Report - Carnegie Year Book 37

Carnegie Year Book 37

http://www.archive.org/details/yearbookcarne37193738carn

CARNEGIE INSTITUTION OF WASHINGTON

YEAR BOOK No. 37

JULY 1, 1937 - JUNE 30, 1938

WITH ADMINISTRATIVE REPORTS THROUGH DECEMBER 9, 1938

Published by Carnegie Institution of Washington

Washington, D. C.

1938

JUDD & DETWEILER, INC. STANDARD ENGRAVING CO.

WASHINGTON, D. C.

Online Editor - Excerpt of only Biology section.

CONTENTS

PAGES

Officers and Staff v-x

Organization, Plan, and Scope xi

Articles of Incorporation xii-xiv

By-Laws of the Institution xv-xviii

Abstract of Minutes of the Thirty-ninth Meeting of the

Board of Trustees xix

Abstract of Minutes of the Fortieth Meeting of the Board of

Trustees xx-xxiii

Report of the Executive Committee xxiv-xxv

Aggregate Receipts and Disbursements xxvi

Report of Auditors and Financial Statement xxvii-xxxiv

Report of the President of the Institution 1-65

Reports on Investigations:

Division of Animal Biology 3-103

Department of Embryology 4-34

Department of Genetics 35-72

Nutrition Laboratory 73-83

Tortugas Laboratory 84-103

Geophysical Laboratory 105-136

Division of Historical Research 137-172

Mount Wilson Observatory 173-208

Division of Plant Biology 209-238

Department of Terrestrial Magnetism 239-293

The Atomic-Physics Observatory, Department of Terrestrial

Magnetism, Washington, D. C facing 239

PRESIDENT AND TRUSTEES

PRESIDENT

John C. Merriam*

BOARD OF TRUSTEES

W. Cameron Forbes, Chairman Walter S. Gifford, Vice-Chairman

Frederic A. Delano, Secretary Thomas Barbour Walter A. Jessup Elihu Root, Jr.

James F. Bell Frank B. Jewett Henry R. Shepley

Robert Woods Bliss Charles A, Lindbergh William Benson Storey Frederic A. Delano Alfred L. Loomis Richard P. Strong

Homer L. Ferguson Roswell Miller Charles P. Taft

W. Cameron Forbes Henry S. Morgan James W. Wadsworth

Walter S. Gifford Stewart Paton Frederic C. Walcott

Herbert Hoover John J. Pershing Lewis H. Weed

Executive Committee

W. Cameron Forbes, Chairman

Robert Woods Bliss Walter S. Gifford Frederic C. Walcott

Frederic A. Delano Walter A. Jessup Lewis H. Weed

Jofin C. Merriam*

Finance Committee

Frederic C. Walcott, Chairman

Walter S. Gifford Henry S. Morgan

Alfred L. Loomis Elihu Root, Jr.

Auditing Committee

Frederic A. Delano, Chairman Homer L. Ferguson William Benson Storey

STANDING COMMITTEES FOR THE YEAR 1939 Committee on Astronomy

Herbert Hoover, Chairman

Walter S. Gifford Elihu Root, Jr.

Roswell Miller William Benson Storey

Committee on Terrestrial Sciences

Frank B. Jewett, Chairman

Frederic A. Delano Henry S. Morgan

Homer L. Ferguson James W. Wadsworth

Committee on Biological Sciences

Lewis H. Weed, Chairman

Thomas Barbour Alfred L. Loomis

James F. Bell Stewart Paton

Committee on Historical Research

Henry R. Shepley, Chairman

Robert Woods Bliss Richard P. Strong

Charles A. Lindbergh Charles P. Taft

• On January 1, 1939, Dr. Merriam becomes President Emeritus, and Dr. Vannevar

Bush succeeds him as President of the Institution.

STAFF OF INVESTIGATORS FOR THE YEAR 1938

DIVISION OF ANIMAL BIOLOGY: George L. Streeter, Chairman

An administrative grouping made effective in 1935, including activities of the following Departments:

Department of Embryology

Organized in 1914; Franklin P. Mall, Director 1914-1917.

George L. Streeter, Director Margaret R. Lewis

Carl G. Hartman Warren H. Lewis

Chester H. Hetjser Charles W. Metz

Department of Genetics

Station for Experimental Evolution, opened in 1904, was combined with Eugenics Record Office in 1921 to form Department of Genetics. Charles B. Davenport, Director 1904-1934.

A. F. Blakeslee, Director E. C. MacDowell

M. Demerec, Assistant Director James S. Potter

H. H. Laughlin, Assistant Director Oscar Riddle

A. G. Avery Sophia Sattna

R. W. Bates J. P. Schooley

A. Dorothy Bergner Morris Steggerda

B. P. Kaufmann H. E. Warmke

Nutrition Laboratory

Organized in 1907, opened in 1908; F. G. Benedict, Director 1907-1937.

T. M. Carpenter, Acting Director Robert C. Lee

REPORTS ON INVESTIGATIONS

DIVISION OF ANIMAL BIOLOGY

George L. Streeter, Chairman

The activities of the component groups under the Division of Animal Biology are reported in the following pages and it will be seen that in general they include related phases of physiology, anatomy, embryology, and heredity of animals. The question repeatedly arises as to whether it would be desirable to bring these researches all under one roof, so to speak, or at least in one locality. Much might be gained by a closer assembly, but it would also involve some sacrifices. Even were such a concentration feasible, careful consideration should be given to all the factors concerned. The present arrangement has been one of natural evolution. The various projects have sprung up where it was thought they could be best conducted; some in large scientific centers, others in isolated locations where environments of a special character were demanded by the nature of the project.

As far as possible, the disadvantage of the dispersion of the research personnel has been counteracted by frequent conferences and exchange of facilities between the individuals of the different groups. As a consequence, it is more and more being found that a community of interest and cooperative endeavor are of mutual advantage and an added source of strength. Inasmuch as the selection and grouping of the researches have been of an evolutionary or empirical nature it is to be expected that their goals and planning will be subject to change and adjustment from time to time. Projects are dropped as they reach a reasonable completion and attention is turned in other directions. Trial and error are dominant features in the organization of the groups and subgroups just as they are in their researches and as they are in nature. The investigator is on the fringe of the unknown where in his uncertain advances he must be alert to advantageous alterations in course. He must have a plan but his plan must not commit him to a program that cannot be terminated when it becomes barren.

One of the projects that were brought to a conclusion during the past year is the survey of the human resources of Connecticut. This was undertaken as a cooperative study by a commission appointed by the governor of the state of Connecticut and the staff of our Eugenics Record Office. The survey was of importance to the state because of the alarming and increasing expenditures necessary for the care and treatment of its socially inadequate classes. It was of interest to the staff of the Eugenics Record Office because such a survey fell directly in the field of eugenic and population studies with which they were especially equipped to deal. Now after two years of concentrated effort the factual data, covering the biological and eugenical aspects of the problem, have been collected, classified, and analyzed and placed at the disposal of the Commission. It is realized that the biological point of view, though a very important one, is but one aspect of this population problem and it happens to be the one to which our facilities are limited.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

DEPARTMENT OF EMBRYOLOGY 1

George L. Streeter, Director

EARLY STAGES OF THE PRIMATE EMBRYO

The Eight-Day Primate Egg

In Year Book No. 33 a report was given of the obtaining of a 10-day macaque embryo and it was pointed out that our vision of the mechanism of development was thereby extended into the 24-hour period preceding any hitherto known primate ovum. The year following we were able to report two 9-day specimens, one just before implantation and one just in the process of attaching. With these two eggs the known territory was extended another 24 hours earlier. During the past year we have obtained an 8-day ovum and again a still earlier 24 hours has been mastered. This carries us back to where, in its histological structure, the primate ovum is close kin to other mammalian forms, and to where they have in common the developmental pattern of a blastocyst. It is of the greatest importance that we have acquaintance with this expression of their common unity in functional requirements and the common way they have of meeting them. In going back to origins, once this unity is arrived at, any of the earlier developmental phenomena are in large part common to all mammals and we can in those early periods study them in any order, genus, or species that suits our convenience.

The 8-day ovum of the macaque consists of a blastocyst having a diameter of 0.175 mm., still enclosed by a disintegrating zona pellucida. This unique specimen was reported upon by Dr. C. H. Heuser before the American Association of Anatomists. An outstanding feature of it is the fact that at the embryonic pole there are still a few cells that are approximately double the size of the others. It is clear to see that they have not divided so many times. These relatively inactive cells appear to be the ones that are destined to form the embryo proper. They are large, few in number, not oriented in position, and show no tendency to unite into a common structure, all of which are characteristic of primitive blastomeres. In contrast to these primordial cells, the other cells are numerous, small, and are rapidly differentiating into special structures which will serve to attach the ovum to the uterus and eventually provide the contained embryo with its nourishment and other physiological requirements. In the point of sequence the stage is set before the embryo makes its appearance. In this blastocyst Dr. Heuser finds that he can see clearly that the materials of the ovum have, already on the eighth day, been segregated into the embryonic and extra-embryonic, or auxiliary, parts of the ovum. Thus the eighth day may be said to mark the completion of the first and hence the most fundamental chapter in the development of the ovum.

1 Address: Wolfe and Madison Streets, Baltimore, Maryland.

Implantation

The second chapter in the history of the ovum is its attachment and implantation in the endometrium of the uterus. But this is a long and complicated process overlapping the subsequent stages and so is not exactly a chapter. Implantation must begin, however, and establish its initial principles before growth and development of the ovum can ensue. Some phases of this process have been described in previous reports. During the past year a study of placentation in the macaque has been brought to a conclusion by the author in cooperation with Dr. G. B. Wislocki and published in its completed form. The study is based on a series of stages representing an almost day by day record from the ninth day when the egg first fastens, through the early and formerly least-known phases of placental development, up to the 35th day, by which time the mature features of the placenta are attained. In fact this constitutes the only relatively complete record in existence of the development of this fetal structure in any primate. The earlier stages are quite unknown in the gibbon and the anthropoid apes and in man our knowledge is fragmentary concerning all events preceding the 14th day.

From the biological standpoint the phenomenon of implantation is of peculiar interest. Here we have the ovum as a minute living organism attaching itself to the surface epithelium of the uterus and, after inducing both stimulative and degenerative changes in it, we see the ovum ingest this altered epithelium, with a corresponding increase in its own mass. The whole picture of this act of parasitism can be followed in its finer cytological details and one can determine the microscopic characteristics of the surrender of one living tissue to another. During the first two weeks this ingestion of maternal cells and intercellular plasma provides the sole source of growth material for the embryo and for a long time it overlaps the materno-fetal vascular exchange which, to a large extent, gradually replaces it.

The placental development is found to pass through three general periods or stages. The first of these is the prelacunar stage, in which trophoblast cells of the ovum erode and ingest the maternal epithelium at the implantation site and thus come in direct contact with the uterine stroma. While this is happening the maternal epithelium of the surrounding area proliferates, thereby building more pabulum for the trophoblast cells to fatten on. Within 24 hours a thick trophoblastic plate is formed at the embryonic pole of the ovum, sealing in the gap created by the disappearance of the maternal epithelium.

A second stage follows during which the trophoblastic plate, in continuing to thicken, develops lacunae into which the maternal capillaries empty and promptly fill with plasma and red cells. These spaces greatly increase the absorptive area and there is a corresponding increase in the amount of trophoblast. The very rapid growth of trophoblast that takes place in three days is shown in the figure on page 6, where B to D represent the prelacunar stage and E to H represent the second or lacunar stage. The third or villous stage follows directly after the above stages. In the available material one can plainly follow the formation of the chorionic villi and the differentiation of cytotrophoblastic columns, the centers of which become transformed into reticular connective tissue and capillary-forming cells and thus compose the cores of the villi. The details of this transformation were previously described by Dr. Hertig as referred to in a previous report (Year Book No. 33). The villous stage begins about the 15th day and progresses to the completion of the placenta, the most important features of which can be seen already on the 35th day.

+++++++++++++++++++++++++++++++++++++++++

Fig. 1. Drawings illustrating the enormous growth that occurs in the trophoblastic wall (solid black) of the ovum in the four days following its parasitic attachment to the uterus. Tissue materials are necessary for such growth and these are supplied by the luxuriant uterine epithelium and the plasma and blood cells of the adjacent tissues. Previous to attachment there is very little actual increase in the mass of the ovum. These eight stages are all from monkey embryos shown at the same enlargement (X 75). Their ages are as follows: A, 9 days; B, 9 days; C, 10 days; D, 10 days; E, 10% days; F, 11 days; G, 12 days; H, 13 days.

Yolk-Sac and Gut Endoderm

Following the attachment of the ovum on the ninth and tenth days we can speak of the third chapter in its development. With the auxiliary parts of the ovum well on their way to differentiation there is an awakening of the cells that are to form the embryo proper. This awakening is expressed by their increase in number and by their orientation and arrangement into an ectodermal disk or embryo disk. In this process the disk becomes set off, both dorsally and ventrally, by fluid-containing spaces from the abutting auxiliary tissues. The exact time relationships vary a little in different specimens and still more so in different mammals, so that in the early stages the appearance at the embryonic pole may be quite different, but apparently the underlying principles remain the same.

During the past year the studies of the writer have been directed more especially to the fluid space which develops ventral to the embryonic disk. Whereas the dorsal space enlarges to become the amnion, the ventral space enlarges to become the combined gut and yolk-sac cavity. The yolk-sac is definitely something more than an embryonic vestige. Furthermore it does not bud off from the inner cell mass in the form of a solid clump of cells, thereafter acquiring a central cavity, as had been supposed. Nor is it at any time an intrinsic part of, or homogeneous with, the gut tract. Instead our specimens show us that the earliest cells of the yolk-sac are differentiated from the primitive endoderm as a thin membrane between which and the gut endoderm there arises the conjoint yolk-sac cavity and gut cavity.

This cavity ventral to the embryo disk is therefore dual in origin. It is bordered on its dorsal part by cells that are to form the gut endoderm, an induced product or migratory element from the disk itself; whereas the space on its ventral part is bordered by the yolk-sac endoderm, which is a derivative of the primitive endoderm and is in fact an auxiliary tissue. The gut and yolk-sac are thus different in origin and are always abruptly demarcated from each other. The one becomes a definite part of the embryo and the other is an auxiliary organ which in primates plays a temporary but apparently very essential role in the metabolism of the embryo up to that time when its functions are taken over by the placenta, a much more elaborate and efficient organ. Thereafter the yolk-sac regresses, although we can usually find its degenerate remnants attached to the fetal membranes at birth.

Young Human Embryos

Among the publications in the last number, volume 27, of the Contributions are three accounts of presomite human embryos. This early period of human development is imperfectly known because of the scarcity of well preserved normal specimens of that age, upon which our understanding must be built. One of these specimens is that of Dr. J. I. Brewer studied in the Anatomical Laboratory of the University of Chicago. This 15-day specimen is certainly normal and is in a good state of preservation and is destined to take its place as a standard 6f orientation in the procession of stages through which the human embryo passes in its development.

A slightly younger specimen has been studied in this laboratory by Dr. E. M. Ramsey. This specimen, known as the Yale Embryo, is estimated to be between 13 and 14 days old. It was obtained at autopsy and a sufficiently large block could be made to show fully the relations of the embryo to the uterine wall. Neither this nor the Brewer specimen has the perfection in histological detail that we are able to secure in our macaque embryos, and neither of them can be relied upon for some of the problems that involve finer cytology and intercellular reactions. The student of human embryology, however, is accustomed to such deficiencies and has been able to piece together his story from specimens, some of which are much less perfect than these. It is to be remembered that at the end of the second week even the gross general anatomy of the human embryo is still more or less obscure.

Among other things Dr. Ramsey analyzes the phenomenon of blood-sinus formation around the newly implanted ovum. The Yale specimen illustrates this particularly well and apparently it always occurs in normal specimens as soon as decidua appears. The sinuses are found to consist of localized dilatations of precapillary venules and their formation is an expression of the marked response that occurs in the endometrial vessels under varying hormonal and environmental conditions. Whether such localized allocations of blood plasma and red cells serve a nutritional function for the ovum or whether the sinuses are a means of lowering blood pressure in the vessels communicating with the lacunae, thereby shielding the embryo from pressure extrusion, remains to be determined. Perhaps the most important part of Dr. Ramsey's study is her analysis of trophoblast development, and her conclusion that there is a normal correlation between it and the decidua and the embryo proper. With the normal variation in these correlations established one is then in a position to recognize the abnormal. It is also noted that it is normal for the trophoblast to develop more luxuriantly on the deeper hemisphere of the ovum where the food is more abundant than on the side toward the uterine cavity. This is responsible for the apparent invasion inward. Besides the correlated growth of the trophoblast relative to other things, there is a regulation of its differentiation into syncytium on the surface and into primitive mesoblast on the inner side, the more primitive cytotrophoblast, later known as Langhans cells, being the germinal bed for both of them. Normally a balance is maintained between these elements. The occurrence of an imbalance between them in a specimen is evidence that it is pathological. Dr. Ramsey illustrates this by a specimen from the Carnegie Collection in which the supply of parent cytotrophoblast is exhausted, being wholly converted into mesoblast on one side and syncytium on the other.

The occurrence of "syncytial wandering cells" in the above specimen has given Dr. Ramsey an opportunity to analyze their origin. There is some evidence that these are not fetal cells but are transformed stroma cells, which are in the process of being converted into pabulum for the advancing trophoblast. On that basis the multinuclear or giant cell character could be regarded as an expression of degeneration. On the other hand, the distribution of the cells and the absence of other signs of degeneration leave the matter in some doubt. Dr. Ramsey's study of this phenomenon serves to call attention to a phase of implantation that has been largely overlooked.

A third human embryo, younger than either of the former, has been studied by Dr. E. Scipiades, Jr., of Budapest, a guest of this laboratory as an exchange student of the Institution of International Education. This specimen carries special interest because of an associated clinical history of hormone and "small dose" X-ray treatment for sterility. The embryo was discovered in curettage material and fortunately the sections pass through the ovum in a direction that discloses the implantation details. The trophoblast is primitive in character and as yet there are no villi. This places the specimen in the group of very early ones.

ORGANOGENESIS

The studies of Dr. E. H. Norris on the parathyroid and lateral thyroid glands will already be familiar to the readers of our embryological reports. There is now to be added to these a reference to his study of the human thymus gland, to which his investigations were extended. Part of his work was done on the embryological collection in Dr. Jackson's laboratory and part on our own collection.

In his study of the morphogenesis of the thymus Dr. Norris comes to conclusions which diverge somewhat from those of Dr. Weller, who also had worked in our laboratory and on the same material. This is wholesome and tends eventually to bring us nearer to the correct solutions of these questions. A point emphasized is the cervical sinus, which Dr. Norris concludes to be the primordium of the primitive thymic cortex and the source of Hassall's corpuscles, two very important assignments. Dr. Weller had concluded that the cervical sinus is a product of the mechanical exigencies of the region and is influenced by, rather than being the cause of, the development of the thymus. It is clear that much study is still needed throughout the region of the branchial pouches before we can unravel its many factors in development. It may be that some help can be obtained through experiment. The branchial clefts have been so heavily loaded with the philosophy of recapitulation that it is difficult to separate out what is the real gill-cleft phenomenon from what is the expression of other developmental factors of the region. It is possible that recapitulation in our ways of thinking is more inexorable than it is in the development of the embryo.

In his conclusions Dr. Norris derives the epithelial elements of the human thymus from two distinct sources both of which are within the third branchial complex. These two sources make it an ectodermal-endodermal structure. The ectodermal source is the cervical sinus, which provides the primitive thymic cortex and the HassaH's corpuscles, as noted above. The endodermal thymus, arising from the third endodermal branchial pouch, gives origin to the syncytial cytoreticulum of the gland. The thymic lymphocytes he finds to be of mesenchymal origin, secondarily invading the gland. Other elements of gland reticulum are derived from connective tissue cells in the adventitia of vessels and from the gland capsule.

Embryologists have utilized the somites very largely as a topographical scale against which the levels of the body are oriented. One difficulty has been the determination of the first or most oral somite, with which they must begin their count. There has been an uncertainty as to whether the first somite of a later stage is the same as the first somite of the earlier stages; that is, are new somites added in front or perhaps does the original first somite disappear? Either of these events would alter the count. An important study covering a survey of these occipital somites in a large number of human embryos has been made by Dr. L. B. Arey of Northwestern University, and the results were reported in the last volume of the Contributions to Embryology. He finds that the first pair of somites usually undergo regression both in size and in structure. By tracing this regression he concludes that dedifferentiation plays a considerable part in the process. The regression may be slow in some cases but usually in embryos of more than 20 somites the first remaining pair are actually second somites. Retarded differentiation and beginning regression can be made out as early as embryos of between 5 and 9 somites. For students of embryonic anatomy Dr. Arey's careful work in this field will be of importance. Also it adds to the general principles of development an instance of over-induction, eventually corrected by the lack of supporting stimuli.

In studying the anatomy of the whale fetus Dr. R. Walmsley has covered a field in which comparatively little systematic work had been done. Although of peculiar interest because of its high degree of specialization, the adult animal from its very size and its inaccessibility has rarely been available for study except in its skeletal form. In the fetus some of the difficulties are obviated and the opportunity of studying four fetal specimens has been well utilized by Dr. Walmsley not only to cover the fetal stages but through them to interpret the conditions and problems of the adult. His oldest specimen was a mid-term fetus, and the other three fell within the first half of pregnancy.

The respiratory mechanisms of the whale in adaptation to its habits of submergence have been of great interest to the anatomist and they have resulted in highly specialized structures that are foreign to other mammals. The most striking peculiarities of the whale, however, are found in its vascular system. Dr. Walmsley's investigations were largely concerned with this system. He made a systematic study of the blood vessels throughout the whole body, both as to their gross anatomy and as to their histology. A striking generalized characteristic of the arteries was the breaking up of what in other mammals would be a single trunk into a series of collateral vessels, a condition adapted to a large volume of total blood together with a low blood pressure. The feature of greatest interest was found to be the retia mirabilia, especially the thoracic, which is associated with decrease in lung volume during submergence, and the cerebrospinal retia, which serves as buffer interposed between the arteries of the main parts of the body and those intrinsic to the central nervous system. Dr. Walmsley concludes that these peculiarities of the whale vascular system are not provisions against a possible shortage of oxygen, for apparently such a shortage does not occur. Instead, this highly specialized system is an adaptation to the differences of surface pressure under which their mode of life requires them to live. It is thus understood why these specializations are more elaborate in the adult than in the fetus and are largest in whales that can remain submerged longest.

PHYSIOLOGY OF THE EMBRYO

Secretion in the Fetal Chorioid Plexus

Continuing his studies on the origin of function of the chorioid plexus in the fetal brain, Dr. L. B. Flexner with the cooperation of Dr. R. D. Stiehler has been able to show that the appearance of secretory activity of this plexus is correlated with the development of a difference of potential between stroma and epithelium and the initiation of an electric current between the two tissues. It had previously been shown that cerebrospinal fluid changes from an ultrafiltrate to a secretion at the end of the first third of pregnancy in the pig. In the new studies attention is turned to the biochemical changes that occur in the plexus at this transition period, when the plexus is changing from a passive state to an active chemical machine.

For determining the potentials of epithelium and stroma use was made of oxidation-reduction indicators and it was found that in the pre-secretory period the epithelium and stroma of the plexus have the same potential. With the onset of secretion, however, the potential of the epithelium rises somewhat and the potential of the stroma falls markedly until there is a potential difference of 0.10 volt. This difference increases until in the last third of the gestation period it amounts to 0.23 volt.

In the secretory plexus, in addition to the difference of potential between epithelium and stroma it was also found that the basement membrane conducts electrons. The electric current so established between epithelium and stroma explains the selective transference of acid and basic dyes across the basement membrane. In the pre-secretory plexus, since there is no difference of potential between epithelium and stroma, there is no electric current and consequently no selective transference of dyes.

In explanation of the changes in potentials of epithelium and stroma, it is pointed out by Dr. Flexner that in the pre-secretory period indophenol oxidase is in low concentration and equally distributed between epithelium and stroma. As secretion begins this oxidase disappears from the stroma and is found in much higher concentration in the epithelium. Indophenol oxidase activates molecular oxygen for biological oxidations and in its presence, other factors being equal, the potential level of a tissue is raised. Thus these changes in potential seem to hinge on the distribution of indophenol oxidase.

In his studies of the biological processes underlying the formation of cerebrospinal fluid by the chorioid plexus, Dr. Flexner has made an analysis of the thermodynamics of ultrafiltration and has verified his theoretical deductions by experiments with sucrose solutions and to some extent with colloidal solutions. In this way he has obtained criteria for distinguishing whether the cerebrospinal fluid is a dialysate in equilibrium with the blood plasma, or an ultrafiltrate of the blood plasma, or a true secretion involving energy expenditure by living cells. In this work Dr. Flexner has found himself in the enviable position of the biologist who on reaching one of his frontiers is able to help himself out by making use of some of the pathways that have been developed in a neighboring discipline.

CHROMOSOME STUDIES

Development of the Salivary Gland Chromosomes

Because of their unique interest an investigation has been made by Dr. J. B. Buck of the embryology of the giant chromosomes which are found in the larval salivary glands of Diptera. Using Sciara larvae he has traced the steps of transformation of these chromosomes from their original state, when in size and appearance they are like ordinary somatic chromosomes, until they attain their enormous size and typical banded appearance. In doing this he made measurements of the chromosomes, nuclei, cells, glands, and larvae in the living state. The finer morphology of the chromosomes was studied on fixed material for each stage of development.

It was found that the salivary gland attains its definitive number of cells soon after it arises in the embryo within the egg. Its subsequent growth is due entirely to increase in cell size and the greater part of this growth occurs in the late larval stages. What is true of the cell is also true of the nucleus and then in turn of the chromosomes, which compose 90 per cent of the nuclear volume. The nuclear volume doubles every 1% days during the period of the 4th to the 17th days, from which time to the beginning of pupation it becomes progressively less.

As to the structure of the chromosomes, it is found that in the early embryonic gland homologous chromosomes are paired, and are merely short slender threads. From the outset, however, each thread exhibits enlargements which correspond to those of its homologue. Each of the threads soon splits into two threads, retaining however some cross-connections. The homologues are thus doubled and shortly before the larva hatches from the egg these doubled homologues begin to twist or coil about each other and fusion between homologous regions begins. There is thus produced a fourstranded helically coiled flattened chromosome as early as 6% days after the egg is laid. Each synapsed pair of doubled homologues soon appears as a slender cylindrical and much elongated strand, showing diffuse crossbands at intervals which foreshadow the "banding" of the fully developed salivary chromosome. During the succeeding larval stages the chromosomes greatly increase in diameter, the banding becomes more pronounced, and the coiling reaches its maximum. Before pupation the coiling relaxes and the chromosomes become straighter, and this is the period of their greatest growth. As pupation sets in, however, these giant chromosomes begin to regress and finally during the pupal stage the larval salivary glands undergo histolytic degeneration.

In reference to the development of the banding, Dr. Buck found that the first ones to appear represent the heaviest bands of the definitive chromosome. They remain relatively unaltered during development, though they tend to darken and some of them separate into doublets. The new, light bands, which become visible as development proceeds, appear in the lengthening spaces between the heavier bands. They apparently do not split off from the latter.

This investigation is being carried further but already it has uncovered some very essential features which will need to be reckoned with by the chromosome cytologists and by the searchers for the gene.

Chromosome Puffing and Chromosome Knots

In his studies on the giant salivary gland chromosomes in Sciara Dr. C. W. Metz has turned his attention to the phenomenon of "puffing" and the structure of such regions. Instead of the characteristic banded structure the chromosome may vary in certain particular regions by becoming greatly expanded or "puffed." In such an area it is uniformly granular, resembling cytoplasm as seen in fixed preparations. It is evident that the segment involved has been increased in volume and that the increase is in the form of small achromatic droplets. Intermediate degrees of "puffing" show bands in various degrees of disintegration. Dr. Metz points out that the material in these "puffed" regions is perhaps comparable to that of the "chromocenter" in Drosophila, and if it is we must conclude that the "chromocenter" is not inert. From the study of living cells in tissue culture we have learned the wide range in form which cells may undergo with different physiological states, and our recent observations on chromosomes and their chromatin content tend to show that here, too, we have structures that are delicately responsive to the state of the nucleus and the circumstances of the environment.

In a former report reference was made to the evidence obtained by Dr. Metz of the presence of an insulating sheath surrounding the individual chromosome. In a review of the occurrence of chromosomal knots he adds further evidence of the existence of such a sheath. Simple knots are occasionally found midway in the giant salivary gland chromosomes of Sciara. It is evident that they arise early while the chromosomes are still small threads, but since the knots always involve both homologues they must have formed after the homologues had fused. The principal growth of the chromosome occurs after the completion of the knot, and this tends to make the knot a snug one. A knot can form only in the presence of free movement of the chromosome. At its initiation there must be a loop and the sliding of one part over another. That fusion does not result during the process is explained by Dr. Metz by hypothesizing an insulating sheath. As the chromosome becomes thicker and the knot becomes tighter it would appear on mechanical grounds that the segments involved would become deleted or, if the tenseness sufficiently overcame the protection of the sheath, fusion would follow and an inversion of segments would occur. Such events would explain some of the well-known genetic experiences.

Multiplication and Reduction of Chromosome Groups

Last year an account was given of the studies of Dr. C. A. Berger on the multiple chromosome complexes found in the larval ileum of the mosquito, where repeated division of the chromosomes occurs without division of the nucleus. That is, during the larval period the epithelial cells of the ileum grow by increase in cell size to three or four times their original volume without an increase in the number of cells. It is not until metamorphosis that division of the cells occurs. As metamorphosis progresses these cells repeatedly divide with corresponding decrease in their size and in the number of chromosomes until we come to the relatively small nuclei of the rebuilt imaginal ileum with its normal diploid number of 6 chromosomes. Here we have compounding of chromosomes which may well be compared to that seen in the giant salivary gland chromosomes. Since last year Dr. Berger's paper has been completed in its final form and has been published in volume 27 of the Contributions to Embryology.

GENETIC STUDIES

Genetic Units

In the last Year Book reference was made to the difficulties met in distinguishing between genie and non-genic material in the chromosome. Dr. C. W. Metz has continued his consideration of this problem and at the end of the additional year he still is unable to place his finger on the gene though in the meantime he has made further additions to our knowledge of the finer anatomy of the chromosome.

The problem has been approached from another aspect, namely, a study has been made of small chromosomal deficiencies, which are either lost or acquired as units and therefore might represent genes. Dr. Metz has thus attempted to identify the smallest possible structure that, as a unit, may be lost or acquired, and at the same time to determine the nature of this structure in terms of its visible chromatic and achromatic chromosomal constituents. For his material he has utilized Sciara ocellaris, which is particularly favorable in that it is characterized by various small chromosomal deficiencies which may be found in ordinary stocks without radiation. These deficiencies appear to be widely distributed in nature.

The general trend of evidence from the salivary gland chromosomes of this form is that any of the transverse disks, including even the thin ones, is divisible into two or more disks sufficiently independent so that one may be lost without the other, or a similar one may be acquired. Dr. Metz has analyzed in detail eight examples of this kind of deficiency and has attempted to determine: (1) what is the smallest unit that can be detected as a deficiency? (2) what is the composition of this unit in terms of chromatic and achromatic materials? and (3) what is thereby revealed as to the relationship of the chromatic disks to the achromatic material? He found that these questions could be answered at least in part. The smallest unit loss or acquisition involves a single chromatic disk and the layer of achromatic material on one side of it. As to the relationship of the chromatic disk to the chromatic material matters remain uncertain. The evidence points toward the chromatic rather than the achromatic materials as the genetically important constituents. Both exhibit considerable variation in appearance, but apparently the variation in the chromatic disks is largely due to variations in the amount and distribution of achromatic materials. Dr. Metz concludes that the heavy walled droplets, thought by some to represent unit genes or chromomeres, are probably not unit structures and do not represent unit loci.

Observations on Sciara Hybrids

For purpose of genetic and cytological studies several wild strains of Sciara have been brought into the laboratory from various localities. Dr. Metz has found among this material a new species of Sciara (S. reynoldsi) which hybridizes with one of the more common laboratory species (S. ocellaris). The two species possess definite morphological distinctions. In general appearance, however, the two species are so nearly identical that their discreteness was not noted until it was found that many of the crossmatings between them failed to give offspring and also that their salivary gland chromosomes differ in pattern. Because of the importance of this new species for genetic studies a taxonomic description of it has been published by Dr. Metz for the information of other workers.

Preliminary hybridization studies of the above two species already have been made by Dr. Metz and Mrs. E. G. Lawrence. Among the results obtained by them are the following: (1) no offspring are secured from matings of S. reynoldsi 9 X S* ocellaris $ ; (2) S. ocellaris 2 X S. reynoldsi S give vigorous, viable offspring in large numbers; (3) apparently all hybrid females are sterile; (4) hybrid males are not sterile; when mated with ocellaris females offspring were obtained in eight out of ten matings; (5) many of the hybrids are gynandromorphs, representing various types of mosaics of male and female parts; (6) abnormal gonads are found very frequently; (7) when yellow (a sex-linked mutant character) ocellaris females were mated to wild-type reynoldsi males, the daughters were found to be "wild-type" and the sons all "yellow"; (8) the males thus transmit only the chromosomes received from their mothers and they thus behave genetically as if they were pure S. ocellaris; (9) the original cross and backcross matings are successful only when ocellaris is used as the female parent; (10) the hybrids tend to be intermediate between the two parents in respect to taxonomic characters; (11) the metaphase chromosome groups of the two species appear to be alike, each consisting of three similar pairs of rod-like chromosomes and one pair V-shaped; (12) in the salivary glands of hybrids the chromosomes are associated in symmetrical pairs, but complete fusion is found only in a few short regions. In their publication the interesting structure and incomplete fusion of the salivary gland chromosomes of these hybrids are fully illustrated.

New Mutants in Sciara

Six new mutant characters in Sciara which have been found in this laboratory among descendants of flies which had been exposed to radium, along with two mutants which appear to be identical or reciprocal to two of them, have been studied by Miss H. V. Crouse and Dr. H. Smith-Stocking. In each instance the origin and description of the character and its genetic behavior have been completely analyzed by them. They have found that the character "stop" is an autosomal dominant in S. coprophila, involving a translocation between chromosomes II and IV. "Yellow" is a sex-linked recessive in S. ocellaris. This character has already been referred to in connection with the hybrids described above. Of the four mutations found in S. reynoldsi, "puff" and "vesiculated" are autosomal dominants that are lethal when homozygous. They may prove to be reciprocal or perhaps identical. "Jagged" is a sex-linked dominant that is lethal when homozygous. "Ruffled" is a sex-linked dominant and closely linked to it is "Yellow," a sex-linked recessive. The "yellow" mutations of S. reynoldsi and S. ocellaris appear to be reciprocal. These investigators conclude that the genetic behavior of these mutant characters is proof of selective segregation in S. ocellaris and S. reynoldsi.

CHEMISTRY OF BODY TISSUES

Improved Methods for Histochemical Analysis of Tissues

Dr. I. Gersh has adapted the Altman, freezing-drying technique for the identification of the presence of chloride, phosphate-carbonate, and potassium in muscle tissue. In this way he avoids diffusion of water-soluble substances which had vitiated the studies of earlier workers. By embedding the frozen-dried material in paraffin and sectioning at known thickness the distribution of the substances can be studied and access of subsequent reagents is facilitated. The paraffin is removed by petroleum ether and this allows silver reagents to penetrate the sections. Silver nitrate solutions were used to precipitate chloride ions, alone or together with phosphate and carbonate ions. The precipitated silver salts were reduced by exposure to an arc light. Potassium was made visible by the use of sodium cobaltinitrite. The details of the procedure as worked out by Dr. Gersh will not be given here. They have, however, been made available to other investigators in the published account of his investigation.

As regards chloride, Dr. Gersh was not able to find any in the muscle cells of the frog sartorius. It was present, however, in the interstitial and collagenous connective tissue. When the muscle chloride concentration was decreased or increased by suitable manipulation there was a corresponding decrease or increase in the amount of chloride visible in the connective tissue spaces. In all cases there was more visible chloride present in collagenous connective tissue than in reticular connective tissue.

In all the muscles examined phosphate-carbonate was seen in the muscle cell, as well as in the connective tissue surrounding it. It was determined by the difference between the color intensities produced by two reagents, one an acidified silver nitrate nearly saturated with silver chloride and the other a silver nitrate nearly saturated with silver chloride and silver phosphate. It was found that the distribution of phosphate-carbonate is uniform throughout the cytoplasm and does not have a linear arrangement as described by earlier histochemists.

Potassium was determined with sodium cobalti-nitrite at a low temperature. The crystals representing potassium are seen uniformly throughout the whole cell cytoplasm. There is some variation in the density with which the cell is packed but it is slight. Perhaps the variations are not easily seen because the crystals are so numerous. Like phosphate-carbonate, potassium is distributed uniformly in the muscle fiber, without relation to longitudinal or transverse striations. Dr. Gersh in these observations has been able to demonstrate the intracellular distribution of these four ions in striated muscle fibers, concerning which we have heretofore depended on deductions and assumptions. He confirms by direct histochemical methods the fact that potassium is present in very large amounts in the cytoplasm, that phosphatecarbonate is present in moderate amounts, and that chloride is completely absent.

Distribution of Chloride in Gastric Mucosa

The methods which we have seen developed in the preceding paragraphs provide a technique which is sensitive, specific, and devoid of diffusion errors. With it Dr. Gersh has attempted to clarify the problem of the site and mode of formation of hydrochloric acid in the stomach. The gastric mucosa of dogs was studied during the resting phase and during the height of activity. He found that during rest chloride is absent from the cytoplasm of all gland cells. Traces of it are, however, present in the secretion granules of the zymogenic gland cells. Chloride is found in the lumen of the gland and in still larger amounts in the gastric pits into which the gland lumina empty. Furthermore, it is found evenly distributed throughout the connective tissue spaces of the mucosa and submucosa.

In the actively secreting mucosa the distribution of chloride was found to be essentially the same as in the resting stage. It was found in the same places but was present in greater amounts. More chloride is visible in the cytoplasmic inclusions of the zymogenic cells and the spaces of the connective tissue. The cytoplasm of all the glandular cells is free of chloride, just as was seen in muscle fibers.

Dr. Gersh finds himself unable to explain the presence of chloride in the secretion granules of the zymogenic cells. Its presence in the connective tissue spaces is naturally correlated with greater amounts of chloride-containing tissue fluid and lymph, formed during glandular activity. It is suggested that the protein chloride is extruded from the parietal cells as rapidly as it is formed and none of the compound is stored in the cytoplasm. It may still be true that the parietal cell is the one whose specific activity results in the secretion of hydrochloric acid, even though our present methods fail to detect it in its early secretory phases.

Histochemistry of the Fate of Colloidal Calcium Phosphate

A series of experiments have been conducted by Dr. I. Gersh dealing with the metabolism of calcium and phosphorus in the blood. Following intravenous injections of rats and dogs, in some cases with calcium and phosphate salts, and in other cases with colloidal calcium phosphate suspended in horse serum, he was able to follow the mechanism existing for the rapid removal of excess colloidal calcium phosphate from the blood, namely, by the phagocytosis of the particles by the macrophages of the liver and spleen and to a less extent of the bone marrow. The phagocytosed particles are retained by the macrophages a relatively short time and are then liberated. The whole mechanism is a transitory one. His experiments show that the degree of phagocytosis is influenced by variations in the amount of colloidal compound present in the blood stream. The macrophages of the lymph nodes take no part in this, as long as the colloidal particles are confined to the blood stream. If, however, colloidal calcium phosphate is injected subcutaneously or into the muscles so that it enters the lymph and tissue fluid, phagocytosis in lymph nodes may then take place.

As has been said, the colloidal particles of calcium phosphate are retained in the cytoplasm of the macrophages but a short time and are then liberated. This subsequent liberation of the calcium and phosphate into the blood appears to be responsible for the delayed rise in serum calcium which is experienced particularly after the intravenous administration of phosphate. It is proposed by Dr. Gersh that the granules of calcium phosphate in the cells may be turned into the constituent calcium and phosphate ions, which leave the cells as fast as they are formed. While this process continues the calcium and phosphate in concentrations in the blood plasma are reduced by their passage through the capillaries into the urine, feces, and tissue fluid.

The histochemical method used by Dr. Gersh for testing the content of phosphate in the sections revealed it as a precipitate of black or brown granules in the phagocytic cells. For testing the presence of calcium in the sections, after the freezing-drying-embedding technique and the removal of the petroleum ether, they are treated with a water solution of sodium alizarine sulfonate, which is drained off and replaced by chemically pure glycerine. Calcium in such sections appears as an orange color and the granules are less numerous and not as discretely separated as after the application of the silver reagent. With this much accomplished it is not too much to hope that a test may be found that will reveal these substances differentially in living cells.

TISSUE CULTURE

Lymphocytes and Monocytes

Using lymph node explants, Dr. W. H. Lewis has studied the emigration from them of lymphocytes and small monocytes. As they first make their appearance, they are about the same size and could be mistaken for each other. Dr. Lewis finds, however, that they are two different types of cells and that the lymphocyte does not become transformed into the monocyte. The two types can be distinguished by their mode of locomotion as well as by their morphological details. The lymphocytes may divide but they rarely live more than a few days. The monocytes, in contrast, increase in size, multiply, and survive for many days.

Cultivation of Cancer Cells

A review has been published by Dr. W. H. Lewis of his studies on the cultivation and cytology of cancer cells. This investigation has extended over a period of several years and includes observations on 17 rat sarcomas and 160 dibenzanthracene mouse sarcomas. He confirms his previously expressed opinion that cancer cells are permanently altered cells. They are new types, derived from normal and usually healthy ones, which have been permanently altered by environmental influences or various sorts of agents.

After the normal cells have undergone the initial alteration into malignant cells the special environmental influences or agents which produced them are no longer necessary for the maintenance of their peculiarities.

Dr. Lewis finds that in tissue cultures the maligant cells are visibly different from normal ones and from each other. They retain their peculiar characteristics from one generation to another when serially transplanted from animal to animal and for months or even years. From six of the tumors malignant cells were cultivated "in glass" for two to over four years. In that way pure colonies of the malignant cells were obtained which retained both their cultural characteristics and their malignancy. When inoculated into animals typical tumors resulted. From these in turn pure colonies of characteristic malignant cells were secured and these also produced typical tumors.

In the dibenzanthracene mouse sarcomas there were not as many varieties of malignant cells as in the rat tumors, nor were they as different from one another as were those of the rat sarcomas. The malignant cells from them were, however, visibly different from normal ones, and they could be followed in subsequent generations by comparison of photographs of the primary tumors with those of descendent cultures.

In roller tube cultures where cells have been carried on for three or four months, there were found many large multinucleated giant cells of skeletal muscle origin in addition to the fibroblast type of malignant cells. Further study has revealed that these giant cells are muscle cells altered by the invasion of the muscle by malignant cells but are not themselves malignant.

Transplantability of Induced and Spontaneous Tumors

A study of the transplantability of sarcoma tumors induced by dibenzanthracene and of spontaneous mammary gland carcinomata in four pure inbred strains of mice has been made by Mrs. M. R. Lewis and Mrs. E. G. Lichtenstein. They found that every one of the 200 tumors they induced in their four inbred strains proved to be transplantable into all the mice of the strain in which the tumor arose. Out of 123 tumors which were transplanted to alien strains only 9 grew, that is, 114 proved to be strain specific. This resistance of mice of one strain to tumors transplanted from another strain was referred to in a previous report (Year Book No. 35). It was also reported at that time that this resistance could be broken down. Since then more has been learned as to the range within which this can occur. Fourteen of the tumors that were strain specific when implanted in normal young mice of alien strains lost their strain specificity when they grew in a repeatedly implanted host of an alien strain, so that they were thereafter transplantable not only into mice of their original strain and of the treated strain, but also into mice of other strains.

In the study of 80 spontaneous mammary gland tumors it was found that they not only were strain specific but also largely retained this specificity even though repeated inoculations were made into hosts of alien strains. There were a few exceptions in which the tumors became transplantable into mice of a closely related strain. Where mice already had a spontaneous mammary gland tumor, they were not immune to the development of additional spontaneous mammary tumors. Some of the mice developed as many as five mammary gland tumors during the course of two months. These same animals, however, were somewhat refractory to the growth of similar tumors grafted from other mice even though of the same strain. The individuality of tumor growth was well shown in a series of experiments in which a carcinoma was implanted on one side of a mouse and a sarcoma on the other. Both tumors grew in such cases and were transplanted through many generations in this manner. They each retained their own characteristics, and microscopic examination showed one to be an adenocarcinoma and the other a large spindle cell sarcoma.

A Useful Culture Tissue

It has been found by Dr. W. Mendelsohn that explants from the testicle of the adult rabbit can be successfully grown in roller tube cultures. From them he obtains extensive growths of epithelium suitable for testing the viability of epithelial cells in the present of arsenical drugs, hormones, toxins, and viruses. These growths also are suitable as a culture medium for the growth of Spirochceta pallida.

Fragments of the testicle are distributed inside the tubes and held in place by heparinized chicken plasma. As a nutrient medium rabbit serum is used with a balanced salt solution. Locke solution can be substituted for the balanced salt solution. After seven days areas of liquefaction occur near the explants but these can be repaired by clots of chicken plasm and fresh nutrient fluid substituted for the old. In this way the cultures can be maintained three weeks or longer if desired. There appears to be no difficulty in making permanent mounts of such growths.

The first outgrowths from the explants are chiefly fibroblasts and endothelial cells. Later broad sheets of germinal epithelium extend from the convoluted tubules. Occasionally Sertoli cells are seen. Apparently spermatozoa may undergo maturation from the spermatocytes in this roller tube culture.

STUDIES OF THE PITUITARY GLAND

Tissue Cultures of the Hypophysis

The study of the pituitary gland in tissue culture has been followed for several years by Mrs. M. R. Lewis. She found that such cultures taken from all the ordinary laboratory animals grow rapidly and abundantly, regardless of the age of the animal from which the gland is obtained. They grow particularly well in tube cultures. The cells retain their differentiation and continue to form their specific granules, though the granules are fewer than in the original tissue.

It was found that cells grown in cultures for as long as fifty days continued to produce blood-pressure-raising and melanophore-expanding hormones. In an effort to correlate cell types with specific hormones animals were sought in which cell types were so segregated that particular regions could be dissected out for transplantation. In the mouse the pars intermedia is free of pars nervosa tissue. In the chicken and armadillo the pars nervosa is free of other lobe tissue, whereas in the dogfish and skate the pituitary cells are segregated into six separate lobes. The melanophore hormone was located in two of these latter lobes and continued to form in cultures. Apparently tissue culture affords a valuable tool for determining specific functions of the pituitary cell groups, at least in the case of the hormones belonging to the posterior lobe. The hormones of the anterior lobe are more complicated in their activity and this has made them more difficult to identify. A review of the present status of her studies was given by Mrs. Lewis before the Association for Research in Nervous and Mental Disease.

Cytology of the Hypophysis

In a previous Year Book an account was given of the studies of Dr. I. Gersh on the relation of the histological structure of the posterior lobe of the hypophysis and the pressor, oxytocic, and antidiuretic hormones which are generally assumed to originate there. He was able to show that the hyaline bodies of Herring, which had been thought to be a secretion antecedent, are an artifact and are not present when the tissue is prepared under the best conditions. This left only neuroglia cells as the local source ; unless one concluded that these three hormones were made elsewhere and transported by the blood stream and deposited through means of a special capillary permeability and stored in the posterior lobe.

During the past year further study of the parenchymatous cells of the posterior lobe in rats has led Dr. Gersh to the conclusion that some of these cells are really glandular and that they produce and secrete the antidiuretic hormone. This posterior lobe glandular cell has been found by him in a wide variety of mammals and in pigeons and in chickens. It is characterized by the presence of either granules or lipoid droplets which can be seen in fresh mounts and which fill the cytoplasm and extend out into the cell processes. In the rat these droplets are rich in neutral unsaturated fats ; in some animals there are no visible lipoids. A characteristic feature of this glandular cell is that it appears early in embryonic life and that the number and size of the cells and of their inclusion bodies increase throughout life, reaching their greatest prominence in rats two and a half years old. Owing to the fact that there is a normal range in the number and size of these posterior lobe glandular cells in any particular gland, Dr. Gersh was able to show that fluctuations occur within this range which are correlated with the dietary intake of water. This fluctuation could be controlled experimentally. After rats have been restricted to a relatively dry diet for a week the glandular cells are present in greater number and are larger. On the other hand when the experimentally dehydrated rats are given free access to water, the number and size of the differentiated cells promptly revert to the normal range prevailing in untreated rats. Thus Dr. Gersh establishes a significant correlation of cellular activity and morphology with the hypersecretion of antidiuretic substances, and perhaps also of oxytocic substances.

It may be added that among the parenchymatous cells of the posterior lobe there are many that are relatively undifferentiated, but which during the hyperplasia, which follows the stimulus to hypersecretion, become transformed into the fully differentiated form. The differentiated and the relatively undifferentiated varieties vary inversely in number.

Nerve Terminations in the Posterior Lobe

Using pyridine silver preparations and fresh pituitaries of young rats perfused with methylene blue, Dr. C. McC. Brooks and Dr. I. Gersh have succeeded in demonstrating nerve fibers which pass down on the hypophyseal stalk to terminate in pericellular baskets around the glandular cells of the posterior lobe described in the preceding paragraphs. The endings almost completely enclosed the cells in a close-meshed network. The fibers belong to the hypothalmic hypophyseal tract. It was found that they are unaffected by the removal of the superior cervical ganglion, which would remove them from the category of the sympathetic chain, if that were necessary.

The demonstration of this nerve supply to the posterior lobe glandular cells gives us an explanation of the phenomena of pseudo-pregnancy, which had been recognized as requiring the transmission of nerve stimuli. It also becomes clear that section of the pituitary stalk would lead to degeneration of these terminations and an accompanying severe diabetes insipidus.

REPRODUCTIVE SYSTEM AND ENDOCRINOLOGY

The Gilfillen-Gregg Skin Test for Pregnancy

A theory has been advanced that the abundant supply of prolan in the blood of pregnant women should render these individuals immune to further injections of urinary prolan, whereas the non-pregnant woman should be prolan-sensitive. Were this true it would constitute a simple and economical test for pregnancy. Two guest investigators in the laboratory, Dr. S. Saglik and Dr. E. Scipiades, Jr., have tested this technique on animals. After failing to obtain differential reactions in rats, guinea pigs, and rabbits, they made experimental injections of prolan in a series of monkeys, in a few of them in the form of antuitrin-S and in others in the form of follutein. Here, too, the results proved negative.

For purposes of control and as a test of the potency of the hormone used on the animals, skin tests were made with follutein in pregnant and nonpregnant women. Of 19 non-pregnant women only 11 gave the postulated reaction, whereas of 23 pregnant women, whose pregnancies varied between 16 and 40 weeks, 6 reacted in a manner expected only in non-pregnant women. It is thus clear that this test is not sufficiently reliable to replace the standard Aschheim-Zondek or Friedman test.

Time of Ovulation

In reporting observations on the formation of uterine epithelial plaques in the process of implantation of young monkey embryos, Dr. C. G. Hartman assembled and reported his records on ovulation. Among 300 ovulations accurately diagnosed, all occurred between days 8 and 16 with the exception of 5, which fell irregularly outside that period. This seems to be a mechanism whose precision is rarely surpassed among biological phenomena.

Alleged Birth of Triplets in the Macaque

Though multiple births may occur in the rhesus monkey it has been shown by Dr. C. G. Hartman that one must be on guard against being misled by a tendency to kidnaping that prevails among certain aggressive mother monkeys. In a case of apparent triplets it was possible for Dr. Hartman to show that two other recent mothers in the same cage had been deprived of their young, and also he obtained the conclusive evidence provided by the fact that the appropriative mother had but one recent corpus luteum.

The Anthropoid Ovary

The opportunity of studying the ovaries of three gorillas, two chimpanzees, one orang-utan, and one gibbon has been well utilized by Dr. S. Saglik, a guest of this laboratory from the Gynecological Clinic, Gulhane Hospital, Istanbul, in giving us an analytic description of the anthropoid ovary. He has also compared these ovaries with those of man and with those of the Old and New World monkeys.

Dr. Saglik finds that primate ovaries can be arranged in a series on the basis of their general similarity to the human ovary and he arranges them as follows: orang-utan, chimpanzee, gorilla, macaque, cebus, gibbon, Allouata, and Ateles. Here then we have another organ which would call for a very different phylogenetic tree from that demanded by the skeleton.

Menstruation

A study of the incidence of menstrual cycles without associated ovulation has been made by Dr. C. G. Hartman on 300 female monkeys of the Carnegie colony, concerning all of which he possessed fairly complete biological records. These animals with few exceptions were purchased from animal dealers and about one-third of them were superior specimens. Such animals ovulate either at once after arrival or at least after a few months' period of acclimatization. Another third of the animals received were inferior ones that either did not menstruate at all or menstruated without ovulation. The remaining third of the animals were intermediate in quality. They remained in excellent health but ovulated less reliably or in some instances never. This material provided Dr. Hartman with the opportunity of studying a very large number of cycles, a sufficient number to determine their principal variations. Of particular importance were his determinations of the frequency of nonovulatory cycles.

On analyzing the non-ovulating monkeys he found that the cycles could be separated into two groups, those in which the non-ovulatory cycles may be regarded as normal, and those in which the occurrence is pathological. His records include 1000 cycles in which non-ovulatory cycles occurred during the non-breeding season, from May to September. This appears to represent normal behavior. Also non-ovulatory cycles are normal in adolescence, of which there were 240 records. Likewise in the pre-climacterium and during recovery from pregnancy and lactation non-ovulatory cycles are normal. On the other hand, in 260 animals there were 1075 cycles in which there was no ovulation and which must be classed as abnormal or pathological. Some of these animals were palpably sick, and either did not menstruate at all or menstruated without ovulating a few times before death. About 30 per cent of the animals fell in this group. Then there are some apparently healthy animals which never ovulate or else only occasionally ovulate. Of the observed non-ovulatory cycles 17 per cent belong in each of these two groups. Some animals are unpredictable and ovulate about one-half the cycles of the breeding season. Such animals yielded a number of our finest embryos. Recently acquired animals are very likely to skip their ovulations for a time after their arrival. Some were with us two years before their ovulations started, some one year, and others began ovulating during the first year. Then among even the best animals there were some who occasionally failed to ovulate at the normal times. About 10 per cent of the nonovulatory cycles fell in this group. From these records it is seen that the non-ovulatory cycle is a very definite thing and becomes a factor that must be reckoned with in consideration of the occurrence of sterility.

Further studies have been made by Dr. Hartman on the hormonal control of menstruation. He has found that by periods of daily administration of testosterone in a monkey which has previously been regular in its menstrual cycles the cycle can be lengthened to 38 days as against 25 to 28 days previous to the experiment. In animals which have been castrated the bleeding which usually follows within a few days was inhibited by daily administration of testosterone. Also the menstrual bleeding which in favorable animals uniformly occurs following the injection of amniotin can be inhibited over prolonged periods. Thus Dr. Hartman shows that testosterone has an action upon menstrual control closely simulating that of progestin, just as it also simulates progestin in stimulating mammary development and in inhibiting the vaginal mucosa.

At this point reference should be made to the light thrown on the menstruation problem by the studies of Dr. J. E. Markee done in cooperation with Dr. Hartman. They made transplants of endometrium according to Dr. Markee's method and were able to follow the vascular changes by direct observation. This work is now in course of final preparation for publication. It will be reviewed in full in my next report.

Hormone Injections in Young Alligators

Before his appointment on the Johns Hopkins staff, Dr. T. R. Forbes had already, under Dr. R. K. Burns, Jr., made his experiments on the induction of a precocious development of the reproductive tract in the immature alligator by the administration of hypophyseal extracts. He had also studied the effects of female sex hormone injections (cestrone) in young alligators and found that it produced a marked hypertrophy of both ovarian and testicular cortex, along with greatly hypertrophied oviducts in the females and some development of the male vestigial mullerian ducts. This work has been published during the past year. Dr. Forbes has continued his hormone studies on the sexually immature alligator. He has investigated the effects of prolonged injections of testosterone in recently hatched animals and found them responsive to this hormone. In 14 females moderate hypertrophy of the oviducts took place, although the change in the ovaries was less definite. In the male the testes were twice as large as those of the control animals and there was definite hypertrophy of the vestigial mullerian ducts and of the penis. The wolffian ducts and the wolffian bodies were found unresponsive in both sexes to injections of these sex hormones. If we accept them as embryonic organs that perform an essential but temporary service for the embryo it removes them from the group of reproductive organs and would explain their failure to respond to either testosterone or cestrone.

Tissue Culture of Endocrine Organs for Purposes of Transplantation

Some tentative experiments have been made by Dr. G. 0. Gey toward obtaining pure strains of thyroid and parathyroid cells in continuous tissue culture from which homologous grafts can be made on individuals who through deficiencies of their own need these specific endocrine secretions. Thus far a number of successful grafts of thyroid cultures have been made in dogs and also a few parathyroid grafts. The advantage of tissue culture grafts is that in this way the desired endocrine cells can be acclimated to a tissue culture medium that is composed largely of the recipients' plasma and serum, thereby increasing the probability of their survival and functional activity.

Experiments on Castrated Animals

In recently castrated young male rats it has been found by Dr. J. Ball that the female hormone estrin, if given in daily injections of proper amounts (50 to 100 rat units), will definitely increase or completely restore their sex activity. The experiments were conducted on six male rats which were castrated at about four months old and the tests were begun two weeks later. Quantitative records were made both of their mating behavior and of their motor activity as registered by the revolving drum. The amount of hormone used was regulated by the response of the individual animal. After the castrate level of sex activity had been determined a sufficient amount of the hormone was given in daily injections to bring out an unquestionable response in each rat. From the results of these experiments Dr. Ball reached a conclusion that will be of interest to students of behavior, namely, that the function of this estrogenic hormone in the adult animal is not so much to organize the mating behavior pattern as it is to activate a pattern already laid down through other influences.

New observations have been made by Dr. C. G. Hartman on the results of castration in pregnant monkeys. Two animals castrated at the end of the third month of gestation carried their fetuses to full term. Three others castrated on the 46th, 35th, and 31st day, respectively, were progressing normally two months later. These results harmonize with the fact observed by Drs. Hartman, Corner, and Bartelmez that the corpus luteum of the rhesus monkey is active only during the first four weeks of pregnancy, at which time it markedly regresses. Theoretically its removal at any time thereafter should not interfere with the continuance of pregnancy and, as seen in these experiments, it does not.

CENTRAL NERVOUS SYSTEM

Cerebrospinal Fluid

In his lectures given at the University College, London, Dr. L. H. Weed pointed out the importance to many fields of anatomical investigation of intimately combining anatomical and physiological thought, and paying equal regard to structure and to function. It was not necessary for Dr. Weed to point this out, since his own example in the study of the coverings of the brain and the specialized fluid that is contained within them had long since provided us with a brilliant demonstration of the advantage of blending these two disciplines.

The researches of Dr. Weed upon the cerebrospinal fluid now reach back 25 years. During that time he, along with a group of able coworkers, has unraveled an important series of fundamental problems regarding this intricate system, concerning which little was known when his investigations were starting. Under his guidance we have seen, as the readers of the Carnegie Year Books will know, the histology and embryology of the meninges clarified. This was followed by the demonstration of the sources of the cerebrospinal fluid and the pathways of its return to the venous system. There then followed the series of experiments which showed that the absorption of cerebrospinal fluid is the product of two factors, a hydrostatic one being the difference between the subarachnoid pressure and intracranial venous pressure, and the other the colloid osmotic pressure of the blood. This promptly led to the revelations regarding the pressure relationships between the cerebrospinal fluid and that in the cerebral veins, under the principle of the bony wall of the cranial cavity and the vertebral canal serving as a rigid container. There was then found the provision of elasticity which permits a dislocation of fluid on change in position by means of compensating dilatation and contraction of the intradural vascular bed, the cerebral venous pressure at the same time remaining constant. Associated with the latter experiments, a better hypothesis could be arrived at as to the primary function of the cerebrospinal fluid. It is owing to these researches of Dr. Weed that we now see it as a means of providing a prompt reciprocal volume and pressure adjustment when changes occur in the volume of the vascular bed or in the nervous tissue.

Effect of Inactivity on Nutrition and Growth of Muscle and Bone

The investigations of Dr. S. S. Tower on the isolation of the lumbar enlargement of the spinal cord in young growing animals has been extended to its trophic effect on the muscles and bones normally innervated from that source. In her experiments the cord was transected above and below the lumbo-sacral enlargement and all its posterior roots cut. When this is done the dependent muscles lose all ordinary activities including muscle tone. It was found that for purposes of these experiments young animals may survive several months and provide us with a method of studying the regressive changes which follow nerve section. Also in such experiments one can discriminate between effects of inactivation and those of nerve degeneration.

The three puppies used by Dr. Tower were studied 2, 5, and 6 months respectively. By the nature of her experiment all ingoing nerve impulses were excluded from the isolated cord, which nevertheless survived along with its dorsal root ganglia and peripheral nerves, without developing within itself any nervous activity. Since Dr. Tower had previously shown that severing of the posterior roots is without appreciable trophic influence on skeletal muscle, any trophic disturbances resulting from these experiments would therefore have to be ascribed to the inactivation.

It was found that atrophy and metaplasia of muscle tissue into fibrous tissue, atrophy and destruction of subsarcolemmal nuclei, and interstitial fibrosis are all characteristic of inactivation of skeletal muscle. Macroscopically such muscles can be seen to be atrophied and they develop contractures. Microscopically the fibers are smaller in all dimensions, pale staining, and in the process of transformation into fibrous tissue, and the interstitial fibrous tissue is increased. The innervation remains largely intact.

All these things occur also in denervated muscle. But when a muscle is denervated it shows a rapid proliferation and change in character of the subsarcolemmal nuclei, which changes do not follow inactivation alone. This specific nuclear proliferation must be attributed to the degeneration of nervous tissue. Dr. Tower thus finds that the trophic control of muscle by the nervous system requires both physical integrity of innervation and nervous activation.

In analyzing the effects of inactivation on the postnatal growth of bone it was found that the long bones of the leg were normal in length and in their general configuration, features which appear to be intrinsic. In thickness and certain details, such as elevations at muscle attachments, they were underdeveloped. These then depend on extrinsic factors which in the above experiments were abnormal because of the presence of muscle inactivity. Special trophic nerves continue to be unnecessary to Dr. Tower for the discussion of the nature of trophic control of tissues.

Electrophysiology of Nerves

It is only recently that any of our group have participated in investigations on the electrical properties of functioning nerve fibers. During the past year Dr. H. A. Howe in cooperation with Dr. D. A. Clark has made observations on fiber action potentials in the fiber tracts located entirely within the central nervous system. Observations on fiber action potentials had previously been restricted to peripheral nerves.

Dr. Howe and Dr. Clark studied the changes produced in the electrical potentials of the tracts within the cervical spinal cord following induction coil stimulation of the pyramidal tracts which lie on the ventral surface of the medulla oblongata, in the cat. They were able to demonstrate potentials somewhat analogous to those characteristic of peripheral nerves. The responses, however, were very complex and evidently included the activity of many fiber pathways. Owing to the structure of the cord it was not possible to determine the correlation between conduction velocity and threshold as has been done with peripheral nerves. Under the conditions of their experiments they obtained potentials which in form, rate of conduction, and resistance to asphyxia gave the picture of neuronic fiber activity without synaptic intervention.

These investigators applied their stimuli to the pyramidal tracts by means of a bipolar electrode having two silver contacts set flush in the end of a bakelite rod. This rod was inserted tightly into a trephine hole through the base of the skull, in a manner that avoided blood loss or leakage of cerebrospinal fluid. The resultant disturbances were registered by coaxial needle electrodes inserted into the cord at different distances from the point of stimulation. The two levels chosen were an upper one at the level of the second cervical vertebra and a lower one at the level of the fifth cervical vertebra.

Dr. S. S. Tower participated in a study of impulses as they pass through a sympathetic ganglion and into the nerves beyond it. She had the privilege of working with Dr. D. W. Bronk and his associates, who are experienced investigators in the field of electrophysiology of nerves. Their experiments consisted in stimulating the preganglionic fibers, rami to the stellate ganglion from the spinal cord, and recording the action potentials in the relatively long inferior cardiac nerve, to which the impulses were transmitted through the ganglion.

It was found that the conduction velocities in the nerve studied had a considerable range (1.4 to 0.6 meters per second), and there is a considerable temporal diversity in the maximum potential. Following the peak there is a positive after-potential which may be increased during the course of a tetanus. Likewise a negative after-potential develops after a tetanus. A rested nerve does not generally show a negative after-potential.

The significance of the action potential records which are yielded by the oscillograph is not altogether clear, and it is necessary at present to study them in all their details and under all possible experimental conditions. Among other things it was found by these investigators that a rapid series of preganglionic impulses initiates a single but dispersed series of postganglionic impulses. The records show that the individual nerve cell discharges but one impulse for each series. The dispersion appears to be due to differences in conduction time for the various fiber pathways through the ganglion. It was also found that the number of ganglion cells that respond to a preganglionic stimulus may be modified by various things. Arrest in the blood circulation decreases the number, whereas repeated stimuli can build up a larger response even in a non-circulated ganglion. Perfusion of a ganglion with drugs also modifies the nature of the responses. Although some of the terminology, the technique, and the character of the records are somewhat confusing to one who is not oriented in such matters, it is to be remembered that the workers in this field are collecting observations that lead to something more tangible as to the nature of the conduction of nerve impulses than the pure speculations which formerly were our sole resort. Some of the ground work in the electrophysiology of protoplasm was done with Valonia, that interesting primitive organism which abounds in the waters at the Tortugas Laboratory.

Regeneration of the Facial Nerve and Associated Tics

In Year Book No. 35 a brief account was given of branched axones of the facial nerve in monkeys following experimental injury of that nerve. The results of those experiments have been published during the past year in final form in the Archives of Neurology and Psychiatry under the authorship of Dr. Howe, Dr. Tower, and the late Dr. A. B. Duel, who were aided by a grant from the Carnegie Corporation. These investigators have established, both physiologically and anatomically, the fact that in regeneration of the facial nerve, following injury, its axones at that site undergo branching and subsequently innervate widely separate muscles and that in this way an irrevocable functional union takes place between muscles which do not normally contract simultaneously. There thus occurs a condition which resembles the tics which in humans follow injury of this nerve, and is characterized by an indiscriminate mass contraction whose proportion varies as the amount of damage to the nerve. These contractions show no tendency to regress, even over a three-year period. In man there is a better expectation of being able to suppress or modify the tic movements through reeducation.

Imitative Behavior in a Monkey

A case of what appears to be imitative behavior in a young rhesus monkey has been studied by Dr. J. Ball. Being caged with a kitten for company, this 11-month-old animal learned to drink liquids by lapping, copying perfectly the technique of its companion. The new method of drinking continued several months, as long as the animal lived. The normal way of drinking and the one originally employed by this animal is a process of sucking. Among 600 rhesus monkeys whose habits have been closely followed in this laboratory, none have ever been observed to drink by lapping in this open-mouthed fashion. Dr. Ball concludes that this case can be interpreted as imitation.

Brain of the Whale

Another contribution to the structure of the brain of the whale has been made by Dr. 0. R. Langworthy in collaboration with Dr. F. A. Ries. Because of its many interesting adaptations to the requirements of marine life, Dr. Langworthy has made the whale brain the subject of several investigations, as the reader of these reports will know. The present investigation includes eight additional brains of the sperm whale, Physeter catodon, making Dr. Langworthy 's collection, now housed in the Department of Neurology of the Johns Hopkins University, particularly adequate for such study.

Defective Brain Development

Dr. P. A. Fitz-Gerald of the Department of Anatomy, University College, Dublin, while a guest of our laboratory during the past year has made a study of the cerebral hemispheres of an eighth month child showing defective brain functioning. In the central and parietal regions the cortex of this child, instead of being properly fissured, was found to be almost smooth, and symmetrically so on the two sides. All the other cortical areas appeared to be normal. Dr. Fitz-Gerald associates the stunting of sulcus formation with a developmental arrest in cortical histogenesis and in this way he is helping to solve the problem which has long confronted the embryologist, of what is the exact nature of the forces that produce fissures and convolutions on the brain surface. He is following his surface survey of the material with a microscopic study of the tissues involved.

MORPHOLOGICAL STUDIES

Clavicles and Long Bones of the Limbs in Man and Apes

In order to obtain suitable data bearing upon variability and asymmetries in higher primates and their comparison with man, Dr. A. H. Schultz has collected measurements and observations on a total of 753 human skeletons belonging to a variety of races, and a total of 530 simian skeletons belonging to all the genera of anthropoid apes and to macaques. The data have been obtained with the same technique and for the most part by himself and with the primary intention of its use for the study of variability and asymmetry. With it he has been able to make comparisons between some civilized and uncivilized races of man and comparisons between man and other primates, particularly the anthropoid apes. Heretofore we have had infinitely more data on asymmetries and variation in man than in other animals. Dr. Schultz has now provided the information regarding other related forms which we needed for evaluation of the data which were already available for man.

Among his observations on asymmetries he found, in comparing their distribution in man with that in apes and the macaque, that in general symmetry is rarer in the former than in the latter and that preference of asymmetries for one side is not nearly as marked in the apes as it is in man. In the macaque asymmetries favor both sides with practically the same frequency. In regard to asymmetries of the lower extremities man does not differ essentially from the other primates. In all primates there is comparatively little preference of one lower extremity over the other. Asymmetries in the lengths of the clavicles favor both sides with practically equal frequency in gorilla and chimpanzee and the right side slightly more frequently than the left in orang-utan, gibbon, and the macaque. This contrasts strikingly with the conditions in man, in whom there is a definite tendency for the left clavicles to be longer. In all the human groups the lengths of the long bones of the arm favor the right side in the great majority of cases, and the asymmetries of the arms favoring the right side are more frequent in females than in males among whites, Negroes, Eskimos, and Indians. Since Dr. Schultz had previously shown a similar prevalence of asymmetries in human fetuses, he concludes that the common preferential use of the right arm in man cannot be held responsible for its definite tendency to be longer. Nor is it likely connected with "right-handedness" since "left-handedness" is much rarer in man than are asymmetries favoring the left arms. Furthermore "lefthandedness" is regarded as hereditary, whereas asymmetries of the human body are thought not to be.

In general the tables of Dr. Schultz demonstrate conclusively that man differs strikingly from apes and monkeys in regard to both the percentage distribution and the relative amount of asymmetries of the clavicles and the long bones of the arms ; and that man and other primates are practically alike in regard to the degree of asymmetries and the difference in the preference of asymmetries for the two sides in the long bones of the legs.

Vertebrae and Length of Spinal Regions in Primates

From observations made on 300 freshly killed catarrhine primates, with measurements made from the centers of the intervertebral disks, Dr. Schultz has been able to show that the cervical and the thoracic vertebrae are proportionately larger in all higher primates than in the lower catarrhines, and that among all primates man has the relatively longest cervical and thoracic regions and the comparatively largest lumbar vertebrae. However, the common evolutionary trend among higher primates to reduce the number of vertebrae and the relative length of the lumbar and caudal regions has gone to greater extremes in some anthropoid apes than in man.

In a series of 80 adult gibbons, having a much greater vertebral variability than man, it was found that a reduction in the number of thoraco-lumbar vertebrae is not accompanied by a corresponding reduction in the relative length of this region. On the other hand, a close correlation exists between decreased numbers of thoraco-lumbar vertebrae and increased numbers of sacral vertebrae. Less frequently the sacrum increases its number of vertebrae at the expense of the coccygeal region.

As compared with the lower catarrhines, man and the anthropoid apes differ not only in possessing fewer thoraco-lumbar and caudal vertebrae and more sacral vertebrae but also in having comparatively longer cervical, thoracic, and sacral regions and much shorter lumbar regions.

Shoulder Architecture

In previous reports reference has been made to the studies of Mr. Brazier Howell on the architecture of the shoulder in the vertebrate classes, including Amphibia and Reptilia. To these may now be added studies on the shoulder region of birds and therian Mammalia. In his description of the domestic fowl Mr. Howell has made a contribution to comparative anatomy by the interpretation of the avian shoulder in terms of the tetrapod animal. As in his other studies, due emphasis has been given to innervation in all questions of muscular homologies.

Among the noteworthy details pointed out by Mr. Howell are the following: the spinal accessory nerve and its associated muscles are absent, being replaced by a suboccipital group; m. levator scapulae is lacking; the rhomboids occur in two layers; the subscapularis is poorly represented, whereas the dorsalis scapulae and deltoid are robust; the large breast muscle represents the pectoralis minor element, the major being small and deep; supraand infraspinati are absent; and the brachialis is a feeble muscle near the elbow.

In the anatomy of the appendages Mr. Howell finds a large break between those of the therian mammals and those of reptiles and even those of prototherians. In fact he finds that the monotremes are more comparable with reptiles than they are with therian mammals. For that reason he omits them from his analysis of the mammalian shoulder. He accounts for the great dissimilarity in the pectoral appendages of the above groups by the differences in the way the limbs are used. He points out the prone position of the reptilian body and the horizontal position of the humerus, with divergent elbows, features which call for a very different skeletal and muscular provision from that with which we are familiar in mammals. The reptilian type of shoulder architecture is quite unsuited for quick movements and longsustained action. In order to change the reptilian into the mammalian plan it was only necessary for reptiles to bring the elbow beneath the body. But this required such complicated skeletal and muscular adjustments about the shoulder joint that only a single reptilian group (Theriodontia) ever succeeded in its accomplishment. The improvement in function which resulted from the invention of the mammalian plan of shoulder seems to have played a large part in the development of the class Mammalia. The new deal provided the mammals with a means of using their limbs in a single plane, for purposes of locomotion, and over extended periods with the expenditure of much less energy. Over and above his interesting interpretations Mr. Howell's study has brought together a large amount of information regarding the shoulder region of mammals that will be of much value to the comparative anatomist.

Muscles of Hip and Thigh

The comparative anatomical studies which Mr. Brazier Howell has been making on the shoulder girdle have been supplemented by a similar method of analysis of architecture of the hip and thigh. His material includes the domestic fowl, the giant Japanese salamander, the reptile Iguana, and many varieties of mammals, a sufficient material for a comprehensive review of the homologies of the pelvic girdle.

It is pointed out by Mr. Howell that though the pelvic and pectoral girdles show many resemblances, yet the structures of neither pair can be properly homologized with those of the other, because of their difference in derivation. The cartilaginous pectoral girdle developed as an adjunct of the membranous girdle and is really a part of the head and axial skeleton. When the limbs became the primary organs of locomotion it was the pectoral limbs that propelled the animal by traction, and the girdle movement accompanying this action was accomplished largely by sidewise movements of the head. The membranous part of the girdle was derived from the posterior margin of the gill basket, whose musculature (trapezius) contributed to the control of the girdle. It had the complication of having a dual origin (membranous and cartilaginous) added to the fact of its location at the anterior termination of the axial musculature. With the pelvic girdle matters were quite different. The latter was initiated without the influence of anchorage to the axial skeleton, inasmuch as the pelvic appendages at first functioned for support only and not propulsion. For a long time they remained free of the axial skeleton, that is, from the viewpoint of the phylogenist. Furthermore the influence of the body muscles upon the pelvis were quite different from that upon the shoulder girdle. The story of the difference in functioning of the pectoral and pelvic limbs in the progress of their assumption of more complicated functions has been worked out in its significant details and Mr. Howell has provided the anatomist with a rationale for this region which is a definite advance over that heretofore available. With his four-group basis as the chief criterion, he has been able in large part to homologize the pelvic muscles of urodeles, lacertilians, mammals, and birds. Even so there remain some specializations which still obscure their precise relationships.

Visceral Anatomy of an Infant Chimpanzee

An infant female chimpanzee, 74 days old, which died of an acute pulmonary infection, provided Dr. W. L. Straus, Jr., with an opportunity to study the thoracic and abdominal regions at this early stage. The body was injected with a 10 per cent formalin solution within one-half hour following death, giving excellent preservation of the tissues. Anatomical data on the viscera of this important primate are relatively scant and this appears to be the youngest infant thus far systematically studied. The sitting height and trunk height, measured after fixation, were 31.5 cm. and 16.8 cm., respectively. On comparing this specimen with an older chimpanzee and with an infant orang-utan Dr. Straus found that there are but small differences in visceral morphology between the two chimpanzee specimens though there is a difference of nearly four years in their age, whereas the contrasts between the infant chimpanzee and infant orang-utan are many and striking, an observation that will not surprise those who are genetically minded. Dr. Straus has made his detailed descriptions and measurements available to other investigators by formal publication.

Branches of the Aortic Arch in the Monkey

In Year Book No. 35 the investigations of Dr. C. F. De Garis on variations in the branches of the aortic arch in the macaque monkey were referred to and at that time the value of having a large number of specimens from a single species was pointed out. To his first series of 115 specimens, he has now been able to add 153 more. This provides a total series large enough for significant statistical treatment and for the consideration of problems of variation, inheritance, and symmetry. The new material when arranged in polygons of frequency further substantiates a norm having a short truncus communis comparable to that often found in man. This norm has a marked modal value which is intermediate between human and mammalian patterns, with almost equal distribution of these patterns on either side of the norm. The next step appears to be the search for correlations of this norm with trunk measurements and visceral structure and the consideration of the influences of body symmetry. This will inevitably lead the investigator back into the fetal period, where a large part of the determination of the vascular pattern takes place.

Asiatic Primate Expedition

Joining forces with Professor H. J. Coolidge of Harvard University and Dr. R. C. Carpenter of Columbia University, Dr. A. H. Schultz participated in an expedition to northern Siam and British North Borneo in search of anthropoid material that is native there, and particularly the gibbon and orang-utan. Their program included comprehensive observations on the behavior and social relations of entire ape families as they live in their native jungles, and on the other hand the collection of skins, skeletons, embryos, parasites, stomach contents, bodily measurements of the dead animals with a view toward species characters, growth before and after birth, variability, incidence of disease and injury, and any facts relating to pregnancy. Through careful preparations for the expedition and through the interest and assistance of the authorities of the countries which they visited they were able to obtain data and specimens for subsequent study at their home laboratories, in amount far beyond their expectations. From his own standpoint the large collection of gibbon specimens that has thus become available more than justifies the time and effort that Dr. Schultz devoted to the undertaking. The skeletons are now being cleaned and prepared for study.

Physiological Observations on Fireflies

As a collateral to his cytological studies on insects Dr. J. B. Buck published during the past year his records on fireflies, an extensive collection of which were obtained by him on a visit to Jamaica in the preceding year while a fellow in the Zoological Laboratory of the Johns Hopkins University. In general he found that each species is rather definitely confined to a particular altitudinal range. A few species were found in a single district but most species are found in multiple regions where the altitude determines the appropriate temperature and moisture.

The observations by Dr. Buck on the spectral composition of the light emitted by the fireflies are of especial interest. Under the spectroscope the light emitted by all the species investigated produces a broad structureless band lying wholly within the visible spectrum. In no case was the light below 5050 or above 6550 Angstrom units. It is found that several species emit light of the same spectral composition, whereas others differ from one another. It is also noted that the spectra of some species present a relatively extensive range, e.g. 5050 to 6450 Angstrom units. Finally, Dr. Buck was able to demonstrate photographically that the apparently different color of the light emitted by the thoracic and abdominal light organs in certain species is due to an actual difference in the color and is not a subjective effect.

DEPARTMENT OF GENETICS

A. F. Blakeslee, Director

CHROMOSOME INVESTIGATIONS

A. F. Blakeslee, A. G. Avery, A. D. Bergner, S. Satina, H. E. Warmke, J. T. Buchholz, J. L. Cartledge, and E. W. Sinnott

Last year we reported the discovery, which had just been made, that treatment of seeds of Datura with the alkaloid colchicine would bring about changes in structure of the seedlings which were interpreted as due to doubling the number of their chromosomes. It was pointed out that if our interpretation were correct and the methods could be used with other forms, a tool of considerable value would be available both to the practical plant breeder and to the plant geneticist interested in problems of evolution. The present year's work has shown that induction of chromosome doubling by chemical treatment is of wide application among flowering plants and enables the investigator of certain problems to work with a measure of precision not hitherto possible. The polyploid series In, 2n, 3n, and 4n have been secured in Datura stramonium and may be expected in other species. At the beginning of the calendar year Dr. H. E. Warmke joined our group and since then has assisted in the polyploidy investigations which have been supported by a grant from the Carnegie Corporation to the Carnegie Institution of Washington. The polyploidy project is a single phase of the chromosome problem which has been occupying our group. Other phases of this larger problem are not being neglected but it seems best to restrict the report this year to a discussion of progress in the polyploidy project.

Methods of Inducing Chromosome Doubling by Treatment with

Colchicine

Soaking seeds in solutions of colchicine of different concentrations for different species is the most convenient method of treatment. Seeds of Portulaca respond to a concentration of 0.0002 per cent for two days by producing seedlings with swollen stems. Seed treatment with 0.4 and 0.8 per cent solutions for 4 to 8 days has been found well adapted to Datura and induces an abundant production of 4n branches. The effect of the drug is first noted in delaying germination and development. When the treatment is severe the stems of the seedlings are strongly swollen and many fail to develop beyond the cotyledon stage. The leaves of affected plants are characteristically roughened owing to the fact that they contain a mixture of 2n and 4n cells. From these "mixochimeras" there ultimately may grow out branches with smooth leaves which either are normal 2n or contain twice the normal number of chromosomes and are therefore 4n. The 4n flowers may be recognized by the larger size of their pollen grains or by the more tedious method of actually counting the chromosomes in young buds which have been fixed and stained by the acetocarmine method.

In addition to the seeds, vegetative parts of the plant may be treated by a variety of methods. The most successful method consists in spraying growing points with solutions or better with emulsions containing colchicine. To our surprise, injecting solutions and allowing solutions to be soaked up through the cut parts of the stem were not successful.

1 Address: Cold Spring Harbor, Long Island, New York.

Species in Which Chromosome Doubling Has Been Induced

One of the early problems in the use of colchicine was the extent to which this drug would be effective in doubling the chromosome number of other forms than Datura. A number of species were selected for testing because of their adaptability to experimental cultivation or because of their relation to special problems involved in polyploidy. Following is a list of species of flowering plants in which at least flowers with doubled chromosome number have been secured through colchicine treatment.

Plants with Chromosomes Doubled by Colchicine Treatment

Those marked * have yielded An seed; those marked ** have given An offspring in second generation; unmarked plants were shown to have An tissue by flowers with 50 per cent or more 2n pollen grains.

Caryophyllace^e Lychnis dioica** Stellaria media** Vaccaria parviflora *

CHENOPODIACEiE

Spinacia oleracea ** Composite

Bidens leucantha **

Cosmos sulphureus **

Rudbeckia hirta Crucifer^e

Raphanus sativa *

CUCURBITACE^

Cucurbita pepo

Mammoth pumpkin

Pear gourd

Small round china gourd*

"Spoon" gourd*

White sphere squash *

Yellow disk squash*

Straight-neck yellow squash * Cucurbita maxima

Blue Hubbard squash*

Warren Essex squash *

Buttercup squash* Cucurbita moschata

Small yellow cushaw squash * Cucurbita maxima X C. moschata * Lagenaria vulgaris

Dipper gourd

Giant bottle gourd*

Knobkerrie gourd *

Hercules Club gourd *

EUPHORBIACE.E

Mercurialis annua Malvaceae Anoda lavateroides

Morace^e

Cannabis sativa** Humulus japonicus

OxALIDACE^E

Oxalis valdiviensis ** Plantaginace^ Plantago lanceolata

POLEMONIACE.E

Collomia coccinea * Gilia abrothanifolia

PORTULACACE^E

Portulaca grand iflora ** Portulaca marginata ** Portulaca oleracese ** Portulaca parana **

SOLANACE.E

Datura ceratocaula ** Datura discolor Datura ferox** Datura innoxia Datura leichhardtii ** Datura metel ** Datura meteloides Datura pruinosa Datura quercifolia * Datura stramonium** 6 main lines and gene types * 8 races with extra chromosomal material * Lycopersicum esculentum * Nicotiana sanderae ** Nicotiana tabacum X N. glutinosa ** Nicotiana glutinosa X N. sylvestris * Petunia axillaris** Violace^e

Viola tricolor, hortensis

The 65 different kinds of plants shown in the list above have all been induced to double their chromosome numbers by treatment with colchicine. They represent 14 different families, 24 genera, and 41 species. Among them are 3 species hybrids.

The species successfully treated represent a considerable number of genera and families, enough to indicate that the method is of wide application. In addition, a number of forms which showed the vegetative peculiarities characteristic of tissue with doubled chromosome number following colchicine treatment were discarded for various reasons before the special treatment required to force out 4n branches had been developed. This was notably true of the grass family, but other investigators who have specialized on this family have recently been successful in securing 4n races of grasses through treatment with colchicine. The fungi seem highly resistant to the toxic action of colchicine. Miss Satina has treated a representative group of fungi with saturated solutions of colchicine (±4.0 per cent) with no obvious effect. She has succeeded, however, in doubling the chromosome number of Marchantia polymorpha by treating the gemmae with colchicine and has thus secured 2n male and female thalli of this liverwort.

Polyploidy in Datura

Most intensive study has been made of the responses of the 10 herbaceous species of Datura to colchicine treatment. All appear to have their dividing chromosomes affected in such a way that cells with doubled and higher number of chromosomes result. D. ceratocaula is especially sensitive and is readily induced to double its chromosome number when treated by spraying with solutions of colchicine, a method not very effective with other Daturas. Certain races of D. metel appear to be somewhat resistant. Treatment of 4n seeds of D. stramonium has given rise to branches with 8n flowers, as has also severe treatment of 2n seeds. Such octoploid flowers have so far failed to produce seed. Hexaploid (6n) flowers have been secured by spraying growing points of 3n plants. A few large seeds from such 6n flowers have been formed but they have not yet been brought to germination.

Haploid (In) plants appear not infrequently in our cultures through parthenogenesis of the reduced egg cells. Though some haploids produce very small capsules and an occasional seed, others are entirely sterile and cannot be induced to set any fruits. By spraying branches of a haploid with colchicine solutions it has been found possible to induce In cells in the buds to double their chromosomes and thus to become 2n. An abundant production of large capsules with 2n seeds results. Methods are being investigated which it is hoped may be effective in inducing In offspring from 2n parents. If they prove successful, simple means would be available of producing homozygous races from highly heterozygous parents (such as species hybrids) through the induction of In seedlings which may be forced to produce normal 2n capsules through chromosome doubling.

Dr. Bergner has been making a study of the chromosomal condition in abnormal branches on plants from seeds treated with colchicine. Chromosomal deficiencies due apparently to elimination of lagging chromosomes during the process of doubling appear to be common. Branches have been identified which lacked from one to six chromosomes but were otherwise 4n. There is suggestion that periclinal chimeras may be present. These findings show that the immediate result of colchicine treatment cannot be depended upon to be balanced 4w tissue with four of each kind of chromosomes present. They furnish added justification for our rule that 4n plants for critical study should only be used from the second generation and not then until cytological study has demonstrated that each of the n sets has four chromosomes. Before another report we expect to have such pure 4n plants for all the 10 herbaceous Daturas as well as Zn individuals and these will be kept in cultivation by grafts and cuttings. Already we have gotten 4n flowers and seeds from 14 different lines of Datura stramonium which we wished to use as tetraploids for special purposes. Among these are a half-dozen pure breeders with extra chromosomal material which we wished to use in cooperative studies with Dr. Sinnott on anatomical effects of polyploidy.

Double Diploidy

A well-established method of evolution both of cultivated plants and of plant species in nature is through the spontaneous doubling of the chromosome number of sterile hybrids between species. Such species hybrids are sterile because the chromosomes of the one parental species are too distantly related to those of the other species to allow the chromosomes to pair, and pairing is necessary for sexual reproduction. When the chromosomes in the hybrid are doubled in number the chromosomes of each species will find duplicates with which to pair and the hybrid will become fertile. It seemed desirable to see if such formation of new species could be induced by laboratory methods. Through the kindness of Dr. F. 0. Holmes material was secured of the species hybrid Nicotiana tabacum X N. glutinosa. This is completely sterile. By treating the hybrid plants with colchicine, fertile flowers have been induced which have produced an abundance of double diploid seed, and by the germination of the latter a double diploid race or new species has been established.

Effects of Polyploidy

Now that a ready method of doubling chromosomes is available, it is possible to learn the effects of the doubled number upon the morphology and physiology of the plants affected. Conclusions regarding the effects of polyploidy have heretofore been too often drawn from polyploids found in nature in which the effects of polyploidy alone could not be easily separated from those due to genes which differed in the two members of the polyploid series compared. In consequence conclusions regarding polyploids have often been conflicting.

Since doubling chromosome number appears to have been a method of evolution in nature it is of importance to learn the effect of this process upon the sex mechanism in plants in which the sexes are separate. Evidence from related species in nature has been interpreted as signifying that doubling the chromosome number of a dioecious species gives rise to a hermaphroditic form. Other investigators believe that the chromosomal mechanism would necessarily work against the establishing of either a hermaphroditic or a purely dioecious race by chromosome doubling. A study is being made of a series of dioecious species some of which are listed above as having had their chromosome number doubled. For such forms we should soon be able to learn the effect of polyploidy on the sex mechanism.

Another problem under investigation is the effect of polyploidy upon self-sterility, which is common among plants. A number of self-sterile species are being tested. In one case at least, preliminary tests seem to indicate that doubling the chromosome number neither increases nor decreases the degree of self-sterility.

In 1934 there were published by the Carnegie Institution the results of a cooperative study with Dr. Sinnott on the anatomy of extra chromosomal types in Datura stramonium. Differences in number and size of cells and in tissue pattern were demonstrated for 2n+l types and also for the balanced polyploid series, In, 2n, 3n, and 4n. It is still unclear how the qualitative differences are brought about between the members of the polyploid series. It is proposed to investigate this problem further than was possible in the preliminary study and to include a larger range of anatomical characters in all the herbaceous species of Datura. The possible antagonism between the influence of doubled chromosome number and the presence of unbalanced extra chromosomal material for which races are being developed by colchicine treatment will be investigated. It is thought that the qualitative effects of polyploids may be due to a differential effect of doubling upon different chromosomes. The effect of polyploidy upon size, shape, and growth rates can be better studied in some ways in fruits of the Cucurbitaceae, a group in which Sinnott has carried on investigations for many years and in which highly inbred lines are available. As shown in the list above, the chromosome number has been doubled in a considerable number of Sinnott's squashes and gourds, so that tetraploids of these races should be ready for detailed study by next season.

Attempts to Double Chromosome Number in Animals by Colchicine

Treatment

In cooperation with Dr. G. C. Embody and with the assistance of A. M. Phillips, attempts were made by the use of colchicine to induce tetraploidy in trout by treatment of eggs with the hope that the fusion nucleus could be induced to double its chromosomes before the first division. In cooperation with Dr. G. K. Noble and with the assistance of J. A. Mathewson, eggs also of frogs (Rana pipiens and R. sylvatica) and eggs of a number of aquarium fishes were treated with colchicine. In all these experiments difficulties were encountered in getting the drug to penetrate the membranes and in consequence the attempts to induce tetraploids in these animals were unsuccessful. It is possible that eggs which develop outside the body are better protected against unfavorable environmental influences than those which develop within the body. In consequence the latter might be more promising material in which to attempt chromosome doubling despite the technical difficulties involved in handling eggs of such species.

40 CARNEGIE INSTITUTION OF WASHINGTON

Tests with Other Chemicals

By the use of Portulaca seedlings which show swellings of the stem due to tetraploidy when the seeds are treated with colchicine, a series of experiments were started in cooperation with Dr. Roger Adams to determine if possible what specific groupings of the colchicine molecule are necessary for the action of this chemical in doubling chromosome number. The following results were obtained with chemicals supplied by Dr. Adams:

Colchiceine — no effect

Trimethylcolchicinic acid hydrochloride — no effect

N-Acetyliodocolchinol — no effect

N-Acetylcolchinol — no effect

Benzoyl colchiceine — no effect

Colchicine salicylate — same reaction as colchicine

It is concluded that probably any modification of the colchicine molecule, even minor in character, causes the disappearance of its activity in doubling the number of chromosomes.

An attempt has been made to find other chemicals which would double chromosome number. A few have been found to react positively to the Portulaca test and to induce doubling in some of the cells, but none are as effective as colchicine. Sodium cacodylate, for example, induces a positive reaction with Portulaca and has induced in flowers in this species, but it is ineffective with Datura.

Work is under way on other problems which have to do with control of genetic behavior and with a better understanding of the nature and effects of polyploidy, especially as it relates to evolution in nature, but an account of these experiments is deferred to next year's report.

THE GENE M. Demerec, B. P. Kaufmann, and Margaret E. Hoover

The Nature of X-Ray Induced Changes

During the past year a cooperative study by H. Bauer, M. Demerec, and B. P. Kaufmann on chromosomal alterations induced by X-rays in Drosophila melanog aster was completed and the data were analyzed. The work consisted of an analysis of salivary gland chromosomes of first generation larvae obtained by mating untreated females with X-rayed males. The chromosome rearrangements which were studied in this experiment were those induced in the treated sperm and not eliminated during the embryonic and early larval development, since glands were selected for cytological study from larvae preparing to pupate. Experiments to be described later show that the death rate in these early stages of the life cycle is high. It is probable that non-viable chromosomal combinations are responsible to a large extent for that high death rate. In cytological analysis of salivary gland chromosomes we are dealing, therefore, with the surviving members of a much larger population. Permanent preparations of 1765 pairs of salivary glands were used and a total of 1038 chromosome breaks were analyzed. The X-ray radiation was applied by a Universal Type Coolidge tube, with a tungsten target, at 85 kilovolts and 7 milliamperes, the dosage being measured by a Fricke-Glasser dosimeter.

The relation between the dosage and the frequency of breaks is shown in the accompanying table.

Dosage

Sperm

Breaks per

r-units

Total

Number with alterations

Per cent with alterations

Total sperm (per cent)

Changed sperm (per cent)

1000

2000

3000

4000

5000

331

277 725 215 217

12

24

193

64

87

3 . 63 + 1 . 03

8.66 + 1.69

26 . 62 + 1 . 64

29.8 +3.12

40.09+3.33

8.16 + 0.02

22.02+0.05

72.00 + 0.05

85.58 + 0.11

125.35±0.12

2.25 + 0.13 2.54 + 0.18 2.70 + 0.09 2.88 + 0.18 3.13 + 0.16

It is clear from these data that there is no simple relationship between dosage and percentage of breaks per total sperm (fig. 1) . The effect observed here is not directly proportional to the dosage, and this relationship is different from that observed in the case of lethal changes where the effect was found to be proportional to dosage. In this case the observed curve at low dosages (1000 and 2000 r-units) fits closely an exponential square curve, while at high dosages the deviation from it is appreciable. For its whole length the observed curve approximates an exponential 1.5 power curve which is intermediate between the first power of dosage curve and the square power curve. A similar relationship has recently been reported by Sax for chromosomal aberrations in Tradescantia and by Muller for genetically detected aberrations in Drosophila. Direct proportionality to dosage indicates that individual changes are induced by single ionizations, while a square power relationship would be expected, at least at dosage levels in which there is a low number of breaks per nucleus, if an individual change, in this case a chromosomal rearrangement, were caused by two independent ionizations.

Two types of regions can be distinguished in salivary gland chromosomes as well as in mitotic chromosomes of D. melanog aster, namely, euchromatic regions and heterochromatic regions. Heterochromatic segments are located in each chromosome near the spindle fiber attachment point and in salivary gland nuclei they come together into what is known as the chromocenter. In proportion to the total length, heterochromatic regions are appreciably longer in metaphase chromosomes than in salivary gland chromosomes. The study of 1038 breaks showed that they are distributed at random throughout the euchromatic regions, with the possible exception of distal sections, where breaks tend to be slightly more frequent. On the other hand, in heterochromatic regions of salivary chromosomes breaks were found to be much more frequent than in euchromatic regions of similar lengths. However, the frequency of breaks is found to be approximately proportional to the length of heterochromatic region as represented in metaphase chromosomes. This finding was discussed in last year's report (Year Book No. 36) and used as evidence to support the hypothesis that breaks are distributed at random per unit length of chromonema, that the chromonema is structurally similar in all chromosomes, and that it may be looked upon as composed of fundamental units or backbones corresponding to the fiber protein molecule pictured by W. T. Astbury. The study of breaks showed also that they are distributed at random among chromosomes of comparable lengths. The analysis of two break cases indicates that these breaks are independent of each other, suggesting that a different mechanism is re

Fig. 1. Percentage of induced chromosome breaks in the total sperm. Curves show observed

and theoretical values at different dosages.

sponsible for reattachments producing induced inversions and translocations than for those connected with crossing-over, where it is known that the position of the second break is influenced by the position of the first break.

The origin of induced chromosomal rearrangements may be accounted for by either of two mechanisms, namely, that a break in the chromosome occurs first and reattachment is accomplished later (Stadler) , or that breakage and reunion are part of one process and occur simultaneously (Serebrovsky) . The latter hypothesis assumes all regions involved in rearrangements to be in contact at the time the reunions occur. In order to account for the rearrangements observed in our experiment it would be necessary to assume that in almost 20 per cent of cases more than two chromosome threads were in contact at one point, and to explain some of the complex rearrangements it would be necessary to assume that as many as six or eight threads were in contact at one point. The improbability of this condition combined with our analysis of the relationship between dosage and the percentage of breaks suggests strongly that X-ray induced chromosomal rearrangements result from such a breakage and reattachment mechanism as Stadler has proposed.

The cytogenetic analysis of the white-facet region of D. melanogaster was continued this year by Demerec and Hoover. A contribution to this study was previously made by Helen Slizynska (Year Book No. 36). With an X-ray treatment of approximately 2500 r-units, 81 separate changes have been induced so far in this region of the X-chromosome. Of these, 54 have been completely analyzed both cytologically and genetically at the time this report is made. These changes may be placed in three groupings, according to their genetic effect, namely, (1) changes that are viable, (2) changes that are lethal in the homozygous or hemizygous condition, and (3) changes resulting in a mottled phenotype.

The viable changes, owing to the method of detection, are all connected with the white locus. Out of a total of 25 such mutant whites, none was found to be connected with a major chromosomal aberration such as a translocation or inversion. Eight of these were studied for the presence of all critical bands and no deficiencies were detected in any. This finding indicates that at least the majority of viable changes are free from chromosomal aberrations and are not associated with minute deficiencies.

In the second group 28 lethal changes have been analyzed; 7 of these were found to be connected with chromosome changes and 21 were free from any rearrangement. This proportion suggests that a majority of lethal changes induced by this treatment are not associated with chromosomal aberrations and, consequently, must originate independently of the chromosome breaks resulting in inversions and translocations. These 21 lethals were subjected to careful salivary gland chromosome analysis. In 2 cases no known bands were absent, but 19 were found to be cytologically detectable deficiencies. Five of the 7 lethals associated with chromosomal rearrangements have been studied so far ; 3 are not deficiencies and 2 are deficiencies.

The changes producing a mottled phenotype form a separate class. The 23 such changes collected so far have all been associated with chromosomal rearrangements. Of these, 13 have been studied for deficiencies and in only one case was such a deficiency detectable.

These results suggest that the majority of viable changes in the white locus are free from association with chromosomal aberrations and are also probably not gene deficiencies; that a high proportion of lethal changes are also free from chromosomal aberrations, but the great majority of these are cytologically detectable deficiencies. On the other hand, of those lethals connected with rearrangements, the minority may be deficiencies. The difference in this proportion may possibly be accounted for by position effect. All mottleds so far obtained are associated with chromosomal changes and a great proportion of these are not deficiencies. All the chromosomal aberrations studied in relation to mottling have involved chromocentral regions. The chromocenters of all the chromosomes seem effective in somehow producing these mosaics. Three cases where rearrangements involving chromocenters have been detected have not shown this spotting, suggesting either that all portions of the chromocentral regions are not equally effective, or that some sort of differential position effect is involved.

It is interesting to note that the size of the deficiencies in this collection cover a fairly large range from single up to 46 bands. In the size range from 12 to 46 bands, the distribution seems at random, whereas there is a piling up of deficiencies of 1 to 4 bands. Out of 23 deficiencies analyzed, 13 were small, 5 of these being single bands. The other 10 were scattered over the larger size range. The study on distribution of breaks in chromosomes reported above shows that chromosome breaks occur at random and that they are independent of each other. Therefore, if deficiencies are caused by two independent breaks and loss of a piece between them, it would be expected that various sizes would occur with equal frequency. The data collected so far indicate that this holds true for deficiencies involving 12 or more bands but that it does not hold true for small deficiencies involving 1 to 4 bands. This suggests that these two classes of deficiencies were differently induced, large deficiencies by two independent changes or ionizations and small deficiencies by a single ionization.

Response of Hereditary Material to X-Ray Treatment

In last year's report (Year Book No. 36) data were presented showing that lethal changes induced in males by simultaneous treatments with X-rays were consistently almost twice as frequent in flies of the Swedish-b stock as in flies of the Oregon-R stock. It has been concluded that a biological factor was responsible for this difference in the effect of X-rays on the hereditary material. To investigate this problem further, another set of experiments was conducted in which the frequency of dominant lethals was studied. Males of Swedish-b and Oregon-R stocks were simultaneously treated with 1000, 2000, 3000, 4000, and 5000 r-units. They were mated to untreated females and the frequency of dominant lethals was found by determining the death rate during ontogeny. It has been observed that the frequency of dominant lethals is higher in the treated Oregon-R series than in the simultaneously treated Swedish-b series. In the case of dominant lethals, therefore, the behavior of these two stocks is just opposite to the behavior observed for sex linked lethals. This can be accounted for by either of the following hypotheses: (1) Dominant and recessive lethals are basically similar changes differing mainly in the degree of effect. The Oregon-R stock, when compared with the Swedish-b stock, is physiologically more sensitive to changes induced by X-ray treatment, and because of that sensitivity a greater proportion of individuals carrying these changes die early in ontogeny, the stock, therefore, showing a higher rate of dominant lethals and a lower rate of recessive lethals. (2) Dominant lethals and recessive lethals are, as far as the mechanism of origin is concerned, two unrelated types of changes ; dominant lethals, for example, may be nonviable chromosomal aberrations while recessive lethals may be mutant types. Then the difference in the behavior of the two stocks may be accounted for by the presence of some biological factor which makes one stock more sensitive to one type of changes and the other stock more sensitive to the other type. In that case, two factors may be responsible for the increased sensitivity. It may be that, owing to a specific genetic constitution, chromosomes are more susceptible to breakages and genes to changes. A parallel case is on record regarding spontaneous mutations which were increased by the presence of a certain gene (Year Book No. 35). Or it may well be that this sensitivity is nothing else but an expression of a physiological condition determining the survival rate of various changes. In that case, the higher sensitivity of the Swedish-b stock to lethal changes would mean that owing to a physiological state certain changes which would be semi-lethal in Oregon-R stock are completely lethal in Swedish-b.

In the Oregon-R stock a 1000 r-unit treatment induced about 22 per cent of dominant lethals, a 3000 r treatment induced 63 per cent, and a 5000 r treatment, 78 per cent. This shows that a very large proportion of induced changes are being eliminated as dominant lethals. Since many of these are undoubtedly connected with chromosomal aberrations, this evidence indicates that probably a large proportion of these aberrations were eliminated before the stage at which the material was taken for the salivary gland chromosome study described earlier in this report.

Cytological Analysis of Induced Chromosomal Changes

An analysis of the numerous salivary glands of male and female larval descendants of irradiated fathers used in the study of break frequency and distribution has permitted diagnosis of various types of chromosomal alterations. These include deficiencies, duplications, inversions, inter- and intrachromosomal translocations, and complex rearrangements involving two or more chromosomes. As many as 13 breaks per treated sperm have been detected. In one rearrangement involving 10 breaks, 2 bound an inverted section ; the other 8 are included in a mutual exchange. About 1 per cent of the rearrangements contained duplicated sections, indicating that the chromosomes of the sperm were longitudinally double at the time of recombination of broken sections. One of these duplications, described by Kaufmann and Bate, is intercalary in the left limb of the third chromosome, and is arranged in the pattern of a "reversed repeat," as may be indicated by the sequence ahcdgfeefghijk. Origin of the duplicated section is attributed to fusion of sister chromatids at the same level, thus: abcd|efg|hijk giving ahcdgfeefghijk

abcd|efg|hijk and probably another strand of the constitution abcdhijk, whose fate remains conjectural. Union of sister chromatids in the manner here indicated offers a possible explanation of the origin in nature of such reversed repeats as occur in the chromosomes of Drosophila. In another duplication Kaufmann found that both chromatids of the right limb of the second chromosome had been broken at one level but that a third break occurred at different levels in each of the two strands. Although it cannot be determined with certainty from this case that the second chromosome was split at the time of irradiation, the rearrangement suggests the possibility that two strands may be broken at the same level by the secondary effects of a single ionization.

A considerable number of alterations showed transfer of the nucleolus from its normal chromocentral position to euchromatic regions of various chromosomes. From the pairing configurations in the salivary glands of individuals heterozygous for such exchanges it could be determined that the nucleolusforming regions exist in the heterochromatic regions of the X- and Y-chromosomes, as had previously been described for mitotic chromosomes. In the X-chromosome the nucleolus is found between the 20B1-2 and 20C1-2 bands.

A cytogenetic study of nine inversions in the X-chromosome was completed by Margaret E. Hoover. These inversions, differing in size and position, were studied with particular attention to their cytological limits as determined in the salivary gland chromosomes. The inversions selected included four already familiar to Drosophila workers, namely, C1B, AM, dl-49 and Dichsete. In addition, an X-ray induced transposition (303-1) was used, three induced inversions associated with the cut locus, and the tandem inversion (268-13) mentioned in last year's report. Crossing-over data showed, as expected, that single crossovers are eliminated within inverted regions and greatly reduced in adjacent sections. Data on the amount of synapsis undergone in the salivary gland chromosomes carrying inversions indicate that even with these configurations a high degree of pairing is maintained. Even at the breakage points complete synapsis was observed very frequently. Eased on 100 random observations, the lowest value for total synapsis through the inversion point was 48 per cent, the highest, 82 per cent. Disregarding lack of pairing for less than 1 to 1 section at these inversion points, the general aspect of the whole chromosome showed complete synapsis in 67 to 86 per cent of cases as compared with 90 per cent for controls. The relatively high frequency of synapsis in chromosomes carrying inversions is thus indicated, suggesting that synapsis occurs in the early stages of salivary gland development, when the thinner chromosomes would encounter less difficulties than the mature chromosomes in forming the inversion loops. The breakage points of the inversions were determined cytologically by salivary gland chromosome analysis. In all, 20 breakage points were studied. At 5 of these breakages, deficiencies were detected, at 13 no deficiencies were visible, and a single band at each of the other 2 remains unanalyzed. Those inversions in which the deficiencies were detected are also associated with lethals, but 303-1, a semi-lethal, and Dichsete and C1B, both lethal when homozygous, appear to be free from deficiencies. This would indicate that genetic-physiological lethals and cytological deficiencies are not necessarily synonymous.

In order to determine the nature of chromosomal alterations in cell generations immediately following radiation, J. G. Carlson, a guest investigator, treated embryos of the grasshopper, Chortophaga viridifasciata, and studied the mitoses in neuroblast and ganglion cells. Chromosomes which were treated in stages between telophase and mid-prophase show, in succeeding phases of the same mitotic cycle, chromosome fragmentation and translocation, chromatid breakage and translocation, and what may be half-chromatid effects. The presence of chromatin bridges at anaphase appears to result from any of three different alterations: (1) chromosome translocation, (2) chromatid translocation, (3) fusion of sister chromatids of the proximal portions of fragmented chromosomes at their broken ends. The formation of such bridges and the persistence of broken ends from one generation to the next through breakage in late anaphase makes possible delayed attachments following irradiation.

The mitotic behavior of some fragments lacking spindle attachments parallels that of unaltered chromosomes in several respects. Such fragments lie in the equatorial plane at metaphase. Their "chromatids" begin to separate at anaphase at almost the same time as do those of the unaltered chromosomes, and they come into intimate contact with the spindle at middle anaphase. Sister "chromatid" fragments usually move toward opposite poles behind the other chromosomes, and so are included at telophase in different daughter cells. The initial separation of "chromatids" of fragments has the form of V's, rings, and pairs of rods. A hypothesis, based on the assumption that broken ends of sister chromatids tend to fuse inter se, is suggested to account for this. Not infrequently fragments are included at telophase in the newly formed cell nucleus. This behavior has a bearing on certain hypotheses of the mechanism of mitosis and on the question of delayed attachments.

EXPERIMENTAL LEUKEMIA

E. C. MacDowell, J. S. Potter. M. N. Richter, J. Victor, M. Bovarnick, M. J. Taylor, E. N. Ward, T. Laanes, and M. P. Wintersteiner

The cooperation of the College of Physicians and Surgeons of Columbia University with the Department of Genetics, which has made possible the establishment and continued progress of the leukemia project, has been broadened this year by the active participation of the Department of Biochemistry as well as the Department of Pathology.

Students of cancer have found that procedures that induce resistance to transplanted tumors of various types have failed to influence spontaneous tumors and they have largely abandoned the search for methods of inducing resistance to spontaneous tumors as a result of the current belief that the unique relationship between an animal and its own tumor cells makes the induction of resistance to these cells impossible. But this belief has been contradicted in the case of mouse leukemia (Year Book No. 35, p. 52) by the demonstration that parallel results are due not to the unique relationship between a mouse and its own leukemic cells but to immunological differences between leukemic cells in spontaneous cases and those carried through long series of transfers from mouse to mouse. Different cells require different conditions for survival or suppression.

From this point one procedure would be to search directly by the method of trial and error for conditions that would suppress spontaneous leukemic cells; another procedure would be to determine (1) the nature of the mechanisms of the successfully induced resistance to the long-transplanted cells, and (2) the basis of the change in leukemic cells during the course of transfer. With this knowledge it would be possible to proceed intelligently to build up conditions to suppress spontaneous leukemic cells. The first or direct approach has been followed unfruitfully by numberless workers; the second or indirect approach represents direct contribution to the major aim of the Institution workers of this group, namely, an understanding of normal life processes as revealed by comparison with their abnormal functioning. Indeed the nature of the evolution of leukemic cells of the spontaneous type into the types found in the various transfer lines stands out as a primary problem both for its own broad significance and as a lead to the nature of the antecedent change of normal cells into spontaneous leukemic cells.

Passive Transfer of Immunity

Certain transplantable tumors are said to induce resistance to themselves. Similarly for several years we have been consistently producing resistance to lethal doses of transplanted leukemic cells by means of sublethal doses of the same cells (Year Book No. 34, p. 45). This year the generality of this phenomenon has been broadened by experiments with still another line of leukemic cells. Such immunity seems to parallel that obtained by antigens and antibodies of conventional immunology, but the very general failure to obtain passive transfer of any tumor-induced immunity has made the identification of antibodies impossible, and leaves a question as to the nature of this induced resistance. During the year actively induced resistance to leukemic cells has been passively transferred to susceptible mice with success in 246 cases in 18 independent experiments with controls, in every experiment, on the lethality of the dose of leukemic cells in the absence of treatment. As in the studies on tumors, serum was ineffective; but immediate protection was afforded by the use of saline suspensions of minced tissue from actively immunized mice. The immediacy of the protection was emphasized by its effectiveness even when the immune tissue was given one to three hours after the lethal dose of leukemic cells. This is probably the first time that any type of induced resistance has completely suppressed a neoplastic growth transplanted before the protecting treatment was given. These results do not prove that an antibody, as usually understood, is concerned, since living cells are present. The protection, however, is not due to the living cells as such, for genetically identical living cells from non-immunized mice give no protection. Passive immunization is usually demonstrated by cell-free extracts. In this case the demonstration depends upon the elimination of any form of active resistance by the immediacy of the protection.

The passive transfer of immunity provides a short cut in the technique of active immunization, as well as a new test for the development of active immunity. The process of actively immunizing a large number of mice has required many weeks and the initial steps have given uncertain results on account of the difficulty of controlling dilutions as great as 10 ~~ 6 of the standard dose. The actual number of cells from the same suspension given to different mice may vary to the extent of killing some while others may not be given enough cells to induce any active resistance, and therefore die from the next higher dose. In such critical thresholds slight variations in the condition of the leukemic cells or of the hosts may have deciding influence. All these difficulties are avoided by starting with one treatment with im

DIVISION OF ANIMAL BIOLOGY 49

mune tissue. The usual lethal test dose (1/64 of the standard dose) is at once resisted by all mice and rapidly repeated massive doses build up the strength of the active immunity.

The process of building up active immunity has been studied by successive tests for passive transfer. After each step, tissue was transferred to still other hosts, which were then given the lethal test dose of leukemic cells. Tissue from a mouse treated only with immune tissue gives no protection; after the first leukemic dose, the tissue will protect some of the test mice; after two leukemic doses, the tissues will protect most of the test mice; while after three doses, the tissue will protect practically all the test mice. The production at a given date of any given number of mice actively immunized by leukemic cells is no longer a difficulty.

A Chemical Approach

A possible lead for a chemical analysis and a clue to the difficulty of obtaining non-cellular substances that resist neoplastic growth have been found in the amazing ease with which the effectiveness of this supposed substance can be destroyed mechanically. The method is simply to force the saline suspension of minced tissue out of a syringe held firmly against the bottom of the vessel. This process almost immediately reduces all discrete bits of the tissue (liver) and in a very few minutes all cells are torn to pieces. Is the protecting substance denatured by the same pressures that destroy the cells, or is it so closely associated with cell structure (possibly as a modification of some normal cell constituent) that cell destruction is responsible for the inactivation? When injected into a mouse the cells of minced immune tissue are destroyed, but the effectiveness of the protection that accompanies them may persist for at least two weeks.

The importance of a chemical approach to questions raised by this project has been repeatedly emphasized. The recent advances in organic chemistry have disclosed such elaborate organizations of protein molecules as to approach the assumed complex "organization" of living matter. The chemist can no longer hold aloof from phenomena depending upon "organized" matter. A specific substance may be associated with living matter in the form of independent molecules operating as units ; again, a special configuration may have no special effectiveness until it becomes established as an organic part of a normal molecule ; or again, still more intimately, a special function may arise from the reorganization of molecular structures already present. Such considerations bear on several problems: on the nature of the change from normal to malignant; on the chemical differences between strains of mice genetically susceptible or resistant to certain transplanted leukemic cells; on the materials responsible for induced resistance, whether occurring naturally or put there by the host in combating growth of leukemic cells.

A significant broadening of the scope of the project has been made possible by the addition to this group of Dr. M. Bovarnick, working in the Department of Bio-chemistry under Dr. Hans Clark at the College of Physicians and Surgeons. During the year, Dr. Bovarnick has been approaching the problem of the mechanism of the resistance to transplanted leukemia that is induced by normal tissue. Normal growth-controlling materials are eliminated because the effect is specific for a given genetic differential between the constitution of the host and of the normal tissue used in the treatment. If antigenic action is concerned, it is unprecedentedly rapid (2 days) and calls for the startling assumption of antigenic similarity between a substance that appears only in highly virulent leukemic cells from one strain of mice and a substance found in all tissues of a certain other strain without leukemia. That some new protective principle should be involved seems entirely possible.

With the three biological variables — susceptible host, leukemic cells, and protecting normal tissue — controlled to the point of giving virtually 100 per cent results, a chemical analysis appears possible. The first step, however, has proved very difficult. Conventional chemical analysis requires separation from the vital organization of the cell; but all familiar procedures known to eliminate living cells have rendered the residues of the normal tissue ineffective. The pressure technique reported above has not been tested on normal tissue.

Previously the normal tissue used was embryonic, and this type of protection has been designated embryo treatment, but this year, adult tissues of the same strain have been found to be equally effective and, being far more available, have been used regularly. By using the same organ (liver) as for the passive transfer of immunity induced by leukemic cells, close comparisons can be made between liver tissue that protects by virtue of its genetic constitution, and liver tissue that protects by virtue of a non-genetic change induced by leukemic cells. In spite of clear differences in the resistance developed by these two different livers, the existence of the essential substance in each case is closely bound up with living cells. Another substance is also intimately bound up with living cells — that responsible for the malignant behavior of leukemic cells. To determine the relationship to living cells in any one of these three cases would serve as an important lead in the study of the others.

Specificity of Inducing Stimulus

Since the interpretation of phenomena occurring before our eyes has been a leading policy, interest has been focused on the mechanism by which the leukemic process, once initiated, is maintained, and upon the extrinsic factors that are in fact active in the production of leukemia in our colony, rather than upon a search for chemicals that might experimentally induce this malignant growth. However, the study of the genetic factors involved in the observed incidence of spontaneous leukemia would be greatly aided by a chemical treatment that would reveal genetic potentialities at an early age. With this purpose, a study of the effect of a highly potent carcinogenic agent, benzpyrene, was undertaken in cooperation with Drs. C. J. Lynch and A. Claude of the Rockefeller Institute for Medical Research. In a strain of mice inheriting a high susceptibility to leukemia would this chemical hasten the appearance of leukemia? The evidence gives no indication of any such hastening in the appearance of leukemia, but benzpyrene does stimulate the occurrence of sarcoma as it does in other strains of mice. While other investigators have reported leukemia as the main result of another chemical treatment, methylcholanthrene, the strain of animals employed did not otherwise show leukemia. The genetic constitution does not merely render certain tissue liable to become malignant, but determines susceptibility of certain tissues to certain agents. A genetic constitution that renders primitive reticular cells susceptible to certain stimuli occurring in our colony leads to leukemia; these cells are not susceptible to the action of benzpyrene; another genetic constitution might make these cells susceptible to the carcinogenic action of benzpyrene, and not to the stimuli responsible for the spontaneous leukemia in our strain C58.

The developmental hypothesis of malignancy reported last year by Dr. Potter has continued productively to guide thought and has been applied to a new series of phenomena. According to this hypothesis malignancy modifies the rate of normal cellular differentiation. This stands in contrast to the view that malignancy depends upon the creation of a new, uncontrollable cell type that reproduces only itself.

Variation in malignancy, according to the new interpretation, changes the rate of cell differentiation so that one or another stage preponderates, but, according to observation, all stages are present and each stage apparently divides at its own normal and characteristic rate. This includes advanced stages no longer able to divide, which accordingly can no longer contribute to progressive growth and so are no longer malignant. The continuation of a leukemic population depends on the presence of incompletely differentiated cells.

Lymphoid vs. Myeloid Leukemia

This hypothesis may now be broadened to cover the direction as well as the rate of development. Under the usual conditions of transfer from mouse to mouse the malignant factor in a given line has led differentiation consistently in the lymphoid direction for many years and through hundreds of normal hosts. But in hosts made incompletely resistant by normal tissue treatment, the direction of development may change and populations appear of leukemic cells in early stages of the myeloid series which terminates in polymorphonuclear leukocytes. Such myeloid populations when transplanted into untreated hosts may immediately develop into lymphoid leukemia with all the specific characteristics shown before. The hypothesis does not imply that early myeloid types have changed their course of development and become lymphoid, but that malignant control has been temporarily modified and, on return to a normal host, again directs the undifferentiated dividing cells in the lymphoid direction, while the myeloid cells are soon left behind by the new growth. In other cases more lasting change in the malignant control is shown by the continued production of myeloid cells in the following transfer or transfers. The outstanding instance of lasting modification showed a gradual return to lymphoid leukemia in the course of numerous transfers. In the tissue lesions of the hosts myeloid leukemic cells were found in the first four transfers and not thereafter; in the peripheral blood they persisted for thirty-four transfers.

Quite apart from the hypothesis regarding malignancy, the facts bring myeloid and lymphoid leukemia very close together ; an unbroken continuum of malignancy yields first one then the other type with return, abruptly or gradually, to the first type. This gives direct support to the theory of a common ancestral cell in normal animals, able to differentiate in either direction according to the needs of the individual.

In man, lymphatic, myeloid, and "mixed" types of leukemia are known, and even within the same case history changes in type have been recorded. But the genetic relationship between the various types has remained unsolved by clinical evidence. If the continuum responsible for malignancy controls both the rate of development and its direction, this continuum is subject to modification by the conditions of its environment. This environment in turn is subject to experimental and clinical control.

Malignant Lymphocytes in Tissue Culture

Through the cooperation of W. H. and M. R. Lewis and the Department of Embryology, living leukemic cells in tissue culture have been studied. Cultures were made of three lines of transplantable leukemia and observations were made microscopically and with the aid of a motion picture camera. At the conclusion of the observations on the living cells the cultures were fixed and stained for further study and reference.

These studies on living cells confirm the conclusion, derived from earlier studies with preserved material, that the malignant lymphocyte has the same morphological characteristics as a normal lymphocyte of the same stage of differentiation, and add the further observation that, according to the Lewis criteria, the method of locomotion of the leukemic lymphocytes studied is the same as that of normal cells.

Differences in transmission lines were evident during the early periods of the cultures ; the most virulent line showed the largest number of blastic cells moving out from the explant. The high mortality of lymphocytes after the first day made further observations of little value. Roller tube cultures kept for 7 and 9 days had very few demonstrable lymphocytes although polymorphonuclear leukocytes were plentiful among the active fibroblastic growth. The malignant nature of leukemic lymphocytes makes their culture in artifical media no less difficult than that of normal lymphatic cells.

ENDOCRINE STUDIES

0. Riddle, R. W. Bates, J. P. Schooley, E. L. Lahr, G. C. Smith, R. A. Miller, and L. H. Elwell

The regulatory powers of the anterior pituitary gland are now actively studied in many laboratories throughout the world, and in this field of study viewpoints may change markedly within a year. It has become clear that the pituitary gland largely provides for the regulation, adjustment, and coordination of the endocrine system, and for certain other activities of body and mind besides. Fitting this generalization into the broader and basic problem of organismal control and regulation, it would now seem possible to conclude that in higher animals and man the brain and the pituitary gland are the two prime sources from which the abilities of an individual are derived.

The investigations reported here, like those of past years, present parts of an effort to learn those endocrine agencies and mechanisms which condition or control functions such as growth (development), reproduction, and regulation in the bodies of higher animals and man. Much of the experience of the year points to the unpredictability of relationships among hormones ; and perhaps the most notable thing observed is the extent to which one hormone may either increase (synergize) or decrease the specific action of another. These studies of Dr. Riddle and his associates have been greatly aided by a grant from the Carnegie Corporation of New York to the Carnegie Institution of Washington.

The Preparation and Assay of Pituitary Hormones

We have now confirmed the report of others that the pituitary product, or products (adrenotropin), which increases adrenal size in rats is not prolactin. Our own preparations of adrenotropin, though containing only insignificant amounts of other hormones, are apparently crude fractions. During the past year all pituitary fractions obtained in connection with our preparation of pituitary hormones have been assayed not only on young pigeons but also on young rats. As a result of these routine assays on rats Bates and Riddle have obtained further evidence that most of our prolactin preparations are essentially free from adrenotropin, folliclestimulating (FSH), and luteinizing (LH) hormones.

Of notable value to the general task of purifying anterior pituitary hormones is the demonstration of the usefulness of copper hydroxide for freeing one or more pituitary hormones from troublesome amounts of prolactin. After demonstration that Cu(OH) 2 precipitates prolactin quantitatively from aqueous media at pH 6-7, this procedure was used to remove residual prolactin from two FSH fractions, one of which was derived from cattle pituitaries and the other from pig pituitaries. In neither of the resulting final preparations was there enough prolactin present in 1.0 mg. of product to provide a response with the local crop-sac test, showing that there remained in these FSH preparations less than 1 part of prolactin to 10,000 parts of the purified substance. Early results further indicate that a separation of thyrotropin from FSH may be accomplished by this same method.

Last year Dr. White of Yale University reported the preparation of prolactin in crystalline form, the first of the anterior pituitary hormones to be obtained in this state. Twenty milligrams of this crystalline prolactin, kindly supplied by Dr. White for our examination, was found to contain about 6.5 Riddle-Bates units per milligram. Some of our own more recent (noncrystalline) prolactin preparations have been found to contain as much as 10 such units per milligram.

Much work involved in the technique of assays has been performed by Mr. E. L. Lahr and Mr. L. H. Elwell. Mr. Graham Erdwurm has assisted in general laboratory work and has shared especially in the numerous assays made on rats.

54 carnegie institution of washington

Factors Affecting the Response to Prolactin

Since experience is proving that the quantitative response of the cropsacs of doves and pigeons is modified by a number of factors (see earlier reports), it is necessary to identify and measure all such factors. Folley and White recently reported that antecedent and simultaneous high dosage with the ovarian hormone, dihydroestrone, markedly reduces the extent to which prolactin normally increases the crop-sac weight. This finding has been confirmed by Riddle and Bates, who obtain this diminished response with purified prolactin, and also with prolactin admixed with all other anterior pituitary hormones ; but in the latter case the decrease is markedly less than in the former. Still other items of information concerning this restrictive factor were obtained which do not lend themselves to brief description here.

Folley and White also pointed out that their data, obtained with subcutaneous injection of ovarian hormone and prolactin, show the existence of a distinct sex difference (quantitative) of crop-sac response. Using intramuscular injections almost exclusively we had never observed a significant sex difference. Repeating the Folley and White experiments, and using the subcutaneous route of injection, the sex difference they describe has also been fully confirmed. In six different series (10 or more pigeons each) we found the male crop-sac weights indicated 1.5 to 4 times larger dosage than that indicated by the crop-sac weights of females.

Relations of Pituitary to Size of Body and Viscera

Observations reported last year concerning the effects of pituitary removal, and of the ability of various pituitary fractions (hormones) to sustain or promote general and localized growth, have been both confirmed and extended. Very little will be noted here concerning this line of study, which has been continued by Drs. Schooley, Riddle, and Bates.

When an unfractionated extract of the anterior pituitary gland is subjected to the action of trypsin it has been found to lose apparently all its power to repair the various disabilities that follow pituitary removal. This result indicates that all the pituitary hormones which participate in maintaining the weights of the body and particular organs studied are of protein nature. Further observations of growth phenomena in organs and in the body as a whole put emphasis on the high efficiency of a mixture of hormones. In pituitaryless pigeons the growth of the body as a whole, and growth in at least several of its parts (intestine, liver, pancreas, crop-glands, adrenal, and gonads) , is accelerated more by mixtures of hormones of the pituitary than by its purified hormones acting singly.

Hormonal Basis of Maternal Behavior and Broodiness

Three years ago, and on the basis of 160 tests, we reported that virgin and adolescent male rats can be made "maternal" in behavior by injections of prolactin, and that other pituitary hormones probably fail to induce this response. A total of 1780 tests have now been made, and the later developments of this study conducted by Riddle and Lahr will be briefly summarized here. The accompanying table lists the hormones or substances found to act positively in inducing maternal behavior in rats of four classes — females, castrate females, males, castrate males. Another list of hormones has been found either to have no similar action or actually to inhibit maternal behavior; of these latter only a word need be said.

"Maternal" behavior induced in young rats by hormones (The values given represent percentage of rats responding positively)

Sex

Control reactors

Prolactin

Progesterone

Testosterone

Intermedin

Whole A. P.

Phenol

Females

Spayed females. . . .

Males

Castrated males ....

(28) (32)

(23) (28)

77 79

56 75

70

78

64 61

60 83

15 53

58

50 25

20 43

40

59 56

24 57

It will be noted that intermedin, presumably produced in the intermediate lobe of the pituitary, induces this behavior; and extracts of the whole anterior pituitary (A. P.), containing prolactin and other hormones, are partially effective. One hormone secreted by the ovary (progesterone) and another by the testis (testosterone) are markedly effective, though apparently in lesser degree than is prolactin, while even carbolic acid seems effective in some cases. Though an adequate explanation of these results is not at hand there is some reason to suspect that progesterone and testosterone, as administered here, cause a diminished production of estrin, and perhaps also an increased release of prolactin from the rat's own pituitary. Conceivably this may apply also to intermedin and phenol, although evidence of depression of gonads by intermedin has been looked for and not found. It is fairly certain that depression of germ gland activity is one condition that favors the onset and exhibition of this parental behavior. Injections of highly potent gonadotropic hormone have repeatedly stopped, or diminished, maternal behavior after its natural occurrence ("reactors"), and also after its induction by prolactin or other hormone. Tests made with adrenotropin, cortin, parathormone, and thyroxine indicate that these hormones have little or no action on the maternal behavior of rats.

The colchicine technique has been used to determine whether the initiation of broody behavior in pigeons is or is not accompanied by a special or unusual release of prolactin from the bird's own pituitary, and thus whether an increased output of prolactin so coincides with the onset of broody behavior as to permit it to be causally related to it. The results show that this coincidence exists. At one day after the beginning of broodiness the preceding normal rate of mitosis in the crop epithelium has already increased by 600 per cent, and this higher rate is maintained and still further increased during the incubation cycle, near the end of which the secretion of crop milk begins. This result, involving a wholly new type of evidence, therefore definitely supports the conclusion that prolactin plays

56 CARNEGIE INSTITUTION OF WASHINGTON

a significant part in the cyclic development of broody and maternal behavior in higher animals.

Still other tests made by Riddle and Schooley relate importantly to this question of the mechanism through which the "broody instinct" arises in birds. They find that the corpus luteum hormone, progesterone, is capable of playing a part directly or indirectly in the initiation of broodiness in ring doves. Utilizing the "pellet implantation" method introduced by Parkes and Deansley, suitable tests were carried out — tests which avoid disturbances incident to repeated handling and injection of the birds. Implantation of such pellets of crystalline progesterone into adult and sub-adult ring doves, mated or unmated and of both sexes, is followed within 2 to 4 days by definite broody behavior. A pair of males thus treated carried out the incubation of eggs and the feeding of young. This ability to feed — the production of crop milk — demonstrates a release of extra or unusual amounts of prolactin prior to the end of incubation ; but it has not yet been proved that in these particular cases (see above for pigeons) this release also occurs at or before the beginning of broodiness. Continuous dosage with progesterone was shown to be unnecessary, since birds completed the brooding cycle after the pellets were removed. It is also notable that immature doves do not react in this manner to this same treatment.

Cytology of the Pituitary and Adrenal

An understanding of the structural basis of hormone production by anterior pituitary cells requires a correct interpretation of the life cycle of these cells, and no interpretation is susceptible of proof by direct observation. Nevertheless the subject is being studied with some success through indirect approach by Drs. Schooley and Riddle. Experimental modification of pituitary structure by treatment with hormonal and pharmacological agents, and by surgical procedures, have given some instances of nearly pure cultures of particular types of pituitary cells which can be correlated with known indices of physiological activity in the body of the animal.

In the pigeon's pituitary the undifferentiated cell (chromophobe) is interpreted as an embryonic element capable of maturing into either acidophil or basophil. These latter cells produce and release the hormones characteristic of the anterior pituitary and then die or undergo dedifferentiation into chromophobes. Indications were obtained of the manner in which the differentiated functional cells accomplish this dedifferentiation. The granulations seen in the pituitary cells seem not to be the active or actual hormone; and therefore the release of hormones does not rid the cell of the accumulation of granules present during active secretion. Apparently, however, such cells throw out this accumulation of granular material and are thus rejuvenated for a new cycle of activity. This discarded material accumulates between and outside the cells and becomes the pituitary "colloid" observed in the pituitaries of most animals. The rejuvenated cells have embryonic characteristics and appear to undergo further division, thereby producing a new supply of chromophobes which take up the work of hormone production as they mature into chromophils.

DIVISION OF ANIMAL BIOLOGY 57

The adrenal cortex produces one or more hormones under the stimulus of one or more pituitary hormones. The cytological changes that occur in the adrenal glands following hypophysectomy, and following administration of various pituitary hormones to normal or hypophysectomized pigeons, have been studied during the past year. A rich supply of adrenal tissue from birds previously subjected to many types of treatment is obtained in connection with other current investigations, and Dr. Richard A. Miller has studied the cytological changes thus induced in the cortical cells. It has been found that in the normal bird these cells contain very little fat; that hypophysectomy is followed not only by the well-known decrease of cortical tissue but by a great accumulation of fat (up to more than half of the cell volumes) in the persisting cells; that this accumulation of fat is progressive during more than 10 days (probably during 20 or 30 days) ; that such accumulation of fat is prevented by administration of whole extracts of anterior pituitary and by pituitary fractions rich in adrenotropin ; and that these two types of pituitary preparations, even in fasting hypophysectomized birds, tend to increase the mitochondria and to restore the Golgi apparatus to or toward the condition found in the normal unoperated bird.

Pituitary Hormones and Carbohydrate Metabolism

Both uncertainty and confusion characterize current reports concerning the pituitary hormones which affect carbohydrate metabolism. Further studies on one aspect of this problem have been made by Dr. Riddle, in association with Dr. Louis B. Dotti of New York Medical College. Prolactin moderately but consistently increases the blood sugar in all types of pigeons, in doves, and in rabbits (New Zealand Whites). In these species prolactin acts upon the blood sugar in much the same way (though in somewhat less degree) as do extracts containing all the pituitary hormones. Adrenotropin also is found to have a somewhat similar action on blood sugar, and still other results provide evidence that at least two different pituitary products affect the metabolism of carbohydrates.

Following Collip's recent report on the glycemic action of intermedin in rats and dogs, a new interest attaches to the action of that hormone. A slight glycemic action of intermedin was observed in normal rabbits and pigeons, but even a slight action was absent in eight tests made on hypophysectomized pigeons treated for 5 to 8 days. Details connected with these studies indicate that it is not possible to ascribe to intermedin the effects on blood sugar and basal heat production of pigeons which we have assigned to prolactin. The prolactin crystals of White, even in the low dosage used, increased the blood sugar of hypophysectomized pigeons.

Basal Metabolism in Functional Regulation

It becomes increasingly important to know which pituitary hormones have an action on basal heat production. More than a year ago O'Donovan and Collip reported that the hormone usually known as intermedin also has power to raise the B. M. R. in rats. That result made it necessary to learn whether intermedin and prolactin have comparable effects on the

58 CARNEGIE INSTITUTION OF WASHINGTON

B. M. R. of pigeons. This and some related problems have been studied by Riddle and Smith during the present year.

Intermedin, prepared and assayed in Hochst, Germany, and obtained through the courtesy of the Winthrop Chemical Company, was injected daily (500-1000 Phoxinus units) for 3 to 10 days into groups of pigeons. Measurements (36) made on normal pigeons 6 hours after a last injection of the hormone gave slightly higher (4 to 7 per cent) values for the heat production; other measurements (10) on hypophysectomized pigeons gave values 2 per cent above their control. The insignificant differences observed seem to demonstrate that intermedin and prolactin do not have comparable effects on the B. M. R. of pigeons.

The action of small amounts of the crystalline prolactin of Dr. White was studied in small groups of hypophysectomized (4) and of normal (8) pigeons. The former group was studied after 3 daily injections of 3.3 units (0.5 mg.) and the B. M. R. was found to be increased by about 14 per cent; the normal pigeons received one-half the above noted dosage during 3 days and at that time showed basal values increased by 12 per cent.

The action of adrenotropin on heat production is being studied. Two groups of normal pigeons given heavy daily dosage (10 mg.) during 3 and 9 days, respectively, increased their heat production values by 9 per cent. Since 10 mg. of this preparation of adrenotropin contains nearly 1 unit of prolactin, and since we cannot now declare it to be entirely free from thyrotropin, these incomplete data indicate that adrenotropin has little or no effect upon the basal heat production of pigeons.

Useful but now little-understood results were obtained from a comparison of the basal metabolism of normal and hypophysectomized young (2-month) pigeons, when both types were continuously fasted during 10 days. Measurements made on small groups of these two types of pigeons gave the unexpected result that at 5, 7, and 9 days after removal of their pituitary glands, and at 6, 8, and 10 days of fasting, these pigeons have a higher metabolic rate (measured at 30° C.) than have unoperated birds similarly fasted. At the above-named periods the operated group produced only 9, 8, and 5 per cent less heat than when unoperated (and fasted for only 24 hours) ; but the unoperated group at these periods of advanced fasting, produced 19, 27, and 31 per cent less heat than when fasted for the standard period of 24 hours. Our study of heat production in these and other related conditions continues to assist an interpretation of the mechanism of pituitary action.

Other Studies and Activities

Last year Dr. Bates reported a rapid method for the determination of the tryptophane content of proteins. Further studies have shown that the intensity of color developed by that method can be increased about 10 per cent and the small but disturbing drift of intensity previously present can be eliminated. These desirable results were obtained by using a smaller amount of HC1 (30 ml./lOO ml. final volume), and by diluting to volume with 50 per cent ethanol instead of with water. With these improvements the method gives excellent and reproducible results.

DIVISION OF ANIMAL BIOLOGY 59

Utilizing the recently introduced technique of McGinty and associates for demonstrating qualitatively the presence of extremely minute amounts of progesterone, Riddle and Schooley have tried to learn whether the corpus luteum hormone, progesterone, is produced in birds. Two types of bird ovaries were taken as possible sources and extracted according to the method of W. M. Allen (1932) ; in each case approximately 15 g. of fowl ovaries (yolk previously removed from follicles) were extracted. One sample of ovarian tissue was obtained from actively laying hens; the other sample, of similar fresh tissue, was from hens in various phases of egg production but previously injected (with prolan) daily for three days in an attempt to luteinize their ovaries. Neither of the resulting purified extracts contained enough (1 gamma) progesterone to produce a positive reaction in the prepared uterus of a suitable rabbit. The best available evidence on this point therefore indicates either that such fowl ovaries do not produce progesterone or that they produce an amount nearly or quite insignificant in comparison with that formed in the ovaries of adult mammals.

Studies initiated but not reported last year showed that a rapid hyperplasia of the crop-sac epithelium of pigeons quickly follows the injection of prolactin, and that this can be clearly and quantitatively measured by the colchicine technique. Details concerning quantities of colchicine (0.6 mg. per 500-g. bird) and time of injections of both the drug and the effective prolactin have been learned and published by Lahr and Riddle. This method was further used in the study of several related problems. In the crop-sacs of adult pigeons it was found that an injection of prolactin (60 units) is followed by (induced) cell divisions within less than 30 minutes; the maximum rate of mitosis occurs after 2 hours, and the effect is not observable after 10 to 12 hours. When a second injection of prolactin is given 8 hours after a first, this has produced a greatly enhanced effect 2 to 4 hours later. Mr. L. H. Elwell has assisted in the determination of the extent of this increase in mitotic rate following the administration of prolactin.

In a study made during a few months of summer, Riddle and Dotti have confirmed and extended their work reported last year on the calciumraising properties of female sex hormones. Sufficiently high dosage has now been shown to be effective in castrate male rabbits; but very high daily dosage (3 mg.) of estrone to rats may be less effective than much lower dosage, and under no dosage hitherto used is the blood calcium of rats greatly increased. Recent results show that estrone is wholly incapable of increasing the serum calcium of parathyroidectomized rats. This result favors the view that the action of the sex hormones on the serum calcium is secondary to their action on the parathyroids.

Several theoretical considerations make it worth while to know which of the many compounds chemically related to the group of sex hormones is capable of causing growth in the uterus. And from a phylogenetic standpoint this same information has added interest when it pertains to the uterus (oviduct) of a bird. A current study of this problem is being made by Riddle and Lahr, to whom Dr. Erwin Schwenk of the Schering Corporation has supplied generous quantities of the crystalline compounds.

The quantitative effect of these substances evidently undergoes unexpected seasonal variations, and this fact precludes a satisfactory description of these results at this time. Dihydroestrone, estrone, progesterone, androstendiol, dehydroandrosterone, testosterone, androsterone, and pregnenolone have all proved effective (and in nearly this order) on the oviducts of ring doves injected daily (1.5 mg. or less) for 7 days; cholestenone was ineffective. The effect of estrone on the dove oviduct is markedly synergized by progesterone. Possible and interesting theoretical implications of this particular fact may be noted. Though progesterone seems not to be produced in birds (see above), their oviducts, like the uteri of mammals, are capable of reacting markedly to the presence of this hormone. This suggests that when evolving mammals produced progesterone for the very first time the chief tissue (oviduct, uterus) upon which the hormone acts was already sensitive to its presence. Thus it is unnecessary to postulate a coincident or concurrent origin of two genetically unrelated things — a new hormone and a new or special sensitivity of a particular tissue to that hormone.

STUDIES IN EUGENICS AND HEREDITY

Harry H. Laughlin

Survey of the Human Resources of Connecticut

Early in the year which ended June 30, 1938, the task allotted in the collection of data, both by first hand field work and by correspondence, was completed, and during the latter half of the year the main efforts of the Survey were expended in the analysis of the returns and in the preparation of the Survey's Report No. 1.

The following progress report reviews briefly several researches conducted by the Survey:

Analysis of the Laws op Connecticut in Direct Reference to the Human Resources

of the State

(a) The constitutional law of Connecticut which defines and limits the function and authority of this sovereign state in its control of population by numbers, race, and inborn quality. This analysis of the legal background of the eugenical function and authority of the state covers the basic items of (1) definition, (2) population, (3) quality, (4) race, color, alienage, (5) migration control, and (6) the census and control of handicapped persons, in the main historical and legal documents which laid the governmental foundation of the present Commonwealth.

(b) Statute-reference and abstract of the laws of Connecticut which bear directly upon the eugenics of the Commonwealth, i.e., upon the establishment and maintenance of racial ideals and family-stock standards, upon the determination and maintenance of the optimum total population number and distribution, and specifically upon handling the inadequate and handicapped members of the population. This analysis covers the General Statutes of Connecticut in force on January 1, 1938; that is, the General Statutes of Connecticut, Revision of 1930, and the Session Acts of 1931, 1933, 1935, and 1937.

DIVISION OF ANIMAL BIOLOGY 61

The present analysis of the statutes of Connecticut presents a new point of view in the study of state law. Its purposes are, first, to enable the student of human resources to locate immediately the statute reference with a short abstract of every law which bears upon any one of the several eugenical subjects listed. The second purpose is to bring together with special clarity, and organized in systematic manner, all laws of the state which bear upon the subject of eugenics or race improvement, and thereby to enable judgment of the completeness, orderliness, and adequacy of the statutes on this subject, with special facility in locating any gaps, contradictions, overemphasis, or repetitions which may exist therein. The third purpose is to indicate a possible codification of the laws of the state in reference to the conservation of its human resources.

The 169 Towns of Connecticut, Each in Relation to Its Problem of Inadequate and

Handicapped Residents

The state of Connecticut is divided into eight counties. These eight counties, in turn, are composed of 169 towns in all. In Connecticut the county government is of little consequence compared with that of the state on the one hand and the town on the other. The real center of local government in Connecticut is the town.

The Survey of Human Resources assembled the basic geographic and demographic statistics of the 169 towns, and added to them the findings of its own first hand researches on the inadequate and handicapped residents of each town. These studies covered the census of inadequate and handicapped individuals resident of the town, the town's responsibility for such residents and how the town discharges such responsibility, and the classes and numbers of such residents who are cared for by the state custodial or residential institutions for the inadequate or handicapped classes.

The Problem of the Feeble-minded in Connecticut

This Survey has given a special consideration to the problem of the feebleminded within the population of the state. Making critical use of the most accurate standards for the mental measurement and the diagnostic classification of the feeble-minded, the present Survey collected and analyzed the case histories of the 11,962 definitely feeble-minded residents of Connecticut. It is estimated that, using the same standards, a complete census of the mentally deficient in the state would have found the total number of such defectives equal to approximately 17,500, or practically 1 per cent of the entire population of the state.

These 11,962 feeble-minded persons — the total number who came under the purview of the Survey — have been studied individually in reference to nine subjects, as follows: (1) sex, (2) age, (3) recidivism, (4) diagnostic class, (5) intelligence quotient, (6) race descent, (7) nativity, (8) citizenship, (9) kin in institutions.

The present Survey has completed twenty-one tables of cross-classification by pairs of the above listed traits and qualities for selected group units and totals within this population.

Cross-Classifications of Traits and Qualities of the Inmates of Institutions and Other Selected Socially Inadequate or Handicapped Population Groups of Connecticut

In this section of the Survey of the Human Resources of Connecticut, 19 portfolios consisting of a total of 346 cross-classification tables have been worked out.

The purpose of this system of classification is to provide more definite facts in reference to the history, characteristics, qualities, and origins of the particular social class or population group which is under analysis. In the current researches particular attention has been paid to the cross-classification of such matters as, for example, race descent against intelligence quotient;

Fig. 1. Sample cross-classification, intelligence quotient against diagnostic class, for the 661 inmates of the Connecticut State Prison for Men at Wethersfield, September 1, 1937

race descent against diagnostic class; or kin in institutions against intelligence quotient.

A representative study of the social inadequates of the state. Besides the inmates of several other institutions, the inmates of the Connecticut State Prison for Men at Wethersfield were cross-classified by pairs of subjects among the above listed nine items. Selections from these cross-classifications give many direct questions and answers such as the following:

What is the comparative intelligence-distribution among those Connecticut men prisoners of Italian descent and those of American blood, for whom I. Q.'s have been computed? Answer :

I.Q.

Prisoners of Italian descent (per cent)

Prisoners of American blood (per cent)

26-50

2.39 32.59 29.02 25.60 10.40

1.91

51-70

20.79

71-80

81-90

26.36 28.36

Above 90

22.68

Total

100.00

100.00

DIVISION OF ANIMAL BIOLOGY

63

What is the distribution of intelligence among the 119 men committed to the Connecticut prison for crimes against chastity — the so-called sex crimes? Answer :

Percentage of prisoners I. Q. committed for sex crimes

26-50 2.5

51-70 22.7

71-80 19.3

81-90 21.0

Above 90 10.1

Unknown 24 . 4

Total 100.0

It is noted that the idiot class, I. Q. 0-25, is not represented among these prison inmates, and that for nearly one-fourth (24.4 per cent) the I. Q. is unknown.

The Direct Cost to the People of Connecticut through Expenditures by the State Government (Exclusive of Expenditures by County and Town Governments and by Private Philanthropy and Charity) for the Maintenance, Care, Training, and Treatment of the Defective, the Dependent, the Delinquent, and the Handicapped

Classes

The Survey made a special analysis of the relative and absolute costs of the inadequate and handicapped population to the state government of Connecticut for the fiscal year 1935-1936 as compared with the fiscal year twenty years previous, 1915-1916.

While the money cost has mounted, the economic drag and the biological cost, as other items of the Survey have shown, have mounted still more rapidly, and the state is thus confronted with serious financial, economic, and biological problems, in the increasing numbers of inadequate and handicapped members of its population by reproduction and net migration, as compared with the competent and adequate members of the state's population and their relatively lower reproductive rates and net migration gains.

Change in twenty years (1915-1916 to 1985-1936), or in but little more than one-half of one

human generation

Estimated population

State governmental expenditures for all purposes

Such per capita expenditures for all purposes . . .

Total expenditures for defective, dependent, delinquent, and handicapped classes

Portion of total state expenditures devoted to defective, dependent, delinquent, and handicapped classes

Per capita cost

1915-1916

1,219,174

,339,428.00

$6.02

,503,022.00

20.5%

$1.23

1935-1936

1,734,000

$47,436,626.94

$27.36

$11,479,544.66

24.2% $6.62

Per cent change

+ 42.23 +546.33

+354.48

+663 . 76

+ 18.04 +438.21

From these data such conclusions as the following may be drawn: (1) In 1937 the people of Connecticut were expending, through their state government (exclusive of county, town, municipal, and private expenditures),

64 CARNEGIE INSTITUTION OF WASHINGTON

more per capita for the maintenance, care, training, and treatment of their socially inadequate and individually handicapped classes than this state government was expending for all purposes twenty years ago. (2) At the present rate every inhabitant of Connecticut is expending, through his state government, five and one-third times as many dollars per year on the socially inadequate and the individually handicapped as the average inhabitant was spending for the same purpose twenty years ago.

Experimental Census and Registry

One unit study in the Connecticut Survey of Human Resources is called "The Experimental Census and Population Registry of One Town." A representative town of about 2500 inhabitants was made the subject of this study, and field work was undertaken to make a census and registry of as large a portion of its population as available field workers would permit.

A card was designed and used as the basis of this experimental population enumeration and registry. The specific additions included such matters as personal identification by fingerprints, classification by race descent, and relationships to near blood-kin and to relatives by marriage. This card, which was used in the present experimental census registry, provides for records which show more definitely than the usual census the status and change of the subject-population in reference to race descent, tongue, literacy, medical case history, reproductive rate, occupation, employment, and economic condition, as well as family connections and conditions.

Studies in Human Heredity: Actuarial Genetics

During the year steady progress has been made on the analysis of stature inheritance in the American population, and concurrently in the development of the actuarial aspect of genetics. In the latter researches, the main question is : "When a given problem in human heredity is presented, accompanied by certain definite data concerning the possession or degree of development of the subject-trait among certain definite near-kin, by what probability will the pre-indicated offspring possess the named quality to the specified degree?" The subject-trait in its somatic development may be almost entirely hereditary; it may be almost entirely environmental; it may be a resultant of these two developmental forces acting in a definite combination ; and the hereditary part may be based upon one Mendelian gene or upon a complex of many. In any case the actuarial probabilities are worked out by comparison of Nature's behavior in the antecedent distribution of the subjecttrait within the subject-family as given with the same trait distribution among definite blood-kin within other families drawn from the same general population, all other factors being randomly represented.

Clinical Service in Human Heredity

Up to the present time the Eugenics Record Office has not served, nor intended to serve, in the capacity of a clinic in human heredity, but in an increasing degree such possible service is being forced upon the attention of the

DIVISION OF ANIMAL BIOLOGY 65

institution. Inquiries, either directly or by reference from other laboratories which were addressed directly, are sent to the Eugenics Record Office by intelligent members of families with specific problems in human heredity — physical, mental, or spiritual. They are sent also by physicians and friends of handicapped subjects, particularly by physicians who have not specialized in human genetics and who, therefore, call upon research students who have become expert in this field to collaborate with them.

In experimental response to these demands for clinical service in the field of human heredity, a new survey of the present status of knowledge of the rules of inheritance of given diagnosable and measurable human qualities has been undertaken, with the possibility of practical clinical service constantly in mind. The individual who applies to a clinic of this sort presents a definite problem in human heredity and desires an answer in the form of a definite "yes" or "no," or, at least, an answer in terms of dependable probabilities — as dependable for instance as the actuarial probabilities computed by a reliable insurance company in its own field.

Genetics of the Thoroughbred Horse Racing Success of Laboratory-produced Thoroughbreds

For many years Walter J. Salmon of New York City, distinguished breeder and racer of the Thoroughbred horse, maintained in conjunction with the Carnegie Institution of Washington a laboratory for research on the measure and inheritance of racing capacity. The scientific work of this laboratory was conducted by the Eugenics Record Office at Cold Spring Harbor, Long Island. A part of this work consisted in assembling and analyzing racing records of individual horses and of groups of definitely related Thoroughbreds. The second part of the laboratory consisted in the breeding farm, which put to experimental test the findings and indications of the researches in the racing and pedigree analysis. This experimental breeding farm was maintained by Mr. Salmon as the Mereworth Stud, in Lexington, Kentucky.

This laboratory in the course of its experimental breeding for the production of high racing capacity produced two of the most successful racing Thoroughbreds of recent years. These horses are:

Discovery, by Display, out of Ariadne; date foaled, 1931

Flat racing capacity, 140.93; number of races run, 63; number of races won, 27; number of races in which the quality of performance was above 1.000, 16.

Battleship, by Man 0' War, out of * Quarantaine ; date foaled, 1927

Flat racing capacity as a 2-, 3-, and 4-year-old, 123.48; number of flat races run,

22; number of flat races won, 10. Won Grand National Steeplechase at Aintree, England, March 25, 1938. Time,

9:29 4/5; distance, 4 miles, 856 yards; number of competitors, 13.

Research Publications

In January 1938, the complete texts for the so-called "Black Book," which is the practical work-volume to accompany volume I on the Measure of racing capacity, and for the so-called "Brown Book," which is the practical work-volume to accompany volume II on the Inheritance of racing capacity,

66 CARNEGIE INSTITUTION OF WASHINGTON

were deposited with the Washington office for publication. These two books contain the formulae and tables necessary for actual use (1) by the judge of racing capacity in the individual horse, and (2) by the practical breeder in planning mate selection for high racing capacity in the foal. The two basic volumes with their respective work books will complete the series of studies as originally planned on the genetics of the Thoroughbred horse. Volumes I and II will give the original data and their analyses which developed respectively the yardstick for the measure of racing capacity and the formula for predicting racing capacity in the foal. The experimental work has been finished, and the texts and figures for volumes I and II are in course of preparation for publication.

INVESTIGATIONS IN PROGRESS

Observations on the marmot. The study of the metabolism of the marmot and of factors related to the metabolism was terminated in the spring of 1937. There remained, however, certain lacunae in the data secured, particularly with regard to the body temperature gradient and the respiratory quotient during hibernation. These two problems have been studied during the past year by R. C. Lee, assisted by G. Lee.

Metabolism of the normal rabbit. The study of the heat production as influenced by differences in size within the same species has been continued under the direction of R. C. Lee (assisted by G. Lee, C. Hatch, and H. B.

DIVISION OF ANIMAL BIOLOGY 75

Lee) and now includes observations on rabbits ranging in weight from 1.5 kg. (Polish) to 7 kg. (Flemish Giant). Emphasis has been laid upon the effect of environmental temperature on the metabolism and the adaptation of the animal to various controlled environmental conditions after living under such conditions for a long time. A special study of the body temperature and the factors affecting it has been made. A group of wild cottontail rabbits have furnished considerable data for comparison of the wild with the domestic animal.

Metabolism of the narcotized rabbit. By the use of nembutal the cell temperature of rabbits has been lowered to approach the condition obtaining with hibernating marmots. With rabbits thus narcotized observations on heart rate, respiration rate, body temperature, respiratory quotient, and metabolism have been carried out by R. C. Lee, assisted by G. Lee, for the purpose of studying the factors that control the heat production and the body temperature.

Basal metabolism in experimentally produced atherosclerosis. For some time it has been known that experimentally atherosclerosis can be produced in animals by the feeding of cholesterol and that if thyroxin or thyroid extract is given simultaneously, atherosclerosis is not so likely to take place or may even be prevented. This suggests that the thyroid gland may play a role in controlling the development of this disease. For a number of years Dr. Timothy Leary, Medical Examiner of Suffolk County, Massachusetts, has been studying experimentally the production of atherosclerosis in rabbits by cholesterol feeding, with special reference to its pathology. During the past year the Nutrition Laboratory has begun a cooperative study with him on the basal metabolism of these animals. The basal metabolism has already been determined of a group of rabbits to be fed cholesterol and a group to serve as normal controls. Repetition of these determinations will be made after several months of feeding cholesterol, as a control on the level of the thyroid activity. The basal metabolism studies have been under the supervision of R. C. Lee, assisted by G. Lee.

Skin temperature of domestic animals. In cooperation with Professor E. G. Ritzman, at the University of New Hampshire, and under his direction, the skin temperatures of a pig, a goat, a sheep, and a ram were measured under various environmental conditions by R. C. Lee and N. F. Colovos. These measurements contribute to a study of heat loss and the conservation of heat by the animal body.

Metabolism studies with domestic animals. The profitable cooperative research on the metabolism of domestic animals, which has been carried on for a number of years with Professor E. G. Ritzman at the University of New Hampshire, was terminated January 1, 1938, so far as direct connection administratively between the Nutrition Laboratory and the University of New Hampshire is concerned. Professor Ritzman was then appointed Research Associate of the Carnegie Institution of Washington.

Metabolism studies on the Macacus rhesus. The metabolism measurements on the Macacus rhesus, which have been carried on for several years with the colony at the Department of Embryology at Baltimore with the cooperation of Dr. George L. Streeter and Dr. Carl G. Hartman, were brought to an end on October 1, 1937. The calculations and completion of the material for a report are in progress at the present time.

Respiratory quotient of the Dalmatian dog. The protein metabolism of the Dalmatian dog is characterized by a much larger urinary excretion of uric acid as an end product of the metabolism than occurs with any other breed of dog. Through the cooperation of Dr. Harry C. Trimble of the Department of Biochemistry of the Harvard Medical School, and with the assistance of B. James and M. Stankard, metabolism studies have been made on a dog of this breed with particular reference to the partition of the total metabolism in relation to the respiratory quotient and the nitrogenous excretion in the urine during fasting and after feeding of meat.

Effect of ingestion of hexoses on the respiratory quotient of the Macacus rhesus. An unusual opportunity for a study of the effect of sugars on the respiratory quotient of primates was afforded by the colony of rhesus monkeys at the Department of Embryology at Baltimore. After the measurements of basal metabolism were terminated, a special intensive series of respiration experiments was made in cooperation with Dr. C. G. Hartman, in which the respiratory quotients were determined after ingestion of glucose, fructose, or galactose. In the experiments with galactose the urinary excretion of the sugar was also determined. The respiration experiments were made by B. James. A preliminary report (see page 82) of this research has been presented to the American Physiological Society.

Effect of ingestion of hexoses on the respiratory quotient of the goat. There is evidence that metabolism of carbohydrates in the ruminant is different from that in other animal species, as is shown by the marked production of methane when ruminants are on full feed. In the general investigation on the effects of sugars on the respiratory quotients of animals it is obvious that this species should be included. Through the cooperation of Professor Ritzman, studies on the effect of ingestion of hexoses on the respiratory quotient of the goat have been initiated and are in progress.

The electrical method of gas analysis. For a number of years it has been evident that the immediate need in the advancement in the technique of metabolism studies is for a more rapid method of determination of the composition of the gases that have to be studied in measurements of respiratory exchange. During the past year an apparatus has been developed and described by Professor A. K. Noyons, of the University of Utrecht, Holland, for the determination of the changes in composition of gases by measurement of changes in resistance of a heated wire due to the changes in the surrounding gaseous atmosphere. Construction of a similar apparatus has been started at the Nutrition Laboratory by V. Coropatchinsky, for the purpose of comparing this apparatus with the standard form of gas analysis apparatus, in which the composition of gases is determined by volumetric analysis.

Prevention of respiratory failure in newborn infants. Dr. Priscilla White, of the Joslin Clinic and the New England Deaconess Hospital, in studying the prevention of respiratory failure in newborn infants of diabetic mothers, has continued the use of the Nutrition Laboratory's helium-oxygen chamber. R. C. Lee and G. Lee have assisted in this work.

Metabolism in diabetes. In cooperation with Dr. Howard F. Root of the New England Deaconess Hospital and with the assistance of R. C. Lee and B. James, a special study was made of a patient who not only had diabetes but also had had acromegaly at some time during the past. The study of the metabolism during diabetic coma is to be continued as opportunity offers.

Maximum temperature of expired air as an index of body temperature. The apparatus for determining the maximum temperature of the expired air (mentioned in Year Book No. 34, p. 65) has been further developed and the relationship of the maximum temperature of the expired air to the body temperature established. This apparatus is particularly adapted for measuring the body temperatures of large groups of humans in that measurements can be made at the rate of four per minute. Numerous measurements of normal subjects were made at the Nutrition Laboratory, and a special series of measurements of patients with artificially produced fever was made at the Boston Psychopathic Hospital through the cooperation of Dr. H. C. Solomon and Dr. I. Kopp. This research was conducted by R. C. Lee, assisted by H. B. Lee.

Effect of ingestion of foods on the human respiratory quotient. The carbohydrates in the human diet are not, for the most part, derived directly from pure sugars but from mixtures of simple and hydrolyzable sugars and starches. Such combinations are found in vegetables, cereals, fruits, and nuts. The rise in the respiratory quotient is one index of the availability of carbohydrates to the body and the rapidity with which they are burned. Experiments were made on the changes in the respiratory quotient with respect to height of the rise and the time relationships, in which single portions of common foods (cooked and raw vegetables, bread, nuts, and fruits) were given that contained approximately 25 grams of carbohydrates each. Control experiments with 25 grams of glucose or of cane sugar served for comparison. Samples of the foods given are to be analyzed for reducing sugars, hydrolyzable sugars, and starches, to compare with the actual changes in respiratory quotient and to determine which of the carbohydrates are most effective in causing a rise in the respiratory quotient. The respiration experiments are being carried out by B. James and the chemical analyses by M. Stankard.

LITERARY WORK

A paper on "The maximum temperature of expired air as a rapid measure of human body temperature" has been written by Dr. and Mrs. Benedict and Mr. and Mrs. R. C. Lee and has been accepted for publication by the New England Journal of Medicine. A manuscript on the body temperature of the normal rabbit and factors affecting it is in the process of preparation by R. C. Lee. The large amount of proofreading and editorial work has had the capable supervision of the editor, Elsie A. Wilson.

PUBLICATIONS

(1) The basal metabolism and urinary nitrogen excretion of Chinese, Manchus, and others of the Mongolian race. Francis G. Benedict, Lan-Chen Kung, and Stanley D. Wilson. Chinese Jour. Physiol., vol. 12, pp. 67-100 (1937).

Basal metabolism measurements on 120 adult Mongolians, chiefly Chinese and Manchus, showed as with other human races an increase in total heat production with increasing weight, a decrease in heat production per kilogram the larger the individual, a decrease in total heat production with advancing age among subjects of the same weight, and a lower metabolism of the women than of the men. Pulse rates and nitrogen output per kilogram of body weight were in general the same as noted with Caucasians. The decrease in total heat production per year increase in age averaged about the same for the male Chinese as for Caucasian men (7 calories) but averaged 5 calories for the Chinese women as compared with 2.3 calories for Caucasian women. No marked influence of Western civilization upon the metabolism of those Chinese who had adopted Western ways of living was apparent. The basal metabolism of these Chinese as a whole was, by every method of comparison, lower than that noted with Caucasians, the difference being more pronounced with the men than with the women.

(2) Die Bedeutung des Korperfettes filr die Warmebildung im Organismus. Francis G.

Benedict and Robert C. Lee. Biochem. Ztschr., vol. 293, pp. 405-409 (1937).

Within the weight range from 2.5 to 7.5 kg. adult geese had a total heat production that increased in a straight-line relationship with the increase in weight. Inasmuch as in the surfeit feeding of adult geese the protein content of the body is only slightly increased, the increase in heat production of these geese with increasing weight was the result of an increased storage of fat. This finding is supported by observations on the basal metabolism and body composition of 8-g., 21-g., and 60-g. mice. Body fat should, therefore, not be considered as metabolically inert but as an energy-demanding substance.

(3) Basal metabolism of rats in relation to old age and exercise during old age. Francis

G. Benedict and Henry C. Sherman. Jour. Nutrition, vol. 14, pp. 179-198 (1937).

With a group of adult rats (non-exercised) of different ages but of the same weights the basal 24-hour heat production was somewhat higher in old age than in middle life. With a group of adult rats (non-exercised) of different weights as well as ages, however, the heat production was relatively constant at the older ages. The total heat production of any one rat decreased slightly with advancing age, but at the same time there was a relatively larger (although also small) decrease in body weight, so that the metabolism per unit of weight and per unit of surface area increased slightly. The body temperature tended to decrease in old age (after 800 days of age), and in very advanced age the decrease amounted to about 2° C. Middle-aged male rats, not previously exercised, could not adjust themselves to strenuous exercise begun so late in life, lost weight rapidly, and usually died. Middleaged female rats, not previously exercised, were apparently benefited by the exercise, which tended to lower their basal metabolism. It is suggested that the organism of the exercised rat is freed by muscular exercise from a middle-age restlessness or chronic useless tenseness and is able to relax better in rest periods.

(4) Race: A factor in human metabolism. Francis G. Benedict. Proc. Amer. Philos. Soc,

vol. 78, pp. 101-110 (1937).

This paper summarizes the metabolic findings of a racial survey made by the Nutrition Laboratory and its collaborators. Oriental races in general were found to have a metabolism somewhat lower than that of Caucasians in the United States. South Indian women in Madras had a metabolism 17 per cent below the Caucasian prediction standards, and this was further depressed about 10 per cent during deep sleep. In striking contrast to the low metabolism of the majority of the Oriental races were the high metabolism of the Maya (+8 per cent) in Yucatan and the Miao males (+16 per cent) in Szechwan, China, although in both instances low pulse rates were observed. With Manchus of the laboring class, both males and females, lower values for total heat production were found than with Chinese laborers of the same weight, age, and sex. This survey has established that there are marked differences in the basal metabolism of different human races, probably ascribable solely to the racial factor and not to differences in climate and diet, and that there may even be a racial difference in metabolism within the Oriental race itself.

(5) Lipogenesis in the animal body, with special reference to the physiology of the goose.

Francis G. Benedict and Robert C. Lee. Carnegie Inst. Wash. Pub. No. 489 (1937) . ix + 232 pp., 30 figs., 35 tables.

To study the metabolism during the transformation of carbohydrate into body fat (lipogenesis) adult geese were fed surfeit amounts of a corn-meal mixture, following which measurements were made of the respiratory exchange and the heat production. The latter was determined directly by means of an emission calorimeter, described in detail. The respiratory exchange measurements, the techniques for which are also described, included the oxygen consumption, the carbon dioxide elimination, the cleavage carbon dioxide, and the respiratory quotient. The physiology of the goose was also studied under normal basal conditions and during prolonged fasting, and observations were made of the rectal temperature, heart rate, respiration rate, insensible perspiration, water-vapor output, chemical composition of the body, and zone of thermic neutrality. By surfeit feeding of carbohydrate the oxygen consumption of the goose may be increased 50 to 100 per cent and the carbon dioxide elimination 100 to 400 per cent. The respiratory quotient may increase to 1.48. Maximum values for all three factors and also for cleavage carbon dioxide occur within 1 to 3 hours after surfeit feeding and persist for about 3 hours, but the usual basal values are found again within 24 hours. According to the data obtained in the calorimeter experiments, when carbohydrate is converted into fat after surfeit feeding and the respiratory quotient remains at a constant high level of about 1.40, the relationship between the simultaneously measured oxygen consumption and the heat production is the same as that during the combustion of pure carbohydrate at a respiratory quotient of 1.00. Hence when the respiratory quotient is above 1.00, the heat production can be calculated indirectly with sufficient accuracy from the measured oxygen consumption by use of the factor of 5.047 calories per liter of oxygen.

(6) Effects of thyroidectomy and thyroid feeding in geese on the basal metabolism at dif ferent temperatures. Milton O. Lee and Robert C. Lee. Endocrinology, vol. 21, pp. 790-799 (1937).

The metabolism (per lOw 2 / 3 ) of twelve normal geese was 25 per cent higher at 11° C. than their basal rate at 23° C. Four of these geese, after thyroidectomy, had a metabolism, on the average, 51 per cent higher at 11° C. than their average level at 23° C. The metabolism of these thyroidectomized geese was 33 per cent lower at 23° than that of the normal birds but only 15 per cent lower at 11° C. The thyroidectomized goose can make metabolic adjustments to low environmental temperatures as great in magnitude as the normal goose. The thyroid gland is, therefore, not necessary to the goose for thermogenesis in adaptation to a cold environment.

(7) Further observations on the physiology of the elephant. Francis G. Benedict and

Robert C. Lee. Jour. Mammal., vol. 19, pp. 175-194 (1938).

Supplementing the detailed study of the physiology of the elephant reported in 1936 (Carnegie Inst. Wash. Pub. No. 474), further observations were made, particularly to fill obvious lacunae in the first investigation. These deal with the muscular activity of the elephant, the sleeping positions assumed, the teeth, the habits of eating and chewing, the hearing, the reaction to rats and mice, the maximum possible weight, measurements of height, microscopic examinations and temperature measurements of feces and urine, skin temperature measurements, determinations of methane in the intestinal gases, and observations on respiration rate and heart rate. Men in charge of zoological parks and circuses are urged to notify scientists when elephants are to be killed, so that further studies can be made when these animals are dissected.

(8) The nutritional physiology of the adult ruminant. Ernest G. Ritzman and Francis G.

Benedict. Carnegie Inst. Wash. Pub. No. 494 (1938). vi + 20O 7 p., 3 pis., 3 figs., 55 tables.

In this monograph are reported the results of a cooperative investigation by the New Hampshire Agricultural Experiment Station and the Nutrition Laboratory, dealing primarily with the nutritional physiology of the cow but supplemented by observations on steers, bulls, sheep, goats, and horses. Eleven cows were studied under maintenance conditions of feeding, over periods ranging from 4 months to nearly 4 years, and with these animals 49 digestion balance experiments (each one month long) were made. At the end of each digestion experiment the respiratory exchange was measured, first with the cows on feed and then on the fourth and fifth days of fasting. The rations fed were hay alone (six kinds, of early and late cutting), concentrate alone (corn meal or linseed oil meal), or a mixture of concentrate and bran. The cows (Holsteins and Jerseys) were all adults and in most instances dry and not pregnant, although a special study was made of two cows during pregnancy and lactation. The first section of the monograph deals with digestibility of roughages and of concentrates, weight and character of fill, methane production, chemical composition of excreta, nitrogen and energy balances, watervapor output, insensible perspiration, heart rate, respiration rate, and body temperature. In the second section the conditions prerequisite for basal metabolism measurements on ruminants are outlined and the results of such measurements on the cows are discussed. Great variability was noted in the metabolism of one and the same cow (dry and not pregnant), changes of 30 to 74 per cent occurring within as short a time as six weeks without marked difference in nutritive condition. The cause of this variability is considered to be of hormone origin, due to selective breeding for milk production. The percentage increases in metabolism above the basal level due to the stimulating effects of the ingestion of the different roughages (predominantly carbohydrate in nature) were as great with these cows as the increases following protein ingestion noted with dogs and humans. In the last section of the book the factors to be considered in evaluating the utilization of food energy by cattle are discussed, especially the dynamic stimulus of food and the great variability in basal metabolism. The digestible energy, the net energy, and the metabolizable energy as measured in the digestion balance experiments are also discussed, and in the concluding pages there is a brief consideration of the efficiency of metabolizable energy for maintenance and for milk production.

(9) Hibernation and marmot physiology. Francis G. Benedict and Robert C. Lee. Car negie Inst. Wash. Pub. No. 497 (1938). x + 239 pp., 2 pis., 11 figs., 58 tables.

A metabolic study of the marmot was made to secure information regarding the physiology of the animal itself and for comparison with other warm-blooded animals that do not hibernate and with cold-blooded animals, for in its awake and asleep periods it resembles somewhat these two types of animals, respectively. Measurements of body weight changes, insensible perspiration, heart and respiration rates, rectal temperature, respiratory exchange, and water-vapor output were made on 48 marmots, at different environmental temperatures, in the non-hibernating and hibernating states, and in the transitional stages of entering and awakening from hibernation. Several marmots were drugged with nembutal and subsequently exposed to a cold environment, others were subjected to carbon dioxide narcosis and cold, and the measurements on these were compared with those on the animals hibernating under normal conditions. The urines voided by four marmots while fasting and hibernating were analyzed to determine the partition of urinary nitrogen. During hibernation there was no marked change in the character of the respiratory exchange, the true combustion respiratory quotient was one of fat, and the protein metabolism underwent no qualitative alteration. When not hibernating, the marmot has a labile basal heat production and rectal temperature, but when its body temperature is 36.9° C. its basal metabolism averages about 400 calories per 10w 2 / 3 per 24 hours. This is much higher than that of any cold-blooded animal of the same size at a body temperature of 37° C. but is considerably lower than the metabolism of other warm-blooded animals of the same size thus far studied. The hibernating marmot resembles the cold-blooded animal in that, when exposed to a low environmental temperature, its rectal temperature becomes very low and its respiration rate slow, but its heat production per lOw 2 / 3 even at the minimum level is two or three times that of a snake of the same size having the same low body temperature. The causes and theories of hibernation are discussed, and a digest of the main findings of the investigation is presented.

(10) Vital energetics: A study in comparative basal metabolism. Francis G. Benedict. Carnegie Inst. Wash. Pub. No. 503 (1938). vii + 215 pp., 46 figs., 4 tables.

The basal metabolism measurements on all the animal species that have been studied at the Nutrition Laboratory during the past three decades are analyzed in this report from the standpoint of comparisons within the same and between different species. Only adult animals are considered. These ranged in size from the 8-g. dwarf mouse to the 4000-kg. elephant. The prerequisites for comparable measurements of basal heat production are outlined, and consideration is given to the bases for intraspecific and interspecific comparisons of animals of different sizes, with particular reference to the metabolically inert factors affecting body weight and the problems of measuring or calculating the true surface area of the body. For each warm-blooded species a chart is given, in which the total 24-hour heat production of each animal measured is plotted with reference to its body weight and a curve is drawn through the plotted data to indicate the general trend of the metabolism with increasing weight. These intraspecific comparisons represent eighteen species of mammals, including humans, seven species of domesticated birds, and several wild birds. The interspecific comparisons are based upon a series of charts for progressive weight groups and also for the entire weight range from 8 g. to 4000 kg., in which are assembled the curves representing the trends of the average metabolism of the different species referred to weight and expressed as total heat production, heat production per kilogram, and heat production per lOw 2 / 3 . The measurements at environmental temperatures of 16° and 28° C. are compared, and there is a critique of the surface area concept. The measurements are also referred to different powers of the body weight. Finally, warm-blooded animals are compared with coldblooded animals both at low cell temperatures and at 37° C. Emphasis is laid upon the value of basing metabolism comparisons upon the total heat production of species of the same body weight, thus eliminating the problem of how to equalize differences in weight. Discussion is given of the factors that may contribute to metabolic differences within and between species, such as body structure, composition, and covering, cell temperature, cell enzymes, brain weight, and blood. It is recommended that one should turn from a consideration of heat loss to a consideration of heat production, that the lack of uniformity in heat production should be recognized, and that the differences in metabolic intensity or vital energetics should be associated with differences in body configuration and composition and with differences in the morphology, chemistry, and particularly the distribution of the blood.

(11) The partition of urinary nitrogen of fasting and hibernating woodchucks (Arctomys

monax). Thorne M. Carpenter. Jour. Biol. Chem., vol. 122, pp. 343-347 (1938).

One specific detailed example is given of the results of a study of the partition of urinary nitrogen in the urine of a marmot. The details of other studies are given in the monograph cited on page 80 of this report.

(12) Effects of hexoses upon the respiratory quotient of the rhesus monkey. Thorne M.

Carpenter and Carl G. Hartman. Amer. Jour. Physiol., vol. 123, p. 32 (1938).

Abstract. (See page 76.)

(13) The effect of urea on the human respiratory exchange and alveolar carbon dioxide.

Thorne M. Carpenter. Jour. Nutrition, vol. 15, pp. 499-512 (1938).

Ingestion of 30 and 40 g. of urea caused marked rises in the alveolar carbon dioxide and in the respiratory quotient during 3 to 3% hours after ingestion, but no change in the oxygen consumption, as compared with these same factors in control experiments. The simultaneous increases in alveolar carbon dioxide and respiratory quotient are the result of the alkalosis following the ingestion of urea. As the alveolar carbon dioxide is also increased by the gastric secretion following the ingestion of protein, there may be two causes for the alterations in the respiratory quotient from the true respiratory quotient of protein, one the gastric secretion containing hydrochloric acid and the other the urea that ultimately results as a metabolic product of the transformations of protein in the body.

(14) The effect of ingestion of alcohol on human respiratory exchange (oxygen consumption

and R. Q.) during rest and muscular work. Thorne M. Carpenter and Robert C. Lee. Arbeitsphysiologie, vol. 10, pp. 130-157 (1938).

Measurements of the respiratory quotient and the oxygen absorption after ingestion of 30 and 50 cc. of alcohol, respectively, indicated that the metabolism of alcohol proceeded at about the same rate, irrespective of whether the subject was resting or exercising on a bicycle ergometer. That muscular work does not increase the combustion of alcohol was also demonstrated by the calculations of the changes in the katabolism of carbohydrate and fat in the periods of work and recovery following alcohol ingestion. There was no summation of the separate effects of alcohol and muscular work when both these factors were superimposed upon the basal metabolism. The efficiency of performance of work (relation between the heat equivalent of the work performed and the energy expenditure) was not the same in experiments with alcohol ingestion as in experiments without alcohol ingestion.

(15) The effect of muscular work on the amounts of alcohol in urine, expired air, and blood,

after its ingestion by man. Thorne M. Carpenter and Robert C. Lee. Arbeitsphysiologie, vol. 10, pp. 158-171 (1938).

Exercise on a bicycle ergometer at varying rates of speed for periods of 30 minutes to 2 hours after ingestion of 30 and 50 cc. of alcohol did not appreciably alter the concentration of alcohol in urine, blood, or expired air, or the amount of alcohol eliminated per liter of carbon dioxide exhaled, as compared with these same factors during rest. The amounts of alcohol in the ventilating air current of the respiration apparatus were greater during exercise than during rest and were greater the severer the work, because of the increased total ventilation of the lungs. However, when the work ceased, the amounts approached those found in the same time interval after ingestion in the rest experiments. In the rest experiments from 0.4 to 0.7 per cent of the total alcohol ingested was eliminated in the ventilating air current and from 0.8 to 1.6 per cent was eliminated in the urine and expired air. In the work experiments these percentages were 0.9 to 1.6 and 1.1 to 2.1, respectively. The disappearance of alcohol through these paths plays only a small role in reducing the amount of alcohol in the body.

(16) The effect of muscular work on the metabolism of man after the ingestion of sucrose and galactose. Thorne M. Carpenter and Robert C. Lee. Arbeitsphysiologie, vol. 10, pp. 172-187 (1938).

Measurements of the respiratory exchange were made in 3-hour experiments after ingestion of 70 g. of sucrose or 50 g. of galactose, with the subject at rest and at work. The exercise (pedaling a bicycle ergometer) resulted in a greater combustion of cane sugar but not of galactose. It is questionable whether muscular work has any effect on the metabolism of galactose except apparently to accelerate the reaction after the ingestion of the sugar. There was no summation of the effect of the ingestion of the sugar (either sucrose or galactose) at rest and the effect of muscular work without sugar when the sugar ingestion was accompanied by work. The efficiency of performance of work was better when sugar was ingested than when it was not and was slightly better after galactose than after sucrose. More of the ingested sugar was utilized in the more intense work.

TORTUGAS LABORATORY

D. H. Tbnnent, Executive Officer

Paul S. Conger, Assistant Executive Officer

During the summer of 1938 the Tortugas Laboratory was open from June 2 to August 9.

The following investigators studied at the Laboratory during the season:

P. L. Bailey, Jr., College of the City of New York. Regeneration in sabellids. June

30 to August 9. N. J. Berrill, McGill University. Budding in polystyelid ascidians. June 16 to June

28. F. J. Brinley, North Dakota Agricultural College. Origin of muscular movements

in fish embryos. June 16 to August 9. Leonard B. Clark, Union College. Swarming of the Atlantic palolo. Habits of the

basket star. June 16 to July 26. Paul S. Conger, U. S. National Museum and Carnegie Institution of Washington.

Investigations on diatoms. June 2 to August 9. B. R. Coonfield, Brooklyn College. The development and coordination of melano phores in embryos of Pomacentrus. June 30 to August 9. Hugh H. Darby, College of Physicians, Columbia University. Continuation of studies

of regeneration in Crustacea. June 30 to August 9. John H. Davis, Jr., Southwestern College. Studies of mangroves and changes of

strand flora. June 16 to June 28 and July 28 to August 9. Walter N. Hess, Hamilton College. Reactions to light and the photoreceptors in the

spiny lobster. Habits of the basket star. Swarming of the Atlantic palolo.

June 16 to July 28. Norris Jones, Swarthmore College. Investigations on ascidians. June 30 to July 26. Balduin Lucke, The School of Medicine, University of Pennsylvania. Studies on

tumors in cold-blooded vertebrates. June 2 to June 28. Gordon Marsh, State University of Iowa. Further studies on the electrical behavior

of Valonia. June 2 to August 9. Paul A. Nicoll, University of Chicago. The response of ascidian larvae toward certain

hormones. June 30 to August 9. Fernandus Payne, Indiana University. Study of the effect of anterior pituitary

hormones on the development and discharge of the sex cells in Ptychodera baha mensis. Observation of Amphioxus larvae. June 30 to August 9. Harold H. Plough, Amherst College. Investigations on ascidians. June 30 to July 28. Gordon A. Riley, Yale University. Study of the quantity of plankton in tropical

waters. July 14 to August 9. Vance Tartar, Yale University. Regeneration in the starfish Linckia and in the

protozoan Condylostoma. June 30 to August 9. D. H. Tennent, Bryn Mawr College. Effect of intensity of light on photodynamic

reactions. June 16 to July 28.

Regeneration in Sabellids P. L. Bailey, Jr.

The problem originally intended was a study of the effect of chemical solutions on regeneration in the littoral oligochaete, Pontodrilus, which has previously been reported as remarkably adjusted to marine existence. The organism proved not to be especially suited to experimental work with solutions. Mortality of individuals kept in sea water, or in sea-water solutions of dinitrophenol, colchicine, and thyroxine, was very great; regeneration of lost parts was too slow to be of value in the time available; and failure of the wound to close properly caused in many control and experimental animals a peculiar bulbous abnormality, without growth of new segments.

The sabellids (genera and species of which I have not completely identified to date) found plentifully in the moat at Fort Jefferson, however, proved to be far more satisfactory. The major portion of the work was spent adjusting different strengths of solution to these worms. Individuals in which 40 to 50 posterior segments had been amputated failed to regenerate lost segments until after 15 days when kept in a solution of 1 grain of colchicine in 1 liter of sea water. Controls in pure sea water began to regenerate lost segments in 3 days. In less concentrated solutions the rate of regeneration was less retarded. Individuals were fixed at various stages of regeneration with the hope that histological study of the effect of the chemical on the cells will reveal further facts concerning the origin of the new tissues in the regenerative process.

A portion of the time was spent in observation and experiments on the tube-forming mechanisms of these same sabellids. It is hoped that a more detailed description of the anatomy and physiology of the process can be given than has heretofore been presented, after further experiments can be made and a histological study of preserved material can be completed.

Budding in Polystyelid Ascidians N. J. Berrill

The investigation relates to the general problem of the nature of inherent organization manifest in bud rudiments and regeneration blastemas. Such anlagen are apparently undifferentiated as wholes and consist usually of unspecialized cells. Polystyelid ascidians represent ideal material for studying this problem inasmuch as on the one hand they form a compact natural group with highly distinctive larval, postlarval, and adult characters, and on the other vary considerably in colony form and size, in size of constituent zooids, and in time sequences of asexual reproduction. Material was collected, studied alive, and preserved for more detailed investigation, of the forms Polyandrocarpa tincta, Symplegma viride, and Botryllus nigra. It was found that the area of the bud rudiment relative to the size of the parent zooid at the time of its appearance was closely related to the size of mature zooids and the general nature of the colony. It is also evident that the bud rudiment must be regarded as an integral part of the individual organization, and as such its development as a part of an individual is strictly comparable with features such as gill slits or heart. Further, in Botryllus at least, the absolute size of a bud rudiment determines the extent to which mature or immature gonads will develop at a time when the rudiment consists of a twolayered disk of unspecialized epithelial cells of ectodermal origin, irrespective of the nutritive supply of the developing bud. It is hoped that further study of this and similar material will show clearly the validity of the field or Gestalt concept of organization when applied to blastemas exhibiting no visible differentiation either as wholes or in their constituent cells.

Study of the Origin of Muscular Movement in Fish Embryos

Floyd J. Brinley

The work here reported was done at the Tortugas Laboratory from June 16 until August 9. It is a continuation of a general problem on the relation of innervation to the origin of the heart beat, peristalsis in the alimentary canal, body and fin movement. The organisms used were the embryos and newly hatched larvae of Pomacentrus leucostictus and embryos of the nurse shark.

Muscular contraction or peristalsis of the smooth muscles of the alimentary canal starts in Pomacentrus embryos two or three days prior to hatching. It is first observed in the fore- and hind-gut and later develops in the mid-gut. Contractions occur in the fore-gut at a definite rhythm of about 24 beats per minute. Peristalsis originates in the anterior region and passes caudad, where it ceases before entering the mid-gut. Contractions of the hind-gut occur at irregular intervals and may originate at the junction with the mid-gut or at the posterior end. Contractions of the mid-gut occur less frequently and at irregular intervals. These three regions of the alimentary canal in the embryos and newly hatched larvae behave as separate physiological units and are independent of each other in regard to muscular activity.

In older larvae, after complete absorption of the yolk, the definite rhythm of contraction of the fore-gut is replaced by a much slower wave-like movement which occurs at irregular intervals. The amplitude of contractions of the mid- and hind-gut in the older embryos is greater than in the younger stages, and peristaltic waves which originate in the mid-gut usually continue to the hind-gut.

Injections of atropine sulfate solution (0.05 to 0.1 per cent) or ergotoxin (0.1 per cent), which depress the parasympathetic and sympathetic nervous systems respectively, have no effect on the muscular movement of the alimentary canal of the embryos or day-old larvae. Injections of atropine into older larvae depress peristalsis and constrict the entire gut. Ergotoxin increases muscular activity in all portions of the alimentary canal in the older embryos and dilates the mid- and hind-gut.

These reactions indicate that muscular contractions of the smooth muscles originate within the muscle but are soon controlled by the ingrowth of the autonomic nerves.

In the nurse shark body movement is first observed in 6-mm. embryos. Microinjections of curare (which blocks the transmission of impulses from motor nerves to striated muscles) does not affect muscular activity until the embryos have reached a length of 18 mm. In embryos from 18 mm. to 48 mm., curare reduces body movement to a series of single twitches at definite intervals varying from a few seconds to two minutes. In embryos larger than 48 mm. the drug completely blocks all impulses or reduces activity to slight twitches at several -minute intervals.

It appears from these results that muscular contraction of striated muscles originates in the muscle and the ingrowth of motor nerves takes place between 18 and 48 mm. During this stage body movement is largely due to impulses coming from the nervous system. When these impulses are blocked by curare, spontaneous contractions of the muscles continue at a definite rhythm which probably is a fundamental property of all muscles. In embryos larger than 48 mm. the stimulant for contraction arises entirely in the nervous system, and conditions within the muscles are not suitable for spontaneous contractions.

A study of the effect of caffeine on the rate of heart beat in shark embryos indicates that the vagus makes contact with the heart when the embryo is about 54 mm. in length. 1 Prior to that stage, caffeine has no effect on the rate of heart beat, but in older embryos the heart shows a decided irregularity in rhythm within 2 minutes after injections of caffeine. Digitalin (0.005 to 0.1 per cent) reduces the rate of the prevagus heart, which results in stoppage within 10 to 15 minutes after injection. However, digitalin produces a slight increase in the rate of the innervated heart immediately after injection. This increase passes into a depression and finally the heart stops in diastole.

A large number of shark embryos were fixed for embryological studies.