# Anatomical Record 11 (1917)

Embryology - 19 Sep 2020        Expand to Translate
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

 A personal message from Dr Mark Hill (May 2020) I have decided to take early retirement in September 2020. During the many years online I have received wonderful feedback from many readers, researchers and students interested in human embryology. I especially thank my research collaborators and contributors to the site. The good news is Embryology will remain online and I will continue my association with UNSW Australia. I look forward to updating and including the many exciting new discoveries in Embryology!
Historic Disclaimer - information about historic embryology pages
Pages where the terms "Historic" (textbooks, papers, people, recommendations) appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

# THE ANATOMICAL RECORD

EDITORIAL BOARD

Irving Hardesty Tulane University

Clarence M. Jackson University of Minnesota

Thomas G. Lee University of Minnesota

Frederic T. Lewis Harvard University

Warren H. Lewis Johns Hopkins University

Charles F. W. McClure Princeton University

William S. Miller University of Wisconsin

Florence R. Sarin Johns Hopkins University

George L. Streeter University of Michigan

G. Carl Huber, Managing Editor

1330 Hill Street, Ann Arbor, Michigan

VOLUME 11 AUGUST, 1916-JANUARY, 1917

THE WISTAR INSTITUTE OF ANATOMY AND BIOLOGY

COMPOSED AND PRINTED AT THE

WAVERLY PRESS

By THE Williams & Wilkins Company

Baltimore, Mi>., U. S. A.

## Contents

### No. 1. AUGUST

Eliot R. Clauk. A study of the reaction of mesenchyme cells in the tad-pole's tail toward injected oil globules. Five figures

Louis H. Kornder. An anomalous urinogenital system in a dog. Two figures 10

W. SoHiER Bryant. Sensory elements in the human cerebral hypophysis. One figure. 25

Emily Ray Gregory. A method for micro-injection. One figure 29

### No. 2. SEPTEMBER

William B. Kirkham. The prolonged gestation period in suckling mice 81

Helen Dean King. On the postnatal growth of the body and of the central nervous system in albino rats that are undersized at birth 41

E. L. JuDAH. Mounting specimens under Petri dishes and clock glasses 5.

### No. 3. OCTOBER

P. E. Smith. The effect of hypophysectomy in the early embryo upon the growth and development of the frog. Ten figures 57

Carbon Gillaspie, Lewis I. Miller and Morris Baskin. Anomalies in lobation of lungs with review of literature. Five figures Go

Carbon Gillaspie, Lewis I. Miller and Morris Baskin. Anomalous renal vessels and their surgical significance. Nine figures 77

Osc.\R Riddle. Size and length relations of the right and left testes of pigeons in health and disease S7

J. A. Long. Hygienic cages for rats and mice. Two figures 103

### No. 4. NOVEMBER

Alwin M. Pappenheimer. The Golgi apparatus. Personal observations and a review of the literature. Twenty-two figures 107

G. Carl Huber. On the form and arrangement in fasciculi of striated voluntary muscle fibers. Four figures 149

G. Carl Huber. A note on the structure of the elastica interna of arteries. One figure 169

G. Carl Huber. A note on the morphology of the seminiferous tubules of birds. One figure 177

### No. 5. DECEMBER

P. G. Shipley and R. S. Cunningham. The histology of blood and lymphatic vessels during the passage of foreign fluids through their walls. II. Studies on absorption from serous cavities 181

Wilbur C. Smith. A case of a left superior vena cava without a corresponding vessel on the right side. Two figures 191

G. S. Hopkins. The innervation of the muscle retractor oculi. One figure 199

Eben Carey. The anatomy with especial consideration of the embryological significance of the structures of a full-term fetus amorphus. Nineteen figures 207

Rat Henry Kistler. The thoracic duct in the rabbit. Six figures 233

Franklin P. Reagan. Some results and possibilities of early embryonic castration. Six figures (four plates) 251

Helen Dean King. The relation of age to fertility in the rat. Three figures. ....... 269

Technique Notes. I. The application of Benda's neuroglia stain. II. Some uses of Mallory's connective tissue stain. By H. M. Kingbry. III. The use of the Van Wijhe method for the staining of the cartilaginous skeleton. By Gustave J. Noback. IV. A convenient method of orientation in paraffin imbedding when paper trays or boxes are used. By B. F. Kingsbury 289

K. Okajima. On the elective staining of the erythrocyte 295

J. B. Johnston. Neutral red as a cell stain for the central nervous system 297

### No. 6. JANUARY

H. H. Donaldson. Biological Problems and the American Association of Anatomists.

Address of the President at the Annual Meeting 299

Proceedings of the American Association of Anatomists, Thirty -third session 311

Proceedings of the American Association of Anatomists. Abstracts 317

Proceedings of the American Association of Anatomists. Demonstrations 439

American Association of Anatomists, Constitution 446

American Association of Anatomists, List of officers and members 449

Proceedings of the American Society of Zoologists, Fourteenth Annual Meeting 467

Proceedings of the American Society of Zoologists, Abstracts 473

American Society of Zoologists, Constitution 543

American Society of Zoologists, By-laws 545

American Society of Zoologists, Historical Review 546

American Society of Zoologists, List of officers and members 570

## A study of the reaction of mesenchyme cells in the tad-pole's tail toward injected oil globules

ELIOT R. CLARK

From the Anatuniical LdJtoralonj uf the University of Missouri

FIVE FIGURES

In their later studies on the mode of development of the lymphatic system, Huntington and McClure have reiterated the view that the hnnphatics are formed by the transformation of mesenchyme cells in the following manner. They hold that fluid accumulates in the tissue spaces, forming small lakelets; that the mesenchyme cells are pressed upon by the fluid collected: and that, as a result of this mechanical pressure, the mesenchyme cells react by flattening out and by forming membranes around the blisters.

The main evidence presented in favor of this view consists in the finding, in microscopic sections, of clear spaces in the tissues, unsurrounded by any membrane, of other spaces which have the appearance of being partly surrounded, and of others with a complete covering. Some of these appear to be completely isolated while others are connected with one another. The numerous possibilities of error in the interpretation of the appearances described have been pointed out by Miss Sabin (1), E. L. Clark (2), and mj^self (3), and will not be reviewed at this time.

The importance laid by these investigators on the part played by the mechanical action of fluid on mesenchyme cells may be seen from the following quotation. Huntington (4) says (page 289):

^ A preliminary report of these studies was published in the Proceedings of the .American Ass. of Anat., Anat. Rec, vol. 10, no. 3, 1916, p. 191.

If two embryonal mesenchymal cells are separated from each other by the accumulation of fluid in the resulting intercellular space, then the opposing aspects of the two cells involved will be subjected to the mechanical and hydrostatic influences of accumulated intercellular fluid, which will react upon the surfaces of the cell still held in syncytial relation to the surrounding mesenchytue. The cells whose opposing surfaces have become freed by the development of an intercellular space, and are subjected to fluid pressure, will react as a whole, become flattened, and be transformed into endothelial cells, forming the parietal limit of an originally intercellular mesenchymal space, which is the font and origin of all vertebrate vascular development.

This should be supplemented by a quotation from McClure (5), for no explanation is given here for the accumulation of fluid in the tissues.

As soon, however, as the haemal vessels begin to function, lymph begins to collect in the intercellular spaces of the embryo and, as we know, is subseciuently collected by a set of newly formed vessels, the l>nnphatics, which convey it to the venous circulation.

Those who maintain that the lymphatics sprout centrifugally and continuously from the veins, would necessarily hold that the lymph in the intercellular spaces patiently awaits the arrival of close and hollow outgrowths from the veins, the lymphatics, before it can be received into any portion of the lymphatic system.

The combined picture, then, is as follows: as soon as the blood-vessels function, "lymph begins to collect in the intercellular spaces." It collects apparently because it cannot get back into the blood-vessels. The mesenchyme cells are subjected to "mechanical and hydrostatic pressure" exerted by the accumulated lymph, and respond by flattening out, and becoming endothelial cells. This comprises the "font and origin of all vertebrate vascular development."

Some of the theoretical objections to this series of assumptions may be pointed out.

First, McClure states that "as soon as the haemal vessels begin to function, lymph begins to collect in the intercellular spaces of the embryo;" to the pressure exerted by this accumulated lymph is assigned the role of acting as the formative stimulus for t lie t r;iiisl()ruuiti(<ii of iiicsoucliyuK' colls into lyiuphatics. Ill allot li(Mi)Iace, McC 'liii('arfi;ues that blood-vessels and lymphatit's differ(Mitiat(> in the same manner. Now if the collection of int(M'c(^llular lyiii])li l)ep;ins only after the blood-vessels be^iii to fiiiictioii, how are we to exi)lain the collections of fluid wJiich ai-e su])])osed to have served as a fonnative stimulus for the (liflereutiatioii of blood-vessels? Afi;ain, if both bloodvessels and lymi)liatics ai'e determined solely by the action of mechanical pressure, we are left with the same puzzling problem which confronted Goette (6) in 1873, namely, wliy the two sets of vessels do not everywhere form comnumications with one anotlu^r. The ])uzzle ditfers only in that (Joette conceived of the circulation fi'om blood to lymph capillary as beinj;- intracellular, and tried to explain why the same mesenchyme cell was not sometimes demanded by both the blood-vessel and the lymphatic, while Huntington and McClure conceive of the fluid as being extra-cellular and must explain why it does not happen that the same lakelet does not connect with a blood capillary on the one side, and with a lymphatic capillary on the other. Goette's puzzle has been eliminated by the proof that in the tail of amphibian larvae, mesenchyme cells take no part in the growth of blood or lymphatic vessels.

Again, it is hardly necessary that "those who maintain that the lymphatic sprout centrifugally and continuously from the veins, would necessarily hold that the lymph in the intercellular spaces patiently awaits the arrival of closed and hollow outgrowths from the veins, the lymphatics," since it was demonstrated clearly by Magendie, over a hundred years ago, and has been proven so many times since that it cannot be questioned, that absorption of substances may take place through the bloodvessels as well as through the lymphatics.

Another assumption which is quite unwarranted on the basis of facts is that with an increase in intercellular fluid, this fluid will collect in definite lakelets, in the intercellular spaces. All our knowledge of intercellular spaces indicates that they form an irregular, intercommunicating net-work of spaces filled with fluid. It would be expected that, if this fluid were increased, the increase in the intercellular spaces would be a general one, the distention, or edema, being regulated in its intensity only by the relative amount of resistance offered by the tissue. This resistance would, of course, be greater in dense tissues, in which the cells are bound firmly to one another, and less in tissues where cells are less firmly attached to one another. In any region, however, where the tissue is uniform the pressure and separation of cells must be uniform, and no accumulation of tissue fluid in the form of lakelets would seem to be possible. Microscopic examination of embryos bears out this theoretical consideration; there are regions in which the intercellular fluid is large in amount, proportionately, and others where it is small. "VMiere large, the mesenchyme or other cells are uniformly separated from one another, as for example in the umbilical cord, and the ventral body wall.

It is unjustifiable from our knowledge, or better, lack of knowledge, to speak positively about the stimuli which are responsible for the primary differentiation of organs or tissues, since with, perhaps, the exception of the formation of the lens as the result of the contact of the optic cup with the epidermis (7), there is hardly an instance in embryology of the satisfactory demonstration of the stimulus responsible for the primary differentiation of any organ or tissue (8). In this connection, results obtained in recent studies made on tissues, grown by the 'tissue culture' method of Harrison, are interesting. Harrison (9) found that primitive nerve cells send out processes into a medium consisting of coagulated lymph. The Lewises (10) obtained similar results with sympathetic nerve cells, in a medium of Locke's solution. Shipley (11) has found that undifferentiated heart muscle cells differentiate and start rhythmic contractions, in a medium of coagulated plasma. In all of these cases the cells were removed from their normal environment. Their continued development makes one sceptical of hypotheses as to the nature of formative stimuli, when such hypotheses are not supported by fact.

The only experimental evidence which has been proposed in support of the hypothesis that collections of lymph furnish

MESENCHYME TELLS IN TAD-POLE S 'I'AIL O

the sliiiiuhis to the lonnatioii of lyin])hati('s, consists of tiie results of studies oil 'ex])('i'iiii('iital niosotlieliiini,' l)y W. C. Clurk (12). This iiivestigiitor found tlint if solid blocks of celloidin are ])la('ed in tlie suheutaneous tissue of dogs, they became surrounded i)y flattened cells, which show, when treated with silver salts, black intei'celhdar lines tj^pical of flattened mesotlielial or (>])ithelial tissues. Solid globules of hard ])araflPin, injected into the cornea of rabbits, were surrounded l)y a layer of flattened cells. Flattened cells were also found to line 'dead spaces' in the tissue and artificial channels, such as may be induced by the ligation of the cystic duct, with formation of a mucous fistula.

Interesting as are these studies, in spite of the absence of evidence as to the source of origin of the cells which formed the flattened lining membrane, there appears to be no justification for the conclusion that we have thereby gained any information as to the mode of differentiation or growth of blood or lymphatic vessels. And 3'et, W. C. Clark concludes (page 316) that "Therefore the second hypothesis, premised in this article, is tenable, namely that the flat cells of serous surfaces and those lining blood vessels may regenerate from deep connective tissue cells, and do not necessarily arise from adjacent intact mesothelial or endothelial cells." Again ^McClure (13) refers to the results of W. C. Clark as bearing out in a most decisive manner" the view that "the gradual increase in the amount of lymph received by the subocular sacs (in the trout) during the stage of their independence, results in the application of a constant and continuous pressure to the mesenchyme cells forming their walls, which in itself must be a positive factor in causing these cells to flatten out and gradually assume an endothelial form."

It is difficult to conceive how the results obtained by W. C. Clark can have anj^ bearing on the problem which confronts McClure, unless the assumption is made that all mesothelial, endothelial, and epithelial membranes which line spaces or ducts have the same properties — an assumption for which facts furnish no justification. Surely blood-vessel endothelium has

6 ELIOT R. CLARK

properties which differ from synovial membranes, peritoneal membranes, or the lining membrane of the urethra, the bile ducts, or the gall-bladder.

There is, perhaps, a suggestion of support for the hypothesis in question in some of the results obtained in tissue cultures. Several observers — Harrison (14), M. R. and W. H. Lewis (15), Lambert (16), W. C. Clark (17) and others — have found that in bits of tissues, explanted to plasma or Locke's solution, membranes may be formed around solid bodies — such as along the cover-slip and around solid threads, as the threads of spider web, used by Harrison, and around droplets of fluid — such as may be formed, occasionally, in plasma preparations, by the retraction of the fibrous threads. The explanation of the formation of membranes around droplets of fluid is associated with the apparent inability of cells to grow into a purely fluid medium, without mechanical support, or as expressed by Harrison, their dependence on 'stereotropism.' The formation of such a membrane in tissue cultures is not to be interpreted as a reaction by flattening out, on the part of the culture cells, but rather, in all probability, as due to the fact that cells grow around the periphery of such a droplet. Moreover, it has not yet been possible to determine the origin of the cells which form membranes in tissue cultures. At present there appears to be no ground for claiming that the formation of membranes in this manner, furnishes us any information as to the mode of differentiation of blood-vessel or lymphatic endothelium.

In order to put to the test the hypothesis that the differentiation of blood or lymph vessel endothelium may be stimulated by the mechanical pressure exerted on mesenchyme cells by accumulations of fluid, and to plan the test in such a way as to make it approach as nearly as possible the actual conditions supposed to exist, the present studies were started. The aim was to inject an inert fluid, in globules of size sufficient to press against the mesenchyme cells, into a region of embryonic tissue, in which the reaction of the cells could be watched in the living animal; to see whether a membrane were formed, and, if so, by what type of cell, and what would be the properties of such a

MESENC'in.Mi; CELLS IN TAD-POLES TAIL 7

monihrano, osperially its roaction towanl hlood-vosscl and lyin])liati(' (Mulothcliuin.

In ordci- to simulate* as nearly as ])()ssU)l(' tiic (liiid wliose })resen('e is thought to excite the traiisfoi'niatioii of inesenchyme cells into lyin])hatics, and at the same thne to have a fluid which would he inert, wliicli would not be absorbed, which w(.uld n.erely exert a iniH'hanical pressure, paraffin oil was selected for injection. The ol)ject chosen was the trans])arent fin expansion of the tail of youn^" fi'oji and toad tad-poles wliere it is possible to see the indi\idual mesenchyme cells, as well as blood-vessels, lym]ihatics and leucocytes, and to watch their reactions in the living- larvae, from qIsly to day. The tad-poles were anaesthetized with chloretone (1 : 4000 to 1 : 5000) and small globules of oil injected into both fins, through fine glass cannulae, under the binocular microscope. The oil was sterilized by heating. In some cases the tadpole was washed in several changes of sterile w^ater, but the results did not difTer materially from those obtained when the only antiseptic precaution consisted in sterilizing the oil. The observations were made by a method previously described in detail (18) — the larva, anesthetized, was placed in a micro-aciuariiun, in chloretone of the proper strength, and the tube of the microscope tilted to the horizontal, to enable the tadpole to retain its normal upright position. The oil globules w^ere of varying sizes, from 20 to 100 micra in diameter.

The larvae used were those of Rana catesbiana (bull-frog) and of Fowler's toad (Bufo lentiginosus Fowleri). The latter have beautifully clear tails, with very few pigment cells at the stages used, while the mesenchyme cells are far apart and stand out most clearly.

Since the oil could not be injected without a certain amount of injury, there were alwaj^s some temporary effects of the injection, not attributable to the presence of the oil. These consisted principalh' of a more or less intense leucocytosis, probably caused by greater or milder degrees of infection. In some cases large numbers of leucocytes gathered about the globules, many of the leucocytes containing pigment. Several of the globules, around which the leucocytosis was most intense

ELIOT R. CLARK

.mi:si;n(Iiv.mi: cklls in tad-i'olk's tail 9

wore oxtnidcd. In raso tliis did not occur, tho loiicocytosis firadiially subsided luitil in most cases tlu'cc or four days after the injection, the i-ej2;ioii surrouii(lin},j; tlie globule contained no more leucocyt(>s than the other ])arts of tlie tail. In some mstances, a sH^litly increased number of leucocytes near the p;lobul(^ continued for several days longer. From now on the globules remained ai)i)arently inert, so far as could be judged from the beluuior of leucocytes in their vicinity. The longest time over which a globule was watched was 12 days.

In a])])(\ii'ance the oil globules, wh(>n "[iresent in the tail, form sphei'es with the central portion clear and transparent and a dark periphery, with a sharp outline. Structures over the central portion can be seen most clearly. Thus it is possible, in case the diameter of the globule is sufficient to distend the skin slightly, to see the nuclei of the cells of the epidermis, and the details of other structures most distinctly. Many of the globules w^ere oval in shape, immediately after injection. Later, after a day or tw^o, they usually rounded up to a spherical shape though sometimes remaining slightly oval.

The beha^'ior of the mesench\nne cells will now be described. In order to follow them with accuracy camera lucida records were made of all mesenchyme cells in the neighborhood of the globules, and their changes from day to day were noted. When a mesenchyme cell happened to be in the outer dark zone of the globule, it was difficult to make out its outlines. Occasionally only one or two of its processes could be clearly seen. The following of such cells, however, was made possible by the fact

Fig. 1 From larva of Fowler's toad. 9 nun. long. Four globules of paraffin oil injected into fin expansion of tail, on Aug. 19. .Much leucocytosis about three of them, and all three extruded within forty-eight hours". There was very little leucocytosis aljout the fourth globule, of which three drawings are shown, two, four, and six daj-s after injection. This globule was in the ventral fin. The mesenchj-me cells in the immediate vicinity of the globule are shown. The letters, a, b, etc., indicate the same cells. * leucocytes against the globule; pig.L., pigmented leucocyte against the globule. The mesenchyme cells were in three different planes; those nearest the observer are represented in solid black, those furthest away are dotted, while those in the midst are cross-hatched. Enlarged 267 times. Drawn with camera lucida.

10 ELIOT R. CLARK

that the globules from time to time shifted their position slightly, so that a cell, at one time not clearly seen, later could be clearly outlined. This applied to only a very few cells, particularly in the toad larvae, because of the relative rarity of their mesenchyme cells. In view of the descriptions in the literature of the reaction of connective tissue cells to the pressure exerted by foreign substances, the behavior of the mesenchyme cells was a great surprise. It was expected that the cells near the globule would flatten out on its surface and form a membrane. On the contrary, the mesenchyme cells apparently paid no particular attention to the globules. They maintained their identity as 'star-shaped' cells, with thickened central portion and branched processes, and their property of slow progression, described in an earlier paper. That the mesenchyme cells are not influenced by the pressure exerted by the globules was brought out quite strikingly in one instance in which the globule shifted its position in such a way as to come to lie against a mesenchyme cell which had been at a slight distance from the globule. For a day or two it was rather difficult to make out the outlines of the cell. The globule then shifted its position in the opposite direction, and the mesenchyme cell could now be seen clearly, apparently unchanged, at a slight distance from the globule. Occasionally there are to be seen, over the clear part of the globule, if the globule is of sufficient size to distend the skin slightly, one or two flattened cells which, at first glance, might be interpreted as cells flattened out by the pressure of the globule. Such cells were seen over only a few of the globules, and their explanation was obvious on studying other parts of the tail, at a distance from the globules. Such flattened out mesenchyme cells appear more or ess evenly distributed, lying just below the epidermis, and are not particularly associated with the oil globules. That they are associated with the skin and not with the oil globules, is shown by the fact that none are present over the majority of the globules, and also by the fact that, if the globules shift, these cells maintain the same position with reference to the skin, while they are left behind by the oil globules (fig. 4). There are no other appearances on

MESENCllVMK CKLLS IX TAI)-1'()LK S TAIL

11

Aug. 4

Fig. 2 From larva of Rana catesbiana, 10.5 mm. long. Two globules of paraffin oil injected into the ventral, and a small globule into the dorsal fin; all three remained. ]\Iuch leucocytosis around each. The first drawing (Aug. 3) was made immediatel}- after the injection. Small mass of cellular debris is shown, in the path of the injection. Mesenchyme nearest the globule lettered as in fig. 1. In drawings Aug. 4, 5, and 7, some of the leucocytes about the globule are shown. In drawing of Aug. 5 are shown the successive positions taken by a leucocyte as it moved to the globule, moved along the surface of the globule a short distance, and then moved away. The shifting of the globule is clearh" seen, by comparing its relation to the blood vessel. The cells fZ and e, which lie close to the globule Aug. 3 and 4, are left behind, while the cells b and c are approached by the shifting of the globule.

Enlarged 267 times. Drawn with camera lucida.

12 ELIOT R. CLARK

the part of the mesenchyme cells which even remotely suggested a flattening out. The 'act that some of the globules watched shifted their position would also indicate that no surrounding membrane had beeii formed, for a membrane would prevent such movement.

The behavior of wandering cells towards the globules was watched with interest. As already stated there was a leucocytosis of greater or less intensity following the introduction of the globules, for the first two or three days. Leucocytes, most of them small and clear, others larger, and containing pigment, collected around the globules. Many of them flattened themselves out on the surface of the globule, or formed irregular humps on the profile. Occasionally such a flattened leucocyte formed a thin, circular structure, with nucleus visible over the clearest, central portion of the globule. Such a cell, coupled with the irregular humps on the profile, if seen only at one stage, and not followed, might well give the impression of membrane formation by wandering cells. When, however, such cells were followed, it was seen that they gradually moved away (fig. 2). The humps on the profile changed shape, with each drawing, even when the records were made several times daily, while the flattened cells on the clearer part moved away. After four or five days most of the globules were quite free from the presence of leucocytes. In order to be sure of the behavior of the wandering cells, some were watched intensively. They were seen to move, w th the typical amoeboid type of progression, up to the oil globule, to flatten themselves out on its surface and again move away. The impression was gained, that, had sections been made at a time when the leucocytes were flattened on the globule, they might have been interpreted as forming a membrane, an interpretation which the study of the living shows would be quite unjustifiable.

Chromatophores which are present in large numbers in the dorsal fins of bull-frog larvae and to a somewhat less extent in the ventral fins of the same larvae, sometimes wrapped around the globules with their long branched processes, when the glob

MESENCHYME CELLS IN TAD-POLE's TAIL 13

ulcs wvw iiij(>('t(Ml Ileal' llicm, hut did iiol I'onii ;i dcfinito UKMiibruiie.

Since the inosonchyiiio colls fnilod to form a ineinbnme around the ^l()i)ulcs, the second ]xirt of tlu^ inciuiry, namely, the reaction of such a membrane, if foinuMl, toward the lymphatics or bloodvessels in their vicinity, could not be followed. It was of interest, however, to observe the reaction of blood-vessel and lymphatic endothelium to the oil globules. In one case a particularly faA'oi'able o])])ortiuiity was afforded to study the reaction of the bl()od-ca])illaiy, since the sl<^^^ule pressed against a blood-capillary, forcing it to make a bend in its course (fig. 4).

Fig. 3 From same Rana catesbiana larva from which figure 2 was taken. To show relation of pigment cells to globule. This small oil globule was injected into the dorsal fin, on Aug. 3; the drawing was made Aug. 5.

Enlarged 267 times. Camera lucida.

The capillary then appeared to wrap around a part of the globule. In this position the capillary showed no tendency to give ofT cells which might grow around the globule, but instead remained as a distinct vessel. The circulation of blood cells through it, which was at first interrupted, was later resumed. Blood-capillaries and lymphatics near the oil globules showed no tendency to grow toward it, or to send out sprouts to it.

The results of this study, then, indicate that, aside from the temporaiy inflammatory reaction, due probably to the injury and to the bacteria introduced at the time of injection, the presence in the fin of the tad-pole's tail of injected globules of paraffin oil, of sufficient size to cause a distension of the tissues, fails

14

ELIOT R. CLARK

to stimulate the formation of membranes about the globules, on the part of mesenchyme cells, wandering cells, or of bloodvessel or lymphatic endothelium.

These results are in disagreement with those of W. C. Clark (19), already mentioned. They are, however, in agreement with the older findings of E. Juckuff (20) who found that soft paraffin, injected subcutaneously, travelled long distances from the injection site, showing that no membrane was formed about

Aug. 20

Fig. 4 Globule of paraffin oil injected into ventral fin of tail of Rana catesl)iana larva on Aug. 13. The globule rested against a blood-capillary, forcing it to bend slight!}'. The two sketches shown, made four and seven days respectively after the injection, show the relation of the globule to the blood capillary, to a nearljy lymphatic, {lijm) and to two mesenchyme cells which happened to be under the epidermis immediately over the globule. Note that the globule has shifted to the right, so that the two mesenchyme cells, which on Aug. 17 are over the right part of the glol)ule, are over the left central portion Aug. 20. Other mesenchyme cells not shown.

Enlarged 267 times. Drawn with camera lucida.

Fig. 5 Same larva as figure 4. Small globule injected in dorsal fin on Aug. 13. Drawing made nine days later — on Aug. 22, d.c, pigment cell. Enlarged 267 times. Drawn with camera lucida.

MKSHXCIIVMK CKM.rt IN TAD-I'OLe's TAIL 15

it — a liinliii^ 'i-iru'(l hy M;ic( 'alhiiii CJI). Il will he rciiiciiiboretl tluit the uttoin])! was made, in selecting tJ»e substance to be injected, to find sonietiiinfi; which would sinuilate tlie supposed lak(>lets of tissue fhiid wliose presence are lield by Huntinji;ton and Ab'("hire to stimulate, mei-ely l)y the mechanical ])r(;ssure whicli tliey are su])])osed to exert, the mesencliyme cells to form membranes. It is oljvious that in point of size and consistency the small globules of oil nmch more nearly reproduce the supposed conditions than the relatively enormous pus ])ockets, or tlie relatively huge solid blocks of celloidin or the globules of hard ])arafHn. It is also obvious that in a transparent object, like the tad-pole's tail, where the individual cells may be seen with great clearness and the reaction process watched in the living animal, the conditions for observing what happens are much more favorable than in the case of the other experiments referred to. Since, then, the action of pressure alone fails to stimulate the formation of membranes on the part of mesenchyme cells, an important link in the argument used in favor of the origin of lymphatics from mesenchyme cells, as presented by Huntington and McClure, drops out. A certain feeling of disappointment must be confessed, that the mesenchyme cells failed to respond to the presence of the oil globules by the formation around them of membranes, as it was expected they would, because of the desire to see w^hat would be the reaction of such membranes toward blood-vessels and lymphatics. It is true that cells derived from the middle layer or mesoderm differentiate at various stages into pavement epithelium, or endothelium, other than that which lines the blood and l^^llph vascular systems. Among such may be mentioned the lining of the large cavities pleural, peritoneal, and pericardial, the lining of bursae and sj^novial membranes, and the outer layers of tendons and fasciae. It should also be remembered that the same middle embryonic layer differentiates into smooth and striated muscle, into cartilage and bone, into blood cells, and other types of tissues. Each of these tissues has certain 16 ELIOT R. CLARK modes of reaction, a specific life history, the property of responding each in its own individual way, to various stimuli. To transfer the modes of reaction of one set of, tissues derived from the mesoderm to another set is quite unjustifiable. To be more specific, it is not justifiable to claim that, if connective tissues, in adult animals, are capable of forming membranes about solid foreign bodies, or large accumulations of fluid, or if membranes form around liquid vesicles in the midst of coagulated lymph in tissue culture preparations, then lymphatics arise as the result of the pressure exerted by accumulated lakelets of Ij^mph. It is even unjustifiable to transfer to lymphatics the properties of structures which resemble them morphologically so nearly as blood-vessels, for, while there are many points of similarity between the modes of reaction of the two, there are also striking differences. It is conceivable that future studies may reveal the various stimuli which are responsible for the primary differentiation of tissues and organs. For the lymphatic endothelium it would seem a more hopeful field to investigate the chemical nature of the intercellular fluid, to see whether any evidence can be gained as to the collection there of especial chemical substances which stimulate its differentiation. To propose such an hypothesis at the present time, however, would be pure speculation. Much confusion has arisen because quite different structures have been grouped together under the name of endothelium or mesothelium. It would seem that the time is ripe to separate these different forms of flattened lining cells under different names. If, for example, we could speak of blood-vessel endothelium as Haem-angiothelium, and of lymphatic endothelium as Lymph-angiothelium , or some equally specific names, and if distinctive names could be selected for the other forms of pavement epithelium, much of the confusion would disappear. In conclusion, it is a pleasure to express my gratitude to the Marine Biological Laboratory at Wood's Hole, where these studies were made, for generously granted laboratory facilities. MKSKXCIIYMK CELLS LV TAI)-1'0LE'.S TAIL 17 liii:ka'iii{|': ciii;!) (1) Sauin, F. R. lOL'J JoliMs Hopkins llosp. U.-p,)ils. Mononiaphs. New Series, no. 5. (2) Clark, E. L. 11)12 Anal. Wi-c, vol. t>. (3) Clauk, K. R. 1911 -Vnal. I{cc., vol. f.. (4) Huntington, C. S. HU I .\iii. .lour. .\iial., vol. 10. (5) McCluue, C. F. \V. I'.llf) Aiiat. ]{rc., vol. <J, no. 4, p. 281. (6) CiOKTTE 1875 Die KntAvickclunfisfrcschichte der Unke. Leipzig. (7) Lkwis, \V. H. 1907 Am. Jour. .\iia(., vol C, p. 473. 1913 Storkard Am. Jour. .Vii.it., vol. !.'>, p. 253. 1910 vol. 10, pp. 393-423. (8) IIkrbst 1901 Fornuitive Reize in der 'I'hierischen Ontogcnese, Leipzig. (9) Hahrison, a. G. 1910 Jour. Exp. Zoo!., vol 9. (10) Lewis and Lewis 1912 .\nat. Rcc, vol. 6, j). 7. (11) Shipley, P. G. 1910 .Vnat. Rec, vol. 10, p. 347. (12) Clark, \V. C. 1914 Anat. Rec, vol. 8. 1910 Anat. Rec. vol. 10. (13) .McCluhe, C. F. W. 1915 Mem. of Wistar Inst., no. 4, p. 29. (14) Harrison, R. G. 1910 Jour. E.xp. Zool., vol. 9. 1911 Science, vol. 34. 1914 Jour. Exp. Zool., vol. 17. (15) Lewis, M. R. and W. H. 1911 Anat. Rec, vol. 5. (16) Lambert, R. A. 1912 Anat. Rec, vol. 6. (17) Clark, W. C. 1910 loc cit., p. 313. (18) Clark, E. R. 1912 Am. Jour. Anat., vol. 13. (19) Clark, W. C. 1910 loc cit. (20) JucKUFF, E. 1893 Archiv. f. Pathol, u. Physiol., Bd. 32, p. 124. (21) Mac C.a.lltjm, W. G. 1903 Johns Hopkins Hospital Bulletin vol. 14. pp. 5 and 0. THE ANATOMIC.M, RECORD, VOL. 11, NO. 1 AX AXO.MALors rHIXOGENITAL SYSTEM LX A DOC i>()ris II. K()K\i)i:i{ From the Ano!o>/iical Lnboralori/ of the A'urthwcslcrn Unu'crsilij Medical School^ TWO FIGl'KES Anomalies of the iiriiiogenital system are frequent and have ceased to attract nmch attention. Few, however, present features of such embryological interest as the following case. Because of this and on account of its value, as illustrating the physiological adaptability of one system to the needs of another, this case seems worthy of mention, ^ly acknowledgment is due Dr. L. B. Arey for valuable suggestions regarding the embryological considerations. On opening the abdomen of a dog it is commonly observed that the bladder is large and lies almost entirely in the abdominal cavity. In a medium sized mature female, selected at random for the purpose of obtaining certain tissues used in a research problem, the bladder lay deep in the pelvic cavity and was rather small, the size and shape being that of a walnut. On palpation it felt extremely firm, much as though it were a solid mass of tissue. A longitudinal incision through its wall revealed but a very small lumen, less than 1 cm. in diameter and 2 cm. in length. Two broad ligaments passed from this bladder over the rectum and gained attachment to the front of the sacrum. One ligament was considerably longer than the other, due to the bladder lying ventral and to the left of the uterus instead of directly ventral as is normally the case. Except for these two ligaments and a slightly shortened urethra which merged into the left wall of the urinogenital sinus, other connections with the bladder could not be established. 1 Contribution Xo. 40, :May 15, 191(3. 19 20 LOUIS H. KORNDER These findings led to an examination of the kidneys, which were found to be normal in shape, size and position. Each possessed one short ureter, the right being 6.5 cm. and the left 7.2 cm. in length. Originating in the pelvis of the kidneys the ureters coursed downward over the psoas muscles and passed one on each side into the horns of the bicornuate uterus. This union occurred about a centimeter below the place where the short Fallopian tubes merge into the uterine horns. The uterine horns and the uterus were not soft and pliable as is usual but were hard and rigid and on making a longitudinal incision through their walls, were found to be filled with debris, composed mainly of desquamated epithelial cells. The structures in this region were surrounded and some deeply imbedded in a mass of fibrous and adipose tissue. This appeared to form a common capsule which covered the union of the ureters with the uterine horns and extended over the Fallopian tubes including the ovaries, becoming at this connection part of the ovarian bursa. The ovarian bursa exists normalh' in the dog as a separate fold of peritoneum covering each ovary. This is usually covered by adipose tissue but opens through a small slit-like opening into the abdominal cavity. It has been mentioned that the urethra was sUghtly shorter than normal and passed from the left into the wall of the urinegenital sinus. This relation of the urethra to the urinogenital sinus and the original location of the bladder explains why in the accompanying figure (fig. 1) the bladder is shown as lying between the uterus and rectum, instead of ventral to the uterus as is normal. Histological preparations of the bladder show a slightly changed epithelial lining, consisting in two to three layers of low cuboidal epithelium. The uterine surface epithelium instead of being high columnar in type is pseudo-stratified. The deeper glandular epithelium, however, is the same as in a normal dog's uterus. The ovaries which were deeply imbedded in their ovarian bursae on sectioning showed nothing atypical, with the exception of ANO.MAI.ors IKIXOdKM TAL SYSTEM IN A DOG 21 ureter \ XjOvarian bursa. entranceop ureter liter ine hornopened opt . uterus , Lateral/ tiaaynent- y of the Ucbdder, .^ bUdde.r..jjf- -^j|^\ cervix.. ■ uretkrou. recuum ji_.jr__»a_ vcuamcL. i* |fl^ ^^ IT Fig. 1 \'entral view. Ureters shown as they enter both horns of uterus. Ovarian bursae and upper end of uterine horns opened. Bladder small and abnormal in location. 22 LOUIS H. KORNDER a slightly more fibrous stroma than is usual. Several large Graffian follicles present indicated a normal functional activity of the ovaries. Histologically, then, practically nothing unusual exists, the anomaly being one of gross anatomy, this consisting in a union of the urinary with the reproductive tract, the fusion of ureters and uterine horns leaving the bladder as a cul de sac which leads through the urethra into the urinogenital sinus. E.AIBRYOLOGICAL CONSIDERATIONS The structures in^'olved here are embryological derivatives of the mesonephric ducts, metanephros, jNIuellerian ducts and cloaca. That the cloaca developed normally is indicated by the presence of a rectum, bladder, urethra and urinogenital sinus. The presence of the uterus, tubes and vagina likewise indicate the normal development of the Muellerian ducts. The anomaly then must be due to a defective embryological growth of the mesonephric ducts and their derivatives, with a deficiency in the development of the nephrogenic cord as a possible causal stimulus. In pig embryos of approximately 5 mm. length the mesonephric ducts give rise to the ureteric anlage of the metanephros where the ducts bend to join the cloaca. But that the ureteric anlages do not always originate at this bend is indicated by the frequency of double or triple ureters. In these instances the first ureteric bend develops usually into the ureter most normal, while the rest show evidence of slowed or mal-development. From these cases it may be assumed that the ureteric anlage need not necessarily arise at a definite location but can occur at any point along the ducts. Figure 2, a very diagrammatic sketch, shows both the Muellerian and Wolffian ducts leading into the urinogenital sinus. The approximate position where normally a single metanephric anlage arises from the Wolffian duct is indicated by (A). However, in man as many as six such anlages have been observed. While in these instances the anlage corresponding to (A) de^'elops into the adult ureter the possibility exists that a more ANOMALors ri{INOGENITAL SVSTKM IN A l)()(i 23 cranial aiilajio. Toi- iiistaii('(> (/>'), may hccdiiic llic functional urct(>r. Kct'crcncc to figure 2 will show tlic lowci- i)art of the Wolffian duct not cross-hatched. Tiiis ])()rtion which extends from the normal ureteric anlaj>;e on downwards is durijij;' further development drawn into the urin()f2;enital sinus. Thi'ouf^h this fusion the uieteis ivceive theii' normal connection with tJie definitive bladder which develo])s ])ai1ly out of this portion of the sinus. Fig. 2 Diagrammatic sketch. M, Muellerian duct; W, Wolffian duct; A, location of normal metanephric anlage; B, possible upwardly displaced metanephric anlage; D, extended ureter merging into the Muellerian duct at [/; N, portion of Wolffian duct taken into wall of urogenital sinus; UgS., cross-hatched portion of Wolffian duct degenerates. If in this particular case, however, the ureteric anlage did not develop low enough to be included in that lower portion of the mesonephric duct then just as soon as the normal degeneration of the upper part of the Wolffian duct occurred, the upwardly displaced anlage (B) which developed into the ureter was without connection with the urinogenital sinus. Being thus isolated it seems probable that the ureter (D) extended to the nearby Muellerian duct and merged into it at (U). This established an outlet into the urinogenital sinus. The above is ofTered as one possibility to which the present anomaly may be due. It is entirely hypothetical as any con 24 LOUIS H. KORNDER sideration of this case must be. Because of this a further explanation may possibly be found in the following. In addition to those known embryological facts mentioned above it should be recalled that in embryos of 8 to 11 mm. length the Muellerian ducts develop caudalward beneath the epithelium of the mesonephric fold. Reference to dissections of the pig embryo show the Muellerian ducts lying very close to the Wolffian ducts so that the occurrence of a more or less complete longitudinal fusion of these two ducts seems not impossible. The establishment of this anomalous union of ureters and uterine horns presumably occurred early in the development of the animal. Since in the female, during the normal development the mesonephric ducts degenerate and disappear almost entirely, it may be assumed that in this case these ducts gradually fused with the Muellerian ducts. It may be that in this way an early connection occurred on either side between the Muellerian duct and an upwardly displaced ureter. The question may be raised why a metanephros thus displaced should have abandoned the mesonephric duct and appropriated a new outlet by way of the Muellerian ducts? It may be that the functional need of maintaining the patency of the mesonephric duct was ineffectual compared with the tendency toward atrophy and consequent occlusion which the cranial portion normally shows. This query becomes all the more pertinent in view of the recent report by Bremer (Jour. Anat., vol. 19, '16) that in the cat the mesonephros maintains its activity until the permanent kidney assumes the excretory function. This being true of the cat it is more than probable that it also exists in the dog since both belong to the Carnivora. SENSORY ELEMENTS IN THE HUMAN CEREBRAL HYPOPHYSIS w. S()hii:r hhvaxt ONE FIGURE 111 the i)ast few years, the ^rt^ater i)art of the work relating to the cerebral hy])()])]i>'sis has been of a therapeutic, a clinical or a surgically experimental nature, and the interest aroused in these aspects of the pituitary has tended to obscure the fact that certain histological elements in the structure of the organ still remain a mystery. The following report re-introduces the subject of the sensoiy elements of the hypophysial cavity, of which I have made a careful examination in human specimens : These sensory elements occur in maculae, which, in sagittal sections of the pituitary are seen situated on the posterior wall of the cavity, and sometimes, apparently, on the anterior wall. The maculae are composed of tall columnar ciliated sensory cells interspersed with bipolar cells, which have their nuclei towards the periphery; whereas in the ciliated cells, the nuclei are near the base which terminates in a caudal prolongation. Between these caudal processes of the ciliated cells, there is a layer of round cells, resting on a thin basement membrane. An area of ciliated cuboidal cells occurs at the margins of the maculae. I have found these sensory cells in all the freshly hardened human hypophyses that I have examined, except those in which the parenchyma had been almost completely replaced by connective-tissue; the sensory cells are moreover encountered even in pituitaries which have undergone very extensive pathological change. In their gross arrangement, the sensory elements of the hypophyseal cavity are suggestive of the sensory elements of the maculae acousticae. Gentes (3), in his examination of the hypophysis of cats and dogs, found in the juxta-nervous layer, a stratified cylinder 25 26 W. SOHIER BRYANT Fig. 1 Sensory epithelium of the hypophysial cavity of a human adult. Stained with hematoxylin-eosin. This material was procured through the kindness of Dr. William Mabon and the assistance of Dx. Clarence O. Cheney, Pathologist of the Manhattan State Hospital; the work was done in the New York Psychiatric Institute. Special thanks are due Dr. Charles Bates Dunlap for his technical assistance and supervision of the work. SENsoKv i:lk.mi;n rs i.\ ckukiucai, ii vi-onivsis 2/ opitliclimii i-('S('ml)liii (lay and ni^ht. hut much more commonly at night or, according to Long and Mark ('1 1), in tlu^ early morning. 1 day })()st-/>(tiii(ni. 11" the female white mouse comes in heat, as is the rule during the warmer part of the year (April to October) and the new born j'oung are not suckled, within twenty-four hours after parturition ovulation and pairing (provided a male is present) will occur. 2 days post-partum. The eggs have been fertilized in the ui)per third of the Fallopian tubes and the first cleavage has occurred. 3 days post-partum. The eggs are still in the two-cell stage. 4 days post-partimi. Cleavage is again in progress and morulas of 8 to 10 blastomeres are found at this time. 5 days post-partum. Morulas of 12 to 16 blastomeres. At the close of this day the morulas develop a central cavity, thus becoming blastulas, and at the same time they pass from the Fallopian tubes into the horns of the uterus. 6 days post-partum. Blastodermic vesicles lie free in the horns of the uterus. 7 days post-partum. The blastodermic vesicles are now implanted in proliferated masses of uterine cells which completely obstruct the lumen. The embryos themselves are in the 'eggcylinder' stage, the 'cylinder' almost filling the vesicle and possessing a single, undivided cavity. 8 days post-partum. The egg-cylinder in embryos of this age has its lumen divided into three cavities. 9 days post-partuin. The embryo now possesses a medullary groove, which is open except at the extreme anterior end. 10 days post-partum. Embryos of this age have the medullary groove closed in its anterior half, and for the first time show a heart. 11 days post-partum. The medullary groove is now closed for more than half its length; the optic vesicles are budding off from the brain; and the auditory vesicles appear as cup-shaped depressions in the ectoderm. 12 days post-partum. The medullary groove has closed except at the extreme posterior end; the auditory vesicles are almost or 34 WILLIAM B. KIRKHAM entirely closed. The fore limb buds are present, together with the first nephric tubules and the anlage of the liver. 13 days post-partum. At this age the embryo is a decidedly complex organism, and from this time on the daily changes are rather matters of detail than the appearance of entirely new structures. The characteristic features of this particular stage of development are these: a well developed cranial flexure; optic vesicles completely separated from the brain, invaginated, and showing the beginnings of lens formation. The nasal capsules are visible; the liver has developed into a distinct organ; hind limb buds are present. The embryo has the anlagen of the lungs, and a few pancreatic tubules; the auditory vesicles have withdrawn from the surface and are connected by nerve fibers with the brain. Along the free border of the kidneys, especially at the posterior end appears the genital ridge with a few large cells with large round nuclei, the primordial germ cells, scattered through a much larger number of smaller, epithelial cells. 14 days post-partum. The eyes have developed to the stage where the lenses have a sohd, clear core. Dense masses of connective tissue foreshadow the future location of the bones of the limbs, girdles, and ribs. Whisker follicles are present; also the semi-circular canals of the ears. The kidneys possess definite boundaries. The nuclei of the red blood corpuscles are smaller and stain less deeply than in earlier stages, while their cytoplasm shows a faint indication of haemoglobin. The first indications of teeth follicles are found in embryos of this age; also the anlagen of the thymus and thyroid glands. The gonads differ from the preceding stage merely in having more of the primordial germ cells in the genital ridges. 15 days post-partum. Embryonic cartilage cells constitute the most striking characteristic of this stage, clearly differentiating embryos of this age from all younger specimens. Other features are the fewer and smaller blood spaces in the liver, as compared with fourteen-day embryos; the deeper straw yellow color in the cytoplasm of the red blood corpuscles, together with a few which are non-nucleated; the well-developed anlagen of GESTATION PERIOD IN SUCKLING MICE 35 the tooth; aiul tlio prosonce of cartilage cells in the floor of the crauiiiin. Sexual difforoiitiation is present in (embryos fifteen days post-partum, male specimens showing gonads in which follicle formation has already started, wliile the female gonads preserve the earlier condition of prunordial germ cells scattered through a mass of epithoiial colls. 16 days -post-partum. Differential characteristics now become still more matters of detail and of direct comparison with earlier stages, however, embryos of this age differ from all younger ones in having a decidedly transparent cornea. The heart has assumed its final shape. The pancreas is a clearly defined organ. The nasal capsules open into the front part of the mouth, while the naso-pharynx is connected with both the nasal capsules and the back part of the mouth. The gonads show no marked change from those of fifteen days embryos. 17 days post-partum. The chief characteristic of embryos of this age is the commencement of ossification around the rib cartilages. Nucleated red blood corpuscles are very scarce. The tongue possesses conspicuous striated muscle cells, stratified epithelium, and at least one circumvallate papilla. The nasal capsules have lost their direct connection with the mouth, but the nasopharynx opens into both the anterior and posterior regions of the mouth. The anlagen of the cartilaginous rings of the trachea are present. The testes have tubules with a peripheral layer of small cells while the larger primordial germ cells occupy the lumen. In the ovaries an ingrowth of connective tissue is noticeable. 18 days post-partum. Ossification of the cartilaginous skeleton is now the striking feature, and the membrane bones of the upper jaw, hard palate and roof of the skull are also being formed. The testes show a considerable amount of connective tissue between the tubules, while the ovaries differ from those of the preceding stage only in that they project further into the abdominal cavity. 19 days post-partum. The ribs of embrj^os of this age have an outer shell of bone, and the underlying cartilage is being torn down to make a marrow cavity. There is a marked spongy 36 WILLIAM B. KIRKHAM structure in the lungs; the eye balls show a differential curvature in the cornea and sclerotic; the naso-pharynx no longer opens into the front of the mouth. The testes show no change, but the ovaries are more spherical, and the ingrowth of connective tissue has forced most of the germ cells toward the periphery. 20 days post-partum. The iris and choroid of the eyes first show pigmentation at this time, and lymphoid tissue appears in the tonsils. One litter of four animals was born twenty days after the birth of a preceding litter and grew to maturity. 21 days post-partum. The feature of this stage of development is the ossification of the metacarpals and metatarsals. The testes have a well organized tunica albuginea and show less space between the individual tubules than in twenty-day specimens, while in the ovaries the primordial germ cells, or oogonia, are of varjdng sizes, the largest of them beginning to form follicles about themselves. 22 days post-partum. The embryos in non-suckling white mice have now completed their intra-uterine development, and parturition occurs. Such, in brief, is the record of development from day to day of white mouse embryos carried by females not suckling young. More extended study and more material would certainly yield many more details, but the above data are sufficient basis for estimating the age of embryos of unknown history, provided onlj'they come from non-suckling females. Also the evidence collected is adequate for comparison with the facts noted below concerning the development of embryos carried by suckling mothers. THE DEVELOPMENTAL CYCLE IN SUCKLING FEMALES White mice are able to become pregnant while lactating but when suckling a litter only about one female in five undergoes a complete pregnancy; those which do not complete a pregnancy either failing to ovulate (the majority of cases) or the fertilized eggs developing normally until shortly after implantation in the uterus, when they die and are absorbed (the minority of GESTATION PEUIOD IN SUCKLINC; MICE 37 cases). In some instances we find auotlier state of alTairs, certain ones of a set of implanted (Mnhryos undergoing normal development wliile others die and are absorbed, a very rare condition in non-suckling females. Suckling females which arc not going to skip an o\ulation cycle shed their eggs within twenty-four hours of parturition as do non-suckling females. These eggs are then fertilized and divide according to the same time scheme as given above for eggs in non-suckling females, being in the two-cell stage on the second and third days following parturition, morulas on the fourth and fifth days, and blastulas lying free in the uterus on the sixth day post-partum. Now comes the point of greatest interest in this investigation. The fertilized eggs in non-suckling females, as stated earlier in this paper, become implanted in the uterus at the close of the sixth daj^ post-partum, the fertilized eggs in suckling white mice, on the contrar}'^, lie free in the lumen of the uterus from the sixth to the end of the fourteenth day following parturition. The material on which this statement is based comprises serial sections of the entire uterus of ten females, suckling from three to eight young, killed at various times from the sixth to the fourteenth day post-partum, all of w^hich show normal blastulas lying free in the uterus, with no sign of any reaction in the adjacent cells of the uterine epithelium (table 1). TABLE 1 Data regarding all suckling white mice from which embryos were obtained from the sixth to the fourteenth day following 'parturition AGE IN DATS POST-PARTUM STAGE OF DEVELOPMENT NUMBER OF SUCKLING TOUNQ 6 Free blastulas in uterus 6 7 Free blastulas in uterus 4 8 Free blastulas in uterus 5 9 Free blastulas in uterus 3 10 Free blastulas in uterus 5 10 Free blastulas in uterus 8 11 Free blastulas in uterus 6 12 Free blastulas in uterus 4 13 Free blastulas in uterus 4 14 Free blastulas in uterus 5 38 WILLIAM B. KIRKHAM Here evidently is one, and perhaps the main cause of the prolonged gestation in suckling white mice. Ovulation, fertilization and early cleavage stages occur at the same time intervals in both suckling '• and non-suckling animals, but in the sue ling females the uterine cells will not react to the eggs and enable the latter to become implanted while the mammary glands are withdrawing the surplus nourishment from the parent organism. In support of this statement it should be said that young mice derive all of their nourishment from the mother for the first ten or eleven days after they are born, and thereafter appear to nurse as much and as often as the female will let them, a fact which undoubtedly accounts for the precise number of suckling young exerting a definite influence on the rate of growth of intrauterine embryos. If only one or two young mice are suckling the development of eggs and embryos proceeds as though none were suckling, but if three or more young nurse the gestation period is lengthened, according to both Daniel ('10) and to the small amount of evidence on this point possessed by the present writer, approximately one day for each animal suckled. This observed fact is, however, very difficult to correlate with the series of embryos obtained from females suckling three to eight young and killed from fifteen to twenty-four days after parturition (table 2) . These embryos show a variation in state of development which appears to vary neither with the number of young suckled nor with the number of embryos carried; in fact it would seem as though in one instance (female killed eighteen days post partum table 2) that the embryos, being as fully developed as in a non-suckling female of the same age post-partum, would have come to birth on the twenty-second day following the previous parturition in spite of there being three young suckling. Even supposing this had happened, were all the facts known such an exception would probably be explainable, since occasional females may reasonably be expected to possess amounts of nourishment far in excess of the average, and some litters of young may start eating grain at an earlier age than usual. More difficult of explanation are such cases as the embryos of a female killed seventeen days GESTATION I'KKIUl) IN SUCKLING MICE 39 TABLE 2 Data regarding alt suckling white mice from which embryos were obtained from the fifteenth to the tievnty-fourth day following parturition. Embryos labeled {small) tvould probably all have been absorbed AGK IN DAYS P08T-P\RTUM DEVELOPMENT EQUAL TO EMURYOB OP NON-8DCKLINO 9 8 OF NUMBER OP SUCKLING YOUNG NUMBER OF EMBRYOS days p. p. 15 8 8 11 16 12 4 8 17 14 5 2 large + 6 small 18 18 3 9 19 12 7 2 large + 4 small 20 8 6 5 21 8 5 9 21 12 8 9 22 8 ? 10 23 12 7 9 24 14 4 6 large + 1 small post-partum (table 2) which, if, as we have every reason to beheve, they became implanted at the close of the fourteenth day post-partum must in the course of the three days following have undergone a development which in embryos in non-suckling females requires eight days to complete, and this in spite of there being five suckling young. At the present time work is in progress with a view to explaining, if possible, these apparent contradictions and until that work is completed, which may not be for some time, it does not seem desirable to attempt any further analysis of the facts presented in table 2. SUMMARY 1. The present work has brought together sufficient data with which to determine, within the possible error of one day, the age of all embryos obtained from non-suckling white mice. 2. Ox-ulation, fertilization, and the early cleavage of the eggs bear the same time relations to parturition and to one another in both suckling and non-suckling white mice except the former are much more apt to skip an ovulation period. 40 WILLIAM B. KIRKHAM 3. Implantation of embryos in the uterus occurs in non-suckling white mice on the fifth day following parturition (provided the female did not skip an ovulation cycle). 4. Implantatioii of embryos in the uterus occurs in suckling white mice, with 3 or more young, on the fourteenth day following parturition (provided the female did not skip an ovulation cycle). In these lactating females the blastulas lie free in the lumen of the uterus from the sixth to the fourteenth day postpartum due supposedly to the activity of the mammary glands. 5. The available material of stages following implantation in suckling females shows no evident correlation with either the number of nursing young or the number of embryos being carried. It also is impossible at present to reconcile the development of these embryos with the observed facts regarding the time of parturition in suckling mice. 6. The conflicting evidence from post-implantation stages in suckling females is at present being subjected to further study. JUNE, 1916 LITERATURE CITED Daniel, J. F. 1910 Observations on the period of gestation in white mice. Jour. Exp. ZooL, vol. 9. King, H. D. 1913 Some anomalies in the gestation of the albino rat (Mus norvegicus albinus). Biol. Bull., vol. 24. KiRKHAM, W. B. 1907 Maturation of the egg of the white mouse. Trans. Conn. Acad., vol. 13. 1910 Ovulation in mammals, with special reference to the mouse and rat. Biol. Bull., vol. 18. KiRKHAM, W. B., AND BuRR, H. S. 1913 The breeding habits, maturation of eggs, and ovulation of the albino rat. Am. Jour. Anat., vol. 15. 1916 The prolonged gestation period in nursing mice. Anat. Rec, vol. 11. Long, J. A. and Mark, E. L. 1911 The maturation of the egg of the mouse. Pub. Carnegie Inst., Washington, D. C. ## On the postnatal growth of the body and of the central nervous system in albino rats that are undersized at birth HELEN DEAN KING The Wistar Institute of Anatomy and Biology Among the newborn young of various species of mammals there occasionally appear individuals that are undersized and have a very small weight at birth, although they are apparently normal in all other respects. Such individuals are very generally called 'runts' and they are usually discarded by breeders at birth, since it is the popular belief that they never attain normal size and that they are always sterile. Little is known of the true status of these animals, as few attempts have been made to rear them for the purpose of studying their growth processes and reproductive capacity. In the course of an extensive series of breeding experiments with the albino rat a number of litters have been obtained in which one or more of the rats was very small at birth and weighed much less than the average birth weight for the young of the species, which is about 4.5 grams for the male and 4.3 grams for the female rat (King, '15 a). An examination of these litters some twenty-four hours after birth has shown, as a rule, that the very small individuals were dead, although all of the other members of the litter were alive and vigorous. Thus even under very favorable environmental conditions rats that are much below the average size at birth have little chance, apparently, of surviving even the first few hours of postnatal life; in a state of nature probably very few of them ever live to reach maturity. In three of the htters examined the undersized young survived the seemingly crucial twenty-four period, and it was found possible to rear them with the other members of the litter and to study their growth in bodj^ weight. In order that the small 41 42 HELEN DEAN KING individuals might have every possible chance for subsequent growth, the young that were not of the same sex as the smallest member were discarded from each litter. This gave two series of female rats, on6 containing three and the other four individuals; and one series of four males. The rats in the first and those in the third series belonged to the twentieth generation of an inbred strain of albinos in which matings had been made only between brother and sister of the same litter in each generation; the rats in the second series were the offspring of an inbred female (nineteenth generation) and a stock male. For convenience litters with a parentage like that of the second series are designated as 'half -inbred' litters. The rats in each series were weighed for the first time before they had suckled and then daily for one week. Thereafter the weighings were made at intervals varying from two to fifteen days until the rats were 150 days old w^hen the experiment was ended. In an investigation of this kind it is impossible to obtain the exact body weights owing to the varying amounts of food in the digestive tract at the time that the animals are weighed. To minimize this source of error as much as possible all of the weight records were taken in the morning before the rats had received their daily ration. Table 1 gives the birth weights of the rats belonging to the three series and also their later body weights at the different ages for which records were taken. There was considerable variation in the birth weights of the rats in each of the three litters, as is shown in table 1, but in no series was the range of variation as great as that known to occur within the species (King, '15 a). In the first series the heaviest rat (no. 3) weighed but slightly more than the average weight for the female albino rat at birth, yet this weight is 77 per cent greater than the weight of rat no. 1 which had the smallest birth weight that has been found, as yet, in any female rat in our colony. The range of variation in the birth weights of the rats in the second series is much less than that in either of the other series, yet the weight of the heaviest member (no. 4) is 46 per cent greater than that of the smallest member. In the < o a 2 >< o n ' a "0 CO QO i •^ «(^ 1(5 00 -f ov o> fl CO -f 1(0 r>. CO CO CI -H CD 00 1^ (M (M ?S (M — eo (M lo J5 CO lO lO iM ^ ^ rs 1 "0 CO IM O ■n (N O fO o CO <o ><o 5§2S CO CO 1^ o o» ?5 ci c3 ( S I-H •— 1 .-1 IM (M CO CO — < CO CO «*< i^ll s CO 00 t^ g »-< ?5 CO I i-l rH f-H O O Ci 1-^ CO (M CO ^ — < iM (M 1 9 ^ ^ ^ Oi 1(0 CO CO r^ o> o -H 93 142 111 162 119 180 125 182 2 00 o C5 5 •"• 00 "-1 03 00 lo CO CO t^ < «> e<5 00 3 t>- O 00 lO t^ CO ^M lO "5 CO t^ r^ -H lo o 00 O -H (M o o o ■*■<*' O CD C5 CO (N CO ? lo ifs c^i f lO CD ^ lO 05 (M CD ■* Tt< »C >0 o CO r^ -H 00 00 00 o < y T}< (M r^ CO (N 00 t~ ci o <o CO 1 to ^ Tt< t lO IC -H Tt< r^ 00 CO CO CO CO o ■^ CO CO lO lO o t 1 J o <o CO O CO 05 ■* IXI C5 Tt* 05 5 OP »o t^ IM CO CO Ci 1^ lO O .— (M (M (M O t — 1 — I -* '^ >o CO J O lO 00 O lO CD ^ o ^ 00 —t c» ■* <r> rt (M (N 00 o ^ 0-1 T-H OJ C-l IM 0> 'fi C5 rfi iM CO CO -41 J (N t^ rf< t^ t- ■* t~ CD t^ CO O t^ -^ CO ^ (M (M ■* o t^ r^ »-i 05 CO CD (N IM CO CO •T J 1 lO •* O o 00 lo ci iQ r^ t^ (M T)4 00 d 1 »—)»—* c^ (M CO ■* ■* t^ CO 00 o —1 (M (M CO J 1 CO O iM lO iM 00 KO CO Ci t^ 00 J c^' d t^ O (M (M CO CO o -Tt< •<*< ^ ^ IM (M c J C2 CI lO 00 t^ ^ (M CO ^ 00 (M 00 c^ ■* 00 O ^ (M O CO -H (M ^ -H !M (M 1^ o »o o (M T}< CO Tt< O 't' 00 05 d C5 d t^ 00 C5 Oi t^ O •* lO T 00 '^ CO CD 1-H CO C5 t^ rf t^ ■* »c 00 o5 O 00 00 00 L':i 00 (M CO T —I o -* Oi O ^H CD C; O CO ^ IQ t^ 00 »0 t^ 00 00 '^ t^ c — < r lo r^ '^ CO 00 ■* uo ■* O CI ■* TJH d t^ lo CO t^ 1^ ■* CD 00 C5 c^ 05 O i-O »o o lo 00 t^ CO o o CO ;d d ■* CD CD CO CO d t^ 00 s rp Tti <r> 00 Oi CO ■* •^ CO ic 00 CO lO o CO ■* lo «io CO -^ o CO X O O <N •^ CO o o (M o 00 o; CO no IC CO -<*< lo lo CO ■* »o ic j: CO Tfi CD ■-I C2 -* IC t^ •* (M CO (M ■* •* CO CO ■* ■* (M CO i-O UO a m a ■< 6 ^.H M CO .-H -M CO '^ ^ (M CO '^ Scries 1 1 (Females) | Series 2 j (Females) ] Series 3 J (Males) 1 1 43 44 HELEN DEAN KING third series the birth weights ranged from 2.7 grams to 5.3 grams, the largest individual being nearly twice as heavy as the smallest. All of these rats followed the normal com'se of the growth in body weight, as already determined by Donaldson ('06), in spite of the very great differences in their birth weights. Increase in body weight was very rapid during the early days of postnatal life. The rate of growth dropped off somewhat abruptly at about thirty days, and again showed a marked decline at seventyfive days. After the rats reached ninety days of age the period of rapid growth was ended, and the rats gained relatively little in body weight up to 150 days of age when the weighings were discontinued. With few exceptions the body weights of all individuals fall within the range of variation in the body weights of stock albino rats of like age (King, '15 b, table 3). None of the rats that were unusually small at birth, therefore, could properly be considered as 'runts' when they became mature. The rats that were undersized at birth never succeeded in attaining a body weight equal to that of the other individuals in the same series at any period of their growth, and at successive weighings the actual weight differences between the individuals that were small and those that were heavy at birth tended to increase. This was true for the individuals in each of the three series, irrespective of sex, as is shown by the data in table 1. At the end of 150 days the rats in each series still maintained the same order with respect to body weight that they had at birth, with the exception of rat no. 3 in the third series. This rat overtook its brother, which had a heavier birth weight, when it was ninety days of age and it subsequently kept the lead in body weight until the end of the experiment. According to the 'standard' tables for the relation of age to body weight and to body length in the albino rat as given by Donaldson ('15), breeding females should have a body length of 193 mm. and a body weight of 186.1 grams when they are 150 days of age; the body length of a male albino rat of the same age should be 207 mm. and the body weight 218.7 grams. Records for the growth in body weight of a selected series of stock, albino rats reared in The Wistar Institute annual colony under nUOWTll OF 1U)1)V AND NERVOUS SYSTEM IN RATS 45 environmental conditions similar to those under which the rats used in the present experiment lived (King, '15 b) show that for this group the average body weight of breeding females at 151 days of age is the same as that given by Donaldson, namely 186.1 grams; the average weight of the males of the same age is 244.8 grams, which is 26.1 grams above the computed weight for the male as given in Donaldson's tables. The a\'erage body weight of all of the females used in this experiment was 146.7 grams, and that of the males was 258.7 grams, when the animals were 150 days old. The females, as a group, are too light in weight for their age, while the males are much too heavy, wliichever of the above series of records is taken as a standard for comparison. The fact that the females were either strictly inbred or halfinbred does not account for their small size, since in the strain of albinos from which these rats were taken inbreeding has increased rather than diminished the average body weight of both males and females (Popenoe, '16). Investigations made by Watson ('05) have shown that female rats that are allowed to breed are heavier at a given age than non-breeding females. It is probable that the low weight of these females is due, in some measure at least, to the fact that the rats were never mated (the stock females whose weights were given for comparison were all breeding animals). The relatively large size of the males in the third series can be attributed to the fact that the animals were from a selected inbred strain. The average daily percentage gain of these rats in body weight during the period covered by the experiment is shown in table 2. The percentage values for weight increase, unhke the weight data, show no definite order with respect to the birth weights of the individuals concerned. At some periods the rat which had the smallest birth weight shows a greater daily percentage gain in weight than smy other member of the same series; at other periods the weight excess is in favor of the individual with the heavier birth weight (table 2). If, instead of considering all of the percentage values in a given series, only the data for the two individuals that had the extreme H D3 2 □ O ra Z w IK <; P3 o g a < z o IB B. >< a ~ >.is o ©3 5476.92 3377.27 3313.04 CO CO CO 00 CO CO lO CO CO C^ CO 00 CO CO (M -^ -t CO CO CO CO 03 r^ t^ c^ t^ 10 CO (N CO CO (M CO C2 CD CO t^ t^ 10 >o O ^ M^ CD O O d CD d -H CO ^ t^ CVJ ^ ^ ,_( d> d d> d> CO C5 (M 10 0^00 d d d d lO Oi Oi o o o (6 c6 G 00 "^ Tt* CO 10 C<1 rt tH d d d d 02 CO r-4 CO CO 10 CO >o d d d d OC C32 00 <6 di (6 c^ 00 -^ ■* .-1 CO CO --I d d d d -rt< 02 lO ^ (M CO ■* CO d d d S ill oi 00 00 lO Tt< »o o d o (M CD 03 i-H T-H CO oq d d d d 10 t^ -^ ^ CO e ^ o c^ •41 O kO S Tt< CO ^ C5 CO t^ '^l 00 (N 00 '« d ^ d> (m' C^ ^ 1-1 1-H ,-1 tOT3 •^ CO •* l>. CO (N t-i 00 Tfi t^ (M C2 10 •* ,-1 ^ (N (M CO CO (M CO CO CO CO 3, 10 10 (M t ,-1 di '-I ^ 10 00 10 t^ as d d d d Ci C2 >0 OQ C5 CO -rf^ d d d d e 2-§ r-H C2 00 CO ^ 1-1 CO <z> ^ 1— 1 ic ci lO C<i Oi ^ t^ r-l ^ CC o CO ■* '^ M — 1 l-H (M (N CO --H II 10 en CO 00 C2 CO Ci '^f 135 C<l CO ■* 05 CO ■^ ■^ CO "^ 1-H CO M rH TfH t^ 00 CO 10 CO CO 51 ^ "* (M r-H '^ CO CO aJ CD lO 01 05 t^ --1 CO r-< 10 05 CD ic 00 CO t^ CD •^ Tfi "^ CO CO t^ CD CO "* 00 01 t^ ^H CO 00 rt 05 CO 05 d CO CD ■^ CO 1— 1 CO ^ CO 10 ^ 00 CO Tt< 00 i-3 10 -^ '^ CO 00 CO "^ "^ 00 00 Tt^ CO ■* kO (M i>i t^ 00 06 .-1 CO --H ■^ 0> 02 kO ^ •*' csi d O St3 H-l CO (M C2 CO (O to in CO CO iO CO iM t^ t^ CO CD CO t^ C2 00 Tt^ ^ CO r^ 00 C5 1^ V >> ^ CO C5 lO 1— 1 00 t^ 00 rt" lO ■* GC T-H 10 CD ^ t^ CO CO 10 01 ^ C2 00 I^ >0 GO 10 00 2 II (M ^ ^ CO CO d !M 10 '^i (M CO --H CO 00 i> CO 10 CO CO ic 10 t^ 00 CO 00 (m' 1-i d 10 1 rt S ci CO r-H r-i 01 (N (M (M Tt< ^ iCi 05 t^ OS b- 05 TfH 1^ CO 1-1 (M '^ c^ r^ 00 CD CO C35 f^ t^ 01 01 r- -H t^ 05 CD CO LO 00 00 10 Pi. CO CO CO 02 CO CD iC (M CO CD 00 (M (M ^ ^ 5& 5O-0 (M C<l — 1 r^ 10 t^ CO d CD i-~ r^ 00 ^ ^_ ^ 00 CO CO (M ^ C^ CO 00 r^ d 02 rH C^ ,-H (N WT3 CO CO CO CO •* 10 CO CO CO iM en CD CD CO CO -* CO r-i C^ •* CD CD 00 CO CD ^ T-H CO d 00 -g •^ S" 00 CD '^ CO CD 00 lO "^i CO t^ CO ■* t^ CO 00 CO I-^ CO CO 02 Tt^ CD 10 00 CO oa t^ mT3 ^ t> t^ -H Ol '^ '^ c^ ■rtH iM CD 05 Cq ^ CO -* 00 IM 10 CO Tt< ^H CO 00 iM <M >0 ^ (M (M OJ 00 ic rt t^ .-H (M ,-H CC 13 >. CO 01 CO CD CO 00 t^ CO lO t^ 05 i-i CO CO 00 10 r^ 10 <M (M d >o cci 10 m c3 00 CO Tt< CO CD 1^ lO CO ^ CD (M CD --H .-1 'f CO r>) lO CO 10 CO UO CO CO 01 CO ^ 00 r^ ^ ^ •ON XVH ^ (M CO ^ (M CO ■* 1 rH (M CO "* I S3uag 4 (saiTJuia^) S sai.iag 5 8 soijag t; HOW I'll OF llODV AND NERVOUS SYSTEM IN HATS 47 birth weights are coinpared, the results ()])taiiietl seem consistent enough to be significant. In the total of 01) records for the three series, 42, or nearly- two-thirds, show that the daily percentage weight increase is greater for the individuals with a low birth weight than for those with a lieavy birth weight. In other words, regardless of sex, rats that are very small at birth tend to grow more rapidly than do rats that have a heavy birth weight, although their actual body weights are always less at any given period. On computing the percentage increase in body weight at 150 days over the birth weight for the various individuals it was found that the rats that were small at birth had gained a greater amount than had the rats that were heavy at birth. For each series, as shown in the last column in table 2, these percentage values stand in inverse order to that of the birth weights, with the one exception in the second series (rat no. 4). The results of this investigation are in accord with those obtained by Dunn ('08) in her study of the weight increase in a 'group' of seven albino rats (three males and four females) w^hich had very unUke body weights when they were fourteen days old. Dunn found that, with, one exception, the order relation of the weight at fourteen days of age was maintained until the end of the experiment ; the rats having the heaviest initial weights were also the heaviest at sixty-sLx days of age when the weighings were discontinued. The lighter rats, on the other hand, while putting on less absolute weight, had gained at the end a greater percentage of their original weight than had the individuals with the heavier initial weight. In addition to the undersized young which, as shown above, are capable of developing into adults which are only slightly below normal, a litter sometimes contains individuals of a much lower grade to which the term 'runt' properly applies. In these individuals, which apparently are indistinguishable from the other young at birth, the normal action of the growth factors is inhibited from the very beginning of postnatal life by unknown constitutional causes, not by environmental conditions. When the young rats are old enough to leave the nest the runts can THE ANATOMICAL RECORD, VOL. 11, N'O. 2 48 HELEN DEAN KING easily be distinguished from the other members of the Utter, not only because of their very small size, but also because of their slower movements and apparent lack of normal vitality. Runts grow slowly for a certain time, but no matter how favorable the external conditions, they never exhibit normal vigor and they are always dwarfed and stunted in their body growth. In such animals growth is not merely retarded, as it is in the case of rats experimentally stunted (Hatai, '07 ; Osborne and Mendel, '14) , but it is permanently checked at an early age. Several attempts have been made in our colony to increase the size of runts by special feeding, and to breed them for the production of a dwarfed race of rats. Only a few Jitters could be obtained from such stock, and these contained a very small number of young which were puny from birth and which died at an early age. The reproductive powers of these animals are apparently never developed in a normal way, as the males rarely mate and most of the females are sterile. All of the rats used in this study were killed at the end of 150 days, the body weights and body lengths determined, and t-he brains and spinal cords removed and weighed. This was done in order to ascertain whether the central nervous system in adult rats that were small at birth bears the same relation to body weight and to body length as that found in adult Individ uals that were of average size, or above, at birth. Table 3 gives the body lengths of the individuals, the observed weights of the brains and of the spinal cords, and the brain and cord weights corrected according to table 68 in 'The rat: data and reference tables' (Donaldson, '15) which gives the computations for the 'standard' weights of the central nervous system in albino rats of various body lengths. In addition to the above data, table 3 shows the percentage deviations of the observed weights of the brains and of the spinal cords from the corresponding standard weights. It is evident, from the percentage values given in the fifth and in the eighth columns of table 3, that 'the observed weights of the brain and of the spinal cord in all of the rats are considerably below the standard weights for these organs in animals GRO^VTH OF 1U)I)V AM) NERVOUS SYSTEM IN RATS 49 Showing the bodij lengths, wilh the observed and the 'standard' weights for the brain and for the spinal cord, of the albino rats whose weight data are given in table 1 ^; 7. X > f. f o D >. X H • K (S ° O M < o r ^5 c E « v. K n « « -is o O B ■< M g RAT NO. o 5 Q e o «2^ a a h OS -! 3-3 a H > H •< K a a u H H O H g U (B " .a a — o •<! r u •^ m O ^ w a; «: > w "a O n 00 0. o 00 «. 1 183 1.461 1.773 -17.5 0.455 0,524 -13.1 Scries 1 (Females) 2 185 1.611 1.782 - 9.5 0.473 0.532 -11.0 3 182 1.679 1.768 - 5.0 0.499 0.520 - 4.0 1 175 1.511 1.735 -12.9 0.418 0.492 -15.0 2 184 1.616 1.778 - 9.1 0.467 0.528 -11.5 Series 2 (Females) < 3 192 1.653 1.814 - 8.8 0.485 0.560 -13.4 4 188 1.643 1.796 - 8.5 0.475 0.544 -12.6 c 1 198 1.697 1.849 - 8.1 0.529 0.556 - 4.8 2 210 1.765 1.903 - 7.2 0.579 0.603 - 3.9 Series 3 (Males) < 3 217 1.955 1.933 + 1.1 0.631 0.630 + 0.2 4 212 1.999 1.911 + 4.6 0.669 0.611 + 9.4 of like bod}^ length, except in the case of the two largest males where the observed weights slightly exceed the standard weights. That the weights of the central nervous system in all individuals of a given litter, having the same sex and about the same body weight, should deviate from the standard weights in the same direction was not an unexpected result. In the rat variation within the litter unit is usually in the same direction and much less than that in the general population as regards body weight (Jackson, '13; King, '15 b), and doubtless this rule holds for the central nervous system and other organs as well. In every series, as shown in table 3, the rats with the smallest birth weights are the ones whose brain and cord weights show the most marked deviations from the standard weights, regardless of their body length and body weight with which the weight of the central nervous system is, as a rule, closely correlated. Thus, in the first series, female no. 1 had a body length of 183 mm. and a body weight of 145 grams while female no. 3 was shorter and heavier, 50 HELEN DEAN KING yet in the former individual the brain was 17.5 per cent and the cord 13.1 per cent below the corresponding standard weights; in the latter individual the brain and cord weights were only about 5 per cent less tl^ian the standard weights. All of the rats in the second series had cord weights that showed relatively greater deviations from the standards than did the brain weights; the lowest weight for both brain and cord being found in the rat that had the shortest body length and the smallest body weight (no. 1). The most interesting result is found on comparing the records for the rats belonging to the third series. The brains of the two rats that had birth weights much lower than the average birth weight (nos. 1 and 2) were each about 8 per cent less than the standard and the cord weights showed a minus deviation of some 4 per cent from the standard, yet in body measurements rat no. 1 was below and rat no. 2. was above the average for stock males of like age. The brain and cord weights of their brothers, each of which was unusually heavy at birth, were above the computed standard weights for the central nervous system in animals of like body length. That the postnatal growth of the central nervous system is influenced to some extent by the factors that determine the size of a rat at birth seems to be a conclusion warranted by the analysis of data given above. SUMMARY AND CONCLUSIONS The results obtained in this investigation seem to indicate that the undersized individuals which are sometimes found in a newborn litter of rats are not necessarily 'runts' in the generally accepted use of that term. Some of these small individuals, as shown above, attain an adult size that not only is within the normal limits of variation in the body weights of standard stock rats of like age, but may even exceed the average body weight of a large number of stock animals (rat no. 2, series 3). All rats in a litter are not born with a like capacity for growth, as the data in table 1 indicate, and even when environmental conditions are as favorable and as uniform as it is possible to make them, individuals having unlike birth weights show marked differences in their rates of growth from birth to the adult state. (;i{()\\Tll OF BODY AND NERVOUS SYSTEM IN RATS 51 In the cases stiidietl the individuals liaving a small birth weight seemed to possess a 'very great capacity for growth from the very beginning of postnatal life. Female no. 1 of the first series increased 2 per cent more in body weight during the first twentyfour hours after birth than did either of her sisters; for the same period the gain in body weight of the two smallest males in the third series was over G per cent more than that of their brothers. In the second series, howe\'er, the weight increase for the first day was greater for the rats that were heavy than for those that were small at -birth. With an early acceleration in the rate of growth there is seemingly correlated an early cessation in the extent of growth, as in the adult state individuals that were undersized at birth are alwa>'s smaller than the other members of the same litter. On the other hand, rats with a hea\'y birth weight tend to grow more slowly at first than do the smaller individuals, but they continue to grow for a longer time and eventually reach a greater size. This rule seems to apply to females as w^ell as to males. Not only does body weight at birth indicate the probable capacity of the individual for subsequent growth, but it also indicates the probable size of the central nervous system, since rats that are undersized at birth tend to have a much smaller central nervous system when they become mature than do other rats. The factors, whatever their nature, that determine the body size of a rat at birth seem to have a marked effect on the subsequent postnatal development of the individual, influencing the ultimate body weight as well as the size of the central nervous system. A very small weight at birth indicates that a rat has a handicap in its organization that environment, however favorable, cannot overcome. Such animals, although they appear vigorous and healthy during their growth period and after reaching the adult state, are unquestionabty sub-normal in regard to the size of the body and of the central nervous system. If allowed to breed these rats would probabty produce young having a weaker constitution than their ow-n, and from such stock one W'Ould ultimately get 'runts' and an increasing tendency towards sterility that w^ould soon bring disaster to the colony. 52 HELEN DEAN KING Judging from the results obtained in this study a newborn litter of rats may contain individuals of three kinds as regards their inherent capacity for body growth. As a rule, only young rats having a normal birth weight and a normal capacity for growth are found in a small or medium sized litter produced by a female rat in good physical condition. Occasionally rats are born which have a very small birth weight, and in these individuals, if they are able to survive, the growth capacity is lessened to some extent but not sufficiently to prevent them from being classed as 'normal' after they have reached maturity. If a litter is very large, or if the mother is not in good physical condition during the gestation period, some of her young may be born with their growth capacity so impaired that it is impossible for them to grow beyond a certain stage. These individuals are true 'runts' and, fortunately, they are lacking in reproductive vigor as well as in growth capacity so that they are usually unable to reproduce their kind and so prove a menace to the colony in which they live. LITERATURE CITED Donaldson, H. H. 1906 A comparison of the white rate with man in respect to the growth of the entire body. Boas anniversary volume, New York. 1915 The rat. Data and reference tables. Memoirs of The Wistar Institute of Anatomy and Biology, no. 6. Philadelphia. Dunn, Elizabeth 1908 A study of the gain in weight for the light and heavy individuals of a single group of albino rats. Anat. Rec, vol. 2. Hatai, S. 1907 Effects of partial starvation followed by a return to normal diet, on the growth of the body and central nervous system of albino rats. Amer. Jour. Phys., vol. 18. Jackson, C. M. 1913 Postnatal growth and variability of the body and of the various organs in the albino rat. Am. Jour. Anat., vol. 15. King, Helen Dean 1915 a . On the weight of the albino rat at birth and the factors that influence it. Anat. Rec, vol. 9. 1915 b The growth and variability in the body weight of the albino rat. Anat. Rec, vol. 9. Osborne, T. B., and Mendel, L. B. 1914 The suppression of growth and the capacity to grow. Jour. Biol. Chem., vol. 18. PoPENOE, Paul 1916 Experimental inbreeding. Jour. Heredity, vol. 7. Watson, J. B. 1905 The effects of bearing young upon the body-weight and the weight of the central nervous system of the female white rat. Jour. Comp. Neur. and Psychol., vol. 15. MOUxTixc spp:cimens under PF/nn dishes and CLOCK GLASSES E. L. JUDAH McGill University, Monlrcnl Since obtainiii};- a suitable cement for the sealing of square museum jars, the mounting of thin sections of pathological and anatomical specimens under Petri dishes and clock glasses has been made both easy and cheap. Several years ago these mounts attracted a great deal of attention, lacing put on the market as paper weights, etc., but were gradually adopted for the exhibition and display of museum specimens. The method, however, being patented was expensive and beyond the reach of the average museum; besides it was necessary to send your material for mounting to the manufacturer. In 1906 Dr. Hutchinson of the Roj^al Victoria Hospital, Montreal, read a paper before the British Medical Association at Toronto on this method; but unfortunately he did not have a suitable cement and the process was slow and laborious. METHOD The fluid used for mounting should be brought to a boil in the same dish that is to be used for mounting and allowed to stand over night, or until cool, to get rid of as much air as possible. The dish should be deep enough to come well up over the mount to allow of easy manipulation of both Petri dish and specimen. Get a Petri dish of suitable size to hold the specimen so that when the sheet of glass which is to form the cover of the mount is in position it will not quite touch it. Great care must be taken that the Petri dish or clock glass fits perfectl}'^ on the base and does not rock. In placing the Petri dish in the mounting fluid, do so without causing any air bubbles. When the dish and fluid are read}^, wash the specimen in several changes of the same fluid that you are mounting in, to get rid of any loose particles or dirt. In the last change of fluid, work out all air and remove quickly to the Petri dish, face downwards, again eliminating air bubbles. A small piece of looking-glass in the bottom of the mounting dish is very convenient, as by tipping the Petri dish over and on its side shghtly, any bubbles under the specimen may be seen. The specimen now being in position, put on the base or cover and remove from the mounting-dish, holding it firmly so that it cannot slip 53 54 E. L. JUDAH and admit air while turning the mount right side up. Over the junction of Petri dish and cover pour hot cement^ to the thickness of about one quarter of an inch, and allow the mount to stand on a flat surface for a few days, or until the cement is perfectly adherent to both Petri dish and cover. If after several days it is desired to finish the mount, remove any bubWes of fluid between the cover and the cement with a very hot knife, working the knife to the outer edge of the base. This must be carefully done, and the knife kept hot enough so that the cement will be kept liquid and not be drawn away from the Petri dish. All the fluid must be removed from under the edge of the Petri dish in this way. The best results are obtained, however, by allowing the mount to stand for several weeks when most of the excess fluid will work out by itself. I usually mount several dozen specimens at a time and let them stand until they are ready to finish. If any air bubbles should happen to get under the edge of the Petri dish they will have to be worked into the mount in the same way that fluid is removed, only work your knife inwards instead of outwards. Even with the greatest care air is often retained in the specimen itself, and only detected after the mount has stood for several days. Should the bubbles be very small they will quite frequently be absorbed by themselves; if not, shake the mount until they are all in one large bubble, and resubmerge in mounting fluid which must be heated as hot as you can comfortably stand your hand in. When the cement has become pliable enough so that it is possible to move the Petri dish with the fingers, insert the point of a sharp knife between the dish and base, gently forcing them apart and allowing a little fluid to enter. The mount must then be tilted on its side to bring the air bubble to the opening, where it will escape when you remove the point of the knife; then press the Petri dish up against the base, expelling all superfluous fluid. Close the hole by pressing the soft cement together with the fingers. Air bubbles may have to be removed several times depending on the specimen. Sections of lung give the most trouble. When sure that there are no more bubbles in the mount and that all fluid has been removed from between the cement base and the Petri dish, coat the cement over to the thickness of about one-eighth of an inch with refined asphalt, being careful that it does not burn; I usually melt it in a tea-spoon over a Bunsen burner. When the asphalt has been all apphed, reheat the whole with a very hot knife and apply a bezel ring which must be heated red hot, and pressed down into the cement so that the lower edge will rest upon the base. Clean with gasoHn and polish with bon ami soap. The Petri dishes used are manufactured specially from 2 to 3 mm. thick, as the ordinary ones sometimes break with the expansion and Sec Muir and Judah, Sealing of museum jars. Bulletin .5, Inter. Assoc. Med. Mus., page 87. MOUNTING SPECIMENS UNDER I'ETRI DISHES OO contraction of tlu> Ihiid. 'I'hc l)t'zcl iin;;,s which arc nscd to f>;iv(! a tinishcil appearance to the ceniont can In* made out of any in(;tal tliat will stand heinj? heated red hot. They arc no't, howevci-, absohitely necessary, as you can use a sheet of cardboartl with a hole cut in the centre and passe partout the edges. Dr. Higgins of the Experimental Farm, Ottawa, has a very convenient cardboard case into which he slips the mount. If a bezel ring is to be used, tjic base or cover should be made out of plate glass with the edges bevelled. Clock glasses are inferior to Petri dishes for this method because they magnify and distort the spochncn. The following list of sizes have been found to answer all requirements. While 36 in number they onl}' require ten different sized bezel rings and plate glass bases, a desired advantage when the glass ware is made to order. Sizes (Outside measurements) SIZE NO. WIDTH HEIGHT SIZE NO. WIDTH HEIGHT cm. cm. C7n. cm. 1 18 3 19 8 2.5 2 18 2 20 8 2 3 18 7 21 8 1.5 4 16 3 22 8 1 5 16 2 23 6 2.5 6 16 1.5 24 6 2 7 14 3 25 6 1.5 8 14 2.5 26 6 1 9 14 2 27 4 2.5 10 14 1.5 28 4 2 11 12 3 29 4 1.5 12 12 2.5 30 4 1 13 12 2 31 3 2 14 12 1 32 3 1.5 15 10 2.5 33 3 1 16 10 2 34 2 2 17 10 1.5 35 2 1.5 18 . 10 1 36 2 1 ANNOUNCEMENT An Authors' Index has been prepared and printed for each of the following journals: ft JOURNAL OF MORPHOLOGY 25 volumes — 1887-1914 Price per copy, 50 cents THE AMERICAN JOURNAL OF ANATOMY 18 volumes — 1901-1915 Price per copy, 50 cents THE ANATOMICAL RECORD 10 volumes — 1906-(July) 1916 Price per copy, 50 cents Sent post-paid to any address THE WISTAR INSTITUTE OF ANATOMY AND BIOLOGY Philadelphia 56 yi THE EFFECT OF IIYPOPIIYSIX'TOMY IX THE EARLY EMiuno UPON THi: (;uo^vT^ and develop:\ient of the frog A PRELIMINARY REPORT p. E. SMITH From Ihc Anatomical Laboratory, University of California TEX FIGURES The extirpation of the hjpopliysis in the adult frog lias not given uniform results. Caselli ('00) and Gaglio ('02) who reported no changes following M^pophysectomies were followed by Boteano ('06) \\'ho reported a neuromuscular asthenia in the operated animals. Houssay ('10) came to the conclusion that the removal of the gland was followed by death. Adler ('14) bm-ned out the hypophysis of a 20 mm. Rana temporaria larvae ^^■ith the electric cautery. Out of the 1200 operated animals thi'ee were found to have been h>T3oplwsectomized, not, however, without great injury^ to the smTounding soft parts, particularly the brain. In not one of those three animals did hind legs develop beyond a small bud, and transformation did not take place, the specimens remaining as neotonic tadpoles. This work was commenced in the Spring of 1914, repeated in 1915, and again in 1916, Diemyctylus torosus, Rana pipiens, and Rana boylei being successively used. In this paper the results obtained with the Cahfornia yellow-legged frog, R. boylei are reported. Shortly after the closure of the medullary plate, Kopsch's stages d-e, was found to be the size in which the hj^ophysial invagination could be most successfully removed. About 200 larvae of this stage were operated upon. In specimens of this size the h^^Dophysis was successfully removed in over 60 per cent of the operated animals. Approximately 30 per cent of those animals in which the gland was extirpated did 57 THE AXATOMICAL RECORD, VOL. 11, NO. 3 OCTOBER, 1916 58 p. E. SMITH not give reliable results in the rate of growth as the mouth was wholly or partially removed thus interfering with feeding. Unoperated animals and those in which the ablation of the gland was unsuccessfully attempted were available for checks. The operation is a simple procedure. The hyj^ophysial invagination can be accurately determined from the pit that it early forms or from its location between the protuberance of the forebrain and the stomadeum, which is just forming. This epithelial ingrowth was removed with some neighboring epithelium. The wound healed within three hours in most cases, less than 1 per cent of the larvae disintegrating after the operation. The operated animals and checks were kept in boiled water for five days and then transferred to' a frog tank where they were in an essentially normal environment. The rate of growth in the hypophysis-free animals has been slower than in the checks. The larger hypophysectomized animals averaged smaller in size than the larger checks, the averages of the two showing a noticeable difference. On June 6 the operated but not hypophysectomized animals had an average length of 40 to 43 mm., the hypophysis-free animals averaging 33 to 35 mm., a ratio constant throughout their growth. The ratio of body to tail length is the same in the two classes, the difTerence in size being uniform for all parts of the animal. The tail fin did not show an increased width or pleating in the hypophysectomized animals as reported by Adler ('14). In activity the two classes of animals showed no marked differences. The hypophysectomized specimens were perhaps slightly more alert, darted more quickly, and consequently were more difficult to capture with the pipette than were the checks. The resistance of the hypophysectomized animals was greater than that of the checks. Towards the close of the experiment the animals were attacked by disease, none reaching the adult stage. The normal specimens succumbed more rapidly to this infection than did the hypophysectomized ones. Some of the intrinsic factors which induce growth of legs and transformation were lacking in the abnormal specimens as will be shown later. The absence of these factors may well be conducive to a greater EKFKc'i' or in i'(M'iivsK( ro.Mv rrox thk fuog 5!) liariliiioss in an animal wJicn (■()ni])ar('(l to tlic iiornial tadjiole in wliic'li tlio usual ra])i(l ('lianf:;(>s arc takiiij;- ])la('('. nirtci'ciiccs ill color l)c<>;aii to be not iccahic hcforc a i('nfi;tli of IT) nun. was jvaciuHi, and IVoni tlicn on the contrast in ])if2;nicntation between the hypopliysectoniized animals and the cliccks was strikhig. Those animals without hypophyses were characterized by a light grayish appearance; however, the dorsal side was more pigmented tlian the ventral (figs. 7, 10). These are referred to as albinos. The checks were a brown-black color often showing a mottling (figs. 8, 9). This color difference was more noticeable over tlie body tlian on the tail, but was evident in both regions and was the most striking feature up to the time of the appearance of the hind legs in the checks. Sections show that these pigment differences are referable chiefly, if not solely, to the condition of the epidermis. Counts of the melanophores of corresponding areas in the albinos and in the checks show that the number of these cells, in the epidermis, are reduced in the former. Further the melanophores of the albino specimens contain fewer pigment granules than do those of the checks and thus have a distinctly lighter appearance. The melanophores are equally expanded in the two types, consequently, the lighter color of the albinos cannot be due to the contracted condition of the chromatophores but must be referred, in part, to the reduced number of melanin granules in the pigment cells of the epidermis. In addition to this the free pigment granules which form a distinct zone in the superficial layer of the epidermis in the normal checks are much reduced in number in the albino specimens (figs. 5, 6). It is surprising that in the albinos the deeper or subcutaneous pigment is present in as great a quantity as in the normal animals, if not greater. The amount and distribution of the retinal pigment seem to be identical in the two. Another important feature was the inhibition in growth of the hind legs of the operated animals. There was only a slight retardation in the time of appearance of the hind leg buds, normalh^, appearing when the tadpole has reached a length of 25 to 27 mm. In the albino, averages show that the hind limb buds appear when the larvae are from 26 to 28 mm. in length. { 60 p. E. SMITH From this state on, however, the hind Umbs in an h^i^ophysectoniized animal grew but httle if at all, although the animal's length increased at a rate but slightly under the normal. The accompanying tabte shows the increase in length of the hind legs in relation to total length for the albinos and for the checks. (See also figs. 7, 8). Average rate of growth in millimeters in terms of total length, of the hind legs of the checks and the albinos H1TOPHYSECTOMIZED AXIMALS CHECKS Total length Hind le? length Total length Hind leg length' 26 barelj' visible 25 barely visible 28 0.1 28 1.0 30 0.1 30 2.0 35 0.1 35 3.0 37 0.12 38 4.0 40 5.0 45 9.0 Only one exception to the rule that no hind legs grew on albinos was found. A 36 mm. albino had hind legs 4.2 mm. long when killed. The above is in accord with Adler ('14) who found that removal of the hypophysis in a 20 mm. stage inhibited the growth of the hind legs. Examination of sections of albino and normal animals shows striking differences in the endocrine glands. The sectioned hypophysectoniized animals show no trace of the anterior lobe of the h>T)ophysis. That part of the floor of the diencephalon which normally abuts against the hypophysis, rests upon the floor of the cranium (fig. 2). This apparently demonstrates conclusively that the entoderm has not the intrinsic power to form a hypophysis. If it enters into the formation of the gland at all it must be considered as a tissue inclusion which became changed through its adaptabihty into glandular parenchyma, a conclusion previously drawn by the ^^Titer, Smith ('14). The infundibulum shows some structural modifications when compared to the checks, although the saccus vasculosus, as far as determined, appears to be normal. In the checks that region of the diencephalon which rests against the pars glandularis is EKFKCT OF ll'* I'til'HYSECTOMY UPON TIIK FUOG (>L "■"^v Fig. 1 A section throufih tlic hypophysial region of a 3S mm. normal tadpole. X 100. Fig. 2 A section through the hypophysial region of a 87 mm. albino. Note the much reduced pars nervosa. X 100. Fig. 3 A sagittal section through a lobe of the thyroid of a 38 mm. check. X 100. Fig. 4 A sagittal section through a lobe of the thyroid of a 37 mm. albino. X 100. 5 6 Fig. 5 A section through the epidermis, in the mid-brain region, of a normal 39 mm. check. The pigment granules are indicated by dots. X 200. Fig. 6 A section through the epidermis, in the mid-brain region, of a 38 mm. albino. A faint melanophore in the left part of the figure. X 200. 62 p. E. SMITH of considerable thickness, that is, in addition to the ependyma there is a rudimentary pars nervosa. Caudad to this the wall is formed almost entirely of ependyma. The pars nervosa is reduced throughout most of its extent to an ependymal layer in the hypoi^hysectomized animals. There may be a small localized thickening but nothing to correspond to the normal animal (figs. 1, 2). The thyroid shows marked modifications in the albinos. In the accompanying table the size of one lobe of the thyroid of a normal 38 mm. tadpole with 4.0 hind legs and of a 37.0 mm. albino with 0.1 mm, hind legs is given. Size in millimeters oj one lobe of the thyroid SS mm. check 37 mm. albino Length 0.6 Length 0.21 Width 0.3 Width 0.15 Thickness 0.16 Thickness 0.04 -#■ 8 *» Fig. 7 Photograph of an albino. X 2. Note the very small hind limb bud. Fig. 8 Photograph of a normal tadpole. Figures 7 and 8 were photograi)hed on the same plate' X 2. Fig. 9 Photograph of a normal tadpole. X 2. Fig. 10 Photograph of an albino. X 2. EFFECT OK m I't >l'in SK( I'c ).M V Ul'ON 'I'lIK FKOCJ G3 Tlic al)^)^■(' tahlc sliows tlial the thyroid of the all)iiio is ap])roximat('ly oiic-thii-d normal size. Tlic contrast is oven more strikinji wlicii the compactness and character of the parenchyiTm is noted. A sagittal section thronfi;li the thyroid of a :iS mm. cliecU sliows on an averago 12 to lo vesicles, many of which are larjicly distended witli colloid, the parenchyma of the whole gland Ixnng compactcMl togetliei'. A sagittal section through the thyroid of a hyp()])hysectomized 37 nnn. s])ecimen shows 6 to 8 atro})hied vesicles containing but a slight amount, or no colloid, and with large spaces between the vesicles. The cells making u]) tlie vesicles of the former are cuboidal and protoplasmicricli, in the latter little but the nuclei remain (figs. 3, 4). The results from exj^erimental feeding of thyi-oid by Gudernatsch and other workers suggests that the non-development of the hind legs in the albinos is due not to the hj^^ophysis but rather to the failure of the thyroid. In this connection the 36 mm. albino with 4.2 mm. hind legs, mentioned above, is of interest. Sections of this specimen sh(nv that the hypophysis was completely ablated but that the thyroid is normal. This specimen thus gives adcUtional evidence that the retarded development of the hind legs must be referred to the thyroid and not to the hypophysis. Also the reduction in pigment is not due to the atrophy of the thyroids. The modifications of the thyroid obtained by Adler ('14) were similar but less striking. An examination of a large number of male and female albinos and checks has, as yet, failed to show any constant variation from the noraial in the sex glands of the hypophysectomized animals. The sex glands of the albinos although varying considerably apparently do not exceed the limit of variation met with constantly in the normal animals. This conclusion stands in contradiction to the results previously adduced by the author and to the results of Adler ('14) in the hypophysectomized tadpole and to the conclusions of Hahn ('12) in the tadpole with hypertrophied hypophysis as well to the results obtained in mammals by pituitary feeding, notably that of Goetsch ('16). The WTiter wishes to express his appreciation to Dr. H. M. Evans for his generous aid. 64 p. E, SMITH LITERATURE CITED Adler, L. 191-1 jMetamorphosestudien an Betrachierlarven. I. Extirpation endokriner Drlisen. A. Extirpation der Hypophyse. Arch. f. Entw. mech. d. Or^anis., Bd. 39. BiEDL, A. 1913 Innere Sekretion. Zweite Aufi. Berlin. BoTEANO, E. R. 1906 Contr. la physiol. glandei pituitare la brosca. These, Bucarest, zeit. n. Paulesco, cit. f. Biedl. Caselli, a. 1900 Influence de la fonction de I'hypophyse suz le development de I'organisme. Riv. sper. di fren., vol. 37, cit. f. Biedl. Gaglio, G. 1902 Recherches zur la fonction de I'hypophyse de cerveau chez les grenouilles. Arch. ital. d. Biol., vol. 38, cit. f. Biedl. GoETSCH, E. 1916 The influence of pituitary feeding upon growth and sexual development. Bull. Johns Hopk. Hosp., vol. 27. Hahn, a. 1912 Einige Beobachtungen an Riesenlarven von Rana esculenta. Arch. f. mikr. Anat., vol. ,80. HoussAY 1910 La hypofisio de la rana. Trabajos de Labor de Univ. Nacional de Buenos Aires, cit. f. Biedl. Smith, P. E. 191-1 The development of the hypophysis of Amia calva. Anat. Rec, vol. 8. ANO.MALIES IN LOBATION OF LUNGS W 11 H REVIEW OF IJTKRATrRE CARBON' (ill.LASPIE, LEWIS I. MILLER, .VXD MORRIS BASKIX Anatoniiail Department, University of Colorado FIVK FIGURES Judging from tlio literature, abnormal lobation of the lungs is relatively rare. Lindsay ('10) di^'ided the abnormalities in the lobation of the lungs into two classes: those in which the normal number is decreased and those in which the number of lobes is increased. The former is due either to a deficiency of the lobes themselves, or to a deficiencj' of the fissures, which normally separate the lobes. The latter is due to an increase of lobes, or to an increase in fissures. Complete absence of lobation in definitely formed lungs is rarely if ever found, though its homologue is to be found in the Orang, with two lungs each existing as single lobes. Rokitansky ('61) showed that arrests of development may occur and that this may lead to complete absence or great deficiency of one or both lobes. This arrest may be so early that the lungs can scarcely be observed as small round bodies situated at the ends of the bronchi. This condition is generally due to contraction of the volume of the thorax. Pontif ('60) in 'Virchows Ai'chives' recorded a case in which the right bronchus was connected wdth an ovoid body, which was imbedded in gelatinous tissue and filled the right half of the thorax. According to Lindsay, cases of class two, that is those which have an excessive number of fissures, are occasionally encountered, and are probably the most common form of abnormahty. They do not appear to present any regularity, and lack the interest which is attached to accessory lobes. There are two 65 66 CARBON GILLASPIE, L. I. MILLER, AND M. BASKIN groups of accessory lobes. One which is of considerable developmental interest, is composed of completely isolated masses of pulmonary tissue formed between the diaphragm and the base of the left lung; occasionally on the right side such masses are found even in the abdominal cavity. Other masses containing arteries, veins, nerves, and bronchial tissue, but devoid of bronchi, are attached to the oesophagus, aorta or other mediastinal structures by a pedicle. These masses are apparently quite functionless. Two such cases are recorded by Vogel ('99), in each of which he found a deficiency in the bronchial tree. Simpson ('99) described a case of a deficient bronchial tree found in a foetus. In the cases with additional fissures there is a normally placed lung presenting an excessive number of lobes. These abnormalities are usually very definite in their position, and occur more commonly on the right side. Wrisberg ('77), who was the first to notice an accessory lobe in the human lung, recorded a most interesting and unique case of an accessory lobe on the left side produced by the left azygous vein; i.e., the superior intercostal vein which preserved its foetal condition and opened into the left innominate vein. Chiene ('76), described a pear-shaped supernumerary lobe, lying between the upper lobe of the right lung and the bodies of the dorsal vertebrae, having its origin from the angle formed by the junction of the upper lobe with the root of the lung. The supernumerary lobe was separated from the upper lobe of the lung by a double fold of the plenral membrane, which descended vertically for seven centimeters from the apex of the thoracic cavity where it was continuous with the pleura costalis. It enclosed in its free border the vena azygous, and formed the outer wall of the cul-de-sac, in which the supernumerary lobe was contained. The left side of the chest was normal; both sides were healthy. E. W. Collins ('88), recorded a case of an accessory lobe immediately above the posterior part of the root in the angle between it and the upper portion of the right lung. This accessory lobe was somewhat pyriform in shape, with a broad pedun .\\(>M \I.IKS IN I.OHA'rio.N OK I.INCS 67 cular at tac'liinciit . In all. ( (tlliiis was al)lc to collect st'X'cii cuses of {icccssory lohcs in lniiiiaii hui^s. A. I'. .Mai> land ('90) descrilxul ahiioiintditios in lohcs of three liiiijj;s. 'Vhv first was a rif»;]it lun^ with no iiuhcations of ;t middle lobe, l)ut a (h^Nclopnient of ;i tliird, or accessoi-y one, on the inner side of the lun^". Tlie second was :l left Inn^- with a sulxlixision of the iii)i)er lobe. The third was a right lunj; with an iiiconiplete separation of a normal middle lobe. ^Maryland states: "Cases of more than four right lobes and three left lobes are exceedingly rare. ' Patterson ('09-' 10) described a condition of two additional lobes. One of these was above the root, and separated from th(> n])i)er lobe by a fissure. This accessory lobe was enclosed within a pleural pouch, w^hich contained the vena azygous. The second additional lobe was below the root, and between the upper and middle lobes. Case I of the present specimens presents features entirely different from those hitherto described. It w^as obtained in the dissecting room during the term of 1915-1916 from a male subject, aged sixty-one. The cause of death as given on the death certificate was dementia. No clinical history was obtainable. Besides the anomalous condition of the lungs, there was a persistent thymus, and a number of anomalous arteries. On gross inspection the lungs presented no pathologic lesions. The left lung apex-base measured 22 cm., dorso-ventrally 20 cm. The normal fissure (FS), which separated the superior from the inferior lobe, was in its normal position, starting at the junction of the antero-inferior border 25 cm. from the apex, and running obliquely upward and backward, dividing the superior and inferior portions completely. The superior portion, however, presented two other fissures, as shown in the diagram, thus dividing this portion into three more or less distinct lobes. Fissure number one (Fl) started at the anterior border 7 cm. below the apex, and ran horizontally backward on the anterolateral surface of the lung for 6 cm. The depth of the fissure was on an average 1.5 cm. 68 CARBON GILLASPIE, L. I. MILLER, AND M. BASKIN Fissure number two {F2) started at the antero-median border 17 cm. below the apex and ran upward and backward for 9 cm. This fissure extended through the lung tissue, completely separating the middle from the inferior division of the superior lobe. Neither of the two mentioned fissures extended as far as the main fissure, which separated the superior from the inferior lobe. Three distinct divisions of the superior lobe were evident from an antero-lateral view. The upper division (L7) was pyramidal in F " 1 F,3 , •U ^ Fig. 1 Anterior aspect shape, forming the apex. It measured 8 cm. antero-posteriorly, and 7 cm. from apex to base. The middle portion {L2) was wedge-shaped, wider on the anterior border than on the posterior. It measured 9 cm. on the anterior border between the first and second fissures, and 17 cm. antero-posteriorly. The lower division {L3) is hngual in shape. This division measured 6 cm. on the anterior border and 21 cm. anteroposteriorly. The inferior lobe presented two more supernumerary fissures as did the superior lobe. .\.\(tM AI.IKS IN I.oHAIlo.N oi' UNCiS 09 Mssurc iiuiiilxM' I'diir (A'./) stiirlcd ,") cm. uiitcro-inlci'jorl^' I'roin tlic sii])(M()-iiitVriur (issiiic ;m(l i':ni ]);ii'all('l wilh (he above fissure tor a (lis(anc(> of 10 ciii. Tliis fissure also exten(l('(l llii-ou^h tile lunji' tissue se])ai'at iii^ the u])j)('i' IVoiii llic ])ostero-iiif('i-i()r divisions. Fissure iiuiiil)er lix'e (Fo) Avas on the lateral surface, extending' ohli(iuely, su])erioi'ly and inferiorly, and incompletely dividing' \ho ])()st(M-o-lat(M-al fi'oin tlie i)()stero-inferior divisions. 2 Fig. 2 Posterior asjiect The inferior lobe, which was triangular, also presented three fairly distinct di\dsions. The upper division {L4), which is oblong in shape, having a small tongue-like projection on the anterior border, ran obliquely upward, and backward, following the direction of the superoinferior, or great fissure. This division measured 5 cm. anteroinferiorly, and 19 cm. antero-posteriorly. The postero-inferior division (Ld) was more or less quadrangular in shape. It measured 8 cm. on the superior border and 12 cm. on the inferior border, i.e., at the base. The postero-lateral division {L6) was irregular in outline. The upper border was made by the supero-inferior fissure, and 70 CARBON GILLASPIE, L. I. MILLER, AND M. BASKIN the lower by a separate fissure between this division and the postero-inferior division. The right lung apex-base measured 19 em,, and dorso-ventrally 18 cm. In this lung the fissures were deei:)er, and went through the lung tissue, completely dividing the lung into distinct divisions. The normal fissure (F'S), which divided the apex lobe and middle lobe from the inferior, or base lobe, was a httle more irregular than the corresponding fissure on the left lung, Fig. 3 Right lung: lateral view and ran obliquely upward and backward, dividing the lung into an upper and lower division. The upper di\'ision presented an ir egular outline from a lateral view. There wei'e two large fissures; one which normally divided the apex lobe from the middle lobe, and the other a supernumerary fissure dividing the middle lobe into two. There was also a small fissure in the apex lobe. Fissure number one (F'l), which divided the apex lobe from the middle on , started 10 cm. from the apex laterally, midway between the apex and the base. ANOMAMKS IN l.( )HA'n( ).\ OF IJ'NCS 71 Plssiirc lumihci' two (/'"J), which (hxidcd the middle lohc into two (h\'isioiis, i-;iii \ crt ically for ;i (Hstaiicc of S cm., slartiiif^ from tJie base and nmiiiiif;" towurd tlie iipvx. The ai)ox lobe (/>'/) was ([uadrilateral in slia])e, Juiving' the upper border narrower than the base and nieasiirinjij 9 cm. ai)exbase, and 8 cm. antero-posteriorly. The inferior di\isi()n (L'2) of tlie n]i])er lobe was lingual in shape, extcMidiiig obli(iuely upward and backward 13 cm. in its long direction, and measuring 4 cm. a])ex-base. ■VVSI«*«M»» 4 Fig. 4 Left lung: lateral view The inferior division (L'S) of the upper half of the lung was elongated in outline measuring 4 cm. in antero-posterior direction, and 13 cm. apex-base. The lower half of the right lung was triangular in shape, having two clearly visible fissures which ran through the lung tissue and divided this portion of the lung into three distinct lobes. There are also two smaller fissures in the inferior border of the lung, merely forming small tongue-like lobules. The fourth fissure {F'4) started 10 cm. from the apex midw^ay between apex-base, running downward and backward for a dis rz CARBON GILLASPIE, L. I. MILLER, AND M. BASKIN tance of 14 cm., and completely separating the upper from the lower division of this half of the right lung. The fifth fissure {F'5) started from the base, or inferior border, and ran vertically upward for a distance of 10 cm. It extended through the lung tissue, dividing this inferior half into two divisions, a posterior and an antero-median. The superior division {L%) of this inferior half of the right lung was triangular in shape, measuring 13 cm. on its inferior border, 8 cm. on its superior border, and 11 cm. on its posterior border. The postero-inferior division {L'5) was quadrilateral in outline, measuring 8 cm. on its shortest border and 14 cm. on its longest, or posterior border. The antero-median division {L'6) had the form of an elongated triangle, measuring 11 cm. on the anterior border, 6 cm. on the median posterior border. At the base of the right lung was a very irregularly shaped division distinctly separate from the rest of the lung as shown ANOMALIES IN LOHATION OK LUNGS 73 in tJie diagram. 'I'liis basal clivlsion (L 7) iiicasunHi 10 cin. in the long direction and 6 cm. in the short direction. This division was separated from the rest of the lung by two distinct fissures {F'6-7) which started at the postero-inforioi- border, and ran for 9 cm. anteriorly and backwards, Case 11 was obtained in the dissecting room during the term 1915 -191(), from a female subject, aged thirty-eight. The cause of tleatli as given on the death certificate was 'septic meningitis.' No clinical history was obtainable. The left lung apex-base measured 20 cm., dorso-ven trail y 18 cm. The right lung apex-base measured 18 cm., dorso- ventral ly 17 cm. The left lung presented upon examination one accessory lobe, one accessor}^ lobule, and three accessory fissures in the inferior portion of the left lung. The right lung presented upon examination two accessory lobes and two accessory fissures. The first accessory lobe was in the upper portion of the inferior division of the right lung. The other accessory lobe was a basal lobe and was identical with the basal lobe seen in the right lung in Case I. X-Ray^ pictures of Case II show separate bronchi going to each of the main lobes including the accessory lobes, as may be seen by referring to the plates. X-Ray views of Case I were unsuccessful on account of a poor bismuth injection. From the foregoing description it is evident that these specimens show no distinct azygous lobe, which is the common accessory lobe described. These lungs retain their normal shape; but they present, besides the normal fissures, a number of accessory fissures which divide the lungs into distinct accessory lobes. Complete absence or deficiency of one or both lobes may be due as Rokitansky points out to arrests of development as contraction of the volume of the thorax. Supernumerary lobes have been variously explained. Lindsay ascribes a slight adhesion of the lungs to the thoracic wall as the cause of the super 1 Stereoscopic X-ray plates show the separate bronchi as described. Reduced prints fail to show the necessary details and, therefore, are not reproduced here. THE ANATOMICAL RECORD, VOL. 11, NO. 3 74 CARBON GILLASPIE. L. I. MILLER, AND M. BASKIN numerary lobe, or possibly an undue curvature of the embryo, so that the Venae Cavae, as it bent down to a position at right angles to its original position, instead of slipping behind the pleura and lung, dragged down a fold of the former and deeply notched the latter. Fischer remarks that it is difficult to conceive of the azygous lobe without an already preexisting anomalous course of the azygous, and makes no attempt to explain the phenomenon. Collins explains the azygous lobe as a persistent foetal condition of the left azygous vein. All previous attempts to explain the origin of the supernumerary lobes apply only to cases in which there is a definite azygous lobe. The anomalous course of the azygous vein has constricted off a portion of the lung tissue, thereby forming a new lobe. This explanation does not apply to cases in which there are no azygous lobes and in which, however, there are other supernumerary lobes, as the azygous lobes are not in reality true lobes. A true lobe as applied to the lung indicates a separate bronchus. Whether the azygous lobe has a separate bronchus, or not, has not been stated in their descriptions. Considered from the viewpoint of their origin, a constricted portion — such as the azygous lobe — is not a separate entity, but a part of the mother lobe from which it has separated. The formation of lobes of the lungs has first been studied by Aeby and was worked out by Narath. Aeby defines lung lobe as follows: A true lobe is never supported by more than a single bronchus and therefore includes no portion of the stem bronchus." Soon after the formation of the first lateral bud in the embryo, each bud becomes marked out upon the surface of the mesodermal anlage of the lung. Before development of lateral bronchi, the surface of this anlage is smooth. Later it becomes almost mulberry-shaped, and secondary elevations are then formed by the budding of the bronchial bud which it contains. The process goes on until the surface becomes covered with fine granules. These disappear with further growth of the Jung — only the first formed furrows persisting normally. AXO.MALIKri 1\ LDHAIION oi' LUNCJS /O Accessory lobes may then be tliie to a retention of tlie foetal condition in which not only the primary divisions persist, but on account of the rather slow growth of the bronchi, the secondary conditions persist — that is to say, accessory lobes are not new formations. Init represent a stage in the normal development of the lung. Accessory fissures dividing the lobe into a number of lobules — as exists in the present specimens — may have been due to folds of splanchnopleura which have been carried down and caused the formation of a groove, lined with pleura, or they may have been caused by incomplete obliteration of the secondary divisions, which are ])roduced in the mulberry stage and which normally disappear. BIBLIOGRAPHY Chiene 1876 Jour. Anat. and Phys., vol. 4. Clel.\nd 1861 Jour. Anat. and Phys., vol. 43. Collins, E. W. 1888 Royal Irish Academy. Fischer 1898-1899 Anat. Anzeig. Keib.\l and jSIall Human Embryology. LiNDS.\Y, M. A. 1910 ^Maritime Med. News, vol. 22. Maryland, A. E. 1890 Jour, of Anat. and Phys. McAllister Text Book of Anatomy, p. 340. Paterson 1909-1910 Jour. Anat. and Phys., vol. 44. PoNTiF 1860 Virch. Arch., vol. 50. Simpson 1899 Jour. Anat. and Phys., vol. 42. Wrisberg 1874 Trans, for the Royal Irish Academy. ANOMALOUS RENAL VESSELS AND THEIR SURGICAL SIGNIFICANCE CARBON GILLASPIE, LEWIS I. MILLER AND MORRIS RASKIN Anatomical Deparlmcnt, University of Colorado NINE FIGURES Abnormalities of the renal arteries occur more frequently, perhaps, than anomalies of any of the other larger vessels. In view of the enormous number of investigations of the different structures of the kidneys recorded in the literature on the subject, it seems strange that only scanty information exists concerning the actual course of the larger blood vessels, and their relations to the pelvis of the kidney. The normal, as well as the abnormal, arrangement of the renal vessels at the hilum is known; the microscopic picture of the vessels in the cortex and pyramids are likewise thoroughly familiar to every student; but as to the form of the pelvis, and the actual course and distribution of the larger vessels around its walls, very vague ideas still prevail. Professor Thane states that irregularities of the renal arteries are met with in about 25 per cent of cases, and that the most comimon irregularity is the presence of an additional vessel in about 20 per cent. Young and Thompson report four cases of anomalous renal arteries: the first is that of multiple renal arteries and malposition of the right kidney; the second that of multiple renal arteries, malposition, and malformation of both kidneys; the third that of a horseshoe kidney with multiple renal arteries; the fourth that of multiple renal arteries, two of which were on the left side, two on the right. In the last case there were also multiple spermatic arteries, and two renal veins on the right side, while the left side was normal. 77 /8 CARBON GILLASPIE, L. I. MILLER, AND M. BASKIN Irregularities in renal vessels have also been mentioned by McAllister ('83) who found anomalous renal arteries in 43 per cent of the cases examined. Levings ('12) reported two cases of anomalous renal vessels going to the lower poles. Harvey ('14) described a case of multiple renal arteries. The specimens of the present description were obtained in the dissecting room through the help of Dr. E. B. Trovillion, Instructor in Anatomy, University of Colorado; and from autopsies through the courtesy of Dr. R. C. Whitman, Professor of Pathology, University of Colorado. We were able to examine, in all, 33 cases of which 22, or 73 per cent possessed anomalous renal vessels. Case I. The kidneys were normally placed in the abdominal cavity. Both were somewhat smaller than normal. The right showed two small cysts on its anterior surface. It lacked the normal large renal artery, but possessed instead two renal arteries of equal size, which were almost as large as the normal vessel should have been. The lower artery arose from the ventro-lateral portion of the aorta, 6 cm. above its bifurcation, and passed to the lower pole where it divided into two vessels, 2 cm. before it entered the kidney substance. . The second artery arose from the aorta 6 cm. above the lower one, and split into two vessels 2 cm. before it reached the kidney substance. There were also two large renal veins, closely accompanying the arteries; one arising from the upper, the other from the lower pole of the kidney. Both veins emptied into the vena cava. The left kidney possessed three renal arteries and one renal vein. The lowest renal artery arose from the left ventral portion of the aorta 4 cm. above the birufcation of the aorta, and supplied the lower pole of the kidney. The second artery arose from the aorta 6 cm. above the first, and supplied the upper pole. The third arose 1 cm. above the second, crossed it and entered the kidney at the hilum. One large renal vein arose from the kidney at the hilum. Case II. Both kidneys were larger than normal, and in the proper position. The right possessed one renal artery and one renal vein. The left kidney, however, possessed three main renal arteries, and one renal vein. The first renal artery, which seemed to be the normal vessel, arose from the ventro-lateral aspect of the aorta just opposite the origin of the superior mesenteric artery. The second renal artery arose 1 cm. above the first. This was a long slender artery having a tortuous direction, and entered the kidney 4 cm. above the inferior border. The third artery was the smallest, and arose from the aorta just lateral to the second. This artery broke up into thi-ee smaller branches, all of which supplied the upper polo of the kidney. The; veins were normal. woMAi.ors I {i:\.\L \ KssELs 7'.) ('.•ISC III. rilis picscritcd two normally placed Uidiicys, (lie ri^litIxMiip; sniallcr than the left. The riu,lit kidiicN' presented upon examination two renal aiteiies and one renal vein. The lower icnal ail(uy arosj' 'J cm. al)ov(> the hifincation of the aorta from its ventral aspect, and entei'cd the kidney at the lower pole I cm. al)o\c the inferior border-. The iip|)er icnal arlei_\-, which was the normal one, arose; from the dorsal aspect of the aorta 1 1 cm. above the bifurcation of the latter, and, aftei' runnin<2; a tortuous coui'se, (Mitered the kidney at \ho. hilum. The left kidney was normal. Two spermatic xcins emptied into the left renal \-ein. Case I\'. This presented upon examination two normally placed kidneys. The li^ht [jossessed two arteries; the normal one arose from the aoi'ta, and entertMJ the kidney at the hilum; the anomalous one arose from the ventro-latcM-al aspcH't of th(> aorta 5 cm. above the normal renal vessel, and entered the upper jiole after it had si)lit into two branches. The left kidney also possessed two arteries; the normal one coming from the ventro-lateral aspect of the aorta just opposite the normal rifiht I'cnal arteiy. and enterin^j; the kidney at the hilum; the accessory renal ai'tery branchin<;' from the superior mesenteric, and entering; the upper pole of th(> left kidney 3 cm. l)elow the superior l)order. The veiiis in both kidneys were normal. Case y. This sul)ject presented two normally placed kidneys of normal size. The right kidney contained a small anomalous artery arising from the aorta, and supplying the upper pole. The left kidney was normal, as were also the veins. Case VI. This presented two normally placed kidneys. The right contained four lai-ge renal arteries and two large renal veins. The first renal artery aros efrom the ventro-lateral aspect of the aorta 3 cm. above the bifurcation, and entered the kidney on the posterior surface of the lower pole 3 cm. above the inferior border. The second and third arteries took origin from the aorta at the ventral surface as a connnon large branch 1 cm. above the bifurcation but immediately divided into two rather large branches, which entered the kidney at the hilum. The fourth artery was a long slender one coming from the coeliac axis, and entering the kidney at the upper pole 4 cm. below the superior border. There were two renal veins present ; a large one entering the vena cava, and another somewhat smaller one also entering the posterior aspect of the vena cava. The left kidney possessed one accessory artery arising directly from the aorta and entering the kidney substance at the lower pole. Case VH. This was a case of a right kidney with a small accessory artery arising just above the normal renal artery, and entering the kidney at the upper pole 3 cm. below the superior border. Case VIII. In this case the left kidney was somewhat larger than normal, and contained three large renal arteries, one of which divided into three smaller branches. The first artery arose from the aorta, and entered the kidney at the lower pole, 4 cm. above the inferior 80 CARBON GILLASPIE, L. I. MILLER, AND M. BASKIN bordei'. The second artery was the largest of the three. It arose from the aorta above the first, and entered the kidney at the hihim. The third artery arose from the aorta above the second, and divided into three smaller branches all of which supplied the upper pole. Two large renal veins were present which arose from the hilum and entered the vena cava. Case IX. Both kidneys were normal in size and position. On the right side one large arterial trunk sprang from the aorta. This divided into two rather long slender branches after it had continued its course for 2 cm. Each of the two branches further divided into two other branches, thus four arteries entered the kidne.v: two at the upj^er pole 3 cm. below the superior border; the other two at the lower pole 4 cm. above the inferior border. Connecting the main renal trunk with the aorta was a plexus of vessels of varying sizes. This plexus, covering the walls of the aorta, appeared to be a persistence of the embryonic periaortic plexus. There was one renal vein which came from the hilum and entered the vena cava. The left kidney possessed two large and two small arteries. The first, which seemed to be the normal one, arose directly from the aorta and entei'ed the kidney at the hilum. The second artery, whidi was a lorig sleilder vessel, came from the aorta and entered the anterior surface of the kidney 4 cm. below the superior border. On the left side two small arteries came from a plexus which resembled the periaortic plexus. On this side, these small arteries entered the kidney substance, while on the right side the plexus merely connected the large renal vessels with the aorta. There were also two large renal veins, two smaller veins and a plexus of still smaller veins. This plexus connected the larger veins with the vena cava. The largest vein, which seemed to be the noi-mal one, came from the hilum of the kidney and entered the vena cava. The second large renal vein, which was a long slender one, connected the left spermatic vein with the anterior surface of the kidney. The plexus not only connected the vena cava with the large renal veins, but also formed an anastomosis between this venous plexus and the arterial plexus, so that the venous and arterial blood had a chance to mix. Cases X, XI, XII, and XIII. These four subjects had kidneys which were similar in most respects. Thei'e were two renal arteries arising from the aorta and entering the hila of the right and left kidneys in each case. The veins were normal. Cases XIV, XV and XVI. These three cases possessed small accessory arteries arising from the aorta and entering the poles of the kidneys. In cases fourteen and fifteen the accessory arteries entered the lower poles, and in case sixteen the small arteries entered the upper pole. Case XVII. This case possessed four arteries to the right kidney and thi-ee to the left, all of which arose dii'cctly from the aorta. There were also two renal veins from the left kidney, both of which arose from the hilum and cntcj-ed the vena cava. AN'OMALOl'S UKN'M. \ KSSELS 81 Case X\'IIT. The lifjht kidney possessed two snudl renal arteries eoniinji; directly- fioin the aorta and enteiiiiK tlie kidney at the hilum. The left kichiey as well as the veins were normal. Case XIX. Both kidneys posse.ssed three renal arteries rather small in size, whieh entered the kidney at the hilum. The veins were normal. Case XX. The rijiht kidney was normal. The left po.sscsscd two renal arteries, both of which were branches of the aorta, and entered the kidney at the hilum. The veins were normal. Case XXI. The right kidney possessed two renal arteries, one of which entered the supiMior, the other the inferior pole. The left kidney possessed three renal arteries, which arose fiom the aorta as separate branches, and entered the kidney at the hilum. The veins were normal. Case XXII. The right kidney possessed three renal arteries, arising from the aorta and entering the kidney from the hilum. The left kidney possessed two renal arteries, one arising from the aorta and entering the kidney at the hilum, the other arising from the coeliac axis and entering the kidney at the superior pole 2 cm. from the upper border. The veins wove normal. The embryological origin of the renal arteries has not yet been satisfactorily explained. The explanation as here offered is after Keibel and Mall. His ('80) first observed multiple branches of the aorta supplying the mesonephi'os in 7 mm. embryos. A more extended account of them has been given by Broman. At first, when the Wolffian bodies are relatively small, the mesonephric vessels are correspondingly small. They come from the middle portion of the aorta (2nd to 8th thoracic). At the end of the first month the mesonephros reaches its greatest development. It receives many direct branches from the aorta at levels cranial as well as caudal to the original ones. In 8 mm. embryos there are twenty mesonephric arteries on each side (8 cervical to 12th thoracic segments). The last vessels to appear grow out from the region between the first and second lumbar segments. These are destined to persist as the remainder atrophy. There are then on each side^ — in maximo — thirty vessels distributed thi'oughout the entire mesonephric area. At first these are entirely distributed to the mesonephros, but later also supply the reproductive glands, suprarenal bodies, metanephi'oi, and diaphragm. These new regions of distribution prevent their com 82 CARBON GILLASPIE, L. I. MILLER, AND M. BASKIN plete degeneration when the mesonephros disappears. A variable number of them persist as phrenic, suprarenal, renal, accessory renal, internal spermatic, accessory spermatic arteries, and as the rami ad lympho-glandulas and ad sympathetic um. The first mesonephric arteries found in embryos 5.3 mm. arise from the lateral surface of the aorta, and pass horizontally to the urogenital fold, reaching the malpighian corpuscles, and terminating in them with an enlargement which usually assumes a spherical shape. Later a network of vessels occupies the place of the enlargement. This network is also connected with the Fig. 1 Mesonephric arteries in human embryo of 18 mm. greatest length. Circles on anterior surface of aorta indicate origin of coeliac, superior and inferior mesenteric arteries (after Keibel and Mall). posterior cardinal vein. Case IX is an example of the persistence of this anastomosis. With increasing age the arteries continually recede into the lumbar segments disappearing from the thoracic ones. The arteries are di\T.ded into three groups bj'^ the suprarenal body: the cranial group, which is dorsal to the suprarenal {1-2); the middle group, whose vessels pass through the suprarenal {R. 3-~4, L. 3-5); and the caudal group whose vessels pass over the ventral side of the suprarenal body {R. 5-6, L. 6-9). The mesonephric arteries {5-9) situated in the angle formed by the reproductive gland ventrally, the mesonephros laterally, and the ANOMALOUS HHXAL VESSELS 83 niotano])lin)s dorsally, form ;i network: tlio rote iirtoriosum urogiMiitalo. 'I'lio ines()ii('])liros, rei)r()tluctivo gland, and the nietanephros are siip])lie(l witli aiterial branches from this network, thus makinfi; tliese ()r}i;ans indei)endent of shigle branches for their blood sui)i)ly. Should one or several roots degenerate, neighboring arteries can take their places. In the above diagram, for instance, the second mesonephric artery has divided into an ascending and descending branch; the ascending one sui)plies tlie entire ii])per lialf of the mesonephros and the reproductive gland, a region that in the young embryo receives its blood from several mesonephric arteries belonging to more cranial segments. Tlie occurrence of this network at once explains why all persistent arteries that arise from the roots of this network show, within certain Umits a variability in the points of their origin from the aorta. Each of the nine to eleven remaining mesonephric arteries may become an internal spermatic artery, since all supply the reproductive glands. This explains the frequently observed multiplicity of these arteries, and the not infrequent difference in the place of origin of the right and left ones. The renal arteries are not new formations, as some have claimed, but each is formed from a mesonephric artery. The kidney climbs upward to the mesonephric artery and as soon as sufficient blood supply is assured cranially, the caudal branches separate from it. ^\Tien the kidney has acquired its definitive position it possesses several arteries, and of these one becomes greath^ enlarged to form the definitive arterj-, while the others either degenerate, or persist as accessory renals. The definitive renal arterj'^ is either the last vessel of the second group, or first of the thkd group. The relations of both groups, i.e., of the second to the suprarenal artery, and of the third to the internal spermatic, explain the variation in which the renal artery arises from a suprarenal or from an internal spermatic. In the first case it may be the principal stem; in the latter, only an accessory renal. The relations between the urogenital rete and metanephros show how the accessory renal arteries ma^^ develop, and explain their varied relations to the kidney. Accessory renal 84 CARBON GILLASPIE, L. I. MILLER AND M. BASKIN arteries from the fii'st group will be branches of the superior suprarenal, and must pass over the dorsal surface of the kidneys. These consequently first reach the kidney on its dorsal surface and there penetrate its cortex. Those from the second group will be branches of either the middle or inferior suprarenal, and will reach either the hilus or the kidney or the medial edge above this. Those from the third group may enter the hilus at the medial edge below it, or on the ventral surface of the caudal half of the organ. Should a caudal branch be retained, it will be drawn upwards by the migration of the kidney. This condition explains why such an artery may cross the principal stem or another accessory. Anomalous renal vessels are not only interesting from a purely scientific point of view, but are also of very great significance from a clinical and surgical standpoint. The Mayos in 20 out of 27 cases operated upon by them for hydronephrosis, found anomalous blood vessels. The obstruction in each case was caused by the blood vessel crossing the uretero-pelvic juncture. The vessels passed to the lower pole of the kidney, and varied from the size of a knitting needle to that of the radial artery. Levings, in a report of four cases on which operations were performed, found anomalous arteries which crossed the ureter, and to which condition he attributed the clinical symptoms of the patients. He also reported post-operative improvement in every case. Rupert in 1913 found anomalous renal vessels in 35 out of 50 cadavers studied. In every case the kidneys were normally placed, and of normal size and shape. He concluded that the usual percentage given is too low, and that anomalous renal veins, while not as common as the arteries, do occur, and that on account of the thinness of their walls and lack of pulsations they increase the hazards of kidney operations. It is very evident from the specimens at hand that a number of these arteries might be overlooked if it were necessary to remove or examine any of these kidneys. In view of the fact that anomalous renal arteries are important it is rather strange ANOMALorS i{i;nai, \ ksskls 85 i. 4 86 CARBON GILLASPIE, L. I. MILLER, AND M. BASKIN that their existence has been neglected in surgical anatomy teaching. As Rupert points out, the most common text books of anatomy and surgery make but brief mention of this condition, and some do not even refer to it at all. The accepted percentage of 20 to 25 as given by Quaine and Gerrish is evidently too low. Senator as recently as 1905 made the statement that RedupUcations of one or both renal arteries is a rare condition and may be dismissed." Rupert found 35 cases or 70 per cent of anomalous renal vessels, while in this present investigation there are 22 cases out of 33, or 73 per cent. In all of these cases the kidneys were normally placed and the anomalous arteries were so , arranged that they would easily complicate surgical procedures. BIBLIOGRAPHY Brodel, M. 1901 Johns Hopkins Hosp. Bull., Baltimore. Broman, F. 1906 Ergeb. der Anat. U. Entw., Bd. 16. Bremer, J. L. 1912 Am. Jour. Anat., vol. 13, no. 2. Clark, E. R. 1912 Am. Jour. Anat., vol. 13. Harvey, R. W. 1914 Anat. Rec, vol. 8. Hill, E. C. 1905-1906 Johns Hopkins Hosp. Bull., vol. 16-17. Jeidell, H. 1914 Anat. Rec, vol. 5. Keibal and Mall Human Embryology. Lewls, F. T. 1902 .\m. Jour. Anat., vol. 13. Lewis, Papez 1914 Abstracts Proceedings, Am. Jour. Anat. Levings, a. H. 1912 Wis. Med. Jour., March. PoHLMAN, A. G. 1905 Johns Hopkins Hosp. Bull., vol. 16. Rupert, R. R. 1913 Jour, of Gyn. and Obst., vol. 17. Wilson, L. B. 1912 Collection of Papers, St. Mary's Hosp. YoxjNG and Thompson 1903 Jour. Anat. and Phys., vol. 38. SIZE AM) LENGTH RELATIONS OL THE RIGHT AND LEFT TESTES OF PIGEONS IN HEALTH AND DISEASE OSCAR RIDDLE Carnegie Station for Experimental Evolution The right ovary luulergoes an early and more or less complete atroplw in most species of birds. Etzold ('91) has shown that in the sparrow the left testis is larger than the right. Firket ('14) and Swift ('15) have shown that in the chick embryo there were more primordial germ cells in the left gonad, and that this gonad is there also distinct^ larger than the right. Allen ('07) found that the sex cells were unequally distributed to the two gonads of the turtle, the left receiving most. In this form only 24-70 per cent of the sex cells ever enter the gonads. Our own accumulation of data on the size and length relations of the two testes of young and adult pigeons show a very decided predominate number of larger right testes; and also a distinct difference in shape of the two glands — the left though actually smaller in size is usually absolutely longer than the right. Changes in the size relation in birds dead of certain diseases — particularly tuberculosis — and in hybrids are also suggested by our data The meaning of this pronounced inequahty in the distribution of the primordial germ cells which is plainly associated with a larger left embryonic gonad, and the finding in adults of two groups of birds of a marked and nearly constant larger gonad, but this a different gonad in the two cases, is by no means clear. But, whatever this may mean, it is probably a situation of importance to the theory^ of sex. We present our present data then with the confession that on the main points the meaning is not clear, but with the conviction that they are not less valuable because of our present inability to clarify the puzzling situation, and hopeful that the data may stimulate the further 87 88 OSCAR RIDDLE accumulation of facts from enough forms, and of such varied kinds, as may lead to a better understanding of the embryonic and adult inequalities of the sex glands of birds. An examination of our data has shown that the measurements of glands of healthy birds should be grouped apart from those dead of disease; and those of pure species should be separated from hybrids. The justification of these separate groupings will appear later. The relative size of the sex glands in healthy common pigeons The weights of 31 pairs of testes from healthy common pigeons are recorded in table 1. In 27 of these pairs the left testis was the smaller; in 4 the left was the larger. In two or three of these latter cases — 12, 15 (22?) — the disparity of the two glands is so great as to make it clear that the smaller gland was wholly abnormal. In healthy common pigeons the right testis is larger than the left in a high proportion of cases. Size relations of the testes of common pigeons dead of disease In tables 2 and 6 the weights of 9 pairs of testes are given. In 7 of these the left gland was the smaller. It was larger in two instances; in one of these irregular cases, the smaller gland was again quite abnormally proportioned in reference to its larger associate Size relations of testes of pure species, healthy and dead of disease The testes of only 5 healthy birds of pure species (dead of cold, exposure, accident) are included in table 6. In all of these cases the right testis was the larger. The data for 46 individuals of pure species dead of disease are available. In table 2, 9 of the 10 individuals listed had larger right testes; the tenth had the two glands of equal size. Eleven further comparisons are supplied in table 3. Of these, 7 right testes are larger, 2 are smaller, and 2 are the size-equivalents of the left. Table 6 gives the data for 30 additional pairs. Of these, 7 of St. risoria all had larger right testes; 2 of T. orientalis both had larger right testes; 13 of Spil. tigrina — mostly not TESTES OF llGEONS IN HEALTH AND DISEASE 89 mature birds luul, ."> larger, 5 siiiall(>r, and '.] ('([uivalont right testes. Four inisccllaiicous birds here had 2 larger and 2 smaller' rig! it testes. Weight of right unit left testes of heallhj/ common pigeons 3 4 5 6 7 8 9 10 11 12 13 14 15 16 April 5. April 5. April 5. April 5. April 7. April 7. April 7. April 7. April 7. April 9. April 9. April 9. April 9. April 9. April 9. April 9. 0.885 0.515 1.475 0.845 1.185 0.990 1.055 0.820 1.410 0.765 1.190 0.975 1.260 1.000 1.235 0.945 1.280 1.225 1.075 0.900 1.025 0.710 0.025 0.715 1.460 1.390 1.010 0.720 0.275 1.425 1.125 0.500 PER CENT OP DIFF. -71.8* -74.6 -19.7 -28.7 -84.3 -22.1 -26.0 -30.7 - 4.5 -19.4 -44.4 +2760.0 - 5.0 -40.3 +418.2 -125.0 April 9. July 5. . July 5. . July 5. . July 5.. July 5. . July 9.. July 9.. July 13. July 13. July 16. July 18. July 20. July 20. July 21 (juv.; WEIGHT PER CENT OF DIFF. R = 1.115 L = 1.015 -9.9 R = 1.010 L = 0.750 -34.7 R = 1.220 L = 1.375 + 12.7 R = 1.140 L = 0.970 -17.5 R = 1.158 L = 0.765 -51.4 R = 0.370 L = 0.720 +94.6 R = 0.855 L = 0.800 - 6.9 R = 1.643 L = 1.030 -59.5 R = 0.571 L = 0.540 - 5.7 R = 0.820 L = 0.631 -29.9 R = 1.820 L = 1.500 -21.3 R = 1.600 L = 0.536 -198.5 R = 0.051 L = 0.040 -27.5 R = 1.390 L = 1.085 -28.1 R = 0.0004 L = 0.0003 -33.3 Left smaller in 27; larger in 4. In calculating percentage differences in these tables the smaller gland is considered as equal to 100 per cent. ' In both of these cases where the right testes weighed less than the left it will be seen that both testes were quite small — so small as perhaps to raise a question as to the reliability of the weights. 90 OSCAR RIDDLE Size of testes in healthy specific hybrids In tables 4 and 5 the data for 30 healthy young hybrids are given. The very, small gonad size of most of these young birds TABLE 2 Weights of right and left testes of various pigeons (classified) dead of disease PER CENT OF DIFF. PER CENT OF DIFF. 1. Common pigeons 32 33^ July 28 September 1 , R = 0.122 L = 0.460 R = 0.580 L = 0.415 +277.0 -39.8 September 28 December 14. = 0.037 = 0.034 = 1.300 = 0.850 (?) -52.9 2. Blond and white wing doves and their hybrids Hybrids, (specific) Pure 36* 37* 38* 39 40 41* June 10 June 27 July 23 September 12 October 17.. October 31 . . . 0.417 0.305 0.050 0.035 0.032 0.046 0.765 0.705 0.0053 0.0067 0.090 0.090 -36.7 42* -42.9 43* +43.8 44* -8 5 45* +26.4 46 = 0.0 47* April 9 April 17 September 26 October 25... October 25 . . October 28... 0.045 0.025 0.060 0.050 0.055 0.055 0.068 0.056 0.046 0.033 0.030 0.027 -80.0 -20.0 0.0 -21.4 -39.3 -11.1 c . Other hybr ds (spec ific, ex. 51, 52 = gen.) 48 49 50 August 28.. June 10.... August 28.. ■{ ■{ ■{ R = 0.845 L = 0.600 R = 0.190 L = 0.160 R = 0.133 L = 0.115 -40.8 -18.8 -15.7 51* 52 53* September 18. I September 23. | September 27. < R = 0.580 L = 0.510 R = 0.040 L = 0.031 R = 0.010 L = 0.012 -13.7 -29.0 +20.0 4. Other pure species 54^ 55^ August 26 November 10 R = 0.040 L = 0.022 R = 0.033 L = 0.023 -81.8 -43.5 56 57^ October 31. November 9. 0.465 0.445 0.015 0.013 4.5 15.4 Tuberculosis found. TESTKS OK PIGEONS IN HH.VLTH .\N1) DISEASE 91 T.\ni.i:3. Weights of testes of doves (classified) dead of disease March 29 to November 25, 1915 PER CENT or DIFF. .NO. PER CENT OV DIPF. Pure species Specific hybrids 59* 60* 61* 62* 63 64* 65 66* 67* 68* 69* March •_".». ... | April 1 I April 2 \ April 16 < September 2. < September 5. . < September 11 f (Juv.) \ September 14 < September 29 < October 3. . . . < November 11. < R = 0.012 L = 0.018 R = 0.008 L = 0.005 R = 0.105 L = 0.105 R = 0.022' L = 0.060 R = 0.1582 L = 0.146 R = 0.032 L = 0.020 R = 0.003 L = 0.002 R = 0.023 L = 0.023 R = 0.138 L = 0.116 R"= 0.045 L = 0.030 R = 0.082 L = 0.067 +50.0 -60.0 = 0.0 + 172.7 - 8.2 • -60.0 -50.0 = 0.0 -18.9 -50.0 — 22.4 76* 77* 78* 79* 80 81* 82* 83* 84* 85* 86* 87* 88 May 1 1 July 23 I July 30 1 August 22 — \ August 29.... \ September 19 < September 28 \ October 7. . . . { October 19 f (Juv.) 1 October 24 . . . < November 21 f (Juv.) 1 November 22. < November 25. < R = 0.030 L = 0.031 R = 0.262 L = 0.212 R = 0.040 L = 0.032 R = 0.076 L = 0.077 R = 0.6052 L = 0.454 R = 0.188 L = 0.152 R = 0.026 L = 0.026 R = 0.025 L = 0.021 R = 0.007 L = 0.006 R = 0.0.30 L = 0.030 R = 0.015 L = 0.020 R = 0.014 L = 0.016 R = 0.018 L = 0.015 + 3.3 -23.6 -25.0 + 1.3 -33.3 -23.7 = 0.0 -19.0 -16.6 - 0.0 +.33.3 Generic hybrids + 14.3 April 2 1 May 30 | June 20 < July 16 1 August 2 < August 17. . . . < R = 0.140 L = 0.125 R = 0.320 L = 0.280 R = 0.082 L = 0.076 R = 0.300 L = 0.298 R = 0.012 L = 0.010 R = 0.040 L = 0.037 -12.0 -14.3 - 7.9 - 0.7 -20.0 - 8.1 -20.0 70* Common pigeons 71* 72 73* 74* 75* 89 90* 91 April 28 1 August 25 ... . < April 5 \ R = 1.260 L = 1.035 R = 0.098 L = 0.087 R = 1.105' L = 0.990 -21.7 -12.6 -11.6 Tuberculosis found. 'The left suprarenal wholly involved in a tubercle nodule weighing nearly 1.0 gr. 2 Healthy. ' A hybrid from a family cross. 92 OSCAR RIDDLE TABLE 4 Weight, length and ividth of testes of young Ring dove (specific) hybrids — killed PER CENT PER CENT so. DATE '• WEIGHT OP DIFFERENCE LENGTH AND WIDTH OF DIFFERENCE 88a December 4, 1915. / I R = 0.035 L =0.030 -16.7 8.0x3.8 S9a December 4, 1915. / I R = 0.007(?) L = 0.005 (?) -40.0 5.1 xl.6 .6x 1.4 -10.9 90a December 4, 1915. R = 0.004(?) L = 0.005(?) R = less than +25.0 4.1 xO.9 4.9x0.9 + 19.5 91a December 4, 1915. < 0.005(?) 5.1 xO.8 L = 0.005(?) + ? 5.1 xl.5 = 0.0 92 December 4, 1915. { R = 0.005(?) L = 0.005(?) = 0.0 4.8x 1.5 5.3 X 1.4 + 10.4 93 December 4, 1915 7 I R = 0.007 (?) L = 0.005(?) -40.0 4.8x 1.6 5.3x0.9 + 10.4 94 December 4, 1915 { R = 0.015 L = 0.010 -50.0 6.6x 1.9 6.6x 1.5 = 0.0 95* December 8, 1915. / I R = 0.025 L = 0.025 = 0.0 8.3x2.2 9. Ox 1.9 +8.4 96 December 8, 1915. { R = 0.037 L = 0.025 -48.0 . 8.4x2.7 7.2x2.2 -16.7 97 December 8, 1915 { R = 0.010 L = 0.007 -42.9 5.0x1.8 5. Ox 1.5 = 0.0 98 December 8, 1915. { R = 0.320 L = 0.280 -14.3 99 Deceml^er 8, 1915. { R = 0.007(?) L = 0.005 (?) -40.0 4.4x 1.7 5.0x1.0 + 13.6 100 December 8, 1915 { R = 0.025 L = 0.020 -25.0 6.8x2.2 6.7x2.1 -1.5 101 December 8, 1915. / I R = 0.010 L = 0.007 -42.9 6.5x 1.9 6.5x 1.5 = 0.0 December 8, 1915. { R = 0.012 L = 0.012 = 0.0 102 Summary : 103 December 8, 1915. { R = 0.035 L = 0.025 -40.0 Left larger i Left smaller n 2 in ... . 13 104 December 8, 1915. / I R = 0.100 L = 0.070 -42.9 Two equal ir Left longer i 1 3 n... 5 105 December 8, 1915. _{ R = 0.017 L = 0.015 -13.3 Left shorter Left equal ii in ... . 3 1 4 Tuberculosis found. TESTES OF PIGEONS IN IlKAl/ni AND DISEASE WA is probal)!} r('s]M)iisil)l(> for (he f;iilui-(' of our wei^Iiings to (lilTci-eutiuto betwoon llie iiuisscs of s(!Vonil jKiirs of the testes. In the 30 pnirs IS right testes were larger. 5 were smaller, 7 were not differentiated by the weighings. Size, of testes in hybrids dead of disease Seven of the 10 .s]^ocific li3^brids of table 2 had larger right testes; 2 had smaller; 1 had the testes of e(iiial size. Of the two generic hybrids (52, 53) represented in this table, one had a larger and one a smaller right testes. Tn table 3 are listed 13 s])orific hybrids; 7 larger right testes, 4 smaller, and 2 equivalents. TABLE 5 Weight, length and width of testes of young Ring dove (specific) hybrids — killed PER CE.\T PER CENT NO. DATE WEIGHT OF DIFFERENCE LENGTH .\ND WIDTH OP DIFFERENCE f R = too small to 3.2 X 1.5 106 Novembor 29, 191.5 J weigh L = 0.005 + ?.o 4.7 .X 1.8 + 46.9 107 November 29, 1915 / 1 R L = 0.007 = 0.007 = 0.0 5.8 X 1.6 5.4x2.0 - 7.4 108 November 29, 1915 R L = 0.009 = 0.010 +11.1 5.8x 1.8 6.4x 1.8 + 10.3 109 November 29, 1915 1 R L = 0.020 = 0.019 - 5.3 8.2x1.9 8.2x1.9 = 0.0 110 November 29, 1915 { R L = 0.018 = 0.020 +11.1 7.4x2.0 7.4x2.0 = 0.0 111 November 29, 1915 ( R L = 0.010 = 0.010 = 0.0 5.9x2.2 7.4x1.6 +25.4 112 November 29, 1915 / \ R L = 0.190 = 0.170 -11.8 17.0x4.5 15.7x4.5 - 8.3 113 December 3, 1915. . 1 R L = 0.007 = 0.005 -40 5.7x 1.5 4.9x0.9 -16.3 114 December 3, 1915. . December 3, 1915. . December 3, 1915. . December 3, 1915. . / I / I / I / I R L R L R L R L = 0.007 = 0.005 = 0.007 = 0.007 = 0.005 = 0.005 = 0.007 = 0.005 -40.0 = = -40.0 5.5x1.4 4.2x1.0 -30.9 11,5 116 117 Summary : Left larger ii Left smaller Two equal ii Left longer i Left shorter 1 3 in 5 1 4 n 3 in 4 Two equal ir 1 2 THE ANATOMIC.VL RECORD, VOL. 11, NO. 3 TABLE 6 Weights and measurements of testes — birds classified as to kind and disease PER CENT PER NO. DATE DISEASE' WEIGHTS OF DIFFERENCE LENGTH AND WIDTH CENT OF DrFFER ENCE Pure species — Spil. tigrina 118 119 120 121 122 123 124 125 126 127 128 129 130 January 7 Januarj' 9 January 11 January 11 January 13 January 15 January 18 January 25 January 29 March 1 May 17 January 28 January 28 Worms Worms (juv.) Liver Liver and worms. Worms Worms liver. Intest. and liver. . ., Intest Worms (old) Intest. (juv.) Sp. Li. (old) Liver and spleen . . . Worms 0.050 0.050 0.012 0.014 0.040 0.045 0.055 0.055 0.080 0.080 0.040 0.045 031 0.027 030 0.045 0.030 0.022 0.004t 0.004 0.047 0.035 0.030 0.025 0.027 0.030 = 0.0 + 16.6 + 12.5 = = 0.0 + 12.5 -14.8 + 16.6 -36.4 - ?.o -34 3 -20.0 + 11.1 10.8 5.3 7.0 7.3 9.0 8.3 8 9 10 6 7 6.8 7.2 6.7 x3.4 x3.0 x2.2 xl.8 x3.0 X 2.8 x3.2 x3.0 X 4.5 x4 3 x3.0 x2.8 x3.4 x2.7 x3.0 x3.1 x2.6 x2.3 x2.1 X 1.9 x3.1 x2.4 x2.5 x2.5 +38.5 +32.1 +23.3 = 0.0 + 13.9 +26.2 +31.4 +21.1 - 8.6 +4.8 -20.8 + 5.9 +31.3 Pure species — T. orientalis 131 132 133 134 135 March 8 March 23 March 23 March 23 March 25 Lu. liver Cold (juv.)... Cold (juv.). . . Cold (juv.). . . Intest. (juv.). R = 0.016 L = 0.015 R = 010 L = 0.008 R = 0.010 L = 0.008 R = 0.011 L = 0.009 R = 0.010 L = 0.007 6 6 25.0 25 22 2 42.9 5.4x 1 3.7 X 1 6. Ox 1 4 7 x 1 7 0x1 6.3 X 1 5.8 X 1.9 5.5 X 1 .6 -45.9 -27.6 -11.1 - 5.5 1 Abbreviations of the names of organs to their first two letters, implies that advanced and very evident tuberculosis was found in those organs (lungs, spleen, liver, joints, mesentery, intestine). Where more than one organ was affected the name of the organ (or organs) apparently most affected is written first. When the word is written out it denotes that this organ was abnormal, but not necessarily tubercular; immature birds are designated — (juv.). 94 TESTKS OI' I'KiKONS IN II KA I/I'll \.\|) DISK TABLK O-CoiitiiuuMl .\SK <).') PEXt CENT OP DIKKBH ENCE LRNQTH AND WIDTH PER CENT OP DIPFER BNOB Pure species — St. risoria 136 Docomhor 4 Sp. li. lu { R = 030 L = 025 -20 137 January 29 Sp. lu. li • ■ 1 R = 053 L = 031 -70 9 9 2x3 3 10. ox 1.9 + 14.1 138 January 29 Sp. lu. li 1 R = 0.015 L = 010 -50 6.0x2.3 6.6x2.0 + 10.0 139 February 4 Tntost. R = 037 L = 020 -85.0 8.4x2 9 8.8x2.3 + 4.8 140 ^Farch 15 Lu. (? liver < R = 0.550 L = 0.401 -37.1 141 May 10 Sp R = 0.023 L = 0.019 -21.1 SO S.O = 0.0 142 May 14 Jo., lu sp. ; liver ■ ■ ■ i R = 0.020 L = 0.017 -17.6 6.7 6.9 + 2.9 Miscellaneous — pure species 143 November 17 Canker, li , sp. lu . / ■ 1 R = 010 L = 0.007 -42.9 144 November 17 Fight and hemorr. / R =0.595 L = 0.475 -25.3 145 November 18 I'nknown (juv.)... ^{ R = 0038 L = 0.0042 + 10.5 146 November 20 Cold {?).. ^ R = 0.165 L = 0.127 -29.9 147 December 3 Liver ■{ R = 0.015 L = 0.016 + 6.7 Common pigeons 148 January 3 Unknown (juv.). . . { R = 0.005(?) L = 0.005(?) 5.4x 1.5 5.2x1.3 - 3.8 149 January 17 L'nknown (juv.). . . / I R = 0.009(?) L = 0.010(?) R = 0.004? + 11.1 5.8x 1.5 6.5x1.7 about 6.0 + 12.1 150 February 15 Li. pleura < L = 0.003? -33.3? mm. long about 5.3 mm. long + 13.2 151 April 13 Hemorrhage, liver. / R = 0.014 L = 0.010 -40.0 6.4 6.4 = 0.0 152 April 17 Appar. healthy. . . . 1 1 R = 1.340 L = 1 315 - 1.1 20.2x12.1 23.9x10.7 + 18.3 153 May 4 Weakling (juv.). . . / I R = 0.006 L = 0.004 -50.0 5.2 5.2 = 0.0 96 OSCAR RIDDLE TABLE 6— Continued WEIGHTS PER CENT OP DIF FEREXCE PER LENGTH CEXT OF AND WIDTH DIFFERENCE Hybrids — from crosses of species 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 January 1 January 4 January 8 January 15 December 3 January 15 January 27 Febi'uary 18 March 6 January 22 January 24 January 27 February 13 March 17 March 18 March 26 March 28 March 29 April 12 April 20 Cold? (juv.). Sp. liver .... (Juv.) Lu Li. (juv.).... Sp. jo I Healthy d w a r f ; / killed \ Sp- 111 I Sp. li., etc i Li., sp \ Lu., liver, spleen. . . Liver, lu. (?) Lu., liver, spleen J (juv.) \ Sp., lu. liver Sp. li., lu Intest. (?) (juv.). Sp. me. lungs . . . . Li. sj)., me Cold, lungs (Juv.) 0.000 0.000 014 0.017 0.000 0.000 0.037 0.035 0.003 0.003 0.027 0.020 0.010(?) 0.013(?) 0.015 0.015 0.028 0.025 007 0.007 0.050 0.050 0.052 0.050 0.002? 0.003+ 0.022 0.021 0.025 0.025 0.002 0.002 0.048 0.042 0.031 0.035 1.015 0.888 0.002(?) 0.002(?) +21.4 = 0.0 -35.0 +30,0? = 0.0 -12.0 = 0.0 = 0.0 - 4.0 +50.0 - 4.7 = 0.0 = 0.0 -14.3 + 12.9 -14 3 = 0.0 4.2 xO.9 5.2x 1.2 6.3x2.1 8.6x2.5 3.8x 1.5 5.7x 1.2 8.7x2.7 8.4x2.8 5.0x2.1 5.6x2.0 9.7x2.9 10.4x2.7 9.0x3.7 11.4x2.9 5.7 5.9 7.6 4.4 3.9 9.9x3.7 10.1 x2.3 8.7x2.9 12.2x2.2 22.6x9.0 23.5x8.2 5.2 5.4 +23.8 +36.5 +50.0 3.6 + 12.0 + 7.2 +26.6 + 3.5 + 5.6 12.8 + 2.0 +40.2 + 3.9 + 3.8 TESTES OF 1'U;E().\S 1.\ llKALTli A\U IJISK TABLIO 0— Coiitiiiufil ASE 1)7 \vki(;hts PER CENT OF UIKKKUE.NCE I.KNOTH .\.ND WIDTH PER CENT or DIFFERENCE Hybrids — from crosses of species (con.) 174 iTo 176 177 178 179 April 23 May (i May 6 May 10 May 11 Mav 16 I>iv(M- (iiitest. ?). . . . '"^P- li- I"' I ^P Li. spleen Sp. li Healthy (juv.) R = 0.054 L = 0.044 R = 0.034 L = 0.032 R = 0.020 L = 0.022 R = 0.042 L = 036 R = 0.021 L = 0.018 R = 0.125 L = 1.25 -22.7 - 6 3 + 10.0 -16.7 -16.7 = 0.0 10.8 10.6 8.2 8.2 7.6 8.0 8.3 8.7 7.2 6.6 11.6x4.5 13.2x4.3 - 1.9 = 0.0 + 5.3 + 4.8 - 9.1 + 13.8 Hybrids — from crosses of genera 180 December 4 Sp., li / R = 017 L = 0.020 + 17.6 7.0x2.0 6.3x2.5 -11.1 181 December 24 Liver, pericard < R = 017 L = 0.020 + 17.6 18? January 8 Cold (?) I R = 005 L = 0.004 -25 3.2 X 1.8 3.0x1.5 - 6.6 183 January 12 Abdom. wall < R = 0.009 L = 0.012 +33.3 7.3 X 1.8 5.7x1.9 -28.1 184 February 8 Intest < R = 0.005? L = 0.004? -25.0 about 4.5 about 4.5 = 0.0 185 February 23 Intest. (?) 1 R = 0.045 L = 0.051 + 12.3 7.3 7.1 - 0.3 186 March 7 (Cold?) lungs 1 R = 0.007 L = 0.005 -40.0 both teste ■ gular-glo like hem s anbular, p seed 187 March 13 Li., sp., lu. (?) 1 R = 0.014 L = 0.019 +35.7 7.0 9.0 +28.5 188 March 29 (Cause?) I R = 0.006 L = 0.007 + 16.7 4.6x1.9 5.3x2.0 + 15.2 189 April 12 Worms, fighting. . . . < V R = 0.256 L = 0.212 -20.8 15.7x5.6 15.4 X 5.1 - 1.9 190 May 15 Lu 1 R = 0.047 L = 0.043 - 9.3 8.8x 8. Ox -10.0 98 OSCAR RIDDLE Six generic hybrids here all show larger right testes. Of 24 birds listed in table 6, larger rights were found in 11, smaller in 5, eqLiivalents ^n 8 cases. In table 6, 11 generic hybrids are listed; the right testis is larger in 5, smaller in 6. A summary representation of the weight relations of the testes of birds belonging to the several preceding groups is given in table 7. Relative lengths of the two testes The length of 78 pairs of testes was ascertained. These were obtained from the testes of birds belonging to all of the groups discussed in the previous section of this paper. The reader is referred to the summary on 'length relations' given in table 7 for a first view of the result. Although a high proportion (126 to 39 for all groups) of the right testes are heavier, a reversal of proportions (24 to 42) is found for the absolute lengths of the two testes. In five of the seven groups of table 7 the left testis is absolutely longer. In one of the two exceptional groups — TABLE 7 Suniiuary of the preceding data CLASS STATE WEIGHT RELATIONS' LENGTH RELATIONS L + L = L L + L = L Pure species < Common pigeons < Specific hybrids < Generic hybrids Healthy . . . Diseased. . . Healthy . . . Diseased. . . Healthy . . . Diseased. . . Diseased. . . 9= 5 2 6 10 7 7 7 11 5 31 27 8 19 24 12 14 1 2 8 15 2 2 o 6 1 1 3 2 1 7 4 6 Total healthy . . 11 7 51 9 6 10 Total diseased 28 18 74 33 6 14 Grand total 39 24 126 42 12 24 1 The cases of larger left testis, are grouped under L+; those of equal size under L= ; those with a smaller left testis under L— . 2 Five of the (9), and 3 of the (6), are from 14 Sp. tigrina — all dead at less than 9 months old. TESTES OF PHIKOXS I\ MKALTH AM) DISEASE 99 'licallliy ])ur(' s])0('i('s'- it is ])()ssil)l(' lluit the siii.'illcr Icti^tli of the left testis is coiuuH'tcd \\\\\\ the nearest a])))i'()\iin;ili()ii to ujiil'onuity of smaller size in this ^rouj). Tn the seeond ^roiip — '(hseascnl <2;ener.i(' hybrids' the testes show the greatest dejmrture in their weight rehitions fioni wliat is elsewhere the rule. An exaniiniition of the detailed measurements and percentage weiglit difTerenees given in tables 4, 5 and 6 is even more convincing tlian the sununary of table 7 on the point that the two testes definitely tend to assume two different shapes — the left to be thinner and more elongate, the right to be shorter and thicker. This difference in form is perhaps not without interest since the only persistent gonad in the female — that of the left side — is characteristically 'thin' and 'long.' The testis that develops on this side is similarly characterized as compared with its mate of the right side. Pure species and hybrids and the relative size of the two testes Table 7 facilitates an expression of a relation which seems to obtain between degree of hydridization on the one hand, and the number of departiu'es from the usual situation — a larger right and a smaller left testis — on the other. There are good reasons for believing that common pigeons — mongrels of many breeds probably not even descended from a single good species — may rightly occupy a place in this classification intermediate to specific hybrids and pure species. There is no question that the generic hybrids are more separated from the pure species than are the specific hybrids. Now, the number of gonad size relations, that depart from the rule, arrange themselves in all of the seven groups of the table, precisely in this order of departure from purity of species. That is, the greatest proportion of larger left testes is found in the group most widely separated from a pure species (generic hybrids) ; and the two intermediate stages of crossing (specific hybrids and common pigeons) show each its appropriately^ smaller intermediate number of larger left testes. 100 OSCAR RIDDLE The influence of disease on the actual and relative size of the two testes The detailed data of the several tables and the summary of table 7 demonstrate that diseased birds furnish a higher proportion of larger left testes — violations of the more general rule — than do healthy birds. Further, the testes of pigeons suffer great reduction in several, or most, forms of disease. Now it probably happens that because of the great reduction in size of the testes in disease that this of itself has rendered the results of a few of the weighings of the smallest glands less certain.^ An examination of the detailed data will, however, leave no doubt that disease is a very real and important reason for the observed differences. It has been found well to designate the cause of death in most of the tables. Advanced tuberculosis is easily,, and with much certainty, diagnosed in pigeons. It is the most common cause of death among the birds of our collection. All deaths from this cause are so designated in the tables. A careful comparison of the size of the testes of birds dead of tuberculosis as compared with those dying of other or unknown causes will show that the gonads of the male suffer greatest reduction under this disease. This is certainly not true for the female gonad ; a point upon which data are still being collected. In this connection Hatai's ('15) observation of the effects of exercise on the size of the ovary and testis of the rat are of interest. Hatai found a like quahtative response — an increase — in both; but quantitatively the response was quite different; the testes increased only 12.33 per cent while the increase in the ovaries was 84.33 per cent. The several sections of table 6 show that in pigeons the spleen and liver are more often affected by tuberculosis than are other organs. It is the spleen too that suffers greatest hypertrophy and most complete transformation under the disease. This is true of both sexes. ^ Because of imperfect or unclean separation of the gland from body wall, and drying during weight, the greatest care will not always obtain perfect weights of the smallest of these glands. TESTES Ol' I'KIEO.NS I .\ 1 1 1: A l.i'll \.\|) l)ISE.\SE lOl Size rcldlions of the testes nt the eominon. juivl In table S arc ^incii tlic few data w(> Ikivc been able to obtain on the Jungle fowl, coninion fowl, and the duck. Since these were — with two exceptions — lioaUliy fowls even the few data indicate a situation different from tliat found in pure species of pigeons, ^^'hether these data are really representative of these form.-;, cannot now be determined. Whether hydridization (or mong"i'(>lization) of these forms is responsibh; for the api)arent predominance of the reverse of the situation fomid in i)igeons is doubtful. They like the sparrows may normally possess a larger left and a smaller right testis. TABLE 8 Weights of testes of Jungle Fowl and common fowl NO. D.\TE WEIGHT PER CENT OP DIPFERENCE NO. DATE WEIGHT PER CENT OF DIFFERENCE Jungle fowl Common fowl 1 2 3 4 5 6 7 July 5 < July 5 \ July 8 < July 10 1 July 12 1 July 16 (inj.i) | July 22 (inj.i) | R = 3.203 L = 4.217 R = 4.700 L = 5.700 R = 4.635 L = 4.280 R = 6.305 L = 6.270 R = 4.500 L = 4.900 R = 3.960 L = 5.105 R = 1.665 L = 1.820 +31.7 +21.3 - 8.4 - 0.6 + 8.9 +28.9 + 9.3 1 2 3 4 5 6 7 8 July 16 1 July 16 1 July 20 1 July 22 1 August 26 < April 21 I December 10 / (roup) \ December 21 / (roup) \ R = 3.734 L = 4.219 R = 5.880 L = 7.000 R = 6.540 L = 5.610 R = 11.660 L = 12.815 R = 8.820 L = 7. .545 R = 11.100 L = 10.350 R = 0.235 L = 0.200 R = 0.590 L = 0.635 + 12.9 + 19.0 -16.6 + 9.9 -16.9 - 7.2 -17.5 + 7.6 Wild Duck (y'g) Novem- / ber 29 \ R = 0.023 L = 0.027 + 17.4 December 28 j (starved) . . \ R = 8.3x3.2 R = 0.040 L = 0.040 L = 8.6x2 0.0 1 These cocks had been given a few injections of ovarian extract during the week preceding the days of killing and autopsy. 102 OSCAR RIDDLE SUMMARY The prevalence of atrophy of the right ovary in birds; the demonstrated differences in number of primordial germ cells in the two glands of the fowl; and the unequal — and opposite — size relations of the two adult gonads of the male, constitute a body of puzzling facts whose elucidation should contribute largely to our knowledge of the nature and basis of sexual difference. The right testis of the pigeon is normally larger than the left. In hybrid pigeons there are more exceptions to the normal size-relations of the two testes than in pure species. The number of the exceptions seems to increase with the degree of hybridization (width of the cross) ; there being fewer in specific hybrids than in generic hybrids. The testes of pigeons suffer great reduction in size in disease — particularly in tuberculosis. It is probable that the right gland suffers greater reduction than the left. The left (persistent) gonad of the female does not suffer a similar reduction in tuberculosis. Season is plainly not the cause of the differences and reductions noted in pigeons. The two testes of the pigeon are characteristically different in their dimensions. The left (like the left ovary) is thinner and more elongate. The right (represented in the female by atrophied ovary) is shorter and thicker. In poultry the few data at hand fail to indicate a constant qr decided predominance of size in either gland. Cold Spring Harbor, L. I., N. Y. June. 1916 LITERATURE CITED Allen, B. M. 1907 A statistical study of the sex cells in Chrysemys marginata. Anat. Rec, vol. 1, pp. 64-65. (Also Anat. Anz. vol. 30, pp. 391-399.) Etzold, T. 1891 Die Entwicklung der Testikel von Fringilla domestica von der Winterruhe bis ziim Eintritt der Brlrnst. Zeit. f. Wiss. Zool., vol. 52, pp. 46-84. FiRKET, Jean 1914 Recherches sur L'organogenese des glandes sexuelles chez les oiseaux. Arch. de. Biologie, T. 29. Hat.\i, S. 1915 On the influence of exercise on the growth of the organs in the Albino rat. Anat. Rec, vol. 9. Swift, C. H. 1915 Origin of the definitive sex-cells in the female chick and their relation to the primordial germ cells. Am. Jour. Anat., vol. IS. Tannenberg 1789 Spielegium observationum circa partes genitales masculas avium. Gottingcn. llYCilKXiC C'ACES FOR HATS AM) MICE J. A. LoXd AiKiloiniciil Ltihiiraturi/, L'nircrsihj of Culifurnia TWO FICUUES The follow iiiu is ;i hrii'f account of cajiX's recently constructed for the Depai'tnuMit of Anatomy of the University of CaHfoi'iiia for housinfi" colonies of i-ats and mice. In ])lannin,<i; tlu'se cap;es the desirability was kept in mind of so design inii' them that not only might the animals be cared for conveniently, but the cages be easily cleaned and comjiletely stei'ilized and the spread of infection prevented. Accordingly they were made entirely of metal: the sides and front of galvanized iron, the former preventing the direct passage of infection from cage to cage; and the lid, top, back, and floor of hardware cloth of |-inch mesh bound ' with strips of galvanized iron. They can be taken apart, packed in a small space for l)oiling, and reassem])led quickly without the use of any screws or ])olts. All parts are interchangeable. The inside dimensions are: floor, 9^ bv 14| inches; height 11 inches; front 2j inches high. They are arranged in groups of 20 (4 rows of 5 each, fig. 1). Each group is supported by a rack made of iron pipe, and 4 pairs of angle irons on which the 4 rows of cages are hung. Below each row is placed a shallow, removable, galvanized iron pan intended to be filled with sawdust for receiving refuse falling through the bottoms of the cages. The racks measure 6^ feet in height, 17^ inches in depth, and 4| feet in width. There is a space of 9^ inches between the lowest pan and the floor, and 3 inches between the floors of the cages and pans. If desired the racks can be continued Upward to carry one or more additional rows. Most of the details of construction can be seen in figure 2 which shows some of the cages taken down. It will be observed that the sides are suspended and in turn furnish support for the rest of the cage except the top. The sides are put into place by slipping the flanges on the upper edges into grooves (gr fig. 2) formed b}" bendingunder the edges of strips of galvanized iron (st). A projection (pr) prevents sliding the sides in too far. The ends of the strips forming the grooves are bent up and over the angle irons and are permanently fastened by means of bolts. These strips also have soldered to their upper sides grooves (Ig) opening laterally formed by strips of metal bent in the form of a narrow trough. Into the latter slide the tops to which the lids are hinged by two rings. The backs when in place 103 104 J. A. LONG HVCIK.MC ( ACKS 1-(»1{ KATs AM) MKIO 105 106 J. A. LONG rest on the flanges on the lower edges of the sides and against the front faces of the back flanges. The upper ends of the backs are held firmly because they pass behind the rear angle irons; at the lower ends pins (p) fit into holes in the bottom flanges (the pins can also be seen on the under side of the upper row of cages). The binding on the lower edge of each back is turned forward at a right angle and together with the flanges on the lower edges of the sides serves to support the floor. The latter are kept in place l)y two pins (p') which fit into corresponding holes (h) in the binding. The front is made of one piece of metal. One may be seen endwise resting on the edge of a tray. The ends are bent somewhat in the from of a letter S to form troughs which fit over the flanges on the front edges of the sides. In assembling the cages the sides are first put into place, then. the backs, floors, front, and top (with lid). It will be seen that the floors may be changed without disturlnng the rest of the cage, or by removing simply the front. A numlier of extra bottoms makes it possible to clean one set and have them ready to substitute for soiled ones every week. The other parts of the cages need cleaning only at longer intervals. For the cages used for rats, floors of h inch mesh are provided. It has been found in actual breeding that 4 and even 6 adult rats can be kept in one cage, and as many as 10 or 12 young rats raised to breeding age in single cages. The construction of this equipment was worked out by Prof. H. M. Evans and the writer with assistance from Mr. H. B. Foster, the University Engineer. ## The Golgi apparatus. Personal observations and a review of the literature ALU l.\ M. PAPPENHEIMER From the Dcpnrtnirnt of Pitlliologij, College of Physicians and Surgeons, Columbia ['ninrxilij; and from the Marine Biological Lahoratorij, Wools Hole, Mass. TWENTY-TWO FIGURES CONTEXTS I. Introchu'tion 107 II. Nomenclature 109 III. Technique 110 IV. The occurrence of the Golgi apparatus in various t3'{)es of tissue cells: 1. Nervous tissue 112 2. Non-nervous tissue 114 a. Epithelial cells 114 b. Connective tissue cells, cartilage, osteoblasts, odonto blasts, striated muscle cells 127 c. Gonads 130 V. The Golgi apparatus in the cells of embryos 132 VI. The Golgi apparatus in protozoa 133 VII. The Golgi apparatus under pathological conditions 133 VIII. General considerations. Relation to centrosomes. Polarity. Physical nature of the structures. Holmgren's trophospongium theory. . 135 IX. Conclusions 138 I. INTRODUCTION It is rather extraordinary that there should be within every cell a structure as conspicuous as the nucleus, and sometimes surpassing it in size, the meaning of which is utterly obscure. One at least of the functions of the nucleus — its role in heredity — is known to us. We have fairly definite ideas as to the role of the centrosomes, and theories aplenty as to the part played by mitochondrial structures, and other types of granules. But as regards the structure to which Golgi has given the name 'Apparato reticolare interno/ we have learned only its appearance, its distribution in different types of cells, and its behavior dur 107 THE ANATOMICAL RECORD. VOL. 11, XO. 4 NOVEMBER, 1916 108 ALWIN M. PAPPENHEIMER ing cell division. One of the most recent papers on the subject — that of Kolster (102) — ends with the statement "These structures undoubtedly have a special significance, but we are ignorant of it." The credit for the discovery of this intracellular organelle — if such it be — undoubtedly belongs to Golgi, and a large part of the work, including the working out of a fairly easy and satisfactory technique for its demonstration — has been done by Golgi himself, and by his pupils and co-workers, Veratti, Perroncito, Pensa, Negri, Gemelli, Brugnatelli and others. A number of papers dealing with the same structures have been published by Ramon-y-Cajal, who, indeed claims priority for their discovery over Golgi, and by his pupils, Sanchez, Fananas, Tello and others. Nussbaum, in Prague, has inspired a series of papers by his pupils (Weigl, Polyescynski, Bialschowska and Kulikowska) dealing chiefly with the appearance of the Golgi apparatus in the ganglion cells of invertebrates. Important papers have been published by v. Bergen, Deineke and by Kopsch, who discovered a new and very simple method for demonstrating the apparatus. The best and most exhaustive general review on the subject is that of Duesberg (48), before the XXVIII Meeting of the Anatomische Gesellschaft at Innsbruck in 1914, and Cajal (31) in his most recent pubhcation ('15) which was not available at the time this study was begun, has contributed a most interesting critical survey of the entire field, and added many new observations. A whole chapter — largely controversial — is that contributed by Holmgren, whose views and their bearing I shall take up later. In this country only Bensley and Cowdry have made contributions to the subject. In reviewing the literature of the subject I found that but few workers, with the exception of Gajal antl his pui)ils, had attempted to study the behavior of the Golgi apparatus under experimental conditions. I planned, therefore, to follow the modifications of the structure in the epithelial cells of the rat kidney, which might be produced by autolytic changes, secretory phases and toxic agents. The choice of material was unfortunat(\ The ( Jolgi ajiparatus of the cc^lls of the renal tubules I'm; (;()i,(ii Ari'AUArrs 109 proved to ho so atypical and variable iii lomi (hat it was iliflieult to (haw int"ei(Mi('es IVoiii \aiia1ions seen under experimental eoiiditioiis. One was tin I her liaiidicapjx'd hy the dillieulties and eaj)i ieiousness of the ini|)i(',ii;naii()ii nictliod, as ap|)li('d to this oi^ai). Ill atlemplin^ to coiitiol tlie techiiiiiite many other tissues were studied, and insofar as tlie ohserxations made dilTci' from tliose of ))re\ious woikiMs, they aic ^iN'en hc^ow. Although 1 was imsueeessfiil in the main jnnpose of my stud\', it seemed that it might he useful at this tinu^ to collate the widely scattered and rather inaccessible literature, and to record my personal obser\'ations, insofar as they supplement or are at variance with those of other workers in this field. II. XO.MKXCLATrPvK Clolgi (01 j, in his original communication before the Aled. Chir. Society of Pa\'ia in 1898, suggested the term 'Apparato reticolare interno,' and this term has naturally been adopted by all the Italian workers. Kopsch (103) proposed the term 'Binnennetz' as the German equivalent, but both of these designations are open to the objection that the structures do not appear in all types of cells, nor under all conditions, as a closed net. Ballowitz (4) in 1899 described a basket-hke structure about the centrosomes of the cells of Descemet's membrane, and suggested for it the name 'Centrophormia.' Later he recognized the homology of the 'Centrophormia' with the Golgi apparatus, and the term has not come into common use. The 'Nebenkern' of Platner (145) and la Valette St. George (106), and the 'Zentralkapsel' of Heidenhain (68) in the sperm cells have been considered by some as related to or identical with the structures demonstrated by the Golgi technique. The terms, however, are not sufficiently inclusive to apply to the structures described by Golgi. Ramon-y-Cajal and his school who agreed with Holmgren in regarding the apparatus as canalicular in nature — referred to it in their earlier publications as the Holmgren-Golgi apparatus. In his latest review, however, Cajal (31) recognizes the very doubtful identity of mam^ of the structures described bj^ Holm 110 ALWIN M. PAPPENHEIMER gren with those brought out bj^ the Silver methods, and therefore refers to them more justly as the Golgi apparatus. Many of the German workers speak of the Golgi-Kopsch net or apparatus. Hohngren (90)', who believes in the identity of the canahculi described by him, with the structures put in evidence by the silver impregnation methods, uses the term 'Trophospongium.' Cowdry (42) and Bensley (14) speak of 'canalicular apparatus.' None of these terms appear to be entirely satisfactory. I shall, therefore, refer to the structures simply as the Golgi apparatus. III. TECHNIQUE The earliest studies of Golgi were made with a modification of his well-known silver chromate method. This gave capricious and inconstant results. Veratti introduced a modification, the essential feature of which was a fixation in osmium platinic chlorid mixture. Kopsch (103) in 1902 showed that prolonged immersion in 2 per cent osmic acid would demonstrate structures identical with those described by Golgi. The two methods now most commonly used are those of Golgi (66) and of Cajal (30), and, for the convenience of those to whom the original articles are not accessible, they are given here: The Golgi method is as follows: I. Fixation: Formalin (20%) 30 cc. Saturated solution arsenious acid (1%) 30 cc. Alcohol (97%) 30 cc. 6 to 24 hours. II. Silver nitrate 1% 1 hour to several days III. Developynent : Hydroquinone 20 gm. Sodium Sulphite 1 gm. Formalin 20 cc. Distilled water ad 1000 cc. Wash in distilled water, dehydrate rapidly and embed in paraffin or celloidin. IV. Toning: Solution 'A' Sodium liyposulphite 30 gm. Annnonium Sulphocyanate 30 gm. Distilled water 1000 cc. Solution 'B' Gold chloride 1% Use equal parts of 'A' and 'B'. Tone to grey tone. 2 to 3 hours THK (i()L(il AlTAltATrs 111 Voratti has dcNiscd tlic folic (wiiiji; proccMlurc toi" riddiiifi; flic j>j(»|)arati(Hi ol' siKcr iticcipitatc at'lcc loiiiii^': 'A' — l\i'i)c:iti'(l wasliinir in disiillcd wiitcr. 'IV — Kapid passable tliioiiuli lollowiiifi solutions: (,1) I'otassimn pcnnanganato — ().."> frm. Sulphuric acud — 1.0 cc. Distilled water— KMM) cc. (2) Oxalic- acid— r; Wasii in distilled water. C'ounter.stain witii alum carmine. Tlic latest Cajal method differs from the Clolgi chiefly in the use of uranium nitrate in place of the arsenious acid in fixation. It is gi\'en as follows: /. Fixation: Uranium nitrate — 1 gm. ] Formol — 15 cc. !• 9 to 11 hours Distilled water— 100 cc. J //. Wash quickly ///. Silver nitrate— 1.5%— 30 to 40 hours IV. Wash quickly V. Reduce in Hj-droquiuone — 2 gm. Formol — 6 gm. Distilled water — 100 cc. Add anhydrous sodium sulphite^O. 15-0.25 gm. so that solution has a yellow color. VI. Dehydrate and embed in paraffin. VII. Toning: 'A' Sodimii hj-posulphite — 3 gm. Ammonium sulphocyanate^S gm. Distilled water — 100 cc. 'B' Gold chloride— 1% Use equal parts. The addition of 30 cc. of ethyl or methyl alcohol to the fixative is recommended by Cajal ('15), as advantageous in the case of nervous tissue. As counterstain I have found a dilute Giemsa sohition to give the clearest pictures. A 1 per cent methjd-green solution ma}" also be used, and it has been found possible to combine also the Altmann mitochondrial stain, as modified by Bensley.^^ The removal of the silver precipitate \\dth permanganate and oxalic acid must be very carefully controlled, as it is easy to bring about a complete decolorization of the Golgi apparatus as well. 112 ALWIN M. PAPPENHEIMER Cajal and others have obtained the most constant results in the tissues of young animals. Because of the rapid occurrence of autolytic changes httle confidence can be placed in the results obtained with tissues from human autopsies. Both the Golgi and Cajal methods are exceedingly capricious; the impregnation is rarely uniform throughout the entire block. The most delicate and important step in these photographic processes, according to several workers, is the initial time of fixation. Each type of cell has its optimum time of fixation, which must be determined experimentally. In many cells, however, as in the lymphocytes, spermatic cells, glomeruli of the kidney, this appears to vary within wide limits. That, at least, has been my experience, and I have obtained identical pictures with fixation varying from 2 to 12 hours. The technique most recently advocated by Holmgren for demonstrating his Trophospongium is a fixation in trichlorlactic acid (6 per cent) and staining in a freshly prepared resorcinfuchsin solution. The 'canals' take a purplish black color. Holmgren also gives methods which show the canals as colorless structures upon a stained background. IV. THE OCCURRENCE OF THE GOLGI APPARATUS IN VARIOUS TYPES OF TISSUE CELLS 1. Nervous tissue The first clear description of the structm^es is that of Golgi in 1898 (61, 62), in the spinal ganghon cells of Strix flammea (Barn-owl); in the same year, he made similar observations upon the spinal ganghon cells of Mammalia; Veratti in 1898 found the same sort of structure in sympathetic ganglion cells. Since these early papers, the Golgi apparatus has been found to be present in many other types of nerve cells — the anterior horn cells (Golgi (66), Cajal (31)), the pyramidal cells of the cortex (Golgi (65), Legendre (107), Collin and Lucien (37), Soukhanoff (167), Cajal (31)), the Purkinje ceUs and other nerve cells of the cerebellar cortex (Golgi (66), Cajal (31)), of the olfactory lobe (Cajal (28) ), the ganglion cells of insects 'I'm; Cdl.lil Al'l'AllA IMS 1 \'.i ( liialkowska aiul Kiilikowska (19)), of the leech and carthworin ( Hialkowska ami Kulikowsku), Cruslacca (Jawai'owsky (98), Monti (127). Poloiizynsky (14()), of ccphalopods (Weigl (180) ). Tlu^ apparatus loaches its ^Toatost fomplcxity and size in llie s|)inal ^an^lion cells of vertebrates, and these have, therefore, been a faxorite object of study. In adult \'ei1ebrates there is shown by the silver or osniic methods, a dehnite netwoik of solid, tortuous varicose fibrils, which vary in thickness with different species. This network may completely or partially smround the nucleus and may be in contact with it in ]ilaces. Where the threads cross or interlace, there are often nodular \-aricosities. In some species there is a sort of lobulation, into three oi' four ])artially separated skeins, and individual filaments may be gi^'en off from the main mass, and ai)i)arenth' end freely in the c^^toplasm. In all cases the peripheral zone of cytoplasm is left free; at no point does the network, or any of its branches reach the surface. ]\Ionti (127), in the ganglion cells of invertebrates (crustacea, arthropods and cephalopods) found a simple ajDparatus in the form of cur^'ed filaments, often bifurcating or anastomosing, but not forming a closed reticulum. V. Bergen (10), working with the Kopsch osmium method, upon the spinal ganglion cells of the hedgehog, cat, rabbit, rat, mouse, and hen, found that not all the cells showed a complete reticulum as described by Golgi, some containing only short filaments, rows of granules or ring forms. Some of the filaments contained a central clear space, and these he interpreted as degeneration forms. This variation in the appearance of the apparatus in different cells in the same preparation v. Bergen interprets as indicating the transitorj^ nature of these structures. He suggests that they are developed from granules, w^hich range themselves into filaments, form more complex networks, and finally undergo central liquefaction with the formation of canaliculi. Other recent workers, however, using the newer methods of Golgi and Cajal have not confu-med y. Bergen's theory, and ascribe the variations to defective impregnation. 114 ALWIN M. PAl'PENHEIMER Cajal (31), like v. Bergen, notes variation (or 'modalities') in the type of net occurring in ganglion cells of the same order and size. He strongly rejects the idea that these variations are due to irregularities in impregnation, since they may be found in adjoining cells at similar depths from the surface. The question has arisen as to whether the Golgi apparatus is identical with any of the other known cytoplasmic constituents of the nerve cell — namely, the neurofibrillae, the Nissl substance or the mitochondria. It seems quite certain, in spite of occasional statements to the contrary, that the Golgi net is unrelated to any of these structures. The net is not continued into the cell processes, as are the neurpfibrillae, and the fibers of the net are much thicker and more varicose. By combining Kopsch's method with Bensley's aniline-fuchsin toluidin-blue stain, as Cowdry (42) has done, the Golgi net, Nissl bodies and mitochondria may all be stained in the same cells, and their independence of one another made obvious. It seems hardly worth while to go further into this discussion. 2. Tissue cells other thaii nerve cells a. Epithelial cells. The presence of a Golgi apparatus was first demonstrated in the squamous epithelial cells of Ammocoetes (Lamprey eel) by Marenghi (120) in 1903, and in Lumbricus by Ramon-y-Cajal in the same year. Since then it has been found in the corneal epithelium by Barinetti (6) and by Deinecke (47). The net is present in all layers. In the superficial cells the net becomes looser, and often almost entirely surrounds the nucleus, whereas in the rete mucosum, it lies at the superficial pole of the nucleus, and in the cells near the surface, only granular bodies are found. This change, therefore, accompanies the aging of the cells, and is characteristic not only of corneal epithelium, but also of skin, oesophageal mucosa, the epidermis of the ducks bill (Deinecke, Kolmer (100) ). Some of our preparations of the mucosa of the renal pelvis show a similar differentiation. A Golgi net has been found also in cells of the epidermal appendages and glands — in the lachrymal gland by Ancona (3) TiiK (;(ii,(;i Ai'i'AH A rrs 1 If) and in the sweat and s('l)a{'(M)us glands hy v. Rci-fijon (10), by Hiz/ozcK) and Hottisdla (21) (who i)i('tui(' a net snii'onnding lilt' nucleus, and also, incidentally, could not <l('ui()nsti'ate it in the epidermal cells), and by Tello (172). Since 1 he liisl paper of Hallowitz (4) in 1S!)S, a net has been found in tlie single layered ei)ithelial cells of Deseeniet's nieuibiane by Totsuka (174), Zawarzin (bS4), and by Deinecke (47). The net in these cells is of special interest because it very cleaily lies in relation to the centrosonies, and because it was discovered independently of Ciolgi's work by Ballowitz. Deinecke in these cells, also, made :i caieful study of the l)ehavion!' of the net durino- mitosis. Numerous obser\'ati()ns conhrm the presence of a Golgi apparatus in the glandular cells of the gastro-intestinal tract. Thus Kamon-y-Cajal ('03) found it in the intestinal epithelium of lumbricus, and of the guinea pig (27); v. Bergen (10) in the chief cells of the fungus region ('04), Golgi (67) in the gastric and intestinal mucosa of frogs, birds and mammals, in the glands of Rrunner and of Lieberkuhn ('09); d'Agata ('43) in the gastric epithelium of triton ('10) and in the gall-bladder epithelium of the guinea pig (44); Weigl (180) and Kolmer (100) in the gastric and intestinal mucosa of various vertebrates; Kolster (102) in the chief and parietal cells of the fundus, in the pylorus and in the cells of Brunner glands ('13). Kolster has made several interesting observations on the behaviom* of the Golgi apparatus in the gastric cells. He found that w^hen the chief cells were successfully impregnated the parietal cells w^ere not. He also showed that, by using the original Golgi silver chromate method, it was possible to impregnate a system of endocellular excretory canals in the chief cells of the fundus, and that these differed, both in then' topography and in their form from the true Golgi apparatus. He noticed also in the pyloric gland cells that the appearance of the net varied with different phases of secretion. In the resting cells the net was quite dense, the meshes small, the form of the w^hole mass spherical; w^hile in secreting cells, the net was rare 116 ALWIN M. PAPPENHEIMER fied (aufgelockert) , elongated and extended to the basal portion of the cell, in close contact with the flattened nucleus. Cajal (31) also describes in great detail a cycle of changes corresponding to different secretory phases in the goblet cells of the alinientarj^ tract. The Golgi apparatus during the earlier phases undergoes an increase in size, later the argentophile substance becomes dispersed amongst the globules of secretion, and completely disintegrates — not as Kolster (102) believes, becoming merely compressed against the nucleus at the base of the cell. The inferences which Cajal draws as to the functional significance of these cyclical changes, will be discussed later. Before having access to Cajal's paper, I had independently observed similar alterations in the mucous glands of the larynx (figs. 1, 2, 3). It seems to be quite clear that the net, which is more distinct and w^ell-formed in cells during the inactive stage, becomes broken up and distributed amongst the globules of mucus in those cells which are actively secreting. In the course of this process, there appears to occur a real quantitative decrease in the amount of the argentophile substance not to be explained merely by its mechanical disruption, and implying some sort of regeneration of the apparatus, after the cell has discharged its secretion and returned to rest. A Golgi net has been demonstrated by numerous observers in the epithelial cells of various glandular organs, and I shall limit myself merely to giving a list of these. The CJolgi net was described in the thyroid by Negri (130) and by Kolster (102); in the adrenal medulla by Pensa (135) and Kolmer (100), and in the cortex by Pilat (144), by Mulon (128) and by Kolmer (100); in the anterior lobe of the hypophysis by Gemelli (60), and in both glandular and nervous portions by Tello (172); in the pancreas by Negri (130), by v. Bergen (10) and by Kolster (102), Kolmer (100) and Cajal (31); in the dog's prostate by V. Bergen (10) and in the hypertrophied human prostate by Verson (179) and by Taddei (171). Von Bergen claims to have been able to recognize the net in unstained scrapings of prostatic epithelium kei)t aHve for a time in the prostatic secretion. TIIK COLCI M'I'AKA ri'S 117 This appeal's lo he the only rccdidcd attciiipl to ohsci'N'c IIk; (loliii apparatus in the li\ iiiji cell.' ■-:?■ ■■c.%%.*fe^ ^«; .iS>, 2 5 r^'^ •^ Fig. 1 Tracheal gland, non-secreting. Cajal. Fig.s. 2 and 3 Tracheal glands, showing fragmentation of Golgi apparatus during secretion of mucus. Fig. 4 Small group of thyroid epithelial cells, one in mitosis. Dittokinesis. Cajal. Fig. 5 Salivary gland of rat. Cajal, anilin-fuchsin-methyl-green. Note relation of Golgi net to Altmann granules. 1 We also have tried to observe it in growing chick embrj-o cells in vitro, both by direct light and using dark-field illumination, but without success. Lewis and Lewis (109) likewise report their inability to see structures corresponding to the Binnennetz in living chick embrj-o cells, nor could they be brought out b}' prolonged osmic acid fixation. 118 ALWIN M, PAPPENHEIMER The net has been found further in the epithehal cells ol the epididymis by Negri (130), by Fusari COS), by Kolster (102) and by Kolmer (100); in the cihated epithelium of the trachea by Kolster (102); in the choroid plexus by Biondi (20) and in the uterine mucosa and chorionic epithelium by Decio (46) and by Acconti (1). Negri (130), Kopsch (103), v. Bergen (10), Kolster (102) and Kolmer (100) and Cajal (31), record the presence of a net in the sali^'ary gland epithelium. My preparations show one or two points which I do not find mentioned in their descriptions. I find in some acini remarkably large, coarse-meshed nets, enveloping the nucleus, joining by stout, varicose filaments with nets in adjacent cells, and not infrequently giving origin to trabeculae which loop about the lumen of the gland. They are thus not confined to a single cell, and anastomose freely one with another. By counterstaining the Golgi preparations by the anilinfuchsin-methyl gieen modification of Altmann's method, one can clearly recognize the independence of the net from the Altmann granules, which are evenly distributed through the entire cell. Indeed, it seems as if these were an inverse relation — that is, those cells in which the granules are poorly marked and absent, show the most conspicuous and clearly defined net, whereas the large cells which are replete with granules, may contain no net at all. Whether this is a constant relation or not, remains to be seen (fig. 5). It is surprising that the literature should contain but two references to the presence of a Golgi apparatus in the liver cells. Stropeni (169) is the only one who has succeeded, and he stated that in mammalian livers he obtained only a partial impregnation. With the livers of lower vertebrates, namely frogs and amphibians (Axolotl) he was more fortunate. He found the net to be definitely localized to the portion of cytoplasm between the nucleus and the bile-canaliculi, occasionally sending prolongations into the rest of the cytoplasm; no continuity with the bile canaliculi could ever be observed, nor was there any striking difference in the appearance of the net in fasting or well-fed animals. TllK COLCA Al'l'AltATL'S iiU KoliiKM- ( 100) says tluit lio succoodotl hut rarely in demonstrating- a net in tlie liver cells. In a new-hoin eat, the liver cells eontaiiieil a simple jiixla-nuch'ar net consist inj^ only of several iiu'siics or soiiK'tiiiics of single polyuoiial nets with one or two loiii>; processes. They had no constant iclation to the nucleus. We also have tried repeatedly to hnd a Clolf4i appai'atus in tlu^ li\er cells of rats, and have been almost uniformly unsuccessful with the (Jolgi or Cajal technique. In only one preparation were there found discontiimous cur\'ed filaments disti-ibuted through the eytoj^lasm, and bearing little resemblance to the complex reticulum found in other epithelial cells. By prolonged fixation in 2 per cent osmic acid, one may, however, demonstrate very clear-cut intracellular filaments and rows of granules, often curved and occasionally branching, but never uniting to form a definite network (figs. 6, 7, 8, 9, 10, 11). These filaments may lie against the nuclear membrane; in a few instances they aj^pear to join filaments in neighboring cells; in no case do they connect with the bile canaliculi, nor do they appear to reach the surface of the cell. \^Tiether these structures are the homologues of the Golgi apparatus in other cells, I am unable to say with certainty. Their resistance to impregnation by the usual methods implies some chemical variation. They disappear rapidly during autolysis and are absent in cells injured by chloroform poisoning. Surprisingly few workers also, have concerned themselves with the Golgi apparatus as it appears in the kidney. Brugnatelli (25), using the Golgi arsenic method has described a net in the ceils of the tubuli contorti and of the tubuh recti of the guinea pig, which coincides perfectly with that of other epithelial cells, especially as regards its localization between the nucleus and the lumen of the tubule. In the collecting tubules the apparatus was much more comphcated and definite than in the cells of the convoluted tubules, in which it invariably presented itself as small, very simple, almost simulated (i.e., 'accenata'). It seemed, he says, as if the more complex structures here (basal-rods, granules) were harmful to a clear-cut demonstration of the reticular apparatus. 120 ALWIN M. PAPPENHEIMER .S^ '^..^ J ^ 10 II Figs. 6, 7 and 8 Liver cells of rat. Kopsch, 2 per cent osmic acid, 12 days. Fig. 9 Liver cells of rat, autolysed 3 hours at 37°. Kopsch. Fig. 10 Liver cells, rat. Cajal. Fig. 11 Ljver cell, rat, vacuolated with localized juxta-iuiclear Golgi apparatus. Cajal. 'niK (;<)i.(;i aim-auai'I's 121 In tlic <;i(»iii('Hili, I lie Mppaiaius is red weed lo ils lowest Ici'ius — soinetiiiu's appearing' as a simple nodule, oi as a small li^ui'O 8, and always lyiiiji; adjacent (o ilie micleiis. l^iu«>;iia(elli j^ained tlie impicssioii ili.-N the net was ri'stiictcd to the cells of epithelial oiijiiii forming' the \isceral la>'ei' of liowmatis caj)sule. He is (luite wroiii>; iii this, as the endothelial cells and the purietal cells of the cai)sulai sj)ac(> also contain an easily demonstrable net (%12). Barinetti ('12) dc^sciihes and pictures a rather complex net in the renal epithelium, and shows its lelation to the centrosomes l)v compaiin^' it with impiegnations in which the centrosome is stained by Benda's method. He omits, however, to mention the portion of the renal tubule to which he refers. San Giorgi (156) studied the alterations of the Golgi apparatus during experimentally produced nephritis in guinea pigs. He used as toxic agents, uianium nitrate, cantharidin, ricin and diphtheria toxin. The modifications observed were a splitting up of the filaments or a granular fragmentation without complete loss of the reticular character. Such modifications w-ere most clearly observed in the tubuli recti of the medulla, in W'liich the net, as Brugnatelli showed, is normally more voluminous and complete than in the cells of the convoluted tubules. Close examination showed relation between alterations of the cells as a w^hole and of the Golgi apparatus. The fragmentation of the net may be marked in some cells of the tubules, whereas others may contain a normal net. One ma}^ criticize San Giorgi's work because of the fact that none of the poisons used produce ob\'ious changes in the cells of the tubuli recti. Kolmer (100) briefly records the presence of a net in various elements of the kidney, but gives no detailed description. We have made numerous preparations of rats' kidneys, both of normal animals and of animals in which a uranium nitrate nephritis had been produced. We have varied the time of fixation from two hours (as recommended by Brugnatelli) to twelve hours, without obtaining striking differences. We ha\'e also 122 ALWTN M. PAPPENHEIMER - > / ^^%. Ui 1 si* 01 ' )^ 4 A TIIK com; I AI'I'Ali' A IMS 123 studied, llioii^li iiicomplcU'ly, the effect of autolysis, and found that ('liaii<i;('s wwv discciual)!^ only after an hour at .'i7° when the kidney was reiiioNcd tVoiii the body immediately after death. A kidney removed from an animal one liour after death, showed about tlu^ same type of structure as the fleshly fixed tissue. The a])paratiis. therefore seems to be somewhat moi'e resistant to autolytic change than the miloehondiia, which after half an hour at 37"^ were broken up into coarse droi:>lets. As regards the appearance of the Golgi apparatus in the cells of the convoluted tubules, my preparations do not coincide at all with those of Bnignatelli (25) or San Giorgi (156). The argentophile structiu'es in these cells take on the most bizarre and \'arying forms. One finds smaller and larger droplets or granules, rings or signet forms often ending in a delicate filament, curved threads, uniform in calibre, or with nodular thickenings and varicosities; and larger, more complicated skeins approaching the reticulum described in other types of epithelial cells (figs. 13, 14, 15, 16). The locati(m of these structures with respect to the nucleus is as variable as their form. Their most frequent site is perhaps at either side of the nucleus, sometimes in contact with it; the filaments in general tend to run at right angles to the basement membrane. Sometimes one finds a cluster of granules and short filaments in the supra-nuclear zone, very rarely between the nucleus and the basement membrane. The appearance varies also from one tubule to another in the same preparation; some tubules may show predominently irregular twisted threads and skeins, others only granulae of uniform or varying size. The appearances are so bewildering in their ^•ariety that it is difficult to draw any conclusions, and we are still experimenting with the technique in the hope of getting more constant pictures. We have about decided, however, that the Golgi apparatus in the convoluted tubules is not the Fig. 12 Rat kidnej'. Glomerulus. Cajal. Fig. 13 Rat kidney. Proximal convoluted tubule. Cajal. Figs. 14, 15 and 16 Rat kidney removed one hour after death. Convoluted tubules. Golgi. Fig. 17 Rat kidney. Large Henle tubule. Impregnation of basal filaments. Golgi. THE ANATOMICAL RECORD, VOL. 11, XO. 4 124 ALWIN M. PAPPENHEIMER clear-cut definite structure which BrugnateUi depicts, but is normally fragmentary and dispersed. We have not been able to prove that these variations in form are correlated with different phases of secretion. In the Henle tubules, the net is obscured by a very constant impregnation of the Stabchen. Sometimes, however, we can distinguish a small, very dense, supra-nuclear network, and occasionally filaments are continued along the lateral aspects of the nuclear membrane (fig. 17). In the cells of the collecting tubules also, a net has been frequently seen, although in the rat's kidney it is a looser and less complex structure than that- described by the Italian workers. In the glomerular cells, both epithelial and endothelial, the small juxta-nuclear net is constantly found, and very sharply defined and striking (fig. 12). In the uranium nitrate kidney I have made only a few observations which seem worth mentioning at this time. In the first place, the more complicated filaments and skeins, usually evident in the normal tubules, tend to disappear entirely in the injured kidney. Even those cells, which in sections counterstained with Giemsa, show little or no obvious damage, rarely show structures other than granules or irregular black or grey-staining clumps. In the totally necrotic and desquamated cells, one often finds a single coarse black clump possibly representing the remains of the argentophile structure. In many of the injured cells I find oval or circular greyish bodies of varying size, many of which contain one or two eccentrically placed black granules (fig. 19). Now starting with these, one can trace transitions both to small solid black granules and to large droplets which have entirely lost their affinity for the silver-stain, taking the eosin of the Giemsa intensely, and resembling in every way the familiar hyaline droplets of degenerating renal cells. These largest droplets accept the acid fuchsin in the Altmann-Bensley stain, but I have gained the impression that they arise from the argentophile droplets rather than from a breaking-up and fusion of the mitochondrial structures. 'I'lii'j (loLcM AIM'. \i(A res o .€>' 18 12.") Fig. 18 Rat kidney-. Two small collecting tubules with delicate supranuclear filaments. Cajal. Fig. 19 Rat kidney. Uranium nitrate, Xephritis. Pro.ximal convoluted tubule containing hyalin droplets with argentophile granule. Cajal. Fig. 20 Frog's kidney. Tubules showing various types of argenthophile structures. Cajal. 126 ALWIN M. PAPPENHEIMER Fahr (52) has also recently made the observation that this type of 'gross-tropfige' degeneration may be produced experimentally in the rabbit's kidney with uranium nitrate, and is disinclined to derive the droplets from the mitochondria of the cell. The appearances observed in the epithelial cells of the frog's kidney are also very puzzling and difficult to interpret. In the glomeruli, a concentrated juxta-nuclear mass is present in all the cells, identical with that described in the rat's kidney. In the proximal portion of the convoluted tubule there is found to either side of the nucleus, but rather nearer the distal pole, a small irregular, granular filamentous or ring-shaped mass, which is quite definitely the homologae of the Golgi apparatus of other cells. Such an epithehal cell, cut in a plane parallel to the basement membrane, shows the nucleus surrounded by a ring of discrete masses, which do not form a continuous skein, but are interrupted (fig. 22A). In frogs injected with trypan-blue this type of apparatus is present in the cells containing the granules of dye, which in general occupy the supra-nuclear zone of the cytoplasm. The dye granules and the argentophile bodies are quite unrelated. In some of my preparations, the cell boundaries are sharply impregnated by the silver, appearing as delicate black lines. The striated border is also sharply brought out (fig. 22). In another part of the tubule, probably the distal portion of the convoluted tubule or Schaltstiick, which contains no blue staining granules, the argentophile bodies are in the form of rounded globules or granules, varying slightly in size and intensity of staining, and occupying a zone in the middle nuclear plane (fig. 22B). The homology of these granules to the Golgi apparatus is not clear, and it is possible that they represent excretory substances of some sort, possibly chlorides or phosphates (Leschke (109) ). Finally, in still another portion of the tubules (corresponding to the Henle loop) there is obtained an excellent impregnation of the basal filaments and mitochondria (fig. 22 C) . The cytoplasm in the supranuclear zone is somewhat more intensely stained, THE COLCl Ai'l'AUATrs 127 Init IK) (Idiiiil.c sti'uctui'cs coiiipaiahlc to the ( loli;! ;i|)|);iraius of otluT cells is hrou^ht out. This ({('script ion corresponds to the a|)pc:irauc('s usually obserx'cd in the tV(ij>;'s kidney. Some oi" our pr('|)aialioiis, however, sliow curious structuK's of very diHereut tyi)e, tlu^ nature of which is entirely obscure. These are limited to certain ])ortions of the secretory tuhuk^s — probably the distal convoluted portion or Schaltstiick, altiiougli it is not possible to be sure of this. They consist of sheaves of filaments, often of great length, sometimes beaded or with nodular varicosities. They run either perpendicular to the basement membrane, along the lateral aspects of the nucleus, or in some instances, lie above the nucleus and have a com'se more or less parallel to the basement membrane (figs. 20-21). These bundles of fibrils do not appear to form closed skeins or to anastomose with one another, but overlie and cross. The individual fibrils are often irregularly fusiform, w-ith tapering ends. They may be quite rigid, almost crystalline in appearance, or more wavy and filamentous. Scattered amongst them are small isolated clumps and granules of varying size. b. In ordinary connective tissue cells. A small juxta-nuclear apparatus has been described by v. Bergen (10), Cajal, Deinecke (47) and Kolster (102) ; in endothelial cells by v. Bergen (10) and by Cajal (28); in smooth muscle cells of blood vessels by V. Bergen (10) ; and in \'arious types of wandering cells by v. Bergen (10), Verson (179) ]Maccabruni (115), Barinnetti (6) (Plasmacells — relation to centrosome, T2) and Fananas (55). In cartilage cells, Pensa (136) first described a distributed apparatus which he considered to resemble the diffuse net found by Golgi in the ganglion cell. Later v. Bergen showed that the Golgi apparatus in cartilage cells, as in most other non-nervous elements, was limited to the juxta-nuclear region, and interpreted Pensa's diffuse apparatus as a chondriom. Pensa (137) has accepted this correction, and v. Smirnow (165), Barinetti (6) and Kolster (102) are in agreement with v. Bergen on this 128 ALWIN M. PAPPENHEIMER / A ^ ^^•* C 22 Fig. 21 Frog's kidney. Tubules showing argentophile filaments. Golgi. Fig. 22 Frog's kidney. A — Proximal convoluted tubule. Golgi apparatus in the form of discrete peri-nuclear filaments, rings and loops. B — Tubule containing argentophile droplets. C — tubule showing impregnation of basalrods. point. ( 'oiiH's (40. 41) still claims t.hat ( J()lj>;i net and iiiitochoiulria in caitila^c cells arc identical structures. The controN I'isy is interesting, because it has brou^lit out the point that the siKci' method is not always specific, and that und(M" certain conditions the mitochondria may be imjirepniJitod. This occurs i'e«i;ularly, as I ha\e mentioned, in the Ilenle tul)ules ol" the kidney. lV)th Pensa (13(3) and Cajal (31) have described an interesting series of changes in the zone of growing cartilage adjoining the line of ossihcation. Following the enlargement of the cells the net loses its localized character, hypertrophies and finally, with the degenei-ation of the cartilage cell, undergoes granular disintegration. Cajal (31) seems to have been the only one to study the behavior of the CJolgi apparatus in osteoblasts. During the period of fimctional activity, the net is very well developed, and large, usually occupying that part of the cell directed towards the osteoid tissue. In the finished bone corpuscle, the net shrinks to a small compact mass. Teeth. Although ]\Iassenti (123), using pig embryos, had previously described a Golgi apparatus in the form of a large skein occupying almost the entu-e cytoplasm of the pulp cells and odontoblasts Cajal (31) depicts structures of a more typical character and localization. As the odontoblast becomes differentiated from the connective tissue elements of the pulp, the net increases in size, and comes to form a large oval rather granular mass, occupjdng the supra-nuclear portion of the cytoplasm. The further fate of the structure could not be followed, since decalcification interferes with the reaction, but Cajal regards the hypertrophy during the secretory phase of the odontoblast as another example of the cyclical changes seen in goblet cells, growing cartilage cells, etc. Muscle fibers. There are a number of papers dealing with the endo-cellular reticulum of striated muscle fibers. The earliest is that of Cajal (26), in which he found in the wing muscle of certain insects a network continuous with the ramifications of the tracheal tubes, and therefore probably a true cana 130 ALWIN M. PAPPENHEIMER licular sj^stem. This was confirmed by Fusari in 1894 for mammalian muscle and by Veratti (178) in 1902. Veratti, however, denied that the endocellular apparatus in insects represented a continuation of the tracheal tubes, and regarded it as composed of solid filaments. Sanchez (155), a pupil of Cajal, confirmed the tracheal origin of the reticulum in insects, looking upon it as a tubular apparatus, probably of importance for the nutrition of the fibers. The system is composed of transverse meshes on either side of the 'bandes claires,' united by longitudinal connections. Two other recent papers, one by Martinotti (122), the other b}'- Fananas (54) have not been available. Quite different is the apparatus described by Luna (113) in the cardiac muscle fibers. Here he finds granules, rods, curved filaments or more complete nets lying at one or both poles of the nucleus. He believes that the granules and rods might possibly be mitochondrial. c. Gonads. The studies upon the gonads are of much more theoretical interest, because of their bearing upon the question as to whether the Golgi apparatus is a permanent structure, and a heritable constituent of the cytoplasm, or whether it is more ephemeral in character, and related to the vegetative activities of the cell. It is also in the sperm cells that the topographical relation of the net to the centrosphere is most evident, so that the structure during spermatogensis might be expected to show interesting modifications. Platner (145), a number of years before the discovery of the Golgi apparatus, had described a structure surrounding the centrosomes of the spermatogonia, which he called a Nebenkern. Heidenhain (68) in 1900, in the sperm cells of Proteus, found with the iron hematoxylin stain, an incomplete basket-work surrounding the centriole. Heidenhain used the term Zentralkapsel or pseudo-chromosomes for these bodies, and believed that they were formed by special differentiation from the Benda mitochondria. This view finds support in the recent observations of Chambers (33) upon spermatogenesis in the grasshopper, in 11 IK (lOLca Al'l'AItATUS 1 -'U wliicli the iiiitoohondi'iii. vitally stained with Jamis ^rooii, ap])oaro(l to enter (lirectly into tlu^ foriiiatioii of the Xeheukeni. Sj(')\all iKll), iisiujr a modilicatioii ol lli<' Kopsch inclliod, dcMiioustrated these struct uies in the sj)ermat()eytes, spermatogonia and spermatids of Lhe white mouse, and concluded that Heidenhain's view of their mitochondrial origin was erroneous. Benda, and especially Weigl (180), have also taken this stand. Sj()vall also found a net in the Sertoli cells. IVrroncito (140, 141), ushig the (lolgi method, carried tiic observations of Sjovall a step further, by describing the alterations of the net din-ing the maturation divisions. He finds in Paludina, that in the prophase, the net breaks up into granular fragments, which he calls dittosomes, and which are distributed equally to the t\\() daughter cells. The skeins are then reformed from the granules, and in the sj^c^rmatids come to occupy theii' usual juxta-nuclear position. This process of fragmentation and distribution during mitosis he calls dittokinesis. The fate of the Golgi apparatus in the adult spermatozoa is unkno^^^l, nor has it; been established that the substance of which it is composed enters the egg during fertilization. The phenomenon of dittokinesis, however, is established, not only for the sperm cells, but also for the somatic cells (fig. 4). Deinecke (47) ('12) has given a very clear description of this process in the flat cells of Descemet's membrane, where the net forms in the resting cell a thick skein of interwoven and anastomosing threads. During karyokinesis the skein becomes looser and gradually surrounds the nucleus, the individual threads grow thicker, loose their connection and break u}) into unequal bent fragments which are heaped up at the poles of the nucleus. During anaphase they change into short thick rods and granules surrounding the nuclei, and lying chiefly in the equatorial plane. The size of these dittosomes is only slightly smaller than that of the chromosomes, their number somewhat greater. With the formation of the diaster, the dittosomes surround the daughter chromosomes, being more densely aggregated at both poles. The region of the spindle remains free of them. 132 ALWIN M. PAPPENHEIMER The new nets are formed by a fusion or sticking together of the granules or rods, but may remain discreet for a time. In this way, it is possible to recognize a recent mitosis, even after the nuclei have reformed. During the monaster stage one sees often a pairing of the granules and double rods. AMiether this indicates a splitting of the dittosomes comparable to that of the chromosomes, Deineke leaves undecided. At any rate, the above series of changes unquestionably leads to an even distribution of the mass to. the two daughter cells. Studies of the Golgi apparatus in the female gonads have been made by Sjovall (164), Weigl (180), Cattaneo (32), and Kulesch (104), and Weigl (180) and Ilirschler (70) in the ovocytes of invertebrates; in the primitive germ-cells of 3-4 day chick embryos by v. Behrenberg-Gossler (16). All these writers are in substantial agreement as to the main facts namely, that in the young ovocytes and in the follicle cells, there is a circumscribed net at one pole of the nucleus, which in the ripe ovum breaks up into filaments and granules (or, according to Kulesch, small irregularly angular rings, bent threads and discs) which are with difficulty distinguished from the mitochondria and other cytoplasmic granulations. V. THE GOLGI APPARATUS IN EMBRYOXAL CELLS The recognition by Golgi (64), Fananas (54) and Cajal (31) that the Golgi apparatus is present in all types of cells, even at a very early stage of development (chick embryos of 30-40 hours — Cajal) seems to establish firmly the principle that the structure is an important and constant component of the cell. In many of these fetal cells — the mesenchyme — the endothelium of the pericardium and of the primitive blood spaces — the cells of the Wollfian ducts and the entoderm of the intestinal tract, the neuroblasts, and even the erythroblasts and wandering cells, the apparatus is highly typical and constant in its relation to the centrosphere. The attempt has been made by Fananas to trace the development of the net from granules and batonnets in the cytoplasm. TlIK (iOLCI AI'l'AKArrS 133 As Cajal (.SI) points onl, howcxci', tlic not iiit'i<'(|U('iit iiii|)i('jz;n:ition of t.lie niitochomlria and of skeletal and snsleiitaiailar structures in eai-|\- enihiyonic cells makes such an interpretation doulitful. One general |)iincii)le can be deduced from a study of the Golgi ajiparatus in enil)ry()nic tissue, and that is the definite polarity of tlie structure in all fixed non-mobile cells. The location of the net in every case, as Cajal has pointed out, is such that it occupies the 'external' part of the cell, — that is, the portion ahove the nucleus directed originally towards the free surface, and opposite that pole which is towards the interstitial tissue and the nutrient supply. This j^olarity is preserved in adult life in the case of epithelial cells lining ducts and cavities, but may be lost in the case of solid glandular organ or tissues which undergo i)rofound modification and dei'angement during development. The signihcance of this ])olarity which would seem to be bound up w^th the relation of the Golgi net to the centrosphere, is by no means clear, but it seems to be one of the most fundamental and most striking characteristics. VI. THE GOLGI APPARATUS IX PROTOZOA The literature contains but one reference to the occurrence of the Golgi apparatus in Protozoa. Hirschler (70) describes in the cytoplasm of ^lonocystis ascidiae, a Gregarine parasite of the ascidian Ciona intestinalis, diffusely scattered ring and half ring forms, demonstrable by prolonged exposure to 2 per cent osmic acid and resistant to turpentine (Sjovall's modification of the Kopsch method). Whether these structures are the homologues of the Golgi apparatus of the metazoan cell, needs further study. VII. THE GOLGI APPARATUS UXDER PATHOLOGICAL COXDITIOXS The modifications which the Golgi apparatus undergo under pathological conditions have been little studied, and, so far, have not added any new suggestions as to the real nature of the structure. 134 ALWIN M. PAPPENHEIMER In the cells of malignant growths, Golgi nets, often atypical, have been found by ]\Ioriani (125) in a human breast carcinoma, by Veratti (177) in a transplantable mouse cancer; by Lucioni (112) in a naevus; by Savagnone (157) in carcinoma of the breast, in a sarcoma of the jaw and in a giant-celled sarcoma; by Tello (172) in carcinoma and adenoma, in epithelioma and in experimental granuloma caused b}^ Kieselguhr injections. Tello studied especially the distribution of the net in foreign body giant cells, where there are multiple nets — one usually in rela-. tion to each nucleus. In tuberculous giant cells, on the other hand, as shown by Fananas (55), the net is usually centrally placed, in relation to the multiple centrosomes. Regressive changes (fragmentation, pulverization, etc.) in the Golgi net have been described by Fananas (55) in caseating giant cells; by Marcora (117) in the ganglion cells of the hypoglossal nucleus, following avulsion or section of the nerve; by Battistessa (8) in the ganglion cells of animals poisoned by lead or strychnin; by San Giorgi (156) in toxic nephritis, and by Del Rio Hortega (97) in the ganglion cells of a case of paralytic rabies. Ver}' interesting are the recent experimental studies of Cajal (31) dealing with the effect of traumatic injury of the nerves upon the Golgi apparatus. An incision of the central nervous tissue brings about complete destruction of the net only in those ganglion cells most severely injured bj^ the trauma. The apparatus of cells near the line of incision, though perhaps slightly compressed or deformed, shows no grave disorganization. This would indicate a considerable fixity of structure, and firmness of texture, since, were the impregnated substance of fluid consistence, one would expect to find it dispersed through the cytoplasm or confluescing into larger droplets. Cajal has further established the fact that section of a peripheral motor nerve had no effect upon the structure of the Golgi apparatus in the central ganglion cell. There does occur a degeneration of the apparatus in the cell of the sheath of Schwann, distal to the section. TllK (JOLCI AlM'AliATlJS 135 TIkmv arc tluis very fow controlled studies of the idteratious of the net under experinicuttd conditions, :ind it seems thii* st)niething- fiu-ther should be added to our kn()wled}i;e in that w;iy. The great difiiculty had been, and will be, the capricious belKivior of inii)i(^*;nation methods; until simpler and more reliable methods shall have been discovered, the interi)retation of sliiiht variation in the morphology of the structures will always be open to considerable suspicion. VI rr. GEXERAL CONSIDERATIONS This completes the list of cells and tissues in which structures of this type have been demonstrated. They may be considered as universally present in every type of cell, although the \'ariety of form which they assume at once brings up the querj^ as to whether they are all homologous structures. That, I think, is almost impossible to answer until we know something of their function and significance. ^Morphologically there seems to be no single character by which we can grou]:) them altogether. The staining reactions are probably not entirely specific; we have found instances — as for example in the Henle tubules and in cartilage cells — in which mitochondrial structures are more or less regularly impregnated by the silver methods. In many types of cells in which the location of the centrosome is known, there is, as Barinetti (6) insists, a topographical relation between Golgi net and cytocentrum. But such a relation cannot be established for the ganglion cells, in which the net completely encucles the nucleus, nor for the muscle cell, nor for the cells of the choroid plexus, still less for the cells of the convoluted tubules, or for the liver cells. So that this criterion is not universally applicable, at least to fully differentiated and highly specialized types of cells. IVIuch of the discussion about the nature and homologies of these things has hinged about the question as to whether they are solid, that is fibrillary; or canals filled with fluid, and made to appear as solid filaments by the metaUic mipregnation methods. With the canalicular theory the name of Holmgren is associated, and though there are many shades of opinion in 136 ALWIN M. PAPPENHEIMER regard to detail the general idea that these nets and filaments represent canals with or without definite walls, has had the support of such expert histologists as Studnicka (170), Retzius (149), Cajal (31), and, in this country, of Bensley (14) and Cowdry (42). Holmgren was not the first to observe 'endocellular canals' or 'Saftkanalchen.' As far back as 1887 Nansen (129) described in the protoplasm of nerve cells of Homarus and in the spinal ganglion cells of Myxina glutinosa, primitive tubes consisting of h^'alin contents enveloped in sheathes of spongioplasm. These were probably identical with the Saftkanalchen. Nelis (132) also described an 'etat spiremateux' in the cytoplasm of certain mammalian nerve cells. He speaks of 'bandes incolores' of about the same diameter, sometimes straight, sometimes convoluted which did not branch, and therefore formed no reticulum. This spireme was not constantly found, and does not seem to be the same thing as the Golgi net. Holmgren (73) made his first contribution to this subject in 1899, a year after Golgi's first paper, and without, apparently, know^ing of the work of Nansen and Nelis. In the spinal ganglion cells of rabbi ts and of Lophius piscatorius he found an endocellular system of canals communicating with the pericellular lymph spaces; and in the following year he published a large monograph on the ganglion cells of Lophius (72). His earlier views, expressed in these papers, that the endocellular Saftkanalchen are to be regarded as lymph channels, and are thus continuous with structures of connective tissue origin, were later abandoned by him. Morphologically Holmgren considered his canaliculi to be identical with the Golgi net, basing his opinion upon a study of his own preparations, and those of Retzius, prepared according to the earlier Golgi technique. In 1900 (76), in a study of the ganglion cells of Helix pomatia Holmgren described a penetration of the nerve cells by cellprocesses from surrounding cells. This view was developed in subsequent studies. The penetrating cells of neuroglial origin he called trophocytes, and to the netw^ork formed by the penetrating cell processes, he applied the term trophospongium. TMK (i()L(il Al'TAKAITS 137 Those piolontiatioiis of tlio ti'()j)li()('yfi('S, he hclicxed, l)oc'niiio canalized by a \':i('uoli/ati()U i)f their c3'to])lasin, (he eonfluescencc of the \ aciioles j^ivins origin to the canal. This was an irreversible process, but the cell prolongation was capable of ainoel)oi<l motion within the host cell. Studies on various tissue cells, which I sliall not review in detail, confirmed him in his view, and lead him to thefollowing generalization. The cells of the bod}' are of two orders of physiological dignity ^ligh and low. Those of exalted function are the nerve cells, muscle cells, sex cells, certain glandular cells. The lower order of cells, which are the trophocytes, function as servitors, looking after the wants of their more specialized neighbours b}- means of their trophospongia. Although this hypothesis is vaguely expressed, and open to obvious criticism, Holmgren has maintained it in a long series of papers (77-96), many of them controversial, and adding no new evidence. The arguments against any such generalized conception are apparent enough. To what order shall w^e assign the leucocytes and other wandering cells? Where are the trophocytes of cartilage cells? Why have the trophocytes about the ganglion cells, as well as all the other types of cells classed with the lower order, endocellular nets, and what cells look after their lowh'- wants? Holmgren has always insisted vigorously upon the identity of his trophospongium wdth the Golgi apparatus; on the other hand, all who have worked with the Golgi or Cajal methods deny that the structures which they bring out reach the surface of the cells or communicate wdth other cells. Even Cajal, who regards the Golgi structures as canaliculi, believes them to be wholly endocellular, except perhaps in the special case of the insect muscles, in which the homology with the Golgi net is not very clear at best. Ross (153) in a recent paper on the trophospongium of the ganglion cells of the crayfish, describes the penetration of the cytoplasm by partitions and fibrils from the surrounding neuroglia, but rejects the idea that these have any relation to the internal reticular apparatus. 138 ALWIN M. PAPPENHEIMER This view, I think, may be accepted without reserve; nor does it seem that Hohiigren's generalizations are based on sound evidence, nor that they have added much of value to the subject. Leaving aside, then, this controversial phase of the subject, one may ask what can be said as to the more intimate physical structure of the Golgi apparatus. It seems to the writer, that the conceptions of Cajal (31) best meet the observed and established facts. Cajal's view is that the apparatus represents a canalicular system, filled with a lipoid-containing substance which reacts to the specific impregnation methods employed. The walls of this system are presumably fairly fixed and rigid in cells of a sedentery habitus, and permanent form, but more plastic in secretory cells and in young cells frequently undergoing mitosis. It seems probable that the quantitative changes observed during activity indicate a using up of a store of stainable material within these canaliculi, and that the re-appearance of the net during the quiescent stage is due to the re-accumulation of the substance within more or less preformed and permanent channels. Wliat purpose this material serves in the cell metabolism, and what is its more intimate chemical structure, are questions unanswerable with the data at hand. IX. SUMMARY AND CONCLUSIONS There is present in the somatic and sex cells of all metazoa, and possibly also, of protozoa, a cytoplasmic structure of considerable complexity and size, demonstrable by prolonged fixation in osmic acid, or by silver impregnation and reduction. The reaction of this structure to osmic acid indicates, of course, a lipoid component, but there are no other data bearing upon its chemical composition. Nor is anything certain known of its physical characters. Its invisibility in the living cell would indicate a low refracti\^e index. The fundamental question as to whether the impregnated structiu^es are canalicular or filamentous remains unsolved. The constant topography in many types of cells, particularly the definite relation to the cytocentrum would favor the idea that the structures are at least in part TllE COLCI APPARATUS 139 solid, rather than casual rifts or fluid-filled canals in the cytoplasm. The fraf2;ineutaiion or tlispersion of the net which occurs duriiiji; certain secretory i)hases, or accompanying pathological changes in the cell, and particularly during cell division, would also suggest a solid or semi-solid consistence. The Ciolgi api:)aratus in the secretory portion of the renal tubules does not conform to the usual closed skein found in many types of glandular epithelial cells, but is dispersed and assumes complex and varying forms. In the glomei-ular cells, on the other hand, and in the epithelium of the collecting tubule and ot the pelvis the structure is more typical both in form and location. Injury to the epithelial cells of the convoluted tubules (uranium nitrate poisoning) is followed by complete disintegration and disappearance of the Golgi apparatus. This appears to precede the complete necrosis of the cell. The large hyalin droplets found during the degeneration of the cells contain an argentophile component possibly derived from the remains of the Golgi apparatus. The structures brought out by the Golgi or Cajal technique are more resistant to autolysis than are the mitochondria. There occurs regularly in the rat's kidney an elective impregnation of the mitochondrial filaments of certain portions of the Henle loops. This illustrates the fact that the method is not absolutely specific. The writer wishes to express his thanks to the Marine Biological Laboratory of Woods Hole for according him the privileges of the Laboratory during the summer of 1915-1916; and to Dr. E. V. Cow^drey for helpful suggestions and criticism. THE ANATOMICAL RECORD, VOL. 11, NO. 4 140 ALWIN M. PAPPENHEIMER BIBLIOGRAPHY (1) AccoNTi, 1912 Di alcuni fini particolarita di struttura della mucosa uterina della decidue e dell' uovo. Boll. Soe. Med. Chir. di Pavia, 25, 125. (2) Arnold, J. 1908 Haben die Labzellen Membranen u. Binnennetz. Anat. Anz., 32, 257. (3) Ancona, 1909 Apparato reticolare nelle cellule della ghiandola lachry male. Diss, di laurea (unedita) Pavia. (4) Ballowitz, 1900 Ueber d. Epithel. d. membrana elastica posterioris des Auges, seine Kerne u. eine merkwerdige Struktur seine grosse Zell-sphaeren, u.s.w. Arch. Mikr. Anat., 56, 230. (5) 1900 Eine Bemerkung zu den v. Golgi und seine Schiilern beschriebenen 'apparato reticolare intorno' der Ganglien u. Driisenzellen. Anat. Anz., 18, 177. (6) Barinetti, 1912 L'apparato reticolare intorno e. la centrospera nelle cellule di alcuni tessuti. Boll. Soc. Med. Chir. di Pavia, 25, 289. (7) 1911 Di una fina particolarita di struttura nelle cellule del epitelio della cornea. Boll. Soc. Med. Chir. di Pavia, 5. (8) Batistessa, 1911 Sulle alterazioni dell apparato reticolare interuo delle cellule nervose dei gangli spinali in seguito ad advelenamento da piombo e stricnino. Riv. ital. di Neuropatologia, 4, 345. (9) Benda, 1902 Weitere Mittheilungen liber die Mitochondrien. Ergebn. d. Anat. u. Entw., 12. 743. (10) V. Bergen, 1904 Zur Kentniss gewisser Strukturbilder (Netzapparate Saftkanalchen, Trophospongien) im Protoplasm verschiedener Zellarten. Arch. f. Mikr. Anat., 64, 498. (11) Besta 1910 Suir apparato reticolare intorno (apparate di Golgi) della cellula nervosa. Anat. Anz., 36, 476. (12) 1911 Ricerche sull reticolo endocellulare degli elementi nervosie nuovi metodi di dimonstrazione. Riv. di patol. nerv. e. ment., 16, 341. (13) Bialkowska and Kulikowska, 1911 Ueber d. Golgi-Kopsch'en Ap parat der Nervenzellen bei den Hirudineen u. Lumbricus. Anat. Anz., 38, 193. (14) Bensley, R. R. 1910 On the nature of the canalicular apparatus of animal cells. Biol. Bull. Mar. Biol. Lab. of Woods Hole, 19, 179. (15) 1911 Studies on the pancreas of the guinea pig. Am. Jour. Anat., 12, 297. (16) Berenberg-Gossler, 1912 Ueber gitterkapselartige Bildungen in den Uhrgeschlechtszellen v. Vogelembryonen. Vorl. Mitth. Anat. Anz., 40, 587. (17) 1912 Verh. d. Anat. Ges. Muenchen., 26, 263. (18) 1912 Die Uhrgeschlechtszellen der Hiihnerembryonea a.m 3ten u. 4ten Bebrijtungstage, mit. besonder Beriicksichtigung der Kernplasma Strukturcn. Arch. f. Mikr. Anat., 81, Abteilung, ii, p. 24. (19) Bialkowska u. Kulikow.ska, 1911 Ueber den feineren Bau der Ner venzellen bei verschiedenen Insckten. Bull. Acad. Sc. Cracovie. CI. Sc. math, et nat. THK (;()l,(il Al'l'MiVTUS 141 (20) BluNUi, I'.Ul Sulla liiwi .stiulliiia ilrll' cpilcslic dci pic.s.si i;(jrui(l(!i. Arch. f. Zcllforscli., (5, ;5S7. (21) BizzozKKo !•: Mottksklla, 190!) Sull' apparato intenio nelle cellule dolle uliiaiulole sudoripare e sebacee. \vv\i. per le Sc. med., 33, 279. (22) Bjokkhnhkim, 1913 (lol-fi's ajjparato roticolaio intoruo in den Pla centar-opithelien. Arch. I". (Jynek., 101), 44(j. (23) BouiN ET Ancel, 190;") Apropos du 'tropiio.sponp;iuin' et de.s 'canali culcs du sue' Coniptcs Rend, dc la Soc. de Biol., 2, 221. (24) Buowicz, 1902 Moine Ansicht iil)er den Hau der Leberzelle. Virch. Arch., 168, 1. (25) Brugnatelli, 1908 Di una fina partioolarita di strutturi degli epitelii dei tubuli renali. Boll. Soc. Med. Chir. di Pavia, 22, 96. (26) Cajal, Ramon y, 1890 Coloration par la mdthode de Golgi des ter minaisons des traehees et des nerfs dans les muscles des ailes des insectes. Zeitsehr. f. Wi.ss. Mikr., 7, 332. (27) 1904 El apparato tubuliforme del epitelio intestinal de los mamiferos. Trab. del Lab. Invest. Biol. Madrid, 3, 35. (28) 1908 Les conduits dc Golgi-Hohngren du protoplasm nerveux at le reseau pericellulaire de la meml)rane. Trab. del Lab. Invest. Biol. Madrid, 6, 123. (29) 1912 EI apparato endocellular de Golgi de la celula de Schwann y algunas observaciones sobre la estructupa de los tubos nervosa. Trab. del Lab. Invest. Biol. Madrid, 10. (30) 1912 F6rmula dc fijacion para la demonstraci6n fdcil del apparato reticolar de Golgi y. apuntes dobre la disposici6n de diche apparato en la retina, en las nervios y algunos estados patologicos. Trab. del Lab. Invest. Biol. Madrid, 12, 209. (31) 1914 Algunas variaciones fisiologicas y patologicas del aparato reticular de Golgi. Trab. del Lab. de Invest. Biol., 12, 127. (32) Cattaneo, 1914 Ricerche suUa struttura dell' ovario dei maminiferi. Arch. ital. de anat. e di embriol., 12, 1. (33) Chambers, 1915 Microdissection studies on the germ cells. Science, NS., 41, 290. (34) CiAccio, 1903 Communicazione sopra i canaliculi di secrezions nella capsula soprarenali. Anat. Anz., 22, 493. (35) CoHN, 1903 Zur Histologic u. Histogenese des Corpus luteum u. des interstitieilen Ovarialgewebes. Arch_ f. Mikr. Anat., 62, 745. (36) CoLSON, 1910 Histogenese et structure de la capsule surrenale adulte. Arch, de bioL, 25, 535. (37) Collin et Luciex, 1909 Observations sur le reseau interne de Golgi dans les cellules nerveuses de mamiferes. Comptes rend, de la Soc. Anat., Nancy. (38) 1914 Algunas variaciones fisiologicas y patologicas del aparato reticular de Golgi. Trab. Lab. de Invest. BioL, 12, 127. (39) 1909 Sur les rapportes du reseau interne de Golgi et des corps de NissI dans la cellule nerveuse. Bibl. Anat., 19, 123. (40) Comes, 1909 Sulla natura mitocondriale dell' apparato reticolare delle cellule cartilagine. Boll. Acad, giolinica di sc. nat. Siena, Ser. II, fasc. 6. 142 ALWIN M. PAPPENHEIMER (41) 1913 Apparato reticolare o condrioma? Condrocinesi o dittocinesi? Anat. Ant., 43, 422. • (42) CowDREY, 1912 The relations of mitochondria and other cytoplasmic constituents in spinal ganglion cells of the pigeon. Int. Monatsch. f. Anat. u. Phys., 29, 1. (43) D'Agata, 1910 Sulle modificacazioni dell' apparato reticolare interno neir epitelia della mucosa gastrica. Boll. Soc. Med. Chir. di Pavia, 25, 517. (44) 1910 Di una fina particolarita di struttura delle cellule epiteliali della cistifellea. Boll. Soc. Med. Chir. di Pavia, 25, 531. (45) 1911 Ueber eine feine Struktureigenthiimlichkeit der Epithelzellen der Gallenblase. Arch. f. Mikr. Anat., 77, 78. (46) Decio, 1910 Sulla minuta struttura dell' epitelio uterino. Boll. Soc. Med. Chir. di Pavia, 25, 476. (47) Deineke, 1912 Das Netzapparat von Golgi in einigen Epithel- u. Bindegewebszellen wahrend der Ruhe und Teilung derselben. Anat. Anz., 41, 289. (48) Duesberg, 1914 Trophospongien und Golgi'schen Binnenapparat. Verh. d. Anat. Ges., 46, 11. (49) 1912 Plastosomen, 'Apparato reticolare interno,' und Chromidalapparat. Erg. d. Anat. u. Entw., 20, 567. (50) Ellermann, 1899 Ueber die Struktur der Darmepithelzellen von Helix. Anat. Anz., 16, 590. (51) Erhard, 1910 Studien iiber Trophospongien. Zugleich ein Beitrag. z. Kenntniss der Sekretion. Festschr. f. R. Hertwig, 50, 133. (52) Fahr, 1914 Zur Frage der Hj^alintropfige Zelldegeneration. Verh. d. deutschen Path. Ges., 17, 110. (53) Fananas, 1912 El apparato endocellular de Golgi de la mucosa y bulbo olfactorios. Trab. Lab. Invest. Biol. Madrid, 10, 253. (54) 1912 Nota preventiva sobre el apparato reticolar de Golgi en el embri6n de polio. Trab. Lab. Invest. Biol. Madrid, 10, 247. (55) 1913 Alteraciones del aparato reticular de Golgi en los cellulas gigantes y altros elementos del tuberculo. Trab. Lab. Invest. Biol. Madrid, 11, 119. (56) Fatjre, Fremiet, 1910 Apropos d'une note de M. Perroncito sur le reseau de Golgi des cellules spermatiques. Bull. Soc. Zool. de France. (57) 1910 Un appareil de Golgi dans I'oeuf de I'Ascaris megalocephala. Response a M. Perroncito. Bull. Soc. Zool. de France. (58) Fedeli, 1912 Apparati retocolari e sarcolemma nella fibra muscolare cardiaca. Rend. d. R. Accad. d. Sc. fis. e mat. di Napoli. (59) Fragnito, 1901 Le developpement de la cellule nerveuse et les canali cules de Holmgren. Bibl. Anat., 9, 72. (60) Gemelli, 1900 Ricerche sperimentali suUa struttura della ghiandola pituitaria nei mammiferi. Boll. Soc. Med. Chir. di Pavia, quoted by Golgi-Arch. per le Sc. Med., 1909, 33, 5. (61) Golgi, 1898 Intorno alia struttura delle cellule nervose. Boll. Soc. Med. Chir. di Pavia. 1898 Arch. ital. de biol., 30, 60. THE GOl.CX APPARATUS 143 (62) ISOS Sulla struttuni delle fcllule nervoso dei Kunuli spiniili. Boll. Sor. I^Icd. Chir. di Pavia. Arch. ital. de hiol., 30, 278. (63) I'JCK) Di nuova suUa struttura delle cellule nervo.se dei ^angli spinali. lioll. Soc. Med. Chir. di Pavia. 18i)!> Arch. ital. de bid., 31, 273. (64) 19(X) Intorno alia struttura delle cellule della corteccia cerebrate. Verh. Anat. Ges. Pavia, 164. (65) 1909 Sulla struttura delle cellule nervose della corteccia del cervello. Boll. Soc. Med. Chir. di Pavia, 23, 341. (66) 1908 Une m6thode pour la pronipte et facile dcjinonstration de I'appareil reticulaire interne des cellules nerveuses. Arch. ital. de biol., 49, 269. (67) 1909 Di una minuta particolarit^ di struttura dell' epitelio della mu cosa gastrica ed intestinale di alcuni vertebrati. Arch, per le Sc. Med., 33, 1. 1909 Boll, della Soc. Med. Chir. di Pavia, 1. (68) Heidexhain, M. 1900 Ueber die Centralkapseln u. Pseudochromoso men in den Samenzellen. Anat. Anz., IS, 513. (69) Henschen, 1904 Ueber Trophospongienkaniilchen sympatischer Gan glienzellen beim Menschen. Anat. Anz., 24, 385. (70) HiRSCHLER, 1912 Ueber die Plasmastrukturen (Mitochondrien, Golgi' scher Apparat u. A.) in den Geschlechtszellen der Ascariden (Spermato- u. Ovogenese). Arch. d. Zellforschung, 9, 351. (71) 1914 Ueber Plasmastrukturen (Golgi'sche Apparat, Mitochondrien, u. A.) in den Tunikaten, Spongien und Protozoenzellen. Anat. Anz. 47, 289. (72) Holmgren, 1899 Zur Kenntniss d. Spinalganglienzellen v. Lophius piscatorius. Anat. Hefte, 12, 71. (73) 1899 Zur Kenntniss der Spinalganglienzellen des Kaninchen's u. des Frosches. Anat. Anz., 16, 161. (74) 1899 Weitere Mittheilungen liber den Bau der Nervenzellen. Anat. Anz., 16, 388. (75) 1900 Noch weitere Mittheilungen liber den Bau der Nervernzellen verschiedener Tiere. Anat. Anz., 17, 113. (76) 1900 Studien in d. feineren Anatomie d. Nervenzellen. Anat. Hefte, 15, 7. (77) 1900 Von den Ovocyten der Katze. Anat. Anz., 18, 63. (78) 1902 Beitriige z. Morphologie d. Zelle. I. Nervenzellen. Anat. Hefte, 18, 269. (79) 1902 Einige Worte liber das 'Trophospongium' verschiedene Zellarten. Anat. Anz., 20, 433. (80) 1902 Weiteres liber das Trophospongium d. Nervenzellen u. der Drusenzellen des Salamanderpankreas. Arch. f. Mikr. Anat., 60, 669. (81) 1902 Ueber die Trophospongien d. Darmepithelzellen, nebst. eine Bemerkung in Betriff seiner v. Prof. Browicz neulich publizierten Abhandlung liber die Leberzelle. Anat. Anz., 21, 477. (82) 1903 Ueber die Saftkanalchen der Leberzellen u. der Epithelzellen der Nebenniere. Anat. Anz., 22, p. 9. 144 ALWIN M. PAPPENHEIMER (83) 1903 Ucber die 'Trophospongien' d. Nebenhodenzellen und die Lebergangzellen v. Helix poniatia. Anat. Anz., 20, 83. (84) 1902 Beitragezur Morphologic d. Zelle. Ergebn. d. Anat. u. Entw., 11, 274. (85) 1903 Weiteres iiber die 'Trophospongien' der Leberzellen u. der Darmepithelzellen. Anat. Anz., 22, 313. (86) 1903 Einige Worter zu der IMittheilung v. Kopsch: "Die Darstellung des Binnennetzes in spinalen Ganglienzellen u. andere Korperzellen mittelst Osmiumsaure." Anat. Anz., 22, 374. (87) 1903 Weitere Mittheilungen liber die 'Trophospongienkaniilchen' d. Nebennieren vom. Igel. Anat. Anz., 22, 476. (88) 1903 Ueber die sogenannten 'intrazellularen Fa den' der Nervenzellen von Lophius piscatorius. Anat. Anz., 23, 37. (89) 1903 Weiteres iiber die Trophospongien verschiedene Drusenzellen. Anat. Anz., 23, 289. (90) 1904 Ueber die Trophospongien der Nervenzellen. Anat. Anz., 24, 225. (91) 1904 Beitrage z. Morphologie der Zelle. II. Verschiedene Zellarten. Anat. Hefte, 25, 99. (92) 1904 Ueber die Trophospongien Centraler Nervenzellen. Arch. f. Anat. u. Phj^s., 15, Anat. Abt. (93) 1905 Zur Kenntniss der Zylindrischen Epithelzellen. Arch. f. Mikr. Anat., 65, 280. (94) 1907 Ueber die Trophospongien der Quergestreiften Muskelfasern nebst. Bemerkungen iiber den allgemeinen Bau dieser Fasern. Arch, f. Mikr. Anat., 71, 165. (95) 1910 Untersuchungen iiber die morphologische nachweisbaren stofflichen Umsetzungen der quergestreiften Muskelfasern. Arch. f. Mikr. Anat., 75, 240. (96) 1915 Die Trophospongien spinaler ganglienzellen. Arch. f. Zool. (Scandinavian), 9 (Hft. 2, Art. 15). (97) HoRTEGA, 1914 Alteraciones del sistema nervioso central en un caso de moquillo de forma paralitica. Trab. Lab. Invest. Biol., 12, 97. (98) Jaworowsky, 1902 'Apparato reticolare interne' von Golgi in Spinal ganglienzellen d. niederen Wirbeltiere. Bull. Acad. Sc. Cracovie, 403. (99) KoiRANSKY, 1904 Ueber eigentlimliche Gebilde in d. Leberzellen d. Amphibien. Anat. Anz., 25, 435. (100) KoLMER, 1916 Ueber einige mit der Ramon-y-Cajal'sche Uransilber methode darstellbaren Strukturen u. deren Bedeutung. Anat. Anaz., 48, 506, 529. (101) K0LOS.SOW, 1902 Zur Anatomie undPhj'siologied. Drtisenepithelzellen. Anat. Anz., 21, 226. (102) KoLSTER, 1913 Ueber die durch Golgi' sArsenik- u. Cajal's Urannitrat silber Methode darstellbaren Zellstrukturen. Vcrh. d. Anat. Ges. Greifswald. (103) Koi'SCH, 1902 Die Darstellung des Binnennetzes in spinalen Ganglien zellen u. anderen Korpenzellen mittelst Osmiumsaure. Sitzungsber. d. k. preuss. Akad. d. Wiss., 40, 1. THE t;OL(Jl .M'l'AKATUS 145 (104) KuLKsrn. I'.H 4 Der Netzapparat v. Tiolf^i in don Zcllon dos Kinrstockes. Arch. i. Mikr. Anat., 84, 142. (105) KiMKowsKA, 1911 Ucber d. Golgi-Kopschon Appamt. in dcii Nerven zcllon dcr Insekten. Festschr. f. J. Nussbaum. (106) La Valktte St. Georue, 18SG iSperinatologische lieitrage. Arch. f. niikr. Anat.. 27, 1. (107) Legendue, 1910 Rei'lierches sur le r6seaii interne do flolgi des cellules ncrvouses des ganglions spineaux. Anat. Anz., 36, 207. (108) 1905 De la nature pathologique des canalicules de Holmgren des cellules nerveuses. Comptes rendues Soc. Biol., 687. (109) Leschke, 1914 Untersuchungen liber den Mechanismus der Harnab sonderung in der Niere. D. Arch. f. Klin. Med., 81, 14. (110) Lewis, M. R., and Lewis, W. H., 1915 Mitochondria (and other cy toplasmic constituents) in tissue cultures. Am. Jour. Anat., 18, 339. (111) LuGARO, 1900 Sulla patologie delle celluli dei gangli sensitivi. Riv. di Pat. Nerv. c ment., 5, 145. (112) LuciONi, 1909 Contributo alio studio dei nevi molli. Arch de Sc. Med., 33, 21. (113) Luna, 1911 Sulla fina struttura della fibre muscolare cardiaca. Arch. f. Zellforschung, 6, 383. (114) 1914 Sulla fina struttura delle cellule endothelial! dell' endocardio e delle cellule que tappezano le fendituri di Henle. Arch f. Zellforschung, 12, 513 (115) Maccabruxi, 1909 Sulla fina struttura dei Megacariociti. Boll. Soc. Med. Chir. di Pavia, 57. (116) 1910 I Megacariociti. Intern. Monatschr., 27, 477. (117) Marcora, 1908 Di una fina alterazione delle cellule nervose de nucleo d'origine del grande ipoglosso, consecutiva alio strappamento ed al taglio del nervo. Boll. Soc. Med. Chir. di Pavia, 22, 134. (118) 1910 SuU' alterazione dell' apparato reticolare interno delle cellule nervose motrici consecutivi a lesioni dei nervi. Riv. di Pat. nerv. e ment., 15, 393. 1910 Arch. ital. de bioL, 53, 346. (119) 1909 Sui rapporti tra apparato reticolare interno e corpi di Xissl negli element! nervosi. Boll. Soc. Med. Chir. di Pavia. (120) MarenghI", 1903 Alcune particolarita di struttura e di innervazione delle cute dell' ammocoetes branchialis. Ztschr. f. wiss. Zool., 75, 421. (121) ]\Iartinotti, 1899 Sur quelques particularites de structure des cel lules nerveuses. Arch, ital de biol., 32, 293. (122) 1904 Contributo alio studio dell' apparato reticolare nei muscoli striati di alcuni mammiferi. Giorn. Accad. med. Torino, 67, 639. (123) Massenti, 1914 L' apparato reticolare interno del Golgi nel germe dentale. Monit. zool. Ital., 25, 107. (124) MiscH, 1903 Das Binnennetz der spinalen Ganglienzellen bei verschied enen Wirbeltieren. Intern. Monatsch. f. Anat. u. Phys., 20, 329. (125) Moriani, 1901 Di un apparato reticolare entre alcune cellule cancer ique. Atti della R. Accad. dei Fisiocr. di Siena fasc. 6. 146 ALWIN M. PAPPENHEIMER (126) 1004 Ueber ein Binnennetz der Krebszellea. Ziegler's Beitriige, 35, 629. (127) Monti, 1915 I condriosomi e gli apparati di Golgi nelle cellule nervosa. Arch. ital. di anat. e di embr., 14, 1. (128) MuLON, 1912 Apparato reticolare et mitochondries dans la surrenale du herisson. Comptes rend. Soc. Biol. Paris, 268. (129) Nansen, 1886 The structure and combinations of the histological elements of the central nervous system. Bergens Mus. Aarsberetning. (Quoted by v. Bergen). (130) Negri, 1900 Ueber die feinere Struktur der Zellen mancher Driisen bei den Saugetieren. Verh. Anat. Ges. Pavia, 178. (131) 1900 Di una fina particolarita di struttura delle cellule di alcune ghiandole del mammiferi Boll. Soc. Med. Chir. di Pavia, 1, 61. (132) Nelis, 1899 Un nouveau detail de structure du protoplasme des cel lules nerveuses (etat spiremateux du protoplasme). Bull. acad. R. Belgique, CI. Sc, Ser. 3, T. 37. (133) NussBAUM, 1913 Ueber die sogenannten inneren Golgi' schen Netz apparat u. seine Verhiiltnisse zu den Mitochondrien, Chromidien, u. andere Strukturen im Tierreiche. Arch. f. Zellforschung, 10, 359. (134) Oppenheim, 1912 Die Nervenzelle, ihr feinerer Bau u. seine Bedeutung. Anat. Anz. 41, 271. (135) Pensa, 1899 Sopra una fina particolarita di struttura di alcuni cellule delle capsule soprarenali. Boll. Soc. Med. Chir. di Pavia, No. 2. (136) 1901 Osservazione sulla struttura della cellule cartilagines. Boll. Soc. Med. Chir. di Pavia, No. 3/4, 119. 1901 Rend. d. R. Inst. Lomb. de Sc. e Lett., Ser. ii., .34, 443. (137) 1913 La struttura della cellula cartilaginea. Arch. f. Zellforschung., 11, 557. (138) Perroncito, 1908 Condriosomi, cromidi. ed apparato reticolare in terno delle cellule spermatiche (Nota preventiva), Rendiconti R. Inst. Lombardi., Ser. 2., 41. (139) 1909 Contributo alio studio della biologia cellulare. II fenomeno dello dictiocinesi. Atti Soc. ital. di patologia, 6. (140) 1910 Mitocondrii, cromidi ed apparato reticolare interno dello cellule spermatiche. Atti. Accad. dei Lincei., Ser. 5, 7. (141) 1911 Beitriige zur Biologie der Zelle (Mitochondrien, Chromidien, Golgi'sohen Binnennetz in den Samenzellen (Autoreferat). Arch. f. Mikr. Anat. 257,311. (142) 1913 Mitochondries et appareil reticulaire interne. (Apropos d'une publication de J. Duesberg.) Anat. Anz., 44, 69. (143) 1913 A proposito di un articolo di S. Comes sulla dittocinesi. Anat. Anz., 44, 78. (144) PiLAT, 1912 Der intracellulare Netzapparat in den Epithelzellen der Nebenniere vom Igel Erinaceus europeus. Arch. Mikr. Anat., 80, 157. (145) Platner, 1889 Beitriige zur Keiintniss der Zelle u. ihrer Teilungser scheinungen. Arch. f. mikr. Anat., 33, 125, 108. THE tiOLCI AIM'AIIATUS 147 (14G) I'ohNKZYNSKi, 1911 UntersuclumKeii iiber dc'in (jolKi-Kopscli'en Apparat u. cinigc andere Strukturen in den Ganglienzellcn d. Crustaceen. Hull. Acjid. Sc. Gracovie, 104. (147) I'opoKK, l'.K)6 Zur Kragc der Homologisierung des Hinncnnetzes d. Clanglienzellen init. den Chroniidion. Anat. Anz., 2!>, 249. (148) Reinkk, 190G I'eber die Heziehungon d. Wanderzollen zu d. Zell briu'ken, Zell-liicken ii. Trophospougiou. Anat. Anz., 28, 369. (149) Rktzius, 1901 Ueber Kaniilchonhildung in don Ricsenzellen des Knoch en-markes. Verh. Anat. Ges. Bonn., 92. (150) RiQuiEK, 1910 L'involuzione dell' apparato reticolare interuo nelle cellule dello corpo luteo. Boll. Soc. Med. Chir. di Pavia, 24, 185. (151) 1913 L'apparato reticolare interno. Rivista critico-sintetica. Riv di patologia nerv. e nient., 18, 314. (152) 1910 Der innere Netzapparat in den Zcllen des Corpus luteura. Arch, f. Mikr. Anat., 75, 772. (153) Ross 1915 The trophospongiuni of the nerve-cell of the crayfish (Cam barus), Jour. Conip. Neur., 25, 523. (154) Rossi 1912 L'apparato reticolare endocellulare di Golgi. Perugia, 8vo. (Abstracted in Anat. Anz., 1912, 44.) (155) Sanchez, 1907 L'appareil reticulaire de Ramon-y-Cajal-Fusari des muscles striees. Trab. Lab. Invest. Biol., Madrid, 5, 154. (156) San Giorgi, 1909 SuU' apparato reticolare interno di Golgi nel epitelio renale in condizioni patologico-sperimentali. Giorn. della R. Accad. di Med. di Torino, 15, 340. (157) Savagnone, 1910 Sur le reseau interne de Golgi dans les cellules des tumeurs. Arch. ital. de bioL, 53, 1. 1909 Also Lo Sperimentale, 63, 574. (158) 1910 Das Golgi'sche Binnennetz in Geschwulstzellen. Virch. Arch. 201, 275. (159) ScHMiNCKE, 1903 Zur Kenntniss d. Driisen der menschlichen Regio respiratoria. Arch. ]\Iikr. Anat., 61, 233. (160) ScHMAtJS TJ. BoHM, 1898 Ueber einige Befunde in d. Leber bei experi menteller Phosphorvergiftung u. Strukturbilder von Leberzellen. Virch. Arch., 152, 261. (161) SiNiGAGLiA, 1910 Osservazioni sulla struttura dei globuli rossi. Arch. pi Sci. Med., 34. (162) Sjovall, 1901 Ueber die Spinalganglienzellen des Igels. Anat. Hefte., 18, 239. (163) 1906 Ueber Spinalganglienzellen u. Markscheiden. Zugleich ein Versuch die Wirkungsweise der Osmiumsaure zu analysieren. Anat. Hefte, 30, 259. (164) 1906 Ein Versuch das Binnenetz von Golgi-Kopsch bei der Spermato- und Ovogenese zu homologisieren. Anat. Anz., 28, 561. (165) V. S.MiRXOW, 1901 Einige Beobachtungen iiber d. Bau der Spinal ganglienzellen bei einem 4-monatlichen menschlichen Embryo. Arch, f. Mikr. Anat., 59, 459. (166) 1906 Ueber die ^Nlitochondrien u. den Golgi' schen Bildungen analogen Strukturen in einigen Zellen v. Hyacinthus orientalis. Anat. Hefte, 32, 143. 148 ALWIN M. PAPPENHEIMER (167) SouKHANOFF, 1901 Rcseau endocellulaire de Golgi dans les elements nerveaux des ganglions spinaux. Rev. neurol., 1228. (168) 1903 Sur le reseau endocellulaire de Golgi dans les Elements nerveaux de I'ecorce cerebrale. Le Nevraxe, 4. (169) Stropeni, 1908 Sopra una fina particolarita di strut'tura delle cellule epatiche. Boll. Soc. Med. Chir. di Pavia, 22, 146. (170) Studnicka, 1899 Ueber das Vorkommen v. KanJilchen u. Alveolen im Korper der Ganglienzellen. Anat. Anz., 16, 397. (171) Taddei, 1910 Suir apparato reticolare interno di Golgi negli elementi epiteliale della prostata ipertrofica. Lo Sperimentale, 64, 434. (172) Tello, 1913 El reticulo de Golgi en las celulas de algunos tumores y en las del granuloma experimental prodicios por el Kiesel guhr. Trab. Lab. Invest. Biol., Madrid, 11, 146. (173) Terni, 1914 Condriosomi, idiozoma e formazioni peri-idiosomiche nella spermatogenessi degli amfibii (Ricerche sul Geotriton fuscus). Arch. f. Zellforschung, 12, 1. (174) ToTSUKA, 1902 Ueber die Centrophormien in dem Descemets'schen Epithel des Rindes. Intern. Monatschr. f. Anat. u. Phys., 19, H 1/2, 68. (175) Vastarini-Cresi, 1903 Trophospongium e canalini di Holmgren nelle cellule luteiniche die mammiferi. Anat. Anz., 24, 203. (176) Vecchi, 1909 Di una fina paricolarita di struttura della cellula cartil deciduale. Anat. Anz., 34, 224. (177) Veratti, 1909 Sulla fina struttura delle celluli di alcuni tumori. Boll. Soc. Med. Chir. di Pavia, 34. (178) 1902 Recherche suUa fina struttura della fibra muscolare striata. Rend. 1st. Lomb. di Scienze e Letter, 19, 6. (179) Verson, 1008 Contributo alio studio delle cellulo giganti tubercolare e di altri elementi cellulare normali e pathologici. Arch, per le Sc. Med., 32, 489. (180) Weigl, 1912 Vergleichende-zytologische Untersuchungen tiber den Golgi-Kopsch'en Apparat u. dessen Verhaltniss zu anderen Strukturen in d. Somatischenzellen, Geschlechtszellen verschiedenen Tieren. Bull. Acad. Sc. Cracovie, 417. (181) 1910 Studien fiber den Golgi-Kopsch'en Apparat u. die Trophospongien Holmgrens in d. Nervenzellen d. Wirbeltierc. Wiss. Arch. 1. (182) VViGERT UND Ekberg, 1903 Ueber binnenzellige Kanalchenbildungen ge wisser Epithelzellen d. Froschnieren. Anat. Anz., 22, 361. (183) 1903 Studien iiber das Epithel gewisser Telle der Nierenkanale von Rana esculenta. Arch. Mikr. Anat., 62. (184) Zawarzin, 1909 Beobachtung an dem Epithel der Descemet'schen Membran. Arch. f. Mikr. Anat., 74, 116. ON THE FORIM AND ARRANCEMENT IN FASCICULI OF STRIATED VOLUNTARY MUSCLE FIBERS A PRELIMINARY REPORT G. CARL HUBIOIl Department of Anatomy, Univemitij of Michigan FOUR FIGURES In the last few decades, relatively little special attention has been given to the form of striated voluntary muscle fibers and to their arrangement in the fasciculi. In the anatomic literature of this period, consideration is given mainly to the structure of the myofibrils, to the relation of the connective tissue of the muscle to the fibers, to the development of muscle fibers and the development of the musculature as a whole. One studjdng current texts is impressed with the unanimity of expressed views concerning the form of muscle fibers and the question is treated as ha\dng received satisfactory solution. Heidanhain^ in 'Plasma und Zelle' treats of the form and length of voluntary muscle fibers as follows: Da der Gegenstand allgemein bekannt ist, konnen wir uns kurz fassen. Es handelt sich um lange faserformige Gebilde, welche in kleinen Muskeln von einem Sehnenende bis zum anderen hindurchlaufen und an diesem immer abgestumpft enden, im Inneren sehr grosser Muskeln hingegen auch frei und zwar unter allmahlicher Verschmalerung mit spitzen Enden auslaufen. Sie werden bei geringer Breite (9-60 fj.) bis zu 12 cm. lang und daher findet man nur in Muskeln, welche, parallel der Fasermig gemessen, die Ausdehnung von 12 cm. iiberschreiten, die erwahnten freien Endigungen. This quotation expresses fairly well, I believe, the current views of the form and mode of ending of striated voluntary muscle fibers. 1 Heidanhain, M., Plasma und Zelle. Fischer, Jean, 1911, p. 529. 149 150 G. CARL HUBER Bardeen,- as a result of a study of teased preparations made from the external oblique of certain mammals, deviates from the current views and gives a much more correct statement as concerns the form and relations of striated voluntary muscle fibers; indeed his brief statement is one of the most accurate I have found and is here given in full. His words read as follows: The individual muscle-fibres either run from one tendon to another or they may end at one extremity or at both within the muscle fasciculi which extend from tendon to tendon. We may therefore distinguish two modes of ending of individual muscle-fibres: the 'intratendinous' , where the tip of the fibre terminates within a definite extension of a well marked tendon; and the Hntrafascicular ,' where the muscle-fibre terminates in the midst of a bundle of other muscle fibres which have a different region of termination. In the former case the musclefibre has a rounded or cone-shaped termination, often swollen in isolated specimens. In the intrafascicular mode of ending the musclefibre gradually becomes more and more narrow until it terminates in a thread-like extremity. In Bardeen's figures 2 and 4, types of these modes of ending and of the form of muscle-fibers are given, figure 4 including spindle-shaped fibers. Bardeen's statement of the form and mode of ending of muscle fibers, however, is antedated by the account given by v. Kolliker in his Handbuch der Gewebelehre des Menschen, which may be added to the references here given. KolHker's^ account reads as follows: Ueber die Gestalt der Muskelfasern haben besonders die Untersuchungen von Herzig und Biesiadecki, dann von mir, W. Krause, Weismann, Aeby und Kiihne Aufschluss gegeben. Nach diesen Erfahrungen kann es wohl als Kegel bezeichnet werden, dass die Muskelfasern im Innern grosserer Muskeln spindelformig sind, die an den Enden dagegen ein inneres spitzes und ein in die Sehne lihergehendes breites Ende besitzen, welches entweder abgerundet ist oder in einige stumpfe Spitzen ausliiuft oder auch wie treppenfcirmige Absatze darbietet. Ausser spindelformigen Fasern kommen ein Innern der Muskel noch manche andere Formen vor, am gewohnlichsten an dem einen oder an beiden Enden stumpfe Fasern. 2 Bardeen, C. R., Variations in the internal architecture of the m. Obliquus Abdominis Externus in certain mammals. Anat. Anz., vol. 23, 1903. 'v. Kolliker, A., Handbuch der Gewebelehre des Menschen. Erster Band. Engelmann, Leipzig, 1889, p. 371. FOKM AND AI{|{A\r,EMENT OF MUSCLK FIHKIIS I'll Kollikor quotes K. II. NW'bor (the ()ri{2;iiutl I liiive been unable to Inul) as rejj;aitling; spinille-sliaped fibers as tlie prevalent form of striated \'ohintary muscle fibers. It is not u\y purj)ose at the ])resent time to enter ui)on a more extended discussion of the literature dealing with the form and arrangement in fasciculi of striated voluntary muscle fibers. It is hoped that the quotations given may suffice to orient the results here to be presented. In the course of this brief report other pertinent literature will be considered as occasion demands. ]\Iy own studies on the form of striated muscle fibers have been made largely on teased preparations. The nmscular tissue was obtained largely from adult rabbits. The maceration preparatory to teasing was by the hydrochloric acid method developed in this laboratory.^ The method as used for the maceration of muscular tissue may be given here in detail in the hope that other workers may feel tempted to make use of it in a further analysis of this tissue, since a correct understanding of the form and arrangement of muscle fibers in fascicuU, as also their length, is of importance in valuating certain fundamental conceptions concerning the functions of muscles. For instance, according to E. Weber's law of the working capacity of a muscle w^e are taught that the lifting powder of a muscle is proportionate to the cross section of its fibers or fasciculi when arranged parallel, whUe the extent of elevation is proportionate to the length of its fibers. The method as used is as follows: After freely bleeding an adult rabbit a cannula was inserted into one of the iliacs central to the inguinal liganaent or into the subclavian before it passes under the clavicle and firmly secured by ligature. A 75 per cent solution of hydrochloric acid was then quickly injected at a pressure of 25 to 30 pounds. The apparatus used in obtaining and maintaining pressure was described by the author in the Am. Jour. Anat., vol. 6.* It is desirable to have the acid injected enter the tissues as quickly and as freely as possible. The pressure is maintained for several minutes. Some 15 to 20 minutes after the injection is com Huber, G. Carl, A method for isolating the renal tubules of mammals. Anat. Rec, vol. 5, 1911. ° Huber, G. Carl, The arteriolae rectae of the mammalian kidney. Am. Jour. Anat., vol. 6, 1907. 152 G. CARL RUBER pleted, the muscles are exposed and separated and removed and placed in a 75 per cent solution of hydrochloric acid. In removing a muscle care should be taken to remove the entire muscle, at least portions extending from tendon to tendon and great care should be taken not to crush the muscle during removal. The muscle pieces or entire muscles remain in the hydrochloric acid for about 3 hours, the period varying a little, depending on the thoroughness of the preliminary injection. After thorough maceration is obtained, the acid is carefully poured off and' distilled water slowly added. The water is reneAved at frequent intervals until it is practically free from acid. In the distilled water the muscle pieces remain about 2^ hours, though a stay of 4 to 6 days is not harmful. After thorough washing in distilled water the larger pieces are usually readily broken up into smaller bundles of fasciculi. Small bundles of fasciculi are nowtransferred to a hemalum solution diluted to one half with distilled water. The transfer from the distilled water to the hemalum solution should be executed with care if one wishes to obtain fasciculi'through their entire length. The transfer is best made with a glass rod, lifting the small bundle carefully as it leaves the water. In the hematoxylin solution the small bundles remain about 24 hours. This interrupts the maceration and stains the fibers. They may be kept indefinitely in the hematoxylin solution and are best stored in this solution for future use. In this solution the muscle bundles become quite hard and brittle and contract to about two-thirds or even to one-half of their former length. The preliminary teasing I have carried on in Esmarch dishes under the stereoscopic binocular and in a 0.5 per cent solution of ammonia water. In the ammonia water the stain attains a purple-blue color and the hard and brittle bundles become soft and pliable. A stay of one half hour to one hour in the ammonia water prepares the bundles for preliminary teasing. In the Esmarch dishes the bundles of fasciculi are with care separated into separate fasciculi. It should be stated, however, that a fasciculus is not a unit of muscle structure. For a distance all of the fasciculi of a bundle are readily separated. However, at one or several points small bundles of fibers or single fibers pass from one fasciculus to contiguous fasciculi. Great care is thus necessary and very careful teasing to completely separate what is known as a fasciculus. Bardeen^ has noted the fact that muscle fasciculi are joined by fibers. After the preliminary teasing resulting in the separation of a single fasciculus, the final teasing is undertaken on a large slide or lantern slide cover prepared as follows. The slide is thoroughly cleaned in acids and alcohol and wiped dry. Narrow strips of wax plates (the plates used in wax reconstructions) 2 mm. thick are cut and placed near the borders of the slide in the form of an oblong and pressed to the slide. The slide is then gently heated until the wax strips adhere. The slide on cooling is roady for use. The shallow well thus formed is filled with ammonia water and an isolated muscle fasciculus transferred to it. The final teasing may then be undertaken. It has repeatedly been possible to separate completely all or nearly all of the muscle fibers of a given fasciculus, even with fasciculi having a length of about 6 cm. The teased fibers may then be arranged in their approximate positions. Only when this has been accomplished can a muscle fasciculus be considered as having been teased. A worker should not attempt this unless he has at his disposal some 3 to 4 uninterrupted hours, and ought to bear in mind that the best results are obtained by 'making haste slowly.' The mounting of such preparations presents many difficulties and discouragements. My procedure FORM AND AllHANCJKMENT OK MUSCLE FIBERS 1 ').S is as follows: A fjisficulus is teasjd until uoarly all its fibers liave been separatctl. The aniinonia water is then very slowly and (iarefully withdrawn by means of a small dropper with point drawn to a capillary tube. Tins is und(!rtak(Mi under the i)inocular, observing the elTeets of (uirrents. The water is withdrawn until only a thin layer remains, only sufficient to enai)Ie moving the fibers on the slide. The final teasing and arranging of fibers may now be undertaken. As the ammonia water evaporates, the muscle fibers begin to adhere to the slide. The wax w'all may now be removed and a larg;e cover glass, on the under side of which a thin layer of glycerin has been spread, is gently lowered over the preparation from one edge. It is necessary to obtain the right degree of drying in order to gain successfully mounted preparations. If not sufficiently adherent to the slide the nuiscle fibeis will move, float and break. If allowed to dry too much the nuiscle fibers, although fixed in place, will appear fragmented. Such preparations are not valueless since the fragments of the fibers are not displaced laterally. Thus a single fiber may readily be traced throughout its whole length. By this method of preparation the muscle fibers show only faint cross striations, though they present a blue color. The nuclei are not evident. Neither has it been possible to locate the place of entrance of the nerve fibers. The sarcolemma seems very resistant to the acid, the neurolemma less so. In the ammonia water the muscle fasciculi appear to regain the length they had prior to staining in the hematoxjdin solution. The exact relation existing between the lengths as presented by teased fibers and living muscle fibers I am unable to determine definitely. The drawings here presented were made from preparations of muscle fascicuh teased completely, and arranged on the slide in their approximate positions, approximate with reference to the ends of the fasciculus teased. The drawings were made with the aid of the camera lucida at a magnification of 50 diameters, and are reduced 10 times in reproduction. The length of the respective fibers is accurately given. The thickness of the fibers is correctly given as pertains to the thicker portions of the fibers. At the attenuated ends the ink lines follow the outer border of the pencil outhnes. The results of these observations may be briefly recorded as follows : It is usually stated that in muscles having relatively short fascicuh the muscle fibers extend from tendon to tendon. This is of course not determined by the size and length of the muscle as a whole since in semipinnate, pinnate and compound pinnate muscles and in muscles where distal and proximal tendons overlap the lengths of the respective fascicuh of a given muscle are much shorter than the length of the muscle as a whole. 154 G. CARL HUBER In figure 1 are presented some muscle fibers . teased from fasciculi taken from the gastrocnemius of an adult rabbit. The fibers in group A, drawn from a completely teased fasciculus taken from the proximal portion of this muscle have an actual length of 1 cm. (these and all measurements given are obtained b}' di\dding the length of the respective fiber as measured in Fig. 1 Muscle fibers from the gastrocnemius of an adult rabbit. Group A, actual length 1 cm., from fasciculus taken from the more proximal portion of the muscle. Group B, actual length 1.5 cm., from fasciculus taken from the more distal portion of the muscle. X 5. the drawing by 50, the drawing having been made at a magnification of 50 diameters). The fibers in group B have an actual length of 1.5 cm. and are from a completely teased fasciculus taken from the more distal part of the same muscle. In all of the fasciculi from this muscle completely teased, the great majority of the muscle fibers extend from end to end or from tendinous insertion to tendinous insertion. In material prepared as above described, at the place of termination of a muscle fiber FORM AN'D ARRANGEMKNT OK MTSCLE FIBERS 155 in tendon, tlio (muI of llio muscle fiber stains more deeply in hematoxylin than does the same filler in close proximity. The tine end of the fiber difTers in appearance from the end of a broken fiber. The tendon ends of various fibers varj slij^htly in shape. They may appear as if cut at right angle to the fiber, as slightly beveled, as slightly rounded, tapering a little or having the form of a blunt cone and now and then as slightly expanded, though this may be due to a slight flattening of the end of the fiber. Now and then tendon ends of muscle fibers are met wdth that give the impression as though the sarcolemma did not enclose the end but terminated ring-shaped at the extreme tendon end of the respective fiber, but the limitations of the method used are such that this question could not be conclusively decided. In this connection it is of interest to note the observations of 0. Schultze,*^ who believes that muscle fibrils and tendon fibrils are parts of a single stiiicture but this observer adds that the behavior of the sarcolemma at the ends of the fibers deserves further study. Also the studies of Bald"win," who regards the sarcolemma as covering the tendon ends of muscle fibers and denies the continuity of muscle fibrillae and tendon fibrillae, and discusses two types of terminations of muscle fibers in tendon; one type in which the long axis of muscle and tendon fibers coincide, the other type in which they meet at an angle. In the former the tendon fibrils are attached to cone shaped processes of the sarcolemma dovetailed into the tendon ends; in the latter type the sarcolemma end is considerably thickened and presents a number of projections into the muscle substance. Digitations or branchings of muscle ends or step formations have not been observed by me in my teased preparations. It should be understood, however, that in successfully macerated preparations the collagenous connective tissue is so completely removed that it is not e\ident on teasing. Out of quite a number of fasciculi with fibers of type B, of figure 1, ^ Schultze, O., tjber den direkten Zusammenhang von Muskelfibrillen und Sehnenfibrillen. Arch. f. Mik. Anat., vol. 79, 1912. ' Baldwin, W. M., The relation of muscle fibrillae to tendon fibrillae in voluntary striped muscle of vertebrates. Morph. Jahrb., vol. 45, 1913. THE A?JATOMICAL RECORD, VOL. 11, NO. 4 156 G. CARL RUBER successfully and completely teased, in only two and in each onlj" one fiber was found which did not extend from tendon end to tendon end. In both of these fibers one end reached the tendon, terminating as adjacent fibers, while the other end reached to about the middle of the fasciculus ending in a fine tapering filament. The fibers in a number of fasciculi having muscle fibers of the length of type B were counted and averaged about 20 fibers to a fasciculus. What length a muscle fasciculus of an adult rabbit may attain and still have the great majority of its fibers reach from end to end is a question I am at present unable to answer definitely. Of the muscles teased, none in which the contained fasciculi reached a length of a little over 2.5 cm. did I find such in which the majority of the muscle fibers reached from end to end. However, samples have not been taken from nearly all of the muscles and it may be that in certain of them fasciculi having a length of over 2.5 cm. in which the majority of the fibers extend from end to end, may be found. In figure 2 are presented type fibers obtained from a completely teased and successfully mounted fasciculus, taken from one of the adductor muscles of the thigh of an adult rabbit. In this fasciculus a single muscle fiber (A) extends from end to end or from tendon insertion to tendon insertion; both extremities showing the characteristic staining and appearance of the tendinous end of a muscle fiber. This fiber has an actual length of 3.64 cm., a length which is regarded as the length of the fasciculus. After final teasing and after withdrawing the ammonia water Fig. 2 Muscle fibers from the thigh adductor of an adult rabbit. Teased fasciculus had a length of about 3.5 cm. The completely teased fibers are in the drawing placed with reference to the ends of the fasciculus. Fiber A, has actual length of 3.G4 cm.; a, 2.1 cm.; b, 1.8 cm.; c, 1.5 cm.; d, 1.3 cm.; e, 2.1 cm.; f, 1.9 cm.; g, 1.7 cm. X 5. Fig. 3 Types of muscle fibers teased from a single muscle fasciculus, having a length of a little over 4 cm., taken from one of the larger thigh muscles of a rabbit. This fasciculus contained 37 fibers. The fibers are arranged with reference to an imaginary line, bottom of figure. The tendon ends of fibers ending intrafascicular are brought to this line. Certain of the fibers sketched have an actual length as follows: a, 2.9 cm.; b, 2.4 cm.; c, 1.92 cm.; d, 0.9 cm.; e, 1.4 cm.; f, 0.14 cm.; g, 2.9 cm., ; h, 3.04 cm.; i and j, 2.9 cm. X 5. 157 158 • G. CARL HUBER from the well on the slide, as explained in the detailing of the method used, I was able to arrange the teased fibers so as to have the tendon ends of the teased fibers reach imaginary lines draT\^l at right angles to the ends of the single muscle fiber which extends from end to end in this fasciculus. The spindleshaped fibers hold approximately the same relative position with, reference to the ends of the fasciculus as before teasing as was determined at the time of teasing. This fasciculus, completely teased, contains 26 muscle fibers, of which as stated one passes from end to end, 10 others reach one tendon end, 12 the other tendon end and 3 are spindle shaped fibers reaching neither tendon end. Of the 26 fibers, 15 type fibers are given in figure 2. This bundle of fibers completely teased is here spoken of as a fasciculus. I have above stated that fasciculi are not units of structure, but that from each small bundles of fibers or single fibers pass from one fasciculus to contiguous fasciculi. A single 'fasciculus' completely separated constitutes thus an artificially separated bundle of muscle fibers. Thoma® has also appreciated the fact that a muscle fasciculus is not a unit of structure. In serial cross sections of the gastrocnemius of the frog, in which, with the aid of the camera lucida, the outlines of the muscle fibers were sketched serially he noted single muscle fibers passing from one muscle fasciculus to another, concluding as follows: "Die einzelnen Muskelfaserbiindel hangen somit vielfach durch Muskelfasern zusammen, welche bald mit dem einen, bald mit dem anderen Biindel sehr innig verbunden sind, und die ganze Muskelmasse bildet ein stark in die Liinge gezogenes Netzwerk von Muskelfasern." The fasciculus above referred to as containing 26 fibers is to be considered in this Ught. In the figure as drawn, at each end 5 fibers begin with blunt ends showing by form, structure and staining that they are muscle fibers ending in tendon. Each of the fibers extends into the fasciculus for a distance which varies for the several fibers, becoming attenutated and finally terminates in a thread like filament having a thickness of 3 ^ to 4 /^. It requires very ' Thoma, R., tJber die netzformige Anordnung der quergestreiften Muskelfasern. Virchow's Archiv, vol. 191, 1908. FORM AND ARRANGEMENT OF MUSCLE FIBERS 159 thoroiifj;h inacenition to eiuible one to separate completely these fiiio, iiitrafusc'icular teriniiuitions of the muscle fibers. The length of the muscle fibers having one tendon end at the end of the fasciculus, the other ending in an intrafascicular filamentous termination varies as follows; fiber a, 2.1 cm.; fiber b, 1.8 cm.; fiber c, 1.5 cm.; fiber d, 1.3 cm. The other fibers of this type sketched are intermediate in length between fibers a and d. The three spindle shaped intrafascicular fibers with both extremities attenuated and neither end reaching the tendon ends of the fasciculus measure as follows, fiber e, 2.1 cm.; fiber f, 1.9 cm.; and fiber g, 1.7 cm. The extent of overlapping of fibers beginning at the tendon end of the fasciculus and ending intrafascicular in fine, attenuated ends may be noted in this figure (2) . Their exact relation cannot be readily seen in a completely teased preparation, with fibers separated and arranged on the slide. To gain their relationship actual teasing is necessary. "WTiile teasing the details of the arrangement of the muscle fibers becomes evident and it is observed that the fine filamentous intrafascicular ends are applied usually to the thicker portions of other fibers, usually not near a filamentous end of another fiber. The same is true of the ends of the spindleshaped fibers reaching neither fascicular end. This figure (2) I regard as representative of the form and arrangement of the striated voluntary muscle fibers in the fasciculi of rabbit muscles having a fascicular length of from about 3 cm. to about 5 cm. Probably the same is true of voluntary muscle of other vertebrates, though my observations have not been extensive outside of rabbits and birds (rooster). In muscle with longer fasciculi the length of the muscle fibers having blunt tendon ends and filamentous intrafascicular terminations varies more than indicated by the measurements above given, and the spindle shaped fibers with intrafascicular position may He nearer one end or the other of the respective fasciculus or occupy a more middle position. This variation in the length of the muscle fibers I have indicated in figure 3, giving t\'pe fibers from a fasciculus having a length of somewhat over 4 cm., and taken from one of the thigh muscles. Un 160 G. CARL HUBER fortunately the specific muscle could not be determined after the maceration. This fasciculus was also completely teased and successfully mounted. It contains 37 fibers of which 8 are spindle shaped and have an intrafascicular position. In it one fiber extends from end to end, through the length of the fasciculus. The fibers could readily have been sketched in approximate relative position with reference to the ends of the fasciculus, but the resulting figure, at the magnification used, would have been too long to admit of publication in the pages of this Journal. The arrangement of the fibers, however, is not unlike that presented in figure 2. For figure 3, type fibers were selected. The single fiber extending from end to end could not be included by reason of its length. The fibers having a tendon end are arranged with reference to an imaginary line, at the bottom of the figure; the tendon end being brought to this line. Of certain of the fibers with tendon ends and intrafascicular filamentous terminations the actual lengths are as follows, fiber a, 2.9 cm.; b, 2.4 cm.; c, 1.92 cm.; d, 0.9 cm.; e, 1.4 cm.; f, 0.14 cm. The spindle shaped fibers sketched with both ends terminating intrafascicular with filamentous endings present the following measurements, fiber g. 2.9 cm.; h, 3.04 cm.; i and j, 2.9 cm. The single fiber extending through the entire fasciculus presents a length of almost 4.5 cm. For rabbit muscle fasciculi having a length of more than 4.5 cm. to about 5 cm., so far as my observations go, there are no muscle fibers that extend the whole length of a respective fasciculus. In some of the longer fasciculi taken from the latissimus dorsi, the pectoralis major and the extensor cruris almost complete teasing was obtained. Many muscle fibers were completely isolated, though never all of the fibers of a given fasciculus. In some of the most successfully macerated fascicrli, their distal ends were slightly crushed during removal, so that not all of the fibers could be traced to their tendinous ends. For final teasing of these longer fasciculi, lantern slide covers answer the purpose of slides very well. In the longer fascicuH, having a length of 6 cm. to about 6.5 cm., in which many fibers were completely isolated, no fibers were found FORM AND ARRANGEMENT OF Mrs( I.K FIBERS KJl roacliiuj!; from end to end. FiixM-s witli blunt tendon ends and tilanientoiis intnifuscicuUir terniinutions, these, severally of varying lengths, and spindle shaped fibers with intrafascicular position, with ends terminating in hair hke processes, constistiited the types of fibers isolated. In these longer fasciculi one end of certain of the spindle shaped fibers reaches nearly to one or the other tendinous end of the respective fasciculus while others of the spindle shaped fibers have a more nearly central position, with reference to the length of the fasciculus. In the longer and longest fasciculi teased, no muscle fibers having a length of more than about 3.5 cm. were observed. Felix ^ is quoted as having isolated striated muscle fibers approaching a length of about 12 cm. In his account stress is laid on the fact that in the macerating fluids used, acids mainly, the muscular tissue contracts by one-third to two-thirds of the original length. In his own material he sought to obviate this contraction by maintaining the original length through tension. I have noted the fact that in the method used, hydrochloric acid is injected into the living muscle while under extension, that during immersion in the hydi'ochloric acid and in the hematoxylin stain, a contraction of the muscle fasciculi to about twotliii'ds to one-half of their original length is obtained, but also that in the ammonia water fasciculi of muscle taken from the hematoxyhn solution extend in length so as to approach very nearly their length in fresh muscl^. Exact measurements I am unable to give since, obviously, it would be necessary to isolate at least small bundles of fasciculi from fresh muscle, and trace them through the various steps, making measurements at various stages. Of the longest fibers isolated by Felix, one from the gracHis of man measured 11.5 cm. and one from the sartorius of man 12.3 cm. ; the latter fiber ha\dng a broken end. Division of fibers was not seldom found. A figure of a single fiber with branchings is reproduced natural size. This fiber in the figure measm^es approximately 12 cm. Concerning this fiber the text speaks as follows: "Die Faser theilt sich, lasst ' Felix, W., Die lange der Aluskelfaser bei dem Menschen und einigen Siiugethieren. Festschrift, Albert von Kolliker, Englemann, Leipzig, 1887. 162 G. CARL HUBER Spaltrilume erkennen, steht mit anderen Fasern in Verbindung, kurz um, das Bild wird durch veilfach abgehende Fasern ein so complicirtes, dass man ein Gewirr von mehreren Fasern vor sich ?u haben glaubt, bis eine genaue mikroskopische Untersuchung ihre Ziisammengehorigkeit nachweist." Felix teased unstained tissue. I have not teased human muscle. However, the figure presented by Felix is not unfamiliar to me. In incompletely macerated tissue such 'fibers' are now and then obtained. However, they are interpreted by me as representing an incompletely teased fiber complex. The fine hair ]ike intrafascicular ends of muscle fibers are so closely applied to the sides of other fibers that the cross diameter of the thicker fiber is scarcely increased. Such a misinterpretation, I can conceive, may readily be made in incompletely macerated and teased muscle tissue. Felix gives data concerning the length of muscle fibers in the rabbit, a tissue with which I am familiar. This observer isolated fibers from the pectorahs, sartorius, latissimus dorsi and extensor cruris of the rabbit. His own words concerning them read as follows: "Hier waren fast siimmtliche Fasern mindestens 5 cm. lang, doch waren unter 6 cm. nur wenige zu erzielen. Die meisten Fasern schwankten zwischen 6.0 und 7.5 cm. Die Fasern zeichneten sich sammtlich durch ihre Starke aus. Die langste Faser isoherte ich aus dem extensor cruris, der am Thiere selbst nur 8 cm. mass, von 8 cm. Lange. Die Dicke war ungemein schwankend, dickere und diinnere Stellen wechselten ab, die diinnste Stelle mass nur 0.0109 mm., wilhrend dickere Stellen 0.111 mm. gemessen wurden. Offenbar sind hier verschiedene Wirkungen der Salpetersaure zur Geltung gekommen. Theilung konnte ich haufig beobachten." An analysis of this statement from Felix in the light of my own investigations leads me to conclude that this observer did not obtain completely teased muscle fibers. Many hundreds of muscle fibers of the rabbit have been completely isolated and in no instance have I observed branching of fibers. Often have I seen apparent branching, but on careful teasing such structures ha^'e been separated into several fibers. The variation in thickness of the long fibers referred to in the above quotation, I be FOK.M AND AilUANGEMENT OF MUSCLE FIBERS 1G3 lieve, is ('xplaiiUHl by a liiikinp; in fhain of sovonil fibers. Even granting thai the faseicuh teased by nie after a stay in ammonia water, some attaining a l(Migtli of about (3.5 cm., had not attained their full, original length, the difference in the length of muscle fibers of the rabbit teased by Felix and by myself is not accounted for. Felix found few attenuated ends of fibers with intrafascicular terminations, while, as my own figures show, these are mnnerous. In the light of these studies I am inclined to regard the measm'ements of the length of striated voluntary muscle fibers as given by Felix as inaccurate and as made on incom})letely teased muscle tissue, and to regard the figures given by earlier observers as more accurate. These, to quote freely from Felix, are for the medium length of muscle fibers of man 2 cm. to 3.5 cm., Krause giving as the longest of the fibers of the sartorius 4 cm. Striated voluntary muscle fibers of other manunals and other vertebrates have thus far been only incidentally teased by me. Bardeen's- figure 4, b, gives a flat band of fibers dissected from the external oblique of a dog, having a length of approximately 15 cm. (figure one-half natural size) with figures of completely isolated fibers; spindle shaped fibers having a length of approximately 8 cm. and fibers with blunt tendon ends and attentuated intrafascicular terminations, varying in length from approximately 4 cm. to 6 cm. The general shape of these fibers appears to me as correctly drawn. Since I have not teased muscle fascicuh of the dog I am unable to verify the accuracy of the measm-ements given. For the dogs muscle fibers Felix gives 3 cm. to 4.5 cm. as common measurements and 5.5 cm. to 6.5 cm. as long fibers. Opportunity presented itself to tease muscle fibers of an adult rooster (Gallus domestica), injected with hydrochloric acid for other purposes. In one specimen, the thigh muscles were well macerated. In figure 4 are sho'WTi four completely teased spindle shaped fibers taken from these muscles. These fibers, some of which are among the longest completely teased, present the following measurements: fiber a, 3.2 cm.; b, 3 cm.; c, 3.2 cm. and d, 2 cm. Several spindle shaped fibers wdth intr^ 164 G. CARL HUBER Fig. 4 Spindle shaped muscle fibers teased from the thigh muscles of an adult rooster (Gallus domestica). Actual length of fibers, a, 3.2 cm.; b, 3 cm.; c, 3.2 cm.; d, 2 cm. X 5. FORM AND AKllANCJKMENT OF MTISHLE FIBERS 1G5 fascicular position, with uiuloiibted brauchiiiK were observed. The thvision extended to about the mitldk; of the respective fibers, the two ]iarts terniinatcd in attenuated, hair hlce fibers. Muscle fibers with blunt tendon ends and filamentous intrafascicular toi'niinations were also observed. It is the purpose, as oportunity presents, to include in this study fibers from different types of muscles from the different classes of vertebrates and to extend the investigation so as to include several different mammals with types of muscle from each. Schiefferdecker^" and certain of his i)upils have spent infinite pains in determining, among other things, the relative thickness of nmscle fibers. The thickness and form of muscle fibers these workers have determined largely in cross sections of various muscles. Each muscle is said to be composed of muscle fibers having specific size and form (cross section) with specific arrangement of connective tissue and elastic fibers. It is recognized that in each muscle, muscle fibers of varying sizes are found. In many muscles this difference in size of fibers is said to be considerable, in others less so. This difference in size of fibers may be ascribed, according to Schiefferdecker, to tw^o possibiUties: 1, the muscle may be composed of fibers which in reahty differ in size; 2, the smaller and smallest cross cut fibers of a given cross section may represent cross sections of the ends of fibers terminating in the muscle. In considering the structure of muscle, he adds, the second possibility plays only an unimportant role, and only as concerns the smallest fibers. The fibers sketched in figures 2 and 3 may serve to show that such contention is difficult to support in the light of this work. Except for muscles in which the fibers of the respective fasciculi extend from end to end, or in which the majority of the fibers do this, the variation in the size of the fibers in a given cross section is largely dependent on the fact that many of the fibers of a given fasciculus do terminate intrafascicularlj^ In order to make the numerous measurements of Schiefferdecker and his pupils of real value, or of similar investigations, it would be necessary to ascertain by means of teasing and complete isola 1" Schiefferdecker, P., Muskeln und Muskelkerne. Barth, Leipzig, 1909. 166 G. CARL HUBER tion of fibers, the arrangement of the fibers in the fascicuh of muscles, the fibers of which are measured in cross sections. MacCallum's^^ investigations led him to conclude, as a result of counting the fibers of the sartorius muscle in man at various ages that the muscle fibers cease to multiply in the fetuses from 13 cm. to 17 cm. in length, and that after that period muscles increase in size by increase in size of individual fibers. This statement, it would seem to me, needs verification and could only be verified by study of muscles in which all of the fibers of the fasciculi extended from end to end or by very careful and painstaking teasing, of fasciculi^ covering the several periods in which the muscle fibers are counted. Myofibrils are usually regarded as extending from end to end in a given muscle fiber. In muscle fibers having filamentous intrafascicular terminations, and this includes the majority in the longer fasciculi, this is obviously not the case. Concerning the relations of the ends of myofibrils not reaching the ends of the respective muscle fibers, my teased preparations give no evidence. The festooning of the sarcolemma, described by certain authors, may perhaps be brought in relation with the ends of myofibrils which do not extend the entire length of the muscle fiber. In this communication the expression "completely teased and isolated muscle fibers" has been repeatedly used. Therefore it will no doubt seem paradoxical, for me to express in this concluding paragraph, even tentatively, the view that striated voluntary muscle is syncytial in character. From the arrangement of muscle fibers in the fasciculi of striated voluntary muscle; from the fact that muscle fasciculi are not units of structure; from the further fact that in teasing muscle fibers there are always found points of contact where the fibers are ultimately separated with great difficulty, I am led to tentatively express the view that striated voluntary muscle tissue presents syncytial character even in its fully developed state, as does involuntary muscle and cardiac muscle, though " MacCallum, J. B., On the histogenesis of striated muscle fibers and the growth of the human sartorius muscle. Johns Hopkins Bull., 1898. FOlt.M AM) ARRANGEMENT OF MUSCLE FIBERS 1()7 not to tlic samo dcf^roo as tlie last nainod. This (lucstion cannot hv finally docidoil hy teasing'. It is not my j)urpose at the present time to enter upon the mooted question of the histogenesis of vohmtary nnisele tissue, nor to consider the extensive literatiu'e invohed. The problem of the syncytial character of \'oluntaiy nniscle is one of histogenesis. Embryological evidence at hand indicates that the histogenesis of voluntary muscle lends support to the view that striated voluntary muscle is syncytial in origin. Material is being collected to determine this question if possible. One of SchiefTerdecker's^" general conclusions reads as follows: "Muskelnetze fanden sich in den untersuchten IMuskeln so vielfach, dass man sie wohl als eine allgemein verbreitete Erscheinung ansehen kann." Thoma^ finds frequent anastomoses between fibers. Reference, however, is not had to anastomoses between fibers such as described by Thoma. This observer finds intimate contact between adjacent fibers, so that for a distance only a single layer of sarcolemma appears to separate them. Myofibrils are not thought to pass from one fiber to another. It has seemed to me that this may be verified in teased preparations. Now and then two fibers adhere together, for a short distance, so closely, that separation, even in well macerated tissue, is impossible; this very generally in thicker portions of fibers. Involuntary muscle, if successfully macerated in potassimn hydrate or by the hydrochloric acid method here detailed is readily teased so as to present spindle shaped cells, although as shown by McGill^^ this muscle develops from mesenchjuie, retaining its syncytial character. The mere arrangement of striated, voluntary muscle fibers in a fasciculus possessing fibers with, attenuated intrafascicular terminations, is such as to suggest the sjTicytial character of this tissue. In partially teased, though w^ell macerated tissue, a mesh work of fibers, with long meshes is now and then evident. It is usually possible to tease the fibers having intrafascicular termination, quite readily, so far as concerns the thicker portions of these fibers and to isolate them to near their thread like terminations 1- McGill, Caroline, The histogenesis of smooth muscle in the alimentary and respiraton- tract of the pig. Monatschrift Anat. u. Phys., vol. 24, 1907. 168 G. CARL RUBER on other fibers. Near their intrafascicular ends they adhere very tenaciously to adjacent fibers. In ammonia water the macerated and stained fibers become quite phable and present an elasticity and a tensile strength which is often surprising. Yet, often the finer ends are broken before they can be detached from adjacent fibers. It is evident that the relations of the intrafascicular ends of muscle fibers to adjacent fibers is different at their attenuated terminations than in course. Their exact relation I am unable to determine in teased preparations, though even the finest ends often present the appearance of a torn sarcolemma which does not extend to the extreme tip. I am unable to state whether the myofibrils extend from the attenuated ends to fibers on which they appear to end. In a number of preparations of rabbit embryos of the tenth day, cut serially in the sagittal plane, sections having a thickness of 2 n and 3 fi, stained in iron-lac-hematoxylin, the syncytial character of the cells from which the voluntary muscle tissue is developing is evident. Conclusive preparations, from embryos varying in ages, have thus far not been obtained. This question shall form the subject of a further study now under way. It may be recalled here that Godlewski^^ considers striated muscle as presenting a syncytial character, basing his deductions on a study of the histogenesis of skeletal and heart muscle. It is impossible at the present time to do more than suggest that striated voluntary muscle, like involuntary and cardiac muscle, presents a syncytial character, evidence of which is seen in its full development. Godlewski, E., Die Entwicklung des Skelet- und Herzmuskelgewebes der Saugethiere. Arch. f. Mik. Anat., vol. 60, 1902. A NOTE ON THE STRUCTURE OF THE ELASTICA INTERNA OF ARTERIES G. CARL HUBER Deparlmenl of Anatomy, University of Michigan ONE FIGURE A comparison of a number of texts, descriptive of the structure of the elastic intima — the fenestrated membrane of Henle — of arteries, reveals the fact that the views concerning the structure of this layer are by no means unanimous. Schafer^ speaks of the elastic intima as follows: "The elastic tissue is represented by one distinct lamina, which is separated from the endothelium by the subendothelial layer. It is, on its outside, in direct contact Avith the non-striped muscle of the middle layer." The 'internal elastic lamina' is spoken of as membranous in character, the membrane is not, however, a continuous one, but is perforated by apertures. In figure 517, of Schafer's text, is show^i a portion of the fenestrated membrane from the femoral arterj^ as figured by Henle. MalP in his study of connective tissue fibrils states that "elastic fibers are composed of two distinct substances — the interior, which stains intensely with magenta, and the membrane, which does not." A study of the membrane of Henle, isolated by boiling in acetic acid or KOH and then stained with magenta or picrocarmine leads him to conclude that "The Henle's fenestrated membrane is therefore composed of three layers — an upper and a lower transparent membrane in which there are no openings, and w^hich is identical with the membrane of elastic fibers; and a middle layer which stains with magenta, and is identical with the interior of elastic ' Schiifer, E. A. Text book of microscopic anatomy. Longmans, Green, and Co., 1912, p. 332. ^ Mall, F. P. Reticulated tissue, and its relation to connective fibrils. Johns Hopkins Hospital Reports, vol. 1. 169 170 G. CARL HUBER fibers. This central portion contains the openings." V. Ebner's' description of the elastic intima concludes as follows, "Uebrigens erscheint dieselbe fast immer als eine sogenannte gefensterte Haut mit verschieden deutlich ausgepragten, netzformigen Fasern und meist kleinen langlichen Oeffnungen, seltener als ein wirkliches, aber sehr dichtes Netz vorztiglich langsverlaufender elastischer Fasern mit engen, langlichen Spalten, und stimmt in ihrem chemisches Verhalten vollkommen mit den elastischen Hauten der Media grosser Arterien iiberein." TriepeP in his account of the elastic tissue in the walls of intracranial arteries considers the elastic intima as a fenestrated membrane, stating that in the smaller arteries the fenestra are so near together that only a felt-work of elastic fibers remains, so that in cross sections a row of adjacent points is observed instead of a membrane. His figure 4 shows this clearly. Schoppler,^ who studied the finer structure of the brain arteries of several mammals, gives especial consideration to the internal elastic membrane, and gives emphasis to closely arranged longitudinal ledges, which have a course parallel to the long axis of the vessels. He recognizes a fibrillar structure in the elastic intima as expressed in these words, "Vielfach zeigt sich auch, dass die Membrana flava interna keine homogene Platte ist, sondern wie die Betrachtung von Schragschnitten bei lOOOfacher Vergrosserung lehrt, aus, sehr feinen elastischen Faserchen besteht. Die erwahnten Leistchen werden durch Ausbildung starkerer nach dem Lumen vorspringender Fasern bedingt." His figure 6, which shows an oblique longitudinal section of a basilar artery presents the fibrillar character of the elastic intima clearly. Diirck" records observations made on connective tissues studied by means of Weigert's iron-hematoxylin myelin sheath staining method. In tissues fixed in formalin and Mtiller's fluid or in formalin, ^ V. Ebner, Victor, Kolliker's Handbuch der Gewebelehre des Menschen, Dritter Band, Zweite Hiilfte. Engelmann, Leipzig, 1902, p. 643. Triepel, H. Das elastische Gewebe in der Wand der Arterien der Schadelhohle. Anat. Hefte, vol. 7, 1897. ^ Schoppler, H. Ueber die feinere Strukture der Hirnarterien einiger Saugetiere. Anat. Hefte, vol. 15, 1900. Diirck, H. Ueber eine neue Art von Fasern im Bindegewebe und in der Blutgefiisswand. Virchow's Archiv, vol. 189, 1907. sria ("IHUK OF KLASTICA IN TKItN A-AK TKUIKS 1/1 mordant (h1 in a coppci' salt aiul stained in ii'on-honiatoxylin, following- the W'ci^cii nictiiod foi' staining myelin sluuiths, ('(M'tain coiuuM'tivc tissiio fibrils were stained hlue-hhick. Certain of these differentially stained fibi'ils were re^aided l)y Diirck as a special tyjje of connective tissue fibrils, others, as yellow elastic fibers. This method as used by this observer, gave, in successful preparations, unusually distinct staining of the elastic intima of vessels. His words read as follows, "Untersucht man zunachst kkune Arterien auf dem Liingsschnitt oder auf Schrilgschnitten, welche das Rohr in langer Ausdehnung tieffen, so erkeimt man an den durch die Intima fallenden Schnitten, dass die Elastica interna hier nicht durch zirkuljire Fasern, Faserbiindel oder Lamellen dargestellt wird, wie man dies gewohnlich abgebildet und beschrieben findet, sondern unmittelbar fiber dem Endothelrohr liegt wie eine Basthiille unter einer Baunninde eine einfache Schicht von straffen Liingsfasern, welche unter sich allerdings durch kurze quere Zwiechenstiicke verbunden sind und so ein Netz mit sehr langgestreckten und liings verlaufenden Maschen darstellen." In cross-sections such fibers appear as points. The method used in staining the sections on which this study was based and from one of which the figure accompanying this note was dra^\^l, was presented by Dr. De Witt^ at the Wisconsin meeting of the American Association of Anatomists in 1907. This differential elastic tissue staining method consists of a modification of Weigert's* iron-hematoxylin van Gieson method. According to Weigert's method two stock solutions are prepared. Solution I Hematoxj-lin crystals 1 gram Alcohol, 96 per cent 100 cc. Solution II cc. Liquor ferri sesquichlorati (U. S. P.) 40 Hydrochloric acid (sp. gr. 1.20) 7 Aqua dist 959 ' DeWitt, Lydia M. Abstracts of papers presented at the 22nd Session Amer. Ass. Anat. Anat. Record, vol. 1, p. 74. Weigert, K. Eine kleine Verbesserung der Hamatoxylin-van Gieson Methode. Zeitsch. f. wissensch. ]\Iik.. vol. 21. 1904. THE ANATOMICAL RECORD, VOL. 11, SO. 4 172 G. CARL HUBER Solutions I and II are mixed in equal proportions just before using. Differentiation is obtained by means of van Gieson's picric acid fuchsin solution, prepared after Weigert as follows: cc. Picric acid, saturated aqueous solution 100 Acid Fuchsin (Weigert), 1 per cent aq. sol 10 In the method as used for elastic tissue staining, stock solutions I and II are mixed in proportion of 3 to 4 parts of solution I to one part of solution II. The sections are stained several hours and after rinsing in distilled water differentiated in van Gieson's picric acid, acid fuschsin solution, prepared as above indicated. The differentiation is controlled from time to time under the microscope. The method is simple and can be used on celloidin sections or paraffin sections fixed to the slide. The yellow elastic fibers are stained blue-black, the collagenous tissue is stained brick-red to pink, depending on the degree of differentiation and the thickness of the sections. The method is not unlike that used by Dlirck, although the differentiation by means of the van Gieson picric acid, acid fuchsin solution has the advantage of counterstaining the collagenous tissue. This differential elastic tissue staining method has been extensively used in the preparation of sections for classes and is recommended as simpler than other differential elastic tissue stains. Numerous sections of arteries varying in size from arterioles with two or three layers of muscle cells to arteries of about 2.5 mm. in diameter, cut in pieces of tissue fixed in formalin, formalin and Midler's fluid, and picro-nitric solution, embedded in paraffin, and sections fixed to slides, were stained after the above mentioned iron-hematoxylin and picric acid, acid fuchsin method. The differentiation in picric acid, acid fuchsin was carried in most sections to an extreme degree, so that only the yellow elastic tissue retained any of the blue-black coloring. Usually four to six sections were fixetl to one slide, the sections ap]3roximating 5/x in thickness. They were cut on the sliding microtome, thus \'aried a little in thickness and gave slightly varying degrees of differentiation. The larger and largest arteries were, owing to want of suitable material, not included in this specipJ STKlt "I'l in; OF KLASTICA I X'l'KKX A- A K l'i;i{IKS I7:i study, tlioutili provioiis inckleiital study of such vessels leads mo 1() l)oli('\(> that tlio elastic iutiina, where present as such, is in general charactcM- like that of the smaller vessels. In no instance \ver(> tilt' aitciies especially studied remoNcd from tlie surromidiujj; connective tissue, so that the staining- of the elastic tissue in the peri\'ascular areolar tissue served as a control for the staining of the elastic tissue in the arterial walls. In all of the successfully stained preparations and in arteries varying in size from the smaller to the larger ones studied, the elastic intinia api)ears, when successfully differentiated, as a P'ig. 1 Elastic intima of deep plantar arteiy, human. Stained in ironhematox\din and van Gieson's acid fuchsin, picric acid solution. X 600. network of elastic fibers, the larger fibers of the network ha\'ing in the main a direction which is parallel to the long axis of the respective vessel. A well stained and well differentiated longitudinal or longitudinal oblique section of an artery including the elastic intima, appears not unlike a successfully teased preparation of yellow elastic tissue from the ligamentum nuchae. In figure 1 is presented a drawing of a portion of the elastic intima of one of the larger deep plantar arteries of a human foot. During fixation the artery had collapsed in such a way that on one side, for a distance, its wall was nearly in a plane. Several sections of a series thus included long stretches of the elastic intima. In this figure only the elastic tissue, which is 174 G. CARL HUBER stained deeply blue-black, is reproduced as drawn with the aid of the camera lucida, using a /o inch oil immersion objective and a No. 4 Zeiss compensation ocular with paper at table level. The network character of the coarser elastic fibers with frequent anastomoses and numerous cross-bridges is faithfully reproduced. It was not possible to draw accurately all of the finest fibrils throughout their entire extent. However, the figure as a whole gives a correct impression of the appearance presented by the section. At both ends (above and below the figure), the intima leaves the plane of section and the elastic fibers, shown as a network in the figure, appear as cross cut or obliquely cut fibers. In numerous other sections of vessels of varying sizes, longitudinally or obliquely cut, including the elastic intima, similar appearances are found. The character of the network varies but slightly, dependent on the degree of extension or distension of the respective vessel. Oblique sections approaching cross sections of vessels are especially instructive. In such sections a side view of the elastic network of the elastic intima with end view of the fibers as seen in cross-cut, is obtained by moving the micrometer screw of the microscope. In cross sections of vessels, in place of the usual line representing the elastic intima as seen after the usual staining, there is observed a row of deeply stained blue-black dots, varying in size with the size of the vessel, with here and there a longer or shorter blue-black dash where a cross anastomosis between fibers is included in the section. Sections of areolar connective tissue, differentially stained for elastic tissue by means of the iron hematoxylin picric acid, acid fuchsin method, present no evidence of an ^outer membrane' for elastic fibers as described by Mall. However, the existence of such a membrane is in no sense denied, since a slight tinging with picric acid would not be evident against the deep blue-black stain of what is probably the 'inner substance,' stained readily in magenta. In certain of the longitudinal sections of vessels including the elastic intima, as for instance in the section from which the figure here presented was drawn, a delicate grey-blue color overlies the elastic network. This is represented in the figure by a liglit wasli of neutral tint. If STIMC rri{K OF KLASTICA 1 .\ TKirN A-AllTEllIES 17.") tliis 1)(> expressive of structiiro it icvcals a lioinojiciieoiis stniftiiro and may j){)ssibly iudicato tlio presouce of a hoinof^eiieous iiieinbraiie. Such a uuMnbrano, however, I have not detected in cross sections of xessels. From this study of tlio elastic intima of aiteries the conclusion seems wairanted that tlio staiiiable substance of this layer consists of a network of yellow elastic fibiMs, with coarser fibers having; in the main a coiu'se which is ])arallel to the lonjy axis of the respective vessel, these fibers piesenting frequent anastomoses and cross bridges, and with numerous finer fibrils which pervade the network. Here and there certain of the fibers of tlie elastic intima may in cross or obHque sections be traced in anastomosis with elastic fibers of the media. It would thus appear desirable to discard the term 'fenestrated membrane,' since this term does not express the structure of this layer. Of previous descriptions, that given by Diirck appears to me the most nearly conforming with observed facts. A xoTi-: ox riiK .morphology of the sKAiiNna':R0us tubules OF BIRDS (;. CARL HUBER Departinoit of Anatomy, Universily of Michigan OXK FIGURE In the coiii'se of a study of the renal tubules of birds, by means of the maceration method devised by Huber/ in which full grown roosters (Callus domestica) were used as material, the injection of the 75 per cent solution of hydrochloric acid was tlu-ough the aorta central to the branches supplying the kidneys and sex glands. In a number of the cases the testes were found to be well injected and w^ere removed and placed in 75 per cent hAxh'ochloric acid with a ^iew of obtaining thorough maceration preparatory' to teasing. Following the method as described, the macerated pieces were washed thoroughly in distilled water, stained in hemalum, softened and cleared in 0.25 per cent to 0.5 per cent ammonia water, in which they were teased. Even during the preliminary teasing of the larger pieces it was noted that the testis of the rooster was not separable into lobular masses, as is the case in mammalian testes, so that it was found impossible to isolate structural units with which the final teasing could be carried out. Huber and Curtis- found that in the mammalian testis the seminiferous tubules presented no blind ends, diverticuli or nodular enlargements but were arranged in the form of an arch or a variable number of linked arches, all of the ends terminating ^ Huber, G. Carl. A method for isolating the renal tubules of mammalia. Anat. Rec, vol. 5, 1911. - Huber, G. Carl, and Curtis, George ^lorris. The morphology of the seminiferous tubules of mammalia. Anat. Rec, vol. 7, 1913. 177 178 G. CARL HUBER in tubiili recti attached to the rete testis. Wax reconstructions made by Curtis confirm the observations made on teased pieparations. Repeated teasings have convinced me that in the adult bird no such arrangement of tubules pertains, but that the seminiferous tubules of the bird are arranged in the form of a network, presenting a varying number of anastomoses found at different levels in the gland substance. For this reason the teasing of these tubules is exceedingly difficult in that it is impossible without breaking or tearing tubules to separate favorable pieces preparatory to final teasing. All of my teased preparations present an endless net, with broken tubular ends as a boundary. Fig. 1 Teased preparation showing a portion of the tubular system of the testis of the bird (Gallus domestica). X 5. In the accompanying figure is presented one of the most successful preparations obtained. The figure was traced and sketched with the aid of the camera lucida at a magnification of 50 diameters, reduced to a magnification of 5 diameters in the reproduction. The portion of the tubular net reproduced presents in all some forty broken ends, and at least three closed rings. Such closed rings I have found in nearly all of my prepaI'ations. Their conii)lete separation I'equires great care and patience since uniting tubular poitions are very easily broken. The clearness of the figure, it is thought, obviates the necessity SK.MlMI'Kltors Tl'mivKS 179 of cxtciulcd (h'scriptioii oi" the cliaiaclcM' of the iicl work formed l)y tlic sciiiiuifcrtms tul)ul('s of hiids. The figure, liowever, sliould 1)0 studied with the mulcisluHdinj;' 1 hal \n tlie mount from wliich the figure was dniwu the teased tulniles were spread out as nmch as possil)le. The fip;uie, tliorefore, does not give spatial relations of the tul)ules. in the gland, as is well known, these luhules foini compact coils, evident somewhat from the extended kinks and bends seen in the figme. In a rather careful search of the literature -1 have been unable to find any description of the form of the seminiferous tubule of birds. Hiis note would thus seem justified. However, the ol)ser\'ations liere recorded would seem to me to have a bearing on i)re\'ious work emanating from this and other laboratories, relating to tlu^ foim of the seminiferous tubules of mammalia. The results heie recorded seem to me to confirm the observations made on teased preparations of the . seminiferous tubules of mammals. The fact that complex anastomoses resulting in closed ring structures have been teased in the bird's testis argues for the possibility of teasing such structures in the adult mammalian testis, did they exist. Bremer,^ as a result of careful wax reconstructions of the tubular system of the human testis, working on embrj^onic and fetal tissue, the oldest stage studied being that of a human fetus of seven months, reached the conclusion that The testis cords, growing from the germinal epithelium of the genital ridge, form a network with three sets of anastomosing l^ranches. After completion, this network breaks down partially, leaving certain cords as persistent stems. The tubules of the adult show, in their course, connection, and position in the testis, traces of this network. Testis tubules may be single, ending blindly, may branch, or may anastomose. In the adult mammalian testis tubules, completely teased, no bhnd endings, buds, nor ring formations were observed, while in the teased preparations of the seminiferous tubules of birds, the remains of the network of tubules as observed in the embryonic and fetal stages and well figured by Bremer, may be ^ Bremer, John Lewis. The morphologj' of the tuljules of the liuman testis and epidich'inis. Amer. Jour. Anat., vol. 11, 1911. 180 G. CARL HUBER noted. In a cryptorchid of the rabbit, as described by Huber and Curtis, extended anastomoses of testis tubules were observed in two regions of the tubule complex, and in two regions, near the periphery of the gland, tubules were joined so as to form two folded rings. The preparations from the crj^ptorchid of the rabbit present appearances not unlike those shown in teased preparations of the seminiferous tubules of the bird. The presence of the remains of the embryonic network of the seminiferous tubules in the cryptorchid of the rabbit and in the bird's testis, postulates a relatively late, complete morphogenesis of the seminiferous tubijles of the mammal. Phylogeny and ontogeny indicate this. In the light of these observations I am of the opinion that Bremer's careful study of the morphology of the seminiferous tubules of the hiunan testis are of value as concerns embryonic and late fetal stages, but may not be transmitted to the adult gland in that in the oldest stage studied, a human fetus of the seventh month, the seminiferous tubules, in all probability, had not completed their morphogenesis. The question is one deserving further study and will form the subject of a future, more comprehensive communication, based on especially prepared and 'timed' material from the rabbit. This form is chosen since the morphology of the seminiferous tubule of the adult rabbit has received special consideration in this laboratory, both by means of teased preparations and reconstructions. /^/ THE HISTOLOGY OF BLOOD AND LYMPHATIC VESSELS DURING THE PASSAGE OF FOREIGN FLUIDS THROUGH THEIR WALLS II. STUDIES ON ABSORPTION FROM SEROUS CAVITIES P. G. SHIPLEY AND R. S. CUNNINGHAM From the Aiiatomical Laboratory of the Johns Hopkins University, Baltimore It is at once evident to any one who studies the enormous literature which has been pubhshed 'during the last century, on the absorption into the body of foreign material from the serous cavities, that the problems presented to the present day investigators in this field may be roughly grouped under foui* heads, as follows: 1) the localization of absorbing surfaces; 2), the definite establishment of the channels of removal; 3) the determination of the forces concerned in the passage of matter through the walls of the serous cavities, and its entrance into and exit from the vessels which distribute it about the body for storage and digestion, or for destruction and excretion; and 4) the location and identification of the organs concerned in the storage or destruction of substances or fluids absorbed. Up to the present time not even the questions which are included under the first and fourth of these heads can be said to have been sufficiently answered; the second has been settled definitely only for a single area and with a limited number of substances, and about the thu'd practically nothing is known. Although a vast amount of work has been done on the localization of absorbing surfaces and then- related organs of storage and excretion, until quite recently but little was known regarding them; and this is so in spite of their immense clinical importance in connection with the postural treatment of the various serousitides. It is owing to the work of MacCallum (1), who made a careful and productive study of the removal of foreign granules from 181 THE ANATOMICAL RECORD, VOL. 11, NO. 5 DECEMBEB, 1916 182 r. G. SHIPLEY AND R. S. CUNNINGHAM the peritoneal cavity through the lymphatic lacunae of the diaphragm, that practically all who have made a study of serous absorption accept it as a fact that a great deal of any solution or foreign body introduced into the peritoneal cavity passes through the peritoneal surface of the diaphragm, and enters the system of the experimental animal or the patient via its vessels. The position of other foci of drainage has remained unknown, and the tendency to deny or ignore their existence has been and is very strong. A few observers have suggested the broad surface of the greater omentum as a possible^ agent in the removal of foreign fluid from the peritoneum, but their assertions have rested upon probabilit}^ rather than definite proof. Rubin (2), who attacked the problem from an experimental basis, showed, however, that less fluid was absorbed from the peritoneal cavities of animals whose omenta he had amputated, than from the peritoneal cavities of normal controls ; and Grouse (3) after careful study concluded that the omentum is an important factor in the mechanism of peritoneal drainage, and hypothecates a protective lymphatic drainage to account for the phenomena which he has observed. The authors (4) have been able to show experimentally that beyond doubt the omentum is a very efficient agent in the drainage of the peritoneal cavity. By drawing the omentum out of the animal's body through a midline incision, and keeping it immersed in a fluid medium under physiological conditions, it was possible not only to isolate the organ and to prevent the experimental fluid from reaching other surfaces, but also to eliminate any influence on absorption which might be exerted by the increased abdominal tension which follows the intra-peritoneal injection of large amounts of the fluid. In spite of conditions which might be supposed to make for secretion rather than absorption, we found that a large amount of the fluid in which these omenta were immersed passed into the omental vessels, and reaching the general circulation, was carried at once by the blood stream to the organs of excretion, from which the test fluid could then be recovered. AHSOIM'TIOX FROM SEROUS ('AVITFES 183 As to tlu> socoiid (lucstioii, \\\o ostablishment of the vascular system conconuMl in tlio drainage, for example, of the peritoneum, Meltzer (a), Museatello (6), and others, held that drainage is accomplished through the lymphatics, while Heidenhain (7), Cohnstein (8), Dandy and Rowntree (9), and others, have shown that much of the fluid absorbed from the peritoneal cavity leaves it through the blood vessels. We have never been able to demonstrate the presence of lymphatics in the omental tissue of the adult cat, and Ranvier (10) claimed that while there are lymphatics in abundance in the omenta of young kittens many of them are obliterated by degenerative changes at the age of three months. If lymphatics exist in the cat's omentum they must necessarily drain in the same direction as those of the gastric system; that is, an omental hnnphatic stream, if such a thing exists, must eventually become tributary to the l^anph content of the thoracic duct. In the experiments mentioned above with the omentum of the cat, the influence of lymphatic vessels was entirely eliminated in many of our experiments by the hgation of the duct. Hence we were able to prove not only that the omentum furnishes a surface where absorption takes place, but, by varying the fluid in which the omenta were immersed, we have shown that the removal of molecular solutions and colloidal solutions and of fine particulate matter in true suspensions may be accomplished through the blood vascular system to a large extent; though we do not by any means deny the probability of drainage through the lymphatic channels in the localities where these vessels exist. But when any attempt is made to ascertain, through the medium of existing literature, the forces concerned in the absorption of foreign matter from serous surfaces one enters at once into a region of guess and hazard, where only a few isolated facts exist as a guide to certain knowledge. We have only just ceased to argue for and against the presence of preformed 'stomata' and 'stigmata,' and to indulge in surmises as to their physiological significance. Students of the phj^siology of absorption are still discussing whether absorbed material passes through or between the lining endotheUum of blood and lymphatic vessels and the 184 p. G. SHIPLEY AND R. S. CUNNINGHAM mesothelial cells of serous cavities. In our own experiments we found that a great deal of fluid may enter the blood stream when the influence of intra-abdominal pressure is removed; and, since our fluids were isotonic with the blood serum of the experimental animal, osmosis as it is generally understood, could have had only a negligible amount of influence upon the phenomena observed. Indeed if osmotic pressure had any influence at all, it would seem that it would have been exerted against rather than for the passage of the experimental fluid into the blood vessels, since even the small amount of fluid lost from the solution by evaporation from exposure to the air, must have changed an originally isotonic to a slightly hypertonic fluid, to which one might expect water to pass from the serum through the vascular wall. If such a passage occurred it in no way interfered with the imbibition of the experimental solution. We know very little about the part played by fluid pressure, the movement of the blood and lymph in their respective vessels, or the influence on serous absorption of the movement of contractile somatic organs, like the diaphragm, or the contraction of the musculature of the vessels themselves. We cannot say what chemical changes accompany or influence the transport of material from cavity to vessel; or whether the cytoplasm of the cells of the serous cavities, or of the blood and lymphatic vessels play any part in the transmission of matter thrc^ugh the vascular or serous walls. And does a disturbed balance of intra- and extra-cellular equilibrium militate for or against absorption? We do not know. It will readily be seen that the examination of histological preparations made from the omentum during active drainage, may be of great value in strengthening the positive evidence for absorption through the blood vessels, and in aiding us to understand the mechanism of the removal of foreign matter through the vascular wall. With this end in view, sections have been made and studied of omenta which, up to the time of fixation, had been exposed to and were absorbing all sorts of material from true solutions to mechanical suspensions. A report of the findings in this material is the purpose of the present paper. AUSOHI'TION FI'lOM SKllOUS ("AVITIKS 185 By far the most valujible preparations were yielded by omental tissue which had been absorbing an isotonic solution of potassium ferrocyanide and iron auunoniuni citrate, and which was fixed immediately upon remoxal from that fluid in hydrochloric acid formalin with a resulting precipitation of prussian blue — the method used by Weetl (11) to study the drainage of the cerebro spinal fluid. An omentum so treated appears in gross to be stained a uniform pnXe blue except for the fat, and is patterned by an irregular nerwork of an intense dark blue color. It is only necessary to examine the spread preparations with a, binocular microscope to be convinced that the network is made up of the omental blood vessels whose lumina are filled -with precipitated Prussian blue; the picture is strikingly suggestive of a complete blood vascular injection of the omentum with a somewhat dilute prussian blue gelatine mass, and, in fact, we are dealing with much the same thing, since the coagulation of the colloidal proteids during fixation of the blood serum causes the same comminution of the nascent dye stuff which follows its precipitation in pectizing gelatin. In contrast to the general tissue which fills the meshes of the vascular net and which is very pale blue, or uncolored, a wide deeply stained zone of thicklj^ precipitated dye surrounds each vessel. The capi laries are all filled with prussian blue, even those supplying the perivascular fat being crowded with the dye and the capillary knots or glomeruli which form the support of many of the taiches laiteuse are completely injected. Here and there a capillary may be seen empty or nearly so, perhaps because contraction of its walls during fixation forced the absorbed dye from its lumen. Of the larger vessels all have a greater or less amount of dye precipitate within the lumen. All are completely filled, but in some the blue color is perceptibly paler than in others. In general the arteries show much less absorption than the corresponding vein, but the depth of the color may not be the same throughout the length of a given vessel. There are often light and dark blue areas present. These preparations show that ac 186 p. G. SHIPLEY AND R. S. CUNNINGHAM tive absorption is going on through the walls of the arteries as well as through the veins, even arteries with thick muscular walls taking part in the general process as will be shown below. The larger vessels are paler than the smaller, probably because less fluid is taken in through their walls. In other words as the vascular size increases there is a gradually decreasing concentration of the intra-vascular dye solution, the significance of which will be discussed below. From the point where the vessels begin to be surrounded by a perivascular sheath of fat the pallor of the precipitated dye in the vascular lumen markedly increases, evidently because the advent of the perivascular fat is accompanied by an increased thickness of the vascular wall and a diminished absorption. That the fat itself can have no effect in the decrease is shown by the intense color of the injection mass in the capillaries supplying the fat; and examination of sections shows that the ferrocyanide solution penetrates easily between the fat cells themselves. Sections of the same material confirm the evidence of toial spread preparations. The capillaries, even those imbedded in and sulpplying the fat, are distended with the blue color, and the veins are full of blue precipitate of varying depth of color. It is however not so easy to see the blue color in the larger arteries in sections. It is possible even in thin sections to distinguish the deep blue perivascular area described above, and to trace its existence to precipitated dye in the intrafibrillar tissue spaces. Individual fat cells are outlined by dye precipitated from the fluid which has worked its way between the cells that it has never penetrated. Coarse precipitates of dye may be seen along the surface of the omentum and adherent to the surfaces of the elastic fibres. In some places the tissue is diffusely stained, and throughout the omentum, cells, probably of the clasmatocyte type are found, like those described by Weed in the meninges, whose cytoplasm is filled with fine granules of Prussian blue, their nuclei however remaining uncolored. This intracellular precipitate is the result of imbibition of fluid by the AHS()KI"I'1(».\ FROM SKltOlS CWI'IMIvS 187 (•()iuuH'ti\(' lisuc cells, an adsoi pi ion |)l»ononieii()n of tlie sanio luituic as the diiukin{2; in of solutions of hi^li niolcciilar vital live stuffs which is lesponsible for the diffuse cytoplasmic coloration seen early in a course of staining. The blood vessels contain many leucocytes, mostly of the monoiuiclear type, embedded in the precipitated blue of the injection mass. Tliis prussian blue precipitate in the blood vessels is not the coarse amorphous mass in which that color is usually seen under the microscope, but because the dye was thrown do^vn in the presence of the colloidal serum proteids it is so finely divided as to appear homogeneous except when examined with the highest power immersion lenses with which its finely granular nature can be ascertained. It is in the same physical condition of finely divided suspension as the silver in silver gelatine mixtures of photographers, or the dye granules in injection masses made by precipitating colors in the presence of solidifying gelatin. The intracellular precipitates and those adherent to the surface of the omentum and its component fibres are much more coarsely granular. The endothelial walls of both capillaries and larger vessels are stained a dark blue. The cytoplasm of the endothelial and mesothelial cells is entirely filled with a fine granular precipitate of the Prussian blue, and in some places dye particles are found appaiently between the cells, though the walls of the cells are in such close apposition that it is difficult to say with certainty that such is the cause. The cell nuclei are uncolored, and the cytoplasm of the serous mesothelium covering the omental surface is filled with the fine granules of dye which show the track of fluid which has passed through their bodies. By the time that the dye bearing serum has reached the visceral blood vessels— liver, lung, etc., — it has become so diluted with blood from non-absorbing parts of the body that it is not possible to follow the course of the chromogen through the body by the examination of sections. Large quantities are present in the kidney tubules, and the presence of the dye in the urine of the animal can easily be demonstrated. 188 p. G. SHIPLEY AND R. S. CUNNINGHAM The same conditions prevail, though they are much more difficult to demonstrate, in preparations made from omenta which have been immersed in strong solutions of trypan blue and collargol and colloidal solutions of other metals. In animals which have been injected intraperitoneally with the chromogen solution — or where certain isolated portions of the peritoneal surface (small intestine or bladder) — have been inmiersed in, or covered with cyanide-citrate solution, and fixed in hydrochloric acid formalin, the blood vessels and lymphatics directly beneath the peritoneal surface, are found on section to be filled with dye precipitate and to have the same appearance as the omental vessels. Moreover it was possible in the gi'oss to trace the stained lymphatic vessels directly to the lymph nodes into which they drained, and to obtain definite macroscopic evidence of the presence of prussian blue in the lymph gland. This feature of peritoneal absorption will be taken up separately in a later communication. It is evident then from these histological preparations that there is very active absorption of foreign fluids through the peritoneal blood vessels, not only those in the omentum, but also through those beneath the peritoneum over the gut and bladder. In all probability fluids may be removed from the peritoneal cavity through any area in which blood or lymphatic vessels lie just beneath the peritoneal surface. Furthermore absorption of fluid obtains not only through capillaries, but through vessels of quite large caliber, and through arteries as well as veins, though not to as great an extent probably, because of the greater obstacles to fluid passage offered by the tissues which go to make up the thicker, denser arterial wall. This is probably the reason for the pallor of the dye mass within the larger vessels, since it would seem reasonable to suppose that their thicker walls would hinder the passage of fluid and make it slower and of less amount. That fluids do pass through is shown by the fact that the wall and its lining endothelium contain granules of stain precipitated from the fluid during its passage into the vessels. That surrounding tissues have no significance in preventing fluid from coming in AHSORPTIOX FROM SEROUS CAVITIES 189 contact with these large vessels is shown, as we have pointed out tibove, by the ease wi'.h which it penetrated between the cells in the i)eriv;iscular fat and iilled the capillaries by which the fat is supplietl. There is of course the possibility of dilution of the chroniogen fluid in the larger vessels as a result of their receiving blood from vessels thi-ough which absorption was not going on, but this is unlikely, since the vessels themselves and their entire tributary area were immersed in the test fluid. The significance of the dark stained areas about the blood vessels is not quite clear. Apparently the solutions are drawn forcibly from the general tissue towards the blood vessels faster than they can be forced through the blood vessel wall, and then, removal being delayed (perhaps by the condensation of the connective tissue in the vascular margin) they are concentrated there. What the forces are which are exerted on the fluid, and what part is played by the movements of the omentum as a whole, the contraction of the blood vessels and the movement of the blood within them, it is impossible yet to say. The material demonstrates also that while some fluid may pass between the lining cells of vessels on its way to their lumen, by far the larger part goes through the cytoplasm of the cells themselves. The sections are also of interest in that they show how little, if at all, the omentum was damaged during the operative procedure which preceded its immersion in the test fluid. Xowhere is there any sign of exudation or haemorrhage; there is no cellular death, as may be seen by the un colored nuclei of the various cells; and, moreover, the mesothelial cells of the serous surfaces show no sign of disturbance or desquamation. 190 p. G. SHIPLEY AND R. S. CUNNINGHAM BIBLIOGRAPHY 1) MacCallum, W. G. 1903 Johns Hopkins Hospital Bull., 14, 105. 2) Rubin, I. C. 1911 Surgery, Gynecology and Obstetrics, 12, 117. 3) Grouse, H. 1912 Bulletin of the El Paso Med. Soc. April. A) Shipley, P. G. and Cunningham, R. S. 1916 Am. Jour, of Physiol., 40,1, 75. 5) Adler, I. and Meltzer, S. J. 1896 Jour, of Exper. Med., 1, 482. 6) MuscATELLO, G. 1895 Arch. f. Path. Anat., 142, 327. 7) Heidenhain, R. 1891 Pfliiger's Arch. f. d. gesammt. Physiol., 49, 209. 8) Cohnstein, W. 1895 Zentralbl. f. Physiol., 60, 484. 9) Dandy, W. E. and Rowntree, L. G. 191 4. Annals of Surgery, 59, 587. (10) Ranvier, L. 1896 C. R. Ac. des Sc, 122, 578. (11) Weed, L. H. 1914 Jour. Med. Research, N. S., 26, 1, 21. A C\^K OF A Lr:FT SUPERIOR VENA CAVA WITHOUT A CORRESPONDING VESSEL ON THE RIGHT SIDE WILBUR C. SMITH The Ariaiomical Laborolon/ of the Wake Forest School of Medicine TWO FIGURES Many instances are recorded in the literature of the presence in human fetuses and adults of two venae cavae superiores, with or without a transverse inter-jugular anastomosis^ The presence, however, x)i a left vena cava superior persisting without a right (the \dscera not being transposed) is comparatively rare. I have studied the original descriptions of all the cases of this nature occurring in the bibliographies by Ancel, P. et Villemin ('08),' Boyd ('93), Halbertsma ('62), McCotter ('16), Nutzel ('14), and Weigert ('81) with the exception of the case of Mausert ('99) which was not available, and find that thirteen such cases have been previously recorded. The subject of the anomaly here recorded is a middle aged, well developed male. The right internal jugular and subclavian veins unite to form a comparatively long innominate vein (referred to in this description as the right innominate) which extends obliquely downward and to the left, ventral to the roots of the arteries arising from the aortic arch and unites with the short left innominate vein to form the left vena cava superior. The left vena cava superior crosses ventral to the arch of the aorta, to the left pulmonary artery, and ventral to the root of the lung as it approaches the dorsal surface of the heart; here 1 The case of Ancel, P. et Villemin ('08) is usually cited in the literature as being one of simple left vena cava superior, is one of a double vena cava. The case of Cheselden (1713) (the correct reference to which is given in the accompanying bibliography), has been difficult to find on account of the frequence with which an erroneous reference has been given. 191 192 WILBUR C. SMITH it reaches the sulcus coronarius and becomes continuous with a large coronary sinus which opens into the right atrium in the usual situation. A careful dissection was made for a vein representing the right vena cava superior, but no trace was found excepting the terminal part of the azygos which represents that part of the vena cava superior developed from the anterior cardinal. The highest right superior intercostal vein (draining the first space) is a tributary of the right vertebral. The azygos vein is somewhat smaller than normal, but receives the usual tributaries. It opens by means of the persisting caudal part of the anterior cardinal into the right innominate vein about one inch from its right extremity. On the left side the highest intercostal is a tributary of the left vertebral vein. The uninterrupted hemiazygos system is a large vein (representing the left superior intercostal, hemiazygos and accessory hemiazygos) which opens into the left vena cava superior. Its caliber is nearly as large as that of the normal internal jugular vein. There are two inferior thyreoid veins (one for each lobe), each of which opens into the right innominate vein. The right internal mammary vein empties into the right innominate vein ventral to the termination of the azygos; the left is represented by two veins, the larger empties into the left vena cava superior, the smaller (representing the pericardiophrenic tributary) into the left innominate vein. The cardiac veins are normal in position and termination. The great cardiac is smaller than usual. The heart is normal in size and position. In the upper dorsal part of the right atrium at the site of the ostium of the vena cava superior, the atrial wall is covered within by musculi pectinati. Below this point the inner surface of the dorsal wall is smooth. The ostium of the vena cava inferior at the lower and dorsal part of the atrium is normal, there is a faint trace of the inferior caval valve. The fossa ovalis and its limbus are normal, a foramen ovale being absent. Between the large coronary ostium and that of the vena cava inferior there is no intervening space, nor is there a coronary valve. Both from the exterior and in LEI-^ SUPEHIOH VENA CAVA W Tl'IK KIT TIIK ItlfillT 193 terior of the atrium, tlie two veins ;i,i)i)eiir to comiiiuuicate with the atrium by a common opening- Tlie tricuspid valve is normal in position and arrangement. The remaining chambers of the heart are normal. Before off(M'ing an explanation for this anomaly, the normal method of ckn^elopment of the veins in (juestion may be briefly recalled. It is well known that the early embryological condition is one in which the veins are sjanmetrical on the two sides. On either side the anterior cardinal vein unites with the posterior cardinal to form the common cardinal vein (duct of Cuvier), and eacli common cardinal opens into the lateral part of the sinus venosus (sinus horn) of its own side. The sinus venosus, which is a transversely mdened chamber, at first- communicates wdth the common atrium by a large foramen, but during the formation of the interatrial septum, the sinus venosus comes to open into the right portion of the dividing atrium. Somewhat later in development the terminal part of the subclavian veins (which at first open into the posterior cardinals) migrate cephalad to become tributary to the anterior cardinals. The large trunk on either side caudad of the confluence of the subclavian and anterior cardinal veins becomes the primitive vena cava superior. Each primitive superior cava consists of two regions, a cephalic part originally derived from the anterior cardinal and a caudal part derived from the common cardinal. The smaller trunks lying cephalad of the anterior cardinal-subclavian junction of either side represent the internal jugular of the adult. Subsequently there is formed upon the medial side of each internal jugular vein, in close proximity to its junction with the subclavian, a vein (Vena thjonico-thyreoidea, Thyng '14) which drains a venous plexus about the developing thyreoid and thymic glands. An anastomosis between these thymico-thyreoid veins e\ddently forms a transverse anastomosis, connecting the right and left jugulars, which normally becomes the vena anonyma sinistra (Szawlowsky '91, Anikew '09, and Thyng '14). This interpretation is substantiated by the fact that the inferior thyreoid and thjinic veins of the adult are usually tributary to the vena anonyma sinistra. 194 WILBUR C. SMITH vjug.int.d. Vfnferc. supnd. Vsubcl.d. Vanom. d. Vmont irttd. v:%m. v: interc. Su.d V&nt. d. v:ca\/. Jnj: -Vjug.ints. VthyKfTied.S. X trans, scap. -V verfeb.s. y subel.s. -V anom.s. -v: peric. phr. v: mam. lots. V cav. ^up. V hem'iaz 1 V interc so. s. -Venf. S. SOi^'/v^ST^ Fig. 1 Ventral aspect of heart and thoracic veins ABBREVIATIONS USED IN FIGURES 1 AND 2 A.pulm., A. pulmonalis Aor.asc, Aorta ascendens Atr.d., Atrium dextrum Au.s., Auricula sinistra Gl.thyr., Glandula thyreoidea V.anom.d., V. anonyma dextra V.anom.s., V. anonyma sinistra V.az., V. azygos V.cav.inf., V. cava inferior V.cav.sup., V. cava superior V.hemiaz., V. hemiazygos V. interc. supr.d., V. intercostalis suprema dextra V. interc. su.d., V. intercostalis superior dextra V.interc.su.s., V. intercostalis superior sinistra V.jug.int.d., V. jugularis interna dextra V.jug.int.s., V. jugularis interna sinistra LEFT SUPEIUOK VENA CAVA WITHOUT THE RIGHT 195 v: jug. int. d VTsubcl.d YinfBrc.supr.d Vmam. intd V.ihym. Aor CSC .v: thyr med.S. Yju^. int.S. V.vertebs. s. scap.s. ubcl.s. V. anom.s. V. penc.phr ^/: mam.ints ,Y hemiaz; -Vinterc.SU.S; SOr^/'lpS Fig. 2 Ventral aspect of thoracic veins. The apex of the heart has been turned toward the right to expose the left vena cava superior. ABBREVIATIONS (CONTINUED) \'. mam. int. d., V. mammaria interna V. thyr. inf. s., V. thyreoidea inferior dextra sinistra V.mam.int.s., V. mammaria interna V.thyr.med.s., V. thyreoidea media sinistra sinistra V.peric.phr., V. pericardiaco phrenica V. trans. scap.s., V. transversa scapulae V.subcl.d., V. subclavia dextra sinistra V.subcl.s., V. subclavia sinistra V.verteb.s., V. vertebralis sinistra V.thytn., V. thymica Vent.d., Ventriculus dexter V. thyr. inf. d., V. thyreoidea inferior Vent.s., Ventriculus sinister dextra 196 WILBUR C. SMITH Normally the blood which reaches the left side of the neck now presumably finds a more favorable course through the transverse inter-jugular anastomosis into the primitive right vena cava superior and thence into the right atrium, the greater part of the sinus venosus by this time having been absorbed into the latter. At any rate, the portion of the primitive left vena cava superior, representing the part of the common cardinal immediately caudal of the termination to the hemiazygos system (posterior cardinal), either atrophies or becomes fibrous. The transverse inter-jugular anastomosis then becomes the left innominate vein of the adult; the terminal part of the left anterior cardinal forms the proximal part of the left superior intercostal, and the caudal portion of the left common cardinal persists as the oblique vein which is tributary of the coronary sinus. The adult vena cava superior formed by the confluence of the right left innominate veins, represents the terminal part of the right anterior cardinal together with the entire right common . cardinal vein. EXPLANATION OF ANOMALY The anomaly above described is apparently due to the fact that subsequent to the formation of the transverse inter-jugular anastomosis, the right common cardinal was obliterated instead of the cephalic portion of the left common cardinal, which here remained intact. This condition may be explained by assuming either that the left thymico-thyreoid vein had a more caudal origin than normally occurs, or that it migrated early in development to a more caudal position than is usual. In either case when the transverse inter- jugular anastomosis was formed; the blood, by following the course of least resistance, must have flowed from right to left instead of vice versa as usually occurs. I wish to take this opportunity to extend my appreciation to Profs. H. D. Senior and F. W. Thyng of the University and Bellevue Hospital Medical College for their kind suggestions during the preparation of this paper. LKFT sri'Kuiou m:.\.\ cwa wrnioi'i' tiik hicht 197 MIl^LloCltAI'llV Anikkw, a. loot) Zur Franc iilxT die iMituicUcliiiiK 'Icr NCiia aiioiivma sinistra. Aiiat. An/.., H.l. M. S. LM-2it. Ancki., r. ET ViLLKMix lOOS Jour. (le I'Anat., vol. 44, pp. 4(H)2. B^DARD, 1892 Vena Cava suporieure situce a gaiicho. Hullotins de la soci6t6 d' anthropologic dc Paris, T. 30, p. 379. Boyd, S. 1893 A case of left .superior cava without traiispo.sition of viscera. Jour. Anat. and Physiol., vol. 27, N. S., vol. 7, p. 20. Charles, J. J. 1889 Note of a case of persistent left superior vena cava, being in great part a fibrous cord. Jour. Anat. and Physiol., vol. 33, X. S., vol. 3, p. 049. Cheselden, W. 1713 Some anatomical observations. Philos. Trans., vol. 27-28, p. 281. Deitrich, a. 1913 Uber ein Fibrioanthrosarkom mit eigenartiger Ausbreitung und liber einc Vena Cava superior sinistra bei dem gleiden Fall. Ar chiv f. Path. Anat., Bd. 212, S. 119-139. Greenfield, W. S. 1876 Persistence of left superior vena cava with absence of the right. Trans. Pathol. Soc. of London, vol. 27, p. 120. Gruber, W. 1880 Vorkoinmen einer Vena cava superior sinistra (bei Abwe senheit der V. cava superior der Norm.) (3. der im Verlaufe von 167 Jahren zur Kenntnis gekommenen Fiille.) Virchows Archiv, Bd. 81, S. 458. Halbertsma 1862 De Afwyking van het Tusschenschot der kammers en der primitive aorta naar links, met hare gevolgen; spater deutsch im Archiv fiir die hollandischen Beitrage zur Xatur-u. Heilkunde. Bd. 3, S. 387. LiNDES, Georg 1865 Ein Beitrag zur Entwicklungsgeschichte des Herzens. Diss. Dorpat. Marshall. J. 1850 On the development of the great anterior veins in man and mammalia. Philos. Trans. Roj'al Soc. vol. 140, part 1, pp. 133-170. Mausert, a. 1899 Zue Casuistik der vena cava superior sinistra und der einen Spitzenlappen der rechten lung abschniirenden anomalie der vena azygos. Diss. Giessen. McCoTTER, RoLLO E. Three cases of the persistence of the left superior vena cava. Anat. Rec, vol. 10, pp. 371-383. Merkel, H. E. 1912 a ]\Iissbildung im Bereich der oberen Hohlvene. Muen chener ^ledizinische Wochenschrift, Bd. 59, S. 615. 1912 1) Missbildung der oberen Hohlvene. Muenchener Medizin ische Wochenschrift, Bd. 59. S. 110. NuTZEL, H. 1914 Beitrag zur Kenntnis der Missbildungen im Bereiche der oberen Hohlvene. Zeitschr. f. Path.. Bd. 15, S. 1-19. ScHROEDER, R. 1911 Uber anomalien der pulmonalvene Zugleich im beitrag zum Cor biloculare. Archiv f. Path. Anat., Bd. 205, S. 122. SzAWLOw.sKi, J. 1891 Zur Morphologic der Venen der oberen Extremitat und des Halses. Dockt.-Diss., St. Petersburg. THE ANATOMICAL RECORD, VOL. 11, NO. 5 19§ WILBUR C. SMITH Thyng, F. W. 1914 The anatomy of a 17.8 mm. hmnan embryo. Am. Jour. Anat., vol. 17, pp. 31-99. Weigert, C. 1881 Ueber einen Fall von links verlaufender Vena cava superior, ' mutmaaslich bedingt (lurch friihzeitige Synostose der Sutura mastoi dea dextra. Virchows Archiv, Bd. 84. S. 184. THE INNERVATION OF TH1<: MUSCLE RETRACTOR OCULI G. S. HOIMvIXS Cornell University, Ithaca, New York ONE FICUUE Til view of the great nunihor of dissections of the cranial nerves of the horse that ])resunial)ly have been made in the ^'eterinary colleges of this countr>' antl of Europe and in view of the probably still greater number of similar dissections of certain of our domestic animals, especially the dog, the cat and the rabbit that have been made in the numerous laboratories of comparative anatomy and physiology, it would seem that nothing further remained to be said concerning the gross anatomy of these nerves. However, after many dissections of the cranial nerves of the horse and certain other of the domestic animals I am convinced that the descriptions of two of these nerves, viz., the N. oculomotorius and the N. abducens as given in many of the standard veterinary and comparative anatomies, are incorrect. The error referred to consists in attributing two sources of nerv^e supply to the M. retractor oculi (retractor bulbi, suspensor oculi, posterior rectus, choanoid) namely, the N. oculomotorius and the N. abducens whereas the muscle is innervated exclusively by branches from the latter nerve. The most common statement as to the distribution of the Nn. oculomotorius and abducens, in quadripeds, is essentially that given by Chauveau as long ago as 1857. According to this author the N. oculomotorius is distributed to the following eye muscles — the dorsal, medial and ventral recti, the obliquus ventralis (or externus), the levator palpebrae dorsalis and the retractor ocuU with the exception of its lateral portion; it also supplies one or more motor roots to the ciliary ganglion. The 199 200 G. S. HOPKINS N. abducens, according to Chauveau, supplies the M. rectus lateralis and the lateral portion of the retractor oculi. (Foltz states that Chauveau subsequently found that the M. retractor oculi was innervated exclusively by the N. abducens). This distribution of the two cranial nerves under discussion is, in the main, correct; the nerve supply to the M. retractor oculi however, as given by Chauveau and a number of other writers is without doubt incorrect. A brief review of the innervation of the M. retractor oculi as given by several writers, in mammals and in some other animals will first be noted. M'Fadyean gives precisely the same distribution as just quoted. Bradley's description is the same as the above with this slight difference, viz., the medial part only of the M. retractor oculi is mentioned as receiving a branch from the N. oculomotorius ; the lateral portion of the muscle, according to Bradley, is supplied by a branch of the N. abducens, as described by Chauveau and M'Fadyean. In the latest American edition of Strangeway's Veterinary Anatomy no mention whatever is made of any portion of the M. retractor oculi being supplied by the N. oculomotorius. But taken in connection with what is said of the muscle "that it completely envelopes and forms a sheath round the extra cranial portion of the optic nerve" and also in connection with what is said concerning the distribution of the N. abducens "it is distributed to the lateral rectus and the lateral portion of the retractor oculi" one may fairly infer that the M. retractor oculi with the exception of its lateral portion, is supplied by some other nerve than the abducens, presumably by the N. oculomotorius. In the first edition of his Veterinary Anatomy Sisson mentions both the oculomotorius and abducens as supplying branches to the M. retractor oculi; one poi'tion of the muscle being supplied by a branch from the dorsal portion of the oculomotorius while the dorsal and lateral parts of the muscle are supplied by the N. abducens. In a subseciuent edition, ll()we^'er. this error is corrected the M. retractor oculi being described as innervated bv the N. abducens onlv. I.\M:K\ A rio.N Ol' Ml SCI, K HKIKA* I'dU OdlJ 201 According to Sliarc-.loiics I lie M. rcti'tictor ociili receives l)r;iiu'li('s iVoiii tliicc (lifTci'eiit soun-es IVoin (lie ociiloinotorius, llu' troclilcafis and I lie ahdufens. Alaitiii and l-'caiick stale llial the M. retractof ociili feceives its iiene sui)i)ly troiii l)()tli the dorsal and Acnlral hranches of tlie oculo-inotorius and from the abtlucens. Ill contrast with this statement of Mai'tin and Franck is that of r]llenberji;er and Baum, also Gurlt, who mention the dorsal brancli only of the oculomotorius and the abdiicens as supplying branches to the M. retractor oculi. Struska agrees in all i-espects with Ellenberger, Baum and Gurlt. Leisering says that the lateral ])ortion of the retractor oculi is su])plied by the aijducens and the remaining portions of the muscle by branches from the oculomotorius. According to Zimmerl the M. retractor oculi is supplied from the dorsal branch of the oculomotorius and from the abducens but he does not mention the particular portions of the muscle supplied from each of these two sources. \^araldi gives practicall}' the same distribution as Zimmerl only he does not state from which branch, dorsal or ventral, of the oculomotorius the filaments to the retractor oculi are given off. In an article on the development of the eye muscles of the pig, Renter describes the M. retractor oculi as receiving its nerve supply from both the oculomotorius and the abducens. In the dog Bradley, Ellenberger and Baum describe both the dorsal and ventral portions of the N. oculomotorius as supplying branches to the M. retractor oculi; they make no mention of any branch from the N. abducens to this muscle. In the second edition of Anatomie des Kaninchen, Krause describes and figures the ventral ramus of the N. oculomotorius as giving off a branch to the M. retractor oculi. Concerning the X. abducens he says that it gives branches to the M. rectus oculi posticus (rectus lateralis). In many mammals, he says the M. retractor oculi is supplied, as in the rabbit, by the N. oculomotorius; in the cat and the calf, however, the N. abducens is the source of the nerve fibers for the same. 202 G. S. HOPKINS Bensley gives precisely the same distribution of these two nerves in the rabbit as does Krause. Reighard and Jennings saj' that in the cat the M. retractor ocidi receives its nerve supply from the N, oculomotorius and that the N. abducens is distributed to the M. rectus lateralis; there is no intimation whatever that the latter nerve gives any branches to the M. retractor oculi. Mivart, Wilder and Gage on the other hand describe the M. retractor oculi of the cat as supplied wholly by branches from the N. abducens. Concerning the innervation of this muscle, Brinton says the muscle which sweeps the broad nictitating membrane across the bird's eye and the funnel shaped or choanoid muscle (retractor oculi) which surrounds the optic nerve and eyeball of many mammalia are both supplied from this nerve (N. abducens), Wiedersheim also describes the N. abducens as supplying the lateral rectus, the retractor oculi and the muscular apparatus of the membrana nictitans in sauropsida, thus agreeing in all respects with Brinton's account of the nerve. From experimental evidence on the live horse and rabbit Foltz asserts that the N. oculomotorius supplies nothing to the M. retractor oculi but that the N. abducens alone supplies this muscle and the lateral rectus. In a note he further states "it is stated in the treaties of Veterinary Anatomy that this muscle (retractor oculi) in the domestic animals is animated chiefly by the common oculomotor." Chauveau has found recently and our experience confirms it, that this muscle is animated exclusively by the oculomotor externus (N. abducens). According to Owen "in lower Quadrumana a few fibers seem to be detached from the inner part of the origin of the recti to be inserted into the sclerotic nearer the entry of the optic nerve. This is the remnant of a stronger muscle which in other mammals, with few exceptions, surrounds the optic nerve, expanding funnelwise, as it approaches the back of the eyeball; it is called the choanoid, nmscle, or suspensor oculi, and is supplied by a branch of the sixth cerebral nerve." INNKHX'A TIO.N Ol MISCLK RKTItACTOK OlTM 203 Montane and Huurdcllc describe and figure the N. ocnlomotorins as su])])lyin^ all tlie muscles of the eye except the external rectus, the ])()st(M-i(>r i-ectus (retractor oculi) and tlie ^reat obli(iue. The N. ahdnceiis they stat(> su])l)lies tlie (>xternal rectus and the ])osterior rectus. Contirniatory evidence of the eiTor of tliose who described the N. t)culoni()tt)ruis as siipi)lying; branches to one or more portions of the ]\I. retractor oculi are found in the distribution of these two nerA'es in some of the reptiles. In a pa])er on the development of the musculature of the head and extremities of reptiles, Corninji; says that in tli(> lizard (Lacerta \-ivipera) the abducens nmscle mass which gives rise to the Mm, retractor oculi and rectus lateralis is supplied by the N. abducens. Precisely the same distribution of these two nerves is given by Hoffman for another lizard (Lacerta agilis) and for the turtle. In the Crocodile also, according to Fischer, the N. abducens is distributed as m the lizards. In an investigation on the development of the prootic head somites and eye muscles in Chel3'dra serpentina, Johnson found that the X. abducens supplies the Mm. rectus lateralis and retractor oculi. No portion of the latter nmscle is innervated by the N. oculomotorius. The writer's conviction of the incorrectness of the descriptions of the distribution of these two nerves, as given bj'" several authors, is based on repeated dissections of the nerves in the horse, ox, sheep, pig, dog, cat, and rabbit ; in the woodchuck and the badger the nerves were dissected but once. In most cases the nerves were traced their entire length, i.e., from their superficial origin from the bram to their respective muscles. All of the dissections were made with the greatest possible care under a binocular dissection microscope. In the pig the nerves were also traced microscopically in a 42 mm. embryo cut into sections of 20 microns. In all of these animals the dorsal branch of the N. oculomotorius was distributed to the Mm. rectus dorsalis and the levator palpebrae; the ventral branch was distributed to the Aim. rectus ventralis, rectus medialis and to the obUquus ventralis. In all cases one or more small branches were given off Fi / . Cut surface of supraorbital process. 2. Frontal sinus. 3. Cut surface of the zygomatic proc ess of the temporal. 4. Cut surface of the malar. 5. Palpebrae. 6. Gl. lacrimalis (somewhat reflected). 7. M. rectus dorsalis or superior. 8. M. levator palpebrae dorsalis or superior. 9. M. rectus lateralis. 10. M. rectus ventralis or inferior. 11. M. rectus medialis. 12. M. obliquus dorsalis or superior. IS. M. obliquus ventralis or inferior. H. IVI. retractor oculi or bulbi. (The greater part of the muscle has been removed; only the two extremities, 14 and 14' are shown). 15. Cut edge of sphenoid. 16. A. maxillaris interna. 17. A. ophthalmica. 18. A. temporalis profunda anterior. 19. A. supraorbitalis or frontalis. 20. Small artery to the mass of adipose tissue in the temporal ossa. 21. A. infraorbitalis. 22. A. orbitalis or malaris. 23. A. l)Ufcinatoria. u. 25. 29. 30. 31. 35. 36. 37. 39. 40. u. 42. u. 45. A. sphenopalatina. N. maxillaris, cut and one end turned aside. N. lacrimalis, cut and turned aside. N. supraorbitalis or frontalis. N. nasociliaris or palpebronasal. N. ethmoidalis. N. infratrochlearis. N. trochlearis. Sensory root of Ganglion ciiiare. and 34. X. oculomotorius, dorsal and ventral branches. Small branch from the N. oculomotorius to the M. levator palpebrae dorsalis. N. abducens. N orbitalis or zygomaticus (the peripheral portion has been removed). Ganglion ciliaris. Nn. ciliares. N. opticus N. sphenopalatinus. Ganglion sphenopalitinum showing many small nerves leaving it. N. palatinus anterior or major. N. palatinus posterior or minor. Cut edge of the periorbita or ocular slicath. 204 INNKItN ATION OF MISCLK KKTKACTOK OCLLI 205 to tlic (.'iliiiry j»;an}ili()n. In no cuse were there found the slif^litest indications of filain(Mits from eitlier tlio dorsal or vential bj'aiiches of tlie N, oculoinotorius to tlie M. retractor ociili as stated by so many. In all cases the X . abducens, figure 1, 36, supplied all portions of the M. retractor octdi. JMost of the statements regarding the form of the AI. i-etractor ocuh and its relation to the optic nerve are somewhat misleading. In some of the domestic animals as the horse, ox, sheep and pig this muscle is not readily divisible hito four distinct portions — dorsal, ventral, medial and lateral as it is in the doji. cat and rabbit, but forms a continuous sheet which surroiuids the posterior part of the eyeball and a part of the extra cranial portion of the optic nerve. The medial side of the optic nerve for a distance of one and one half centimeters from the apex of the orbit, in the horse, is entirely uncovered by this muscle all the fibers of which are attached to the lateral side of the optic nerve as shown in the figure {14') LITERATURE CITED Bexsley, B. a. lyiU Practical anatomy of the rabbit. P. Blakiston's Son and Company, Philadelphia. Br.\dley, O. C. 1897 Outlines of veterinarj- anatomy. London. Also W. R. Jenkins, Xew York. 1912 A guide to the dissection of the dog. London, Longmans, Green and Company. Brixton, Wm. 1847—19 C3'clopaedia of anatomy and physiology, vol. 4, part I, p. 622. Chauveau, a. 1857 Traite d' Anatomic Comparee Animaux Domestiques. Corning, H. K. 1900 Ueber die Entwicklung der Kopf und Extremitateu Muskultur bei Reptilien. Morpho. Jahrbuch, Bd. 28, pp. 28-104. Ellenberger u. Baum 1891 Anat. des Hundes. Berlin. 1908 Handbuch der Vergleichenden Anat. der Haustiere. Berlin. Fischer 1890 Bronn Klassen des Thier Reichs Yl, III, 16-42. Reptilien, p. 753. FoLTZ, J. C. E. 1862 Recherches d' Anat. et de Physiologic Experimentale sur les Voies Lacrj-malcs Jo. de la Physiologic, Tome V, pp. 226-247. Franck, L. 1894 Handbuch der Anat. der Haustiere. Stuttgart. GuRLT, E. F. 1873 Vcrgleichende Anat. der Saugethiere. Hoffmann, C. K. 1890 Bronn Klassen des Thier Reichs, VI, III, I, 15, Reptilien. 206 G. S. HOPKINS Hopkins, G. 8. 1913 Directions for the dissection and study of the cranial nerves and blood vessels of the horse. Published by the author. Johnson, C. E. 1912-13 The development of the prootic head somites and eye muscles in Chelydra serpentina. Am. Jour. Anat., vol. 14, pp. 119-186. Krause, W. 1881 Die Anat. des Kaninchens. Leipzig. Leisering. 1899 Atlas der Anat. des Pferdes und der ubrigen Haustiere. Leipzig. Martin, P. 1904 Lehrl)uch der Anat. der Haustiere. Stuttgart. MiVART, St. G. 1889 Lessons in elementary anatomy. Macmillan and Company, New York. ^PFadyean, J. 1902 The anatomy of the horse. W. R. Jenkins Company, New York. AIoNTANE AND BoTJRDELLE. 1913 Anatomic Regionale des Animaux Domestiques, vol. i, Paris. Owen, R. C. 1868 Comp. Anat. and Physiol, of Vertebrates, vol. 3, pp. 258259. Reighard AND Jennings. 1901 Anatomy of the cat. Heniy Holt and Company. Reuter, K. 1897 Ueber die Entwickelung der Augenmuskultur beim Schwein. Anat. Heft., Bd. 9, pp. 367-389. Share-Jones, J. T. 1906 The surgical anat. of tlie horse. London, part I, p. 154. SissoN, S. 1914 The anatomy of the domestic animals. Strangeway, T. 1909 Veterinary anatomy, 12th Am. ed. Strtjska, J. 1903 Lehrbuch der Anat. der Hausthiere. Wien u. Leipsig. Varaldi, L. 1839 Anatomia Veterinaria, vol. 2. Milano. Wiedersheim, R. 1902 Vergleichende Anat. der Wirbelthiere. 5th ed. Wilder and Gage. 1886 Anatomical technology as applied to the domestic cat. ZiMMERL. U. 1909 Tratte di Anatomia Veterinaria, vol. 3. Mihiuo. THE ANATOMY WITH J!:SPECLVL CONSIDERATIUX OF THE EMBRYOLOGICAL SIGNIFICANCE OF THE STRUCTURES OF A FULL-TERM FETUS AMORPHUS EBEN CAREV' De pa rime lit of Histology and Embryology, Creighlon Medical College, Oinalta, N^ebraska XINETEEX FIGURES IXTRODCUTION Although teratological Uterature is replete with descriptions of the various types of the fetus amorphus the specimen here studied is of sufficient interest to record, for no description, to which the author had access, was found that entirely coincides with this case. The outstanding characteristics of this interesting group of monsters are according to Ahlfeld ('04) the absence of a heart, normal bilateral symmetry and generally the brain. If the latter is present to any extent at all it is verA' rudimentary. The subcutaneous tissue is usually hypertrophied, cystic and oedematous. The cord possesses but one artery and one vein. A partial list of the authorities consulted is the following: Ahlfeld, Ballantyne, Beardsley, Bevill, Brodie, Charlton, Claudius, Embleton, Hall, Herholdt, Houston, Hicks and Baukart, Hirst and Piersol, Jacobi, J. Jackson, J. S. B. Jackson, Le Cat, Lusk, Mall, Marchand, ^Meckel, Rauber, Schatz, Schwalbe, Simpson, Tiedmann, Von Winckel, A'rolik, Willey and Windle. The specimen that is the subject of this paper was received by the author, upon taking charge of the department in 1914, from Dr. J. S. Foote, professor of histo-pathologj", Creighton Medical College. Dr. Foote came into possession of the monster in 1901 shortly after it was born. It was well preserved in Kaiserling's fluid. The co-twin was a normal female fetus. The period of gestation lasted the full nine calendar months. The 207 208 EBEN CAREY normal fetus was the first to be expelled at parturition. No other clinical data were recorded; the condition of the fetal membranes and maternal deciduae were not tabulated. Ten years ago Dr. Foote heard that the normal co-twin was a healthy, robust girl of five years of age. GENERAL APPEARANCES, WEIGHT AND DIMENSIONS A systematic description of each aspect of the fetus follows in the text. The general appearance of the monster was that of an irregularly rounded, potato-shaped, skin-covered mass with the external indications of the two lower limb buds; a well marked outgrowth for the right upper limb and a slight protuberance over the region of the left upper limb. The ventral surface presented a thoracic and a hypogastric elevation. Upon the cephalic slope of the latter the umbilical cord was located which contained but one artery and one vein. The parietal, occipital and mid-dorsal regions carried hair. There was no external indication of a neck. No skeleton would be suspected by palpation. Upon percussion a tympanic note was elicited in the left lower thoracic region. There proved to be a large cavity, {11, fig. 18.) in this area, outside the skeleton, in the subcutaneous tissue. The facial and genital aspects, the limb buds and right glandular area were interesting from the embryological standpoint. The specimen weighed 1130 grams after excellent preservation in Kaiserling's solution for fifteen years. The greatest length from the tip of the frontal region to the end of the rump was 19 cm. {4 to 16, fig. 2). The crown-rump measurement was 17 cm. (1 to 10, fig. 3). The greatest width through the upper limb buds was 13.5 cm. (7, fig. 4). Through the lower limb buds the width was 6.4 cm. The greatest thickness 8 cm. (5, fig. 3) was at the mid-ventral point on the line of greatest width. At the location of the external genitalia, the fetus amorphus was 4.5 cm. in thickness. On the ventral surface there were two elevations with an intervening furrow {6, fig. 3). The apex of the caudal elevation {8, fig. 3) was just below the umbilicus in the hypogastric region. The thickness of the embryo at this point was 6 cm. AXATO.MV OF A FULL-TKJl.M FKIM S AMOlill 1 1'S 2(M) s\si'iv\i Aiic i)i:s('ini'i'i<)\ <)!' 'I'lii-; r<)i't>i;i; \nn Riijhl nnlro-ldhrdi as/xcl {Ji(j. I ) The ('c|)haji(' ikuI of the fotiis is incliiuHl ohlicnicly doisud, so as to l)i-iii<!; more promiiuMitly iiilo xicw and lo sliow llic contiiuiit_\- and fclatioiiships of llic lower liiiil) l)uds. /J and 13, and external genitalia; .9, body of the clitoris, 10, ^lans clitoi-idis, 11, ostium urogenital, 15, right labia majora, of the ventrocaudal asj)e('t of the fetus, witli the rest of the ventral surface Fig. 1 (Al)out one-half natural size.) 1, left globular process; 2, left maxillary process; 3, oral pit; 4, groove at ventral junction of mandibular arches; 5, depression of right glandular area; 6, right upper extremity; 7, umbilical cord; 8, the left ventro-latero-dorsal fissure; 9, body of the clitoris; 10, glans clitoridis; 11, ostium uro-genital; 12, left lower limb bud; 13, right lower limb bud; 14, imperforate anus; 15. the right labia majora; ventral aspect. i^ 210 EBEN CAREY of the body. The embryological consideration of the structures of the external genitalia will be reserved until plate 2 is described. Directly cephalad from the genitals is seen the remnants of the umbilical cord 7, which contains only one umbilical artery instead of two as are normally found. In the right cephalo-lateral aspect is seen the marked protuberance of the right upper extremity 6. The slight protuberance, directly across the body on the left side, in the region of the left upper extremity is seen to a better advantage in figure 4, no. 7. The location of the imperforate anus is marked by ^4 The interesting fact to be considered in studying this aspect is the depression 5, of the right glandular area. No similar depression is seen on the left side of the body. This depression is a persistency of the depressed right glandular area, which appears in an embryo of about 25 cm., according to Pinkus ('10) and gradually deepens until it is well marked in a fetus of eight months. At this time the nipple is also supposed to be partly formed, but no sign of a nipple is seen in this specimen. The genetic significance of the structures of the facial aspect; 1, left globular process, 2, left maxillary process, 3, oral pit, 4) groove at ventral junction of mandibular arches, will be discussed when plate 1 is considered, figures 6 to 10. Left ventro-lateral aspect (fig. 2) This view of the fetus is presented in order to show the continuity of the rudimentary facial structures with the ventral aspect of the thorax and abdomen. The cephalic part of the fetus is inclined obliquely ventrad, in contra-distinction to the position as shown in figure 1, in which the inclination was obliquely dorsad. The most important fact clearly presented is the absence externally of the neck separating head and thorax. Dorsal aspect (fig. 4) The left frontal region 1, is prominently marked. The hair is characteristically arranged over this area in converging whorls. On the other hand the hairs over the crown or vertex 4, are ANATOMY OF A FT LL-TKKM FKTIJS AMORPHUS 211 arranged in tli\(M«i;in<2; wliorls, the direction of tlie hairs being toward the ri^lit, over tlie loft side of crown as can be made out from the i)liotograi)h. The hterature on the subject of the whorls of the hair, together with a complete bibliogi'aphy, is presented by Pinkiis ('10). The external indication of the location of tlio siipciioi sagittal siitm'o is seen as a sulcus, S. The Fig. 2 (About one-half the natural size.) 1, right nasal pit;^, right globular process; 3, sulcus between right and left globular processes; 4, ventral indication of prosencephalon; 5, left globular process; 6, left nasal pit; 7, palpebral folds of left eye; 8, tongue in oral pit; 9, left maxillary process; 10, ventral junction of mandibular processes; 11, depression for right glandular area; 12, right upper extremity; 13, left ventro-latero-dors'al fissure; 14, umbilicus; 15. left lower limb bud; 16. external genitalia, left ventro-lateral aspect. 212 EBEN CAREY right frontal and right crown regions are not so clearly differentiated morphologically in the fetus as the left regions. However, the converging and diverging whorls of the frontal and crown areas respectively, are seen, although they do not show so clearly on the right side as upon the left. The normal and abnormal facial aspects {figs. 6, 7, 8, 9, and 10) The abnormal facial structures of the monster are exceedingly^ interesting, due to the fact that they retain the external facial Fig. 3 (Aljout one-half natural size.) 1, frontal area; i, palpebral folds of left eye; 3, occipital area; 4, protuberance over left upper extremity; 5, cephalothoracic elevation; 6, ventral depression; 7, left ventro-latero-dorsal fissure; 8. ventre caudal elevation; 9, left lower limb bud; 10, imi^erforate anus; left lateral aspect. \\\I'().MV OF A ITLL-TKUM FKTCS A.M( (Itl'll IS '2\'A characterislio, h()W(>\(M' considorably distortod. of an embryo of the second montli. In tlie li^ht of normal facial develop.nient, as first satisfactorily studied by His ('80J, and later by Keibel and Klze ('08), Hertwig ('06), Mall ('91 and '93), Retzius ('04) and othei-s we are enabled to interpret the facts here presented. Although considerably distorted, the two large globular processes 5 and 10, developed from the lateral aspects of tlie median Fig. 4 (About one-half natural size. ) ;, left frontal region ; 2, palpebral folds of right eye; 3, superior sagittal sulcus; 4, occipital region; 5, deep furrow dorsomesiad of right shoulder; 6, right upper extremity; 7, protuberance of left upper extremity; 8, dorsal line of the body; 9, ventro-latero-dorsal fissure; 10, right lower limb bud; //, imperforate anus; dorsal aspect. THE ANATOMICAL RECORD, VOL. 1!, KO. 5 214 EBEN CAREY Fig. 5 (About one-half natural size.) 1, apex of right maxillary process; 2, right upper extremity; 3, depression of right glandular area; 4, ventro-cephalic elevation; 5, tip of the right upper extremity; 6, umbilical cord; 7, ventral depression; 8, ventro-caudal elevation; 9, button-shaped right lower limb bud; 10, location of external genitalia; right lateral aspect. Figs. 6-10 The indication numbers of figures 6, 7 and 10 mark out similar structures. 1, crown; 2, frontal area; 3, sulcus between the right and left globular processes; 4, palpebral folds of right eye; 5, right globular processes; 6, nasolacrimal furrow; 7, nasal pit; 8, palpebral folds for left eye; 9, left nasal pit;i<?, left globular processes; 11, left maxillary processes; 12, left mandibular processes; 13, ventral furrow intervening between mandibular arches; 14, right maxillary process; 7', in figures 6 and 7, lateral nasal processes. Figures 6, 7, 8 and 9. Development of the face of the human embryo by His taken from Heisler: figure 6, embryo of about twenty-nine days. The nasal frontal plate differentiating into processes globulares, towards which the maxillary processes of first visceral arch are extending. Figure 7 of about 34 days: the globular, lateral, frontal, and maxillary processes are in apposition; the primitive opening is now better defined. Figure 8 embryo of about the eighth week: immediate boundaries of mouth are more definite and the nasal orifices are partly formed, external ear appearing. Figure 9, embryo at the end of the second month. ANATOMY OK A FULL-TEIiM FETUS AMOIil'HUS 215 216 EBEN CAREY nasal process are seen separated by a well marked fissure S. The right and left nasal pits 7 and 9 respectively are shown as cicatricial depressions. The lateral nasal processes are not distinct on either side. However, their location is differentiated from the maxillary processes by the persistent naso-optic fissures best marked on the right side by 6. The palpebral folds of the right and left eye, 4 and 8 respectively, are clearly made out. The right is better developed and shows a fissure as does also the left, between upper and lower eyelids. At the inner angle of the right orbit a groove or furrow continuous with the fissure between the two primitive eyelids is well marked, not so clearly seen on the left side, which courses to the lateral aspect of the anterior nares and is no doubt the persistent naso-optic fissure. The fissure is normally present as seen in figures 6 and 7, No. 6, but later becomes obliterated as seen in figure 9. On the left side the naso-optic fissure is not so well marked. The right and left maxillary processes, 14 and 11 respectively, are seen as prominent lateral bulgings of the cheeks. The external indication of the left mandibular arch 12, is apparent, separated by a narrow shallow furrow, 13, from the external indication of the right mandibular arch. This groove 13, leads cephalad to the distorted tongue, in the oral pit, shown darkly colored and protuberant in the photograph. The primitive oral cavity is pentagonal in outline, bounded cephalad by the left enlarged globular process, caudad by the two indistinct external indications of the mandibular arches, and laterad by the external indications of the right and left maxillary processes. The facial aspect thus presents a double hare-lip due to lack of complete fusion of the maxillary and globular processes of their respective sides and the complete split-nose 5, due to lack of union of the right and left globular processes. No bony nor cartilaginous structures were found in the mesoderm immediately underlying the superficial facial structures considered in this topographical description. No ears were present. ANATOMY OF A KUr.L-TKIlM FETUS AMORPHUS 217 The nonnal and abnorfnal ijeniUd aspects {figs. 11, 12, 13, 14 and 15) The structures of the genitalia of the monster are interesting, for to a certain degree they retain the characteristics of an enibiyo, however, considerably enlarged, at the beginning of the third month. At this time the indifferent phallus, by the direction which it takes, shows a transitory sexual differentiation. According to Herzog ('04), the phallus in the female bends downwards and in the male it is perpendicular to the long axis of the body. By now referring to figure 15, we see an abnormally enlarged phallus 3, which has practically differentiated into the clitoris. The female phallus at the beginning of the third month may even be larger than that of the male; the downward direction as seen is diagnostic of the female embryo. Surmounting the phallus is the glans 4> at the base of which is an encircling fold, the praeputium. The urogenital sinus 6, is seen patent as a groove; the genital folds which bound this groove have receded inwards due to the overgrowth of the genital tubercles or swelling which now form the labia majora, 5. The urogenital opening distally, towards the glans clitoridis, on the anal slope of the phallus has closed and has formed the urethral gi'oove. Proximally, towards the anus, the urogenital opening 6, remains patent, which is also diagnostic of the female embryo. INTERNAL ANATOMY Subcutaneous and muscular tissues The fetus was opened by a median incision running to the left of the umbilicus. It was then observed that the skin was firmly adherent to the underlying dense oedematous connective tissue and fat. This tissue was less fibrous as the bony structures were approached. Frozen sections of the tissue contiguous to the skeleton were stained with hematoxylin and eosin. Isolated fibers of voluntary muscle were found intermingled with the connective tissue. There were no well defined groups of Figs. 11-15 (Fig.s. 11, 12, IS, uiid 14. /, phallus; 2, glans elitoridis; 3, ostium iirogenital; 4, labia majora; 5, anus; 6, coccygeal eminence; 7, labia minora. Fig. 15 1, ventral surface of fetus amorphus; 2, umbilical cord; S, body of the I)hallus; 4, glans of elitoridis; 5, left labia majora; 6, urogenital sinus; 7, right lower limb bud; 8. dorsal aspect of right upper limb; 9, imperforate anus. 218 A.\\rt>MV OF A Kl'LL-TKliM FETUS AMOItl'UUS 219 nuistnilatuii'. In tlic irrosularly arranged muscular fibers were seen various sized vacuoles, Avhieh wheii stained in an alkaline alcoholic solution of scarlet red fat stain, which is Bell's modification of the Herheimer method, proved to be fat. Counter staining was made with Delafield's hematoxylin and the sections were mounted in glycerin. This method ])ro\-ed an excellent one for the detection of fat within the muscular fibers. The fat vacuoles were found in the angles of Cohnheim's areas and in the spaces between the in(li\idual sarcostyles. Blood vascular system (fig. 16) The chief features of the blood vascular system of the fetus amorphus are the absence of a heart and, according to Ahlfeld ('80 and '82), the reversal of the flow of blood. In the diagram the arrows indicate the reversed flow of blood according to Ahlfeld's theory. However this idea is refuted by Breus ('82). The blood enters the body of the fetus amorphus through its single umbilical artery 31, figure 16. As the latter turns to become continuous with the dorsal aorta, there are two branches given off, 35 and 36, the left and right common iliacs, respectively. The single umbilical artery is seen to belong to the right side. There is no left umbilical artery present. The right common ihac 36, branches into the internal and external iliac arteries, 37 and 38 respectively. The internal iliac artery ends as shown diagrammatically in figure 16 in a dilatation representing the capillaries of the pelvic viscera. This vessel evidenth" carries the blood supply to the viscera as no left internal iliac vessel is present. The right external iliac is also shown diagrammatically as ending in a dilatation representing the capillar^^ system of the right caudal limb bud. The left common iliac 35, pursues a straight course to the left caudal limb bud where it breaks up into a capillar^' network. The next vessel given ofT as we follow the blood current up through the aorta is the right renal artery 30, next the left renal 29, and lastly the hepatic artery 26, which comes off on the ventral surface of the aorta. At the same level, on the 220 EBEN CAREY AXATOMV OF A I T LI.- TKK.M KKTIS AMOItl'MUS 221 left lattMal aspect, to the oii<i;iii of tlio hopalic art(^ry, the superior luoseutci ic artery .<!7, is foiiiul. Tliero was no vessel comparable to the inferior mesenteric. The coeliac axis is represented by the single trunk of the hepatic artery. The superior mesenteric artery is a good sized trunk and evidently supplied the regions of the intestine that are normally su])]^lied by the coeliac axis and tlie inferior mesenteric arteries as well as its own region. At S, the aorta is seen to bifurcate. The right trunk is very much the larger of the two. The branches given off from these two trunks are comparable on each side. The first branch on the left side is the common carotid artery 3, which bifurcates into the external 1, and internal 2, carotid arteries. Caudad the intern.al mammary ai'tery is given off, 7, from the subclavian, 6'. From the dorsal aorta the paired intercostal and lumbar arteries which are not represented in the diagrammatic reconstruction, are given off. The blood is returned from the left and right sides of the head and neck and upper extremities through the left and right superior vena cava 15 and 16, respectively. The left external 4 and internal -5, jugulars unite to form a common trunk which in turn with the left subclavian vein 12, unite to form the left superior vena cava 15. At the junction of the two veins menFig. 16 1, external carotid artery (left); 2, internal carotid artery (left); S, common carotid artery (left); 4, left external jugular vein; 5, left internal jugular vein; 6, left subclavian arterj^; 7, left internal mammary artery; 8, dorsal aorta; 9, vertebral vein; 10, point of fusion of left and right azj^gos veins; U. trunk of left common jugular vein; 12, left subclavian vein; 13, left internal mammary vein; 14, common trunk of azygos veins; 15, left superior vena cava; 16, right superior vena cava; 17, common trunk of inferior vena cava; 18, left azygos vein; 19, left inferior vena cava; 20, right inferior vena cava; 21, thoracic aorta; 22, umbilical vein; 23, an anastomotic vein between left and right inferior venae cavae; 24, diagonal branch of vunbilical vein draining the intestinal pelvic, and left lower limb buds; 25, the hepatic vein; 26, the hepatic artery; 27, superior mesenteric artery; 28, left renal vein; 29, left renal artery; 30, right renal artery; 31, umbilical artery; 32, superior mesenteric vein; 33, vein draining pelvic region into right inferior vena cava; 34, continuation of femoral vein with the right inferior vena cava; 35, left common iliac artery; 38, right common iliac artery; 37, internal iliac artery; 38, external iliac arterj^; 39, vein draining pelvic area into diagonal branch of umbilical vein; 40, vein draining left lower limb bud into diagonal branch of umbilical vein; 4^, left inferior vena cava. 222 EBEN CAREY tioned above to form the left superior vena cava, the left internal mammary vein 13, joins the left subclavian. The branches forming the trunk of the right superior vena cava are similar to those forming the left superior vena cava but they are not labeled in the diagram. The left vertebral vein is represented by 9; the right is not labeled. The regions drained by the left superior intercostal vein and the azygos system of veins are represented hj 18. The right side is not labeled but is similar to the left. The vessels of the two sides after receiving the left and right vertebral veins form a common trunk 14, which- flows into the arch of the umbilical vein. There are no pulmonary veins as well as no pulmonary arteries. The blood is returned from the pelvic viscera through a vessel 33, which empties into the right inferior vena cava and a vessel 39, which empties into a diagonal branch 24, which in turn pours its blood into the umbilical vein shortly before it enters the umbilical ring. The blood is returned from the left caudal limb bud by two veins, one 40, which goes to form with the left pelvic visceral vein 39, the diagonal vein 24, and one 41, which is continuous with the left inferior vena cava. From the right caudal limb bud there is but one vein 34, which joins the vein 33, draining the pelvic viscera. These two latter veins unite to form the right inferior vena cava and no doubt represent the right and left common iliac veins. Upon the left side the left inferior vena cava does not bifurcate into right and left common iliac veins as is found on the right side. The right inferior vena cava receives the right renal and hepatic ^'eins, 25. The left inferior vena cava receives the left renal vein 28. The two inferior venae cavae 19 and 20 unite to form a single trunk which empties into the dorsal aspect of the arch of the umbilical vein. Before these two veins unite there is an anastomotic branch 23, coursing cephalad from left to right. The blood draining the intestines empties through the superior mesenteric vein 32, into the diagonal vein 24, which in turn pours its blood into the common vein of exit, the umbilical vein, 22. AXATOMV OK A Kl'I.L- TKICM FKTl'S AMOItl'UrS 223 Tlic al)iu)iiii;il rcxcrsal of llic direction oi' ihc hlood stream, through a sini^lc uiiil)ili('al artery, with its siil)se(iu<Mit aeccntiiatiou of ('(M'taiii arteries and degeneration of others, increases tlie difhcuhy of interpreting the significance of the vessels dissected. The arch of the aorta is obhterated and at its location we find a bifurcation, with the right integer which supplies mainly the right upper limb, consideral)ly the larger of the two. The left 15, and right 16, superior venae ca\'ae are the persistent left and right anteri(^r cardinal veins. Instead of the two posterior cardinal veins joining their respective anterior cardinal veins, they join each other to form a common trunk H, which empty into the umbilical venous arch, and is no doubt a persistency of the sinus venosus of the embryonic heart. The inferior vena cava is double 19 and 20, up to within one inch of its termination, in the umbilical venous arch where its two components of the right and left sides fuse to form a common trunk, 17. These tw^o posterior venae cavae are undoubtedly the persistent remains of the two subcardinal veins. There is no portal system comparable to that found in the normal adult. The small rudimentary liver possesses but one arterj' and one vein. The diagonal vein 2J^, draining the blood from the intestines, pelvic \dscera and left limb bud is evidently the remains of the portal system. The blood through this vein empties directly into the umbilical vein. Respiratory system {fig. 17) The lung tissue was completelj^ degenerated. In the thorax there was found a putty-like mass of degenerated tissue through which the oedematous connective tissue was growing. Broken off portions of the cartilaginous bronchi were found throughout the above mass. The small bronchi w^hich contain portions of cartilage and those in the stage of pre-cartilage, proved to be very^ resistant. The trachea 3, is seen tapering to its bifurcation into the right and left bronchi at 5. There w^ere rudimentary respiratory^ passages wdth anterior and posterior nares leading into the dorsal aspect of the deformed mouth. The lar\Tix 1, is considerably dilated as seen in figure 17. 224 EBEN CAREY Digestive system (fig. 17) The alimentary tract began at the mouth, and at the caudal part of the pharynx dorsad to the larynx the esophagus 2, figure 17, began. A small dilatation 6, is detected in the region of the stomach. A duodeno-pyloric flexure is made and at the base of the umbilical ring 7, and extending into the umbilical cord, a few coils of small intestine 9, were unraveled. The small cecal evagination 8, was found at the base of the umbilical ring. The large intestine had already made the primary twist across the small intestine. From the cecum the large intestine 13, is traced to its termination in the rectum 25. The descending and part of the transverse colons are marked out at this stage of development. The relationships of the intestinal coils to the umbilical vein 10 and umbilical artery 11, and urachus 12, at the base of the umbilical ring, are also shown in figure 17. There was no pancreas present. The common bile duct was absent. The rectal cul-de-sac is connected to the utero-vaginal tube by a fissure-like opening 22. Uro-genital system (fig. 17) The urinary apparatus is separated from the genital structures as shown in the figure. Both kidneys 16, are tri-lobed, the left being slightly larger than the right one. A ureter 17, leads from each to the bladder, 21. From the bladder leading to the uro-genital sinus is the urethra 23. From the tip of the bladder there is a patent tube, the urachus 12, which leads up to and out through the umbilical cord. Fig. 17 1, The larynx; 2, the oesophagus; 3, trachea; 4. broken off bronchi in thorax; 5, bifurcation of trachea into the right and left bronchus; 6, spindle shaped stomach; 7, umbilical ring; 8, cecum; 9, loops of small intestine; 10, umbilical vein; 11, umbilical artery; 12, urachus; 13, descending colon; H, left renal vein; 15, left renal artery; 16, tri-lobed metanephros; 17, left ureter; 18, fimbriated end of left Fallopian tube; 19, left Fallopian tube; 20, dichotomus division of the uterus into the two Fallopian tubes; 21, bladder; 22, opening between rectal cul-de-sac and the utero-vaginal tube; 23, urethra; 2Ji., vaginal portion of ureto-vaginal tube; 25, distal end of rectal cul-de-sac. ANATOAIV OF a KILI -TI.'i;\i i 225 FIC17 226 EBEN CAREY The genital system is composed of a iitero-vaginal tube W to 24, continuous with the two oviducts which branch immediately in a Y-shaped manner. The latter end in a fimbriated dilation 18, the left sHghtly larger than the right. There were no ovaries. The alimentary tract is connected with the uterovaginal tube at the location marked 22, which represents the persistent connection between the alimentary tract and the urogenital systems in the cloaca. Nervous system There was an entire absence of the brain and cranial nerves. The upper part of the cord was absent as far as the seventh cervical nerve. From this point to the end of the lumbar vertebra the cord was present. There was a rudimentary cauda aquina extending from the caudal end of the cord. There were irregular net works of nerves in the regions comparable to the brachial and lumbo-sacral plexuses. Skeletal system From the tip of the frontal bone, the highest point on the skull, to the tip of the fourth sacral vertebra the bony skeleton measured 14.5 cm. (2-12, fig. 19) From the tip of the left shoulder girdle to the tip of the rudimentary right ulna along indication line 9, figure 18, the width was 8.5 cm. A comparison of the intact fetus amorphus can easily be made with its skeleton when it is remembered that before the dissection the monster measured 19 cm. greatest length and 13.5 cm. the greatest width. At least 2.5 to 3 cm. of dense connective tissue and fat were interposed between the tip of any bony structures and the outer skin which precluded the palpation of the underlying skeleton. The skeleton was composed of a deformed cranium which was composed of, a fused occipital, sphenoid, ethmoid, temporals, and frontals. The two parietals were respectively distinct Many of the bones of the face and those of the bony vault of the cranium were fused into a distorted mass and could h(^. made Fig. IS Skiagraph No. 1. /. right parietal bone; 2. riglit temporal bone; 3, occipital bone; 4, atlas bone; 5, spine of .scapula; 6, glenoid fossa; 7, right clavicle; 8, right humerus; 9, left scapula; 10, pro.ximal end of ulna; 11, left subcutaneous cavity; 12, first lumbar vertebra; 13, liver tissue; H, first sacral vertebra; 15, center of ossification of right ilium; 16, fourth sacral vertebra; Ic, first cervical vertebra; Id, first dorsal vertebra; IL, first lumbar vertebra; Is, first sacral vertebra. 227 FIG 19 Fig. 19 Skiagraph No. 2 /, fused mandible and temporal bones; 2, frontal bone; 3, right parietal bone; 4, anterior fontanelle; 5, left parietal bone; 6, occipital bone; 7, styloid process of left temporal bone; 8, left clavicle; 9, liver tissue; 10, first lumbar vertebra; //, first sacral vertebra; 12, center of ossification of left ilium. 228 ANATD.MV OF A FULL-TERM FETUS AMOHIMIUS 229 Dill only with diHiciilty und then only by location and vague siniihiritios to the normal bones. The nasal, lacrimal and vomers were completely fused and these hi turn were fused to the midpohit of the mandible by the plowshare of the vomer. The mandible was completely continuous by bony union with the temporal bones on each side of the cranium. The paired maxillae, zygomatic and palate bones, six in all, were entirely absent. A rudimentary hyoid bone was present. The vertebral column presented complete spina Ijitida from the atlas to the fourth sacral vertebra 16, figure 18. The laminae were widely separated which deformity is well seen in figure 18 in the dorsal view of the skeleton. In the cervical, most of the dorsal, lumbar and sacral regions the laminae were fused to each other. The atlas was completely fused with the occipital bone at the base of the skull. As can readily be seen by reference to figure 18, the vertebrae presented right lateral scoliosis with the vertex of the curvature at the ninth dorsal vertebra. There is also a slight dorsal kyphosis, sharply angular, from the second to the sixth dorsal vertebrae, somewhat obscured by the opacity in figure 19. The last sacral and coccygeal vertebrae were absent. Upon the left side of the thorax the first nine ribs were fairly well formed with the remaining three merely attenuated cartilaginous rods. Upon the right side the first seven ribs were fairly well formed and the remaining five represented only by fine rods of cartilage. Upon the ventral aspect the thorax presented a complete cleft sternum. The cartilaginous ribs over the cardiac area presented a distinct bulging. Upon the left side the upper extremity presented a compound clavicle and scapula the remaining bony structures being absent. Upon the right side in addition to the distinct clavicle and scapula (which are not compound as on the left side) there was a well formed humerus 3 cm. long and the proximal end of the ulna 0.5 cm. long. No radius nor other bones of the upper extremity were present. Both clavicles were completely ossified. The left scapula presented ossification along the acromial process, w^here it is continuous with the left clavicle. The spine THE ANATOMICAL RECORD, VOL. 11, NO. 5 230 EBEN CAREY and supra-spinous fossa were also completely ossified. The corocoid process was not present. The right scapula was completely cartilaginous except the acromial process and contiguous parts of the spine and corocoid process. The pelvic girdle presented centers of ossification for the ilium, ischium and pubis. The sacro-iliac synchondrosis was cartilaginous on its iliac and sacral aspects. Both pubic elements of the symphysis were cartilaginous. Besides the pelvic girdle there were no other bones of the lower extremity present. In conclusion, I wish to express my gratitude to Dr. J. S. Foote for the fetus amorphus recorded in this article; to Prof. J. C. Heisler and the Saunders Publishing Company for their permission to reproduce figures 6, 7, 8, 9, 11, 12, 13 and 14 of this article from Heisler's Embryology; and to Dr. A. F. Tyler, professor of Roentgenography, Creighton Medical College, for the skiagraphs herein reproduced. ANATOMY OF A rULL-TEIlM FETUS AMORPHUS 231 liti;i{ai'i;rk cited Ahlkkm) 1SS0-S2 Die Missil)il(liiMnt'n dcs Mcnschcii. Leipzig, Parts I and II. Hallantvnk 18!)4-93 Journal of Anatomy and Physiology, vol. 29, p. 466. Citod by Winillo (Report on Recent TeratoloKieal Literature.) Journal of Teratologia. no. 1. 1904 Antenatal Pathology, 2 vols. Edinljurgh. Hkaudsley 1858-59 An acephalous fetus. Boston Medical and Surgical Journal, vol. 59, p. 39. Bevill 1SS5 A case of acephalous, with spina bifida. St. Louis Cour. Med. vol. 13, p. 2S0. Breus 18S2 Zur Lehre von dem Acardiacus. Med. Jahrb., VVien, pp. 57-72. Buddie 1819 Account of the dissection of a human fetus, in which the circulation of the blood was carried on without the heart. Phil. Tr. London, 99, p. 161-168. Charlton 1910 Cited by Adami. Principles of Pathology. Plate VII, vol.1 Claudius 1859 Die Entwicklung der herzlosen Missgeburten. Kiel. E.MHLETf:N 1865 Case of a human monstrosity, with sketch; absence of heart and lungs. Edinburgh Medical and Surgical Journal, vol. 56, p. 423-27. Hall 1843 On the circulation in the acardiac foetus. London and Edinburgh Monthl}-, Journal of Medical Science, vol. 3, p. 541-7. Herzog 1904 Beitrage zur Entwicklungsgeschichte und Histologic der Mannlichen Harnrohre. Arch. f. Mikr. Anat. u. Entw., vol. 63. Herholdt 1889 Reference Handbook of the Medical Sciences, vol. 7. Cited by Fischer in his article on Teratology. HorsTON 1S36 An account of a human fetus without brain, heart, or lungs; with observations on the nature and cause of the circulation in such monsters. Dublin Journal of Medical Sciences, vol. 10, pp. 204-20. Hertwig 1906 Handbuch, vol. 1, chapter 6. Hicks axd Baukart 1866-67 Dissection of acephalous monsters without head, heart, lungs or liver. Guy's Hosp. reports. Hirst and Piersol 1891 Human monsters. 4 vols. Philadelphia. Hi.s 1880- 85 Anatomic menschlicher Embryonen. Leipzig. 1904. Die Entwicklung des menschlichen Gehirns Leipzig, Jacobi 1873 Report of Committee on case of acardiac monsters. American Journal of Obs., vol. 6, p. 631. Jackson, J. 1843 On circulation in acardiac fetuses. London Medical Gazette, vol. 37, p. 467. Jackson, J. S. B. 1837-8 Case of monstrosity in which the brain, heart, lungs, stomach, liver, spleen, pancreas and right kidney were wanting. American Journal of Medical Sciences, Philadelphia, vol. 21, p. 36268. Keibel and Elze 1908 Normentafeln zur Entwicklungsgeschichte des Menschen. Jena. 232 EBEN CAREY Le Cat 1767 A monstrous human fetus having neither head, heart, lungs, stomach, spleen, pancreas, liver nor kidneys. Translated from the French by M. Underwood. Philosophical Transactions, London, vol. 57, p. 1-20. LusK 1874 Case of acardia. New York Medical Journal, vol. 19, p. 176-79. Mall 1891 A human embryo twenty-six days old. Jour. Morph., vol. 5. 1893 A human embryo of the second week. Anat. Anzeiger, vol. 8. 1900 Pathology of early human embryos. Johns Hopkins Hospital Reports, vol. 9, p. 57. 1908 A study of the causes underlying the origin of human monsters. Jour. Morph., vol. 19, No. 1. Marchand 1897 Missbildungen. Eulenburg's Real Encyclopedia, vol. ■>. Third edition. Meckel 1902-1905 Handbuch d. pathol. Auatomie. Vols. 15-19. PiNKus 1910 The development of the integument. Keil)el and Mall, Human Embryology, vol. 1. Rauber 1878 Die Theorien der excessiven Monstra. Arch. f. path. Anat., Berlin, vol. 78, p. 551-94, and vol. 74, p. 66-125. Retzius 1904 Biologische Untersuchungen, neue Folge, vol. 11. ScHAT? 1901 Archiv f. Gynak. ScHWALBE 1906-1907 Die Morphologic d. Missbildungen des Menschen und der Thiere. Jena, Pt. I, 1906. Pt. II, 1907. Simpson 1874-77 Description of an acardiac fetus. Transactions of Edinburgh Obstetrical Society, vol. 4, p. 384-89. TiEDMANN 1828 Observations sur I'etat du cervau et des rerfs dans les monsters. J. cumpl. du diet. d. sc. med. Paris, vol. 31, p. 142. Von Winckel 1904 Ilebei: die menschl. Missbildungen. Samml. klin. Vortrage, Leipzig. Vrolik, W. 1811 Beschrij ving eeniger merkwaardige mi.sgeboorten. Verhandel v. h. Genootsch t. Bevord. Cited by Fischer, Reference Handbook of Medical Science. Willey 1869-70 Birth of an acephalous. California Medial Gazette. San Francisco, vol. 2, p. 20. WiNDLE 1890-1910 Report on recent teratological literature. Journal of .\natomv and Phvsiologv. THE THORACIC DUCT IN THE RABBIT RAY HENRY KISTLER From titc Diuision of Anatomy of the Stanford Medical School SIX FIGURES Notwithstanding the above title I intend to describe not only the thoracic duct, its beginning and termination, but also the large lymphatic vessels in the abdominal cavity which are its real beginning. In my description I will also take into account the main lymph channels from the brim of the pelvis. The injections were made on a series of rabbits, the exact species of which because of cross breeding I was unable to determine. They were bought at random from rabbit farms where they had been bred promiscuously. I used 26 adult animals about two years of age, sixteen males and ten females. Immediately after killing by illuminating gas the animals were taken from the jar and placed on an animal board. The hind and front foot pads were immediately injected by means of an ordinary hypodermic syringe with a medium-sized needle. I used India ink for the injection mass and found that by inserting the needle just under the skin in the pad of the foot, this fluid would immediately travel on up the leg in the lymphatic vessels. Sometimes the ink would go only as far as the nodes in the popliteal fossa, while in other animals it would go clear up into the abdominal vessels, and even into the thoracic duct. Out of the 26 animals I succeeded in getting the injection fluid to go from the pad of the hind foot clear up the thoracic duct in five cases. In all the rest it stopped at the popliteal nodes. In three of the five cases I found that there was an anastomosis of the vessels around the popliteal node but in the other two cases it looked as though the injection had gone right through the node. The inner part of the node was usually darker than 233 234 RAY HENRY KISTLER the periphery and in one instance just one portion of the node was blackened. Whether this can be explained by a difference in the arrangement of the lymphatic vessels or is due to the effect of the strong pressure created by the syringe I cannot say. However, my classmate, Clattenberg, found that young kittens and guinea pigs, in same way, always show the injection clear up the thoracic duct. It occurred to me that perhaps I did not have enough pressure to force the injection on through. I therefore used a larger syringe, in a few cases, and injected as much as I could into the subcutaneous tissue, but the results were as before for I did not get the injection to go beyond the popliteal nodes in any of these cases. After injecting the pad of all four feet in this manner I tied the animal to the board and made an incision through the skin from the mandible to the xiphoid. I then dissected out the two external jugular veins, which are very large in the rabbit and ligatured them close to the clavicle, and opened the thorax and tied off the two superior venae cavae. In this way I prevented the injection mass from either going on into the heart or cranially through the large veins. Next I skinned the hind leg and dissected out the pophteal nodes. There may be one or two of these and they are of considerable size. Upon injecting these nodes the ink usually went through into the thoracic duct very easily if the leg was repeatedly flexed and extended. I also always opened the abdomen and injected the large group of mesenteric nodes in order to distend the duct and vessels so that they stood out distinctly. Injections into the pads of the anterior extremity usually stopped in the axillary nodes. I succeeded in injecting these nodes only a few times and not" as easily as the popliteal nodes. In almost every case about 14 or 15 cc. of ink was used per rabbit. Of this 4 cc. was injected from the hind and 3 cc. from the front pads, 4 cc. from the popliteal nodes, 1 to 2 cc. from the axillary nodes and 3 cc. from the mesenteric nodes. Since I was not able to dissect the vessels very well in the fresh specimen the whole animals were preserved in a dilute solution of formalin. After this I could THORACIC DUCT IN TIIK UAHIUT 235 finisli my dissection easily especiiilly that of the duct at its terinin.-itioti into tlio fj;reiit veins of the neck. The injection went through the nodes at the bifurcation of the abdominal aortu to the vessels above. This raises the question as to whetlier an injection can go through a node. This Baum '11 answered in the affirmative. Baum also always found connections of the lymphatics with the peripheral veins but this was never noticed in my injections. In my series of animals I found that the thoracic duct and abdominal vessels were very constant and similar in distribution and appearance for only minor variations were observed. Hence a single scheme can represent the arrangement in all the specimens injected with the exception of a few variations which will be spoken of later. From 3 to 5 lymph nodes were always found at the brim of the pelvis or better at the bifurcation of the large abdominal blood vessels. The more usual number was three nodes. One lay on each common iliac and the third ventral to the bifurcation of the abdominal aorta. How^ever, the position of these nodes varied somewhat for the third or central one was sometimes found in the pelvis just caudal to the bifurcation of the aorta and the iliac nodes also sometimes lay more caudal. The nodes lay in the subperitoneal tissue always ventral to the vessels and could be moved about very easily. They varied in size from 0.5 cm. to 1.2 cm. in diameter and were connected together by vessels thus forming , kind of a plexus. From this group of nodes a group of 3 to 7 lymph vessels started cranially but in every case small branches of these vessels would connect with the adjoining vessels or a more dorsal vessel so as to form a sort of plexus. Nevertheless I could almost always distinguish and trace a certain number of distinct and parallel Vessels w^hich ran cranially as far as the renal vein. This group of hnnph vessels was always mainly ventral to but partly surrounded the great abdominal blood vessels. The number and exact position of these lumbar lymph vessels and their relation to the vein and artery I have tabulated in the accompanying table. The average number of lumbar 236 RAY HENRY KISTLER UWABJO -qns puB Jt;i -nStit (BUJO) -xa JO uoijounf •[BUJajni pOB IEUJ84 -xa JO uopounf + + + + + + + + + + + + ^ T3 m o o t-. ^ > o o -tJ t»-l ^ b(l C a o a a cS j- O =3 .2 O) 00 O 03 -S ^ to o3 "^ 03 O 00 ^ -p ■> -s Q ' ^ o 03 o3 S 03 _ H O -C c — I — ' ^ -t-' ^ O .2 -r; -r; M O .Jl, o a n y a> o cc n j3 > o 3 o S c3 O -3 o o > 03 ^ -(-> 03 Se? I I + I + I + O ^ f ° « S ■* 5 HU S a ^ s S H Q K ■Bnjaj -Sra IBIUBJQ (M (N r-H 0% I B S J O Q i-H <M <M o^ jBj^aaj\ ^CM OJ--H CO(M i-H (M uiaA O^ [■Bl'i'uO\ Oi y-i !M r-l IM ^ CQ snxajd jbu -toiopqB uj siAaad JO wiHa iv 8aaoM JO aaoKnN avwiNV JO HaawQN ■*■* lO-rt^ l>.iO -^ -^co CO CO CO-* coco CO -^IM TFIOKACIC DUCT IN THE RABBIT 237 + + + + + ++ + ++ + + ++ 2 Q >-> + + ++ + a o ■ Q Q Q h:; ^ j CO £ -D -< ^ +-> (>J T-l H-! f-H b -(-> CO •— t Cl ■4-3 CQ ID a 73 ^ > fi c ^ C3 rt r" +J ^ «J CO --I CQ 4 ++ III I ++ 01 !M ^^ CM CM '-I oi T-l ^^ (M .— 1 1— ( 01 .-H .—1 I— 1 CM (M ■M —1 M <M <N ^ CM CM CO CO r-l CM CO <N (M I— 1 (M (N CM .-H C^l ^ Ol CM CO lO -* •-O ^ »C -rt* 10 »0 CO i-O ■:? 00 Tf< C^ CO (M CO CM CO CO CO CO ^ -* 238 RAY HENRY KISTLER lymph vessels was about four. Some might he ventral and to the right of the vena cava, while others in the same group layon a level dorsal to the vein but ventral to the artery. Still others lay dorsal to the artery. These vessels could be easily moved and rolled with the blood vessels and their number and position determined although the same vessel changed its relative position in many cases running more dorsal around the vein or artery or more to right or left. At the level of the left renal vein the group of lymph vessels joined to form two or three vessels which always ran dorsal to the left renal vein and to the left suprarenal gland where they formed the cisterna chyli at the level of the second lumbar vertebra. In this region the mesenteric, diaphragmatic and hepatic lymph trunks join the main lymph vessels. In some cases these join by a common trunk. Generally they flow into a large dilation to the right of the aorta just cranial to the renal veins which is connected to the cisterna chyli by a single trunk or a separate vessel and extend cranially from it and running along the right crus of the diaphragm join the thoracic duct above. In some cases a trunk from the mesenteric nodes also runs caudally passing ventral to the renal veins and joins the lymph vessels before they reach the cisterna chyli which was almost always distinct. It was a well marked dilatation partly surrounding the aorta, found between the second and third lumbar vertebrae or on the body of the second. It usually lies mainly ventral to the aorta but curves dorsally around the right side. In some cases it was formed by two dilatations one ventral and to the right and the other dorsal and to the left of the aorta which were connected by a large trunk. However, in two cases out of the series of 26 rabbits instead of a distinct cisterna a dense plexus was present. The dilatation into which the vessels from the abdominal viscera terminated was usually located a little cranial and ventral to the cisterna chyli which formed the beginning of the thoracic duct proper. The latter usually left the cisterna by one large trunk which lay dorsal to the aorta and ran in this position through the aortic hiatus in the dia THOHAC'IC DUCT IN THE UAUIUT 239 phrn^iii u]) into the thorax always lying to the left and dorsal to the aorta. In some cases a plexus took the place of a single vessel and in others there was a second trunk which started from the ventral part of the cisterna or from the dilatation made by the combination of the visceral lymphatics. This vessel passed to the right side of the aorta along the right crus and at about the 11th intercostal space usually turned dorsally and to the left to join the main trunk. If it did not do this it continued on up on tliat side of the aorta to a plexus always present at the level ot the tenth intercostal space dorsal to the aorta. The thoracic duct runs up dorsal to the aorta to the level of the 11th rib where it turns to the right and as it crosses always breaks up into a plexus of vessels which lie dorsal to the aorta. The occurrence of this plexus dorsal to the aorta at the level of the 11th rib may be explained by the fact revealed by Lewis ('05) in his work on the Ijaiiphatic development in the rabbit. Lewis says thiit the thoracic duct arises from a plexus of lymphatics surrounding the aorta. Hence it may be that in the rabbit only this part of the plexus persisted into adult life. The duct reappears as a single vessel to the right of the aorta at the level of the tenth rib or a httle cranially to it. The dorsal plexus which is rhade up of from 3 to 5 vessels is always present. As the duct extends craniall}^ on the right side of the aorta and dorsal to it, it lies ventral to the azygos vein. It thus extends to the level of the third rib where it comes to lie on the longus colli muscle. At the level of the second rib it begins to cross to the left and passes dorsal to the aortic arch as one vessel. It then runs up in the interval between the innominate and the left subclavian arteries to its termination into the left external jugular vein at the angle of junction of the internal jugular where the presence of nodes is very common. Hence forming the main lymph channels of the trunk we find, as shown in figure 1, a group of vessels running up through the abdomen around the large blood vessels and with the visceral vessel forming a cisterna at the level of the second lumbar vertebra. From this on extends one duct which Hes to left of aorta to the tenth intercostal space where it crosses forming a plexus dor 240 Fig. 1 Thoracic duct. Figs. 2, 3, and 4 Modes of termination of the thoracic duct. ABBREVIATIONS I.J., internal jugular E.J., external jugular T.S., transverse scapular vein Sub., subclavian vein A, aorta T.D., thoracic duct n., gland A.V., azygos vein I.V., innominate vein N , node S.R., suprarenal gland L.P., lumbar gland K, kidney THORACIC DUCT IN TIIE RAHBIT 241 sal to the aorta. From this plexus a single trunk extends cranially voiitnil to the azygos vein and to the right of the aorta to the second rib. At this level it crosses to the left again dorsal to the arch of the aorta and extends on up to its termination mto the veins to the left as a single vessel. Variations from the type of duct just described were very few. In 2 out of the 26 cases nos. 13 and 18 I found a plexus in place of a cisterna chyli. The plexus in each case surrounded the aorta and was longer than the average cisterna found in the other cases. In case of one of these the plexus extended on cranially as a narrow plexus on the dorsal side of the aorta up to the plexus at the 10th intercostal space which was described before as being always present. In 9 cases out of the 26 small vessels were seen iji the lower thoracic region running laterally at right angles from the duct to the right and dorsalwards, some lying ventral to the azygos, while others were dorsal. The usual place for these was at the plexus found dorsal to the aorta where the duct crosses over, or slightly above in the eighth or ninth intercostal space. There Avere never more than four and in most of the cases only one or two vessels. They came off at irregular intervals some close together while others were in the intercostal spaces. In 5 out of these 9 cases they were traced to small nodes found in the intercostal spaces near the sympathetic ganglia. In the other 4 the vessels were small and extended dorsally but it was not possible to determine distinctly where they went and in no case could I trace these to the azygos vein as reported by Witzer ('34) iQ a human subject and by Boddaert ('99) in the rabbit. Indeed in only one case a vessel was given off from the duct to the left at the level of the twelfth rib. It terminated in a small node. Another variation found was the presence of two ducts in two cases. In each of these instances two ducts came off from the retro-aortic plexus. In one case it came off from the plexus at the tenth intercostal space and ran to the left of the aorta to its termination. This vessel passed through a large group of nodes in relation to the left vena cava, from which it received tributaries and continued on up running dorsal to the 242 RAY HENRY KISTLER left subclavian artery and also dorsal to the main thoracic duct and terminated cranially to it at the junction of the external and internal jugular veins. The main duct on the other hand terminated at the innominate vein farther down and received a communicating vessel from the accessory duct as it passed ventral to it. In the other specimen the accessory duct at first turned to the right of the main duct for a length of two centimeters, then curved back to the left and dorsal to the main duct and aorta and ran up on the left side receiving vessels from nodes around the left vena cava and terminating caudal to the main duct in the innominate vein. Another variation was the presence of a vessel extending from the thoracic duct over to a node of variable size which usually lay under the right innominate vein. This occurred in 10 out of the 26 cases and hence was fairly frequent. The vessel usually came off from the duct at about the level of the third rib and ran cranially and to the right on the longus colli muscle to the node which lay at the first rib. In one case two vessels came from or joined this node one on each side of the right vena cava. The presence of small communicating loops was noticed in many cases. These were formed by a small vessel branching off from the thoracic duct and then running cranially a little distance and again joining the duct. This occured in about 9 cases at a region between the retro-aortic plexus and the aortic arch. In some instances the loop was longer than in others. In one instance it was about 4 cm, in length but others were shorter even as short as 5 mm. Most of these loops occurred where the duct lay on the longus colli muscle just before passing dorsally to the arch. In some cases small vessels were given off from these loops to surrounding nodes and in one case also ventrally to the aorta. Figure 2 illustrates the termination of the thoracic duct in rabbit no. 12. It has a single termination at the junction of the left internal with the left external jugular. It divides and is joined by vessels from surrounding nodes which are always present and finally empties by a single trunk into the large vein. TIIOIIACIC DUCT IN THE IIAHMIT 24'.^ 'I'liis is the type of teriniimtion found in IG out of the 20 rabbits. In some instances instead of dividing just before reaching the termination tlie chict formed a large dilation and in others it passed directly through nodes lying in this region. In every case the hiuph nodes especially those on tlie left side and around the three great vessels here illustrated, were completely injected. In some of the cases vessels could be seen coming from the nodes and joining the duct just before it terminated. These were most likely efferent vessels which must have been injected back from the duct. In one case a node fully 2 cm. in length lying dorsal to the left innominate vein was completely filled by the black injection fluid. This indicates that the valves in these vessels were either incompetent or had broken down. In figure 3 I have illustrated the type of termination found in 5 out of the 26 cases. The termination is double. One tnmk tenninates at the junction of the internal with the external jugular and the other joins the common jugular vein further caudally. In two of the cases it was hard to determine whether the most cranial duct terminated in the internal jugular itself before the junction was reached or whether it terminated at the junction of the jugulars. In all the five cases it divided after it had emerged from the dorsal side of the aortic arch. Injected nodes were also seen in this region. The teraiination of the second duct varied a little in its position being close to the main duct in some cases while in others it was a considerable distance caudally. Figure 4 represents the type of termination found in three out of the 26 rabbits. The termination was single in these cases and was found at the external jugular cranial to its junction with the internal jugular as a rule and usually running dorsal to the internal jugular in its upward course. In two of these cases a large dilation was found just before the duct terminated. From this dilation a lymph vessel always ran up the neck along with the internal jugular vein. Figure 5 represents the mode of termination of one of the double ducts found in this series. The left duct came off from the plexus dorsal to the aorta at the tenth intercostal space and ran 244 KAY HENRY KISTLER up on the left side. These ducts empty separately. The one at the junction of the jugulars and the other into the common jugular. After passing through a group of nodes the left duct passes on up dorsal to the subclavian artery and to the main duct to which it sends a communicating vessel and then forms a large dilation which joins the junction of the jugulars. The main duct takes the usual course and empties into the common jugular below. Figure 6 represents the other case of a double duct found in this series. The accessory duct branched off from the main duct at the eighth rib extending to the right and gradually E.J I.J I J. E.J. T.S. .. Sub N. Sutx Figs. 5 and 6 Modes of termination of the thoracic duct. curving back to the left, dorsal to the main duct and the aorta. It then passed through a group of nodes to the left and terminated in the innominate vein while the main duct followed the usual course and terminated at the junction of the external jugular with the subclavian vein. These five t>T)es here described include all of my series. AIcClure and Silvester ('09) have illustrated the termination of the duct in four rabbits and their results are very similar to those obtained by me. In two out of their four cases the termination was single and into the junction of the internal and external jugular veins. This was the predominating place of termination in my series. In their rabbits they found two with double ducts which are similar to those cases in which I found two ducts. THORACIC DUCT IN THE UAHHIT 245 Upon compiirinj!; tho tcnnination of the thonicic duct in the nibbit witli otlier animals we find a considerable difTerence. In monkeys as fomid by McClure and Silvester the most general termination was a single trunk tenninating in the jugulo-subclavian junction. According to McClure and Silvester the two jugulars and the subclavian veins in monkeys usually join at a single point. According to Ellenberger and Baum the termination in the dog is usually double. One duct joining the junction of the left internal and external jugulars but the other the external jugular, a condition very similar to the rabbit. . The relative size of the position and junctions of the veins is also very much like that found in the rabbit. But according to ]\IcClure and Silvester and of Sisson ('11) and of Chauveau ('10) the thoracic duct of the cat usually terminates in the ex-^ ternal jugular by two trunks. According to Sisson the thoracic duct in the horse takes quite a tortuous course and usually terminates in one trunk which joins the anterior vena cava just behind the angle of junction of the two left jugular veins. The duct in the horse is said to develop from a plexus and hence varies considerably. Often there are two ducts. In the ox the duct is said to be very variable, hardly ever being single but often double or plexiform. Its termination also varies being single or double but it usually empties into the junction of the jugular and brachial veins. Parsons and Sargent ('09) who have investigated about 40 cases in man found that the duct in 75 per cent of the cases terminated as a single trunk. This trunk joins the internal jugular below the left flap of the valve where the jugular joins the left subclavian vein. In only 7.5 per cent of the cases was the termination at the junction. In 22 per cent of the cases the termination was by two trunks into the last centimeter of the left internal jugular vein. In many of the cases the duct bifurcated and then reunited sometimes only in the wall of the vein. In 2 cases out of the 40 multiple terminations were met with but never more than four. Clattenberg found the termination in the guinea pig usually single and into the innominate vein just at the junction of it with the internal jugular. THE ANATOMICAL RECORD, VOL. 11, NO. 5 246 RAY HENRY KISTLER Nevertheless the duct is usually double up to within a very short distance of its termination. The arrangement of the great vessels is quite different in the guinea pig than in the rabbit. According to Job ('15) the left duct of the rat empties into the jugulo-subclavian junction but is said to carry lymph from only about half of the body. Upon comparing my results with those found in other animals I find that there is a considerable difference in some respects while in others the differences are slight. In the horse, it is said, the cisterna chyli is about 10 cm. in length, ampullated and Hes dorsal to the aorta and to the right opposite the first and second lumbar vertebrae. It is always very definite and is formed by two trunks from the viscera and one or two lumbar trunks. It seems that the lumbar trunks are not regular vessels but contain nodes of considerable size and are also joined at intervals by lymph vessels. In the rabbit, on the other hand, a group of parallel vessels always run through the abdomen. These are never interrupted by the presence of nodes along their course. They combine and with those from the viscera and the diaphragm form a cisterna chyli which is not always very definite. From the cisterna in the horse the duct runs up through the aortic hiatus and forward on the right of the median plane between the azygos vein and the aorta. It is generally single and extends up to the sixth or seventh thoracic vertebra where it crosses ventral to the oesophagus and then runs to the left of the aorta to its tennination. In case of the rabbit the duct first lies to the left of the aorta and as it passes through the aortic hiatus dorsal to it. It then crosses to the right as described before at the tenth intercostal space and as it does so breaks up into the plexus dorsal to the aorta. This plexus, it seems, is never found in the horse. Cranially from this plexus the duct in the rabbit, lies in the same position as in the horse but crosses to the left, again much higher up and dorsal to the oesophagus. Comparing the duct in the rabbit with that in the ox we find some interesting variations. Instead of the group of lumbar vessels found in the rabbit there is usually only one large duct THORACIC DUCT IN TIIK UAMlU'l' 247 extending up from the lumbar glands and joining the large gastrointestinal tnink caudal to the renal veins dorsal to which it passes to form a small cistcrna. From the cisterna in the ox there may be one or two ducts which may pass through the aortic hiatus or there may be several anastomosing ducts. There hardly ever is only one duct as is typical in the horse. These ducts extend up and the one on the right crosses over at varying heights usually joining the left and terminating as one. In the ox the duct is more variable than in the rabbit. Comparing the thoracic duct of the human being with that in the rabbit we note few differences. In the abdominal region of man we have two trunks which extend up from the pelvis and lower extremities and the lumbar glands which he separated by the large abdominal vessels in the midline. These vessels pass dorsal to the renal veins and are joined by a large intestinal tnink which also passes dorsal to these veins and by two descending tioinks from the thorax which pass through the ventral part of the aortic hiatus. All of these vessels go to form the cisterna chyli which is about 5 to 6 cm. in length and Ues to the right between the aorta and the lower part of the vena azygos, posterior to the right cms of the diaphragm and opposite the first and second lumbar vertebrae. According to Davis ('15) a definite cisterna is only present in 50 per cent of cases in man. From the cisterna the human duct is single and extends cranially through the aortic hiatus. Lying dorsal and to the right of the aorta but ventral and to the left of the vena azygos, the duct tenninates singly in 89 per cent of the cases into the left subclavian vein, and in 22 per cent of the cases into the junction of the jugular and subclavian. In the rabbit, on the other hand, we find a group of 4-6 anastomosing lumbar vessels closely grouped around the large blood vessels with a relatively smaller and indistinct cisterna. From the cisterna in the human being one duct usually extends cranially through the aortic hiatus, lying dorsal and to the right of the aorta but ventral and to the left of the vena azygos. The vessel runs cranially in this position to about the fifth thoracic vertebra. Here it crosses over to the left dorsal to the aorta 248 RAY HENRY KISTLER and the oesophagus and from there extends out of the thorax on the left side of the aorta. In the rabbit the duct crosses the vertebral column, once at the tenth intercostal space in the form of a plexus and once above at the second vertebra. This crossing is very similar to that in the man but a plexus is never present in the latter where the duct crosses the vertebral column. Of the many variations which have been found in the human subject none known to me have the characteristics found typical in the rabbit. Comparing the cisterna of the rabbit with that of the cat we find that in the cat the cisterna is formed by the intestinal trunks and two or three lumbar trunks. It is definite and lies opposite the second lumbar vertebra. From the cisterna the thoracic duct which may be looped extends up as a single duct but always lies to the left and terminates in the left external jugular vein. In Bos taurus according to Baum, the cisterna is not very large. It is formed by two lumbar trunks which join the visceral trunks caudal to the renal vessels. These vessels join and form one large trunk which runs dorsal to the renal vessels to the cisterna opposite the second lumbar vertebra. From the cisterna the duct runs to the right of the aorta to the fifth thoracic vertebra. Here it crosses dorsal to the aorta and extends up on the left and terminates into the left external jugular vein. In the dog the cisterna is relatively very much larger than in the rabbit. It is long and ovoid and even extends into the thoracic cavity between the crura of the diaphragm. From the cisterna the duct is usually single, extends up on the right and then crosses over at about the fourth or fifth thoracic vertebra to terminate in the left external jugular vein. In the guinea pig according to Clattenberg a definite cisterna is not always present. The abdominal vessels are plexiform extending up in the midhne. At the level of the renal vessels a number of nodes are usually found in which the plexus is lost. From these nodes extend vessels which form a dilation opposite the first and second lumbar vertebrae dorsal to the blood vessels. THORACIC" DUt'T IN TIIIO IlAUiilT 249 Cninially from tlie cisterna another plexus of vessels may leave the nodes just mentioned and pass dh'ectly into the plexus above the cisterna. This ])lexus is usually wide and extends up into the thoracic cavity to the level of the (!i^!;hth rib. From the eighth rib on u]) natteul)er}>; ahnost jilways found two separate ducts one on each side of the aorta. The one on the right side crosses over at the third rib and forms another small plexus with the left. Cranially from here a single duct which terminates at the junction of the innominate and the internal jugular veins on the left is found. This form of duct is much like the embryonic plexus found in most mammals. In conclusion I wish to thank Professor Meyer for the help he gave me in the course of this investigation. LITERATURE CITED. Batjm, H. 1911 Kcinnen Lymphgefasse dirckt in Venen einmiinden. Anatomischer Anzeizer, vol. 39. 1911 Konnen Lymphgefasse ohne einen Lyniphknoten passiert zu haben in den Ductus Thoracicus einmiinden. Zeitschrift fiir Infections krankheiten der Haustiere, vol. 9. BoDDAERT, Richard 1899 Etude sur une communication exceptionelle entre le canal thoracique et la veine azygos chez le lapin. Annales de la societc medecine de Gand, Tome 78. BoHNE, A. 1907 Zwei Falle von Verletzungen des Ductus Thoracicus. Deutsche Zeitschrift fiir Chirurgie. Leipzig Band 87. Butler, C. S. 1903 On an abnormal thoracic duct. Journal of Medical Research, Boston, vol. 10. Ch.\uveau, A. 1910 The comparative anatomy of the domesticated animals. New York and London. Davidson, A. 1910 Mammalian anatomy with special reference to the cat. Davis, Henry K. 1915 A statistical study of the thoracic duct in man. Am. Jour. Anat., vol. 17. Gerh.\rdt, W. Das Kaninchen zugleich eine Einfiirhung in die Organisation der Siiugetiere. Leipzig. Job, Thesle T. 1915 The adult anatomy of the lymphatic system in the common rat. Anat. Rec, vol. 9, no. 6. JossiFow, G. M. 1906 Der Anfang des Ductus Thoracicus und dessen Erweiterung. Arch, fiir Anat. u. Phys. Anat. Abt. Kampmeier, Otto F. 1912 Thoracic duct development in the pig. Am. Jour. Anat., vol. 13. Lewis, Frederic T. 1905-08 The development of the lymphatic system in rabbits. Am. Jour. Anat., vol. 5. 250 RAY HENRY KISTLER Miller, Adam M. 1913-14 The thoracic duct in the chick. Am. Jour. Anat., vol. 15. MoRAN, H. 189-4 Note sur uiie anomalie du canal thoracique. Comp. Rend. Soc, de Biol. Paris et Nancy. 10 S., I. Meyer, A. W. 1915 Spolia anatomica. Addenda I. An unusual thoracic duct. Anat. Rec, vol. 9, no. 7. McClure, C. F. W., and Silvester, C. F. 1909 A comparative study of the lymphatico-venous communications in adult mammals. Anat. Rec, vol. 3. McClure, C. F. W. 1908 The development of the thoracic and right lymphatic ducts in the domestic cat. Anat. Anz., Jena, vol. 32. Pensa, a. 1908 Osservazioni suUa morfologia della cisterna chili a del Ductus Thoracicus. Review in Schwalbes Jahresberichte, Band 14. Parsons, T. G. and Sargent, Percy, W. G. 1909 On the termination of the thoracic duct. Lancet. Patruban, Carl von 1845 Ueber die Einmtindung eines Ljanphaderstammes in die linke Vena Anonyma. Arch, ftir Anat. und Phy. SviTZER, 1845 Beobachtung einer Theilung des Ductus Thoracicus. Arch. fur Anat. und Phys. Stromsten, F. a. 1912 On the development of the prevertebral duct in turtles as indicated by a study of injected and uninjected embryos. Anat. Rec, vol. 6. Sisson, S. 1911 Veterinary Anatomy. Philadelphia and London. WuTZER, C. W. 1834 Einmiindung des Ductus Thoracicus in die Vena azygos. Arch, fiir Anat. und Phys. SOME RESULTS AND POSSIBILITIES OF EARLY EMBRYONIC CASTRATION FRANKLIN P. REAGAN Department of Comparative Anatomy, Princeton University SIX FIGURES (four PLATES) The works of Eigenmann (7, 8), Hoffmann (11), Nussbaum (13), and Beard (3) have supported the proposition that the germ-cells in several vertebrates are precociously segregated, and that their locus of origin is not necessarily the 'gemiinal epithelium' or even its immediate region. The classic case of eai'ly segregation of the germ-tract of a vertebrate is that described by Eigenmann, who was able to trace the lineage of the sex cells of Micrometrus aggregatiis probably as far back as the fifth cleavage; this is the farthest that any vertebrate germcell has ever been traced. Hoffmann discovered the existence of primitive ova in the mesenchjmie of a number of bird embrj^os at a time prior to the establishment of genninal epithelia in those embryos. For a review of the works of Beard (3), Nussbaum (13), Rubaschkin (14), von Berenberg-Gossler, Allen (4, 1, 2) and Fuss (9), and others, the reader is referred to the introductory discussion by Swift (18). For a discussion of the segregation of the germ-cells in the invertebrates, Hegner's 'Germ-Cell Cycle' raiw be profitably consulted. The work of Hoffmami and others seems to be borne out by the highly interesting work of Swift (18). The latter has not only strengthened our belief in the extra-regional origin of the sex-cells in the chick, but he has offered a solution of the question why the primordial germ-cells of the chick had never been found prior to the twenty-two somite stage. He maintains that the sex-cells originate in a crescent-shaped area of the extraembryonic blastoderm anterior to the body-axis at the line of 251 252 FRANKLIN P. REAGAN demarcation between the areas pellucida and opaca (see his figure 15) ; that these primitive sex cells reach the gonad partly by their own wandering, but principally by way of the bloodstream which transports them either to the gonad where they continue to develop, or to some other region where they soon degenerate. He makes the highly interesting suggestion that the large wandering cells which Dantschakoff (5) found to degenerate in the blood-stream following the twenty-two somite stage are really germ-cells which failed to become incorporated into the gonad. At a very short time subsequent to the publication of the work of Swift, Professor McClure in 1914 suggested to one of his students that this work of Swift's could be verified or disproved by the early excision of this crescent-shaped area described by that author, provided the operation be done at a time before the germ-cells have wandered away from this area of proliferation. The work was attempted but no results were obtained. Since then the writer has spent a great deal of time trying to develop a technique such that the embryos might be maintained despite the extremely low viability of the experimental material. In this work, some few thousand chick embryos have been sacrificed. During the first year of the work it was possible to rear a few embryos to the seventy-two hour stage; later than this, two embryos survived to the age of five days. These were, however, discarded on the assumption that they were unfit for critical study owing to the fact that they had died a short time before they were preserved. This is explained by the fact that such embryos were always allowed to develop as far as they would. It was not always possible to find them in the act of dying. It was found, however, that in normal individuals which were bled to death and then allowed to stand many hours before preservation that the germ cells were quite easily distinguishable in these corresponding stages; material which has been dead a few hours is of some value, although it is unfavorable for the distinguishing of such fine structures as mitochondria. I have tried to rear the embryos to the more advanced ages for the reason that my chief interest lies not in UE8ULTS OF EARLY EMIJUYONIC CASTRATION 253 the C()ii(iniuition or (lisi)rov;il of the work of Swift, but rather ill dot(Miiiiiiiii{!; the later effects of this extraordinary early castration. It occurred to the writer that if this crescent-shaped area is the seat of the keinibahn, its removal would afford a means of the earliest castration ever yet performed on a vertebrate; it would give an opportunity for studying in pure culture the interstitial cells of the gonad — particularly their supposed effect on the secondary sexual characters. If, as castration of animals subsequent to hatching or to birth has shown, the secretion of the gonad profoundly affects the secondary sexual characters, it is reasonable to suppose that it might affect some of those sexual characters which are usually considered as primary — such as the persistence or degeneration of the Wolffian and the Mtillerian duct. If the latter be true, early removal of the primordial sex cells might possibly cause the retention of the original ground-plan of the urogenital system, namely the simultaneous existence of both of Wolffian and Mtillerian ducts (or at least one of the latter), if however the development and normal functioning of the interstitial cells were in no way impaired by the absence of sex cells, their secretions would cause the alteration of the ground-plan if this were their normal function. But on the other hand, if the gonad were incapable of producing a secretion in the absence of sex-cells, opportunity would be afforded of studying such response or irregularities as might be observed in the development of the other ductless glands. One castrated embrj'o was recently reared to an age at which one or the other set of genital ducts should have been ehminated. The ground-plan was found to be persistent when the embryo was dissected, but there is a possibility that an arrest of development due to operation played a part in the result. An ideal way to confirm the work of Swift would be to transplant a 'crescent' from an embry^o of one pure breed to the mesenchjmie of an embiyo of a different breed but of the same sex. If such an embryo could be reared to the adult stage its germcells could be tested by breeding. I have been able to rear an embryo into which a crescent was transplanted, to the age of twenty-one days when it died the day on which it should 254 FEANKLIN P. REAGAN normally have hatched. Dissection of this embryo showed it to be a male. Artificial hermaphroditism would be produced in fifty per cent of such transplantations as this just described. These are some of the problems raised by Professor McClure's suggestion, and which are now in progress of investigation. Some results of this work at its present status may be of interest. I shall describe at this time the results of some experiments by means of which I have convinced myself, at least, that the work of Swift has given us a more correct and more ultimate solution of the origin ,of the germ-cells than any other yet given in case of the chick; that his work is of the very greatest importance owing to the experimental possibihties which have arisen from it. The description consists of a comparison of a few individuals in which castration was complete or at least very nearly so, with normal individuals at corresponding stages. According to Swift, most of the germ-cells have assembled at the base of the mesentery back of the twenty-second somite when the embryo has reached the stage of thirty-three somites. In this position they remain until the germinal epithehum begins to thicken. They then migrate into the gonad. This is seen to be in process in my figure 1. In most of the adjoining sections, primitive ova are being incorporated into this thickened epithelium. The sections of this embryo are only five micra in thickness, yet it is impossible to find a section in the gonad-region which fails to exhibit sex cells. Some of these cells are found far ventrally in the mesentery. The figure shows the sHght protuberance of the right gonad; the region illustrated is very similar in location to that of figure 3 and the small rectangle in the keyplate for the latter. In the left upper corner of figure 1, a few erythrocytes have been inserted; they are drawn to the same scale as the rest of the figure. In every case, eiythrocytes were used for comparison with the size of the sex-cells. Figures 2 and 3 are from a section of the left gonad of an embryo slightly older than that from which the preceding figure was made. The gonads of the two embryos were, however, about equally prominent. The germinal crescent of this em RESULTS OF EARLY EMBRYONIC CASTRATION 255 bryo Avas excised :it ji time ])revi()iis to tlie establishment of uny intersomitic grooves and at Avlucli the neural fohls of the brain were first indicated. The embryo was killed at the age of ninety-four hours. In this section it will be seen that the germinal epitheliimi is very Uttle thickened. As in all other sections of this embryo, germ-cells are entirely absent. Only a few cells in the mesenchyme approach in size the erythrocytes shown in the figure; these are not germ-cells. None can be found which exceeds the size of the erythrocytes to any such extent as do the germ-cells which are invariably present in sections of the gonads and mesentery of embiyos of this age,. and which can be most readily identified, no matter what may be the technique of fixation or staining employed. Figure 4 is a section through the left gonad of a nonnal chick embiyo at the age of one hundred and nine hours. This excellent series was prepared about fifteen years ago by Prof. A. M. Miller. The gonad protrudes considerably into the coelom. Among the very compact interstitial cells are numerous lightly staining germ-cells of large size. The mesothelium of the mesenteiy adjacent to the gonad is very greatly thickened, darkly staining, and as in all normal individuals of this age, contains numerous ova which cause local protrusions into the coelom. It will be noted that these germ-cells which have been incorporated into the mesothehum are shghtly larger than those which have reached the gonad. Lying in the mesenchyme of the mesentery are large germ-cells. These can usually be found in any section of a mesenteiy at this age. Pictures quite similar to this are readily obtained in embryos one hundred and twenty-five hours old except that the sex cells found in the gonad are relatively somewhat smaller. It will generally be found that this size-relationship obtains in normal embryos. The germcells which remain in the mesenchyme are usually the largest. It may be that those in the gonad undergo more rapid multiplication. A few erythrocytes sketched indiscriminately from the dorsal aorta are inserted for comparison of size. Germ-cells can be found in the mesenchyme of the mesentery as late as one hundred and seventy-five hours. This is the oldest material which I have examined for this particular point. 256 FRANKLIN P. REAGAN In these older stages, many instances will be found in which nuclear division in such germ-cells has not been accompanied by cytoplasmic division, so that a large multinuclear cell results, having a diameter about equal to that of the mononuclear germ-cells of the thirty-three somite stage. Such a condition is shown in figure 5. A very interesting condition is found here; it will be noted that a large binucleate cell-body projects from the mesenteric surface and touches the gonad from which there is likewise a slight protrusion and on which there is also an interruption of the mesothelium. These are undoubtedly germ-cell nuclei. If this is a ,case of chemotaxis it is a very remarkable one. It is entirely possible that such proliferations from the coelomic wall might sometimes be misinterpreted as giant blood-cells. In the gonad will be seen a germ-cell which is smaller than those lying in the mesenchyme. We may now consider the conditions which obtain in a castrated embryo five days old. The operation was performed just prior to the estabhshment of the first intersomitic groove. If castration was not complete, it was very nearly so. In this embryo I have found about five cells which are considerably larger than the average erythrocyte, and several which are a little larger. I am convinced however, that these are not germcells. If one examines Swift's (18) figure 6, for instance, he will find at least two mesenchyme cells in the left upper corner of the figure which are larger than the erythrocyte in the small endothehal cavity at the lower right corner. There would be little danger of confusing these with the large germ-cells present in the figure. I have made no attempt to diagnose germcells on the basis of the form of the mitochondria, since there is not perfect agreement on the question whether the germ-cell mitochondria of the chick are chr.racteristic. See Rubaschkin and Tschaschin (14, 15, 16, 17, 19) and Swift (18) p. 496. The conditions in this five-dry castrated embryo are illustrated in figure 6. The section passes through the right gonad. The mesothelium of the mesentery consists of a single kyer of cells. In all nonnal individuals at this age this mesothelium near the gonad is found to be greatly thickened, staiiis darkly and contains sex-cells. There are no germ-cells in the mesenchyme of RESULTS OK KAliLV EMBRYONIC CASTRATION 257 the mcsontory, w Iumviis tlioy iirc iiivuriahly present in the normal iiulividiULl. The gomid contains only interstitial cells, so far as those a\ lio have examined the material have been able to detect. The ^onad tissue is greatly vacuolated, while in normal embryos e\'en iit* younger stages the gonad is quite compact and remains so even through rough histological treatment. The interstitial tissue displays solid darkly staining lines of intercellular substance bordered by interstitial cells the arrangement of which gives a foUage-like appearance. In some cases the plane of section passes through an interstitial cell which projects into a large vacuole. When the diameter of such a vacuole is about that of a germ-cell the picture is somewhat similar to that which a germ-cell in a normal gonad would present if the cytoplasm were dissolved out. Professor E. G. Conkhn has examined this material and has pronounced the fixation to be sufficiently perfect that germ-cells should be easily distinguished if they were present. If any are present they are certainly very few. In the foregoing account, the term 'stroma tissue' might well have been used instead of the term 'interstitial tissue.' It seems reasonable to believe that the abnormal conditions recorded and the early removal of the crescent-shaped piece of blastoderm are causally related. In conclusion I wish to thank Professors C. F. W. McClure and E. G. ConkUn for the aid which has made this work possible. SUMMARY 1. The extra-regional origin of the germ-cells of the chick may be regarded as highly probable. 2. The early location of the germ-tract on the yolk-sac gives opportunity for the earhest castration yet performed on a vertebrate. It makes possible an analysis of the functional activity of the sex-cells and the interstitial cells in the production of internal secretions, the effects of the latter on the sexual characters and on other characters. It makes possible the production of artificial hermaphroditism, provided the germ-cells of one crescent can be made to enter the gonad of an embryo of the opposite sex. The work of Swift is of very great importance. 258 FRANKLIN P. REAGAN LITERATURE CITED (1) Allen, B. M. 1906 The origin of the sex-cells of Chrysemis. Anat. Anz., Bd. 29. (2) 1909 The origin of the sex-cells of Amia and Lepidosteus. Anat. Rec, vol. 3. (3) Beard, J. 1904 The germ-cells. Part 1. Jour. Anat. and Phys., vol. 38. (4) VON Berenberg-Gossler 1912 Die Urgeschlechtszellen des Hiihner embryos aus 3 und 4 Bebriitungstage. Arch. Mik. Anat., Bd. 81. (5) Dantschakoff, W. 1908 Entwicklung des Blutes bei den Vogeln. Anat. Ilefte, Bd. 37, S. 471. (6) DoDDS, G. S. 1910 Segregation of the germ-cells of the teleost Lophius. Jour. Morph., vol. 21, p. 563. (7) EiGENMANN, C. H. 1891 On the precocious segregation of the sex-cells in Micrometrus aggregatus, Gibbons. Jour. Morph., no. 5, p. 481. (8) 1897 Sex differentiation in the vi^'^parous teleost Cymatogaster. Arch. f. Entw'raech., Bd. 4, p. 125. (9) Fuss, A. 1912 tJber die Geschlechtszellen desMenschenundderSaugetiere. Arch, fur Mikros. Anat., Bd. 81, Ileft 1. (10) 1911 tJber Extraregionare Geschlechtszellen bei einen Menschlichen Embryo von vier Wochen. Anat. Anz., Bd. 39. (11) Hoffmann, C. K. 1893 Etude sur le developpement de I'appareil uro genital des ois#.ux. ■ Verhand. der Koninklyte Akademie von Wetenschoppen, Amsterdam, Tweedie Sectie, vol. 1. (12) Meves, F. 1908 Die Chrondriosomen als Trager erblicher Anlagen. Cytologische Studien am Hiihnerembryo. Arch, fiir mikr. Anat. und Entw., Bd. 72. (13) NussBAUM, M. 1901 Zur Entwicklung der Geschlechts beim Huhn. Anat. Anz, Bd. 19. (14) RuBASCHKiN, W. 1907 a tJber das erste Auftreten und Migration der Keimzellen bei Vogelembryonen. Anat. Hefte., Bd. 39. (15) 1907 b Zur Frage von der Entstehung der Keimzellen bei Saugetierembryonen. Anat. Anz., Bd. 31. (16) 1910 Chondriosomen und Differensierungsprozesse bei Saugetierembryonen. Anat. Hefte, Bd. 41. (17) Semon, R. 1887 Die indifferente Anlage der Keimdriisen beim Hiihn chen und ihre Differenzierung zum Hoden. Jena Zeitschr. Naturwiss., Bd. 21. (18) Swift, C. H. 1914 Origin and early history of the germ-cells in the chick. Am. Jour. Anat., vol. 1.5, no. 4. (19) TscHAscHiN, S. 1910 liber die Chondriosomen der Urgeschlechtszellen bei Vogelembryonen. Anat. Anz., vol. 37. (20) Waldeyer, W. 1870 Eierstock und Ei. Leipzig, Englemann. (21) Woods, F. A. 1902 Origin and migration of the germ-cells in Acanthias. Am. Jour. Anat., vol. 1, p. 307. PLATES 259 260 RESULTS OF EARLY KMHUYONIC CASTUATION KHANKLIN P. KEAUAN I'l^TK » ^ \v '^j ,%•... ^. ^m ^«N. _ • % A ■^ ^ "" • • 261 THE ANATOMICAL EECOBD, VOL. 11, NO. 5 PLATE 2 EXPLANATION OP" FIGURES 3 Section through the left gonad of a ninety-four hour chick embryo which was castrated prior to the formation of any intersomitic grooves. Note the absence of germ-cells and the thinness of the germinal epithelium. The section is viewed from its posterior surface. Benda's fixation followed by Meves' iron-Jiaematoxylin stain. X 800. Endih., endothelium: Erth., erythrocytes; G. eplh., germinal epithelium. Msth,. mesothelium. 262 RKSri/rSKF I.AKI.V KMHUYONIC C:ASrUATI()N MIANKI IN P. KK'AIIAN PLA ri'; 2 <S:z>- ^^' ^ Erfh. Endth 263 PLATE 3 EXPLANATION OF FIGURES 4 Section through the left gonad of a normal chick one hundred and nine hours old, showing germ-cells in the gonad, and in the mesothelium and the mesenchyme of the mesentery. Zenker fixation. Delafield's haematoxylin and orange-G. Princeton Embryological Collection, series no. 46. X 800. 5 Section through the left gonad of a normal chick one hundred and sixtytwo hours old, showing primitive ova in the gonad and in the mesentery. Fixation and stain as in figure 4. Princeton Embryological Collection, series no. 41. X 800. CoeZ., coelom; Gon., gonad; .1/.si., mesentery; Msth., mesothelium; Pr.O., primitive ova. 264 KKSULTS OF KAKLY KMBKYONIC CASTUATION IKANKI.IS P. ItKAOAN 'KATE 3 265 PLATE 4 EXPLANATION OF FIGURE 6 Section through the right gonad of a one-hundred-and-twenty-hour chick which was castrated at a time prior to the establishment of the first intersomitic groove. The mesentery, the mesothelimn and the gonad are devoid of sex cells. Note the peculiar foliage-like appearance of the gonad tissue. A few erythrocytes drawn to scale were inserted into the lower part of the figure. Bouin's fixTation followed by iron-haematoxylin and eosin. The left gonad would have afforded a still better comparison with figure 4. The right one was chosen for the reason that the stroma-tissue was less dense than in the left. Both gonads are almost the same size. The mesothelimn near the left gonad is very little thicker than that on the right. The left gonad contains no germ cells. X 800. CoeL, coelom; £rt/i.. erythrocytes; (7on., gonad; A/.sL, mesentery; Msth., mesothelimn of mosenterv. 266 KESULIS or KAKLY KMBUYONIC CASTRATION h'HAN'KI.IN' P. UEAliAN I'l.A'JK * 267 THE RELATION OF vVGE TO FERTILITY IN THE RAT HELEN DEAN KING The Wislar Institute of Anatomy and Biology THREE FIGURES Datii have recently been obtained that show the complete breeding history of a considerable number of female rats. An. analysis of these data with reference to the question of fertihty and its relation to age seems desirable, since literature dealing with litter size in rodents (bibhography in 'The Rat,' Donald-, son, '15) gives very little information on this point and fails to record the entire litter production of even one pair of animals. The breeding records of seventy-six females that produced a total of 585 litters are used in this study. The majority of the females (50) were piebald or 'hooded' rats; the rest were either 'extracted' albinos (15) or 'extracted' grays (11). All three strains were derived from the F2 generation of a cross between the \nld Norway rat (AIus norvegicus) and the domesticated albino (Mus norvegicus albinus). Mention is made of the kind of rats used merely as a matter of reference. The conclusions drawn from the results are doubtless applicable also to other strains of rats. All of the females lived to be at least sixteen months of age, the oldest dying at the age of twenty-three months. Under the conditions existing in the animal colony of The Wistar Institute a rat is usually in its prime at the age of seven or eight months, and after reaching twelve months of age it is classed as 'old.' Very few individuals live for more than twenty months, although all animals are kept under environmental conditions that are seemingly well suited to their needs. The relatively early death of the animals is due, in part at least, to the fact 269 THE ANATOMICAl, BECOBD, VOL. II, NO. S 270 HELEN DEAN KING that seasonal changes in temperature in the region of Philadelphia render old animals very susceptible to pneumonia, the disease that invariably, proves fatal to a rat of any age. In a more equable cUmate, hke that of California, rats have been kept in good physical condition until they were four years old (Slonaker, '12). In the rat the menopause usually appears at the age of fifteen to eighteen months (Donaldson, '15, p. 21). Data covering the litter production during the first sixteen months of life, therefore, may be assumed to show the actual fertility of the great majority of females. The word 'fertility' is here used as defined by Pearl and Surface ('09) to designate: the total actual reproductive capacity of pairs of organisms, male and female, as expressed by their abihty when mated together to produce (i.e., bring to birth) individual offspring." FertiUty, according to this definition, is a much more comprehensive term than fecundity with which it is often confused. The latter, as suggested by Pearl and Surface should properly be used to signify only "the innate potential reproductive capacity of the individual organism as denoted by its ability to form and separate from the mature body germ cells." Litter data for the three strains of rats are shown in table 1. TABLE 1 Showing litter data for the three series of rats §^ ^ H fa K „ O H faO oo K H J H H ^ pa « ^ fa ce < a « V E R A ( NUMBER YOUNG P LITTER 2 •' s z z o w < a) w ■< a o «, H H K te J H H <l la J S s i "^ g S fa z Z < H ■s fa z Piebald series 50 15 406 88 6.9 6.2 2798 548 1447 279 1351 269 107.1 Extracted albinos 103.7 Extracted grays 11 91 6.7 609 310 299 103.6 76 585 6.7 3955 2036 1919 108.1 As table 1 shows, the corresponding records for the three series are very similar. The differences in regard to litter size and to the relative proportion of the sexes that are found are well within the limits of the variation .that is always to be ex UKLAl'IOX OK A(iE TO FEUTH.ITV IN RAT 271 pectocl when the iiiimher of records is eomjKinitively siiklU. For further uiudysis, therefore, tlie thitn for the tliree stniins have been combined. Tlie entire series of records, arranged according to tlie age of the mothers when the litters were cast, is given in table 2. Tlie 'mean age of the females,' as given in the first column of table 2, is the median point of a thirty day period in the life of TABLE 2 Showing the number of litters in the combined series, together with the sex ratios and the coefficients of variation for litter size. Data arranged according to the age of the females when the litters were cast MEAN AGE OF NUMBER AVERAGE TOTAL NUMNUMBER OF COEFFICIENTS OF FEM LES IN OF LITTEHS NUMBER OF VOUNQ PER UTTER BER OF IN MALES FEMALES MALES TO 100 FEMALE VARIATIONS DAYS CAST DIVIDUALS FOR LITTER SIZE 90 38 6.9 264 126 138 91.3 34.9 120 49 7.9 389 207 182 113.7 32.3 150 57 7.6 433 215 218 98.6 26.5 ISO 60 7.8 472 252 220 114.5 31.5 210 01 7.7 471 243 228 106.5 25.2 240 46 7.3 337 155 182 85.1 36.8 270 44 7.0 314 163 151 107.9 36.2 300 49 7.4 363 187 176 106.2 35.7 330 41 6.0 246 138 108 127.7 37.8 360 31 6.1 191 97 94 103.1 36.5 390 35 5.1 179 104 75 138.6 51.2 420 26 4.5 118 64 54 118.5 41.8 450 18 4.3 79 35 44 79.5 51.5 480 13 3.2 42 23 19 121.0 73.3 510 10 3.4 34 17 17 100.0 36.3 540 6 3.6 22 9 13 69.2 47.7 570 1 1.0 1 1 585 6.7 3,955 2,036 1,919 106.1 38.0 each animal, except in the two cases noted below. For example, the mean age '120 days' includes the records for all litters produced by females that were from 105 to 135 days of age when parturition occurred. The ninety day group is one exception to the above rule; it comprises litter records for a twenty day period only, as the youngest mother in the series was eightysix days old when her first Utter was cast. One female gave bu'th to a litter of one when she was 594 days old. For the 272 HELEN DEAN KING sake of uniformity this record is put under the mean age '570 days' which is thus extended to include a period of forty-four days. The majority of female rats that are in good physical condition cast their first litters when they are about three months old. Tliirty-eight of the seventy-six breeding females bore young before they were 105 days old; all of the remaining females, Fig. 1 Graph showing, for the entire series, the relation of the age of the mother to litter production (data in table 2). with four exceptions, threw litters before they reached the age of 135 days. As table 2 shows, the number of litters cast increased with the age of the mothers until the females attained the mean age of 210 days. After the age of maximum fertihty was passed the number of litters cast decreased rapidly, and only a small proportion of the females bore young after they had reached the age of fifteen months. The graph in figure 1, constructed from the litter data in table 2, shows the relation of the age of the mother to litter production. RELATION OF AGE TO FERTILITY IN RAT 273 Tlu' ^ni])!! in figure 1 starts relatively high and rises rapidly to its niaxinunn which conies at the 210 day ])cri()(l. The decline of the f>;r:!,])h is much more gradual than its rise, and not until n(>ar the 3()0 th),y pei'iod does the graph drop to the level at which it starts. From this point the fall is more rapid, and the gnijjh reaches zero after the females have attained the mean age of 570 days. Fecundity in the rat, measured solely by the number of litters cast by the females at different age periods, is thus found to accord remarkably well with the law formulated by Marshall ('10): "The fecundity of the average individual woman may be described, therefore, as forming a wave which, starting from sterility, rises somewhat rapidly to its highest point, and then gradually falls again to sterility." There can be no doubt that annuals, in general, tend to follow a similar law, as the Utter records for various species collected by Marshall, by Pearl ('13) and others have already shown. Judging from the data in table 2 a female rat reaches the height of her reproductive capacity when she is about seven months of age. This age represents also the median point in the animal's breeding career. That is, one-half of the total number of her offspring are produced by the time she has reached this age and one-half are produced afterwards. When the females have reached the age of eighteen months their reproductive activity is usually at an end, as the data in table 2 indicate. Donaldson ('06) has shown that the first year of a rat's life is approximately equal to thirty years of human life. On this assumption a female rat that is eighteen months of age corresponds physiologically to a woman of fortyfive. The menopause evidently takes place in these t^vo forms at about the same period in the life span of the individual, but there is no corresponding likeness as regards the age of puberty or of maximum fertihty; both of these processes take place in the rat at a relatively much earUer period. The third column of table 2 shows the average size of the litters cast by the females at different age periods. The fitters of very young females contained an average of 6.9 young per litter. This is a smaller average number of young than is found 274 HELEN DEAN KING for anj group of litters until the females have past the zenith of their reproductive activity. Such a result was to be expected, since a number of investigations, for instance those of Minot ('91) on guinea pigs and of Hammond ('14) on rabbits and pigs, have shown that the number of offspring produced by young animals breeding for the first time is usualty below the number that is considered normal for the species, and also that litter size tends to increase for a time with the age of the female. The largest litters in the series were those produced by females with a mean age of 120 days. Litter size remained close to the maximum until the females M^ere eight months old when a slight diminution in the number of offspring was noticed. A further decrease to an average of only six young per litter was found in the litters thrown by females that were one year old. Each succeeding month added to the female's life seemed to lessen the number of her offspring to a marked extent, and after the females were fifteen months old the mean size of the litters cast was only about three young per litter. Not infrequently the offspring of old females were born dead or soon died from neglect as the mothers seemed unable to suckle them. There is, as yet, no standard for litter size in 'extracted' strains of rats with which the present series of records can be compared. Miller ('11) and Crampe ('84) give 10.5 as the average number of young in a litter of wild gray rats ; but Lantz ('10), on examining a large series of animals, found an average of only 8.1 embryos in pregnant gray females. According to Crampe the average litter of albino rats contains 6.3 young; data for over 1000 litters, collected by King and Stotsenburg, give the mean number of young in albino litters as 7.0. According to the above observations litters of gray rats contain a greater average number of young than do those of albino rats. The 585 litters used in the present investigation contained an average of only 6.7 young. This seems to indicate that litter size in 'extracted' strains of rats is less than that in either of the pure strains from which the animals were derived. It must not be forgotten, however, that the litter size for the pure strains, as given above, was not obtained from the complete breeding rec RELATION OF ACiE TO FERTILITY IN RAT 275 ords of ii luiinber of females but from u random collection of litters cast by females of unknown a^(\ Litter size in various strains of rats cannot be pi-operly comparetl until litter records for the several strains have been collected in a similar manner. The relation between the age of the mother and litter size is shown by the graph in figure 2. The data used in constructing this graph are given in table 2. The graph reaches its maximum when the females are practically at the beginning of their reproductive activity (i.e., at Z40 270 300 33D 3S0 330 430 450 460 Fig. 2 Graph showing, for the entire series, the relation of the age of the mother to the average size of the litter (data in table 2). 120 days of age), and then declines very gradually approximating zero when the females are eighteen months old. Fertility in the rat, measured by the size of the litters cast, is thus found to be correlated with the age of the mother at the time that parturition occurs. There is a possibihty that the number of the pregnancy is a factor that influences the size of the litters cast. In order to analyze the data on this basis the records have been arranged according to the position of each litter in a Utter series and are given in table 3. When the data are arranged as in table 3 it is found that the second litter is the largest of the series. This result is in accord with the observations of Crampe ('84) and of King and Stotsen 276 HELEN DEAN KING TABLE 3 Shouting the number, the average size of the litters and the sex ratios when the data are arranged according to the position of each litter in a litter series POSITION OP THE H LITTER IN A LITTER SERIES TOTAL NUMBER OF LITTERS AVERAGE NUMBER OF YOUNG PER LITTER TOTAL NUMBER OF INDIVIDUALS MALES FEMALES NUMBER OF MALES TO 100 FEMALES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 76 76 76 74 66 56 43 36 30 21 14 9 7 1 7.2 7.7 6.9 7.3 7.0 6.8 6.9 5.8 5.0 4.8 4.8 3.8 3.5 2.0 553 591 531 543 463 384 ,300 210 151 100 67 35 25 2 290 292 276 270 235 197 157 118 89 57 31 15 9 263 299 255 273 228 187 143 92 62 43 36 20 16 2 110.2 97.6 108.2 98.9 103.0 105.3 109.7 128.2 143.5 132.5 86.1 75.0 56.2 585 6.7 3,955 2,036 1,919 106.1 burg ('15) on the albino rat. The number of the pregnancy, up to five, does not seem to have a very marked effect on Htter size. The first five groups of fitters have an average of 7.2 young per fitter, which is above the average size of fitters of albino rats (7.0 young per litter) and considerably greater than the mean size of all the litters in the present series (6.7 young per litter). A slight decrease in size is noted in the sixth litter group, and in the succeeding litters the number of young diminishes steadily. Only exceptionally vigorous females are able to produce more than ten litters and these later litters rarely contain more than one to three young. As a rule female rats begin breeding when they are three months old, and they will produce a litter each month for several succeeding months if they are in good physical condition. The second litter is cast, therefore, when the female is about four months old and the fifth litter is born when the mother is seven or eight months old. On referring to table 2 it is found that litters born when the females are four months old have a RELATION OF AGE TO FEIITILITV IN HAT 277 jj;reiit(M' iiverap;e size thiin litters cast at any other age period, ajicl tliat females reach the chniax of their reproductive activity at about seven niontlis of age. In botli tables there is a rapid decrciise in the size of the litters towards the end of the series. As far as the question of litter size is concerned the two tables are in complete agreement. Sucli a litter series as that in table 3 is necessarily an age series, and it is very probable that it is the age of the female and not the number of the pregnancy that is a determining factor for litter size. The size of a newborn Utter of rats depends, primarily, on the number of ova shed at a given period of ovulation that are capable of fertilization. Litter size, however, is not always indicative of the actual fecundity of the female, since the offspring born represent only that portion of the fertilized ova that were capable of normal development. Not infrequently the examination of a gravid female will show one or several fertilized ova in the uterus that are more or less atrophic and so incapable of developing into normal embryos (Huber, '15). Such ova are usually absorbed in situ, and only very rarely are monstrosities found among the normal newborn young. According to Hammond ('14), the lower fertility of young sows as compared with that of adult animals is due to the fact that not so many ova are shed at each period of o\'ulation. This explanation for the change in the fertility of swine is doubtless appUcable also to a similar change in the fertility of rats and of other animals. Very probablj'^ the lessened fertility of old animals as compared with that of animals in their prime is due to the same cause. Whether abnormal ova are more frequent in old females than in young ones and so help to diminish the fertility in later life has not, as yet, been determined. The last column of table 2 gives the coefficients of variation for the size of the litters cast by the females at different age periods. These coefficients show that size variation is considerably greater in the litters thrown by very young females than in the Utters produced by females at the height of their reproductive activity when they are seven months of age. The latter 278 HELEN DEAN KING group of litters has the lowest coefficient (25.2) in the entire series. As the number of litters cast after the females were a year old was relatively small, the coefficients for later litter groups can have little value. There seems, however, to be a very marked tendency for litters cast by older females to exhibit a greater range of variability in size than is shown by the litters of young females, the maximum variability appearing in the litters produced by females when they were about sixteen months old. The entire series of litters gives 38.00 as the coefficient of variation for litter size. This coefficient is practically the same as that for litter size in the mouse, which is 37.5 according to the records collected by Weldon ('07), but it is 10 points less than the coefficient for the number of human offspring (Powys, '05). The coefficient of variation for fertility is very high in all mammals, apparently, being at least 25 per cent in the several cases where it has already been determined (Surface, '08). Different females — even sisters from the same litter — show marked variations in the number and in the size of the litters they produce. Whether such differences depend upon the inheritance of various fertility factors, or whether they are due to environment or to individual peculiarities of the females themselves remains to be determined. Table 4 shows the number of Utters produced by the seventysix females whose breeding records are used in the present study. As shown in table 4, the range of variation in the number of Utters produced by different females was from three to fourteen mth an average of 7.7 litters per female. One of the two females that cast only three Utters did not breed until she was six months old when she gave birth to a litter of seven. A second Utter, with nine young, was born when the mother was eight months old, and a final Utter, containing seven young, one month later. This female lived to be seventeen months old and she appeared to be in good physical condition until shortly before her death. The other female casting only three litters had a very similar breeding history. Some diseased RELATION OK ACE TO FERTILITY IN RAT 279 SliuwiiKj the liUtr proilurlion of 7G fcmnlc riitH Nl'MIlEH Of UKKKDINU NUMBEIi OF LITTEKS TEMALKS CAST 2 3 8 4 10 5 13 6 7 7 6 8 8 9 9 10 4 11 2 12 6 13 1 14 76 585 condition of the generative organs was doubtless responsible for the small number of litters produced by these two females, as investigations being carried on in the animal colonj^ of The Wistar Institute by Dr. Stotsenburg show that sterihty in a female rat is usually due to the formation of ovarian cysts or to degenerative changes in the uterus. According to Crampe ('84), female albino rats, as a rule, do not produce more than four or five Utters: records collected by Miller show that the wild gray rat has relatively more Utters than the albino rat. The average of 7.7 litters per female, found in the present series of animals, is undoubtedly too high for the general run of females. Twenty-thi'ee of the seventysix breeding females in this series had a total of five or six litters only, and it seems probable that this is about the average munber of litters produced by female rats in general. While six females had thirteen Utters each, only one female gave birth to fourteen Utters. This latter case is so unusual that it seems worthy of special note. The complete Utter data are given in table 5. This female, a piebald, gave birth to her fii'st Utter on February 7 when she was ninety-five days old. This litter was ex 280 HELEN DEAN KING TABLE 5 Shoiving the litter production of a female piebald rat, that ivas born November 4, 1913, and died June U, 1915 LITTER SERIES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 DATE OF BIRTH February 7, 1914. . . March 11, 1914 April 3, 1914 April 30, 1914 May 23, 1914 June 20, 1914 July 14, 1914 August 12, 1914.... September 10, 1914 October 15, 1914... November 23, 1914. January 28, 1915... March 26, 1915 April 28, 1915 NUMBER YOUNG 11 13 8 9 9 10 11 6 10 10 4 3 3 2 109 59 FEMALES 50 ceptionally large for the first litter of so young a female as it contained eleven young. The second litter, with thirteen young, was cast the following month. It is rather remarkable that both of these litters should be so much larger than normal, since, as a rule, a very large first litter is followed by a comparatively small one, unless at least two months intervene between the bu'th of the litters. The female cast two litters in April, and subsequently she gave birth to a litter each month until she was twelve months old. With one exception each of these litters was larger than the average litter of albino rats. A marked decUne in fertility was noted after the female was a year old: the intervals between litters became longer and the size of the litters decreased. The fourteenth litter, which contained only two young, was cast when the female was about seventeen months old, and although the female lived to be nearly twentytwo months old she did not breed again. During this long period of reproductive activity a total of 109 young were born, 59 males and 50 females. The median point in this female's breeding career was the same as that for the entire group of RELATION OF A(JE TO FERTILITY IN RAT 281 feiiuiles, nnnioly seven months, and she produced an average of 7.8 young in each Utter. An examination of the individual records for each of the remaining females in the series that gave birth to a very large number of litters i.e., from eleven to thii'teen, shows that in every uistance the first litter cast was large, containing from nine to eleven individuals. In those cases where females produced less than sLx litters the first litter cast, with one exception, never contained more than seven young. The number of records is so small that no definite conclusions can be drawn from them, but they seem to indicate that the size of the first litter cast is somewhat of an index of the fertility of that particular female : a large first litter indicating that the female, if she keeps in good physical condition, will produce more litters than the average run of females. Crampe states that the second of a rat's litters is always the 'best' and that this litter is indicative of the size of subsequent litters. This observation has been confirmed only in part by the present series of records: the second litter is the largest of the series, but the size of this Utter is not as indicative of the later fertility of the female as is the size of the first litter cast. Individual rats show as marked differences in the number of young produced at one birth as they do in regard to the total number of litters cast. Litters cast by some females are almostalways relatively large. The female whose Utter record is given in table 5, for example, cast but one litter in the first ten that contained less than seven young. Some females never have a litter that contains more than seven young, while others females cast a large and a smaU litter alternately. The litter frequencies in the three series of rats are shown in table 6, the range in litter size being from one to sixteen. In table 6, as in table 1, there are slight differences in the corresponding data for the three series of rats that may or may not prove to be significant when larger series of records are analyzed. Litters of eight young were most frequent in the piebalds and in the extracted grays, while six was the most common number of young in the litters of extracted albinos. The data for the 282 HELEN DEAN KING litter frequencies in the combined series is shown in the form of a frequency graph in figure 3. The graph in figure 3 has two modes, one at the point of six and the other at the point of eight young per litter. The graph thus appears to be compound, and it is possible that one of the two modal points corresponds to the degree of fertility normal for the wild Norway rat and the other to the degree of fertility that characterizes the albino rat, since these are the two strains from which the animals used for this study were derived. As the material is probably heterozygous as regards the factors for litter size, it does not seem advisable to attempt any analysis of the curve. It is of interest in this connection to note that the graph for litter frequencies in swine, as given by Went Showing TABLE 6 litter frequencies in the hree series SIZE OF LITTER 1 2 3 4 5 6 7 8 9 10 n 12 13 14 16 Piebalds 6 20 5 3 32 8 9 35 7 6 31 9 13 56 20 12 40 14 10 71 8 15 48 12 11 28 4 9 17 1 1 12 1 8 1 1 1 Extracted albinos Extracted grays 6 28 49 48 53 88 64 94 71 41 19 13 9 1 1 worth and Aubel ('16), has three modal points; one at four, a .second at eight, and a third at twelve pigs per litter. The first mode corresponds to the degree of fertility in the wild hog, the third is close to that of the most fecund of the domestic breeds of swine, and the third probably represents a heterozygous condition. Evidence regarding the relation of the age of the mother to the sex of her offspring is conflicting. Statistics collected by Bidder ('78) and by Punnett ('03) show that there is a great excess of boys among the children of very young mothers, the relative number of boys decreasing at subsequent births until the mother is thirty. Among children of old mothers (i.e., over forty) the sex ratio is again very high. In the horse Wilchens ('86) found a relation between the age of the dam and the sex of her offspring very similar to that existing, apparently. RELATION OF AGE TO FERTILITY L\ RAT 2S3 in the human vdvv. On the other liund, Schultze's ('03) mvestigations on mice indicate that the age of the mother has stH^mingly no influence whatever on the sex of her young. According to the observations of King and Stotsenburg the normal sex ratio in the albino rat is about 107.5 males to 100 females. As there arc no a\'ailable data regarding the normal sex ratio in other strains of rats the sex ratio in the albino rat is here taken as the standard with which to compare the sex ratios found in the present series of animals. i iii:UiTmiilMiH-riimj 10 II 12 13 l« IS IS Fig. 3 Graph for the frequencies of b'tter size in the entire series (data in table 6). Table 2 gives the sex ratios for the various litter groups when the data are arranged according to the mean age of the females at the time that the litters were cast. The sex ratios in litters belonging to closely related groups are so unlike that it would appear that there is no relation w^hatever between the age of the mother and the sex of her offspring. The sex ratio for the entire series of 3955 individuals is 106.1 males .to 100 females. This shows that in the strains of rats used for this study the normal proportion of the sexes is about the same as that in the pm-e albino strain. When the litter data are arranged according to the position of each litter in a litter series (table 3), the sex ratios obtained 284 HELEN DEAN KING for the individuals in successive groups of litters are not quite as diverse as those for related litter groups as shown in table 2. The sex ratio among the individuals belonging to the first litters of the series is higher than the standard, and in subsequent litter groups, up to the fifth, there is seemingly a tendency for the number of male offspring to decrease. A similar change in the sex ratios from the first to the fourth litter was noted by King and Stotsenburg in a series of litters cast by twenty-one albino females. Beginning with the fifth litter the sex ratios rise gradually until a maximum of 143.5 males to 100 females is reached at the ninth Utter of the series. For the eleventh and subsequent litters, however, the sex ratios are much lower than the standard. From the sex ratios as given in table 2 it would appear that among the individuals of a litter series the sex ratio might be expected to start relatively high and then fall steadily until about the fifth litter, rise again gradually to a maximum at about the ninth or tenth litter and subsequently drop to a low level which is maintained until the female reaches the menopause. The records under consideration are a special group selected solely because they cover the complete breeding history of a number of females that lived to an advanced age. Perhaps, therefore, they cannot be used legitimately to give evidence regarding the possible effects of the age of the mother on the sex of her offspring. From the data as given the only conclusion that can be drawn is that the age of the mother is not a dominant factor in determining the sex of her young. If, as Riddle ('16) maintains, sex is determined by the 'level of metabolism' in the fertilized egg, there is a possibility that the age of the mother may indirectly influence sex through its effects on the metabolic processes in the egg. Age has a profound influence on every tissue in the body, and its effects on the germ cells is a problem that must be attacked from a chemical standpoint, since it can never be solved by sex statistics however extensive they may be. UKl.A'l'lOX OF ACJE T(^ FERTILITY IN HAT 2S5 SUMMARY 1. Litter datsi covering the entire breeding history of seventysix fenuile rats are given in the present paper. All of the females belonged to 'extracted' strains that were derived from the Fo generation of a cross between the wild Norway rat and the domesticated albino. 2. The material used comprises the data for 585 litters containing 3955 individuals, 2036 males and 1919 females. The average number of young in each litter was 6.7. 3. Fertility in the rat, measured by the total number of litters cast, increases with the age of the female up to the time that the animal is seven months old. There is a sharp decline in fertility after the female is a year old and, except in rare instances, the menopause has appeared by the time that the female is eighteen months of age. 4. Female rats reach the height of their reproductive activity when thej^ are about seven months of age. This age also represents the median point in the animal's breeding career. 5. The age of the mother is a factor in determining the size of the litter cast. Litters of very young mothers are relatively small, and later litters are large, until the female reaches seven months of age. Litter size diminishes with the reduction in the number of litters cast, and litters of veiy old females rarely contain more than three young. 6. The second litter is the largest of the series, the third and fourth litters are usually a little larger than the first. 7. The serial number of the pregnancy, up to the fifth, does not seem to alter the size of the litter to any great extent. The sixth litter cast, however, is smaller than the preceding ones, and the number of offspring decreases rapidly as the position of the litter in the litter series advances. It is very probable that it is the age of the mother, not the number of the pregnancy, that influences the size of the Utters. 8. Coefficients of variation for litter size show that the litters cast by very young females have a greater range of variation in size than have the litters cast by females at the height of THE ANATOMICAL RECORD, VOL. 11, NO. 5 286 HELEN DEAN KING their reproductive activity. From this point the range of variation in Utter size appears to increase as the female grows older, and to reach its maximum in the litters cast when the females are sixteen months old. 9. For the entire series of litters the coefficient of variation for litter size is 38.00. 10. The total number of litters produced by different females varied from three to sixteen, with an average of 7.7 litters per female. 11. The majority of female rats probably produce from five to six litters only. 12. The size of the first litter cast seems to be somewhat of an index of the fertility of the female. If the first litter is very large the female will probably cast more litters than the average iiin of females, provided she remains in good physical condition. 13. The range in litter size was from one to sixteen. Eight was the most frequent number of young in the litters of the piebalds and of the extracted grays, while six was the most common number for the litters of the extracted albinos. 14. The sex ratio for the 3955 individuals in the series was 106.1 males to 100 females. This sex ratio is very close to the normal sex ratio for the pure albino strain (107.5 males to 100 females) . 15. The sex ratios obtained for the various litter groups (tables 1 and 2) do not indicate that the age of the mother is a dominant factor in determining the sex of her offspring. Old females, however, seem to produce relatively more females than male young. LITERATURE CITED Bidder, F. 1878 Ueber den Einfluss des Alters der Mutter auf das Geschlecht des Kindes. Zeitschr. Geburtshiilfe und Gj'nakologie, Bd. 11. Crampe, H. 1884 Zucht-Versuclie mit zahmcn Wanderratten. II Resultate der Kreuzung der zahmen Ratten mit wilden. Landwirthschaftliche Jahrbiicher, Bd. 13. Donaldson, H. H. 1906 A comparison of the white rat with man in respect to the growth of the entire body. Boas Anniversary Volume, New York. UK1-A'I'I().\ OK M,K TO FKIt'l'I M TV I.V HAT 287 h..\M.i).s.).N.ll.ll. l\n:> riu- Hm. MfinuirsofriicWistHrrnstituto of Anatomy jmd HioloK.V, No. 0, Philadelphia, 191o. Hammond, John lOl-l On some factors controlling fertility in dotnestic animals, .lour. Afiri. Sci., vol. G. Mriiiu, (5. C\u\. lid.") The development of the albino rat, Mus norveKicus albinu.s. 11. .\i. normal ova; end of the first to the end of the ninth day. Jour. .Morph., vol. 26. KiNu, H. D. AND ST«)T.sKNnrnG, J. M. 1915 On the normal se.\ ratio and the size of the litter in (he albino rat (Mus norvegicus albinus). Anat. Rec., vol. 9. L.x.NTz, D. E. 1910 Natural history of the rat. Bull. Public Health and Marine Hospital Service. Govt. Printing Office, \Yashington, D. C. Maksu.^i.l, F. H. a. 1910 The physiology of reproduction. Longmans, Creen and Co., London. MiLi.EH, N. 1911 Reproduction in the brown rat (Mus norvegicus). Amer. Nat., vol. 45. MixoT, C. S. 1891 Senescence and rejuvenation. I. On the weight of guinea pigs. Jour. Phys., vol. 12. Pkahl, R. 1913 Note regarding the relation of age to fecundity. Science, vol. 37. Vkaul, R., axd Sukface, F. M. 1909 Data on the inheritance of fecundity obtained from the records of egg production of the daughters of '200egg' hens. Bull. Me. Agri. E.xper. Station, no. 166. Powvs, A. O. 1905 Data for the problem of evolution in man. On fertility, duration of life and reproductive selection. Biometrika, vol. 4. PuxNETT, R. C. 1903 On nutrition and sex-determination in man. Proc. Cambridge Phil. Soc, vol. 12. Riddle, O. 1916 Sex control and known correlations in pigeons. Amer. Naturalist, vol. 50. ScHiLTZE, O. 1903 Zur Frage von den Geschlechtsbildenden Ursachen. Arch. mikr. Anat., Bd. 43. Sloxaker, J. R. 1912 The normal activity of the albino rat from birth to natural death, its rate of growth and the duration of life. Jour. Animal Behavior, vol. 2. Surface, F. M. 1908 Fecundity in swine. Biometrika, vol. 6. Weldon, W. F. R. 1907 On heredity in mice. 1. On the inheritance of the sex-ratio and of the size of the litter. Biometrika, vol. 5. Wextworth, E. N. and Aubel, C. E. 1916 Inheritance of fertility in swine. Jour. Agri. Research, vol. 5. \\iLCKENS, M. 1886 Untersuchungen ueber das Geschlechtsverhaltniss und die Ursachen der Geschlechtsbildung bei Haustieren. Biol. Centralbl. . Bd. 6. TECHNIQUE NOTES From the Dcparlminl of Histdlogy and Embryology, Cornell University, Ithaca, New York I. THE APPLICATIOX OF BENDA'S NEUROGLIA STAIN II. M. KIXGERY This note embodies the results of soinc experiments with Benda's stain for ncurogHa cells and fibers. The stain is, as is well known, the basic anilin dye, toluidin blue, used after a double mordantage of the sections in ferric alum and sodium sulphalizarinate. A previous mordantage of the tissues in a chrome solution, a 'chromation,' is an essential part of the method. Benda himself has employed two methods for producing this 'chromation.' The first (Benda, '00) was the use of Weigert's chrome alum bath followed b}' chromic acid, with tissue fixed in formalin. In the second method (Benda, '10), tissue fixed in alcohol was placed in 10 per cent nitric acid, then in 2 per cent potassium dichromate, and finally in 1 per cent chromic acid. jM}' results show that this 'chromation' may be obtained by using a chrome fixer, or better, by a dichromate bath after fixation; and the best results are produced by the use of a fixer containing a dichromate, followed by a dichromate mordantage. This is essentially the method used in preparing tissues for Weigert's copper hematoxjdin, and it may be said that in general, sections which will take a good Weigert stain for myelin will also give good results with Benda's neuroglia stain. Tissues fixed in Zenker's fluid, in Zenker's followed by Mailer's, in Helly's (Zenker-formol) followed by Miiller's, in Muller's alone, in formalin followed by copper dichromate, and in the copper dichromate-sublimate-acetic mixture as used by Kingsbury, followed by copper dicliroinate, all have given good preparations with Benda's stain. It is worthy of note that the tissue fixed in Zenker's fluid alone gave the poorest neuroglia stain, and that good Weigert preparations were obtained from all the tissues, except this same one. A number of sections of the spinal cord of an animal ( a skunk happened to be available at the time) were fixed in copper dichromatesublimate-acetic and mordanted in copper dichromate for periods from two to twenty-five d&ys to determine the length of time required for the best 'chromation.' These all gave good results w th Benda's stain but the best preparations were obtained from tissues mordanted four to six days. Sections of this same material were also stained with Weigert's copper hematoxylin, and good preparations were obtained 289 290 H. M. KINGERY from the tissue mordanted two days. After a longer stay in the copper dichromate sohition, the myehn failed to stain with the copper hematoxylin, remaining yellow, while the neuroglia cells and fibers stained blue. With Benda's stain, the myelin after a short mordantage stains a light reddish brown, but after a longer mordantage remains yellow ; unaffected by the stain. From this it would seem that a short mordanting in dichromate solution is best for staining the myelin and a longer is best for neurogha. After a prolonged mordantage the axis cylinders are brought out more clearly, apparently as a result of their 'chromation' for they did not stain with the copper hematoxylin. This does not quite agree with the results of Smith, Mair, and Thorpe ('08) who state that the order is myelin, axis cylinders, neuroglia. This method of fixation — Helly's fluid (Zenker-formol) followed by Muller's or copper dichromate-sublimate-acetic followed by copper diclii'omate — has given excellent results with tissues from a number of animals: cat, dog, man, mouse, rat, skunk. The spinal cord was used in each case, and in addition, in one form (dog) the optic nerve was taken. These results would show that for the 'chromation' four to six days in copper dichromate (2.5 per cent solution) is equivalent to two to four weeks in Muller's fluid or plain potassium dichromate solution. There is apparently little choice between them; possibty the tissues mordanted in copper dichromate are a trifle less brittle. A 2 per cent solution of potassium dichromate works as well as Muller's fluid. My results with material fixed in this way (Muller's after Helly's or copper dichromate after the copper dichromate-sublimate-acetic mixture) have been better than with tissues treated as Benda recommends. The preparations correspond to his description; neuroglia fibers and nuclei of neuroglia cells, deep blue; cytoplasm of neurogha cells, paler blue or purplish; myelin, reddish brown; axis cylinders darker red; connective tissues light red or pink; nerve cells reddish or purplish; and Nissl bodies darker purple. Benda's method of staining was used, and is given for reference. Paraffin sections 5 to 8 /x are treated as follows : 1. 4 per cent ferric alum, twenty-four hours. 2. Running water, ten-twenty minutes, followed by several changes of distilled water. 3. Amber-yellow solution of sodium sulphalizarinate, twenty-four hours (saturated solution of sodium sulphalizarinate (Kahlbaum) in 70 per cent alcohol, 1 cc, distilled water, 100 cc). 4. Several changes of distilled water, which is then absorbed with tissue paper. 5. 0.1 per cent aqueous solution of toluidin blue, heated (on the slide) until it steams, and allowed to cool, fifteen minutes or more. 0. After rinsing in distilled water the sections are treated for a few seconds with acidulated alcohol (70 per cent alcohol, 100 cc, concentrated hydrochloric acid, 6 drops). The length of time required varies TECHNIQUE NOTES 291 with the (HffercMit nietliocls of inorclantiiifi;, mul is best (Icitoniiined by trial. 7. After the acid alcohol is absorbed with tissue paper, the sections are rapidly dehydrated with al)solute alcohol. 8. The sections are then differentiated with creo.sote under control of the microscope. This usually takes several minutes. If more than ten seems necessary, it may be well to remove the creosote with absolute alcohol and treat with acid alcohol again for a few seconds; then the sections maj' be dehydrated and differentiated with creosote as before. 9. The creosote is absorbed with tissue paper and after several changes of toluene or xylene, mounted in balsam or damar. To prevent the fading of the stain, it is necessary to remove the creosote pretty thoroughly. REFERENCES Benda, C. 1900 Erfahrungen liber Neurogliafarbungen und eine neue Far bungsraethode. Neurol. Centralbl., Bd. 19. 1910 Neurogliafiirbung. Enzyk. d. mikr. Tech., 2te Aufl., Berlin, pp. 308-311. HuBER, G. Carl 1903 Studies on neuroglia tissue (No. 2), neuroglia cells, and neuroglia fibers of vertebrates. Contributions to medical research (dedicated to Victor C. Vaughan). Wahr, Ann Arbor, Mich., pp. 578 620. Kingsbury, B. F. 1912 Cytoplasmic fixation. Anat. Rec, vol. 6, pp. 39-52. Smith, J. L., JNIair, W. and Thorpe, J. F. 1908 An investigation of the principles underlying Weigert's method of staining medullated nerve. Jour. Path., vol. 13, pp. 14-27. II. SOME USES OF MALLORY'S CONNECTIVE TISSUE STAIN H. M. KINGERY Mallory's anilin blue connective tissue stain is of course well known for the purpose for which it was intended, but it has been found quite useful for other purposes as well. As usually employed for collagen fibers, Zenker material is used. I have found that this stain used with tissue fixed in picro-aceto-formal (Bouin's fluid) gives a very pretty differentiation for skeletal muscle, distinguishing clearly the isotropic and anisotropic bands. In practice, a thin muscle is moderately stretched and pinned out flat on a cork which is then floated upside down on the fixer. The picric acid is sufficiently removed after washing about a week in alcohol. Thin paraffin sections (4-6^) are stained according to Mallor3^'s directions (five minutes in the acid fuchsin solution and then twenty minutes in the anihn blue solution; differentiation in 95 per cent alcohol, dehj'dration in absolute alcohol, toluene or x>'lene, balsam or damar). Differentiation proceeds rather slowly and may be watched under the microscope. In a finished preparation the dark band (anisotropic) is stained blue and the light band (isotropic) is light red or pink. Hensen's disc (M) appears light in the middle of the blue band and Krause's membrane (Z) is deep red in the middle of the pink band. 292 GUSTAVE J. NOBACK This method of fixation and staining when applied to insect materia differentiates very nicely chitinised from non-chitinised cuticula. The chitinised is stained red and the non-chitinised is a clear blue. This method was appMed particularly to the intestine of the grasshopper and besides differentiating the cuticula it brought out the striations of the muscle fibers of the muscular coats with almost diagrammatic clearness. With this same fixation (picro-aceto-formol) Mallory's stain also brings out the connective tissue very clearly. And I have found that the stain may be used with good results after a number of fixers — alcohol, Carnoy's 6-3-1, formalin — if the sections are placed for a short time in "picro-aceto-formol and then washed, before staining. Aqueous and alcoholic solutions of picric acid also give fairly good results as 'mordants,' but not so good as picro-aceto-formol. The results obtained by this method compare very favorably with those secured after Zenker fixation. The formulae follow. Picro-aceto-formol Saturated aqueous solution of picric acid 75 parts Formalin 25 parts Glacial acetic acid 4 parts Mallory's anilin blue stain Solution A. Acid fuchsin 2 gram Distilled water 100.0 cc. Solution B. Grubler's water soluble anilin blue. .'. 0.5 gram Orange G •. 2.0 gram 1 per cent aqueous solution of phosphomolybdic acid 100.0 co. III. THE USE OF THE VAN WIJHE METHOD FOR THE STAINING OF THE CARTILAGINOUS SKELETON GUSTAVE J. NOBACK While the Schultze and the Spalteholtz clearing methods for bone are well known and frequently used, it would seem to the writer that the van Wijhe staining and clearing method for cartilage, which nicely complements the Spalteholtz method in the study and demonstration of the development of the skeleton, is not so well known and appreciated. I venture therefore to call attention briefly to its value, having used it during the past three years with satisfactory results. The method is very simple; embryos or other material to which it is to be applied should be preserved in alcohol or (better) formalin. The specimen is next placed in 67 per cent alcohol with 1 per cent of hydrochloric acid added, for several days or a week. It is then transferred to the same solution plus 0.25 per cent of methylene blue in which stain it remains for a week or two weeks, until thoroughly stained. Toluidin blue may be used instead of methylene blue if preferred (Lundvall TECHNIQUE NOTES 293 '0-t,'12). It is then retmnsfcrred to tlie acid alcohol, whicli is changed at intervals of one or two days or when markedly colored. In this the sj)ecinien remains until the color is nearly entirely removed from all parts save the car(ilafi;e which remains deep blue. To remove the acid it shoulil be washed for several days with changes of 82 per cent (8") per cent) alcohol and then dehydrated l)y passinji; up through 95 per cent alcohol, absolute alcohol, eciual parts of absolute alcohol and benzene, into benzeiu^, changed at least once, in which it may remain, or it may be mounteil in xylene damar or Canada balsam. The method is particularly serviceable in demonstrating the development of (a) the sternum and ribs, (b) the auditory ossicles, Meckel's cartilage and Reichert's cartilage, (c) the chondrocranium, (d) the cartilage in the developing bones of the extremities, etc. As has been indicated at the beginning, the method supplements satisfactorily the Spalteholtz method in which the bone has been stained red in the usual way by means of alizarin. Two embryos, or the two halves of the same embryo carefully cut as nearly as possible in the median plane, may be run through, for bone (according to the Spalteholtz alizarin method) and for cartilage respectively, the skin, central nervous system and viscera having been removed to clarify the view. One ami or leg may be stained for bone and the other for cartilage, etc. A solid mounting medium offers so many advantages over a liquid one that where possible it was used, damar balsam in xylene solution being preferred to Canada balsam because of its lighter color. By using glass supports for the cover-glass, such as small pieces of glass rod, etc., and using care in adding successive amounts of xylene damar, solid mounts of quite large specimens may be made on glass slides or plates. In this way specimens such as arms of the same embryo stained for bone f.nd cartilage respectively may be mounted side by"^side or in parallel series to show advancing stages of development. Glycerinjelly may be used as a solid mounting medium in the Schultze method, or with alizarin stained bone, but it may not be used with specimens stained for cartilage with methylene or toluidin blue. In the use of liquid mounting media, it w^as found that the benzyl benzoate and oil of wintergreen mixtures used in the Spalteholtz method were not so useful for mounting specimens stained for cartilage by the van Wijhe method, since fading was apt to result. Benzene was^ found to be the most serviceable mounting medimn. Lundvall has used for larger specimens benzene four parts and carbon disulphide one part (see also Lundvall, 1912). REFERENCES LuNDWALL, H. 1904 Ueber Demonstration embryonalerKnorpelskelette. Anat Anz., vol. 25, pp. 219-222. Lundvall, H. 1905 Weiteres liber Demonstration embryonaler Skelette Anat. Anz., vol. 27, pp. 520-523. Lundvall, H. 1912 Ueber Skelettfarbung und Aufhellung. Anat. Anz., vol 40, pp. 6.39-646. Shipley, P G. and C. C. Macklin '1916 The demonstration of centers of osteoblastic activity by use of vital dyes of the benzidene series Anat. Rec, No. 9, July 20, vol. 10, pp. 597-599. 294 B. F. KINGSBURY Spalteholz, W. 1914 Ueber das Durchsichtigmachen von menschlichen und tierischen Praparaten. Leipzig, Ed. 2, 1914. VAN WiJHE, J. W. 1902 A new method for demonstrating cartilaginous mikro skeletons. Koninkl. Akad. van Wetenschappen te Amsterdam, Proc. meeting, May 31, 1902. IV. A CONVENIENT METHOD OF ORIENTATION IN PARAFFIN IMBEDDING WHEN PAPER TRAYS OR BOXES ARE USED B. P. KINGSBURY The following simple method has been used by the writer occasionally for the past fifteen years or so and since I do not recall ever having seen it mentioned, a brief description may not be out of place. The paper box method of paraffin imbedding which I use quite generally is of course well known and described in most books of technique (cf. Lee, Vade-mecum, p. 77). The procedure in obtaining the orientation is the following. By means of a moderately soft lead pencil direction lines or, if desired, the outline of the specimen or embryo (traced from a x 1 photograph), are marked on the inside of the bottom of the box chosen or usually on the paper before it is folded. The box is then floated on cold water and the melted imbedding paraffin poured in. The specimen is then at once transferred from the melted infiltration paraffin to the box. By the time this has been done a thin translucent layer of solidified paraffin covers the bottom of the box, the orientation lines upon the paper showing through. The specimen is arranged as desired according to the orientation lines or upon the outline and the paraffin allowed to cool, etc. Subsequently after removing the paper from the cold block it will be found that the pencil mark is on the paraffin block which then may be trimmed and by it easily oriented for sectioning. ON THE ELECTIVE STAINING OF THE ERYTHROCYTE K. OKAJIMA Kyoto, Japan Today we are in possession of numerous excellent staining methods applicable to microscopic researches of the blood. The anilin dyes have been especially serviceable in the coloration of the red blood cells, and among these Eosin and Orange G deserve special mention. Solutions combhiing a number of dyes, for instance the triacid mixture (Khrlich-Biondi), eosin methylblue (Maj'-CJiriiuwald), methylazure methylblue eosin (Giemsa), have rendered excellent service in the hands of many investigators. However these dyes are designed more particularly for the staining of film specimens of blood, since in staining sections they color not only the er\'throc>'tes, but also the plasmic substances; in a strict sense they are, therefore, not the elective stains for erythrocytes. The need of an elective stain for the hemoglobin bearing erythrocytes is often felt in researches dealing with the genesis of blood cells and in many other types of the histological investigations. I have recentty discovered a method of the elective tinction of the erythrocyte. The finding is based on the fact that the phosphomolybdic acid lac of alizarin stains among several tissue elements only haemoglobin. After mordanting with phosphomotybdic acid the great majority of the tissues of the animal bod}' lose the property of staining with molybdenum alizarin lac while the er3^hrocyte or more particularly haemoglobin is colored with it. In this regard the method here recorded may be regarded as a new method for the microscopic determination of haemoglobin. The various steps of the method are as follows : The material ma}' be fixed in for- 4. Stain in following mixture for malin, sublimate, potassium bi- 20 minutes to 20 hours. Sodium chromate, etc. sulfalizarinate, saturated aqueous 1. The sections are transferred solution — 100 cc. 10 per cent phosto distilled water. phomolybdic acid, aqueous solu 2. Mordant in 10 per cent phos- tion — 30 cc. (10-50 cc). phomolybdic acid solution for 30 5. Wash in water, seconds to 2 minutes. 6. Alcohol. 3. Wash in water. 7. Xjdol, balsam. It is not necessar}^ to prepare the staining solution a short time before using. A solution kept for one-half 3'ear, exposed to daylight, gave excellent results. On mLxing the phosphomotybdic acid and sodium sulfalizarinate solutions the yellowish brown color first observed later changes to one of bright orange red. Attention is called to the fact that on staining sections according to this method, the er\-throc34es of vertebrates, the nuclei of the erythroC3i:es excepted, are durably stained a light to dark orange red, other tissues remaining unstained. The method is thus differential. Some 295 296 K. OKAJIMA times the nuclei and protoplasmic substances, especially in materials fixed in bichromate of potassium, take the stain a little, but it is easy to distinguish the bright orange red color of erythrocyte from the dirty yellowish brown color of nuclear chromatin or from other protoplasmic substances, for in the latter the color bleaches gradually. The connective tissue fibrils and osseous tissue are colored a bluish tinge, increased in intensity by longer staining so that excellent double staining, with brilliant contrast in orange red and blue may be obtained. It remains to be considered whether the molybdenum alizarin lac stains haemoglobin or some other structure of the erythrocyte. To determine this question the following experiment was undertaken. On a spot on two slides the chemical pure hemoglobin (Merk, Darmstadt), was spread and near it a section of liver fixed in formalin. Both were allowed to dr3^ One slide was now mordanted in the phosphomotybdic acid solution, the other not. Both were then stained in the alizarin molybdenum lac. On the slide mordanted we observe that the liver cells were not stained while the haemoglobin was colored a very dark orange red. On the unmordanted slide the liver cells were colored brilliant red and the haemoglobin a deep orange red. This observation may serve to show that the erythrocyte or haemoglobin represent the substance which stains after mordanting of phosphomolybdic acid by molybdenum alizarin lac. From the facts given it would seem that molybdenum possesses the property of effecting animal tissues so that mordanting by it diminishes or entirely deprives them of their staining property, this with the exception of haemoglobin. It is a question whether the molybdenum alters the majority of tissues, haemoglobin excepted, or whether there exists a peculiar affinity between the haemoglobin and this lac. It may be of interest to determine precisely the chemical relations of both. It is recommended that before or after the staining by the solution given the sections be treated with some nuclear dye. Haematoxylin may be used for this purpose, the section being first stained in this dye. On the subsequent use of haematoxylin, staining of connective tissue fibrils is obtained by reason of the formation of an haematoxylin molybdenum lac as in the Mallory's stain. In conclusion it may be stated that as with the solutions so with the stained preparations, they are durable. The slides made one year ago and kept in the half dark are as yet unbleached. After the present work was completed and ready for publication, attempts to make alcoholic solutions were undertaken. The alcoholic saturated solution of the sodium sulfalizarinate changes a little its color on the addition of the phosphomoly])dic acid. The procedure is the same as described for aqueous solutions. By this modification the length of time required for staining has been considerably shortened and the coloration seems more certain. On the durability of both solutions and preparations, when alcoholic solutions are used, a future communication will give information. The alcoholic stain has been prepared by mixing the two following solutions: Sodium sulfalizarinate saturated alcoholic solution, 100 cc; 10 per cent phosphomolybdic acid aqueous solution, 1 to 2 cc. NEUTRAL RED AS A CELL STAIN FOR THE CENTRAL NERVOUS SYSTEM J. B. JOHNSTON University oj Minnesota The method (lescril)e(l below has been used for eight years for the study of tlic tigroid content of cells as well as the size, form and grouping of cell bodies. As a 'Nissl method' it gives better results with formahn material than methylene blue or toluidin blue. The method is simple and the stain is permanent and is suitable for photography. The stain is made up in a 1 per cent aqueous solution and kept for months or years until thoroughly ripened. The ripening process is hastened by exposure to the air or by boiling but I know of no way to obtain a satisfactory staining solution in a few days. A good stain can be had after a few months; a better one after three or four years. A 1 per cent solution is diluted as required and the dilute stain may be used over and over again. The stain is used with either celloidin or paraffin sections. Differentiation is carried out in alcohol. The lower grades of alcohol differentiate slowly enough to allow one to examine the sections under the microscope. The higher grades wash out the stain and the sections should remain in these grades only as long as necessary to secure dehydration. If differentiation has been completed in 70 per cent alcohol, it may be carried too far while the sections are being dehydrated in the higher grades. The clearing of the sections requires care, as any alcohol left in the sections will remove the stain. Also, some of the common clearing agents injure the stain while others soften celloidin. For paraffin sections xylol is safe and satisfactory'. For celloidin sections castor oil and bergamot oil are the only common clearing agents which have proved constantly reliable. Carbol-xylol ruins the stain at once and cedar oil, clove oil, cajepat oil, oil of thjane, and anihn oil have all been unsatisfactory for one reason or another, at least in certain samples. After clearing in bergamot or castor oil it is well to rinse the sections in xylol. The mounts harden more rapidly and seem a little more brilliant and transparent. If castor oil is used the sections can be handled much more easily if the oil is thinned bj- addition of one part of xylol to two or three of castor oil. The xjdol also counteracts any tendency of the castor oil to soften the celloidin. It is most convenient in handling large sections to float them on glass slides in the last grade of alcohol and pass them through the clearing agent on the 297 298 J- B. JOHNSTON slide. In all these particulars the method is more difficult with large and, especiall}^ thick sections. Sections 50 microns in thickness thi-ough the whole brain of a newborn babe have been perfectly stained by this method, Avhile sections 100 microns thick through the entire adult human brain could not be differentiated and cleared uniformly. The steps in the process may be stated as follows: Material fixed in formalin or alcohol. Old formalin material gives good results. Paraffin or celloidin sections. Aqueous neutral red Griibler (well ripened) diluted to one-fourth or one-tenth of 1 per cent. Use warm 10 to 30 minutes, or cold 12 to 24 hours. Differentiate in 50 or 70 per cent alcohol. Dehydrate rapidly in higher grades. Clear paraffin sections in xylol; celloidin sections in castor-xylol or bergamot oil. Mount in dammar or balsam. Clearing and mounting media must be neutral. Neutral red does not work well after chromic salts and hence can not be used as a secondary stain after Weigert. The stain does work admirably, however, after either the Cajal or Bielschowsky process on formalin or alcohol material. Some beautiful preparations have been obtained by treating with neutral red as above sections in which the silver stain has attacked chiefly the fibers. This enables one to study the relations of cell masses to fiber bundles and to see the origin of a silver-coated axone from the cell-body stained by neutral red. f X'l"^ BIOLOGICAL PROBLEMS AND THE AMERICAN ASSOCIATION OF ANATOMISTS HENRY H. DONALDSON Address of the President at the meeting of the American Association of Anatomists, December 27, 1916 ' The purpose of this address is to discuss the relations of anatomy — as represented by our Association — to the problems of biology, and to consider our future course in the light of our past record. My opportunity to deal with these matters has come as a result of 3^our kindness in placing me in the presidential chair. I wish to express my appreciation of this honor which you have conferred, and also to venture the hope that this address may serve to indicate my interest in our common development. Let me begin with a few words on the historic relations of anatom3^ It is easy to see that the divisions of biology have been named in a rather incongruous way. For example: Zoology is defined by its material ; animals. Physiology, by a great domain ; nature. Patholog>% by a state; disease, and Anatomy, by a mode of procedure; dissection. Strictly speaking therefore the anatomist is one who cuts up things — with the common connotation that his dissection is applied to the adult human body. But, as these instances show, derivations are hardly illuminating. As we know, all workers in science are arbitrarily labelled, for as they enter the hall of science they find at the very threshhold a robing room. Here hang the gowns mostly of ancient cut; bearing still more ancient labels, and clothed in one of these they pass, each to his appointed place. 299 / THE ANATOMICAL RECORD, VOL. 11, NO. 6 J.4.NUARY, 1917 300 AMERICAN ASSOCIATION OF ANATOMISTS It is an arrangement of convenience mainly, but the dead hand of these labels often lies heavy on us and may become even a misdirecting influence. WTien we turn to human anatomy as a practice, we find that it began without technological affiliations — as a pure science so to speak — and only later became fundamental to surgery. When that branch of medicine developed, and — despite the absence of anesthetics or the control of sepsis — was pushing forward, with speed in operation as the great desideratum, the cultivation of gross human anatomy as a body of ever ready knowledge was most intense. With the dawn of the modern era in surgery, its importance as a mass of minute information carried with much effort, diminished, for the facts came to be excellently recorded; the surgeon and the professional anatomists worked together and speed ceased to have its former significance. Moreover anatomy had reacquired something of its earlier and broader point of view. From the nature of the case human anatomy is purely descriptive, and for various reasons has remained almost exclusivelj a medical cult. Let us see then how our Association of Anatomists stands in relation to this subject, when it is narrowly defined. Out of 323 members we have 146, or 45 per cent, connected with departments of human anatomy in the strict sense Going a step further and examining the first fifteen volumes of The American Journal of Anatomy — ^where the larger portion of the purely anatomical papers is printed — we find that there have appeared 300 papers in all. These I have analyzed by groups into those dealing with gross human anatomy; with mammalian embryology; and with other topics. Subdividing this last group into those purely descriptive and those including experimental work. By this treatment it appears that 11 per cent are on gross human anatomy; 40 per cent on mammalian embryology and 49 per cent on other topics. Thirty-five per cent within the last group being purely descriptive, 14 per cent including experimental work. PRESIDENT S ADDRESS 301 In this connection it should be remenibered that while The American Journal of Anatomy contaijis almost all of the papers on human anatomy which are presented to us yet the papers on other topics tend to appear in different journals. That the fore^oinp; ])i-()p()rti()ns represent fairly well what has been going on is indicated Ijy the fact that an analysis of our programs for the same period shows a similar distribution of the papers. In all, there have been 558 titles presented at our meetings and the proportion for human anatomy rises, by virtue of the presentation of several papers on the brain and on anthropometry, to 16 per cent. Taking then these two determinations for the work on human anatomy at their face value it appears that this subject is represented in The American Journal of Anatomy by 11 per cent of the papers, and at the meetings of the Association by 16 per cent of the titles— there being 45 per cent, or nearly half, of the members of the Association who are also members of departments of human anatomy. The reason for this seeming disparity is not far to seek, for even a hundred years ago we find men like old John Barclay of Edinburgh rather desperate over the anatomical situation. Before going further however, let me add here a few words intended to forestall any misinterpretation which might be made of the preceding paragraphs. These paragraphs are not intended to be in dispraise of human anatomy, nor as questioning either its paramount importance, nor the debt we all owe to those who have put the facts in order and made them accessible. Neither by my statistics would I imply that the anatomical members of this Association have failed in loyalty to their subject. The implications of the old name do hamper us however, and I wished to point that out, by showing, as I have done, how one could apparently demonstrate that the Anatomical Association was not attending to anatomy. On the other hand it was important to emphasize the accepted fact that human anatomy, in the historic sense, has been so carefully worked over and is so strictly descriptive, that today the human 302 AMERICAN ASSOCIATION OF ANATOMISTS body, though very completely known, is not the material with which we can work most advantageously in the solution of many biological problems. Appreciation of this fact has led to action and there is therefore a tendency to overstep the earlier and narrower bounds, to escape the ancient barriers and to annex the surrounding country. This tendency appears in our programs; but of these matters — later. The fairest conclusion from such a survey, is this: To any group of active biological workers it is hardly possible to give a descriptive name — fixed by statute, so to speak — which shall hold good for any length of time and be an indication of their intellectual aims for the guiding interests of such a group are bound to change from decade to decade, if not more rapidly. This Association illustrates most clearly such a shift of interest. Those who founded it (1888), and those who watched over its earlier development, felt constrained to retain the term anatomy — but at the same time sought to give to this term a wide and elastic meaning — a meaning, if you will permit the paradox, almost too useful to be formulated. As the association grew, both comparative anatomy, mammalian embryology, and physical anthropology were brought in to broaden the field of work and widen the scope of interest. Through embryology the common origin of organs and systems is revealed and the form relations of the adult explained. Comparative anatomy has shown the correlation of form with function, and by it the phylogenetic relationships of man have been made evident. But aside from advances of the sort just cited, the chief virtue of this more generous interpretation of anatomy has been to emphasize the fact that the animal body is in a continual state of flux and change, even from the structural standpoint. The moment attention is directed to the fact of change — be it ontogenetic or phylogenetic — the inevitable tendency is to endeavor to determine how such change occurs — and this leads to experimental work with the attempt to relate the character of the process to the details of the accompanying structural alteration. Fortunately studies which include experimental PKESIDENT S ADDRESS lUVA iiu)(lifi(':it ions luive been welcomed at our meetings. Tlie analysis of the programs shows an average of 14 per cent of exi)erimental papers up to 1910 and from that date on, an avejage of 32 per cent — a very notable increase. We shall agree, I believe, that a broad policy should be maintahied and the pj-esentation of work of this character encouraged. Such an attitude, however, desirable as it may be, does not jneet our needs completely. It is also important for us to i)ush further on in the direction in which these first steps have been taken, and I regard this occasion as an opportunity to say something concerning tlie course and character of this coming effort. We shall all admit that the ultimate problems of biology lie in the field of function. The investigator aims to control, to explain and to foretell activities, and in this connection it is fitting to recall that it comes down to us from Aristotle that the most important question concerning an animal is how it lives — not what it looks like. The functional phenomena are naturally the most important, first because of our deep interest in the dynamics of human life, and second because as we progress in the study of function, we pass from the cell — the unit of structure — to the more precise units of chemical activity, and these latter studies are thus made in the terms of the most fundamental masses yet recognized. When functional processes are expressed by a formula, the animal sometimes appears a bit superfluous, but a description of the structures involved is generally required to complete the picture. All physiological reactions imply structure— for the cells are the loci of the activity. To be sure only a portion of the structural peculiarities of cells is ever visible, but when studying the activities of an animal or its organs or tissues, it is of fundamental importance for us to possess the most detailed information available concerning the structural make up of that which is being tested. Thus interest in function normally carries with it a like interest in structure, and conversely structures which have known 304 AMERICAN ASSOCIATION OF ANATOMISTS functions become those worthy of the most intense study from the standpoint of anatomy. Apropos of this aspect of the question before us, permit me a word on the subject of general physiology. The study intended to discover the laws of living substance, wherever found, was well started on the continent in the early seventies — but just at that time the many problems in morphology and phylogeny brought into prominence by the work of Darwin, diverted the young men of the period — and nearly a generation had to pass before the subject was again taken up and developed as we find it today. For us the interest of this earlier endeavor lies in the fact that it was based on a sound knowledge of structure and represented the sort of appreciation of the inter-relations of form and function — for which I would here make a plea. Of course, since anatomy and physiology received their present elaboration, it hardly has been possible for individual workers to command, even in limited fields, either the experience or the information that would make their results equally valuable from both points of view. In a sense this state of affairs has come to stay — but on the other hand it is mitigated by the fact that scientific progress is accompanied by something more than the heaping up of many details. Definite conclusions emerge and generalizations are established, so that considered as usable information, the accumulated data gradually become more and hiore available for those not immediately occupied with the initial experiments or observations. As students of structure it is important for us to utilize the physiological information thus prepared, for in many cases it gives the significance to structural features which is needed to make those features intelligible. Despite the close interdependence of form and function, there often appears a curious antagonism between the workers in these two fields, an antagonism that amounts almost to a contempt for the kind of work done by the other man. When each is following his own line narrowly — be he concerned with function or with structure — I am not sure but that the antagonism is justified. Yet all work is not necessarily as restricted president's address 305 as the convejition.il labels might suggest, jind so far as any work in either field is hi-oad, it is not o])eji to hostile criticism solely by reason of its major content. If then, we break through historical limitations, and as anatomists extend our interest to the entire vertebrate series — including in the case of each form everj^thing between the germ cells and tlie senile animal, and if in our descriptions we recognize functional phases which are more or less under control, and on occasion exercise such control, we have marked out for study a field which includes all of the general biological problems, and which also may be approached from the angle of our special interest. Speaking in concrete terms we may put the case as follows: Given an animal the biological questions which it raises can be enumerated thus : Its systematic position; its place in the palaeontological series; its distribution; its ecology; its behavior; its heredity; its functions in the terms of tissues and organs — including the chemistry of the energy interchanges; its structural features from the ovum to the senile animal, in terms not only of gross anatomy, but also in the terms of constituent cells and cell structure, as well as in physical or chemical terms. These represent the concrete problems while over and above them floats the great cloud of general questions. The answers to all the questions thus indicated are to be sought with the idea of comparing the data gathered from the form examined with those gathered from other forms, for the purpose of obtaining general conclusions. Such a program calls for a good deal of downright description. Every biological investigation falls somewhere within this general frame, but in each instance has its own peculiar affihations and its special goal. If we turn now to the question of the direction in which our work as an association might move with advantage — it is necessary to clear the ground a bit before passing to particulars. In the first place it is self evident that biological work is bound to produce results applicable to man— and these results will 306 AMERICAN ASSOCIATION OF ANATOMISTS some time or other be added to our store of common knowledge. All our work moves that way just as steadily as our solar system moves towards the constellation of Lyra — though unfortunately not so fast. This definite trend need not worry us, for it has no bearing on what we sometimes call pure science. Pure science is the work of those who endeavor, as they go, to clear up the underlying problems brought to light by their special studies, rather than to press the immediate application of crude results. The practice is largely a question of temperament, and the definition one of degree. We have seen that traditional anatomy has been vivified by the inclusion of comparative anatomy, physical anthropology and embryology — ^all of them still very live subjects — but this is no reason why we should not look about for other fields which might also be cultivated with advantage. Doubtless each of us could present a list of biological topics that might be added to those already enumerated — and it is probable that such lists, while they would largely agree, would always express something of the personal preoccupations of the men who made them. Thus, were I asked to name some directions in which we might extend our work I should naturally lay weight on post-natal growth in the terms of cell multiplication and cell structure, with its many subsidiary problems, and also on the need for more precise information touching the chemical constitution of tissues and organs as modified by advancing age — for these changes must be of prime importance from the standpoint of function. This illustrates my point. Details in this direction are however not in place at this time, but it is perJiaps worth while to emphasize the value to each one of us of formulating with some exactness for himself the general lines of work which might be followed — and of determining their relations to each other — whatever his particular program may be — for when an investigation is to be carried out, these formulations offer the investigator something sohd on which he can lay hold. The question of the direction in which one should move is intimately related to the larger question of what things are worth rUESl dent's ADDltESS 307 while. A coimnon answer to this latter question is that one cannot tell what will be worth while— but I wonder whether this a<in()sti(' attitude is quite justified. If we look on science as cooperative, it follows that our individual woj'k should have some recognizable relation to the lai-ger problems or special interests of our own time, and the worth of our personal activity can, in one way, be measured by the degree to which the worker himself can link with his own activities those of his contemporaries. I put the statement in this form because others may not at once see the significance of an investigator's work and may even require to have it hanmiered in, but if the man himself sees it, and in the end can make it visible, he has arrived. This problem of worth links itself closely with the broad question of the way in which the body of knowledge constituting any science is built up; the question of "the work that lives." The answer to this last question is simple. The work that lives is that which is useful to other investigators — whether it be an hj^jothesis, a method or an observation. Other work no matter how brilliant in itself or how remarkable the technical abiUties were that lead to its accomplishment, has but an ephemeral existence. There is on the other hand a sort of ferment action exercised by thorough researches, for when an investigator reaches the point where he is able to offer generalizations, he specially contributes to the advancement of learning, for it follows as one result of his activity that his colleagues will be fairly certain to hunt assiduously^ for weak points in his work, and thus he acquires in the scientific world the honorable position of an organizer of labor. If you will look over the titles of any list of classical memoirs in science which have been thought important enough to be reprinted, you will find not only in biology, but in the whole realm of natural science, the sort of work to which I would direct attention. It is of interest to note that among the investigations thus honored, there is no stereotyped work, characterized by the 308 AMERICAN ASSOCIATION OF ANATOMISTS accumulation of results without improvements in method, or studies wdth a time clock attachment or investigations that are prescribed. The moral appears to be that the effective investigator is a free man: Yet along with this essential freedom must go limitations which are self imposed. As there is no such thing as an isolated idea it follows that in the world of thought it is possible to start from any point and by successive steps of association pass to any other point whatsoever. The investigator is continually called upon therefore to curb his interests or rather his activities — while pushing forward towards some goal which he himself clearly perceives. Thus it happens that the mature research always appears framed and oriented, and the author takes pains to inform us of the direction in which he is moving. Each fundamental science has at its core a few more or less precise concepts in relation to which it grows. A derived science like biology also has its own central concepts. To add to these concepts or to modify them by extension or rectification is to contribute. To this end descriptive and technical work are both important and necessary, but in the case of biology, there is also the functional side which the student of structure needs to appreciate. Something is gained if he admits it is there, but more if he takes it into serious consideration. I conclude therefore that although we are entrusted with the grave responsibihty of preserving and transmitting anatomical knowledge, and are designated by the ancient and honorable title of Anatomist, yet we are at liberty to reject anything historically imphed by that title which might prove hampering to our present work. Nevertheless our group is represented by those trained primarily in the study of structure and such studies must remain our chief occupation. It is not possible however to contribute effectively to the solution of the larger biological problems unless the functional responses associated with structures are kept clearly in view — or even examined. rUESlDENT'jS ADDRESS 309 So loiip; then as animals or their parts are studied comprehensively by us, we may hope to keep intellectually alert, and it is my conviction that our scientific worth as an association will iji the main depend on the persistence with which we follow an inclusive plan of investigation and maintain the broader view. rilOC'EKDIXCiS OF THE AMEUIC AN ASSOCIATION OF ANATOMISTS THIRTY-THIRD SESSION Anatomical Laboratories of Cornell University, New York Universitij and Columbia University, New York City, December 27, 28 and 29, 1916 Wednesday, December 27, 9.00 a.m. The thirty- third session of the American Association of Anatomists was called to order by President Henry H. Donaldson, who appointed the following committees: Committee on Nominations for 1917: J. P. McMurrich, chairman; G. Carl Huber, George A. Piersol. Auditing Committee: F. T. Lewis, chairman; S. W, Ranson. The morning session for the reading of papers concluded with an address by the president, Prof. Henry H. Donaldson, on Biological Problems and the American Association of Anatomists." Thursday, 12.30 p.m. Association Business Meeting, President Henry H. Donaldson, presiding. The Secretary reported that the minutes of the Thirty-second Session were printed in full in The Anatomical Record, volume 10, number 3, pages 133 to 269, and asked whether the Association desired to have the minutes read as printed. On motion, seconded and carried, the minutes of the Thirty-second Session were approved by the Association as printed in The Anatomical Record. Prof. F. T. Lewis reported for the Auditing Committee as follows: The undersigned Auditing Committee has examined the accounts of Dr. Charles R. Stockard, Secretary-Treasurer of the American Association of Anatomists and finds the same to be 311 312 AMERICAN ASSOCIATION OF ANATOMISTS correct with proper vouchers for expenditures and bank balance on December 23, 1916, of 1264.34. (Signed) F. T. Lewis, S. W. Ranson. The Treasurer made the following report for the year 1916: Balance on hand December 21, 1915, when accounts were last audited 264.09 Receipts from dues 1916 2222.88 Total deposits 2486.97 Expenditures for 1916: Postage and telegrams 40.55 Printing and stationery 68 . 15 Collection and exchange ' 2.53 Expenses of Secretary-Treasurer, New Haven Meeting... 17.80 Wistar Institute for subscriptions Journal of Anatomy, Anatomical Record, etc 2064.00 Stenography-typewriting 29 .60 Total expenditures 2222.63 Balance on hand 264.34 Balance on hand, deposited in the name of the American Association of Anatomists in the Corn Exchange Bank, New York City. On motion the reports of the Auditing Committee and the Treasurer were accepted and adopted. The Secretary announced that the Committee on Nominations, through its Chairman, Prof. R. R. Bensley, places before the Association the following names : For members on the Executive Committee, terms expiring in 1920, Prof. Franklin P. Mall and Prof. James Playfair McMurrich. On motion the Secretary was instructed to cast a ballot for the election of the above named officers. The Secretary presented the following names recommended by the Executive Committee for election to membership in the American Association of Anatomists. Allen, Erza, A.M., Ph.D., Professor of Biology, Philadelphia School of Pedagogy, 125 Thompson Ave., Ardmore, Pa. Amsbaugh, a. E., A.B., Student of Medicine, University of California, Berkeley, Calif. Bailey, Percival, B.S., Assistant in Anatomy, Northwestern University Medical School, Chicago, III. PROCEEDINGS 313 Byrnes, Cuaulios M., B.S., M.D., Instructor in Neurulogy, Johm Hopkins Medical School, 207 E. Preston St., lialtimore, Md. Carey, Eben J., Instructor in Anatomy, Creightun University Medical Department, Omaha, Neh. Carver, (Jail L., A.H., A.M., Profe.s.sur of Hiolo{i;y, Mercer University, Macon, Ga. CuM.MiNs, Harold. A.B., Instructor in Histology and Embryology, Vanderbilt University Medical School, Nashville, Tenn. DuBREUiL, G., M.D., Professor of Anatomy, Institut d' Anatomie, Universile de Bordeaux, Bordeaux., France. Eaton, Paul Barnes, A.B., M.D., 1306 W. Lexington St., Baltimore, Md. Fisher, Homer G., A.M., Student, Johns Hopkins Medical School, Baltimore, Md. Gee, Wilson, M. A., Ph.D., Professor of Biology, Emory University, Oxford, Ga. Gibson, G. H., M.D., Wailangi, Chatham Islands, Wellington, New Zealand. Holt, Caroline M., A.B., Ph.D., Assistant Professor of Biology, Simmons College, Boston, Mass. Johnson, Sy'dney E., Ph.D., Instructor in Anatomy, Northwestern University Medical School, Chicago, III. Keegan, John J., A.M., M.D., Instructor in Anatomy, University of Nebraska Medical College, Omaha, Neb. Koch, John C, B.A., Student of Medicine, Johns Hopkins Medical School, Baltimore, Md. KuNiTOMO, Kanae, M.D., Professor of Anatomy, Nagasaki Medical School, Nagasaki, Japan. Latimer, Homer B., A.M., Associate Professor of Zoology, University of Nebraska, 1909 South 27th Street, Lincoln, Neb. Lewis, Margaret R^ed, M.A., Collaborator, Department of Embryology, Carnegie Institution of Washington, Johns Hopkins Medical School, Baltimore, Md. Morris, Margaret, B.A., Ph.D., Osborn Zoological Laboratory, Yale University, New Haven, Conn. Murray, H. A., Jr., A.B., Student, Columbia University, College of Physicians and Surgeons, 437 West 59th Street, New York City. NoRRis, Edgar H., B.S., A.M., Assistant in Anatomy, University of Minnesota, Minneapolis, Minn. Pfeiffer, John A. F., M.A., M.D., Senior Asst. Physician and Pathologist, Government Hospital for the Insane, Washington, D. C. Rasmussen, Andrew T., A.B., Ph.D., Instructor in Neurology, University of Minnesota, Minneapolis, Minn. Ringoen, Adolph R., Assistant in the Department of Animal Biology, University of Minnesota, Minneapolis, Minn. Robertson, Albert Duncan, B.A., Professor of Biology, Western University, London, Ontario. Canada. Rose, Frank H., A.B., Austin Teaching Fellow, Harvard Medical School, Boston, Mass. Schultz, Adolph H., Ph.D., Collaborator in Embr3'ology, Carnegie Institution, Johns Hopkins Medical School, Baltimore, Md. Sharp, Clayton, A.B., M.D., Instructor in Anatomy, Columbia University, College of Physicians and Surgeons, New York City. 314 AMERICAN ASSOCIATION OF ANATOMISTS Smith, H. P., A.B., Student of Medicine, University of California, Berkeley, Calif. Smith, Wilbur Cleland, M.D., Assistant Professor of Anatomy, Tulane University, New Orleans, La. Swindle, Gaylord, Ph.t)., Instructor in Anatomy, Washington University Medical School, St. Louis, Mo. Turner, C. L., B.A., M.A., Instructor in the Department of Anatomy and Biology, Marquette University School of Medicine, Milwaukee, Wis. Wheeldon, Thomas Foster, A.B., A.M., Austin Teaching Fellow, Department of Anatomy, Harvard Medical School, Boston, Mass. Whittenborg, a. H., M.D., Professor of Gross-Anatomy, College of Medicine, University of Tennessee, Memphis, Tenn. Williams, James Willard, B.A., M.A., Professor of Biology, College of Yale in China, Changsha, China. {Care of G. H. Malone, Nanking, China.) On motion, the Secretary was instructed to cast a ballot for all the candidates proposed by the Executive Committee. Carried. Motion was made by Prof. S. W. Ranson, that a committee of three be appointed by the President to consider the matter of nomenclature relating to the Sympathetic Nervous System. Seconded and carried. The Association passed the following resolution regarding the National Research Council: Be it resolved: that the American Association of Anatomists hereby registers its approval of the coordination and federation of the research agencies in the country undertaken by the National Academy of Sciences, and expresses its willingness to join with and assist the National Academy in accomplishing the above federation. In connection with this, it was further moved by Professor Harrison that the Association cooperate with the Committee of One Hundred on Research and the National Research Council and that the Chair designate a Committee to work jointly with other members appointed by the Committee of One Hundred and the National Research Council. Seconded and carried. On motion the Association adjourned. Friday, December 29. A short business session followed THE Scientific Session. President Donaldson announced the appointment of the following committee to consider the matter of nomenclature relat PROCEEDINGS 315 ins to tlio Symixithetic Nervous System— Prof. G. Carl Huber, ch:iirin;iii; Profs. S. Walter Hanson and Irving Hardesty. It was then moved and voted that the Association express tlirouuii the Secretary appreciation of the valuable aid The Wistar Institute is rond(M-inp; the Association and the progress of anatomy in this country by its most generous and efficient publication of the Anatomical journals. Further The Wistar Institute has contributed greatly to the success of the present and past two meetings by its prompt publication and distribution of the abstracts of the conuiiunications presented. It was moved and voted that the Association express through the Secretary its thanks and appreciation to Cornell University Medical College, New York University and Columbia University for the cordial hospitality and the manner in which the Association has been accommodated and entertained. Charles R. Stockard, Secretary of the Thirty-Third Session of the American Association of Anatomists THE ANATOMICAL RECORD, VOL 11, NO. 6 ABSTRACTS OF PAPERS PRESENTED AT THE THIRTY-THIRD SESSION OP THE AIMERICAN ASSOCIATION OF ANATOMISTS December 27, 28 and 29, 1916 AT NEW YORK CITY (All papers marked with an asterisk (*) were read only by title) 1. The Golgi apparatus in the cells of the distal glandular portion of the hypophysis. William H. F. Addison, University of Pennsylvania. The Golgi apparatus in the cells of the distal glandular portion (pars anterior) of the hypophysis of the albino rats is demonstrable by several methods. It is shown in thin sections (3^/x) in the basophilic and acidophihc cells. After castration the basophilic cells undergo distmct changes (Anat. Rec, January, 1916) but the Golgi apparatus persists throughout these changes, apparently functioning as a definite cell organ. The methods used have been the neutral formol-bichromate-sublimate fixation of Bensley, (Biol. Bull., 1910), the neutral formol-bichromate procedure of Cowdry, as well as the special methods — the osmic acid method of Kopsch, and the method of Golgi as given by Kulesch (Arch. f. Mikr. Anat., 1914) and Riquier (Arch. f. Mikr. Anat., 1910). The pictures of the Golgi apparatus given by these several methods are quite comparable although of very different appearance. In the preparations made by the Bensley or Cowdry fixations, the general appearance of the Golgi apparatus may be studied after the usual staining methods — ]\IaUory's aniline blue-orange G., iron hematoxylin, or hematox3din and eosin. It is of larger size and so is more readil}^ seen in the large basophilic cells. Here it shows as a round or oval condensed spot, surrounded by a lighter ring, situated within the cytoplasm near the nucleus. The surrounding lighter zone is no doubt accentuated by 317 318 AMERICAN ASSOCIATION OF ANATOMISTS shrinkage due to difference in density between the substance of the Golgi apparatus and the surrounding cytoplasm. In a three-month normal rat hypophysis, prepared by one of these methods, the basophiles measure 14 /x x 11 yu, with nuclei 6.7 ^t x 5/x. Here the Golgi apparatus measures on the average 5.5 /x x 3.8/1. In three-month animals which had been castrated two months previously, the basophiles measured 19yu x 15.5^1 and the Golgi apparatus 8fx x 5 fx. Thus with the increase in size of the basophiles, the Golgi apparatus has also increased. In the basophilic cells of animals which had heen castrated more than two months, large vacuoles developed. These vacuoles press aside the nucleus and the Golgi apparatus, bat the latter retams its definite structure although often somewhat flattened by the enlarging vacuole. In the acidophilic cells, which are smaller in size than the basophiles, the Golgi apparatus is also smaller and hence not so conspicuous. Gemelli, an early observer (Boll. Soc. Med. Chir. Pavia, 1903) pictures much larger structures, which extend as a network throughout a considerable portion of the cytoplasm, but his experimental animal and technique were different. With the silver methods of Golgi the deposits of silver are within the spots. Often they show as a locahzed condensed mass but in other parts of the same preparation the black deposit takes on a network aspect. With the osmic acid method of Kopsch, the network appearance is often more sharply defined. What the nature of this appearance may be, is at present a topic of debate, but the fact that it persists when the cell is undergoing cytomorphic changes may be put forward as evidence that it is here a definite cell organ. S. The behavior of the interstitial cells of the testis towards vital dyes. William H. F. Addison and J. Monroe Thorington, University of Pennsylvania. On examining sections of testes of animals {e.g., white mice and white rats) which had been previously injected with trypan blue, distinct scattered blue spots are seen between the seminiferous tubules. This appearance was interpreted by Goldmann ('09) as being that the interstitial cells of Leydig had collected the blue within themselves. From our. studies it would seem that the blue is not within the glandular interstitial cells, but within other cells, which are of connective tissue types, and which normally compose part of the intertubular cell-masses. Before Bouin and Ancel ('03 and later) brought forward strong evidence for the hypothesis that the interstitial cells of the testis represented a gland of internal secretion, these cells were regarded as trophic structures, with the function of passing on nutrient material from the blood vessels to the various cells within the seminiferous tubules. In this latter view, Goldmann concurred and believed that he could follow the processes of the interstitial cells forcing their way into the tubules and so purveying granular nutritive substances directly either to the Sertoli cells or to the spermatids. H. M. Evans (Science, '14) PllOCEEDINGS 319 found in tli(« t(>stis tho iiiterstitiul ('(^IIs of Ijoydif:; aro stained brilliantly alth(niji;li the granules liore arc sinf;iilarly rof^ular and internu^diate in size between those possessed by cells of types one and two in the skin. In addition, true vitally stained connective tissue cells of type two are present between the seminiferous ducts." In our seiies of experiments on male white mice and white rats, in wliicii trypan blue was injected subcutaneously, the reaction to the stain was tyjiical. When thin sections (3-5 /jl) of the testes were examined, the l)lue coloration was seen to be confined within a small ninuber (2-3-()) of cells in each intcrtubular mass. These cell-masses are often trianf^iilar in shape, as seen in cross-section of the testes, and the blue-containing cells were disposed singly and situated, for the most part, at the periphery of, and often one at a corner of the intertubular cell-masses. By counterstaining it was seen that the glandular interstitial cells were free from the dye. The cells containing the ])lue were of two appearances. Some were elongated in the form of fibroblasts, but the greater number were rounded or polyhedral. These latter from their finely vacuolated cytoplasm and often slightly irregular oval nucleus, might be regarded as clasmatocytes. The nuclei of these were smaller and of more homogeneous appearance than the nuclei of the glandular interstitial cells. According to Evans' definition, both types of cells containing blue would be included under the term macrophage. Thus by the use of trypan blue it is possible to distinguish between glandular interstitial cells and macrophages in the intertubular cellmasses of the testes of mice and rats. 3. On the origin and fate of the osteoclasts {lantern). Leslie B. Arey, Northwestern University ]\Iedical School. Since their discovery bj' Robin in 1849, the poh'karyocj^tes of developing bone have been regarded commonly as the agents of bone resorption. For this reason these multinucleate cells have been termed 'osteoclasts.' Views as to the origin of the osteoclasts are not in accord. Kolliker maintains that the}- arise from osteoblasts by repeated nuclear division; Howell infers an origin by osteoblastic coalescence; Bredichin derives them from fused bone cells; Ranvier, Duval, and Bohm from lymphoid marrow cells; Mallory from fused endothelial leucocj'tes; Wegener and Schaffer from the endothelium of capillaries; Kaczander from cartilage cells. The results of Jackson, Danchakoff and Maximow agree in tracing the origin of the first osteoclasts in the early stage of bone development to enlarged reticular cells of the bone marrow. These cells possess at first but tw^o or three nuclei and the cytoplasm is basophilic. Later their cytoplasm becomes oxyphilic and the nuclei may become extremely numerous. A variance of opinion exists also as to the resorptive potentialities of the osteoclasts and the manner in which the nuclei increase in number. According to Kolliker and Jackson these cells actively resorb 320 AMERICAN ASSOCIATION OF ANATOMISTS the bone matrix, while the number of nuclei is increased by nuclear division. Bredichin views the osteoclasts as transitional stages in the transformation of bone matrix into marrow and granulation tissue, with a coincident multiplication of the nuclei of the component bone cells. Danchakoff speaks of the confluence of mesenchymal cells. Maximow believes that large osteoclasts arise at the expense of smaller ones, the multinuclear cell-masses thus formed being ameboid phagocytes. He never observed nuclear division either by mitosis or amitosis. F. T. Lewis emphasizes the absence of direct evidence respecting a resorptive activity on the part of the osteoclasts and rejects their origin by cell fusion. They are rather to be regarded as degenerating cells "produced by those conditions which lead to the dissolution of bone." Concerning the ultimate fate of the osteoclasts, Kolliker believed that they might resume an osteoblastic function after their resorbfcive activity had ceased. Jackson rejects this view and maintains that both osteoclasts and bone cells return to a reticulum similar to that from which he holds they took origin. Maximow differs with Jackson in that some osteoclasts are said to be destroyed through extreme degeneration. The untimely death of Prof. C. W. Prentiss interrupted an investigation which he had been pursuing on the origin and fate of the osteoclasts, and relative to which he had published a brief note. A reinvestigation and extension of these observations, made by the writer, form the basis of the present preliminary communication. Observations have been made on membrane-bone of human, and especially of pig embryos. A favorable site for study is found about the walls of the dental alveoli where active bone resorption is preparing for the accommodation of the rapidly growing teeth. Here osteoclasts appear in large numbers. In regions where bone is actively forming the osteoclasts are columnar and distinct with basophilic cytoplasm. During development the c}i:oplasm diminishes in amount, and in older regions the still basophilic osteoblasts flatten out and form syncytial masses. While the osteoclasts may arise from reticular cells in the early stages of bone development, my observations indicate that in later stages they take origin from the osteoblastic syncytia just described. There were found all transitional stages between these syncj^ia with basophihc cytoplasm, staining blue with hemotoxylin, and typical oxyphilic osteoclasts staining red with eosin. Furthermore, osteoclasts were seen frequently continuous at either end with basophilic osteoblasts and particularly with osteoblastic syncytia. According to these observations, therefore, the osteoclasts arise from depleted osteoblasts which have first formed a syncythmi before being transformed into the oxyphilic osteoclasts. Nuclear division by mitosis or amitosis has never been observed, although mitoses were not uncommon in the nearby germinative layer of the epidermis. As bone resorption continues, new osteoblasts come into relation with PROCEEDINGS 321 tlie osteoclasts ami aro incorporated into them, so that in fi;eneral, the hirger an osteoclast, the more numerous its nuclei and tin; more extensive its prohaMe bone resorptive history. But the osteoblasts, as such, are not the only source irom which the nuclei of the osteoclasts arc recruited. Bone cells, eml)edded in the matrix, are laid bare by the resorptive processes and are invested by the osteoclasts followiiifr in the wake. All intermediate conditions may be found between the initial and final stages of inclusion. P'urthermore, bone cells are surrounded normally by a capsule which resists the action of strong hydrochloric acid. Encapsulated and distinctly stellate cells, which resemble bone cells, may occasionally be found embedded in the osteoclastic cjioplasm. Such cells are interpreted as l)one cells whose capsules are still resistant to cytoplasmic digestion. From the relative infrequency with which persistent capsules are seen it is probable that the enclosed bone cells are eventually lil)erated. Ingested bone cells must contribute in substantial numbers to the formation of osteoclasts. Here it appears that the degree of multinuclearity is an index of the number of osteoblasts and bone cells entering into the composition of the osteoclasts. There is no direct evidence as to how bone matrix (inorganic and organic) is resorbed. One may assume that it is essentially a process ot decalcification and digestion through the agency of an acid and an enzyme The relation of the osteoclasts to these changes can only be inferred. The facts that osteoclasts are plentiful w^here resorption is going on and disappear when resorbtion ceases, that they are applied closely to the surface of the bone, that they often wrap themselves around irregular spicule-processes or lie in distinct pits (Howship's lacunae), that the surface applied to the bone possesses at times a striate border, and that the staining reaction differs from osteoblasts all suggest but do not prove a resorptive activity. A word as to the fate of osteoclasts. That these cells may be resolved finally mto osteoblasts and again act as bone-formers seems mprobable. My preparations show nothing in favor of such a cycle whereas pictures of degeneration in varying degrees are abundant' lUe onset of cytoplasmic degeneration before the cessation of boneresorptive activity seems to be not uncommon. At times degenerating osteoclasts, smgly or in nests, are found stranded in the marrow tissue lliese often exhibit extreme vacuolization and other degenerative changes To a certain extent they probably atrophy and disappear altliough indications of a transformation into marrow reticulum are not lacking Large osteoclasts have been observed rarely in the blood vessels of the bone marrow. That such gain admittance, and do not arise in situ is supported by their degenerate appearance, due to cytoplasmic vacuohzation, granular degeneration and pyknotic nuclei Nummary. 1. While osteoclasts may form in the early stages of bone development from reticular cells of the marrow, in later stages they arise from syncytia of depleted osteoblasts. 322 AMERICAN ASSOCIATION OF ANATOMISTS 2. The numerous nuclei of large osteoclasts are derived (a) from the constituent osteoblasts, and (b) from bone cells which are ingested as the bone matrix is resorbed. 3. Only indirect evidence points to the osteoclasts as the active agents in bone-resorption; they may also be interpreted as degenerating osteoblasts. 4. Eventually the osteoclasts either atrophy and disappear or are resolved into the reticulum of bone marrow. 4- 071 the relation between neural and intermediate portions of the hypophysis. (Lantern.) Wayne J. Atwell, Department of Anatomy, University of Michigan. In a study of the development of the hypophysis in the rabbit an intimate association of neural portion and intermediate lobe has been observed. This imion of the two parts is first seen in rabbit embryos of fifteen days' development. Not all embryos of this day present the phenomenon, bat every sixteen days' embryo examined shows one or more definitely circumscribed regions in which the two parts are in very intimate relation. The conditions found in rabbit 8 A (sixteen day embryo; 5^ series) may be taken as typical. The hypophysis region of this embryo was carefully reconstructed in wax and then the neural and intermediate parts were separated. On viewing the model of the neural lobe from the surface which was in apposition to the intermediate part, four distinct areas of contact may be seen. They are arranged in two pairs, one of which is near the caudal, or free, end of the lobe. The other pair lies about midway between the free end of the lobe and the place of its attachment to the brain wall. The combined areas of these four contact regions approximates one-third of the entire surface of the lobe facing the intermediate portion. One of the caudal areas is the largest of the four and no two are of the same size. Embryos which present only one region of contact have that one near the caudal end of the lobe. From sections it may be seen that the basement membranes of the two parts are lacking at the places of contact. For this reason it is not easy to determine whether the contact is due to the active ingrowth of one part into the other or merely to the passive fusion of the two parts. There seems to be considerable evidence, however, from staining reactions and from the arrangement of cells that the contact is due to the active penetration of the neural portion into the intermediate portion. These contacts, or ingrowths, are, of course, not to be confused with the growth of connective tissue into the gland, which takes place, in the intermediate lobe, about the nineteenth or twentieth day of development. While every sixteen-day embryo examined presents at least one region of contact, older embryos do not all show them. In certain eighteen, nineteen and twenty-day embryos the two portions of the hypophysis are entirely separate, each showing its own, uninterrupted PROCEEDINGS 323 l)as(MU(Mit membrane'. Such of the older embryos as do show contact rej^ions have them re(hiced both in size and in ninnber. One further observation should be here recorded. On tlie surface of the intermethate part facinji; the resichial lumen of Jiathke's pocket may be seen a slijj;ht dejjre.ssion to correspond with the center of each contact area foimd on tlu^ oi)))ositi' surfaces of the part. In at tempt in ji; to interpret the abov(!-r(>corded observations for the rabbit the foUowinji' possibilities present themselves: 1. The contacts or ingrowths may be an attempted ontogenetic repetition of the condition which is normal for certain fishes. 2. An opportunity is ofll'ered for the early intermingling of elements of neural and intermediate portions of the hypophysis. On this account, and because of the conflicting statements concerning the achilt structure of the part, a careful study of the histogenesis of the intermediate lolie in mammals is very much to be desired. 5. Studies of the cortex of the sheep brain. Charles Bagley, Jr., Phipps Psychiatric Clinic, Johns Hopkins University. The purpose of this investigation was to outline clearly the cell and fiber architecture of the sheep brain, with the hope of establishing a foundation for further experiments. The brain of the sheep was selected for the following reasons: 1) Those interested in cerebral localization have scarcely touched upon the whole group of ungulates and there has been no attempt made to present a complete study of the brain of this member of the group. 2) To, facilitate the use of the brain in teaching the anatomy of the central nervous system. 3) The animal lends itself well for experimental study, is easily obtained and can be cared for with but little more difficulty than the more common laboratorj^ animals. 4) The specific starting point was von jMonakow's claim of the great size of the frontal lobe and its relation to the red nucleus — a contention which we soon found untenable. Methods. The plan of attack has been : (a) the study of the brain in phases of development; (b) survey of the cell and fiber structure of the normal adult brain, and (c) experimental ablation. For the embryological studies serial sections were made of brains of embr\'os varying from 2 to 47 cm., the latter being about full term. The architecture of the normal adult brain was first studied in a series of frontal sections of the brain and later in a series of sections made after individual dissection of the gyri — the latter having the advantage of eas}' orientation and accurate right angb cutting of the laminae. Experimental ablation of areas previously outlined histologically was done in five lambs (three to five weeks). These brains are now being sectioned and will he stained by the Pal-Weigert method for the stud}' of fiber changes and with various cell stains, to demonstrate alteration in the nuclei of the brain-stem. 324 AMERICAN ASSOCIATION OF ANATOMISTS Observations. In the embryological series we have outhned the order of development of the sulci and gyri, the period at which cell differentiation is sufficient to present lamination in various portions of the cortex, and the order of myelinization. The studies of lamination demonstrate five principal areas, connecting which there are narrow zones of cortex of transitional type. The fiber structure also permits of division into areas, but to a less definite degree. The results of the ablation experiments cannot be stated until further work is done on adult brains, followed by Marchi stains. 6. The effect of ultra-violet light rays upon the development of the frog's egg: (2) the artificial production of folded (u-shaped) embryos. W. M. Baldw^in, Albany Medical College (Union University). This paper is the second of a series on the effects of radiation on certain areas of the fertilized and undivided ovum of the frog by means of ultra-violet light rays. The source of energy utilized was an electric arc provided with iron electrodes and actuated by a high-frequency direct current. The illumination of small surface areas of the eggs was brought about by the use of a perforated tinfoil diaphragm. The eggs measured 1.7 mm. in diameter, while the perforations varied from 0.3 to 0.4 mm. in diameter. Consequently, but a comparatively small surface area was influenced in any one experiment. The intensity of radiation was such that an exposure of from twenty to thirty seconds was sufficient to produce developmental defects. The area studied by this method and reported in this paper extended from the region of the equator up to the animal pole, but was exclusive of the pole itself and of a narrow median strip connecting this pole with the equator. The developmental end-product was, in every instance, what is best described as a U-shaped embryo of that general type in which the two folded body-parts lay upon the same horizontal plane. A singular result of the study of the sectioned specimens, in addition to the striking feature of the uniformity of production of thin type of deformity, is the absence of defects in the anatomy of the developed tadpole. Comparatively little has been ascertained concerning the chemical effect brought about by the action of the rays upon the egg mass. Notwithstanding, the conclusion was drawn from studies upon these defective eggs during the process of development that the chemical composition of the cell-mass thus radiated was so altered that it lagged behind in what might be termed the chemical development of the ovum and though ultimately participating in the developmental processes exercised a retarding influence upon some of the gross mechanical shiftings of embryonic anlage. The altered mass, located as it was lateral to the definitive median plane of the embryo, retarded, in the first place, the medianward shifting of the neural tube-half on its own side and induced, in the second place, an exaggerated migration of the opposite tube-half medianward and beyond, towards the affected area, before coaptation with the neural tulxvhalf of that side with the production of a whole tube could be l)rougiit about. PROCEEDINGS 325 7. The development of the serous glands (von Ehner's) of the vallate papillae in man. E. A. Baui\u;autner, Washiiifrton University Medical School, Serial sections of the vallate papillae in fetuses and new-born form the basis for this i)aj)er. Wax rcconstnictions of the papillae in the smaller fetuses W(>re made and studied. In a reconstruction from an 8.5 cm. fetus (crown-rump length) the epithelial down-growth forming the Innits of the papilla is complete; growing inward from this are three well defined solid bodies, the early gland anlagen, together with less well defined outgrowths. These gland anlagen in a 10 cm. fetus present enlarged ends and slightly constricted stalks, the beginnings of terminal acini and ducts of later stages. The papilla in front of the foramen caecum has many glands extending both from the lower margin and lateral walls of the circumscribing epithelium; at the caudal end of the papilla these glands project backward and downward. The enlargement of the ends and constriction of the stalks is more pronounced in this papilla than in those fmther forward. The glands of a lateral papilla of an 11.5 cm. fetus are more numerous in connection with the lateral side than with the lower margin of the circumscribing epithelium, whereas in a 12.5 cm. specimen glands extend downward and backward from the lower border alone. A wax reconstruction of a right anterior papilla shows the glands to be more numerous and longer than in papillae located further caudalward. In one instance a gland is divided near its origin, one of the branches being subdivided. In several instances in a 15 cm. fetus the enlarged oval ends of glands have divided into stalks and end pieces. Some of the stalks are hollow at their origins from the trench which is now present. The trench surrounding the suiface of the papilla is well defined although the shape of the papilla can be readily recognized in younger stages by a slight depression. A model of a papilla made from a 19 cm. fetus shows the trench well developed and gland ducts extending from its lower rim and lateial walls. One exception was noted, namely a gland duct arising from the inner wall of the trench. Some glands are very short, appearing either as knob-like outpouchings or possessmg a short, well developed duct and an expanded but still sohd end. The ends in some glands are beginning to show superficial subdivisions separated by grooves, indicating acini. Two or three very long ducts extend downward and outward on the lateral side of the papilla at different levels that give off short tubules which terminate in acinar outgrowths, some hollow, others sohd. Some ducts divide immediately into two to four branches. No anastomosing ducts were found. Older specimens (23 cm.) showed ducts spreading laterally, others extending deeply into the muscular tissue before breaking up into glandular tissue. 326 AMERICAN" ASSOCIATION OP ANATOMISTS 8. The weights of the organs in relation to type, race, sex, stature and age. (From 115 post mortem examinations at the Charity Hospital, New Orleans, La. Prehminary report.) Robert Bennett Bean, Anatomical Department, University of Virginia. . The types may be classified in three groups, the hyper-ontomorph at one extreme, the meso-onotmorph at the other, and an intermediate in between. The organs are invariably smaller in the hyper-ontomorph than in the meso-ontomorph, and this difference may be expressed by a number that represents the relative number of large and small organs in each group. This number is a factor that represents the ratio of difference between the hyper-ontomorph and meso-ontomorph in relation to the organ. TABLE 1 The size of the organs in relation to type LIVER KIDNEYS BRAIN SPLEEN HEART APPENDIX PANCREAS 6.14 4.10 4.50 4.70 19.00 3.38 4.50 There is greater difference in the heart of the hyper-ontomorph and the meso-ontomorph, and less difference in the appendix. The difference is about the same for the kidneys, brain, spleen and pancreas, but greater than these for the liver. The differences due to race are determined for the wliite male and the negro male only. These differences are less than the differences due to type, except in the pancreas. The size of the TABLE 2 organs in relation to race LIVER KIDNEYS BRAI>f SPLEEN HEART APPENDIX PANCREAS 1.00 0.12 2.33 2.00 0.40 0.50 4.50 The organs are larger in the white male than in the negro male. The numbers given in this and all the tables are obtained by the same method, and indicate relative differences in size. The number is a rough ratio of difference, and any two may be compared with each other in any of the tables. The differences due to sex are determined for the negro male and the negro female only. These differences are less than those due to type except for the kidneys. TABLE 3 The size of the organs in relation to sex LIVER KIDNEYS BRAIN SPLEEN HEART APPENDIX PANCREAS 2.86 3.40 0.60 2.00 14.40 1.30 1.50 PROCEEDINGS .S27 The ()rt>;niis arc larj:;or in ilio iiopiro in;ilo tliaii in i.lio iioffro female. The ilift'cnMicc is <;;r(>att's( in llio luniit, and least in tlic^ brain. 'Vho (lilTcrences due to age are less than the differences due to type except in the s])Ieen. TABLE 4 The size of the organs in relation to age LIVER KIDNEYS BH.VIN HEART APPENDIX PANCREAS SPLEEN 3.50 0.000 1.40 5.15 0.20 0.50 7.75 The greatest difference due to age is in the spleen, and the least in the Iddneys. The heart and liver also show differences due to age. More small than large livers, spleens and appendices are found in the old than in the young, and more largo than small brains, hearts and pancreases are found in tlu; old than in the young. The brain and heart continue to grow in size with increasing age, whereas the liver and spleen atrophy. The differences due to stature are considerable, especially for the brain and appendix, in which the differences are greater for stature than for type. The differences in the heart due to stature are unexpectedly small. TABLE 5 The size of the organs in relation to stature LIVER KIDNEYS BRAIN SPLEEN HEART APPENDIX PANCREAS 4.85 3.17 13.68 4.87 1.00 6.27 2.40 After all, however, the test of the value of these differences lies in their application to the individual. Is it possible to approximate the size of any organ in the living, if one has the type, race, sex, stature and age of the individual? Let us try a few cases at random. Here is a male negro hj^per-ontomorph, aged 52, stature 165 cm. Is the liver large or small? The liver of the negro male should not be so large as that of the white male nor so small as that of the negro female, therefore it should be intermediate in size by race and sex. The age and stature are both intermediate, therefore the type is the most distinctive characteristic. This individual is a hyper-ontomorph and the liver of the hyper-ontomorph is small, therefore the liver of this individual should be small. It weighs 1150 grams which is 350 grams less than the weight of an average or intermediate liver. Here is a negro male meso-ontomorph, aged 31, stature 169. Is the liver large or small? The race and sex may be disregarded, as before. The stature is only a little above the intermediate, therefore it may be disregarded. The age and the type both indicate a large liver, and the actual weight is 1580 grams. 328 AMERICAN ASSOCIATION OF ANATOMISTS Here is a white male hyper-ontomorph, aged 50, stature 157. Here the Hver should be small but it weighs 2270 grams. The pathological record reads, "amebic abscess of liver," "syphilis of liver." Either one would make the liver weigh more than normal. Here is a white male meso-ontomorph, aged 56, stature 170. The liver ought to be large and it weighs 2020 grams. A negro female hyper-ontomorph, aged 36, stature 150, has a liver that weighs 1480 grams which is a little larger than expected. A negro female meso-ontomorph, aged 18, stature 173, has a liver that weighs 1750 grams, which is about what to expect. I have gone over the records of the 115 post mortems in this way and in only a few cases was there no explanation of the weight of the organ. If all the factors in each case could be known the weight of each organ might ba determined with a fair degree of accuracy during the life of the individual, and the most important factors for such determination are given above. 9. A case of a persistent vitelline vessel in a human adult. Alexander S. Begg, Harvard Medical School. . Attention has repeatedly been called to the persistence of the vitelline vessels as slender cords in certain mammals, particularly in Carnivora, but strange as it appears, reports of cases of such persistence in the human adult are very few. In the time at my disposal, I have found reference to but two cases comparable with the present one. Of these, the first was that described by Spangenberg in 1819, and the second by Hyrtl in 1859. Spangenberg describes his case as showing a vem, while Hyrtl's case, in a child, showed an artery and an accompanying vein. The above papers, together with those of Meckel and Fitz, that of Broman and the more recent exhaustive work on "The umbiHcus and its diseases" by CuUen, have been consulted. The present case was found in an adult female subject in the dissecting rooms of the Harvard Medical School and was kindly given me by Professor Warren for further study. Upon opening the abdomen a strand was noted which ran free from the inner side of the ventral body wall, 5 cm. below and slightly to the right of the navel, down into the pelvis and up amongst the coils of the intestine. The strand was 30 cm. in length and throughout the greater part of its extent was only of the thickness of a coarse thread. The inner end of the strand was attached to the ventral surface of the mesentery at the level of the sacral promontory. Upon taking that portion of the ileum which is 60 cm. from the colon and drawing it forward, the free strand was seen to enter the mesentery midway between the intesthial tube and the dorsal attachment of the mesentery at the level of the promontory. From the mesentery the strand ran downwards crossing in front of the ileum and to the left of its terminal portion, and then turned upwards to join the ventral body wall as stated above. The cord was visible beneath the parietal peritoneum and was seen to terminate in relation to the obliterated right umbilical artery, near the umbilical ring. At PROCEEDINGS 329 both Tii(*s(»nt(M'ic :iii<l umbilical oxtrciuitics tho straiul hccamc dociclcdly tiiiokoiiod. Tlnis if the slciulcr luitldlL' ])()rti()n had l)0(;<>iiu; ohlit(!rated, thoro would have reMuained an apjx'iidix uieso-ilei, 3 cm. Ions, and an appendix imibilicalis of 12 cm. (nsin<^ th(^ terminology pro])ose(I by Bronian). Tho intestine showed no diverticulum, but the strand was directed towards a point on the iloinn approximately 500 cm. from the pylorus and 60 cm. from tho colon, a region whore the remains of the vitelline duct might bo expected. There was no trace of an appendix mesoduodeni (Broman) to indicate the former position of the vitelline vein. In fact, no other abnormalities were observed except certain pathological adhesions antl the tissin-es on the under side of the liver indicated that tho umbilical veins had been normal. Upon injection of the superior mesenteric vessels a small i)ranch of the artery was found to enter the proximal end of the strand, but the mesenteric vein showed no corresponding branch. An attempt at injection of the distal end of the strand, through the epigastric artery, was unsuccessful. In HjTtl's case there was an anastomosis between the persistent vitelline artery and the deep epigastric artery, but such a connection could not be demonstrated in the present case. An examination of the models made by Dr. Papez seems to show that the strand in the specimen under discussion represents an uncomplicated persistence of the sheath of the vitelline artery, together with a portion of the vessel itself. In Hyrtl's case the artery in a corresponding strand was accompanied bj^ a branch of the superior mesenteric vein, which w^ould seem to be a new formation rather than a persistent vitelUne vein. In Spangenberg's case a strand apparently in the same situation is said to have contained a vein only, but this does not accord wdth normal embryological conditions. The specimen here described is therefore particularh' satisfactory since it can be fully interpreted embrj'ologically. 10. Vestigial gill filaments in chick embryos. (Lantern.) Edward A. BoYDEN, Harvard MecUcal School. Since the researches of von Baer, the brancliial region in the Amniota has interested biologists as supplying the most conspicuous evidence that the higher vertebrates recapitulate, in modified form, stages in the life-historj' of their ancestors. Apparently no record has yet been made of structures on the branchial arches of higher vertebrates which could in any way be interpreted as functional or rudimentary gill filaments. While studying the anatomy of the S-daj^ chick my attention was attracted to ectodermal proliferations, protruding from behind the hyoid arch, which seemed to be involved in the obliteration of the cervical sinus. To Prof. F. T. Lewis I am indebted for the suggestion that these projecting cell clusters might be brought into Une with the gill filaments of amphibians and fishes. Subsequent stud}^ of older and j^ounger chick embryos, together with the finding of similar structures in turtle embryos, seems to warrant a presentation of this material from a phjdogenetic standpoint. 330 AMERICAN ASSOCIATION OF ANATOMISTS The life historj'- of these filaments, covering a period from the fourth to the eighth da^^ embraces nearly one fifth of the total period of incubation. Throughout this time the epithelium of the filaments themselves as well as the branchial epithelium which gives rise to them is characterized by the presence of what appear to be degeneration vesicles. These accompany, and thus may be said to register, an activity of the epithelium of which the filaments seem to be the fruition. They first appear, as early as the 76-hour stage, in the posterior walls of the first three pharyngeal pouches at a time when the first three gill clefts have broken through and the fourth pouch touches the ectoderm. When complete, each vesicle is a clear, spherical cyst embedded within the epithelium, containing pycnotic nuclei and cellular fragments. Favorable sections indicate that these cysts result from the nearly simultaneous disintegration of several adjacent epithelial cells. Following their first appearance, scattered vesicles may arise in the walls of all the pharyngeal pouches and in the ectoderm between the clefts on the outside. About the end of the fourth day they are most numerous in the ectoderm between the second and third clefts and are sufficiently abundant to give it a punctate appearance. Eventually some of them are crowded downward into the underlying tissues. A second, less conspicuous concentration area occurs in the ectoderm between the third and fourth clefts. It is of interest to note that it is the ectoderm of these two arches in frogs and toads which forms the first external gills. Again, recalling the fact that vesicles in the chick first appear in the entoderm, it is well known that a proliferation of entodermal cells in the anura precedes the formation of the gills, but instead of producing vesicles as in the chick, tends to spread out beneath the ectoderm as a secondary layer. At the beginning of the fifth day the ectoderm covering the third arch, where the vesicles are most abundant, has become considerably thickened and is beginning to produce filaments. As viewed from the outside this arch is wedge-shaped and its downward directed point is the first part to develop filaments. Later the upper portion will also give rise to tufts of cells. In a 12-mm. embryo slightly older than the last, the middle of this wedge actually forms an outpocketing which contains a mesodermal core, thus almost reproducing the early formation of external gills in the Amphibia. In this same embryo the ectoderm of the fourth arch forms an evagination, which however is solid and much smaller, and tends to fuse with that from the third arch. During the remainder of the fifth day the hyoid arches meet in the mid-ventral line and begin to grow backward over the posterior arches, as in the Anura, in the form of a gill-cover or operculum. It carries with it, on its under surface, the tufted epithelium of the third arch, and from now on the filaments of this region of fusion will appear to come from under the surface of the operculum. The filaments attached to the operculum will always be the largest of the series and persist the longest. Differentiation lateral to this point on each side continues PROCEEDINGS 331 slowly, \iiitil, toward the end of the sixth (hiy, there is a transverse line of filaments on either side extending in a dorsal direction part way across the neck. In the next few hours this may he Hiipi)lementcd by a ventral extension of filaments along the under surface of tha operculum until in some specimens they nearly reach the mid-ventral line. It is al)out this time that the larger filaments regularly show l)ranching. This ]uM-iod of maximmu extent in the first quarter of the seventh day is followed by a rath(>r rapid decline, imtil, at the end of this day, all trace of opercuhnu and filaments has disappeared. Coincident with this decline is a curious median proliferation of epitrichial cells in the cardiac region. This is being studied further, but apparently has no connection with the subject under discussion. By way of a sunmiary the life history of these vestigial structures may be divided hito five stages : 1, the appearance of degeneration vesicles in the branchial epithelium; 2, the concentration of these in the ectoderm covering the third arch and, to a lesser extent, that covering the fourth; 3, the thickening of the ectoderm of these two vesiculated areas into tufted epithelial mounds, and, in the case of the first, an apparent evagination of the ectoderm with a mesodermal core; 4, a gradual differentiation of these areas (now crowded into one and fused with the sides of the backward growing operculum) into a transverse line of filaments on each side of the neck; 5, a rather rapid reduction of this line and the eventual suppression of both filaments and operculum. In conclusion, attention should be called to the recent experiments of Ekman ('13) which have a bearing on this problem. He was able to show^ by transplantation methods that the ectoderm of the branchial region of frogs and toads has a certain specificity for building gill filaments not possessed by the remaining ectoderm of the embryo; that a polarity of this ectoderm could be demonstrated; and that even when the entoderm and mesoderm underlying the future gill region were removed the ectoderm alone could produce abortive filaments. It is the ectoderm of this same region in reptiles and birds which produces rudimentary filaments and they bear a striking resemblance to some of the abortive structures produced experimentally in Amphibia by Ekman. In the case of higher vertebrates the process never passes beyond the initial stages, as evidenced by the early appearance of degeneration vesicles and the failure of blood vessels to participate in gill formation. 1 1 . Development of the preoptic part of the forebrain of Amia calva. Chas. Brookover, Universitj' of Arkansas, Little Rock. In connection with a study of the adult brain of Amia it was thought best to model some stages of the development of the forebrain in order, if possible, to bring it and the closely related divergent forebrain of the teleosts into closer harmony with the ancestral vertebrate brain. The early stages, as expected, show a simpler and more diagrammatic condition than the adult. THE ANATOMICAL RECORD, VOL. 11, NO. 6 332 AMERICAN ASSOCIATION OF ANATOMISTS The earliest stage modeled is 5 mm. long, taken at a time just iDefore hatching. The embryological condition of an epithelial tube exists. It is bent so as to throw the pineal anlage to the anterior and is but slightly thickened in two places on either side of its basal part. The anterior swelling into the neural canal is beneath the fibrous connection with olf actor}^ placode and goes later into the formation of the olfactory bulbs. The posterior one is the derivative from which comes the remainder of the preoptic forebrain. A fibrillar zone at the periphery permits mapping out the anterior commissure, the olfactory tracts, the optic tracts and perhaps the habenular tracts (fimbria). On the ventricular surface the slightly thinner pallial portion is marked by upper and lower parallel lines running just dorsad of the lateral olfactory tract. The ventricular sulcus bends caudoventrad to flatten out and be lost in the diverticulum of the optic stalks. Larval stages of 8 and 10 mm. total length show a previous rapid development of the olfactory centers. The thickness of the olfactory bulbs shows in an external swelling and an internal rhinocoele has been developed. Neuroblasts for the formation of mitral cells have migrated toward the periphery. Posteriorly in the olfactory lobes the cells for the formation of the lateral olfactory area have proliferated. This makes a dorsal swelling. Ventrally the cells keep for a longer time their original position and largely epithelial condition along the walls of the common forebrain ventricle. This ventricular lumen extends forward somewhat beneath the olfactory builds but not so far as in the earlier stages where it produces what has been called an unpaired olfactory placode. The above mentioned cells along the ventral portion of the forebrain ventricle are continuous without line of demarcation into the rhinocoeles and here the cells of the nucleus olfactorius anterior of authors are proliferated. In the forebrain this undifferentiated epithelial zone extends posteriorly over the commissure into the thalamus. The medial olfactory tract somewhat belated in its development comes into relation with the anterior part which develops into the nucleus of the precommissural body. This is not exclusively olfactory in function. The nervous terminalis passes through it and the ascending fibers to the olfactory bulbs originate from it in some forms. Golgi preparations of larval Amia at this age show ascending as well as descending fiber connections between this region and the thalamus. Fishes of 22 mm. and of 47 mm. length have a ventricular sulcus e^itending from the rhinocoele dorsocaudad to the neighborhood of the velum transversum. This sulcus (limitans telencephali) becomes less evident with age and the pressure of opposite sides of the hemisphere on each other. This pressure is due to the large lateral and medial olfactory areas lying dorsal and lateral to the sulcus, as well as to the striatum (paleostriatum) lying laterally. At the anterior end of the sulcus limitans the cells of the islands of Calleja originate and migrate laterally into position between the two olfactory tracts. Adjacent and continuous with these within the riiocEEDiXGs 333 olfactory lmll)S the miclcus olfactorius anterior proliferate. Tiie olfai'tory lnill)s remain in eiui)ryonic (])riinitive) i)roxiinity to the forebrain. In its nii<l(lle jiortion the cells of the region just ventral to the sulcus liniitans j:;ive rise to the i)al('ostriatuni. At its caudal end near the vehuu transvcrsuni three nuclei connected to the hahenula oripnate from the ventral edge of the sulcus, viz., the nucleus theniae carried laterally in the eversion of the hemispheres, the nucleus lateralis commissural is (Sheldon on the carp), and a thalamic nucleus remaining ill its jn-imitive position near the ventricle. The second may be connected only with the ventral lobe of the habenula. The forebrain of Amia is primitive in the proximal position of its olfactory bulbs and the location of its nonolfactory centers near the ventricle but specialized in its hypertrophied olfactory nuclei with strong habenular connections. 12. Histological differences between certain muscles of the cat as related to physiological and chemical differences. (Lantern slides.) H. Hays Bullard, Pathological Laboratory, Johns Hopkins University. The object of this paper is to call attention to some morphological differences between certain muscles of the cat which are believed to correspond to a number of the physiological and chemical differences in the same muscles set forth by Lee, Guenther, Meleney, Scott and Colvin in a series of interesting papers appearing in The American Journal of Physiology, May, 1906. In one of these papers (Lee, Guenther and Meleney) it is mentioned that the authors could find in the literature no comparative histological study of the muscles under discussion, namely, diaphragm, extensor longus digitorum, sartorius, and soleus. From their own histological observations they were not able to point out any striking differences between the given muscles. In the present experiments these four muscles from twelve cats (young and adult) have so far been examined. The tissue was fixed in 20 per cent formalin and frozen sections were stained with Sudan III by Herxheimer's method. This method is well adapted to the demonstration of the dual structure of muscle as seen in the so-called hght and dark or cloudy fibers. The occurrence of an intermixture in varying proportions of these two tj'pes of fibers in the skeletal muscles of many animals, including probably all vertebrates, has been known for a long time (Knoll). As is also well knowm light fibers usually have a few small fatty droplets in their cytoplasm. Dark fibers have many larger fatty droplets. Intermediate fibers also occur which are likewise fatty and probably belong to the dark type. The four different muscles of the cat may be characterized, briefly, as follows: Diaphragm. Light and dark fibers in about equal number, small number of intermediate fibers. Average diameter of fibers less than in other three muscles. Extensor digitorum. Light fibers form more than half the total number, dark and intermediate in about 334 AMERICAN ASSOCIATION OF ANATOMISTS equal number. Average diameter of fibers greater than in diaphragm and less than in soleus. Sartorius. Dark fibers predominate, intermediate and light fibers in about equal number. Average diameter of fibers greater than in diaphragm and less than in soleus. Soleus. Intermediate fibers predominate, dark fibers in considerable number, light fibers either absent or of very infrequent occurrence. Average diameter of fibers greater than in the other three muscles. In all four muscles light fibers are of greater diameter than dark. When these four muscles, only, are considered the microscopical picture of each is usually sufficiently characteristic to permit of its identification. From the predominance of intermediate and dark fibers it is always possible to identify the soleus, in transverse section, from a single low power field. The predominance of light fibers serves for the easy identification of the extensor digitorum. The diaphragm and sartorius are frequently identified only after several fields are examined and, exceptionally, one of these muscles is mistaken for the other. The extensor digitorum and sartorius are macroscopically pale muscles. The soleus is a red muscle and the diaphragm is intermediate in color. It is not true, however, that the dark fibers of the soleus are the cause of the dark or red color of that muscle. Dark fibers have their characteristic appearance only when seen by transmitted light under the microscope. By reflected light they appear light. It follows that dark fibers tend to make the muscle appear pale and the dark red color of the soleus is not due to its dark fibers. Also it is known that many pale muscles are composed largely of fibers that are microscopically dark. The extensor digitorum, according to Lee, is characterized by great irritability, a short latent period and quick contraction and relaxation while the soleus is less irritable and after a longer latent period it contracts and relaxes slowly. As mentioned above the histological picture shows a great predominance of light fibers in the quick muscle* (extensor) and a predominance of intermediate and dark fibers, with almost total absence of light fibers, in the slow muscle (soleus). The diaphragm and sartorius occupy an intermediate position both in these physiological properties and in the relative number of light and dark fibers. It is not probable that these facts are unrelated. In the present experiments the histological findings in respect to the relative fat content of the four muscles agree with the chemical analyses of Lee. The entensor digitorum, both chemically and microscopically, shows considerably less fat than do the other three muscles. Moreover the quantity of fat microscopically demonstrable in each of the muscles is so great that it would appear to account for the entire quantity shown by chemical analysis. This is not in accordance with the accepted view that much of the fat in the tissues is chemically combined and not capable of microscopical demonstration. In fasting animals, Lee has shown that the total working power of the diaphragm is reduced by 44 per cent while the working power of PROCEEDINGS 335 the oxtonsor is rcMluccd by 10 ]n)v cent. It is known that f:it {z;ra(huilly disappears from tlie fatty dark imiscic fibers of fastinfj animals (Knoll, Bell, Billiard). As we have seen the diaphraj^m of the cat contains many more fatty fii)ers than does the extensor. When the fat is removed i)y starvation one would expect the diaphragm to underf^o a greater ])ro|)oi tional re(hu'tion of working power. It is scarcely necessary to mention that the microscopical differences between the four muscles under discussion are not here considered exhaustively. Interesting details respecting nuclear and myofibril characters, occurrence and distribution of mitochondria and glycogen, blood and nerve supply, and connective tissue distribution, all remain to be investigated. A study of the length of the muscle fibers by the methods recently described by Huber might also prove of considerable interest. There is good reason to believe that the striking physiological and chemical differences which Lee and his collaborators have demonstrated in these four muscles are paralleled by no less remarkable morphological differences. 13. Some factors regulating growth. Moxtrose T. Burrows, Department of Pathology, Johns Hopkins University. The problems under consideration in this paper are (1) the nature of the immediate conditions which lead to the failure of scar formation in many woimds or following many extensive inflammatory processes and (2) the general nature of the conditions which inhibit or allow the growth of connective tissue. It is well known that the most extensive inflammations of epithelial surfaces, as pneumonia, are most often followed by complete healing, while inflammations of the deeper connective areas are most frequentlj' followed by the formation of a scar. In cancerous processes the connective tissue cells grow wi'dly at the expense of other parts. Hertzler, a few years ago, came to the conclusion that the fil)rmous exudates which forms in a wound, is the direct stimulus for the .lowth of the connective tissue cells. He noted that skin grafts take only when they become imbedded in a layer of coagulable exudate. He also thought that the fibrin fibrils were transformed directly into the extracellular connective tissue fibrils. He had come to this last conclusion by means of a careful chronological study of intestinal adhesions and wounds induced bj^ mechanical means in young rabbits of the same litter. He noted that previous to healing the fibria is laid down in the form of fibrils. Connective tissue cells migrate amo g these fibrils. At a later period he found the fibrous tissue -ri'.s t ) occupy the same position and have the same arrangement as the fiijrin fibrils. The wounds and adhesions had been removed at reg ilar intervals, sectioned and stamed. He was unable to see any evidaaca of the disappearance of fibrin, and the laying down of the c"> : .active tissue fibrils. Similar experiments were also made with W'lids In the summer of 1908, the author had the opportunity to sv.dy these experiments with Dr. Hertzler. 336 AMERICAN ASSOCIATION OF ANATOMISTS In the early studies of tissue culture made at Cornell University Medical College, it was noted that the fibrin fibrils in many of the cultures, after a considerable growth of connective tissue cells, stain the characteristic pmk color, of white fibrous tissue with Von Gieson stain. These observations were communicated to Dr. Hertzler, who reported them with his own studies ('13). Recently, Baitsell ('16), in Harrison's Laboratory, has made similar observations in tissue cultures. He did not observe the characteristic color reaction with Van Gieson stain. Whether or not these pink staining fibrils that had been observed in the tissue culture could be considered as true fibrous tissue, or merely structures simulating these fibers in their ability to absorb dyes, was a problem of interest. One of the objections to accepting them was the inconstancy of the appearance of pink staining fibrils in many of the cultures. In later studies of the growth of tissue in vitro several facts have been found, however, which would tend to substantiate this particular view. The first is that clot contraction or the formation of fibrin fibrils in the cultures of chicken tissue in plasma occurs only in the presence of connective tissue cells. Chicken plasma, when carefully prepared, free from previous tissue contammation, clots with the addition of a fragment of tissue to form a practically structureless jellylike mass which has the same volume as the original fluid plasma. Any type of tissue conditions this primary clotting. In the presence of living connective tissue cells the clots later undergo contraction, while in the presence of epithelium they may undergo contraction but later liquefaction. Leucocytic or lymphocytic cells cause only slight Hquefaction of these clots and very little, .if any, contraction. The second fact is that this contraction takes place only in the presence of living connective tissue cells and then only after a considerable latent period. It fails entirely when the oxygen is replaced by nitrogen or hydrogen. It also fails when the tissue fragments have been heated for five minutes at 60°C. In other words, it is evident that clot contraction is a phenomenon instituted by conditions quite different from those of primary clotting and it is a phenomenon which is brought about by the action of the products of metabolism of the connective tissue cells. At another time the author studied more carefully the properties of the connective tissue cells. He has found that the cells of higher animals are not highly organized, but fluid-like systems. Their various manifestations, of life such as movement, growth, etc., are differential surface tension phenomenon under the control of a specifically organized environment. The food materials or energy producing substances in the cultures are not derived from the medium, but from the cells within the fragment. The growth that one observes in the cultures is none other than a simple transfer of materials from the cells of the center of the fragment or in a less favorable environment to those on the periphery or those which have been carried out into the medium through the interchange of substances between the fragment and the PROCEEDINGS 337 ino(li\im. This was shown by the fact that the cells can ho srown in snnple salt solution and, in the plasma cultures, growth cease after a few transplants, the sum of the total {growth bein^ les.s than the original mass— or it represents what on.^ niisht assume the original mass mnnis the enerjry of transfer. The cells that tend to hieak down in the Ira^inent and lead to the greater growth of the connective tissue cells are not the connective tissue cells but the e})ithelium, muscle cells, etc. Agam It was noticed that this growth takes place only in the presence of oxygen. It conuuences after a given latent period in the case of connective tissue, subsequent to the contraction of the clot). Ihe cells grow actively for a time gradually to come to rest. This reaction is one which apparently commences subsequently to the slow ditlusion of substances between the fragment and the medium and proceeds until an equilibrium is established. In other words, it follows the curve of reaction of a heterogeneous physical chemical system. 1 he cells at the end of this reaction do not undergo, at least for a considerable time, any further change. They show no immediate disintegration. That this cessation of growth is not due to the exhaustion of oxygen or food materials is further shown by the failure ot any change m the cells following the introduction of fresh air into the cu ture chamber, and by the fact that activity is again seen when the cells are transplanted to a new culture medium. On the other hand, that it is due to the accumulation of waste products is shown by the tact that the cells which tend to survive for the longer periods of time are invariably those cells which had grown out into the clot and have become completely surrounded by the contracted fibrin. It was ot general interest m making these observations to note that this equilibrium which had been established in the tissue culture did not alone concern the cells which had migrated out and grown in the culS-?uTu ^^"^' . likewise those which remained within the fragment With the cessation of growth of cells in the outer medium, disintegration with the liberation of energy-producing substances in the fragment a so ceases This is especially true when the fragments have been placed m thick layers of plasma, so that they have become likewise completely surrounded by contracted clots. Many such cultures were kept for as long as six months at incubator temperature and in an ample supply of oxygen before any disintegration became apparent, beveral were transplanted at this time and an active growth of cells was observed. The growth of the connective tissue cell is apparently a tissue and not a cellular reaction. The failure of the connective tisfhjfLA t^sol^-e the fibrin and their abihty to transform it into fibrils, leads to the behef that these fibrin fibrils foim the superstructure upon which or out of which, the connective tissue fibrils are built V\ e observed pink stamingfibrils only in cultures of skin of foetal chickfu VV^ethe^' tl^e formation of the connective tissue fibril is a bodv rather than a connective tissue, cell reaction is a question for solution It is^well knovvm that the growth of cells in the animal organism IS not determmed alone by food and oxygen but bv other unknown 338 AMERICAN ASSOCIATION OF ANATOMISTS conditions. The question naturally arises, have these unknown conditions been found. Are the actual waste products of metabolism of these cells, substances which are insoluble in body flaids and is the cessation of growth of a part, the result of the accumulation of these substances? One may assume that the fibrin fibrils are formed by the action of the products of metabolism of connective tissue cells on the coagulable exudate, and that the cessation of growth in the wound is due to the accumulation of these substances in and about the growing cells. From these observations, one might readily reduce stimulation as any condition which would lead to the reduction of concentration in these substances. The stimulating action of fibrin is due to its ability to absorb these substances. To prove this more completely the author studied rhythmical activity in heart muscle cells as well as the growth of these and of connective tissue cells in plasma cultures so arranged that the media could be conthmously washed with a stream of serum. The rhythm of heart muscle fragment in simple hanging drop cultures is invariably intermittent. In the body it is a form of activity which continues throughout the life of the individual. In the cultures, where the media was continuously washed by a stream of serum, the rhythm was not only greatly prolonged up to the time of complete exhaustion of the cells but it continued regular, while the growth of the cells was not changed but similar to that seen in the simple hanging drop cultures. In a former paper before this Society, the author presented facts to show that the contracting, embryonic heart-muscle cell has an organization which one might readily assume capable of splitting these insoluble waste products into simpler substances and of transforming the energy liberated with their formation into work of contraction. Certain rapidly growing tissues, such as embryonal and rapidly growing tumors grow readily in liquid medium. This growth takes place only near the surface of the liquid. Cells suspended in liquid invariably round off and show no activity. The cells do not grow out into the Uquid. It is of interest to note that adult tissues do not, however, grow readily in liquid medium while, on the other hand, they grow actively in plasma. It was in the light of these facts and the more careful study of the properties of epithelial as well as leucocytic and lymphocytic cells that the general deductions as to the cause of the faihire of scar formation in superficial inflanmiations of the epithelial surfaces was derived. Epithelial cells invariably bring about a rapid dissolution of the fibrinc lots. When occurring in considerable numbers in a fragment of tissue, they invariably prevent entirely a growth of the connective tissue cells. That the appearance of organization in the pneumonic lung probably indicates a complete destruction of the epithelium rather than the lack of leucocytes in the exudate was further suggested by the fact that the leucocytic infiltration in deep seated inflammation is frequently as great as in the superficial ones. It is true that the leucocytes of man are licher in ferments than those of PROCEEDINGS 339 lower Hiuiiials; the failure to observe any extensive liquefaction about the leucocytes of chickens would not indicate that this difl not occur in Ininian beings. On the other hand, it has been fo\ind that frap;nients of human connective tissue containinfr leucoc>^es grow readily in plasma clots when they are removed after twenty-four hours from the first culture to a droj) of fresh plasnia. This is not true of epithelium. The cells continue to liquefy the plasma, even after many transplants, or until they are dead. These observations are reported not onlj'- for the general bearing that they have on the nature of stimulation and the significance of extracellular substances in their relation to life processes, but also for their immediate significance for the better understanding of the actual conditions which regulate the growth of body cells. If these experimejits are sul:)stantiated, showing as it is believed they do, that growth is inhibited by the accumulation of insoluble waste products and permitted to proceed only by their removal, then the problem of the growth of the cell is brought into the domain of chemistry. Thus problems, such as those that confront us in cancer, are narrowed. 14- Preliminary report on the normal unequal growth and degeneration in the early ossification centers in the diaphyses of femora of the pig. Eben Carey, (introduced by F. W. Heagey), Department of Anatomy, Creighton ^Medical College. The normal unequal growi,h of the subperiosteal osseous tissue, and of the degeneration of the hyaline cartilage cells and matrix in the early advancing diaphyseal center of ossification, as in the femur of the pig, have not been made an intensive problem by investigators. As a consequence descriptions of the early development of the diaphyses of long bones are so worded as to avoid clear statements on certain fundamental points. This paper w^ill be limited to the time when the primary subperiosteal osseous lamina, or as it here w^ill be designated, the lamina prima, is fairly well differentiated. 1. Material and methods. The pig was selected for this problem on account of the abundance of material procurable by the proximity of the laboratory to the South Omaha abattoirs. The femur was chosen as a prototj-pe of long bones with two epiphj^ses; it was also chosen because the ventral, lateral and mesial walls of the diaphysis of the adult femoral bone primarilj^ present a laminar type of formation and not the characteristic Haversian system tj-pe as found in the human femur (Foote '11-' 16). This report is a"^ part of a more extended study considering the complete embrs-olog}- of the pig's femur. The embryos were hardened by both the alcohohc and Zenker's fixation methods. In the former no decalcification takes place and it was used to ascertain the time when the precipitation of calcium salts precludes sectioning. By the latter method the earliest center of ossification is easily decalcified without subsequent subjection to a stronger decalcifjdng agent. The center of ossification of older bones, 340 AMERICAN ASSOCIATION OF ANATOMISTS after Zenker's fixation, are prepared for sectioning by von (Ebner's) decalcifying fluid which was used with excellent results. With the latter reagent less distortion, due to the swelling of the collagenous fibers, took place. Both cross and longitudinal serial sections were prepared of the femora. Some of the sections were stained with Delafield's haematoxylin counterstained with alcoholic eosin, others were well defined with IMallory's connective tissue stain. 2. Inception of period of ossification. In embryos with a crownrump measurement of between 20 to 22 mm. the cartilaginous femur is approximately 2 mm. in length. It is well outlined, shows certain adult characteristics, and cavity formation has just begun in the tissue lying between the cartilaginous floor of the acetabulum and the head of the femur. In the first step of the formation of the cavity at the knee joint, there is a condensation of the capsular tissue immediately bordering upon the joint and of the perichondral tissue which at this stage covers the cartilages on their articular surfaces as well as elsewhere. The cartilage of the shaft is of the cellular variety and presents an epithelioid appearance. At either extremity the precartilaginous cells predominate. It is at this stage that the first changes of osteogenesis are noticed. There is an unequal growth of the osteogenetic cells of the Cambium layer of the periosteum, which encircles the shaft. More cells are proliferated and the constriction is slightly deeper on the ventromesial aspect than elsewhere. The cellular cartilage begins to appear slightly atrophied and vesiculated, as evidenced respectively, by the more shrunken granular cells and by a few hollow vesicles, immediately underlying the zone of more vigorous proliferation of osteogenetic cells. The osteoblasts ultimately form the lamina prima or the first embryonic osseous layer encircling the degenerating cartilaginous shaft. This process also begins on the ventro-mesial aspect. The unequal growth of this lamina prima was followed through the 3 mm., 3.5 mm., 4 mm., 4.5 mm. and 5 mm., lengths of the femur, in embryos with a respective crown-rump measurement of 28 mm., 30 mm., 34 mm., 38 mm., 41 mm. These measurements are the mean computed from five series of 8 to 10 embryos in each series. 3. Conclusions, a. On the ventro-mesial aspect of the central osteogenetic cellular constriction there is a more active proliferation of osteoblastic cells than elsewhere. Immediately underlying this zone the processes of atrophy and vesiculation in the cellular cartilage are begun. 6. The lamina prima first differentiates in a 3 mm. femur on the ventro-mesial aspect, quickly encircling the center of the degenerating cartilaginous shaft as evidenced by the growth in a 3.5 mm. length of the femur c. There is a more rapid advance of the developing lamina prima towards the proximal epiphyseal line at the head end of the femur than towards the distal epiphyseal line at the condylar extremity of the shaft. PROCEEDINGS 341 (/. Calcitication and vcsieulation of the cartilaginous matrix and cells, respect ivoly, also advances at a moie rapid rate towards the proximal head end of the fennir keeping slightly in advance to the rai)i(ll>' progressing lamina prima. e. The lamina i)rima on the ventro-mesial aspect reaches the proximal epii)hyseal line i)efore that on the dorso-lateral aspect. This is the case in a 5 mm. feimir and the lamina prima is also considerably thicker on the former than on the latter aspect. /. A lamina secunda is differentiating, in the 5 mm. length of femur, at the central primary constriction peripherad to the lamina prima. g. A significant fact is that the nutrient artery enters the adult bone in the upper one-third of the diaphysis on its ventral aspect and is directed distally towards the condyles. REFERENCES FooTE, J. S. 1911 The comparative histology of femoral bones. Trans, of the - Amer. Micros. .Soc, vol. 30, pp. 88-140. IfllG A contribution to the comparative histology of the femur. Smithsonian Contributions to Knowledge, vol. 35, no. 3. Edited by Ales Hfdlicka. 15. Microdissection studies. Cell and nuclear division. (Lantern). Robert Chambers, Jr., Cornell Medical College, New York City. Cortical changes in the egg cell on the approach of cell division can. be demonstrated with the needle. In polar body formation the change appears to be due to the presence of the nucleus for, on removal of the nucleus to another region of the cell, a corresponding change in the cortex occiu's. Preparatory to division the bulk of the cytoplasm undergoes temporary gelation and constriction takes place in the region where liquefaction sets in. The resting nucleus is an optically homogeneous body in which a granular network is often produced on the slightest injury. It is fluid in consistency in egg cells and possibly in other cells. Formation of the nuclear spindle was studied in germ cells and in egg cells during polar body formation and segmentation. In no case could spindle fibers be demonstrated. The nuclear substance assumes a spindle shape under the influence of the centrospheres. Liquefaction of the cytoplasm in the equatorial region of the cell is accompanied by a constriction which shapes the nuclear spindle into a dumb bell during the ana-and telo-phases. As the daughter nuclei draw away from one another they are connected by a strand which is of a sufficient consistenc}^ to distort the nuclei if caught and pulled by the needle. The prechromosomal filaments appear in the prophase out of the hyaline nuclear matrix through the precipitation of granules in more or less irregular masses closely investing long slender cylindrical strands of a gel like consistency. Many of the diplotene filaments figured in fixed material may posssibly be due to the collapse of these strands into ribbon like structures wdth an accumulation of the granules along the edges. Shortening of the prechromosomal filaments is ac 342 AMERICAN ASSOCIATION OF ANATOMISTS companied by a fusion of the granules thus forming the definite hyaline, viscous and comparatively rigid chromosomes. The chromosomes are imbedded in the nuclear spindle mass and undergo movement^ resembling those in drops which are undergoing changes in surface tension. The constriction of the nuclear spindle at its middle possibly aids the migration of the chromosomes by pushing them to the poles. Many if not all of the structures which appear during cell and nuclear division may be explained as reactions in a mixture of colloidal materials producing local liquefactions together with precipitations and gelatins which may be made to disappear on changing the reaction. 16. A study of the reaction of lymphatic endothelium and of leucocytes in the tad-pole's tail toward injected fat. Eliot R. Clark and Eleanor Linton Clark, Department of Anatomy, University of Missouri. The present investigation is part of a series undertaken in order to stud}^ the growth and reactive powers of living tissues and cells, in the tad-pole's tail, by observing their response toward various injected substances. One of the authors has reported the results obtained by injecting small globules of paraffin oil. He showed that these globules were not absorbed and that, aside from a transitory leucocytosis, the various tissues and cells showed no reaction to the injected paraffin oil. In the present study, small amounts of various fatty substances were introduced subcutaneously, with the especial object of studying the reaction of the lymphatic capillaries. The substances injected were olive oil, oleic acid, cream and yolk of egg. 1. dive oil. The oil was injected in the form of single globules, measuring about 50 /j. in diameter. The globules diminished in size during the period of observation but were not completely absorbed after nineteen days. Soon after the injection, clear leucocytes were seen to pass through the walls of nearby blood-vessels and to approach the oil. Here they flattened out and formed a ring of leucoc^-tes around the periphery of the globule. Soon after coming in contact with the oil, they became pigmented and were seen to contain small drops of oil. The lymphatic capillaries were evidently attracted by the presence of the olive oil; they grew toward it, even bending out of their course in some instances. Upon reaching the globule, the lymphatic tip remained in close contact with the rim of pigmented leucocytes for several da3^s. In some cases, the tip of the lymphatic later extended beyond the oil globule. No pigmented leucocytes were seen to enter a lymphatic capillary. 2. Oleic acid. Immediately after injection, the oleic acid changed from a clear refractile globule to an opaque granular mass which was brown by transmitted and white by reflected light and rescmljled closely the sodium soap of oleic acid. The leucocytes responded to this substance more qaickly and in larger numbers than in the case of the olive oil. They formed a ring around the injected mass several layers deep and all f)ecame deeply pigmented. On the day following the PROCEEDINGS 'M'.^ iiijt'ction, many snuil! rcfniclilc droijlcts could he .seen, .sciittorcd thr()\i{;li tlu' hrown siihstaiicc. Tlio absorption procccilod more quickly than in the case of the ohve oil hut was not complete after ten days. The lymjihatics responded, as in the case of the olive oil, by growing toward the injected sni)stance. Pigmented leucocytes were observed to move away from the injected mass and to come into close contact with the tip or wall of nearby lym])hatic capillaries. After remaining for five or ten minutes in close pro.ximity to the lymphatic, they moved away again and, shortly before or at the time of wandering away, these leucocytes lost their pigment and became clear. 3. Cream and yolk of egg. Both of these substances consist mainly of an emulsion of very small droplets. The leucocytes were attracted to the injected cream or yolk in large numbers and actively took up the minute drops of fat. The leucocytes containing fat were then seen to wander off and to come into close relationship with nearby lymphatic capillaries. After remaining close to the lymphatic for a few moments, they became clear again. At the end of twelve hours, most of the injected droplets had been taken up by leucocytes, and after twenty-four hours all had been. On the second day, only a few pigmented leucoc>ies marked the site of injection. In spite of the rapidity of the absorption in the case of the yolk and cream injections, a definite growth of lymphatic capillaries toward the injected region was observed in some instances: Conclusions, a. Lymphatics reacted to the injected fat by sending out sprouts which grew toward it. b. LeucocA^es responded quickly to the injected substances, migrated tow^ard them in large numbers and actively engulfed the fat. c. The fat appeared to be absorbed through the combined efforts of leucocytes and lymphatics. d. Mesenchvme cells and blood-vessels did not respond to the mjected fat. e. The rapidity of the absorption depended upon the size of the iat droplets: the fine emulsions of cream and yolk were taken up very much more quickly than the single relatively large globules of olive oil and oleic acid. /. The fat appeared to be changed within the leucocytes and to be absorbed in a soluble form by the lymphatics. 17. Some points on the urogenital system of myxinoids. J. L. Conel (introduced by H. D. Senior), University and Bellevue Hospital Medical College. A study of the adult structure of the urogenital systems of Myxine and Bdellostoma. The work was done in the Zoological laboratories of the University of Illinois. A central duct is present in the pronephros of both Bdellostoma and Myxine, but it is in a state of degeneration in both animals. This degeneration is more advanced in M3'xine than in Bdellostoma, and includes the inner ends of the pronephric tubules and all parts of some of the largest tubules. 344 AMERICAN ASSOCIATION OF ANATOMISTS The ]Malpighian body of the pronephros in both Bdellostoma and Mj'xine is located at the posterior end of the pronephros, and in appearance and structure closely resembles the Malpighian bodies of the mesenephros. It is formed by the fusion of two or more glomeruli. The glomerulus of the pronephros in young Mj^xinoids is exposed to the pericardial cavity through a large opening, thus resembling a glomus. This opening becomes constricted to a small tubule in adult animals. The tubules of the mesonephric Malpighian bodies in Bdellostoma . are structurally of two types. Neither Bdellostoma nor Myxine is a protandric hermaphrodite. 18. On the lipoidal nature of structures in the corpus luteum cells of swine. George W. Corner, Hearst Anatomical Laboratory, University of California. In a previous paper (The Corpus Luteum of Pregnancy as it is in Swine, Carnegie Institution of Washington, Contributed to Embryology, 5) the author called attention to ceitain structures appearing in the lutein cells of swine after fixation in formol, in the form of small hollow spheres lying in spaces in the outer part of the cj'toplasm. They are probably similar to bodies found in the lutein cells of rabbits by Franz Cohn (Arch. f. Mikr. Anat., 62, 1903). In sows they occur onlj' during the first half of pregnancy, and the changes leading to their disappearance from the cells are so characteristic that by examining the corpus luteum it is possible to estimate the stage of the pregnancy with some accuracy. It was tentatively suggested, in the former paper, that these bodies represent an elaborate modification of the Golgi-Holmgren intracellular canalicular apparatus. However, preparations smce made, by Cajal's uranium nitrate technique, show the above described structures and the Golgi net to be present in the same cell and independent of each other. It has been found that they are not present in the fresh tissue, appearing only after several hours in formol, Zenker's fluid, or other aqueous fixatives. They are stained blue by >sile-blue sulphate, pale brick-red by neutral red, give a brown lake with Weigert's hematoxylin, stain positively with Ciaccio's method, and are not anisotropic when seen under the polariscope. They do not stain with Nile-blue after treatment with alcohol of 65 per cent or stronger, and stain but faintly or not at all after a day's immersion in acetone, chloroform, ether, xylol, and benzene. It seems clear, therefore, that the appearance of the spherules is produced by the swelling into rounded masses, in the presence of water, of a lipoid, which is present in the luteum cells and which is probably a phosphatid, to judge from its microchemical reactions. Apparently, however, in the process of swelling of the lipoid material some proteid is carried with it, and is precipitated in the spherules during fixation, for even after removal of the lipoid substance by alcohol, the skeletons of the spheres remain, and may be stained by appropriate dyes, especially Mallory's con PROCEEDINGS 345 lUH'tivc tissiio stain. 'l\ic hollow center in such stained preparations is due to tlie fact tliat the splierules usually form about tlu; fi;lol)ulcs of neutral fat which the lutein cells of early pregnancy contain in jrreat niuubers, and the spaces about them are due to shrinkage of the cytoplasm. The discovery that the appearance of these bodies after fixation is due to an artefact does not detract from their interest, since they enable us to detect the presence of a constituent of the cell which may be of importance in the economy of the corpus luteum, and to follow histologically the changes in amount of this substance. 19. Oestrus and ovulation in swine. George W. Corner and A. E. Amsb.\ugh, Hearst Anatomical Laboratory, University cf California. In order to obtain a point of departure for studies on the changes of the uterus and ovaries during the reproductive cycle in swine, we have undertaken to determine what relation exists between oestrus and ovulation. Breeders agree that oestrus occurs every eighteen to twenty-one daj's, usually lasting three days. (One of our animals showed signs of heat on four successive days.) Heat is not terminated 1)3' copulation, which may occur repeatedl5^ We have found that animals killed during this period usually show recently ruptured Graafian follicles, and in such animals we have been able to recover the ova b}' washing out the Fallopian tubes. Rupture of the follicle is spontaneous, occurring even in the absence of the boar. In one animal in which oestrus was noted sixteen hours before killing, copulation almost certainl}' having occurred during the interval, rupture of the follicles had not yet taken place, although sections showed an intact follicular wall with a normal ovum in which the first polar body was forming. It is clear, therefore, that ovulation occurs during oestrus, and probal^ly on the first or second day of the period, since we find regularly that ovulation has taken place when sows are killed on the third day of oestrus. The unfertilized ripe ovum of the sow, as found in the tube, measures 155 to 165 /x in diameter. The zona pellucida is 7 to 8 ^ thick, enclosing a 3^olk heavily laden with fat globules, obscuring the nucleus. The polar bodies are often clearly seen in the fresh ovimi. Study of a small series of ova which have been cut into serial sections seems to show no deviation from the stages reported in other mammals; the first polar body is formed within the follicle just before rupture, the second in the tube. Entrance of the spermatozoon and fusion of the pronuclei occurs in the tube, 20. Potentialities of the lyni'phoid hemoblasts of the adult spleen. (With demonstrations.) Vera Danchakoff, College of Physicians and Surgeons, Columbia Universit}'. The adult spleen is a lymphopoietic organ. Under normal conditions numerous small lymphocytes are developed in its follicles. The stem cells of the small lymphoc^-tes are morphologically similar to tlie stem 346 AMERICAN ASSOCIATION OF ANATOMISTS cells of other blood cells, either granuloblasts or erythroblasts. The morphological structure of all these stem cells or lymphoid hemoblasts is permanently retained by the organism from the time of the first appearance of blood. Embryogenetic studies have shown that the various differentiations of the hemoblasts is associated with different environmental conditions. A hemoblast under definite environmental conditions can split off only' one kind of blood cells. What is the action exerted by the environmental conditions on the hemoblasts? Do they alter the metabolism of the hemoblast irreversibly, definitively narrowing its potencies to the limits of its prospective value, or do they merely condition its differentiation for the time in which they act, temporarily inhibiting some of its potencies, which may subsequently become active in other environments? If the normal differentiation inta a small lymphocyte is forced upon a hemoblast in the follicle of the spleen by definite environmental conditions, other potentialities should be realized by the same cell under other environmental conditions. A mesenchymal cell in the allantois responds to a stimulus or to an irritation in a characteristic manner, it readily proliferates; if situated in a well vascularized region, it rounds up, becomes mobile, transforms into a hemoblast; finding itself outside of the vessels it differentiates into a granuloblast and finally into a granular leucocyte. The stem cells in the lymphoid follicle of an adult spleen do not differentiate under normal conditions into granular leucocytes. Would this differentiation take place if the follicle of an adult spleen were transplanted into environmental conditions, favorable to the granuloblastic differentiation? If the potencies of the stem cell in the follicle of an adult spleen, besides its own prospective value, include the prospective values of the stem cells of granuloblastic tissue, then such differentiation of necessity must occur. Cultures of spleen tissue of different animals were made on the allantois of the chick embryo. The cultures grew well and formed tumors of nearly 1 centim in diameter. Both embryonic tissue of the allantois and adult tissue of the spleen participated in the formation of the tumor. In many cases the tumors in well advanced stages consisted of a uniform granuloblastic tissue more or less richly vascularized, in which no distinction between adult and embryonic tissues could be drawn. The study of early and intermediate stages gives, however, precise information concerning the respective role, assigned to each tissue in the formation of the tumor. The lymph follicle of the adult spleen is easily recognized by its arterial vessels and by the presence of small lymphocytes, which are lacking in the allantois of the embryo. Usually follicles, which in the cultures are immediately adjacent to the allantois persist and thrive. Several types of cells constit\iting the follicle react in definite and differing ways. The fate of the following cells of the spleen follicle in the culture could be determined: PROCEEDINGS 347 1. Snuill Iviiiphocylcs. 2. Cells fonuinji, the rcliciiluni. '.]. Slom cells of tho small lym))lu)c\'t.es. Numerous small lymphocytes emigrate out of a follicle into the loose tissue of the allantois. Some of them remain in the follicle. They do not show here any sifj;ns of activity but remain inert. Very rarely can a mi((jtic fi^in-e in a small lymi)hoc>le be recofAiiized. The small lymphocytes }i;ra(hially die out and are inj^ested by adjacent cells; ill their cytoi)lasm remnants of ingested small lymphocytes may be discovered a long time afterwards. Plienomena of a substantially different order are observed in the two other characteristic structures of the lymph foUicle, in the hemoblasts and in the cells forming the reticuhuu. They are of a progressive order and reveal in these cells a liigli ]iower of activity. Transplanted into a loose embryonic mesenchj'-mal tissue the ]yiui)h follicle loses gradually its dense appearance; numerous cells of the reticulum hy])ertropliy, become free and together with the existing stem cells of the lymph follicle form large groups of amoeboid cells. The stmcture of these cells correspond in all details to that of the numerous hemoblasts which sinuiltaneously develop at the expense of the embryonic mesenchymal cells in the allantois around the culture of the spleen. If left in undisturbed relations with its normal environment the lymphoid hemoblasts of the adult spleen would naturally develop into small Iymphoc>'tes. But now when subjected to conditions, in which an embryonic lymphoid hemoblast displays an intensive granuloblastic differentiation, the hnnphoid hemol)lasts of the adult spleen follow exactly the same lines of differentiation, they become granulocytoblasts. New environmental conditions disclose in the lymphoid hemoblast of the adult spleen a potency to differentiation which was inhibited by its previous environmental conditions. The lymphoid hemoblasts of the adult spleen (and this applies as well to the mesenchymal cells of the reticulum) are morphological units, the development of which into small lymphocytes under normal conditions is but the realization of one possibility out of its larger range of potentialities. 21. Differential factors of erijthro-granulopoiesis in the embryo of mammals and sauropsids. (With demonstration of hemotopoiesis in the yolk sac of birds.) Vera Danchakoff and Clayton Sharp, From the Anatomical Laboratory of Columbia University. The development of the hemotopoietic function in the mesenchyme seems to be closely connected with the development of vascular channels in more than one respect. The first appearance of blood corpuscles is inseparably connected with the first development of vessels. Even the endothelial wall seems to become m Sauropsids, a differentiating factor in the development of the common mother cell. Again, the presence of a large sinus like venous capillary net favors highly a hematopoietic differentiation THE ANATOMICAL RECORD, VOL. 11, NO. 6 348 AMERICAN ASSOCIATION OF ANATOMISTS in the mesenchyme. The effect upon the stem cells of separation by an endothehal wall, the mfliience exercised by the intravascular conditions within a sinus like net are factors, which determine, together with the cell complex of the cell its further differentiation. A comparative study of the development of the first hemotopoietic organs in mammals and Sauropsids is of interest when connected with the study of its determining factors. The first striking difference in the hemotopoiesis of these two classes of annuals is the difference in localization of hemotopoiesis daring a great part cf their embrj'onic life. In mammals hemotopoiesis is localized in the liver, while in Sauropsids, it is locaUzed in the appendages of the yolk sac. If hemotopoiesis is an activity of the mesenchyme, induced by a definite complex of environmental conditions, there must be found a striking analogy between conditions encountered in the liver of mammals and those prevailing in the appendages of the yolk sac. It is of course easy to understand, that hemotopoiesis cannot take place in the walls of the mammalian yolk sac, foi there is not space in this region for a hemotopoietic organ to develop. Why, however, is this process transferred to the liver? May the circumstance of its entodermal origin- constitute a factor in the determination of this organ as a hemotopoietic center? It does not seem probable, for the definite organ of erythro-granulopoiesis associates with the bone marrow, which is not an entodermal derivative. It has been pointed out, that the large quantity of food material, present in the yolk sac, is one of the chief factors in determining the development of erji:hro-granulopoiesis in this region. Is this requirement fulfilled in the mammalian liver? It is, for the umbilical veins, carrying the resorbed food from the placenta, traverses the liver. The presence of this large quantity of food can be effective, only if the blood carrying it flows slowly through large channels. Such conditions are found in the hemotopoietic organ of the birds. These conditions repeat themselves in the liver of the mammalian embryo, when the developing liver converts the lumen of the umbilical veins into sinusoids. The existence of conditions favoring the absorption of definite substances from the blood stream, seems so important, that hemotopoiesis does not occur where these conditions are absent, e.g., in the placenta whose vessels are merety thoroughfares for the blood. If the mere opportunity for resorption, the presence of food material and favorable conditions, was sufficient for the development of hemotopoiesis, mesenchyme in the liver of Sauropsids would be induced to undergo a hemotopoietic differentiation, as it has in the liver of mammals ; for the omphalo-mesenteric veins in birds carry large quantities of food material from the yolk sac and are transformed also by the growing liver into sinusoids. But, the mesenchyme of the bird liver does not develop a hemotopoietic function. There must be an essential difference between substances carried by the blood of the umbilical veins in the mammalian embrj^o and those PROCEEDINGS 349 carried by the l)lo()(l comiiij;- tlir()ii};li the omjjhalo-mesciitoric veins of tlio S:uiroi)siils. Tlioujiih l)()th are rich in iood necessary fot embryonic development, the blood of the om])hal()-mesenteric veins of the Sauropsids has jiassed thro\ip;h a venous ])lexus, in the appendages of the yolk sac, which are enj:;orped by multiplying and differentiating blood cells. These cells are the first to profit and to abstract from the resorbed food certain substances necessary for their multiplication and specific hemoto])oietic differentiation. The l)lood reaching the liver in the Sauropsids, therefore, has been filtered of definite substances necessary to the mesenchyme for its differentiation into erythro-leucopoietic tissue. The mesenchyme of the liver in Sauropsids, though surrounded by sinusoids, in which the resorption of food material may easily occur fails of specific development because of the lack of specific substances in this food material. 22. Further verification of functional size changes in nerve cell bodies by the use of the polar planimeter. (Lantern.) David H. Dolley, Laboratory of Pathology, Universitj^ of Missouri. Kocher (Jour. Comp. Neur., vol. 26, 1916, 341) severely attacks the work of the writer on the basis of area measurements of nerve cells from undisturbed and exercised animals by means of the polar planimeter. I have tested the planimeter method, not by averaging all cells measured from any region irrespective of their functional stage, as Kocher did, but by applying it to the same number of cells from every functional stage. Data of average diameter measurement were obtained concurrently from all series, and the two methods compared graphically. The two methods afford results absolutely identical in every detail both in size and nucleus-plasma relations, thereby verifying previous findmgs. The area method has no special or superior value in the case of uniformly shaped cells, though it would be well to use both methods in the case of irregular contours. Though exact in itself, the planimeter gives only two dimensions, with smaller variations than in volume calculations, and hence may be quantitatively misleading. Kocher finds only slight variations in area size between control and exercised animals. The essential fallacy in denying any functional size differences from the results of comparative measurements when all stages are averaged together depends upon the tendency to a general uniformity of absolute size of corresponding nerve cells among animals of the same species. Further, when the third dimension is ignored, and Kocher 's only reference thereto is in the heading of a table "volume expressed as square inches," it follows that there is less range of variation between functioning type sizes than if volumes were used, and the averaging tends better to equalize differences due to unequal distribution of various-sized tj-pes. Kocher's results are just what might be expected, and instead of confounding the writer in respect to functional size changes, tend only to support his induction of a uniformity of size relations as a general rule for a species. 350 AMERICAN ASSOCIATION OF ANATOMISTS Kocher confirms the existence of all the stages interpreted as functional by the writer. However, he admits no significance to them, because he finds them all in undisturbed as well as exercised animals. His conclusion from this of absence of qualitative differences is not, as he ignorantly thinks, destructive to me, but is the first induction I should wish to be confirmed since it throws nerve cell function solely on the fundamental quantitative principle. He denies such quantitative differences in comparative differential counts, but since for one thing certain stages as diagnosed do not accord with his text statements and are unlikely upon the basis of relative differentiation, these counts need not be taken seriously. Kocher's sweeping criticism falls into absurdity, being based on objective results in the above respects confirmatory of my own conclusions. 23. 'Histiocytes^ and 'macrophages' and their relations to the cells of normal blood in animals stained intra vitam with acid colloidal dyes. Hal Downey, Department of Animal Biology, University of Minnesota. Various colloidal substances, such as aqueous suspensions of lithium carmine, of the azo dyes Pyrrolblau, Trypanblau, etc., of collargol, India ink, and of various other non-toxic colloidal substances injected into the veins of animals are taken up very rapidly and stored in the form of granules by certain cells of the hemotopoietic organs, liver, and of the loose connective tissue. Repeated injections of the same animal result in an increase in the number of 'dye granules' in the cells, and also in the number of cells which contain the dye. Cells containing the dye granules have been termed 'histiocytes' by Aschoff-Kiyono, 'pyrrol cells' by Goldman, 'macrophages' by Evans. Schulemann ('12) believed that the dye granules represent a combination of dye with a reaction body of the cell, and also that various preformed structures, such as plasmosomes, secretory granules, etc., can be made visible by means of these dyes. V. Mollendorf, Pappenheim and Tschaschin also agree with this earlier view of Schulemann. At present Schulemann and Evans believe that the dye granules are merely concentrations of the colloidal suspensions which are located within cytoplasmic vacuoles. They are not combined with any constitutents of the cell. The formation of dye granules is therefore a physical process and not a chemical union of preformed structures or receptors of the cell with the dye. In his latest paper Schulemann ('16) claims to have demonstrated the physical nature of the granules experimentally. If the results of these experiments are correct then we must agree with Evans that the fine particles of the injected suspensions are phagocytosed by the cells which are able to store the dye in the form of coarse granules. Phagocytosis, therefore, is the process which quickly cleans out the particles of the injected suspensions from the fluids of the body, and cells which phagocytose the suspended particles should obey the general laws of phagocytosis. PROCEEDINGS 351 The cells of tlic circtilating l)lootl do not t;ikc \i]) these substances, even thonjih the hitter are injected (Hrcctly into the blood stream, However, histiocytes :ii)i)ear in the blood stream after rei)eated injections of the colloichd substances when the animal becomes 'overloaded,' and AscholT-Kiyono have demonstrateil them in sections of the veins of the hemato])oietic organs after a sinj^le injection of lithium carmine. They do not take up the colloidal substance directly from the circulation, but become loaded with it while in the tissues, especially of the hematopoietic organs. They are never abundant in the general circulation but may be quite numerous in the veins of the liver and hematoi)oietic organs. The fact that Aschoff-Kij'ono nuist resort to sections in order to demonstrate the presence of the histiocytes seems to favor the view of Alitamura and Masanori that the anesthetic, or the manipulation dm-ing the operation, or premortal agony cause the separation of endothelial histioc>i;es from the hematopoietic organs and their entrance into the jjlood stream. These authors find the histiocA-tes to be very rare in the veins of living animals, but quite numerous in dead animals. These facts seem to indicate that the passage of histiocji^es into the blood is more or less accidental, a view which is further strengthened by the observation of Aschoff-Kiyono that they are rapidly filtered out of the circulation by the capillaries of the lung. This latter fact shows that the dye cells act as foreign bodies when they reach the blood stream, and it also accounts for the small numbers of histiocytes in the general circulation. The few dye cells which get into the circulation in overloaded animals, and the free histiocj-tes of the serous cavities, taches laiteuses, and many of those of the loose connective tissue are identical morphologically with the larger lymphocytes and large mononuclears of the normal blood, excepting that they contain the dye granules while the latter do not. For this reason, and because of the fact that the reticular cells, especially those lining the sinuses, and their free derivatives in the hematopoietic organs, the stellate cells of the sinusoids of the hver, and the clasmatocytes or restmg wandering cells of the loose connective tissue are the cells which show the greatest preference for the dyes, it has been claimed by Goldmann, Aschoff-Kiyono, Pappenheim, Schulemann-Evans that the dye cells are always of tissue origin. In other words, true blood cells will not take up the dyes, and we therefore have a sure method for distinguishing between blood cells and tissue cells. The free wandering cells in the loose connective tissue which take up the dyes, and those about the taches laiteuses of the omentum and in the serous cavities are tissue cells although they are identical in structure with the lymphoid cells which we see in the circulating blood of normal animals. The above authors, excepting Pappenheim, also claim that the histoic3d;es, both fixed and free, are closely related cells which constitute an mdependent cell-line which is absolutely distinct in origin and function from an3i;hing found in the circulation of normal animals. 352 AMERICAN ASSOCIATION OF ANATOMISTS The only worker in this field to seriously oppose the idea of the independence of the histiocytes is Tschaschin, who believes that in experimental inflammation of the loose connective tissue in vitally stained annuals hanphocytes leave the vessels in great numbers and increase rapidly in size to form 'potyblasts' which take up the dye and store it in the form of the t3T3ical dye granules of the histioc3i;es. The results of the following simple experiments are opposed to the theory of the mdependence and close relationship of all the dye cells. They also show that blood cells under proper conditions will take up the dyes as rapidly as do the so-called histiocytes. Rabbits and albino rats were the experimental animals, and Pyrrolblau, one of the acid azo dj^es, was used as the colloidal suspension. Intramuscular injection of a 1 per cent suspension of the dye in water caused immediate migration of polymorphonuclear leucocytes to the site of injection. More poljanorphonuclears were obtained if a suspension of aleuronat was injected first. When the animals were killed twenty-four hours after the injection sections of the muscle showed that the polymorphonuclears were gorged with the dye which was stored in the form of coarse granules identical with those of the histiocytes of animals stained mtra vitam by intravenous injections of the dye. None of the leucocytes of the circulation contained the dye granules. The most convincing results were obtamed when the dye was injected directly into a segment of the femoral vein which had been isolated from the general circulation by a double ligature. The vein was tied off first and then the dye injected into the vessel between the hgatures. Twenty-four ' hours later the ligatured portion of the vein was removed, fixed and sectioned. It contained great numbers of polymorphonuclear leucocytes which were loaded with the typical dye granules. All of the larger lymphocytes and large mononuclears present also contained the dye granules, the latter cells being identical in structure with the free histiocytes of the peritoneal cavity of animals stained intra vitam by peritoneal injections of the dye. The presence of the dye in the ligatured vessel has attracted great numbers of polymorphonuclears from the vessels of the surrounding tissue or from the vasa vasorum, but it seems to have had no influence on lymphoid cells other than those which were already contained in the ligatured vessel. The lymphoid cells are not more numerous than would be expected in the amount of blood contained in the vessel. The fact that polymorphonuclear leucocytes have wandered into the vessel does not destroy the value of the experiment, the most significant feature of which is the storage of dj-e in the form of typical dye granules by cells that are primarily blood cells, and which under ordinary circumstances do not take up the injected dyes. The lymphoid cells containing dye granules are those which were already present in the vessel, and therefore they are also blood cells. These results seem to indicate that the reaction of cells to colloidal suspensions is not sufficiently specific to be used as an index of celllineage. Mechanical conditions and nature of the surrounding medium PROCEEDINGS 353 uiuloubtedly i)l:iy an imi)ortaiit rule in these reactions. In the hlood stnvini the ci'lls hardly j^t't a chance to phagoc>i,ose the parti(!les of tlic injected colloidal suspensions, for the latter leave the circulation almost as soon as they are injected. The presence of dye granules in the reticular cells of the lymph nodes, especiallj- in those lining the sinusus, within a few minutes after injection indicates that the suspensions pass very quickly into the lymph stream to be removed from the latter by the reticular cells. That the reticulo-endothelial cells should show great voracity in the taking u]) of the dye is not at all sur])rising, for we know that they are among the niost active phagoc>^es of the body. This is also true of the free cells Avhich are cut off from the reticulum; they will phagoc^-tose erythrocj-tes, bacteria or any other foreign substance about as readily as they do the injected dye. The Ijnuphocj-tes of the nodes do not take up the dye nor any other foreign substance, such as living bacteria, carmine, etc. This is because they are inmiature cells having relatively small amounts of cytoplasm. Tschaschin has sho\\ii that when they migrate into inflamed areas they rapidly develop into protoplasmic phagocj^tes which take up the dye, and the experiment with the ligatured vessel shows that the larger and more mature lymphocj-tes which are caught in isolated segments of the circulation, where they remain in contact with the dye for some time, will also take up the suspension. In the free circulation neither the l3miphoc}i:.es nor the polymorphonuclear leucoc^'tes take up the d3'-e, probably because of the fact that the dye is eliminated from the circulation so quickly that the phagocytic cells are not in contact with it for sufficient length of time to be able to take it up. The mechanical agitation of the blood in the free circulation probably also has something to do with the prevention of phagocytosis. That mechanical conditions are a factor in phagoc3'tosis is indicated b}' the results of intravenous injection of ordinary carmine ground up in salt solution. The particles of this suspension are very much coarser than those of the colloids used. Hoffmann and Langerhans have shown that tliis substance remams in the circulation as late as 148 days aftei the injection, and that it does not enter the lymph nodes until 3 days after injection. Within the circulation the carmine is phagocjiiosed very freety by both the polymorphonuclears and the macrophages. It takes several days, however, before the last of the carmine particles are included in phagocytes (6 grams of carmine at one injection!). The entu-e reaction can be explamed b}^ the slow diffusibihty of this substance; it is elimmated from the circulation very slowly, and the phagocji^ic cells of the blood have abundant opportunity to take it up. The more diffusible colloids are also taken up by phagocji^es, but isolation from the general circulation is necessary before phagocytosis can take place. A further illustration of tliis rule is to be found in the work of Kline and Winternitz on experimental pneumonia in animals stained intra vitam with Trypanblau. Poljinorphonuclears contain 354 AMERICAN ASSOCIATION OF ANATOMISTS ing tjq^ical dj'e granules were found in the alveoli, bronchioles and blood vessels of the involved lung but not in the general circulation. Injection of the blood vessels of the involved lung showed that they were cut off from the general circulation by plugs of fibrin in the capillaries. Pappenheim and Fukushi also reported dye granules in polymorphonuclears and in lymphoid cells of peritoneal exudations. Rosenthal has shown that even living organisms are rarely phagocytosed withm the circulatmg blood. A virulent bacteria injected into the tail vein of mouse were phagocytosed by 'endothelial' cells, chiefly the stellate cells of the liver. Wandering cells of the blood became active only when the bacteria were so numerous that the 'endothelial' cells could no longer take care of them. Werigo found anthrax bacilh. to disappear from the circulation within a few minutes. They were held bj^ phagocytes in the spleen, liver and lungs. The WTiter's experiments with Pyrrolblau have shown clearly that the type of phagocj'-tic cells involved in the process of taking up the dyes depends altogether on the conditions of the experiment. According to Buxton and Torrey this is also true when living organisms are used. Typhoid bacilli and staphylococci injected into the peritoneal cavity appeared within the macrophages of the mediastinal lymph nodes within one hour. They conclude that the macrophages have seized the organisms because they were present before the arrival of the polymorphonuclears. Later the latter invaded the nodes in great numbers and seized whatever organisms were left. The power to phagocytose may vary greatly in cells of the same type (Hektoen, Rosenow), and cells which are not ordinarily phagocytic may become so under certain conditions, as shown by Achard, Raymond and Foix, who reported a case in which the eosinophils of a pleural exudate were very active as phagocytes. Phagocytic eosinophil leucocytes have also been reported by Lattan-Larrier and Parvu, Wendenburg, and liy Weinberg-Seguin. It is evident that phagocytosis is a physiological process which is not confined to any one type of cell, and that the material to be phagocytosed which under ordinary conditions is taken up by a certain type of cell may, under slightly different conditions, be taken up by cells which genetically and structurally are quite distinct from the first type. The taking up of colloidal dj^es seems to be, in many cases at least, very similar to phagocytosis as we have known it before the advent of the azo dyes. Dye granules have been reported in the following cells: reticular cells, endothelial cells, clasmatocytes, lymphocytes, polymorphonuclear leucocytes, hepatic cells, cells of the adrenal and epithelial bodies, interstitial cells of the testis, cells of the Graafian follicles, hypophysis, plexus chorioideus, ectoderm cells of the placenta and the giant cells of this organ. The granules of some of these cells may be vitally stained preformed structures. Nevertheless, it is evident that the reaction is not specific for any one line of cells, and all attempts to classify cells according to their reactions to these dyes must be regarded as failures, especially since Kiyono, one of the most PROCEEDINGS 355 ardciit (k'ftMulers of the theory of the specificity of the reaction, is forced to :i(hnit that lymphocytes of lymph nodes in the later stages of aseptic inflammation may enlarge and develop relatively more protoplasm, when it is impossible to distinguish them from the true histiocytes of the organ, and impossible to show that the}' do not take up the dye. 24- '^ human embryo of seven to eight somites. H. M._ Evans and G. ^^^ Hahtelmkz. Dci)artment of Embr3'ology, Carnegie Institution of Washington and Department of Anatomy. University of Chicago. This eml)rvo was obtained from an aborted ovum measuring 18.0 by lo.O by 10.8 mm. including the villi, fixed intact in 10 per cent formalin. The age was estimated clinically as three weeks. The embryo measured 2 mm. in length in formol and belongs in the group with the Mall embryo no. 391 (described by Dandy), the seven somite Spee emliryo, the 2.11 mm. embryo of Eternod and the R. Meyer embrj-o no. 335. The embryo was found attached directly opposite the chorion laeve, projecting into the extracmbrvonic coelom at right angles to the chorionic wall, supjjorted by a few strands of magma. The embryo lies flat upon the yolk sac, with head and tail folds rising above it. The height of the head fold may have been slightly increased b}- distortion in the fixative but this region could not possibl}- have been bent ventrally over the yolk sac as it is in the Keibel embryo of 6 somites (Normentafel no. 3), and the above mentioned Eternod specimen. The amniotic cavity is large and the amniotic duct extends along the belly stalk. In dorsal view the embryo appears somewhat like a spoon, the expanded cephalic neural folds corresponding to the bow^l. The neural tube is closed from the middle of the hind bi-ain to the level of the seventh somite. Caudally the neural folds gradually flatten out and pass over into the primitive streak at Henson's node. The neurenteric canal has ah-eady closed. It is possible to dehmit forebrain, midbrain and three hind brain neuromeres since the cerebral flexure has just begun to appear, the neural crest is actively proliferating and the otic plate and ganglion are well defined. The as^Tumetric forebrain is bent almost at right angles to the hindbrain, the midbrain fonning the knee. In the forebrain two shallow sulci can be distinguished converging rostrally: they are the earliest stage of the optic sulci yet described in man. The neural crest cells are migrating from the dorsolateral region of the folds in the midbrain and hind brain as far caudally as the VIIVIII ganglion. This is a bulbar swelhng of the neural fold dorsally, h'ing opposite the thickened otic plate. The pharynx is intemiediate in development between that of 391 in the ]Mall collection and 335 of R. ]\Ieyer. Its epithelium is in contact with the ectoderm of the oral membrane, the first visceral pouch is well developed and its dorsal diverticulum touches the thickened ectodemi at one point. The second pouch is beginning to fonn and there is an asjTnmetric thyroid anlage. The hind gut extends but five sec 356 AMERICAN ASSOCIATION OF ANATOMISTS tions caudal to the origin of the allantois and there is no eloacal membrane. The chorda begins near the upper end of the pharynx as a thickened ridge and is everjnvhere incorporated in the entoderm except in the region of Henson's node. The heart is an ahiiost bilaterally sjinmetrical tube formed by the union of the vitello-umbiUcal veins lying ventral to the pharynx. A pan- of delicate vessels, the first aortic arches pass around m front of the first pah of ^^sceral pouches from the bulbar end of the heart to the greatly dilated cephalic ends of the dorsal aortae. The latter have four well developed pairs of dorsal rami, the first two of which are growing in the direction of the fifth and eighth ganglia respectively. Caudally the aortae break up into a plexus on the dorsal wall of the yolk sac. which plexus in turn gives rise to the umbilical arteries at the beginning of the belly stalk. More rostrally the vitelline vein is beginning to differentiate from the plexus on the. yolk sac. A compai'ison of the various embryos of this stage shows that the different systems of organs do not develop pari passu and it is impossible to arrange them correctly in a series by referring to a single character such as nmnber of somites or greatest length. In the present case the nervous s^^stem is relatively more differentiated than any other. It is necessary to seriate a limited group of embryos separately for each organ system. 2o. Endothelium and wandering connective tissue cells as seat of origin of hemoglobin. J. H. Globus, Cornell Medical School, New York City. In the course of an investigation of the blood elements in a large number of invertebrates, the writer selected three species of annelides as best adapted for the study of the origin of hemoglobin. The intimate relationship of this respiratory pigment with er^^throcytes had lead many investigators to believe that the solution of the problem of the origui and differentiation of red blood corpuscles would also uncover the ultimate source of origin of hemoglobin. This, perhaps, holds true of vertebrate blood, but in invertebrates and, especially, in annelides it is possible to find species with hemoglobin circulating free in the blood plasma. Here the way is open to trace the respiratory pigment to the group of cells, tissue or organ responsible for its foraiation. The three annelides studied show wide variations in the organization of their vascular system, while they retain great similarity in the rest of their internal structures. Arenicula Cristata a marine annelide, has a veiy highly developed vascular system, with pulsating heart, numerous large and small blood vessels, fonning a net-work about the gut and sending in branches into the body wall. Diopatra dibrahchiata, another marine annelide, shows a less complex vascular system, the vessels being arranged mainly about the gut and forming by their smaller branches a plexus. In Rhyncobolus, on the other hand, we find only a rudimentary ventral vessel, the blood fluid occilating in the haemocoele. PROCEEDINGS 357 In Arenicula oiu' liiuls in the intiM-niuscular connective tissue, a large ninnl»er()t' p,\iini cells loaded with brownish j-ounded jjif^incnt frranulcs. These cells when fixed are fountl to be ai'rested in various irrcfi^lar forms, indicatinji; amoeboid movements. Their i)roKress seemed to be in the direction of blood vessels. These blood vessels are surrounded by similar lar^e pigment cells which seem to form their limiting wall, as no other tissue, fibrous or muscular is there discernible. Some of these cells are arrested by the killing fluid in the process of emptying their granular contents into the lumen of the blood vessel. Their granules when stained with iron-detecting reagents indicate a rich iron content. They also react readily to Eosin and other acid dyes in a way similar to the hemoglobin containing plasma and erythrocytes in vertebrates. Other staining reactions indicate that (in some stages) these granules are modified nuclear derivatives. In some cells there are discernible three types of granules showing a transition from a nuclear to a plasma reaction, with the indifferent stage of brownish granules intervening. Thus, it is reasonable to believe that these cells are concerned with the manufacture of the respii'atory pigment. Shnilar conditions exist in Diopatra, with the only difference that the wandering mcsenchpnal cells are not as numerous, and that the endothelial cells are not as large. It is interesting to note that while in the above two species which have rather highly developed vascular channels, the hemoglobin is held in solution in the plasma, and can be traced to the cells lining the walls of the blood vessels in Rhj-ncobolus we have only a rudimentary vessel, and here the hemoglobin is incorporated in nucleated red blood corpuscles. Thus the origin of hemoglobin in this species is obscured and its development must be worked out along the lines of origin of the hematids. There is, however, sufficient ground to believe that these hematids are derived from the mesodermal layer of the wall of the gut. £6. Studies on internal secretion IV. Treatment of tadpoles with thyroid and thymus extracts. J. F. Gudernatsch, Department of Anatomy, Cornell University Medical College, New York City. In some previous experiments on growth and differentiation portion of fresh thj^roid, thjTiius and other glands were given as food to developing tadpoles. At the same time the animals lived in aqueous extracts of the entire glands, some of the tissue juices, of course, going into solution. In the experiments reported here an attempt was made to study the influence upon the development of tad-poles of several distinct constituents of the thyroid and thymus. By a detailed chemical procedure each organ was split into seven products, all in aqueous solutions.' 1 The solutions were prepared bj' the Department of Experimental Medicine. 358 AMERICAN ASSOCIATION OF ANATOMISTS Six of these portions contained each one 'agent/ the seventh, termed the alcohol soluble portion, combined two 'agents,' precipitate and filtrate, before the process of precipitation. The seven isolated portions were: Thyroid or thymus 1 Nucleo-proteins. 2. Globulins. 3. Coagulable proteins Residue (not used as such) 4. Alcohol soluble portion. 5. Alcohol insoluble portion 6. Filtrate 7. Precipitate The nitrogen content of the solutions was determined and the same concentration used for each set of tadpoles. An ordinary meat and vegetal diet was given. The thyroid and several of its constituents proved the most powerful stimulants for accelerated differentiation, while growth was suppressed almost entirely. When graded according to their action on differentiation the thyroid portions give the following list : (Thyroid, entire gland, dessicated; Parke, Davis and Company) Thyroid nucleo-proteins no growth Thyroid globulins Thyroid acohol insoluble portion Thyroid coagulable proteins Thyroid alcohol soluble portion Thyroid filtrate Thyroid precipitate greatest growth Control The last three products hardly showed any effect on differentiation, the tadpoles treated with the precipitate running ahnost parallel to the control. Tadpoles treated with the first four portions, however, metaihorphosed from five to six weeks before the control, while they never reached the size of the control anunals. Nucleo-proteins and globulins checked growth entirely. Three sets of the experiment started at different ages of the tadpoles —each interval being about four weeks— showed strikingly that older animals react to the thyroid treatment faster than younger, being, of course, nearer the stage of metamorphosis. In the youngest set the nucleo-proteins required twenty days to bring about metamorphosis, in the next set six days and in the oldest four days. It has been claimed that the amount of iodine present in the thyroid is mainly responsible for its action. It might be possible to determine this question by using the several thyroid products in the same iodine concentration. In the present experiment they were equalized as to PROCEEDINGS 350 their iiitroi!;oM contont. Cradod in regard to the io(hne present they riinj;e as follows: Nuolco-protcins Co:iKulal)le proteins and alcohol soluble portion Cllol)iilins Alcohol insoluble portion Filtrate Precipitate This coknnn does not run i:)arallel with the previous one, although the nucleo-protcins and the precipitate occupy the corresponding phices. Considering the action of the th^iiius as a whole, viz., retardation of differentiation, these experiments furnished some bewildering results. Only two of the seven products of the thymus delayed differentiation considerably, the alcohol soluble portion and the precipitate. The rest brought about differentiation simultaneously with or ahead of the control. It is impossible at present to explain these facts. It becomes necessary to assume either that the thjnnus furnishes several products of different physiological properties or that five of the seven constituents of the fresh gland are inactive. The latter assumption does not seem plausible; for the mass of the nucleo-proteins alone, for instance, is far greater than the a4cohol soluble portion. Beginning with that thymus product most active in retardation a graded list of the seven constituents would read as follows: Thymus precipitate Thymus alcohol soluble portion slowest growth (Thymus entire gland, dessicated; Parke, Davis and Company) Thymus filtrate Control Tliymus alcohol insoluble portion Thymus globulins Thymus nucleo-proteins fastest growth The expectation that the corresponding constituents of the two glands would be most active in counteracting each other was not realized; for it will at once be seen that in general the serial arrangement of the thymus products is ahnost the opposite of that of the thjToid series. It is impossible here to give all the growth curves. In the thyroid group the nucleo-proteins (strongest differentiator) permitted no growth, the filtrate and precipitate (weakest differentiators) allowed the tadpoles to grow to nomial size in about the normal time. Th\Toid filtrate and precipitate may well be considered inactive as far as differentiation is concerned. They may have other physiological properties not disclosed in these experiments. In fact, injection experiments on mammals revealed such properties, especially in the case of the filtrate. In the th^^nus group the nucleo-proteins gave most, the alcohol soluble portion least rapid growth. Both have their position at opposite ends of the above list showing their influence on differentiation. 360 AMERICAN ASSOCIATION OF ANATOMISTS 27. The effect of feeding Sudan III to albino rats. S. Hatai, The Wis tar Institute of Anatomy and Biology. The primaiy object of feeding Sudan III was to determine whether the mA^ehn sheaths could be stained as they were formed. This attempt failed entirelj^ but the alterations noted in the growth of the body and organs suggested the desirability of further studying these responses. The followmg is a brief summary of the results so far obtained by the use of 135 rats, representing 19 litters. 1. The amount of Sudan III given was 0.008 grams per rat per daj^ The Sudan was mixed with 1.5 cc. of ohve oil and given with the other food. 2. It has been found that rats less than 30 grams in weight do not grow at all when given the Sudan III in oil, while rats more than 30 grams but less than 50 grams in weight show an increase in weight for the first few days of feeding, but if the Sudan is continued they begin to emaciate. Rats which weigh more than 50 grams appear highly resistant to this dye and it takes several weeks of feeding before they show the effects seen in the younger animals. We find no noticeable injurious influence due to the oil, when given in the quantity here used. 3. When the effect of the Sudan III is marked, the thymus and sex glands show considerable diminution, while the liver and pancreas show noticeable increase in weight. In several cases the thymus gland had disappeared completely'. 4. The percentage of water in the liver and brain diminished, while that in the lungs, pancreas and blood increased. An increase in the alcohol-ether extracts was found in the liver, brain and blood. 5. As determined by a color test, a trace of Sudan was recovered from the lungs, pancreas and kidneys, while more than a trace was found in the hver. Complete absence of the dye was noted in the brain, spleen and heart. 6. All the Sudan rats were in a state of profound anemia. In one case nearh^ 40 per cent reduction of erythrocytes was found. 7. The present investigation shows that retardation of growth by the feeding of Sudan III can not be explained as due to a simple starvation, owing to the reduced availability of the fat after staining, as Riddle concludes, as there is clear evidence of more extensive pathological alterations such as anemia, fatty liver, nephritic kidney's, etc. Since Sudan III fails to enter and stain the myehn, it is highly probable that this substance is not obtained directly from the ingested food, but is very probably developed in situ. 28. The ganglionic crest and mesectoderm of the chick in relation to the closure of the neural tube. Francis W^ Heagey, Creighton Medical College, from the Department of Anatomy, Columbia University. The ganglionic crest of the trunk has long been known to differ from the mode of fomiation of that of the head. Neumayer in the Crocodilian embryo, Schulte and Tilney in cat embryos have shown that the ganglia of the cranial nerves were wholly derived from the PROCEEDINGS 301 nouraxis without tiie participation of the ectodenn so that the crest was not an element intermediate between the neural tube and the somatic ectoilerm. In view of this it was su^^csted that in more cci)halic portions of the neural tube where ganglia were not formed that tlie ganglionic element might form a permanent constituent of the wall of the brain. In an effort to see whether there was an equivalent mode of fonnation in the sauropsids and what part the mcsectoderm plays in the formation of the ganglionic crest I incubated and sectioned forty chick embryos, exercising great cai'e in the determination of the age of each embryo. Weigert's hcmotoxylin was used for the staining of the sections. There were several marked differences from the mammalian type of the ganglionic crest fonnation. The medullary plate imlike that of the cat, showed no p\aee of sudden transition from the neural to the somatic portions, i.e., there was no sharp neuro-somatic junction. Whereas in the mammal the cranial nerve anlage appeared as derivatives of the neiu'axis in the angle between somatic ectodenn and the neural tube before the closure of the latter; in the sauropsid there was no evidence of any similar outgrowth until after the closure of the neural tube and the separation of the latter from the ectoderm. In the formation of the neural tube the amount of the medullary plate which was assimilated by the neural tube in the head region varied considerably from embryo to embryo. As a consequence the area of the surfaces of the medullary folds opposed to fonn the neural tube varied and the resulting stem which connected the neural tube with the ectodenn was either long or short depending upon the above conditions. Then bilateral cleavage planes appeared which separated the neural tube from the ectoderm and in so doing divided the stem into two portions, one a group of cells connected with, the ventral surface of the ectodenn and the other portion was connected with the dorsal fusion point of the neural tube; from the fonner the mesectodemi is derived exclusively. The mesectoderm was present onh^ from six to nine somites and in the prequintal region exclusively. It had no connection with the neural tube or relation with the ganglionic crest. Its presence depended upon the amount of the medullary plate included in the neural tube and the level of the cleavage planes through the stem of the neural tube. The portion of the stem of the neural tube remaining attached to the dorsal fusion point of the neural tube assumed different shapes depending upon the region and the relations to the cranial nerve anlages. In the intervals between the cranial nerve anlages it was merely a wedge of cells inserted between the dorsal extremities of the neural tube. In the region of these nerve anlages it formed a prominent mound of cells from whose sides lateral processes were seen to project. The cells of the dorsal fusion point of the neural tube are a part of a true ganglionic crest because of their position in the neural tube, their relation to the cranial nerve anlage and their intimate connection with 362 AMERICAN ASSOCIATION OF ANATOMISTS the lateral processes on the dorsum of the neural tube in the quintal and the aeustico-facial regions. Based on these facts it is reasonable to conclude that the ganglionic crest of the sauropsid like the mammalian forai is inherent in the neural tube and, that the mesectoderm does not participate in the ganglionic crest or the cranial nerve anlage but contributes solely to the head mesenchyme. 29. Maturation and initiation of development in Cumingia. L. V. Heil BEUNN, (introduced bj^ A. C. Eycleshymer,) Department of Anat oray, University of Illinois, College of Medicine. The egg of the mollusc Cumingia is immature when shed into the sea-water, and it remains immature unless it is fertilized. Soon after sperm entrance, the first polar body is given off, and this is the first step in the developmental process^ Artificially, polar body formation can be induced in any one of three ways. In the first place, if the surface tension of the vitelline membrane is markedly lowered, it rises from the egg and maturation follows. This effect is in general produced by all substances which markedly lower surface tension, irrespective of their chemical constitution. Although many such substances were tested, only one or two failed to produce maturation, and in these cases the egg was very evidently injured by the reagent. Secondly, if the stiff vitelline membrane of the egg is caused to absorb water and swell, then maturation also occurs. Such a swelling effect is produced by certain salts, and by dilute acids and alkalies. Finally, if the vitelline membrane is removed by shaking, or if it is ruptured bj^ placing the eggs in diluted sea-water, the egg can be made to undergo maturation. All three of these treatments have one effect in common. The stiff vitelline membrane which encloses the egg is replaced by a much more plastic film. Thus in the Cumingia egg, as in the sea-urchin egg,' development can only occur if the egg is freed from the constraint of a stiff vitelline membrane. Moreover, the same types of reagents pror duce membrane elevation and membrane swelling in both eggs. But in Cumingia, unlike Arbacia, it is a membrane swelling rather than a membrane elevation which is the normal process. Of the two leading theories of initiation of development, neither could be applied to the Cumingia egg. No evidence could be found for increase of permeability. On the other hand, the maturation process takes place normally in concentrations of potassium cyanide two hundred times as great as those supposed to check oxidations. Although polar body formation can readily be produced by the methods just described, in the great majority of cases, segmentation does not follow. As in the sea-urchin egg, a gelatinization or coagulation can be demonstrated to precede segmentation. iHeilbrunn, '15, Biol. Bull., vol. 29, p. 149; v. especially p. 183. PROCEEDINGS 3G3 Thus it is apparent that the theories advanced by tlie writer to exphiin the initiation of development in the sea-urchin egp;, are cUrectly apphcahle to the e^^ of Cuniingia. The only essential difference between the e^f2;s lies in the fact that in the Cuminf:;ia e^fl, cortical change sthnulates to maturation, whereas in the sea-urchin efrg the maturation process has already been completed before the egg is fertilized. SO. On thijroidedonnj in amphibia. E. R. Hoskixs and Margaret Morris, New York and Yale Universities. Experiments performed at the Yale Zoology- Department. (Presented by Dr. Hoskins.) With due care to technique it was possible to remove successfully, the anlage of the thyroid gland, from young growing larvae of R. sylvatica and Amblystoma. The stage best suited for this experiment is that just preceding the beginning of the circulation of the blood. At this time there is no danger of hemorrhage and the chances of regeneration of the removed gland are fewer than with younger larvae. Chlorotone in salt solution was used to produce anesthesia. The experiments are to be repeated on an extensive scale the coming season, as many more data are needed for final conclusions. The following results were obtained from thyroidectomy in 40 R. sylvaticae and 50 Amblystomae checked against an equal number of control anhnals. A few of the thyroidectomized frog larvae developed abnormally shaped external gills in some of which, no circulation was to be seen. This was evidently due to injuiy to the vascular system. One animal developed no external gills although it lived and grew through the period during which external gills nomially persist. From time to time larvae were killed and fixed. A number of them, both control and experimental died and were lost. The operated animals grew less rapidly than the controls. Only one control and one experimental animal survived the nomial period of metamorphosis. Of these the control showed hind legs two months after the operation and the other had not developed legs four months after the operation. The operated lar^-ae showed no marked tendency toward albinism. Serial sections were made of eight experimental larvae. The operation was seen to have prevented development of the thjToid gland in all but one case. The hypophysis as compared with that of the controls showed no changes in size or structure to be attributed to loss of the thyroid gland. Among the amblystomae none developed abnormal gills. The average growth rate of the experimental lar^^ae was less than that of the controls, but of the fourteen which were aUve, after three months, the largest had had the thyroid removed. In none of the thirteen operated animals that were sectioned was there any regeneration of the thjToid. There were no changes in the hypophysis nor in the pigmentation of the skin following the thyroidectomy. THE ANATOMICAL RECORD, VOL. 11, NO. 6 364 AMERICAN ASSOCIATION OF ANATOMISTS 31. Preliminary remarks on a collection of eleven gorilla brains recently acquired by the Division of Physical Anthropology, U. S. National Museum. (With demonstration of casts.) A. Hrdlicka, Smithsonian Institution, United States National Miisemn. Through fortunate circumstances the Division of Physical Anthroi:)olog>', U. S. N. M., has recently acquired, from the Kameroons, a pi-ecious collection of 14 brains of anthropoid apes. Of these brains no less than 11 are those of gorillas, while 3 are chimpanzees; and of the gorilla brains, 5 are adult, the rest ranging from young to practically full grown. Of the chimpanzees, 1 is adult, 2 young. The whole collection is. in a remarkably good state of preservation, with the exception of two brains of the young, and offers unprecedented opportimities for study. The form of the brains, the fissure pattern and the variations, are of much interest. S2. The morphological basis for the dominant pulmonary asymmetry in the maynmolia. Geo. S. Huntington, Columbia University. The development of an eparterial bronchus only on the right side, and the pulmonary as>amiietry resulting therefrom, is prevalent in at least 95 per cent of the mammalian genera and species whose intrapulmonary architecture is known, and the etiological factors responsible for this condition have been much discussed. The as^nmnetry has in a general sense been ascribed to the left-sided position of the heart and aorta in the mammalia and to the resulting cm-tailment of the left thoracic space available for pulmonary extension. The difference in the mechanics of the intrapulmonary circulation in the two lungs during the placental period, due to the retention of the dorsal terminal of the sixth left aortic arch as the Botallian duct, has also been cited as a possible cause contributor}^ to as^T^nmetrical pulmonary development. In the early ontogenetic stages the I'otation of the stomach has also appeared as offering a bar to the caudal progress of the primitive left lung-tube. Aeby ('80) considered the dominant asjamiietrical bronchial type of the mammalia to be derived from an ancestral bilaterally s^aimietrical eparterial condition by the phylogenetically acquired reduction and su})sequent complete elimination of the left eparterial component, the right eparterial l^ronchus alone persisting and producing the asjmimetry. Aeby offered no morphological evidence in support of this assumption. Such evidence was apparently supplied in 1896 by d'Hardi^iller, who described in rabbit embryos of the thirteenth day a transient bronchial vesicle arising from the left stembronchus, in the same position and in the same relation to the pulmonary artery as the eparterial bronchial bud of the right side. This left eparterial bud appears at the beginning of the thirteenth day, develops into a distinct hollow epithelial vesicle, whose lumen connects by a narrow canal with that of the left stembronchus, and then i-etrogrades rapidl}'. By the end of the thir PROCEEDINGS 365 tocMith day it is riHlucod to a solid epithelial hutloii, havinj^ lost its lumen and tlic open connnunication with the stenil)ionduis, to which it is now attached solely by a solid epithelial jjedicle. By the fourteenth day it disajipcars entirely, leaving no trace of its ephemeral existence. D'Hardiviller finds in this discovery the absolute ontogenetic proof establishing Aeb>'s hypothesis of an archeal bilateral symmetrical epartcrial component of the bronchial tree, from which the dominant modern asjnnmetrical type, with the epartcrial element confined to the right side, evolved through the jihylogenetic loss of the left epartcrial bronchus. The latter ajipears in the ontogeny for only a very short period as the temporar}' derivative from the left stembronchus above described. D'Haidiviller does not mention the number of 13 day embryos in which he found this evanescent vesicle. The context of his publication implies, however, that it is of constant occurrence. No subsequent confirmation of d'Hardiviller's observ^ation has been made. In no embryos of the critical stages accessible to me, either of the rnljbit, or of other mammalian fomis (cat, rat), was there any trace of the structure in question. On the other hand in a series of 70 adult rabbit lungs, which I examined by corrosion for the occurrence of bronchial variants, I found the cephalic pole of the left lung, normally supplied by the ascending branch of the first left ventral hyparterial bronchus, supplied in one individual by an atypical first side branch of the left stembronchus arising dorsal to the pulmonary arter}' and corresponding in position to the larger right epartcrial bronchus. Nai'ath, in describing the bronchial variants in a series of 39 adult rabbits, has reported somewhat similar conditions in two individuals. The conclusion appears justified that, on the evidence at present available, d'Hardiviller's observation was made on one or more variant embryos which, if development had proceeded, would have j'ielded atypical adult individuals possessing the above described left bronchial variant, but that no warrant is given for the assumption that such embryonic variants have a phylogenetic significance in the interpretation of the normal intrapulmonary architectonics. A detailed study of the topographical relations of the developing mammalian lung in the critical stages reveals clearly the reason for the prevalent right-sided epartcrial development, and at the same time shows the possibility of the occasional sprouting of a homologous bud from the left stembronchus. The conditions, while strongly favoring the development of the epartcrial component of the right bronchial tree, are all against a corresponding bronchial development on the left side. The possibility of such an occurrence exists, as shown by the comparatively rare variation of the left epartcrial bronchus, both in the embryo and in adult individuals of mammalian fomis with typicall}^ as^Tnmetrical bronchial tree (Echidna, Choloepus, Lepus, Tragulus, Homo), and by the normal bilateral epartcrial type found in cer 366 AMERICAN ASSOCIATION OF ANATOMISTS tain limited mammalian groups (some aquatic Carnivores and Rodents; some Cetaceans; the Camelidae among Ungulates, and the genus Cebus among Primates; Elephas, Hj^rax). Rabbit embryos of 10 mm. and 11 mm. show in transverse section the ventral wall of the oesophagus with the two vagi as the background against which the tracheal bifurcation, the stembronchi and their primary buds are placed. The pulmonary arteries accede to the ventrolateral cu-cumference of the trachea, along which they at first descend fau-ly s\Tnmetrically on each side. In approaching the tracheal bifurcation two of the extra-pulmonary structures gradually change their position relative to each other and to the pulmonaiy tube. The left vagus, which here is larger than the right nerve and forms a massive cord, turns ventrad. This is the expression of the rotation of the foregut to the right through which the left side of the gastric enlargement becomes directed ventrad. At the same time the left pulmonary artery turns caudo-dorsad in obedience to the sinistral axial twist of the heart and of its arterial pedicle. As the result of these two rotations in opposite directions of the fore-gut and of the heart, the left vagus and left pulmonary artery approach each other and are, at the level of the tracheal bifurcation and of the origin of the stembronchi, crowded closely together. The reverse obtains on the right side. The right pulmonary artery turns more and more ventrad in descending, while the right vagus moves dorsad. An arterio-neural interval is thus opened up on the right side toward which the lateral circumference of the right stembronchus faces directly and into which it sends the bronchial bud responsible for the unfolding of the right eparterial bronchus. The latter thus comes to be placed between the right vagus and the right side of the oesophagus dorsally and the right puhnonary artery ventrally. On the left side this arterioneural portal for eparterial development has been blocked by the approxmiation of left vagus and left pulmonary artery, or definitely narrowed to such an extent that it no longer affords a favorable path for extension from the left stembronchus. The first side-branch of the latter is hence forced to pass in front of the artery and thus becomes the first ventral hyparterial bronchus of the left "side. This supplies by means of its large ascending branch the cephalic pole of the left lung, the same area which on the right side receives the eparterial bronchus. The narrow interval between left vagus and left puhnonary artery might occasionally suffice for the passage of a bud from the left stenibronchus contributing the cephalic portion of the left lung. In such a case, the adult variants above mentioned and the embryonic left eparterial bronchial bud described by d'Hardiviller, on which they are based, would be found. The ascending branch from the first left ventral hyparterial bronchus, usually supplying this area, would then be correspondingly reduced. It is significant to note in this connection that in the rabbit, in which form both the adult variations above mentioned and their embrvonic PROCEEDINGS 3(37 aiiluKo have licen fouiul, the cephalo-ventral puhiionary extension fonnuifr the upper pole of tJio left hirifr, is very muikedlv reduced con'ipared with tlie ri.i>ht side. Tlie impulse to send ;in eparteiial hud from the left stembront'hus throu^li the narrow va^o-iirtcrial interval, in spite ot the small available si)ace, would he fostered in this form hv the reilucod area it would he called upon to serve. I believe that the developmental conditions just outlined furnish the juleciuate explanation of the prevalent asymmetry of the mammalian l)ronchial tree. This asymmetry is primarily founded on the different opportunity for cephalo-ventral pulmonary expansion afforded usually to the right and the left lung respectively, in consequence of the cardiac and oesophageal rotation in opposite directions. The sinistral tui-n of the heart and its influence on the initial direction taken by the pulmonary- arteries is probably dependent in the first instance on the development of the left-sided mammalian aorta and its association with the left Botalhan duct. This cardiac turn directs the left pulmonary arten- dorsad and favors the caudal extension of this vessel in the left mediastinal background. By contrast, on the right side, the cardiac rotation directs the pulmonary arterv ventrad, unimpeded by the retention of a communication with the systemic arterial arches, and this forward thrust carries into the caudal extension of the vessel. The enonnous mechanical force exerted by the Botallian duct on the confonnation of the adult arterial pattern and heart is strikingly demonstrated in human major arterial variation by the figure-of-eight twist of a right thoracic aorta with the left subclavian arteiy as its last primary branch, arising from the descending aorta via a retained left dorsal aortic arch, with a left ductus arteriosus in place of a right, as should occur in right aortic arch development. This cardiac rotation, which imparts the initial direction to the further groMih caudad of the pulmonarv arteries, is met, in its influence on pulmonary development, by the rotation of the fore-gut in the opposite direction, carrying the vagi with it. These two factors cooperate m detemiinmg the plan of pulmonarv development. Whereas on the right side the vago-arterial interval is widened as a result of these rotations and the avenue for the right eparterial expansion is opened up, the path on the left side is blocked by the approximation of vagus and left pulmonary artery. It is especially the massive trunk ot the nerve which would stand in the way of a bud advancing from the left stembronchus dorsal to the artery, or would, at the most, permit its development only rarely, as a variant, supplying a limited peripheral pulmonary area. Absolute proof of the above outlined genesis of mammalian pulmonary- asAnnmetry could of course only be obtained by the study of the proper stages on embryos of mammalian forms possessing normally the bilateral s^inmetrical eparterial type of bronchial tree. This material has so far not been attainable. I have observed, however, that in the Sirenia and in some Cetaceans and pmmpede Carnivores the heart in the adult is practically median 368 AMERICAN ASSOCIATION OF ANATOMISTS in position with slight, if any, axial rotation. The apex, fonning the ventro-caudal point of the heart-cone (bifid in Manatus), is nearly in the median line, and the right and left ventricles have approximately an equal share in the ventricular area of the sterno-costal surface. The stomach of Phoca vitulina is almost vertical with the omental borders approxmiating the median plane, in marked contrast to the transverse position of the completely rotated t^'pical mammalian stomach, whose original left surface is directed obliquely ventro-cephalad in distension. I am inclined to believe that the embryos of bilateral eparterial forms will show a diminished degree of both cardiac and gasti'ic rotation, with consequent greater equality of the vago-arterial interval and a more even opportunity for the development of eparterial bronchial components on both sides. I am deeply indebted to Prof. F. T. Lewis of Harvard University, for the oppoi'tunity of studying three rabbit embrvos of the Harvard Embryological Collection (Series 155, 1327 and 1658). These preparations, perfectly fixed, stained and sectioned, proved of the utmost value. SS. Effects of inanition and refeeding upon the growth and structure of the hypophysis iii the albino rat. C. M. Jackson, Institute of Anatomy, University of Minnesota. Sections (usually serial) of 3 yu to 5 /x were made of the hypophysis from 91 i-ats, of both sexes. These included 44 normal (newborn to one 3'ear), 15 held at maintenance (constant bodv weight) for various periods by underfeeding beginning at the age of three weeks, 6 adults subjected to acute inanition and 5 to chronic inanition, and 21 .young rats refed for various periods after being held at maintenance from the age of thi-ee to twelve weeks. The material was fixed in Zenker's fluid, and usually stained with haematoxylin-eosin, occasionally by other methods. Volumetric data were obtained for the parts (lobes) of the hvpophysis (29 cases), and, in the pars anterior (distalis), for the vessels and stroma, and the nuclei and cytoplasm of the parench-\Tna (8 cases). The method used was by projection upon paper, the desired areas being cut out and weighed, and the corresponding volumes computed. Direct measurements were also made with the filar micrometer. 1. Volumes of the parts (lobes) of the hypophysis During normal postnatal growth, there is considerable individual variation in the relative volumes of the various lobes; but on comparing the younger (newborn to three weeks) with the older (ten weeks and above) it appears that in general the pai'S anterior (distalis) becomes relatively larger, and the pars nervosa correspondingly smaller, the pars intermedia remaining about the same in relative (percentage) volume. PROCEEDINGS 3G9 That the hypophysis is rclativoly heavier in the female rat is already known. The present data indicate that this is due (though perhaps not entirely) to a larger pars anterior in the female. 4 males, average. . . 4 females, average. per cent 82.0 86.4 per rent 9.7 G.7 PARS NERV. per cent 8.3 7.0 Thus in the female the pars anterior appears to have gained, with a relative decrease in the pars intermedia and, to a lesser extent, in the pars nervosa. In three rats held at maintenance (nearly constant body weight) from age of three to ten or twehe weeks, the pars anterior appears slightly reduced in relative volume, the partes intermedia and nervosa correspondingl}' larger. In an adult rat with chronic inanition, the partes anterior and intermedia appear relatively reduced, the pars nervosa increased. In two adults with acute inanition, the pars anterior appears slightly increased, the pars intennedia correspondingly decreased, and the pars neivosa unchanged in relative size. In several rats refed one-half week, one week, two weeks and four weeks after maintenance from tln-ee to twelve weeks of age, there is some variability, but in general a gradual return to the normal proportions in the lobes of the hypophvsis, the pars anterior becoming relatively larger, the partes intermedia and nervosa smaller. In two adult rats which were refed six or seven months after a long period of maintenance (three weeks to five months of age), although the hypophysis as a whole and the body weight are nearly normal, the relative volumes of the lobes are abnormal. The pars anterior is rclativel}' low and the pars nervosa high. This resembles the change found after maintenance and chronic inanition, and may be a persistent effect of the prolonged inanition. 3. Relative size of the components in the pars anterior (Distalis) In the pars anterior of the normal newborn rat, the vessels and associated stroma form 6.7 per cent bj'- volume, increasing to 9.G per cent at three weeks, and to 10.6 per cent at ten weeks. In 3'oung animals held at maintenance, the volimie of the vascular stroma usually increases to about 13 per cent, and in adults under acute or chronic inanition to about 17 per cent. The parench\Tna is of course correspondingly reduced in relative volume. In the parenchjTiia of the pars anterior, the nuclei form about 34 per cent of the total cell volume in the newborn, decreasing to about 24 per cent at three weeks, and 20 per cent at ten weeks. The cytoplasm undergoes a corresponding increase in relative volume. During inanition, the loss is usually relatively greater in the cytoplasm, the 370 AMERICAN ASSOCIATION OF ANATOMISTS nuclei thereby increasing to 26 to 28 per cent of the cell volume in the young at maintenance, and to 23 to 26 per cent in adults with chronic or acute inanition. From the volumetric data, the (calculated) average diameter of the parenchAmia cells increases from 10.1 ^ in the normal newborn to 11.9 /J, at three weeks and 13.6 /j. at ten weeks. In the young rats at maintenance the average cell diameter is 9.7 to 10.2 /x and in the adult starved rats 10.0 to 11.0 m- The nuclei average 5.9 /i in the normal newborn; 5.8 /x at three weeks; 6.0 /j, at ten weeks. In the young rats at maintenance the average nuclear diameter is reduced, being 4.9 to 5.3 fx; in starved adults, 5.3 to 5.5 /jl. Direct measurements of the nuclei with filar micrometer give results in fair agreement with those above. The nuclei of eosinophiles are usually slightly below the general average. Nuclei of the pars intermedia are usuallj^ near those of the, pars anterior in average diameter. For all data there is a considerable amount of variability, both in various individuals and in various parts of the gland in the same individual. S. Histological structure Mitoses. In forty cases, the number of mitoses in entire coronal sections of the hypophysis was counted. Marked individual variations occur. In no case was amitosis observed. In the newborn, mitoses are very numerous throughout, the average being 62 in each section of the pars anterior, 9 in the pars intermedia and 7 in the pars nervosa. In the pars nervosa, mitoses soon disappear. At seven days, they are rare, and none were found later (growth thereafter consisting chiefly in the increase of intercellular substance). In the pars intermedia mitosis continues at a diminished rate, the average being 1 in each section at three weeks. At ten weeks and later, they are very scarce. In the normal pars anterior, the rate of mitosis likewise decreases, the average number being 62 at birth, 18 at one week, 7 at three weeks, 2 at ten weeks, and rare in adults. In young rats held at maintenance from three to ten weeks of age, mitosis has nearly ceased. No mitoses were found in the partes nervosa and intermedia, although in the pars anterior they still occur occasionally, even in rats nearly dead from inanition. No mitoses were observed in the starved adults. In the young rats refed after the maintenance period, mitoses reappear promptly in the pars anterior, the average number per section being about 2 after one-half week of refeeding, 7 after one week to two weeks, decreasing to an average of 3 after four weeks of refeeding. Mitoses were observed but rarely in the pars intermedia and never in the pars nervosa. The rate of mitosis in the refed rats therefore corresponds in general to that of younger normal rats of similar body weight. Cell-structure. In the pars nervosa, the only change noted during inanition is a vaiialjle degree of hyperchromatism in the nuclei, which rarely may become shrunken and pycnotic. PROCEEDINGS 371 In the pars intcnuedia, the cells usuall}' suffer relatively little cliange in structure during inanition. The nuclei have a variable tendency to hypcrchroniatisni, occasionally becoming p^^notic. The cytoplasm tends to lose its granular structure, becoming more homogeneous in appearance, often finely vacuolated. Around pycnotic nuclei, it is usually moie deeply l)asoi)hilic. Occasionally the cytoplasm may be greatly atrophied and reduced in volume, leaving the nuclei very closely packed. In the pars anterior, the changes during inanition are variable. Some areas may remain nearly normal, while others in the same gland show extreme changes of atrophy and degeneration. The cytolasm is greatly reduced in volume (as shown above), and is fi-equently vacuolated. There is a marked tendency to loss of the specific staining reactions, so that sti'ongly chromophilic cells become weakly chromophilic or even chromophobic. The nuclear changes in the pars anterior are likewise variable in extent and character. There is, however, a veiy general tendency to hype.rchromatism, often reaching marked pycnosis. Karyorrhexis and karyolysis are rare. Upon refeeding one-half week after the maintenance period (three to twelve weelcs), the hj-pophysis retains the typical inanition structure, although mitosis has begun. After one week of refeeding some areas have become nearly normal, and after two weeks the normal structure pre])onderates. After four weeks, the hypophysis has usually become nearly normal for the most part, although more or less extensive atrophic areas may persist for indefinite periods. 34- The later development of the lobule of the pig's liver. (Lantern slides.) Fraxklix p. Johxsox, University' of ^Missouri. Up until late fetal life the lobules of the pig's liver are fused together and form a continuous mass of liver cells, a condition not especially unlike that found in most adult mammals. It is in stages just before bii'th before any evidence of a segmentation of the liver parench>Tna becomes apparent, and the completion of the formation of connective tissue septae is not fully accomplished until several months after birth. In the separation of the liver parench>Tna the liver cells themselves apparently take some part. The cells along the boundaries of the lobules become granular and stain more readily than the cells elsewhere. Soon they become arranged in parallel rows, or rather sheets, extending from one branch of a portal vein to another, and thus definitely mark out the future lobules. The rows of cells become split apart by a thickening of the reticulum between them. Collagen fibers gradually spread into this thickened reticulum from around the portal veins. The question of growth of the lobules is of particular interest. Mall (Am. Jour. Anat., vol. 5, 1906) states that "in the pig the lobule measures 0.8 mm. in embryos 4 cm. long until a number of months after bii'th," adding that "in the adult they are 1.4 cm. in diameter." IVIy own observations are at variance with these. Thus I find beginning 372 AMERICAN ASSOCIATION OF ANATOMISTS with pmbrvos old enough in which the lobule boundaries are easily recognized, that the lobules gradually increase in size. The following table gi^'ing the average diameter of the largest lobules demonstrates this point: AGE DIAMETER AGE DIAMETER 229 mm. 254 mm. 1 day 3 days mm, 0.35 0.43 0.43 0.45 3 weeks 4 weeks 2 months Adult mm. 0.49 0.51 0.59 1.2 That there is an actual increase in the number of lobules even after the connective tissue septae are formed appears evident from estimates of the total number of lobules based upon volumetric calculations. The additional lobules are formed by a splitting up of certain large lobules. In sections one finds evidence of this in incomplete septae, developing appai-ently after the manner of the earliest ones. The conception that the lobules of the liver conform to a general shape and size can be easily disproven. If small blocks of pig liver are treated with 20 per cent nitric acid at 50°C. for several hours, the connective tissue septae are destroyed. The lobules can then be readily teased apart without injuring them. Thus seen, they present a great variety of shapes, some with rounded surfaces and borders, others with flat or concave suifaces and angular borders, some approaching prisms, pyramids and polyhedrons, but the majority are so irregular thev liken none of the regular geometrical solids. In size the lobules likewise present great valuations within the same liver, some being at least five or six times as large as others. In the adult pig the lobules average about 1.8 cubic mm. in volume, at two months, 0.31 cubic mm., and at four weeks, 0.098 cubic mm. 35. Aortic cell clusters in vertebrate embryos. H. E. Jordan, Department of Anatomy, Univei'sity of Virginia. Aortic cell clusters were first described by Maximow in 1909 in rabbit embryos (Arch. f. mikr. Anat., Bd. 73, p. 517). Minot subsequently ('12) described similar structures in human embryos of from 8 to 10 mm. length and in rabbit embryos (Keibel and Mall, Human Embr3'Olog\^ p. 523). Emmel reported ('15) aortic cell clusters in rat embryos, rabbit embryos, and in pig embryos of from 6 to 15 mm. (Anat. Roc, vol. 9, p. 77). Joi-dan discovered these clusters in pig embryos (10 to ]2 mm.) at about this same time, and reported their presence also in mongoose and turtle embryos (Anat. Rec, vol. 10, p. 417). Emmel later ('16) published a detailed description of the aortic clusters of the pig embrvo (Am. Jour. Anat., vol. 19, p. 401). Meanwhile I had observed them also in the chick embryos of three to four daj's' incubation. Danchakoff had already ('07) reported similar struc PROCEEDINGS 373 tiires ill the chick (I<oha Hacniatoloffica, vol. 4, p. 159). Aortic coll ehisters would seem to be a coininon feature of certain early stages of vertebrate developinent. All of thv ab()\o-iiamecl investigators agree in interpreting the constituent cells of the clusters as henioblasts. Minot alone (1 c.) disagrees with the otheiwisc unanimous conclusion that they fepresent endothelial differontiation products. That they are endothelial derivatives, however, rather than accretion products of henioblasts from the circulating embryonic blood is easy of demonstration. Aortic cell clusters represent one phase of the general hemogenic capacity of embryonic endothelium. The doctrine of the partial origin of henioblasts from embryonic endothelium lias become associated witli the monophyletic hypothesis of blood cell origin (Maximow I.e.) ; that of the non-hemogenic capacity of endothelium with the polyphvletic and diphvletic hvpotheses (btockard, Am. Jour. Anat., vol. 18, p. 227). The question of the genetic relationship between endothelium and certain cellular elements of the embryonic blood touches also the 'angioblast' theory of His. The two chief tenets of this theory are: 1) the inability of intraembryonic mesenchyme to produce blood vascular tissue, and 2) the incapacitv of endothehum to differentiate nonnally into blood cells. Abandonment of the first tenet has been forced largelv through the e.xpei'imental work of Hahn (Arch. Entwickmech., Bd. 27, 1909) of Miller and Mc^^horter (Anat. Rec, vol. 8, 1914), of Reagan (Anat. Rec vol. 9, 19lD), and of Stockard (1. c), and the morphologic studies of bchulte on the cat embr\^o (Mem. Wistar Inst., No. 3, 1914) those of McClure on the trout embryo (Mem. Wistar Inst., No. 4, 1915) and the studies of Huntington on the development of the lymphatics in amniotes (Am. Jour. Anat., vol. 16, 1914). The disproof of the second tenet is the chief burden of this investigation. The material studied includes three mongoose embryos of 5, 6 and 7 mm. respectively, a series of pig embrvos of from 8 to 12 mm' chick embr^'os of the third and fourth dav of incubation, and a series of twenty loggerhead turtle embryos ranging from the second to the thirty-second day of incubation. These embryos are variously preserved and stained, the several methods including fixation with Helly's fluid and staining with Giemsa's solution. This description confines itself ahnost exclusively to the 5 mm mongoose embryo and the twelve-day turtle embryo. This selected material IS at just the proper stage of development to furnish the key for the correct interpretation of the larger aortic clusters of the 10 'mm pig embryos. The study was approached by way of the volk-sac of the mongoose embryo. The endothelial origin of henioblasts can here be readily demonstrated. These observations on the mongoose volk-sac confirm my previous findings regarding the hemogenic role of yolk-sac endothelium m the 10 mm. pig embryo (Am. Jour. Anat., vol. 19, p. 277) 374 AMERICAN ASSOCIATION OF ANATOMISTS The second step involved a search for similar intraembryonic phenomena. It seemed reasonable to expect that, since the yolk-sac mesench>Tne could differentiate directly into blood cells and into endotheUimi. and since the endothelium could subsequently transform into blood cells, then the same order of events should probably follow also in the intraembryonic mesenchyme; and further, since mesenchyme is the fundamental hemogenic tissue, and since both endothelium and mesothelium in the embryo are only slightly modified mesenchyme, then embryonic mesothelium and endothehum should both, in the only slightly differentiated condition, be capable of producing cellular blood elements. That mesothelium can differentiate into vascular tissue has been' shown by Bremer in the case of the body stalk of a 1 mm. human embryo (Am. Jour. Anat., vol. 16, p. 447). Examination of the intraembryonic endothelium in the pig and mongoose revealed, in the smaller pericerebral blood channels, an occasional endothelial cell rounding up and taking on hemoblast features and finally separating from the endothelial wall; and led to the discovery and detailed study of the aortic clusters of hemoblasts, with the origin and significance of which this study is largely concerned. Moreover, investigation of the pericardial mesothelium disclosed very similar clusters, both attached to the visceral and the parietal pericardium and lying free within the pericardial cavity. Emmel has recently described comparable structures in the 12 mm. pig embryo (Am. Jour. Anat., vol. 20, p. 73). Occasional individual cells can also be seen in process of separation from the visceral pericardium in the mongoose embryo. As regards the aortic cell clusters, the 5 mm. mongoose embryo shows admirably various early stages in their origin and development, and so furnishes the key to the interpretation of the later products. And the 12 day loggerhead turtle embryo shows besides, the peculiar intravascular encapsulated cell clusters, and the endothelial strands, recently noted also for a 12 mm. pig embryo by Emmel, in a footnote to his paper (p. 407) on aortic cell clusters in mammals; and the conditions in this respect also are such as appear to solve the mystery of their genetic significance. The aortic cell clusters in the mongoose embryo of from 5 to 7 mm. range from such as are composed of only a single cell to those composed of a score or more. Single cells or groups of two or three can be seen separating from the endothelium at any point, even along the mid-dorsal line. Larger groups are found only in the ventral and ventrolateral portions, frequently in more or less close relation to the mouths of the lateral mesonephric branches or the ventral intestinal rami. This proliferative activity of the aortic endothelium is present only in the abdominal portion of the aorta, approximately coextensive with the mesonephroi. Single endothelial cells may round up and take on hemoblast features and separate from the wall in exactly the same manner as that by which the hemoblasts are derived from the endothelium of the yolk-sac vessels and in the pericerebral vascular channels. PROCEEDINGS 375 The process is tlic same in the yolk sac and the embryo, and indicates a common hemogenic capacity of eml)ryonic endothelium. The mongoose material shows also the initial stages in the formation of the larger cell clusters. Throughout the ventral half of the alxlominal aorta, the endotlielhnn at certain points appears to buckle into the hmien. This invaginated area may be more or less extensive, and may include a considerable portion or none of the subjacent mesenchyme. The cause of the buckling remains obscure, though the suggestion lies close to hand that it may be related to the caudal shifting of the embryonic representatives of the celiac, superior mesenteric and inferior mesenteric arteries; a process dependent in part at least upon the presence of a less rigid and less differentiated endothelium ventrally, permitting thus of an inequality of growth as between the ventral and dorsal walls or allowing for the formation of successively lower connecting vascular segments for the migrating definitive stems. The endothelium seems to be lacking centrally underneath the cell clusters. This is explained by the fact that the larger clusters arise by an invagination of the endothelium over an area of some extent rather than by process of proliferation of one or several differentiating endothelial cells. Proximally the clusters show transition stages between endothelial cells and hemoblasts (laterally) and between mesenchymal cells and hemoblasts (centrally). Within the clusters some of the cells are in mitosis, while the nuclei of others may appear at some phase of amitotic division; and an occasional cell may show phagocytic properties. Sometimes the core of the cluster shows transition stages between endothelium or mesenchyme and hemoblasts. Many of the nuclei subjacent to the cluster appear to be at some phase of amitotic division. The aortic cell clusters of the mongoose embrj^o originate from the cells of an invaginated area of endothelium; they enlarge by intrinsic growth and differentiation, not by accretions from the circulating blood. Similar clusters appear also in the superior mesenteric artery. In a 10 mm. pig embryo a large aortic cluster, 130 n in diameter, occurs near the mouth of the superior mesenteric artery and consists of a hundred or more cells. Clusters occur also along the greater length of this definitive aortic stem. In the twelve-daj^ loggerhead turtle embryo, encapsulated clusters and extensive strings of hemoblasts attached to the endothelium appear in the inferior vena cava, near the point of fusion of the original paired subcardinal veins. The endothelial strands, some of the cells of which bear hemoblast features, are most probably only another aspect of the general hemogenic capacity of young endothelium. Similar strands appear also in the jugular vein. Emmel (1. c.) saw similar strands in the proximal portion of the left umbilical artery, and in the aorta of this level, in two 12 mm. pig embryos, and suggests that they may be related to the fusion of the two original dorsal aortae. In the case of the development of the inferior vena cava, the coincident fusion between the originally separate post- and sub-cardinal veins in 376 AMERICAN ASSOCIATION OF ANATOMISTS volves the formation of young, less differentiated, endothelium and so offers a favorable site for hemoblast production by endothelium. The encapsulated clusters present in this same region of the inferior vena cava may be explained as follows: Subjacent to such clusters the mesenchyme appears to be differentiating into hemoblasts ; this observation may give the clue to the correct interpretation of these clusters. If the invaginating area of endothelium included a considerable portion of such differentiating (vascularising) mesenchyme, then the peripheral cells might possibly be so far outstripped in the expression of their hemogenic potentiality as to be forced, perhaps principally by reason of internal pressure from the differentiating and proliferating cells, to continue development along the line already begun, namely into definitive endothelium. Emmel (1. c.) interprets the endothelial and mesothelial desquamation products, both cells and clusters, in terms of the stimulative effect of a pathologic factor upon the endothelium ; a toxin whose source is in the degenerating cells of atrophying redundant ventral aortic rami, and in degenerating erythrocytes in the serous cavities in the case of the mesothelia. That atrophying vascular stems are present at this stage, both in relation to the aorta, and the inferior vena cava, cannot be disputed. In the 7 mm. mongoose embryo solid regressive ventral aortic stems are especially conspicuous. At least a portion of the caudal shifting of the three large aortic rami is due to a progressive atrophy of upper portions of a connecting net of vessels. But coincident with this phase of a regressive development among the upper roots, there may possibly be a new formation of lower roots. I incline to see the cause of cluster formation in the latter possibility rather than in the former fact. Great stress is laid by Emmel upon the structure of the atrophying rami. Some of these are occluded by intravascular collections of hemoblast-like cells, both in the 10 mm. pig embryo and in the 7 mm. mongoose embryo. With these intra-arterial cell masses some of the aortic clusters are intimately related. Emmel ascribes the presence of this intra-arterial mass to the stimulative action of a dilute toxin, presumably liberated by the regressive aortic branches. This explanation is suggested by an alleged comparable pathologic process where endotheHum is believed by certain pathologists (e.g., Mallory) to be stimulated to the formation of 'endothelial leukocytes' ('large mononuclear leukocytes') by dilute toxins such as are produced by typhoid and tubercle bacilli. A more likely interpretation, it seems to me, would attribute the presence of the intra-arterial cells mass of the smaller rami to the relatively slightly differentiated character of the endothelium. The occlusion of the rami and the degeneration (karyorrhexis) of the cells would thus be a secondary effect of the constriction of the regressive atrophying vessels. In other words, the intra-arterial cell mass is not the result of the action of a toxin; l)Ut the occlusion and degeneration (and the possible formation of a 'toxic substance') are all PROCEEDINGS 377 tlic ivlati'd coHimoM sc(iuahio of the .shrinkiiifr of the atropliying vessel uroiiiul a previously present, normally produced, mass of hemoblasts. It may he emphasized that as regards the endothelial origin and the composition of the aortic cell clusters, and as rc^gards the mesothcHal ori«;in of cellular elements of the serous fluids, lOnnnel and I arc in essential agreement. But l']mmel views these structures as the r(>sult of the presence of a stunuilatinp; toxin; I see in them only the expression of a normal inherent capacity of embryonic endothelium to produce blood cells. The explanation of the limited distribution of the clusters is to be found in a relationship to young or newly formed, only tlightly differentiated, endothelium, rather than in a connection, with regressive blood vessels and an associated toxic substance. All the facts seem to fit better the hypothesis that the hemogenic activity of embryonic endothelium is a normal function at a certain stage of emlnyonic development, than that the causative stimulus is a 'toxin derived from degenerating vascular tissues." S6. The formation of the anterior hjmph hearts and neighboring ly7nph channels in biifo. Otto F. Kampmeier, Anatomical Laboratories, School of Medicine, University of Pittsburgh. Because a detailed narrative of the origin and development of the lymph hearts, ducts and sinuses in anuran Amphibians will be published in a comprehensive treatise during the coming year, at this time only the mam steps in the formation of the anterior lymph heart and of that part of the lymphatic channel sj'stem situated in its immediate vicuuty will be presented. This can be most clearly demonstrated by a comparison of six or seven reconstructions of that region in consecutive genetic stages. For the purpose of orientation, it should be remarked that the anterior lymph heart during development occupies a position, at the level between third and fourth spinal ganglia, in the triangular territory bounded medially by myotome, dorso-laterally by epidermis, and ventrally the pronephros and its surrounding venous smus. Beginning in 3 mm. toad embryos (European common toad), a series of venous tributaries, following one another in metameric sequence, are formed in connection with the pronephric venous sinus and the postcardinal vein. They are the intersegmental veins located at the intervals between- consecutive muscle segments and confluent ventrally with the cardinal venous trunk, the first three joining the pronephric sinus and the remaining ones the post-cardinal. In 4 mm. embryos we have the first intimation of the future lymph heart. Associated with the third intersegmental vein near its junction with the pronephric sinus, a simple venous plexus appears, as yet crude and indefinite. In the next stage, a 5 mm. embryo, this plexus, the anlage of the lymph heart, has become better defined, and not only does it include the prox-'mal segrnent of the third intersegmental vein, but it is also united by short channels with the preceding and succeeding intersegmentals, the second and the fourth. . Gradually the lymph heart plexus becomes sharp 378 AMERICAN ASSOCIATION OF ANATOMISTS ly circumscribed, its meshwork disappears by the expansion and fusion of its channel components, and at the time when the 6 mm. stage is reached, the tymph heart anlage is practically an ininterrupted chamber whose outlmes already suggest its definite condition. Coincident with the genetic changes of the lymph heart anlage, a more or less intricate network is established between the anterior intersegmental veins, that is from the first to the fourth, so that the criginal metameric condition becomes obscured and finally lost. This plexus, which we may call a veno-lymphatic one in view of its later character, progressively (in 7 and 8 mm. embryos) loses all continuity with the pronephric venous sinus and unites anteriorly with the great cephahc lymph smus and so becomes intercalated in the lymph system; it thus loses its venous function to assume that of a lymphatic, carrying the lymph stream from the head to the lymph heart. But the original connections of this transformed venous or lymphatic plexus with the heart do not persist to become the permanent oices; they break away from the heart at the same time when the connections with the pronephric venous sinus disappear, so that temporarily the heart is completely isolated except for its confluence on its ventral side with the pronephric sinus by means of a short venous stalk, the beginning of the anterior vertebral vein. During subsequent stages of development, in 10 and 11 mm. toad embryos, the lymph heart and the dorsal longitudinal channel of the lymphtic plexus again are brought close together, evidently by the expansion of both structures as well as by the relative reduction in extent of the region between the dorsal surface of the tadpole and the phronephros. The approximation between heart and respective segment of the lymph duct becomes more intimate, as is indicated by the fact that the duct comes to lie in a grove-like depression in the dorsal wall of the heart. It is along this line of contact that secondarily an opening or tap is now established, simultaneously with a simple valve, between duct and heart. In the meantime a similar valve has appeared at the lymphatico-venous junction, the connection of lymph heart and anlage of the anterior vertebral vein. The presence of these valves and the pulsations of the heart, which begin approximately at this time, consequentlj^ determine the direction of the lymph flow from duct to heart and thence to vein. Other and later developmental changes as well as a discussion of the homology between the process of lymph heart formation in Amphibia and that in other vertebrates will be considered in the larger paper. PROCEEDINGS 37!) S7. A com paratirc stud !i of the roof of the fomih ventricle. .)..). Keeoan, (Iiitroclueed by C\ \\ . M. l^)yMtc•r,) University of Nebraska Medical Cllef2;e, Omalia.

At the lueetiii}:: of tlie association last year Weed reported a series of injections into the brain cavities of living pig embryos. Independently at that time I iiad made a similar series of injections into the brain ct'vities of living rabbit embryos. The technique employed was exposure of the embryo l)y a small incision through the uterine wall of the anesthetized rai)l)it and injection of a suspension, solution or dye into the lateral ventricles l)y means of the finest j^ossible glass capillary tube and bulb pressmv. The quantity injected Avas small and was controlled by an estimate of the rate of flow imder a constant pressure. The most useful injection substances were the double solution of 1 per cent ferric anuuonium citrate and potassiun ferro-cyanide fixed in 10 per cent formol and 1 per cent hydrochloric acid for the precipitation of the Prussian blue, as recommended by Weed, and the 1 per cent ammonium citrate solution alone, fixed in the acid formaldehyde with the addition of potassium ferro-cyanide which caused a precipitation of the Prussian l)lue at the site of the citrate solution. The living condition of the embryo was observed in each case by the continuance of the heart beat at the time of inunersion in the fixing fluid.

The anatomical findings from these injections, combined with a study of a large number of sectioned pig and sheep embryos and several human embrj'os, corroborated very closely Weed's results, showing a very early development (pig embryo 6 mm.) of a thin pavement-cell oval area in the roof of the fourth ventricle, which later becomes modified b^^ the development of the transverse fold of the choroid plexus and extends to the area posterior to the plexus.

This area undoubtedly plays an important part in the early escape of the cerebrospinal fluid, as disclosed by the denser collection of protein coagula within the ventricle in contact with this area, its early intimate ec odermal and vascular relation, its later relation to the first region of dilatation of the meningeal spaces and, in all early embryos injected with the double solution, by the collection of the precipitated Prussian blue granules as a dense mass in contact with the inner surface of the membrane.

Injection with the double solution in rabbit embryos up to the age of seventeen days showed no extra ventricular spread of the fluid, which was rather surprising in view of the fact that the choroid plexus is quite well developed at this age. Even in the earlier stages before the development of the plexus there is present a rather typical cuboidalcell layer of ependyma surrounding the membranous area, indicating a possible slight secretion at this age.

For comparison with the double solution injections, which appeared to act in the manner of suspensions or colloids, a series of injections of the ferric ammonium citrate solution alone was made with strikingly dissimilar results. In the stages before the development of the choroid plexus there was a diffusion of the solution into the loose mesenchymal

THE ANATOMICAL RECORD, VOL. U, NO. 6

380 AMERICAN ASSOCIATION OF ANATOMISTS

tissue about the rhombencephalon and only a faint indication of a condensation of the blue granules in contact with the inner surface of the membranous area. After the development of the choroid plexus there was a very rapid absorption into the circulation.

A similar series of injections was made into the midbrain cavity of chick embryos from the four day to the nine day stage, during which time the choroid plexus develops (sixth to seventh day). Similar to the rabbit embryos, in no case did the double solution reach the exterior of the brain cavities or enter the circulation. This was determined by removal and fixation of the entire embryo with its membranes. The same manner of condensation of blue granules in contact with a central membranous area of the roof of the fourth ventricle was noted. In the four day stage this was a rather small oval area in the center of the widely expanded tela of the fourth ventricle. In the seven day stage the tela is constricted at its center by the development of the internal ear and the collection of blue granules is most dense at a small central spot in the anterior half, with a more diffuse collection in contact with the posterior half. On section these areas appear as typical pavement-cell membranous areas similar to the mammalian embryos.

The citrate injections in the chick always diffused into the mesenchjTiial tissue m the region of the roof of the fourth ventricle and after the development of the choroid plexus (sixth to seventh day) very rapidly entered the circulation. At the seven day stage a considerable portion remained within the ventricle and as a very evident blue coloration in the mesenchymal tissue over the rhombencephalon and mesencephalon. In contrast with this, the double solution injection, remaining an hour in the living embryo, showed no escape from the ventricles.

In the eight and nine day chick the escape of the citrate solution was even more rapid, practically all leaving the ventricles within ten to twenty minutes. Embryos of the same age injected with the double solution and remainmg aUve for an hour showed no escape from the ventricle.

As this is only a preliminary report of a feature of a more extensive comparative study of the development of the tela choroidea and cranial meningeal spaces in various vertebrate types, final conclusions cannot be given, but the inferences from the material examined are that this membranous area of the roof of the fourth is non-permeable to the double solution in the early embryo stages while it is permeable to the citrate solution; that a slight escape of the cerebrospinal fluid occurs before the development of the choroid plexus; and that the collection of protein coagula and Prussian blue granules in contact with the inner surface of the membranous area represents a dialysis phenomenon of this semi-permeable membrane.

PROCEEDINGS 381

3S. The stnidiirc of (In- skull of Ziphius. .loii.v D. Kkhnan, Jr., Aiui tomical Laboratory of Columbia University.

For convonieuec tlic skull of Zii)hi\is is usually described as consisting of the rostrum and the cranium i)roper, the line of division being drawn tangent to the rostral borders of the maxillarj tuberosities. It is well to note, iiowevcr, that structurally any such division is arbitrary. The rostriun exjiands at its base both in the vertical and transverse diameters. The transverse expansion, effected by a broadening of the maxillae, reaches its extreme in the region of the maxillary tuberosities. So their rostral borders indeed furnish a valid line of demarcation l)etwen rostrum and cranium in the transverse plane. On the other hand, the vertical expansion, formed in the dorsal aspect of the skull bj^ the premaxillae, on the ventral aspect by the broad pterygoids, extends nuich further caudad and reaches the occipital. Thus it is difficult to settle on a point where rostrum ends and cranium begins.

The rostnnn is long and exceedingly^ massive in its structure. It is of interest to consider the mutual relations of the rostrum and cranium. The former is in reality merely the apical portion of a pyramidal mass which rests its base agamst a ring formed by the supraoccipitals, parietals and the basisphenoid. The junction of the occipital ring and rostral base results in the formation of the great transverse crest of the cranium. From the crest, both caudad and rostrad, the bone falls away toward the foramen magnum on the one hand and rostrum on the other, so establishing the caudal and ventral walls of the cranium. It is the wide expansion of the rostrum at its base which results, in spite of the great length of the structure, in the firmest security against all manner of strains.

In the vertical plane forces are transmitted from rostrum to the occipital ring chieflj^ through the massive frontal portions of the premaxillae. It must be noted that the chief strength is not gained through direct contact, for only the right premaxilla reaches the crest of the frontal, and that by a narrow process. Between premaxillae and the frontals are interposed the massive nasal bones, and it is through them that thrusts must be in the main transmitted. The surfaces of contact between nasals and premaxillae are broad and the articulations are of great strength.

In union with the premaxillae, the maxillae turn dorsad and reach the transverse crest. This portion of the maxilla, however, is merely a thin sheet of bone and can contribute no great strength in this direction.

On the ventral aspect of the skull the lines of the transmission of force pass through the pterygoids. These bones have a broad articulation with the base of the rostrum, chiefly with the palates which are interposed between them and the maxillae. There is also a small direct articulation with the maxilla and a very sUght contact with the vomer. The union then is one of great strength. Caudally the pterygoids reach the bases of the basioccipital processes of the occipital region.

382 AMERICAN ASSOCIATION OF ANATOMISTS

The ventral brace thus formed is much less firm than the dorsal, since the pterygoids are less massive than the premaxillae, and the structure of the bone is less dense. Moreover, dorsal strains are transmitted through a continuous line of bone, whereas the ventral strains must pass through three articulations. We may conclude then that strains in this direction are not so severe as those in the opposite. However, in the adult skull provision for greater firmness is made by the expansion laterad of the borders of the pterygoids which thus secure a very firm hold on the base of the skull.

Lateral strains are provided against by the expansion of the base of the rostrum in the transverse plane. This is effected, as already stated, by the spreading out of the maxillae. Through them forces reach the occipital ring along two lines. One is an inner, along the lateral cranial wall, formed chiefly by the parietals, which articulate directly with the exoccipitals. There is in addition an outer line of transmission through the postorbital processes of the frontals to the jugal processes of the squamosals, and hence to the exoccipitals. That the lines of greatest strain lie through the lateral margins of the bones is at once evident from their formation. Mesally both the maxillae and the frontals are thin sheets which lie in contact and form an articulation of great strength preventing displacements indeed, but offering no great resistance to compression. Their lateral margins on the other hand are thicker. The frontals form the massive orbital processes. These thrust forward strong preorbital processes which are locked to the maxillae by the lachrymals. Lighter postorbital processes reach the jugal processes of the squamosals which interlock with the exoccipitals.

Additional security is given to the rostrum by the character of the articulations of the bones forming in with those of the cranium. On the dorsal aspect the chief of these is that between maxilla and frontal. Both of these bones expand into broad sheets of bone, deeply concave rostrally, the maxillae fitting into the concavity of the frontal. The premaxillae embrace over the mesal margins of the maxillae and secure this border to the nasal and frontal. On the ventral aspect the pterygoid has a broad contact with the base of the rostrum and spreads over the base of the cranium to the basioccipital. Two characteristics then distinguish these articulations, their breadth and a certain amount of interlocking.

To sum up, the relation of the rostrum to the cranium proper, depends for its security on the expansion of its base, and the broad interlocking character of its articulations with the cranium. The central feature of the cranium proper is a great transverse crest, the core of which is formed by the frontals; against it in front rests the base of the rostrum. Its caudal face supports thrusts from the condyles through the supra- and exoccipitals. These bones are reinforced by certain thicknesses which may here be noted as follows: A dense ring of bone above the foramen magnum; a thick crest seen on the inner surface of the skull which passes in the midline from the foramen magnum to the

PROCEEDINGS 383

tnuisviTsr crest of the slaill; two heavy ridges from the caudal ends of tins crest ruiininfr in a latero-rostral direction to tlu; lateral terminations of the transverse crest. Tims the central ridf^e antl the two lateral ridges make together three re-inforcing Hues of bone, which transmit the thnists of the vertebral column to the great transverse crest of the crannnn.

39. The laws of bone architecture. John C. Koch, (introduced by Dr. \\ arren H. Lewis,) Department of Anatomy, Johns Hopkins Medical School.

Wolff's law of the functional form of bone and the functional pathogenesis of defornuty has had as its sole mathematical basis, the analogy first noted by the mathematician, Culmann, between the position of the trabeculae seen m frontal, longitudinal sections through the head and neck of the human fennir, and the paths of the maximum tensile and compressive stresses computed for the Fairbairn crane carrying an over-hanging load somewhat as does the upper femur. The Fairbairn crane analyzed l)y Culmann w^as assumed to be solid, with a circular cross section and an outline roughly corresponding to that of the human femur with both trochanters removed so as to present a smooth, curved form. The paths of the maximum internal stresses computed in this solid crane (a body entirely different from the femur in shape and in the disposition of the material) were assumed to explain the position of the trabeculae in the upper femur, and to be conclusive proof that the trabeculae in bone are laid down in exact accordance with mathematical laws. Although suggestive, such analogy has never been accepted as conclusive mathematical proof of the law formulated by Wolff.

Wolff's law, though originally based upon faulty evidence from a mathematical point of view, is proved for the normal, human femur

J* u^^^^ mechanical analysis of the inner structure of the femurand other important relations between the structure and the stresses due to the preponderant load on the femur-head, are shown by the study presented in this paper.

That human bone obeys the laws governing elastic bodies when carrying a load has been demonstrated by numerous investigators, iheretore the laws of mechanics applicable to elastic materials also Hold lor bone. After a preliminary study of some thirty femurs, the writer undertook the analysis of the mechanics of the femurs from a do-year-old subject, who w^eighed 200 pounds and was in good health at the time of his death by accident.

The object of this work was the exact mechanical analysis of the structure of the normal, whole femur and the determmation of the relations between structure and function at every point. The femur was studied m detail m much the same manner as a structural element ot a machine or a member of a truss, where the cross section varies greatly from point to point.

384 AMEKICAN ASSOCIATION OF ANATOMISTS '

The left femur was used for the purpose of studying the bone in longitudinal section, the right femur was used for the analysis of the transverse sections which were made at intervals of one-fourth inch measured along the axis of the bone. The shape of these transverse sections was extremely variable and conformed to no geometrical forms, so that the formulas of calculus could not be applied directly to the mechanical analysis of these sections. Accordingly each transverse section was covered by a series of squares formed by two series of parallel lines at right angles to each other, drawn at intervals of onetwentieth of an inch from the center of gravity of the section. In this manner the area of the compact and the spongy bone was accurately found by the summation of the squares covering these areas : the integration of the product of the area of each strip one-twentieth of an inch wide by its distance from the neutral plane gave an accurate value of the statical moment of the section: the integration of the product of each strip one-twentieth of an inch wide by the square of its distance from the neutral plane yielded an accurate value of the moment of inertia of the section analyzed. These integrations are practical applications of calculus and are absolutely necessary for a mathematical anaylsis of any structure subjected to bending stresses similar to those in the femur. These calculations were made in detail for the femur at intervals of h inch in the upper and lower extremities and at intervals of one inch in the shaft. The total number of calculations required in the preliminary studies was about 50,000, and, in the final analysis of the normal femurs studied in this paper, an additional 70,000 calculations were necessary for the analysis of the inner architecture and the computation of the maximum stresses produced in the femur by the load on the femur-head.

An assumed load of 100 pounds was considered as acting on the femur-head in the same direction as the weight of the body, under normal conditions, and the axial load, vertical shearing force and the bending moments were computed for each section in accordance with the principles of mechanics. The amounts of the stresses were then computed at the sections of the femur which had been previously analyzed. As the intensity of the stresses varies directly with the load assumed, the stresses due to any load can be determined by simple proportion once the stresses have been found for a given load._

PROCEEDINGS 385

The analysis of strength as outlined agrees with the actual breakingstrengths of femurs made by Messerer for loads on the femur-head and also for cross-breaking loads. Statistics of the location of fractures in several independent series of fractures of the femur agree closely with the proportionate distribution of fractures according to the laws of probability, as applied by the writer to the femur.

The writer believes that the evidence ' presented warrants the folio wing conclusions:

1. The normal external form and internal architecture of the human femur results from an adaptation of form to function.

2. The proportions of the femur are everj-where such as to show a definite mathematical relationship between the body weight and the internal structure of the bone: there is a definite relation between the structure and the stresses at everj^ section.

3. Spongy bone is homogeneous with compact bone as a structural material and differs from it mechanically only in possessing smaller strength approximately in proportion to its relative density as compared with compact bone.

4. The femur has a factor of safety of 5.68 for the stresses due to running, 11.36 for walking and 30.30 for standing.

5. The structure of the femur is based upon the mathematical requirements of mechanics and the inner architecture is such as to produce great strength with a relatively small amount of material; the material is arranged to correspond with the stress requirements existing at every section.

6. The adaptation of form to function, proved mathematically for the normal, human femur by exact mechanical analysis, is the general law of normal bone.

386 AMERICAN ASSOCIATION OF ANATOMISTS

7 . The thicloiess and closeness of spacing of the trabeculae in bone varies directly with the intensity of the stresses transmitted by them.

Jfi. A comparison of the Herzog and Strahl-Beneke embryos. (Lantern.)

Frereric T. Lewis, Harvard Medical School.

The Herzog and Strahl-Beneke embryos are of particular value to embrj'-ologists since they are the youngest human embryos of which satisfactory drawings of serial sections have heretofore been published (22 sections of the Herzog embryo and 60 of the Strahl-Beneke specimen). Of the Peters embryo, apparently only a single section is thus available, which has been reproduced everywhere; and although 30 sections of the Bryce-Teacher specimen have been published, they are fragments defying interpretation, at least apart from the specimens themselves. Bryce and Teacher estimate that their embryo is three days younger, and Peters two days younger, than the Beneke specimen; but that these figures are merely approximations is indicated bj' the fact that Bryce and Teacher consider Peters embryo to be thirteen and one-half to fourteen and one-half days old on p. 53 of their work, but fourteen to fifteen days on p. 59. Although the difference in age of the four specimens under discussion is perhaps not greater than three days, there is the widest difference in the amount of information available about their structure. The Strahl-Beneke and Herzog embryos are the ones to which at present we must resort for an approximately adequate description.

Herzog's embryo is not now in a good state of preservation and its reconstruction is in part a matter of restoration. Thus the detached tube in the body-stalk, which was originall}^ interpreted as an allantois, is clearly an amniotic duct. The place where it was torn off from the amnion can be definitely located. (This fundamental correction, reported at the 1913 meeting of this association, involves a change in labelHng rather than any other modification of the model figured by the writer in Keibel and Mall's Embryology, vol. 2.) The StrahlBeneke embryo is in far better condition, and helps to explain the Herzog specimen, whereas the latter is the most interesting commentary on Strahl and Beneke's important publication. This was appreciated by Strahl and Beneke who expressed regret that Herzog's paper was received too late for them to make a detailed comparison between his sections and their own, and they merely note that his sections differ in many respects very strikingly from theirs. We can now supply such a comparison and find that the embryos are so nearly alike that they may be said to establish the essential features of a certain early stage in human development. The conditions in younger human embryos are still subjects for diagram and conjecture.

41. Further observations on the longitudinal muscle of the pig's colon. P.

E. LiNEBACK, Atlanta Medical College, Emory University.

The longitudinal muscle of the pig's colon develops from an accumulation of myoblasts, which are earliest seen in a 39 nun. embryo, at the

PROCEEDINGS 387

lucscnicric arc, and rapidly spread around the circumferenee of the l)o\vel. The niesenterie arc also first shows fully formed muscle fibers which are j-rouped into a crescentic form, and from the tips of the crescent jj;rowth exKMids laterally till the whole circU; becomes chanjrcd into a layer ot nnisck. fiJKTs; completed at IK) mm. All the while the crescentic thickeninj^ is the most consi)icu()Us i)art of the muscle a I'ondition m harmony with what F. T. Lewis finds in the liuman emoryomc colon.

Up to the IK) mm. stage no accumulations of fibers appear which can be inteipreted as taeniae. After this time two thickcninRs develop m the muscle, one at either side of the intestine, equidistant from the mesentery, which receive the greater number of mesenteric vessels and become the permanent taeniae; the only bands which tlie pig's colon possesses. The mesenteric arc never becomes a taenia except in the caecum where it extends from the tip of this pouch to the ileum at tJie line ot attachment of the ileocaecal membrane. '

1^2 Observations on the effect of Madder-feeding, especially with regard to deposits of calcium-salts. C. C. Macklin, Johns Hopkins Medical V;^?"o, Baltimore, and Wistar Institute of Anatomy and Biology Philadelphia. -^ '

It has recently been shown that the physical and chemical processes underlying the formation of normal bone and of pathological calcific deposits are similar, if not identical. Furthermore, it has been known tor niany years that the dyestufT, madder, when fed to animals, will specifically stain the calcium-salts of bone which are being laid down during the time that the coloring matter is present in the circulating l)lood, and that this is true for the callus formed in the repair of injuries as well as for the bone of normal development. It seemed reasonable, therefore, to suppose that the same staining reaction toward the dyes of madder would be evidenced by calcaerous tissue of a pathological character as is shown by developing bone. Accordingly, madder was fed to animals which possessed calcium-salt concrements of dilTerent kinds, and the results were studied. In general it may be said that .T r"^^'^^^ o^ pathological calcium-salt deposits toward the dyestutts of madder was identical with that of bone.

The work was carried on principally with the white rat. Experimental calcification of the kidney was initiated in the well known way by unilateral ligation of the renal vessels. After a short time the kidney, which is revascularized through the adhesions of the surrounding peritoneum, becomes the seat of a process of calcification, so tliat upon cutting through the organ numerous hard nodules of calcium salts are found m its interior. During a part of the time when this deposit of calcmm-salts was forming madder was fed, with the result that the deposit was stained. Not all of the calcium-salt aggregation was colored, however, the nodules being composed of both white and red granu ar masses. These latter were probably laid down during the time that madder was being fed to the animal. Thus the picture

388 AMERICAN ASSOCIATION OF ANATOMISTS

presented b}'- such a stained concretion recalls that of madder-stained groo'ing bone.

]\Iadder was fed to a young rat which presented a double cataract, udth the result that a distinct pinkness of the eyes was noticed in three days. The animal was killed after six days' feeding, and the lenses cleared in oil of mntergreen. It was then seen that the red staining was at the periphery. This was also evident in the sections. Most of the calcific deposit was unstained, the red areas representing the part of the cataract formed in the six daj's during which madder was being fed. Normal lenses do not stain with madder, even after five months' continuous feeding.

From these results it is believed that all pathological calcific deposits in process of formation will stain with madder.

Smce calcium-alizarate, an insoluble precipitate of alizarin, has been shown to be responsible for most of the staining of bone in the madder-fed animal, it may also be regarded as the principal staining agent in the case of the pathological concretions.

This identical staining reaction towards madder-dyes on the part of both developing bone and developing pathological deposits of calcium would seem to strengthen the conception that the processes involved in their formation are largely identical.

It was noted that the dyes of madder pass easity through epithelia of different characters. The observation that the bones of the fetus are stained in utero by feeding madder to the mother was confirmed, and it was also noted that the milk of the madder-fed mother was colored pink, and readily stained the bones of the young suckling animal. Furthermore, madder is freely excreted by the kidney. The urine is yellow or orange when passed, but this color is turned to a red upon the addition of an alkali, thus indicating the presence of alizarin. Madder -dyes may be absorbed by serous membranes and excreted b}^ them. Proof of the former statement is afforded by the staining of the bones in the usual way following the introduction of a sterile madder-decoction into the peritoneal cavity. The latter assertion is illustrated by the pink color shown by the hydrocele fluid of an animal fed with madder.

In animals fed for three months continuously with madder the cartilage was permanently stained red. Transient staining of the lining of the cardiac end of the stomach was noted even after only a few days' feeding. This coloration was found in the superficial layers of the keratin-hke material which covers the squamous epithelium in this region. It is due to direct contact with the dyestuff.

Madder-feeding was without apparent effect on growth and reproduction.

43. Influence of heat on the eggs of Cumingia. Margaret Morris, Osborn Zoological Laboratory, Yale University. (By invitation.) The following report is a continuation of the cytological study of

the influence of heat on the eggs of Cumingia. In the report made last

PROCEEDINGS 389

year, it was shown that if tho cp;p;s of this inolhisc are siibjoctcd to certain t(Mn])t«ratur('s ihcy \iii(l('rp;o a sort of solf-fcrtiHzatioii. 'J'ho first polar imdi'tis is sc])arat('il from tho o^'^ inicloiis, but is retained in the cytoplasm of the e^ji, and the two nuclei fuse to form a cleavage nucleus. In the ensuinjj cleavajj;e the form and munher of the chromosomes is al)normal. Instead of 30 long threads such as are present in tho normal egg, there are 50 or GO short rods. It was supposed when the retention of tho polar body was first observed that this process would result in tho restitution of the normal chromosome number, but though the normal amoimt of chromatin is undoubtedly present, the nmnber of chromosomes is nuich larger than the diploid number.

It has boon found in more recent experiments, that if the eggs are fertilized and subjected to heat immediately after fertilization, the chromosomes of the first polar spindle behave as they do when the unfertilized egg is heated. They divide, as in the normal egg, and go to tho poles of tho spindle, where they form a number of small vesicles. These fuse gradually and two large nuclei are formed. This takes place without an}^ c}i:oplasmic cleavage, so that here, as in the parthenogenetic egg, the egg nucleus and the polar nucleus are both present. In the meantime, however, the male pronucleus has undergone perfectly normal development, so that the egg contains three nuclei instead of two. Each of these nuclei contains the haploid number of chromosomes: they come to lie close together and fuse in most cases to form a single cleavage nucleus. In some instances the male and female pronuclei do not fuse before the cleavage spindle begins to form, but there is never any discarding of nuclear material. The first cleavage spindle, then, should have a plate of 54 chromosomes or three times the haploid number. Here, as in the case of the parthenogenetic eggs, the theoretical expectation is disappointed. The first cleavage spindles have about 60 chromosomes — one plate that has been counted shows 62, another 58. Although the number is about the same as that found in the parthenogenetic eggs and the oultines of the chromosomes are similar, a great difference in size is noticeable, for the chromosomes of the fertilized egg are much larger than those of the parthenogenetic one. Evidently a larger amount of chromatin is present in the fertilized egg divided into about the same number of bodies.

It was thought that perhaps the abnormal shape and number of the chromosomes in these eggs was an effect of heat which would be reproduced in normally fertilized eggs which had formed polar bodies. If this effect is one of heat only, such eggs should show equatorial plates like those of the parthenogenetic (self-fertilized) eggs. Experiments were made in which the eggs were fertilized and allowed to form polar bodies, and heated before the first cleavage had taken place. The cleavage spindles in the eggs treated in this way show a condition entirely different from that of the parthenogenetic egg. The chromatin is present in the form of large irregular masses which give the spindle an appearance similar to that of the first maturation spindle of the normal egg. The presence of the two polar bodies shows, however, that we have here an abnormal cleavage spindle.

390 AMERICAN ASSOCIATION OF ANATOMISTS

The amount of chromatin present in the egg is an important factor, for it has been shown that eggs having less than the diploid number of chromosomes are incapable of development. The addition of extra chromatin, on the other hand, is not injurious to the egg, for those that have three times the haploid number develop fairly regularly. The number of chromosomes present in cleavage varies under experimental conditions. It is interesting to note that the mean about which this number varies is not the diploid number, but is nevertheless the same in eggs which have the diploid and in those which have a triploid amount of chromatin.

44. Studies on the mammanj gland. J. A. IVIyers, Institute of Anatomy, University of Minnesota.

1. The fetal development of the mammary gland in the female albino rat

Henneberg ('00) made a careful study of the development of the mammary glands in the albino rat from the earliest appearance of the glands through the conditions found in 16 day fetuses. Also the postnatal (birth to 10 weeks) development of these glands has been investigated (Myers '16). Heretofore the developmental conditions between 16 day fetuses and newborn rats have presented a gap in our knowledge of the mammary gland. The object of the present investigation is to fill up this gap, thus completing the history of the mammary glands in the albino rat to ten weeks of age.

In 17 day and 2 hour fetuses slight mammary pits appear on the surface of the skin over the mammary gland areas forming the typical 6 pairs. The primary ducts are sohd epithelial invaginations measuring only about 0.05 mm. in length. Secondary ducts have not yet appeared.

Wax reconstructions and serial sections show very definite mammary pits in 18 day 9 hour fetuses. The primary ducts remain unbranched except in the abdominal glands where short secondary ducts appear. A small end-bud is present at the terminal end of each duct.

At 20 days, 6 hours the mammary pits are still present but in the deepest part of each pit is a small eminence surrounded by a shallow furrow. The eminence corresponds to the future nipple. Deep to the shallow furrow there is a short ingrowth of epithelium which forms the epithelial hood described m postnatal stages. Secondary ducts are present in all of the glands. In the abdominal and first inguinal glands tertiary and terminal ducts were observed. The terminal ducts present well developed end buds. In many of the terminal ducts lumina are quite well formed. The lumina begin at the distal end of the tertiary ducts and terminate about 30--40 n from the distal end of the terminal ducts.

PROCEEDINGS 391

f. .1 coinixirisoii of the tumumnnj ijbmih in male and female albino ruts

111 llic jxist natal stages observed inainmary gland nipples have not api)earetl in the male rat, although as previously shown they are very consi)ic\i(nis in the female.

Male fetuses of 18 days 9 hoiu's present no mannnary pits. Instead there is eorresjKJnding to each gland a slight eminence covered with thickened e])itheli\nn. The ducts are similar in size and form to those of the female fetuses of the same age.

At 20 days and 6 hoin-s there is neither pit nor eminence over the mammary gland areas but the ducts come directly to the surface, the epithelium of which is slightly thickened. The epithelial hood found in female fetuses of this stage is absent. The ducts of the male have branches and lumina corresponding with those of the female.

45. The history of the eye muscles. (Lantern.) H. V. Neal, Tufts College.

The attempt is made in this paper to demonstrate on the basis of embryological evidence the exact homology of the first three permanent myotomes of Amphioxus, Petromyzon, and Squalus and to describe the more important stages in the phylogenesis of the eye muscles.

The evidence is presented for the first time to support the assertion of Dohrn ('04) and the writer ('07) that the second as well as the third myotome participates in the formation of the external rectus muscle. In the light of the evidence given the familiar text-book formula for the ontogenesis of the eye muscles should be amended as follows:

From the first myotome (pre-mandibular head-cavity) arise the muscles innervated by the oculomotor, viz., the Mm. recti superior, internus, and inferior, and the M. obliquus inferior;

From the second myotome (mandibular head-cavity) develop the

M. obliquus superior and the ventro-lateral portion of the M. rectus externus ;

From the*third myotome (hyoid head-cavity) arises the dorso-median portion of the M. rectus externus.

46. The early morphogenesis of the thyroid gland in Squalus acanthias. E. H. NoRRis, (Introduced by C. M. Jackson) Institute of Anatom}^, University of Minnesota.

The anlage of the thyroid gland in Squalus acanthias makes its appearance in embryos of- approximately 4.0 mm. in length, as a solid epithelial bud from the floor of the posterior pharynx. This bud, which is at first little more than a thickening of the entodermal lining of the pharynx, is placed just ventral to the point at which the oesophagus leads from the pharynx, and in the region inferior and caudal to the ventral extremities of the first two gill pouches. The bud increases in size and extends caudally. It assumes a pedunculated form, being suspended from the floor of the pharynx by a rather extensive but very narrow neck. By the time the embryo has attained a length of 19 mm.

392 AMERICAN ASSOCIATION OF ANATOMISTS

the gland has severed its connection with the pharynx, and has the form of a cohimn with rounded ends and Avhose cross-section is ovoidal. At 28 mm. the gland has once more changed its form, appearing at this time as a relatively thin diamond shaped plate. At 36 mm. the gland has increased remarkably in length and is divisible into an anterior, diamond shaped corpus and a posterior, elongated, narrow cauda. Durmg the previous stages the gland has presented surfaces which have been quite smooth, but at 48 mm. the surfaces begin to lose their smoothness and by 100 mm., which approximately marks the end of the prefoUicular period, the surfaces have been greatly altered by the development of great numbers of small ridges and furrows.

While these changes have been altering the external gross form of the gland, various other changes have been in process withm the gland mass. Two of these deserve particular note. During the period immediately preceding the separation of the gland from the pharnyx a large amount of pigment has been deposited in the gland. Although this pigment is not confined to any particular part of the gland, by far the largest part of it is found in the narrow and constricted neck which suspends the gland from the pharyngeal floor. The pigment disappears after the gland has gained its mdependencefrom the pharynx. The second process is that which has to do with the development of completely closed cavities within the gland mass. It is important to note that these cavities are quite distinct from the follicular cavities which develop later. They finally open to the outside and are invaded by the vascular mesenchyme. These cavities are produced by the vital activity of the cells which rearrange and redistribute themselves about the forming spaces. By the appearance of these intraglandular spaces the gland is transformed into a number of irregular epithelial plates, two cells in thickness, in which the primary follicles develop.

These findings, as regards the development of the primary follicles from the two-celled epithelial plates, are of particular interest in the light of recent work upon the early development of the human thyroid.

47. On the position of the vitelline arteries in human embryos. (Lantern.) James W. Papez, Atlanta Medical College, and Frederic T. Lewis, Harvard Medical School. (Presented by Professor Papez.) This paper is a continuation of the study of the mesenterium commune reported at the last meeting of the association. The relation of the mesenteric artery to the rotation of the intestine can now be more fully considered. Broman, in 1914, declared that the proximal part of the definitive vitelline artery arises, as is well known, by the fusion of a pair of ventral aortic branches. Evans ('12) reviewed the evidence that instead of fusing, either the right or left members of an original pair may shift to the median line and become the persistent vessel, and he concluded that "the question is an open one." The formation of the very short median stem which by great elongation becomes the superior mesenteric arterj', requires an anastomosis or par

PROCEEDINGS 'A^'A

ti;il fiision Ix'Uvccii the i'ij;ht and left vessels, hul not necessarily a complete hision. Sections of the successive median roots of the mesenteric artery in a shiulo embryo seem to shoAv that sometimes the vessels fuse from the aorta outward, and sometimes one member of a pair enlarpies as its mate becomes obliterated. After the short median vitelline tnud-cs have been formed and have anastomosed antero-i)osteriorly to make the final superior mesenteric artery, that vessel increases rapidly in length and caliber. In an embny'o of 4.6 mm. its tmnk divides into several branches which encircle the intestine and re-imite as they proceed to the yolk-sac. The vessels crossing the intestine are remnants of the considerable row of vitelline arteries of earlier stages, and by persistence on one or the other side, they would cause the mesenteric artery to cross on the left or right of the ileum respective!}'. Usually the persistent vessel crosses the intestinal tube on the right side as seen in a 7.5 mm. embryo.

At 10 nun. the intestinal loop has dropped downward over the artery, which then appears below the ileum; and at 14.5 and 22.8 mm. the artery is seen on the left side of the small intesthie. Nevertheless it is morphologically on the right side, as is made apparent by undoing the effects of intestinal rotation". In all these stages it does not become free from the mesentery until it is about to cross the intestinal tube. (The same is true of the vitelline vein, but since the latter crosses the duodenum, it has a long course as a free vessel within the abdominal cavity).

In an embryo of 44.3 mm. the intestines are entirely within the abdomen. The vitelline arter}^ has now become free from the mesenterj' for a considerable distance before reaching the ileum. Thus it comes out from the left side of the mesentery as a strand which extends to the umbilicus. Owing to further rotation, the intestinal tube seems to encircle it completely; but again by undoing the intestinal rotation the artery is found to be morphologically on the right side. An unusual case in which the left half of the peri-intestinal arterial ring persisted, so that the vitelline artery crossed the intestine on the left, was found in an embryo of 42 mm. This shows conclusively that the artery may persist on either side of the intestine without modifying the direction of the intestinal rotation, and the same is true of the vitelline vein, which in an abnormal embryo modelled by Begg crossed the duodenum on the right. Usually the vein persists on the left side of the tube and the artery on the right side, but these are not essential factors in directing the intestinal rotation.

The vitelline artery, as found by Dexter in the cat ('00) anchors the ileum to the umbilical hernial sac so that the jejunum re-enters the abdomen first and the ileum last of all, stretching the artery to a filamentous strand. This order of return has been confirmed by Broman for the seal, and is probabh^ applicable to man. After the arterial strand ruptures (which usually takes place in embryos somewhat older than those studied, though the strand may exceptionally persist until birth, and rarely through adult Ufe) the place of its at

304 AMERICAN ASSOCIATION OF ANATOMISTS

tachment to the mesentery may be indicated by an appendage of the mesentery, which Broman finds invariably in seals and names the appendix meso-ilei. According to Fitz, who made an important clinical study of omphalo-mesenteric remains in man ('84), Riige described such appendages as frequent, but Fitz admits that he himself sought for them with only indifferent success. The models of the 42- and 44mm. embryos indicate the position, well on the side of the mesentery, where such appendages may be expected, and it seems quite probable that they have been frequently overlooked.

Jf8. The notochord of an East Indian scorpion. (Lantern.) William

Patten. Dartmouth College.

In a large scorpion, Heterometrus sp.?, recently obtained in Java there is an unusuallj^ well developed notochord which presents important new characters of general interest.

A. Embryology. The details of its development are too intricate to be considered here, owing to the presence in it of several different cellular constituents, and owing to its diverse local relations to the spinal cord, to the brain, to their neurilemma sheaths, and to the vascular system.

The main point in its development is that it arises, in close association with the medullary plate, as a median axial cord extending forward from the cephalic end of the 'primitive streak.'

In this important respect it agrees with the notochord of vertebrates and differs fundamentally from the so called notochord of Balanoglossus and of Cephalodiscus. This fact, when its significance is fully appreciated, should effectively^ dispose of some of the false ideas that have been entertained, since the earliest days of embryology, concerning the 'gastrula,' the 'blastophore,' and the 'archenteron' of vertebrates, and especially of the idea that the notochord is in some way derived from an outgrowth of the alimentary canal.

B. Structure in the adidt. General relations. The notochord of Heterometrus is a cylindrical tube, much larger than the spinal cord, extending more than one third the length of the body, that is, from near the root of the tail to the posterior part of the head, or cephalothorax. It lies on the haemal side of the nerve cord, between it and the alimentary canal. The notochord is firmly attached to the nerve cord, its peripheral cells being continuous with those that constitute the neurilemma, and its thick walls, which are relatively firm and elastic, serve to keep the more delicate nerve cord straight and to hold it in place.

Relations to the primordial cranium. In the embryo, the notochord tissue extends as far forward as the forebrain, ending just behind the stomodaeal (infundilmlar) infolding. The more cephalic portion, however, atrophies, so that, in the adult, the cylindrical notochord appears to terminate abruptly in a point in the hind brain region, although it is continued forward for some distance as a minute, ill defined filament. The cephalic end of the definite notochord lies on

PROCEEDINGS 395

the tloor of tlu' oiuUH-raniiuM attached to the basilar plato, just inside the occijiital ring. In this respect it resembles the anterior end of the notochord in the Cyclostonies.

Relations to the vascular system. - The notochord of Heterometrus contains a blood sinus that opens at several points into the vascular system. One opening, at the posterior end, leads into a caudal artery, the others lead into vertical vessels that pass through the floor of the nerve cord, between the longitudinal connectives.

('. Minute structure. The principal part of the notochord is a thick wall of lymphoid tissue containing numerous minute, deeply stained nuclei iml)edded in a sharply defined reticulum. Here and there are a few, much larger nuclei, of a very different character, completely filled with very fine chromatin granules.

An ill defined layer of cells forms an outer wall continuous, more especially in the early stages, with the neurilemma of the nerve cords.

A delicate endothelial layer, usually well separated from the notochord tissue, lines the central cavity.

Summanj. The available evidence now indicates that the notochord does not make its first appearance in the vertebrates. It first appears in the arthropods, where it occurs sporadically in widely different groups, under widel}' different structural disguises, and in widely different degrees of development.

While it is to be regarded primarily as a continuous axial organ it may in the same animal be imequally developed in different regions of the body, or present various segmental modifications opposite the ganglionic and interganglionic regions of the nerve cord.

In the vertebrates, the nerve cord becomes more widely established and more uniform in character; that is, it assumes a more constant degree of development throughout the whole class, it extends over a larger part of the body in a given animal, and it is more uniform in character from its anterior to its posterior end.

It will be recalled that the scorpions are the modern survivors of the giant marine Curypterids of the Cambrian and Silurian periods, and that the Curyperids are to be regarded as the probable ancestors of the ostracoderms, and through them of the true vertebrates.

The presence of a well developed notochord in one of the largest of living scorpions is, therefore, a fact of special significance, the more so in view of the facts that in its anatomical structure and in its morphological relations to the alimentary canal, to the nerve cord, and to the cartilagenous cranium, it is in fundamental agreement with the notochord of vertebrates.

Although we have a considerable knowledge of the structure and development of the notochord of -arthropods, we have no definite knowledge of its underlying, or primary, function. It is, however, clear in view of its diversified structure in different regions of the same animal and its unequal specialization in different groups of arthropods, that it cannot everywhere have the same function.

THE ANATOMICAL RECORD, VOL. 11, NO. 6

396 AMERICAN ASSOCIATION OF ANATOMISTS

In the arthropods, two different functions appear in special cases to be exercised. In some cases the notochord serves as a skeleton, or as supporting tissue; in others it has some obscure relation to the vascular system; or both functions may be exercised, at the same time, in the same animal.

Jf9. The pathological development of a young human embryo. C. W. M.

PoYNTER, Anatomy Department, University of Nebraska Medical

College, Omaha.

Doctor Mall has classified pathological ova in four groups as; 1, vesicular forms; 2, ova with neither amnion nor embryo; 3, ova with amnion but no embryo; 4, embryo present but showing more or less degeneration.

This ovum belongs to the fourth group. The chorion, which measures 10 by 8 by 6 mm., was not ruptured; it was mounted in parafRne and sectioned entire.

The embryo is nearly 2 mm. in length, but on account of the unusual development this cannot be taken as an index of age. Estimating from the menstrual history the embryo is about five weeks old.

If atypical environmental factors are responsible for the defective development these factors were probably of a mechanical character, for the chorion seems to be in all respects normal.

The shape of the embryo suggests a development of about 7 somites (Dandy). The neural tube is entirely closed but the differentiation of the brain and cord resemble that of a 3.2 mm. embryo figured by His ('04), except that there is no cephalic flexure.

The body stalk is very much enlarged and represents an outgrowth on the dorsal surface, which is attached to the caudal end of the embryo and extends upward into the amniotic cavity for a distance equal to almost half the length of the embryo.

The circulatory system presents some interesting irregularities. The heart is very much enlarged, extending forward farther than the head, it is roughly 'U' shaped. Only one aortic arch is present on each side and the dorsal aortae are separated throughout their entire length. The right cardinal veins are enormously dilated and there is free communication between the arteries and the veins in the body stalk.

The histological picture is somewhat changed through disassociation (Mall) and infiltration of the tissues with red blood cells.

The embryo furnishes a valuable example of early pathological changes, and in conjunction with a complete clinical history of the pregnancy suggests injury to a normal embryo in the third week of development followed by pathological development for a period for two weeks before death occurred.

60. Some observations on the ossification of the bones of the hand. (Lantern). J. W. Pryor, State University of Kentucky, Lexington. It is my purpose to call your attention briefly to some of the following observations :

PROCEEDINGS 397

1. 'Hie process i)f ossification is inau^iiratcil iinicli sooner than hitherto sup|)osecl.

2. Tiie hones of the female ossify in advance of the male. This is measured at first hy tlays, then months, then years.

3. The ehronolojiieal oriler in whicii the hones of the carpus are ossified is different from that formerly supposed.

4. The bones of the first child, as a rule, ossify sooner than those of subsequent children.

5. Regardless of the variations (normal) the ossification is bilaterally symmetrical.

6. The union of the epiphyses with the shaft takes place much sooner than formerly supposed.

7. Variation in the ossification of bones is a heritable trait.

51. On the use of the word 'sympathetic^ in anatomical and 'physiological

nomenclature. S. W. Ranson. Northwestern University Medical

School.

Confusion and disorder exist in the literature on visceral innervation because of the lack of a satisfactory and comprehensive terminology which meets the needs both of the anatomist and physiologist. The most important contributions to this difficult subject have been made by Langley, who is also responsible for much of the confusion through the unfortunate use of the term sympathetic. Defined in accordance with the accepted anatomical nomenclature, the sympathetic nervous system is that aggregation of plexuses, nerves and ganglia especially concerned in the innervation of the viscera, glands and smooth muscle. Defined in accordance with physiological usage, the term sympathetic nervous system includes only the visceral efferent fibers of the white rami and their postganghonic connections. The terms used by English and German phj'siologists are given in parallel columns in table 1. Note the divergent use of the word autonomic.

The choice of the adjective sympathetic is unfortunate in any case having in its favor only the advantage of estabHshed usage. It is doubly unfortunate that the term should be employed in two such different senses. If it were possible it would be desirable to drop the word entirely and substitute others in its place. As a first step towards adopting a satisfactory terminology it is necessary to have clearly before us the correct interrelation of the parts to be named as well as a statement of what terms may rightly be regarded as synonymous. These relations are expressed in table 2. In the first column the word sjnnpathetic is retained but with the use of the qualifying terms major and minor. In the second column that word is eliminated with the mtroduction of one new term "the plexiform nervous sj'stem." The third column is the same as the second except that the official term systema nerv'orum sympatheticum" is substituted for the "plexiform nervous system."

The major sjinpathetic system is that aggregation of plexuses, nerves and ganglia especially concerned in the innervation of the vis

398 AMERICAN ASSOCIATION OF ANATOMISTS

cera, glands and smooth muscle. It is obvious that some term is needed to designate this anatomical complex. While estabUshed usage may make it necessary to retain the word sympathetic in this connection,, it would be desirable to have a term drawn from some obvious gross characteristic of the system. This would suggest the use of the word plexiform. In any case the name chosen is not of so much importance if we understand that it applies to a gross anatomical complex and carries no implication concerning internal structure and function. The word autonomic cannot properly be used in this connection because it refers to a purely efferent system, while all agree that the complex under discussion contains afferent fibers on their way from the cerebrospinal ganglia to the viscera.

It is therefore necessary and proper to have under the major sympathetic or plexiform nervous system a subdivision including the visceral afferent components. Although the only afferent elements that have been satisfactorily demonstrated are fibers having their cells of origin in the cerebrospinal ganglia, this subdivision would give a place for any other sensory elements that might later be shown to exist.

The autonomic nervous system according to English usage includes all general visceral efferent elements both pre and post ganglionic. It is obvious that the term autonomic cannot be used as an equivalent for major sympathetic or plexiform, because it excludes the afferent fibers and because the autonomic fibers extend beyond the plexiform system through the cerobrospinal nerves into the cerobrospinal axis. The autonomic nervous system includes for instance the cells of the intermediolateral group of the spinal cord and the visceral efferent fibers in the ventral roots. It designates a functional group of neurones which are partly within and partly without the cerobrospinal nervous system. It is therefore correct to say that the major sympathetic or plexiform nervous system contains autonomic components though it does not contain all of the autonomic sj^stem. The preganglionic autonomic fibers leave the cerebrospinal axis in three streams: the cranial, thoracicolumbar and sacral. For many reasons, anatomical, physiological and pharmacological, it is desirable to group the cranial and sacral streams together and contrast them with the thoracicolumbar.

The thoracicolumbar autonomics include the visceral efferent fibers of the white rami and their postganglionic connections. This group Langley has called the sympathetic nervous system and we suggest that if this word is to be retained at all the designation be the minor sympathetic nervous system.

The other group of preganglionic visceral efferent fibers make their exit along the III, VII, IX, X and XI cranial and the II, III and IV sacral nerves. These with their postganglionic connections constitute the craniosacral autonomics or the parasympathetic nervous system.

In addition to the afferent and efferent components connecting the viscera with the cerebrospinal axis, there is probably present in the

PROCEEDINGS

399

gastroiiitostinal tract a mechanism for local reflexes. Such local reflexes, if tiiey exist, must clepeiul upon a mechanism different from that of the autonomic system. In order to give a place in our classification for these myenteric reflex arcs we have added a third subdivision for which we have adopted Langley's designation — the enteric nervous system. This should be understood to include only those elements in the gastroenteric plexuses which are involved in the myenteric reflex. While it may be necessary to retain the word sympathetic in some anatomical names, it is desirable to restrict its use as far as possible. Such terms as sympathetic ganglion or sympathetic fiber should never

TABLE 1

English and German nomenclature contrasted

AUTONOMIC NERVOUS SYSTEM

DAS VEGETATIVE NEBVENSYSTEM

Parasympathetic system Sympathetic system

Das autonome system Das sympathische system

TABLE 2

Showing the correct relations of the various parts of tJie peripheral nervous system primarily concerned with visceral innervation giving synonymous terms in parallel columns •

MAJOR SYMPATHETIC SYSTEM

Visceral aflferent components Autonomic components^ Minor sympathetic

components Paras3'mpathetic components Enteric components

PLEXIFORM NERVOUS SYSTEM

Visceral afferent components Autonomic components^ Thoracicolumbar components Craniosacral components Enteric components

SYSTEMA NERVORUM SYMPATHBTICUM

Visceral afferent components Autonomic components^ Thoracicolumbar components Craniosacral components Enteric components

^ The autonomic nervous system and its two major divisions, the thoracicolumbar or minor sjonpathetic system and the craniosacral or parasympathetic system, include cells in the brain and spinal cord and fibers in the cerebrospinal nerves and cannot therefore be classed under the major sympathetic or plexiform nervous system. Yet the latter is made up in large part of components belonging to one or the other divisions of the autonomic system.

be used. Since the ganglionated cord is associated only with the thoraciolumbar autonomies or minor sympathetic system we may with good reason retain for it the name — sympathetic trunk. When it is necessary to designate the ganglia of this system by a general term they should be called autonomic ganglia, since they are all essentially relay stations in the autonomic system. In the names of the individual ganglia, plexuses and nerves no qualifying adjective is needed, as for example, the superior cervical ganglion, the carotid plexus and the carotid nerve. Since the Latin name, systema nervorum sympatheticum has never been used in the same loose way as its English equiva

400 AMERICAN ASSOCIATION OF ANATOMISTS

lent it might be used without danger of confusion if the Latin form were always retained and the Enghsh equivalent never used, even to designate the thorocicolumbar autonomics. All things considered the third column of table 2 seems to offer the best solution of the problem, although the synonyms suggested in the other columns may occasionally be useful.

More important than the selection of satisfactory terms is the recognition of the correct relationship of the various parts to be named. It is clear that there is a gross anatomical complex, the systema nervorum sympatheticum or plexiform nervous system. It is also clear that this complex contains autonomic components but not the entire autonomic system and that it also contains visceral afferent fibers. The clear recognition by both anatomists and physiologists of these apparently self evident facts would make possible the adoption of a consistent and uniform nomenclature.

52. Studies in elastic tissue. III. The behavior of the elastica in arteries

following ligation and in the organization of thrombi which ensue.

(Lantern). J. Parsons Schaeffer, The Daniel Baugh Institute

of Anatomy, Jefferson Medical College, Philadelphia.

In earUer communcations of this series of studies an account of the behavior of elastica in the occlusion and the obliteration as such of the ductus arteriosus (Botalli) in man and pig was presented. In both instances elastica was found to play an important role in the occlusion of the lumen of the postfetal ductus. Many new elastic fibrils were demonstrated; some, doubtless, the product of preformed elastica, others apparently the product of protoplasmic activity of cells in loco and of certain wandering cells which found their way into the occluding mass.

It is a well known fact that elastic tissue behaves diversely m the various pathologic states. Even in the organization of thrombi elastic tissue presents bizarre appearances which seem dependent upon the etiological factor back of the thrombus. In view of the latter, an investigation of the behavior of elastic tissue in arteries after ligation was undertaken. Opportunity was also afforded in this study for observing the behavior of elastic tissue in the organization of thrombi following ligation. This work is merely in a preliminary state and is briefly abstracted herewith.

The rabbit was the animal used for the experimental work. Twentyfive animals were operated upon after the usual technic preparatory to surgical procedures and anaesthetization with ether. Twenty-four animals recovered rapidly' from the operation with no apparent ill effects. One animal died of an infected wound.

The common carotid artery of eHher the right or the left side was secured and ligated at two points, an inch or less of artery intervening between the two ligatures. The distal ligature was iisually placed first, thus insuring a })lood-filled segment between the ligatures. The ligatures were placed variously: In some instances the endothelial

PROCEEDINGS 401

walls wi'iv l)rc)\i<!;ht into iiktc ap]iosition without injury, in otlu^rs tlio ligatiiros were tightly placed with obvious endotiielial injury. In most instances the hlood was left in the artery segment, in others a slit was cut and the hlood was allowed to escai)e. Again tiie artery segment was consiilerahly traumatized, in others handled with a minimum traimui. At selected periods after the operation the animals were again etherized and the ligated vessels secured for study.

It is a well known fact that clots may form within hlood vessels upon the introduction of foreign material. Injured endothelial cells may act as a foreign substance. In the present series it was found that in those cases in which the artery was ligated gently, i.e., the endothelial walls brought into mere apposition, the blood between the ligatures remained fluid for a relatively long time. However, when the artery was considerably traumatized and the ligatures tightly placed so as to injure the endothelium, intravascular clotting between the ligatures ensued nuich sooner. Again, the blood became thrombosed in one portion of the artery segment between the ligatures and at another it would remain fluid much longer. Indeed, at some points the blood elements would undergo disiiitegration and no thrombosis or organization occur in that position. The subendothelial stratum would thicken instead and gradually encroach upon the lumen. Later the cone of the thrombus from another portion of the segment of the artery would grow into the narrowed lumen.

After an interval of from five to tM'elve days following the ligation, there was found here and there a positive cellular thickening of the subendothelial stratum. It is well to recall that in the common carotid artery of the rabbit, the endotheHum at many places rests directly on the inner elastic lamina. At other places a few scattered connective tissue cells intervene. In other words, the subendothelial stratum IS for the most part wanting. It seems scarcely possible that the cells in the thickened subendothehal stratum all have their source from the few connective tissue cells in loco. Furthermore, the endothelium is for a considerable time intact and continuous. This is especially so in those cases in which the intravascular clotting is dela3^ed. It is not believed, therefore, that the endothelium contributes to this initial cellular thickening. There is another possible source, viz: the connective tissue cells of the media and adventitia. It is probable that some of these cells wander mto the subendothelial stratum and contribute to the early cellular thickening. Indeed, serial sections reveal occasionally connective tissue cells on their way through the inner elastic lamina and at points one can demonstrate a number of cells near the mner elastic lamina looking vertically towards the lumen of the vessel. Of course, m those cases in which the blood early becomes thrombosed and organization sets in, the endothelial lining soon breaks up and undergoes prohferation and doubtless aids in the organization of the thrombus. In such instances the endothelial cells become modified and seem to assume a fibroblastic role.

402 AMERICAN ASSOCIATION OF ANATOMISTS

When the blood remains fluid for a long time, the subendothelial stratum becomes thicker and thicker at the expense of the lumen. The enclosed blood is encroached upon. Usually the cone of a growmg thrombus from another portion of the artery segment, pushes its way into the constricted lumen. In cross-section the cone of the thrombus is entirely free. Serial sections show, however, it to be a part of a thrombus from another portion of the artery segment undergoing organization.

It is the very earljf but obvious thickening of the subendothelial stratum that is of prime interest here. Careful study of thin sections reveals in the outlying portion of the exoplasm of the cells in the thickened endothelial stratum very delicate, granular-appearing elastic fibrils. These fibrils in a sense immesh cells, and their appearance, position, and relation indicate that they are the product of protoplasmic activity. At this time the inner dastic lamina appears normal and healthy. The delicate, granular elastic fibrils have no connection with it.

Subsequently the inner elastic lamina, shows here and there a splitting into several layers. In these positions the lamina becomes fragmented, the fragments projecting into the subendothelial stratum and seemingly undergo proliferation in this position. The inner elastic lamina in such positions appears 'moth-eaten.'

The thickened subendothelial stratum is, therefore, made up of connective tissue cells and collagenous and elastic fibrils, the latter of a dual source. When the blood remains fluid for a long time the subendothelial stratum will continue to thicken by a multiplication of these elements and at the expense of the lumen. It reminds one of an obliterating arterio-sclerotic process.

If intravascular clotting early ensues as is often the case the subendothelial stratum has little time to thicken. The endothelial lining is broken up. There is a rapid migration, especially along fibrin bridges, of the elements of the subendothelial stratum and of endothelial cells into the thrombus to aid in its organization. It appears that the endothelial cells assume a fibroblastic role in the rapid organization of the thrombus.

The newly formed elastic tissue extends ring-like into the organizing thrombus, forming a net-like mass. This gradually pervades the whole thrombus. Newly formed blood vessels within the thrombus alwaj^s have very definite inner elastic lamina.

In presenile gangrene, the result of a thrombo-angiitis obliterans probably due to an infection, there is always an absence of elastic fibers save a few around the larger canalizing vessels. Here we have a definite and positive organization of a thrombus with little participation of the elastica. In an arterio-sclerotic process (without thrombosis) in which the lumen of the vessel may become obliterated, we have an abundance of elastic fibers aiding in the occlusion. In the present study where thrombi were cxperinu^ntally produced under aseptic conditions, elastic tissue plays an important role in thrombus organization.

PROCEEDINGS 403

In a few iiistancos a lonp;itiHliiial slit was cut into the artery segment between the Hffatiires and the blood allowed to escape. Where the separation of tlie cut edf;es was not too Ki"<it, the elastica aided in till' rei)air. New elastic fibrils were formed from the cut ends of the ch-cular fil)ers of th(> inner elastic lamina, in an effort to establish continuity of structure of elastic librils. However, where the separation of the cut edges was fairly great, the hunen of the vessel was rapidly encroached upon by endothelium and connective tissue elements. New elastica aided in the occlusion of the lumen, b\it the cut ends of the circular fibers of the inner elastic lamina did not throw out new fibrils to the same extent as in the positions where the cut edges of the vessels were separated to a less degree. Seemingly when the cut ends of the circular fibers of the inner elastic lamina are less separated, some stimulus is forthcoming to l)ring about proliferation of the cut ends in an attempt to establish conthniity of structure of elastic fibrils. Where the separation is too great, the cut ends of the circular fibers of the inner elastic lamina not only fail to give rise to new fibrils, but they actually seem to undergo disintegration. A study in the larger series is, of course, necessary to draw conclusions. The behavior of the longitudinal fibers of the inner elastic lamina were not studied.

53. The early stages of the development of the great veins and of the hepatic circulation in the cat. H. von W. Schulte, Anatomical Laboratory

of Columbia University.

That the duct of Cuvier has a more complicated history than is expressed in the familiar statement that it serves to connect the preand post-cardinal veins with the sinus venosus, has long been known to embryologists. A model of the great veins of a cat embryo of 5 mm. by Dr. Huntington, shows the participation in its formation of a plurality of elements. Of this it has been stated that "the duct of Cuvier is formed through the confluence of the precardinal, the postcardinal, the omphalo-mesenteric and the umbilical veins." (Huntington and McClure: Am. Jour. Anat. vol. 10, 1910, p. 231 and fig. 25). It is now possible, through the opportunity of suitable material, to record the stages antecedent to the condition above described. In embryos having 12 to 14 pairs of mesodermic somites the umbilical vein extends to the anterior limb bud and receives in addition to plexiform tributaries from the somatopleure, the precardinal vein. This primitive drainage of the precardinal has previously been described by His and by Lewis (Proc. Ass. Am. Anat., 17th Sess., 1903). In the cat at 14 somites the umbilical crosses the omphalo-mesenteric vein dorsally and here opposite the interval between the third and fourth, somites anastomoses with the latter (Mem. Wistar Inst., No. 3, fig. 16).

Subsequently with the lengthening of the foregut ancl the development of a ventral body wall, the omphalo-mesenteric vein for a segment of its course becomes parallel to the umbilical which has been carried ventrad and now lies against the lateral side of the omphalomesenteric. As the umbilical extends to the limb-bud a considerable

404 AMERICAN ASSOCIATION OF ANATOMISTS

segment of it extends cephalad of its communication with the omphalo-mesenteric vein. With this segment the precardinal retains its connection, a portion of the umbihcal extending beyond the junction. Additional taps are estabHshed between the parallel segments of the umbilical and omphalo-mesenteric veins and a portion of the umbilical is gradually incorporated into the latter vessel. This is accomplished in embryos of 4 to 5 mm. The umbilical now appears as a tributary of the omphalo-mesenteric vein and its cephalic segment gradually falls in line with and forms the continuation of the precardinal. Meanwhile the postcardinal has increased in size and with the resolution of its plexiform connections with the umbilical establishes a secondary anastomosis with the precardinal. The blood from both cardinal veins from this period is carried to the heart through the proximal segment of the precardinal, a portion of the umbilical and finally a portion of the omphalo-mesenteric, the resulting vessel constituting the duct of Cuvier.

The earliest capillaries of the septum transversum make their appearance in close proximity to the entoderm of the foregut in the interval between the omphalo-mesenteric veins and immediately caudal to the sinus venosus. Angiogenesis is active and prior to the appearance of the hepatic diverticulum a plexus is formed which becomes connected at the side with the omphalo-mesenteric veins and cephalad with the sinus venosus. These capillaries are preceded by the formation of angiocysts among which are scattered small blood islands. Many of the vessels are of considerable size with wide irregular dilatations. Subsequently this plexus becomes connected with that about the lung bud and is evidently an accelerated part of the general peri-oesophageal or splanchnic plexus (Davis, Brown).

The hepatic diverticulum is present in embryos of 19 and 20 pairs of mesodermic somites, and naturally is in immediate contact with this plexus of the septum transversum. The ventro-lateral sprouts of the liver invade the septum transversum which increases rapidly in size and continues to be actively angiogenetic. An invasion of the omphalo-mesenteric and umbilical veins is not present until the stage of 4 to 4.5 mm. At this period on the right side a mass of liver sprouts fills the angle of confluence of these vessels and a long falciform process of liver grows out upon the dorsum of their fused segment. Here upon follows rapidly the resolution of this segment into sinusoids and a reduction in the size of the distal segment of both veins. The left omphalo-mesenteric has increased greatly in size, the left umbilical moderately. In the ventral region of the liver the capillaries are abundant and continue to be connected with both omphalo-mesenteric veins and with the sinus venosus. Many of them are separated from immediate contact with the liver sprouts by the interposition of a moderate amount of young connective tissue. This topography and their antecedence in time of the hepatic sprouts, substantiates Mollier's statement of the independent formation of capillaries in the septum transversum and indicates that a considerable portion of the hepatic

PROCEEDINGS 405

vessels :ire formed before and independently of the resolution of the onii)hal()-niesenteric and umbilical veins hito sumsoids.

J4. Observations on the osteology of the porcupine fish. (Diodon hystrix) . R. W. SiiUFELDT, Washington, D. C.

Several years ago I made disarticulated skeletons of Diodon hystrix, and compared them with others I had prepared of the Burr fishes (Chilomycterus schoeppi); a skull of some species of Ovoides; the skeletons of several species of Spheroides, popularly known as Swell fishes, and of two or three of the trunk fishes, as Lactophys triqueter, L. tricornis and others. Also with the Butterfly-fishes (Chaetodon) and Angel-fishes (Angelichthys), and, finally, with several species of the File fishes of the family Monacanthidae, including Monacanthus hispidus and two or three species of Alutera, the last kindly presented me by Dr. Francis E. Sumner, of Woods Hole, Massachusetts.

This material is before me at the present writing, as well as a series of photographs I have made of not a few of these skeletons.

Jordan and Evermann, in the second part of their most useful work on the "Fishes of North and Middle America," place all of the abovenamed forms in groups more or less nearly allied to each other. For example, in the Suborder Squamipinnes (or the Scalj'-fins) we find the Chaetodontidae, the species of the genus Angelichthys (Angelfishes), and various others, as the Doctor-fishes, which I have likewise osteologically examined (Teuthis). Following the Squamipinnes, we have the Suborder Sclerodermi, in which we have the File-fishes (Monocanthidae); the Suborder Ostracoclermi, created to contain theTrunkfishes, and, finally, the Suborder Gymnodotes, which includes the families Tetraodontidae or the Puffers, the family Canthigasteridae or Sharp-nosed Puffers, and, lastly, the family Diodontidae or Porcupine fishes, in which group we find the subject of my researches of which the present abstract is a brief resume.

All the Diodontidae are very sluggish fish, and they usually remain near the bottom, slowly swimming about amidst the marine vegetation and the various kinds of corals there found. Their armor-spines are far more formidable than we find them in their relative, the Puffers, and the bones entering into the formation of the mandibles unite so completely in the adult fish, that the sutures almost, if not entirely disappear. This has led some ichthyologists to believe that each jaw is composed of a single bone.

This remarkable fish, with its curious coat-of-mail bearing from head to tail the long, bony, and very sharp spines, may grow to be a yard long, or even slightly longer in some specimens which have been met with off the Bermudas. It is very abundant everywhere in tropical seas, especially in the Hawaiian Islands, coast of Lower California, and from the Carolinas all the way around the eastern coast of North America. It is useless as a food fish, while its odd-looking, spiny skin is frequently found in various places and in collections, blown up and dried as a curiosity; mounted specimens also are preserved in

406 AMERICAN ASSOCIATION OF ANATOMISTS " ,

a similar manner. There are those who beUeve that the flesh of the Porcupine fish is poisonous if eaten; but there is absolutely no truth in this idea.

]\Iany are familiar with the habit of the Puffers or Swell-fish of blowing themselves up when caught and taken out of the water. A slight scratching of the abdomen will induce the fish to do this. In Diodon this capacity of inflation is very much feebler, though the fish has the power of gulping in sufficient air to cause it to float, belly upwards, on the surface of the water. Many years ago, I saw one floating in this manner in the Florida Straits, when the surface of the water was smooth and at rest.

Apart from the dense and massive mandibular bones and a few of the smaller bones of the cranium, the entire skeleton of Diodon, as in the case of its near allies, differs from that of most true teleosteans, in that the bones composing it are quite elementary in appearance and texture— as though they were composed of a whitish paper pulp rather than of true bone. This does not apply, however, to the spines of its coat-of -armor; they are especially hard and dense, and as stiong as steel. When my final paper on the osteology of this fish is published, there will appear in it an account of the most interesting articulations among these spines, which are imbedded in the integuments so that the fish can elevate or depress them at will._ After the heavy and massive jaws have been removed, the cranium is seen to be of a sub-cubical form, being almost as broad and as deep on its anterior facial aspect as it is upon its dorsal superficies.

The superior mandible is, as just stated above, a very heavy and massive bone, being composed of the thoroughly fused maxillaries and pre-maxillaries, the sutures between the two being plainly visible, but only faintly in the median line between the premaxillaries. As one bone, it has the form of a broad U, and the corrugated dental plates are similar in the two jaws. On the superior edge of either maxillary there is a free admaxillary.

Viewing the cranium from in front, it is to be noted that it is the nasals that have, in chief, been modified to articulate with the upper jaw; and from them, upon either hand and separated by a wide interval, descends a long column of bone to include the mandibular articulation below. In either one of these columns we have an extraordinary arrangement of the quadrates and entopterygoids.

This is difficult to make clear without the use of a figure of the skull: but such a figure and others have already been made, and will illustrate my final contribution to this subject.

On dorsal aspect, the skull of the Porcupine fish is quadrilateral in outline, the anterior moiety being formed by the large and spreading frontals. All the bones on the roof of the cranium are firmly united together through a peculiar overlapping articulation. Most of the sutures can be made out; while, in the case of others, the boundaries have nearly disappeared in the adult.

PROCEEDINGS 407

Thoro is a small cranial capacity, and tlic cranial casket is entirely open anteriorly; while anterior to its inferior border the presphenoid passes forwards as a tnmipet-shaped hone, w'th the small end attached to the hasisphenoid.

On either side of the skull a rather large symplcctic is present, making; its usual articulations with the surrounding bones.

The remaining i)ones of this part of the skull will be fully described later on in my formal monograph on the osteology of this truly remarkable fish.

As in the case of the upper jaw, the mandible is a very massive, U-shaped structure, supporting a row of small teeth, and a somewhat removed, medium, subelliptical plate of others, quite similar to the dental armature in the upper jaw. The articulation with the quadrate is quite extensive.

The bones of the branchial apparatus and tongue are all handsomely developed, but cannot be well described here without the use of figures.

Passing to the spinal column, we find the vertebrae large and unusually well developed for a fish having a coat-of-mail. In the Burr fishes for example, this is not the case, and they possess a very highly developed armor (Chilomycterus). Diodon has the leading eight (8) vertebrae almost devoid of prominent lateral and inferior apophyses. In the ninth vertebrae, however, these suddenly appear, to become broad and spreading in the 12 to 15 vertebrae. There are 20 vertebrae all told in the spine of this fish, and they present some very remarkable characters. The terminal one is much compressed laterally, and very large. It supports, upon either side, a prominent hji^ural spine, and offers a long shape margin for articulation with the caudal fin, or peduncle. The neural spines supporting the dorsal fin fuse below with the neurapophyses of the 12 to 16 vertebrae, which have their neural spines powerfully compressed laterally, and otherwise modified to receive them. The raj^s of all the fins are well ossified, and the actinosts thoroughly developed, in some instances of considerable size. The important comparisons of the skeleton of Diodon with those of its near relatives will appear in my formal contribution, as they are of too extensive a nature to be incorporated in the present abstract.

o5. The ovarian cycle in mice. H. P. Smith, Anatomical Laboratory, University of California.

During the last year a study of the ovulation cycle in mice has been undertaken in conjunction with Dr. J. A. Long, a preliminary report of which has been published elsewhere. On that occasion we presented the results derived from the study of serial sections through the oviducts and ovaries of 61 mice which had been isolated from their litters and from males immediately upon parturition, and killed at varying intervals thereafter. In this waj- we avoided any possible effect of lactation or sexual excitement due to the presence of the male. That study, which has been carried out over a period of 91 days fol

408

AMERICAN ASSOCIATION OF ANATOMISTS

CALCULATED

LENGTH OP TIME

LENGTH OF TIME

SERIES

BETWEEN PARTURITION AND

POSITION OF EGGS IN OVIDUCT

BETWEEN PARTURITION

KILLING

AND OVULATION

121

20 hours

Ovarian one-third

20 hours

186

33 hours

One-third way down

20 hours

216

40 hours

Nearly one-half way down

20 hours

488

48 hours

In second fifth

20 hours

448

48 hours

In ovarian fifth

20 hours

489

60 hours

Two-fifths way down

20 hours

483

72 hours

In last fold

20 hours

490

96 hours

At entrance to uterus

20 hours

491

96 hours

At entrance to uterus

20 hours

414

96 hours

Near entrance to uterus

20 hours

492

108 hours

In last fold

20 hours

494

6 days 415

6 days

^

460

7 days 416

8 days 485

8| days 486

9 days

In loop leading to last fold

6| days

463

9 days 417

10 days 418

m days 419

11 days 420

12 days 450

12 days 496

12i days 421

14 days 497

15 days 422

16 days 498

17i days 462

18 days 423

18 days

Uterine one-third

16i days

447

18i days 432

m days 459

18^ days

Uterine one-fourth

16^ days

484

18f days

One-fifth way down

18 days

461

19 days 424

19 days

Ovarian one-fourth

18 days

434

19 days 500

19 days 501

19 days

One-fourth way down

18 days

502

19 days

One-fifth way down

181 days

487

195 days

At ovarian end

19 days

504

19^ days 439

19| days 505

19i days 506

20 days 425

20 days

Little over half way down

181 days

507

201 days 426

20i days

Uterine one-fourth

18| days

509

21 days 427

2U days 465

2U days 428

22 days

PROCEEDINGS

409

lowing parturition, pointed to the conclusion tluit ovulation in mice occurs spontaneously at intervals of ahotit ]7\ (h\ys.

The present report of contiiuied work mulertaken at the suggestion of Dr. Long, represents an attempt to examine a greater series of cases during the first 22 days post partum in order to detect variations which might occur as regards the time of occurrence of ovulation.

Fifty-two cases are tabulated on p. 92. It will be seen that it is possible to state that the next spontaneous oculation following upon the one which so quickly succeeds parturition occurs between 16.J and 19 days after parturition. The average of the nine such cases observed in this study is 18 days after parturition or in other words a few hours less than 17 days after the ovulation following littering.

A surpi'i.sing result of the above table must now be alluded to, i. e.: the spontaneous ovulation in question occurred in onl}^ 42 per cent of the cases in which it would be expected (9 cases in 21 cases). No explanation is offered for this irregularity. ^ Possibility of individual cases of still greater variation is indicated by one instance in which a second spontaneous ovulation occurred at 6^ days post partum (case 486).

Furthermore, the early part of the present series gives us a good indication of the rate of migration and time of survival of unfertilized eggs in the oviduct of the mouse. A part of the above table may be presented again in the form given below. From it may be inferred that the ovum of the mouse consumes approximately two days in LENGTH OF TIME BETWEEN ESTIMATED OVULATION

SERIES

(time of PARTURITION

position OF EGGS IN OVIDUCT

PLUS 20 HOURS

-LONG AND MARk2) and

killINQ

hours 121

Ovarian third

186

13

One-third way down

216

20

Nearly half way down

488

28

In second fifth

448

28

In ovarian fifth

489

40

Two-fifths way down

483

52

In last fold

490

76

At entrance to uterus

491

76

At entrance to uterus

414

76

Near entrance to uterus

492

88

In last fold

^ Since one case of estimated ovulation at 19 days was found it will be well to recognize that other instances of as late ovulation could conceivably be represented by cases 439, 462, 447, and 432, killed prematurely.

Precisely the time of occurrence of ovulation did not interest Long and Mark, but from their careful table of stages of ovulation observed in 19 mice killed from 14| to 28? hours post partum one may conclude that in over 75 per cent of all cases ovulation had occurred by the twentieth hour after parturition. (Long and Mark. The maturation of the egg of the mouse. Publ. Carnegie Institution of Washington No. 142, p. 20, table 4, 1911.)

410 AMERICAN ASSOCIATION OF ANATOMISTS

traversing the greater part of the oviduct but that it also waits in the last uterine loop or portion of the oviduct approximately one day or more so that three days must be allowed for the completion of migration.

It was in fact by the use of the second table that estimations of time of ovulation were recorded in table 1 .

'56. The effect of hypophysectomy upon the subsequent growth and development of the frog {Rana hoylei.) P. E. Smith, University of California, Berkeley.

In the operated specmiens the hypophyseal rudiment was removed soon after it had commenced to invaginate, that is shortly after the closure of the medullary tube. Controls consist of specimens upon which the operation was unsuccessfully attempted, and of unoperated animals. All animals were reared under identical conditions.

In the hypophysectomized specimens particular attention is called to the non-development of the hind legs; to the pronounced decrease in the size, parenchyma, and colloid of the thyroid gland; and to the profound reduction of the epidermal pigment.

Only one specimen of the group in which the hypophyseal rudiment was ablated, developed legs at the normal rate. Sections show that, although the glandular portion of the hypophysis was totally extirpated, yet the thyroid developed normally and the epidermal pigment was typically reduced.

57. Changes in the relative weights of the various parts, systems and organs of very young albino rats urtderfed for various periods. C. A.

Stewart, Institute of Anatomy, University of Minnesota.

Forty-three rats have been used, including 20 controls and 23 test animals. Of the controls 6 were dissected at an average net bodyweight of 9.9 grams, 4 at 12.9 grams and 10 at 14.7 grams. The test rats were starved for intermittent periods starting a few hours after birth, and were dissected at the age of three weeks (7 rats), six weeks (7 rats), and ten weeks (9 rats), the average net weight at each age iDeing 10.1, 12.5, and 14.9 grams. On account of the normal variability, and the small number of observations, conclusions are somewhat uncertain in some cases.

As to the body proportions, the weights of the head, trunk and extremities are practically normal in the test rats as compared with the (younger) controls of the same weight. The existing small differences may appear more significant upon the addition of further data.

Of the systems, the skeleton and visceral group (as a whole) show a considerable increase in weight in the test rats. There is also a slight increase in the musculature. The integument is variable, showing an increase in the rats underfed three weeks, but a marked decrease in the later periods. The 'remainder' appears to decrease greatly in the rats underfed three weeks, while in the rats underfed six and ten weeks the change is less marked.

PROCEEDINGS 411

Of tlu* individual vit?c'(M*a, tlu* spinal cord, eyeballs, liver, stomach and intestines (empty), suprarenals, kidneys, testes, hyi)ophysis and ovaries show a definite increase in weif:;ht in all three fi;roiips. The brain 'and epididymi show a marked increase in the rats imderfed the shorter periotl (3 weeks), but are doubtful later. The heart antl spleen are variable, each apparently losing weight during the earlier fasting period (8 weeks), but increasing in weight so as to surpass the controls during tlie later periods.

The hmgs sufYer a considerable loss in weight in all three groups and likewise the thymus (especially at 10 weeks).

There is apparently no marked change in the weights of the thyroid and pineal l)ody.

In general the results agree fairly well with those oljtained by Jackson (Jour. Exp. Zool., vol. 19, no. 2, '15) in rats held at maintenance for various periods beginning at 3 weeks of age. There are, however, several differences found in tlie rats of the present experiment, in which the underfeeding was begun shortly' after birth. These differences are probably due to the varj-ing tendencies to growth and maintenance among the various organs at this earlier period.

oS. The existence of a typical oestrous cycle in guinea-pigs and its histology. Charles R. Stockard and George N. Papanicolaou, Cornell University Medical College, New York City. Normal guinea-pigs of our control stock possess a regular periodic procestrum, occurring every fifteen or sixteen days. This fact was ascertained by examining the vaginae by means of a speculum every day during different seasons of the year. The flow, which marks the prooestrum activity, is not very abundant and consists largely of desquamated epithelial cells and some mucous secretion.

In the first stage there is a flow of a mucous fluid filled with superficial squamous cells of the vagina. A few hours later a thick cheeselike substance occurs in the vagina. This consists almost entirely of the deeper epithelial cells which preserve their eipthelial structure and often remain together in groups or actual pieces of epithelial tissue. This thick vaginal substance after a few hours becomes more fluid in consistency and pus-like in appearance. A microscopical examination at this time shows a very large amount of polymorphonuclear leukoc}i:es among the epithelial cells and the beginning of an active phagocytosis. The result of this phagocytosis is that in a few hours the vagina is almost completely cleaned and no longer contains the menstrual substance except for a little fluid containing leucocytes and a few broken down epithelial cells. The occurrence of either red blood corpuscles or hemoglobin in the menstrual flow takes place only in the later stages.

The active phagocytosis may probably account for the fact that none or very little of the pus-like fluid ever flows out from the vagina. The leucoc>i:es, migrating from the subepithelial capillaries of the uterus and the vagina, as microscopical sections show, attack the desquamated

THE ANATOMICAL RECORD, VOL. 11, NO. 6

412 AMEKICAN ASSOCIATION OF ANATOMISTS

epithelial cells within the lumen of the uterus and vagma and there begin to destroj' them. The entire process of menstruation is not long, its duration bemg less than twenty-four hours. But during this time the entire genital tract is inflamed by a very active circulation of blood.

The anoestrous period of the uterus and vagina is characterized by the absence of the secretions and the constant presence of leucocytes. Especialh^ the first week after prooestrum, the vagina is very clean and dry. During the second week and particularly a few days before the next prooestrum there is a little mucous fluid in the vagina, and this con tains some leucocytes and a few squamous cells. The massive desquamation and the abundant thick secretion, however, occurs very regularly every fifteenth or sixteenth day.

The menstruation or prooestrum seems to be closely followed by an ovulation. At this stage of our preliminary study of the correlation between menstruation and ovulation observations mdicate that the ovulation occurs about eighteen hours after the height of menstruation (the presence of the abundant thick secretion).

59. The morphological changes of the idiosome during the spermatogenesis of the guinea-pig. C. R. Stockard and George N. Papanicolaou,

Cornell Medical School, New York City.

La Valette St. George was the first to describe the idiosome under the name 'Nebenkern' in 1865-1867. Since then this body has been described under many different names in a great number of papers treating the processes of spermatogenesis and oogenesis in different animal classes. Meves ('99), described the structure during the spermatogenesis of the gumea-pig and called it the 'idiozom' (from idios — own, and zoma — belt-). Regaud ('10) proposed a modification of this term to idiosome (from idios -own and soma-body). We accept the modification as better fitting our conception of this structure.

Not very much specific study has been devoted to the idiosome probably on account of the fact that its close position in relation to the centrosomes has made some observers fail to realize that it is in independent body, having its own peculiar history. There is certainly at particular stages a very close relation as to position between the idiosome and the centrosomes, but the relation is temporary. In only one stage in the spermatogenesis of the guinea-pig, are the centrosomes really in very close connection with the idiosome, and this is in the primary spermatocytes. During this stage the centrosomes are enclosed in the center of the idiosome. But as soon as the spireme begins to form they come out of the idiosome to play their usual role during the nuclear division, while the idiosome continues its independent development with a surprising succession of highly specialized morphological changes. In the later stages, the secondary spermatocytes and spermatids, no connection is to be observed between the centrosomes and the idiosome.

rUOL'EEDINGS 413

The uliosonu* in iho male f^erin cells of the fdiinca-pifi; is present even in the spernuitof^oiiia. At this sta^e the ichosome is of somewhat irrei^iTilar shape and ai)pearance and its internal structure is not clear and detinite. First in the jjrimary spermatoc>'tes, the idiosome takes a repihir sjiherical form and shows a clear differentiation into two zones, a peripheral, the idioectosome, and a central, the idioendosome. The latter zone is completely enclosed by the former. When the prophase beghis in the tlivision of the primar}^ spermatocytes the idioendosome breaks up into a ninnber of granules the idiogranulomes. During the progress of tlivision, the idioectosome also breaks into smaller pieces, which, together with the itliogranulomes are dispersed throughout the protoplasm. In this way a uniform distribution of the idiosomatic material is secured din-ing the division process.

At the end of the division the idiogranulomes and the pieces of the idioectosome beghi to flow together near the daughter-nuclei and thus form two daughter-idiosomes. In the secondary spermatocytes the idioectosome takes a regular spherical shape, while the idiogranulomes form a group of closely arranged granules hi its center, as if they were preparing to fuse. A fusion does not take place, however, probably on account of the short existence of the secondary spermatoc}i,es.

During the division process of the secondary spermatoc\'tes the idioectosome again breaks up into small pieces to be distributed in the CA'toplasm together with the liberated idiogranulomes. In this way repeating the phenomenon which occurred in the primary spermatocytes. In the telophase a new flowing together of the idiogranulomes and of the pieces of the idioectosome takes place near the new daughter-nuclei, forming the idiosomes of the spermatids.

The idiosome of the spermatids thus has the same formation as that of the secondary spermatocytes, consisting of a large idioectosome containing a great number of small idiogranulomes. The idiogranulomes are each enclosed in a small vacuole, the idiogranulotheca. Soon after the spermatids are formed the idiogranulomes and idiogranlothecae fuse together into larger and larger masses and vacuoles until finally they form a smgle big body, the idiosphaerosome surrounded by a large vacuole the idiosphaerotheca. The idiosphaerosome, as soon as formed, begins to secrete on its surface furthest from the nucleus a new substance, showing a different color reaction and a vacuolar consistency. This substance, the idiocalyptrosome, covers in a cap-like fashion the remains of the idiosphaerosome now called the idiocr^^ptosome. At this time the idioectosome forms a cap above the idiosphaerotheca and later on when the secretion of the idiocalj^ptrosome is about complete the idioectosome becomes detached as a separate bodj'. It then moves along the nuclear membrane and finally goes over into the remains of the protoplasm on the posterior pole and is eliminated.

In later stages of development the idiocrATDtosome comes to He in the form of a small cap on the anterior pole of the nucleus. The ealyptrosome grows into a very large body, alwaj-s showing a vacuolar consistency. In a still later stage, when the spermatid comes into con

414 AMERICAN ASSOCIATION OF ANATOMISTS

nection with a Sertoli cell, the idiocryptosome and the idiocalj^ptrosome are elongated in the form of two cones, the one enclosed by the other. During the final metamorphosis of the spermatid, the cryptosome again assvmies the form of a cap, covering the anterior part of the head of the sperm, while the calyptrosome loses its original vacuolar consistency and forms an outer larger cap covering over the inferior crjTDtosomal cap. The calyptrosome cap and part of the head of the spermatozoon are covered by the idiocalyptrotheca, which is the fully developed and transformed idiosphaerotheca.

A point of special mterest is the appearance of small granules, comparable to the idiogranulomes, m the nucleus of the germ cells during all stages of development. These we have termed caryogranulomes. Such granules are to be seen in the nucleus of the spermatogonia, the primary and secondary spermatocytes and the spermatids. The caryogranulomes are usually of small size about that of the smallest idiogranulomes, but in some stages they also seem to fuse together forming large granules. Their appearance, their color reactions and their tendency to fuse all show great similarity to the idiogranulomes, yet no genetic relation was directly observed. The caryogranulomes persist up to the time of the metamorphosis of the spermatid into the spermatozoon. Then they dissolve in the head of the sperm in the same way as does the chromatic material of the nucleus.

The new points brought out by this study are the following: First, The recognition of the idioendosome. Second, The description of the formation of the idiogranulomes through the breaking up of the idioendosome. Third, The persistence and the behavior of the idiogranulomes during the divisions of the primary and secondary spermatocytes. Fourth, The existence of the caryogranulomes and their development. Fifth, The exact manner of the formation of the calyptrosome and its vacuolar structure. Sixth, Certain peculiarities in the development of the cryptosome. Seventh, The double nature of the idiosome, consisting of an ectosomatic and an endosomatic substance, each having an independent development. Eighth, the role of the granulation of the idioendesome in serving to distribute the idioendosomatic substance during each division.

60. Some studies on the venom gland and its excretory duct in crotalus horridus. Vivian Strahm. Department of Anatomy, University of Kansas. (Introduced by J. Sundwall.)

The venom gland has been studied in several species of reptile. It has been reported by Leydig for Vipera berus (Schultze's Arch., Bd. 9, pp. 598-652); by Emery for Naja-haje (arch, fur Mik. Anat., Bd. 11, pp. 561-569); and by Holm for Heloderma suspectum (Anat. Anz., Bd. 13, pp. 80-85). Dr. Mitchell in his researches on the chemical and physical properties of the venom in Crotalus inchided a description of the venom apparatus (Smithsonian Contributions to Knowledge, vol. 12).

PROCEEDINGS 4 1 5

l'\)r my stiulics, IrvuIs from twtt six'cimcn.s of C'rotalus liorridiis were used. The reptiles were killed the latter part of January, after about three months captivity, and during their normal hibernation period.

The reptiles were cliloroformed and the heads severed about an inch behind the jaw. At once, the skin over the glands was reflected, one ghuul from each head removed, and cut into small pieces. Part of these l)its were fixed in neutral form-Zenker, the remainder in Bensley's aceticosmic, bichromate mixture. The heads with the remaining glands imdistm-betl wen» fixed in acetic Zenker, and run to 70 per cent. The small pieces were used for histological demonstrations; the heads, for gross structures and relationships.

The venom gland, a flat, smooth oval body, tapering at the anterior end to join the excretory duct, lies on the side of the head above the jaw and behind the eye. The posterior border extends 30 or 40 mm. behind the commissure of the lips. The anterior end lies just below and behind the eye. The gland is thus bounded by the skull behind the eye, the anterior and the middle temporal muscles, the external pterygoid nuiscle and the skin below and in front of the anterior temporal muscle. The gland is covered with a single layer of dense, fibrous connective tissue. This layer, which is continuous with the layer covering the duct, gives rise to three ligamentous bands and attachment to one.

The excretor\' duct throughout its course is imbedded in the fascia between the skin and the skull. As it leaves the gland, it is pointed forward and downward. In front of the eye it turns upward, rises to the level of the orbit and again turns sharply downward. It passes under the fossa, around the superior maxillary bone and ends at the base of the poison fang.

The outer surface of the duct is of uniform diameter throughout its length except for a distance of 35 or 40 mm., as it crosses the lateral side of the superior maxillary bone. Here it increases slightly in diameter. Dr. jMitchell noted this thickening and believed it was due to the presence of a sphincter muscle. I found a small mucous gland in this region surrounding the duct. The capsule of the duct also forms the capsule of the gland. No smooth muscle fibers occur in this region nor in any other portion of the duct or gland. This observation is not in accord with that of Dr. Mitchell.

Veno7n gland and excretory duct

The venom gland is made up of wide branching tubules lined with a single layer of columnar epithelium which is supported by a connective tissue framework. These tubules are short and straight. They are directed backward and outward from their openings. The lumina of the tubules are verv^ wide throughout. The number of tubules seen in any cross section increases rapidly on passijig from the anterior to the posterior portion of the gland.

416 AMERICAN ASSOCIATION OF ANATOMISTS

In the anterior third of the gland, on the medial side, a wide lumen forms a receptacle which serves to store collecting secretion. It is continuous with the lumen of the excretory duct.

The lumen of the duct is wide, unobstructed and roughly circular through the three flexures. After passing the third flexure, the walls of the duct are crowded and much folded by the enlargement of the mucous gland which appears in this region,

Leydig has noted a difference between active and resting glands in his specimens. The resting glands resemble the jaw glands while the active glands show very wide lumina and connective tissue trabeculae reduced to thin sheets.

Much the same conditions appear to exist in Crotalus. Of the two individuals studied, one shows glands with very wide lumina filled with coagulated secretion. The connective tissue layers are very thin, widening in many places sufficiently to admit of the passage of a blood vessel. The epithelium is simple low columnar with coarsely granular C3i;oplasm. Nuclei are round or shghtly oval, have scant chromatin and lie near the base of the cell.

The second individual shows connective tissue layers greatly thickened with the cells forming a loose network with many open spaces. The lumma of the tubules are greath' contracted and corrugated. The epithelial cells here are compressed and elongated and several layers are present as a result of the contracted lumina. These cells have dark, coarsely granular cytoplasm and a single nucleus. The nuclei are large, oval bodies with pale karyoplasm, scant chromatin and one or more distinct nucleoli. They lie at different levels though the greater number appears between the middle portion and the base of the cell. The lumina of the tubules and excretory duct contain masses of loose cells. Neutral and other granule stains fail to show the presence of secretion granules in either the active or the resting glands.

The epitheUum of the duct is identical with that of the gland tubules.

Interstitial granular cells

This resting gland is characterized by the presence of two types of interstitial cells, which appear only infrequently in the active gland. The first are large conspicuous cells with coarse granules, which are irregular in size and shape. The granules are usually present in such quantities as to completely obscure the nucleus. The granules are deeply stained in all acidophilic dyes utilized, such as picric acid, acid fuchsin, Congo red, erythrosin, eosin, methyl orange, orange G., aniline blue, acid green and acid violet. It was found that these granules stain more quickly when the staining solution is placed in the thermostat at 60°C. The acid radicle in such stains as Wright's blood stain and Bensley's neutral gentian affects the granules. In the former, they are stained brilliantl}^ red, while in the latter they are stained orange. They are also stained deeply in iron alum haematoxylin.

PROCEEDINGS 417

'Phf lUK-lei of tliese cells when seoii arc nnuul or slightly oval and li(> luvir the cell ineinhrane. A chromatin network is seen.

Tiiese cells are irregularly distrilnited throughout the coiniective tissue sei)ta. They are fcnuid most numeroush' stu'rounding tiie blood vessels. They are seen also between the epithelial cells of the tubules and in some instances, processes project out into the lumen. The majority, however, are seen in close i)roximity to the basement membrane at the basal ends of the cells. They are conspicuous between the ei)ithelial cells of the excretory duct, and a few appear in the connective tissue of the mucous gland. These cells have also been observed in alnuulance in the connective tissue, especially surrounding cai)illaries, and between the epithelial cells of the vagina dentis.

This jncture strongly suggests that these more or less acidophilic cells are passing from the blood vessels through the connective tissue se])ta and betAveen the epithelial cells into the lumen of the gland, and consequently contributing to the elements of the secretion.

The second type of interstitial granular cells, which appear in much smaller numbers than the first type, are mast-cells and show the structure and staining reactions characteristic of mast cells.

The mucous gland

The mucous gland is entirely distinct and separate from the venom gland. It is, like the venom gland, made up of short, simple columnar epithelium. The excretory duct of the venom gland passes through the center of this gland, and the gland tubules, which lie next to it, empty mto it. The mam excretorj- duct of the mucous gland itself is, however, a crescent shaped duct and lies on the under side of the gland mass next to the capsule. Near the external opening which connects with the fang these two ducts join and empty through a common onfice.

The restmg cells are tall columnar, with granular cytoplasm. The nuclei are oval with well defined nucleoli and scant chromatin. In some active cells only a small globule of mucin appears near the base of the cell. Agam, the whole cell is loaded with secretion.

These secretion laden cells are scattered in the upper end of the tubule but increase rapidly in numbers as the outlet of the tubule is reached. In some tubules, near the outlet, every cell is filled with mucin.

### Histogenesis of the otic capsule

Streeter GL. Histogenesis of the otic capsule. (1917) Anat. Rec. 11: 417-418.

61. Histogenesis of the otic capsule. (Lantern.) G. L. Streeter,

Carnegie Laboratory of Embryology, Baltimore.

The cartilaginous capsule of the ear is a favorable place for studying the histological features of the growth of cartilage and its associated tissues, particularh' because of two reasons; in the first place, there are, on account of the mtricacy of form of the labyrinth, many kinds of cartilaginous changes found there that are necessary to accommodate its groAvth, including the deposit of new tissue and the removal of old tissue; and in the second place the topography is so well marked by known landmarks that all of these changes as well as the location and direction of growth can be easily followed. From such a study one is forced to conclude that the tissues of the otic capsule are capable both of differentiation and dedifferentiation throughout a considerable period of their development. This progressive and retrogressive adaptabihty of the cartilaginous tissue makes possible those changes that are necessary in the growth and alteration in form of the labyrinth.

In the earlier stages the precartilage tissue abuts directly against the epithelial labyrinth. Subsequent!}^ the periotic reticulum, beginning along the central borders of the canals, becomes established and spreads at the expense of the temporary precartilage thereby forming a crescentic shaped area of reticulum entirely inclosing the membranous canal. The invasion of the reticulum into the surrounding area of precartilage is brought about, at le/ist in the later stages, by a dedifferentiation of the latter into the former.

At the same time that the precartilage is reverting into reticulum the inner margin of the cartilage that surrounds the canal is dedifferentiated into precartilage, so that a new and more peripheral area of precartilage becomes established as the old area disappears. In this way the margin of the true cartilage gradually recedes from the epithehal canal.

In human embryos 30 mm. long the cartilaginous labyrinth has attained approximately the adult form. Its subsequent development consists primarily of an increase in size to accommodate the growing membranous labyrinth. However, if one compares the superior canal of an 80 mm. fetus with that of a 30 mm. fetus it will be found that the canal has doubled its diameter and has trebled its length and furthermore its linear curvature corresponds to an arc with a longer radius. In reality therefore the developing cartilaginous labyrinth is continuously undergoing considerable changes both in size and form. The enlargement of the cartilaginous canals and their alterations in form and position involves both the excavation of cartilage and also the laying do%vn of new cartilage, the excavation being accomplished by its dedifferentiation into precartilage and reticulum, and the new cartilage being built up, through a precartilage stage, from the periotic reticular tissue. Throughout the entire period of growth of the cartilagenous canals the elements of this continual transformation exist along their margin. The margin during this period is in a state of temporary equilibrium and is capable of advancing or receding as the conditions determine.

The first and relatively the major part of the hollo wing-out of the cartilag^ious canals is complete before the perichondrium makes its appearance. The perichondrium is formed as a membrane-like condensation of the periotic reticulum which can be first recognized in fetuses of about 70 mm. CR length. In its histogenesis it is analogous to the membrana propria of the epithelial canals.

### The genesis of the membrana tectoria and its anatomical substratum

62. The genesis of the membrana tectoria and its anatomical substratum.

O. Van deu Stricht (By invitation), University of Ghent, Fellow

in Cytolofiy of the Western Reserve University.

In the present paper I should like to present some preliminary results of my investigations upon the histogenesis of the superficial epithelium of the crista spiralis, the sulcus spiralis and of the membrana tectoria of the cochlea.

I had at my disijosal cochleae from pig embryos of GO.O, 93.5, 127.0, 137.0, 150.0 and 190.0 nun., from a new born dog and from the following adult manunals: bat, dog, rat and mouse, fixed and stained in various ways.

Crista spiralis. The youngest pig embryos exhibit at the surface of the future crista spiralis and in the greater epithelial ridge, an epithelial layer rather thick, described by many authors; it consists of multiple rows of nuclei, the cells of which trasverse its whole breadth, but the nuclei are arranged at various heights. In reality the cells represent a simple cohunnar epithelium. The apices of the cells, well visible on tangential sections, figure a very elegant mosaic or pavement, the numerous fields of which are separated by the terminal bars or closing Lines (Schlussleisten) and contain each of them, two central corpuscles or diplosomes.

Many authors have shown how the subjacent mesenchyme gradually penetrates between the different epithelial cells of the crista spiraHs in more advanced embryos and form a special layer partially epithelial and partially connective. N. Van der Stricht demonstrated that despite this extension the primitive superficial mosaic remains intact, even in the adult organ. Never do the connective tissue elements traverse the system of terminal bars.

In order to get a true picture of the connections between the epithelium and the proliferating subjacent tissue sections tangential to the surface are needed and must be compared with transverse sections. Series of preparations show the gradual thickening of the intermediate ground substance, but in such a manner that rarely is an isolated epithelium cell completely surrounded by the connective tissue. On the contrary, the epithelial cells are arranged and pressed together so that they constitute a sj^stem of bands (or sheets) ramified and anastomosed at the surface of the smallest most axial segment of the vestibular lip, near the attachment of Reissner's membrane. The bands are more or less parallel at the surface of a more lateral and larger segment which represents the substratum of the future Buschke's teeth, the axis of the bands being directed from the axial region towards the free edge of the lip. This axis exactly represents the future furrow, separating two developed "teeth." The spaces between the parallel cellular sheets, occupied by the proliferating connective tissue rapidly enlarge, reach their definitive extension and constitute the 'teeth.'

A slighth' oblique tangential section at the surface of the crista displays a series of interesting figures, varying according to the depth cut by the knife. From the surface to the depth one sees:

(a) The superficial epithelial mosaic.

(b) Cj-toplasmic sheets, parallel on the lateral and anastomosed on the axial segment of the vestibular lip.

(c) A little deeper the protoplasmic bands become nucleated and their nuclei are pressed together in such a way that often no trace of CA-toplasm is perceptible between them.

(b) The vascular tissue, areolar in the young, embryonic organ and fibrous in the adult. The existence of such cellular bands suggests two special questions.

1. At first the possibility of the genesis by fusion of epithelial cells, primitively separated, of a vast syncytium, unexampled in other organs: formed by cells entirely fused along their length but absolutely distinct at their apices where I find, even in the adult persisting distinct centrosomes. Comparing the transverse with the tangential sections already mentioned I have no doubt about the presence of such a syncytial formation at least at some places of the organ. Nevertheless great caution is to be observed before drawing final conclusions for besides very demonstrative figures one sometimes observes sites where primitive intercellular boundaries are persisting partially or even entirely.

2. A second question seems to me at the present time more difficult to solve. Do these multiple nuclei so numerous in this syncytium represent those of the orginally separated cells or do they partially result from the mitosis or amitosis of these nuclei? From the stages of 93.5 mm. (and probably earlier) no mitosis can be found. Should new nuclei appear later it must be admitted that they are formed by a process of direct nuclear division. Although my preparations exhibit some few signs of nuclear amitosis, I am not able to give a final answer to this question.

The structure of the greater and the lesser epithelial ridges is loiown. I will only point out that the greater thickening from rather early stages shows three more or less distmct segments:

1. A lateral sensorial; very small, beside the lesser ridge; in which appear the inner auditory and inner supporting cells;

2. An axial; the columnar cells of which possess a clear protoplasm and represent the future elements of the sulcus spiralis;

3. An intermediary; between the two preceding; the largest, consisting of prismatic cells, the protoplasm of which is rather dense and dark. The last mentioned cells will provide the epithelial elements covering the region of the foramina nervina.

Tangential sections at the surface of the greater ridge exhibit a very regular mosaic formed by the apices of the columnar cells with their contained diplosomes and the terminal bars separating the cellular fields.

I will not speak of the structure of the lesser ridge studied by N. Van der Stricht who gives an accurate description of its constituents and superficial mosaic.

MiDibrana Ucloria. 1 luive clcscrilx-d tlic jicncratiii};- suhstnitliin of tiiis organ, the crista spiralis and tho groat opitholial ridgo and chiefly their superficial mosaic. The last mentioned varies somewhat in aspect according to the site. On the surface of the V(>sti(;ular lip its fieUis are large and the closing lines an; thin hut a little thicker than in the areas covered by indifferent ('i)ithelhun. (e.g., area of Henle's cells). In the sensorial sections (organ of ('orti), tluy exactly represent the apices of the auditoiv and sustentacular cells and the tenninal bars become very thick (N. Van dcr Stri(!ht) although they are thin at the l)eginning of the genesis of the membrana tectoria.

On the surface of the greater ridge the fields are of equal size in the earliest stages but much smaller than on the vestibular lip and the terminal bars are always relatively thick. When the axial segment which first becomes detached from the membrane and inactive (Hardesty) may be distinguished from the intermediate, the fields on its surface are much smaller than in the latter segment.

This system of lines and polygonal fields interferes with the genesis of the membrane tectoria. In previous papers mj^ pupils and I have referred to the paramount importance of the closing lines in the formation of the membi-ana reticularis of the auditory epithelium, of the membrana limitans olfactoria and of the membrana limitans externa of the retina. I will add here that the zona pellucida of the ovum of dog, bat and other mammals according to my preparations has a similar origin.

]\Iy preparations of the duct of the cochlea show that the terminal bars, that is the superficial intercellular cement, condensed, becoming denser and more intensively staining, take part in the genesis of the membrana tectoria in such a manner that they generate the dense part whereas the apices of the cells the more fluid part of the membrana tectoria. But the chemical composition of the dense part is quite different from that of the bars for they take up different stains.

At the surface of its matrix the membrane is formed by a S3^stem of delicate lamellae the sections of which represent a reticulum of lines, filaments, in direct continuity with the bars Ijying over and reproducing exactly in a tangenital section the system of the closing lines. The spaces between these lamellae are filled by a paler substance which seems to be elaborated by the superficial protoplasm of each cell always provided with the centrosome. This fluid like substance and the lamellae may suffer more or less shrinkage and distortion provoked by the fixing agents so that indeed the meshes become smaller or larger and irregular.

In very successful preparations I perceive a similar but irregular network in the sections at the surface of the sensorial epithelium of the organ of Corti. I am thus induced to conclude that the membrane is foiined by the same histogenetic process.

This network, first origin of the membrana tectoria, has been described by other authors but erroneously considered as produced by the cytoplasmic summits of the cells (Hardesty, Coyne and Cannieu, Prentiss). A series of preparations fixed by the new Cajal method shows the tenninal bars stained black and the lamellae of the membrane colored in the same manner.

For the wealth of bat material the study of which in the present connection has been veiy useful to me I am indebted to the generous interest of ]\Ir. W. G. Marshall.

### Some anatomical features of the bird hand

63. Some anatomical features of the bird hand. (Lantern.) R. M. Strong, Anatomical Laboratories, Vanderbilt University Medical School.

This work is part of a comparative studj^ of bird anatomy giving especial attention to the Tubinares. The anatomy of the bird hand has been found to need more elaborate treatment than appears in the literature. Fiirbringer's exhaustive work dealt especially with the brachial plexus and the muscles of the breast, shoulder, and upper arm.

On this occasion I shall present some details in the anatomy of the albatross hand, with the aid of lantern slides. The mechanism for controlling the flight feathers will be described.

What I gave called the palmar aponeurosis is a part of this apparatus and has not been described in the literature, for birds, to my knowledge. It is continuous at the radial border of the wrist with a dorsal aponeurosis. The palmar aponeurosis merges with fascias of the upper ann, and it gives off slips to the quills of the distal secondaries and most of the primary feathers. It is also attached at points to the phalanges and to the carpo-metacarpus. Further details are not' given here as illustrations are needed. This work is planned for publication elsewhere later.

### On the nature of basal striations in salivary ducts

64. On the nature of basal striations in salivary ducts. John Sundwall,

Department of Anatomy, University of Kansas.

The presence of basal striations in the salivary ducts, which are characteristic of these ducts, was early noted by Pflliger, R. Heidenhain and Miiller, and have been described also by Merkel, Zerner, R. Krause and others, in various annuals. That these ducts are concerned in salivaiy secretion was maintained by Zerner, Eckhard, Milawsky and Smirnow, Solger, and R. Krause among others.

The development of new technique, particularly that for the specific demonstration of mitochondria, has made it unperative that the various glands be made the object of reinvestigation. Much investigation is now going on regarding the presence of and the role that mitochondria phw in cell cytoplasm. Meves, Bensley, Cowdry, Duesberg, Champy, Hoven, and Arnold, among others, have contributed much to our knowledge of these structures. Whether mitochondria are directly concerned in the production of secretion substances still remains within the field of speculation, notwithstanding that Hoven, Champy, and Arnold hold that such is their special function in gland cells.

Bcnslej' has observed that basal striations in the acinous cells of the pancreas are primarily due to mitochondrial filaments. I also have observed that Avlien striations were seen in the lachrymal jfland, they were dejiendcnt upon tlie preservation of these structures. These observations naturally sufjjsested that the basal striation in the salivaiy duets niif>;ht i)elon,i!; to the same cate{j;ory.

With a view of dotormininjj; the nature of basal striations in the salivary ihu^ts, I applied Bensley's acetic osmic bichromate, anilin fuchsin, methyl f2;reen teclmi(iue in the preiniration of salivary p;lands taken from the ojwssum. The mitochondria appear as rods or filaments in the base of the cell, occupying approxhnately the basal one-third of the cell and tenninatinp; as a rule at the base of the nucleus. They are arranged perpendicularly to the long axis of the cell and appear as basal rays in cross sections of the duct. Collectively the basal striation formed by the mitochondria is of the same length in the various cells that make up the salivary ducts. So that in cross sections of these ducts an even deep red fuchsinophilic border is prominent. The border is regular in width and occupies the outer third of the radii of the ducts.

The individual mitochondria making up the basal striations may be generally classified into three groups:

1. Long filaments or rods which extend the entire perpendicular length of the basal striations. As a rule these tend to taper off slightly at both ends and are thickest in the middle. However, many are approximately the same width throughout and temiinate bluntly.

2. Short bacillus-like rods of irregular lengths an-anged in rows.

3. ]Minute granules or coccus-like mitochondria arranged in rows or seen more frequently associated with the short rod forms.

Groups 1 and 2 are the types most frequently seen making up the striations. The mitochondrial rods are indeed numerous and are very closely appressed so that under low power observation the characteristic more or less red homogeneous border is noted, especially when sections of 10 or more fx are examined. Observation of sections under 5 /j by means of high power lenses — oil emersion, reveals the rows of mitochondria and individual mitochondria making up these perpendicular striations. The intermitochondrial basal cytoplasm is seen as very narrow strips between the mitochondrial rods. This undifferentiated cytoplasm appears either unstained or as slightly greenish stained clefts between the mitochondrial rods.

In the middle and proximal zones of the cell, the mitochondria are present in the form of minute granules or cocci irregularly distributed throughout the cytoplasm. In some cells they are found much more numerously than in others. Two cells may be seen side by side in one of which they are abundantly present, while in the other only a few are seen.

When the animal has been subject to pilocarpinization, as manifested by the extraordinary secretion of saliva, an interesting phenomenon is observed in the salivary ducts. The mitochondrial rods making up the basal striations show a tendency to lose their perpendicular arrangement and break up into the granular or coccus-hke forms. In some ducts all evidences of the perpendicular arrangement of the mitochonedria have disappeared and the mitochondrial rods have disintegrated into granules. Granules of mitochondria, irregularh^ distributed, now occupy the basal portion of the cell as well as the proxunal portion. The granules are more congregated than in the nonstimulated gland, so appear more numerous, and a granule-free zone of cytoplasm is noted between the mitochondria and the cell membrane.

This description of the stimulated glands represents the extreme picture of the effects of pilocarpin. In many of the stimulated glands the change in the arrangement of the mitochondria is not so marked, and in some the basal striations are only partially disintegrated.

Perhaps the chief value of these observations is the detennination of the nature of basal striation of salivary ducts. That this is formed by two elements is apparent — mitochondrial rods perpendicularly arranged (the so-called filaments of Michaelis and Altmann) and an intennitochondrial undifferentiated cytoplasm (the basal filaments of Solger or the ergastoplasmic filaments of Prenant, Garnier, and Bouin) . The latter are especially prominent when the mitochondrial elements have been destroyed by the fixation fluids. Bensley differentiated these two elements in his studies of the pancreas.

These observations also explain the inconstant results obtained by investigators and teachers of histology in respect to the basal striations of salivary ducts and other gland ducts. Naturally when the mitochondria are destroyed by the action of certain fixing fluids, such as those containing considerable quantities of acetic acid, alcohol, corrosive sublimate, etc., the basal striations are not distinctly conserved and consequently the sections are of little value for specific teaching purposes.

### Early development of the cartilaginous ethmoidal skeleton in cat

65. Early development of the cartilaginous ethmoidal skeleton in cat. R.

J. Terry, Washington University Medical School, St. Louis.

The first evidence of chondrification in the ethmoidal region was observed in embryos of 12 mm. (crown-rump), prepared by van Wijhe's method; in the ventral part of the future septum nasi are two deeply stained parallel streaks separated across the mid-plane by a lightly stained interval, extending forward from the travecular plate. In embryos of 5 mm. the septum is a single, high median cartilage dividing dorsally into a pair of processes, the parieto-tectal cartilages, which arch over the anterior part of the nasal cavity of each side. A second region of chondrification was found next to that diverticulum of the cavum nasi which later gives rise to the frontal air sinuses. Here a pair of paranasal cartilages is formed independently of the parietotectal cartilages behind which they lie. A third contribution to the early ethmoidal skeleton appears in embryos of 17 mm. as a pair of small cartilaginous plates, laminae antorbitales, at the very back of the cavum nasi on either side of the nasal septum. At this stage there is no cartilaginous et>Tnoidal floor. A floor is first indicated in embryos of about 20 mm. b}- that portion of the paraseptal cartilage which stands in relation to Jaculjson's oif^an and !)>• the antt'iior transverse lamina. The posterior portion of the pai-aseptal and the nasopalatine cartilages appear relatively late. All of these cartilaj^inous centers, with the exception of the parieto-tectaland possibly the anterior transverse lamina and nasopalatine cartilafucs arise independently of one another. Nearly complete fusion takes place secondaiily between tlie a<ljacent edges of the parieto-tectal and paranasal cartilages resulting in the formation of the crista semicircularis within the nose and leaving a small space, the foramen epiphaniale for the passage of the nasociliaiy nerve. Fusion of the neigiiboring edges of the paranasal andantorljital plate also occurs; at their junction internally, a small ridge is found where ethmoturbin al I is later developed. Union of the antorbital plate with the septum nasi completes the posterior transverse lamina and posterior cupola of the nasal capsule. The anterior transverse lamina is, at an early stage, continuous with the parieto-tectal cartilage, if indeed it is not originally an extension of it.

### The development of the hemal channels in the central nervous system of the albino rat

66. The development of the hemal channels in the central nervous system of the albino rat. Frederick Tilxey and Louis Casama.'or, Departments of Anatomy and Neurology, Columbia University. The manner in which the central nervous system acquires its vascular channels is a process presenting several distinctive features. In the first place the neuraxis is devoid of mesench^anal elements for a relatively long period in its ontogenesis. There is a boundary line between mesench^aiie and neural ectodemi during this period which is sharp and unmistakable, namely the external limiting membrane. In consequence of this separation of mesench^ine from neural ectoderm the neuraxis is lacking for a considerable time in the angiogenetic elements necessary to vascular formation. How, therefore, are these angiogenetic derivatives of the mesenchyme acquii-ed by the central nervous system and what, after their acquisition, are the details of the vasofactive process within the neural tube?

The material selected in the investigation of this problem was the albino rat from somite stages to teiTn and later. As the injection method has been shown to be inadequate for the demonstration of conditions in the earlj^ development of the vascular system, this method was not employed. The stud}' was based upon serial sections cut at 5 /x after fixation in Bouin's fluid and stained according to the Mann-Regan method. Reconstructions were also made at magnifications varsdng from 300 to 1100 diameters.

In the hemal vascularization of the central nervous sj'stem of the albino rat, it is possible to discern two more or less distinct phases, the one characterized b}' the exuberant formation of blood vessels, the other, making its appearance somewhat later, characterized by the selection of the definitive vascular fines and the gradual disappearance of the excess in the earlier, exuberant formation of hemal channels. The dominant feature in the first of these phases is an angiogenetic process while the second phase, although still angiogenetic to a marked degree, is rendered conspicuous b}^ an ajigiolytic process. During the phase of vascularization characterized b.y angiogenesis several stages may be distinguished. For example, in the early somite period of development, not only the neuraxis but its iimnediate mesenchymal environment are lacking in any evidence of vessel formation. From this absence of all angiogenetic activity, the conditions may be described as the prevascular stage. This stage is limited in its duration for embryos as 3'oung as 2 to 3 mm. manifest a decided angiogenetic activity in the mesenchjTue immediately surrounding the neural tube. The mesenchjTiial cells are collected in scattered groups which border upon the external limiting membrane. Near the center of many of these groups erythroblasts in the several stages of their evolution are seen. In many instances the hemopoietic mesenchymal elements are still in part attached to the surrounding mass of mesenchjTue. In other instances the er\i;hroblasts lie wholly detached within well-formed endothelial spaces. During this stage it is possible to recognize in the perineural mesenchjaiie blood islands and isolated endothelial spaces containing blood cells. Thus the vasofactive process, giving rise to isolated endothelial spaces, goes hand in hand with the formation of perineural blood islands from which latter the endothelial spaces and erythroblasts are derived. It is essential to note that at this particular period there is no evidence of vascularization in the neural tube and for this reason the conditions may properly be described as the stage of perineural angiogenesis.

In embryos of 3 to 4 mm. length a notable and critical change occurs. The mesench^ane assumes a definite attitude with reference to the nervous system in that many mesenchymal cells begin to make their way through the external limiting membrane into the neural tube. This wandering into the neuraxis of mesench^nne cells constitutes a veritable mesenchjonal invasion. Cells of the mesenchyme, singly and in strings of two or three, make their way from the perineural mesenchyme into the neural ectoderm. These invading cells are not sprouts from the already formed endothelial spaces of the perineural mesenchyme. In many instances they proceed inward from regions in which neither blood islands nor isolated endothelial spaces are to be found. They may be seen either partially within the nervous system or their general course after entrance may be traced as strands extending toward the central canal. Coincident with this invasion there is a distinct advance in the perineural angiogenesis, for during this period the blood islands have given place to many isolated endothelial spaces, some of which have already become confluent with others in their immediate vicinity. Inasmuch as the outstanding feature at this time of development is the invasion of the neural tube by mesenchymal cells, this condition may be described as the stage of ynesenchymal invasion.

Invasion of the tube by cells from the mesenchyme does not cease at the stage just mentioned, but continues until a much later period. Its fundamental importance in this early period lies in the fact that the neuraxis thus acquires the angiogenetic elements which serve as the anhij^o for its vascular system. In eiuhrycjs from 4 to 5.5 mm. in leiif^th the mesenchymal cells in the neural tul)e give evidence that the basofactive process is well under way. Numerous areas are observed which have all the appearances of the perineural blood islands, namely clusters of cells, some of which seem to l)e assuming endothelial characters while others, more central in position, present many features wiiich ally them with the erythroblasts seen elsewhere. These latter cells are in many instances still attached to the mesench>anal clusters of which they fonn a part. In other areas the vasofactive process has advanced a step further and in consequence not a few isolated endothelial spaces containino; free erythroblasts have made their appearance. The position occupied by the blood islands and isolated endothelial spaces within the neuraxis is noteworthy. At this time the neural ectoderm, in the myelencephalic region is eight to ten cells deep. The spaces and blood islands are situated near the central canal, being separated from it usually by two and never b}^ more than three layers of cells. This position removes the spaces and blood islands within the neural tube as far as possible from the perineural endothelial spaces. The latter, by a process of confluence, have fomied a well defined perineural plexus which, however, has no connection as yet with any of the endothelial spaces within the tube. The most conspicuous feature in embryos of this size is the inception of vascular fonnation within the neuraxis and the conditions may be described as the stage of entoyieural angiogenesis.

In embryos from 6 to 7 mm. in length the number of entoneural endothelial spaces has increased and many of them have become confiuent to form a relatively rich ento-neural plexus. As this plexus first takes fomi it presents no connection with the perineural plexus. Gradually, however, these two plexuses establish communication with each by the process of confluence. At first these communications are few in number but as the entoneural plexus increases in complexity the communications become more diffuse and more numerous until the period of full efflorescence in the vasofactive process is reached at 8.5 mm. when ahnost every section (cut at 5 n) shows two or more connections between the perineural and entoneural plexuses. In embryos of 8.5 mm. the vasofactive process is at its height since the entoneural plexus is then more extensive, the size of its hemal channels larger and its connections with the perinem-al plexus more numerous than at any other period. From this time on the entoneural plexus gradually becomes reduced in richness as well as in the size of its channels.

In embryos from 9 to 24 mm. vascular formation passes into it by actual angiolysis. This latter process confines itself to the apparen-t excess of the early exuberant entoneural plexus. The prominent features of the angiolytic phase may be summarized as follows: (a) The partial solution of the rich entoneural plexus during which only the more radially disposed channels withstand the retrograde process and are thus selected to participate in the permanent vascular pattern, (b) the deflorescent channels revert to mesenchyme, '(c) the deposition of such mesenchymal cells among the neural elements of the central nervous system and (d) the breaking down of many erythroblasts excluded from the circulation. From our observations we have drawn the following conclusions;

1. The development of the hemal channels of the central nervous system depends upon the foniiation of two separate plexuses, i.e., the perineural plexus and the entoneural plexus.

2. The perineural plexus is developed from the perineural mesenchyme first as a series of blood islands which give rise to isolated endothelial spaces, the latter in turn becoming confluent to form a plexus.

3. The entoneural plexus is developed from derivatives of the perineural mesench\ane, which, in the early stages, invade the neural tube, there giving rise to blood islands. From these blood islands are fonned isolated endothelial spaces which become confluent to form a plexus.

4. The perineural and entoneural plexuses establish communication with each other by a process of confluence.

5. The entoneural plexus attains its highest development in the relatively early stages and then undergoes deflorescence. As a result of this latter process a considerable portion of the primitive vascular tissue within the neural tube reverts to mesenchyme cells which remain as constituents of the neuraxis.

### On the pineal region in human embryos

67. On the pineal region in human embryos. John Warren, Harvard Medical School.

The object of this communication is to call attention to three special features in the development of the pineal region in human embryos.

1. The primary arches in the roofoftheforebrain. In an embrj^o of 10 mm. all these arches are clearly differentiated. A low velum separates a well marked paraphysal arch from a relatively short and thick-walled post velar arch. The epiphysal arch is sharply defined with also rather thick walls and is succeeded by a relatively long pars intercalaris. This part of the brain roof forms a low arch in the posterior end of which the posterior commissure can be seen. The commissure seems to appear in this portion of the forebrain before invading the roof of the midbrain. In an embryo of 15 mm, the arches are more fully developed. A slight median thickening in the paraphysal arch marks the anlage of the paraphysis. The posterior commissure now occupies a larger part of the pars intercalaris and has developed backward into the midbrain. The primary arches as first described by Professor Minot in Acanthias can therefore be demonstrated in human embryos and in addition the pars intercalaris forms an arch of relatively great length as compared with its appearance in lower vertebrates.

2. Paraphysis. In the embryos of the Harvard Collection the earliest trace of the paraphysis can be noted as a slight thickening in the paraphysal arch in the embryo of 15 mm. mentioned above and in two others of 16 mm. No traces could be found however in any of the embryos of from 17 to 22 mm. with the possil)le exception of one of 19 mm. In an embryo of 23 mm. sagittal series it appears as a tiny hollow elevation similar to its earliest form in lower vertebrates. In another embryo of 2."^ mm. transverse series it is a little larj^er and in one of 25 mm. it forms a j;oocl sizeil oval outi;i()\vth with a cavity opening into that of the l)rain in front of the velum. Above this stage it could be demonstrated in an embryo of 81 nun. as a mere tiny elevation and also in one of 30 mm. where there was a slight cavity attached by a solid stalk to the brain wall. The oldest embryo in which any trace could be made out was one of 44.3 mm. where a tiny conical elevation in the paraph>'sal arch was all that could be seen. The paraphysis therefore does exist in certain human eml)ryos but it is a rudimentary and inconstant structure. A short l)ut well defined paraphysal arch could be followed in all the embryos studied.

3. Post velar arch. A complicated prolongation of the anterior end of this arch just behind the velum forms a striking feature in many embryos. This outgrowth appears either as a median projection or as a bilateral fonnation on either side of the median line coming into intimate relation with the vessels over the brain roof. As the outgrowth becomes more complicated, tubules are given off in a rather bewildering manner, which may become detached and appear as blind vesicles bmied in the midst of this tubular fonnation. The paraphysis is more or less covered by this projection which overhangs to a large extent the paraphysal arch. The fonnation begins in embryos of 23 mm. as a simple median outgrowth. In an emryo of 25 mm. the outgrowth is a double one. In an embryo of 31 mm. two closed vesicles can be seen on either side of the middle arch, but still in contact with the brain wall. In an embryo of 36 and 44 mm. where the formation is extremely complex, one or more of these vesicles are found completely detached from the brain. Their walls are usually thinner than those of the other tubules among which they lie. Attention is called to this formation especially to the vesicular portion as it is such a striking feature in all the embryos from 23 mm. up to 44 mm.

68. Progressive movements in decerebrate kittens. Lewis H. Weed, Anatomical Laboratory, Johns Hopkins University. In a series of forty kittens subjected to decerebration, different reactions of a rhythmic character were obtained. In all, an essentially similar ablation was performed with removal of cerebral hemispheres and basal ganglia. The brain-stem was cut through just anterior to the superior corpora quadrigemina, sloping forward in the line of the bony tentorium. The kittens continued to breathe spontaneously and as soon as the anesthesia had passed away, showed active reflexes and rhj^thmic progressive movements.

Differing from the reactions of adult animals, these decerebrate kittens did not exhibit an invariable extensor rigidity. All, however, gave typical scratch reflexes; they responded to both dorsal and ventral excitation, and reacted to trauma to the tail. In general the animals after decerebration were much more active than are the customary adult preparations. Practically all of these animals showed rhythmic progressive movements; they seem to fall into two classes in this reaction, the groups being determined by the length of time the beats continued.

Twelve of these decerebrate kittens, on recovery from the anesthesia, showed, when suspended, typical prolonged walking movements of all four legs. These rhythmic progressive beats could be initiated by almost any stimulation and were continued for varying periods. In some of the animals the movements, after the initial excitation, were prolonged uninterruptedly until exhaustion occurred; in others the movements ceased after a few minutes. In the more active kittens, a smooth surface placed beneath the four feet of the animal occasioned continued progressive movements.

Separated from this group of kittens which showed prolonged rhythmic movements of progression are the other kittens of the series. These on appropriate excitation, made rhythmic movements lasting not over thirty seconds after the cessation of the stimulus. While the differentiation of the two groups is wholly arbitrary, the reactions of the animals are so different in this temporal relationship that the two classes seem indicated.

The reactions of the group of decerebrate kittens showing prolonged progressive beats are those of an adult cat in which only the cerebral hemispheres have been removed. The animals of the second group, however, are somewhat more active than are adult decerebrate cats. In the kittens of the first group, an extensor rigidity was absent in all but three; in these, a questionable rigidity occurred soon after the ablation and was quickly abolished by the onset of the rhythmic beats of all four legs. As the tendency to progression decreased, extensor rigidities became more pronounced, affecting in some only the forelegs. A reciprocal relationship between the prolonged progressive movements and the extensor rigidities has been indicated by these experiments.

The kittens which when decerebrated showed the phenomenon of prolonged progression varied in age from one hour to sixteen days (in length, from 150 to 220 mm.). Kittens of approximately the same age from the same litters in general gave similar reactions but considerable variation could be noticed. In any one litter, as the kittens grew older, the tendency toward prolonged progression grew less and the likelihood of extensor rigidities grew greater. Age alone, however, is, as judged by these observations, not an index of the reactions.

69. The coronary vessels of the heart of the 20-mm. pig embryo. (Lantern.)

Thomas Foster Wheeldon, Harvard Medical School. (Introduced

by Franklin P. Johnson.)

The present article is an extract from a larger paper entitled "The heart of the 20 mm. pig embryo." It is a study which was carried to completion at the University of Missouri under the direction of Professor Johnson.

Arteriae coronariae. As in the adult, the coronary arteries are two in number, a right and a left. They appear as small rounded vessels with relatively thick walls. They spring fioin the superior aortic sinus and the left inferior aortic sinus res|)ectivoly, and are distributed almost entirely to the heart, hut a few small branches are given off to the roots of the great vessels.

The left coronary artery gives off a l)ranch of particular interest which accompani(>s the atrio-ventricular bundle through the moderator band, and breaks up into capillaries in the right wall of the right ventricle.

Attention must be called, also, to a plexus to be found on the diaphragmatic surface of the heart. This plexus is fonned by the anastomosis of the terminal branches of the right and left coronary arteries.

Sinuf! coronarius et venae cordis. Owang to the fact that the venal hemi-azygos of the pig opens into the coronary sinus, the relations of tlie entering cardiac veins are somewhat different from those found in man. There is, strictly speaking, no definite coronary sinus, fo ■ the cardiac veins all enter the terminal portion of the vena hemi-azj^gos. This portion of the vena hemi-azygos, which lies in the inferior portion of the coronary sulcus, is somewhat enlarged, and represents the coronarj-- sinus of the human heart. In my description, I have used the temi 'coronary sinus' to designate the tenninal portion of the vena hemi-azygos.

The -coronary- sinus receives five large veins. They are from left to right; the vena hemi-azA'gos, the great cardiac vein, the inferior cardiac vein of the left ventricle, the middle cardiac vein, and the small cardiac vein. In addition, it receives the smaller anterior cardiac veins. It empties into the sinus venosus just ventral to the inferior vena cava. There is as yet no subdivision of the valve of the venous sinus to mark out the Thebesian valve, which in the adult pig guards the orifice of the coronary sinus.

The small cardiac vein, instead of arising as a single trunk as it does in the adult pig, arises as two definite vessels, one of which receives branches from the right atrium, while the other collects blood from the right wall of the right ventricle.

Little seems to have been published on the development or observation 'of the coronary vessels in embryos younger than 20 mm., and it was quite unexpected to find so complete a system in a pig of this size.

70. Study of a human spina bifida monster with encephaloceles and other

abnormalities. Theodora Wheeler, Carnegie Institute of Em bryologj^ Baltimore.

The specimen was accompanied by only a meager clinical history; the child was illegitimate, was bom spontaneously at full term, and lived onh' a few hours.

The external form is fairly typical of a class often called inencephalic monsters. There is extreme dorsal flexion and shortening of the trunk. The head is drawn back close to the sacral region. At the back of the head three encephaloceles project. The shoulders he very high, pushed close to the cheeks. The neck is obliterated, the chin and chest tying in one plane. The ears are distorted; the anthelix is pushed out so that it is very prominent; the tragus is shifted medially and upwards so that it lies opposite the concha; the antitragus lies below the tragus pushed against the cheek; the lobule is narrowed. These deformities are evidently caused by pressure on the external parts of the ear and twisting of these parts during their early development by the backward bent head and the shoulders which He close to it on each side. In the region of the encephaloceles there is some normal scalp and also various cutaneous anomalies. On the lower border of the left sac there is a rounded bleb of porous, wrinkled skin, 1 cm. in diameter. A section through this region shows the bleb to He over a funnel shaped canal which extends into the cerebro-spinal cavity. The covering of the central cncephalocele is formed largely by a leather-Uke tissue. This consists of a very vascular connective tissue over the surface of which is a very thin layer of epidermis. There are numerous sweat glands in this region but no hair follicles. Near the tip of the sac there are two naevi; these consist of areas where the epidermis is lacking and the vascular tissue extends to the surface.

A sagittal section was made of the specimen. The central nervous system was removed and a wax model made of the cerebro-spinal cavity. The specimen was then dissected.

The skeleton shows marked maldevelopment. The arches of all the veretebrae are defective; they are open posteriorly and are flattened outwards. In the cervical and thoracic regions the bodies of the vertebrae are fused, shortened, and dorsally flexed, so that the spine is bent almost double. The occiput rests on the gaping vertebral arches and actually fuses with them on both sides. The lower two-thirds of the squamous portion of the occiput is defective. This forms a very large foramen magnum through which much of the brain has slipped. There is slight scoliosis of the vertebral plate. The lengths of the presacral vertebral parts are: cervical 15 mm.; thoracic 45 mm.; lumbar 43 mm. Comparing these with Aeby's figures for average lengths of these parts in the normal newborn, cervical 45.1 mm.; thoracic 83.9 mm.; lumbar 47.5 mm., it is seen that while the lumbar portion of this specimen is of nearly normal length, its cervical portion is less than half, and its thoracic portion a trifle more than half that long. The vertebral bodies themselves have become somewhat widened. The ribs have undergone considerable disturbance. There are twelve ribs on each side. On the right the first six are fused near their origins; the second rib terminates at the end of its proximal third in a plate of bone joined to the first and third ribs. On the left the fifth to ninth ribs are crowded together in their proximal half; the fifth and sixth ribs have but one costal cartilage between them. The sternum is well formed. There are but six costal cartilage connections on each side. There is a persistent episternum. The two scapulae are defective along their vertebral margins. The left scapula is a much wider bone than the right owing to a prolongation of the medial extremity of the spinous pro(;oss. This prolongation is joined to a rod of bone which in turn is attached to the cvortotl vertebral arclics of that .sul(\ This stimulates a condition frequently associated with Sprenfiic'l's deforniity, confi;enital elevation of the scaimla. The cause of the hunched position of the shoulders, so prominent externally, shows clearly in the skeletal condition. The cervical and \i])i)er thoracic vertebrae lie cnun])led to half their normal extent under the scapulae ami completely change the normal relationshi])s of these parts.

The foUowinfji; imiscles have undergone disturbance. The trapezei are reduced to thin strap-like bands. The rhoml)oidci are shortened to 3 mm.; they arise from connective tissue over the everted arches of the thoracic vertebrae and are inserted in fascia along the inferior vertebral borders of the scapulae. The posterior superior serati could not be identified. The anterior serati are probably represented by scattered tissue on both thoracic walls. The sacro-spinalis and short back muscles lie as two separate muscle bundles, one on each side of the everted vertebral arches.

The soft palate is not formed. There is a bilateral anlage of the uvula on the sides of the pharynx. The right lung is formed of but one lobe.

The central nervous system is very much distorted. The three encephaloceles lie below the foramen magnum. There is considerable cerebral tissue in the left and middle encephaloceles while the one on the right side contains cerebellar tissue. At the base of the brain, of which a large part has slipped below the foramen magnum onto the thoracic vertebrae, the cranial nerves can all be identified; they are much elongated. Owing to a Z-shaped bend of the cord and brain stem the fourth ventricle has been completely inverted and lies on top of the spinal cord. From the floor of the fourth ventricle some cerebellar tissue is drawn back as a flattened sheet to join the rest of the cerebellar tissue in the right encephalocele. The spinal nerves though much crowded are all present, emerging from a fiat spinal cord.

71. Variations in the wall and epithelium of the stomach and esophagus in normal distention. (Lantern.) H. 0. White, Anatomical Laboratory of the College of Physicians and Surgeons, University of Southern California, Los Angeles.

From observations and comparisons derived in the laboratory course of normal histology during the least few semesters, it appeared that the marked changes evident in the musculature and epithelium of the stomach and esophagus merit detailed investigation. To accomplish this the stomach and esophagus of two freshly killed cats were utilized in the normally distended and collapsed conditions. In order to fix the organs in a state of approximate!}' normal distention, saturated aqueous solution of mercuric chlorid was injected into them, and after securely Hgating to prevent the outflow of the fluid, were immersed in the same solution for further fixation. Longitudinal and transverse frozen sections from the contracted and distended organs were made, after proper washing in running water and subsequently in iodized

434 AMERICAN ASSOCIATION OF ANATOMISTS

V

70 per cent alcohol. Details of histological variations were best obtained b}' staining the sections with hematoxylin and Picro-acidfuchsin. Preference was given to the latter stain due to its brilUancy with which it brings out muscle and fibrous tissue. The following results were obtained after measurements taken in micra.

Under the same pressure, the esophageal wall in nearly normal contraction and distention, shows a greater decrease in thickness than that of both parts of the stomach wall under the same conditions.

Comparing the limit of distensibility of the stomach and oesophageal walls, under the same pressure, I found that it is more definite in the latter than in the former, due in all probabilities to the greater amount of supporting and muscle tissue, and also to the varied directions of distribution of the muscle layers of the stomach. In the mucous coat the papillae-like projections of the tunica propria, or corium upon which the epithelium of the esophagus rests are, in consequence of normal distention entirely obliterated, to such an extent that the tunic represents a smooth layer of areolar tissue supporting an equally smooth layer of stratified epithehum, and the tunic is markedly reduced in thickness. The tunica propria of the normally distended stomach, though very appreciably reduced in thiclaiess, does not lose its characteristic appearance, due undoubtedly to the greater abundance of areolar tissue in that tunic and also to the greater folding of the entire mucosa of the stomach.

While in the contracted state only the innermost cells of the epithelium of the esophagus are flattened, and the flatness gradually changing as the middle layers are approached until finally the outermost layer of cells is rather of a simple columnar shape and the nuclei round or oval, in the normally distended esophagus, not only the famiUar shape of the entire epithelium changes, but is also much reduced in thickness, every cell is flattened out and hence elongated; the nuclei in consequence assuming a distinct spindle-shaped appearance.

May not this diminution of the epithelium in thickness (about 65 per cent) together with the flattening of the cells in consequence of normal distention possibly suggest an actual gliding of one cell upon the other, and hence a displacement to some extent? As to the epithelium of the stomach, while the entire mucosa is reduced in thickness (about 40 per cent) in normal distention under the same pressure as in the esophagus, the epithelium itself is only slightly compressed, due probably to the fact that the tunica propria, normally prolonged into the lumen of the stomach above the gastric glands supporting the single layer of columnar epithelium, forms projections of considerable length and these, during normal distention, become imbricated, thereby preventing material changes in the epithelium of the viscus.

The musculature of normally distended stomach and esophagus share relatively an equal reduction in thickness, the longitudinally disposed muscle layers, however, participating more in the thinning process than the circular layers, due evidently to the fact that normally the former is much thinner than the latter. On the other hand,

PROCEEDINGS 435

since the luusciihiris nuK't)s;i of tlie csoijliajius nornially docs not form a eoiiiplete contimious luyor, hence the fascicAiU of tliis tunic, in normal (Hstention, l)ecome widely separated, Avhile the sanu; tunic of the stomach, because of its continuity as a comjilete layer froin one end of the stomach to the other, is only apjireciahle reduced in thickness (about 15 per cent) under identical conditions.

72. Some peripheral relations in the cranial yicrves of reptiles. William A. WiLLARD, Department of Anatomy, University of Nebraska,

Collet>;e of Medichie.

This i:)aper deals with the peripheral distribution of the preauditory cranial nerves of the common garter snake, Eutania sirtalis, with some comparisons with the same regipn in the lizard, Anolis carolinensis, which has i^reviously been described. The points presented represent a i)artial re])ort on the stud}' of the Ophidian head which embraces all the cranial nerves with the attempt to analj^ze them into their functional components. While the following descriptive account is inclusive of the main points only, it is based upon a detailed study of two excellent scries of sections, one of a full grown specimen the other of a young snake about six inches in length. The only noticeable difference was one in myelination and this was so slight that it may be due to technique instead of difference in age.

The nerves to the nmscles of the eyeball are well developed and their superficial origin from the brain corresponds to the condition found elsewhere. Their course to the orbit is intracranial. The}^ converge to unite with each other and with the ophthlamic branch of the trigeminal. The identity of the IV nerve is preserved in section although included in the same sheath with the others. That of III and VI is entirely lost in V for a short part of the course. At the level of the posterior side of the orbit this common trunk again resolves itself into its component parts, VI and part of III first separating from the dorsal side of the trunk to innervate the posterior and inferior rectus muscles respectively. The remainder of III still closely applied to the ventral side of V is marked by the deeper impregnation of its fibers. As this part separates from the ophthalmic the short root of the ciliary nerve is given off from the latter. A segregation of unmyelinated and very hghtly mj^elinated fibers which make up this root is observable in the ophthalmic portion of the common trunk before III separates from it. No direct contribution of fiber from III is recognized either before or after its separation from V. The ciliary root passes at once into the small cilary ganglion from which a single cilary nerve passes into the eyeball. The muscles of the orbit are limited to the four recti and two oblique, there being no accessory muscles. The six muscles are innervated in the usual manner.

The trigeminal nerve is represented in the snake by two anatomically distinct nerves. The ophthalmic nerve has a course cephalad within the bony and membranous cranium to a point opposite the middle of the orbit where it gives off a ramus frontalis to the interorbital integ

436 AMERICAN ASSOCIATION OF ANATOMISTS

iinient while the remainder of the nerve as the ramus nasahs after its separation from III and VI, and from the cihary root has the same distribution as in AnoHs to the epithehum of the nasal capsules and the integument of the preorbital region. It sends no fibers to the mucous membrane of the mouth. The ophthalmic ganglion is large and its root enters the brain just cephalad to the roots of the maxillo-mandibular division of the trigeminal. The maxillo-mandibular division of the Gasserian ganghon is relatively about twice the size of the same structure in the lizard and the nerves entering the ganglion correspondingly large. The maxillary ramus is formed by the union of an infraorbital and a superior labial branch. The first innervates the integument covering the superior labial glands, the second, in combination with the palatine branch of VII is limited to the mucous membrane of the palatine and maxillary regions of the mouth.

The mandibular ramus after giving off nearly all of its motor fibers to the muscles of the head, passes into the bony mandible. In its cephalad course it divides its fibers in a manner similar to that of the maxillary ramus, between the mtegument and the mucous membrane of the mouth. In addition a large Imgual branch takes a recurrent course to reach the base of the long tongue sheath. With this lingual division an extremely fine chorda-tympani nerve unites.

The geniculate ganglion and the facial nerve when separated from the trigeminal, as is only possible bj^ means of sections, present about the same condition as that found in Anolis. The two component roots are distinguishable within the brain. They leave the cranium through a special foramen, which is covered externally by the manibular wing of the Gasserian ganghon. This results in the fusion of the geniculate with the latter ganglion, although its limits are clearly marked in the sections. From the geniculate ganglion a fine fibered palatine ramus passes cephalad intracranially, and a mixed fine and coarse fibered hyomandibular ramus passes caudad. The palatine ramus does not give off any branches until after it has been united anterior to the orbit by means of sympathetic ganglia with the median or infraorbital branch of the V, and there seems to be evidence in the large number of unmyelinated fibers which it contains that it represents, in part, a sympathetic pathway between these ganglia and the brain. The hyomandibular ramus gives off nearly all its fine fibers, most of which are not myelinated to form a communicating ramus with the lower sympathetic ganglia. This ramus is not joined as in the lizards by an external communicating ramus from the orbital plexus. The few remaining fine fibers in the hyomandibular ramus are given off some distance caudad. This extremely small ramus enters the lower jaw and eventually joins the lingual branch of the mandibular ramus of the V. It has been termed the chorda-tympani on the basis of its course and connections, although it has the appearance of a fine unmyelinated sympathetic ranms. There is no middle ear chamber here to modify its course as a post-trematic branch and its position is relatively farther caudad than in the lizard.

PROCEEDINGS 437

Tlic ji;nnips of syinpatlictic }«;aiiglion colls which hi other reptiles have Ix'tMi iiichidcil uii(l(>r tlio terms, palatine, infraorbital, ethmoidal and mandil)\ilar j;;in^lia are reco{>;nized in the snake in iiu^'eased degree and in addition other ganglia are found not appearing in the lizard. The ciliary ganglion and nerve are smaller owing to the less specialized development of the e^'^e.

The cutaneous sense organs arc dermal corpuscles projecting into the ei>idermis. These are very alnnidant along the jaws. Special epithelial sense buds are found in the mouth along the palatine and maxillary and mandibular dental areas. These have the appearance of taste buds but their undoubted innervation by fibers carried in the branches of the trigeminal nerve suggests the possibility of a tactile function. Further study may disclose different types of sense organs in the mouth.

7S. A new method for the study of the development of the lymphatic system. G. B. WiSLOCKi, Department of Anatomy, Johns Hopkins University.

A clear idea ^f the origin and growth of the lymphatic system was first obtained by Sabin, who combined the study of sections of embryos with the injection of the lymphatics with india-ink. This conception comprises the fundamental truths, that the first lymphatics arise by budding from the veins, that the lymph-channels grow from center to periphery (the theory of lymphatic and non-lymphatic zones) and lastly, that the lymphatic capillaries are closed channels lined by endothelium which grow by sprouting and budding. The theory of the venous origin of the lymphatic system is in direct accord with the earlier work upon the development of blood-vessels by His, Hoyer, Prevost, Rouget and Mall and with the work upon the nature of lymphatics by von Recklinghausen and upon their growth by Ranvier.

Schwann's discovery of the vascular channels in the tadpole's tail made possible the first studies of living growing vessels, and since his time the tails of amphibian larvae have been studied again and again in hope of solving the questions of the mode of growth of lymphatics. The difficulty of this method of studjdng the growth of lymphatic vessels is shown by the diversity of the views to which such a study has led different observers, and it was not until Clark devised an anaesthetizing chamber that accurate and useful observations could be made. He was able to show definitely that lymphatic vessels grow by sprouting, and not by the addition to them of changed blood-vesselsor tissuespaces or the transformation of mesenchymal cells into endothelium, and his observations tend to support the argument for the venous origin of the lymphatics system.

In spite of the enormous amount of evidence in its favor, there are many who are unwilling to accept the theory of the venous origin of lymphatics as outlined by Sabin and her co-workers. It seems pertinent therefore at this time to report briefly and present a few of the results of a new technique for studying the growth of lymphatics,

438 AMERICAN ASSOCIATION OF ANATOMISTS

which, combined with the Clark method and the study of sectioned material promises to throw additional light upon the growth and origin of lymphatics.

The method consists m vitalty staming amphibian larvae and observmg the Ij-mphatics, whose endothelium stains quite specifically intra vitam with acid azo-dyes, by means of a Clark chamber and in sections.

Frog eggs were collected and brought into the laboratory, where they were hatched and the larvae kept in aquaria until used. Initial