-/? •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