation  Biology;  and  Hershey  on  the 

The  Lecture  Hall  was  also  utilized,  in  editorial  or  advisory  boards  of  Genetics, 
the  summers  of  1961  and  1962,  for  Virology,  and  the  Journal  of  Molecular 
lectures  and  seminars  of  the  series  of  Biology.  Kaufmann 's  term  as  a  member 
advanced  courses,  offered  by  the  Bio-  of  the  Executive  Committee  of  the 
logical  Laboratory  to  research  workers,  Division  of  Biology  and  Agriculture, 
entitled  Bacterial  Genetics,  Bacterial  National  Research  Council,  terminated 
Viruses,  and  Microbiology  of  Vertebrate  on  June  30,  1962,  after  many  years  of 
Cells  and  Quantitative  Animal  Virology,  active  association.  As  past-president  of 
From  August  29  to  September  1,  1961,  the  Genetics  Society  of  America,  he  has 
the  annual  Phage  Meeting,  organized  by  continued  to  serve  on  the  Executive 
Dr.  Hershey  and  Dr.  Burgi,  was  attended  Committee  of  the  Society, 
by  about  100  workers  engaged  in  bac- 
terial virus  research.  Abstracts  of  the 
talks  presented  at  that  meeting  were 
mimeographed  at  the  Department  and  As  staff  members  in  the  Genetics 
issued  in  the  form  of  a  Phage  Information  Research  Unit  of  the  Institution,  Dr. 
Service  bulletin.  The  1962  Phage  Meeting  McClintock  and  Dr.  Hershey  are  con- 
was  scheduled  to  be  held  in  Cold  Spring  tinuing  to  carry  on  their  work  in  the 
Harbor  from  August  21  to  24.  laboratories  at  Cold  Spring  Harbor. 

The  departmental  program  was  aug-  Dr.  Gay,  while  remaining  on  the 
mented  during  the  year  by  the  presence  Genetics  Research  staff  of  the  Institution, 
of  several  fellows  and  guest  investigators,  transferred  her  program  in  cytogenetics 
Dr.  Fred  R.  Frankel,  Postdoctoral  Fellow  and  electron  microscopy  to  the  LTniversity 
of  the  U.  S.  Public  Health  Service,  of  Michigan,  where  she  has  been  accorded 
continued  to  work  in  Dr.  Hershey 's  the  title  Professor  of  Zoology.  During  the 
group,  which  was  joined  in  October  1961  summer  of  1962  she  attended  the  Second 
by  Dr.  Edward  Goldberg,  Fellow  of  The  International  Congress  of  Radiation  Re- 
National  Foundation.  Dr.  C.  C.  Das,  on  search,  at  Harrogate,  England,  and 
leave  from  Allahabad  University  in  India,  participated  in  the  International  Sym- 
held  an  appointment  from  the  Biological  posium  on  Repair  from  Radiation  Dam- 
Laboratory  as  an  exchange  visitor  to  age  and  Differential  Radiosensitivity  in 
work    in    Dr.    Kaufmann's    laboratory.  Germ  Cells,  in  Leiden,  the  Netherlands. 


I 

! 


Terminal  Activities 


DEPARTMENT    OF   GENETICS  441 

After   his   retirement   Dr.    Kaufmann  assistant  librarian  during  the  extra  hours 

traveled  to  Europe  to  participate  in  the  of  coverage, 
international    congresses    on    radiation 

biology.  When  he  returned  to  the  United  D          ,   n 

c,    .       i       ii                        i         i--i-  Hesearch  Froqrams 
States    he    took   up    research    activities 

at  the  University  of  Michigan,  where  he  Kaufmann  and  Gay  and  their  associ- 

had    been    appointed    to   the    posts    of  ates  have  continued  with  cytogenetic  and 

Professor  in  the  Department  of  Zoology  cytochemical    studies    of    the    changes 

in    the    College    and    Senior    Research  occurring  in  the  organization  of  chromo- 

Scientist  in  the  Institute  of  Science  and  somes  and  cytoplasmic  organelles  during 

Technology.  growth   and    differentiation   of   cells   in 

higher   organisms.    Several   new   experi- 

1  he  Library  mental    approaches    have    shown    that 

Mrs.  G.  C.  Smith,  Librarian,  reports  chromosomes  behave  at  times  as  inte- 

that  between  July  1,  1961,  and  June  30,  grated  fabrics,   although  their  essential 

1962,    256    books    were    added    to    the  coding  material  is  DNA.  Agents  reacting 

Department  library.  Of  these,  53  were  specifically    with     DNA — the    enzymes 

purchased,  22  were  received  by  gift  or  DNase  I  and  II,  and  the  base  analogue 

exchange,  and  181  were  volumes  of  newly  5-bromodeoxy uridine — have  been  found 

bound  periodicals.  The  number  of  books  to  induce  mutations  and   chromosomal 

now   catalogued,   exclusive   of   unbound  aberrations;  but  a  refined  analysis  of  the 

publications,  exceeds  20,000.  results  suggests  that  the  whole   DNA- 

The  recent  acquisition  of  volumes  I  RNA-protein  chromosomal  complex  re- 
(1857)  and  III  (1860)  of  Contributions  to  sponds  as  a  unit,  not  the  DNA  alone. 
the  Natural  History  of  the  United  States,  The  finding  that  far-red  radiation 
by  Louis  Agassiz,  which  were  presented  increases  the  frequency  of  double-cross- 
to  the  library  by  Mr.  Walter  K.  Earle  of  over  types  in  Drosophila  has  been 
Oyster  Bay,  completes  a  set  of  four  interpreted  by  this  group  as  evidence  of 
volumes  of  this  historic  work.  Volumes  the  essential  role  of  RNA  in  crossing  over. 
II  and  IV  represent  an  earlier  gift  from  The  marked  changes  in  basicity  of  the 
Mr.  Earle.  DNA-associated     protein     during     late 

The  library  served  investigators  and  cleavage  mitoses  and  during  spermatele- 

visitors  at  the  Biological  Laboratory  as  osis  in  Drosophila  appear  to  be  significant 

well    as    those    at    the    Department    of  indices  to  functional  states.  Other  changes 

Genetics.   In  addition,  interlibrary  loan  of  function  have  been  detected  in  studies 

facilities  were  extended  to  Adelphi  Col-  of    cytoplasmic    fine    structure    in    the 

lege,  the  Canadian  Department  of  Agri-  vegetative  cell  during  microsporogenesis 

culture  in  Ottawa,  Dartmouth  College,  in    Tradescantia.    The   highly   metabolic 

Glen  Cove  Community  Hospital,  Haskins  vegetative  cell  increases  greatly  in  volume 

Laboratories,  Inc.,  New  York  Botanical  during  this  process,  and  the  cytoplasmic 

Garden,  Republic  Aviation  Corporation,  organelles — endoplasmic  reticulum,  mito- 

State  University  College  on  Long  Island,  chondria,    and    Golgi    bodies — increase 

Time  and  Life  magazines,  Tufts  Univer-  greatly  in  numbers.  The  changes  observed 

sity,  University  of  Vermont,  and  Wilbur  in  the  Golgi  body  show  that  this  organelle 

Cross  Library  in  Storrs,  Connecticut.  has    different    forms,    depending    on    its 

In  the  summer  months  of   1961   and  functional  or  developmental  state. 

1 962,  supplementary  library  services  were  In  continuation  of  her  studies  of  DNase 

provided  by  the  Biological  Laboratory  on  II,  McDonald  has  noted  that  there  is 

evenings  and  weekends,  for  the  benefit  of  more  than  one  enzyme  of  this  type  in  calf 

students    in    the    Laboratory's    summer  spleen  but  apparently  only  one  in  salmon 

courses.    Miss    Ann    Carroll    served    as  testes.  The  salmon-testis  enzyme  has  a 


442 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


molecular  weight  of  about  52,000.  It 
degrades  heat-denatured  DNA  at  only 
one-tenth  the  rate  of  its  degradation  of 
native  DNA;  nevertheless,  its  action  on 
heat-denatured  DNA  appears  to  be  due 
to  an  intrinsic  property  of  the  enzyme 
and  not  to  an  impurity. 

Many  samples  of  crystalline  RNase 
contain  a  nonenzymic  component  which, 
on  the  basis  of  its  ultraviolet  absorption 
spectra  and  dialytic  properties,  is  believed 
to  be  polynucleotide  in  nature.  McDonald 
has  isolated  this  component  from  various 
RNase  samples  and  is  now  characterizing 
it.  She  has  found  that  the  sugar  compo- 
nent is  primarily  deoxyribose. 

McDonald  and  Gay  have  shown  that 
the  ability  of  crystalline  RNase  to  reduce 
methyl  green  stainability  of  fixed  bio- 
logical sections  without  impairment  of 
Feulgen  colorability  is  due  not  to  RNase 
per  se,  nor  to  the  polynucleotide  impurity, 
but  to  a  heretofore  unrecognized  contam- 
inating protein.  This  material  has  been 
separated  from  the  bulk  of  the  RNase  by 
ion-exchange  chromatography,  but  its 
chemical  action  has  not  yet  been  identi- 
fied. Until  that  has  been  established,  the 
previous  hypothesis  of  these  workers, 
that  chromosomes  contain  a  type  of 
structural  RNA  connecting  chains  of 
DNA,  should  be  held  in  abeyance.  On  the 
other  hand,  McDonald  and  Kaufmann 
have  noted  that  even  "chromatograph- 
ically  pure"  cytochrome  c  is  capable  of 
reducing  pyronin  stainability,  a  phenom- 
enon for  which  they  still  have  no  reason- 
able explanation. 

McClintock  reports  that  some  mech- 
anisms associated  with  control  of  gene 
action  appear  to  be  similar  in  maize  and 
in  bacteria.  Certain  gene-control  systems, 
in  both  these  organisms,  are  composed 
basically  of  two  elements,  an  operator  and 
a  regulator.  In  bacteria,  both  elements 
may  be  located  near  the  gene  controlled 


by  the  system  (structural  gene);  or  only 
the  operator  may  be  located  there,  and 
the  regulator  elsewhere  in  the  genome.  In 
maize,  a  number  of  examples  of  the 
second  type  have  been  detected  among 
both  the  Spm  (Suppressor-mutator)  and 
Ac  (Activator)  systems.  Sometimes,  how- 
ever, in  each  system,  it  is  evident  that 
the  regulator  element  initially  occupied  a 
position  close  to  the  structural  gene.  If 
the  circumstances  are  similar  to  those  in 
bacteria,  the  operator  element  likewise 
should  have  been  present  near  the  gene 
locus.  Experiments  have  been  conducted 
to  test  this  inference,  and  all  the  results 
support  the  supposition  that  in  such 
instances  both  the  operator  and  regulator 
elements  of  a  system  are  originally 
located  close  to  the  structural  gene. 

Hershey  has  found  that  measurement 
of  the  shearing  forces  required  to  break 
DNA  molecules,  combined  with  chro- 
matographic measurement  of  the  break- 
age, apparently  offers  a  general  method 
for  determining  molecular  weights  of 
DNA.  This  method  reveals  for  T5  DNA 
a  molecular  weight  of  84  million,  indi- 
cating that  there  is  one  DNA  molecule 
per  phage  particle. 

The  DNA  of  phage  lambda  is  charac- 
terized by  an  unusual  intramolecular 
heterogeneity  of  interbase  bonding 
strengths;  it  also  exhibits  readily  demon- 
strable intermolecular  interactions. 
Hershey  and  his  colleagues  think  these 
properties  may  be  related,  and  may  offer 
a  clue  to  the  forces  that  operate  in 
synapsis  of  chromosomes. 

Shearing  forces  that  break  DNA  mole- 
cules can  also  produce  local  defects  in 
structure.  The  two  phenomena  occur 
independently,  however,  and  stirring  at 
low  temperatures  and  high  salt  concen- 
trations permits  breakage  without  de- 
naturation,  as  required  in  biological 
experiments  with  molecular  fragments. 


DEPARTMENT    OF    GENETICS  443 

GROWTH   AND   INHERITANCE  IN   BACTERIOPHAGE 

A.  D.  Hershey,  Elizabeth  Burgi,  Fred  Frankel,  Edward  Goldberg,  and  Laura  Ingraham 

Several    methods    applicable    to    the  As  a  preliminary,  a  sample  of  DNA 

characterization  of  DNA  molecules  have  isolated  from  phage  T2  is  stirred  under 

been  developed  in  recent  years.  Examples  conditions    that    produce    single,    clean, 

are    the    optical    analysis    of    thermal  transverse    breaks    near   the    centers    of 

denaturation,  chromatographic  analysis,  molecular  length.   The  DNA  is  labeled 

measurement  of  fragility,  measurement  of  with  radiophosphorus,  because  the  tracer 

buoyant  density,  and  specific  enzymatic  permits    analysis    of    extremely    dilute 

tests.  Such  methods  make  it  possible  to  solutions  (less  than  0.1  fxg  of  DNA  per 

distinguish  DNA's  from  different  sources  ml)  in  which  molecular  interactions  can 

and  to  detect  alterations  produced  experi-  be  neglected.   The  stirred  solution  now 

mentally.   They  do  not,  however,  yield  consists  of  fragments  of  DNA  molecules 

direct  information  about  molecular  struc-  ranging  in  length  from  about  }/&  to  %  of 

ture.  Structure  is  a  more  or  less  plausible  the  length  (50  /z)  of  unbroken  T2  DNA 

inference   that   serves   to   unify   diverse  molecules.  The  fragments  are  next  sorted 

measurements,  as  did,  for  example,  the  out  into  length  classes  by  chromatography 

notably  successful  model  of  Watson  and  on  a  column  of  methylated  bovine  serum 

Crick.  albumin.     The    separation    is    possible 

These  remarks  sufficiently  explain  why  because  the  basic  protein  in  the  column 

our  work,  though  directed  toward  rather  acts  as  an  ion  exchanger  from  which  the 

specific  biological  goals,  is  for  the  moment  acidic  DNA  fragments  are  removed  at 

devoted  to  the  exploration  of  physical  different  salt  concentrations,  depending 

techniques.  on  their  length.  The  resulting  fractions 

mi     -nr  i      7     ttt  ■  i      r  m    nir  i  are   listed   in   table    1,   where   they   are 

The  Molecular  Weight  of  T5  DNA  characterized    by    their    sedimentation 

The  above  generalities  are  illustrated  coefficients  and  corresponding  molecular 

in  experiments  of  the  following  type,  by  weights  (Year  Book  60). 

which  we  arrive  at  an  estimate  of  the  The    DNA   of   phage    T5   exhibits   a 

molecular  weight  of  the  DNA  of  T5  and,  sedimentation  coefficient  of  48.5  S.  From 

more  important,  evaluate  a  novel  tech-  table  1  we  see  that  fragments  of  T2  DNA 

nique.  sedimenting   at   the   same   rate   have   a 


TABLE  1.     Equal  Fragility  under  Hydrodynamic  Shear  of  T5  DNA  and  Fragments  of  T2  DNA  of 

Similar  Sedimentation  Coefficient 


Chromatographic 

Sedimentation 

Molecular 

Breakage  at 

Sample 

Interval, 

% 

Coefficient, 

S 

Weight,  X  10"6 

630  rpm,  % 

T2  DNA  frag- 

ments 

tube  30 

10.6 

31.8 

37 



tube  31 

23.4 

36.2 

47 



tube  32 

36.6 

38.6 

55 



tube  33 

49.4 

42.4 

64 

0 

tube  34 

60.7 

44.1 

71 

15 

tube  35 

69.8 

44.9 

73 

27 

tube  36 

77.9 

47.2 

79 

51 

tube  37 

83.8 

48.5 

82 

60 

tube  38 

88.5 

49.5 

86 

73 

tube  39 

92.3 

51.0 

89 

95 

T5DNA 



48.5 

? 

65 

444 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


molecular  weight  of  82  million.  It  does 
not  immediately  follow  that  this  is  the 
molecular  weight  of  T5  DNA,  however, 
because  sedimentation  constants  depend 
on  molecular  shape  as  well  as  molecular 
weight,  and  are  also  influenced  by 
molecular  interactions  that  may  differ 
from  one  DNA  to  another. 

To  circumvent  the  difficulties  last 
mentioned,  we  apply  a  very  different 
criterion  to  the  same  materials.  The  last 
column  in  table  1  gives  the  results  of 
fragility  tests,  in  which  we  measure  the 
fraction  of  DNA  broken  when  very  dilute 
solutions  are  stirred  for  30  minutes  at 
630  rpm.  From  the  results  with  the  T2 
fragments,  it  is  clear  that  molecular 
fragility  is  a  sensitive  index  of  molecular 
weight.  T5  DNA  exhibits  a  fragility 
corresponding  to  that  of  T2  fragments  of 
molecular  weight  lying  between  82  and 
86  million.  In  other  words,  T5  DNA 
matches  practically  the  same  fragments 
of  T2  DNA  either  in  terms  of  sedimenta- 
tion coefficient  or  in  terms  of  fragility. 
The  agreement  permits  the  following 
conclusions. 

1.  A  single  relation  between  molecular 
weight  and  sedimentation  velocity  applies 
to  both  T2  DNA  and  T5  DNA. 

2.  A  single  relation  between  molecular 
weight  and  fragility  under  hydrodynamic 
shear  applies  to  both  DNA's. 

3.  The  bonds  broken  by  stirring  must 
be  of  equal  strength  in  both  DNA's. 

4.  The  molecular  weight  of  T5  DNA  is 
about  84  million,  as  compared  with  130 
million  for  the  DNA  of  T2. 

5.  Since  T2  DNA  shows  about  the 
proper  mass  per  unit  length  for  a  double 
helical  structure,  the  DNA  of  T5  must 
also  have  this  structure. 

6.  Since  particles  of  phage  T2  contain 
a  single  molecule  of  DNA,  the  same  must 
be  true  of  T5  particles,  which  are  some- 
what smaller. 

These  conclusions  are  interrelated  in 
such  a  way  that  all  must  be  correct  or  all 
or  most  of  them  incorrect.  That  they  are 
correct  follows  as  a  plausible  inference 
from  the  data   presented,   although   no 


single  measurement  forces  this  conclusion. 
The  alternative  can  be  rejected  on  the 
basis  that  it  would  require  an  improbable 
set  of  coincidences. 

Finally,  independent  checks  of  any  of 
the  stated  conclusions  reinforce  them  all. 
Such  checks  will  be  reported  in  another 
publication. 

The  DNA  of  Phage  Lambda 

Comparisons  of  the  sort  illustrated  in 
table  1  can  also  bring  to  light  DNA's 
having  unusual  properties.  The  DNA  of 
phage  lambda  is  an  example.  It  is  unusual 
in  the  following  respects. 

1.  It  does  not  emerge  from  our 
fractionating  column  in  a  single  band,  but 
trails  over  a  wide  range  of  salt  concen- 
trations and  fails  to  elute  completely. 

2.  It  shows  a  broad  range  of  denatura- 
tion  temperatures,  similar  to  that  of  the 
bacterial  DNA's  and  in  contrast  to  the 
exceedingly  narrow  range  characteristic 
of  other  phage  DNA's.  A  comparison  with 
the  DNA  of  phage  Tl,  chosen  for  its 
similar  composition  and  molecular  size,  is 
made  in  figure  1.  The  flatter  curve  for 


CD 

o 

c 
o 
.o 

o 

CO 

o 

> 

cr 


Temperature,  °C 

Fig.  1.     Thermal  denaturation  of  DNA  from 
phages  Tl  and  lambda. 


lambda  DNA  probably  reflects  the  char- 
acter of  the  individual  molecules,  though 
the  possibility  that  the  DNA  is  composed 
of  a  mixture  of  species  with  different 


DEPARTMENT    OF    GENETICS 


445 


melting  temperatures  has  not  been  ex- 
cluded. 

3.  Lambda  DNA  forms  a  broad  or 
double  boundary  in  the  optical  ultra- 
centrifuge.  Zone  centrifugation  according 
to  methods  developed  by  Britten  and 
Roberts  at  the  Department  of  Terrestrial 
Magnetism  confirms  that  even  in  very 
dilute  solutions  the  DNA  exists  in  two 
or  more  differently  sedimenting  forms. 

4.  On  stirring  at  low  speeds,  too  low  to 
break  DNA  molecules  of  the  maximum 
molecular  weight  (50  million)  represented 
by  the  DNA  content  of  the  phage 
particle,  lambda  DNA  is  converted  into 
a  form  homogeneous  by  ultracentrifugal 
criteria  and  exhibiting  the  sedimentation 
coefficient  (32  S)  corresponding  to  the 
slowest  form  present  in  unstirred  DNA. 
The  behavior  on  heating  and  chroma- 
tography is  not  affected  by  this  treatment. 

These  properties  suggest  a  diversity  of 
molecular  shapes  or  aggregation  products, 
together  with  and  perhaps  related  to 
other  structural  peculiarities  responsible 
for  the  unusual  response  to  heating.  It  is 
evident  that  different  DNA's  do  exhibit 
idiosyncrasies  not  yet  accounted  for  in 
structural  terms,  and  it  is  interesting  that 
an  example  should  be  found  in  phage 
lambda,  itself  a  biological  curiosity  in 
several  respects. 

A  cursory  examination  of  DNA  from 
phages  Tl  and  P22,  as  well  as  more 
thorough  studies  of  T2  and  T5,  has  not 
revealed  comparable  unexpected  prop- 
erties. Since  lambda  is  a  phage  of  well 
known  and  complex  genetic  properties, 
we  propose  to  continue  our  physical 
studies  in  the  hope  of  reaching  conclusions 
of  biological  pertinence. 

The  Nature  of  Self-Protection 

Some  years  ago  we  noted  that  the 
breakage  of  DNA  by  stirring  depends 
strongly  on  the  concentration  of  the 
solution.  In  more  recent  work  we  find 
that,  as  more  and  more  dilute  solutions 
of  T2  DNA  are  subjected  to  stirring, 
susceptibility  to  breakage  increases  to  a 
maximum  at  a  concentration  lying  be- 


tween 0.2  and  0.1  jug/ ml.  Further  dilution 
is  without  effect.  At  0.2  /zg/ml,  the 
molecules  are  separated  by  an  average 
distance  of  about  28  /x,  or  half  the  actual 
length  of  the  molecules.  Self-protection  is 
therefore  a  remarkably  sensitive  index  of 
molecular  interactions. 

Such  interactions  could  involve  specific 
intermolecular  forces  of  biological  inter- 
est. We  have  looked  for  evidence  of  such 
forces  by  testing  the  ability  of  T2  DNA 
to  protect  that  of  T5,  and  vice  versa.  We 
find  that  the  protection  is  not  specific  and 
therefore  does  not  involve  specific  inter- 
molecular bonding. 

The  protective  effect  greatly  depends 
on  the  length  of  the  molecules.  We 
suspect  that  it  results  simply  from  their 
viscous  drag,  which  may  reduce  the 
maximum  local  shear  rate  in  the  stirring 
vessel  without,  of  course,  affecting  the 
average  rate  of  shear  in  the  vessel  as  a 
whole. 

Local  Denaturation  by 
Hydrodynamic  Shear 

For  a  number  of  experimental  purposes 
it  is  desirable  to  be  able  to  fragment  DNA 
molecules  without  producing  unwanted 
side  effects.  In  the  course  of  our  work  with 
T5  DNA  we  found,  however,  that  stirring 
can  produce  denaturation  as  well  as 
breakage.  In  order  to  understand,  and  the 
better  to  avoid,  such  denaturation,  we 
studied  it  in  some  detail.  The  principal 
facts  are  the  following. 

1.  When  T5  DNA  is  stirred  in  salt 
solution  at  25°C  at  speeds  just  sufficient 
to  initiate  breakage  of  the  molecules, 
subsequent  chromatography  reveals  that 
a  considerable  fraction  of  the  DNA  has 
been  altered  in  such  a  way  that  it  attaches 
irreversibly  to  the  basic  protein  in  the 
column. 

2.  Such  denaturation  can  be  demon- 
strated also  by  the  action  of  Lehman's 
phosphodiesterase,  a  bacterial  enzyme 
that  acts  on  denatured  but  not  native 
DNA. 

3.  The  extent  of  denaturation,  meas- 
ured by  either  of  these  methods,  increases 


446 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


as  the  salt  concentration  at  the  time  of 
stirring  is  lowered  or  the  temperature 
raised.  Salt  concentration  and  tempera- 
ture do  not  affect  breakage,  however.  At 
5°C  and  0.6  M  NaCl,  or  at  25°C  and 
2.6  M  NaCl,  DNA  can  be  broken  with 
very  little  denaturation. 

4.  The  speed  of  stirring  is  also  critical, 
depending,  of  course,  on  the  other 
variables  mentioned.  At  25°  in  0.6  M 
NaCl,  little  denaturation  is  produced  at 
speeds  too  slow  to  cause  breakage. 
Remarkably,  rapid  stirring  also  fails  to 
produce  denaturation,  which  requires  a 
critical  stirring  speed  just  sufficient  to 
break  the  molecules  slowly.  At  higher 
temperatures  and  lower  salt  concentra- 
tions, denaturation  can  be  produced 
either  without  breakage,  by  stirring  at 
low  speeds,  or  in  spite  of  rapid  breakage 
caused  by  stirring  at  high  speeds. 

5.  The  existence  of  a  critical  stirring 
speed  for  denaturation  is  explained  by 
the  fact  that  DNA  fragments  produced 
without  denaturation  by  stirring  at  a  low 
temperature  are  relatively  resistant  to 
denaturation  on  restirring  at  a  higher 
temperature. 

6.  The  critical  speed  at  which  denatur- 
ation occurs  varies  with  DNA  concen- 
tration in  the  same  way  as  the  speed 
required  to  produce  breakage:  high 
concentrations  protect  against  both  de- 
naturation and  breakage. 

7.  If  a  sample  of  DNA  is  stirred  under 
conditions  that  produce  both  partial 
denaturation  and  partial  breakage,  the 
denaturation  can  be  demonstrated  in  both 
broken  and  unbroken  molecules.  Simi- 
larly, molecules  previously  denatured 
without  breakage  can  be  shown  to  exhibit 
about  the  same  susceptibility  to  breakage 
on  restirring  as  their  native  counterparts. 
Breakage  and  denaturation  are  inde- 
pendent events. 

8.  The  denaturation  produced  by  stir- 
ring does  not  alter  the  characteristic 
melting  curve  of  the  DNA. 

9.  When  DNA  denatured  by  stirring  is 
restirred  at  low  temperature  to  reduce 
the  size  of  the  fragments,  the  fraction 


that  will  subsequently  pass  a  column  is 
greatly  increased  but  the  susceptibility  to 
phosphodiesterase  is  not  affected. 

10.  This  type  of  denaturation  cannot 
be  repaired  by  a  regime  of  gentle  heating 
and  slow  cooling. 

1 1 .  Denaturation  showing  all  the  above 
characteristics  can  also  be  produced  by 
heating  DNA  in  the  absence  of  shear  to 
temperatures  (84°  to  88°)  near  the  mid- 
point of  the  melting  curve. 

These  results  can  be  interpreted  in  the 
following  way. 

When  a  DNA  solution  is  heated  to  a 
characteristic  range  of  temperatures,  the 
weak  bonds  responsible  for  maintaining 
the  two-stranded  configuration  are  loos- 
ened and  local  separation  of  strands  is 
permitted,  even  though  a  sufficient 
number  of  bonds  persists  to  hold  the  two 
chains  in  apposition.  This  effect  can  be 
measured  by  the  increased  absorption  of 
ultraviolet  light  as  the  temperature  is 
raised.  A  major  factor  in  the  process  is 
the  purely  mechanical  effect  of  Brownian 
motion.  When  the  solution  is  cooled  at 
this  stage,  the  original  structure  is 
regained  as  far  as  can  be  determined  by 
optical  measurements.  By  chromato- 
graphic or  enzymatic  tests,  however,  local 
molecular  lesions  can  be  shown  to  persist. 
If  the  molecules  are  now  fragmented 
without  further  denaturation,  the  dena- 
tured and  undenatured  regions  of  indi- 
vidual molecules  are  separated  and  the 
weight  fraction  of  the  DNA  that  can  pass 
a  column  is  increased. 

Subjecting  the  molecules  to  shearing 
forces  can  produce  the  same  result  at  low 
temperatures :  in  effect  such  forces  merely 
lower  the  temperature  of  thermal  de- 
naturation, as  does  decreasing  the  salt 
concentration.  The  stresses  generated  by 
shear  are  of  a  special  character,  however, 
in  that  they  depend  not  only  on  the  speed 
of  stirring  but  also  on  the  length  of  the 
molecules.  They  are  localized,  moreover, 
near  the  centers  of  molecular  length. 
Thus  if  a  collection  of  fragments  of 
different  lengths  is  subjected  to  stirring 
at  a  given  speed,  the  shortest  ones  will 


DEPARTMENT    OF    GENETICS  447 

survive  because  subjected  to  little  stress,  and  then  only  if  the  measurement  is  made 
the  longest  will  quickly  be  broken  to  at  about  45°C.  By  chromatography  or  by 
fragments  that  are  likewise  resistant  to  tests  with  phosphodiesterase,  however, 
denaturation,  and  only  those  of  a  critical  complete  denaturation  can  be  demon- 
length  will  resist  breakage  and  at  the  strated.  Thus  denaturation  produced  at 
same  time  undergo  the  repeated  mechan-  100°C  is  irreversible  on  rapid  cooling,  but 
ical  distortions  that  eventually  produce  optical  criteria  are  poor  measures  of  it. 
permanent  local  denaturation.  Only  if  the  As  is  well  known,  measurements  of  the 
original  population  consists  of  molecules  absorption  of  ultraviolet  light  serve 
of  uniform  length  can  a  majority  of  them  admirably  in  following  the  progress  of 
be  denatured  in  this  way  at  low  tempera-  denaturation  during  the  actual  heating 
tures,  and  then  only  at  the  critical  speed  (fig.  1).  According  to  this  criterion,  T5 
of  stirring  appropriate  to  that  length,  DNA  is  half  denatured  at  85°C.  However, 
namely,  the  maximum  speed  that  just  if  the  solution  is  heated  to  85°  and  then 
fails  to  cause  rapid  breakage.  If  the  cooled,  either  rapidly  or  slowly,  little 
temperature  is  sufficiently  low  and  the  denaturation  is  detectable  by  enzymatic 
salt  concentration  sufficiently  high,  the  tests.  What  little  there  is  is  due  to  small 
stresses  required  for  denaturation  exceed  imperfections  in  many  of  the  molecules, 
those  required  for  breakage,  and  denatur-  which  prevent  them  from  passing  a 
ation  cannot  be  observed.  fractionating  column. 

As  far  as  we  can  surmise,  permanent  Thus  denaturation  on  heating  occurs 
denaturation  of  this  type,  whether  the  simultaneously  in  all  the  molecules  but  is 
consequence  of  gentle  heating  or  of  largely  reversible  unless  a  temperature  of 
stirring,  must  result  when  locally  sepa-  about  90°  is  exceeded.  The  irreversible 
rated  strands  rejoin  out  of  register,  event  occurring  at  this  temperature  is  not 
producing  unpaired  loops  or  similar  reflected  in  the  optical  density  measure- 
structural  irregularities.  As  can  be  demon-  ments;  it  undoubtedly  consists  in  the 
strated  with  fully  denatured  DNA,  such  unwinding  of  the  polynucleotide  chains  of 
structures  have  a  strong  affinity  for  the  the  helix,  which  have  already  largely 
basic  protein  of  our  fractionating  columns  separated  at  lower  temperatures, 
and  are  the  natural  substrate  for  Leh-  The  behavior  described  above  is  pre- 
man's  phosphodiesterase.  sumably  characteristic  of  a  molecularly 

In  short,  local  denaturation  of  DNA  homogeneous     DNA     and     explains     a 

molecules    can   be    produced    by    either  puzzling  feature  of  experiments  performed 

heating  or  stirring,  the  effects  of  which  in  the  past  with  inhomogeneous  DNA 

are  identical  except  in  one  respect.  When  preparations.  In  the  earlier  experiments 

hydrodynamic  shear  is  a  dominant  cause  it  was  found  that  progressive  denatura- 

of   denaturation,    the   molecular   lesions  tion  involved  the  irreversible  "collapse" 

tend  to  be  central  with  respect  to  molecu-  of    individual     molecules     at     different 

lar  length.    Heating  in  the   absence   of  temperatures,  indicating  that  the  dena- 

shear,  on  the  contrary,  tends  to  produce  turation  of  a  given  molecule  was  an  "all 

terminal  defects.   These   differences   are  or  none"  event. 

readily  accounted  for  theoretically  and  It   is   now   clear   that   it   is   not   the 

can  be  demonstrated  experimentally.  denaturation  measured  by  optical  means 

(i.e.,  the  loss  of  regular  structure)  that 

mi     „         .7  „.,      .  „7          .  _              ,.  constitutes  the  all-or-none  event,  but  the 

The  Reversibility  of  Thermal  Denaturation  subgequent  separation  0f  chains.  And  the 

When  a  solution  of  T5  DNA  is  heated  failure  to  observe  partly  denatured  mole- 
to  100°C  for  10  minutes  and  then  quickly  cules  after  heating  to  intermediate  tem- 
cooled  in  ice  water,  relatively  little  peratures  and  subsequent  cooling  (by 
denaturation  is  evident  by  optical  criteria  electron  microscopy,  for  example)  is  due 


448 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


to  the  fact  that  partial  denaturation  of 
individual  molecules  is  reversible  on 
cooling.  Our  experiments  now  show  that 
the  recovery  is  often  imperfect,  but  the 
resulting  local  imperfections  that  we  see 
were  not  detected  by  the  means  previ- 
ously employed. 

It  may  be  added  that  the  reversibility 
of  denaturation  of  a  sample  of  DNA 
heated  to  the  middle  range  of  its  melting 
curve  should  provide  a  new  criterion  of 
molecular  homogeneity  applicable,  for 
instance,  to  the  DNA  of  phage  lambda. 

Replicating  DNA  of  Phage  T2 

DNA  undergoing  replication  during 
growth  of  phage  T2  can  be  studied  in  two 
ways:  first,  by  infecting  bacteria  with 
isotopically  labeled  phage  particles  and 
then  examining  the  labeled  DNA  subse- 
quently isolated  from  the  infected  cells; 
second,  by  labeling  DNA  synthesized 
after  infection,  which  is  exclusively  viral, 
and  examining  it.  In  either  method 
chloramphenicol  can  be  added  to  prolong 
the  period  during  which  DNA  synthesis 
can  be  observed  without  complications 
due  to  the  re-formation  of  phage  particles. 

Experiments  of  this  type  reveal  at 
least  two  forms  of  DNA  that  differ  from 
the  finished  molecules  finally  incorporated 
into  phage  particles.  These  are  as  yet 
poorly  characterized,  and  little  can  be 
said  about  them  except  that  they  do  not 
include  any  appreciable  fraction  of  low- 


molecular- weight  DNA,  either  single  or 
double  stranded. 

On  the  other  hand,  the  cells  always 
contain  a  considerable  fraction  of  their 
total  DNA  in  a  form  indistinguishable 
from  that  found  in  finished  phage 
particles.  This  finding  shows  that  a 
mechanism  for  the  preservation  and 
determination  of  molecular  length  oper- 
ates continuously  during  replication,  and 
not  only  at  some  terminal  stage  in  the 
formation  of  a  phage  particle — a  conclu- 
sion pertinent  to  several  hypotheses 
concerning  genetic  mechanisms. 

Conclusion 

The  phage  DNA's  provide  favorable 
materials  for  investigation,  for  three 
reasons.  First,  they  can  be  isolated  in  a 
molecularly  homogeneous  state,  a  circum- 
stance that  permits  for  the  first  time 
satisfactory  correlations  between  gross 
structure  and  biological  function.  Second, 
the  synthesis  of  phage  DNA  can  be 
studied  in  infected  cells  that  have  proved 
amenable  to  metabolic  experimentation 
in  the  past.  Third,  intensive  genetic  study 
of  a  few  phage  species  is  yielding  results 
to  which  physical  and  chemical  findings 
can  be  constantly  referred. 

In  the  present  state  of  knowledge 
concerning  both  genetic  mechanisms  and 
molecular  structure,  work  of  a  purely 
exploratory  nature  is  called  for  as  well  as 
efforts  directed  toward  well  defined  goals. 


TOPOGRAPHICAL   RELATIONS   BETWEEN   ELEMENTS 
OF   CONTROL  SYSTEMS  IN    MAIZE 

Barbara  McClintock 


In  Year  Book  60,  parallels  were  drawn 
between  gene-control  systems  in  maize 
and  those  in  bacteria.  In  both  organisms, 
the  comparable  systems  are  composed, 
basically,  of  two  genetic  elements:  an 
"operator"  element,  located  adjacent  to 
the  structural  gene(s)  and  directly  con- 
trolling genie  activity;  and  a  "regulator" 
element  that  in  turn  controls  the  func- 


tioning of  the  operator  element.  Each 
operator  responds  only  to  a  specific 
regulator.  Therefore,  each  operator  with 
its  corresponding  regulator  represents  a 
gene-control  system.  In  bacteria,  the 
position  of  the  regulator  element  on  the 
chromosome  is  not  the  same  in  all 
examined  systems:  it  may  be  located 
either  near  by  or  at  a  distance  from  the 


DEPARTMENT   OF   GENETICS  449 

operator.  Probably  the  elements  of  a  directly  modify  the  action  of  a  gene  when 
control  system  in  maize  express  similar  located  close  to  it,  and  does  the  two- 
topographical  relationships ;  evidence  sup-  element  system  that  subsequently  ap- 
porting  that  probability  will  be  presented  pears  result  from  an  alteration  of  the 
here.  regulator  element  which  converts  it  into 

In  maize,  the  controlling  elements  of  a  an  element  that  thereafter  behaves  as  an 

system  are  transposable.  Therefore,  dif-  operator? 

ferent  genes  may  come  under  the  control  Experiments  eliciting  direct  answers  to 

of  the  same  system,  or  the  same  gene  the  above  questions  may  not  be  conducted 

under  the  control  of  different  systems,  readily  with  maize.  Nevertheless,  certain 

Inception  of  control  of  gene  action  by  a  relevant  facts  are  known.  Although  it  is 

particular  system  sometimes  occurs  when  possible  to  consider  that  insertion  of  a 

the  operator  of  the  system  is  inserted  known    regulator    element    close    to    a 

near  the  locus  of  the  gene,  the  regulator  structural  gene  may  initiate  control  of 

element  being  located  elsewhere  in  the  action   of  the   gene,   and   also   that   an 

chromosome  complement.  At  other  times,  operator  element  may  originate  by  some 

inception  of  control  is  associated  with  the  modification  of  the  regulator,  no  certain 

appearance  of  the  regulator  near  the  locus  evidence  has  yet  been  obtained  of  the 

of  the  gene.  In  each  such  example  that  reverse  process,  that  is,  conversion  of  an 

has  been  examined  adequately,  however,  operator    into    a    regulator.    Moreover, 

a  clearly  expressed  two-element  system  examination  of  the  behavior  of  each  of 

has    subsequently    arisen,    the    operator  the  elements  of  a  two- element  system  has 

element  residing  near  the  gene  locus  and  shown  that  transposition  of  both  elements 

responding  to  the  regulator  element,  now  to    new    locations    in    the    chromosome 

located    elsewhere    in    the    chromosome  complement  may  occur  coincidently  in  a 

complement.  single  nucleus,  so  that  the  operator  is 

With  advancing  knowledge  of  control  removed  from  the  locus  of  the  gene  and 
systems  in  bacteria  and  of  the  topograph-  the  regulator  is  moved  to  a  new  location 
ical  relations  among  individual  compo-  in  the  chromosome  complement.  In  other 
nents  of  a  system,  the  cases  in  maize  in  cells,  on  the  other  hand,  only  one  of  the 
which  the  regulator  element  of  a  known  elements  is  transposed  at  a  given  time, 
system  initially  occupies  a  position  close  Now,  if  it  is  assumed  that  both  an 
to  the  structural  gene  require  close  operator  and  a  regulator  are  located 
scrutiny.  In  maize  genetics,  the  devised  adjacent  to  the  structural  gene  in  those 
test  methods  allow  ready  detection  of  the  instances  where  only  the  presence  of  the 
presence  of  the  regulator  element  of  a  regulator  at  that  location  can  be  deter- 
system,  either  when  it  is  located  close  to  mined  with  certainty,  the  following 
the  gene  under  the  control  of  the  system  possibilities  are  predictable.  The  operator 
or  when  it  is  located  elsewhere.  When  the  and  regulator  might  undergo  simultane- 
regulator  is  close  to  the  gene  locus,  the  ous  transposition,  leaving  the  structural 
concomitant  presence  of  the  operator  gene  with  neither  element  adjacent  to  it. 
element  is  not  so  readily  detected.  Never-  Other  transpositional  events  might  re- 
theless,  as  was  stated  above,  a  clearly  move  only  one  of  the  elements,  the 
expressed  two-element  system  of  control  complementary  element  of  the  S}^stem 
of  gene  action  does  subsequently  appear,  remaining  in  location.  With  regard  to 
The  origin  of  such  a  system  poses  several  control  of  gene  action,  the  effects  pro- 
critical  questions:  Are  both  elements  of  duced  by  these  kinds  of  transposition 
the  system  initially  located  close  to  the  would  lead  to  three  distinctly  different 
structural  gene,  and  does  the  two-element  consequences.  Removal  of  both  elements 
system  arise  by  removal  of  the  regulator  would  release  the  structural  gene  from 
only?  Or  can  a  regulator  element  alone  control    by    the    system    to    which    the 


450  CARNEGIE     INSTITUTION      OF      WASHINGTON 

elements    belonged,    and    genetic    tests  ~  .  .     .               ,    .     „       „7 

would  reveal  the  absence  of  the  regulator  0n<>m  P^i  ^  of  a  Two-Element 

element  from  the  vicinity  of  the  gene.  ^ 

Removal  of  the  operator  element  only  The  original  isolate  of  a im~r°  had  an  Spm 

would  likewise  release  the  structural  gene  element  located  close  to  the  A  i  gene ;  the 

from  control  by  the  system  to  which  it  degree  of  closeness  was  not  made  apparent 

belonged;    but    the    regulator    element  in  the  initial  tests  of  28  plants  carrying 

would  still  be  located  close  to  the  gene  aim~b.  In  18  of  these  plants,  more  than 

locus.  Removal  of  the  regulator  alone,  two  Spm  elements  were  present;  6  had 

however,   would   give   rise   to   a   clearly  two  Spm  elements,  one  obviously  linked 

expressed  two-element  system  of  control  with  aim~5;  and  4  had  one  Spm  element, 

of  gene  action— the  same  system  that  was  linked  with  aim~b.  The  location  of  Spm 

operating  before  the  transposition.  More-  in  the  immediate  vicinity  of  the  A\  gene 

over,  the  location  of  the  regulator  element  was  not  recognized  in  the  tests  conducted 

at  some  distance  from  the  structural  gene,  with  these  last  4  plants,  because  frequent 

or  in  another  chromosome  of  the  comple-  transpositions  of  Spm,  occurring  in  some 

ment,  could  readily  be  determined.  sporogenous  or  presporogenous  cells,  re- 

The    Cold    Spring    Harbor    cultures  suited  in  the  production  of  a  number  of 

include  five  identified  instances  in  which  gametes  in  which  Spm  occupied  a  new 

the    initial    location    of    the    regulator  location  in  the  chromosome  complement, 

element  of  a  known  control  system  close  The  intimate  proximity  of  Spm  to  the  A  i 

to  a  structural  gene  resulted  in  control  of  locus   was   made   evident,    however,    in 

the  gene's  action  by  the  system  to  which  tests  of  certain  progeny  of  these  plants, 

the    regulator    belonged.    In    the    three  in  which  transpositions  of  Spm  occurred 

instances    that    have    been    adequately  so  late  in  development  that  few  or  no 

examined  so  far,  all  three  of  the  expected  gametes  carried  a  transposed  Spm  ele- 

consequences  of  transpositions  outlined  ment. 

above  have  been  confirmed.  The  evidence  With  respect  to  A\  gene  action,  trans- 
is  reviewed  below.  position  of  Spm  away  from  the  locus  of 

Three  of  the  five  examples  involve  the  aim~b  leads  to  one  or  the  other  of  two 

Ac  (Activator)  system,  and  two  the  Spm  quite  different  results:  either  release  of 

(Suppressor-mutator)    system.    Insertion  gene  action  from  the  control  of  the  Spm 

of  Ac,  the  regulator  element  of  the  Ac  system,   or   continued   control   by   that 

system,  close  to  the  locus  of  the  bronze  system.  Approximately  half  of  the  trans- 

(Bz)    gene    in    chromosome    9    initiated  positional  events  that  effect  release  from 

control  of  action  of  this  gene  by  the  Ac  control  of  the  Spm  system  result  in  an  A  i 

system.   The  modified  locus  was  desig-  gene  capable  of  a  high  level  of  action.  The 

nated  bzm~'2,  and  some  discussion  of  it  other  half  of  these  events  bring  about  a 

appears  in  Year  Books  54  and  55.  Two  much  lower  level  of  A\  gene  action,  or 

independently  occurring  insertions  of  Ac  occasionally  the  absence  of  such  action, 

close  to  the  locus  of  the  Wx  (waxy)  gene  in  both  plant  and  kernel, 

in  chromosome  9  resulted  in  control  of  That  the  Spm  system  can  continue  to 

action  of  this  gene  by  the  Ac  system,  control  A\  gene  action  after  some  trans- 

These    modified    loci    were    designated  positions  of  Spm  away  from  the  aim~5 

wxm~7  and  wxm~9.  Insertions  of  Spm,  the  locus  was  discovered  in  two  ways.  One  of 

regulator  element  of  the  Spm  system,  these  utilized  selected  kernels  on  ears  of 

near  the  locus  of  the  Ai  (anthocyanin)  plants  of  the  constitution  aim~b  Sh2/ai  sh2 

gene  in  chromosome  3  gave  rise  to  two  (sh2,  shrunken  endosperm;  ah  standard 

modifications  designated  aim~2  and  aim~5.  recessive   allele   of   Ai)    that   had    Spm 

Extensive  examinations  have  been  made  located  close  to  aim~5,  produced   by  a 

only  of  bzm~2,  aim~2,  and  aim~5.  cross  with  plants  homozygous  for  a\  and 


DEPARTMENT    OF    GENETICS  451 

sh2  and  having  no  Spm.  Plants  were  The  second  demonstration  that  not  all 
grown  from  29  Sh2  kernels  that  exhibited  transpositions  of  Spm  away  from  the 
uniform  anthocyanin  pigmentation  in  the  locus  of  aim~h  release  the  genie  action  from 
aleurone  layer,  intense  in  some  kernels  the  control  of  the  Spm  system  was 
and  pale  in  others.  The  plants  were  tested  provided  by  an  examination  of  plants 
for  presence  or  absence  of  Spm  and,  if  it  derived  from  other  selected  Sh2  kernels 
was  present,  for  the  number  of  Spm  from  ears  produced  by  the  same  type  of 
elements  and  their  relative  locations  in  cross  as  that  producing  the  29  kernels 
the  chromosome  complement.  When  Spm  whose  constitutions  are  described  above, 
was  absent,  it  was  introduced  by  means  Each  of  these  kernels  exhibited  a  marked- 
of  a  cross  into  the  endosperm  nuclei  of  ly  altered  pattern  of  pigmented  and 
kernels  on  the  ears  produced  by  these  nonpigmented  areas,  as  compared  with 
plants,  in  order  to  test  the  expression  in  that  of  kernels  carrying  the  original  state 
its  presence  of  the  Ai  gene  in  the  Sh2-  of  ctim~b.  It  was  suspected  that  each  of 
carrying  chromosome.  The  tests  indicated  these  kernels  had  received  an  a\m~b  locus 
that,  in  28  of  the  29  plants,  action  of  the  whose  state  had  been  altered  in  a  cell  of 
gene  A\  was  no  longer  under  the  control  the  aiw-5-carrying  parent  plant.  To  test 
of  the  Spm  system;  the  same  level  of  this  assumption,  plants  derived  from  10 
genie  expression  appeared  both  in  the  such  kernels,  each  selected  from  a  differ- 
presence  and  in  the  absence  of  Spm.  In  ent  ear,  were  examined,  and  extensive 
some  of  these  28  plants,  no  Spm  was  tests  were  subsequently  conducted  with 
present.  In  others,  one  or  more  Spm  the  progeny  of  4  of  them.  All  10  plants 
elements  were  present  but  were  not  carried  a  modified  state  of  aim~b,  and  also 
located  close  to  the  A  i  gene.  In  2  plants,  an  Spm  element.  In  3  of  the  plants,  the 
however,  Spm  was  found  to  be  located  single  Spm  had  remained  in  intimate 
very  close  to  the  gene,  even  though  genie  association  with  the  a\m~b  locus;  that  is, 
action  had  been  released  from  the  control  the  event  responsible  for  the  alteration  of 
of  the  Spm  system.  state  had  not  resulted  in  its  removal  to  a 
The  remaining  plant  of  the  29  was  more  distant  location.  In  the  other  7 
derived  from  a  pale-pigmented  kernel,  plants,  however,  Spm  was  located  else- 
Tests  for  the  presence  of  Spm  were  where  in  the  chromosome  complement, 
negative.  Nevertheless,  the  action  of  the  and  in  6  of  them  it  was  not  linked  with 
A  i  gene  in  the  $/i2-carrying  chromosome  aim~b;  in  the  seventh,  it  was  located 
remained  under  the  control  of  the  Spm  approximately  30  crossover  units  from 
system,  as  was  shown  when  Spm  was  a\m~b.  Thus  a  typical  two-element  system 
introduced  by  a  cross  of  this  plant  with  of  control  of  gene  action  was  operating  in 
one  carrying  Spm.  The  response  of  the  each  of  these  7  plants,  and  Spm  was  its 
gene  to  Spm  was  similar  to  that  given  by  regulator. 

the  class  II  states  of  a2m~i  described  in         The  illustrations  given  above  show  that 

Year  Book  57.  With  this  state  of  aim~5,  a  with  a\m~b  the  three  anticipated  conse- 

medium  level  of  A  i  gene  action  is  ex-  quences  of  different  types  of  transposition 

pressed  in  plants  and  kernels  that  have  of  elements  of  the  control  system  were 

no  active  Spm  in  their  nuclei,  but  gene  observed:  (1)  Release  of  A\  gene  action 

action  is  suppressed  if  an  active  Spm  is  from  control  by  the  Spm  system,  associ- 

present.    Here,    then,    the   Spm   system  ated    with    removal   of   Spm   from   the 

continued  to  control  the  action  of  the  A  i  immediate    vicinity    of    the    gene.     (2) 

gene  although  Spm  no  longer  occupied  a  Release  of  such  control,  not  accompanied 

position  close  to  it.  A  typical  two-element  by  transposition  of  Spm.  (3)  Continued 

system   of   control   of   gene   action  had  control  of  Ai  gene  action  by  the  Spm 

evolved  from  an  apparently  one-element  system,  after  removal  of  the  Spm  element 

system.  from  the  immediate  vicinity  of  aim~b. 


452  CARNEGIE     INSTITUTION      OF      WASHINGTON 

,      7     .     ,         9  mutants  and  the  second  as   "mottled" 

Analysts  of  a\m~2  ,      , 

u        J  mutants. 

The  analysis  of  aim~5,  just  described,  Control  of  gene  action  at  a\m~2  by  the 

proceeded  rapidly  as  soon  as  plants  had  Spm  system  is  quite  different  from  the 

been  isolated  that  carried  a  single  Spm  control  exercised  by  that  system  when 

element  located  close  to  the  A  i  gene.  The  the  Spm  element  is  not  located  near  the 

types  of  gene  action  produced  by  the  controlled    gene.    For   example,    in   the 

stable   mutations   were   not   difficult  to  modified   loci   aim~l   and   a2m~1,   in   the 

interpret;     they    appeared     to     express  above-described  derivatives  of  a,im~5,  and 

different    levels    of    standard    Ai    gene  in  wxm~s,  gene  action  is  suppressed  by  an 

action.    The   behavior   of   the   modified  active  Spm  element  but  is  expressed  in 

states    was    also    readily    inter  pre  table,  its  absence  or  when  it  is  present  but 

Those  that  were  associated  with  the  two-  inactive.    With    aim~2    (original    state), 

element  system  of  control  of  gene  action  however,  the  reverse  is  true:  when  Spm 

whose  origins  are  described  above  fol-  is  inactive,   the   action   of  the   gene   is 

lowed  the  same  rules  that  had  previously  suppressed;  when  it  is  active,  gene  action 

been  established  for  the  Spm  system.  is  expressed.  This  fact  could  be  deter- 

Analysis  of  aim~2,  on  the  other  hand,  mined  because  it  was  possible  to  select 

has    been    complicated.    Although    the  some  ai™~2-carrying  plants  in  which  Spm 

original  state  is  similar  to  the  original  was  in  an  inactive  phase  of  long  duration, 

state  of  aim~5,  in  that  Spm  resides  close  and  others  in  which  Spm  was  in  an  active 

to  the   A  i  locus  and  the  Spm  system  phase    of    long    duration.    Some    tests 

controls   A  i  gene   action,   the   types  of  conducted  with  plants  having  Spm  in  an 

expression   resulting   when   the   gene   is  active  phase  of  long  duration  will  be 

released  from  the  control  of  the  Spm  considered  first. 

system  are  distinctly  different.  There  are  Location  of  Spm  before  and  after  release 

two  classes  of  mutants.  The  first  has  a  of  control  of  gene  action  at  a im~2  by  the  Spm 

phenotype  resembling  that  produced  by  system.  With  the  original  state  of  aim~2, 

the  standard  Ai  gene.  The  other  class  is  the  location  of  Spm  close  to  the  Ai  gene 

composed  of  a  series  of  alleles,  distin-  was  established  by  several  types  of  test, 

guished    from    the    first    class    by    the  commencing  with  a  cross  of  plants  of  the 

distribution  and  intensity  of  pigment  in  constitution  aim~2  Sh2/ai  sh2  by  plants 

plant    and    kernel.    In    the    kernel,    the  that  had  no  Spm  and  were  homozygous 

intensity  of  pigmentation  in  the  aleurone  for  a  specially  selected  state  of  aim~l  and 

layer   is   not   uniform,    so   that   kernels  also   for   sh2.    Ai  gene   action   in   these 

appear  somewhat  mottled.  The  different  last-named  plants  was  under  the  control 

alleles  in  this  class  may  be  distinguished  of  the  Spm  system,  involving  an  operator 

from    one    another    by    the    degree    of  element  located  close  to  the  Ai  gene  and 

intensity  of  kernel  pigmentation,  which  an  Spm  element  located  elsewhere  in  the 

ranges  from  very  faint  to  fairly  dark.  The  chromosome  complement.  With  this  se- 

plants  also  are  pigmented,  but  the  color  lected    state    of    aim~l,    gene    action    is 

develops  slowly  and  is  markedly  affected  expressed  in  the  absence  of  Spm  (or  when 

by  sunlight:  the  parts  of  a  plant  exposed  it    is    present    in    an    inactive    phase), 

to     direct     sunlight     become     intensely  Anthocyanin   pigment   appears   in   both 

pigmented,  whereas  parts  not  so  exposed  plant  and  kernel.  In  the  kernel,  pigment 

remain   light   in   color.    Although    some  of  medium  intensity  is  uniformly  distrib- 

pigment  develops  in  the  mid-rib  of  the  uted  over  the  aleurone  layer.  W^hen  an 

leaf  and  at  its  edge,  very  little  or  none  active  Spm  is  present  somewhere  in  the 

develops  in  the  leaf  blade.  The  two  classes  chromosome  complement,  gene  action  is 

of  mutants  are  thus  readily  distinguished,  suppressed  until  there  occurs,  in  some 

The  first  will   be  referred  to  as   "ii"  cells,  a  response  of  the  operator  to  Spm 


DEPARTMENT    OF    GENETICS  453 

that  effects  a  release  of  gene  action  from  a  direct  relationship  between  the  per- 

the  control  of  the  Spm  system.   These  centage  of  kernels  in  the  Sh2  class  that 

releases  occur  in  a  relatively  few  cells  late  received  a  germinal  mutant  of  a\m~2  and 

in  the  development  of  plant  and  kernel,  the  percentage  of  kernels  in  the  sh2  class 

and  most  of  them  lead  to  an  expression  that  received  Spm.  On  ears  in  which  all 

of  A\  gene  action  resembling  that  of  the  the  Sh2  kernels  had  received  unmodified 

standard   A\  gene.   Consequently,   they  a\m~2  there  were  no  kernels  in  the  sh2 

give  rise  to  a  distinctive  pattern  of  deeply  class  that  carried  Spm.  Among  the  ears 

pigmented  dots  in  the  kernel  and  small  bearing  kernels  that  expressed  germinal 

pigmented   streaks   in   the   plant,    both  mutations  of  aim~2  the  percentage  of  such 

appearing    on    a    nonpigmented    back-  kernels  and  the  percentage  of  sh2  kernels 

ground.  In  plants  and  kernels  carrying  with  Spm  were  directly  related.    (This 

the  original  state  of  aim~2,  on  the  other  correlation    was    exhibited    among    the 

hand,  release  of  control  of  gene  action  by  kernels    on   ears   having   no   detectable 

the  Spm  system  may  occur  in  many  cells,  sectors   derived   from   cells   in   which   a 

both  early  and  late  in  development.  As  is  stable  mutation  of  a\m~2  had  occurred 

described  above,  such  release  may  result  early   in  development.    Ears   with  such 

either  in  a  high  level  of  Ai  gene  action,  a  sectors  are  not  included  in  the  table.)  The 

lower  level  that  produces  the  "mottled"  correlation    suggested    that    Spm    was 

phenotype,  or,  rarely,  a  null  expression  located  very  close  to  a\m~2  in  the  hetero- 

of  the  gene.  Thus,  kernels  carrying  the  zygous  parents  and  that  its  removal  from 

original  state  of  a\m~2  and  a  fully  active  this   location   was   associated   with   the 

Spm  exhibit  both  large  and  small  pig-  origin  of  many  of  the  stable  mutations, 

mented  areas  of  various  intensities.  This  possibility  was  also  suggested  by 

In  the  above-described  cross,  nearly  all  the  phenotypes  of  kernels  on  ears  pro- 

the  Sh2  kernels  on  an  ear  receive  from  the  duced  by  test  crosses  conducted  with  other 

heterozygous   parent   either   unmodified  plants  having  the  constitution  a\m~2  Sh2/ 

a\m~2  or  a  modified  derivative  of  it,  and  a-F~x  sh2.  Ears  of  33  plants  of  this  consti- 

nearly   all   the   sh2   kernels   receive   the  tution  were  utilized  in  crosses  with  plants 

standard  «i  allele.  This  happens  because  homozygous  for  a\m~l  and  sh2  and  having 

crossing  over  between  the  locus  of  a\m~2  no  Spm,  and  also  with  plants  homozygous 

and  that  of  Sh2  is  very  infrequent,  not  for  ai  and  sh2  and  having  no  Spm.  Table  3 

exceeding  0.12  per  cent.  The  presence  or  shows   the   phenotypes   of  kernels   that 

absence  of  active  Spm  can  be  detected  appeared  on  ears  produced  by  the  second 

readily  in  the  sh2  kernels  on  the  ears  cross.  Again,  a  direct  relationship  will  be 

produced  by  the  cross,  and  also  in  Sh2  noted  between  the  percentage  of  kernels 

kernels    that    have    received    a    gamete  that  received  a  germinal  mutant  of  a\m~2 

carrying    a    stable    mottled    mutant    of  and  the  percentage  of  kernels  in  the  sh2 

aim~2.  If  an  active  Spm  is  present  in  one  class  that  received  Spm. 

such  kernel,  the  distinctive  pattern  of  The  cross  that  produced  the  kernels 

deeply  pigmented  dots  produced  by  the  entered  in  table  2  was  conducted  after  the 

response  of  aim~l  to  Spm  appears  in  a  above-described    correlation    had    been 

mottled  background.  If  Spm  is  absent,  recognized.  The  ear-bearing  parent  plants 

these  dots  are  absent  and  the  kernels  in  cultures  7979 A  and   B,   7980 A,   and 

exhibit    only    the    mottled    phenotype.  7981 A  were  derived  from  variegated,  Sh2 

Table  2  lists  the  phenotypes  of  kernels  kernels    on    ears    of    plants    7799B-1, 

that  appeared  on  some  ears  produced  by  7799B-6,  and  7800 A-5  of  table  3.  Plants 

the  cross.  The  ratios  of  kernel  types  were  were  grown  from  kernels  in  the  underlined 

not  the  same  on  all  these  ears.  Neverthe-  classes  in  table  3  in  order  to  test  the 

less,    except    on    ears    of    plants    whose  conclusion    that    Spm    resides    close    to 

numbers  are  printed  in  italics,  there  was  unmodified  a\m~2  and  that  many  of  the 


454 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


TABLE  2.     Phenotypes  of  Kernels  on  Ears  of  Plants  of  the  Constitution  axm~2  Sh2/ai  sh2 
Produced  by  a  Cross  with  Plants  That  Were  Homozygous  for  diTO_1  and  sh2  and  Had  No  Spm 

Phenotypes  of  Kernels 


Plant 
Number 


Sh2  Class* 


sh2  Classf 


Germinal  Mutations 


\ 


Ai 


Mottled 

No  A  i  Dots   A  i  Dots 

(No  Spm)       (Spm) 


Variegated  p  Dots  ofii  p 

for  A  i  and  ,~        .     ,         Pale  in  Colorless  '  .,, 

Mottled  ^/T!         (No  5pm)  Background  af  wlth 

Spots  Mutations  {Spm)  Spm 


7979A-7 

0 

0 

0 

57 

0 

73 

0 

0 

A-8 

0 

0 

0 

233 

0 

189 

0 

0 

B-l 

0 

0 

0 

239 

0 

219 

0 

0 

A-6 

0 

3 

1 

69 

5.4 

59 

1 

1.6 

A-3 

1 

4 

1 

71 

7.7 

64 

2 

3.0 

A-12 

3 

13 

11 

98 

21.6 

114 

12 

9.5 

A-l 

4 

25 

15 

105 

29.5 

118 

20 

14.5 

B-4 

4 

36 

13 

109 

32.7 

148 

29 

16.3 

A-2 

3 

33 

25 

113 

35.0 

148 

25 

14.4 

A-13 

3 

35 

22 

106 

36.1 

142 

17 

10.6 

A-10 

9 

46 

47 

169 

37.6 

232 

51 

18.0 

A-ll 

7 

61 

57 

115 

52.0 

171 

53 

23.6 

A-9 

10 

26 

62 

134 

42.2 

97 

157 

61.8 

B-3 

25 

44 

82 

83 

64.5 

96 

137 

58.8 

7980A-9 

11 

32 

22 

149 

30.3 

177 

31 

14.9 

A-7 

1 

5 

5 

24 

31.4 

22 

6 

21.4 

A-3 

5 

20 

6 

53 

36.9 

76 

13 

14.6 

A-4 

6 

38 

21 

110 

37.1 

163 

30 

15.5 

A-l 

6 

62 

41 

169 

39.2 

263 

64 

19.5 

A-2 

10 

55 

43 

110 

49.5 

190 

49 

20.5 

A-8 

12 

33 

80 

111 

52.9 

121 

129 

51.6 

7981A-1 

3 

49 

24 

151 

33.4 

183 

22 

10.7 

A-4 

5 

51 

33 

114 

43.8 

108 

41 

27.5 

A-5 

10 

64 

33 

136 

44.0 

199 

37 

16.1 

A-7 

6 

41 

66 

68 

62.4 

157 

53 

25.2 

A-8 

9 

63 

66 

59 

70.0 

144 

59 

29.0 

*  In  addition  there  was  one  pale,  Sh2  kernel. 

f  In  addition  there  were  three  sh2  kernels  that  received  aim~2. 


germinal  mutations  arise  when  Spm  is 
transposed  to  a  new  location  in  the 
chromosome  complement.  Among  the  26 
plants  listed  in  table  2,  23  had  one  Spm 
element  in  the  cells  that  gave  rise  to  the 
testcross  ear  and  3  (whose  numbers 
appear  in  italics)  had  two  Spm  elements. 
There  were  4  additional  plants,  each  also 
derived  from  an  Sh2  kernel  whose  endo- 
sperm was  variegated.  A  mottled  pheno- 
type  was  expressed  in  these  4  plants 
rather  than  the  phenotype  produced  in 
plants  that  commence  development  with 
unmodified    aim~2.    The    presence    of    a 


mottled  mutant  in  these  plants  was 
confirmed  by  the  kernel  types  that 
appeared  on  a  testcross  ear  of  each  (rows 
1-4,  table  4).  Since  the  endosperm  of  the 
kernel  from  which  each  of  these  plants 
arose  started  development  with  unaltered 
dim~2,  the  event  that  produced  the 
mottled  mutant  must  have  occurred 
during  development  of  the  female  game- 
tophyte  in  the  parent  plant,  or  in  the 
kernel  early  in  development  of  the 
embryo.  All  4  plants  had  one  or  more 
active  Spm  elements.  In  2  of  them 
(7981  A-3  and  7981  A-6),  one  Spm,  not 


DEPARTMENT   OF   GENETICS 


455 


TABLE  3.     Phenotypes  of  Kernels  on  Ears  of  Plants  of  the  Constitution  a,im~2  Sh2/a\m~l  sh2 
That  Had  One  Active  Spm,  Produced  by  a  Cross  with  Plants  Homozygous  for  ai  and  sh2  and 

Having  No  Spm 


Phenotypes  of  Kernels 

Plant 
Number 

Sht 

Class 

sh 

2  Class 

Germinal  Mutations 

Variegated 

for  Ai  and 

Mottled 

Spots 

Percentage 
Germinal 
Mutations 

At 

Pale 
(No  Spm) 

Dots  of  A  i 
in  Colorless 
Background 

(Spm) 

Percent- 
age with 
Spm 

Ax 

Mottled 

7799B-1 

0 

0 

201 

0 

0 

214 

1 

0.46 

7799B-6* 

2 

50 

176 

22.8 

1 

216 

28 

11.4 

7800A-5 

6 

60 

153 

30.1 

1 

180 

21 

10.4 

7984   -7 

6 

64 

131 

34.8 

0 

172 

31 

15.2 

7984   -4 

2 

60 

109 

36.2 

0 

150 

27 

15.2 

7984   -3 

4 

62 

99 

40.0 

1 

131 

30 

18.5 

7799A 

14 

95 

119 

47.8 

2 

160 

44 

21.3 

In  addition  there  was  one  colorless,  sh2  kernel  on  this  ear.     The  plant  derived  from  it  had  no  Spm. 


linked  to  the  mutant  locus,  was  present 
in  the  cells  that  produced  the  testcross 
ear.  In  the  other  2,  two  Spm  were  present 
in  the  cells  giving  rise  to  the  testcross 
ear — neither  element  linked  with  the 
mutant  locus  in  plant  7980 A-6,  but  one 
linked  with  it  in  plant  7981A-2. 

The  plant  grown  from  the  single  sh2 
kernel  containing  Spm  on  the  ear  of  plant 


7799B-1  (row  1,  table  3)  proved  to  be 
a\m~l  shz/ai  sh2  in  constitution  and  had 
two  independently  located  Spm  elements 
in  the  cells  that  produced  each  of  its 
tested  ears. 

Nine  plants  derived  from  the  mottled 
Sfi2  class  of  kernels  on  the  ear  of  plant 
7799B-6  (row  2,  table  3)  were  also  tested 
for  Spm  constitution.  No  evidence  of  its 


TABLE  4.     Phenotypes  of  Kernels  on  Ears  of  Plants  That  Were  Mottled-Mutant  Sh2/a\  sh2  in 
Constitution,  Produced  by  a  Cross  with  Plants  Homozygous  for  aim~l  and  sh2  and  Having  No  Spm 


Phenotypes  of  Kernels 

Plant 

Mottled  Shi 

Class 

sh2  Class 

Number 

No  Dots  of 

Deep 

Dots  of  Deep 

Deep-Pigmented 

Pigmentation 

Pigmentation 

Pale 

Dots 

in  Colorless 

(No  Spm) 

(Spm) 

(No  Spm) 

Background  (Spm) 

7980A-6 

86 

162 

73 

149 

7981A-3 

120 

107 

120 

111 

7981A-6 

82 

81 

74 

82 

7981A-2 

92 

125 

111 

55 

7980B-3 

91 

85 

103 

80 

7980C-2 

60 

141 

166 

45 

7980B-4 

30 

211 

95 

127 

7981B-1 

20 

219 

254 

22 

7981B-6 

0 

274 

255 

1 

7981B-8 

80 

126 

149 

71 

7981C-3 

47 

175 

51 

180 

456  CARNEGIE     INSTITUTION     OF      WASHINGTON 

presence  was  shown  by  the  kernels  on  the  Spm.  The  agreement  in  distribution  of 
ears  of  6  of  these  9  plants;  but  it  was  Spm  to  the  mutant  class,  demonstrated 
present  in  the  cells  that  gave  rise  to  the  by  these  two  types  of  test,  is  good, 
testcross  ear  in  the  remaining  3  plants  Further  confirmation  that  Spm  was 
(7980B  and  C,  table  4).  Plant  7980B-3  located  close  to  a^~2,  and  that  its 
had  one  Spm,  not  linked  with  Sh2;  plant  removal  from  that  location  was  related  to 
7980C-2  had  one  Spm,  linked  with  Sh2;  the  origin  of  the  stable  mutants,  was 
and  plant  7980B-4  had  two  Spm,  one  provided  by  tests  of  the  progeny  of  an 
linked  with  Sh2.  Testcrosses  conducted  aim~2  Sh2/ai  sh2  plant  produced  by  cross- 
with  8  of  the  12  plants  derived  from  ing  this  plant  with  one  that  was  homozy- 
mottled  Sh2  kernels  on  the  ear  of  plant  gous  for  aim~l  and  sh2  and  had  no  Spm. 
7800 A-5  (row  3,  table  3)  produced  no  A  very  large  sector,  present  in  the  a\m~2- 
evidence  of  the  presence  of  Spm.  It  was  carrying  plant,  was  derived  from  a  cell  in 
present,  however,  in  the  remaining  4  which  a  mutation  to  a  stable  mottled 
plants,  as  indicated  in  table  4.  Very  close  allele  had  occurred.  On  the  ear  of  the 
linkage  of  Spm  with  the  locus  of  the  described  testcross  there  were  306  mot- 
mottled  mutant  was  exhibited  by  plant  tied,  Sh2  kernels,  of  which  150  carried 
7981B-6.  Linkage  of  Spm  with  the  locus  Spm  and  156  had  no  Spm.  Only  18  Sh2 
of  the  mutant  was  expressed  in  plants  kernels  on  this  ear  had  received  unmodi- 
B-l  and  B-8.  The  ratio  of  kernel  types  fied  aim~2.  Among  the  315  sh2  kernels  on 
on  the  ear  of  plant  7981C-3  (355  with  this  ear,  177  were  uniformly  pigmented 
Spm/98  with  no  Spm)  suggests  the  (no  Spm)  and  138  had  dots  of  deep 
presence  of  at  least  two  Spm  elements,  pigmentation  in  a  colorless  background 
not  linked  with  Sh 2,  in  the  cells  that  gave  (Spm  present).  Ten  plants  grown  from 
rise  to  this  ear.  mottled  Sh2  kernels  carrying  Spm,  and  10 
All  together,  42  plants  derived  from  plants  from  Sh2  kernels  that  had  received 
mottled  Sh2  kernels  on  ears  of  aim~2  unmodified  a\m~2,  were  tested  for  Spm 
plants  that  had  one  Spm,  located  close  to  number  and  location.  On  testcross  ears 
aim~2,  have  been  tested  for  Spm  consti-  produced  by  the  10  plants  derived  from 
tution  and  location.  Twenty-six  of  the  the  mottled  kernels  the  ratio  of  kernel 
plants  showed  no  evidence  of  the  presence  types  indicated  that  9  of  them  had  one 
of  Spm.  Twelve  plants  had  one  Spm:  in  Spm,  not  linked  with  Sh2,  and  that  two 
2  of  them  it  was  situated  very  close  to  Spm  elements,  not  linked  with  Sh2,  were 
the  locus  of  the  mutant ;  in  2  others  it  was  present  in  the  cells  that  produced  the  ear 
linked  with  the  mutant  locus;  and  in  the  on  the  tenth  plant.  On  testcross  ears  of 
remaining  8  there  was  no  evidence  of  such  the  10  plants  derived  from  kernels  having 
linkage.  Three  plants  had  two  Spm;  unmodified  aim~2,  the  ratios  of  kernel 
neither  was  linked  with  the  locus  of  the  types  were  similar  to  those  entered  in 
mutant  in  2  of  the  plants  but  one  Spm  table  3:  one  Spm  was  present  in  each 
was  linked  with  it  in  the  third.  The  plant,  and  it  was  located  close  to  a\m~2. 
remaining  plant  of  the  42  had  three  Spm  It  may  be  concluded,  then,  that  the 
elements,  none  of  them  linked  with  the  aiTO_2-carrying  parent  of  these  20  plants 
mutant  locus.  Thus,  Spm  was  present  in  commenced  development  with  a  single 
only  16  (38  per  cent)  of  the  42  plants  Spm,  located  close  to  a\m~2.  Early  in 
derived  from  kernels  in  which  a  chromo-  development  of  that  plant,  transposition 
some  carrying  a  germinal  mutation  was  of  Spm  to  a  new  location,  occurring  in  one 
received  by  both  the  endosperm  and  the  cell,  led  to  the  origin  of  the  stable  mottled 
zygote  nuclei.  Among  the  1288  mottled  mutant  that  was  present  in  all  descend- 
Sh2  kernels  in  table  2  that  appeared  on  ants  of  the  cell. 

ears  of  plants  having  one  Spm,  552  (43  From    the    above-described    series    of 

per  cent)  carried  Spm  and  736  had  no  tests,  it  is  evident  that  the  origin  of  many 


DEPARTMENT    OF   GENETICS  457 

of  the  stable  mutants  of  a \m~2  is  associated  the  nonvariegated  phenotype  with  pr,  it 
with  transposition  of  Spm  to  a  new  is  concluded  that  gene  action  at  this 
location  in  the  chromosome  complement,  modified  a\m~2  locus  is  under  the  control 
In  some  of  the  stable  a\m~2  mutants,  on  of  the  Spm  system,  although  Spm  no 
the  other  hand,  Spm  continues  to  occupy  longer  resides  close  to  the  locus, 
a  position  close  to  the  locus  of  the  The  modified  aim~2  locus  present  in  the 
modified  A\  gene.  Thus,  in  these  respects,  plant  just  described  could  have  arisen 
a\m~2  and  aim~b  are  comparable.  from  removal  of  only  the  Spm  element 
Origin  of  a  two-element  system,  of  control  from  the  vicinity  of  the  a\m~2  locus,  the 
of  gene  action  at  a\m~2.  Although  the  tests  operator  element  remaining  in  location, 
aimed  at  identifying  events  in  which  a  That  inactivation  of  the  Spm  element 
clearly  expressed  two-element  system  of  was  responsible  for  the  modification  is 
control  of  gene  action  arises  from  a\m~2  not  probable,  as  it  is  well  established  that 
have  not  yet  been  completed,  one  example  such  inactivation  results  in  suppression  of 
may  have  been  found.  A  kernel  with  a  the  action  of  the  A\  gene  (see  below); 
distinctive  phenotype  appeared  on  an  ear  much  pigment  appeared  in  the  plant 
of  a  plant  that  was  a\m~2  Sh2/a\  sh2  in  having  this  modified  a\m~2  locus,  and 
constitution  and  had  one  Spm  located  some  pigment  appeared  in  the  aleurone 
close  to  aim~2,  after  it  was  crossed  with  a  layer  of  the  kernels, 
plant  of  similar  constitution.  The  selected  Inactive  Spm  at  the  locus  of  a\m~2. 
kernel  was  weakly  and  irregularly  pig-  Testcrosses  conducted  with  plants  in 
mented,  with  no  spots  of  deep  pigmenta-  which  Spm  was  in  an  inactive  phase, 
tion  in  its  aleurone  layer.  The  phenotype  changing  to  an  active  phase  only  in  a  few 
of  the  plant  grown  from  this  kernel  was  cells  very  late  in  development,  also 
similar  to  that  expressed  by  many  of  the  served  to  place  Spm  close  to  the  locus  of 
stable  mottled  mutants  of  a\m~2,  since  unmodified  a\m~2.  Two  types  of  testcross 
anthocyanin  pigment  appeared  in  the  were  performed.  When  plants  with  in- 
same  regions  of  the  plant.  Testcrosses  active  Spm,  a\m~2  Sh2/ai  sh2  in  constitu- 
conducted  with  this  plant  gave  no  evi-  tion,  were  crossed  with  plants  homozy- 
dence  of  the  presence  of  Spm.  Its  consti-  gous  for  a\  and  sh2  and  having  no  Spm, 
tution,  however,  proved  to  be  a\m~2  Sh2/  all  the  kernels  on  some  ears  were  colorless. 
a\  sh2.  The  kernels  on  one  of  its  ears  were  On  other  ears,  however,  a  few  kernels  in 
produced  by  a  cross  with  a  plant  homozy-  the  Sh2  class  had  a  sector  containing 
gous  for  ai  and  sh2  and  having  no  Spm.  pigment  of  light  intensity,  and  some  of 
All  the  Sh2  kernels  on  this  ear  exhibited  these  sectors,  in  turn,  also  displayed  small 
the  same  phenotype  as  that  shown  by  the  dots  of  deep  pigmentation.  Occasionally, 
kernel  that  gave  rise  to  the  plant.  The  the  entire  aleurone  layer  of  an  Sh2  kernel 
kernels  on  a  second  ear  of  the  plant  were  exhibited  such  dots  on  a  lightly  pigmented 
produced  by  a  cross  with  a  plant  that  was  background.  Also,  an  occasional  Sh2 
homozygous  for  ai  and  sh2  and  carried  kernel  was  variegated  throughout  its 
one  Spm  closely  linked  with  the  Pr  marker  aleurone  layer,  with  large  as  well  as  small 
in  chromosome  5  (Pr,  purple  aleurone;  pigmented  areas  of  various  intensities. 
pr,  recessive  allele,  red  aleurone).  Of  the  Tests  conducted  with  plants  derived  from 
217  Sh2  kernels  on  this  ear,  102  were  such  variegated  kernels  indicated  that 
variegated  for  pigmented  areas  of  differ-  the  phenotype  of  the  kernels  was  pro- 
ent  intensities,  in  a  pattern  resembling  duced  by  a  change  in  phase  of  activity  of 
that  produced  by  unmodified  a\m~2.  The  Spm,  from  inactive  to  active,  occurring 
phenotype  of  the  remaining  115  Sh2  in  a  cell  late  in  the  development  of  the 
kernels  was  similar  to  that  on  the  first  aim~2-carrying  plant, 
ear,  just  described.  From  the  close  linkage  The  second  type  of  test  utilized  the 
of  the  variegated  phenotype  with  Pr  and  same  plants  as  those  described  above,  as 


458 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


ear-bearing  parents  in  crosses  with  plants 
that  were  homozygous  for  a\m~l  and  sh2 
and  had  no  Spm.  On  some  of  the  resulting 
ears,  all  kernels  in  both  the  Sh2  and  sh2 
classes  were  uniformly  pigmented  and 
showed  no  evidence  of  the  presence  of 
Spm.  On  other  ears,  a  few  of  the  kernels 
in  the  Sh2  class  exhibited  sectors  of  much 
lower  pigment  intensity,  and  some  of 
these,  in  turn,  had  small  deeply  pigmented 
spots.  Occasionally,  the  whole  aleurone 
layer  of  an  Sh2  kernel  exhibited  this 
small-spotted  phenotype,  or  one  with 
large  as  well  as  small  areas  of  different 
grades  of  pigment  intensity.  Kernels 
having  these  last  two  phenotypes  would 
be  expected  to  appear  if,  in  the  a\m~2- 
carrying  plant,  the  inactive  Spm  under- 
went a  change  to  the  active  phase  in  some 
cells,  late  during  development  of  the 
ovule  or  in  the  female  gametophyte. 

When  plants  of  the  constitution  aim~2 
Sh2/a,i  sh2  that  carried  an  inactive  Spm 
were  crossed  with  plants  homozygous  for 
di  and  sh2  that  carried  one  or  more  active 
Spm  elements,  all  or  nearly  all  kernels 
that  received  a\m~2  from  the  heterozygous 
parent  and  no  Spm  from  the  homozygous 
parent  were  colorless.  In  contrast,  all 
those  that  received  active  Spm  from  the 
homozygous  parent  exhibited  many  mu- 
tant areas,  in  a  pattern  resembling  that 
produced  by  a\m~2  when  the  Spm  adjacent 
to  it  is  in  an  active  phase. 

Conclusions  derived  from  the  study  of 
aim~2.  Results  of  the  described  tests  with 
plants  having  unmodified  aim~2,  and  with 
others  having  modified  derivatives  of  the 
locus,  indicate  that  certainly  two  and 
probably  all  three  of  the  predicted 
consequences  of  transpositional  events, 
outlined  early  in  this  report,  have  been 
observed. 

In  this  report,  some  aspects  of  the 
analysis  of  a\m~2  have  been  considered  in 
detail,  not  only  in  order  to  develop  the 
thesis  stated  earlier  but  also  to  indicate 
the  nature  of  the  evidence  that  makes  it 
possible  to  relate  the  mode  of  control  of 
the  Spm  system  to  that  controlling  the 
alternate  action  of  the  duplicate  genes, 


Hi  and  H2,  associated  with  flagella 
antigen  formation  in  the  bacterium 
Salmonella.  In  maize  plants  that  are 
aim~2/aim~l  in  constitution,  with  an  Spm 
element  located  close  to  aim~2,  which  of 
the  alleles  will  be  active  and  which 
inactive  is  determined  by  the  phase  of 
activity  of  the  Spm  element.  In  Salmo- 
nella, which  of  the  two  duplicate  genes 
will  be  active  and  which  inactive  is  deter- 
mined by  the  phase  of  activity  of  the 
controlling  element  Vh,  located  close  to 
the  H2  gene. 

The  Derivatives  of  hzm~2 

Early  studies  of  bzm~2  were  reported  in 
Year  Books  54  and  55.  It  was  shown  that 
Ac,  the  regulator  of  the  Ac  control 
system,  resides  close  to  the  locus  of  the 
bronze  gene  in  chromosome  9,  and  also 
that  the  Ac  system  controls  the  action  of 
this  gene.  The  behavior  of  unmodified 
ozm-2  was  examined,  initially,  in  172 
plants  carrying  bzm~2  in  one  chromosome 
9  and  the  standard  stable  recessive,  bz,  in 
the  homologue,  as  well  as  in  13  plants 
homozygous  for  unmodified  bzm~2.  Sub- 
sequent studies  were  conducted  with 
plants  carrying  modified  derivatives  of 
bzm~2.  Since  the  information  obtained  is 
both  diverse  and  extensive,  the  present 
report  will  be  confined  to  summary 
statements  pertinent  to  the  topic  in  hand. 

In  a  cross  of  62m~2-carrying  plants  to 
plants  homozygous  for  standard  bz, 
kernels  that  had  received  a  modified 
derivative  of  bz m~2  appeared  on  some 
ears.  Most  of  these  kernels  exhibited 
either  a  null  level  or  a  high  level  of  gene 
action  at  the  bronze  locus.  It  was 
suspected  that  in  them  the  action  of  the 
bronze  gene,  derived  from  bzm~2,  had  been 
released  from  the  control  of  the  Ac 
system.  To  test  this  conjecture,  selections 
were  made  of  35  independently  occurring 
examples  of  change  of  bzm~2  to  an 
apparently  stable  null-expression  allele, 
and  of  14  independently  occurring 
changes  to  an  allele  expressing  a  high 
level  of  gene  action.  It  could  be  deter- 
mined  readily   that,    in   33    of   the   35 


DEPARTMENT    OF   GENETICS 


459 


selected  examples,  release  of  gene  action 
from  control  by  the  Ac  system  was 
associated  with  the  origin  of  a  stable  null 
expression  of  the  bronze  gene.  In  19  of 
these  33,  A  c  was  absent  from  chromosome 
9  in  the  original  plant  carrying  the 
modified  bzm~2  locus,  although  it  was 
present  elsewhere  in  the  chromosome 
complement  in  6  of  the  19  plants.  In  the 
rest  of  the  33  plants  (14),  Ac  was  present 
in  chromosome  9.  It  was  located  close  to 
the  bronze  gene  in  2  of  them,  and  at 
positions  away  from  the  locus  in  3  others ; 
but  its  exact  location  was  not  determined 
in  the  remaining  9  plants,  1  of  which  had 
two  Ac  elements,  one  in  chromosome  9 
and  one  elsewhere. 

The  2  remaining  kernels  of  the  35  that 
were  selected  for  a  stable,  null  expression 
of  the  bronze  gene  produced  plants  in 
which  no  Ac  was  present;  both  plants 
were  totally  bronze  in  phenotype.  When 
they  were  crossed  with  plants  carrying 
Ac,  it  was  learned  that  the  bronze  gene, 
derived  from  bzm~2,  was  under  the  control 
of  the  Ac  system.  The  manner  of  its 
response  to  Ac  was  similar  to  that 
observed  in  the  many  other  examples  of 
two-element  control  systems  in  which  Ac 
is  the  regulator  element.  In  both  these 
plants,  and  in  a  third  plant  derived  from 
a  kernel  selected  in  a  different  manner,  a 
two-element  system  of  control  of  gene 
action  had  arisen  from  bzm~2.  Although 
Ac  was  no  longer  located  close  to  the 
bronze  gene,  it  continued  to  be  the 
regulator  of  the  system  controlling  its 
action. 

The  14  original  kernels  selected  for  the 
presence  of  a  derivative  of  bzm~2  that 
expressed  a  high  level  of  gene  action  gave 
rise  to  8  plants  having  Ac  and  6  having 
no  Ac  in  their  nuclei.  Seven  of  the  8 
Ac-carrying  plants  had  one  Ac  element. 
In  4  plants  it  was  not  linked  with  markers 
in  the  short  arm  of  chromosome  9 ;  in  the 
other  3  it  was  linked  with  such  markers, 
being  located  close  to  the  i?z- expressing 
gene  in  2  of  them  and  proximal  to  the 
locus  of  Wx  in  the  third.  The  eighth 
Ac-carrying  plant  had  two  Ac,  one  located 


close  to  the  Bz  gene  and  one  not  linked 
with  markers  in  the  short  arm  of  chromo- 
some 9.  Tests  of  all  14  plants  and  their 
progenies  indicated  that  the  presence  of 
an  Ac  element  in  the  nucleus  did  not 
effect  a  modification  in  action  of  the 
^-expressing  gene  derived  from  bz m~2. 
Action  of  this  gene  appeared  to  have  been 
released  from  the  control  of  the  Ac 
system. 

A  modified  state  of  bzm~2  was  recog- 
nized early  in  the  study  of  that  locus. 
The  alteration  at  the  bronze  locus  that 
produced  this  state  did  not  remove  Ac, 
which  remained  close  to  the  locus  of  the 
gene,  and  the  Ac  system  continued  to 
control  gene  action.  In  contrast  to  the 
original  state  of  bzm~2,  this  altered  state 
is  characterized  by  a  high  level  of  bronze 
gene  action.  Some  of  its  Ac- controlled 
modifications  result  in  recognizable 
changes  in  level  of  gene  action.  Others 
result  in  release  of  the  gene  from  control 
by  the  Ac  system,  and  such  release  is 
often  associated  with  maintenance  of  a 
high  level  of  gene  action.  Still  other 
modifications  give  rise  to  further  altered 
states.  One  of  these  resembles  the  initial 
state  of  bzm~2,  and  another  has  proved  to 
be  instructive  for  the  thesis  of  this  report, 
for  it  allows  ready  selection  of  kernels 
produced  from  cells  in  which  Ac  no 
longer  occupies  a  position  close  to  the 
bronze  gene.  This  state  was  recognized, 
initially,  in  a  single  kernel  on  an  ear.  The 
aleurone  layer  of  the  kernel  exhibited 
many  deeply  pigmented  spots  in  a  lightly 
pigmented  background. 

Tests  of  the  plant  derived  from  this 
kernel,  and  of  its  progeny,  showed  that 
Ac  occupied  a  position  close  to  the  locus 
of  the  modified  bronze  gene  and  that  the 
observed  changes  in  action  of  the  gene 
were  expressions  of  control  by  the  Ac 
system.  On  ears  produced  by  a  cross  of 
plants  carrying  this  state  with  plants  that 
were  homozygous  for  the  standard  reces- 
sive bz  and  had  no  Ac,  kernels  that 
received  the  state  exhibited  deeply  pig- 
mented areas  in  a  lightly  pigmented 
background.  A  few  kernels  on  some  ears, 


460 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


however,  showed  only  the  light  back- 
ground pigmentation,  with  no  deep- 
colored  spots.  Plants  derived  from  5  such 
kernels  were  examined.  Tests  were  made 
for  the  presence  or  absence  of  Ac  in  them, 
and  for  the  expression  of  the  weak  bronze 
allele  in  their  progeny  produced  by  a 
cross  with  plants  that  were  homozygous 
for  standard  bz  and  had  either  no  Ac  or 
one  or  more  Ac.  It  was  learned  that  no 
Ac  was  present  in  these  plants.  In  the 
absence  of  Ac,  the  expression  of  the 
bronze  gene  is  constant,  and  it  behaves 
as  a  weak  allele  of  Bz.  When  Ac  is  present, 
however,  deeply  pigmented  spots  appear 
in  a  lightly  pigmented  background.  It 
could  be  determined  readily  that  these 
spots  arise  through  the  response  of  an 
operator  element  at  the  locus  of  the  weak 
Bz  allele  to  the  presence  of  Ac.  Thus,  a 
two-element  system  of  control  of  gene 
action  was  expressed  in  each  of  these  5 
selected  examples,  and  Ac  was  the 
regulator  of  the  system. 

The  above-described  sequence  of  events 
affecting  bzm~2  may  be  interpreted  in  the 


following  manner.  Insertion  of  the  oper- 
ator and  regulator  elements  of  the  Ac 
system  close  to  the  locus  of  the  standard 
Bz  gene  gave  rise  to  bzm~2,  which  exhibits 
a  null  base  level  of  gene  action.  Thus,  the 
original  bzm~2  locus  may  be  symbolized  as 
bz(op)Ac.  Since  only  the  Ac  element  at 
the  locus  of  the  gene  has  been  determined, 
the  symbol  for  the  operator  element  (op) 
is  shown  in  parentheses.  Removal  of  both 
the  operator  and  the  regulator,  Ac,  from 
the  locus  of  the  gene,  or  removal  of  the 
operator  alone,  releases  the  gene  from 
control  by  the  Ac  system.  Removal  of  Ac 
alone  allows  the  presence  of  the  operator 
to  be  recognized,  and  the  locus  can  be 
given  the  symbol  bz-op.  A  change  at 
bzm~2,  assumed  to  be  induced  by  the 
operator  in  response  to  the  regulator  Ac, 
produces  a  new  state  characterized  by  a 
high  level  of  gene  action,  which  is 
symbolized  Bzs(op)Ac.  A  subsequent 
modification,  again  assumed  to  be  induced 
by  the  operator,  gives  rise  to  another 
state  characterized  by  an  intermediate 
level  of  gene  action,  which  is  symbolized 


TABLE  5.     Response  of  the  Bronze  Gene  to  Ac  in  Selected  Derivatives  of  bz  (op)  Ac,  the  Original 
State  of  bzm~2  (Part  I),  and  in  Two  of  Its  Modified  States,  Bz'(op)Ac  (Part  II)  and  Bzw(op)Ac 

(Part  III) 


Expression  of  Bronze  Gene 
in  Selected  Kernel 

Response  of  Bronze 
Gene  to  Ac 

Symbol  for 
Bronze  Gene 

No.  Cases 
Examined 

Parti 

Null  expression;  stable 

Negative 
Negative 
Negative 

bz 

bz-Ac 

bz;  Ac  in  chromosome 

22 

2 

9;  position  not 
determined 

9 

Positive 

bz-op 

3 

High  level  of  gene  action;  stable 
High  level  of  gene  action;  unstable 

Negative 
Negative 
Positive 
Part  II 

Bz" 
Bza-Ac 

Bz"(op)Ac 

11 
3 
1 

High  level  of  gene  action;  stable 

Return  to  bzm~2  expression 

Weak  expression  of  gene;  unstable 

Negative 
Negative 
Positive 
Positive 
Part  III 

Bz' 
Bz'-Ac 

bz(op)Ac 
Bzw(op)Ac 

5 

7 
10 

1 

High  level  of  gene  action;  stable 
Weak  expression  of  gene;  stable 
Return  to  high  level  of  gene  action; 
unstable 

Negative 
Positive 

Positive 

Bz' 
Bzw-op 

Bz'(op)Ac 

3 
5 

1 

DEPARTMENT    OF    GENETICS 


461 


Bzw(op)Ac.  Removal  of  only  Ac  allows 
the  presence  of  the  operator  to  be  recog- 
nized, and  in  this  event  the  gene  locus  is 
given  the  symbol  Bzw-op. 

Table  5  summarizes  the  evidence  for 
the  considerations  outlined  in  this  section. 
The  symbols  in  the  table  are  the  same  as 
those  just  described.  A  symbol  that  does 
not  include  op  or  Ac  indicates  that  no 
evidence  has  been  obtained  of  the 
presence  of  either  element  at  the  locus  of 
the  gene. 

The  findings  presented  in  this  report 
are    sufficiently    extensive    to    leave    no 


doubt  that  a  two-element  system  of 
control  of  gene  action,  composed  of  an 
operator  element  at  the  locus  of  the  gene 
and  a  regulator  element  located  else- 
where, may  arise  at  a  gene  locus  that 
initially  carried  the  regulator  of  the 
system.  Although  in  the  examples  studied 
the  origin  of  the  operator  element  has  not 
been  determined  directly,  it  is  neverthe- 
less evident  that  the  predicted  conse- 
quences of  removal  of  either  or  both 
elements  from  the  locus  of  the  gene,  on 
the  assumption  that  both  were  present 
initially,  have  been  confirmed. 


ENZYMOLOGY 

Margaret  R.  McDonald  and  Anne  K.  Carhart 


Since  1945  we  have  intermittently 
attacked,  in  a  joint  endeavor  with  Drs. 
Berwind  P.  Kaufmann  and  Helen  Gay, 
the  problem  of  the  "submicroscopic" 
organization  of  the  chromosome  by  means 
of  "enzymatic  dissection."  It  was  ap- 
parent from  the  beginning  of  this  program 
(Year  Book  46)  that  to  obtain  decisive 
results  we  must  work  with  enzymes  that 
were  pure,  were  specific  in  their  action, 
and  had  access  to  their  substrates;  and 
much  effort  has  been  expended  during  the 
intervening  years  in  the  fulfillment  of 
these  requirements.  In  addition,  the 
biological  materials  to  be  tested  must 
have  the  enzyme's  substrate  in  a  proper 
condition  to  be  acted  on,  and  must  be 
free  of  endogenous  enzymes  capable  of 
acting  before  or  after  the  exogenous  one 
and  thus  confusing  the  results. 

An  excellent  example  of  such  confusion 
was  noted  in  1950,  when  it  was  observed 
that  RNase,  although  freed  of  all  meas- 
urable traces  of  DNase,  appeared  under 
certain  conditions  to  degrade  DNA  in 
fixed  tissues,  as  evidenced  by  reduction 
in  Feulgen  stainability.  This  anomaly 
was  eventually  traced  to  the  presence  in 
the  tissues  of  a  DNase  capable  of  rapidly 
depolymerizing  isolated  DNA  but  unable 
to  depolymerize  intracellular  DNA  unless 
the  cells  had  been  treated  previously  with 


RNase  (Year  Book  51).  Several  other 
unexpected  results  had  presented  them- 
selves throughout  the  years,  which, 
although  investigated  intensively  as  they 
were  noted,  had  remained  mysteries. 

Cognizant  of  Dr.  Kaufmann's  inevi- 
table retirement  in  July  1962,  and  of  the 
contemplated  changes  in  the  organization 
of  genetics  research,  we  decided,  in  the 
time  left  at  our  joint  disposal,  to  reexamine 
these  unexplained  phenomena  by  means 
of  new  tools  acquired  last  year.  The 
results  of  the  investigations  are  described 
below,  together  with  a  few  new  findings 
in  our  studies  of  DNase  II,  which  have 
been  pursued  simultaneously.  We  are 
indebted  to  Dr.  Helen  Gay  and  Miss  Ann 
Weingart  for  all  the  cytochemical  analy- 
ses, details  of  which  can  be  found  in  the 
report  by  Kaufmann  et  al.  in  this  volume. 

Ribonuclease 

The  location  of  DNA  in  fixed  tissues 
can  be  determined  cytochemically  by 
means  of  two  color  reactions:  that  pro- 
duced by  the  Feulgen  test,  which  is 
regarded  as  specific  for  deoxyribose;  and 
that  produced  by  purified  methyl  green, 
which  colors  highly  polymerized,  but  not 
depolymerized,  DNA.  Treatment  of  tissue 
sections  with  either  DNase  I  or  DNase  II 


462 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


prevents  both  Feulgen  and  methyl  green 
staining  of  the  chromosomes,  as  would  be 
anticipated  from  the  known  specificities 
of  these  enzymes.  An  unexpected  result 
was  observed,  however,  when  fixed  tissue 
sections  were  treated  with  high  concen- 
trations of  crystalline  RNase  free  of  all 
measurable  traces  of  DNase.  Such  treat- 
ment rendered  the  chromosomes  non- 
stainable  with  methyl  green  without 
reducing  the  intensity  of  the  Feulgen 
reaction  (Year  Book  49).  This  failure  to 
stain  with  methyl  green  did  not  appear 
to  be  due  to  combination  of  RNase  as  a 
basic  protein  with  the  DNA  of  the 
chromosomes  (although  such  combination 
undoubtedly  occurs),  since  other  basic 
proteins,  such  as  lysozyme,  cytochrome  c, 
and  chymotrypsinogen,  tested  under 
identical  conditions,  did  not  alter  methyl 
green  colorability  (Year  Books  54,  60). 
Furthermore,  the  reduction  in  methyl 
green  stainability  of  fixed  sections  treated 
with  RNase  was  dependent  on  time, 
temperature,  and  concentration  (Year 
Book  60)  and  thus  definitely  suggestive  of 
an  enzymatic  reaction. 

These  results  were  explainable  on  the 
supposition  that  chromosomes  contain  a 
complex  nucleic  acid  composed  of  both 
deoxyribo-  and  ribonucleotides.  If  the 
ribonucleotides  occupied  intercalary  posi- 
tions, and  the  number  of  adjacent 
deoxy ribonucleotides  was  large,  removal 
of  the  ribonucleotides  by  the  specific 
action  of  RNase  would  depolymerize  the 
DNA  sufficiently  to  impair  its  ability  to 
stain  with  methyl  green  without  impair- 
ing its  Feulgen  stainability.  Such  an 
explanation  was  offered  in  Year  Book  49, 
and  subsequent  developments  in  other 
laboratories  regarding  RNA-DNA  com- 
plexes tended  to  support  such  a  hypothe- 
sis. 

Nevertheless,  the  large  amounts  of 
RNase  necessary  to  reduce  methyl  green 
stainability,  and  the  variations  in  relative 
effectiveness  more  recently  noted  among 
different  samples  of  RNase,  were  sug- 
gestive of  an  impurity  of  unknown 
enzymatic    behavior,     which    we    were 


unable  to  assay  chemically  because  of 
lack  of  knowledge  of  its  substrate, 
essential  cof actors,  and  so  forth.  There- 
fore, when  "chromatographicaHy  pure" 
RNase  became  available  commercially 
last  summer,  we  immediately  tested  it  for 
action  on  methyl  green  stainability  and 
found  it  to  be  completely  ineffective.  But 
chromatography  of  crystalline  RNase 
yields  several  fractions  capable  of  de- 
grading RNA,  only  one  of  which  was 
available  commercially.  So  it  became 
necessary  for  us  to  chromatograph  crys- 
talline RNase  and  analyze  the  various 
fractions  for  ability  to  render  chromo- 
somes unstainable  with  methyl  green. 

Several  examples  of  crystalline  RNase 
(chosen  to  represent  the  extreme  varia- 
tions among  our  samples)  were  chromato- 
graphed  on  the  carboxylic  acid  cation 
exchange  resin  Amberlite  IRC-50,  ac- 
cording to  the  method  of  Hirs,  Moore, 
and  Stein,  with  sodium  phosphate  buffer 
(0.2  M,  pR  6.47)  as  the  eluant,  or  on 
carboxymethyl  cellulose  ion-exchange 
columns  buffered  with  tris(hydroxymeth- 
yl)aminome thane-hydrochloric    acid    at 


0.8  - 


£   0.6 

Q> 
"O 

"5 

o 

q.  0.4 

o 


1         1         1 

-t  1 

i         l         l 

1      1 

I           1 

I            1 

-     \       \ 

\      \            #•♦*» 

\     \       "              \ 

\     \  J*                     \ 

*    V                    6 

>         \ 

*•£                       0 

k         \ 

>        \ 

\       Y> 

*      \ 

\      \ 

*     \ 

—                                   n         r$ 

\     \ 

t    \ 

\    \ 

t   \ 

*   V. 

*  R 

•  Polynucleotide 

\\ 

*  \ 

o  Z  protein 

k\ 

*R 

mm 

>  \ 

v\ 

\V 

\o 

i         i         i 

*  _ 
1         1         1         1 

0.2 


250  270  290 

Wavelength,  m// 

Fig.  2.     Ultraviolet  absorption  spectra  of  two 
impurities  present  in  crystalline  ribonuclease. 


DEPARTMENT    OF    GENETICS 


463 


pH  8.0,  according  to  the  method  of 
Taborsky,  with  0-0.1  M  sodium  chloride 
as  the  eluant.  Each  sample  yielded 
polynucleotide  material  and  three  protein 
fractions  having  RNase  activity.  Cy to- 
chemical  analyses  of  these  fractions, 
singly  and  in  all  possible  combinations, 
proved  that  none  of  them  was  capable  of 
effacing  methyl  green  stainability.  It  was 
apparent  that  the  material  in  crystalline 
RNase  responsible  for  that  activity 
either  (a)  was  present  in  amounts  too 
small  to  be  detected  with  the  monitoring 
system  employed,  (6)  had  been  inacti- 
vated by  the  fractionation  procedures,  or 
(c)  had  remained  on  the  column.  Tests  of 
these  three  possibilities  proved  the  last 
to  be  true. 


By  increasing  the  sodium  chloride 
concentration  of  the  eluting  medium  in 
Taborsky's  procedure  to  1  M,  we  ob- 
tained a  fourth  protein  (fraction  Z,  fig. 
2),  the  concentration  of  which  could  be 
correlated  with  the  ability  of  the  sample 
of  RNase  to  eliminate  methyl  green 
stainability  (table  6) ;  and  treatment  of 
fixed  sections  of  onion  root  tips  with  this 
protein  did  indeed  render  them  nonstain- 
able  with  methyl  green  without  reducing 
their  capacity  to  stain  with  the  Feulgen 
reagent.  The  reduction  in  methyl  green 
stainability  effected  by  fraction  Z  is 
dependent  on  both  time  and  concentra- 
tion. The  new  protein  shows  slight  traces 
of  RNase  activity;  its  specific  activity  is 
approximately    1/25   that   of   chromato- 


TABLE  6.     Summary  of  Assays  of  Various  Samples  of  Crystalline  Ribonuelease 

Assays:  I.     Relative  RNase  activities,  with  the  most  active  (chromatographed  RNase,  A,  accord- 
ing to  Hirs,  Moore,  and  Stein)  as  100. 

II.  Total  phosphorus,  expressed  as  per  cent  of  weight  of  air-dried  enzyme.     Essentially  no  in- 
organic phosphorus  was  detected. 

III.  Deoxyribose,  measured  in  terms  of  DNA-P  and  expressed  as  per  cent  of  weight  of  air-dried 
enzyme. 

IV.  Effect  on  methyl  green  stainability  of  fixed  onion-root-tip  chromosomes.     No  measurable 
effect  =  0;  maximum  effect  =  +  +  +  +  +• 

V.  Z  protein,  expressed  as  the  percentage  of  the  total  optical  density  at  278  rn.fi  placed  on  column 
recovered  in  1  M  sodium  chloride  fraction. 


I 

II 

III 

IV 

Effect  on 

V 

Sample 

RNase 

Phosphorus 

Deoxyribose 

Methyl  Green 

Amount  of 

Activity 

Content 

Content 

Stainability 

Z  Protein 

A 

100 

0 

III* 

99 

0.007 

0 

1 

95 

0.020 

0.005 

+ 

0.05 

2 

94 

0.021 

0.003 

+ 

E 

90 

0.062 

0.028 

+  +  (  +  ) 

Wf 

90 

0.037 

+  + 

G 

86 

0.045 

0.020 

+  +  +  + 

51 

86 

0.016 

0.009 

+  + 

D 

82 

0.058 

0.023 

+  + 

7F 

73 

0.071 

0.028 

+  +  +  + 

H 

72 

0.063 

0.031 

+  +  +  (  +  ) 

8F 

69 

0.099 

0.049 

+  + 

2.3 

11F 

69 

0.063 

0.025 

+  +  +  + 

60 

66 

0.058 

0.020 

+  +  +  + 

4F 

60 

0.116 

0.051 

+  +  +  +  + 

11.7 

70 

56 

0.029 

0.007 

+  +  +  +  + 

14.6 

*  Chromatographed  according  to  Taborsky,  type  III.     Purchased  from  Sigma  Chemical  Company: 
lot  R22B-70. 

f  Purchased  from  Worthington  Biochemical  Corporation:  lot  597-L. 


464  CARNEGIE     INSTITUTION      OF      WASHINGTON 

graphed  RNase.  This  activity  is  probably  fixed  tissue  sections,  was  reported  in  Year 
due  to  contamination  with  residual  RNase  Books  53  and  54 .  Concurrent  with  the 
from  the  column,  but  a  definitive  answer  reduction  in  basophilia  was  a  reduction 
has  not  yet  been  obtained.  So  far  we  have  in  absorption  of  light  in  the  ultraviolet 
been  unable  to  demonstrate  any  enzy-  range  (wavelength,  about  260  nuz),  corre- 
matic  reaction  between  isolated  highly  sponding  to  "loss"  of  purines  and 
polymerized  DNA  and  fraction  Z,  as  pyrimidines.  There  was  also  an  increased 
measured  either  by  increase  in  ultraviolet  absorption  of  404  nux,  indicating  combi- 
absorption  of  the  DNA  or  by  its  ability  nation  of  cytochrome  c,  or  at  least  its 
to  combine  with  methyl  green.  Tests  are  heme  component,  with  the  materials  of 
still  being  made  by  adding  various  the  sections,  and  an  increase  in  stain- 
activators  and  coenzymes  that  could  be  ability  with  acidic  dyes  such  as  fast  green, 
present  in  fixed  biological  materials  but  Similar  studies  with  other  proteins  (lyso- 
absent  from  isolated  DNA.  Until  they  are  zyme,  chymotrypsinogen,  hemoglobin, 
completed  it  seems  futile  to  speculate  egg  albumin,  and  serum  albumin)  indi- 
about  the  mechanisms  whereby  this  cated  that  combination  of  basic  proteins 
protein  reduces  methyl  green  stainability,  per  se  with  cellular  RNA  was  not  the 
or  even  about  the  mechanics  of  methyl  primary  factor  in  reducing  basophilia;  it 
green  staining.  could  account,  however,  for  the  increased 

Chromatography  of  crystalline  RNase  stainability    with    acid    dyes    of    tissue 

in  other  laboratories  has  usually  shown  a  sections  treated  with  basic  proteins.  The 

leading,    enzymatically    inactive    band,  results  with  hemoglobin  were  essentially 

composed    of    material    believed    to    be  similar    to    those    obtained    with    cyto- 

polynucleotide  on  the  basis  of  its  ultra-  chrome  c. 

violet  absorption  spectrum  and  its  reten-  Since  chromatography  of  cytochrome  c 

tion  within  the  membrane  sac  on  dialysis,  had   been   reported   to   yield   a   rapidly 

Our  results  have  confirmed  these  findings  moving  component  containing  no  heme 

(fig.  2)  and  have  shown  that  the  amount  (probably   globin   from   myoglobin),    as 

of    this    contaminant,    as    judged    from  well  as  reduced  cytochrome  c,  oxidized 

phosphorus  analysis,  varies  enormously  cytochrome  c,  and  a  firmly  bound  oxi- 

in  different  preparations   (table  6) .  We  dized  cytochrome  c  that  could  be  removed 

have  analyzed  the  material  for  ribose  by  only  at  high  pH  values,  and  since  we  had 

the  orcinol  and  phloroglucinol  procedures,  found  that  chromatographed  RNase  did 

for  deoxyribose  by  Burton's  modification  not   have   the   same   unusual   effect   on 

of  the  diphenylamine  reaction,  and  for  methyl   green   stainability   as   the   non- 

deoxyribonucleosides    by    the    microbio-  chromatographed  enzyme,  we  decided  to 

logical  assay  procedure  of  Hoff-J0rgensen.  investigate  the  possibility  that  the  unex- 

The  results  indicate  that  the  polynucleo-  pected  effect  of  cytochrome  c  in  reducing 

tide    fraction    is    essentially    polydeoxy-  basophilic  staining  might  also  be  due  to 

nucleotide    rather   than   polyribonucleo-  an    impurity.    Chromatography    of    the 

tide.  But  it  should  be  noted  that  not  all  commercial  sample  of  cytochrome  c  that 

the  phosphorus  is  accountable  as  DNA-  had  been  used  in  our  earlier  work,  on 

phosphorus  (table  6).  Further  analysis  of  Amberlite  IRC-50,  according  to  Paleus 

this    fraction    awaits    stockpiling    of    a  and  Neilands,  yielded  the  expected  four 

supply.  fractions.     The    nonheme    protein    was 

found  to  have  no  effect  on  the  pyronin 

Cytochrome  c  stainability  of  fixed  tissue  sections;  the 

An  enigmatic  action  of  cytochrome  c  three  cytochrome  fractions  seemed  to  be 

(and    other    compounds    containing   the  equally  effective  in  reducing  it.  Appar- 

heme  molecule),  namely,  a  simulation  of  ently,  therefore,  the  simulation  of  ribo- 

RNase  in  reducing  cellular  basophilia  in  nuclease    by    cytochrome    c   in   effacing 


DEPARTMENT    OF    GENETICS 


465 


TABLE  7.     Comparison  of  Rates  of  Hydrolysis  of  Native  and  Heat-Denatured  Deoxyribonucleic 
Acid  by  Salmon-Testis  Deoxyribonuclease  of  Various  Degrees  of  Purity 

Experimental  procedure:  Denatured  DNA  was  prepared  by  heating  a  0.03G  per  cent  solution  of 
native  calf-thymus  DNA  in  0.15  M  acetate  buffer,  pH  5.1,  for  15  minutes  at  98°C,  then  cooling  it 
rapidly  to  1°C.  The  preparation  of  the  DNase  samples  is  described  in  Year  Book  60.  A  and  B 
signify  the  rates  of  formation  of  acid-soluble  split  products  from  native  and  from  heated  DNA, 
respectively. 


Fraction 


Specific  Activity 


B/A 


G 
H 

K 
N 
16* 
NHf 


59 
86 
270 
580 
700 
116 


0.103 
0.100 
0.103 
0.107 
0.103 
0.104 


*  Fraction  16,  plate  2B,  Department  of  Genetics,  Year  Book  60. 
f  Fraction  N,  after  standing  for  10  minutes  at  75°C,  pH  4.5. 


E 
o 

00 
CVJ 

o 

>% 

"in 
c 
a> 

x> 

o 
o 

O 


3.0       - 


rt- 


0.3 


o 

o 

a> 

a> 

o 

o 

.0 

ZJ  

0.2 

3 
C 

o 

O 

JO 

XI 

w. 

^_ 

>» 

X 

«4— 

o 

o 

a> 

•o 

t/> 

o 

i/> 

0 

c 
3 

0.1 

-  0 


10 


20 


30  40 

Fraction  number 


50 


Fig.  3.  Electrophorogram  of  partly  purified  calf-spleen  deoxyribonuclease  II.  Plot  of  5-ml 
fractions  from  descending  electrophoresis  of  40  mg  of  heated  fraction  of  crude  spleen  deoxyribonuclease 
(Year  Book  53)  in  a  0  to  50  per  cent  sucrose  gradient.  Buffer,  0.03  M  borate,  pH  8.5;  time,  21  hours; 
temperature,  3°C;  current,  600-625  volts,  12-14  milliamperes.     Arrow  indicates  the  origin. 


466 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


pyronin  stainability  is  indeed  due  to 
cytochrome  and  not  to  an  impurity  in  the 
preparation.  The  mechanism  of  the  action 
remains  unsolved. 

Deoxyribonuclease  II 

Methods  for  the  preparation  of  highly 
purified  DNase  II  from  salmon  testes,  and 
some  properties  of  the  enzyme,  were 
described  in  Year  Book  60.  As  reported 
there,  it  degrades  heat-denatured  DNA 
only  one-tenth  as  fast  as  native  DNA, 
whereas  pancreatic  DNase  I  degrades 
heat-denatured  DNA  at  a  rate  only 
slightly  slower  than  that  of  its  degrada- 
tion of  native  DNA.  We  suggested  that 
the  testis  preparation  might  contain  two 
enzymes,  one  attacking  only  native,  the 
other  only  heat-denatured,  DNA.  Results 
of  experiments  conducted  this  year  to 
test  that  hypothesis  appear  to  negate  it. 
As  is  evident  from  the  data  in  table  7, 
the  ratio  of  the  two  rates  of  degradation 
remains  constant  throughout  the  later 
stages  of  the  fractionation  procedure. 
Furthermore,     the    ratio    of    the    two 


activities  remains  unchanged  on  heat 
denaturation  of  the  enzyme. 

Preliminary  determinations  of  the 
molecular  weight  of  salmon-testis  DNase, 
based  on  ultracentrifugation  data  sup- 
plied by  Dr.  A.  D.  Hershey,  indicate  a 
value  of  about  52,000. 

To  judge  by  our  results  with  salmon- 
testis  DNase,  zonal  density  gradient 
electrophoresis  can  provide  both  useful 
analytical  information  and  preparative 
methods  for  purification  of  enzymes 
( Year  Book  60) .  The  results  of  application 
of  this  technique  to  partly  purified 
samples  of  DNase  II  from  calf  spleen  (of 
which  we  had  accumulated  a  stock  during 
earlier  attempts  to  purify  that  enzyme) 
are  shown  in  figure  3.  They  strongly 
suggest  that  spleen  contains  several 
RNase's,  and  at  least  two  DNase  IFs 
(cf.  Year  Book  54)  with  similar  specific 
activities.  Modifications  of  the  technique 
will  undoubtedly  enable  us  to  separate 
these  activities  further,  should  it  prove 
necessary  to  add  to  the  battery  of 
DNase's  available  for  dissecting  DNA  in 
studies  of  its  structure. 


ORGANIZATION   OF   CELLULAR   MATERIALS 

B.  P.  Kaufmann,  Helen  Gay,  Jennie  Buchanan,  Ann  Weingart,  Keizo  Maruyama,  and  Alice  Akey 


Much  of  the  work  of  the  past  year  has 
been  devoted  to  completion  of  projects 
already  under  way,  rather  than  the 
initiation  of  new  ones,  in  preparation  for 
the  dissolution  on  June  30  of  the  research 
group  named  above.  The  rationale  of  the 
program  in  its  entirety  and  the  signifi- 
cance of  the  individual  experimental 
components  were  stated  in  Year  Book  60 
and  will  not  be  repeated  here. 

In  studies  of  the  organization  of 
chromosomal  and  cytoplasmic  organelles, 
which  have  represented  our  primary 
endeavors  over  the  years,  we  have  been 
aided  since  1946  by  a  research  grant 
(RG-149)  from  the  National  Institutes  of 
Health,  U.  S.  Public  Health  Service,  and 
since  1957  by  an  additional  grant 
(RG-5336)  from  the  same  agency,  this 


one  sponsored  by  the  Long  Island  Bio- 
logical Association.  Grant  RG-5336  has 
supported  during  the  past  year  the  work 
of  Mrs.  Gloria  Gillies  and  Dr.  C.  C.  Das. 

Miss  Myrna  Thomas,  a  graduate 
student  at  Adelphi  College,  completed  a 
thesis  project  in  the  Department  under 
the  supervision  of  Dr.  Gay,  and  received 
her  M.S.  degree  in  June  1962.  The  work 
is  summarized  later  in  this  report. 

Drs.  Gay  and  Kaufmann  continued  as 
guest  investigators  at  Brookhaven  Na- 
tional Laboratory  and  as  editors  of 
Biological  Abstracts.  On  June  30,  Kauf- 
mann completed  a  term  of  six  years  as  a 
member  of  the  Executive  Committee  of 
the  Division  of  Biology  and  Agriculture 
of  the  National  Research  Council.  As 
president    of    the    Genetics    Society    of 


DEPARTMENT    OF    GENETICS 


467 


America  during  1961,  and  since  then  as 
past-president,  he  has  served  on  the 
Executive  Committee  of  that  Society. 
He  has  also  continued  as  an  associate 
editor  of  The  Nucleus  and  of  the  Inter- 
national Journal  of  Radiation  Biology. 

By  the  time  this  report  is  published, 
most  of  the  individuals  named  above  will 
have  moved  away  from  Cold  Spring 
Harbor.  Miss  Akey  has  joined  the 
research  group  of  Dr.  Philip  Woods  at  the 
University  of  Delaware.  Miss  Thomas 
will  continue  her  graduate  studies  at 
Temple  University  in  Philadelphia.  Miss 
Weingart  plans  to  move  to  the  University 
of  Colorado  Medical  School  in  Denver  to 
work  with  Dr.  Richard  Franklin,  a 
summer  associate  at  the  Biological  Lab- 
oratory. Dr.  Gay  and  Mr.  Maruyama 
will  transfer  their  program  in  cyto- 
genetics to  the  University  of  Michigan, 
where  Dr.  Gay  will  continue  as  a  Staff 
Member  of  the  Institution  and  has 
been  accorded  the  title  Professor  of 
Zoology  by  the  University.  Dr.  Kauf- 
mann  also  has  been  appointed  to  a 
professorship  in  that  University. 

Mutagenicity  of  Agents  Interacting 
with  DNA 

Experiments  designed  to  determine  the 
mutagenic  properties  of  agents  capable  of 
modifying  specifically  the  deoxyribonu- 
cleic acid  (DNA)  of  spermatogenous  cells 
of  Drosophila  have  now  largely  been 
completed. 

The  data  reported  in  Year  Book  60 
showed  that  pancreatic  deoxyribonucle- 
ase  (DNase  I),  when  injected  into  the 
abdomens  of  adult  D.  melanogaster  males, 
is  an  effective  mutagenic  agent,  capable 
of  inducing  both  lethal  mutations  and 
chromosomal  rearrangements.  During  the 
past  year  Kaufmann  and  Buchanan,  with 
the  assistance  of  Mrs.  Gillies,  have  tested 
extensively  a  sample  of  deoxyribonu- 
clease  extracted  by  Dr.  Margaret  Mc- 
Donald from  salmon  testes  (DNase  II). 
This  enzyme  also  has  proved  to  be  a  mild 
but  effective  mutagen.  Viewed  chem- 
ically, these  results  are  not  surprising, 


since  both  DNase  I  and  DNase  II  act 
specifically  to  depolymerize  DNA,  the 
first  to  yield  polynucleotides  with  5'- 
monoesterified  phosphate  end  groups,  the 
second  to  yield  polynucleotides  with 
3'-monoesterified  phosphate  end  groups. 
Thus,  at  face  value,  the  tests  for  muta- 
genic activity  would  suggest  that  the 
direct  action  of  the  enzyme  on  DNA 
causes  both  point  mutations  and  gross 
chromosomal  alterations.  From  a  bio- 
logical viewpoint,  however,  the  results 
are  less  meaningful,  since  our  experiments 
have  shown  that  another  protein  of  about 
the  same  molecular  weight  as  DNase  I 
but  lacking  enzymatic  activity,  namely, 
bovine  plasma  albumin,  also  acts  as  a 
mild  but  effective  mutagen. 

In  the  light  of  these  observations,  we 
directed  our  efforts  toward  an  assay  of 
the  mutagenic  potential  of  an  agent  that 
is  capable  of  altering  DNA,  not  by 
disrupting  the  phosphate-sugar  helices 
but  by  effecting  modification  of  base 
sequences.  The  agent  tested,  5-bromode- 
oxyuridine,  is  selectively  incorporated  in 
DNA  as  a  substitute  for  thymine.  When 
injected  into  adult  males,  it  induced 
sex-linked  lethals  but,  in  contrast  to  the 
DNases,  no  chromosomal  rearrangements. 
The  viable  types  detected  in  these  studies 
represented  but  a  residue  of  the  changes 
originally  induced.  Only  about  one-third 
of  the  males  injected  with  5-bromodeoxy- 
uridine  survived  the  treatment,  and  they 
had  low  fertility.  Moreover,  the  Fi 
progeny  had  poor  viability.  Presumably 
these  responses  reflected  a  high  frequency 
of  damage  to  the  germ  cells  of  the 
injected  individuals. 

To  obtain  a  more  direct  measure  of  the 
effect  of  5-bromodeoxyuridine  on  chromo- 
somes, a  cytological  analysis  of  its  action 
on  mitotic  chromosomes  of  the  meristem 
of  Tradcscantia  root  tips  was  undertaken 
by  Miss  Akey  and  Dr.  Gay.  Preliminary 
results  show  that,  in  the  first  anaphase  or 
telophase  after  administration  of  the 
analogue,  chromosomal  fragments  and 
micronuclei  are  present — an  indication 
that  direct  chromosomal  breakage  does 


468 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


occur.  The  complete  analysis  of  this 
experiment  will  include  scoring  of  frag- 
ments found  in  the  second  mitosis 
following  treatment,  as  well  as  a  com- 
parison of  the  effects  of  several  concen- 
trations of  the  agent. 

Effect  of  "Near-Infrared"  Radiation 
on  Crossing  Over 

A  study,  reported  in  Year  Book  57,  of 
the  effect  of  near-infrared  radiation  on 
crossing  over  suggested  that  the  part  of 
the  electromagnetic  spectrum  around  1  fi 
was  effective  in  increasing  double  cross- 
overs by  reducing  or  eliminating  inter- 
ference. In  subsequent  experiments  we 
attempted  to  measure  the  effects  of  more 
restricted  regions  of  the  spectrum.  The 
results  presented  in  Year  Book  60  indi- 
cated that  radiation  passing  through  a 
Corning  Glass  filter  (C.S.  7-57,  whose 
maximum  transmission  zone  falls  between 
1  and  2.5  ju)  did  not  significantly  increase 
the  frequencies  of  double  crossovers  as 
compared  with  those  in  the  unirradiated 
controls.  We  stated  in  that  report  our 
intention  to  test  a  Corning  Glass  filter 
(C.S.  7-69)  that  covers  the  range  from 
0.72  to  1.0  fi  but  affords  over  75  per  cent 
transmission  at  0.8  p.. 

Those  tests  have  now  been  completed 
in  collaboration  with  Miss  Weingart.  The 
analysis  utilized  a  series  of  six  markers 
spread  out  along  the  euchromatic  part  of 
the  X  chromosome  of  D.  melanogaster, 
namely,  yellow,  apricot,  cross veinless, 
vermilion,  forked,  and  Bar.  Some  13,000 
flies  were  examined  in  the  "infrared- 
treated"  and  control  series.  The  fre- 
quency of  double  crossovers  was  higher 
in  the  treated  than  in  the  control,  and 
statistical  evaluation  indicates  that  the 
difference  is  significant  at  the  2  per  cent 
level. 

The  results  suggest  that  the  part  of  the 
spectrum  around  0.8  /z  can  modify  intra- 
cellular phenomena.  Our  original  investi- 
gation (Year  Book  44)  showed  that 
pretreatment  with  near  infrared  increased 
by  about  50  per  cent  the  frequency  of 


chromosomal  rearrangements  induced  in 
mature  spermatozoa  by  a  given  dose  of 
X  rays.  The  filters  we  were  then  employ- 
ing assumedly  afforded  maximum  trans- 
mission in  the  1  /jl  region ;  but  more  recent 
findings  of  Withrow  and  Mow,  and  the 
Kleins,  suggest  that  the  "far  red,"  with 
peak  wavelengths  in  the  0.76-0.78  ix 
range,  is  probably  the  more  effective 
component.  Our  recent  data  tend  to  sup- 
port the  findings  of  those  investigators, 
although  it  should  be  emphasized  that  we 
have  not  tested  the  potentiating  effect  of 
far-red  light  on  the  action  of  another  type 
of  radiation,  but  rather  its  direct  influence 
on  crossing  over,  which  occurs  under 
normal  conditions  in  females  of  D. 
melanogaster.  Apart  from  this,  we  believe 
that  our  findings  now  give  us  an  im- 
portant lead  in  efforts  to  elucidate  the 
mechanisms  of  recombination  in  higher 
plants  and  animals.  Many  lines  of  evi- 
dence suggest  that  ribonucleic  acid  may 
be  implicated  in  processes  of  structural 
rearrangement,  as  we  noted  in  Year  Book 
57.  The  definitive  experiments  necessary 
to  test  the  validity  of  that  assumption 
can  now  be  fruitfully  undertaken. 

Chromosomal  Changes  during  Cleavage 
Mitoses  in  Drosophila 

In  our  efforts  to  define  normal  and 
aberrant  cell  function  in  terms  of  specific 
chromosomal  materials  and  gene  loci,  we 
have  carried  the  analysis  in  recent  years 
to  the  level  of  resolution  afforded  by  the 
electron  microscope.  Electron  micro- 
graphs show  that  chromosomes  contain  a 
multitude  of  fine  fibrillar  elements,  but 
the  chemical  nature  and  the  patterns  of 
association  of  the  discernible  units  have 
not  been  characterized  satisfactorily. 
What  relationship,  for  example,  do  the 
helically  disposed  100  A  strands,  apparent 
in  an  electron  micrograph  of  a  somatic 
prophase  chromosome,  bear  to  the  half- 
chromatids,  whose  existence  was  dis- 
covered by  cytological  observation  and 
confirmed  experimentally?  How  are  nu- 
cleic   acids    and    proteins   associated   in 


DEPARTMENT    OF    GENETICS  469 

these  strands,  and  what  changes  occur  in  (at  pH  8.0-8.2)  or  acidic  bromphenol  blue 

their    patterns    of    organization    during  (at  pH  2.3).  Before  the  tenth  cleavage 

somatic  and  meiotic  mitoses?  the  nuclei  do  not  stain  with  fast  green, 

In  the  search  for  answers  to  these  and  although   they   color   very   faintly   with 

related     questions,     a    combined    cyto-  bromphenol  blue.  They  begin  to  stain  at 

chemical  and  electron-microscopical  study  the  tenth  division,  when  the  nuclei  have 

of  the  chromosomes  of  D.  melanogaster  moved  to  the  periphery  of  the  egg,  and 

has  been  undertaken  by  Dr.  C.  C.  Das,  they  stain  even  more  deeply  during  the 

working  in  collaboration  with  Kaufmann.  eleventh  cleavage  and  thereafter.  In  the 

In  the  first  phases  of  this  work,  attention  light     of     available     information     these 

has    been   directed    to    analysis    of   the  results  suggest  that  a  type  of  protein 

cleavage    mitoses    up    to    the    time    of  appears  during  the  tenth  and  eleventh 

blastoderm  formation.  A  stock  carrying  a  cleavages  that  was  not  present  at  earlier 

ring-X  chromosome  was  chosen,  primarily  stages.    The    known    specificity    of    the 

for    the    advantages    to    be    gained    in  alkaline  fast  green  method  indicates  that 

studying  problems  of  replication,  since  this   protein  is  a  histone   (the  somatic 

the  chromatids  of  the  ring  X  appear  to  adult  histone) ;  and  its  synthesis  does  not 

separate  during  mitosis  while  maintaining  appear  to  be  restricted  to  the  nucleus, 

the  form  of  a  closed  circle.  This  aspect  of  inasmuch  as  a  comparable  increase  in 

the  study  remains  to  be  explored;  our  stainability  occurs  in  the  yolk  spherules, 

initial  efforts  have  been  concerned  with  located  in  the  cytosome.  Stainability  of 

cytochemical     determinations     of     the  the  chromosomes  with  bromphenol  blue 

amounts  of  DNA  and  proteins  in  the  during  early  cleavages  is  assumedly  due 

chromosomes    during    early    embryonic  to  the  presence  of  another  basic  protein 

development.  (the  so-called  cleavage  histone),  which  is 

Amounts  of   DNA  have  been  deter-  neither   a   protamine    nor   an   arginine- 

mined    spectrophotometrically    by    the  containing  protein,  since  stainability  is 

two-wavelength  method,  from  Feulgen-  lost  after  deamination. 

stained  sections  of  material  fixed  in  either  The  significance  of  these  findings  with 

acetic  acid-alcohol  or  formalin.  Since  the  respect   to   problems   of   differentiation, 

nuclei  divide  synchronously,  and  cells  are  particularly  as  related  to  the  appearance 

not  delimited  until  the  onset  of  blasto-  of  nucleoli  (which  are  not  produced  during 

derm  formation  after  the  eleventh  cleav-  early  cleavages),  the  migration  of  nuclei, 

age,  it  has  been  possible  to  determine  and  the  formation  of  independent  cells 

cytologically    with    great    precision    the  (replacing  the  syncytial  state) ,  cannot  be 

exact  stage  of  development.  The  results  discussed  at  this  time  but  will  be  con- 

of  the  measurements  indicate  clearly  that  sidered    when    the    complete    data    are 

there   is    no   significant   change   in   the  published, 
amount    of    DNA    per    diploid    set    of 

chromosomes    from    the    first    to    the  Cytochemical  Studies 

eleventh  cleavage  mitosis.  The  amount  of  Modification  of  chromosomal  stainability 

DNA  also  corresponds  closely,  as  would  with  ribonuclease.  Last  year  (Year  Book 

be  anticipated,  with  that  present  in  the  60)  we  described  experiments  designed  to 

diploid  nucleus  of  the  larval  neuroblast,  test   whether   the   reduction   of   methyl 

All  these  observations  are  in  harmony  green  stainability  in  chromosomes  effect- 

with  the  tenets  of  the  constancy  hypoth-  ed  by  ribonuclease — a  change  implying 

esis.  modification  of  DNA  but  not  loss  of  this 

To  determine  the  distribution  of  pro-  nucleic  acid,  since  the  Feulgen  reaction 

teins    and    possible    changes    occurring  remains  unimpaired — is  indeed  dependent 

therein,   sections  of  formalin-fixed  eggs  on  the  enzymatic  activity  of  ribonuclease. 

have  been  stained  with  alkaline  fast  green  We  concluded  that  the  loss  of  colorability 


470  CARNEGIE     INSTITUTION      OF      WASHINGTON 

is    not    due    to   a   nonspecific    effect    of  but  a  brief  description  of  our  progress  in 

ribonuclease,  acting  as  a  protein,  but  does  analyzing  the  problem  is  presented  below, 

depend  on  the  concentration,  tempera-  The  chemical  aspects  of  analyses  of  the 

ture,  and  pH  of  the  enzyme  solutions  and  ribonucleases  are  described  elsewhere  in 

on  the  length  of  the  reaction  time.   It  this  volume,  in  the  report  of  Dr.  Margaret 

seemed,    therefore,    that    hydrolysis    by  McDonald. 

ribonuclease  degrades  an  RNA  which  is  Fourteen  different  samples  of  crystal- 
intercalated  in  chains  of  DNA,  so  that  it  line  ribonuclease  prepared  over  a  period 
becomes  depolymerized.  Since  these  find-  of  years  by  Dr.  McDonald  were  tested 
ings  offered  a  basis  for  an  interpretation  cytochemically  by  Dr.  Gay  and  Miss 
of  considerable  interest  in  the  present  era  Weingart  to  determine  their  effect  on  the 
of  DNA  coding,  that  is,  the  cytochemical  stainability  of  chromosomes  by  methyl 
demonstration  of  a  DNA-RNA  complex  green.  The  samples  reacted  differently 
in  chromosomes,  it  was  thought  advisable  when  tested  under  the  same  conditions: 
to  pursue  the  study  with  the  purest  some  caused  very  little  reduction  in 
ribonuclease  available.  colorability,   whereas  others  completely 

Until  recently,  crystalline  ribonuclease  eliminated     it.     Chromatographic     and 

free  of  proteolytic  and  deoxyribonuclease  chemical  analyses  of  the  samples  showed 

activity,    prepared    by    Dr.     Margaret  that  they  contained  different  amounts  of 

McDonald,  had  been  the  most  rigidly  a  nucleotide  fraction   (eluted  from  the 

analyzed    and    purified    sample    of    this  column    before    the    enzyme);    but    the 

enzyme  available  to  us.  With  the  advent  amount   of   this   fraction   present   in   a 

of    column    chromatography,    however,  sample  could  not  be  correlated  with  the 

ribonuclease    has    been    separated    into  degree    of    methyl    green    reduction    it 

several  fractions,  and  as  of  last  summer  produced. 

we  have  been  able  to  obtain  commercially  Further  study  of  the  chromatographed 

from  Sigma  Chemical  Company  a  chro-  ribonucleases  showed  that  none  of  the 

matographed,  "essentially  homogeneous"  fractions,  either  separately  or  combined — 

ribonuclease.    When    this    enzyme    was  and  not  even  a  combination  of  all  the 

tested  cytochemically  under  the  condi-  fractions    to    reconstitute    the    "whole" 

tions  of  our  former  experiments  (6  mg/ml  enzyme — would  cause  the  cytochemical 

in  water  at  pH  6.0  for  2,  4,  or  8  hours),  no  reduction  of  chromosomal  methyl  green 

reduction  of  methyl  green  stainability  of  stainability.    Dr.    McDonald    concluded 

chromosomes  occurred.  that  the  factor  responsible  for  modifica- 

Although  this  finding  raised  serious  tion  of  chromosomal  staining  might  have 
questions  about  the  validity  of  our  remained  on  the  chromatographic  col- 
conclusion  that  an  RNA-DNA  complex  umn.  Elution  with  1  M  NaCl  released  a 
exists  in  chromosomes,  it  also  brought  protein-containing  fraction  from  the  col- 
into  question  the  contention  of  our  umn,  and  cytochemical  tests  showed  that 
critics  that  the  ribonuclease  reduction  of  it  did  impair  methyl  green  stainability. 
methyl  green  stainability  in  our  earlier  This  fraction  is  currently  being  analyzed 
experiments  had  been  due  to  blocking  of  by  Dr.  McDonald  to  determine  whether 
the  stainable  groups  on  the  DNA  mole-  the  reduction  is  referable  to  enzymatic 
cule  through  formation  of  a  nonstainable  activity  and,  if  so,  what  the  substrate  is, 
DNA-ribonuclease  complex.  As  a  result,  whether  DNA  or  RNA. 
we  became  interested  in  determining  Pending  receipt  of  this  definitive 
what  was  present  in  the  crystalline,  information,  we  must  suspend  efforts  to 
protease-free  ribonuclease  samples  that  interpret  elimination  of  methyl  green 
could  effect  a  change  in  methyl  green  stainability  by  ribonuclease  in  terms  of 
staining  of  chromosomes.  Up  to  the  organizational  patterns  of  chromosomal 
moment  the  puzzle  has  not  been  solved,  RNA  and  DNA. 


DEPARTMENT    OF    GENETICS  471 

DNA  content  during  microsporogenesis  cantia  plants,  grown  in  the  greenhouse, 

in   Tradescantia.   Miss   Myrna   Thomas,  were  clonal  descendants  of  a  single  plant, 

who    received    the    M.S.    degree    from  Measurements  of  DNA  in  microspores 

Adelphi  College  in  June  1962,  had  com-  before,  during,  and  after  the  postmeiotic 

pleted    her    research    problem    in    our  mitosis,  which  leads  to  production  of  the 

laboratory  during  the  summer  and  fall  of  generative   and   vegetative   nuclei,   were 

1961  under  the  guidance  of  Dr.  Gay.  By  made    in    cells    from    buds    of    a    single 

cytospectrophotometric  methods,  she  de-  inflorescence.   These    values    were    then 

termined  the  DNA  content  of  the  nuclei  compared  with  the  haploid  DNA  value, 

during  microspore  development  in  Trad-  C,     which     had     been    determined     by 

escantia    paludosa.    Since    we    had   been  measurement   of  late-anaphase   or   telo- 

investigating  by  electron  microscopy  the  phase  chromosomal  groups  of  the  quartet 

ultrastructural   modifications  undergone  or  tetrad  (four-microspore)  stage, 

during  microsporogenesis,  a  clear  under-  The   amounts   of   DNA  measured   in 

standing  of  the  behavior  of  nuclear  DNA  nuclei  of  the  two-  and  four-cell  stages  and 

in  the  course  of  that  process  was  needed,  the  pollen-grain  mitosis  all  conform  with 

Several   previous   studies   had   shown  values    expected    on    the    basis    of    the 

that  in  general,  throughout  meiosis  and  DNA-constancy  hypothesis,  which  pre- 

the    postmeiotic    development    of    the  diets  a  doubling  of  DNA  amount  during 

gametophyte,  the  amount  of  DNA  con-  the  interphase  before  each  mitosis,  except 

forms  with  that  expected  according  to  during  interkinesis,  which  is  the  phase 

the  "constancy  hypothesis."   Some  dis-  between  the  two  meiotic  divisions  when 

crepancies  among  the  results  of  various  reduction    takes    place.    Miss    Thomas 

investigations  had  been  reported,  how-  found  that  in  the  generative  nucleus  of 

ever,    particularly   with   respect   to   the  the  binucleate  pollen  grain,  very  soon 

quantity  of  DNA  in  the  generative  and  after  its  formation,  DNA  is  synthesized 

vegetative  nuclei  just  before  germination  to  the  2C  value.  The  generative  nucleus 

of  the  pollen  grain.  Moses  and  Taylor,  in  is  thus  prepared  well  in  advance  for  the 

1955,    found    a    slow    and    incomplete  mitosis  that  will  form  the  two  sperm 

synthesis    of    DNA    in    the    vegetative  nuclei  in  the  pollen  tube.   Contrary  to 

nucleus,    although    that    nucleus    never  expectation,  she  found  that  during  this 

divides.   Earlier,  however,  one  of  these  same    period    the    vegetative    nucleus, 

authors    suggested,     on    the    basis    of  which  will  not  divide  again,  shows  a  slow 

P32-incorporation  experiments,  that  the  increase  in  amount  of  DNA.  Although 

tube  nucleus  does  not  take  up  radioactive  the  observed  increase  was  not  so  great  as 

phosphorus  and  therefore  does  not  syn-  that    reported    earlier    by    Moses    and 

thesize   DNA.    Results   of   still   another  Taylor,  leading  only  to  a  1.3C  value  as 

spectrophotometric   analysis,   by   Wood-  compared  with  their  1.8C,  it  was  statis- 

ard,  hinted  at  an  increase  in  DNA  in  the  tically  significant. 

vegetative  nucleus,  but  these  data  were  We  conclude  that  DNA  synthesis  in 

inconclusive.  Since  differences  in  methods  T.    paludosa,    throughout    meiosis    and 

of  preparation  and  microspectrophotom-  during  postmeiotic  pollen-grain  develop- 

etry  made  comparisons  among  the  several  ment,    is    consistent    with    expectation, 

Tradescantia  studies  difficult,  a  reinvesti-  except    that    synthesis    occurs    in    the 

gation  of  the  problem  seemed  warranted,  vegetative  nucleus  at  the  time  wThen  it  is 

Amounts  of  DNA  were  determined  by  no   longer  undergoing  mitosis.    Because 

the  two-wavelength  method,  which  facil-  this  phenomenon  has  been  observed  in 

itates  measurement  of  irregularly  shaped  other  studies  and  is  statistically  signifi- 

nuclei.    Feulgen-stained   smear   prepara-  cant  in  the  present  analysis,  it  is  suggested 

tions    ensured    the    inclusion    of    whole  that   the   amoeboid    nucleus   may   form 

nuclei  in  the  determinations.  The  Trades-  some  DNA  whose  function  is  concerned 


472  CARNEGIE     INSTITUTION     OF      WASHINGTON 

with    differentiation    rather    than    with  (Epon    812).    Electron    micrographs    of 

replication.  We  must  now  investigate  the  sections   of   this   material   give   a   clear 

exact  nature  of  this  DNA  increase,  its  picture  of  the  single  membrane  bounding 

variability  in  different  inflorescences,  and  the  dense  bodies  and  of  its  direct  conti- 

its  relation  to  metabolic  cellular  changes,  nuity    with  the  endoplasmic  reticulum. 

They  leave  little  doubt  that  the  dense 

Electron-Microscope  Studies  bodies  fre  ff  med1  b^  accumulation  of 

material    withm    the    cisternae    of    the 

Electron-microscope  investigations  endoplasmic  reticulum,  and  that  they 
during  the  year  have  focused  on  the  fuse  to  produce  the  large  vacuole, 
modifications  in  cytoplasmic  organelles  The  origin  of  vacuoles  from  the  endo- 
throughout  differentiation  and  growth  of  plasmic  reticulum  has  been  reported  by 
two  types  of  Tradescantia  cells.  Mr.  the  French  cytologist  Buvat  and  his 
Maruyama,  in  collaboration  with  Dr.  co-workers,  and  by  Whaley  et  al.  The 
Gay,  has  completed  a  study  of  ultra-  French  workers  have  shown  that  small 
structural  changes  in  the  stigma  cell  and  vacuoles  arise  as  swellings  of  the  endo- 
is  currently  engaged  in  observing  changes  plasmic  reticulum  in  Elodea  buds  and 
that  occur  during  microsporogenesis.  Triticum  roots.  Whaley's  group,  observ- 
Preliminary  results  of  the  stigma-cell  ing  dense  "lipid  bodies"  as  well  as 
study  were  described  in  Year  Book  60;  the  vacuoles  with  contents  of  low  electron 
findings  are  brought  up  to  date  in  the  density  in  maize  roots,  postulated  that 
following  paragraphs.  The  initial  studies  the  dense  bodies  were  precursor  vacuoles 
of  changes  in  fine  structure  of  the  pollen  containing  substances  in  high  concentra- 
grain,  after  the  differential  mitosis  which  tion.  Our  observations  of  Tradescantia 
produces  the  vegetative  and  generative  stigma  cells  confirm  the  conclusions  of 
nuclei,  were  concerned  mainly  with  these  workers  and  extend  their  findings, 
formation  of  the  wall  between  the  two  In  Tradescantia  stigma  cells  the  origin 
cells.  Analysis  of  the  changes  in  cyto-  of  the  vacuolar  system  from  the  endo- 
plasmic organelles  during  this  period  has  plasmic  reticulum  can  be  more  clearly 
now  been  completed.  Observations  have  visualized  than  in  the  types  of  cells 
also  been  made  of  the  earlier  develop-  studied  by  others.  In  the  first  place,  the 
mental  changes  during  the  two  meiotic  vacuole  arises  in  a  cytoplasm  devoid  of 
divisions  leading  to  formation  of  the  any  vacuoles.  Second,  the  cisternae  of  the 
microspore.  endoplasmic  reticulum  that  are  destined 

Fine  structure  of  the  developing  stigma  to   become   vacuole   precursors   develop 

cell.  At  certain  stages  in  the  growth  of  the  more  or  less  synchronously,  so  that  the 

Tradescantia   stigma   cell,   ellipsoidal   or  sequence  of  events  in  vacuole  formation — 

irregularly  shaped  electron-dense  bodies,  that  is,  swelling,  accumulation  of  dense 

limited  by  a  single  membrane,  appear  in  material,  subsequent  dilution  of  contents, 

the  cytoplasm.  As  the  cell  grows,  these  and  fusion  into  a  large  vacuole — is  very 

"dense  bodies"  gradually  increase  in  size,  noticeable.   Last,  these  changes  can  be 

become  less  dense,  and  fuse  to  form  a  clearly  associated  with  cellular  growth, 

single    large    vacuole.    Observations    re-  since  the  age  of  the  cell  being  studied  is 

ported  last  year  suggested  that  the  dense  determined  on  the  basis  of  cytological  or 

bodies  arise  from  the  endoplasmic  reticu-  morphological  characteristics  of  the  bud 

lum  of  the  cell,  and  therefore  that  the  or  flower,  not  merely  by  its  own  appear- 

vacuole  itself  originates  from  this  cyto-  ance.  Thus,  the  origin  of  the  large  vacuole 

plasmic  membrane  system.  from    its    earliest    precursors    can    be 

Better-preserved  potassium  permanga-  unequivocally  traced, 

nate-fixed  stigma  cells  have  now  been  Unfortunately,  we  have  not  been  able 

obtained  by  imbedding  in  an  epoxy  resin  to  discover  the  nature  of  the  electron- 


Abbreviations  on  plates:  gn,  generative  nucleus;  vn,  vegetative  nucleus;  g,  Golgi  bodies;  p,  plastid; 
m,  mitochondrion;  er,  endoplasmic  reticulum;  s,  spherosome;  cw,  cell  wall  between  the  generative  and 
vegetative  cells;  i,  intine,  and  e,  extine,  of  pollen  wall.     Solid-line  marker  indicates  1  m- 


Plate  1 


Department  of  Genetics 


er 


gn 


m 


Pollen  grain  of  Tradescantia  paludosa,  just  after  formation  of  generative  and  vegetative  nuclei. 
Cell  wall  between  the  two  cells  is  continuous  with  the  intine  of  the  pollen  wall  (arrow).  The  two  dense 
lines  bordering  the  cell  wall  are  continuous  with  the  line  that  borders  the  intine;  each  line  may  rep- 
resent the  plasma  membrane  of  one  of  the  two  cells.  Golgi  bodies  are  composed  of  one  or  a  few  con- 
centric cisternae. 


Plate  2 


Department  of  Genetics 


Pollen  grain  of  Tradescantia  at  a  somewhat  later  stage  than  that  shown  in  plate  1.  Golgi  bodies 
have  larger  numbers  of  paired  concentric  layers  than  those  in  previous  micrograph.  Apparently, 
growth  of  the  pro-Golgi  bodies  occurs  through  an  increase  in  number  of  concentric  cisternae. 


Plate  3 


Department  of  Genetics 


P 


vn 


Pollen  grain  whose  generative  nucleus  is  crescent  shaped.  Golgi  bodies  have  assumed  their  typical 
form,  probably  by  an  opening-out  of  the  concentric  cisternae  of  the  pro-Golgi  apparatus. 


Plate  4 


Department  of  Genetics 


Golgi  bodies  of  adult  form,  at  a  higher  magnification.  They  are  now  composed  of  stacks  of  straight, 
parallel  cisternae. 


DEPARTMENT    OF    GENETICS 


473 


dense  vacuolar  substance  of  the  Trades- 
cantia  stigma  cell.  Our  studies  show  that 
fixation  with  osmium  tetroxide  or  forma- 
lin does  not  preserve  this  material. 
Identification  of  the  vacuolar  contents 
would  enable  us  to  determine  objectively 
whether  the  substance  in  young  and  old 
vacuoles  is  the  same.  It  might  also  help 
us  understand  more  fully  the  functional 
significance  of  these  special  cisternae  of 
the  endoplasmic  reticulum. 

Our  findings  indicate,  then,  that  the 
vacuolar  system  in  plants  originates  from 
the  endoplasmic  reticulum  and  conse- 
quently that  the  tonoplast  of  the  Trades- 
cantia  stigma  cell  consists  of  a  single 
membrane.  The  real  challenge  of  these 
findings  consists  in  the  problem  they 
raise  as  to  the  predisposing  factors 
responsible  for  the  differentiation  of  parts 
of  an  apparently  morphologically  homo- 
geneous cytoplasmic  membrane  system 
into  a  vacuolar  system. 

Fine  structure  and  development  of  the 
pollen  grain.  1.  The  cell  wall.  Last  year, 
when  studies  of  the  ultrastructural 
changes  during  microsporogenesis  in  Trad- 
escantia  were  just  getting  under  way, 
we  reported  our  early  findings  about  the 
nature  of  the  thin  cell  wall  that  separates 
the  generative  and  vegetative  cells.  We 
noted  a  continuity  between  the  newly 
formed  wall  and  the  intine  (the  innermost 
layer  of  the  pollen  wall) ,  and  a  similarity 
in  structure  and  density,  observable 
shortly  after  formation  of  the  vegetative 
and  generative  nuclei  (pi.  1).  Because  of 
these  relationships  it  was  assumed  that 
the  cell  wall,  like  the  PAS-positive  intine, 
is  carbohydrate  and  probably  made  up  of 
cellulose  and  pectin,  which  have  low 
electron-scattering  power.  In  a  recent 
study,  Bopp-Hassenkamp  notes  the  two 
dense  membranes  lying  on  either  side  of 
the  lighter  middle  layer  (see  pi.  1),  and 
suggests  that  together  they  form  the 
plasmalemma  of  the  generative  cell 
rather  than  a  true  cell  wall.  Our  interpre- 
tation is  that  the  dense  membranes  are 
the  plasma  membranes  of  the  generative 
and  vegetative  cells,  which  are  separated 


by  the  true  cell  wall,  the  lighter  middle 
layer. 

During  the  past  months  we  have  been 
analyzing  the  formation  of  the  cell  wall 
when  the  cell  plate  is  laid  down  at  the 
telophase  of  the  microspore  mitosis. 
Endoplasmic  reticulum  cisternae  are  very 
closely  involved  in  this  process,  as  other 
workers  also  have  noted.  Vesicles,  sup- 
posedly pectin  containing,  with  an  inter- 
nal content  whose  density  is  similar  to 
that  of  the  thin  separating  cell  wall 
described  above,  are  contiguous  with 
endoplasmic  reticulum  vesicles  in  this 
region.  Mr.  Maruyama  is  currently 
investigating  the  extent  to  which  endo- 
plasmic reticulum  cisternae  contribute  to 
pectin  vesicle  production  and  plasma 
membrane  formation. 

2.  The  Golgi  apparatus.  A  study  has 
been  made  of  the  submicroscopic  struc- 
ture of  the  pollen  grain  from  the  time  of 
the  microspore  mitosis  until  the  time  of 
pollen-grain  maturity,  which  occurs  in 
Tradescantia  as  the  flower  opens.  The 
analysis  has  been  restricted  to  changes  in 
the  cytoplasm  of  the  vegetative  cell,  since 
the  differential  mitosis  leaves  virtually  no 
cytoplasm  around  the  generative  nucleus. 
Development  and  multiplication  of  cyto- 
plasmic organelles  proceeds  throughout 
the  period  studied,  reflecting  a  high 
metabolic  activity.  The  endoplasmic  re- 
ticulum increases  in  amount,  particularly 
just  before  the  flower  opens,  when  the 
lamellae  become  arranged  in  more  or  less 
parallel  bundles.  Plastids  multiply,  en- 
large, and  accumulate  starch.  Mito- 
chondria also  multiply,  and  in  later  stages 
of  development  become  longer.  Their 
internal  organization  is  modified  so  that 
cristae,  which  were  previously  short  and 
perpendicular  to  the  outer  membrane, 
become  long  and  parallel  to  the  long  axis 
of  the  organelle.  These  findings  and  their 
implications  will  be  discussed  in  a  future 
publication.  We  shall  restrict  the  discus- 
sion here  to  changes  in  the  Golgi  appa- 
ratus, since  they  represent  new  observa- 
tions not  described  by  others. 

Electron  microscopy  in  the  past  few 


474 


CARNEGIE     INSTITUTION      OF     WASHINGTON 


years  has  revealed  that  a  Golgi  apparatus, 
similar  in  structure  to  that  first  demon- 
strated in  animal  cells,  occurs  in  many- 
kinds  of  plant  cells.  The  Golgi  bodies  are 
composed  of  stacks  of  cisternae  with 
associated  small  vesicles,  and  are  usually 
dispersed  at  random  within  the  cyto- 
plasm. Most  published  reports  suggest 
that  there  is  no  very  great  change  in  these 
structures  as  differentiation  proceeds. 
Some  workers  have  observed  increased 
dilation  of  the  ends  of  the  cisternae,  and 
others  have  noted  an  increased  number  of 
small  vesicles,  but  no  one  has  reported 
any  major  modifications  of  the  whole 
organelle. 

We  find  that  in  Tradescantia,  during 
the  period  from  microspore  mitosis  until 
shortly  after  pollen  maturity,  all  the 
Golgi  bodies  in  the  vegetative  cells 
undergo  structural  changes  more  or  less 
synchronously.  Because  of  this  syn- 
chrony, all  Golgi  bodies  within  a  cell  look 
approximately  similar,  and  their  morpho- 
logical modifications  as  they  change 
together  become  conspicuous.  The  se- 
quence of  developmental  changes  can 
readily  be  reconstructed,  since  the  stages 
of  cell  growth  are  identified  by  well  known 
nuclear  events.  Just  after  microspore 
mitosis,  no  Golgi  bodies  of  the  charac- 
teristic type  (i.e.,  stacks  of  cisternae)  are 
found.  Instead,  seen  in  micrographs  of 
ultrathin  sections,  are  concentric  multi- 
layered  ring  structures  (pi.  1).  The 
spacing  of  the  rings  makes  them  appear 
to  be  paired.  The  number  of  paired  rings 
increases  as  the  cell  grows  (pi.  2),  a 
maximum  of  seven  pairs  constituting  the 
largest  complex,  which  is  about  1  /z  in 
diameter.  Sometimes  the  last  two  or  four 
outer  layers  do  not  form  complete  rings 
but  are  continuous  with  each  other,  thus 
forming  apparent  cisternae. 

At  a  later  stage  of  development,  when 
the  generative  nucleus  is  crescent  shaped, 
the  Golgi  bodies  assume  their  typical 
form,  stacks  of  straight  or  slightly  curved 


parallel  cisternae,  seven  to  twelve  in 
number  (pis.  3  and  4).  Finally,  in  the 
pollen  grain  of  the  open  flower,  the  num- 
ber of  flattened  cisternae  is  reduced  and 
the  ends  of  some  of  the  sacs  are  promi- 
nently dilated.  These  observations  sug- 
gest that  the  structural  modifications  we 
have  seen  may  be  related  to  growth  of 
the  Golgi  apparatus.  The  pro-Golgi 
appears  in  cross  section  as  one  or  two 
pairs  of  concentric  rings.  Although  we 
have  not  yet  reconstructed  a  model  from 
serial  sections,  the  spacing  and  shape  of 
the  rings  suggest  that  the  young  Golgi 
apparatus  must  be  either  spherical  or 
disclike.  Since  thin  sections  of  these 
bodies  always  show  circular  rings,  we  are 
inclined  to  favor  a  spherical  shape.  In 
thick  sections,  on  the  other  hand,  the 
appearance  of  these  bodies  is  reminiscent 
of  the  "osmiophilic  platelets"  described 
by  Bowen  and  postulated  by  him  to  be 
the  Golgi  of  plants.  Whatever  its  three- 
dimensional  form,  the  organelle  appears 
to  increase  in  size  by  the  addition  of 
concentric  cisternae.  We  believe  that 
these  arrays  open  out  to  form  the  stacks 
of  parallel  straight  cisternae,  which  then 
decrease  in  number  by  vesiculation  of  the 
cisternal  elements  themselves. 

We  shall  not  discuss  here  the  mecha- 
nisms proposed  by  others  to  explain 
multiplication  of  the  stacks  of  cisternae 
in  Golgi  bodies.  If  our  observations 
concerning  the  growth  of  this  organelle 
are  correct,  it  must  be  concluded  that  the 
Golgi  body  does  arise  and  develop  from  a 
simple  structure,  circular  in  cross  section, 
into  a  multilayered  concentric  structure, 
and  finally  into  the  familiar  stacked  form. 
If  other  mechanisms  do  obtain,  we  must 
conclude  that  this  cytoplasmic  organelle 
may  reproduce  in  several  different  ways. 
In  our  preliminary  study  of  the  earlier 
stages  of  microsporogenesis,  that  is, 
premeiotic  mitosis  and  meiosis,  the  same 
series  of  changes  in  the  Golgi  apparatus 
has  been  noted. 


DEPARTMENT    OF    GENETICS 


475 


BIBLIOGRAPHY 


Akey,  A.,  and  H.  Gay,  Ribonucleic  acid  in 
heterochromatin  (abstract),  Rec.  Genet.  Soc. 
Am.,  31,  68-69,  1962. 

Balbinder,  E.,  The  fine  structure  of  the  loci  tryC 
and  tryD  of  Salmonella  typhimurium,  I, 
Determination  of  the  order  of  mutational  sites 
by  three-point  transduction  tests,  Genetics, 
47,  469-482,  1962. 

Balbinder,  E.,  The  fine  structure  of  the  loci  tryC 
and  tryD  of  Salmonella  typhimurium,  II, 
Studies  of  reversion  patterns  and  the  behavior 
of  specific  alleles  during  recombination, 
Genetics,  47,  545-559,  1962. 

Burgi,  E.,  and  A.  D.  Hershey,  A  relative  molec- 
ular weight  series  derived  from  the  nucleic 
acid  of  bacteriophage  T2,  J.  Mol.  Biol.,  8, 
458-472,  1961. 

Burgi,  E.,  and  A.  D.  Hershey,  Specificity  and 
concentration  limit  in  self-protection  against 
mechanical  breakage  of  DNA,  J .  Mol.  Biol.,  4, 
313-315,  1962. 

Cairns,  J.,  An  estimate  of  the  length  of  the  DNA 
molecule  of  T2  bacteriophage  by  autoradiog- 
raphy, J.  Mol.  Biol.,  8,  756-761,  1961. 

Gay,  H.,  Cytochemical  localization  of  cellular 
nucleic  acids  and  proteins  and  determination 
of  their  patterns  of  association,  Ann.  histo- 
chim.,  4,  467-476,  1961. 

Gay,  H.,  see  also  Akey,  A.;  Larsen,  V.  R. ; 
Maruyama,  K.;  Perreault,  W.  J.;  Woods,  P.  S. 

Hershey,  A.  D.,  see  Burgi,  E.;  Rubenstein,  I. 

Kaufmann,  B.  P.,  see  Maruyama,  K. 

Larsen,  V.  R.,  and  H.  Gay,  A  motor-controlled 
device  for  slow-motion  cutting  with  the 
ultramicrotome,  Stain  Technol.,  87,  211-215, 
1962. 


McClintock,  B.,  Some  parallels  between  gene- 
control  systems  in  maize  and  in  bacteria, 
Am.  Naturalist,  95,  265-277,  1961. 

McDonald,  M.  R.,  Deoxyribonuclease  from 
salmon  testes,  I,  Purification  and  properties, 
J.  Gen.  Physiol,  45,  77-92,  1962. 

Maruyama,  K.,  Electron  microscope  observa- 
tion on  the  development  of  chloroplasts  of 
Avena  and  chlorophyll-deficient  mutants, 
Cytologia,  26,  105-115,  1961. 

Maruyama,  K.,  H.  Gay,  and  B.  P.  Kaufmann, 
Development  of  the  Golgi  body  in  the  Trades- 
cantia  pollen  grain  (abstract),  Am.  J.  Botany, 
49,  662,  1962. 

Maruyama,  K.,  H.  Gay,  and  B.  P.  Kaufmann, 
Development  of  the  vacuole  in  the  stigma 
cell  of  Tradescantia  paludosa  (abstract),  Am. 
J.  Botany,  49,  662,  1962. 

Perreault,  W.  J.,  and  H.  Gay,  The  DNA  of 
Drosophila  spermatocytes  (abstract),  Am. 
Zoologist,  2,  436-437,  1962. 

Rubenstein,  I.,  C.  A.  Thomas,  Jr.,  and  A.  D. 
Hershey,  The  molecular  weights  of  T2 
bacteriophage  DNA  and  its  first  and  second 
breakage  products,  Proc.  Natl.  Acad.  Sci. 
U.  S.,  47,  1113-1122,  1961. 

Sengiin,  A.,  see  Woods,  P.  S. 

Simon,  E.  H.,  DNA  synthesis  in  the  development 
of  resistance  to  ultraviolet  irradiation  in  T2- 
infected  E.  coli,  Virology,  51,  237-244,  1961. 

Thomas,  C.  A.,  Jr.,  see  Rubenstein,  I. 

Woods,  P.  S.,  H.  Gay,  and  A.  Sengiin,  Organiza- 
tion of  the  salivary-gland  chromosome  as 
revealed  by  the  pattern  of  incorporation  of 
H3-thymidine,  Proc.  Natl.  Acad.  Sci.  U.  S., 
47,  1486-1493,  1961. 


PERSONNEL 

Year  Ended  June  30,  1962 


Bocskay,  Elizabeth  M.  (Mrs.),  Stenographer- 
Typist;  Chief  Clerk 
Buchanan,  Jennie  S.  (Mrs.),  Research  Assist- 
ant; Curator  of  Drosophila  Stocks 
Burgi,  Elizabeth,  Associate  in  Microbiology 
Caldarelli,  Donald,  Maintenance  Man 
Carhart,  Anne  K.  (Mrs.),  Research  Assistant 
Carley,  Catherine,  Switchboard  Operator  and 

Computer 
Das,  C.  C.,1  Research  Assistant 

1  USPHS  research  grant  RG-5336,  adminis- 
tered by  the  Biological  Laboratory. 


Fisher,  Agnes  C,  Secretary  to  Director;  Editor 
Frankel,  Fred  R.,  Postdoctoral  Fellow,  U.  S. 

Public  Health  Service 
Gay,  Helen,  Cytogeneticist 
Gillies,  Gloria  (Mrs.),1  Research  Assistant 
Goldberg,  Edward,  Postdoctoral  Fellow,  The 

National  Foundation 
Hershey,  Alfred  D.,  Microbiologist 
Jones,  Henry  H.,  Photographer 
Kaufmann,  Berwind  P.,2  Director 

2  Retired  June  30,  1962. 


476 


CARNEGIE     INSTITUTION      OF      WASHINGTON 


Klees,  Bertha,  Dormitory  Cook 
McClintock,  Barbara,  Cytogeneticist 
McDonald,  Joseph  L.,  Janitor 
McDonald,  Margaret  R.,  Chemist 
McDonald,  William  T.,  Janitor 
Maruyama,  Keizo,  Research  Assistant 
Mosig,    Gisela,    Postdoctoral    Fellow,  U.  S- 

Public  Health  Service 
Peckham,  Leslie  E.,2  Senior  Clerk 
Rogers,  Claude  F.,2  Chief  Clerk 
Smith,  Guinevere  C.  (Mrs.),  Librarian 
Van  Houten,  William  B.,  Engineer 
Weingart,  Eleanor  Ann,  Research  Assistant 
White,  Harry  S.,  Superintendent  of  Buildings 

and  Grounds;  Chief  Mechanic 
Wilson,  Carole  E.  (Mrs.),  Technical  Assistant 

2  Retired  June  30,  1962. 


Temporary  and  Part-Time 

Ahlers,  Paul  de  Wolff,  Maintenance  Man 
Akey,  Alice  L.,  Research  Assistant 
Carroll,  Ann  C,  Library  Assistant 
Cassle,  Marietta  M.,3  Research  Assistant 
Champney,  Scott,  Research  Assistant 
Coyne,  Mary  T.  (Mrs.),  Laboratory  Assistant 
Ingraham,  Laura  J.  (Mrs.),  Research  Assistant 
Kurshan,  Jane,  Research  Assistant 
Lutjen,  George  P.,  Jr.,  Maintenance  Man 
Olsen,  Kirsten,3  Research  Assistant 
Perreault,  William,  Research  Assistant 
Sepe,  Domenico,  Maintenance  Man 
Thomas,  Myrna  C,  Research  Assistant 
Treanor,  Ellen,2  Dormitory  Housekeeper 

3  Biological  Laboratory  Undergraduate  Re- 
search Participation  Program,  sponsored  by  the 
National  Science  Foundation,  summer  1961. 


Bibliography 

July  1,  1961  -  June  30,  1962 


PUBLICATIONS  OF  THE  INSTITUTION 

Year  Book  60,  1960-1961.  Octavo,  xi  +  535 
pages,  29  plates,  202  figures.  December  11, 
1961. 


619.    Mayapan,  Yucatan,  Mexico.  Quarto,  v  + 

515   pages,    frontispiece,    90   figures,    two 

folding  maps.  June  1962. 

Introduction.  H.  E.  D.  Pollock.  Pages 
1-22. 

Part  1.  Ralph  L.  Roys.  Literary  sources 
for  the  history  of  Mayapan.  Pages  23-86, 
frontispiece. 

Part  2.  Tatiana  Proskouriakoff.  Civic  and 
religious  structures  of  Mayapan.  Pages 
87-163,  frontispiece,  figures  1-12. 

Part  3.  A.  Ledyard  Smith.  Residential 
and  associated  structures  at  Mayapan. 
Pages  165-320,  figures  1-23. 

Part  4.  Tatiana  Proskouriakoff.  The  arti- 
facts of  Mayapan.  Pages  321-515, 
figures  1-53. 

621.  Contributions  to  Embryology,  volume  xxxvii. 
Quarto,  iv  +129  pages,  94  plates,  20  text 
figures.  March  1962. 

252.  George  W.  Bartelmez.  The  prolifera- 
tion of  neural  crest  from  forebrain 
levels  in  the  rat.  Pages  1-12,  8  plates, 
7  text  figures. 

253.  George  W.  Bartelmez  and  A.  S. 
Dekaban.  The  early  development  of 
the  human  brain.  Pages  13-32,  30 
plates. 

254.  Bent  G.  Boving.  Anatomical  analysis 
of  rabbit  trophoblast  invasion.  Pages 
33-55,  15  plates,  1  text  figure. 


255.  L.  E.  DeLanney  and  J.  D.  Ebert  in 
collaboration  with  C.  M.  Coffman 
and  A.  M.  Mun.  On  the  chick 
spleen:  origin;  patterns  of  normal 
development  and  their  experimental 
modification.  Pages  57-85,  14  plates, 
1  text  figure. 

256.  Pieter  A.  de  Vries  and  John  B.  de 
C.  M.  Saunders.  Development  of  the 
ventricles  and  spiral  outflow  tract  in 
the  human  heart.  Pages  87-114,  9 
plates,  9  text  figures. 

257.  Roberto  Narbaitz.  The  primordial 
germ  cells  in  the  male  human  embryo. 
Pages  115-119,  5  plates. 

258.  John  McKenzie.  The  development  of 
the  sternomastoid  and  trapezius 
muscles.  Pages  121-129,  13  plates,  2 
text  figures. 

622.  Steinhart,  John  S.,  and  Robert  P.  Meyer, 
with  contributions  by  William  E.  Bonini, 
T.  Jefferson  Smith,  and  George  P.  Wool- 
lard.  Explosion  Studies  of  Continental 
Structure.  Quarto,  xiii  +  409  pages,  142 
figures,  2  plates  in  back  cover  pocket. 
July  1961. 

PUBLICATIONS  BY  THE  PRESIDENT 
OF  THE  INSTITUTION 

Caryl  P.  Haskins 

(With  Edna  F.  Haskins,  John  J.  A.  Mc- 
Laughlin, and  Richard  E.  Hewitt)  Poly- 
morphism and  population  structure  in 
Lebistes  reticulatus,  an  ecological  study. 
Pages  320-395  in  Vertebrate  Speciation,  a 
University  of  Texas  Symposium,  edited  by 
W.  Frank  Blair.  University  of  Texas  Press, 
Austin.  August  1961. 


477 


478 


CARNEGIE     INSTITUTION     OF      WASHINGTON 


A  flower  where  the  roads  divide.  Proceedings 
of  the  American  Philosophical  Society, 
volume  105,  number  4,  pages  452-458, 
August  1961. 

Report  of  the  President.  Carnegie  Institution 
of  Washington  Year  Book  60,  pages  1-54. 
Carnegie  Institution  of  Washington,  Wash- 
ington, D.  C.  December  11,  1961.  Excerpts 
reprinted  under  title  Indigenous  science  for 
new  nations,  in  Current,  March  1962,  pages 
27-29. 

Technology,  science  and  American  foreign 
policy.  Foreign  Affairs,  volume  40,  number 
2,  pages  225-243.  January  1962. 


PUBLICATIONS  BY  THE  EXECUTIVE 
OFFICER  OF  THE  INSTITUTION 

Edward  A.  Ackerman 

Reasons  for  research  and  development  on 
water  desalting.  Statement  prepared  for  the 
Desalination  Research  Conference,  June  19 
to   July    14,    1961,    National   Academy   of 


Sciences — National  Research  Council  and 
Office  of  Saline  Water,  Woods  Hole,  Mass. 
Included  in  Saline  Water  Research  and 
Development  Program,  Hearings  before  the 
Subcommittee  on  Irrigation  and  Reclama- 
tion of  the  Committee  on  Interior  and 
Insular  Affairs,  House  of  Representatives, 
87th  Congress,  1st  Session,  on  H.  R.  152, 
H.  R.  431,  H.  R.  949,  and  H.  R.  2991;  H.  R. 
595  and  H.  R.  5883;  H.  R.  3089;  H.  R.  4721, 
H.  R.  4757,  and  H.  R.  4759;  and  H.  R.  7916. 
March  17,  June  26  and  27,  and  July  17  and 
18,  1961.  Serial  no.  7.  Pages  253-261. 
Printed  for  the  use  of  the  Committee  on 
Interior  and  Insular  Affairs.  U.  S.  Govern- 
ment Printing  Office,  Washington,  1961. 
(With  Marion  Clawson)  Toward  a  permanent 
farm  policy.  Pages  65-71  in  Man  and  His 
Environment,  Proceedings  of  the  Second 
National  Congress  on  Environmental  Health, 
Ann  Arbor,  Michigan,  June  6-8,  1961. 
Continued  Education  Series  95,  distributed 
by  Continued  Education  Service,  School  of 
Public  Health,  University  of  Michigan. 


Administrative  Reports 


Report  of  the  Executive  Committee 


To  the  Trustees  of  the  Carnegie  Institution  of  Washington: 

Gentlemen:  In  accordance  with  the  provisions  of  the  By-Laws,  the  Executive  Committee 
submits  this  report  to  the  annual  meeting  of  the  Board  of  Trustees. 

During  the  fiscal  year  ending  June  30,  1962,  the  Executive  Committee  held  four  meetings. 
Printed  accounts  of  these  meetings  have  been  or  will  be  mailed  to  each  Trustee. 

The  estimate  of  expenditures  for  the  fiscal  year  beginning  July  1,  1962,  has  been  reviewed 
by  the  Executive  Committee. 

Two  vacancies  exist  in  the  membership  of  the  Board  of  Trustees,  resulting  from  the 
deaths  of  James  F.  Bell  in  May  1961  and  of  Robert  Woods  Bliss  in  April  1962.  The  office 
of  Secretary  of  the  Board  is  vacant  because  of  Mr.  Bliss's  death. 

The  terms  of  office  of  the  Chairmen  of  all  Committees  of  the  Board  expire  on  May  11, 
1962.  A  vacancy  has  occurred  in  the  Executive  Committee  by  reason  of  the  resignation  of 
Henry  R.  Shepley.  The  terms  of  the  following  members  of  Committees  also  expire  on 
May  11,  1962: 


Executive  Committee 
Robert  A.  Lovett 
James  N.  White 


Retirement  Committee 
Omar  N.  Bradley 
Henry  S.  Morgan 


Finance  Committee 
Richard  S.  Perkins 
Elihu  Root,  Jr. 
James  N.  White 


May  11,  1962 


Nominating  Committee 
Walter  S.  Gifford 
Henry  S.  Morgan 

HENRY  S.  MORGAN,  Chairman 


481 


Report  of  Auditors 


Lybrand,  Ross  Bros.  G.  Montgomery 


To  the  Auditing  Committee  of  Carnegie  Institution  of  Washington: 

We  have  examined  the  statement  of  assets,  liabilities  and  fund  balances 
of  Carnegie  Institution  of  Washington  as  of  June  30,  1962,  and  the  related  sum- 
mary statement  of  changes  in  funds  for  the  year  then  ended  and  the  supporting 
exhibits  and  schedules,  which  have  been  prepared  on  the  general  basis  of  cash 
receipts  and  disbursements  and  accordingly  do  not  reflect  accrued  income, 
accounts  payable  nor  provision  for  depreciation.    Our  examination  was  made 
in  accordance  with  generally  accepted  auditing  standards,  and  accordingly 
included  confirmation  from  the  custodian  of  securities  owned  at  June  30,  1962, 
and  such  tests  of  the  accounting  records  and  such  other  auditing  procedures 
as  we  considered  necessary  in  the  circumstances. 

In  our  opinion,  the  accompanying  financial  statements  and  supporting 
exhibits  and  schedules  present  fairly  the  assets,  liabilities  and  fund  balances 
of  Carnegie  Institution  of  Washington  at  June  30,  1962,  and  the  changes  in 
funds  for  the  year  then  ended  on  a  basis  consistent  with  that  of  the  preceding 
year. 


ok^wCsw^    Sorvz  /lb 


<7. 
/ 


Washington,  D.  C. 
August  9,  1962 


483 


STATEMENT   A 


ASSETS,    LIABILITIES,    AND    FUND   BALANCES 
JUNE    30,    1962  AND    1961 


JUNE    30 


1962 


1961 


ASSETS 
Operating  Funds: 

Cash 

Advances  

Securities  -  Schedule  2  (See  Note)    

Prepaid  insurance      

Restricted  Grants: 

Cash 

Endowment,  General  Reserve,  and  Special  Funds: 

Cash  awaiting  investment 

Advances  -  Building  program    

Investments: 

Savings  account 

Securities  -  Schedule  2  (See  Note)    

Buildings,  Land,  and  Equipment  (At  Cost) 

Total  Assets 

LIABILITIES   AND    FUNDS 
Operating  Funds: 

Income  taxes,  etc.,  withheld 

Operating  Funds  Balance  -  Exhibit  1 

Restricted  Grants  -  Exhibit  2    

Endowment,  General  Reserve,  and  Special  Funds  -  Exhibit  3 

Buildings,  Land,  and  Equipment  Fund  -  Exhibit  4    

Total  Liabilities  and  Funds 


$      512,904.28 

17,800.78 

490,865.29 

38,162.66 

$   1,059,733.01 


$       390,903.30 


$         96,925.10 
5,089.56 

1,477,156.66 
65,973,429.04 

$67,552,600.36 
$  6,044,814.64 
$75,048,051.31 


$  783.40 

1,058,949.61 

$   1,059,733.01 

390,903.30 

67,552,600.36 

6,044,814.64 

$75,048,051.31 


$       477,300.80 

28,920.45 

442,501.55 

51,858.57 

$   1,000,581.37 


$       409,578.63 


$       318,121.47 
266,171.11 


63,036,527.74 


$63,620,820.32 


$   5,807,296.03 


$70,838,276.35 


$  286.30 

1,000,295.07 

$   1,000,581.37 

409,578.63 

63,620,820.32 

5,807,296.03 

$70,838,276.35 


Note:   Approximate  market  value  of  all  securities  at  June  30,  1962  -  $80,867,556 


484 


STATEMENT   B 


SUMMARY    STATEMENT    OF    CHANGES    IN    FUNDS 
FOR   THE    YEAR    ENDED   JUNE    30,    1962 


Endowment,  Buildings, 

Operating            Restricted          General  Reserve,  Land,  and 

Funds                   Grants             and  Special  Funds  Equipment 

(Exhibit  1)            (Exhibit  2)                 (Exhibit  3)  (Exhibit  4) 


Total 


Balance  July  1,  1961     $1,000,295.07        $409,578.63 


$63,620,820.32         $5,807,296.03         $70,837,990.05 


Additions: 

Investment  income 

Realized  capital  gain  (net)  . 

Restricted  grants 

Dormitory 

Sales  of  publications    .... 
Other  income,  gifts  and 

bequests    

Expenditures  capitalized: 

Current  year    

Prior  years 

Gift  capitalized: 

Prior  years 

By  transfer: 

Budget  appropriation  - 
July  1,  1961  to 

June  30,  1962    

American  Geophysical 

Union    

Embryology  Building 

Program 

Harkavy  Fund  -  Income  . 

Harry  Oscar  Wood  Fund  - 

Income    ;  .  . 


10,267.36 
14,127.78 

7,213.42 
22,774.41 

1,362.26 


$202,699.89 


2,848,480.00 

2,250.00  (2,250.00) 

1,000.00 
1,000.00 

24,133.95 


$   3,004,340.32 
4,104,400.40 


15,829.88 


(2,848,480.00) 


(1,000.00) 
(1,000.00) 

(24,133.95) 


$     214,594.66 
35,200.00 

17,585.00 


3,014,607.68 

4,118,528.18 

202,699.89 

7,213.42 

22,774.41 

17,192.14 

214,594.66 
35,200.00 

17,585.00 


$2,932,609.18        $200,449.89 


$  4,249,956.65         $    267,379.66        $  7,650,395.38 


Deductions: 

Expenditures    $2,873,954.64  $219,125.22  $       318,176.61                  ...  $  3,411,256.47 

Disposition  of  equipment  .  .              ...  ...                            ...  $      29,861.05  29,861.05 

$2,873,954.64  $219,125.22  $       318,176.61  $       29,861.05  $   3,441,117.52 

Net  change  during  the  year.  .  .        $       58,654.54  ($   18,675.33)  $  3,931,780.04  $     237,518.61  $  4,209,277.86 


Balance  June  30,  1962 $1,058,949.61         $390,903.30  $67,552,600.36         $6,044,814.64         $75,047,267.91 


485 


EXHIBIT    1 


CHANGES   IN   OPERATING   FUNDS   FOR 
THE    YEAR   ENDED  JUNE    30,    1962 


Balance  July  1,  1961 


$1,000,295.07 


Additions  -  Statement  B: 


Distributions: 

Investment  income     $      10,267.36 

Realized  capital  gains,  net    14,127.78 

Dormitory 7,213.42 

Sales  of  publications 22,774.41 

Other  income 1,362.26 

Transfers: 

General  Reserve  Fund-  Budget  Appropriations  July  1,  1961  to  June  30,  1962.  2,848,480.00 

American  Geophysical  Union 2,250.00 

Embryology  Building  Program     1,000.00 

Harkavy  Fund  -  Income    1,000.00 

Harry  Oscar  Wood  Fund  -  Income    24,133.95 

Total  available  for  expenditure    


2,932,609.18 


$3,932,904.25 


Expenditures: 

Salaries $1,525,418.82 

Equipment 200,018.06 

Laboratory     , 185,553.32 

Buildings  -  fuel,  lights,  etc 111,221.45 

Shop 12,432.96 

Travel 80,754.62 

Dormitory 10,391.84 

Operating 125,127.96 

Financial  Administration  -  investment  and  custody  fees 63,010.71 

Insurance  Premiums 24,131.02 

General  Publications 61,177.80 

Other  publication  expense 25,571.07 

Fellowships 101,646.50 

Awards 24,000.00 

Retirement  Plan  Contributions     203,633.91 

Pensions 46,791.05 

Hospitalization  Plan  and  Collective  Insurance 42,009.81 

Social  security  taxes 31,063.74 

Total  expenditures 


2,873,954.64 


Balance  June  30,  1962 


$1,058,949.61 


486 


CM 

to 

a> 

o 

1 

rt 

CO 

0-1 

0) 

c 

3 

•"9 

^  ^  O  K) 


eMr-'ro^ocoo 


t-  t-i  m  co 


t-H    f  m 


W 
OS 

B 

w 
3 

H 
OS 

o 

In 

co 
H 
Z 

o 


CM 

IO 

to 

© 

eo 

oo 

CO 

CO 

CM 

o 

<J> 

m 

o 

o 

o 

c~ 

O 

O 

T 

CX> 

00 

o 

CM 

co 

CM 

CM 

CO 

00 

in 

■v 

eo 

o 

in 

•-1 

o 

o 

in 

eg 

CO 

in 

CM 

co 

CO 

T 

oo 

CO 

in 

in 

to 

o 

o> 

TT 

o 

o 

co 

CM 

oo 

OS 

•  oa 

t- 

CO 

O 

o 

r- 

oo 

O 

to 

•  o 

CO 

00 

o 

o 

o 

00 

00 

■"»> 

.    CO 

o 

C- 

CM 

00 

■<*< 

.  c- 

CO 

CO 

m 

in 

rH 

lT5 

* 

eo 

<-i 

m 

CM 

1— 1 

to 

CO 

t- 

r-l 

■f 

in 

w 

CO 

H 

J~ 

o 

i— i 

ctt     ^ 

a 

03  £• 

H 

3 

CO 

►o 

W 

K 

to 

O 

o 

n 

o 

o 

oo 

o>      . 

o 

in 

oo 

in 

a> 

c- 

t- 

00 

00 

CO 

<*► 

~ 


O  O  -"f 

in  in  cm 

CM  CM  f 


co"  cm" 


o  ©  o  o 

o  o  o  o 

d  d  d  o 

o  o  o  o 

in  o  o  c- 


8 


to 

O 

o 

o 

CO 

o 

o 

o 

«— 1 

in 

m 

in 

t-H  CM  r-  O  CO 
00  O  O  r}<  to 
CO    CO    T    CM    00 


CO 


8 


& 


o 

<u 

u* 

o 

_ 

o 

a 
a 
a 

r 

ctt 

s 

S5 

o 
in 

o  o 


c  >  >  > 

O  u  u  u 

Tl  cu  CU  CU 

rt  CO  CO  CO 

"§555 


DC  DC  XI 

o  o  o 

3  3  3 

3  3  3 

P-i  Cm  Dh 

CO*  CO*  CO* 


c 

p 

O      M 

M       ? 

■S  s 

cu 


_  o 
o  .2 


o 


>>    C 

A3 

.    CO 


5  3 


Bj  -a 


to  CO  00  c- 

O  Ol  CT>  to  CM 

c-  co  in  in  cm 

t-  CO  TT  Ol  CM 

O  i-H  i-H  i-H  CM 

O  O  O  O  O 

c  c  c  c  c 


c    c 

o   o   6   6   b   b 


*>    co   ,J5  ^5   .^5   .j;   .^ 


a 


o   o 


o   o    o 

tin       tM       Lt, 


CM  O 

in  oo 

Oi  CM 

fc.  Ph 


CT> 
O 
t*i    '^ 

T?  « 

«■  Q 

cq  a 

cu    cu 
u    u 


5  5  5     5 


cu  cu  cu  cu  cu  cu  cu 
cj  o  o  a  o  o  cj 
c    c    c    c 


(h     > 
CU     ti 

E™1    3    cu    cu    cu    cu 


jQ    ctt    ctt    rt    rt 


cucucucucucucuf-, 
_    o    o    o    o    u    u    u 

rtWCOCOCOCOCOCOCQ 


X  X 


cu    cu 

DC  DC 


c 

'Eo 

(Tt 

rt 

rt 

m 

rt 

rt 

rt 

C 

C 

C 

c 

C 

C 

C 

H 

O 

O 

o 

o 

0 

o 

O 

ctt 

a, 

CU 

rt 

rt 

rt 

rt 

rt 

rt 

rt 

a 

2 

Z 

Z, 

Z. 

2 

2 

£ 

S  2  3  2  3  2  S    u      -Scoco 


«       ti 


^     CJ     CJ     o  o 

o  ^C  ^C  ^C  ^C 

_    J2    XI    XI  J2 

■g     D     3     3  3 

<D  a,  Pm  a,  a, 

co'  co" 

p  p" 


p  p 


s  5 


Ch 


.a*  a        cu 


o  a 


o 

o 

E 
m 

o 

rt 

U 

0) 

J 

cu 

u- 

— 1 

c 

ri 

c     «    S 

cu    & 

ja  (^  -2 


u 


— i  ii  co 


cu    u    S 


O 


n   oj 


2  5  B  H 

rt    rt    cu 
2  2  H 


487 


^ 

co 

0) 

Ol 

o 

c 

rt 

o 

•— t 

rt 

CQ 

<d 

3 

<~3 

•* 

•H 

i-i 

o 

in 

CO 

t- 

co 

»-H 

t~- 

CO 

CM 

co 

1ft 

CO    CO 

en  co 


lO^tDOJHWl-NNTt1  00  CO 

o>»-«iftcscooeoot-c-  oo  oo 

UJNMt-NlflMNWN  CO  CM 

offiinNO)inN<oooi  »m  c-^ 
csT  oo"  oo"  c-*  T-*    _ 

-J  t-  tH 


•7     -< 


<a> 

s 

T5 

to 

o 

C 

->-> 

o 

rrt 

5 

to 

3 

J-. 

» 

0) 

5 

O 

CQ 

o 

o 

o 

o 

o 

o 

o 

CO 

o 

CO 

o> 

oo 

OJ 

Ol 

•<* 

o> 

co 

oo 

CO 

?H 

■<* 

i-i 

tH 

oo 

1-4 

2 


li 


*oiH(o«co«oioao 

OCNCOOOi-tCOE-r-im 

cao^j»co        co  co  o>  »-t  o 


CN  «-H  »H 


o 

© 

CO 

Tj< 

*r 

o 

CO 

o 

H 

•<* 

fH 

•^ 

o 

1-1 

^r 

€«• 

8 


oo 

cq 

rH 

CO 

o 

o 

CO 

Tl< 

CO 

CO 

CO 

f 

>-H 

o 

o 

CO 

<&■ 

a  - 


I    2 


u 


"S  "°  2 


c   o, 


S   S 

eg    o 
co    u 

2  & 

w   bo 
-m    C 

bfi  2 
•o  :3 

3     3 

CQ  CQ 


HN<OMHt-00<*flO(0 
t-OJC-OOt-COCO'^'Ot- 

c-mc-oo^roic-oooco 


.   "O 


■=!    u  "S  *  °  CQ  °  S 


.5    o    <u    rt 

cq  u  o  ac 


*■"•   .**   *—•   r*i 
Jh    aJ    cu    c 


488 


ia 


CO 

73 

C 

s 

•a 

J 

•o 

3 

s 

CJ>  i-l 


«-»  CO 


CO 

00  ,| 

CO 

» 

CO* 

oi 

lO 

O 

o 

** 

o 

00 

CO 

00 
CM* 
09- 

p 


«     tH 


£ 


o 

■* 

CD 

CO 

tfi 

■>*< 

00 

r- 1 

o 

00 

CO 

Tj< 

CM 

■<* 

o> 

o 

CO 

w- 

r-l  CM 


2  w 

CD 

j  w 

PQ  H 

-3   o 

5 

CO 


w 


»-H 

CO 

<u 

Oi 

o 

G 

a 

1-H 

fa 

>> 

"3 

"j 

•-«  o 


«-«  os 


Tf      O  <-l 


o 

o 

o 

co 

o 

O 

o 

CO 

o 

o 

in 

oi 

o 

o 

CO 

c- 

CO 

CO 

in 

eo 

^-1 

en 

V- 

c- 

CM 

^H 

T-H 

CO 

eg 

c-  »■« 


o 

eo 

to 

O 

CT> 

CO 

CM 

Oi 

in 

CM 

CM 

C- 

O 

o 

OS 

CO 

IO 

m- 

2  .«»     ^ 


CO    "2 

O  o 


•a  <u 

jj     CO 

3  ft 


£  O 


S  £ 

i)  co 

ri  > 

a,  P" 


^5  "O 

o  c 

o  qT 

e  o 

2  S 

S  -S 


Q 


2  o 

2  Q 

g  . 

S  o 

v  co 


c 
a> 

S 
o, 

3 

a* 

w 


<U     CO 

c  ° 

CJ    T3 


co   3 
§   & 

ft  o  » 


K 


3 
(5*   ? 


a   <u 

«-    c? 
O 


o 


& 
H 


ft  o 


489 


a 

< 


9- 

a> 

c 


•2  «  ^  "S 


c  "2 

o    6 


a>   -r- .    a 
>  3   o 


w    rt 


J3  O    c 


^  73 


O 


Ol 

CM 

CM 

r~ 

rH 

to 

r- 

oo 

in 

CO 

CO 

c- 

CO 

r- 

CO 

CO 

o 

rH 

oo 

o 

CO 

o 

o» 

CM 

Ift 

oj 

CO 

OJ 

CM 

e- 

cj 

CO 

co 

CM 

rH 

r- 

co 

CO 

CM 

-*1 

e- 

CO 

t-« 

in 

T-t 

rH 

CO 

CO 

o 

O 

co 

OJ 

C~ 

in 

o 

in 

o 

oo 

CM 

OS 

CO 

o 

CO 

o 

cj 

CO 

o> 

CO 

CO 

OJ 

o 

o» 

O 

OJ 

f 

r- 

CM 

o 

to 

rH 

rH 

r- 

o 

in 

lO 

o» 

m 

rH 

■<r 

^J< 

CJ 

rH 

CO 

CO 

m 

o 

o 

•<!« 

CO 

o 

lO 

Tj< 

r- 

en 

o> 

m 

rH 

O    CO 

in  co 
in  oo 


o 

CM 
OJ 
CM 


0) 

u 

75  5 

05     -rH 

H    5 


*3< 

CO 

I> 

CO 

o 

co 

i-< 

CM 

rH 

l-< 

rH 

rH 

O 

o 

co 

m 

CO 

■"f 

CM 

m 

CM 

CO 

o 

HO 

•<!> 

l- 

o 

OJ 

■"*< 

CM 

OJ 

Oi 

CO 

in 

rH 

CO 

CO 

CO 

in 

CM 

OJ 

CO 

CO 

00 

^ 

CO 

CO 

T 

CO 

to 

rH 

T 

CO 

CO    00  CO 

in  co  co 

in  co  to 

rH    CO  CO 

to  in  o 

rH  CM 


rH  CJ    CO 


I5 

u 

Q. 
< 


CO  CO 

in  in 

cj  o  © 
o  oo  oo 

CO  CM 
ID  rH 

in 
in 

CO 

CO  CO  CM  o 
in  c-  co  O 

oo  *r 
oo  co 

rH  OJ 
00  CM 

CO 

rH 

OJ 

in 

CM 

CO  ^ 

co  o> 

rH  r-  to 

to  ^<  CO 

cj  in  in 

CO  rH 
t-  00 

00  C- 

rH 
CO 
CM 

in 

oo  to  CM  o 

CO  00  CO  o 

tji  cm  t-  in 

cj  TJ< 

oj  cj 
to  co 

OJ  OJ 
O  O 
O  CM 

tn 

to 

o 

OJ 

00  CO 
CM  CM 
rH  CM 

CM  CM  CO 
rH  CM  CO 

•*r  co 

rH  CO 
•<J<  CO 

in  co 

•<* 

o 

co 
in 

C-  CM  CM  C- 
Tji  O  to  CM 

rH  rH  CM  CM 

OJ  rH 

(O  CO 

rH 

CM  rH 

■*  CO 

OJ 

to 

CO 
00 
CM 

CM 
CO 
OJ 

V* 

CM 

CO 

CO 

-rp    © 

to  ^r 
in  cm 

to  oo 

CM  rH 

CO 

to 

CO  -*• 
tO  rH 

00 
CO 

rH  eo 

rH  OJ 
CO  CM 

OJ 

o 

o 
o 

OJ 

o 

^  o 
Tf  in 

rH  O 

CO  CO 

•  o  c- 

.  CO  00 

O  OJ 

oj  in 

"»«  rH 

o 

CM 

o 

TJ<  ID 
CO  f 
OJ  rH 

in 

♦  CO   • 

.to 

O  TT 
O  OJ 
CO  O 

OJ  OJ 

o  o 

OJ  t- 

OJ 
in 

o 

CO 
CO 

OJ 

rH 

•  co  m 

rH  E- 
•f  CO 

CM 

00 

oj  m 
to  o 

CO 

♦  to 
o 

rH 

m  oj 

rH 

rH  CM 

CM 
CO 

OJ 
CM 

CM 

m 

(to 

e> 

>m_. 

v> 

s_ 

€» 

W 

i 

s 

u    w 

3 

a,  c 

.5 

o   o 

c 

S*  rt 

o 

Q. 

1 

O 

< 

CM 

<J 

u 

(U  OJ 
■O  rH 
C    - 

a 

o 

c 

o 

a>  o 

on 

rt 

«*-i 

0J 

u  n 

T> 

S-. 

?i  c 

3 

D, 

>>    3 

m 

•-3 
03 

C  rH 

o  to 

TJ 

X3  o> 

c 

rt  "-• 

ft 

a» 

2  >> 

P 

a  3 
< 

rH        .CO 


00 

to 

rH  O 

OJ 

o 

OJ 

o 

m 
■n 

•   •  •>*  O   •      .   •      .   • 

.  .  t"  o  .     .  .     .  . 
t-  o 

350 

o 
m 

CO 

t- 

CM  rH 
CM 

CO 

CO 

«r> 

w 

«^ 

o 
o 

o 
o 

O 
o 

o 
o 

o 
o 

o 
o 

o 
o 

o 

o 

o 
o 

o 
o 

o 
o 

o 
o 

o 
o 

© 
o 

© 
© 

o 
o 

o 
o 

o 
o 
o 

o 
o 

CO 

o 
o 
o 

o 
o 

o 
o 

OJ 

o 
o 

CO 

o 

CO 

o 

00 

o 

o 
o 

to 

o 
o 
o 

o 
o 
o 

o 
o 

o 
o 
m 

o 
o 
o 

o 
o 
o 

o 
o 

rH 

o 
o 
in 

CO 

cn 

CM 

CO 
CO 
CO 

in 

CO 

o 

CO 

CM 

in 

CM 

OJ 

to 

rH 

to 

CD 
CO 
CO 

o 

rH 

CM 

CM 
rH 
CM 

to 

rH 

00 
CM 

o 

CO 
CM 

ifr 

te- 

oj  co  c-  n<  o 
o  m  to  cm  co 

CM  r*< 
■«3<  OJ 

OJ 

rH 
CM 

CM  CM  f  O 
C-  CO  to  O 

C-  rH 

■"*•  to 

CO 

rH 
OJ 

o 

Ol  TJI  C-  T  rH 

»  *  *  if  O 
K5  Tf  t>  rl  t 

CO  rH 
CO  1t< 
C-  f 

in 
m 
in 

O 
CM 
CM 

Tf  CM  CO  O 

CO  rH  OJ  O 

m  in  co  o 

OJ  OJ 

OJ  OJ      .   . 

CO  c- 

00 

to 

CO 
CM 
00 

in 
o> 

CM 

CO  OJ  CO  CO  OJ 

rH  rH 

Cfi- 

OJ  rr 
CO  f 

to 

to 

W3- 

to 

eo  oj  co  in 

rH  tO  CO  rH 
rH      r-i 

00  TT      .   . 
"»•  CO 
r-l 

00 

to 

CM 
CO 

o 
o 

1-i 

a  .2 

O  CQ  o 

o  c  S 

QJ  Cm  O 


C     C 

B   B 


a  a 
Q  Q 


3  'S 


w     q, 


"3  o 


2      « 


S    -M     S    ^ 


S  2  ^  5 

S  «  ^  S 

Q   &  I  a 

a>  o  5> 

O  S  Q 


c  is 


5       S 


Q  S 

3  o 


c  .2 


to"    • 

a  w 
!c  c 
»  2 

11 

OJ  >H 

-  o 

o    — 

c3 


rt    rt 


m     cj    .~   to    to 
P.  -ri    o  r  r*  *. 


U 


2  Ss 
pi    a  3 


.5 
S 

<  a  o 


QJ      01 

c    c 
m    o> 


c-i  o> 

3      >»  rH  c 


o 

u 

a 

a 

"^ 

c 

.2 

c 
i3 

(-. 

43 

rH 

o 

3 

ei 

)-. 

3 

x; 

Ol 

0) 

rH 

3 

3 

U 

H 

01 
0) 

to 

s 

0) 

s 

< 

c 
o 
to 
c 

5 
a 

«5 

C 

CO 

0) 

0) 

d> 

0) 

o 

o 

CO 

tf 

tf 

OS 

a 

X 

o 


490 


SCHEDULE   2 


SECURITIES,   JUNE   30,   1962 
AND  INCOME   RECEIVED  DURING   THE   YEAR 


Bonds: 

United  States  Government 
Foreign  and  International  Bank. 

Public  Utility 

Communication 

Railroad    

Railroad  Equipment  Trust 
Industrial  and  Miscellaneous  .  . 

Total  Bonds . 


Per  Cent  of  Total 

Investments 

r 

Approximate 

1 

Approximate 

Market 

Book 

Market 

Income 

Book  Value 

Value 

Value 

Value 

Received 

$  3,686,131.68 

$  3,720,192 

5.55 

4.60 

$     110,457.88 

2,641,408.22 

2,571,828 

3.97 

3.18 

112,610.02 

10,122,191.59 

9,674,380 

15.23 

11.96 

381,809.58 

3,907,499.98 

3,597,188 

5.88 

4.45 

145,665.40 

368,218.96 

317,610 

.55 

.39 

16,088.97 

76,685.33 

79,485 

.12 

.10 

3,318.75 

18,174,738.94 

18,048,824 

27.34 

22.32 

778,336.78 

$38,976,874.70 

$38,009,507 

58.64 

47.00 

$1,548,287.38  (ai 

Stocks: 

Preferred    

Common 

Total  Stocks 
Total  . 


$  1,619,411.87  $  1,506,863  2.44  1.87  $      71,650.00 

25,868,007.76  41,351,186  38.92  51.13  1,372,513.64 

$27,487,419.63  $42,858,049  41.36  53.00  $1,444,163.64 

$66,464,294.33  $80,867,556  100.00  100.00  $2,992,451.02 


(a)  After  deducting  bond  premium  amortization  of  $21,521.05 


491 


SCHEDULE    OF   SECURITIES 


Principal 
Amount 


$     100,000 

160,000 

1,105,000 

1,470,000 

403,000 

450,000 

$3,688,000 


$  250,000 
500,000 
150,000 
137,000 
250,000 
125,000 
125,000 
250,000 
150,000 
200,000 

500,000 
$2,637,000 


250,000 
125,000 
250,000 
237,000 
300,000 
300,000 
300,000 
300,000 
150,000 
203,000 
300,000 
500,000 
200,000 
250,000 
400,000 
100,000 
200,000 
300,000 
250,000 
87,000 
50,000 
500,000 
207,000 
250,000 
200,000 
400,000 
500,000 
250,000 
250,000 
200,000 
191,000 
500,000 
265,000 


Description 


Maturity        Book  Value 


Approximate 
Market 
Value 


United  States  Government  Bonds 

United  States  of  America,  Treasury  Bills    7-26-62 

United  States  of  America,  Treasury  Bills    10-25-62 

United  States  of  America,  Ctf .  of  Ind.,  3*s 1963 

United  States  of  America,  Treasury  Notes  3f  s    .. 1964 

United  States  of  America,  Treasury  Notes  4  Is 1963 

United  States  of  America,  Treasury  Notes  5s 1964 

Total  United  States  Government 


Foreign  and  International  Bank  Bonds 

Aluminum  Co.  of  Canada,  Ltd.,  S.  F.  Deb.  3|s 

Aluminum  Co.  of  Canada,  Ltd.,  S.  F.  Deb.  4^s 

Australia  (Commonwealth  of)  4|s 

Australia  (Commonwealth  of)  5s 

British  Columbia  Power  Commission,  S.  F.  Deb.  Series  "L"4fs 

Intl.  Bank  for  Reconstruction  &  Development,  3s 

Intl.  Bank  for  Reconstruction  &  Development,  3|s 

Intl.  Bank  for  Reconstruction  &  Development,  4|s 

Noranda  Mines  Ltd.,  S.  F.  Deb.  4f  s 

Shawinigan  Water  &  Power  Co.,  1st  Mtg.  &  Collat.  Tr.  S.  F. 

Series  "M"  3s 

Toronto  (Municipality  of  Metropolitan),  S.  F.  Deb.  5s 

Total  Foreign  and  International  Bank 

Public  Utility  Bonds 

California  Oregon  Power  Co.,  1st  Mtg.  3ls 

Columbia  Gas  System,  Inc.,  Series  "B"  3s 

Columbia  Gas  System,  Inc.,  Series  "F"  3ls 

Columbus  &  Southern  Ohio  Electric  Co.,  1st  Mtg.  3  is 

Commonwealth  Edison  Co.,  1st  Mtg.  Series  "R"  3|s 

Consolidated  Edison  Co.  of  N.Y.,  1st  &  Ref.  Mtg.  Series  "L"3fs 
Consolidated  Edison  Co.  of  N.Y.,  1st  &  Ref.  Mtg.  Series  "N"  5s  . 

Consolidated  Natural  Gas  Co.,  Deb.  2fs 

Consumers  Power  Co.,  1st  Mtg.  4s 

Consumers  Power  Co.,  1st  Mtg.  4|s 

Florida  Power  Corporation,  1st  Mtg.  3|s 

Illinois  Power  Co.,  1st  Mtg.  3|s 

Minnesota  Power  &  Light  Co.,  1st  Mtg.  3|s 

Niagara  Mohawk  Power  Corp.,  Gen.  Mtg.  3§s 

Niagara  Mohawk  Power  Corp.,  Gen.  Mtg.  4ls 

Ohio  Power  Co.,  1st  Mtg.  3{s 

Pacific  Gas  &  Electric  Co.,  1st  &  Ref.  Mtg.  Series  "X"  3|s  .  . 
Pacific  Gas  &  Electric  Co.,  1st  &  Ref.  Mtg.  Series  "Y"  3|s  .  . 
Pacific  Gas  &  Electric  Co.,  1st  &  Ref.  Mtg.  Series  "BB"  5s  .  . 

Panhandle  Eastern  Pipe  Line  Co.,  S.  F.  Deb.  3js 

Philadelphia  Electric  Co.,  1st  &  Ref.  Mtg.  2js 

Philadelphia  Electric  Co.,  1st  &  Ref.  Mtg.  4§s 

Philadelphia  Electric  Power  Co.,  1st  Mtg.  2fs  Guar 

Potomac  Electric  Power  Co.,  Deb.  4§s 

Public  Service  Co.  of  Indiana,  1st  Mtg.  Series  "F"  3}s 

Public  Service  Co.  of  Indiana,  1st  Mtg.  Series  "L"  4^s 

Public  Service  Electric  &  Gas  Co.,  1st  &  Ref.  Mtg.  4^s 

Southern  California  Edison  Co.,  1st  &  Ref.  Mtg.  Series  "G"3|s 
Southern  California  Edison  Co.,  1st  &  Ref.  Mtg.  Series  "H"4is 
Southern  California  Edison  Co.,  1st  &  Ref.  Mtg.  Series  "J"  4ls 
Tennessee  Gas  Transmission  Co.,  1st  Mtg.  Pipe  Line  3s    .... 

Tennessee  Gas  Transmission  Co.,  5s 

Tennessee  Gas  Transmission  Co.,  1st  Mtg.  Pipe  Line  5{s .... 


$ 


99,770.00 

157,705.60 

1,105,000.00 

1,471,370.80 

402,285.28 

450,000.00 


$ 


99,785 

158,479 

1,105,691 

1,480,106 

412,068 

464,063 


$3,686,131.68         $3,720,192 


1970 

$  251,561.49 

$  247,813 

1980 

508,695.20 

500,000 

1971 

147,750.00 

145,500 

1972 

137,000.00 

137,514 

1987 

245,000.00 

240,000 

1976 

125,000.00 

108,438 

1975 

123,125.00 

115,938 

1977 

250,000.00 

252,500 

1968 

151,399.03 

133,875 

1971 

203,240.00 

161,500 

1979 

498,637.50 

528,750 

$2,641,408.22        $2,571,828 


1986 

$  252,622.54 

$  229,688 

1975 

126,745.43 

108,750 

1981 

245,937.50 

240,313 

1970 

242,328.42 

218,633 

1986 

300,612.54 

264,375 

1986 

303,222.91 

266,250 

1987 

301,969.20 

315,375 

1968 

300,231.22 

283,125 

1986 

151,194.42 

143,438 

1987 

204,090.77 

210,866 

1986 

301,796.95 

280,875 

1986 

497,937.50 

458,750 

1975 

202,126.87 

174,750 

1986 

252,704.50 

224,688 

1987 

402,938.35 

418,000 

1968 

101,500.00 

96,125 

1984 

201,283.28 

164,000 

1987 

305,584.79 

257,625 

1989 

251,660.34 

263,125 

1973 

87,677.86 

80,258 

1978 

49,687.50 

41,813 

1987 

500,000.00 

514,688 

1975 

209,236.62 

174,915 

1982 

255,185.60 

254,375 

1975 

202,220.21 

176,250 

1987 

400,000.00 

413,000 

1987 

503,995.49 

522,500 

1981 

247,765.00 

223,750 

1982 

251,484.40 

247,813 

1982 

201,763.11 

208,500 

1969 

192,856.18 

178,585 

1982 

505,000.00 

502,188 

1977 

265,000.00 

274,275 

492 


Principal 
Amount 


$  500,000 
265,000 
235,000 
300,000 

$10,065,000 


$  150,000 
350,000 
800,000 
500,000 
400,000 
200,000 
100,000 
200,000 
300,000 
250,000 
300,000 
300,000 

$  3,850,000 


$       100,000 
267,000 

$      367,000 


SCHEDULE    OF    SECURITIES-Continued 


Description 


Public  Utility  Bonds— Concluded 

Union  Electric  Co.,  1st  Mtg.  3f  s 

United  Gas  Corp.,  1st  Mtg.  &  Coll.  Tr.  2f s 

Virginia  Electric  &  Power  Co.,  1st  &  Ref.  Mtg. Series  "M"4is 
Washington  Water  Power  Co.,  1st  Mtg.  4|s 

Total  Public  Utility 

Communication  Bonds 

American  Telephone  &  Telegraph  Company,  Deb.  2|s    .  .  . 

American  Telephone  &  Telegraph  Company,  3js 

American  Telephone  &  Telegraph  Company,  Deb.  3 is .... 
American  Telephone  &  Telegraph  Company,  Deb.  4|s .... 
Illinois  Bell  Telephone  Co.,  1st  Mtg.  Series  "E"  4{s  .... 
Mountain  States  Telephone  &  Telegraph  Co. ,  Deb.  3|s    .  .  . 

New  York  Telephone  Co.,  Ref.  Mtg.  Series  "E"  3|s 

Pacific  Telephone  &  Telegraph  Co.,  Deb.  3£s 

Pacific  Telephone  &  Telegraph  Co.,  Deb.  4|s 

Southern  Bell  Telephone  &  Telegraph  Co.,  Deb.  4s 

Southern  Bell  Telephone  &  Telegraph  Co.,  Deb.  5s 

Southwestern  Bell  Telephone  Co.,  Deb.  3|s    

Total  Communication 

Railroad  Bonds 

Chesapeake  &  Ohio  Railway  Co.,  Gen.  Mtg.  4|s 

Fort  Worth  &  Denver  Railway  Company,  1st  Mtg.  4|s  Guar 

Total  Railroad    


Maturity 

Book  Value 

$   500,089.75 
265,000.00 
238,742.34 
300,000.00 

$10,122,191.59 

Approximate 
Market 
Value 

.  .  .     1986 

$  458,750 
243,138 

...      1967 

4£s      1986 
.  .  .     1987 

228,831 
312  000 

$9,674,380 

Railroad  Equipment  Trust  Bonds 

$        50,000  Chicago,  Burlington  &  Quincy  Railroad  Co.,  Eq.  Tr.  2*s  Guar. . 

30,000  Pennsylvania  Railroad  Company,  Eq.  Tr.  Series  "S"  2fs  Guar. 

$        80,000  Total  Railroad  Equipment  Trust 

Industrial  and  Miscellaneous  Bonds 

$      250,000     Aluminum  Co.  of  America,  S.  F.  Deb.  4{s 

234,000     Bristol-Myers  Co.,  Deb.  3s 

550,000     C.  I.  T.  Financial  Corp.,  Deb.  4|s 

400,000     Commercial  Credit  Co.,  Notes  3§s 

400,000     Continental  Oil  Company  (Del.),  S.  F.  Deb.  3s 

105,000     Corn  Products  Co.,  Sub.  Deb.  4|s 

500,000     Crown  Zellerbach  Corp.,  Prom.  Note  4|s 

400,000     Federal  Farm  Loan  Consolidated,  4|s 

200,000     Federal  Farm  Loan  Consolidated,  4|s 

285,000     Federal  Farm  Loan  Consolidated,  4|s 

1,025,000     Federal  National  Mortgage  Association,  4|s 

465,000     Federal  National  Mortgage  Association,  4|s 

400,000     Federal  National  Mortgage  Association,  4§s 

500,000     Federal  National  Mortgage  Association,  5|s 

500,000     Food  Machinery  &  Chemical  Corp.,  S.  F.  Deb.  3.80s 

364,000     Four  Corners  Pipe  Line  Company,  Sec.  Note  5s 

500,000     General  Electric  Credit  Corp.  (N.  Y.)  Prom.  Note  5s 

200,000     General  Motors  Acceptance  Corp.,  Deb.  3^s 

480,000     General  Motors  Acceptance  Corp.,  Deb.  4s 

200,000     General  Motors  Acceptance  Corp.,  Deb.  5s 

200,000     General  Motors  Acceptance  Corp.,  Deb.  5s 

150,000     General  Portland  Cement  Co.,  Conv.  Sub.  Deb.  5s 

275,000     Goodrich  (B.  F.)  Company,  1st  Mtg.  2|s 


1975 

$  150,956.25 

$  127,875 

1984 

358,963.68 

300,125 

1990 

817,827.48 

736,000 

1985 

504,805.20 

502,500 

1988 

404,471.60 

397,000 

1978 

200,770.00 

169,250 

1978 

100,739.59 

85,375 

1978 

202,428.04 

174,250 

1988 

305,984.22 

300,750 

1983 

251,044.30 

238,750 

1986 

305,509.62 

316,500 

1983 

304,000.00 

248,813 

$3,907,499.98 

$3,597,188 

1992 

$   99,464.29 

$   96,000 

1982 

268,754.67 

221,610 

$  368,218.96 

$  317,610 

1963 

$   47,839.05 

$   49,485 

1962 

28,846.28 

30,000 

$   76,685.33 

$   79,485 

1982 

$  250,000.00 

$  250,000 

1968 

234,310.66 

226,103 

1970 

536,937.50 

570,625 

1976 

406,626.56 

364,000 

1984 

403,707.00 

336,000 

1983 

109,557.13 

109,200 

1981 

500,000.00 

476,250 

1969 

394,000.00 

406,000 

1964 

201,714.29 

203,375 

1970 

284,330.96 

290,700 

1970 

1,015,070.31 

1,014,750 

1971 

466,003.98 

460,350 

1970 

394,500.00 

409,000 

1972 

498,125.00 

532,500 

1981 

500,000.00 

465,000 

1982 

364,000.00 

383,110 

1975 

500,000.00 

510,000 

1972 

202,454.62 

184,000 

1979 

435,037.50 

450,000 

1977 

195,000.00 

210,000 

1981 

199,000.00 

211,000 

1977 

153,375.09 

163,500 

1965 

275,144.85 

268,125 

493 


SCHEDULE    OF   SECURITIES- Continued 


Principal 
Amount 


$      662,121.59 
553,035.63 
289,007.18 
352,773.34 
516,359.57 
400,000 
300,000 
236,000 
200,000 
195,000 
488,000 
150,000 
100,000 
250,000 
300,000 
525,000 
300,000 
300,000 
215,000 
300,000 
250,000 
905,926.01 
451,000 
346,000 
400,000 
250,000 
500,000 

$18,318,223.32 

$39,005,223.32 


Description 


Maturity 


Industrial  and  Miscellaneous  Bonds— Concluded 

Instlcorp,  Inc.,  Collat.  Tr.  Note  Series  A- 16 1991 

Instlcorp,  Inc.,  Collat.  Tr.  Note  Series  A- 19 1991 

Instlcorp,  Inc.,  Collat.  Tr.  Note  Series  A-21 1991 

Instlcorp,  Inc.,  Collat.  Tr.  Note  Series  A-23 1991 

Instlcorp,  Inc.,  Collat.  Tr.  Note  Series  A-36 1992 

Intl.  Harvester  Credit  Corp.,  Deb.  4|s 1979 

Kaiser  Aluminum  &  Chemical  Corp.,  1st  Mtg.  5|s 1987 

Lorillard  (P.)  Company,  Deb.  3s    1963 

Montgomery  Ward  Credit  Corp.,  Deb.  4£s 1980 

National  Dairy  Products  Corp.,  Deb.  2|s 1970 

Phillips  Petroleum  Company,  S.  F.  Deb.  2|s 1964 

Quaker  Oats  Co.,  Deb.  2§s    1964 

Riegel  Paper  Corp.,  S.  F.  Deb.  3|s     1981 

Scovill  Mfg.  Co.,  Deb.  4|s 1982 

Seagram  (Joseph  E.)  &  Sons,  Incorporated,  Deb.  2^s 1966 

Sears  Roebuck  Acceptance  Corp.,  Sub.  Deb.  4|s 1977 

Sinclair  Oil  Corporation,  Conv.  Sub.  Deb.  4fs 1986 

Superior  Oil  Company,  The  (California),  Deb.  3|s 1981 

Talcott  (James),  Inc.,  Senior  Note  5|s 1966-80 

Texas  Corporation,  Deb.  3s 1965 

Tidewater  Oil  Company,  S.  F.  Deb.  3§s 1986 

Trailer  Train  Company,  4js 1976 

Tremarco  Corporation,  1st  Mtg.  Series  "E"  5s 1983 

Union  Oil  Co.  of  California,  Deb.  2f  s 1970 

Westinghouse  Electric  Corp.,  Deb.  2§s 1971 

Whirlpool  Corporation,  S.  F.  Deb.  3|s 1980 

Woolworth  (F.  W.)  Co.,  Prom.  Note  5s 1982 

Total  Industrial  and  Miscellaneous 

Bonds  —  Funds  Invested 


Approximate 

Market 

Book  Value 

Value 

$  640,004.46 

$  660,466 

534,736.60 

550,270 

278,891.94 

289,007 

347,058.58 

351,891 

493,984.64 

496,996 

398,000.00 

415,000 

300,000.00 

315,000 

236,000.00 

233,640 

199,000.00 

207,000 

196,043.94 

176,719 

488,488.91 

480,680 

148,922.50 

145,688 

100,000.00 

95,000 

246,250.00 

247,500 

298,500.00 

288,000 

511,505.00 

525,000 

314,788.96 

295,875 

300,000.00 

280,500 

212,850.00 

224,944 

302,071.76 

294,000 

250,000.00 

212,500 

905,926.01 

917,250 

451,000.00 

466,785 

350,074.87 

311,400 

401,745.32 

346,000 

250,000.00 

215,625 

500,000.00 

512,500 

$18,174,738.94 

$18,048,824 

$38,976,874.70 

$38,009,507 

Number 

of 
Shares 


1,500 
1,500 
3,800 
1,900 
800 
2,000 
1,300 
3,100 

15,900 


5,000 
28,100 
14,350 
19,373 
11,500 

6,200 
14,100 
10,000 

6,000 
12,000 

2.872 


Preferred  Stocks 

Appalachian  Power  Co.,  4|%  Cum.  Pref 

Bethlehem  Steel  Corporation,  7%  Cum.  Pref 

Carrier  Corporation,  4^%  Cum.  Pref 

Consolidated  Edison  Co.  of  N.  Y.,  $5.00  Cum.  Pref 

National  Distillers  and  Chemical  Corp.,  4£%  Cum.  Conv.  Pref. 

Niagara  Mohawk  Power  Corp.,  3.60%  Cum.  Pref 

Ohio  Power  Co.,  A\%  Cum.  Pref 

United  States  Steel  Corporation,  7%  Cum.  Pref 

Total  Preferred  Stocks 

Common  Stocks 

Aetna  Casualty  &  Surety  Co 

Aluminium  Ltd 

American  Electric  Power  Co.,  Inc 

American  Telephone  &  Telegraph  Company 

Arizona  Public  Service  Co 

Armco  Steel  Corporation    

Armstrong  Cork  Company 

Atchison,  Topeka  &  Santa  Fe  Railway  Co 

Campbell  Soup  Company 

Caterpillar  Tractor  Co 

Chase  Manhattan  Bank,  N.  Y 


159,000.00 

$  142,500 

183,637.50 

211,875 

197,931.28 

180,500 

202,815.50 

198,550 

80,000.00 

69,000 

207,990.00 

146,000 

144,630.02 

123,663 

443,407.57 

434,775 

$1,619,411.87    $1,506,863 


192,782.20 
791,574.11 
186,637.38 
963,574.35 
432,901.08 
240,383.71 
231,516.80 
166,256.21 
362,612.33 
96,913.60 
81,118.78 


306,250 
540,925 
808,981 
2,009,949 
296,125 
305,350 
697,950 
228,750 
534,000 
382,500 
198,886 


494 


SCHEDULE  OF  SECURITIES—  Concluded 


Number 

of 
Shares 


4 

800 

9 

800 

18 

000 

2 

500 

1 

500 

11 

294 

9 

500 

11 

474 

7 

257 

12 

000 

14 

700 

3 

700 

31 

100 

28 

000 

2 

700 

16 

500 

21 

712 

18 

606 

4 

340 

14 

000 

22 

000 

4 

000 

9 

500 

13 

800 

8 

500 

6 

800 

1 

025 

4 

600 

13 

000 

5 

000 

5 

800 

10 

100 

2 

100 

8 

600 

9 

000 

17 

000 

1 

800 

8 

000 

5 

000 

27 

000 

10 

098 

5 

500 

14 

100 

32 

400 

12 

000 

10 

200 

18 

500 

15 

500 

29 

704 

500 

7 

600 

6 

000 

30 

750 

14 

200 

7 

000 

13 

800 

783 

455 

Description 
Common  Stocks— Concluded 

Christiana  Securities  Co 

Coca  Cola  Company  (The)  . 

Continental  Oil  Company,  (Del.) , 

Corning  Glass  Works „ 

E.  I.  du  Pont  de  Nemours  &  Co 

Eastman  Kodak  Company 

Falconbridge  Nickel  Mines,  Ltd 

Farbenfabriken  Bayer  AG-ADR  Par  50  Deutschemark 

First  National  City  Bank  of  N.  Y 

Florida  Power  &  Light  Co 

Ford  Motor  Company 

General  American  Transportation  Corp 

General  Electric  Company    

General  Motors  Corporation 

General  Reinsurance  Corp 

Gillette  Company 

Goodyear  Tire  &  Rubber  Company 

Gulf  Oil  Corporation    

Home  Insurance  Co.  of  N.  Y. 

Household  Finance  Corp 

Illinois  Power  Co 

Insurance  Co.  of  North  America 

International  Business  Machines  Corp 

International  Nickel  Co.  of  Canada,  Ltd 

Kellogg  Company 

Kennecott  Copper  Corporation 

Litton  Industries,  Inc 

Marquette  Cement  Manufacturing  Co 

Mead  Corporation 

Merck  &  Co 

National  Cash  Register  Co 

Niagara  Mohawk  Power  Corp 

Norfolk  &  Western  Railway  Company    

North  American  Aviation,  Inc 

Northwest  Bancorporation 

Ohio  Edison  Co 

Otis  Elevator  Company 

Panhandle  Eastern  Pipe  Line  Co 

Philip  Morris  Incorporated 

Philips'  Incandescent  Lamp  Works,  Ltd.  (N.V.Philips'  Gloeilampen- 

fabrieken),  Par  25  Florin 

Pittsburgh  Plate  Glass  Co 

Republic  Natural  Gas  Co 

Revere  Copper  &  Brass,  Inc 

Royal  Dutch  Petroleum  Co 

Scott  Paper  Company 

Shell  Oil  Company 

Standard  Oil  Co.  (New  Jersey) 

Stevens  (J.  P.)  &  Co 

Texaco,  Inc. 

Texas  Instruments  Inc 

Texas  Utilities  Co 

Travelers  Insurance  Co 

Unilever  N.  V.,  Par  20  Florin 

U.  S.  Plywood  Corp 

United  States  Steel  Corporation 

Virginia  Electric  &  Power  Co 

Total  Common  Stocks 

Common  and  Preferred  Stocks  -  Funds  Invested 

Aggregate  Investments  (Bonds  and  Stocks) 


Approximate 

Market 

Book  Value 

Value 

$   356,143.00 

$   787,200 

628 

984.09 

749 

700 

176 

753.10 

864 

000 

59 

631.83 

290 

625 

61 

220.33 

295 

781 

134 

396.92 

1,007 

990 

550 

837.50 

424 

820 

655 

244.34 

622 

465 

348 

095.50 

605 

052 

148 

863.69 

636 

000 

849 

086.16 

1,131 

900 

153 

433.14 

201 

650 

647 

197.55 

1,850 

450 

968 

006.87 

1,354 

500 

220 

697.46 

425 

250 

401 

418.90 

577 

500 

480 

153.74 

681 

214 

95 

259.49 

665 

165 

273 

523.65 

209 

134 

659 

231.85 

507 

,500 

589 

126.02 

731 

,500 

42 

590.65 

297 

,000 

148 

783.26 

3,222 

,875 

379 

279.52 

791 

775 

332 

482.68 

461 

,125 

446 

112.26 

481 

950 

122 

512.02 

94 

,300 

190 

140.68 

158 

,700 

614 

427.23 

446 

,875 

93 

798.41 

325 

,000 

530 

904.80 

437 

,900 

458 

553.32 

405 

,263 

217 

512.31 

183 

,750 

559 

647.34 

491 

,275 

231 

895.01 

333 

,000 

587 

855.31 

675 

,750 

97 

869.13 

93 

,150 

431 

553.54 

417 

000 

493 

240.88 

364 

,375 

784 

874.79 

1,213 

313 

713 

845.30 

489 

753 

13 

406 

665 

214.01 

481 

163 

1,190 

625.97 

1,162 

350 

53 

041.98 

346 

500 

170 

667.87 

321 

300 

392 

518.12 

925 

000 

476 

576.67 

476 

625 

297 

924.93 

1,444 

357 

63 

483.24 

31 

063 

163 

042.97 

304 

000 

452 

662.70 

789 

000 

1,162 

805.47 

1,149 

281 

697 

928.16 

610 

600 

160 

003.06 

308 

875 

240 

058.49 

674 

475 

$25,868, 

007.76 

$41,351 

186 

$27,487 

419.63 

$42,858 

049 

$66,464, 

294.33 

$80,867, 

556 

495 


SUMMARY   OF   SECURITY   TRANSACTIONS   JULY    1,    1961    TO   JUNE    30,    1962 

Cash  awaiting  investment  -  July  1,  1961    $      318,121.47 

Sales  and  Redemptions 

Gain                     Loss  Book  Value 

Bonds $      12,349.41              ...  $  5,517,353.28 

Common  Stocks 4,089,270.39             ....  4,683,491.66 

Sale  of  Stock  Rights 16,908.38 

$4,118,528.18              .  .  .  $10,200,844.94 

Net  Gain  -  Statement  B ...  $4,118,528.18  4,118,528.18 

$4,118,528.18  $4,118,528.18 

Total  Sales  and  Redemptions $14,319,373.12 

Income  applied  to  amortization  of  bond  premium 21,521.05 

Gifts  and  bequests 700.00 

$14,659,715.64 
Cash  transferred  from  investment 1,355,159.51 

Total $13,304,556.13 

Acquisitions 

Bonds $  6,121,410.73 

Common  Stocks 7,086,220.30 

Total  Acquisitions $13,207,631.03 

Cash  awaiting  investment  -  June  30,  1962    $        96,925.10 


496 


Abstract  of  Minutes 

of  the  Sixty-Fourth  Meeting  of  the  Board  of  Trustees 


The  annual  meeting  of  the  Board  of  Trustees  was  held  in  the  new  laboratory  building  of 
the  Department  of  Embryology,  Baltimore,  Maryland,  on  Friday,  May  11,  1962.  Mr. 
Henry,  Chairman  of  the  Board,  presided. 

The  following  Trustees  were  in  attendance:  Amory  H.  Bradford,  Omar  N.  Bradley, 
Vannevar  Bush,  Caryl  P.  Haskins,  Barklie  McKee  Henry,  Alfred  L.  Loomis,  Keith  S. 
McHugh,  Henry  S.  Morgan,  William  I.  Myers,  Garrison  Norton  (Secretary  pro  tern), 
Richard  S.  Perkins,  Elihu  Root,  Jr.,  Charles  P.  Taft,  James  N.  White,  and  Robert  E.  Wilson. 

The  minutes  of  the  Sixty-Third  Meeting  were  approved. 

With  unanimous  consent,  Crawford  H.  Greenewalt  and  Juan  T.  Trippe  were  reelected 
members  of  the  Board  of  Trustees. 

The  Chairman  notified  the  Trustees  of  the  death  of  Robert  Woods  Bliss.  Mr.  Root  spoke 
of  the  Trustees'  high  esteem  for  Mr.  Bliss  and  of  his  many  contributions  to  the  Institution. 
Mr.  Root  proposed  the  following  resolutions,  which  the  Trustees  adopted  unanimously: 

Be  It  Resolved,  That  the  Trustees  of  the  Carnegie  Institution  of  Washington  desire  to  record 
their  deep  sense  of  loss  at  the  death  of  their  distinguished  fellow  member,  Robert  Woods  Bliss. 

And  Be  It  Further  Resolved,  That  these  resolutions  be  entered  on  the  minutes  of  the  Insti- 
tution and  a  copy  be  sent  to  Mrs.  Bliss. 

The  annual  report  of  the  President  was  accepted. 

The  reports  of  the  Executive  Committee,  the  Finance  Committee,  the  Retirement  Com- 
mittee, the  Auditor,  and  the  Auditing  Committee  were  accepted. 

To  provide  for  operation  of  the  Institution  for  the  fiscal  year  beginning  July  1,  1962,  and 
upon  recommendation  of  the  Executive  Committee,  the  sum  of  $3,116,993  was  appropriated 
from  the  General  Reserve  Fund. 

Carl  J.  Gilbert  and  William  W.  Rubey  were  elected  members  of  the  Board  of  Trustees. 

Garrison  Norton  was  elected  Secretary  of  the  Board  of  Trustees  to  fill  the  unexpired 
term  of  the  late  Robert  Woods  Bliss. 

Vacancies  in  standing  committees,  including  one  resulting  from  the  resignation  of  Henry 
R.  Shepley  as  a  member  of  the  Executive  Committee,  were  filled  as  follows:  Amory  H. 
Bradford  and  Robert  E.  Wilson  were  elected  members  of  the  Executive  Committee  for 
two-year  terms,  and  Robert  A.  Lovett  and  James  N.  White  were  reelected  for  three-year 
terms.  The  following  were  reelected  for  three-year  terms:  Richard  S.  Perkins,  Elihu  Root,  Jr., 
and  James  N.  White  as  members  of  the  Finance  Committee;  Omar  N.  Bradley  and  Henry  S. 
Morgan  as  members  of  the  Retirement  Committee;  and  Charles  P.  Taft  as  member  of  the 
Nominating  Committee.  The  following  were  elected  or  reelected  for  one-year  terms:  Henry 
S.  Morgan  as  Chairman  of  the  Executive  Committee,  James  N.  WThite  as  Chairman  of  the 
Finance  Committee,  Keith  S.  McHugh  as  Chairman  of  the  Auditing  Committee,  Omar  N. 
Bradley  as  Chairman  of  the  Retirement  Committee,  Amory  H.  Bradford  as  Chairman  of 
the  Nominating  Committee,  and  Richard  S.  Perkins  as  member  of  the  Nominating  Com- 
mittee. 

497 


Articles  of  Incorporation 


Public  No.  260.  An  Act  to  incorporate  the  Carnegie  Institution  of  Washington 


Be  it  enacted  by  the  Senate  and  House  of  Representatives  of  the  United  States  of  America 
in  Congress  assembled,  That  the  persons  following,  being  persons  who  are  now  trustees  of  the 
Carnegie  Institution,  namely,  Alexander  Agassiz,  John  S.  Billings,  John  L.  Cadwalader, 
Cleveland  H.  Dodge,  William  N.  Frew,  Lyman  J.  Gage,  Daniel  C.  Gilman,  John  Hay, 
Henry  L.  Higginson,  William  Wirt  Howe,  Charles  L.  Hutchinson,  Samuel  P.  Langley, 
William  Lindsay,  Seth  Low,  Wayne  MacVeagh,  Darius  0.  Mills,  S.  Weir  Mitchell,  William 
W.  Morrow,  Ethan  A.  Hitchcock,  Elihu  Root,  John  C.  Spooner,  Andrew  D.  White,  Charles 
D.  Walcott,  Carroll  D.  Wright,  their  associates  and  successors,  duly  chosen,  are  hereby 
incorporated  and  declared  to  be  a  body  corporate  by  the  name  of  the  Carnegie  Institution 
of  Washington  and  by  that  name  shall  be  known  and  have  perpetual  succession,  with  the 
powers,  limitations,  and  restrictions  herein  contained. 

Sec.  2.  That  the  objects  of  the  corporation  shall  be  to  encourage,  in  the  broadest  and 
most  liberal  manner,  investigation,  research,  and  discovery,  and  the  application  of  knowledge 
to  the  improvement  of  mankind;  and  in  particular — 

(a)  To  conduct,  endow,  and  assist  investigation  in  any  department  of  science,  literature, 
or  art,  and  to  this  end  to  cooperate  with  governments,  universities,  colleges,  technical 
schools,  learned  societies,  and  individuals. 

(b)  To  appoint  committees  of  experts  to  direct  special  lines  of  research. 

(c)  To  publish  and  distribute  documents. 

(d)  To  conduct  lectures,  hold  meetings,  and  acquire  and  maintain  a  library. 

(e)  To  purchase  such  property,  real  or  personal,  and  construct  such  building  or  buildings 
as  may  be  necessary  to  carry  on  the  work  of  the  corporation. 

(/)  In  general,  to  do  and  perform  all  things  necessary  to  promote  the  objects  of  the 
institution,  with  full  power,  however,  to  the  trustees  hereinafter  appointed  and  their  suc- 
cessors from  time  to  time  to  modify  the  conditions  and  regulations  under  which  the  work 
shall  be  carried  on,  so  as  to  secure  the  application  of  the  funds  in  the  manner  best  adapted 
to  the  conditions  of  the  time,  provided  that  the  objects  of  the  corporation  shall  at  all  times 
be  among  the  foregoing  or  kindred  thereto. 

Sec.  3.  That  the  direction  and  management  of  the  affairs  of  the  corporation  and  the 
control  and  disposal  of  its  property  and  funds  shall  be  vested  in  a  board  of  trustees,  twenty- 
two  in  number,  to  be  composed  of  the  following  individuals:  Alexander  Agassiz,  John  S. 
Billings,  John  L.  Cadwalader,  Cleveland  H.  Dodge,  William  N.  Frew,  Lyman  J.  Gage, 
Daniel  C.  Gilman,  John  Hay,  Henry  L.  Higginson,  William  Wirt  Howe,  Charles  L.  Hutchin- 
son, Samuel  P.  Langley,  William  Lindsay,  Seth  Low,  Wayne  MacVeagh,  Darius  0.  Mills, 
S.  Weir  Mitchell,  William  W.  Morrow,  Ethan  A.  Hitchcock,  Elihu  Root,  John  C.  Spooner, 

499 


500  CARNEGIE     INSTITUTION      OF      WASHINGTON 

Andrew  D.  White,  Charles  D.  Walcott,  Carroll  D.  Wright,  who  shall  constitute  the  first 
board  of  trustees.  The  board  of  trustees  shall  have  power  from  time  to  time  to  increase  its 
membership  to  not  more  than  twenty-seven  members.  Vacancies  occasioned  by  death, 
resignation,  or  otherwise  shall  be  filled  by  the  remaining  trustees  in  such  manner  as  the 
by-laws  shall  prescribe;  and  the  persons  so  elected  shall  thereupon  become  trustees  and  also 
members  of  the  said  corporation.  The  principal  place  of  business  of  the  said  corporation 
shall  be  the  city  of  Washington,  in  the  District  of  Columbia. 

Sec.  4-  That  such  board  of  trustees  shall  be  entitled  to  take,  hold,  and  administer  the 
securities,  funds,  and  property  so  transferred  by  said  Andrew  Carnegie  to  the  trustees  of 
the  Carnegie  Institution  and  such  other  funds  or  property  as  may  at  any  time  be  given, 
devised,  or  bequeathed  to  them,  or  to  such  corporation,  for  the  purposes  of  the  trust;  and 
with  full  power  from  time  to  time  to  adopt  a  common  seal,  to  appoint  such  officers,  members 
of  the  board  of  trustees  or  otherwise,  and  such  employees  as  may  be  deemed  necessary 
in  carrying  on  the  business  of  the  corporation,  at  such  salaries  or  with  such  remuneration 
as  they  may  deem  proper;  and  with  full  power  to  adopt  by-laws  from  time  to  time  and 
such  rules  or  regulations  as  may  be  necessary  to  secure  the  safe  and  convenient  transaction 
of  the  business  of  the  corporation;  and  with  full  power  and  discretion  to  deal  with  and 
expend  the  income  of  the  corporation  in  such  manner  as  in  their  judgment  will  best  pro- 
mote the  objects  herein  set  forth  and  in  general  to  have  and  use  all  powers  and  authority 
necessary  to  promote  such  objects  and  carry  out  the  purposes  of  the  donor.  The  said  trustees 
shall  have  further  power  from  time  to  time  to  hold  as  investments  the  securities  hereinabove 
referred  to  so  transferred  by  Andrew  Carnegie,  and  any  property  which  has  been  or  may 
be  transferred  to  them  or  such  corporation  by  Andrew  Carnegie  or  by  any  other  person, 
persons,  or  corporation,  and  to  invest  any  sums  or  amounts  from  time  to  time  in  such 
securities  and  such  form  and  manner  as  are  permitted  to  trustees  or  to  charitable  or  literary 
corporations  for  investment,  according  to  the  laws  of  the  States  of  New  York,  Pennsylvania, 
or  Massachusetts,  or  in  such  securities  as  are  authorized  for  investment  by  the  said  deed 
of  trust  so  executed  by  Andrew  Carnegie,  or  by  any  deed  of  gift  or  last  will  and  testament 
to  be  hereafter  made  or  executed. 

Sec.  5.  That  the  said  corporation  may  take  and  hold  any  additional  donations,  grants, 
devises,  or  bequests  which  may  be  made  in  further  support  of  the  purposes  of  the  said 
corporation,  and  may  include  in  the  expenses  thereof  the  personal  expenses  which  the 
trustees  may  incur  in  attending  meetings  or  otherwise  in  carrying  out  the  business  of  the 
trust,  but  the  services  of  the  trustees  as  such  shall  be  gratuitous. 

Sec.  6.  That  as  soon  as  may  be  possible  after  the  passage  of  this  Act  a  meeting  of  the 
trustees  hereinbefore  named  shall  be  called  by  Daniel  C.  Gilman,  John  S.  Billings,  Charles 
D.  Walcott,  S.  Weir  Mitchell,  John  Hay,  Elihu  Root,  and  Carroll  D.  Wright,  or  any  four 
of  them,  at  the  city  of  Washington,  in  the  District  of  Columbia,  by  notice  served  in  person 
or  by  mail  addressed  to  each  trustee  at  his  place  of  residence;  and  the  said  trustees,  or  a 
majority  thereof,  being  assembled,  shall  organize  and  proceed  to  adopt  by-laws,  to  elect 
officers  and  appoint  committees,  and  generally  to  organize  the  said  corporation;  and  said 
trustees  herein  named,  on  behalf  of  the  corporation  hereby  incorporated,  shall  thereupon 
receive,  take  over,  and  enter  into  possession,  custody,  and  management  of  all  property, 
real  or  personal,  of  the  corporation  heretofore  known  as  the  Carnegie  Institution,  incor- 
porated, as  hereinbefore  set  forth  under  "An  Act  to  establish  a  Code  of  Law  for  the  District 
of  Columbia,  January  fourth,  nineteen  hundred  and  two,"  and  to  all  its  rights,  contracts, 
claims,  and  property  of  any  kind  or  nature;  and  the  several  officers  of  such  corporation,  or 
any  other  person  having  charge  of  any  of  the  securities,  funds,  real  or  personal,  books,  or 
property  thereof,  shall,  on  demand,  deliver  the  same  to  the  said  trustees  appointed  by  this 
Act  or  to  the  persons  appointed  by  them  to  receive  the  same;  and  the  trustees  of  the  existing 
corporation  and  the  trustees  herein  named  shall  and  may  take  such  other  steps  as  shall  be 
necessary  to  carry  out  the  purposes  of  this  Act. 

Sec.  7.  That  the  rights  of  the  creditors  of  the  said  existing  corporation  known  as  the 
Carnegie  Institution  shall  not  in  any  manner  be  impaired  by  the  passage  of  this  Act,  or 


ARTICLES    OF   INCORPORATION  501 

the  transfer  of  the  property  hereinbefore  mentioned,  nor  shall  any  liability  or  obligation 
for  the  payment  of  any  sums  due  or  to  become  due,  or  any  claim  or  demand,  in  any  manner 
or  for  any  cause  existing  against  the  said  existing  corporation,  be  released  or  impaired; 
but  such  corporation  hereby  incorporated  is  declared  to  succeed  to  the  obligations  and 
liabilities  and  to  be  held  liable  to  pay  and  discharge  all  of  the  debts,  liabilities,  and  contracts 
of  the  said  corporation  so  existing  to  the  same  effect  as  if  such  new  corporation  had  itself 
incurred  the  obligation  or  liability  to  pay  such  debt  or  damages,  and  no  such  action  or 
proceeding  before  any  court  or  tribunal  shall  be  deemed  to  have  abated  or  been  discontinued 
by  reason  of  the  passage  of  this  Act. 

Sec.  8.  That  Congress  may  from  time  to  time  alter,  repeal,  or  modify  this  Act  of  incor- 
poration, but  no  contract  or  individual  right  made  or  acquired  shall  thereby  be  divested 
or  impaired. 

Sec.  9.     That  this  Act  shall  take  effect  immediately. 

Approved,  April  28,  190 If. 


By-Laws  of  the  Institution 


Adopted  December  13,  1904.  Amended  December  13,  1910,  December  13,  1912,  December 
10,  1937,  December  15,  1939,  December  13,  1940,  December  18,  1942,  December 
12,  1947,  December  10,  1954,  October  24,  1957,  May  8,  1959,  and  May  13,  1960. 


ARTICLE  I 

The  Trustees 

1.  The  Board  of  Trustees  shall  consist  of  twenty-four  members  with  power  to  increase 
its  membership  to  not  more  than  twenty-seven  members.  The  Trustees  shall  hold  office 
continuously  and  not  for  a  stated  term. 

2.  In  case  any  Trustee  shall  fail  to  attend  three  successive  annual  meetings  of  the  Board 
he  shall  thereupon  cease  to  be  a  Trustee. 

3.  No  Trustee  shall  receive  any  compensation  for  his  services  as  such. 

4.  All  vacancies  in  the  Board  of  Trustees  shall  be  filled  by  the  Trustees  by  ballot  at  an 
annual  meeting,  but  no  person  shall  be  declared  elected  unless  he  receives  the  votes  of 
two-thirds  of  the  Trustees  present. 

ARTICLE  II 

Officers  of  the  Board 

1.  The  officers  of  the  Board  shall  be  a  Chairman  of  the  Board,  a  Vice-Chairman,  and  a 
Secretary,  who  shall  be  elected  by  the  Trustees,  from  the  members  of  the  Board,  by  ballot 
to  serve  for  a  term  of  three  years.  All  vacancies  shall  be  filled  by  the  Board  for  the  unexpired 
term;  provided,  however,  that  the  Executive  Committee  shall  have  power  to  fill  a  vacancy 
in  the  office  of  Secretary  to  serve  until  the  next  meeting  of  the  Board  of  Trustees. 

2.  The  Chairman  shall  preside  at  all  meetings  and  shall  have  the  usual  powers  of  a  pre- 
siding officer. 

3.  The  Vice-Chairman,  in  the  absence  or  disability  of  the  Chairman,  shall  perform  the 
duties  of  the  Chairman. 

4.  The  Secretary  shall  issue  notices  of  meetings  of  the  Board,  record  its  transactions,  and 
conduct  that  part  of  the  correspondence  relating  to  the  Board  and  to  his  duties. 

ARTICLE  III 

Executive  Administration 

The  President 

1.  There  shall  be  a  President  who  shall  be  elected  by  ballot  by,  and  hold  office  during 
the  pleasure  of,  the  Board,  who  shall  be  the  chief  executive  officer  of  the  Institution.  The 
President,  subject  to  the  control  of  the  Board  and  the  Executive  Committee,  shall  have 
general  charge  of  all  matters  of  administration  and  supervision  of  all  arrangements  for 

503 


504 


CAKNEGIE     INSTITUTION      OF      WASHINGTON 


research  and  other  work  undertaken  by  the  Institution  or  with  its  funds.  He  shall  prepare 
and  submit  to  the  Board  of  Trustees  and  to  the  Executive  Committee  plans  and  suggestions 
for  the  work  of  the  Institution,  shall  conduct  its  general  correspondence  and  the  corre- 
spondence with  applicants  for  grants  and  with  the  special  advisers  of  the  Committee,  and 
shall  present  his  recommendations  in  each  case  to  the  Executive  Committee  for  decision. 
All  proposals  and  requests  for  grants  shall  be  referred  to  the  President  for  consideration 
and  report.  He  shall  have  power  to  remove,  appoint,  and,  within  the  scope  of  funds  made 
available  by  the  Trustees,  provide  for  compensation  of  subordinate  employees  and  to  fix 
the  compensation  of  such  employees  within  the  limits  of  a  maximum  rate  of  compensation 
to  be  established  from  time  to  time  by  the  Executive  Committee.  He  shall  be  ex  officio  a 
member  of  the  Executive  Committee. 

2.  He  shall  be  the  legal  custodian  of  the  seal  and  of  all  property  of  the  Institution  whose 
custody  is  not  otherwise  provided  for.  He  shall  sign  and  execute  on  behalf  of  the  corporation 
all  contracts  and  instruments  necessary  in  authorized  administrative  and  research  matters 
and  affix  the  corporate  seal  thereto  when  necessary,  and  may  delegate  the  performance  of 
such  acts  and  other  administrative  duties  in  his  absence  to  the  Executive  Officer.  He  may 
execute  all  other  contracts,  deeds,  and  instruments  on  behalf  of  the  corporation  and  affix 
the  seal  thereto  when  expressly  authorized  by  the  Board  of  Trustees  or  Executive  Committee. 
He  may,  within  the  limits  of  his  own  authorization,  delegate  to  the  Executive  Officer 
authority  to  act  as  custodian  of  and  affix  the  corporate  seal.  He  shall  be  responsible  for  the 
expenditure  and  disbursement  of  all  funds  of  the  Institution  in  accordance  with  the  directions 
of  the  Board  and  of  the  Executive  Committee,  and  shall  keep  accurate  accounts  of  all 
receipts  and  disbursements.  Following  approval  by  the  Executive  Committee  he  shall 
transmit  to  the  Board  of  Trustees  before  its  annual  meeting  a  written  report  of  the  operations 
and  business  of  the  Institution  for  the  preceding  fiscal  year  with  his  recommendations  for 
work  and  appropriations  for  the  succeeding  fiscal  year. 

3.  He  shall  attend  all  meetings  of  the  Board  of  Trustees. 

4.  There  shall  be  an  officer  designated  Executive  Officer  who  shall  be  appointed  by  and 
hold  office  at  the  pleasure  of  the  President,  subject  to  the  approval  of  the  Executive  Com- 
mittee. His  duties  shall  be  to  assist  and  act  for  the  President  as  the  latter  may  duly  authorize 
and  direct. 

5.  The  President  shall  retire  from  office  at  the  end  of  the  fiscal  year  in  which  he  becomes 
sixty-five  years  of  age. 

ARTICLE  IV 

Meetings 

1.  The  annual  meeting  of  the  Board  of  Trustees  shall  be  held  in  the  City  of  Washington, 
in  the  District  of  Columbia,  in  May  of  each  year  on  a  date  set  by  order  of  the  Executive 
Committee,  unless  the  date  and  place  of  meeting  are  otherwise  set  by  order  of  the  Executive 
Committee. 

2.  Special  meetings  of  the  Board  may  be  called  by  the  Executive  Committee  by  notice 
served  personally  upon,  or  mailed  to  the  usual  address  of,  each  Trustee  twenty  days  prior 
to  the  meeting. 

3.  Special  meetings  shall,  moreover,  be  called  in  the  same  manner  by  the  Chairman  upon 
the  written  request  of  seven  members  of  the  Board. 


ARTICLE  v 

Committees 

1.  There  shall  be  the  following  standing  Committees,  viz.  an  Executive  Committee,  a 
Finance  Committee,  an  Auditing  Committee,  a  Nominating  Committee,  and  a  Retirement 
Committee. 

2.  All  vacancies  occurring  in  the  Executive  Committee,  the  Finance  Committee,  the 
Auditing  Committee,  the  Nominating  Committee,  and  the  Retirement  Committee  shall  be 


BY-LAWS    OP   THE   INSTITUTION  505 

filled  by  the  Trustees  at  the  next  regular  meeting.  In  case  of  vacancy  in  the  Finance  Com- 
mittee, the  Auditing  Committee,  the  Nominating  Committee,  or  the  Retirement  Committee, 
upon  request  of  the  remaining  members  of  such  committee,  the  Executive  Committee  may 
fill  such  vacancy  by  appointment  until  the  next  meeting  of  the  Board  of  Trustees. 

3.  The  terms  of  all  officers  and  of  all  members  of  committees,  as  provided  for  herein, 
shall  continue  until  their  successors  are  elected  or  appointed. 

Executive  Committee 

4.  The  Executive  Committee  shall  consist  of  the  Chairman,  Vice-Chairman,  and  Secretary 
of  the  Board  of  Trustees  and  the  President  of  the  Institution  ex  officio  and,  in  addition, 
five  trustees  to  be  elected  by  the  Board  by  ballot  for  a  term  of  three  years,  who  shall  be 
eligible  for  re-election.  Any  member  elected  to  fill  a  vacancy  shall  serve  for  the  remainder 
of  his  predecessor's  term. 

5.  The  Executive  Committee  shall,  when  the  Board  is  not  in  session  and  has  not  given 
specific  directions,  have  general  control  of  the  administration  of  the  affairs  of  the  corporation 
and  general  supervision  of  all  arrangements  for  administration,  research,  and  other  matters 
undertaken  or  promoted  by  the  Institution.  It  shall  also  submit  to  the  Board  of  Trustees  a 
printed  or  typewritten  report  of  each  of  its  meetings,  and  at  the  annual  meeting  shall  submit 
to  the  Board  a  report  for  publication. 

6.  The  Executive  Committee  shall  have  power  to  authorize  the  purchase,  sale,  exchange, 
or  transfer  of  real  estate. 

Finance  Committee 

7.  The  Finance  Committee  shall  consist  of  not  less  than  five  and  not  more  than  six 
members  to  be  elected  by  the  Board  of  Trustees  by  ballot  for  a  term  of  three  years,  who 
shall  be  eligible  for  re-election. 

8.  The  Finance  Committee  shall  have  custody  of  the  securities  of  the  corporation  and 
general  charge  of  its  investments  and  invested  funds,  including  its  investments  and  invested 
funds  as  trustee  of  any  retirement  plan  for  the  Institution's  staff  members  and  employees, 
and  shall  care  for  and  dispose  of  the  same  subject  to  the  directions  of  the  Board  of  Trustees. 
It  shall  have  power  to  authorize  the  purchase,  sale,  exchange,  or  transfer  of  securities  and 
to  delegate  this  power.  It  shall  consider  and  recommend  to  the  Board  from  time  to  time 
such  measures  as  in  its  opinion  will  promote  the  financial  interests  of  the  Institution  and  of 
the  trust  fund  under  any  retirement  plan  for  the  Institution's  staff  members  and  employees, 
and  shall  make  a  report  at  each  meeting  of  the  Board. 

Auditing  Committee 

9.  The  Auditing  Committee  shall  consist  of  three  members  to  be  elected  by  the  Board 
of  Trustees  by  ballot  for  a  term  of  three  years. 

10.  Before  each  annual  meeting  of  the  Board  of  Trustees,  the  Auditing  Committee  shall 
cause  the  accounts  of  the  Institution  for  the  preceding  fiscal  year  to  be  audited  by  public 
accountants.  The  accountants  shall  report  to  the  Committee,  and  the  Committee  shall 
present  said  report  at  the  ensuing  annual  meeting  of  the  Board  with  such  recommendations 
as  the  Committee  may  deem  appropriate. 

Nominating  Committee 

11.  The  Nominating  Committee  shall  consist  of  the  Chairman  of  the  Board  of  Trustees 
ex  officio  and,  in  addition,  three  trustees  to  be  elected  by  the  Board  by  ballot  for  a  term 
of  three  years,  who  shall  not  be  eligible  for  re-election  until  after  the  lapse  of  one  year. 
Any  member  elected  to  fill  a  vacancy  shall  serve  for  the  remainder  of  his  predecessor's 
term,  provided  that  of  the  Nominating  Committee  first  elected  after  adoption  of  this.  By-Law 
one  member  shall  serve  for  one  year,  one  member  shall  serve  for  two  years,  and  one  member 
shall  serve  for  three  years,  the  Committee  to  determine  the  respective  terms  by  lot. 

12.  Sixty  days  prior  to  an  annual  meeting  of  the  Board  the  Nominating  Committee  shall 


506  CARNEGIE     INSTITUTION     OF      WASHINGTON 

notify  the  Trustees  by  mail  of  the  vacancies  to  be  filled  in  membership  of  the  Board.  Each 
Trustee  may  submit  nominations  for  such  vacancies.  Nominations  so  submitted  shall  be 
considered  by  the  Nominating  Committee,  and  ten  days  prior  to  the  annual  meeting  the 
Nominating  Committee  shall  submit  to  members  of  the  Board  by  mail  a  list  of  the  persons 
so  nominated,  with  its  recommendations  for  filling  existing  vacancies  on  the  Board  and 
its  Standing  Committees.  No  other  nominations  shall  be  received  by  the  Board  at  the 
annual  meeting  except  with  the  unanimous  consent  of  the  Trustees  present. 

Retirement  Committee 

13.  The  Retirement  Committee  shall  consist  of  three  members  to  be  elected  by  the 
Board  of  Trustees  by  ballot  for  a  term  of  three  years,  who  shall  be  eligible  for  re-election, 
and  the  Chairman  of  the  Finance  Committee  ex  officio.  Any  member  elected  to  fill  a  vacancy 
shall  serve  for  the  remainder  of  his  predecessor's  term. 

14.  The  Retirement  Committee  shall,  subject  to  the  directions  of  the  Board  of  Trustees, 
be  responsible  for  the  maintenance  of  a  retirement  plan  for  staff  members  and  employees 
of  the  Institution  and  act  for  the  Institution  in  its  capacity  as  trustee  under  any  such  plan, 
except  that  an}'  matter  relating  to  investments  under  any  such  plan  shall  be  the  responsi- 
bility of  the  Finance  Committee  subject  to  the  directions  of  the  Board  of  Trustees.  The 
Committee  shall  submit  a  report  to  the  Board  at  the  annual  meeting  of  the  Board. 

ARTICLE  VI 

Financial  Administration 

1.  No  expenditure  shall  be  authorized  or  made  except  in  pursuance  of  a  previous  appro- 
priation by  the  Board  of  Trustees,  or  as  provided  in  Article  V,  paragraph  8,  hereof. 

2.  The  fiscal  year  of  the  Institution  shall  commence  on  the  first  day  of  July  in  each  year. 

3.  The  Executive  Committee  shall  submit  to  the  annual  meeting  of  the  Board  a  full 
statement  of  the  finances  and  work  of  the  Institution  for  the  preceding  fiscal  year  and  a 
detailed  estimate  of  the  expenditures  of  the  succeeding  fiscal  year. 

4.  The  Board  of  Trustees,  at  the  annual  meeting  in  each  year,  shall  make  general  appro- 
priations for  the  ensuing  fiscal  year;  but  nothing  contained  herein  shall  prevent  the  Board 
of  Trustees  from  making  special  appropriations  at  any  meeting. 

5.  The  Executive  Committee  shall  have  general  charge  and  control  of  all  appropriations 
made  by  the  Board.  Following  the  annual  meeting,  the  Executive  Committee  may  allocate 
these  appropriations  for  the  succeeding  fiscal  year.  The  Committee  shall  have  full  authority 
to  reallocate  available  funds,  as  needed,  and  to  transfer  balances. 

6.  The  securities  of  the  Institution  and  evidences  of  property,  and  funds  invested  and 
to  be  invested,  shall  be  deposited  in  such  safe  depository  or  in  the  custody  of  such  trust 
company  and  under  such  safeguards  as  the  Finance  Committee  shall  designate,  subject  to 
directions  of  the  Board  of  Trustees.  Income  of  the  Institution  available  for  expenditure 
shall  be  deposited  in  such  banks  or  depositories  as  may  from  time  to  time  be  designated 
by  the  Executive  Committee. 

7.  Any  trust  company  entrusted  with  the  custody  of  securities  by  the  Finance  Committee 
may,  by  resolution  of  the  Board  of  Trustees,  be  made  Fiscal  Agent  of  the  Institution,  upon 
an  agreed  compensation,  for  the  transaction  of  the  business  coming  within  the  authority 
of  the  Finance  Committee. 

ARTICLE  VII 

Amendment  of  By-Laws 

1.  These  by-laws  may  be  amended  at  any  annual  or  special  meeting  of  the  Board  of 
Trustees  by  a  two-thirds  vote  of  the  members  present,  provided  written  notice  of  the  pro- 
posed amendment  shall  have  been  served  personally  upon,  or  mailed  to  the  usual  address 
of,  each  member  of  the  Board  twenty  days  prior  to  the  meeting. 


Index  of  Names 


Numbers  in  italic  type  refer  to  pages  in  the  Report  of  the  President. 


Abbe,  Cleveland,  107 
Abbot,  Charles  G.,  56,  79 
Abbott,  Edith,  93 
Abel,  John  J.,  108 
Abell,  George  O.,  39-40 

Abelson,  Philip  H.,  viii,  74,  80,  85,  185,  188,  189, 
190,  191,  192,  193,  208 

publications,  197,  201,  202 

report  of  Director  of  Geophysical  Laboratory, 
51-208 

studies,  53-54,  56,  179-184 
Abercrombie,  Michael,  105,  371,  432 

studies,  396-397 
Aberle,  Sophie  D.,  91 
Abetti,  Giorgio,  79 
Abrams,  Leroy  R.,  76 
Abt,  Helmut  A.,  9 

publications,  47 
Ackerman,  Edward  A.,  x 

publications,  478 
Acree,  Solomon  F.,  105 
Adams,  Charles  Francis,  89 
Adams,  Eleanor  B.,  91 
Adams,  Ephraim  D.,  89 
Adams,  Frank  D.,  48,  86 
Adams,  H.  S.,  244 

publication,  288 
Adams,  Henry  C,  93 
Adams,  Leason  H.,  85 
Adams,  Oscar  S.,  108 
Adams,  Robert  M.,  Jr.,  91 
Adams,  Walter  S.,  78,  5 
Adams,  William  E.,  87 
Adelberg,  Edward  A.,  439 
Adler,  Cyrus,  111 
Agassiz,  Alexander,  vii,  499 
Agassiz,  Louis,  441 
Agrell,  S.  O.,  193 
Akerberg,  Erik,  328 


Akey,  Alice,  476 

publication,  475 

studies,  466-474 
Albrecht,  Sebastian,  99 
Aldrich,  L.  Thomas,  viii,  80,  291 

publication,  288,  289 

studies,  45-47,  173-179,  221-239 
Alfven,  H.,  241 
Allan,  Frank  D.,  432 
Allen,  Eugene  T.,  85 
Allen,  Samuel  J.  Mcintosh,  81 
Allende,  Ines  de,  87 
Aller,  Lawrence,  79,  10,  40 

studies,  60 
Allfrey,  V.  G.,  439 
Allison,  Ira  S.,  110 
Allison,  Samuel  K.,  86 
Allison,  William  H.,  89 
Alvarez,  H.,  292 
Amaldi,  E.,  81 
Ambler,  Charles  H.,  93 
Ames,  Joseph  S.,  106 
Amsden,  Monroe,  91 
Andersen,  Olaf,  85,  108 
Anderson,  Carl  D.,  106,  3 
Anderson,  Edgar,  83 
Anderson,  Ernest,  76 
Anderson,  J.  A.,  78 
Andervont,  Howard  D.,  87 
Andrade,  Manuel  J.,  91,  111 
Andrews,  Charles  M.,  89 
Andrews,  E.  Wyllys,  91 
Andrews,  John  B.,  93 

Appleman,  D.  E.,  publication,  196-197,  201,  202 
Argyris,  Bertie  F.,  87,  371,  409,  432 

studies,  384-390 
Argyris,  Thomas  S.,  87,  371,  405,  432 

studies,  408-415 
Armsby,  Henry  P.,  100 


507 


508 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Arnold,  Ralph  G.,  104,  193 

publications,  199-200,  201 
Arnold,  William  A.,  xi,  76 
Aronson,  Arthur  I.,  102 
Arp,  Halton  C,  viii,  58,  74,  78,  28,  29,  46 

publications,  47 

studies,  61,  15,  26,  27,  31-32 
Arpigny,  Christine,  46 
Arpigny,  Claude,  8 

publication,  47,  48 
Asada,  Toshi,  102,  291 

publications,  288,  290 

studies,  221-234 
Ash,  James  E.,  100 
Ashby,  Eric,  76 
Atwater,  Wilbur  O.,  109 
Atzeni,  A.,  112,  113 
Auerbach,  Robert,  430 
Ault,  J.  P.,  80 
Axelrod,  Daniel  I.,  76 

Baade,  Walter  A.,  57,  78,  16,  26,  27,  28,  29,  35 

publication,  47 
Baadsgaard,  H.,  208,  292 

studies,  234-239 
Babcock,  Ernest  B.,  76 
Babcock,  Harold  D.,  56,  78,  7,  46 
Babcock,  Horace  W.,  viii,  56,  78,  3,  7,  39,  46 

publication,  47 

studies,  21,  22,  23 
Badcock,  E.  B.,  72 
Bahner,  K.,  publication,  47 
Bailey,  D.  Kenneth,  104,  53,  208 

studies,  91-96 
Bailey,  Irving  W.,  76 
Bailey,  Percy  L.,  Jr.,  96 
Baird,  John  W.,  109 
Baker,  Herbert  G.,  102 
Baker,  Oliver  Edwin,  93 
Balbinder,  Elias,  439 

publications,  475 
Balch,  Emily  Greene,  93 
Baldwin,  F.  Spencer,  93 
Baldwin,  George  J.,  vii 
Baldwin,  James  Mark,  109 
Ball,  Ernest,  83 
Ball,  Stanley  C,  96 
Ballagh,  James  C,  89 
Bancroft,  Wilder  D.,  108 
Bandelier,  Adolf  F.  A.,  89 
Banta,  Arthur  M.,  82 
Bappu,  M.  K.  Vainu,  102 
Barbour,  Thomas,  vii 
Barghoorn,  E.  S.,  185,  190 
Barker,  Eugene  C,  89 
Barker,  G.  F.,  106 
Barlow,  Horace  B.,  101 
Barlow,  Robert  H.,  105 
Barnard,  Edward  E.,  55,  79 
Barnard,  J.  Lynn,  93 
Barnes,  Hubert  L.,  104,  194 
Barnett,  Samuel  J.,  80,  106 


Barnhart,  Philip  E.,  42 

Barry,  Alexander,  87 

Bartelmez,  George  W.,  86,  371,  432 

publications,  430 

studies,  427-428,  477 
Bartels,  J.,  81 
Barth,  Tom.  F.  W.,  85,  121 
Barton,  Paul,  193 
Bartsch,  Paul,  96 
Barus,  Carl,  81 
Baskerville,  Charles,  105 
Bass,  Manuel  N.,  102,  237 
Bassett,  F.  B.,  108 
Bassett,  John  S.,  89 
Bast,  T.  H.,  87 
Bate,  John  Pawley,  111 
Bates,  Robert  W.,  82 
Bauer,  Ailene  J.,  x 
Bauer,  Hans,  83 
Bauer,  Louis  A.,  4^1  80 
Baum,  William  A.,  viii,  58,  78,  29,  33,  46 

publications,  47,  288 

studies,  62,  30-31,  34,  35,  295-301 
Baumgartner,  E.,  242 
Baxter,  Gregory  P.,  105 
Beach,  Liselote,  292 

publication,  290 
Beadle,  George  W.,  83 
Beams,  Jesse  W.,  81 
Bean,  George  L.,  108 
Beattie,  J.  C,  81 
Becker,  G.  F.,  48,  86 
Becker,  W.,  79,  43 
Behr,  Gustavus  E.,  105 
Bell,  Earl  H.,  110 
Bell,  Herbert  C.  F.,  89 
Bell,  James  F.,  vii,  481 
Belling,  John,  33,  72,  76,  82,  437 
Belov,  N.  V.,  132 
Bemis,  Samuel  F.,  89 
Benedict,  Cornelia  Golay,  100 
Benedict,  Francis  G.,  100 
Benfield,  Adalbert  E.,  107 
Benioff,  Hugo,  108 
Bennett,  Ralph  D.,  81,106 
Bensley,  Edward  H.,  100 
Benson,  C.  C,  100 
Bergen,  Henry,  111 
Berger,  Jacques,  19,  46 

studies,  18 
Berger,  Jeanne,  46 
Bergeron,  Tor,  107 
Berkebile,  Wilma  J.,  47 
Berkner,  Lloyd  V.,  80 
Berlin,  Heinrich,  105 
Berrill,  Norman  J.,  96 
Bertani,  Guiseppe,  103 
Bessey,  Charles  E.,  76 
Bessman,  Maurice,  439 
Bethe,  Hans  A.,  81 
Beyer,  Herman,  91 
Bhattacharya,  Prabhat  K.,  102 


INDEX    OF    NAMES 


509 


Bichowsky,  F.  Russell  von,  85 

Biesele,  John  J.,  83 

Biggers,  J.  D.,  430 

Billings,  John  S.,  vii,  499,  500 

Birch,  F.,  228 

Bishop,  Alvard  Longley,  93 

Bishop,  David  W.,  ix,  64,  86,  369,  370,  432 

publications,  431 

studies,  390-392 
Bishop,  Norman,  360 
Bishop,  Timothy,  432 
Bjerknes,  J.,  107 
Bjerknes,  V.,  107 
Bjorkman,  Olle  E.,  309,  310,  312 

studies,  320-323 
Blaauw,  A.,  25 
Blackmar,  Frank  W.,  93 
Blackwelder,  Eliot,  107 
Blake,  Marion  E.,  Ill 
Blakeslee,  Albert  F.,  72,  82,  437 
Blieden,  H.  R.,  243 

publication,  288,  290 
Blinks,  Lawrence  R.,  36,  96,  343 
Bliss,  Robert  Woods,  v,  vii,  68-69,  481,  497 
Blom,  Franz,  91 
Blood,  Alice  F.,  100 
Boas,  Franz,  111 
Bodenstein,  Dietrich,  83 
Boezi,  J.  A.,  publication,  288 
Bogart,  Ernest  Ludlow,  93 
Boise,  James  W.,  x 
Bolton,  Ellis  T.,  viii,  80,  219,  291 

studies,  43-44,  244-288 
Bolton,  Herbert  E.,  89 
Bond,  Beverley  Waugh,  93 

Bonini,  W.  E.,  publications,  288,  289,  290,  477 
Bonsack,  Walter  K.,  44 
Booker,  Henry  G.,  81 
Borhegyi,  Stephen  F.  de,  91 
Boschma,  H.,  96 
Boss,  Benjamin,  99 
Boss,  Lewis,  56,  99 
Bouchat,  William,  432 
Boving,  Bent  G.,  ix,  86,  371,  432 

publications,  431,  477 

studies,  419-423 
Bowditch,  Henry  P.,  109 
Bowen,  Ira  S.,  viii,  74,  78,  3,  46 

publication,  47 

report    of    Director    of    Mount    Wilson    and 
Palomar  Observatories,  3-50 
Bowen,  Norman  L.,  51,  85,  54,  60,  69,  70,  75, 

85,  92,  95,  96,  97,  98,  108,  117 
Bowie,  William,  108 
Bowles,  Edward  L.,  81 
Bowman,  Harold  Martin,  89 
Bowman,  Howard  H.  M.,  96 
Boyce,  Joseph  C.,  81 
Boyd,  Francis  R.,  Jr.,  viii,  85,  61,  66,  67,  192,  208 

publication,  201 

studies,  52,  53,  54,  68-75,  107-112 
Boyd,  J.  D.,  87 


Boyd,  William  K.,  93 

Boyden,  Alan  A.,  96 

Boyden,  E.  A.,  87 

Boyle,  James  E.,  93 

Bradford,  Amory  H.,  v,  vi,  497 

Bradford,  Lindsay,  vii 

Bradley,  Omar  N.,  v,  vi,  481,  497 

Brainerd,  George  W.,  91 

Bramhall,  E.  H.,  80 

Brandt,  John  C.,  27,  46 

publications,  47,  48 
Branner,  J.  C.,  86 
Breder,  Charles  M.,  Jr.,  96 
Brehme  (Warren),  Katherine  S.,  103 
Breit,  Gregory,  6,  80 
Brenner,  Sydney,  72,  83 
Brett,  P.  R.,  104,  208 

studies,  55,  155-160 
Bretz,  Julian  P.,  89 
Bridges,  Calvin  B.,  72,  85 
Brierly,  J.  Leslie,  111 
Briggs,  Winslow  R.,  323 
Brigham,  William  T.,  Ill 
Briles,  Elwood,  396 
Brinley,  Floyd  J.,  96 
Britten,  Roy  J.,  viii,  80,  291,  373,  445 

publications,  289 

studies,  43-44,  244-288 
Britton,  Nathaniel  L.,  76 
Brode,  Robert  B.,  81 
Brodfiihrer,  Ursula,  76 
Brody,  Marcia,  352 
Brody,  Samuel,  100 
Brookings,  Robert  S.,  vii 
Brooks,  William  K.,  96 
Brouwer,  Dirk,  79 
Brouwer,  L.  E.  J.,  86 
Brown,  Amos  P.,  105 
Brown,  Donald  D.,  86,  285,  369,  432 

studies,  64,  370,  372-384 
Brown,  Dugald  E.  S.,  96 
Brown,  Ernest  W.,  100 
Brown,  Jeanette  S.,  ix,  76,  310,  346,  354,  365 

publications,  366 

studies,  37,  40,  307,  350-353 
Brown,  Louis,  102,  291 

studies,  242-244 
Brown,  Marion  E.,  329 
Brown,  W.  L.,  publication,  194,  201 
Brown,  William  L.,  439 
Brownlee,  Robert,  13 
Brueckel,  Frank  J.,  46 
Bruhn,  John  M.,  100 
Bryan,  Kirk,  91 
Bryson,  Vernon,  83 
Buchanan,  Jennie  S.,  438,  475 

studies,  466-474 
Buchholz,  John  T.,  83 
Buck,  Solon  J.,  93 
Buddington,  A.  F.,  85 
Bull,  M.  Lucien,  100 
Bullard,  W.  R.,  91 


510 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Bullington,  Walter  E.,  96 

Bundgaard,  Robert  C,  107 

Buno,  Washington,  87 

Buravas,  Saman,  236 

Burbidge,  E.  Margaret,  publication,  48,  50 

Burbidge,  Geoffrey  R.,  102 

publication,  48,  50 
Burd,  Sylvia,  22,  46 
Burgess,  Charles  F.,  106 
Burgi,  Elizabeth,  ix,  440,  475 

publications,  475 

studies,  443-448 
Burke,  Bernard  F.,  viii,  80,  291 

publications,  289 

studies,  214-221 
Burkenroad,  Martin,  96 
Burks,  Barbara  S.,  82 
Burlew,  John  S.,  85 
Burnet,  Sir  Macfarlane,  83 
Burnett,  Edmund  C,  89 
Burnham,  Charles  W.,  104,  62,  193,  208 

studies,  55,  132-139 
Burns,  Robert  K.,  ix,  69,  70,  86,  372,  432 

publication,  431 

studies,  415-416 
Burr,  William  H.,  107 
Buscombe,  William  A.,  102 
Buseck,  Peter  R.,  104,  208 

studies,  55,  150-151,  152-154,  161-163 
Bush,  Vannevar,  v,  vii,  7,  73,  497 
Butterfield,  Kenyon  L.,  92 
Buwalda,  John  P.,  110 
Byerly,  Perry,  108 
Byrd,  Richard  E.,  81 

Cabre,  R.,  S.J.,  231,  292 
Cadwalader,  John  L.,  vii,  499 
Cady,  Walter  G.,  100 
Cairns,  Hugh  J.,  103 

publication,  475 
California  Institute  of  Technology,  5,  18,  20,  3 
Callaway,  Morgan,  111 
Cameron,  A.  G.  W.,  15 

publication,  48 
Campbell,  Douglas  H.,  76 
Campbell,  Frank  L.,  192 
Campbell,  Ian,  107 
Campbell,  William,  106 
Campbell,  William  W.,  vii 
Cannon,  William  A.,  76 
Carhart,  Anne  K.,  475 

studies,  461-466 
Carhart,  Henry  S.,  106 
Carlson,  Anton  J.,  109 
Carlson,  Gerald  L.,  64,  87,  390,  432 

studies,  370-371,  384,  392-393,  417-419 
Carlson,  J.  Gordon,  83 
Carnegie,  Andrew,  3-4,  68,  500 
Carnegie  Collection  of  Embryos,  425,  426-427, 

428 
Carpenter,  Frank  M.,  110 
Carpenter,  Thome  M.,  100 


Carroll,  Ann  C,  441,  476 
Carroll,  John  A.,  79 
Cartledge,  J.  Lincoln,  83 
Carty,  John  J.,  vii 
Carver,  Thomas  N.,  93 
Cary,  Lewis  R.,  96 
Casaverde,  Mateo,  102,  292 

publication,  290 
Case,  Ermine  C,  110 
Caso  y  Andrade,  Alfonso,  91 
Caspari,  Ernst,  82 
Cassle,  Marietta  M.,  476 
Castle,  William  E.,  72,  83 
Caston,  J.  Douglas,  369,  432 

studies,  64,  370,  372-384 
Catcheside,  David  G.,  72,  83 
Cathcart,  E.  P.,  100 
Cattell,  J.  McKeen,  111 
Catterall,  Helen  T.,  89 
Cavalieri,  Liebe  F.,  439 
Cayrel  de  Strobel,  Guisa,  publication,  48 
Cesar,  H.  Lenz,  236,  292 
Chaddock,  Robert  E.,  93 
Chamberlain,  Robert  S.,  91 
Chamberlin,  Rollin  T.,  107 
Chamberlin,  Thomas  C,  86 
Chambers,  Edward  L.,  96 
Chambers,  Robert,  96 
Champney,  Scott,  476 
Chandra,  Subhash,  46 
Chaney,  Ralph  W.,  76 
Channell,  R.  B.,  studies,  333-334 
Chapman,  Frank  M.,  96 
Chapman,  Kenneth  M.,  91 
Chapman,  Sydney,  81 
Charles,  Donald  R.,  83 
Chariot,  Jean,  91 
Chayes,  Felix,  viii,  85,  193,  208 

studies,  54,  112-126 
Chen,  Shao-lin,  102 
Child,  Clement  D.,  106 
Chinner,  G.  A.,  104,  57,  59,  61,  193 
Chittenden,  Russell  H.,  108 
Chovnick,  Arthur,  439 
Chowning,  Ann,  91 
Choy,  Jai  H.,  46 
Churchill,  William,  111 
Chutna\  J.,  391,  430 
Clair,  Robert  A.,  310,  317 

studies,  319-320 
Clark,  Eleanor  L.,  63,  87 
Clark,  Eliot  R.,  63,  87 
Clark,  Hubert  Lyman,  96 
Clark,  John  B.,  93 
Clark,  L.  A.,  194 

publications,  198,  201,  202 
Clark,  Leonard  B.,  96 

Clark,  Sydney  P.,  Jr.,  viii,  85,   135,   146,   149 
193,  208,  231 

publications,  194,  196-197,  202 

studies,  50,  52-53,  53,  55,  56,  59-68,  165-172 
Clark,  Victor  S.,  92 


INDEX    OF    NAMES 


511 


Clarke,  Frank  W.,  96 
Claude,  Albert,  83 
Clausen,  Jens  C,  74,  76,  310 

publications,  366 

studies,  309,  312-313,  325-334 
Clawson,  A.  B.,  109 
Clawson,  Marion,  publication,  478 
Cleary,  William  I.,  423,  432 
Clement,  J.  K.,  85 
Clements,  Frederic  E.,  76 
Cleveland,  Frederick  A.,  98 
Cleven,  N.  Andrew  N.,  112 
Cloud,  P.  E.,  Jr.,  publication,  197,  202 
Coble,  Arthur  B.,  107 
Coblentz,  William  W.,  106 
Code,  Arthur  D.,  78,  40 
Coffeen,  Mary  F.,  46 
Coffman,  C.  M.,  publication,  431,  477 
Cohen,  Georges  N.,  81 
Cohen,  I.  Bernard,  91 
Cold  Spring  Harbor  Symposia  on  Quantitative 

Biology,  439-440 
Cole,  Fay-Cooper,  91 
Cole,  Leon  J.,  96 
Cole,  Whitefoord  R.,  vii 
Coleman,  John  R.,  64,  371,  416,  432 

publications,  431 
Coleman,  R.  G.,  193 
Collins,  G.  W.,  91 
Colman,  John,  96 
Colwin,  A.  L.,  430 
Commons,  John  R.,  92 
Compston,  William,  102 
Compton,  Arthur  H.,  81 
Compton,  Karl  T.,  81 
Condit,  Carlton,  110 
Conger,  Paul  S.,  96 
Conklin,  Edwin  G.,  96 
Conklin,  Marie  E.,  83 
Connell,  Frank  H.,  91 
Conway,  Thomas,  Jr.,  93 
Coolidge,  Mary  Roberts,  93 
Coomber,  Janice  E.  B.,  366 

publication,  366 
Coonfield,  Benjamin  R.,  96 
Cooper,  Lane,  111 
Cooper,  W.  G.,  publication,  431 
Cooper,  William  S.,  76 
Corey,  Robert  Brainard,  20-21 
Corner,  George  W.,  70,  86 
Corner,  George  W.,  Jr.,  87,  371,  432 

studies,  423-425 
Coropatchinsky,  V.,  100 
Correns,  Karl  Erik,  10,  72 
Couch,  J.  N.,  83 
Coulombre,  A.  J.,  430 
Coulter,  John  Lee,  93 
Coville,  Frederick  V.,  76 
Cowdry,  E.  V.,  87 
Cowie,  Dean  B.,  viii,  80,  291 

publications,  288,  289 

studies,  43-44,  244-288 


Cowles,  Rheinart  P.,  96 
Cowling,  T.  G.,  81 
Cox,  A.  N.,  13 
Cox,  Isaac  Joslin,  89,  112 
Coyne,  Mary  T.,  476 
Cragg,  Thomas  A.,  6,  46 

publications,  48 
Crampton,  Henry  E.,  72,  109 
Crane,  F.  L.,  360 
Crawford,  John  W.,  426 
Creaser,  E.  H.,  102 
Cressman,  Luther  S.,  91 
Crew,  Henry,  106 
Crick,  F.  H.  C,  21,  443 
Crofts,  Elizabeth  E.,  100 
Crook,  James  Walter,  93 
Cross,  Ira  Brown,  93 
Cruz,  Maria  Victoria  de  la,  87 
Csapo,  Arpad,  64,  86,  424 
Cuajunco,  F.,  87 
Cummins,  Harold,  87 
Cushman,  Joseph  A.,  109 
Cutright,  Paul  R.,  96 
Czyzak,  Stanley  J.,  40 

Daggett,  Stuart,  93 

Dahl,  O.,  80 

Dahlgren,  Ulric,  96 

Dalton,  H.  Clark,  108 

Daly,  Reginald  A.,  96 

Damon,  P.  E.,  237 

Danchakoff,  Vera,  87 

Danesino,  Vittorio,  105 

Dansereau,  Pierre,  76 

Darby,  Hugh  H.,  81 

Das,  C.  C,  440,  466,  469,  475 

Daugherty,  Lyman  H.,  110 

Davenport,  Charles  B.,  SO,  82,  71,  72,  82,  437 

Davenport,  Frances  G.,  89 

Davidson,  George,  107 

Davies,  F.  T.,  80 

Davis,  Carl  Lawrence,  87 

Davis,  Edwin  A.,  102 

Davis,  Gordon  L.,  viii,  46,  85,  190,  194,  208 

publication,  200,  202,  289 

studies,  56,  173-179,  234-239 
Davis,  John  H.,  Jr.,  96 
Davis,  Paul  B.,  105 
Davis,  Ranice,  425,  426 
Davis,  William  Morris,  107 
Dawson,  Edgar  M.,  93 
Day,  Arthur  L.,  6,  48~49,  85 
Day,  Clive,  93 
Day,  David  T.,  93 
Dean,  A.  L.,  85 
Dean,  Bashford,  109 
Decker,  Floyd  F.,  107 
Decker,  John  P.,  77 
De  Feo,  Vincent  J.,  87 
DeHaan,  Robert  L.,  ix,  64,  86,  369,  432 

publications,  431 

studies,  400-404 


512 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Dekaban,  Anatole  S.,  87,  371,  432 

publication,  430,  431,  477 

studies,  427-428 
Dekker,  Arentje,  371,  432 
de  Kouchkovsky,  Y.,  310,  334 

studies,  37 
DeLanney,  Louis  E.,  105,  432 

publication,  431,  477 

studies,  404-406 
Delano,  Frederic  A.,  vii 
Delaporte,  Berthe,  103 
de  Lapradelle,  Albert  G.,  Ill 
de  Laubenfels,  May  W.,  97 
Delbriick,  M.,  72,  83 

Demerec,  Milislav,  33,  69,  71,  72,  74,  82,  437 
de  Neufville,  John,  104,  208 

studies,  52,  53,  53,  56-68 
Denham,  Robert  N.,  Jr.,  93 
Denison,  John  H.,  Jr.,  91 
Dennis,  Louis  M .,  105 
de  Paula  Souza,  G.  H.,  100 
de  Renyi,  George  S.,  97 
de  Sitter,  William,  79 
De  Terra,  Hellmut,  110 
Deutsch,  Armin  J.,  viii,  78,  46 

publications,  48,  49 

studies,  21 
de  Vries,  Hessel,  86 

de  Vries,  Hugo,  10,  22,  30,  71,  72,  83,  437 
de  Vries,  Pieter  A.,  publication,  431,  477 
Dewey,  Davis  R.,  92 
Dice,  Lee  R.,  77 
Dickson,  Leonard  E.,  107 
Didusch,  James  F.,  86,  425 
Dill,  David  B.,  100 
Doak,  J.  B.,  292 
Doan,  Charles  A.,  87 
Dobzhansky,  Th.,  72,  83,  313 
Dodd,  R.  T.,  237 
Dodd,  Walter  F.,  89 
Dodge,  Cleveland  H.,  vii,  499 
Dodge,  Raymond,  100 
Dodge,  William  E.,  vii 
Doe,  B.  R.,  46,  104,  194,  208,  292 

studies,  234-239 
Doermann,  A.  H.,  103 
Dole,  Richard  B.,  97 
Donaldson,  Henry  H.,  97 
Donnan,  Elizabeth,  89 
Donnay,  Gabrielle,  viii,  85,  193,  208 

publications,  194,  200,  202 

studies,  55,  130-132,  139 
Donnay,  J.  D.  H.,  86,  208 

publications,  194,  200,  202 

studies,  55,  130-132,  139 
Donner,  Martin  W.,  371,  432 

studies,  423-425 
Dorf,  Erling,  77 
Dorsey,  N.  Ernest,  81 
Dorsey,  George  A.,  Ill 
Doten,  Carroll  W.,  93 
Doughty,  Howard  W.,  105 


Douglass,  A.  E.,  77 
Dove,  W.  N.,  292 
Doyle,  William  L.,  97 
Drew,  George  Harold,  97 
Drew,  Gilman  A.,  97 
Drury,  Newton  B.,  77 
Du  Bois,  Eugene  F.,  100 
Dubois,  W.  E.  B.,  93 
Duerksen,  J.  D.,  102 
Duesberg,  Jules,  87 
Duggar,  Benjamin  M.,  77 
Dunham,  Theodore,  Jr.,  78,  5 
Dunn,  L.  C,  72,  83 
Durand,  William  F.,  107 
Du  Toit,  A.  L.,  110 
Duysens,  L.  N.  M.,  102 
Dyer,  Edward  R.,  Jr.,  102 


Eames,  Wilberforce,  111 

Earle,  Walter  K.,  441 

East,  Edward  M.,  6 

Ebert,  James  D.,  ix,  63-64,  67,  86,  371,  430,  432 

publications,  431,  477 

report  of  Director  of  Department  of  Embry- 
ology, 367-433 

studies,  404-406 
Eckel,  Edwin  C,  93 
Ecklund,  E.  T.,  292 
Eddy,  Walter  H.,  108 
Edmonds,  H.  M.  W.,  80 
Edmondson,  Charles  H.,  97 
Edmunds,  L.  N.,  Jr.,  292 
Edsall,  David  L.,  100 
Edwards,  H.  T.,  100 
Eggen,  Olin  J.,  viii,  78,  18,  45,  46 

publications,  48 

studies,  60-61,  12-13,  15-17,  25-26,  44 
Ehrendorfer,  Friedrich,  77 
Eigenmann,  Carl  H.,  109 
Einstein,  Albert,  79 
Elliott,  Ruth  F.,  310 

studies,  309,  323-325 
Ely,  Richard  T.,  93 
Emerson,  Robert,  36,  77,  346 
Emerson,  Rollins  A.,  91 
Emery,  K.  O.,  181 
Emmons,  William  H.,  86 
Enders,  Robert  K.,  87 
England,  Joseph  L.,  viii,  85,  61,  66,  67,  74,  75,  208 

studies,  52,  54,  107-112 
Ephrussi,  Boris,  83 
Epstein,  Paul  S.,  106 
Erdtmann,  G.  E.,  77 
Erickson,  William  C,  103 
Ernst,  W.  Gary,  104 

publication,  197,  202 
Errico,  James,  405,  432 
Eskola,  Pentti,  85,  121,  178 
Eugster,  H.  P.,  85,  90,  101,  102,  105,  194 

publications,  198-199,  202 
Evans,  Herbert  M.,  87 


INDEX    OF    NAMES 


513 


Fahlquist,  D.  A.,  publications,  289,  290 

Fairbairn,  H.  W.,  237 

Fairchild,  Fred  Rogers,  93 

Fairchild,  Henry  Pratt,  93 

Falconer,  John  I.,  93 

Falta,  W.,  100 

Fano,  Ugo,  82 

Farlow,  William  G.,  77 

Farnam,  Henry  W.,  92 

Farr,  Gertrude  A.,  100 

Farr,  Shirley,  89 

Farrand,  Max,  89 

Farrar,  Clarence  B.,  109 

Faul,  Henry,  86,  208 

Faust,  Albert  B.,  89,  93 

Fawcett,  Jeff  J.,  104,  208 

studies,  54,  88-91 
Feige,  Jacques,  18 
Feinendegen,  Ludwig  E.,  439 
Fenner,  Charles  P.,  vii 
Fenner,  Clarence  N.,  85 
Fen  wick,  Charles  G.,  Ill 
Ferguson,  Homer  L.,  vii 
Ferguson,  W.  Richard,  371,  432 
Ferguson,  William  S.,  89 
Fernandez,  Gonzalo,  103 
Ferris,  Roxana  S.,  313 
Field,  George  B.,  103 
Field,  Richard  M.,  97 
Findlay,  J.  W.,  103 
Finger,  I.,  430 
Firor,  John  W.,  viii,  80,  291 

publications,  289 
Fish,  Carl  R.,  89 
Fish,  Harold  D.,  83 
Fisher,  Agnes  C,  475 
Fisher,  Fulton  J.  F.,  102 
Fisk,  H.  W.,  80 
Fleischer,  Michael,  85 
Fleming,  John  A.,  80 
Fleming,  Walter  L.,  93 
Flexner,  J.  B.,  285,  292 
Flexner,  Louis  B.,  xi,  87,  285,  292,  432 
Flexner,  Simon,  vii,  109 
Fliigel,  Ewald,  111 
Folkers,  Karl,  360 
Folse,  J.  A.,  106 
Forbes,  Alexander,  1 09 
Forbes,  George  S.,  105 
Forbes,  Thomas  R.,  87 
Forbes,  W.  Cameron,  vii 
Forbush,  Scott  E.,  viii,  74,  80,  291 

publications,  289,  290 

studies,  42,  239-241 
Ford,  C.  E.,  430 
Ford,  W.  Kent,  Jr.,  viii,  45,  80,  288,  291,  300, 

301 
Ford,  Worthington  C,  89 
Fork,  David  C,  ix,  76,  310,  346,  365,  366 

publications,  366 

studies,  37-38,  40,  305,  334-345 
Forrestal,  James,  vii 


Fowler,  William  A.,  publications,  48 

Fox,  Allen,  439 

Fox,  Dixon  R.,  89 

Fox,  J.  D.,  publication,  288,  289 

Frankel,  Fred  R.,  35,  440,  475 

studies,  443-448 
Franklin,  Kenneth  L.,  103 
Franklin,  Richard,  467 
Franklin,  William  S.,  106 
Franz,  Shephard  I.,  109 
Frazer,  Joseph  C.  W.,  105 
Fredericq,  Pierre,  83 
Fredrick,  L.  W.,  293,  300 
Free,  Edward  E.,  77 
Frei,  Yael  F.,  publication,  366 
French,  B.  M.,  208 
French,  C.  Stacy,  ix,  76,  310,  317,  341,  354,  366 

publications,  366 

report  of  Director  of  Department  of  Plant 
Biology,  303-366 

studies,  37-38,  40,  319-320,  345-350 
Freudenberger,  L.  A.,  106 
Freundlich,  E.,  79 
Frew,  William  N.,  vii,  499 
Frick,  G.,  291 

publication,  289,  290 
Frothingham,  Arthur  L.,  Ill 
Fryklund,  V.  C,  193 
Fuchs,  Fritz,  87 
Fulk,  Harriet  M.,  310 
Fuller,  John  M.,  100 
Furlong,  Eustace  L.,  110 
Fuscaldo,  Kathryn,  439 


Gage,  Lyman  J.,  vii,  499 
Gaige,  F.  W.,  91 
Gale,  Henry  G.,  79 
Gamble,  James  L.,  100 
Gamburtsev,  G.  A.,  226 
Gamow,  George,  22,  81,  282 
Gardner,  Ernest  D.,  87 
Gardner,  Henry  B.,  92 
Garven,  H.  S.  D.,  100 
Gasic,  Gabriel,  83 
Gast,  Paul  W.,  178,  179 

publications,  197-198,  202,  289,  290 
Gates,  H.  S.,  19 

publication,  48,  50 
Gault,  Robert  H.,  109 
Gaviola,  Enrique,  81 
Gay,  Helen,  ix,  33,  71,  82,  438,  439,  440,  461,  475 

publications,  475 

studies,  441,  442,  466-474 
Gazin,  Charles  Lewis,  108 
Gee,  A.  Haldane,  97 
Gehlen,  K.  v.,  86,  208 

studies,  55,  154-155 
Geiling,  E.  M.  K.,  87 
Gerould,  John  H.,  97 
Gersh,  Isidore,  87,  97 
Gettens,  Rutherford  J.,  91 


514 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Gey,  G.  O.,  87 

Gibbs,  George,  107 

Gibbs,  Josiah  Willard,  13 

Gibbs,  Martin,  77 

Gibson,  James,  publication,  47,  48 

Gibson,  Ralph  E.,  85 

Giesecke,  A.  A.,  Jr.,  292 

Giesecke,  Albert  A.,  94 

Gifford,  Walter  S.,  v,  vi,  481 

Gilbert,  Carl  J.,  v,  73,  497 

Gilbert,  Cass,  vii 

Gilbert,  Christine,  105 

Gilbert,  Grove  Karl,  86 

Gilbert,  Perry  W.,  105 

Gilborn,  Steven  N.,  310 

Giles,  Norman  H.,  Jr.,  83 

Gillespie,  E.  Clark,  105 

Gillett,  Frederick  H.,  vii 

Gillies,  Gloria,  466,  467,  475 

Gillin,  John  P.,  91 

Gillman,  Joseph,  87 

Gilman,  Daniel  Coit,  vii,  3,  499,  500 

Gilmore,  Eugene  A.,  94 

Gish,  Oliver  H.,  80 

Gitlin,  G.,  87 

Glattfeld,  John  W.  E.,  77 

Glock,  Waldo  S.,  76 

Glover,  Timothy,  87,  392,  432 

Goddard,  Robert  H.,  106 

Godske,  C.  L.,  107 

Goedheer,  Joop  C,  102 

Goldberg,  Edward,  439,  440,  475 

studies,  443-448 
Goldberg,  I.,  190 
Goldberg,  Leo,  79 
Goldenweiser,  E.  A.,  94 
Golder,  Frank  A.,  90 
Goldforb,  Abraham  J.,  97 
Gollnow,  H.,  40 
Gomberg,  Moses,  106 

Gomes,  Alercio  M.,  publications,  48,  49,  50 
Goodrich,  Hubert  B.,  97 
Goodwin,  Richard  H.,  77 
Goranson,  Roy  W.,  85 
Gordon,  Myron,  97 
Gortner,  Ross  A.,  83 
Goss,  Charles  M.,  87 
Goss,  Richard  J.,  105 
Goss,  William  F.  M.,  107 
Gots,  Joseph  S.,  83 
Goubaud,  Antonio,  91 
Gowanlock,  James  N.,  97 
Gowen,  John  W.,  83 
Graf,  Ludwig  von,  83 
Graham,  A.  F.,  439 
Graham,  John  W.,  80,  103 
Gramm,  Charles  E.,  42 
Grant,  Verne  E.,  77 
Graton,  L.  C,  94 
Grave,  Caswell,  97 
Gray,  Donald  J.,  87 
Gray,  George  M.,  97 


Green,  Ronald,  103,  291 

publication,  289 

studies,  221-234 
Greene,  Evarts  B.,  90 
Greenewalt,  Crawford  H.,  v,  vi,  497 
Greenstein,  Jesse  L.,  viii,  78,  3,  12,  36,  46 

publications,  48,  49 

studies,  60,  8,  10-11,  17,  18,  19-20,  34 
Greenwald,  Gilbert  S.,  87 
Greenwood,  Hugh  J.,  viii,  85,  90,  193,  194,  208 

publication,  197,  202 

studies,  54,  82-88 
Gregory,  George  D.,  Ill 
Gregory,  Herbert  E.,  108 
Gregory,  Paul  W.,  87 
Gregory,  William  K.,  110 
Greig,  Joseph  W.,  85,  75,  160 
Griffin,  L.  E.,  63 
Griffin,  Mary  F.,  89 
Griffin,  Roger  F.,  102 

publication,  48 
Griffith,  Elmer  C,  94 
Grill,  Richard  D.,  433 
Groat,  George  Gorham,  94 
Grodzinski,  E.,  87 
Gross,  Samson  R.,  439 
Grun,  Paul,  76 
Gucker,  Frank  T.,  Jr.,  107 
Gudger,  Eugene  W.,  97 
Gum,  Colin  S.,  102 
Gump,  Dieter,  391 
Gunn,  James  E.,  11,  46 
Gunn,  Ross,  81 
Gustafson,  Florence,  100 
Gustafsson,  Ake,  83 
Gutenberg,  Beno,  108 
Guthe,  Carl  E.,  91 
Guthrie,  Charles  C,  83,  109 
Guttmacher,  Alan  F.,  88 
Guttmacher,  Manfred  S.,  88 

Habermann,  Helen  M.,  77 
Hackett,  Charles  W.,  90 
Hafstad,  Lawrence  R.,  80 
Hagerty,  James  Edward,  94 
Haible,  William  E.,  370 
Haig,  Robert  M.,  94 
Hale,  George  Ellery,  5,  8,  55,  78,  5 
Hales,  Anton  L.,  81 
Halferdahl,  S.  B.,  104 
Halkka,  Olli,  83 
Hall,  Harvey  M.,  76 
Hall,  John  S.,  81 

publication,  288,  289 

studies,  295-301 
Hall,  Richard  C,  103,  291 
Hall,  S.  Stanley,  109 
Halldal,  Per,  77 
Halpern,  Abraham  M.,  91 
Halvorson,  H.  O.,  publication,  289 
Hameka,  H.  F.,  132 
Hamilton,  Tom  S.,  100 


INDEX    OF    NAMES 


515 


Hammond,  Matthew  Brown,  94 
Hancock,  Glover  D.,  94 
Handwerker,  D.,  134 
Haney,  Lewis  Henry,  94 
Hanke,  Lewis  U.,  91 
Hanna,  Hugh  Sisson,  94 
Hardin,  Garrett  J.,  76 
Hargitt,  George  T.,  97 
Haro,  Guillermo,  79,  18 
Harrington,  Mark  R.,  91 
Harris,  J.  Arthur,  83,  437 
Harris,  Jerome  S.,  105 
Harris,  John  E.,  97 
Harris,  John  W.  S.,  88,  371,  432 

studies,  425-426 
Harrison,  Fielding  H.,  Ill 
Harrison,  J.  A.,  97 
Harrison,  Ross  G.,  70,  109 
Hart,  Pembroke  Jones,  103 
Hart,  Richard  W.,  310 

study,  365 
Hart,  Stanley  R.,  viii,  46,  80,  222,  291 

publication,  289 

studies,  173-179,  234-239 
Hartman,  Carl  G.,  87,  424 
Hartman,  Frank  A.,  109 
Hartmeyer,  Robert,  97 
Harvey,  E.  Newton,  97 
Hashimoto,  Kazuo,  103 
Haskins,  Caryl  P.,  v,  vi,  x,  497 

publications,  477-478 

Report  of  the  President,  1-112 
Haskins,  Charles  H.,  90 
Haskins,  Edna  F.,  publication,  477 
Haskova,  V.,  391,  430 
Hasse,  Adelaide  R.,  94 
Hatai,  Shinkishi,  97 
Haxo,  Francis  T.,  77,  343 
Hay,  John,  vii,  499,  500 
Hay,  Oliver  P.,  110 
Hayase,  L,  292 

studies,  46,  234-239 
Hayes,  Donald  S.,  46 
Hayes,  Frederick  R.,  97 
Hayford,  John  F.,  106 
Haynie,  William  H.,  42 
Hazlehurst,  John,  46 
Heard,  Osborne  O.,  87 
Hebel,  John  W.,  Ill 
Heidel,  William  A.,  91 
Heifer,  H.  Lawrence,  103,  10,  11,  45 

studies,  60 
Helwig,  Edwin  R.,  97 
Hempl,  George,  111 
Henard,  Kenneth  R.,  x 
Hendler,  R.  W.,  285,  292 
Hendricks,  Sterling  B.,  85 
Hendrix,  Don  O.,  68,  69,  45,  46 
Henize,  Karl  G.,  102 
Henry,  Barklie  McKee,  v,  vi,  497 
Herb,  Raymond  G.,  81 
Herbig,  George,  79 


Herrick,  Myron  T.,  vii 
Herriott,  Frank  I.,  94 
Herrmann,  Heinz,  370,  371 
Hershey,  Alfred  D.,  ix,  71,  72,  82,  259,  269,  373, 
381,  438,  439,  440,  466,  475 

publications,  475 

studies,  33-35,  442,  443-448 
Hertig,  Arthur  T.,  88,  432 
Hertzsprung,  Ejnar,  57,  59,  79 
Herz,  N.,  237 
Herzog,  Emil,  46 

publication,  48 

studies,  32-33 
Hess,  Harry  H.,  86,  62 
Hess,  Victor  F.,  81 
Hess,  Walter  N.,  97 
Hester,  Joseph  A.,  Jr.,  105 
Heuser,  Chester  H.,  87,  432 
Hewitt,  Abram  S.,  vii 
Hewitt,  Richard  E.,  publication,  477 
Heydenburg,  Norman  P.,  viii,  80,  291 

publication,  288,  289 

studies,  242-244 
Hibbard,  Benjamin  H.,  94 
Hibben,  James  H.,  85 
Hickox,  Joseph  O.,  6,  46 
Hicks,  C.  S.,  100 
Hiesey,  William  M.,  ix,  76,  310,  366 

studies,  40,  308-309,  311-320,  323-333 
Higgins,  Harold  L.,  100 
Higginson,  Henry  L.,  vii,  1,  499 
Hill,  George  W.,  107 
Hill,  John  H.,  42 
Hill,  Robert,  77 
Hill,  Roscoe  R.,  90 

Hiltner,  W.  A.,  publications,  47,  48,  49 
Hinds,  Norman  E.  A.,  107,  110 
Hines,  Marion,  88 
Hitchcock,  Ethan  A.,  vii,  499 
Hitchcock,  Fred  A.,  100 
Hitchcock,  Henry,  vii 
Hoagland,  Henry  E.,  94 
Hobbs,  William  H.,  108 
Hodder,  Frank  H.,  90 
Hodell,  Charles  W.,  Ill 
Hodge,  Paul  W.,  46 

publications,  48 

studies,  30 
Hoering,  Thomas  C,  viii,  85,  188,  189,  193,  208 

studies,  54,  56,  184-187,  190-191 
Holden,  Roy  Jay,  94 
Hollaender,  Alexander,  72,  88 
Holland,  John  H.,  xi,  107 
Holland,  Thomas  Erskine,  111 
Hollander,  Jacob  H.,  94 
Hollinger,  J.  W.,  220,  292 
Holmberg,  Erik,  79 
Holmes,  William  H.,  Ill 
Holmgren,  K.  Paul,  309,  310,  312 

studies,  320-323 
Holt,  A.  Stanley,  77 
Homans,  John,  100 


516 


CAKNEGIE    INSTITUTION     OF     WASHINGTON 


Hooker,  Davenport,  97 
Hoover,  Herbert,  vii 
Hopkins,  Dwight  L.,  97 
Hornblower,  Marshall,  x 
Horvath,  Beni,  105 
Hoskins,  Roy  G.,  100 
Hostetter,  John  C,  85 
Hotta,  Y.,  285,  292 
Houziaux,  Leo,  102,  7,  46 

publications,  48,  49 

studies,  18 
Howard,  Edgar  B.,  91,  110 
Howard,  Hildegarde,  110 
Howard,  Leland  O.,  109 
Howard,  Robert  F.,  viii,  78,  7,  39,  45,  46 

publication,  49 

studies,  56 
Howard,  William  T.,  91 
Howe,  Henry  M.,  106 
Howe,  William  Wirt,  vii,  90,  499 
Howell,  B.  F.,  Jr.,  publications,  289,  290 
Howell,  William  H.,  109 
Hoyle,  Fred,  78 

publications,  48,  49 
Hu,  Hsen  Hsu,  110 
Hubble,  Edwin  P.,  8,  61,  78,  30,  31 
Huber,  P.,  211,  242 
Huebner,  Solomon  S.,  94 
Huggett,  A.  St.  G.,  88 
Hughes,  D.  S.,  228 
Hughes-Schrader,  Sally,  83 
Hull,  William  I.,  90 
Humason,  Milton  L.,  78,  19,  46 

publications,  49,  50 
Hunt,  John  M.,  181 
Huntington,  Ellsworth,  77 
Hurley,  P.  M.,  237 
Huskins,  C.  Leonard,  84 
Hutchinson,  Charles  L.,  vii,  499 
Hyde,  Walter  W.,  Ill 
Hytonen,  Kai,  104,  56-57,  65,  208 

Iben,  Ikco,  13 
Iddings,  Joseph  P.,  86 
Imlay,  Marjorie  E.,  192,  208 
Ingalls,  Walter  Renton,  94 
Ingerson,  Earl,  85 
Ingraham,  Laura  J.,  476 
studies,  443-448 

Jack,  Theodore  H.,  94 

Jackson,  E.  Dale,  193 

Jackson,  Robert  Tracy,  97 

Jacobs,  Merkel  H.,  97 

Jager,  Emilie,  86 

Jahns,  R.  H.,  193 

James,  H.,  191 

James,  Herman  G.,  90,  112 

Jameson,  John  F.,  89 

Jeans,  Sir  James  Hopwood,  79 

Jeffers,  Hamilton  M.,  publication,  47,  49 

Jenkins,  Heroy,  99 


Jenks,  J.  W.,  92 
Jennings,  Herbert  S.,  109 
Jennings,  Jesse  D.,  91 
Jernegan,  Marcus  W.,  90 
Jessup,  Walter  A.,  vii 
Jewett,  Frank  B.,  vii 
Jochelson,  Waldemar,  111 
Johansen,  Donald  A.,  77 
Johnson,  Allah  C.,  Ill 
Johnson,  Ellis  A.,  80 
Johnson,  Emory  R.,  93 
Johnson,  P.  A.,  292 
Johnson,  Thomas  H.,  81 
Johnston,  Ivan  M.,  77 
Johnston,  John,  85 
Jones,  D.  H.  P.,  17,  46 

publication,  49 
Jones,  Edward  D.,  94 
Jones,  Harry  C.,  106 
Jones,  Henry  H.,  475 
Jones,  Norris,  97 
Jones,  T.  J.,  94 
Jordan,  Harvey  E.,  97 
Jorgensen,  E.,  97 
Jorgensen,  Erik  G.,  77 

publication,  366 
Joslin,  Elliott  P.,  100 
Journal  of  Geophysical  Research,  192 
Joy,  Alfred  H.,  78,  46 

publications,  49 
Jugaku,  Jan,  46 

publications,  49,  50 

studies,  60,  8-10,  36 
Jungwirth,  Christoph,  439 


Kafer  (Boothroyd),  Etta,  103 
Kaiser,  Irwin  H.,  88 
Kaler,  J.,  44 
Kamitsuki,  A.,  292 
Kapteyn,  Jacobus  C,  79 
Karsner,  Howard  T.,  100 
Katem,  Basil  N.,  46 
Kato,  Yoshihiro,  105 
Katsh,  Seymour,  88 
Kaudewitz,  Fritz,  84 
Kaufmann,  Berwind  N.,  439,  440 
Kaufmann,  Berwind  P.,  ix,  33,  69-70,  71,  72,  82, 
439,  440,  461,  475 

publications,  475 

report  of  Director  of  Department  of  Genetics, 
435-476 

studies,  441,  442,  466-474 
Kearns,  Charles  E.,  19 

publication,  49 
Keck,  David  C,  76 
Keenan,  Philip  C,  79 

publication,  48,  49 
Kegel,  W.,  44 
Keibel,  Franz,  88 
Keith,  Mackenzie  L.,  10 4 
Kellner,  Carl,  97 


INDEX    OF    NAMES 


517 


Kellogg,  Louise  P.,  90 

Kellogg,  Remington,  110 

Kemp,  Tage,  84 

Kempner,  Ellis  S.,  103 

Kempton,  J.  EL,  92 

Kendrick,  Benjamin  B.,  90 

Kennedy,  Arthur  G.,  112 

Kennelly,  Arthur  E.,  81 

Kertz,  Walter,  241 

Kew,  William  Stephen  Webster,  108 

Kidder,  Alfred  V.,  90 

Kilborn,  Leslie  G.,  101 

Kimmel,  Charles  B.,  432 

King,  Arthur  S.,  78 

King,  Clyde  L.,  94 

King,  Robert  B.,  78 

King,  T.  J.,  430 

Kingsbury,  Benjamin  F.,  88 

Kippenhahn,  Rudolph,  46 

studies,  37 
Kistiakowsky,  George  B.,  106 
Klein,  Julius,  94 
Klinger,  F.  L.,  185 
Knight,  Wiley,  Jr.,  310 
Kofoid,  Charles  A.,  109 
Kok,  Bessel,  102 
Kokrikos,  Efstathios  J.,  105 
Koller,  P.  C,  84 
Konigsberg,  Irwin  R.,  ix,  87,  369,  432 

publications,  431 

studies,  64-67,  370,  397-400 
Kopac,  Milton  J.,  97 
Koranda,  John,  312,  313 
Korff,  Serge  A.,  81 
Kouvo,  Olavi,  47,  178,  179 

publications,  197,  198,  202,  289,  290 

studies,  56,  178,  179 
Kowal,  Charles  T.,  17,  46 

studies,  60 
Kracek,  Frank  C,  85 
Kraft,  Robert  P.,  viii,  74,  78,  28,  46 

publications,  47,  49 

studies,  11,  20,  25 
Krauss,  Robert  W.,  77 
Krishnan,  T.,  292 
Krizenecky,  Jaroslav,  84 
Kroeger,  H.,  430 
Kuiper,  Gerard  P.,  79 
Kulangara,  Abraham  C,  88,  420,  421 

publication,  431 
Kullerud,  Gunnar,  viii,  85,  139,  157,  158,  193, 
194,  208 

publications,  198,  199,  202 

studies,  50,   55,    143-152,    154-155,    160-161, 
163-165 
Kulp,  J.  L.,  237 

Kundu,  M.  R.,  publication,  289 
Kunkel,  Beverly  W.,  97 
Kuo,  Zing  Yang,  101 
Kupke,  Donald  W.,  76 
Kupres,  F.  J.,  394 
Kurshan,  Jane,  476 


la  Cour,  D.,  81 
Laetsch,  W.  M.,  323 
LaFleur,  Virginia,  430,  432 
LaMer,  Victor  K.,  84 
La  Motte,  Robert  Smith,  110 
Lancaster,  Henry  C,  112 
Landauer,  Walter,  101 
Landolt,  Elias,  77 
Laney,  Francis  Baker,  94 
Langley,  Samuel  P.,  vii,  499 
Langworthy,  Orthello  R.,  88 
Lapp,  John,  94 
Larsen,  Esper  S.,  Jr.,  85 
Larsen,  Victor  R.,  439 

publication,  475 
Larson,  Laurence  M.,  94 
Lash,  J.  W.,  430 
Lashley,  Karl  S.,  97 
Latarjet,  Raymond,  84 
Latimer,  Paul  H.,  102,  310 
Laufer,  Hans,  88 
Laughlin,  H.  H.,  83 
LaVelle,  Arthur,  371,  432 

studies,  393-394 
Lawrence,  Ernest  O.,  vii 
Lawson,  Andrew  C,  108,  176 
Leahy,  John  J.,  103 
Learned,  Marion  D.,  90 
Le  Compte,  Marius,  97 
Lee,  Milton  O.,  101 
Lee,  Robert  C,  100 
Lehmer,  Derrick  N.,  107 
Leighton,  Philip  A.,  106 
Leighton,  Robert  B.,  79,  3,  41 
Leitch,  James  L.,  97 
Leith,  C.  K.,  94 
Leland,  Waldo  G.,  89 
Le  Maitre,  Abbe,  79 
Lenhoff,  Howard  M.,  103 

publication,  289 
Lenz,  Andrew  N.,  310 
Lessof,  Maurice,  390 

publication,  431 
Leuschner,  A.  O.,  79 
Levan,  Albert,  84 
Levinthal,  Cyrus,  84 
Lewis,  Charlton  M.,  77 
Lewis,  Harlan,  77 
Lewis,  Ivey  F.,  97 
Lewis,  Margaret  R.,  64,  87 
Lewis,  Warren  H.,  64,  87 
Libby,  Orin  G.,  90 
Libby,  Willard  F.,  86 
Lillie,  Frank  R.,  67,  70,  97,  369 
Lillie,  Ralph  S.,  109 
Lincoln,  J.  Steward,  92 
Lindbergh,  Charles  A.,  vii 
Lindblad,  Bertil,  79 
Lindsay,  William,  vii,  499 
Lindsley,  Donald  H.,  104,  208 

studies,  54,  100-106 
Linton,  Edwin,  97 


518 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Lipman,  Charles  B.,  98 
Lippincott,  Isaac  P.,  94 
Little,  C.  A.,  Jr.,  292 
Little,  C.  C,  72,  84,  437 
Littlehales,  George  W.,  90 
Livingston,  Burton  E.,  35,  76 
Lloyd,  Francis  E.,  35,  76 
Lockhart,  Oliver  C,  94 
Lodge,  Henry  Cabot,  vii 
Loeb,  Leo,  109 
Loeffler,  Josef  E.,  102 
Lombard,  Robert  H.,  85 
Long,  Esmond  R.,  77 
Long,  Joseph  A.,  109 
Longley,  William  Harding,  96 
Longyear,  John  M.,  Ill,  92 
Loomis,  Alfred  L.,  v,  vi,  497 
Loos,  Isaac  A.,  94 
Lorentz,  H.  A.,  81 
Lothrop,  Samuel  K.,  92 
Louderback,  George  D.,  86 
Loughlin,  Gerald  Francis,  94 
Lovett,  Robert  A.,  v,  vi,  481,  497 
Lovtrup,  Soren,  103 
Low,  Seth,  vii,  499 
Lowe,  Elias  A.,  112 
Lowell  Observatory,  295,  297 
Lowen,  A.  Louise,  46 
Lucke,  Balduin,  98 
Lundell,  Cyrus  L.,  92 
Lundholm,  Helge,  101 
Lundmark,  Knut,  79 
Lunn,  Arthur  C,  107 
Luria,  S.  E.,  83 
Lutz,  Frank  E.,  83 
Luyten,  W.  J.,  79,  18 
Lynden-Bell,  Donald,  14,  46 
studies,  61,  25-26,  37 

McCabe,  David  A.,  94 
McCarthy,  Brian  J.,  viii,  80,  291 

publications,  289 

studies,  43-44,  244-288 
McCarthy,  Charles,  94 
McClelland,  James  Farley,  94 
McClendon,  Jesse  F.,  98 

McClintock,  Barbara,  ix,  71,  72,  82,  438,  439, 
440,  476 

publication,  475 

studies,  31-82,  442,  448-461 
McClung,  Clarence  E.,  109 
McClure,  Frank  T.,  81,  292 

publication,  289 
McCollum,  Ruth  L.,  x,  69,  70-71 
McConnell,  Anne,  47 
McCoy,  Oliver,  98 
MacCurdy,  Hansford,  109 
McCutchen,  George,  94 

MacDonald,  Gordon  J.  F.,  86,  168-169,  170,  208 
McDonald,  Margaret  R.,  ix,  88,  82,  438,  439, 
467,  470,  476 

publication,  475 

studies,  441-442,  461-466 


MacDougal,  Daniel  T.,  35,  76 
MacDowell,  E.  C,  72,  82,  437 
McGee,  J.  D.,  xi,  81 
MacGinitie,  Harry  D.,  110 
McHugh,  Keith  S.,  v,  vi,  497 
Mcllwain,  C.  E.,  239 

publication,  289 
McKee,  Edwin  D.,  110 
McKee,  Marguerite  M.,  89 
McKenzie,  John,  88 

publication,  431,  477 
MacKenzie,  William  S.,  85 
McLaughlin,  Andrew  C,  89 
McLaughlin,  John  J.  A.,  publication,  477 
McLean,  S.  J.,  94 
MacLeod,  Grace,  101 
McLeod,  Guy  C,  102 
McMath,  Robert  R.,  79 
MacMillan,  William  D.,  107 
McNamara,  D.  H.,  41 
McQuillen,  Kenneth  R.,  81,  285,  292 
McQuown,  Norman  A.,  92 
MacVeagh,  Wayne,  vii,  499 
McVey,  F.  L.,  94 

Macdowall,  Fergus  D.  H.,  102 

Macelwane,  James  B.,  S.J.,  108 

Macfarlane,  John  M.,  77 

Madsen,  Axel,  77 

Maestre,  Leonard  A.,  publication,  49 

Maguire,  John  D.,  112 

Mahan,  Alfred  T.,  90 

Makemson,  Maud  Worcester,  92 

Maling,  John,  354,  360,  362 

Mall,  Franklin  P.,  62,  63,  86 

Mandell,  Joseph  D.,  103,  259,  269,  373,  381 

Mane,  J.  Ignacio  Rubio,  91 

Mann,  Albert,  96 

Manning,  William  R.,  90 

Manning,  Winston  M.,  76 

Manter,  Harold  W.,  98 

Margolin,  Paul,  439 

Mark,  Edward  L.,  84,  109 

Markee,  Joseph  E.,  88 

Marsh,  Gordon,  98 

Martin,  James  C,  98 

Martin,  Paul  S.,  92 

Martin,  Percy  A.,  112 

Marton,  L.  L.,  publication,  288,  289 

studies,  295-301 
Maruyama,  Keizo,  476 

publications,  475 

studies,  466-474 
Marvin,  Cloyd  Heck,  98 
Marvin,  Horace  N.,  84 
Mason,  Eleanor  D.,  101 
Mason,  Herbert  L.,  77 
Mast,  Samuel  O.,  98 
Mathews,  Jon,  20 
Mathis,  John  S.,  27 
Matsushima,  Satoshi,  46 
Matthai,  George,  98 
Matthews,  R.  E.  F.,  277,  285 


INDEX    OF    NAMES 


519 


Matthews,  Thomas  A.,  102,  34-35 

Maurette,  C,  228 

Maxson,  John  H.,  108 

Mayall,  N.  U.,  79,  32 

Mayer,  Alfred  G.,  96 

Means,  James  H.,  101 

Medes,  Grace,  98 

Meek,  Seth  E.,  98 

Mehrizi,  Ali,  432 

Mellon,  Andrew  W.,  vii 

Mendel,  Lafayette  B.,  62-63,  108 

Mendez,  Manuel  E.,  46 

Meng,  John  J.,  90 

Merriam,  C.  Hart,  109 

Merriam,  Charles  W.,  98 

Merriam,  John  Campbell,  vii 

Merrill,  Paul  W.,  78 

publications,  48,  49 
Merwin,  Herbert  E.,  85 
Metz,  Charles  W.,  88,  87,  437 
Meyer,  Arthur,  88 
Meyer,  B.  H.,  93 
Meyer,  Mary  Henderson,  101 
Meyer,  Robert  P.,  publication,  289,[290,  477 
Michelson,  A.  A.,  8,  11,  79 
Midgley,  John  E.,  103,  291 

studies,  43-44,  244-288 
Mihalas,  Dimitri  M.,  46 

publication,  49 
Miles,  Mary  Jane,  192 
Miles,  Walter  R.,  100 
Miller,  Carey  D.,  101 
Miller,  Dayton  C,  79 
Miller,  E.  T.,  94 
Miller,  Harry  M.,  Jr.,  98 
Miller,  Margaret  Carnegie,  v,  vi 
Miller,  Roswell,  vii 
Miller,  William  C,  46 
Millikan,  Robert  A.,  81 
Millis,  H.  A.,  94 
Mills,  B.  Y.,  82 
Mills,  Darius  O.,  vii,  499 
Mills,  John  W.,  96 
Milner,  Harold  W.,  ix,  76,  310,  366 

studies,  40-41,  308-309,  311-312,  313-319 
Milner,  Max,  77 
Milner,  Peter,  xi,  109 
Minkowski,  Rudolph  L.,  78,  26,  33,  34 

publications,  49 
Miranda,  Maury,  280,  285 
Mitchell,  S.  A.,  79 
Mitchell,  S.  Weir,  vii,  109,  499,  500 
Mitchell,  Walter  E.,  Jr.,  41,  42 
Mitchell,  Wesley  C,  94 
Mitra,  S.  K.,  82 
Moenkhaus,  William  J.,  84 
MofTett,  Ben  C,  Jr.,  105 
Moh,  Giinter,  104,  208 

studies,  55,  151-152 
Montague,  Andrew  J.,  vii 
Montalenti,  G.,  84 
Moody,  Sheila  J.,  371,  432 
Moore,  Blaine  F.,  94 


Moore,  Eliakim  H.,  107 

Moore,  George  T.,  77 

Moravek,  Vladimir,  77 

Morey,  George  W.,  6,  85,  194 

Morgan,  Henry  S.,  v,  vi,  481,  497 

Morgan,  Thomas  Hunt,  85 

Morgan,  W.  W.,  79,  25 

Morgulis,  Sergius,  98,  101 

Mori,  Tom,  88,  432 

Morimoto,  Nobuo,  104,  193,  194,  208 

publications,  194,  198,  201,  202 

studies,  55,  139-141,  143-144 
Morison,  George  S.,  107 
Moritz,  Charles  E.,  98 
Morley,  E.  W.,  8 
Morley,  Frank,  107 
Morley,  Sylvanus  G.,  90,  91 
Morris,  Ann  Axtell,  92 
Morris,  Earl  H.,  90,  91 
Morrow,  William  W.,  vii,  499 
Morse,  Albert  P.,  109 
Morse,  Harmon  N.,  106 
Mortensen,  Theodor,  98 
Moser,  Hermann,  103 
Mosig,  Gisela,  476 
Mossman,  Harland  W.,  88 
Moulton,  Forest  Ray,  86 
Moyse,  A.,  310 
Muan,  A.,  101 
Muchmore,  William  B.,  88 
Mudd,  Seeley  G.,  v,  vi 
Mukai,  Frank  H.,  103 
Muller,  H.  J.,  84 
Miiller,  W.  Max,  111 
Mulnard,  Jacques,  88 
Mun,  Alton  M.,  88,  432 

publication,  431,  477 

studies,  404-406 
Munch,  Guido,  viii,  74,  78,  46 

publication,  49 

studies,  56-57,  6,  7-8,  25,  26,  27,  35,  37,  38-39 
Munch,  Luis,  26,  46 
Munroe,  Charles  E.,  94 
Muntzing,  Arne,  328 
Muratori,  G.,  88 
Murlin,  John  R.,  101 
Murphy,  Thomas,  108 
Murray,  Bruce  C,  57,  6,  42 
Murray,  C.  A.,  45 
Murschhauser,  Hans,  101 
Mussen,  Aubrey  T.,  109 
Musse}^,  Henry  R.,  94 
Muster,  John  *C,  41,  42 
Myers,  Jack  E.,  77,  344,  346,  350 
Myers,  William  L,  v,  vi,  497 

Nabours,  Robert  K.,  84 
Nagata,  T.,  82 
Nanda,  Jatinder  Nath,  103 
Narbaitz,  Roberto,  88 
publication,  431,  477 
Nardin,  W.  T.,  94 
National  Bureau  of  Standards,  139,  159,  168,  295 


520 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


National  Geographic  Society-Palomar  Observa- 
tory Sky  Survey,  18,  24 
Naughton,  M.  A.,  393 
Naval  Observatory,  U.  S.,  295 
Neel,  James  V.,  84 
Neher,  H.  Victor,  106 
Neill,  Catherine,  88 
Neuvonen,  Kaarlo  J.,  104 
Newcomb,  Simon,  56 
Newcombe,  Howard  B.,  84 
Newhouse,  Walter  H.,  108 
Nichols,  Earnest  F.,  79 
Nichols,  Edward  L.,  106 
Nichols,  Richard  F.  F.,  x 
Nicholson,  Frank,  310 

studies,  312-313,  323-325 
Nicholson,  Seth  B.,  78,  5,  46 

publications,  49 
Nicolaysen,  Louis  Otto,  104,  122,  190 
Nicoll,  Paul  A.,  98 
Niggli,  Paul,  86 
Nipher,  Francis  E.,  106 
Nirenberg,  M.,  282,  285,  287,  292,  384 
Nishida,  Minoru,  46 

studies,  36 
Nitzulescu,  Julius,  101 
Noble,  Levi  F.,  108 
Nobs,  Malcolm  A.,  76,  ix,  310 

publication,  366 

studies,  41,  309,  311-319,  320-323,  325-333 
Noguchi,  Hideyo,  109 
Nomoto,  M.,  243 
Nordenskiold,  Hedda,  77 
Nordmann,  Martin,  88 
North,  S.  N.  D.,  93 
Norton,  Garrison,  v,  vi,  497 
Norton,  Robert  EL,  14,  46 
Noyes,  Arthur  A.,  106 
Noyes,  Robert  W.,  41 
Nur,  Uzi,  333,  334 
Nygren,  Axel,  77,  309,  310,  312 
Nys,  Ernest,  112 


O'Dell,  Charles  Robert,  102,  24,  28,  46 

Ohman,  Y.,  79 

Oke,  J.  Beverley,  viii,  78,  11,  46 

publications,  49,  50 

studies,  12,  19,  32,  35-36,  39 
Olsen,  Kirsten,  476 
Ondik,  Helen,  139 
O'Neale,  Lila  M.,  92 
Oort,  Jan  H.,  xi,  79,  26,  35,  46,  215,  216 
Oosterhoff,  P.  Th.,  79 
O'Rahilly,  Ronan,  64,  88,  371,  432 

publication,  431 

studies,  427 
Orts  Llorca,  F.,  88 
Orville,  Philip  M.,  104 
Osborn,  Elbert  F.,  85 
Osborn,  Henry  Fairfield,  110 
Osborn,  William  Church,  vii 


Osborne,  Thomas  B.,  108 
Osburn,  Raymond  C,  98 
Osgood,  Charles  G.,  112 
Osgood,  Herbert  L.,  90 
Osterbrock,  Donald  E.,  78 
Osterhout,  Winthrop  J.  V.,  77 
Ovary,  Zoltan,  373,  375,  392 
Overman,  Ralph  T.,  192 
Overton,  James  B.,  77 
Owren,  Leif,  103 


Packard,  Earl  L.,  110 
Pahlow,  Edwin  W.,  90 
Paigen,  Kenneth,  103 
Painter,  Theophilus  S.,  84 
Palmeira,  R.  A.  R.,  219 
Papaconstantinou,  John,  88,  371,  407 

studies,  407-408 
Pardee,  Arthur  B.,  430,  439 
Parker,  David  W.,  89 
Parker,  Edward  W.,  93 
Parker,  Patrick  L.,  viii,  85,  193,  208 

studies,  53-54,  56,  181-184,  187-190 
Parker,  Robert  A.  R.,  46 

studies,  60,  10-11,  12,  25 
Parkinson,  W.  C,  80 
Parkinson,  W.  D.,  103 
Parmelee,  James,  vii 
Parsons,  Wm.  Barclay,  vii 
Paton,  Stewart,  vii 
Patten,  William,  110 
Patterson,  C.  C,  86 
Patton,  Donald  J.,  x 
Paul,  W.  M.,  88 
Pauling,  Linus,  20-21 
Paullin,  Charles  O.,  89 
Pawsey,  J.L.,  82 
Paxson,  Frederic  L.,  90 
Payne,  Fernandus,  98 
Peabody,  Francis  W.,  101 
Pearl,  Raymond,  84,  110 
Pearse,  Arthur  S.,  92,  98 
Pease,  F.  G.,  78 
Peirce,  George  J.,  77 
Pender,  Harold,  107 
Peoples,  J.  A.,  Jr.,  192 
Peoples,  Rowena  E.,  192 
Pepper,  George  W.,  vii 
Perek,  L.,  79 
Perkins,  Henry  F.,  98 
Perkins,  Richard  S.,  v,  vi,  481,  497 
Perlman,  Selig,  94 
Perlmann,  P.,  430 
Perreault,  W.  J.,  476 

publication,  475 
Perret,  Frank  A.,  86 
Pershing,  John  J.,  vii 
Persons,  Warren  Milton,  94 
Pestell,  Wilbur  A.,  69,  71,  310 
Petrik,  Josef  M.,  101 
Petterssen,  Sverre,  108 


INDEX    OF    NAMES 


521 


Pettit,  Edison,  78,  5 

publication,  49 
Phage  Information  Service,  440 
Phelps,  Earle  B.,  109 
Philbrick,  Francis  S.,  90 
Phillips,  Alexander  Hamilton,  98 
Phillips,  Gerald  C,  103 
Phillips,  John  B.,  95 
Phillips,  John  C,  110 
Phillips,  Ulrich  B.,  95 
Pickard,  Greenleaf  W.,  82 
Pidgin,  Charles  ¥.,95 
Pieper,  George  F.,  103 
Pierson,  William  W.,  Jr.,  112 
Piggot,  Charles  S.,  85 
Pinchot,  Gifford,  77 
Pinson,  W.  H.,  237 
Pitts,  Robert  F.,  98 
Pizzella,  G.,  239 

publication,  289,  290 
Plaut,  L.,  79 
Plehn,  Carl  C,  95 
Plough,  Harold  H.,  98 
Pogo,  Alexander,  78,  91 
Pohn,  Howard  A.,  42 
Pollock,  Harry  E.  D.,  xi,  90 

publication,  477 
Popenoe,  Wilson,  92 
Popper,  Daniel  M.,  42 
Porter,  James  P.,  109 
Posnjak,  Eugene,  85 
Potapenko,  Gennady  W.,  106 
Potter,  J.  S.,  72 
Potts,  Frank  M.,  98 
Poulson,  Donald  F.,  88 
Powell,  Fred  Wilbur,  95 
Powell,  Wilson  M.,  82 
Powers,  Philip  B.  A.,  98 
Prasil,  Anthony  J.,  42 
Pratt,  Henry  S.,  98 
Pratt,  Joseph  H.,  101 
Pratt,  Joseph  Hyde,  95 
Pray,  Thomas  R.,  77,  310 
Prentis,  Henning  W.,  Jr.,  vii 
Preston,  George  W.,  Ill,  102 

publication,  49 
Price,  A.T.,82 
Price,  Llewellyn  I.,  110 
Pritchett,  Henry  S.,  vii,  1 
Proskouriakoff,  Tatiana,  x,  91 

publication,  477 
Puckett,  E.  P.,  95 
Pumpelly,  Raphael,  111 
Purgathofer,  A.  Th.,  44 
Putnam,  Herbert,  111 

Rabi,  I.  I.,  106 
Ramberg,  Hans,  86 

Ramdohr,  Paul,  xi,  86,  101,  149,  153,  185,  193, 
208 

publication,  201,  202 

studies,  50,  55-56,  163-1G5 


Ramsey,  Elizabeth  M .,  ix,  87,  425,  429,  432 

publications,  431 

studies,  423-425,  42G-427 
Ramsey,  Norman,  82 
Rankin,  George  A.,  7,  85 
Raper,  Charles  Lee,  95 
Ratcliffe,  J.  A.,  82 
Rawles,  Mary  E.,  ix,  87,  414,  432 

studies,  406-407 
Rawles,  William  A.,  95 
Rawson,  D.  E.,  193 
Raymond,  Harry,  99 
Redfield,  Robert,  92 
Reeves,  Jesse  S.,  90 
Reeves,  Ruth,  92 

Reichen,  L.  E.,  publication,  199-200,  201,  202 
Reichert,  Edward  T.,  109 
Reid,  Harry  Fielding,  108 
Reighard,  Jacob  E.,  98 
Reinke,  Edwin  E.,  98 
Remsen,  Ira,  106 
Rentschler,  Gordon  S.,  vii 
Rever,  Arthur  G.,  423,  433 
Reynolds,  Samuel  R.  M.,  87,  424,  432 
Rhoades,  Marcus  M.,  84 
Richards,  Donald  L.,  xi 
Richards,  Oscar  W.,  98 
Richards,  Theodore  W.,  106 
Richardson,  Robert  S.,  78 
Richter,  Charles  F.,  108 
Richter,  Curt  P.,  88 
Richter,  Maurice  N.,  84 
Rick,  Charles  M.,  313 
Ricketson,  Oliver  G.,  Jr.,  91 
Riddle,  Oscar,  83,  437 
Ries,  Heinrich,  95 
Riley,  Gordon  A.,  98 
Riley,  Thomas  James,  95 
Ripley,  W.  Z.,  93 
Ritchey,  G.  W.,  78 
Ritzman,  E.  G.,  101 
Robbins,  C.  R.,  studies,  54,  106-107 
Robertis,  Eduardo  de,  88 
Roberts,  Henry  B.,  91 
Roberts,  Howard  S.,  85 

Roberts,  Richard  B.,  viii,  80,  219,  291,  373,  382, 
430,  445 

publications,  289,  290 

studies,  43-44,  244-288 
Roberts,  Stuart  L.,  39,  46 
Robertson,  James  A.,  90 
Rock,  John,  88 
Rockefeller,  David,  vii 
Rodgers,  Alexander  W.,  102 

publication,  49 
Rodriguez  B.,  A.,  231,  291 

publications,  290 

studies,  221-234 
Roedder,  Edwin  W.,  104 
Roesel,  Catherine,  103 
Rogers,  Malcolm  J.,  110 
Rogers,  Robert  W.,  90 


522 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Rogerson,  John  B.,  Jr.,  102,  7 

Rojas,  Hector  R.,  103,  291 

Roll,  Ronald  E.,  42 

Rolph,  F.  W.,  101 

Romberg,  Arnold,  108 

Rooney,  William  J.,  80 

Roosen-Runge,  Edward,  88 

Root,  Elihu,  vii,  499,  500 

Root,  Elihu,  Jr.,  v,  vi,  481,  497 

Root,  Howard  F.,  101 

Rosales,  Juan  de  Dios,  92 

Rose,  Joseph  N.,  77 

Roseboom,  Eugene  H.,  Jr.,  104,  153 

publication,  200-201,  202 
Rosenwald,  Julius,  vii 
Ross,  Frank  E.,  79 
Rosseland,  S.,  79 
Rossi,  Bruno,  82 
Roth,  Paul,  101 
Rowe,  Leo  S.,  112 
Rowse,  John,  432 
Roy,  Arthur  J.,  99 
Roys,  Ralph  L.,  91 

publication,  477 
Rubenstein,  I.,  publication,  475 
Rubey,  William  W.,  v,  73,  497 
Rucker,  Sir  Arthur,  82 
Rudin,  Hermann,  103,  291,  292 

studies,  242-243 
Rudnicki,  Konrad,  19,  32,  46 

publication,  48,  49 
Ruger,  Henry  A.,  109 
Ruggles,  Clyde  Orval,  95 
Ruhimas,  Jorma  J.,  103 
Rule,  Bruce,  46 
Runnstrom,  J.,  430 
Ruppert,  Karl,  91 
Russell,  George  Oscar,  109 
Russell,  Helen  E.,  292 
Russell,  Henry  N.,  57,  59,  79 
Ruth,  Royal  F.,  87 

studies,  394-396 
Ryerson,  Martin  A.,  vii 


Saa,  German,  S.J.,  231,  292 

Saari,  John  M.,  5 

Sabels,  Bruno  E.,  104,  208 

Sabin,  Florence,  88 

Sager,  Ruth,  102,  285,  292,  353 

Saha,  M.  N.,  82 

Sahade,  Jorge,  43 

Sahama,  Th.  G.,  104 

St.  John,  Charles  E.,  78 

Sakolski,  Aaron  M.,  95 

Salas  del  Carpio,  P.,  292 

Salgueiro  P.,  D.  R.,  231,  291 

publication,  290 
Salisbury,  W.  W.,  292 
Salomonsson,  Gosta,  181 
Salpeter,  Edwin  E.,  79 


Sandage,  Allan  R.,  viii,  58,  74,  78,  3,  17,  30,  46 

publications,  47,  49,  50 

studies,  60,  11-12,   12-13,  14-15,   17,  25-26, 
28,  34-35,  38 
Sanders,  William  T.,  105 
Sandstrom,  J.  W.,  108 
Sanford,  Roscoe  F.,  78 
Saposs,  David  J.,  95 
Sargent,  Porter  E.,  110 
Sargent,  Wallace  L.  W.,  46 

publications,  49,  50 

studies,  60,  8-10,  20-21 
Sarton,  George  A.  L.,  91 
Satina,  Sophie,  83 
Saunders,  George  M.,  92 

Saunders,  John  B.  de  C.  M.,  publication,  431,  477 
Saxen,  L.,  430 
Schabtach,  Gretchen,  432 
Schaeffer,  Asa  A.,  98 
Schafer,  Joseph,  90 
Schairer,  J.  Frank,  viii,  85,  53,  88,  98,  193,  208 

publications,  195,  198,  200,  202 

studies,  52-53,  53-54,  56-82,  91-98 
Schatzman,  Evry,  46 
Schein,  Marcel,  82 
Schiff,  Jerome  A.,  102 
Schilt,  Jan,  79 
Schlegel,  Jorgen  U.,  88 
Schmidt,  A.,  82 
Schmidt,  Hubert,  111 
Schmidt,  Maarten,  viii,  78,  46 

publications,  50 

studies,  25,  27,  34,  35,  37 
Schmitt,  Waldo  L.,  98 
Schmucker,  U.,  292 
Schofield,  Brenda,  105 
Scholes,  France  V.,  91 
Schonland,  B.  F.  J.,  82 
Schrader,  Franz,  84 
Schreyer,  Werner  F.,  104,  194,  208 

publications,  195,  196,  200,  202 
Schultz,  Adolph  H.,  87,  92 
Schultz,  Jack,  85 
Schwachheim,  G.,  219,  292 
Schwarzschild,  Martin,  79 
Scott,  George  W.,  90,  111 
Scott,  James  Brown,  112 
Scott,  W.  E.,  292 
Scripture,  E.  W.,  112 
Scroggs,  William  O.,  95 
Sear,  R.  Owen,  432 
Seares,  Frederick  H.,  78 
Searle,  Leonard  T.,  11,  46 

publications,  50 

studies,  60,  8-10,  35-36 
Sears,  Paul  B.,  110 
Sears,  Robert  E.,  47 
Sechaud,  Janine,  103 
Seidel,  F.,  430 
Sengel,  P.,  430 
Sengiin,  Atif,  103 

publication,  475 


INDEX    OF    NAMES 


523 


Senior,  Harold  D.,  88 
Sensenig,  E.  Carl,  105,  432 
Sersic,  J.  L.,  28 
Setchell,  William  A.,  98 
Sej'mour,  Thomas  D.,  Ill 
Shane,  C.  D.,  79 
Shapley,  Harlow,  78 
Shappell,  Maple  D.,  108 
Sharpless,  Stewart  L.,  102 
Shattuck,  George  C,  92,  101 
Shaw,  Eugene  W.,  98 
Shaw,  James  B.,  107 
Sheeley,  Joyce  E.,  46 
Sheldon,  A.  E.,  95 
Shenstone,  Allen  G.,  106 
Shepard,  Anna  O.,  x,  91 
Shepherd,  Earnest  S.,  6,  85 
Shepherd,  William  R.,  90 
Shepley,  Henry  R.,  v,  481,  497 
Sherman,  Henry  C.,  101 
Shibata,  Kazuo,  102 
Shoemaker,  Clarence  R.,  98 
Shook,  Edwin  M.,  91 
Shorthill,  Richard  W.,  5 
Shreve,  Forrest,  76 
Shull,  George  H.,  6,  31,  72,  88,  437 
Silver,  Peter  H.  S.,  105,  371,  432 

publication,  431 

studies,  407-408 
Silverman,  Alexander,  108 
Silvester,  Charles  F.,  98 
Simon,  E.  H.,  publication,  475 
Simon,  George  W.,  41 
Singer,  A.  D.,  192,  208 
Singer,  Ronald,  88 
Sinnott,  Edmund  W.,84 
Sinton,  William  M.,  5 
Sioussat,  St.  George  L.,  95 
Sirlin,  J.  L.,  439 

Skinner,  B.  J.,  publication,  196-197,  202 
Slade,  William  A.,  90 
Slichter,  C.  S.,  86 
Slizynski,  B.  M.,  84 
Slocum,  Frederick,  79,  82 
Smith,  A.  Ledyard,  91 

publication,  477 
Smith,  Edgar  Fahs,  106 
Smith,  Evelyn  E.  B.,  84 
Smith,  F.  Graham,  82 
Smith,  Gilbert  M.,  77 
Smith,  Guinevere  C,  441,  476 
Smith,  H.  G.,  98 
Smith,  H.  Monmouth,  100 
Smith,  J.  Russell,  95 
Smith,  J.  V.,  74,  194 

publication,  194,  201,  202 
Smith,  James  H.  C,  76,  307,  310,  352,  366 

publications,  366 
Smith,  James  R.,  104 
Smith,  Joseph  L.,  92 
Smith,  Joseph  Victor,  104 


Smith,  Lewis  L.,  46 

publication,  50 

studies,  11-12,  14 
Smith,  Philip  E.,  92 
Smith,  Robert  E.,  91 
Smith,  Sinclair,  78 
Smith,  T.  Jefferson,  viii,  80,  291 

publication,  290,  477 
Smith,  Theobald,  vii 
Snowden,  Yates,  95 
Snyder,  Laurence  H.,  84 
Sokoloff,  Alexander,  313 
Solberg,  H.,  108 
Sommer,  H.  Oskar,  112 
Sosman,  Robert  B.,  85 
Sowers,  Don  C,  95 
Sparrow,  Arnold  H.,  84 
Spencer,  Warren  P.,  84 
Spinrad,  Hyron,  6,  8,  31,  43 

studies,  56-57 
Spitzer,  Lyman,  Jr.,  79 
Spoehr,  Herman  A.,  36,  76 
Sponsler,  George  C,  103 
Spooner,  John  C,  vii,  499 
Sprague,  Robert  James,  95 
Stadelman,  R.,  92 
Stager,  Henry  W.,  107 
Stanier,  Roger  Y.,  77 
Stebbins,  G.  Ledyard,  Jr.,  77 
Stebbins,  Joel,  79 

Stedman-Parmenter,  Hazeltine  L.,  101 
Steemann-Nielsen,  E.,  publication,  366 
Steggerda,  Morris,  83 
Steinberg,  Malcolm  S.,  105 
Steiner,  W.  F.,  293 
Steinhart,  John  S.,  viii,  80,  291 

publications,  289,  290,  477 

studies,  45-46,  221-234 
Steinlin,  Uli,  43 
Steinmetz,  Charles  P.,  107 
Stelter,  Benjamin  F.,  112 
Stenstrom,  Nils,  101 
Stephens,  S.  G.,  84 
Stern,  Curt,  72,  84 
Stevens,  Nettie  M.,  110 
Steward,  Frederick  C,  98 
Stewart,  David  B.,  86 
Stieglitz,  Julius,  106 
Stock,  Chester,  110 
Stock,  Leo  F.,  89 
Stockard,  Charles  R.,  98 
Stoek,  Harry  Harkness,  95 
Stone,  Alfred  Holt,  93 
Stone,  Ormond,  107 
Stone,  Raymond  G.,  98 
Storey,  William  Benson,  vii 
Stormcr,  Carl,  80 
Stojranow,  Alexander  A.,  110 
Strain,  Harold  H.,  76 
Stran,  Herbert  M.,  423 
Straus,  William  L.,  Jr.,  88 
Strauss,  Fritz,  88 


524 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Street,  J.  C,  82 
Streeter,  George  L.,  86 
Strehler,  Bernard,  77 
Streisinger,  George,  82 
Striebel,  H.  R.,  242 
Strieck,  F.,  101 
Stromberg,  Gustaf,  78 
Stromgren,  Bengt,  80 
Stromsten,  Frank  A.,  98 
Stromsvik,  Gustav,  91 
Strong,  Richard  P.,  vii 
Strong,  William  W.,  106 
Struve,  Otto,  viii,  78,  45,  46 

studies,  24 
Stryker,  Lucile  B.,  x,  192 
Stuart,  Graham  H.,  112 
Stuart,  L.  C.,  92 
Stull,  Wilfred  N.,  106 
Sturgis,  Somers  H.,  88 
Sturtevant,  A.  EL,  85 
Sugiura,  M.,  103 
Sullivan,  James  F.,  x 
Sumner,  James  B.,  106 
Suzuki,  Yoshio,  104,  208 

studies,  54,  112,  123-130 
Sverdrup,  H.  U.,  82 
Swann,  W.  F.  G.,  82 
Swett,  Francis  H.,  88 
Swingle,  Walter  T.,  35,  77 
Swings,  Pol,  80 
Swope,  Henrietta  EL,  78,  29,  46 

publication,  47,  50 
Sydenstricker,  Edgar  M.,  95 
Sykes,  Godfrey  G.,  76 
Sype,  Nancy,  405,  433 

Taft,  Charles  P.,  v,  vi,  497 

Taft,  William  H.,  vii 

Takata,  Chinami,  371,  405,  432 

Takeuchi,  Ikuo,  105 

Talbot,  Fritz  B.,  101 

Tamiya,  Hiroshi,  77 

Tandy,  Geoffrey,  98 

Tardent,  Pierre,  88,  405 

Tarkowski,  A.  K.,  430 

Tartar,  Vance,  98 

Tashiro,  Shiro,  99 

Tate,  John  T.,  82 

Tatel,  Howard  E.,  80,  222,  225,  226 

Tatlock,  John  S.  P.,  112 

Taub,  Stephen,  439 

Tavcar,  A.,  84 

Tax,  Sol,  91 

Taylor,  Charles  V.,  99 

Taylor,  Henry  C,  95 

Taylor,  William  R.,  99 

Teas,  Howard  J.,  84 

Teeple,  John,  92 

Tejeda  F.,  Antonio,  92 

Teller,  Edward,  82 

Temmer,  Georges  M.,  viii,  80,  291 

publications,  288,  290 

studies,  242-244 


Tennent,  David  Hilt,  96 

Tennent,  David  M.,  99 

Thackeray,  A.  D.,  80 

Thayer,  William  S.,  vii 

Thomas,  C.  A.,  Jr.,  publication,  475 

Thomas,  D.  M.,  208 

Thomas,  D.  Y.,  95 

Thomas,  John,  313 

Thomas,  Myrna  C,  439,  466,  467,  476 

studies,  471-472 
Thomas,  Rene,  84 
Thome,  H.  J.,  340 
Thompson,  Donald  E.,  92 
Thompson,  Henry  J.,  310 
Thompson,  J.  David,  111 
Thompson,  J.  Eric  S.,  74,  91 
Thompson,  Raymond  H.,  105 
Thurston,  Robert  H.,  107 
Tifft,  William  G.,  publication,  50 
Tigerstedt,  Carl,  101 

Tilley,  C.  E.,  xi,  86,  59,  82,  88,  95,  98,  121,  194, 
208 

studies,  52,  53 
Tilton,  George  R.,  viii,  46,  80,  85,  190,  194,  208 

publications,  197-198,  200,  202,  289,  290 

studies,  56,  173-179,  234-239 
Timofeeff-Ressovsky,  N.,  84 
Tingle,  John  Bishop,  106 
Tisdale,  Edwin  W.,  78 
Tolman,  William  H.,  95 
Tomizawa,  Jun-ichi,  84 
Toosy,  M.  H.,  88 
Torreson,  Oscar  W.,  80 
Torrey,  Harry  Beal,  99 
Torrey,  Theodore  W.,  88 
Toulmin,  Pete,  193 
Tower,  Walter  Sheldon,  95 
Tower,  William  L.,  84 
Traving,  Gerhard,  44 

publication,  50 
Treadwell,  Aaron  L.,  99 
Trelease,  Sam  F.,  78 
Trik,  Aubrey  S.,  92 
Trimble,  Harry  C,  101 
Trippe,  Juan  T.,  v,  vi,  497 
Troften,  Per-Fredrick,  104 
Tumin,  Melvin,  92 
Tunell,  George,  85 
Turner,  Abby  H.,  101 
Turner,  Frederick  J.,  90 
Turnock,  Allan  C,  104,  88,  194,  208 

studies,  54,  82 
Tuttle,  O.  Frank,  85,  96,  117,  126,  127 
Tuve,  Merle  A.,  viii,  6,  44,  67-68,  80,  291,  295 

publications,  288,  290 

report  of  Director  of  Department  of  Terres- 
trial Magnetism,  209-293 

studies,  45-46,  214-234,  295-301 
Tyler,  David  B.,  87 

Udden,  Johan  August,  86 
Uhler,  Horace  S.,  106 
Ulehla,  Vladimir,  78 


INDEX    OF    NAMES 


525 


Umbarger,  H.  E.,  439 
Unsold,  Albrecht,  80,  25,  44 

publication,  49,  50 
Uotila,  U.  U.,  88 
Urbani,  E.,  419 
Urey,  Harold  C,  106 
Urry,  William  D.,  85 
Ursprung,  H.,  439 
Usher,  Robert  James,  95 
Utter,  Merwyn  G.,  6,  46 

Vaillant,  George  C.,  92 

Vainio,  T.,  430 

Valentine,  Joseph  M.,  99 

Vallarta,  Manuel  S.,  82 

Vallentyne,  J.  R.,  104 

van  Agt,  S.  L.  Th.  J.,  29 

Van  Allen,  James  A.,  80,  239 

Van  de  Hulst,  H.  C.,  80 

Van  Deman,  Esther  Boise,  111 

van  Doorenmaalen,  W.  J.  van,  88 

Van  Hise,  C.  R.,  86 

Van  Holde,  K.  E.,  285,  292 

Van  Ingen,  Gilbert,  99 

van  Lear,  Arnold  J.  F.,  90 

van  Leersum,  E.  C.,  101 

van  Maanan,  Adrian,  78 

Van  Niel,  Cornelius  B.,  78 

Van  Orstrand,  C.  E.,  86 

van  Overbeek,  J.,  84 

Van  Royen,  William,  108 

Van  Tyne,  Claude  H.,  90 

Vardbasso,  S.,  112 

Varnum,  William  B.,  99 

Varsavsky,  Carlos  M.,  102 

Vaughan,  T.  Wayland,  99 

Velde,  Bruce,  10 4 

Venkatesan,  D.,  239 

publications,  289,  290 
Venkatesh,  Chakrauarti  S.,  78 
Vestine,  Ernest  H.,  80 
Vickery,  Frederick  P.,  108 
Vickery,  Hubert  B.,  108 
Vickery,  Robert  K.,  312 
Vieira,  A.  H.  G.,  219,  292 
Villa  Rojas,  Alfonso,  91 
Virgin,  Hemming  I.,  102 
Vishniac,  Wolf,  78 
Visscher,  J.  Paul,  99 
von  Bar,  Ludwig,  112 
von  Borstel,  Robert  C,  103 
von  Herzen,  R.  P.,  168 
von  Koenigswald,  G.  H.  R.,  110 
von  Neumann,  John,  82 
von  Wettstein,  Diter,  78 

Waddington,  C.  H.,  84 
Wadsworth,  James  W.,  vii 
Wait,  George  R.,  80 
Walburn,  Marjorie  H.,  x 
Walcott,  Charles  D.,  vii,  499,  500 
Walcott,  Frederic  C,  vii 
Walcott,  Henry  P.,  vii 


Waldo,  Leonard,  107 

Walker,  Francis,  95 

Walker,  Merle  F.,  102 

Wallace,  Bruce,  82 

Wallace,  W.  Seward,  99 

Waller,  John  C,  99 

Wallerstein,  George,  10,  11,  12,  36,  43,  45 

studies,  60 
Walmsley,  R.,  88 
Walter,  Heinrich,  78 
Ward,  William  Hayes,  111 
Wardlaw,  H.  S.  Halcro,  101 
Warmke,  H.  E.,  83 
Warsh,  K.  L.,  243 

publication,  290 
Wartman,  William  B.,  99 

Washington,  Henry  S.,  85,  112,  113,  114,  115,  122 
Watson,  James  D.,  20-21,  443 
Watson,  John  B.,  99 
Watson,  Patricia  J.,  publication,  366 
Wauchope,  Robert,  92 
Way,  Royal  Brunson,  95 
Weaver,  Ellen  C,  76,  310,  366 

studies,  39-40,  307,  353-365 
Weaver,  Harold,  103 
Weaver,  Harry  E.,  366 

studies,  307 
Weaver,  John  E.,  78 
Weed,  Lewis  H.,  vii,  88 
Weidemann,  Volker,  46 

studies,  36 
Weingart,  Eleanor  Ann,  461,  476 

studies,  466-474 
Weinman,  James  A.,  103 
Welch,  William  H.,  vii 
Wells,  Benjamin  B.,  83 
Wells,  Harry  W.,  viii,  80,  291 
Wells,  John  W.,  99 
Wells,  L.  J.,  105 
Wells,  Roger  C,  99 
Wenner,  Frank,  108 
Werber,  E.  I.,  99 
Wesson,  Laurence  G.,  101 
West,  Andrew  F.,  Ill 
Westergaard,  Mogens,  72,  84 
Westlake,  John,  112 
Weston,  Nathan  Austin,  95 
Wetherill,  George  W.,  80,  173,  178,  179,  194,  237 

publications,  197-198,  200,  202,  289,  290 
Wettstein,  Fritz  von,  84 
Wheeler,  James  R.,  Ill 
Whitaker,  Douglas  M.,  99 
Whitbeck,  Ray  H.,  90 
White,  Andrew  D.,  vii,  499,  500 
White,  David,  110 
White,  Edward  D.,  vii 
White,  Henry,  vii 
White,  James  N.,  v,  vi,  481,  497 
White,  M.  J.  D.,  84 
White,  Paul  Dudley,  101 
White,  Priscilla,  101 
White,  Walter  P.,  85 
Whitehead,  Dexter,  103 


526 


CARNEGIE    INSTITUTION     OF     WASHINGTON 


Whitehead,  John  B.,  106 
Whitehorn,  John  C,  101 
Whitford,  Albert  E.,  80 
Whiting,  P.  W.,  84 
Whitney,  Walter  T.,  108 
Whittinghill,  Maurice,  84 
Wickersham,  George  W.,  vii 
Wieland,  George  R.,  78 
Wiggins,  Ira  L.,  78 
Wiik,  H.  B.,  107 
Wilbur,  Paul  C,  78 
Wilczynski,  Ernest  J.,  107 
Wild,  P.,  32-33 
Wilde,  Walter  S.,  87 
Wildey,  Robert  L.,  46 

publication,  50 

studies,  15,  42 
Wildt,  Rupert,  80 
Wilgus,  Horace  L.,  95 
Wilkie,  Douglas  R.,  105 
Willard,  Hobart  H.,  106 
Willcox,  Walter  F.,  93 
Williams,  Francis  H.,  101 
Williams,  G.  D.,  101 
Williams,  Henry  S.,  110 
Williams,  Howell,  92,  110 
Williams,  J.  L.,  99 
Williams,  Robert  R.,  108 
Williamson,  C.  C.,  95 
Williamson,  Erskine  D.,  85 
Willier,  Benjamin  H.,  70,  99,  370 
Willis,  Bailey,  108 
Williston,  Samuel  W.,  78,  110 
Wilson,  Albert  G.,  78 
Wilson,  Calvin  Dill,  95 
Wilson,  Carole  E.,  476 
Wilson,  Edgar  B.,  106 
Wilson,  Edmund  B.,  72,  84,  110 
Wilson,  Henry  V.,  99 
Wilson,  Ian  B.,  105,  371,  432 

studies,  421-423 
Wilson,  J.  M.,  99 
Wilson,  Karl  M.,  89 
Wilson,  Olin  C,  viii,  78,  44,  46 

studies,  13-14,  25 
Wilson,  Ralph  E.,  78 
Wilson,  Robert  E.,  v,  vi,  497 
Wilson,  Robert  W.,  110 
Wilson,  Stanley  D.,  101 
Wilt,  Fred  H.,  105 
Winternitz,  Milton  C,  89 
Wislocki,  George,  89 
Wissler,  Clark,  92 
Witkin,  Evelyn  M.,  xi,  82 


Witschi,  Emil,  89 

Witt,  H.  T.,  310,  338 

Witte,  Edwin  E.,  95 

Wolf,  Frederick  T.,  78 

Wones,  David  R.,  104,  102,  105,  194 

publications,  198-199,  202 
Wood,  Francis  Waverly,  103 
Wood,  Harry  O.,  108  " 
Wood,  Robert  W.,  106 
Woodring,  Wendell  P.,  110 
Woods,  P.  S.,  publication,  475 
Woodward,  Robert  Simpson,  vii,  48 
Woollard,  George  P.,  publication,  290,  477 
Woolley,  R.  v.  d.  R.,  80,  45 
Wright,  Carroll  D.,  vii,  92,  499,  500 
Wright,  Fred.  E.,  7,  85 
Wright,  Herbert  Francis,  112 
Wright,  Irene  A.,  90 
Wright,  Kenneth  O.,  36,  45,  46 
Wright,  R.  R.,  Jr.,  95 
Wulf,  Oliver  R.,  publication,  49,  50 
Wurm,  K.,  25 
Wyckoff,  Ralph  W.  G.,  85 

Yagi,  Kenzo,  104,  53 

studies,  96-99 
Yamada,  T.,  371 
Yanofsky,  Charles,  84 
Yatsu,  Naohide,  110 
Yerkes,  Robert  M.,  101 

Yoder,  Hatten  S.,  Jr.,  viii,  85,  69,  151,  193,  194, 
208 

studies,    52,    53,    54,    75-82,    88-91,    96-98, 
106-107 
Yonge,  C.  M.,  99 
Young,  Allyn  A.,  95 
Young,  Frederic  G.,  95 
Young,  Violet  (Koski),  76 
Younkin,  Robert,  6,  8 

studies,  56-57 
Yund,  Richard  A.,  104,  145,  148,  151,  155,  157, 
158,  160,  194,  208 

publications,  196,  199,  202 

Zahm,  Albert  F.,  106 

Zeleny,  Charles,  99 

Zies,  Emanuel  G.,  85,  208 

publications,  200,  202 

studies,  54,  112-118 
Zirin,  H.,  publication,  289,  290 
Zubay,  Geoffrey,  439 
Zwicky,  Fritz,  viii,  78,  46 

publications,  48,  50 

studies,  74,  18,  19,  20,  32-33