py v4 © oo Digitized by the Internet Archive in 2009 with funding from Ontario Council of University Libraries http://www.archive.org/details/anatomicalrecord16bard hd Bt. . a vibe o& gee THE ANATOMICAL RECORD EDITORIAL BOARD IRVING HarpDEsTY WaRREN H. Lewis Tulane University Johus Hopkins University CLARENCE M. Jackson Cuar.es F. W. McCuure University of Minnesota Princeton University Tuomas G. LEE Witui1aM S. Mituer University of Minnesota University of Wisconsin Freperic T. Lewis FLORENCE R. SABIN Harvard University Johns Hopkins University GrorGE L. STREETER University of Michigan G. Cart Huser, Managing Editor 1330 Hill Street, Ann Arbor, Michigan VOLUME 16 MARCH-AUGUST, 1919 ot PHILADELPHIA THE WISTAR INSTITUTE OF ANATOMY AND BIOLOGY PROTA [ . : ay ' S0/ : set Cises | $$ TADTIFO / © renawa TT Wag 4 4S” . “2 - L Mem y i s vi /6 - 7 ores’ foram eviell rear wt - af uo CONTENTS 1919 NO. 1. MARCH C. W. M. Poynter. Some observations on wound healing in the early embryo. Twelve EM ra Wi ow nace se cece pee eee PUN wold sw eWbils CE OEMS UML ees ee et 1 Ezra ALLEN. A technique which preserves the normal cytological conditions in both germinal and interstitial tissue in the testis of the albino rat (Mus norvegicus albinus), I SEULAD 200). UD. 202. U2 RUA, ONL eee de te eale ee 25 Hareison R. Hunt. The variations of the inferior thyroid vein of the domestic cat. Seven figures.......... aero OEMS Us «LALA Sh aU baw s'26)). HLS oa eee ie 39 NO. 2. APRIL Esen J. Carey. Teratological studies. A. On a phocomelus, with especial reference tothe extremities. B. The external form of an abnormal humanembryo of 23 days. C. The anomalies of an anencephalic monster. Complete craniorrhachischisis D. A second anencephalic monster. Complete craniorrhachischisis. Seventeen ee emese wen Iduiliel. pissss. jest te bevliow slides that the value of this method is realized. It must alsobe remembered that this method gives added value to the stained sections, for a process can be watched till it reaches the desired stage or one difficult of interpretation and then fixed. WOUND HEALING IN THE EARLY EMBRYO 3 After injury changes take place very rapidly, so it is desirable to place the hanging-drop preparation under the microscope as soon as possible. If a wound is examined a minute or two after it has been made the edges appear ragged and broken. The border is made up of fragments of cells and free cells. Within five minutes the borders of the wound have become smooth in outline due to the readjustment of cells. The broken cells are earried away and the free cells move about till all irregularities are overcome. The cells from one germ layer do not become incorporated in the other germ layers, but these readjustments take place independently in each layer. In observing the behavior of the cells during this period one is reminded of the repulsion and attraction which electric poles have for a pith ball. There is a. noticeable current in the fluid of the drop directed away from the wound; this is probably an out-pouring of the fluids of the tissues and is undoubtedly one of the factors involved in the movement of the debris away from the wound margin. In the case of certain detached cells there is an attrac- tion sufficiently strong to overcome the movement of fluid away from the wound and cause them to move back again in contact with the undisturbed cells and become incorporated in the growing margin. This behavior of the unattached cells suggests the experiments of Driesch (’96) who, by shaking, displaced the primary mesenchymal cells of Echirus. In the course of a few hours these scattered cells rearranged themselves so that devel- opment proceeded in a normal way. Both are examples of what Roux (’96) called ‘cytotaxis’ and which he suggested was due to local differences in the superficial tension of the cells. I have not seen injured cells, i.e., cells which have had part of their cytoplasm cut away, incorporated in the growing wound margin, but think farther observations should be made on this point. Observations on the extra embryonal blastoderm show that all three germ layers take part in the healing process. The ecto- derm and the entoderm are somewhat more active than the mesoderm. Within thirty minutes after the operation the borders of the wound are noticeably thicker than the region farther back and the rent is narrower than in the beginning 4 C. W. M. POYNTER (fig. 1). The piling up of the cells at the edge of the wound is a constant feature of all wounds observed in the extra-embryonal blastoderm. It would seem to be an expression of the difference between the gradient resulting from wound stimulus and the force of surface tension of the germ-layer mass. At this time all of the cells of the wound margin are round and more transparent than those of the undisturbed blastoderm. They may be said to have taken on an indifferent character for the individual cells in their morphology furnish no indication of — the germ layer from which they have been derived. I must make an exception to this in the case of the cells of the entoderm of the area opaca which may still be identified by .the yolk granules present in the cells. The cells of the different germ layers are alike only in appearance. Dedifferentiation has occurred to the extent that the characteristics of the parent germ layer are lost and to this extent the cells are in an indif- ferent state. They are, however, only relatively indifferent for, at least as long as normal stimuli operate, each cell will remain in the germ-layer group of common origin and later differen- tiates, or redifferentiates, into the germ-layer type from which it sprung. At no time during the healing process are the cells distributed evenly over the wound, but those derived from each layer remain together to build up, or advance, that layer. I have never observed any cell from one layer become incorporated with those of another layer, even when the relations seemed most favorable for such an adjustment. For example, when examin- ing a border in which the ectoderm was slightly behind the other layers none of the cells pushed in to fill the gap. This is illus- trated in figure 2. The process of healing, or building up of the wound margin, is difficult to interpret, notwithstanding that the process takes place immediately under the eyes and can be readily observed. At first, as cited above, there is an adjustment of cells on the edge of the wound so that the surface becomes regular and in the case of syncytium an adjustment of cytoplasm about the border nuclei with the formation of recognizable cell membranes (fig. 1). The cells near the border seem to become loosened WOUND HEALING IN THE EARLY EMBRYO 0 from each other and approach spherical shape. The cytoplasm increases, becomes less granular and the nuclei and nuclear membrane almost indistinguishable. There is a forward move- ment of the entire tissue mass and frequently at some distance from the wound border individual cells may be seen moving toward the border more rapidly, edging their way among the other cells till they reach the margin where they take on the characters of the border cells. The healing process is quite rapid; a fissure 2 mm. wide had entirely closed in two hours and twenty minutes, leaving only a slightly more transparent area to indicate the line of juncture. As soon as the two wound borders come in contact the cells of the thickened borders begin a read- justment which soon leaves the cell layers of the same uniform thickness they were before the wound was produced. ‘The cells of the borders gradually resume the typical appearance of those in the undisturbed layers and within four hours no visible evi- dence of a wound remains. The activity of the border distal to the embryo seems to be as great as that next to the embryo. An observation of the wound margin for some time suggests that growth activity is not constant for the whole. This produces a slightly wavy appearance; for example, an area was noticed which advanced for a time more rapidly than that on either side of it; then a period of decreased activity ensued of such duration that the area became the most retarded in the region and ap- peared as a depression; later growth activity was resumed. I can only suggest that possibly the products of cell metabolism, which on account of the great activity are not eliminated, act as an inhibitor to cell movement. Injuries of the embryo present a slightly different picture. If the wound borders are close together, in contact, there is a shifting of the cells of the wound margin and the appearance of the clear cells already described which come in contact with those of the opposite side and so produce union per primam intentionem. All of the germ layers seem to take part in this process and union is brought about through cell dedifferentia- tion and adjustment. 6 Cc. W. M. POYNTER When tissue is cut away, as in the removal of a lateral portion of the embryo, there is a readjustment of cells similar to that described for the extra-embryonal blastoderm, then the process becomes very much slower. During a four-hour period of obser- vation the ectodermal and entodermal borders could be seen to advance slowly over the exposed mesenchyme by the process noted above, with the exception that the ectoderm was much more active then the entoderm and the cells of the border did not tend to pile up. The only response of the mesenchyme to wound stimulus seemed to be a shifting of cytoplasm so that a continuous cytoplasmic border could be seen. This border was like a cell membrane, but, except in the case of the mesothelium, was never observed to produce a distinct layer of surface cells. The behavior of the spinal cord was observed in various regions, but may be best studied in transverse sections. As in the other tissues, immediate proximity of the cut surfaces seems to stimulate cell activity, and primary union is observed to take place in two to three hours. In the case of an exposed surface reaction is very slow. Adjustment of cells closes the end of the canal and covers the surface with closely arranged cells. This surface layer is derived by the shifting of cells. It seems that certain cells which are exposed by the cut do not take part in this general movement, but are finally covered by other cells which come to occupy their place in the superficial layer. Care- ful observation has failed to discover any distinguishable differ- ence between cells which are active in forming the surface layer and those which are covered by it. From the most careful observations I was able to make I did not discover an instance of indirect cell division. The healing process seems to be accomplished through a general movement of cells and cell layers and changes in the cells which, with modi- fication, may be called dedifferentiation; only at a later period can we consider normal cell multiplication as a factor in wound repair. WOUND HEALING IN THE EARLY EMBRYO 7 STUDY OF SECTIONS - Figure 1 is taken from experiment 170. The age of this embryo at the time of operation was about forty hours. The operation consisted in making a long cut in the outer border of the area pellucida, about opposite the heart and to the right of the embryo. The wound margins separated widely, due to the pressure of the yolk. After incubating for two hours the blasto- derm was fixed for examination. The edges of the wound show the piling up of dedifferen- tiated cells at the border of each germ layer as already described in the preceding section. ‘These cells are large clear cells with distinct cell membrane, they take the stain lightly and those of one cell layer are indistinguishable from those of the other layers. All examples of the early periods of the healing process show that the activity of the three germ layers of the extra-embryonic blastoderm is nearly the same. There are no evidences in any of these sections of indirect or direct cell division. Figure 3 is taken from experiment 172, in which the wound was a tear made with a fine needle with the object of removing only the ectoderm layer. The experiment was fairly satisfac- tory, for the major portion of the wound consisted of an area of from 4 to 2 mm. broad denuded of ectoderm; at one end of the wound the needle entirely pierced the membranes, which was an advantage, for it permitted the study of different degrees of injury for comparison. The figure shows an area in which there has been a union of the two torn margins of ectoderm. All of the wound margin is made up of dedifferentiated cells, and when, as in the figure, the borders have fused there is as yet no evidence of redifferentiation. In this experiment the healing process was allowed to proceed for seven hours. A study of these sections shows that the wound stimulus is not transmitted from an injured to an uninjured cell layer. When, as in the figure, the ectoderm is destroyed there is a reaction of the border cells of this layer but none of the meso- derm although it is exposed. Wound stimulus is apparently directly related to the injury or separation of cells in any cell 8 Cc. W. M. POYNTER layer, the different layers behaving for the time being as sepa- rate individuals. It seems that the ectoderm is uninfluenced in the speed of its reaction by the presence of the other layers. A wound, for example, 1 mm. broad involving the ectoderm alone closes in the same time as one involving all three layers, other conditions being the same. Experiment 173 was an operation similar to those described above, on an embryo forty-eight hours old with five hours as the duration of the healing process. This wound was almost 2 mm. broad with rough edges and included all three germ layers. When examined at fixation, the wound was found to be entirely closed. The wound area was very easily made out, for the juncture of the two margins was marked by a thickening due to the large number of indifferent cells present. The division between the ectoderm and mesoderm is not distinguishable nor is it possible to discover any line of juncture of the wound bor- ders. All of these dedifferentiated cells appear so much alike that there are no physical characters to suggest from which germ layer any of them have sprung. At this stage, while there is no indication of a wound border it has been found that the wound can be easily reopened. Repeatedly in experiments showing perfect union at fixation, the most careful handling has not prevented a ruinous tear before the material could be safely embedded. . Experiment 201 is illustrated in figure 5. A blastoderm about twenty-four hours old had a large hole torn in the extra-embry- onic blastoderm. The egg was resealed and allowed to incubate for twenty-four hours. When the blastoderm was removed from the yolk a large hole still remained, but the edges of the opening appeared smooth and healed. The stained sections. show that the wound margins are filled with indifferent cells and that through these the germ layers are fused. Except in primary union, I have not found the germ layers of the border united in this way earlier than eighteen hours after the injury was inflicted. Figure 6 is taken from experiment 11 in which the conditions were the same as in experiment 201 except that the healing proc- WOUND HEALING IN THE EARLY EMBRYO 9 ess was allowed to proceed for twenty-eight hours. In this experiment the germ layers are still separated and the picture very much resembles figure 1, only there are more redifferen- tiated cells at the border of each layer. It seems that the three germ layers remain separate at the wound margin, as in figure 6, till such time as the freedom from embryonal dominance pro- duced by wound stimulus is overcome, then the indifferent. cells fuse into one mass and redifferentiation begins. The mass of dedifferentiated cells seldom exceeds in size that seen in the fizure. In wound margins of this type, which are twenty-four hours old or older, the outer cell layer of the indifferent mass frequently shows degenerative changes or even distintegration. - These changes are probably due to lack of nutrition. Experiment 28, shown in figure 7, was an operation similar to that of experiment 201, in a blastoderm of twenty-hours’ incu- bation. The healing process was allowed to proceed for fifty- four hours. In this, as in all experiments when the healing proc- ess has been allowed to continue for forty-eight hours or more, the fusion of the border layers is complete and redifferentiation has occurred. The ectoderm joins the entoderm so that the line of juncture is hardly discernible and the mesoderm fuses with mesoderm to form a single continuous plate. Wounds of the amnion were observed to behave in the same way as those I have just outlined, so it is unnecessary to repeat the observations made on this membrane. Barfurth, 02, and Lillie, 03, have both observed the closure of wounds of the amnion. Experiment 212 was made on an embryo about thirty-four hours old. A cut was made parallel to the long axis of the embryo and a short distance lateral to the spinal cord, com- pletely dividing all of the tissues. The wound edges were sepa- rated widely, the egg resealed and allowed to incubate for five hours. The sections show the difference in reaction between the embryo and the extra-embryonic blastoderm. The denuded mesenchyme shows a definite border where the readjustment already described had occurred, but there has been no active dedifferentiation of cells. The advancing border of ectoderm 10 Cc. W. M. POYNTER shows dedifferentiated cells at its edge, but there is no tendency for these to pile up in the way observed above. The entoderm «shows much less reaction to the wound stimulus than the ecto- derm. At later stages when the ectoderm has almost covered the exposed mesenchyme of the somatopleure there is a shifting of cells of the mesenchyme, so that the coelom becomes closed through the fusion of the two mesodermal plates. The process seems to be brought about, at least in part, by shifting and fusion of dedifferentiated mesothelial cells. Figure 8 is taken from this experiment. Figure 9 is taken from experiment 54 in which conditions were the same as in experiment 212 except that healing was allowed to proceed for twenty-four hours. This shows the two plates fused, covered by ectodermal epithelium and the healing proc- ess complete except for the redifferentiation of mesodermal cells where the two mesenchymal plates joined. If a large area has been denuded of its epithelium it will take more than twenty-four hours for it to become covered. In all wounds of the embryo the greater activity of the ectoderm as compared with the entoderm is very noticeable. The tendency of dedifferentiated cells to pile up along the advancing border of either ectoderm or entoderm is very much less than in the extra-embryonic blastoderm. After about twenty-four hours mitotic figures can occasionally be seen in the cells of the wound margin, but never in sufficient numbers to warrant the con- clusion that cell division is more active here than in regions more remote, nor that cell division, even at this time, plays a major role in the healing process. A study of wounds of the spinal cord shows that there is no apparent change after the first readjustment of cells observed in the section above. After a longer or a shorter time the ecto- dermal epithelium covers the exposed cord tissue in the same manner as that just described for the mesenchyme. Figure 10 is taken from an oblique section cutting through a transected cord in which the healing process had proceeded for twenty-four hours. The section shows the advancing border of the epi- thelium and the reaction of the cord cells on the wound margin. WOUND HEALING IN THE EARLY EMBRYO 11 Figure 11 taken from another experiment on the cord is shown to illustrate the closure of the neural canal. I have already spoken of the early closure of the canal by a shifting of cells. This is followed by a collapse of the canal for a distance of from 0.3 to 1 mm. from the transection. The two lateral walls of the cord come into contact as if through lateral compression and the canal is entirely obliterated. There is, however, no true fusion of the cells of the two lateral halves of the cord and a ae tinct line of separation can always be made out. All of the examples I have used in this study have been from incised or torn wounds. The experiments in which the hot needle was used were not entirely satisfactory. It was necessary to work very rapidly with the needle or it would become too cool to cauterize properly. This made it impossible to do careful operations or exactly limit the amount of damage to the tissues. If the needle was too hot it tended to stick to the tissues and in freeing it much unintentional damage was inflicted. In the experiments studied there was the same type of reaction to wound stimulus as when a simple incision was made, but the process was much more difficult to interpret because of the injured cells and necrotic tissue present. Figure 12, taken from experiment 62, shows the effect of forty-eight hours’ healing of a wound of the somatopleure produced with a hot needle. The experiments cover observations on embryos as young as the ten-somite stage and as old as the twenty-nine-somite stage. No differences in the degree or type of reaction have been noted for the different ages. Observations were made up to 120 hours after the injuries were inflicted. This is a longer time than is necessary for processes covered in this paper, for redif- ferentiation is complete in most cases by the end of sixty hours. _ The later observations suggest that the chick may survive very Serious mutilation and go on developing in a normal way but that it is incapable of regeneration in the sense that the term has been used for the simpler animals. Simple wound healing seems to be the only reaction to wound stimulus of which warm- blooded animals are capable and in this respect the embryo, at least the chick embryo, reacts as does the adult. 12 Cc. W. M. POYNTER DISCUSSION The reaction, common to all animals, by which wound sur- faces become closed through a process called healing has long been recognized. Within the last two hundred years observa- tions and experiments on the possibilities of regeneration have broadened our knowledge of this important reaction and sug- gested problems of the greatest biological interest. The earlier studies concerned themselves with the behavior of organs and tissues of adult vertebrates. More recently the ability of dif- ferent animals to regenerate lost parts or to regenerate an entire individual from a fragment has received the major attention of biologists. The general questions concerning the extent of regeneration, its histogenesis and the conditions effecting it furnish a field for research of the most fundamental importance. Wound healing seems to be one step or phase of the general process of regeneration. In the lower animals the closure of the wound is followed by continued local growth leading to a more or less complete restoration of lost parts. The higher we go in the animal scale the greater the complexity of organization and the less the power of a part to reproduce the whole. Among the higher vertebrates almost all that remains of the process of regeneration is the reaction of wound repair. Wound healing, particularly in man, has been very extensively studied. There are, however, many unsolved problems still confrontingus. A comparative study of the regenerative process in embryos and adults of the higher vertebrates has received very little atten- tion. The reaction of the embryo to wound stimulus offers an opportunity to study the process under somewhat simplified conditions and not only adds to our knowledge of the problem of regeneration, but throws light on the general growth question as well. Fraisse (’85) studied the regeneration of epidermis on the tail of amphibian larvae and gave an historical account of epithelial regeneration for vertebrates. He concluded that, in adults, the epithelium regenerated from epithelium by cell proliferation. This was contrary to the generally accepted theory that epi- WOUND HEALING IN THE EARLY EMBRYO 13 thelium might arise as the result of a metamorphosis of leuco- cytes and contrary to his own observations on larvae, or rather in spite of the fact that he saw no evidence of cell division nor new cell formation. Barfurth (’94) studied regeneration of the germ layers of amphibian larvae. He noted that the time of closure of the wound was directly dependent on its size and that the ectoderm reacted more rapidly than the entoderm, due, he believed, to its greater elasticity and the firmer manner in which the cells are held together. He made a sharp distinction in the healing time between a cut and a tear, finding that a clean cut healed more rapidly. In the chick there is very little difference in the reac- tion time of the different germ layers of the extra-embryonic blastoderm, and except in the time of adjustment, which is negligible, it seems to make no difference whether a wound is incised or torn. Barfurth noticed the piling up of cells, ‘‘con- sisting solely of cells torn from the layer,” which he called, after Roux, Extraovata. He said, ‘‘The extraovata after reaching a certain size comes under dominance of the embryo. If it passes beyond the dominance of the embryo it will develop into a sepa- rate embryo.” It is probable that the extraovata is of the same character as the mass of indifferent cells I have noted on the border of the wound in the chick. In the chick, however, the mass does not reach a size nor behave in a way to suggest an extraovata. The higher organization of the chick and the con- sequently greater dominance of the embryo probably accounts for the difference between the two forms. Born ( 96) also studied amphibian larvae and concerned him- self with the way in which the cicatrix was covered by the epi- thelium. He observed (p. 579): ‘‘On account of the time in which the epithelial covering is completed, mitotic division of cells is not to be thought of. It appears to me, that the epi- thelium as a whole is concentrically shifted (vergeshoben) over the wound surface,—for the picture does not suggest an active wandering out of the individual cells.’ As already pointed out, I am in entire agreement with the conclusions that there is a general shifting of the epithelial layer and absence of cell divi- 14 Cc. W. M. POYNTER sion, but in the chick the border of the wound suggests the locus of greatest cell reaction and I see no evidence for the conclusion (p. 572) that the movement of the whole layer is due to the vital effort of the individual cells to flatten themselves over the great- est possible surface. It is impossible to observe the loosening up of cells near the border and the behavior of the cells at the extreme margin without being impressed with the fact that the advance is an active, not a passive one. I must conclude, as Rand (’05) has for the earthworm, that ‘‘We are compelled to look in the individual cell itself for the immediate source of activity.” Oppel (’13) and Osowski (14) studied the behavior of the epithelium on explants of frog larvae and concluded that the movements of the epithelium are responsible for the covering of the wound, not through some pressure behind, but as a direct result of the activity of the cells themselves. Osowski speaks of the action as due to ‘Massenbewegung.’ He observed no pseudopodia, consequently does not look on the movement as amoeboid in character. Holmes (’14) repeated the work of Osowski but reached a different conclusion concerning the way in which the cells moved. He decided that ‘‘The extension of epidermis in both larval and adult forms is due to the amoeboid activity of the hyaline pro- toplasm along the margin of the extending mass.” I was un- able to observe pseudopodia on any of the advancing cells, but this is valueless as negative evidence, for Holmes pointed out that, ‘‘The pseudopodia of epithelial cells of amphibian larvae are so short and so fine that it would scarcely be possible to detect them when they are extended over other parts.’ To morpholo- gists who find satisfaction in tangible structure, the discovery of pseudopodia will offer additional proof of individual cell activity. Harrison (’14) has shown that cells growing in vitero possess stereotropism and suggests that this may explain in a measure cell movement on wounds. The phenomena of stereotropism and cytotaxis, already alluded to, suggest that the movement of epithelial cells is in response to a direct stimulus of chemico-— WOUND: HEALING IN THE EARLY EMBRYO 15 physical nature. We have already seen that the behavior of the ectoderm of the extra-embryonic blastoderm is different from that on the embryo. If the behavior were the same we would not have a gap in the blastoderm bridged, but the ectoderm would immediately advance over the mesoderm to unite with the entoderm and a healed border would result; a reaction which does not take place till several hours have elapsed, and then only in extensive wounds. May it not be that in tissues remote from the embryo the wound stimulus, for a time, frees the cells of the border from the natural gradients and makes them behave as independent individuals; later as the stimulus is exhausted the normal gradients are reestablished and the ectoderm behaves as it does on the embryo? Lillie (03) studied the powers of regeneration of various organs in the chick. He was interested primarily in correlative differentiation, but recognized the existence of wound repair in these embryos. Shorey (’09) also studied regeneration in the chick. Both observers concluded that, beyond wound repair, no regeneration is to be expected for the chick. I have referred to the changes which take place in cells of the wound margin by which they lose their distinguishing germ- - layer characters. These cells may be said to have dediffer- entiated or become indifferent, but they always take on again later, when normal ‘conditions are reestablished, the type form they had before the wound stimulus was operative. So much has been written of dedifferentiation in recent years that it is not necessary to review the literature. In lower animals, as those in which a whole part or individual may regenerate from a fragment of tissue, the commonly held theory seems to be that the new part or individual is developed by the dedifferentiation of the old tissue cells and their redifferentiation into tissues of the new individual. Minot (08) presented a different view, he said “If the head or tail of a planarian is cut off the part lost is regenerated, not by ‘growth of the old tissue, but perhaps wholly by multiplication and differentiation of ‘formative’ cells which migrate to the place where they are needed and produce the structure required.” THE ANATOMICAL RECORD, VOL. 16, No. | 16 Cc. W. M. POYNTER Again he called these cells ‘‘of embryonal type, that is to say, cells of the young type.’”’ It would seem that there is much experimental evidence lacking to establish the theory of ‘forma- tive’ cells, and while there are many connective-tissue: cells which are difficult to classify, their behavior is not sufficiently understood to warrant the conclusion that they are ‘young cells’ capable of differentiation in any direction. If we accept the theory that cells under certain conditions are capable of dedifferentiation and redifferentiation it may be qualified, for an examination of the experimental work in this field indicates that the potentialities of dedifferentiated cells are influenced by certain known factors. The cut end of a planarian may be made to grow either a head or a tail at the will of the experimenter; but as we advance to animals of more complex organization the extent of regeneration becomes less and less. Child (15) has pointed out, ‘‘That the inhibition or retardation of new individuation by the dominant region of an individual occurs when the original gradient is sufficiently fixed in proto- plasm.’”’ That would mean that the further we advance in the animal scale the greater the dominance of the ‘original gradient,’ hence the more limited the process of regeneration. If the original gradient limits the process of regeneration, it probably does so through determining the type of redifferentiation of the dedifferentiated cells and through limiting the extent of dediffer- entiation. The indifferent cells which appear on the wound border of the chick, as the result of wound stimulus, are dom- inated by the original gradient or gradients to the extent that they redifferentiate only into the germ-layer type from which they sprung. ’ WOUND HEALING IN THE EARLY EMBRYO 17 SUMMARY Wounds of the chick blastoderm heal with great facility and the process can be watched for a number of hours in hanging- drop preparations. In wounds of the extra-embryonic blastoderm all three germ layers take part in the healing process, but the ectoderm and the entoderm are somewhat more active than the mesoderm. As the result of wound stimulus the cells dedifferentiate and all take on the same general appearance and stain reaction. The dedifferentiated cells pile up along the border of the wound, but these indifferent cells of the three germ layers do not fuse and form a healed margin till many hours after the wound has been made. In the interval, before fusion, the masses of the three layers remain separate and advance by amoeboid movement of the individual cells till the opposite wound border is encountered when the two fuse. Later the dedifferentiated cells rediffer- entiate into cells of the type from which they sprung and wound healing is complete. 5 Wounds of the embryo heal by dedifferentiation of the ecto- dermal epithelium and the migration of these cells over the cicatrix. The process is through dedifferentiation, migration by amoeboid movement and redifferentiation into epithelium. There is no regeneration of the underlying parts and the ento- derm takes very little part in the process. The directive stim- ulus which causes the migration of the epithelial cells is probably of a chemico-physical nature and the covering of the wound is effected without the occurrence of cell proliferation. The wound stimulus is not transmitted from the injured to the uninjured tissue layers in the chick, and even dedifferentiated cells are so limited in their potentialities that regeneration, in the sense that it occurs in the lower animals, is not observed. It would seem that wound repair is a step or phase of the process of regeneration and that the embryonic dominance is so pro- nounced that it prevents the wound stimulus from carrying the process beyond this phase. ‘ 18 Cc. W. M. POYNTER LITERATURE CITED BarrurtH, D. 1894 Experimentelle Untersuchung iiber die Regeneration der Keimblatter bei den Amphibien. Anat. Heft., Bd. 3, S. 309-351. Barrurtu, D., Dietrich UND DraGEenporFrF, O. 1902 Versuche tiber Regen- eration des Auges und der Linse beim Hiihnerembryo. Anat. Anz., Ergiinz. zum Bd. 21, Ver. der anat. Ges. auf der 16 Versamul. in Halle. Born, G. 1896 Ueber Verwachsungsversuche mit Amphibienlarven. Arch. f. Entw-Mech., Bd. 4, S. 349-465; 517-623. Curtp, C. M. 1915 Individuality in organisms. Univ. Chicago Press, Chi- cago. Driescu,H. 1896 Die taktische Reizbarkeit der Mesenchymzellen von Echinus microtuberculatus. Arch. f. Entw-Mech., 8S. 362-380: Fraisse, P. 1885 Die Regeneration von Geweben und Organen bei den Wirbel- thieren, besonders Amphibien und Reptilien. Theodor Fischer, Cassel und Berlin. Harrison, R. G. 1914 The relation of embryonic cells to solid structures. Jour. Exp. Zool., vol. 17, pp. 521-544. Hotmes, T. J. 1914 The behavior of epidermis of amphibians when cultivated outside the body. Jour. Exp. Zool., vol. 17, pp. 281-296. Lewis, R. M., anp Lewis, W. H. 1911 The cultivation of tissues from chick embryos in solutions of NaCl, CaCl, KCl, and NaHCO. Anat. Rec., vol. 5, pp. 277-293. Linu, F. R. 1903 Experimental studies on the development of the organs in the embryo of the fowl. Biol. Bull., vol. 5, pp. 92-123. McWuorter, J. E., anp WuippLe, A.O. 1912 The development of the blasto- derm of the chick in vitro. Anat. Rec., vol. 6, pp. 121-139. Minor, C. 8. 1908 The problem of age, growth and death. G. P. Putnam’s. Sons, New York. OpreL, A. 1913 Demonstration der Epithelbewegung im Explantat vom Frosh- larven. Anat. Anz., Bd. 45, S. 173-185. Osowsk1, H. E. 1914 Ueber aktive Zellenbewegung im Explantat vom Wer- beltierembryonen. Arch. f. Ente-Mech., Bd. 38, S. 547-583. Rano, W. H. 1905 The behavior of the epidermis of the earthworm in regen- eration. Arch. f. Entw-Mech., Bd. 19, 8. 16-57. Roux, W. 1894 Ueber die Selbstordnung (Cytotaxis) sich ‘‘beriihrender”’ Furchungszellen des Frocheies durch Zellenzusammenfiigung, Zellen- trinnung und Zellengleiten. Arch. f. Entw-Mech., Bd. 3, 8. 381-468. Suorey, M. L. 1909 Differentiation of neuroblasts. Jour. Exp. Zool. vol. 7, pp. 25-63. Pe 4¢ i ws. 3 wv 2 ¥q . Ce te a] east PY ee « sl tf ge "? ey" ia $= celts Pc wae P = PLATE 1 EXPLANATION OF FIGURES 1 Section through the wound border of the area pellucida. From exper- iment 170 on chick forty hours old, wound healing of two hours’ duration. Left margin shows the indifferent cells of the three cell layers. The tendency to pile up is most noticeable in the ectoderm. X 450. 2 Section through the same wound shown in figure 1, but at a different level. When the three layers have relations which are favorable for fusion, they remain independent of each other. The clear indifferent cells are a constant feature of this stage. 3 Section through wound of experiment 172 in which healing process has proceeded for seyen hours. The ectoderm only was wounded and the fissure has filled with indifferent cells so that the line of juncture cannot be discovered. The other cell layers have not reacted to the wound stimulus of the ectoderm. 4 Section through a wound of the extra-embryonic blastoderm which has just closed. Experiment 173 allowed to heal for five hours. Indifferent cells so packed together that it is difficult to distinguish those of the mesoderm. x 450. 5 Section through wound of extra-embryonic blastoderm of experiment 201. Healing has progressed twenty-four hours. Somatopleure and splanch- nopleure have fused and entoderm is continuous with ectoderm. At this stage the cell layers have lost the seeming repellence which acted to keep them apart in earlier stage shown in figure 1. 6 Section through wound of same region shown in figure 5, experiment 11; wound repair time, twenty-eight hours. The two plates of mesoderm have fused, but the ectoderm and entoderm are still separate. Except for the migra- tion of the wound border and a somewhat larger number of dedifferentiated cells, the figure is the same as figure 2, indicating a slowing up of the reaction produced by wound stimulus. 7 Section through healed wound from experiment 28. The process of repair has been allowed to continue for fifty-four hours. This is a later stage of the conditions shown in figure 5. The dedifferentiated cells have redifferentiated and it is impossible to discover the line of juncture of the two plates. 20 WOUND HEALING IN THE EARLY EMBRYO PLATE 1 Cc. W. M. POYNTER PLATE 2 EXPLANATION OF FIGURES 8 Photomicrograph showing transverse section of wound cutting through the splanchnopleure and somatopleure of embryo. The denuded mesenchyme has not yet been covered by the ectoderm and the coelom is not closed. Time of healing, five hours. Reduced in publication. X 500. 9 Photomicrograph from experiment 54, in which healing has proceeded for twenty-four hours. Region same as figure 8. Wound is closed and covered by ectoderm, a few indifferent cells can be seen where the two mesenchymal plates join. 10 Photomicrograph of a transection of the spinal cord after twenty- four hours. The ectoderm has only partially covered the cicatrix of the cord. 11 Photomicrograph of an oblique section of a transected cord. This picture is shown to illustrate the way in which the neural canal closes for some distance beyond the wound. The-cells lining the neural tube come in close contact but the line of juncture is always very distinct. 12. Photomicrograph of a cauterized wound, experiment 67, after forty- eight hours. Section is of the embryonic somatopleure and shows a largeamount of dead tissue which has not as yet been eliminated. ‘ 22 WOUND HEALING IN THE EARLY EMBRYO PLATE 2 Cc. W. M. POYNTER Resumido por el autor, Ezra Allen. Un método para la fijacién de los testiculos de la rata, mediante el cual se conservan los detalles citol6gicos asi como las relaciones normales entre los tibulos y los tejidos intersticiales. La relacién normal entre el tejido intersticial y los tuibulos en el testiculo de la rata puede conservarse inyectando los vasos san- guineos con el fijador ‘‘B-15” (descrito en mi trabajo acerca de los Experimentos sobre Técnica, (16) ), después de expulsar la sangre con solucién salina normal o con solucién de Locke. Una presion de 15 a 20 mm. de mercurio es suficiente para la inyeccién. Para conocer la presién se une el frasco de presién con un mandmetro de mercurio, comprimiendo el liquido por medio de una pera de goma, vertiendo agua gota a gota o medi- ante el aire comprimido. Tan pronto como los testiculos se en- durecen, se separan del cuerpo del animal y se colocan en el liquido fijador templado, durante treinta a sesenta minutos, al cabo de los cuales se cortan en rodajas de 2 a 4 mm. de espesor. Tanto el animal como los liquidosa inyectar deben conservarse a una temperatura de unos 38°C. durante el proceso de la inyec- cion. La deshidrataci6n, aclaramiento e infiltracién se realiza- rin por cambios muy graduales de los liquidos empleados, que se verterdin gota a gota. Este método fija muy bien todos los detalles citolégicos, incluso los cromosomas en los estados en que tienden a aglomerarse. Otros O6rganos se fijan también muy bien para trabajos citolégicos. Para expulsar toda la sangr> de los rifiones se necesitarAé una presién algo mayor que la indicada. Translated by Dr. José Nonidez Columbia University AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, FEBRUARY 24 A TECHNIQUE WHICH PRESERVES THE NORMAL CYTOLOGICAL CONDITIONS IN BOTH GERMINAL AND INTERSTITIAL TISSUE IN THE TESTIS OF THE ALBINO RAT (MUS NORVEGICUS ALBINUS) EZRA ALLEN The Wistar Institute of Anatomy and Biology and the Zoological Laboratory of the University of Pennsylvania ELEVEN FIGURES In arecent paper (Allen, ’16), I described a method by which the cytological details of the germ cells in the albino rat might be demonstrated. While this method is successful for the purpose named, it does not preserve the normal relationships between the tubules and the interstitial tissue. This tissue is torn away from the tubules and distorted. These effects are shown in figures 3, 5,and 7. The normal conditions appear in figures 4, 6, and 8. Interstitial tissue in the rat testis is much less in relative quan- ity than in most mammals, and is so delicate that if the notor- iously impermeable tunica albuginea is ruptured to admit the fixing fluid freely, the interstitial tissue is badly torn. The purpose of this paper is to describe a method by which the normal histological and cytological relationships of the two tissues involved may be preserved. It is published with the hope that it may also be of service in suggesting a solution of similar prob- lems in other tissues. A list of reagents and apparatus will be given and then a description of their use. An extended ex- perience has shown that no detail may be omitted in the proc- ess without danger to the material, and for that reason the description of reagents and processes is full. 25 26 EZRA ALLEN REAGENTS AND APPARATUS Reagents Washing fluid for removing the blood: either 0.9 per cent salt or Locke’s solution. Fixing solution: A. Picrie acid, saturated aqueous solution................ 75.0 ec. Fornial, @hemicasly pure... 1)... Vi Bb 25.0 ce. Glacial GR6GG QARE>. Sas. . «0 a0 50 dh pia neel ves 2 10.0 ce. B. Chrovitp acid, crests C 2.5 oo ue os Se 1.5 grams Pk apa) *)) as oe See 2.0 grams To prepare: Mix the reagents under A and warm to 38°C. ina closed vessel. Then stir in the chromic acid until completely dissolved; after which add the urea, stirring while it is being added. The resulting fluid should be transparent and rather dark brown in color. If a white precipitate forms, the difficulty is doubt- less with the formalin. Ordinary commercial formalin is almost certain to produce this result. That put up by Schering never gave this trouble. The representatives of this firm are now putting out a product that seems about equal for this purpose to that formerly imported. If the solution is turbid, the fault may lie with either the formalin or the chromic acid. This latter should be as nearly equal in quality to the Kahlbaum as may be obtained. Deep red crystals have proved satisfactory. After standing an hour or thereabouts, the fluid will turn green on account of the formation of chrome acetate, when it is not as effective for fixation as before. Other reagents 5 per cent, 10 per cent, 50 per cent, and 70 per cent alcohol. Saturated aqueous solution of lithium carbonate. Analin oil, C.P. Synthetic oil of wintergreen (methy] salicylate), C.P. 52° or 56° paraffin—the lower temperature is preferable. The anilin oil should be nearly colorless. If doubt exists as to its purity, it should be distilled, when it will be practi- cally colorless. 667. ~“ 36 CYTOLOGICAL CONDITIONS IN TESTIS OF ALBINO RAT EZRA ALLEN PLATE 1 a Resumido por el autor, Harrison H. Hunt. Las variaciones de la vena tiroidea inferior del gato doméstico. La vena tiroidea inferior vierte la sangre que conduce en cual- quiera de las siguientes venas: innominada izquierda, yugular interna izquierda, innominada derecha, yugular externa derecha, yugular interna del mismo lado y precava. En uno de los casos la vena tiroidea inferior se dividia en su extremo posterior en dos ramas una de las cuales entraba en la vena innominada derecha, la otra en la izquierda. Pr6ximamente en la mitad de los gatos estudiados la tiroidea inferior se vertia en la innominada izquierda. En su extremo anterior la vena en cuestién, generalmente, aunque no siempre, se divide dicot6micamente. Generalmente esta ramificacion estd situada entre los l6bulos de la glandula tiroides, aunque en algunas ocasiones, se presenta en el mismo nivel que el extremo anterior de la glindula tiroides 0 posteriormente a dicha glandula. Translated by Dr. José Nonidez Columbia University AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, FEBRUARY 24 THE VARIATIONS OF THE INFERIOR THYROID VEIN OF THE DOMESTIC CAT HARRISON R. HUNT West Virginia University SEVEN FIGURES The inferior thyroid veins of man vary considerably. The following observations, made on thirty-three domestic cats selected at random, show that the same is true of this vein in the cat. Each of the accompanying figures represents, some- what diagrammatically, the conditions in a single animal. These seven animals suffice to give a fairly complete idea of the variations in the remaining twenty-six. Figure 1 shows the inferior thyroid vein (/) communicating anteriorly with the left internal jugular, receiving branches from only the left lobe of the thyroid gland (14), then passing ob- liquely backward across the trachea (13) to join the right innominate vein (/0). The veins labeled 2 and 3 in this and the following figures were not homologized with certainty with the human superior and middle thyroid veins. Two other dis- sections resembled figure 1 very closely, except that in one of them the inferior thyroid vein joined the external jugular vein at a. Figure 2 resembles figure 1 in some respects. However, the inferior thyroid vein (1) in figure 2 receives a branch (a) from the right lobe of the thyroid gland (1/4), empties into the right internal jugular vein (4), but in the dissection did not appear to communicate directly with the left internal jugular. The inferior thyroid vein in a second cat joined the jugulars at m 39 40 HARRISON R. HUNT (fig. 2), and its transverse branch (fig. 2, a) could not be traced as far as the right lobe of the thyroid gland. In a third indi- vidual, vessels a and 3 (fig. 2) were connected by a conspicuous vessel running along the dorsal surface of the right lobe of the thyroid gland, and the inferior thyroid vein entered the right innominate at n (fig. 2). In all other respects the inferior thyroid veins in these three cats were very similar. In one case (fig. 3) the inferior thyroid vein (1) was formed by the union of a branch (2) from each of the internal jugular veins (4 and 5). The inferior thyroid, after receiving branches from the right lobe of the thyroid gland (7/4), passed obliquely backward, joining the left innominate vein (//) near its union with the precava. . The resemblance between the inferior thyroid veins (/) in figures 3 and 4 is evident. However, in figure 4 the inferior thyroid apparently was unconnected, at its anterior end, with the internal jugular veins; it lay close to the left lobe of the thyroid gland; and its junction with the left innominate vein was considerably anterior to the junction shown in figure 3. A small vein (fig. 4, 75) ran along the dorsal side of the left lobe of the thyroid gland parallel to the inferior thyroid vein (7), con- necting anteriorly and posteriorly with the latter vessel. One other cat showed practically the same conditions as those repre- sented in figure 4, though the vein 1/5 (fig. 4) was not found, and the inferior thyroid vein emptied into the innominate at a (fig. 4). In figure 5 the two vessels uniting to form the inferior thyroid vein (/) passed backward a short distance before coming together. The inferior thyroid vein was median in position, receiving small side branches from both lobes of the thyroid gland (14). Except for the fact that 16 and 15 (fig. 4) lay on opposite sides of the trachea, their locations and courses were quite similar. In four other individuals the inferior thyroid vein strongly resembled the same vein in figure 5, though in one case it joined the precava at e (fig; 5). INFERIOR THYROID VEIN OF DOMESTIC CAT 41 EXPLANATION OF FIGURE NUMERALS 1, inferior thyroid vein; 2 and 3, branches of the internal jugular veins; 4 right internal jugular vein; 6, left internal jugular vein; 6, right external jugular vein; 7, left external jugular vein; 8, right subclavian vein; 9, left subclavian vein; 10, right innominate vein; 11, left innominate vein; /2, precava; 13, trachea; 14» thyroid gland. (For the significance of 14, 16, and the letters, see text.) 42 HARRISON R. HUNT In the animal from which figure 6 was drawn a vein ran backward near the medial margin of each lobe of the thyroid gland (14), the two veins uniting at the level of the isthmus of the gland. One of these vessels (2) branched off from the right internal jugular vein. The inferior thyroid vein (1) emptied into the innominate where the external jugular and subclavian veins joined (8). The variations of the inferior thyroid vein in thirteen more cats can best be described by reference to the lettering in figure 6. In twelve of these animals the vein branched at approximately the following points: in one case at a (the branching thus closely resembling the branching of the inferior thyroid vein in fig. 5), in four cases at c, in six animals at the place where the inferior thyroid vein branches in figure 6, and in one animal as far back as d. The approximate points at which the inferior thyroid vein in these thirteen cats emptied into the innominate and external jugular veins varied greatly (fig. 6). In four cases this union was at o, in four cases at n, in two at 1, and in three cases at the points e, h, and 7, re- spectively. In two of these individuals the left anterior branch could be traced to the left internal jugular vein; in two cases, including the animal shown in figure 6, the right anterior branch came from the right internal jugular vein. Incomplete records of the inferior thyroid veins in four animals show that the vein emptied on the right side in three of them, and into the left internal jugular in the fourth animal. Figure 7 represents the posterior end of the inferior thyroid vein (/) in one animal. The anterior portions of the vein were not drawn. The vessel divided into two branches which joined the innominate veins (J0 and 1/1) at the places shown in the figure. Possibly some very small veins emptying into the inferior thyroid vein were not well injected in all the cats examined, thus escaping observation. This might explain the failure in many cases to find connections between the inferior thyroid and internal jugular veins near the anterior end of the thyroid gland. INFERIOR THYROID VEIN OF DOMESTIC CAT 43 TABLE 1 NUMBER OF CATS Inferior thyroid vein unbranched at its anterior end.................. 3 Inferior thyroid vein divided into two branches at the anterior end of NEE clo co ohne Sone eee eee ee eee nee 6 Inferior thyroid vein divided into two branches at varying positions between tuevopes of the thyroid gland........:................65 17 Inferior thyroid vein dividing into two branches posterior to the aan 1 TABLE 2 | NUMBER OF CATS Inferior thyroid vein emptying into left innominate................. 17 Inferior thyroid vein emptying into the left internal jugular......... 1 Number of cases in which the vein emptied on Ree Sides sees x... ey il: Inferior thyroid vein emptying into right innominate................. 6 Inferior thyroid vein emptying into right external jugular............ 3 Inferior thyroid vein emptying into right internal jugular............ ] Inferior thyroid vein emptying on the right side (exact position not SURE LL) co ces cee ctl oes Oe ee 3 Number of cases in which the vein emptied on the right side......... 13 Inferior thyroid vein emptying into precava...................-.005: 1 Inferior thyroid vein emptying into both innominate veins........... 1 SUMMARY The results of this investigation may be summarized best in tabular form. Table 1 shows the variation in the dichotomous branching of the inferior thyroid vein at its anterior end. Table 2 summarizes the variations of the point at which the vein emptied posteriorly into the larger veinous trunks. Thus the inferior thyroid vein in the majority of cases branches between the lobes of the thyroid gland and enters the larger veins on the left side. Morgantown, West Virginia, November 22, 1918. Resumido por el autor, Eben James Carey. Estudios teratolégicos. A. Sobre un phocomelus, con especial mencién de las extremi- dades. El cardcter principal de esta clase de monstruos es un acortamiento anormal y cesacién del desarrollo de algunos o todos los huesos largos de las extremidades. El presente estudio revela el hecho de que en ausencia completa o desarrollo rudi- mentario de una parte del esqueleto, se encuentra también una falta completa o parcial de los mtsculos relacionados con él. Lo inverso es también cierto, pues un excesivo desarrollo de las partes esqueléticas esta’ acompafado por un grado mayor de desarrollo en los mtisculos relacionados con ellas. B. La forma ex- terna de un embrién humano anormal de veintitrés dias. El autor da una detalla da descripcién de la forma externa de este embrién y una descripcién de la interesante malformacidén de la regién cer- vical. El estado de desarrollo de la forma exterior coloca a este embrion entre el descrito por Bremer, de 4 mm. de longitud y 21 dias de edad, y el descrito por Mall, de 7 mm. y 26 dias, de modo que la edad probable del que se describe es 23 dias. C. y D. Las anomalias de los monstruos anencefdlicos: cranioraquis- quisis completa. El hecho més interesante con relacién a los monstruos anencefdlicos, ya notado por varios observadores, es que generalmente pertenecen al sexo femenino. Las anomalias bien marcadas, que se describen en los presentes estudios, son: la falta de cerebro, generalmente también de médula espinal y la falta de desarrollo de los huesos que integran la béveda craneal y la lamina de la columna vertebral. La boca tipicamente abierta de los monstruos anencefdlicos esté relacionada con la falta de la béveda craneal y la pérdida correspondiente de la porcién ante- rior del mtisculo temporal. Las disecciones revelan también la existencia de una interrelacién definida entre el desarrollo de los tejidos 6seo y muscular. Translation by José F. Nonidez Columbia University AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, MARCH 17 TERATOLOGICAL STUDIES A. ON A PHOCOMELUS, WITH ESPECIAL REFERENCE TO THE EXTREMITIES B. THE EXTERNAL FORM OF AN ABNORMAL HUMAN EMBRYO OF TWENTY-THREE DAYS C. THE ANOMALIES OF AN ANENCEPHALIC MONSTER. COMPLETE CRANIORRHACHISCHISIS D. A SECOND ANENCEPHALIC MONSTER. COMPLETE CRANIORRHACHISCHISIS EBEN J. CAREY Department of Anatomy, Creighton Medical College, Omaha, Nebraska SEVENTEEN FIGURES ACKNOWLEDGMENTS I wish to express my sincere thanks to Drs. Alonzo Mack, J. 8S. Foote, T. J. Dwyer, and C. J. Nemec, for the specimens herein studied,. and to Dr. A. F. Tyler, for the skiagraphs of the skele- tons, and would also acknowledge the helpful interest and sug- gestions of Prof. H. von W. Schulte, Director of the Department of Anatomy in this school. A. ON A PHOCOMELUS WITH ESPECIAL REFERENCE TO THE EXTREMITIES The term phocomelus is derived from the Greek ¢ & x 1, seal, and np éXos, limb. The chief characteristic of monsters belong- ing to this class is an abnormal shortening and arrest of develop- ment of some or all of the long bones of the extremities. The feet and hands are usually composed of the normal number of skeletal elements, but generally appear to arise directly from the 45 46 EBEN J. CAREY pelvic and shoulder-girdles, respectively, which lends them the fantastic appearance of a seal’s flippers. The specimen described was obtained by Doctor Mack, Profes- sor of Obstetrics, Creighton Medical College, February, 1917. It was a full-term, still-born fetus, and parturition was marked by excessive dystocia. The weight was 2500 grams and the crown- rump measurement 25cm. Through the shoulder and pelvic re-. gions it measured 12 and 10 cm., respectively. External form The marked umbilical hernia, in which the coils of the intes- tines show through the attenuated walls at the base of the umbili- cal cord, is readily apparent in figure 1. This is due to arrested development. Normally, five or six primitive intestinal loops, by rapid elongation in embryos between 17 and 20 mm. in length, push their way into the umbilical coelom, producing the normal hernia funiculi umbilicalis physiologica, where they remain un- til the embryo reaches a length of between 35 and 45mm. Soon after the latter period the intestinal loops return to the abdominal cavity proper. The lateral contour of the abdominal and thoracic regions is exceedingly convex, depending upon the excessive size of the liver. Normally, the liver at term is one-twentieth of the total body weight in the still-born, Jackson (’09). The liver in this case, however, weighs 490 grams, or about one-fifth of the total body weight. The excessive size of the liver is due to two factors: first the umbilical hernia; second, the cleft sternum, both operating to release the liver from the confinement which is normally pres- ent. The liver in coelossomic monsters, like the brain in cases of hydrocephalus, shows a tendency to assume an unnatural bulk when relieved from the normal pressure of relational structures. In the embryo the ventral wall of the trunk is at first very thin, and the heart with its various parts as well as the liver and other viscera may be seen through it. From the sides the anlagen of the skeleton and musculature grow into the walls: arrest of this growth may occur and produce ectopia cordis and fissura sterni, TERATOLOGICAL 1 Phocomeius, 2 Phocomelus, 3 Phocomelus, 4 Phocomelus. STUDIES 48 EBEN J. CAREY which depend upon the disturbance in the thoracic region. A portion of the intestine, as noted above, normally projects into the umbilical cord at the time that the outward growth of the abdominal walls occurs. Normally, the intestinal hernia recedes into the abdominal cavity with the further development of the abdominal walls, but if the latter are arrested in their develop- ment, the hernia persists, as is seen in this specimen. The upper extremities are much distorted and resemble in no way, except in the contour of the shoulder and in the digits, the extremities of a normal full-term fetus. The palmar surfaces of the hands are turned mesiad, a retention of the position which is normal in embryos of 18 to 25 mm. in length. The median raphe extending from the anus to the scrotum is well marked. The scrotum is composed of two fat-filled pouches which do not enclose the testicles. These organs are found in the inguinal canal about ready to emerge at the external abdominal ring. The lower extremities also show the degree of rotation char- acteristic of a normal embryo 18 to 25 mm. in length. The knees are directed ventrolateral, while the plantar surfaces of the feet are turned mesial and consequently are opposed to each other. The dorsum of each foot is unusually high. The abducted position of the great toe and the metatarsal pads are especially clearly seen in the right foot. This is characteristic of embryos of about 25 mm. in length. The metatarsal pads normally un- dergo a gradual retrogression and their outlines become indistinct during the fourth and fifth months. The phalangeal pads of the great toe, and in a less degree the second and third toes, of the right foot are clearly shown. The right upper extremity is best seen in figure 2. The shoul- der and wrist are recognizable and a slight groove dorsally in the region of the axilla marks the position of the elbow. The limb is so placed that the extensor surface is directed laterad. The five digits are distinct; the thumb is certainly rudimentary. The hand is unusually broad toward the base of the fingers; this is another early fetal characteristic which has been retained. TERATOLOGICAL STUDIES 49 In the lower extremity the heels are well marked and the foot is extended as a result of the contracture of the gastrocnemius muscle; this occasions the talipes equinus variety of club-foot. In addition, the foot is inverted so that the soles are opposed to each other by the contracture of the tibialis anticus muscle, re- sulting in the talipes varus variety of the club-foot. This is mere- ly arrested development, however, for the position described above is a normal phase of limb rotation and is regularly pres- ent in embryos of 35 mm. The head is abnormally large; the forehead is especially pro- tuberant. This is partly due to arrested development and part- ly to its hydrocephalic condition. The anterior and posterior fontanelles and sagittal suture gape in an abnormal manner be- cause of the distended cerebral hemispheres. The protuberant forehead resembles the condition found in prosencephalic mon- sters. In addition to hydrocephalus, there is an extensive meningeal hemorrhage incident to labor. The root of the nose is deeply depressed, and the nose as a whole is very low and broad. The upper lip projects, whereas the lower one recedes. The depression between the root of the nose and forehead is normal in embryos between 18 and 42 mm, in length, but later this character is effaced. The shoulders are distinctly marked, as seen in figure 3. The protuberance due to the acromial process of the scapula is es- pecially distinct on the left side. The disproportion between the segments of the limbs is striking. In each lower limb a dorsal groove is seen which corresponds to the popliteal space. It is at once evident (fig. 4) that the region of the thighs is greatly shortened, and a corresponding shortening of the proximal seg- ment of the upper limb is also apparent. The left lateral aspect is reproduced to show the symmetry of the surface abnormalities (compare figs. 2 and 4). Externally a tendency is detectable to subdivision of the upper extremity in- to arm, forearm, and hands by grooves which limit these regions. Rudiments of the finger nails may be detected, but these struc- tures have not broken through the overlying epidermis. 50 EBEN J. CAREY In the lower extremity the regional outlines are not as distinct as in the upper. The area corresponding to the knee is directed cephalolateral, but there is no definite demarcation between the dorsum of the foot and the leg. The toe nails and toes are not as well developed as the finger nails and the fingers. The rotundity of the cheeks is quite marked (fig. 4), and well- developed sucking pads were found on dissection. The symmetrical arrest of development of the lower ex- tremities is clearly seen in figure 5. Internal anatomy The section of the alimentary canal, contained in the umbilical hernia, consists of the cecum, appendix, 2 cm. of the ascending colon, and 10 em. of the ileum. The latter possesses a marked Meckel’s diverticulum, from the apex of which a fibrous strand extends into the cord for some distance, eventually to be lost in its connective tissue. The remaining coils of the small intes- tine are arranged in the normal manner. Coil no. 1 forms the duodenum; the secondary derivatives of coils nos. 2 and 3 are found in the left hypochondriae region, those of coil no. 4 are found in the right hypochondriae region, while those from coil no. 5 are located in the left iliac fossa. So far the arrange- ment of the small intestine is normal; however, the derivatives of coil no. 6, instead of occupying the hypogastric region, extend directly to the base of the umbilical cord and enter into the umbilical hernia as noted above. The liver is nearly three times the size of the organ normally found in still-born infants by Jackson (’09), and the spleen is nearly six times its normal weight. The kidneys and splee 1 are also found to be overweight. The pancreas, bladder, prostate, and testicles are about the normal size. The heart is hypertrophied; it is about double the average size. The two lungs are much compressed, being only of about one- half the normal size. The thymus is double the average weight, whereas the thyroid is about normal. To facilitate comparison with conditions in still-born infants of the tenth month, the ) TERATOLOGICAL STUDIES 51 Fig. 5 Phocomelus, caudal view. Fig. 6 Phocomelus, skiagraph from in front. Natural position of lim! Fig. 7 Phocomelus, skiagraph. The arms abducted to show curvature the radii. or Fig. 8 Phocomelus. Skeleton of the upper extremity. 52 EBEN J. CAREY TABLE 1 Relative sizes of the fetal organs of the still-born phocomelus compared to those in still-born infants by Jackson MALE STILL-BORN PHO- MALE STILL-BORN TENTH!} COMELUS TENTH MONTH. MONTH (JACKSON) TOTAL WEIGHT 2500 GRAMS Bi Sn or | Percent Weight. | Percent grams BYAIN Sco: & Po eee ey ee 71 12.91 450.00 | 18.00 THptttunt.osc6 - eee tea oe Sem erg erie 65 0.296 12.5 0.5 Heart. nee kre eRe i as 80 0.69 30.0 1.2 La Abd 6 Fel G2 by eee Re es, Ao Airy eR 69 0.98 10.0 0.4 Dettwdune:<.. 4: baw eee ee aoe 69 0.79 10.0 0.4 PAVED SOR ete Soe ae CE Oe. sR 71 4.81 300.0 12:2 Spleennest s- ssotn Peek Ae ee 70 0.27 40.0 1.6 Hight $adney ss v5 cae cto hae een oa eae 9 0.367 25.0 1.0 Tet ioane ys: cess sue oe Se See eee 9 0.341 25.0 1.0 Riehpsuprarenaleeceeecc eee eee 2 0.101 3.0 0.125 Lei sUpraren al. ccc aces he. eee 2 0.111 3.0 0.125 SED YROUd (bia tes eee oe eee 26 0.111 4.0 0.16 weights of the several organs and their percentage of the total body weight are given in table 1, to which are added the findings of Jackson in his series of still-born infants. Skeletal and muscular systems The scapula is peculiar in that it possesses no glenoid articular cavity; in its place there is a large rounded protuberance which fits into a corresponding cartilaginous depression of the humerus. The body of the scapula is ossified, but the vertebral border, in- ferior angle, coracoid process, and acromion are cartilaginous. A supraspinous muscle arises from the corresponding fossa, but no omohyoid nor levator anguli scapulae muscles are pres- ent. The trapezius and rhomboid muscles are normally lo- cated. From the infraspinous fossa arises an infraspinatus and from the vertebral border the teres major and minor take origin. The fibers of the deltoid and trapezius muscles are, for the most part, directly continuous over the spine of the scapula, but there TERATOLOGICAL STUDIES 53 is a deep fibrous inscription which unites this complex muscle mass to the spine. There is a well-marked cartilaginous sup- raglenoid tubercle for the attachment of the long head of the bi- ceps; the short head of this muscle arises in common with the coracobrachialis muscle from the soracoid process. Both heads of the biceps, the coracobrachialis and the deltoid muscles, fuse to form one large complex which is inserted into the radius, no fibers whatever finding attachment on the humerus. From the infraglenoid tubercle, which is cartilaginous, arises the middle or long head of the triceps; this is the only representative of the normal triceps muscle, the other two heads being absent. The humerus is represented by a small, all but shapeless mass of soft cartilage. On its superior aspect it presents a deep articular cavity into which fits the head on the scapula described above. This head is firm and calcified, but no secondary ossi- fication center is present; absolutely no calcification is found in the humeral mass. No doubt this difference in density is the immediate cause for the reversal of curvature at the shoulder- joint, determining the presence of a scapular head and a humeral articular cavity. Caudal to this cavity the humerus is produced into a rounded process of cartilage surrounded by a dense mass of fibrous tissue, upon which is inserted the teres major and min- or, supraspinatus and pectoralis minor muscles. No brachialis anticus muscle is present. A few fibrous strands extend from the pectoralis major and latissimus dorsi, to the dense perichondrium of the humeral mass, but the major part of the insertions of these two muscles are by tendon into the proximal end of the radius. The common origin of the flexor group of muscles of the fore- arm, for the most part, is from the proximal end of the radius. There is a small direct continuity, however, on the part of the flexor muscles with the biceps by means of muscular slips. Simi- liarly, the extensor group, arising in the main from the proximal end of the radius, has direct muscular continuity with the triceps. The radius is the largest and longest bone of the upper extrem- ity. It is well ossified and bowed in adaptation to the abnor- mal stresses and strains to which it is subjected. At its upper end it presents a concavoconvex articular surface which artic- 54 EBEN J. CAREY ulates with the lower convex surface of the humerus. The ulna does not enter into the formation of the elbow-joint and is merely a cartilaginous bar extending from the upper extremity of the radius to the carpus. The carpal elements, navicular, lunate, triquetral, pisiform, greater multangular, lesser multangular, capitate and hamate, are each definable. They are, however, nothing but cartilaginous nodules presenting but a very faint resemblance to the normal components of the carpus. In the metacarpus and phalanges the normal number of ele- ments are present, but they are abnormally shortened, especially the metacarpals. All are in a cartilaginous state except the ter- minal phalanges in which ossification is beginning at the distal extremities. The proximal articular ends of these phalanges are cartilaginous. The wrist is extended and the hand is adducted towards the ulnar side. It is interesting to note at this point that the flexor carpi ulnaris and the extensor carpi ulnaris are practically one muscle. At their origins these muscles are inseparable. The for- mer is inserted partly into the pisiform and partly into the ulnar side of the base of the fifth metacarpal element. The latter mus- cle is inserted also, on the ulnar side of the base of the fifth meta- carpal element. The greater muscular mass of the extensor group combined with the physiological unity of the flexor and extensor carpi ulnaris muscles explains the extended position of the wrist and the inclination of the hand to the ulnar side. The sternum is widely cleft, indeed, union is present only at its cephalic extremity. This point evidently represents the per- sistent episternal band which has become chondrified. Through this band the clavicles are in direct continuity across the ventral median line. The extent of development of the thoracic walls is comparable to that of an embryo of about 17 mm. in length (Muller, ’06). The sternal ends of the lower eight cartilaginous ribs do not extend medialward beyond the midaxillary line, and in this connection it is interesting to note that both the recto- abdominales are absent. It is highly probable that their develop- ment was initiated, but that subsequently they degenerated, for TERATOLOGICAL STUDIES 09 a rectus sheath was found on dissection containing a mass of adi- pose tissue. The oblique and transversalis are imperfect especi- ally towards the thorax. The nerves of the region appear normal and have the usual course and distribution, so the defect in the skeletogenous tissue would seem to be the important factor in the development of these muscles. Towards the pelvis the muscu- lature is more nearly normal. Two small pyramidalis are pres- ent, extending from the pubis to the rectal sheath, and the caudal portions of the flat muscles are readily defined, but less developed than in a normal fetus at term. Ventrally between the condyles is a smooth cartilaginous ele- vation attached to the femur at the site of the patellas trochlea. The inhibition of development here present would seem to be asso- ciated with tke failure of limb rotation and in particular to the im- perfect condition of the quadriceps extensor, which has alike failed to detach the patella and to bring the limb into normal position. The muscle is represented by a small rectus, associated with which are a few fasciculi on each side corresponding to the vastus medialis and lateralis. With the latter the gluteus maxi- mus is continuous. No traces were found of either the vastus intermedius or the subcrureus. The gluteus medias and gluteus minimus are inserted into the greater trochanter on its lateral as- pect, into its ventrocephalic.surface a fused muscle mass, repre- senting the pyriformis, obturator internus and gemelli, is inserted. The adductors magnus, longus, and brevis are small; separable at their origins, they insert by a common tendon into a ridge im- mediately above the medial condyle. The psoas iliacus inserts into the lesser condyle; the pectineus is attached immediately distal to it. No popliteus nor plantaris muscles are present. The gastrocnemius is a very large mass, and no septal division of this mass is found which would reveal an underlying soleus. The former muscle arises primarily from the dorsal aspect of the tibia, however, a few muscular and fibrous prolongations are found at- tached dorsally to the distal extremity of the femur immediately cephalad to each condyle. dle tibi>, like the radius of the upper extremity, is abnormally curved, with the convexity directed ventrad, but not to such a 56 EBEN J. CAREY degree as the radius. The tibial diaphysis is completely ossified, being separated from the epiphyses by intermediate zones of grow- ing cartilage. There is no fibula. . As a result of the persistent continuity of the patella with the femur, there is no retropatellar extension of the cavity of the knee- joint. Into the cartilaginous knob, which represents the patella, are inserted a few fibrous bundles from the tendon of the very small rectus femoris. Here again emphasis should be placed on the fact that the quadriceps extensor muscle is represented chiefly by a rudimen- tary rectus femoris. The vastus internus and externus possess . but a few muscular strands, which arise from the mesial and lat- eral aspects, respectively, of the greatly shortened femur. The vastus intermedius and subcrureus are absent. We have already noted the fact that the adductors, although present, are very small. Here again the normal stimulus to muscular differentia- tion and development is either absent entirely or minimal. ill the nerves are present. The only absent element which we can discover in the thigh is the diaphysis of the femur. Evidently, then, the teratological evidence indicates that the more rapidly developing skeletal (blastemal-chondrous or osseous) axial zone is the normal stimulus in muscular development, and if it is ab- sent entirely or very much reduced, we find also retarded develop- ment in the musculature. This fact further explains the failure of separation and rudimentary condition of the patella. as cor- related to the inhibited development of the quadriceps extensor. The flexor muscles of the leg form a larger mass than the ex- tensors. This directs the convexity of the tibial bow ventrad. The large gastrocnemius inserts into the cartilaginous calcaneus. The tibialis posticus and the flexor digitortum longus are fused at their origins, but separable at their insertions. The latter mus- cle also gives rise to the tendon of the flexor hallucis longus, which has no independent belly. The tibialis anticus is a large muscle and not only has its nor- mal insertion but distributes the tendons normally belonging to TERATOLOGICAL STUDIES 57 the extensor longus digitorum. These tendons, however, are small; the latter muscle is absent as well as the three peronei mus- cles which normally find their origin on the fibula. In the tarsus the calcaneus, astragalus, cuboid, and navicular are discrete cartilages, but the cuneiform elements form a fused mass. The metatarsals and phalanges are represented by short, thick cartilage rods. The terminal phalanges have ossification centers at their distal ends. This is comparable to the terminal phalanges of the upper extremity (fig. 8). The intrinsic muscles of the plantar surface of the foot were not well defined and formed a common muscle mass in which con- siderable fatty degeneration had taken place. The condition of club-foot in this monster is readily under- stood in reference to the extent of development and contraction of the muscles; those belonging to the tibia are well developed, but the fibular muscles are absent except for some of their ten- dons of insertion which have become amalgamated with the tibial muscles, in consequence the latter group of muscles are practi- cally unopposed in their action. The tibialis anticus manifests its action by inverting, the strong gastrocnemius by extending the foot, resulting in a talipes equinovarus. In concluding this section I wish again to emphasize the fact brought out in this study, that in the complete absence or rudi- mentary development of a part of the skeleton, we also find a com- plete defect or rudimentary development of the related muscles. The converse is also true that an overdevelopment of the skeletal parts is accompanied by a greater degree of development of the related muscles. B. THE EXTERNAL FORM OF AN ABNORMAL HUMAN EMBRYO OF TWENTY-THREE DAYS This specimen was given to the writer by Dr. C. J. Nemec, Instructor in Surgery, Creighton University Medical School, Jan- uary 31, 1917, three hours after miscarriage. The history of the pregnancy is as follows: 58 EBEN J. CAREY The woman, twenty-one years of age, Bohemian, had been married five years. She began menstruating at twelve years of age and invari- ably suffered from dysmenorrhea. Leucorrhea was always evident two to four days before each period. Her time was irregular; it was not unusual for her to miss a period completely. She had given birth pre- viously to three sickly children, the second of which died two months after birth. The last child was born July 31, 1917. During lactation there was the usual condition of amenorrhea. The nursing child was weaned January 6, 1917. Her first coitus since the birth of her last child was on January 7, 1917. On January 25, she noticed a slight flow, the first since her last conception. This condition of metrorrhagia con- tinued until January 31, 1917. At first she thought this was the re- appearance of her normal menstruation. But, on the latter date, a hemorrhagic mass was passed, and then I was called on the case. I saw immediately that the uterine decidua had been passed and I had no difficulty in finding the chorionic sac imbedded in this bloody mass. I put the entire mass in 10 per cent formalin, within one hour after it had been expelled, when I returned to my laboratory. The chorion was found covered with villi 2 mm. in length. The chorionic sac measured 20x 19x10 mm. _ I opened it and found the abnormal embryo with its neck extended contained in an am- nion with the normal amount of liquor amnii. The chorion was sectioned and was found to be normal, save that it was covered with necrotic syncytium. In this syncytium there was a slight round-cell infiltration. The embryo was unbent in the cervical region obliterating the nape flexure. Especial care was taken to preserve the embryo in the exact position in which it was found. It remained in Zenker’s fluid eighteen hours, and excellent preservation was obtained, the surface relief standing out prominently. ° Age. No approximate age in days can be computed from the menstrual history, since the woman was in a condition of amen- orrhea at the time cohabitation and fertilization occurred. The former took place January 7, and the miscarriage occurred Janu- ary 31. Undoubtedly, a living ovum was in the outer third of the Fallopian tube at the time of cohabitation. Allowing twenty- four hours until the time that fertilization occurred, the aborted ovum would be approximately twenty-three days old. The em- bryo is certainly not over one month old, as the external anatomy agrees very closely with the description of embryos between the TERATOLOGICAL STUDIES Fig. 9 Phocomelus. Skeleton of the lower extremity. Fig. 10 Twenty-three-day embryo. A pointer is in the amniotic cavity. Fig. 11 Twenty-three-day embryo. Fig. 12 Twenty-three-day embryo. Fig. 13 Twenty-three-day embryo. THE ANATOMICAL RECORD, VOL. 16, No. 2 Chorion (above) and deciduae Ventral view. Right lateral view. Left lateral view. (below). 6 EBEN J. CAREY twenty-first and the twenty-third days. The embryo stands . very close to the embryo a of His’s normentafeln; embryo 112 of Keibel’s collection, normentafeln of Kiebel and Elze, and embryo G 31 of the Anatomical Biological Institute of Berlin. The speci- men is younger than the twenty-six-day-old normal embryo de- scribed by Mall (91). In Mall’s embryo the eyes are further de- veloped, the liver swelling is more prominent, the branchial arches and clefts are more differentiated, especially the maxillary process of the mandibular arch; the nasal pits are larger, and the limb buds more extended from the body wall. The monster un- der consideration, however, is older than the 4mm. embryo de- scribed by Bremer, which is estimated approximately at twenty to twenty-one days old. In Bremer’s embryo there is no sur- face marking for the eye and no posterior limbs. The shape, size, and degree of development are midway between Bremer’s 4- mm. embryo, aged twenty-one days, and Mall’s 7-mm. nape- breech embryo, twenty-six days old, and twenty-three days in all probability is its age. The measurements of the embryo be- fore fixation are as follows: mm, Height of yolk-sac when it projects from. un bilicus............... 0.9 Maximal height of yolk-sac.... 2.2... --. 22251 oe eee 2.5 Length: of -yolk-sa¢...¢ 0.0. 5c c4.. Se a ne cle eal 4.0 Length of posterior portion of body, measured from the point of emergence of yolk-sac..i...4...-. 2s: 2.76 eee cee eee 0.9 From fore-brain to tip of coccyx following curvature.............. 13.0 Straight line from boundary between neck and thoracic regions to twelfth thoracic segment (nape-breech)...................-.+- 40.0 —_Zrmus Greater length in a straight line (crown-breech).................. 6.0 From vertex to behind the mandibular process..................-. 0.9 Fram vertex to behind the heart........-) 2... sen eaee eee 3.3 External form It is readily apparent from the photographs that the cervical flexure of the embryo has been abnormally unbent. This is fur- ther seen in the sharp groove on the nape separating the neck from the back. There is a very marked degree of curvature no- ticed in the dorsal, lumbar, and coceygeal segments. If the neck TERATOLOGICAL STUDIES 61 and head had not unbent, the embryo would form almost a com- plete circle; the tail would lie in close proximity to the head, if not actually touching it. The body is bent anteriorly and at the same time spirally twisted about its axis so that the head is turned slightly to the right and the pelvic end to the left. The first, second, third, and fourth arches are clearly defined, especially the nodular ventral end of the mandibular arch (fig. 13). The maxillary processes are perceptible on both sides, not to the extent, however, found in Mall’s twenty-six-day old embryo. The second bar is not so bulbous as the first nor the third so prom- inent as the second. The region of the sinus praecervicalis is distinct, but not so depressed as in Mall’s embryo. The fourth arches are barely perceptible on both sides. They lie deep in the groove of the sinus praecervicalis and are covered by the third arches. The clefts begin to show a slight irregularity. Above the branchial region (approximately above the second visceral groove) on both sides there is a small oval depression immediately above the otic vesicle measuring 0.25 mm. in diameter. The head shows the outline of the brain and the marked ele- vation over the region of the Gasserian ganglion. The shape of the cerebral hemisphere, the interbrain, midbrain, and afterbrain are plainly recognizable, and the boundaries of the fourth ven- tricle are sharply defined. From the dorsolateral aspect I was able to define the neuromeres in the lateral walls of the fourth ventricle; these appeared as bilaterally symmetrical transverse folds. The optic vesicles are circular in form on each side and measure 0.3 mm. in diameter. The nasal pits are oval and shal- low, but not as large as Mall found them in his embryo. The mouth is a large shallow pentagonal depression bounded above by the nasofrontal process; below and lateral by the nodular mandibular processes on both sides; and lateral, the minute ele- vations craniad to the mandibular processes, the incipient maxil- lary processes. It is readily apparent that a line drawn ver- tically through the ventral ends of the four visceral arches would be approximately straight and would cut the forebrain some dis- tance in front of the optic vesicles. The marked prominence of the forebrain noticed in the phocomelic monster we see is a very characteristic human feature in early embryos. O2 EBEN J. CAREY ig. 14 Craniorrhachischisis. Ventral view. 15 ( raniorrhachischisis. Right lateral view. ig. 16 Craniorrhachischisis. Skiagraph. Dorsal view. ig. 1i I hig I I Craniorrhachischisis. Skiagraph. Lateral view. TERATOLOGICAL STUDIES 63 The anlage of the heart projected ftom the ventral surface of the body as a large nodular swelling; its prolongation on the right side extends forward as the aortic bulb. If the neck was normally bent, the relief of the aortic bulb would extend to the edge of the mandibular arch. The atrial portion of the heart is seen as a protuberance on the lateral wall through the thin wall of the pericardial cavity. The atrial swelling is more marked on the left side and the swelling of the aortic bulb on the right. Caudal to the heart the vitelline vesicle projected from the um- bilicus; this vesicle was shrunken and pear-shaped. The liver swelling is poorly developed and the tail is curved to the left between the cardiac swelling and the body stalk. The marked coccygeal and pelvic curve is seen from the left as a hook-like process. The tail is conspicuous as is usual in human embryos of this age. In the posterior region of the trunk four parallel ridges are pres- ent; two belong to the axial zones, the medullary and somitic ridges; two belong to the parietal zone, the Wolffian and mar- ginal ridges. There are thirty somites present. The upper extremities are simply oval mounds; they have not become plate-like as yet. The lower extremities are but ill-de- fined ridges. The umbilical cord is large and lies on the left side; a similar condition is found in the embryos described by Mall, Waldeyer, and Janosik, a departure from the usual right-sided position of the cord in human embryos. C. THE ANOMALIES OF AN ANENCEPHALIC MONSTER—COM- PLETE CRANIORRHACHISCHISIS The specimen represented in figures 14, 15, 16, and 17, was given to the writer by Dr. J. S. Foote, Professor of Pathology Creighton University Medical School, in 1914. No clinical data were obtainable. The specimen, approximately eight months, old, weighed 1500 grams, and was 20 cm. in length from the breech to base of skull at its dorsal aspect. The interesting fact in regard to the anencephalic monsters is that they are usually of the female sex; the monster under con- 64 EBEN J. CAREY sideration was a female. he marked abnormalities are the ab- sence of the brain, spinal cord usually, and lack of development of the bones of the vault of the cranium and of the lamina of the vertebral column. Surface anatomy Ventral aspect (fig. 14.): The arms are in an unnatural position- They were forcibly drawn lateral to the lower limbs in order to procure a clearer view of the face and ventral regions. The striking feature is the attitude of the head. It is sunk between the shoulders and extended. Owing to the absence of the cran- ial vault, the face is very prominent. The tongue protrudes from the mouth, the eyes project markedly from their sockets and look upward. This is due to the fact that the forehead is abnormally sloped backward, the supraorbital plates are rudi- mentary and are necessarily tilted in the same direction as the forehead. The nose is broad and flat and the mouth is partly open. The broad shoulders and the general plump appearance of the trunk and the overdeveloped upper extremities present a curious contrast to the deformed head. The excessive development of the shoulders and upper limbs usually gives rise to serious dys- tocia. The abnormal shape of such a head generally leads to face presentation. Owing to the exposed condition of the base of the brain, there is frequently a marked increase in the amniotic fluid. There is a partial development of the frontal, parietal and occipital bones towards the narrow base of the skull. These rudiments slant mesiad and are not prominent. The brain is rep- resented by a conglomerate mass of membranes, blood-vessels, and connective tissue. There is absolutely no trace of nervous tissue of the cerebrospinal axis. These rudiments of the central nervous system, just enumerated, entitle this monster to be classed as a pseudoencephalus, according to Geoffroy Saint-Hilane, who reserves the term anencephalus for those monsters in which ab- solutely no membranous rudiment of any kind is present. TERATOLOGICAL STUDIES 65 No neck is definable. The umbilical cord possessed one vein, but only a single artery, a fact noted by Gillaspie and Henston (17) in the anencephalus described by them. Right lateral aspect (fig. 15). The partial development of the frontal and parietal bones and their marked slope inward is well shown. The exposed membranes, connective tissue and blood- vessels are seen as a protuberance on the dorsocephalic aspect of the head. The well-developed and plump appearance of the trunk and upper limbs are seen in this view, and forms a decided contrast to the abnormal head. The slit on the lateral aspect of the right thigh was made in taking out the femur. Dorsal aspect (fig. 16). | The condition of craniorrhachischisis is apparent. The lack of development of the calvarium, consist- ing of the squamous part of the occipital, parietal bones and fron- tal bone, is well shown. The membrane and connective tissue over the dorsal aspect of the base of the skull and floor of the ver- tebral canal, which is wide open, were left intact. The edges of the peduncles of the vertebra form a continuous ridge on both sides.of the wideopen vertebral canal. These are seen as light linear ridges on both sides of the dark groove. The broad well- developed shoulders stand out prominently. Left lateral aspect (fig. 17). The more marked bulging of the left eye is better seen in this view. The lack of development of the left supraorbital ridges together with its acute slope in- ward causes this eye to look upward and to protrude more than the right eye. The eversion of the right foot and inversion of the left are manifest. The club-foot condition of the former is of the talipes caleaneus variety; of the latter, of the talipes varus variety. Internal anatomy The topographical relations within the thorax were normal. The lungs and thymus showed no marked abnormalities. The heart showed a high degree of defect in its septa both in the atrium and in the ventricle. The septum primum and septum se- cundum of the auricular partition were rudimentary, leaving an abnormally enlarged foramen ovale. The ventricular septum 66 EBEN J. CAREY was represented by mere ridges. The auricular-ventricular valves were quite inadequate and did not function as valves at all. There was evidently marked cardiac incompetency. In the abdomen the intestines, liver, pancreas, spleen, kidneys, and suprarenals were normal. It has previously been pointed out that there was but one umbilical artery. Upon dissection this proved to belong to the left side. A persistence of but one umbilical artery belonging to the right side was pictured and de- scribed by Gillaspie and Henston (’17). In their specimen the uterus was displaced to the left, owing to the fact that the aorta was directly continuous in the median line with the right hypo- gastric artery. In my specimen a similar arterial condition existed in the pelvis with the difference that the persistent um- bilical artery belonged to the left side which in turn caused a displacement of the uterus to the right instead of to the left. Although the muscular system was dissected and studied, no detailed report will be made here except to state that a sternalis muscle was found on both sides in this specimen, but not in the second monster. The arteries to the limbs were normal as well as the peripheral nervous system. The condition of the skele- ton is well depicted in the skiagraph, figures 16 and 17. Note especially absence of the spinal lamina as well as the cranial vault in the dorsal aspect (fig. 16). The base of the skull was acces- sible to the examining finger. The sella turcica and the anterior and posterior clinoid processes were easily palpated. The styloid process of the right side is precociously ossified and throws a dark shadow in the skiagraph. In consequence of the ill-development of the laminae of the cervical vertebrae, there is a lordosis in this region (lateral aspect skiagraph). It is also definitely seen that there is a compensa- tory kyphosis of the upper thoracic vertebrae, extending the head and allowing it to sink between the shoulders, and giving the fe- tus an attitude characteristic of many forms of deficient head and spine development. There is also present a marked kypho- _ sis in the lumbar region beginning at the twelfth thoracic verte- bra and extending to the first sacral vertebra. TERATOLOGICAL STUDIES 67 The position of the mandibula and the protrusion of the tongue, which are really an exaggeration of the first act of deglutition, are to be attributed to the imperfect development of the temporal muscles, allowing the depressors to predominate. Of the tem- poral muscles only the fasciculi arising from the lower part of the fossa are present; these insert upon the coronoid process. They thus are largely representative of the posterior part of the muscle, which is active mainly in retracting the jaw. The large anterior portion, which elevates the jaw, is absent. All the mandibular depressors are present, with only the small masseter and the internal pterygoid to oppose their action. Ac- cordingly, the typical open mouth of anencephalic monsters is correlated to the defect in the cranial vault and the associated loss of the anterior portion of the temporal muscle. D. A SECOND ANENCEPHALUS MONSTER—COMPLETE CRANIOR- | RHACHISCHISIS This specimen was given to the writer in December, 1917, by Dr. T. J. Dwyer, Associate Professor of Surgery, Creighton Uni- versity Medical School. It was prematurely born at eight months and at first presented by the face. However, because of the - overdeveloped shoulder, even more marked than on the forgoing specimen, an obstruction was presented to the descent of the child which called for podalice version. There was a marked condition of hydramnios. Before birth, at six months, the child had been © predicted by Dr. Dwyer to be an anencephalos. This condition was suggested by the hydramnios and the exaggerated intensity of the fetal movements which were also irregular and spasmodic in character. The monster was a female. The condition of craniorrhach- ischisis was more extensive than in the former specimen. The spina bifida extended through the coceyx. Absolutely no rem- nants of the calvarium were present. No membranous rudiments were found and only a slight amount of connective tissue covered the base of the skull. Both eyes bulged even more prominently than the left eye of the first specimen because of the greater slant and arrest of development of the supraorbital ridges and plate. 68 EBEN J. CAREY Two arteries and one vein were found in the umbilical cord. No sternalis muscle was found. Outside of these differences the description of the first specimen holds for the monster under con- sideration; but the abnormalities of the heart are if anything more extensive. Ablfield (’80) assigns as the direct and immediate cause of anen- cephalic monsters, hydrocephalus. If the serum accumulates early within the ventricles, the brain and its covering are ruptured at about the fourth week of embryonic life, they atrophy and dis- appear and the result is anencephalous. The accumulation of serum may make it impossible for the bony case of the brain to enclose the cranial cavity, causing thus varying defects of the skull through which the membranes and their contents protrude. If the serous effusion affects the spinal region as well as the cranial cavity before closure of the neural tube, which is thus prevented, there is an associated spina bifida resulting therefore in the condition of craniorrhachischisis. The primary cause of hydrocephalus can only be surmised. Many have made a vague reference to the already overworked amniotic bands. It is interesting to note that Morgan has ex- perimentally produced spina bifida in the tadpole of frogs by sub- jecting the eggs to a 0.6 per cent solution of common salt. This retards development and results in posterior spina bifida. In re- gard to the underlying cause of anencephalic monsters with spina bifida, Mall (’10) concludes: ‘‘It is no longer*necessary for us to seek mechanical obstructions which may compress the umbilical cord, such as amniotic bands, for it is now clear that the impair- ment of nutrition which naturally follows faulty implantation or the various poisons which may be in a diseased uterus, can do the whole mischief.”’ : TERATOLOGICAL STUDIES 69 LITERATURE CITED AHLFELD, A. 1880-82 Die Missbildungen des Menschen. Parts I and II. Leipzig. BeLuarD, EuGENE G. 1882 Contribution A l’etude des monstres celosomiens. Lille. BrrmMincHaM, A. 1889 On the nerve supply of the sternalis in an anencephalous foetus. Trans. Royal Acad. of Med. of Ireland, vol. 7, Dublin. Bremer, J. 1906 Description of a 4-mm. human embryo. Am. Jour. Anat., vol. 5. Bormer, Emit C. 1887 Anatomische Untersuchung eines Kindes mit Phoco- melie. Marburg. Broca, P. 1882 Note sur les monstres ectromelus. Rev. d’Anthrop., T. 10, Paris. Cuaron, E. 1880 Monstre ectromelien, se rapprochant du phocomelie. Journ. de med., chir. et pharmacal, T. IXX. Bruxelles. Monstre ectromelien, se rapprochant du phocomelie. Presse med. Belge, T. 23, Bruelles. Davis, E. W. 1885 A child born without arms. Med. Herald, Louisville, vol. 4. GILLASPIE AND Henston. 1917 Study of monster with craniorrachischisis. Anat. Rec., vol. 13, no. 5, pp. 289-295. Hater, E. 1847 Monsters with eventration. Edinb. Med. and Surg. Journal, vol. 68. Herve, G. 1886 Sur un cas d’ hemimelle. Bull. Soc. d’ Anthrop. de Paris, hae Hirst, B. C. 1889 A phocomelie monster. Univ. Med. Mag., Phila., vol. 2, p. 151. Hucues, A. W.. 1887 The central nervous system and axial skeleton in anen- cephalous monsters. Lancet, vol. 2, p. 1212, London. Jackson, C. M. 1909 On the prenatal growth of the human body and the relative growth of the various organs and parts. Am. Jour. Anat., vol. 9. Lavuaicne, J. 1883 Contribution 4 l’etude de l’anencephalie; diagnostique pendant la grosse. ‘ Macpovueatt, J. 1878 Foetal monstrosity (phocomelus). Trans. Edinb. Obstet. Soc., vol. 4. Matt, Be P. 1891 A human embryo twenty-six days old. Journ. Morph., vol. 5. Mayer, E. 1882. Acranial monsters with report of a case. Amer. Journ. Med. Sci. Phil., N. S., 83. Mitts, T. W. 1880 Case of congenital ectopia of abdominal organs. Canada Journ. Med. Sci., vol. 5, Toronto. Paterson, A. 1878 Notes of a case of anencephalous foetus born co-twin with a healthy: child. Trans. Edinb. Obstet. Soc., vol. 4. RreBert, H. 1883 Beitrag zur Entstehung der Anencephalie. Arch. f. path. Anat., Berlin, Bd. 93, S_ 396-400. 70 EBEN J. CAREY Sentex, L. 1886-87 Phocomelie accompagneed’entrodactylie. Journ. de Med. de Bordeaux, T. 16. SuepHerD, F. J. 1884 The musculus sternalis and its occurrence in (human) anencephalous monsters. Journ. of Anat. and Physiol., London, vol. 19, pp. 311-319. 1885 On the musculus sternalis occurring in anencephalous monsters. Trans. Acad. Med. of Ireland, vol. 3, pp. 439-446. Dublin. Westsrook, B. F. 1879-80 Microcephalus. Proceedings of Med. Soc. County of Kings, vol. 4, p. 275. Brooklyn. vy. Leronowa, O. 1890 Ein Fall von Anencephalus. Uber den feinen Bau des Riickenmarkes eines Anencephalus. Arch. f. Anat. und Entwicklungs- geschichte. Bd. 10, S. 403-422. ¢ ‘ 1 Ss ~*7) ~ < ‘ ee é . : ~ - .¢ . . > a s 4 Se Q ; ' % ’ ; ~ bad * “ ‘ i ‘- ; 2 % - 7 = 5 ” { 5 . , ’ . Resumido por la autora, Mary Drusilla Flather. La irrigacién sanguinea de las dreas de Langerhans; estudio comparativo del pancreas de los vertebrados. El presente trabajo es el primero de una serie de estudios comparativos sobre la irrigacién sanguinea especializada en las dreas de Langerhans. La introduccién contiene un corto resu- men de los trabajos ya verificados en el campo general sobre el origen y funcién de las dreas insulares y su estructura histo- l6gica. La autora hace un estudio comparativo de la disposi- cién de las células y vasos sanguineos en las dreas insulares del aligator, opossum, caballo, racoon (Procyon lotor), badger (Taxidea taxus), skunk (Mephitis putida),.conejo y conejillo de Indias. Los resultados obtenidos llevan a la conclusién de que, mientras que las dreas vasculares son en extremo variables, in- cluso en el mismo individuo, hay ciertos rasgos distintivos— forma, tamano, red sinusoidea, ete.,— que caracterizan a los islotes de las diferentes especies de vertebrados. El trabajo esta ilustrado con ocho figuras de los ejemplares examinados, dibu- jadas con la cimara clara. La técnica empleada en la obtencién de las preparaciones se describe en el texto. Translation by José F. Nonidez Columbia Univers ty AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, MARCH 17 THE BLOOD SUPPLY OF THE AREAS OF LANGERHANS, A COMPARATIVE STUDY FROM THE PANCREAS OF VERTEBRATES. (PRELIMINARY PAPER.) MARY DRUSILLA FLATHER Bryn Mawr College EIGHT FIGURES INTRODUCTION Since 1895, when the islets of Langerhans were declared by L. A. Shaeffer to be endocrinous glands, secreting a substance capa- ble of modifying the metabolism of carbohydrates in the tissues, ' the study of these organs has been provocative of deep interest and much controversy. It is my purpose in this paper to present a purely comparative study of the specialized blood supply in the islets. Therefore, in my consideration of the work already accomplished in the general field, I shall mention only those facts which are necessary for an adequate comprehension of my prob- lem. A detailed summary of the literature up to 1906 is given by Laguesse in La Revue Generale d’Histologie, vol. 2, 1906-1908. Two important contributions since then are the papers of Lydia M. Dewitt in The Journal of Experimental Medicine, 1906, vol. 8, and of R. R. Bensley in The American Journal of Anatomy, 1911-1912, vol. 12. It is generally conceded that islets are groups of internally secreting glands embedded in the pancreatic tissue of all species of vertebrates. Their origin is still a matter of controversy, although the careful work of Dewitt and of Bensley seems to prove conclusively that the islets and acini arise from common anlagen, later becoming differentiated and incapable of trans- formation one into the other. The cells, varying in form and structure, are always arranged in cords or masses separated by 71 72 MARY DRUSILLA FLATHER large anastomosing blood-vessels. The nature of this vascular network is analogous to that found in other endocrinous glands, especially the thyroid and suprarenals. According to Jordan and Ferguson, certain arterial branches enter the islets and form a plexus of large capillaries from which the blood is drained through the venous system. Dewitt, however, claims that the sinusoids communicate intimately with an interacinar capillary plexus and with larger vessels of venous origin only, basing her theory upon her inability to find near the islets any of the char- acteristic arterial endothelium. The islets vary in size and shape according to their vascular content and according to whether they are singular or compound. Harris and Gow find three distinct types of islets—those which are lymphoid in appearance with many small, deeply stained nuclei in a syncytial mass of tissue; those having distinct cell outlines, and those consisting of compound cell groups divided into smaller areas by strands of connective tissue. It is probable that the first two types represent physiological differentiation only. Usually the cells toward the periphery of the islet are massed, while those in the center are arranged in irregular cords one or two cells in depth resting directly on the walls of the capillaries. As a rule, a connective-tissue sheath surrounds the islet, sometimes penetrating within, and frequently failing to separate the acinous and islet cells completely. It is often diffi- cult to distinguish the islet from the peri-insular zone, but in general the islet cells may be recognized by their polymorphism, their slight colorability, their small size, and the large chromatin content of their nuclei. In addition to the rich blood supply there is also in the islet a plexus of nerve fibers, which was shown by Pensa to pass along the blood-vessels and in between the cells. TECHNIQUE In my own investigations on the histology of the islet cells I am much indebted to Dr. Frederick M. Allen for supplying me with the greater part of the material which I have used. The specimens of alligator, opossum, horse, coon, badger, and skunk BLOOD SUPPLY AREAS OF LANGERHANS io pancreas which I obtained from him had been fixed in either chrome sublimate or Zenker’s solution and embedded in paraffin. From these blocks I cut sections 3 to 5 u thick, and stained them all in Mallory’s connective-tissue stain before mounting them for observation. Fresh material was taken from a guinea-pig and a rabbit. Following Bensley’s directions, I employed all four of his methods of fixation, and used the neutral gentian stain. The most successful results were from the Zenker bichro- mate sublimate fixation. Both neutral gentian and Mallory’s connective-tissue stain proved excellent in the differentiation of the islets and the surrounding tissue. The drawings were made with a camera lucida, using a Zeiss oil-immersion lens and no. 4 ocular, resulting in a total magnification of 1250. THE BLOOD SUPPLY OF THE AREAS OF LANGERHANS A cursory glance at the accompanying figures shows how varied. is the arrangement of the vascular areas in the islets of Langer- hans. To a certain extent there may be variation even in the istets from the same individual, but I shall endeavor to show that there are certain distinctive features which characterize islets in the different species of vertebrates. | In the alligator, figure 1, the islets are noticeably large, com- pound, and syncytial in appearance, with a very appreciable granular content due to great physiological activity. This latter feature may be due merely to the youth of the specimen. With the physiological vascular injection produced by congestion it is easy to see how the distended blood-vessels surround the area, rarely penetrating within it, except in the connective-tissue sheaths, and forming almost a complete barrier between the islet and acinous cells. There is no differentiation between the periph- eral cells and those of the interior of the islet, and no indi- cation of a capillary network. In ten islets from the same individual there was no appreciable variation except in size. Many of the islets were larger than the one represented here. The islet from the opossum, figure 2, presents a quite different appearence with distinct cell divisions, a sinusoidal arrangement THE ANATOMICAL RECORD, VOL. 16, NO. 2 74 MARY DRUSILLA FLATHER of: blood-vessels in the interior of the area, no encircling vessels, and no definite capsule or sheath to separate the islet from the acinous cells. A peculiarity of the cell arrangement not found in any other pancreas observed is the radial grouping around capillaries, which is most suggestive of the radial form of acmous cells about their lumina. These essential features were found in ten islets from the same pancreas. The islet from the horse, figure 3, is definitely ovoid in shape, and is surrounded by a frame of blood-vessels and connective tissue, which in places penetrate within the area. The cells are clearly outlined, but show no differentiation between the central and peripheral grouping in any of the ten islets studied. The distinctive feature of the raccoon islet, figure 4, is the com- pound form and lobular appearance. The smaller masses are separated by a network of blood-vessels and connective tissue which also lies between the acinous cells and the large islet area. Another characteristic is the extensive penetration of the lobules by the capillaries. The cell outlines are fairly distinct, but again there is no differentiation, even where the blood-vessels have invaded the central mass. There was no variation worthy of note in ten islets from the same pancreas. As might be expected from their close relationship, the skunk and badger, figures 5 and 6, have islets with many similarities of structure. In both, the blood-vessels only partially separate the acinous from the islet cells. There is a definite sinusoidal network running through the central mass. This makes it possi- ble for nearly every cell to come in contact with the capillaries, a feature which should greatly facilitate the circulation of the secretion. Figure 6 shows a large accumulation of interlobular connective tissue at one side where the area of the islet reaches the periphery of the lobule. From a study of ten islets from Fig. 1 Island of Langerhans from young alligator. Aq. chrome sublimate Mallory. X 417. Fig. 2 Island of Langerhans from opossum. Zenker Mallory. X 417. Fig.3 Islandof Langerhansfrom horse. Aq.chomesublimate Mallory. X 417. Fig. 4 Island of Langerhans from racoon. Zenker Mallory. X 417. a, acinous cells; b, islet cells; c, blood vessels; d, conneetive tissue. BLOOD SUPPLY AREAS OF LANGERHANS 49 76 MARY DRUSILLA FLATHER —— BLOOD SUPPLY AREAS OF LANGERHANS (4 each individual it was concluded that the islet areas were smaller in the skunk than in the badger pancreas. In the islet of the rabbit, figure 7, the line of demarcation between the islet and acinous cells is difficult to define be- cause the former frequently extend into the acinous area and there is no circle of blood-vessels or connective-tissue sheath to mark the division. The blood supply is sinusoidal in nature. In this form there is a slight indication of cord-like cell grouping and a massing of cells with larger nuclei toward the periphery. Un- like Dewitt, I found in the ten islets examined no radial arrange- ment similar to that which was observed in the opossum. The guinea-pig islet, figure 8, shows a fairly regular contour, a connective-tissue sheath separating the islet and acinous areas, . and an irregular network of insular cells frequently but one cell in depth surrounding the large and abundant sinusoids. After careful observation of islets from two individuals and with due allowance for faulty fixation, I decided that these islet areas were the most sponge-like of any that I have examined. From the islets which I have described I feel that there is an arrangement of the cells and blood-vessels which may be regarded . as characteristic of a species. Within certain limits there may be variation in size and in abundance of capillaries with a conse- quent rearrangement of the islet cells. However, I believe that the special features can be proved peculiar to the species. I realize that the proof is inadequate as yet owing to the fact that with one exception I have studied islets from only one individual of a species. It is my intention to continue the investigation with many more species and more individuals of the species. I am greatly indebted to Dr. David H. Tennent for his helpful supervision of the work. Fig. 5 Island of Langerhans from skunk. Chrome sublimate Mallory. xX 417. Fig. 6 Island of Langerhans from badger. Chrome sublimate Mallory. X 417. Fig. 7 Island of Langerhans from rabbit. Zenker neutral gentian. % 417. Fig. 8 Island of Langerhans from guinea-pig. Zenker neutral gentian. X 417. a, acinous cells; b, islet cells; c, blood vessels; d, interlobular connective tissue. Resumido por la autora, Inez Whipple Wilder. Una anomalia de la circulacién de la porta en el gato. En un gato macho de gran tamafo y aproximadamente de un ano de edad, un espacioso canal sanguineo colateral, formado por la anastomosis de los tributarios de Ja porta con la vena frénica izquierda, hacfa posible el paso directo de la sangre desde dichos tributarios a la vena postcava, evitando de este modo el trayecto normal a través del higado, si bien este trayecto estaba abierto. Con esta anomalia estaban asociados: un aumento de tamano de los rihones y una disposici6n irritable en extremo, por parte del animal. . Translation by José F. Nonidez Columbia University AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, MARCH 17 AN ANOMALY IN THE PORTAL CIRCULATION OF THE CAT INEZ WHIPPLE WILDER Department of Zoology, Smith College, Northampton, Massachusetts FOUR FIGURES While injecting the circulatory system of a cat recently, I noticed that the injection of the systemic veins through the right femoral resulted in nearly filling the hepatic portal tributaries, so that when I came to make the usual yellow injection of the portal system through one of the mesenteric veins, I found these already filled with the blue venous injection mass. Upon dissecting this specimen I found that the hepatic portal vein was small, while there was a very large collateral connection between the hepatic portal system and the posteava. A compari- - son of this aberrant condition (figs. 1 and 2) with the normal condition (figs. 3 and 4) makes it evident that this collateral is formed by the anastomosis of the coronary veins of the stomach with the left phrenic vein, so that the collateral vein thus formed extends along the lesser curvature of the stomach, and enters the postcava at the level of the diaphragm. A voluminous anasto- mosis of the gastrosplenic vein with this collateral furnishes a very direct channel into the postcava from the whole system of mesenteric tributaries, as well as from the gastrosplenic vein itself. Near the junction of the collateral with the main portal vein, the collateral is joined by the combined pancreaticoduodenal and gastro-epiploic veins, thus completing the direct connection of all of the portal tributaries with the postcava through the coronary collateral. The blue injection mass had thus backed into the portal system directly from the postcava and had entered not only all of the portal tributaries, but the portal vein itself and all of its branches to the various lobes of the liver. 79 SO INEZ WHIPPLE WILDER Parsdenal.._ / Gastrs—__ epiplore : = Fragment of great omentum. Fig. 1 Ventral view of dissection of anomalous cat showing the anastomosis of the coronary and gastrosplenic veins with the left phrenic, resulting in the for- mation of a direct collateral drainage from the portal system into the postcava. The diaphragm is represented as slit from the midventral line to the postcava, the liver is lifted, and both liver and diaphragm are drawn anteriorly to display the relationships of the blood-vessels. PORTAL CIRCULATION OF THE CAT S1 It seems probable from the large size of the collateral vein as compared with the unusually small size of the main portal vein, that a large proportion of the blood had habitually escaped its normal course through the liver capillaries, and had entered directly into the main circulation, carrying with it continually an excess of nutritive material and unconverted nitrogenous wastes. | 3 i ee: thea: i 3 Y i, 7} Pren : we S. \ \ Gastro_ oe am +S x Pyloric epi ploic.. ntreatito. ff de ton 0 DN stimach SY Dorsal Surface y)) of Stomach //g Fig.2 Details of the connections of the portal system with the collateral chan- nel in the anomalous individual, shown by lifting the pyloric end of the stomach and carrying it anteriorly and to the left. It could scarcely be imagined that such a condition would not be accompanied by other abnormalities if not by actual patholog- ical conditions. There was, however, nothing unusual in the appearance of the freshly killed specimen, which was a rather large male, well developed, but not unusually fat. Unfortu- §2 INEZ WHIPPLE WILDER nately, the abnormal condition of the circulatory system was not discovered in time to make any histological study of liver or kidneys. In fact, no examination of these organs was made while the specimen was fresh, and as the preserving fluid had not well penetrated the anterior abdominal organs, it was impossible to Poncreatico- duodenal ___' Gostro Memale 3) 5 04 eee eee 470 40.0 8.5 HEMAIOTN fc). eo cee koe eee 415 30.0 oa 2 Female!) Of) See ee 475 38.5 8.1 Memale > Se. tas cee ee 480 37.0 Y DPM AIRS bak hoes Oy «ice ate ee 450 43.0 9.6 | Gravid (ad-_ vanced) WeMMe So. oes cee ee ee 430 Sis0 8.7 Male ts.c2 se ee 480 49.0 10.2 Memalec. |... |. a. ss0stee eee 450 36.0 8.0 | Gravid (ad- vanced) Averike........6t00 eee ee 38.5 8.43 Abnormal individual EE Ee ae ae | 500 | 60.0 | 12.0 | It will be noted that although there is a considerable range of variation in both the actual and the proportionate length of kidneys of the normal individuals measured, in none of these does either the actual or the proportionate length equal that of the abnormal individual. . Inquiries were made to determine, if possible, whether there had been anything abnormal in the behavior of the cat or any indi- cations in its history of a pathological condition. It was learned that the cat was about a year old and had always been peculiarly active and irritable. Even as a kitten it had never tolerated petting, and, to quote the informant, a member of the family of the donor, “it was the strangest acting cat’? he had ever seen. It was at first denied, however, that the cat had ever been patho- PORTAL CIRCULATION OF THE CAT 85 logical, but the admission was finally made that it had had a ‘fit’ a short time before it had been donated to the laboratory, and that this fact, together with the increasing excitability of the animal, had led to its being donated to the laboratory. This account points rather significantly to an inability of the kidneys to cope fully with the extra work devolving upon them as a result of the interference with the full function of the liver. This case would seem also to have some embryological signifi- cance, since, as pointed out by Huntington and McClure (’07) in referring to conclusions based upon the dissection of 605 cats by Darrach (’07) although ‘‘the average individual assumes the venous and lymphatic type considered normal for the species,’”’ by the well-known process of selection and continued develop- ment of certain embryonic pathways while others undergo degen- eration, it has been found possible to ‘‘interpret all of the observed adult variants as examples of atypical persistence of early channels normally destined to disappear in the course of further development, but capable, by continued and unusual growth, of affording all of the variations of the adult venous system observed in the cat.” So far as I know, an anomaly of this particular type has not before been reported. Undoubtedly, however, when the full report of the embryological evidence collected by Huntington and McClure is published, this case will fall into its proper place as one of the variants due to the atypical persistence of embryonic channels. BIBLIOGRAPHY DarracH, WILLIAM Variations in the postcava and its tributaries as ob- served in 605 examples of the domestic cat. Anat. Rec. vol. 1, p. 30. HUNTINGTON, GEORGE S., AND McCuure, C.F. The interpretation of variations of the postcava and tributaries of the adult cat, based on their develop- ment. Anat. Rec., vol. 1, p. 33. Resumido por el autor, Harrison R. Hunt. Anomalias vasculares en un gato doméstico (Felis domestica). Las anomalias vasculares descritas a continuacién han sido observadas en un gato adulto. Las del sistema venoso eran: Una posteava izquierda, venas renales dobles en cada lado, un orificio en la vena iliolumbar izquierda a través del cual pasa la arteria correspondiente, vena espermitica izquierda ramificada desde la posteava y la misma vena del lado derecho ramificada desde una de las venas renales derechas. El uréter estaba rodeando a la posteava. En el sistema arterial, el arco aértico estaba situado en el lado derecho y la arteria innominada en el izquierdo; desde esta Ultima se ramificaba la arteria carétida comin y la sub- clavia izquierda, mientras que la subelavia derecha arrancaba de la base del cayado de la aorta. Translation by José F. Nonidez Columbia University AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, MARCH 17 VASCULAR ABNORMALITIES IN A DOMESTIC CAT (FELIS DOMESTICA) HARRISON R. HUNT West Virginia University ONE FIGURE Recently the writer dissected a male cat which presented so many interesting vascular anomalies that publication of the facts seemed justified. The accompanying figure is a semidiagram- matic representation of the main blood-vessels of this animal as seen from the ventral side. Posterior to the superior mesenteric artery (/2) the postcava’ (9) was situated at the left of the aorta (10).1. The left ureter (28) looped around the postcava in the manner shown in the fig- ure. The position of the spermatic veins was the reverse of the normal position, the left spermatic (23) branching from the post- . cava (9), while the right (20) emptied into the posterior right renal vein (17). Other observers have reported similar abnormalities in the cat. Darrach (’07) has described three cases in which the relations of the postcava, ureter, and sex veins were practically the same as in this individual. McClure (’00) figures (fig. 3) a case in which each ureter looped around the persistent postcardinal vein, as the left ureter passed around the postcava in the accompanying fig- ure. Hochstetter (93) mentions one cat in which the postcava lay at the left of the aorta posterior to the superior mesenteric artery, and a second case (having two persistent postcardinal veins) in which each ureter looped around a posteardinal. Two renal veins (7/7 and 19) drained each kidney (74). Double renal veins were observed by McClure also (’00) in the cat. 1 Unfortunately, my records do not show whether the superior mesenteric and coeliac arteries were on the right or the left side of the postcava. 87 o 0 HARRISON R. HUNT The left iliolumbar artery (26) passed dorsally through a fora- men in the left iliolumbar vein. Such venous foramina have been reported by several other observers (Darrach, 07; MeClure, 00; Treadwell, 96; Weysse, ’03; Smallwood, ’06). The explanation for these anomalies in the venous system must be sought in embryology. Most of the distal portion of the left posteardinal vein in the cat embryo normally degenerates an- teriorly to the level at which the left spermatic vein branches off (Hochstetter, 93, ’06). On the other hand, the right post- cardinal vein persists, becoming part of the postcava. Normally, the right spermatic vein permanently maintains its embryonic connection with the right postcardial. But apparently in this animal the postrenal part of the left postcardinal, instead of the right, persisted as part of the postcava. Consequently the left spermatic vein retained its embryonic relation with the left post- cardinal. Probably the right postcardinal of this animal degenerated anteriorly as far as the right spermatic vein, so that the latter became a branch of the right renal. Hochstetter’s (’93, ’06) observations on the development of the cat’s veins show clearly why the left ureter encircled the postcava in this anima]. In the cat embryo each ureter is surrounded, at a certain stage, by a venous island, consisting dorsally of a supra- cardinal vein, ventrally of the postcardinal (Prentiss, °15, fig. 274). In the cat, according to Hochstetter, and in other mam- mals only the dorsal, or supracardinal, limb of this island persists as a part of the posteava. In this particular cat, and in similar cases which have been reported, doubtless the supracardinal limb of the island degenerated, and the postcardinal limb survived as a part of the posteava, causing the ureter to pass around the dorsal side of the postcava (Metcalf, ’18). Probably the most infrequent abnormality in this cat was the position of the aortic arch. Normally it lies on the left side of the animal, but in this case it was on the right side. The left subclavian artery (4), instead of connecting directly as usual with the aortic arch, came from the distal end of the unusually short innominate artery. The right subclavian (6) branched VASCULAR ABNORMALITIES IN DOMESTIC CAT 89 Fig. 1 Ventral view of the arteria! and venous systems. The hepatic veins have been omitted. 1, left innominate vein; 2, precava; 3, left common carotid artery; 4, left subclavian artery; 5, right common carotid artery; 6, right sub- clavian artery; 7, azygos vein; 8, heart; 9, postcava; 10, aorta; 11, c-eliac axis; 12, superior mesenteric artery; 13, right adrenolumbar vein; 1/4, left kidney; 15, left adrenolumbar vein; 1/6, right renal artery; 17, right renal veins; 18, left renal artery; 19, left renal veins; 20, right spermatic vein; 21, right spermatic artery; 22, left spermatic artery; 23, left spermatic vein; 24, inferior mesenteric artery; 25, iliolumbar vein; 26, iliolumbar artery; 27, right ureter; 28, left ureter. THE ANATOMICAL RECORD ,VOL. 16, No. 2 QO HARRISON R. HUNT from the aortic arch at the base of the innominate. In man also the aortic arch occasionally occurs on the right side (Cunning- ham, ’09). Mammalian embryology furnishes an explanation for the anomalous position of this arch. In the normal development of mammals the right fourth arch degenerates, leaving the left fourth arch to carry the blood from the heart. Probably the left fourth arch of this animal degenerated, as in birds, leaving the arch on the right side as the permanent blood channel. It would be interesting to know whether these vascular abnor- malities were due to hereditary or environmental influences, to one cause or to a chance combination of several independent causes. All these anomalies were not produced by the disap- pearance or unusual modification of the normal activity of one Mendelian factor, for all these abnormalities seldom occur to- gether in one animal. Nevertheless, the riormal and abnormal conditions of these vessels may possibly be Mendelian characters, the occurrence of so many abnormalities in one animal being a very unusual chance combination of characters. Or, possibly, the developing embryo was subjected to unusual environmental conditions, such as abnormal amounts of certain substances in the mother’s blood. These conditions may have slightly modified the development of the body as a whole, but produced most pronounced abnormalities in the blood system. Investigations in genetics or experimental morphology might furnish a satisfactory explanation. Morgantown, West Virginia, December 6, 1918 VASCULAR ABNORMALITIES IN DOMESTIC CAT 91 LITERATURE CITED CunnincHaM, D. J. 1909 Text-book of anatomy. Third edition. Wm. Wood & Co. DarracH, W: 1907 Variations in the postcava and its tributaries as observed j in 605 examples of the domestic cat. Anat. Rec., vol. 1, p. 30. HocustetTer, F. 1893 Beitrige zur Entwickelungsgeschichte des Venensys- tems der Amnioten. III Sauger. Morph. Jahrb., Bd. 20, S. 543-648. 1906 Die Entwickelung des Blutgefasssystems. In Handbuch der Vergleich- enden und Experimentellen Entwickelungslehre der Wirbeltiere. G. Fischer, Jena. Bd. 3, Teil 2, S. 21-166. McCuvre, C. F. W. 1900 On the frequency of abnormalities in connection with the postcaval vein and its tributaries in the domestic cat (Felis domes- tica). Am. Nat., vol. 34, pp. 185-198. Mertcatr, H. E. and K. D. 1918 Persistence of the posterior cardinal veins in an adult cat. Anat. Rec., vol. 14; no. 1, pp. 123-126. Prentiss, C. W. 1915 Text-book of embryology. W. B. Saunders Co. Smatiwoop, W. M. 1906 Some vertebrate abnormalities. Anat. Anz., Bd. 29, No. 16 und 17, S. 460-462. TREADWELL, A. L. 1896 An abnormal iliac vein in a cat (Felis domestica). Anat. Anz., Bd. 11, No. 23 und 24, S. 717-718. Weysse, A. W. 1903 The perforation of a vein by an artery in the cat (Felis domestica). Am. Nat., vol. 37, pp. 489-492. Resumido por el autor; Ezra Allen. Degeneracion en el testiculo de la rata albina a consecuencia de una dieta deficiente en la vitamina soluble en el agua, con una comparacién de una degeneracién semejante en ratas tratadas de un modo diferente y una con- sideraciOn sobre el tejido de Sertoli. Las ratas albinas sometidas por Osborne y Mendel a una dieta deficiente en la vitamina soluble en el agua son estériles. El] eximen de los testiculos ha demostrado la degeneracién com- pleta de las células germinales. Tan solo persisten en los tubu- los las células de Sertoli, si bien sus nicleos aparecen muy con- traidos. El tejido intersticial estaba hipertrofiado. Estos car- acteres son los mismos que se encuentran en los testiculos de otros mamiferos sometidos a la accién de los rayos X. Una de- generacion semejante ha sido observada también en algunas ratas alcoholizadas por MacDowell, degeneracién que se pre- sentaba también, aunque en menor grado, en los hermanos de dichas ratas que no fueron sometidos a la aeccién del alcohol. Bajo estas condiciones el tejido de Sertoli.revela una estructura sincicial que el autor del presente trabajo considera como el estado normal, como demuestra ‘el material bien fijado. Translation by José F. Nonidez Columbia University AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, MARCH 17 DEGENERATION IN THE ALBINO RAT TESTIS DUE TO A DIET DEFICIENT IN THE WATER-SOLUBLE VITAMINE, WITH A COMPARISON OF SIMILAR DE- GENERATION IN RATS DIFFERENTLY TREATED, AND A CONSIDERATION OF THE SERTOLI TISSUE EZRA ALLEN The Wistar Institute of Anatomy and Biology SEVENTEEN FIGURES CONTENTS ELD DALE 2 eS Pe ee ee be 93 SPS TL Ss eR a eS 94 A eee int SP ee 95 4. Obsérvations upcn the Osborne and Mendel rats....................... 96 _ FUEL USELESS ee ne ee 96 The Sertoli tissue in the degenerate tubules........................: 97 Other structures...... ee eR A BIS a diol -! Sis aha: eigen HS shcreve’c EMe + oes 98 MCR ENeECdS OF GEPCNELALION.. 00. 2... cee nese boc eawevucecbclesess 100 2 LLU RS ee ee er 101 caopservations upon the MagDowell rats............0:.0.eccececc ec ceees 102 mereHersiiow im the MacDowell rats... 2.05.00. ck deca k cee ee cece eeeee 102 The order of degeneration among the germ cells..................... 104 ate RAPE Ores T UCN HIS IER 2 25 AS cies. Sass (hs Allo sw huals Siam loc ce Cov veccedes 104 6G. Conclusions with regard to the two sets of rats................-.-..0-. 104 7 PELE doce 2 ead BR Sil eel eRe eae 105 SMe rMenpOle hinge 48 9 SYNCY IW gsc. 5 ccc... cs ccc e cscs e eee cece enews 107 Steen MMIII GUS 98 em. a cla a oa sh Sinise sucrose wdc esos es ceees 108 SR Mies PEI TRSSUIE | | ae Sa rales cote ch aoe o's ws tse ee oe oe see cene went 110 Pe MER CONSIGETALIONS.......< ++ sscccsacincsaescevccescccacesese 110 SUMNER hk eo oa dav evecs cee ce tees Se. SE EA Pee 111 RRC eae t,o es Pa eins ceed c ieee eveuns 112 MRE aEIOUR I BOG DIATES:, 1... c a od cv ciece moos seem ceemeeweresccceeeecdes 113 INTRODUCTION This paper deals with a similar type of degeneration found in the testes of two groups of albino rats which had been subjected to very different treatment. The histological and cytological 93 94 EZRA ALLEN conditions revealed upon examination are of sufficient interest for description not only for the light they throw upon certain problems connected with the Sertoli tissue, but also because they indicate the value of further experiments along the lines of their causes. THE MATERIAL I am indebted to Professors Osborne and Mendel for one lot of rats and to Dr. E. C. MacDowell, of the Carnegie Station for Experimental Evolution at Cold Spring Harbor, for the other. There were four in the first lot and ten in the second group. For convenience I will refer to the two lots as the Osborne and Mendel and the MacDowell rats. The first lot were chosen at random from a larger group which had been subjected to a diet deficient in the water-soluble vitamine. All of these rats, both male and female, had proved sterile. Aside from their sterility they were very well developed. The data of especial interest as to their development will be found in table 1. The MacDowell rats came from a lot which have been referred to,in a paper published by MacDowell and Vicari (17) dealing with reduction in fertility among rats subjected to alcohol. Five of the alcoholized rats, from five different litters, and their normal brothers were sent to me. The alcoholics had been made drunk daily beginning at twenty-eight days of age. Each treatment was continued long enough to produce not only TABLE 1 Showing the body measurements, comparison of weight of testes with standard on body length, and age at death of the Osborne and Mendel rats : STANDARD wewpen | wetomr | J O,| boo” | “aemee” [OPHSCESON| equa | LENGTH grams So eke 4s mm. aioe a errs months 3554 316 400 230 0.975 2.278 24 3610 325 438 237 0.690 2.926 22 3756 348 397 218 0.851 2.525 20 3896 288 414 222 1.334 2.609 18 DEGENERATION IN THE RAT TESTIS 95 inability to stand, but frequently to render the rats absolutely motionless. This meant long treatments, since as they became habituated they could stand larger and larger amounts before they succumbed. All ten had been staid to a special diet for fifteen days, beginning when they were seventy days of age. It consisted solely of white bread soaked in fresh milk. The rats were allowed to eat of it for thirty minutes the first two days, for fifteen minutes the next five days, and for five to ten minutes the next eight days. Data for these rats, corresponding to those for the Osborne and Mendel lot, are to be found in table 2 TECHNIQUE The two sets of rats were treated substantially alike, although the MacDowell rats were killed in May, 1917, and thé Osborne and Mendel in April, 1918. After etherization they were measured and weighed. The testes were then removed and, with the exception of Osborne and Mendel’s nos. 3554 and 3756, dropped into the fixative, B-15 at about 38°C., after being cut - into small pieces by scissors. The fixative had been weighed previously, so that a second weighing with the testes in it gave data for determining their weight. The two rats, nos. 3554 and 3756, were injected by the fixative after washing out the blood- vessels with Locke’s solution. For the sake of uniformity in weights, the testes of all the Osborne and Mendel rats were weighed after fixation. — The fixative was replaced with 70 per cent alcohol by the drop method, the picric acid washed out with the help of lithium carbonate in 70 per cent alcohol, and dehydration completed in anilin oil. Clearmg was by oil of wintergreen. Both oils were added by the drop method. Infiltration by paraffin was brought about very gradually. Details of this treatment are described in my paper on technique (Allen, ’16). Sections were cut at 7u and 10, and stained with iron haematoxylin and acid fuchsin or orange G, 96 EZRA ALLEN OBSERVATIONS UPON THE OSBORNE AND MENDEL RATS In the case of all four rats the tunica albuginea was found to have an excess of a clear, serum-like fluid, which flowed out as soon as the tunic was ruptured. The solid part of the testes was much shrunken. The fluid had distended the organs so that they looked quite normal before the tunic was ruptured. Upon examination of the sections certain conditions common to all four were revealed. Some minor variations in the different individuals were present. The common conditions will be de- scribed first. These may be enumerated as 1) no mature sper- matozoa; 2) almost no normal germ cells in any stage; 3) generally speaking, an absence of spermatogonia, spermatocytes, and spermatids; 4) the Sertoli tissue the most prominent of any tissue in the tubules, and 5) an increased quantity of interstitial tissue as compared with the normal. An idea of the general appearance of the tubules and the interstitial tissue may be had by examining figures 2 and 9. It will be noted that the germinative epithelium is almost entirely lacking or very abnormal, the latter state due to the presence of numerous cavities. Figures 5 and 6 show these conditions better. Sometimes the tubules are apparently solid, no lumen appearing, and the cavities just referred to either lacking en- tirely or, if present, very small (figs. 1 and 8). In either case the contents of the tubules are composed chiefly of Sertoli nuclei and their syncytial cytoplasm. Since this tissue is the most prominent, it will be described first, after a brief considera- tion of its normal structure. . Normal Sertoli tissue In normal tubules the syncytium is difficult to see, on account of the closely packed germ cells, but under favorable conditions in the section very small areas may be discerned. Close to the Sertoli nuclei the cytoplasm is clearly observable, but no cyto- plasmic walls are to be seen. In character the syncytial cytoplasm in both the normal and in these degenerate conditions resembles the substance which fills the tubules in the early DEGENERATION IN THE RAT TESTIS 97 stages of their development. See my figures 2 to 8, Allen (18). In the adult it is best seen when the groups of spermatozoa have nearly matured and are migrating toward the lumen from the immediate neighborhood of- the Sertoli nuclei along the basement membrane. These groups of spermatozoa then seem to have dragged with them some of the Sertoli cytoplasm, which under favorable conditions may be seen extending nearly across the. entire width of the germinal epithelium. It is distinguished by its coarse nature and its heavy staining character. Even then no cytoplasmic wall is visible. This appearance is shown in figure 14. The Sertoli nuclei are nearly always in close relationship with the basement membrane (fig. 14). They may be distinguished by the nucelolus, which is well illustrated in figures 14 to 17, and by the less active reaction of the nuclear plasm to haematoxylin as compared with that of the spermatogonia. The Sertoli tissue in the degenerate tubules To return now to the degenerate tubules under consideration: the nuclei are irregular in outline, due to their membranes having wrinkled and formed grooves of various sizes (figs. 11 to 13). Several small bodies staining like chromatin are present. Two, twin-like, are the most prominent. These are shown in figures 11 and 13. In the normal Sertoli nuclei these bodies are very unequal in size, as shown in figures 14 to 17. Inter- mediate stages have been traced between the normal and degenerate condition with respect to this duplex body, so we may be confident that such nuclei as shown in figures 11 and 13 are Sertoli cell nuclei. In such a stage of degeneration as shown in figure 6, these nuclei are very numerous and are scattered throughout the tubule. In the more advanced stage, as illustrated in figure 5, they are confined to the periphery, and are reduced in size and . in number. A rough idea of this diminution may be obtained by a comparison of the tubules photographed for figures 5 and 6. They are from the same section, which was cut at 7x. QS EZRA ALLEN In the tubule shown in figure:6, fifty-one nuclei are to be counted in the section; in the one shown in figure 5, thirteen are present. In neither case are any other kinds of cells to be seen within the tubules. The true germ cells have completely disappeared. The cytoplasmic substance stains with orange G and with acid fuchsin more lightly than the same substance in the young tubule, but seems very similar in structure. It may be de- scribed as a loose meshwork, the substratum of which seems to be minute, irregular ‘granular’ bodies, the same sort of sub- stance as found in the normal Sertoli cytoplasm. It is either continuous, except as interrupted by the nuclei, or broken by cavities, figures 3 B, 6, 8, 9, and 11. In one respect it differs somewhat from the cytoplasm of the normal Sertoli tissue, or of the young tubule, in that it seems somewhat stringy or thready. This appearance is partly due to degenerate spermatozoan tails and partly to the tendency of the syncytial substance itself to form a more or less fibrous structure, which bears some resemblance to the fibers of connective tissue, but is much more delicate and less sharply defined. In figure 1], this characteristic is to be noted, particularly in the lower left- hand corner. In the normal tissue, when the spermatozoan ~ heads have migrated well toward the lumen, this appearance is suggested to the observer, although the stringiness is much less well defined. Other structures From the quantitative. point of view, the next structures to be described are the immature spermatozoa. These may be so abundant as to divide the space about equally with the Sertoli tissue or they may be relatively few. In the former condition they give to the tubule the appearance illustrated in figures 1 and 7. With higher magnification such a tubule appears to be filled with a more or less fibrillar mass intermingled with ‘a granular substance, the whole taking any stain lightly, scattered through which the shrunken Sertoli nuclei are quite abundant (fig. 6). The fibrillar portions resolved into tails of spermatozoa, the heads of which, only partially formed, do not DEGENERATION IN THE RAT TESTIS 99 stain prominently with nuclear stains. These immature sper- matozoa are in bundles or groups, but so intermingled that it is out of the question to determine whether the number in each bundle is less or greater than normal. In this type of degenerative condition, cavities appear which vary in size from one equal to two or three Sertoli nuclei to one whose diameter is half that of the tubule or even greater (figs. 6 and 7). Usually, however, with the increase in size of these cavities the immature spermatozoa disappear. There are also to be seen degenerate nuclei of a type other than those pre- viously described. These are of various sizes, some smaller and some larger than the Sertoli nuclei. They stain yellowish in the iron haematoxylin and acid fuchsin preparations. Their con- tents consist of granular masses of unequal size, more or less densely aggregated. The medium-sized ones may have a clear space between such a unified mass and the nuclear membrane. No cytoplasm is discernible. Their identification is doubtful. They may be either remains of spermatid nuclei or degenerating Sertoli nuclei. That they may be the latter is indicated by the numerical relationships already referred to on page 98. In this connection an individual difference in the case of one rat, no. 3554, may be noted. Degeneration seemed uniform but incomplete. In nearly every tubule the condition was that just described, that is, the tubules filled with the mixture of Sertoli tissue, the immature spermatozoa, and degenerate nuclei which stain yellowish. It is typically shown in figures 1 and 8. Scarcely a tubule was found showing degeneration to the degree illustrated in figure 5, and only a few with cavities as large as in figure 6. No spermatogonia, spermatocytes or spermatids were found. Along the basement membrane, however, in a -very few tubules dividing cells are present, one ‘or two to a tubule per section. These cells resemble dividing spermatogonia or the undifferentiated embryonic nuclei. In the dividing cells just referred to the chromosomes and spindle are normal. One case of another type of dividing cell was seen, situated similarly, in which the chromosomes while abnormal resemble the first spermatocyte metaphase forms, 100 EZRA ALLEN although the stage of division cannot be positively determined. No idiosome nor chromatoid body appears. In the case of the other rats, cell division was rare in no. 3610, and abnormal; it was more abundant in no. 3896, and often normal. It was found in the ceils along the basement mem- brane. Some cell division was observed in no. 3756, also along the marginal layer of cells. The chromosomes were normal and characteristic of spermatogonia in anaphase. In a tubule show- ing degeneration advanced to that typical of no. 3554, one such cell was found. The degeneration in rat no. 3554 is much more uniform than in the others. The least uniform was no. 3756. Figure 3 shows that adjoining tubules may differ widely in this testis. Tubule A is practically normal; B has reached the stage a little later than that characteristic of no. 3554, while C is in the last stage observed. Only four tubules as nearly normal as A were found. This condition is therefore exceptional. In no. 3896 more spermatocytes in early stages of development were found than in any other. These cells seemed to be normal. Growth has in some cases reached the leptotene stage. In the tubules showing these spermatocytes no immature spermatozoa were to be seen, but the wrinkled Sertoli nuclei were scattered freely in their characteristic cytoplasmic substance, and in some cases many cavities of large size appeared. Occasionally pyecnotic nuclei have been found in both the germinal and the interstitial tissue, but these are likely to be found in normal tissue. It is difficult to say whether the number of these in this degenerating tissue is more or less than the normal, as considerable variation exists in the normal. It has not impressed me as abnormal. In fact, I have been surprised that degeneration has taken that form so seldom. The progress of degeneration From the foregoing observations it would seem that degenera- tion had begun in the different individuals at different times or that different tubules reacted very differently to the diet. The condition found in no. 3554 would indicate that degeneration DEGENERATION IN THE RAT TESTIS 101 had not begun until the spermatozoa had started to form, after which all the other germ cells had degenerated. Whether this lot of immature spermatozoa represent the first crop, which has persisted through the degeneration of the other germ cells, or the second or a later crop cannot be determined from the con- dition of the testes at death. The uneven degeneration found in no. 3756 indicates that certain tubules were almost immune to the condition which brought’ about degeneration. These tubules are, however, very few indeed, so few that perhaps they may almost be neglected. At the same time the tubule shown in figure 3, A, shows all stages of spermatogenesis up to the spermatid formation. A similar stage in young material is normal. In the case of rat no. 3896, in which complete degeneration of | the germ cells and great reduction of the Sertoli nuclei were the rule, some tubules showed spermatocytes as far advanced as the leptotene stage. This last-named condition would indicate that for some reason these particular spermatocytes had been pre- served during the degeneration of the Sertoli nuclei. It is not likely that they could have advanced much farther in their development. Certainly, the spermatozoa could not mature upon such pathological nurse cells. The interstitial tissue As previously noted, the interstitial tissue is relatively much greater than in the normal. This abundance is well shown in figures 1 and 2. The only cytological abnormality is that many of the glandular nuclei are slightly irregular in outline, often shrunken considerably. They react equally to the stain, while often in normal testes the reaction is quite unequal. Rat no. 3504 showed the shrunken and irregular nuclei much more abundantly than any of the others. In all the rats the amount of connective tissue seems about the same and quite normal. The endothelial nuclei are also normal. The interstitial tissue is unequally distributed, in some cases occurring in large masses. See figure 1 near the blood-vessel. Dividing cells are occasionally found. 102 EZRA ALLEN OBSERVATIONS UPON THE MacDOWELL RATS These rats, like the Osborne and Mendel lot, were in general about normal in development, with the exception of the testes of the five alcoholics, three of which, as shown by table 2, were under weight, one about normal, and one (no. 764) considerably over weight. In body weight based upon body length, there is a slight variation from the standard, from eleven under to fifteen grams over weight among the entire number. ,On the whole, the alcoholics ran a little over weight. The only two under weight were nos. 687 and 767, respectively eleven and five grams. Degeneration in the MacDowell rats The degeneration in these rats, while similar in kind to that of the Osborne and Mendel lot, varied greatly in degree in the different rats. Some testes were almost or quite normal, others slightly abnormal, while one showed almost as advanced a stage as that of the Osborne and Mendel rats. Unlike the first lot, however, degeneration of any degree had not extended equally throughout the organ. It was confined to certain portions only, like islands in normal tissue, these varying in size. They were most numerous and largest in no. 704. In fact, very little of this rat’s germinal tissue was normal. The final stage, when present, is confined to small portions of the tubule concerned, unless degeneration has progressed to a very advanced stage, as in no. 704, in which case considerable lengths of the tubule are involved. This last-named condition is shown in figure 7, the tubules in which are representative of the testis from which it was taken, no. 704. In the early stage, the signs of degeneration are confined to small cavities which appear in the germinative wall, such as shown in figure 10, C. At a little later these cavities have enlarged and appear as in figure 7, C. At a later stage still, the cavities have enlarged, the germ cells are scattered through the lumen and are abnormal in various ways. The Sertoli nuclei usually remain near the basement membrane, but in advance stages are always to be found there. The Sertoli cytoplasm is shreddy and fills the DEGENERATION IN THE RAT TESTIS 103 TABLE 2 Showing in A the body measurements, comparison of weighi of testes with standard on body length, length of alcohol treatment, age when begun, age at death, and degree of degeneration of the MacDowell rats, and in B the corresponding data for the normal brothers | | : STAND- AGE | LENGTH ‘ LENGTH z WEIGHT ARD WHEN LEN OF AGE AT | DEGREE OF NUMBER|/ WEIGHT |OF BODY) erie] OF WEIGHT | ALCOHOL z : |OF BODY =: fi | 4 TREAT- | DEATH DEGEN- + TAIL TESTES |ON BODY, WAS aneatea ASTON LENGTH BEGUN _ ae grams mm. mm. grams grams days days days 614 198 391 201 | 2.052 | 2.094 43 155 225 | Slight 686 | 171 372 192 | 2.124 | 1.964 28 97 162 | Slight 704 | 122 334 163 | 1.914 } 1.313 38 83 158 | Extreme 725 | 142 347 176 | 1.963 | 1.609 28 83 148 | Medium 764 | 191 370 196 | 1.778 | 2.951 28 74 142 | Medium B Normal brother of 618 | 279 422 220 | 2.075 | 2.567 614 225 | Normal 687 | 269 410 222. | 2.574 | 2.609 686 162 | Slight 707 | 211 380 202? | 2. 822"| 2.181 704 158 |. Medium 727 | 185 350 192 | 2.240 | 1.964 727 148 | Slight 767 | 210 394 206 | 2.049 | 2.267 764 142 | Slight interstices between the germ cells, just as in the Osborne and Mendel specimens, presenting the same appearance as that shown in figure 6, or, for the most advanced stages, as that shown in figure 5. The Sertoli nuclei are not as shrunken and wrinkled as in the Osborne and Mendel specimens, nor is the nucleolus so markedly different from the normal, seldom showing an appearance like that in figure 13. In many places cells in various conditions appear, some normal, some polynuclear, some degenerating by pycnosis and others by a process in which the chromatin occurs in very small, lightly staining granules, yellowish in color with haematoxylin. Many of these may be identified as first spermatocytes, some in the growth stages and others in division. Giant cells occur frequently in the earlier stages of degeneration. 104 EZRA ALLEN ‘The tubules which show advanced degeneration are smaller in diameter than the normal, although this difference in size is not always as great as shown in figures 7 and 10. The order of degeneration among the germ cells In order of degeneration the spermatocytes seem to be the first affected, the spermatids next, and the spermatogonia last, although in some places the early growth stages of the first spermatocytes are found with spermatogonia, but no spermatids. In many places the spermatogonia persist along the basement membrane along with the Sertoli nuclei, but all other germ cells have disappeared. Under these conditions these spermatogonia nuclei are somewhat shrunken and wrinkled. I have not determined whether the second spermatocytes are more sus- ceptible than the first. The interstitial tissue In these animals the quantity of interstitial tissue does not seem to be increased, nor have any shrunken muclei been noted even in the most advanced degenerate conditions of the tubules. Further experiments are needed to demonstrate why this differ- ence in this respect should appear between the two sets of rats when the germinal tissue is affected in a like manner. CONCLUSIONS WITH REGARD TO THE TWO SETS OF RATS The facts just set forth show that certain agents have a selective action upon the development of the germ cells. From the Osborne and Mendel rats we must conclude that for normal growth of these cells a diet containing a generous supply of the water-soluble vitamine is an essential. Just how much remains to be determined by further experiment. From the MacDowell rats studied, a conclusion is not so easily drawn since two factors entered into their history. They were all subjected to a reduced diet for fifteen days, beginning when they were ten weeks of age, while half of them were also subjected to the fumes of alcohol for a period long enough to DEGENERATION IN THE RAT TESTIS 105 produce complete intoxication daily during a time varying from seventy-seven to one hundred and fifty-five days. Neither of these treatments interfered with these rats’ producing offspring, although the whole group of alcoholics from which they were taken showed a considerable reduction in fertility (MacDowell and Vicari, ’17), to the extent that twenty-nine pairs of normals produced 300 young, while during the same time thirty pairs of alcoholics produced only 108 young. How far the male is responsible does not yet appear, but it is clear that in the ten rats of the series which came to me the degeneration is con- sistently greater in the alcoholics than in their control brothers, both from the extremes and from the average, as is brought out in table 2. Under the circumstances, it is unwise to draw any general conclusions as to the cause of the degeneration in these MacDowell rats. What does seem clear from a comparison of the two lots under consideration is that similar conditions of degeneration may arise in the testes of rats subjected to widely different treatments, and that the immediate causes affecting growth and cell division in the germ cells may be identical. DISCUSSION, The discussion is divided naturally into anatomical and physiological considerations. With respect to the former, the chief interest centers about the type of degeneration and the light it throws upon the Sertoli tissue. The type of degeneration is not new. Regaud (’01) de- scribes similar conditions in the tubules of the white rat near their distal extremities, but does not state a cause. The same author (10) and Barratt and Arnold (’11) find a similar type of degeneration in mammalian testes which have been subjected to x-rays. Colwell and Russ (715) have assembled the effects of x-ray treatment upon tissues and note nothing different in connection with the testis. They quote chiefly Regaud and the work of Barratt and Arnold, just referred to, and reproduce the latter’s figure 30 as their figure 40. THE ANATOMICAL RECORD, VOL. 16, no. 2 106 EZRA ALLEN Regaud (°10) worked with rabbit, guinea-pig, mouse, cat, and rat, and obtained essentially similar results with all. His figure (from cat) shows a condition practically identical with my figure 5. Barratt. and Arnold (’11) used the rat. Their figures 30 and 31 represent the same phenomena as my figures 11 and 6, respectively. The descriptions given by the three authors just quoted indicate that there is no difference in the final histological picture—total destruction of the germ cells, the Sertoli tissue alone remaining within the tubules. Barratt and Arnold claim an increase in the number of Sertoli nuclei found in the final stages as compared with the earlier, but give no data upon which such a conclusion has been based. My own observations indicate the reverse, as already noted under ‘Observations.’ Barratt and Arnold indicate that this increase is accomplished by amitotic cell division, a process which I believe does not exist in the tissues which I have examined, either Sertoli or germinal. Barratt and Arnold note a diminution in size and edematous state of the testes, the clear fluid running freely upon incision of the tunic, two conditions which I found in the Osborne and Mendel rats. With regard to the order of cell degeneration, Regaud (’10) finds that the very young spermatocytes and the last generation of spermatogonia are very sensitive so that they are entirely destroyed. Further than this he does not analyze the order. He does note in the final stage the persistence of certain large cells long the basement membrane, which resemble the “males ovules, or better the oviform spermatogonia of the prepubertal animal.’ These are probably the same cells to which I have referred, and which I am inclined to interpret in the same way. Barratt and Arnold (’11) indicate that the second spermato- cytes are the first cells to begin degeneration, as they state that amitotie division was observed in them twenty-four hours after the application of x-rays (p. 261), while in the first spermato- cytes necrosis was not observed until after the third or fourth day, necrosis of the spermatids begins after the fourth day and is marked by the ninth day, whereas the spermatogonia ceased DEGENERATION IN THE RAT TESTIS 107 to be recognizable after the fourth day. These observations would indicate that the spermatids were the most resistant. These changes are not in the same order as I found them in the MacDowell rats, where the spermatogonia are the last to degenerate. The Sertoli tissue as a syncytium With rgeard to the Sertoli tissue, my studies confirm Regaud’s conclusion (’01) that this tissue is a syncytium. In order to satisfy myself that this is its normal state I reexamined many slides of well-fixed rats testis which I had used in my work on spermatogenesis, and studied for the first time the Sertoli nuclei and cytoplasm with great care. I found many places where the cytoplasm could be traced from one Sertoli nucleus to an ad- joining one without encountering a cytoplasmic wall. In some poorly fixed material prepared while I was experimenting upon technique, I did find suggestions of a wall, but such appearances seem better interpreted as a thickening of the cytoplasm inci- dent to the shrinking action of the fixative. This condition is quite likely to occur when Flemming’s fluid is used, if during dehydrating or infiltrating too abrupt changes are made in the strengths of the reagents. My series of developing tubules enabled me to trace the Sertoli tissue from the very young tubules in which there is only one type of nucleus present up to the condition where all the various stages of the germ cells are present. In the young tubule a syncytium is plainly the rule. One can readily see when the first cytosomes are differentiated. These are early growth stages of the first spermatocytes, and are shown in figure 5 of my spermatogenesis paper (Allen, ’18). As these cells continue to increase in number, they differentiate cyto- plasmic walls, but I have not been able to find any Sertoli cytosome thus developed. The distinctive nucleolus of the Sertoli cell enables one to be certain whether, in a particular case, he is dealing with a germ or nurse cell. Consequently, determination of the stage when the Sertoli cell first differentiates is an easy matter. It does 108 EZRA ALLEN not appear until after the first spermatocytes are well developed. In fact, I did not find it until the spermatids had formed. When I made this observation I was unaware of Regaud’s statement in his 1901 paper: ‘La premiére apparation des noyaux de Sertol dans |’épithelium séminal a lieu, chez le Rat, au moment de la puberté, ou plus exactement au moment ot sont formés les premiéres spermatozoides normaux’ (p. 374, 11, 15 to 19). By the time this stage of growth is reached the germinative wall is closely packed with germ cells, as shown in figure 4, so that the determination of cytoplasmic walls in Sertoli cells is very difficult. It would appear that the original syncytium of the young tubule persists as such, within which the germ cells lie enmeshed. In the degenerate conditions described in this paper and by Regaud, and Barratt and Arnold, and others, no new syncytial tissue is developed, but the Sertoli syncytium is simply revealed by the loss of the germ cells, which under normal conditions obscure it. The Sertoli nucleolus The character of the Sertoli nucleolus needs further considera- tion. Previous reference has been made to the duplex nature of this body. Under normal conditions it consists of a large nearly spherical body close beside which is a much smaller one, also approximately spherical. For the sake of convenience, I shall refer to this latter as the paranucleolus.' It is not wholly unique with the Sertoli nucleolus, as a similar body is to be seen in well-fixed preparations of first spermatocytes in their late growth stages. In my paper on spermatogenesis (Allen, °18), the nucleolus is shown as a simple spherical body in figure 27. The cell from which this drawing was made had been fixed in Bouin’s fluid, not by the fluid which gave the best fixation, B-15. Before publishing the paper to which reference has just been made I did not study the nucleolus carefully in my prepa- rations fixed with the improved fixative. Since then, in con- nection with these Sertoli studies, I have discovered from this better-fixed material that its true structure is bipartite, and DEGENERATION IN THE RAT TESTIS 109 unequally so. However, in the first spermatocytes the smeller body is more closely united with the larger than in the case of the Sertoli nucleolus. scholarship and the personal fitness of its students are more care- fully scrutinized and where students who do not maintain a given standard are either forced to withdraw from the school or repeat the entire year. We owe this primarily to Doctor Sheldon. It was he who set the standards of attainment for the first-year students; it was he who brought the names of the delinquents before the executive authorities, and it was he who always un- flinchingly stood for the best. He may have appeared to some as a heartless judge, but yet beneath that sternness was a melt- ing heart. He expected the student to do well the work allotted to each course. If he failed to do so, it meant repetition of the course, and this frequently entailed the repetition of the whole year. ‘This seemed to him such a hardship that he voluntarily without any financial reward gave summer courses for those IN MEMORIAM 123 students who were unable to attend regular summer schools in order to enable them to continue with their classes. When war broke out Doctor Sheldon, like all true patriots, wanted to tender his services. He was urged by Major Bagley to apply for a commission. He was torn between two duties, and chose the harder one—he stayed at home. He, neverthe- less, offered the services of this department to the Council of National Defense and began some very important work on stain- ing methods for formalin-fixed material. The first report was sent to Washington a few weeks before his death. The methods he devised were then used in several laboratories in the neuro- pathological service. We meet here in the department which at every turn bespeaks of his activities. In the dissecting-room the tables, book racks, museum cases, cadaver tanks, all were designed in every detail by him. In the histological laboratory the desks with the inter-— changeable units of drawers, slide holders, and reagent racks, in the private rooms the slide and stock cases were built according to his specifications. A case full of charts, nearly all of which were drawn under his direction, give further evidence of his - zealous attempts to serve the student and the medical school. The anatomical library was built up by him, and considering the short time he was at the helm, the war conditions, and the not too ample funds, it shows better than anything else how he had built for permanence. The library is not filled with antiquated text-books, but with most of the current anatomical periodicals. What cannot be seen but is nevertheless felt by those who have been associated with him is the highly scientific and moral at- mosphere which he had created in the nine years he was a teacher in this institution. He was a most indefatigable worker and a most meticulously careful man, and as is so often the case with men of this type, he expected everyone else to do as much as he did. His em- ployees dared not shirk their tasks. Every day he quickly passed through all the rooms of the department and noted everything that was done or undone. It should not be judged from this that he attempted to run the department single- 124 RALPH EDWARD SHELDON handed. It was his aim to train the members of his staff to do the right kind of work and then give them responsibilities. He conferred and directed but allowed his assistants to work out the details. The result was that these men acquired an in- dependence of thought and action as is rarely found in an institution that is guided by a youthful hand. His activities did not cease with the work he found to do in his department. His unbounded enthusiasm and determination caused his colleagues to throw many more burdens on his shoulders. One has but to look over the documentary file of this department to realize that not one single phase of medical school and university administration had escaped his personal attention. Elaborate tabulations of curricula, schedules, bud- gets, staffs, and equipment of every prominent medical school in the country were orderly filed. Though the youngest pro- fessor, he was chosen the representative at large of the medical school to sit in the council of the University. Committee meet- ings upon committee meetings consumed his energy, of which he seemed to have an absolutely unlimited amount. The reason he devoted so much time to these activities is well expressed in a letter written on December 1, 1915. It seems to me that every individual in the service of the University should be made to feel that the interests of the University must always be paramount to those of the individual. A department for instance is not the personal possession of its chief. He is given certain rights, privileges and facilities in order that he may use these for the interests of his school and the University as a whole. To the extent which he fails to do this, he is not living up to his obligations to the institution. I feel, of course, that a considerable leeway must be given in the inter- pretation of this in order that full advantage may be taken of individual initiative, and that, therefore, it is not possible to lay down hard and fast rules regarding the conduct of individual departments. If it could, however, be indicated to every member of the teaching staff that loyalty to the department, the school and University would be an im- portant factor in evaluating his work, and if the individual could feel that all were on an equal footing in this regard, I believe it would do more than anything else to establish, in the institution, a spirit of affection and loyalty. In this evaluation a wide acquaintance of the work of all departments of the University and high ideals of attain- ment are necessary in order that work of individuals and departments IN MEMORIAM 125 along widely divergent lines shall be fully appreciated and that sub- stantial work for the benefit of the institution receive due credit as compared with that which happens to receive publicity. As previously stated, he was intensely interested in scientific work, and his published papers show that he was as capable a scientist as he was an administrator. In his paper on ‘‘The Nervus Terminalis in the Carp,” he gave the first account of this nerve in the teleosts. ‘‘The Olfactory Tracts and Centers in Teleosts’ is his magnum opus of scientific research. It rep- resents the most detailed and accurate analysis by an anatomi- eal method of the functional localization of tracts and centers of any vertebrate hitherto described. He carried the work to the last refinement permitted by a combination of all available strictly anatomical methods, and the work is a model of its kind. It was only after his death that I learned that Doctor Sheldon was very much interested in Chemistry. It is therefore not sur- prising that the interest in this science should have led him to apply it to neurology. In “The Reactions of the Dogfish to Chemical Stimuli” he showed that the skin of fishes is exceedingly sensitive to some chemicals, even more so in some cases than the taste-buds. In “The Sense of Smell in Selachians” and ‘‘The Sense of Smell in Teleosts’”’ he intended to test physiologically the anatomical work of his doctor’s thesis. These five papers represent an ideal of combined anatomical and physioiogical work on a definite program such as has rarely been attained by any investigator. “The Phylogeny of the Facial Nerve and Chorda Tympani” is a valuable summary illustrating the value of comparative study in solving vexed problems of mammalian anatomy. Phylogenetic history, experimental physiology, and pathological anatomy are brought to bear on the problem of human periph- eral conduction paths bringing about noteworthy results in clarifying practical problems of surgery. He summarized in “The Paraffine-Weigert Methods for the Staining of Nervous Tissue, with Some New Modifications” his own and many others’ extensive experience in the difficult problem of getting the most possible out of the Weigert method in the study of both human 126 RALPH EDWARD SHELDON and comparative brains. It is one of the most helpful contri- butions to technique in the literature. ’ » When he was but twenty-six years old he projected a text-book on Neurology. This undertaking was his life of the last two years and it was his death. The book was practically written with 950 pages of manuscript six years ago. Being inexperienced, he thought it a comparatively easy matter to have the illustra- tions made and the manuscript set to type. Before long he was sadly disillusioned. When I came to this department a little over two years ago I was asked to look over a few chapters of the book. I was amazed to find that the book which I had expected to use for my classes the previous year had not progressed any farther, but it did not take me long to find the cause of this delay. It was not teaching, it was not administrative work that was the cause of the delay, but his endeavors to do the impossible—write a — perfect text-book. His publishers rightly urged and urged. Letters, telegrams, and representatives called for haste. He finally realized that perfection was not attainable and more rapid progress was made. He was relieved of most of his teach- ing duties, and then it was book from early morning to late at night. His almost indecipherable handwriting made it diffi- cult for his stenographers, and to facilitate matters he used the dictating machine. This is not the place to review the bobk. It represents the first attempt to present in a text-book the subject of Neurology from a functional point of view. It is intended to give the medi- cal student a broad conception of the fundamental principles that underlie the structure and function of the nervous system, but at the same time pointing out the paths for future inves- tigations. The reason he undertook this work may be best stated in his own words found in the preface written many years ago. To the older anatomists the nervous system was only an anatomical structure, to be dissected out and studied in relation to the other organ systems, such as the bones, blood-vessels, muscles, ete. This attitude of mind led to the development of a school of neurologists, both human and comparative, who devoted themselves to the study of the gross IN MEMORIAM 127 morphology of the brain and peripheral nervous system in the most minute detail, identifying and comparing every depression, protu- berance, and membrane and their relations to the surrounding tissues of a different kind. Although his interests are different, the worker of to-day, with a wealth of technical methods at his disposal, must always look with amazement and admiration at the results which these men secured with the crudest of methods. ' At present there is no treatise available, which, in addition to the gross relations, will give to the student of anatomy, neurology, medi- cine or to practicing physicians, a complete presentation of the func- tional relations of the nervous system. This book represents an en- deavor to fill this gap and to present in an adequate fashion for such workers the gross and microscopic anatomy of the nervous system with all the more important functional relations. I cannot help but speak from the very depth of my emotions. The two years that I was associated with him I treasure as two of my most valuable ones. There was no barrier between us and I felt I had his utmost confidence as he had mine. We had similar ideals and it was a distinct pleasure to be able to work harmoniously by his side. Not once during these years did we part in an argument or discussion but that I felt a greater ad- miration for his personality. The sternness which he felt he had to assume on account of his youth would melt in a smile © that made you feel happy in his presence. He seemed to fairly radiate good-will and energy. I am glad that a month before his death I had coerced him to take an auto trip with me through Maryland. We both were boys again. Oblivious of everything that might burden the heart of a man we were for two days as care-free as the birds that flitted about. He vowed that he never had had such a good time and promised that before the summer was over he would take his family over the same trip. A month later and his bright career came to an end. On Sunday he worked to get off some drawings to the publishers. On Monday he felt ‘grippy’ and thought he would rest up a bit. On Tuesday morning, July 9, he got up and thinking that it might dispel his lassi- tude he took a hot bath. When he stepped out of the tub, weakness was manifest in one foot. Before the hour was over both feet were paralyzed. The paralysis rapidly crept upward, and by midnight his spirit had departed. It is curious irony of ~ fate that Dr. Sheldon should have succumbed to a nervous dis- 128 RALPH EDWARD SHELDON ease, Landry’s acute ascending paralysis, the nature of which we understand so little. It is probable that he thought the paralysis a transient one. He was hopeful to the end. The influence Dr. Sheldon exerted during his brief career will live a long time. His scientific work has placed him among those that have given substantial contributions to the ,advance- ment of knowledge. The influence he exerted as an educator will long be felt by his students and by his students’ students. He founded a department which it is hoped will ever reflect credit to his name. It is also hoped that, realizing the debt the University and the anatomical world owes him, a way will be found to bring before the world the book which will always stand a monument to industry. PUBLICATIONS The participation of medullated fibers in the innervation of the olfactory mucous membrane of fishes. Science, vol. 27 pp. 915-916. 1908. An analysis of the olfactory paths and centers in fishes. Proc. Assn. Amer. Anat., Anat. Rec., vol. 2, no. 3, pp. 108-109. 1908. The nervus terminalis in teleosts. Proc. Assn. Amer. Anat., Anat. Rec., vol. 3, no. 4, pp. 257-259. 1909. The nervus terminalis in the carp. Jour. Comp. Neur. and Psychol., vol. 19, no. 2, pp. 191-201, figs. 1-7. 1909. The reactions of the dogfish to chemical stimuli. Jour. Comp. Neur. and Psy- chol., vol. 19, no. 3, pp. 273-311, figs. 1-3. 1909. The phylogeny of the facial nerve and chorda tympani. Anat. Rec., vol. 3, no. 12, pp. 593-617, figs. 1-6. 1909. The sense of smell in selachians. Jour. Exp. Zodl., vol. 10, no. 1, pp. 51-62. 1911. Some new laboratory furnishings. Anat. Rec., vol. 5, no. 10, pp. 483-490, pls. 1-4. 1911. The olfactory tracts and centers in teleosts. Jour. Comp. Neur., vol. 22, no. 3, June, pp. 177-255, pls. 1-42. 1912. The sense of smell in fishes. With G. H. Barker. Bulletin of the U. S. Bureau of Fisheries. Vol. 32, 1912, Document No. 775, May 3, 1913, pp. 3546. Some new dissecting-room furnishings. Anat. Rec., vol. 7, no. 10, pp. 369-370. 1913. Paraffine-Weigert methods for the staining of nervous tissue, with some new modifications. Folia Neuro-Biologica, Bd. 8, Nr. 1, 8. 1-28. 1914. Some new receptacles for cadavers and gross preparations. Anat. Rec., vol. 9, no. 4, pp. 323-827, figs. 1-8. 1915. PROCEEDINGS OF THE AMERICAN ASSOCIATION OF ANATOMISTS THIRTY-FIFTH SESSION Medical School of the University of Pittsburgh; Pittsburgh, Pennsylvania April 17, 18 and 19, 1919 THURSDAY, APRIL 17, 9.30 A.M. The thirty-fifth session of the American Association of Anato- mists was called to order by President Robert R. Bensley, who appointed the following committees: Committee on Nominations for 1919: Professor J. Playfair MeMurrich, Chairman; and Professors George 8. Huntington and Florence R. Sabin. Auditing Committee: Professor Eliot R. Clark, Chairman; and Professor Frederic T. Lewis. The morning session was’ devoted to the presentation of scien- tific papers followed by a paper in memory of the late Professor R. E. Sheldon, presented by Professor Robert Retzer. Follow- ing this, the firial feature of the morning programme was an address by the President of the Association, Professor Bensley on “Anatomy, a Science or a Curriculum?” Fripay, 11.30 a.m. Association Business MEETING, PREsI- DENT RoBEeRT R. BENSLEY, PRESIDING. The Secretary reported that the minutes of the Thirty-Fifth Session were printed in full in The Anatomical Record, volume 14, number 1, pages 19 to 23, and read the minutes as printed. On motion, seconded and carried, the minutes of the Thirty- fourth Session were approved by the Association as printed in The Anatomical Record. Professor E. R. Clark reported for the Auditing Committee 129 130 AMERICAN ASSOCIATION OF ANATOMISTS as follows: The undersigned Auditing Committee has examined the accounts of Doctor Charles R. Stockard, Secretary-Treasurer of the Association of Anatomists and finds the same to be correct with proper vouchers for expenditures and bank balance on January 8th, 1919, of $211.59. (Signed) Exior R. Ciarkx, Freperic T. Lewis. The Treasurer made the following report for the year 1918: Balance on hand December 19, 1917, when accounts were last AUCIGOG ca. «oie a fe depen oe GE We niee » chlo Sets « “1h . tia’ r = ~ i a d - 7 ¥ Seine Rs ) ie hd bet hail > 2 = ws hives a a ee , . - _ - PS ean ty ~ ’ Sey iy ‘ | a - we Re ae x ‘ x 7 bd a i” s > = __mne t 4 ab RE) © Diie tat S38) ya’ Baatasider:.- Leto Resumido por el autor, Frank Blair Hanson. Demostracién de las capas germinales. El autor propone los estados denominados A-H por Balfour en | el desarrollo del embri6én de tibur6n como el mejor material facil de obtener para demostrar a los estudiantes de cursos elementales el origen de las capas germinales. Los embriones correspondi- entes a dichos estados se cortan transversalmente, seleccionando cierto numero de cortes practicados a diferentes niveles del cuerpo del embrién. Las capas germinales estan bien separadas unas de otras en la cavidad de segmentaci6n y, por consiguiente, se evitan las dificultades o confusiones con que tropieza el estudiante cuando intenta analizar las capas germinales de los anfibios. Translation by José F. Nonidez Columbia University AUTHOR’S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, APRIL 7 ON TEACHING THE GERM LAYERS FRANK BLAIR HANSON Zoological Laboratory of Washington University FIVE FIGURES The course in Comparative Embryology for premedical stu- dents as usually organized around the chick and the pig seems to lack on the laboratory side material for certain fundamental conceptions of the earliest development, without which the student finds himself at sea when attempting to study later stages, such as the 33-, 48-, and 72-hour chick and the 10-mm. pig. That this lack is felt is evidenced by the introduction at the beginning of many courses of the frog’s egg or the egg of some teleostean form for the study of the cleavage stages and the rise of the germ layers. That these are sufficient for the cleavage stages is undoubted, but when the beginning student attempts to unravel the germ layers of the amphibian embryo and keep track of the history of the various cavities therein (segmentation, archenteron, and coelome), his conclusion is apt to be that the germ-layer theory is more theory than fact. The writer wishes to call attention to Balfour’s stages A to H of the shark embryo as probably the best material available for teaching the germ layers. In our Comparative Embryology class at Washington University the cleavage stages are taught from Amphioxus slides, also the blastula and gastrula. Then the shark, Balfour’s stages, are taken up in toto mounts and transverse sections. The transverse sections are made up into sets, each set containing a slide from each stage through the same region of the body. For instance, one set (figs. 1 to 5) is through the middle of the body; another set gives the stages of the brain; a third near the anterior end of the foregut, and the last through the tail. Four sets are used at these four levels, others could be added or substituted. 193 194 FRANK BLAIR HANSON In the set herein illustrated there is never any confusion in the student’s mind as to the identity of the germ layers, for there is no confusion of the germ layers themselves. ‘They are perfectly separate and distinct, lie at considerable distance from each other, and the first three cavities of the embryo are always definitely demarkated and followed from stage to stage without difficulty. The origin of the mesoderm from the entoderm of the gut is conclusively shown, for this form at least. The schizocoele is formed before the student’s eyes, and a few words of explanation contrasts this with an enterocoele. The development of the spinal cord in its earliest stages, the infolding of the neural plate, ON TEACHING THE GERM LAYERS 195 its origin from the ectoderm, are all told simply, quickly, and clearly in the set of five slides indicated. The notochord is seen to arise out of the dorsal midline of the entoderm of the digestive tract, and its subsequent history is all contained within this same set of slides. The shark has been used here for several years and with uni- formly good results. A diagrammatic series of figures could hardly be devised that would more clearly show the steps of these early and important stages, while these have the advantage of being the sections of an actual animal. From this an easy transition is made to the chick and later the pig. It has been the writer’s experience that, judging from the final examination papers, this part of the work relating to the rise of the germ layers based upon shark material has left one of the most vivid impressions of the entire course. It has been likewise his experience that when only amphibian material was - used, this remained the muddiest part of the course. By the use of specific colors for each germ layer, the differentiation and - contrast is heightened, and the student soon comes to associate automatically each germ layer and its respective color. Balfour’s stages of the Elasmobranch embryo may be secured from several dealers in such supplies, and are not expensive if one cuts his own sections. To install sets of the stages outlined above costs approximately one dollar per student. Since the same slides are used year after year, in subsequent years there is no expense, the student paying for all breakage. Figures 1 to 5 are camera-lucida drawings of the set of slides through the midsomites. It has been found advantageous to have the student study successively sections from all the stages at the same level of the body, rather than study the different levels in each stage before passing to the next. Resumido por el autor, Frank Blair Hanson. El coracoides de Sus scrofa. El proceso coracoideo falta en el cerdo. Existe una porcién subcoracoidea que participa en la constitucién de la fosa glenoidea y es homdloga del subcoracoides del hombre. El subcoracoides se ha considerado como una simple epifisis, pero puede muy bien corresponder a la epifisis del metacoracoideo, como ha indicado Gregory. En la escdpula del cerdo hay solamente un centro de osificacién en la parte coracoidea. Translation by José F. Nonidez Columbia University AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, APRIL 7 q THE CORAGOID OF SUS SCROFA FRANK BLAIR HANSON Zoological Laboratory of Washington University SIX FIGURES The problem of the coracoid is one of the unsolved questions of vertebrate morphology. Its homologies have been described from every possible viewpoint, yet there remains to-day as much confusion and divergence of opinion concerning this structure as in any past period. This paper is unlike most in that it does not attempt to offer a new solution of the old problem. It is a description of the developmental stages of the subcoracoid as found in the pig. The material consists of a series of seven scapulae ranging in age from three weeks to adult life and a fairly complete series of sections of embryonic stages. No embryonic stages are figured, for in general they are not essentially different from figure 1 of the suckling pig. Although the coracoidal part of the pig scapula is never a separate cartilage, it may be identified as early as the 26-mm. stage of the embryo as a distinct, but blunt knob, on the an- terior side of the glenoid surface. This is not, however, sepa- rated from the cartilage of the glenoid, but is continuous with it, and is a constituent part of the articular surface. Following the history of this portion through close stages of the embryonic and foetal life discloses no essential change of relations or appear- ance until we pass from foetal to postnatal life. Figure 1 is the glenoid end of the scapula of a pig two weeks old. The rela- tively large cap of cartilage is one homogeneous whole, and has been so since first recognizable in the embryo. The shaft of the scapula has been cut in the median plane in an anterior-pos- terior direction, as have also the other stages herein illustrated. Figure 2 is the scapula of a pig three months of age. During 197 THE ANATOMICAL RECORD, VOL. 16, No. 3 FRANK BLAIR HANSON ( \) i ON: ° e 4 j ~ . ~ 4 peur t te : , ¥ Buss | i> a) (fe caves. “ieee ‘ ? + By ¥ aie ek a ; ; a ad : - x e ‘a ar, ‘ a Resumido por el autor, Harvey Ernest Jordan. Estudios sobre la estructura estriada de los miusculos. IV. Discos interecalados en el mitsculo estriado voluntario. El autor describe discos intercalados tipicos en los mtsculos de la pierna del hombre, semejantes a ios sencillos discos en ‘‘forma de banda” que existen en el mtsculo cardfaco. La presencia de tales discos en el mtsculo estriado voluntario esta de acuerdo con la hip6tesis que les supone como los representantes de bandas de contracci6én irreversibles que se han modificado, y, ademas, sus- ministra una prueba adicional en contra de su interpretacién como los limites de las células del miocardio. El autor resume y discute los cambios estructurales que sufre la fibra muscular estriada durante la contraccién, los cuales se manifiestan por cambios en sus reacciones colorantes. Translation by José F. Nonidez Columbia University AUTHOR’S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, APRIL / STUDIES ON STRIPED MUSCLE STRUCTURE IV. INTERCALATED DISCS IN VOLUNTARY STRIPED MUSCLE H. E. JORDAN Laboratory of Histology and Embryology, University of Virginia ONE FIGURE INTRODUCTION In a series of papers! on the intercalated discs of cardiac muscle, the hypothesis was developed that these structures repre- sent modified irreversible contraction bands. As stated in an earlier article (5), if this interpretation of the intercalated discs of heart muscle is correct, then we should expect to find similar structures also in skeletal muscle, under certain conditions. Continued search had, however, until recently been only re- warded by negative results. While giving the course in His- tology at the College of Physicians and Surgeons, Columbia University, during the summer session of 1918, I noticed in the sections of human voluntary striped muscle (sec. no. 294) in- cluded in the students’ loan sets of slides, certain peculiar structures which seemed very suggestive of the intercalated discs characteristic of cardiac muscle. Dr. George 8. Huntington has very kindly given me permission to use this material for study and description. The data now available only give the information that the specimen came from a leg removed at operation by Dr. W. C. Clarke, that the tissue was fixed in formalin, imbedded in celloidin, and stained with Maliory’s phos- photungstic-acid hematoxylin. The preparations are superb. ‘The only reference to intercalcated discs in skeletal muscle which I have been able to find appears in a monograph by 1 See bibliography in article by Jordan and Banks. Am. Jour. Anat., vol. 22, p. 285, 1917. 203 204 H. E. JORDAN Dietrich (1), who mentions that H. B. Schmidt demonstrated at the meeting of the Pathological Society in Erlangen in 1910 preparations of voluntary striped muscle in which occurred bands which Schmidt regarded as having the same structure as the intercalated discs of heart muscle. Dietrich, however, after careful study of Schmidt’s preparations, disputes the accuracy of such an interpretation. Dietrich’s description of these struc- tures as “‘irregular stripes which cross the muscle bundle in suc- cessive waves,” and his identification of them with the abundant contraction waves frequently seen in the cardiac muscle of human cadavers (p. 11), make it quite certain that there is no close similarity between the structures in our specimen and those in Schmidt’s material. I feel the more convinced of this conclusion since certain sections through the upper third of the cat’s esophagus in my collection show bands which correspond exactly to those described by Dietrich in Schmidt’s sections; these bands have nothing in common in the way of essential details with those in the sections of human leg muscle. In the latter material occur definite and regular discs related to the telophragmata in a manner identical with the intercalated discs of Limulus and vertebrate heart muscle; in the former material occur irregular contraction waves, spanning the entire width of the tunica muscularis in the case of the esophagus and involving a variable number of sarcomeres. The discs of the leg muscle are modified single contraction bands, which apparently failed or were unable to reverse in certain sharply limited locations; the bands of the esophageal muscle are widespread contraction waves fixed either in rigor mortis or by the action of the © preserving fluid. It would be of prime interest to know the morbid conditions which necessitated the removal of the limb from which our specimen was taken. Such information might indicate the specific factors operating in the production of intercalated discs. At present we can only surmise, in accordance with our general theory regarding the formation of such dises in heart muscle, that the special conditions under which dises originate include relatively violent or prolonged tensions or exceptional strains, STRIPED MUSCLE STRUCTURE 205 such as may accompany long-continued rhythmic contraction. The correlation of such dises with specific physiologic, patho- logic, and-experimental conditions would seem to offer a worth- while field for future investigations. In order to determine whether these peculiar structures in this specimen might possibly be a common characteristic of human leg muscle, I prepared sections from three other amputated legs which had been preserved in formalin. For two of these speci- mens I am indebted to Dr. W. C. Clarke, of Columbia Univer- sity; for the third to Dr. Stephen H. Watts, of the University of Virginia. None of these specimens contained similar discs; only occasional areas of fibers were at a midphase of contraction; the sections indicated that these muscles were for the most part in a condition of repose. In order to test the further possibility that the discs of our specimen might be the result of the special staining technic, sections of these three specimens of leg muscle © were also stained with the phosphotungstic-acid hematoxylin. These preparations again gave no evidence of similar discs. In addition, pieces of striped voluntary muscle of the frog were fixed in strong Flemming’s solution, in the picro-acetic solution, and in 95 per cent alcohol, and stained respectively with iron- hematoxylin and phosphotungstic-acid hematoxylin. These six sets of slides showed essentially identical conditions: the Q-disc was bisected by an H-dise of considerable width, but nothing suggestive of intercalated discs was discernible. It is required first to establish the homology between these discs in our specimen of leg muscle and those of cardiac muscle, vertebrate and Limulus. This is a relatively easy matter. The description of these discs will show a detailed similarity to cardiac intercalated discs amounting practically to a morphologic identity. While it is of much theoretical interest to have dis- covered genuine intercalated discs in skeletal muscle, and while the discovery adds support to our interpretation of these discs in cardiac muscle as modified irreversible contraction bands, it is clearly recognized that it does not prove that the theory is entirely correct. But a prediction made on the basis of this theory has now been fulfilled. The theory, however, involves 206 H. E. JORDAN the whole question of the mechanism of muscular contraction, concerning which there is much diversity of opinion. This fact necessitates an attempt to formulate a theory of contraction which can embrace consistently the recorded morphologic data relative to striped muscle in the various phases of contraction and extension. It is recognized also that our theory concerning the intercalated dises of heart muscle has won only limited and generally qualified acceptance. Very recent revisions of text- books of general anatomy still describe heart muscle as com- posed of distinct cells, interpret the intercalated discs as cement lines, and ignore the histologic evidence that during contraction and extension some substance moves within the myofibril from mesophragma to telophragma, and in the reverse direction, re- spectively, as demonstrated by various staining methods. That the intercalated discs in cardiac muscle are not cement lines or cell boundaries is definitely proved by the fact that they do not extend completely through a fiber in the manner of a cell mem- brane, but are more or less deep peripheral structures. They do not extend centrally beyond the innermost myofibrils in cardiac muscle, and an occasional disc lies superjacent to a nucleus. The relative sparsity of typical intercalated discs in the heart of Limulus also contravenes any interpretation of adult cardiac muscle in terms of discrete cells. Since our theory interprets intercalated discs of striped muscle as modified irreversible contraction bands, it is required that the precise morphologic features of contraction, which serve as the basis of the theory, be definitely established. It may suffice at this point to state that by ‘contraction .band’ we mean the deeply staining area which appears on either side of the telo- phragma during muscle contraction (in contrast with a similar area on either side of the mesophragma when the muscle is in repose), as first clearly illustrated by Rollet (8) in his figure of the striped muscle of Cassida equestris (fig. 126, Jordan and Ferguson’s Text-book of Histology). By ‘contraction wave,’ as seen, for example, in our specimen of the cat’s esophagus, we understand a much wider irregular condensed area of adjacent groups of fibers; such ‘waves’ in any one limited area may in- STRIPED MUSCLE STRUCTURE 207 clude many ‘contraction bands.’ In Rollet’s figure the ‘wave’ would include eight ‘bands. An intercalated’ disc is in its simplest condition (ontogenetically and phylogenetically) funda- mentally identical, structurally and tinctorially, with such a contraction band. In mature heart muscle it has become secondarily modified through the influence of mechanical and probably chemical factors, and the increase of tissue fluid among the elements. Certain pathologic conditions (e.g., hypertrophy and atrophy) are characterized by specific varieties of discs. DESCRIPTION All of the fibers of the entire section, 15 mm. by 5 mm. in extent, have a fairly uniform structure. A deeply staining Q-dise alternates regularly with a lightly staining J-disc. These two dises are of approximately equal thickness. The Q-disc is bisected by a lightly staining H-disc. The resulting subdivi- sions of the Q-disc have in general approximately the same thick- ness as the dividing H-disc. However, there is some variation in the thickness of the H-dise in different fibers. The J-disc is bisected by a distinct membrane, the telophragma (fig. 1, 7’). A mesophragma (M-membrane) is only discernible in occasional small areas of several of the fibers. The Q-dise is colored dark purple; the J and H dises stain a light pink color; the telo- phragma takes a light purple stain. This description could apply equally well to a stretched fiber or to one at midphase of contraction. The available data are not sufficiently precise to warrant a final conclusion regarding the actual functional con- dition of the fiber, that is, whether contracting or passing into repose or stretched. All things considered, however, I incline to regard the muscle as fixed in a slightly stretched condition. The illustration shows three intercalated dises—one isolated, two arranged in a pair. Very rarely a group of three or even four successive discs occurs. Nowhere in this section, barring groups of three or four discs, are the dises more numerous than indicated in this illustration; in general they are less closely spaced, the distribution varying in different portions of the section. 20S H. E. JORDAN The dises consist of a row of bacillary elements in close latera- juxtaposition, joined together through the middle by a telol phragma. These bacillary elements are portions of the included myofibrillae. The interbacillary tissue. fluid takes a_ slightly fei : eT Soe ee et tn ore . 2 ennanoMa TD. SUTRA terse me HAN Otero em Meee H Fig. 1 Drawing of portion of longitudinal section of voluntary striped muscle fiber of human leg (formalin fixation, Mallory’s phosphotungstic-acid-hematoxy- lin stain). J, telophragma (ground membrane, Z-line); J. clear (isotropic) dise; Q, dark-staining (anisotropic) disc; H, Hensen’s disc; J.D., intercalated disc. 7 stains lightly purple, Q and J.D. deeply purple, J and H light pink. The portion between the two nuclei was drawn by aid of a camera lucida; the portion below the lower nucleus, including the pair of intercalated dises, was added free hand from an adjacent fiber. Magnification 1500 diameters. re STRIPED MUSCLE STRUCTURE 209 deeper stain than the substance of the H-dises or the uninvolved J-dises. Taken as a whole the discs resemble a double-comb structure, the ‘bar’ of the ‘comb’ being formed by the telo- phragma. The discs are more or less deeply placed peripheral structures. Careful focusing reveals the fact that they do not completely span the fiber. The isolated disc is typical; it is less dense laterally, showing thus the bisecting telophragma; a few dises are of equal and maximum density throughout. Many of the discs are in the condition illustrated in the lower pair of discs; that is, drawn out for some distance on one side, this condition generally being confined to alternate sides of the discs in a pair. These discs are comparable to ‘contraction bands’ as defined above; as such they might be in late phases of contraction or early phases of extension, that is, in incomplete contraction or extension; or they might be contraction bands drawn out laterally by a stretching of the fibers. The stretch-_ ing of the discs may have been the result of the relatively greater contraction on the part of adjacent portions.of the unmodified fibers during fixation or normal function. If these ‘discs’ actually represented unmodified contraction bands, then we should expect to find them more abundantly and more exten- — sively arranged in groups, the latter in lateral alignment in the form of contraction waves. In a contracting or relaxing muscle we should not expect to find the bands so generally isolated and so sharply localized. In view of these facts, it seems more likely that the discs have become modified through tension and that perhaps the entire fiber was in a stretched condition. Moreover, the fibers are frequently distorted, that is, sharply bent or drawn out in the vicinity of the discs, facts which indicate that the dises are levels of relative weakness. In my earlier studies of the intercalated discs of cardiac muscle I have shown that the simplest type of disc, both onto- genetically and phylogenetically, is a band form bisected by the telophragma. Primary variations of this original type produce discs that are bounded on one side by the telophragma, and such as are bounded on both sides by telophragmata. Further sec- ondary modifications resulting from the operation of mechanical 210 H. E. JORDAN factors, including chiefly oblique tensions in consequence of the branched condition of the fibers of the heart musculature, lead to the various terraced forms. Hypertrophied muscle, char- acterized by longitudinally splitting myofibrillae, contains only serrated types. The dises of our specimen of leg muscle corre- spond, then, to the original band type with bisecting telophragma. In accordance with my hypothesis that intercalated dises are modified irreversible contraction bands, the discs bounded on only one side are interpreted as contraction bands only half of which became irreversible; those bounded on both sides by a telophragma as fused adjacent irreversible halves of successive contraction bands. Viewing the fiber of the illustration as in midphase of con- traction, we may more closely examine the isolated dise. The condition of this portion of the fiber may be indicated by the following formula: T + J/2 + Q/2 + H + Dise (Q/2 + T + Q/2) + H + Q/2 + J/2 + T = 2 sarcomeres. In other words, the dise includes adjacent halves of successive Q-portions fused along the telophragma. The structure in question corre- sponds to a contraction band and to a simple type of inter- calated disc. As such it consists of deeply staining, laterally aligned, portions of adjacent myofibrils bisected by a telophragma. Since these contraction bands are apparently irreversible as indi- cated by their localized distribution in the absence of contraction waves, and slightly modified as indicated by the cloudy appear- ance of the lateral ‘stretched’ portions, they would seem to be typical intercalated discs. As shown in a previous paper (6), the original dises (irreversible contraction bands) are modified in part also by the accumulation of tissue fluid in the intervals of the adjacent portion of the myofibrils as indicated by the considerable precipitation of silver nitrate in these locations. The occurrence of typical intercalated dises in voluntary striped muscle, as here described, fulfills the prediction made by de- duction from the hypothesis constructed from the histologic data derived from the study of cardiac muscle, and to this extent it gives support to this hypothesis. STRIPED MUSCLE STRUCTURE 211 DISCUSSION The above-outlined interpretation of the discs as modified irre- versible contraction bands calls for a brief discussion of the histologic features of contraction. The histologic evidence shows that contraction and extension are accompanied by the passage of a deeply staining substance from the mesophragma to the telophragma and vice versa, respectively. This statement calls for evidence of the existence of a mesophragma in striped muscle. It must be admitted that such a membrane is under usual con- ditions not discernible with our present means of microscopic observation. In an earlier paper (4) on the leg muscle of the sea-spider, I showed that a mesophragma becomes discernible in the contracting portion of a fiber which has at the same time become stretched, because of the already fully contracted con- dition of one end of the fiber. Assuming that the mesophragma is too delicate to come within the limits of visibility under ordinary conditions, stretching of the fiber would result in lengthening the sarcomeres and in reducing their diameter, thus causing a relaxation of lateral tension and permitting the meso- phragma to coarsen to a point where it comes within the limits of microscopic vision. This is in essence the idea of Heiden- hain (3), who claims to be able to see a mesophragma even in human cardiac muscle. The behavior of the lightly stained portion of the Q-dise (i.e., the transient H-disc) under con- ditions of contraction and stretching points also to the presence of a dividing mesophragma in the Q-disc. The deeply staining portion of the Q-disc under these conditions divides exactly along a middle plane; such precise division of this disc could hardly occur so uniformly unless the disc were bisected by a true mem- brane. A small portion of at least one fiber of this section also shows a mesophragma with great clearness. This fiber is at midphase of contraction and the portion showing the meso- phragma is apparently under some tension. That the meso- phragma of a sarcomere, however, differs to some extent from the terminal telophragmata is indicated by the facts that festoons are not formed in the sarcolemma at its levels, the THE ANATOMICAL RECORD, VOL. 16, No. 3 212 H. E. JORDAN myofibrils are not bent along its levels in distorted fibers, and the nuclear wall is not drawn out into ridges at its levels. This would seem to indicate that the mesophragma, in contrast to the telophragma, is not attached to sarcolemma, myofibrils and nuclear membrane. However, the essential difference between these two membranes may be simply one of relative strength or rigidity. The mesophragma may be sufficiently elastic to be able to yield without rupture to distorting forces, or it may be too delicate to withstand the strains that operate in fixed tissues to demonstrate the relationship between the telophragma and the myofibrils, sarcolemma and nuclear membrane. It would seem, therefore, that any discussion of muscle con- traction may confidently proceed upon the basis of the existence of two membranes, the terminal telophragmata of the sarcomere and the bisecting mesophragma. Since Merkel (7) first called attention to the ‘reversion of the striae’ during contraction, this phenomenon has been repeatedly described in many different specimens of striped muscle. It was especially well shown in the illustration of Rollet (8). Englemann (2) criticised this in- terpretation on the basis of his demonstration that, as revealed by the polariscope, the anisotropic substance of the Q-dise does not move during contraction, but remains permanently segre- gated more or less closely about the mesophragma. This con- clusion is supported by the observations of Schaefer (9) on muscle treated with Rollet’s gold chloride method.? Englemann (2) would seem to interpret the phenomenon of an apparent reversal of striae, as shown in stained sections, on the assump- tion that the absorption of the isotropic substance of the J-disc by the anisotropic substance of the Q-dise (upon which con- traction is supposed to depend) causes a relative condensation of the sarcoplasm about the telophragma and thus produces a relatively deeper-staining area in this location. Schaefer (9), however, attributes the phenomenon to ‘‘the deeply moniliform * It is now clear to me, however, that the sarcostyle of the wasp’s wing muscle which Schaefer regards as in the contractéd condition is one swollen and shortened by the action of the hypotonic formic-acid-watér solution used in Lollet’s technic. STRIPED MUSCLE STRUCTURE 213 shape of the fibrils’? (p. 184), which he believes ‘‘tends to cause the constricted parts to appear dark.’’ In other words, Schaefer interprets the apparent reversal of the striae during contraction as an optical illusion. He states further that “In alcohol preparations (both of the wing muscles and of the ordinary or leg muscle) in which the sarcous elements have been stained, __ there is no appearance of reversal of striation; the darkly colored sarcous element always occupies the central or bulged part of the sarcomere, and the unstained hyaline substance occupies the constricted parts of the sarcostyle’ (p. 185). The latter state- ments seem to me in direct contradiction to the facts. In alco- hol-fixed tissue stained with iron hematoxylin the striae do reverse during contraction, the stainable substance of the Q-disc moving towards the telophragma; and the deeply staining discs of contracted muscle, that is, the ‘contraction bands,’ are bisected by the telophragma, where the constriction, if any © occurs, is uniformly located in contrast with the deeply staining band of extended muscle which is bisected by the mesophragma. Furthermore, that the movement of material during con- traction is not solely from the isotropic to the anisotropic sub- stance, that is, from the telophragma to the mesophragma, but largely in the reverse direction is shown by the fact that the median portion of the Q-disc, the portion farthest removed from the isotropic substance, becomes pale first; this gradual paling or loss of tingibility passes distally in opposite directions from the mesophragma towards the telophragmata, and the myofibrils become swollen distally and assume a knobbed appearance as if material were actually accumulating terminally in the sarco- “meres. These knobbed ends stain even more deeply than the general Q-substance of the relaxed fiber. This phenomenon was especially well shown in the sea-spider muscle and has been illustrated by various investigators on the histology of con- tracting muscle. If contraction only involved an absorption of isotropic by anistropic substance, by reason of which the striae apparently reversed due to a change of relative density of the J and Q discs, then the terminal portion of the original Q-dise 214 H. E. JORDAN should become pale first instead of the farthest removed portions (from the nearest point of alleged absorption) or the middle portion of the Q-dise. To summarize, these are the fundamental histologic data upon which an adequate theory of muscle contraction must be based or with which it must harmonize, and which it must consistently embrace. ‘The occurrence of a mesophragma and a telophragma as constituent elements of a sarcomere, the telophragma at least being intimately connnected with the myofibrils, the sarcolemma, and, where a nucleus intervenes, with the nuclear membrane; changes, as evidenced by an alteration in staining capacity, within the myofibrils, these changes involving a movement of a deeply staining substance from the mesophragma to the telo- phragma, apparently without change in position of the aniso- tropic granules of the original Q-dises; these changes produce a ‘contraction band,’ that is, a deeply staining ‘disc’ bisected by a telophragma; such discs, failing to ‘reverse’ locally, correspond to the simplest type of intercalated discs of cardiac muscle, and may become further modified (physically, chemically and mor- phologically, involving the accumulation of tissue fluid) to form genuine intercalated discs; the appearance of contraction bands is coincident with a shortening and thickening of the myofibrils and of the muscle fibers and of the muscle as a whole; con- traction and mechanical tension produce a comparable effect upon the Q-disc, a separation along the level of the meso- phragma; when constrictions supervene in-a contracted fiber they are at the levels of the contraction bands, that is, at the levels of the deeply staining portions, and are due at least in part to the fact that here the bisecting telophragma is attached to the sarcolemma, between two successive points of attachment — to which the sarcolemma is festooned permitting thus a bulging of the intervening sarcous substance at this level. The above-detailed facts are not in conflict with the newer theories of muscle contraction that attempt to explain con- traction in terms of surface tension or electrical phenomena among the ultramicroscopic sarcous particles, in so far as these theories do not postulate an absorption of the isotropic sub- STRIPED MUSCLE STRUCTURE 215 stance of the J-disc by the anisotropic substance of the Q-disc or a passage of substance from the region of the telophragma towards the mesophragma. Both the change in structure and in staining reaction of the involved portions of the myofibrils demonstrate that the movement of material is largely in the reverse direction during contraction, from the mesophragma towards the telophragma. Such movement is coincident with contraction. How these two concurrent phenomena are funda- mentally or causally related is a question whose answer lies outside the scope of the present investigation. The chief aim of this investigation, finally, concerns a demon- stration that genuine intercalated discs occur in voluntary striped muscle under certain conditions, and the further support of our hypothesis that the intercalated discs of cardiac muscle are in essence modified irreversible contraction bands. If the conclusion that these discs of human leg muscle are comparable with the intercalated discs of heart muscle is correct, as seems incontrovertible, then the latter discs can for an additional reason be no longer regarded as cell boundaries. [LITERATURE CITED Dietricu, A. 1910 Die Elemente des Herzmuskels. G. Fischer, Jena. 2 Encuemann, T. W. 1893 Uber den Ursprung der Muskelkraft. W. Engle- mann, Leipzig (eited from Heidenhain). 3 HerrtpenHAIN, M. 1911 Plasma und Zelle. G. Fischer, Jena. 4 Jorpan, H. E. 1916 The microscopic structure of the leg muscle of the _ sea-spider, Anoplodactylus lentus. Anat. Rec., vol. 10, p. 493. 5 1917 The microscopic structure of striped muscle of Limulus. Pub. 251, Carnegie Institution of Washington, p. 273. _ 6. Jorpan, H. E., anp Banks, J. B. 1917 A study of the intercalated discs of the heart of the beef. Am. Jour. Anat., vol. 22, p. 285. 7 Merxet, F. 1872 Der quergestreifte Muskel. I. Das primitive Muskelele- ment der Arthropoden. Arch. f. mikr. Anat., Bd. 8. 8 Rotter, A. 1891 Uberdie N-Streifen (Nebenscheiben) das Sarkoplasma und die Kontraction der quergestreiften Muskelfasern. Arch. f. mikr. Anat. Bd. 37. : 9. Scuarrer, E. A. 1912 Text-book of microscopic anatomy. Longmans, Green & Co. _ 4 i? iii Oty ii} tat’ dT | niu rey Te nisi Seta ¥ Phat) * hank t ’ yi Mo a ] u ®, if cer lee ek » Py | ab 1a Sud ” saitecil uliqoreni a () iron ; » wenn hy wal organ fipal: . eo feos iniadogs ten Sik Y iho eee ‘Ke : yj fin i ati eee vy 4 nc alone guile y Tite? hot +744 ott i ‘ fers nine = ah ee ae Muar ‘ oe rT { ve We, t. fh et asl «tet 2s . tts | awa) i ee orth ut is > vara ‘tote shel ied pojnc 1 ae ty ii % ited HAE, t 1, ‘wap aud Greats 2nd) Wil ;? va if ee ek ah ; i} 7 Le ae ail Ait e 218 H. E. JORDAN LEG MUSCLE The description of the comparative structure of the wing and leg muscles begins most conveniently with the latter variety, since sarcosomes do not occur here to obscure the several stria- tions in their alterations during contraction. As will appear below, leg and wing muscles, aside from the presence of numer- ous interfibrillar sarcosomes in the latter have an essentially identical microscopic structure. In the extended or relaxed condition, the leg-muscle fiber presents an alternation of light and dark discs, the former approximately twice the width of the latter (fig. 1). The telo- phragma is distinctly seen as a deep-staining granular mem- brane, bisecting the lighter disc. The granules of the telo- phragma are swellings at the points of attachment of the myofibrils. That the connection between the myofibrils and the telophragmata is very intimate is demonstrated by con- ditions in distorted fibers, where the fibrils are held rigidly in place along telophragma levels regardless of modifications in the normal relations of the myofibrils within the sarcomeres. The conditions here are practically identical with those originally described for the skeletal muscle of Limulus.?° The fiber is enveloped by a robust sarcolemma, frequently festooned between, and firmly united to, the telophragmata (fig. 6). The nuclei are peripherally located (figs. 4 and 6). Their number is increased relatively towards the point of in- sertion into the exoskeleton. In transverse section the fibers have a generally polygonal outline. The myofibrils are strap- like peripherally; centrally they have a cylindric form; many of the peripheral fibrils show a radial split, indicating longitudinal division. There is no suggestion of a mesophragma at any stage in the contraction process. Certain fibers with the same general features as those de- scribed for the extended or resting fiber present an additional striation on either side of the telophragma, namely, an acces- sory disc of Merkel or the so-called N-stripe (fig. 2). As the fibers enter contraction, the dark or Q-dise becomes bisected STRIPED MUSCLE OF MANTIS 219 by the appearance of the H-disc, as illustrated in figure 3. This figure shows also the N-disc; in the lower portion of this figure, Q and N are seen in process of fusion. A subsequent stage is illustrated in figure 4. Here Q and N have fused throughout, and the resulting composite dark discs are approaching the telo- phragmata. The union of two such composite discs with an intervening telophragma produces a contraction band (fig. 5). Figure 6 is at approximately the stage illustrated in figure 5. The difference in appearance is due to a difference in degree of destaining. Figure 7 also corresponds to the stage shown in figure 5; it differs in appearance somewhat from the fiber of figure 5 probably by reason of having become stretched. In figure 8 is illustrated a completely contracted fiber, in which deep-staining contraction bands alternate with lght-staining intermediate bands of approximately equal thickness. The foregoing description shows that a contraction band consists of the fused opposite halves of two consecutive Q-discs, plus the two intervening N-discs and the included telophragma. It shows, moreover, that a deep-staining substance of the Q-dise of the fiber in repose actually moves toward the telophragma during contraction. WING MUSCLE As can be seen from figures 9, 10, 11, 12 and 13, the wing muscle during contraction presents the same alterations in the striations as the leg muscle. However, nothing strictly comparable to the N-dise of leg muscle seems to occur in wing muscle. Figure 11 illustrates especially well the passage of material from Q to the telophragma.! The fibers are relatively enormous structures, irregularly polygonal in outline in transverse section (fig. 23). 1The slightly greater length of the sarcomeres of certain contracting fibers, as compared with those of the resting fibers, frequently seen in histologic prepa- rations, may be explained as the result of the superposition of a condition of stretching (figs. 1, 3 and 8; and figs. 9, 11 and 13). It would seem that those portions of a fiber which are at midphase of contraction are more responsive than those at rest to the tensions incident to the death and fixation of con- tracting fibers where one end is in full contraction (see paper on leg muscle of sea-spider’). 220 H. E. JORDAN The nuclei are generally peripherally located (figs. 21 and 22), but occasional nuclei lie in the region between the center and the pe- riphery (figs. 23 and 24). The peripheral myofibrils have the form of broad lamellae, those more centrally placed are cylindric in form. Adjacent fibers differ considerably with respect to the form and arrangement of the myofibrils. (Compare figs. 22, 23, and 24.) The feature of greatest interest about the wing muscle pertains to the interfibrillar granules or sarcosomes (figs. 14 to 20). These are preserved about equally well in the formalin- and the Flem- ming-fixed tissues. In alcohol-fixed tissue they cannot be dis- cerned. Alcohol destroys the bodies, leaving a granular débris. Their reaction to the several fixing fluids suggests that they are largely - lipoid in chemical constitution. Since they cannot be seen in formalin-fixed leg muscle, I conclude that sarcosomes do not occur in this leg muscle. In the wing muscle the sarco- somes are apparently scattered at random in the interfibrillar sarcoplasm, including the peripheral and perinuclear regions (figs. 28 and 24). They are apparently unconnected with the myofibrils and disconnected among themselves, for in teased fresh material they readily separate out from among the fibrils. In the sectioned material the sarcosomes, especially the larger, have an oval shape; in the isolated condition in teased fresh material they are spherical in shape. These observations indicate that the sarcosomes have a semifluid consistency, their oval shape when confined within the fiber being due to com- pression between adjacent muscle columns. In certain por- tions of the fixed tissue the sarcosomes appear to consist of hollow vesicles (fig. 15), only a robust shell (‘membrane’) persisting. Under these conditions, the sarcosome ‘negatives’ are more nearly spherical in shape. The adjacent sarcostyles are curved around these vesicles, and are firmly held in place by the telo- phragmata. Moreover, in almost any area, sarcosomes of dif- ferent consistency can be seen; some stain deeply, others only lightly. In general, the smaller stain more deeply. The vesicu- lar sarcosomes are generally spherical in form (fig. 14). These observations indicate that the sarcosomes undergo changes in —_ ., STRIPED MUSCLE OF MANTIS 221 physical and chemical constitution, and suggest their transient character. As regards the morphologic changes during con- traction, the sarcosomes appear to have a passive role. Figures 17 and 18 represent fibers (destained to a point where the Q-dise no longer shows) in the condition of repose or ex- tension. Here the granules are aggregated within the region of the Q-disc. The majority have an oval form, the long axis of the granule being parallel with the long axis of the fiber. Figure 19 represents a fiber at about midphase of contraction, comparable with the more deeply stained fiber of figure 11 and the leg fiber of figure 4. Here the smaller spherical and more deeply staining granules have become aggregated along the telophragmata. Figure 20 represents a fiber in prattically complete contraction. The destaining has been carried to a degree which no longer leaves the contraction band conspicu- ous. The fiber corresponds to a stage close to those illustrated | in figures 12 and 13. The most interesting fact relates to the altered shape of the sarcosomes; these elements are still oval in shape, but in a fiber in this condition the long axis of the oval granule is placed at right angles to the long axis of the fiber. This alteration in the axes of the oval granules is presumably due to pressure exerted by reason of the closer apposition of the successive telophragmata in the contracted fiber. Similar mechanical factors no doubt operate in the production of the occasional angular, cup-form, collapsed and the other irregular types of sarcosomes. The foregoing description of the movement of the sarcosomes during muscle function discloses two other important facts: 1, The effective confining capacity of the telophragmata and, 2, the ineffective confining capacity or absence of mesophragmata. The sarcosomes apparently cannot pass through the telophrag- mata, and there is apparently nothing to prevent their passage through the area occupied by a mesophragma in certain other insect muscle fibers. A matter which presents great difficulty of explanation is the considerable variation in density of closely adjacent fibers. A comparison of figures 22, 23, and 24 and of figures 14 and 20 222 H. E. JORDAN will give some idea of this variation. In figure 22 the lamellar sarcostyles are of very irregular form; in figure 23 the lamellar sarcostyles are quite regular in form and are numerically greatly preponderant; in figure 24 only a narrow peripheral layer is composed of only relatively delicate lamellae, the cylindric type greatly preponderating. In figure 14 the lamellae are widely separated and the telophragmata are ruptured; in figure 20 the telophragmata are all intact and hold the myofibrillae firmly in place. Figures 14 and 24 are in similar condition, also figures 20 and 23. At first sight the difference between such fibers as are illustrated in figures 14 and 20, that is loose, and compact types, might seem to be due to a relative abundance of sarco- somes. Careful microscopic examination, however, makes it clear that the sarcosomes are about equally numerous in both types. The difference in general appearance is due primarily to a rupture of the telophragmata in some fibers, permitting a separation of the myofibrils and thus bringing into sharper view the included sarcosomes. This matter then resolves itself into the question as to why the telophragmata are ruptured in certain fibers. Since intact fibers occur in all phases of con- traction, a condition like that illustrated in figure 14 is not related to a specific functional stage. (Compare figs. 11, 17, 18, 19, and 20.) Nor is the condition characteristic of smaller fibers or of the central portion of fibers, as might perhaps be inferred from figures 23 and 24. While the looser and more compact fibers generally occur in separate groups, such segregation is by no means invariable. Occasionally a loose fiber may lie among a group of compact fibers or a compact fiber may become loose toward one end in the section. These observations suggest that the ruptured telophragmata are not fixation artifacts. But the data are not sufficiently precise to warrant a final conclusion. The condition might possibly represent a disintegration of certain fibers. But the normal condition of the sarcosomes (compare figures 14, 15 and 19) would seem to be conclusive evidence against this inter- pretation. Whether the condition illustrated in figure 14 is a fixation artifact or a phase of regression must remain undecided STRIPED MUSCLE OF MANTIS 223 for the present. The condition, however, is of much importance from another view-point, as will be made clear in the subsequent discussion when the question of the character of the membranes, telophragmata and mesophragmata, in wing muscle will be discussed. The meaning of the difference in shape and size of the sarcosomes, and their segregation during contraction (compare figs. 11, 17, 19, and 20), will also be considered in the subsequent discussion. Certain of the peripheral fibers which remained covered by chitin during fixation in Flemming’s fluid were only poorly pre- served. The lamellae appear swollen and broken and in part more or less fused into an irregular network (fig. 30). The general appearance is that of muscle tissue fixed by Meves’ technic for mitochondria. Such fibers after staining with iron- hematoxylin are seen to contain filar and bacillary mitochondria in the interlamellar sarcoplasm. MUSCLE-TENDON CONNECTION The data regarding the mode of muscle connection to the chitinous exoskeleton are so clear and definite in this material that it seems desirable to give a brief description. I shall only describe the condition here presented without discussion or review of the pertinent literature. The literature on the subject of muscle-tendon connection has been most recently assembled in a bibliography and fully reviewed by Downey.‘ In the mantis the muscle and tendon fibrils are continuous. A similar condi- tion was previously described for the skeletal muscle of the sea-spider? and the scorpion.!' This mode of muscle-tendon connection accords with that described by O. Schultze?° for vertebrate, including human skeletal muscle generally. Downey, however, claims that in the crayfish the muscle fibers end ab- ruptly and become ‘dovetailed’ into the tendon, the muscle fibrils thus being non-continuous with the tendon fibrils. In the mantis the muscle fibers, wing, abdominal and leg, are inserted directly into the chitin. In the abdomen certain muscle fibers can be followed for long distances in their passage 224 H. E. JORDAN between the hypodermal cells, where they are still clearly stri- ated, but apparently lack sarcosomes, to points where they separate into delicate fibrils which are inserted into the chitin. The evidence is most clearly presented in the case of the muscles related to the ovipositor in a newly moulted specimen of 38-mm. length (fig. 29). The hypodermal cells, with indistinct bound- aries, here form a relatively thick layer, upon which rests young chitin. The muscle fibers show only a faint striation and con- tain relatively very many nuclei. Transverse sections of these fibers show that the sarcostyles are delicate lamellae (fig. 26). The nuclei at this stage multiply both by mitosis and amitosis (fig. 25), later only by amitosis. The muscle fibrils pass be- tween the hypodermal cells into the superjacent chitin. Among the hypodermal cells the muscle fibrils become compacted into bundles. Where they pass into the chitin they spread out again in fan-like fashion before they terminate. Where they pass among the hypodermal cells, the myofibrils stain intensely black. This difference in staining reaction is a physical phenomenon rather than an indication of a chemical change, for within the chitin the fibrils again stain only lightly in a manner similar to that before they enter the hypodermis. Moreover, in certain regions the hypodermal fibril bundles stain lightly and so appear no different from their condition before entering the hypoderm and after passing beyond it. Acid-fuchsin counterstain reveals no sharp difference between the myofibrils before and after they enter the hypodermis. The fibrils among the hypodermal cells and within the chitin are actually muscle fibrils which have lost their cross-striated condition. In this sense, that is, calling the modified muscle fibrils within the hypoderm and chitin ‘tendon,’ muscle fibrils and tendon fibrils are strictly continuous. Below the hypodermis occurs a more or less complete connective- . tissue layer, forming a loose fibrillar and nucleated ‘basement membrane’ for the hypodermal cells. i) bo Or STRIPED MUSCLE OF MANTIS DISCUSSION The lamellar type of sarcostyle is characteristic of arthropod muscle.2 A question of much theoretical importance concerns the intimate structure of this element. Is the lamella the ulti- mate morphologic unit or is it a composite structure? In an attempt to answer this question, I dissected fresh wing muscle of the mantis in diluted glycerin. Lamellae could readily be separated from their neighbors within the fiber. Such lamel- lae were easily divided into coarser and finer fibrils. In trans- verse sections of fixed and stained material certain lamellae appear perfectly homogeneous; others seem to contain con- stituent fibrils. In longitudinal sections also certain lamellae seem to contain a central more condensed and deeper-stain- ing core or fibril. These observations suggest that the lamellae include fibrils which are imbedded in a homogeneous lamellar sarcoplasm. The histogenetic process gives confirmatory data. The young fibers (fig. 26) contain delicate lamellae and cylin- dric fibrils. At later stages the fibrils are largely coarse lamellae, some of which may be seen splitting longitudinally (fig. 27). This process of myofibril increase by longitudinal splitting, | radial and tangential, still obtains in adult fibers (figs. 23 and 24). The complete process follows this order: The first formed myofibrils are cylindric; subsequently appear lamellae; these latter produce other lamellae and central cylindric fibrils by longitudinal fission, In adult fibers some of the lamellar sar- costyles instead o¥ splitting off cylindric myofibrils retain them within their substance, at least for a time, thus producing com- posite myofibrils in the form of lamellae. The evidence, then, indicates that the lamellar sarcostyles contain ultimate genuine myofibrils. ‘The condition of the fiber at its hypodermal termi- nal, where it becomes resolved into fine fibrils which pass in groups between the hypodermal cells to their attachment with the chitin, also demonstrates the essential composite fibrillar nature of the lamellar sarcostyles. 2 The wing muscles of certain insects, e.g., diptera, hymenoptera and cole- optera, present striking exceptions. 226 H. E. JORDAN The similarity between the adult structure of insect muscle and the embryonic structure of vertebrate muscle is a fact of much interest. We have here a specially beautiful illustration of the law of biogenesis (‘recapituation theory’). For the pur- pose of more clearly presenting this point I have added a figure of a developing muscle fiber of the newly hatched rainbow trout (fig. 28). As can be seen by comparing figures 23 and 28, the structural similarity is very close. As in the adult wing muscle of the mantis, the skeletal muscle fiber of the trout embryo con- tains peripheral lamellar and central columnar sarcostyles. The latter are derived from the former by central tangential splitting. The peripheral lamellae split also radially into daughter lamellae. At an earlier stage the muscle cell of the trout embryo contains a single, deeply staining primitive myo- fibril lying close against the nuclear wall. The genesis of the original myofibril could not be determined. Heidenhain® has recorded similar observations on trout-embryo muscle, but he also was unable to determine the origin of this initial myofibril. The subsequent history, however, is clear. The original myo- fibril (primitive sarcostyle) splits radially into four primitive lamellae, each one of which again splits longitudinally, the repetition of this process leading to a condition illustrated in figure 28. Later steps in the continuous longitudinal splitting (radial and tangential) lead to the adult condition of vertebrate striped muscle with its definitive structure of ultimate myo- fibrils (definitive sarcostyles) collected into Kélliker’s columns (Cohnheim’s areas in transverse section). This line of evidence suggests that the lamella of the insect muscle corresponds rather to a K6lliker’s column of vertebrate muscle than to an ultimate fibril or genuine sarcostyle. And, as stated above, certain lamellae clearly reveal included fibrils. In the mantis material, accord- ingly, the term ‘sarcostyle’ is strictly synonomous neither with ‘myofibril’ or ‘Kélliker’s column,’ as these terms are used in connection with adult vertebrate skeletal muscle. The myo- fibrils of mantis correspond to the sarcostyles (synonomous with myofibril) of vertebrate skeletal muscle; the sarcostyle of mantis consists of a collection of myofibrils in’ the shape of a STRIPED MUSCLE OF MANTIS 227 lamella or a cylinder. Groups of such sarcostyles correspond to a Kdlliker’s column of vertebrate muscle. The sarcosomes are therefore strictly intersarcostylic in distribution. Thulin”: has recorded the observation that the wing muscles of certain insects (coleoptera, hymenoptera, and diptera), and the analogous pectoral muscles of birds and bats, lack both the telophragmata and the mesophragmata. Holmgren® likewise believes that telophragmata are lacking in the wing muscle of certain insects. His illustrations of the wing muscle of Libel- lula,? however, show these membranes very conspicuously. A comparison of Thulin’s illustrations of the wing muscle of the wasp and of the bumblebee with the several types of fibers in the mantis wing muscle, shows that his material corresponds with the looser variety of mantis muscle. Certainly, the com- pact wing-muscle fibers of mantis contain definite and con- spicuous and perfectly typical telophragmata. The suggestion presents itself that possibly Thulin had conditions for the basis of his description like those represented in figure 14 where the telophragmata had become ruptured and had apparently very generally disappeared by reason of a parallel orientation of the fragments with the myofibrils. Absence of telophragmata in these muscle fibers would therefore seem to be secondary to changes, either artificial or such as are related to regressive processes, in the fibers resulting in a rupture of those membranes. As regards the mesophragmata, however, the case is different. I have previously shown that in the leg muscle of the sea-spider a mesophragma comes into view under certain conditions.® These conditions include a stretching superimposed upon a midphase of contraction. A contracting fiber which is firmly attached to the chitin at both ends may become modified in this way. One end may be in full contraction while the middle portion is still at midphase. This middle is then put under tension, the diameter of the fiber is reduced, and the mesophrag- mata, which under ordinary conditions are beyond the limits of microscopic vision coarsen and are brought within visual limits. The presence of an H-disc at midphase is also a factor in rendering the mesophragma more conspicuous. These ob- 228 H. E. JORDAN servations would seem to confirm Heidenhain’s® claim that a mesophragma is universally present in striped muscle, but may be invisible by reason of its extreme tenuity. Such ex- planation may apply also to the apparent absence of a meso- phragma in the leg muscle of the mantis. However, in the wing muscle of the mantis a mesophragma seems actually to be lacking. The only plausible alternative interpretation would be one postulating a fenestrated condition of this membrane, permitting a free passage of the sarcosomes. For while the sarcosomes are effectively barred in their move- ments by the telophragma, no such barrier occurs in the region where the mesophragma, when present, is located. On the other hand, the fact that the deeply staining portion of the Q- dise of the wing-muscle fiber divides equally during contraction, and the halves move in opposite directions toward the telo- phragmata, would seem to demonstrate the presence of some sort of membrane sufficiently complete to initiate the division of the Q-substance at certain definite and regular levels. The evidence, then, seems to indicate that if a mesophragma is actu- ally present it must be fenestrated, or at least be of such a nature as to be unable to offer a barrier to the free movement of the sarcosomes. If the evidence is adjudged to indicate the absence of a mesophragma, the equal division and opposite movement of the Q-substance during contraction, and in a stretched fiber, present a problem for which no very plausible explanation is at present at hand. In this case we would seem compelled to postulate a membranous partition in the Q-sub- stance of the sarcostyle which has no representative, contrary to the case of the telophragma, in the intersarcostylic regions. The foregoing discussion leads logically to a consideration of the phenomena of contraction. Englemann and Schaefer's assume a movement during contraction of fluid from the iso- tropic J-dise to the anisotropic Q-dise. Contraction is explained as depending upon a shortening and widening of the sarcomeres following a swelling of the anisotropic granules of the Q-dise by reason of absorption of fluid from the J-dise. Englemann has demonstrated by means of the polariscope that the aniso- STRIPED MUSCLE OF MANTIS 229 tropic granules do not change their location during contraction (cited from Schaefer). The apparent reversal of striations fol- lowing contraction, first described by Merkel and most clearly illustrated by Rollet,!® is believed by Englemann and Schaefer to be an optical illusion due to the movement of fluid from J to Q, producing a condensation of the former disc and a rarefac- tion of the latter. Granting that the anisotropic granules re- tain their original segregation in Q during contraction, and even that fluid may pass from J to Q, there still remains the fact of an additional movement of some deeply staining constituent of Q to the telophragma of J, which is the chief factor in the formation of the contraction band as seen in stained sections, and which effects a true reversal of striations during contraction with respect to this deeply staining constituent of the intra- fibrillar sarcoplasm. This is clearly shown in figure 11. The Q-portions of the adjacent myofibrils lengthen, at the same time becoming lighter-staining and more slender medially, and deeper- staining and knobbed terminally. This terminal knobbed con- dition of the Q-segments of the myofibrils during contraction demonstrates that some substance actually moves towards the z-membrane. If the only movement of substances concerned in contraction were from J to Q, by which Q is caused to appear lighter, then the terminal portions of the Q-segments of the fibrils should become lighter first and the middle portions only sub- sequently. On the contrary, the portion nearest the point of alleged absorption, that is, the level between J and Q remains dense and deep-staining while the farthest removed level, namely, the midportion of Q, becomes light first. No attempt will here be made to construct a modified theory of muscle contraction in accord with this important morpho- logic datum, but the point may be emphasized that any final theory of contraction must be able consistently to include this fact. A reversal of striations, in so far as the deep-staining constituent of the substance of the Q-disc of a muscle fiber in repose is concerned, seems demonstrated by the alterations in the chemical constituents, as indicated by alterations in stain- ing reactions, of the several major stripes, including the oblitera- 230 H. E. JORDAN tion of the N-stripe of extended muscle. Holmgren’ claims that the contraction bands result from the aggregation of the J-sarcosomes about the telophragmata (p. 610), a position sup- ported also by Retzius* and by Heidenhain.® Such claim is conclusively contravened by the fact that typical contraction bands appear both in fibers which normally lack these sarco- somes (fig. 8) and those in which they have been destroyed by alcohol (fig. 13). This leads to the questions regarding the constitution and the significance of the accessory disc or N-stripe. This stripe has been described in the muscles of many insects. M. Heidhain illustrates it even in the voluntary striped muscle of man (posterior crico-arytenoid; Plasma und Zelle, fig. 358, p. 622). In the leg muscle of the sea-spider this dise occurs in the form of a relatively pale band about midway between Z and Q, as pre- viously described.’ In the leg muscle of the mantis it appears with exceptional clearness (figs. 2 and 3). It could not be de- tected in the wing muscles. The chief point of discussion con- cerns the location of its constituent elements, that is, whether the granules whose lateral juxtaposition results in this stripe are inter- or intrafibrillar (sarcostylic). Retzius’ and Heiden- hain® claim that the N-disc is composed of interfibrillar J-sarco- somes. Rollet'® interprets it as an intrafibrillar anisotropic constituent. That it does not consist of sarcosomes in the mantis muscle is demonstrated by the fact that it is present in the leg muscle of mantis after fixation in alcohol, which technic destroys sarcosomes, quite as definitely and distinctly as in muscle fixed in formalin, by which technic the sarcosomes are fully preserved. Moreover, in foramlin-fixed wing muscle the smaller spherical intersarcostylic sarcosomes are segregated at certain phases of contraction close to the telophragma in the J- disc, but do not closely resemble the N-disc of the leg muscle. In the leg muscle it can be clearly seen that the N-dise con- sists of intrafibrillar elements (figs. 2 and 3). The interfibrillar so-called J-granules of the wing muscle do not produce a genuine N-dise. It is, however, easy to see how confusion has arisen - regarding the constitution of this N-disc. During contraction STRIPED MUSCLE OF MANTIS 231 the N-dises fuse with the deep-staining Q-substance in its passage to the telophragmata, and so contribute to the formation of the contraction bands of Rollet. Regular aggregation of J-sarco- somes may simulate N-discs, but genuine intrafibrillar N-discs and contraction bands are essentially independent of the in- terfibrillar sarcosomes. The data are not yet sufficient to permit of definite conclusions regarding the significance of the N-dise in relation to contraction, but that the facts of its oc- currence and movement must be included in any complete theory of muscle contraction is obvious. The matter of prime interest in the wing muscle of mantis concerns the sarcosomes. We should like to know their com- plete history, including origin, function, and fate, as well as structure and position. The study of their origin will be re- served for a separate paper. Their structure and their altera- tions in form and position at various phases of contraction have already been described. Holmgren’ has described similar ele- ments in the wing muscle of Libellula and in the bumblebee and in the heart muscle of the crayfish. He figures and describes delicate connecting threads between successive sarcosomes. I have nowhere been able to detect such connecting fibrils in the mantis sections. Occasionally an underlying or overlying myo- fibril simulates a connecting thread, but the ease with which the granules separate both in the freshly dissected fibers and in the looser fibers of the sections, argues against the occurrence of such connecting fibrils in this tissue. Holmgren classifies the sarcosomes of Retzius as J- and Q- granules. This raises one of the most difficult questions con- nected with this study. Do only sarcosomes of a certain specific type occupy the J-dise and only sarcosomes of a different type occupy the Q-disc? In an attempt to solve this problem areas like the one illustrated in figure 16 were carefully studied. Here the telophragmata are ruptured, the myofibrils widely spaced, and the sarcosomes lie freely within the interfibrillar areas. It is assumed that the sarcosomes maintained their normal mutual relationships during these modifications of the fiber. If this is an accurate assumption, it follows from the illustration that THE ANATOMICAL RECORD, VOL. 16, No.4 232 H. E. JORDAN there is no definitely regular succession of the several types of granules throughout as many as six successive sarcomeres. Large and small sarcosomes appear to be distributed indiscrimi- nately. Moreover, in the intact fibers during repose (figs. 17 and 18) the several forms of granules are mingled in an area about midway between successive telophragmata, that is, in the region of the Q-dise. This is true also of their distribution in the contracted fiber (fig. 20). . But during midphase of con- traction (figs. 11 and 19), when Q becomes widened out, the two extreme size variations of sarcosomes are apparently segregated in different regions; the smaller, generally deep-staining, spheri- eal granules are aggregated close about the telophragmata, while the larger, oval, generally lighter-staining granules are scattered in the newly formed H-dise of the dividing Q-substance. This segregation is the crux of the problem, and the explanation of this separation would seem to promise the clue regarding the genetic relationship of the several size variations of sarcosomes. Reactions of the sarcosomes to alcohol and formalin and their staining reaction to iron-hematoxylin suggest that the sarcosomes of the several sizes are chemically similar, and that the larger are developed by growth from the smaller. The smaller, how- ever, stain somewhat more intensely, and at the same time are more susceptible to the destructive action of water and glycerin in the fresh condition. The fact that the smaller sarcosomes move toward the telophragmata during contraction, considered in relation to the fact that during these stages there occurs a movement of intrafibrillar substance in the same direction, might seem to indicate that the smaller sarcosomes are actually intrafibrillar. But the direct histologic evidence clearly shows that these granules are actually interfibrillar. The explanation that seems most plausible, in view of all the available evidence, with regard to the segregation of the J- and Q-sarcosomes during contraction, is one expressed in terms of place of origin and the operation of mechanical factors. Assuming on the basis of admittedly meager histologic data (the mingling of the two types of sarcosomes in the Q-dise during certain functional phases) that the larger oval sarcosomes are later growth stages of the STRIPED MUSCLE OF MANTIS 233 smaller spherical sarcosomes, it may be that the newly formed young sarcosomes appear first in the outlying regions of the mass of Q-granules—that is, closest to the telophragma along which membrane, due to its connection with the sarcolemma, the elements from which the granules are formed presumably enter the fiber—and in consequence are forced against the telophrag- mata through the approximation of these membranes during contraction and the crowding of the larger oval Q-sarcosomes in the region midway between successive telophragmata. This is in accord with Holmgren’s’? demonstration that the telophrag- mata are in close spatial relationship also with the terminals of the tracheae, through which oxygen is supplied to the muscle tissue, and with his interpretation of the Z-membrane as a path for the entrance and exit of products of metabolism. Holm- gren,’ however, regards the J-granules as ‘secondary degenerative fragmentation products’ (p. 620), presumably resulting from the disintegration of the Q-sarcosomes whose function he conceives to be related to the supplying of the energy needs of the active muscle. In my opinion, the evidence supports better the inter- pretation which regards the J-sarcosomes as the precursors of the Q-sarcosomes than as products of their disintegration. Bullard? has shown that lipoid granules constitute a normal sarcoplasmic element of mammalian, including human, heart muscle. The question arises whether the sarcosomes of insect wing-muscle are the homologues of the lipoid granules of mam- malian heart muscle. There are a number of correspondences. A quotation from Bullard’s paper? will indicate the close simi- larity: ‘Fat droplets of normal cardiac muscle are arranged in longitudinal and transverse rows in the sarcoplasm between the myofibrils or muscle columns. Large droplets are in the Q-band, smaller droplets in the J-band’ (p. 28). Holmgren’ concludes for an identity both in the case of the heart of the crayfish and of mammals (p. 610). Retzius’ has described J- granules also in skeletal muscle of certain arthropods and even of the rabbit. In the latter material they are described as ex- clusively of the small spherical type. The fact that where only * one type of granule is present it is of the smaller spherical type 234 H. E. JORDAN speaks in favor of the above interpretation of the growth relation- ship between the smaller spherical and the large oval sarcosomes. The further fact that in these instances, as in certain skeletal muscles, the exclusive smaller sarcosomes are nearer the telo- phragma is in accord with our interpretation of their origin at this level. Heidenhain® regards the sarcosomes as ‘vegetative organelle,’ places them in the category with secretory and pig- ment granules, and interprets them as reservoirs of the carbo- hydrates which are necessary for muscle function (p. 638). Before discussing further the nature and function of the sar- cosomes, we may consider an aspect of their microchemical re- actions especially emphasized by Holmgren.’ Certain Q-gran- ules stain more intensely peripherally where they abut upon the adjacent lamellae; the portion of the myofibril in contact with the sarcosome stains in a similar intenser manner. Holm- gren interprets this phenomenon to mean that a substance passes from the Q-granules to the Q-segments of the myofibril. This seems plausible, but the possiblity of an artifact may not be ignored. But Holmgren claims further that this condition cor- responds to the extension phase; and that a very different specific condition, where the Q-granules shade laterally into wide deep- staining branches or wing-like processes, corresponds to the con- traction phase and indicates an imbibition of the deep-staining substance from the more peripheral nucleated portion of the fiber. In the wing muscle of the mantis I can detect no sarco- somes with long wing-like processes such as Holmgren describes for the wing muscles of certain neuroptera. The only precise correspondence between sarcosome morphology and functional phase of the fiber is that represented by the change in the direc- tion of the long axis of the oval granules between extension and contraction. However, accepting tentatively Holmgren’s inter- pretation of his observations, we might conclude, on the basis of similar evidence, that the deep-staining substance which passes from the sarcosomes to the myofibrils is identical with that which passes from the Q-dise to the telophragma during contraction and so contributes to the formation of the contraction band, whence it is eliminated via the telophragma. Assuming STRIPED MUSCLE OF MANTIS Zao that the sarcosomes consist of a combination of carbohydrate and lipoid elements, the deep-staining substance passing through the fiber to the Z-membrane might be thought to be in part the lactic acid resulting from the oxidation of the carbohydrate element of the sarcosome. The ‘reversal’ of the contraction band would, according to this interpretation, mean simply the elimination of the lactic acid at the telophragma and its reforma- tion through oxidation of carbohydrates in the Q-disc. But this explanation, taken as a whole, seems too naive to correspond faithfully to the series of undoubtedly very complex chemical reactions. Moreover, it ignores the fact that the same altera- tions occur during contraction in the stripes of muscle fibers which lack typical sarcosomes; and it is unable to adjust satis- factorily the discrepancy between the rapidity of muscle con- traction and the time obviously demanded for the projection and retraction of long wing-like processes by a relatively rigid sarcosome. ‘There is little doubt that the sarcosomes are related to the metabolic needs underlying the production of energy in active wing muscle, but the histologic picture probably can- not reveal all of the subtle physical and chemical changes involved in this activity. In an attempt to arrive at conclusions regarding the function of the sarcosomes on the basis of their morphologic and micro- chemical alterations, the fact must be kept in mind that all striped muscle fibers have essentially an identical fundamental structure at the several functional phases whether the inter- fibrillar sarcoplasm contains definite sarcosomes or not. For the present, however, it cannot be definitely stated whether all striped muscle does or does not contain granular homo- logues, possibly much smaller and more susceptible to the de- structive action of ordinary histologic technics, of the specific sarcosomes of insect wing muscle. The so-called sarcosomes of heart muscle may represent intermediate stages between the larger more resistant sarcosomes of insect wing muscle and their possibly more elusive homologues of skeletal muscle. Without discussing at this point Bullard’s implied identifica- tion of sarcosomes with mitochondria, we may consider his 236 H. E. JORDAN suggestion that ‘the anisotropic property of segment Q is de- pendent upon the presence of the phospholipines of the true interstitial granules’ (p. 27). This would mean that the aniso- tropic condition of Q results from the presence of interfibrillar elements, instead of from intrafibrillar granules, as commonly assumed. I have carefully studied the wing muscle of mantis with the micropolariscope in an effort to determine the refrac- tive properties of the sarcosomes. The fiber as a whole is dis- tinctly anisotropic. But there seems to be no sharp segregation of the anisotropic substance in the Q-dises. It can, however, be easily determined that the Q-sarcosomes are not anisotropic. This demonstrates only that the anisotropic substance is extra- sarcosomic; but since the compact skeletal muscle fibers of mam- mals, in which relatively little besides myofibrillae is present, also are anisotropic as a whole, the conclusion seems to follow that the anisotropic materials are exclusively intrafibrillar. In mammalian heart muscle Bullard? has described two dis- tinct types of ‘interstitial granules’ of Koelliker: fat droplets, and ‘true interstitial granules’ or mitochondria. The former he believes consist of neutral fat, and he interprets them as ‘reserve foodstuff.’ The fat droplets are said to correspond to the ‘liposomes’ of Bell;! the mitochondria to the J- and Q-granules of Holmgren and the sarcosomes of Retzius. Bullard, accord- ingly, would seem to interpret the sarcosomes of insect wing muscle as mitochondria. The accuracy of such interpretation is contravened by the fact that filar and bacillary mitochondria, such as are typical for differentiated somatic cells generally and for developing striped muscle fibers of vertebrates, are present in addition to sarcosomes (fig. 30). That sarcosomes and mitochondria have in part a similar chemical constitution is demonstrated by their similar microchemical reactions. Thulin! also has called attention to the fact that the sarecosomes of Hydrophilus react in typical fashion to Benda’s mitochondrial technic. Such similarity of reaction is probably due to the presence of a predominant lipoid constituent. The evidence seems rather to favor interpreting the sarcosomes of insect wings muscle as analogous to the fat droplets of mammalian heart STRIPED MUSCLE OF MANTIS 237 muscle, their function being related to the supply of nutrient materials for the muscle fiber. This leads to the question of the relationship between mito- chondria and fat droplets (and sarcosomes). The relation of the mitochondria to the sarcosomes in insect wing muscle will be the subject of a separate paper. But here it may be pointed out that Bullard regards his evidence as opposed to the view that true intestitial granules (mitochondria) meta- morphose into fat droplets. On the contrary, however, Schreiner'’ has very clearly shown that in the subcutaneous tissue of the hag-fish (Myxine) embryo, fat globules of developing fat cells do arise from mitochondria. The chromatic nucleolus here expels granules (‘chromidia’) which pass into the cyto- plasm, where they become transformed into rodlets (mito- chrondria); these segment into secondary granules (liposomes) which liquefy to form fat vesicles, the latter subsequently co- — alescing to form the fat globule of a definitive adipose cell. Rus- so’s 17 experiments with rabbits are also of much interest in this connection. By the administration of lecithin (a typical phos- pholipine) he claims to have increased the mitochondrial con- tent of the odplasm, and coincidently the size of the eggs and the quantity of their deutoplasmic constituents; and he records histologic data in favor of the deutoplasmic transformation of the mitochondria. That the injection of a phospholipine (leci- thin) should increase the amount of mitochondria (largely phospholipines) is very significant from the view-point of mito- chondrial function and their relation to nutrient fat. These important observations of Schreiner and Russo, how- ever, seem sharply contradicted by the results of the recent investigations of Lewis.* In cultures of young chick embryos, grown in Locke’s solution to which yolk stained with Sudan III had been added, Lewis finds no evidence indicating any association between the mitochondria and the accumulating fat droplets in the living cell. The conflicting, very positive conclusions especially of Schreiner and Lewis cannot at present be harmonized. The genetic relationship of mitochondria to liposomes (fat droplets) and to the sarcosomes of insect _ wing muscle still remains an open question. 238 H. E. JORDAN SUMMARY 1. The salient differential features between the leg and wing muscles of mantis pertain to a conspicuous N-dise in the leg muscle fiber and sarcosomes in the wing muscle fiber. 2. The sarcostyles are of lamellar form peripherally and cylin- dric form centrally. Both types contain constituent myo- fibrils. The sarcosomes are distributed in the intersarcostylic sarecoplasm. The myofibrils are directly continuous with the tendon fibrils. 3. The N-dise is formed by the lateral juxtaposition of in- trafibrillar constituents. It fuses with a substance which moves from the Q-dises to the telophragma in the formation of a con- traction band. 4. A contraction band (disc) includes the deep-staining sub- stance of opposite halves of successive Q-dises, the two inter- vening N-dises and the inclosed bisecting telophragma. Both N-dises and contraction dises are independent of the sarcosomes. 5. The telophragma presents an effective barrier against the movement (passive) of the sarcosomes. No such barrier exists in the region commonly occupied by the mesophragma. Either a mesophragma does not occur in the wing muscles or it lacks an intersarcostylic component, that is, it may be fenestrated. 6. During certain functional stages, especially the midphase of contraction, the sarcosomes are more or less sharply segregated into J and Q groups. 7. The sarcosomes have an initial spherical shape. This may become modified by pressure into oval or irregular forms. They have a semifluid consistency and are enveloped by a more con- densed peripheral ‘membrane.’ They have in large part a lipoid chemical constitution, as indicated by their reactions to fat solvents and stains. ‘The smaller J-sarcosomes are apparently the precursors of the older larger Q-sarcosomes. The aggre- gation of the J-sarcosomes nearer the telophragma is to be interpreted in terms of their origin from materials transported by the telophragma from the interfiber tissue spaces. The final phase of the sareosomes as hollow vesicles, frequently STRIPED MUSCLE OF MANTIS 239 collapsed, indicates a transient nature and suggests a nutritive significance. The persistence of a granular débris after treat- ment with alcohol indicates that they contain an albuminoid or carbohydrate constituent in addition to the major lpoid element. 8. The sarcosomes present chemical and staining reactions similar to mitochondria. Their cytoplasmic constitution is obviously closely similar. But sarcosomes are more closely analogous to fat globules than to mitochondria. The genetic relation between mitochondria, sarcosomes, and fat globules remains undetermined. LITERATURE CITED 1 Bet, E. T. 1911 The interstitial granules of striated muscle and their relation to nutrition. Internat. Monatsch. f. Anat. und Physiol., Bd. 28. 2 Buruarp, H. H. 1912 On the interstitial granules and fat droplets of striated muscle. Am. Jour. Anat., vol. 14, p. 1. 3 1916 On the occurrence and physiological significance of fat in the normal myocardium and atrioventricular system (bundle of His), in- terstitial granules (mitochondria) and phospholipines in cardiac mus- cle. Am. Jour. Anat., vol. 19, p 1. 4 Downey, H. 1912 The attachment of muscles to the exoskeleton in the crayfish, and the structure of the crayfish epiderm. Am Jour. Anat., vol. 13, p. 381. 5 Herpennarin, M. 1911 Plasma und Zelle. G. Fischer, Jena. 6 1913 Uber die Entstehung der quergestreiften Muskelsubstanz bei der Forelle. Beitrige zur Teilkérpertheorie, II. Arch. f.mikr. Anat., Bd. 83, nu. 4. 7 Hotmcren, E. 1907 Uber die Sarkoplasmakérner quergestreifter Muskel- fasern. Anat. Anz., Bd. 31, s. 609. 8 1910 Untersuchungen iiber die morphologisch nachweisbaren stofflichen Umsetzungen der quergestreiften Muskelfasern. Arch. f. mikr. Anat., Bd. 75. 9 JorpaNn, H. E. 1916 The microscopic structure of the leg muscle of the sea-spider, Anoplodactylus lentus. Anat. Rec., vol. 10, p. 493. 10 1917 The microscopic structure of striped muscle of Limulus. Pub. no. 251, Carnegie Inst. of Wash., p. 273. 11 1917 Studies on striped muscle structure. III. The comparative histol- ogy of cardiac and skeletal muscle of scorpion. Anat. Rec., vol. 13, p.1. 12 1919 Studies on striped muscle structure. IV. Intercalated dises in voluntary striped muscle. Anat. Rec., vol. 16, no. 3. 138 Kopguiiker, A. 1889 Gewebelehre. 6, Aufl. 1. 240 H. E. JORDAN 14 Lewis, M. R. 1918 The formation of the fat droplets in the cells of tissue cultures. Science, vol. 48, p. 398. 15 Rerzius, G. 1890 Muskelfibrille und Sarkoplasm. Biol. Unters. Stock- holm. N.F. 1 (cited from Holmgren). 16 Rotter, A. 1891 Uber die N-Streifen (Nebenscheiben), das Sarkoplasma und die Kontraction der quergestreiften Muskelfasern. Arch. f. mikr. Anat., Bd. 37, s. 654. 17 Russo, A. 1909 Studien iiber die Bestimmung des weiblichen Geschlechtes. G. Fischer, Jena. pp. 1-105. 18 Scuarrer, E. A. 1912 Text-book of microscopic anatomy. Longmans, Green & Co. 19 Scureiner, von K. E. 1915 Uber Kern- und Plasmaveriinderungen in Fettzellen wirend des Fettansatzes. Anat. Anz., Bd. 48, s. 145. 20 Scuuttrze, O. 1912 Uber den direkten Zuzammenhang von Muskelfibril- len und Sehnenfibrillen. Arch. f. mikr. Anat., Bd. 79, s. 307. 21 Tuuuin, I. 1915 Ist der Grundmembran eine konstant vorkommende Bildung in den quergestreiften Muskelfasern? Arch. f. mikr. Anat., Bd. 86, s. 318. EXPLANATION OF FIGURES The drawings were made with the aid of an Abbe camera lucida. Unless otherwise specified, the tissues illustrated are from adult individuals, fixed in strong Flemming’s fluid, and stained with iron-hematoxylin; and the magnifica- tion is 1500 diameters. PLATE 1 EXPLANATION OF FIGURES Leg muscle 1 Longitudinal section of portion of fiber, in the extended or resting con- dition. z, telophragma. Formalin fixation. 2 Similar fiber, more deeply stained, showing the accessory disc, or N-line. Formalin fixation. 3 Similar fiber in early phase of contraction, showing the additional H-dise and the E-dise. Alcohol fixation. 4 Later phase of contraction. The accessory dises (N) have fused with the separating halves of Q. Nu., nucleus. Alcohol fixation. 5 Slightly later stage, showing the beginning of the formation of the con- traction band (C.B). Formalin fixation. 6 Similar fiber, less deeply stained, showing the festooned sarcolemma (8S) and an underlying mucleus. Formalin fixation. 7 Later stage in contraction. Alcohol fixation. 8 Fully contracted fiber. C. B., contraction band. Formalin fixation. STRIPED MUSCLE OF MANTIS H. E. JORDAN bea ote pepe apasteposse +Z seanenqnoet-Q eves enon. P ane cbeseses ‘wae see cose vopanapaend wus eet Ae incbam be isthe geceee? we emene ey ible oseeet? ee Maes atte 4 aT Q Vedra atime TEs badass} yyeee tem verti Ogee OL opel tines a" beso N Ate eee atch E s1e4e HHS Maer iatitia Reds, ttt Pratt i. Wy nithinn PLATE 1 wana 4: mun rQ —~<-Z TL} 8 eteetne PLATE 2 EXPLANATION OF FIGURES Wing muscle 9 Longitudinal section of portion of fiber in extended or relaxed condition, lightly stained. The sarcosomes are not shown. Formalin fixation. 10 Similar fiber in early phase of contraction. Formalin fixation. 11 Later stage in contraction. This fiber, which is slightly stretched, shows also the so-called J- and Q-granules or sarcosomes. Formalin fixation. 12 Beginning formation of contraction band (C. B.) The sarcolemma (S) is festooned between the telophoragmata (Z). Flemming fixation. 13 Fully contracted fiber, showing contraction bands (C. B.) and the festooned sarcolemma (8S). Flemming fixation. 14 Portion of fiber in which the sarcostyles have become widely separated and the telophragmata (Z) ruptured, showing the several types of sarcosomes: oval and spherical; larger and smaller; deeply and lightly staining; solid and vesicular. 15 Portion of fiber in which the telophragmata-sarcostyles connection re- mains intact. The sarcosomes appear as hollow spheroidal globules, the sar- costyles curving around them between successive telophragmata. This fiber is in the contracted condition. 16 Four adjacent interfibrillar spaces, showing the irregular distribution of the several types of sarcosomes, including modified cup-shaped forms. 17 and 18 Lightly stained fibers in the extended condition, showing the more regular type of sarcosome distributed midway between successive telophragmata. 19 Similar fiber in early stage of contraction, showing the distribution of the J- and Q-sarcosomes. (Compare with figure 11.) 20 Lightly stained fiber in the contracted condition. The oval sarcosomes, with their long axis previously parallel to the long axis of the fiber, have changed to an oval condition in which the long axis is now at right angles to the long axis of the fiber, as if modified by a pressure resulting from the closer apposition of successive telophragmata. STRIPED MUSCLE OF MANTIS PLATE 2 H. E. JORDAN sn Muu iis Wes a mes ily fan ' Uhoeapueregens Chee etme ee ew <_Z Vegatverpiaetts HI Vermenetyeray ae ~ } o-q—= . eocenqnaide 1 aad a qecre 6. %% DAY, Loe ) Tit @) Qe, O10 OHO now) 45 0.8, 010000) hp ie! Og) ary V8} ag ye te 202 oss } z< she so sie BT i @| a ea Be '@ - - r wore e*® eee 2 ap ore = abil A pat TA nee hay aim ‘eg @ eo @0 6 oo OS 8G, dedvisewer® S3%F 6 io .f 1 at ate @., ) OL AP et 1 © Be ear it he aio 18 19 20 PLATE 3 EXPLANATION OF FIGURES 21 Longitudinal section of portions of two adjacent fibers. The fibers are only lightly stained. The nuclei (Nw) are very numerous and peripherally placed. S, sarcolemma; Z, telophragma. Drawing by Mr. Massie Page. X 600. 22 Transverse section of four adjacent fibers. The peripheral sarcostyles are of the lamellar type, the central of the cylindric type. Ec, ectoderm; S, sarco- lemma. Drawing by Mr. Massie Page. > 600. 23 Transverse section of a fiber, showing one peripheral and one subcentral nucleus, also a connective-tissue nucleus. The peripheral sarcostyles (L) are chiefly lamellar, the more central cylindric. Many of the peripheral lamellae are in process of longitudinal radial division. Scattered among the sarcostyles, through the central, peripheral, and perinuclear sarcoplasm, are many sarco- somes (Ss.). X 1000. 24 Transverse section of a smaller looser fiber, in which the nucleus is cen- trally placed, the sarcostyles are almost exclusively of the cylindric type (only the peripheral border containing narrow lamellar sarcostyles), and the sarco- somes are more abundant and conspicuous. X 1000. 25 Longitudinal section of small area of muscle connected with the oviposi- tor, showing nuclear multiplication both by the mitotic and the amitotie (X) modes. These fibers show faint cross striations, the telophragmata. Aceto- sublimate fixation. 26 Transverse section of similar group of young muscle fibers, showing the central location of the nuclei and the narrow lamellar form of the initial sarcostyles. 27 Transverse section of a group of older muscle fibers. The nuclei are begin- ning to move peripherally, and the sarcostyles are coaser lamellae. Ct., connec- tive-tissue nucleus. 28 Transverse section of young muscle fiber of newly hatched rainbow trout. The sarcostyles include peripheral lamellae and central cylinders, both in proc- ess of longitudinal division. Nu., nucleus; Mi., mitochondria. The similarity between young vertebrate voluntary striped muscle and adult insect muscle (fig. 23) is striking. Meves’ technic. X 1000. 29 Longitudinal section of group of young muscle fibers connected with the ovipostor. The nuclei increase in number towards the point where the muscle fibrils pass between the hypodermal cells (Zc) to become inserted into the chitin (C). There is here strict continuity between muscle fibrils and tendon fibrils. C.T., basement membrane. %X 1000. 20 Transverse section of wing-muscle fiber, showing filar and bacillary mito- © chondria (Mi.) among the sarcosomes (Ss.) and myofibrils. X 1000. . . STRIPED MUSCLE OF MANTIS H. E. JORDAN PLATE 3 ees) ny”? Fag iak oy Ae \ Ow Resumido por el autor, Hubert Dana Goodale. Las células intersticiales en las gonadas de la gallina doméstica. El autor ha encontrado abundantes células con el tamano, forma y reacciones colorantes de las llamadas células intersticiales del ovario de la gallina, en el timo, en el testiculo del macho bajo ciertas condiciones y en la sangre. Con menos frecuencia tales eélulas se han observado en otros 6rganos. Las células inter- sticiales, por consiguiente, son probablemente, leucocitos eosiné- filos. Translation by José F. Nonidez Columbia University AUTHOR’S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, MAY 1 INTERSTITIAL CELLS IN THE GONADS OF DOMESTIC FOWL H. D. GOODALE Massachusetts Agricultural Experiment Station FOUR FIGURES The literature on interstitial cells of the reproductive organs has been adequately reviewed by Boring and Pearl (717). In the same paper they have reexamined the question and reach the conclusion that the ‘true’ interstitial cells, while always present in the ovary and usually, if not always, absent from the adult testis, are not causally concerned with the control of the secondary sexual characters. This conclusion is substantiated by later papers, viz., Boring and Pearl (‘18) and Pearl and Boring (‘18). Light is thrown on the nature of these cells by certain observa- tions made in this laboratory, in the course of which several differential stains, including those employed by Boring and Pearl, were used. They indicate that the granule containing cells, to which the term ‘interstitial’ is limited by Boring and Pearl, are eosinophils. If this view is substantiated by further studies, it will explain both the irregularities in their distribution in the ovary and the discrepancies in the accounts of various authors as to their occurrence or non-occurrence in the testis. The evidence for the view that the granular interstitial cells are leucocytes is as follows: Cells of the same size, shape, having granules of the same size, and taking the same stains as do the ranular cells of the chicken ovary are to be found in other organs, abundantly in some, rare in others. They may occasionally be observed interspersed among the erythrocytes in blood-vessels. The most favorable material encountered, aside fron the ovary, is the active thymus of two molting drakes. Here these cells 247 THE ANATOMICAL RECORD, VOL. 16, NO. 4 248 H. D. GOODALE are located in great abundance along the trabeculae (fig. 1). A few are found distributed among the lymphoid cells, but in any case they are sharply differentiated from the latter. In other instances, these cells are absent or few in number. Eosinophils are mentioned as normal constituents of the thymus in various text-books of histology. In the tunica albuginea of the testis of the same individual a few of these cells were observed, though absent from the tubules. In figure 2 are shown the same sort of cells in the connective- tissue covering of several tubules of the testis of an old hen- feathered Silver Spangled Hamburg cock, which are undergoing cystic degeneration. They are absent from normal portions of the same testis. Figure 3 illustrates the same sort of cells in the epididymis of a drake’s testis, both in the connective tissue and in the blood stream. Figure 4 shows the granular cells near the surface of a portion of a hen’s ovary. Thyroid, pineal, and pituitary glands have been stained and search made for cells with the same characters. They have been found, though they are not common. Doubtless they could be found in other organs, but no search has been made. The relative abundance of granular interstitial cells in the ovary can easily be understood if these cells are really leucocytes, since the ovary is the active seat of both progressive and regres- sive processes. Their presence in the degenerating tubules of the Hamburg cock’s testis, but not in the normal portions, ean be explained on the same basis as well as the observed irreg- ularities in their occurrence in the thymus. Such gatherings of leucocytes are a well-known phase of their behavior. While these granular cells are usually very abundant in the ovary, one specimen of ovary from a nine-year-old hen with well-de- veloped spurs is noticeable for an almost complete absence of such cells. This may be due to the fact that the greater part of the tissue consists of luteal cells. 1All figures, though outlined with a camera, have been somewhat schematized. The cells are relatively too large. INTERSTITIAL CELLS IN THE GONADS 249 Fig. 1 Thymus of molting drake. /, interstitial cells; BV, blood vessel; P, parenchyma. X 120. Fig. 2 Cystic spermatic tubules of hen-feathered Silver Spangled Hamburg cock. J, interstitial cells; BV, blood vessel; Sz, spermatozoa; Sm; spermato- gonia; C, connective tissue. X 120. Fig. 3 Diagrammatic representation of epididymis, Ep, and adjoining sper- matic tubules, St, of adrake. J, interstitial cells. X 80. Fig. 4 Portion of ovary near the surface. St, stroma; J, interstitial cells xX 452 250 H. D. GOODALE In addition to cells that appear to be ordinary connective- tissue cells, the stroma of the chicken’s ovary contains a cell type that stains poorly and which have been called clear cells by some authors, but which Pearl and Boring designate as luteal cells. Although no granules have been demonstrated, the pos- sibility must be recognized that these cells may be true interstitial cells, at least from the physiological standpoint, i.e., they may furnish an internal secretion such as is demanded of the ovary.’ The fact that they are not demonstrably granular does not militate aginst this view, since special treatment may be re- quired to demonstrate granules (for example, Bensley, ’16.) or the secretion may never be stored in granular form. There are two further observations that bear on the relation between the gonad and the secondary sexual characters. The first is that the various sorts of cells observed in the normal ovary are found in ovarian tissue transplanted into castrated males with resulting feminization of the subject. The second is the presence of yellow pigment, quite like that found in the ovary, in the epididymis of a drake. It is possible that this pigment may have a causal connection with the development of the summer plumage. These matters will be dealt with later. LITERATURE CITED Benstey, R. R. 1916 The mode of secretion of the thryroid gland. Am. Jour. Anat., vol. 19. BorinG, AtIcE M., AND PEARL, RayMonp 1917 Sex studies. IX. Interstitial cells in the reproductive organs of the chicken. Anat. Rec., vol. 13. 1918 Sex studies. XI. Hermaphrodite birds. Jour. Exp. Zool., vol. 25. PeARL, RAYMOND AND Borinc, ALtice M. 1918 Sex studies. X. The corpus luteum in the ovary of the chicken. Am. Jour. Anat., vol. 23: *After the manuscript of this paper was completed, my attention was drawn to the presence of luteal cells in the testes of hen-feathered males by Professor Morgan. An examination of the testes of a young Brown Leghorn male (as adult Brown Leghorns are cock-feathered) showed similar cells, as did pieces of testes taken from a transplant beneath the skin of a castrated Brown Leghorn male, which at maturity was cock-feathered. A further study of such cells and their distribution is being made. ; ii? R4 4 5! . 7 | engiies vii Ais 4h Coutrais cle de Pod at 5 aan x4 ab ~ Hoy / bo aoe 3 ‘ ; Gaia at bers ite et TE ie al " paintG AB OnTOo ‘Roel itr wh ABC DDL: adi ait 4ans nod ia Peyeriite 4G imiuxrs Been. var he Be (310T bu: sis an AME» ee 2 + bisyaisnion pbissioiut ithe i¢ si yet2 pavelp 7369 gol pe rts itidant nisi Ro mialexr 8) CUTE: dheniiiis. te cd VEN aatlarist: ey BTS a a ce spit Webasto sei Bab ailiud! once fh at 25D Resumido por la autora, Ruth Rand Atterbury. La bolsa y ténsila faringeas; nota sobre sus relaciones en el embrién de vaca. El presente estudio es preliminar de una investigacién sobre el origen de la ténsila faringea. Schwabach (1888) y Huber (1912) han demostrado que dicha estructura se desarrolla en el hombre en el punto en que estd colocada la bolsa faringea. Schwabach interpreta la bolsa farfngea como el primer esbozo de la t6nsila y es equivalente al seno tonsilar de la t6nsila palatina. Segtin Huber, la bolsa farfngea del hombre nace por la persis- tencia, en un cierto sitio, de la conexién primitiva entre el noto- cordio y el epitelio faringeo. Una investigacién verificada por la autora sobre el embri6n del cerdo (Rand 1917) ha demostrado que los contactos notocordales notados por los autores citados son secundarios y no estén relacionados, en modo alguno, con el orfgen de la bolsa farfngea, sino que los diverticulos faringeos se desarrollan en intima asociacién con la fascia faringo-basilar. Las observaciones llevadas a cabo sobre el embrién de vaca demuestran: 1) Que no hay nada que corresponda a los contactos notocordales primarios descritos en el embrién humano ni con los contactos secundarios observados en el cerdo. 2) Que la fascia faringo-basilar, en general, no viene a ponerse en relacién directa con el epitelio farfngeo comose ha descrito en el cerdo. 3) La tonsila farfingea se desarrolla con completa independencia de los diverticulos farfingeos. De aqui que la bolsa faringea no pueda considerarse como esencial, o directamente relacionada, eon la aparicién de la t6énsila faringea. Esta ultima, en la vaca, se desarrolla en un drea que coincide generalmente con las fibras terminales de la fascia faringo-basilar. Translation by José F. Nonidez Columbia University AUTHOR’S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, MAY 1 BURSA AND TONSILLA PHARYNGEA; A NOTE ON THE RELATIONS IN THE EMBRYO CALF! RUTH RAND ATTERBURY Department of Histology and Embryology, Cornell University, Ithaca, New York EIGHT FIGURES CONTENTS O) cE Sousa ee oe ees a & ea ae 251 ~ oki) Ti Se ene SS ee ae pe a ee eee ee 5 eee EEG RMMURTINIMIRGIOM Sets ose eas eco ieee osc soccaccsseasess DOA en ed ots) 5) SSO E). SISOS. LL OS SUSIE UE we A 262 ENGI set oe. fits th. bigs 6 otha gtd os 100. the fibers seem to appear in the human embryo ata relatively earlier stage thay in the calf and the pig. In conclusion, it is advisable to emphasize again the me- chanical nature of the pharyngeal outpocketings in the calf (when present) pig, and human embryos. These structures are merely mechanical expressions of the growth conditions of the pharyn- geal region, arising in accordance with the presence of the me- chanical factors determining them. The bursa pharyngea of man therefore cannot be considered a structure of fundamental sig- nificance, essential to the development of the pharyngeal tonsil, BURSA AND TONSILLA PHARYNGEA 263 as Schwabach maintained. If it were essential, its presence might reasonably be expected in the pharyngeal roof of all forms which possess a well-developed pharyngeal tonsil. It has been shown that the calf is a form possessing a well-defined pharyngeal tonsil, but in the calf a pharyngeal pocket corre- sponding to the bursa pharyngea of man does not regularly develop. There is, however, one outstanding feature common to the pharyngeal region of the calf, pig, and human embryos, Fig. 8 Calf embryo, 190 mm. total length. Drawing of a section near the median sagittal plane, showing the location of the developing pharyngeal tonsil in its relation to the fascia pharyngobasilaris, pharyngeal musculature, and pharyngeal epithelium. Glands are drawn in only as they occur in the region of the tonsil. 7J., pharyngeal tonsil; L., larynx; #., esophagus. X 10. namely, a close relationship between the fibers of the fascia pharyngobasilaris and the pharyngeal epithelium. Figure 8 of the pharyngeal region of a 190-mm. calf embryo shows the pharyngeal tonsil developing as diffuse tonsillar tissue in the pharyngeal mucosa, over an area roughly coextensive with the terminations of the radiating fibers of the fascia pharyngobasilaris. 264 RUTH RAND ATTERBURY LITERATURE CITED, Gaap, Susanna Puetps 1906 The notochord of the head in human embryos of the third to the twelfth week, and comparison with other verte- brates. Science, N. S., vol. 24, pp. 295-296. Huser, G. C. 1912 On the relation of the chorda dorsalis to the anlage of the pharyngeal bursa or reeessus medius pharyngis. Anat. Rec., vol. 6, pp. 373-404. Kitiran, G. 1888 Uber die Bursa und Tonsilla pharyngea. Morph. Jahrb., Bd. 14, S. 618-711. 7 Meap, C. 8. 1909 The chondrocranium of an embryo pig, Sus scrofa. Am. Jour. Anat., vol. 9, pp. 167-208. Meyer, R. 1910 Uber die Bildung des Recessus medius pharyngeus, Bursa pharyngea, in Zusammenhang mit der Chorda des menschlichen Em- bryonens. Anat. Anz., Bd. 37, S. 449-453. Miuter, M. M. 1915 A study of the hypophysis of the pig. Anat. Rec., vol. 10, pp. 226-228. Minot, Cuarues 8. 1911 A laboratory text-book of embryology. P. Blacki- ston’s Son & Co., Phila., pp. 62, 237. : Raprorp, M. 1913 A note on the development of the pharyngeal bursa in the ferret embryo. Anat. Anz., Bd. 44, 8. 371-377. Ranp, Rurw 1917 On the relation of the head chorda to the pharyngeal epi- thelium in the pig embryo: a contribution to the development of the bursa and tonsilla pharyngea. Anat. Rec., vol. 13, pp. 465-491. Scuwapacw 1887 Zur Entwickelung der Rachentonsille. Arch. f. Mikr. Anat., Bd. 33, S. 187-213. Tovrnevux, J. P. 1911 Base cartilagineuse du crane et organes annexes., Tou- louse Théses, vol. 71. : ares i193 Mihail PhS 5rTk Hit iS he he fragt te; u ‘ Pahed ss ifoh youre ar ; > < ond ibbarbiles vir ES Ole} oily tzid fof esidol Ww mapayees if Laity) fo pts. +i | BRDET eT iA (inte A, rg i ‘ a) - ~ af »: 4 * a af? ~ ee - - or on . alin ~ ~ ~~ = . Grit Sr die piri s 7 Tabane wet 8h ay irc 7 . yee a : 7 ad DH 7T : St] y isa. Be) te fa i OT aba eh, ee ne ee Mee ra t64 ~ Resumido por el autor, C. V. Morrill. Reversi6n de la simetria e imagenes reflejadas en los monstruos de trucha, con una comparacién de condiciones seme- jantes en los monstruos dobles humanos. En una serie de truchas recién salidas del huevo, que contenfa muchas monstruosidades dobles del tipo dicéfalo, el autor ha encontrado un cierto ntimero en el cual las visceras abdominales de uno de los componentes estaban invertidas simétricamente en cierto grado, produciendo de este modo una imagen compara- ble a la originada por reflexi6n en un espejo. El autor describe estas formas compardindolas con monstruos humanos dicéfalos que presentan la misma condicién. También describe un nuevo caso de ménstruo humano del tipo dicephalus tribrachius dipus con visceras reflejadas y dd una lista de ménstruos semejantes tomada de la literatura sobre este punto. Discute con algtin detenimiento las cuestiones generales de reversién de simetria y produccion de imagenes reflejadas, junto con algunas considera- ciones sobre la relacién de la asimetria visceral y el patr6én de la segmentacion. Translation by José F. Nonidez Columbia University AUTHOR’S ABSTRACT OF THIS PAPER ISSUED By THE BIBLIOGRAPHIC SERVICE, MAY 1 SYMMETRY REVERSAL AND MIRROR IMAGING IN MONSTROUS TROUT AND A COMPARISON WITH SIMILAR CONDITIONS IN HUMAN DOUBLE MONSTERS C. V. MORRILL Department of Anatomy, Cornell University Medical College, New York City EIGHT FIGURES (FOUR PLATES) In the extensive literature on the development of monsters there are frequent references to reversed symmetry and mirror imaging of unpaired organs. This latter condition is most fre-— quently seen in double monsters of the dicephalous type, one component of the monstrosity often exhibiting a partial or com- plete situs inversus viscerum. As far as I am aware, mirror imaging in the viscera has only been found, or at least described, in human monsters, although there is no morphological reason why monsters in the lower vertebrates should not occasionally exhibit this condition. Recently an opportunity occurred to examine the question. In a collection of newly hatched trout! containing many double monsters a number were found in which the abdominal viscera of one component showed reversed sym- metry in some degree. The mirror imaging was practically per- fect in some cases, while in others it was only slightly indicated or irregularities appeared which made the interpretation diffi- cult. The conditions found in these fish, although not entirely novel, seem sufficiently interesting to merit a brief description. Their theoretical bearing will be considered in conjunction with similar conditions in higher forms. 1] am indebted to Professor Stockard for the use of this material which he has under investigation from a somewhat different standpoint. 265 266 C. V. MORRILL SYMMETRY REVERSAL IN MONSTROUS FISH The collection of monsters under consideration exhibited various degrees of doubling. The less pronounced cases showed externally a double-headed condition with the rest of the body single. At the other extreme were duplicate twins attached face to face on a single yolk. Between were the varying degrees of anterior and posterior duplicity. In the figures (pls. 1 and 2) all the specimens are viewed from the ventral or ventrolateral surface with the yolk dissected away. The normal asymmetry of the viscera is shown in each of the twin fish of figure 6. The stomach first bends slightly to the left, then sharply to the right, passing to the pyloric end where it turns posteriorly into the intestine. The swim bladder (S.B.) lies dorsal to and slightly to the left of the stomach. The liver (L.) lies to the right and in the hollow formed by the bend- ing of the stomach. The urinogenital system need not be considered here. In the monsters the doubling of the viscera corresponds in degree with the external duplicity. In the specimens shown in figures 1, 2, 4, and 5 where the head and a considerable part of the trunk are double, there are two stomachs, two swim bladders, and two livers. The intestines are separate anteriorly, but unite posteriorly to form a common rectum which opens in the usual position through a single vent. Also, though not brought out in the figures, there are two hearts and two complete pairs of pectoral fins. In the specimen shown in figure 3, the duplicity does not extend as far posteriorly as in the foregoing. In this case there are two stomachs and two swim bladders, but the intestines are united immediately beyond the stomachs into a common bulbous enlargement (C.J.) from which a single tube leads straight backward to the vent. The liver (L.) is a single, irregular mass nearly twice the size of a normal liver and prob- ably formed by the union of two separate liver buds. It lies on the right side of the monster considered as a unit. The swim bladder corresponding to one of the components (lower in the figure) passes backward on the dorsal side of the liver SYMMETRY-REVERSAL IN MONSTERS 267 and is almost concealed from view. The tip of it can be seen at S.B. The swim bladder of the other component is in the usual position. The specimens shown in figure 6 are twins which have been dissected away from a single yolk uniting them face to face. Each has the normal complement of organs. It will be convenient to designate the two components of a monster as A and B. The lower component in each figure is A and the upper is B. If one imagine the figures turned so that the heads are toward the top of the page then A is on the left of the observer as he faces the ventral surface of the specimen and B is on the right. The scheme then conforms to -that which Wilder (’04, ’08, ’16) has adopted for human monsters. If figures 1, 2, and 4 are now examined, it will be seen that in each case the asymmetry of one component is the reverse of the other, as far as the principal abdominal viscera are concerned. Com- ponent B (upper or right-hand) has the normal asymmetry—the stomach bends first to the left, the swim bladder (S.B.) dorsal and slightly to its left, the liver (L.) on the right side.2 Com- ponent A (lower or left-hand), however, has reversed asym- metry—the stomach bends first to the right, the swim bladder (S.B.) dorsal and slightly to its right, the liver (Z.) on the left side. The transposition of viscera in one component is practically complete in the three specimens shown in figures 1, 2, and 4. In several other specimens (not figured) a partial reversal was indicated. Figure 3 shows a case in which the doubling involves only the head and extreme anterior end of the trunk. There are two hearts, as the figure shows, and four complete pectoral fins. In the abdominal cavity the two stomachs are crowded together and very much contorted, though both seem to bend toward the same side. There is a single large liver (L.), probably formed by the union of two liver buds. The swim bladder of component B (upper or left-hand) is the normal position. Its mate of the opposite side, for the most part hidden by the liver, 2 The rights and lefts are used here with reference to one component, not to the entire monster. 268 C. V. MORRILL does not exhibit the relations to be expected if mirror imaging were present, as a comparison with the position of the swim bladder in the corresponding component of figures 1 and 2 will show. It is difficult to draw any positive conclusion from this specimen, owing to the crowded condition in the abdominal cavity; but from the position of the stomachs and swim bladders, I am inclined to think that no reversed asymmetry is present in either set of organs. It is to be noted that the monsters showing complete reversal of symmetry on one side are all in the same stage of duplicity (figs. 1, 2, and 4). No sign of reversal was found in the more complete stages of doubling. In the twins from the same egg (fig. 6) each has the same (normal) asymmetry. In the case described in the preceding paragraph, where the degree of dou- bling was less marked (fig. 3), all indications were contrary to the idea of symmetry reversal. The mirror imaging described above, whether complete or partial, is not by any means the rule in these monsters. Even in the particular stage of doubling in which it occurs the major- ity of the specimens show the normal asymmetry in both sets of visceral organs. Figure 5 illustrates this latter condition. Here it is obvious that both stomachs bend first toward the left, forming a bay, which opens toward the right. The liver (L.) lies in the bay; that is, on the right side of the component to which it belongs. The swim bladders (S.B.) lie dorsal and slightly to the left of their respective stomachs. Summariz- ing the facts briefly, the abdominal viscera are mirror images of each other in some cases (figs. 1, 2, and 4) and not in others (fig. 5), though most of the specimens present the same degree of dou- bling. Discussion of this point may be conveniently postponed until the conditions in human monsters have been described. Despite the very simple condition of the gastro-intestinal tract in fish, the normal asymmetry is very well marked, and any change from this strikes the eye immediately upon opening the abdominal cavity. In view of this, it is curious that neither Windle (’95) nor Gemmill (’01, ’12) observed any reversal of sym- SYMMETRY-REVERSAL IN MONSTERS 269 metry in their work on monstrous fish. Gemmill especially, in his study of the anatomy of double monstrosities in trout (’01), made a careful examination of the internal organs, but apparently saw no changes in symmetry, though some of his specimens exhibited the same degree of duplicity as those described in the present paper. Possibly reversal of symmetry is rare in fish, but the total number of specimens examined is too small to form any definite conclusion on this point. SYMMETRY REVERSAL IN HUMAN MONSTERS It is well known that some types of human diplopagi exhibit symmetry reversal and consequent mirror imaging in the un- paired viscera. At the time the present study was begun my attention was drawn to a double-headed human monster which had been kept in a museum jar for a number of years in our laboratory. It belongs to the dicephalous variety, which is the one most likely to show mirror imaging in the viscera, judg- ing by previous reports.? A photograph of the monster is shown on plate 4 (fig. 8). Using Fisher’s (’66) classification, it would be a dicephalus tribrachius dipus. The name is so descriptive - that nothing more need be said of the external configuration. It is similar to the Barkow fetus, no 66 in Fisher’s list, except that in the present specimen the hands of the median arm are placed palm to palm and the heads are somewhat nearer to- gether. The sex, as in Barkow’s case, is female. The organs of the upper abdomen, with the exception of the liver, show complete mirror imaging (fig. 7). The stomachs are placed with the pyloric ends pointing toward each other. There are two spleens (Sp.), two gall-bladders (G.B.), two com- mon bile-ducts, and two pancreases (Pan.), each set showing the proper relations to the corresponding stomach. (The he- patic ducts are not shown in the diagram.) The liver forms a large compound mass with many irregular lobules (not figured). The small intestines are separate to within two feet of the caecum, at which point they unite to form acommon ileum. The 3 Fisher (’66); Eichwald (’70); Hirst and Piersol (’93). 270 Cc. V. MORRILL large intestine, including caecum (Cae.) and appendix, is single. The caecum lies in the right iliac fossa, from which the colon (A.C.) ascends in the usual way to the liver. From here the transverse colon crosses to the hypochondrium of the opposite side, the splenic flexure lying in relation to the spleen of the B-component (right-hand, as one faces the monster). The de- scending colon has the usual course and relations. There is a single pair of large, lobulated kidneys, each with a normal ureter. The uterus, tubes, ovaries, bladder, and rectum are normal in size and position for a single individual. The viscera are shown schematically in figure 7. It will be observed that the organs of the right-hand component (B) are more normal in shape and larger than those of the opposite side and that they have the normal situs. There is a disparity in size also in the thoracic organs (to be described below), again in favor of component B. The two heads, however, are prac- tically the same size though one, again the right-hand (compo- nent B) is placed a little more in the direct line of the compound body. The thoracic cavity contains two complete sets of organs. There are two hearts inclosed in a single pericardium. They were pressed close together, both apices directed forward and down- ward. In figure 7 the apices have been widely separated to show the medial surfaces of both hearts. It is obvious that there is transposition in the left-hand heart (component A). The left atrium in this case receives venous blood from the superior vena cava (S.V.C.) and hepatic veins (Vv.h) and de- livers it to the left ventricle from which the pulmonary artery (P.A.) springs. The pulmonary veins here empty into the right atrium and the aorta (A.) springs from the right ventricle. The reversal of symmetry is thus complete. The right-hand heart shows the normal symmetry and need not be described. The lungs consist of two pairs corresponding to the two tracheae. Those of the right-hand, or B-component, are normal in lobu- ‘ A detailed description of the vessels in this monster will appear in a forth- coming paper by Mr. H. B. Sutton, of our laboratory. SYMMETRY-REVERSAL IN MONSTERS 271 lation and nearly so in size, while the other pair are much re- duced.> There is a fair-sized thymus for each side. The diaphragm is extremely incomplete. Both stomachs with their adnexa are herniated high into the thorax, especially in component A, the fundus of whose stomach lies almost in the root of the neck. Summarizing the more important features of the monster: The viscera of the right-hand component (B) are the more nor- mal in size and shape and they have the normal situs. Those of the other component (A) are, generally speaking, reduced in size or irregular in shape and display situs inversus. Externally the head and neck of component B are more nearly in line with the axis of the trunk. Obviously it is always the same compo- nent (the left-hand or A-component according to the scheme adopted in the present paper) which exhibits transposition of the viscera whether in man or in fish (compare figs. 1, 2, 4, and 7).. This point will be discussed in a later part of the paper. It has been frequently assumed, as pointed out above, that monsters of certain types, especially the dicephali and ischio- pagi, may be expected to exhibit mirror imaging. The number | of cases on record where actual examination disclosed this con- dition are however, very few,° and most of them are cited by Fisher (’66) in his very comprehensive paper on diploteratology. From his list I have collected the following cases: Dicephali Case 50. Ritta-Christina, a dicephalus tetrabrachius dipus (fe- male) which lived about eight months. At autopsy it was found that the pericardium was single, but enclosed two hearts, which were right and left, touching at their apices. The stomachs, spleens, and pan- creases were right and left, and placed so that the pyloric ends of the stomachs faced each other, the adnexa conforming as in the specimens described above. The livers, also right and left, were fused, and there >It was impossible to determine the lobulation of this pair of lungs as they were partly destroyed when the corresponding head was removed from the body during delivery. ® Bateson (94) states that Eichwald (Pet. med. Zeitsch., 1870) found some transposition of viscera in thoracopagi, though to a varying extent. The original paper was not available to the writer. 272 Cc. V. MORRILL were two gall-bladders which occupied a median position. Other de- tails need not be given here. Case 60. Dicephalus tribrachius tripus (male). This specimen had two hearts, one right and one left, in a single pericardium; two aortae, one transposed, i.e., lying on the right of the vertebral column. The liver consisted of three portions, two lateral, each of which corre- sponded to the right lobe of a normal liver, one of them reversed, and a median lobe corresponding to the left lobes of normal livers fused. Each lateral lobe had ‘a bile duct, gall-bladder, common duct, and portal vein symmetrically placed. There were two stomachs with pyloric ends turned toward each other; the fundus of that belonging to ‘A’ was in the right hypochondrium and therefore reversed, while that of ‘B’ had the usual position in the left. A spleen was connected with each. % Case 71. Dicephalus dibrachius dipus (Gruber); sex not stated. There were two food passages; two stomachs with the fundus of each turned outward, and two intestines to within five inches of the lower end of the ileum. There was a large compound liver, two gall-bladders, and two bile-ducts; no pancreas; one spleen on left stomach; two hearts, the right small and imperfect; two sets of lungs and tracheae; uro- genital organs single and normal. Case 74. Dicephalus dibrachius dipus (Horner); sex, male. The thorax contained a compound heart. There were normal right and left lungs and a third compound lung due to the coalescence of adjacent lungs of different foetuses. The liver was single, but compound with increased number of lobes. The gall-bladder was double with a com- mon duct which terminated in two orifices, one for each duodenum. There were two stomachs, one on the right, the other on the left, having their pyloric orifices pointing towards each other. The two small intestines, more or less aherent, finally blended into a single tube. The colon was single. There were two pancreases, but only one spleen, which was attached to the larger left stomach. The kidneys were a single large pair; the bladder and genitals were single. Case 102. Dicephalus monauchenos (White); sex, female. There were two stomachs, the left in the usual place, the right reversed, its larger extremity towards the right. The two were united at the py- lorus and opened into a common duodenum. The liver was single and very large. One further specimen may properly be placed with the foregoing five, namely, a dicephalus dibrachius dipus (female) described by Fisher (case 76) which possessed a single globular stomach with right and left fundus resulting from the fusion of two stomachs. An oesophagus from each mouth entered the compound stomach nearly at the same point. The liver and intestines were single. The cases cited above together with the one given in the present paper, are the only definitely described cases of mirror SYMMETRY-REVERSAL IN MONSTERS 273 imaging in the dicephali. Unfortunately, the position of the viscera is not stated in the reports of Barkow’s fetus (tribrachius dipus) and Ruggles’ fetus (dibrachius dipus). Both of these had two stomachs, and it seems almost certain that one was transposed. We have, then, six certain cases of transposition and one indication of this condition in the fetus having a com- pound stomach with right and left fundus. Incidentally, one rather striking fact is brought out in look- ing over the various reports. In human monsters the amount of doubling in the viscera does not necessarily correspond with the amount of external doubling, as was the case in trout. Fisher cites one case (no. 64, from Benedina), a dicephalus tribrachius tripus (male) in which the gall-bladder, stomach, pancreas, spleen, and intestines were all single, although there were two hearts, two uriary bladders, and two pairs of kidneys. Com- pare this with the two cases of dicephalus dibrachius dipus (nos. | 71 and 74) in which the digestive systems were double as far as the lower part of the ileum. It must be admitted that Bene- dina’s case, if correctly reported, is very unusual. Among other classes of diplopagi in which the two compo- nents are more widely separated, it is difficult to find definite — information on the position of the viscera. Bateson (’94) quotes Eichwald (l.c.) to the effect that the thoracopagous monsters examined by him showed, in almost every case, some transposi- tion of the viscera of one of the bodies, though to a varying extent. The pygopagous ‘Carolina twins,’ Millie-Christina (colored), were examined while living, and it was reported that “the apex of Christina’s heart is on her left side while that of Millie is distinctly felt in the right side.””, Gemmnill (’02) reports a case of ischiopagus tripus (human) in which modified transpo- sition occurred in the liver. His figure 14 seems to indicate transposition of the thoracic viscera of one component as well, but the author does not comment on it. Windle (94) gives a report on the ‘Orissa sisters,’ Radica-Doodica, who were united in the thoracic region (xiphopagus or thoracopagus). Regard- ing the position of the viscera, he states that authorities differ as to whether one was situs inversus viscerum. In the case of 274 Cc. V. MORRILL the famous Siamese twins, one of them is stated to have had a partial reversal of viscera. These few reports, meager as they are, show that some trace of visceral transposition or symmetry reversal may occur in monsters other than dicephali. In the syncephali, including Janus monsters, transposition of the viscera in one component apparently does not occur, though it seems to me in one case a slight indication was observed. Wilder (’08), in describing a case of this kind (the “Baldwin synote’) makes the following statement:— ‘The common oesophagus leads into a common stomach, though evidently one formed of two components, since it presents two cardiac enlarge- ments one on either side of the oesophagus” (italics mine). ‘‘The outline of the stomach is thus heart-shaped, but is not quite symmetrical, since the cardiac lobe of component A is a little larger than that of Component B.” With regard to the re- maining organs the author states that there is no trace of ‘look- ing-glass symmetry.’ The stomach of this synote is thus simi- lar to that of Fisher’s dicephalus (case 76) mentioned above. Among other mammals a number of syncephali have been de- scribed: kitten, McIntosh (’68); cat, Reese (’11); pig, Carey (17), but none apparently showed any trace of mirror imaging. Kaestner (’07) has described in detail several syncephalous chick embryos, with especial reference to the heart region, but they were not far enough advanced in development to show the position of the abdominal viscera. Bishop (’08) gives an ac- count of the heart and anterior arteries in several dicephalous reptiles, but as no pronounced asymmetry of the heart is vis- ible in this class of vertebrates, there is little opportunity to look for mirror imaging. In cases where two hearts were pres- ent, both aortic arches developed on each side. It is unfortunate that among the large number of double monsters reported so much attention has been paid to external features and so little to the position of the abdominal viscera. SYMMETRY-REVERSAL IN MONSTERS 275 DISCUSSION The question of symmetry reversal and mirror imaging has ‘been discussed most recently by Wilder (’04, ’16), Bateson (’16), and Newman (’16, 717). It seems to be generally agreed that transposition of the viscera does not occur in human duplicate twins. In armadillo quadruplets Newman finds, after examina- tion of a considerable number of sets, that no symmetry reversal is present in the viscera. The same is true in the duplicate twin trout (fig. 6) described in the present paper. Some mirror imaging, however, does occur in human duplicate twins and armadillo quadruplets, but it is confined to the integumentary structures (friction-skin patterns in the former case, arrange- ment of the scutes and bands in the latter). The integument of young trout, unfortunately, does not present any regular pat- tern of asymmetry, at least none could be detected, and thus yields no information on this point. In double monsters, how- ever, it is admitted that a certain amount of symmetry reversal in the viscera is to be expected, although it may not occur in every case. Fisher, in 1866, clearly expressed this opinion, and is quoted by Wilder (’04) to this effect. Wilder, though also quot- ing Bateson’s (94) opinion, in agreement with Fisher, seems un- willing to admit the importance of this phenomenon and gives little space in his earlier paper (l.c.) to its discussion. In a later paper (’16), however, he discusses a very interesting case of mirror imaging in the friction-skin patterns of a human diplopage. Newman, (716, ’17) has given perhaps the fullest discussion of symmetry reversal, both in multiple births and in monsters. The relations of symmetry observed in armadillo quadruplets are, he considers, “‘the results of an intricate interplay of three grades of successively operating symmetry systems, the later tending to obliterate the effect of the earlier, but not always suc- — cessfully.”” This conclusion is based on the nature of the poly- embryonic development observed in these animals and is ex- plained by Newman as follows: ‘‘When the primary outgrowths are formed (i.e., fission in the blastocyst stage), they are the product of the antimeric halves of the first embryo and should THE ANATOMICAL RECORD, VOL. 16, No. 4 276 Cc. V. MORRILL therefore show mirror-image relations. But a partial physio- logical isolation of the two halves permits a certain reorganiza- tion, or regulation of new symmetry relations, which tends more or less completely to destroy the original symmetry, yet often leaving a trace of the latter. Similarly, when the second- ary outgrowths arise between the primary ones a certain residuum of the primary symmetry may be carried over that frequently manifests itself in mirror imaging between twins derived from one-half of the original embryo. Finally, when each secondary outgrowth organizes its own bilateral symmetry, it tends to lose, partially at least, the earlier symmetry relations and to estab- lish its own mirror imagings of right and left sides” (third grade of symmetry). It must of course be remembered that in arma- dillos, mirror imaging between twins is confined to integumentary structures. In the case of duplicate twins and double monsters, there would be according to Newman’s conclusion, only two ‘grades of symmetry systems.’ Any mirror imaging present in a monster would thus be evidence of the potency of a primary symmetry which had not been overcome by the secondary sym- metry acquired later by the separate components. If physio- logical isolation occurs in a comparatively early stage, there will be, he thinks, very little mirror imaging, as the secondary symmetry will have more time to operate. Conversely, if it appears somewhat later, there will be more mirror imaging. In consequence, double monsters probably arise somewhat later in ontogeny than duplicate twins, since the former more often show evidence of mirror imaging. Newman’s suggestion regarding primary and secondary sym- metry systems is to some extent supported by the conditions found in trout monsters. However, by far the greater num- ber of these monsters, of whatever degree of doubling, show no influence of a primary system of symmetry, that is a symmetry of the monster taken as a unit. On the contrary, each com- ponent develops its own system (secondary, according to New- man) as if it were entirely disconnected from its mate (fig. 5), and this symmetry (asymmetry), moreover, is the same as that of a normal fish. It is interesting to note that in the type of SYMMETRY-REVERSAL IN MONSTERS PAS double monster known as autosite-and-parasite, a number of which occurred in the present collection, the parasite, whenever it was of sufficient size to possess a complete set of abdominal organs, always exhibited its own (secondary) symmetry and never appeared as a mirror image of the autosite. It is only in a small proportion of the monsters that the primary symmetry of the whole is still potent, in which case mirror imaging appears in the viscera (figs. 1, 2, and 4). Newman’s further suggestion that there is a direct relation between the occurrence of mirror imaging and the period in ontog- eny at which doubling takes place, does not accord with what seems to be the mode of origin of monsters in fish. In this form, the initial doubling probably always occurs at the same period of development, regardless of the degree of separation of the two components. This period corresponds with the first appear- ance of the embryonic anlage at the circumference of the blas-. toderm, as Kopsch (’99) concluded in his analysis of the causes of fish monsters.7? In the case of double monsters, two em- bryonic anlages are formed at the same time. The degree of doubling will then depend on how near the two anlages lie to each other. On this view, mirror imaging and the time at which doubling first appears cannot be causally related. Nor is there a very precise relation, it seems to me, between the amount of separation of the two components and the occurrence of mirror imaging. It is true that there is a stage of doubling more favor- able than others for exhibiting symmetry reversal in one com- ponent, but only a small proportion of the monsters even then show any evidence of this condition (compare figs. 4 and 5). Furthermore, specimens showing less separation than in the stage just mentioned might be expected to exhibit more evi- dence of primary symmetry (symmetry of the monster as a whole) and therefore more mirror imaging, while in point of fact the contrary is true (p. 268). It was pointed out (p. 271) that in both fish and human mon- sters it is always the same component (the left-hand or A-com- 7 This view apparently originated with Lereboullet. Kopsch has developed it in considerable detail in the paper referred to above. 278 Cc. V. MORRILL ponent) which exhibits transposition of the viscera. In this the writer agrees with Eichwald, as quoted by Bateson (’94), except that the latter uses the term ‘right twin’ for what is here called left-hand or A-component. Bateson himself is in doubt on this point and quotes Kiichenmeister® to the effect that in xiphopagous twins it may not be possible to say which is the right and which the left. This objection, however, does not apply to dicephalous forms, whether fish or human. Here the undivided portion of the monster obviously has dorsal and ventral surfaces and these may be traced without interruption into the corresponding surfaces of the two components which usually face each other to some extent. The right-hand and left-hand components are thus easily distinguished. In cases where mirror imaging occurs, the arrangement of the two sets of organs is always the same*—the stomachs bend first toward the Jateral borders of the monster (taken as a unit), then toward the median plane (plane of union) so that their pyloric ends face each other; the livers lie close together or are fused. It is difficult, however, to find an explanation for this fact, for even if the reverse arrangement occurred, there would still be mirror imaging—the fundus of one stomach facing that of the other, the pyloric ends pointing in opposite directions, the livers lying on the lateral borders of the monster. It has sometimes been assumed that in normal development the direction of growth taken by the liver bud determines the plan of asymmetry of the remaining viscera. In the case of monsters having either two livers or a composite liver, it might be further assumed that the two liver buds, having formed independently, were drawn to- gether by some sort of mutual attraction. If such a movement took place, the anterior ends of the two intestines together with ® Die. angeb. Verlagerung d. Eingeweide d. Menschen, Leipzig, 1883. The original was not available to the writer. * An exception to this appears to have been found in the famous Siamese twins where it was Chang, the left twin (right-hand or B-component of the present paper), in whose body there were indications of situs inversus (Kiichen- meister, quoted from Bateson). This would give the converse of the usual arrangement. The writer has not had access to the original description of these interesting twins. SYMMETRY-REVERSAL IN MONSTERS 279 the pyloric ends of the stomachs would be drawn with the livers toward the plane of union. This would result in the arrangement found in practically all monsters in which mirror imaging occurs. While the above assumptions do, to some extent, account for the facts, there is some evidence to show, as will be pointed out below, that the factors controlling asymmetry are located in the primitive gut and become operative before the liver bud has developed. It must be admitted that we are still in the dark regarding the causal factors underlying the conditions of mirror imaging found in some types of monsters. The question here arises, why, in a certain stage of doubling, should mirror imaging occasionally appear and not always? One might assume that the rate of development in one component of a monster occa- sionally becomes a little slower than in its mate so that it tends to fall behind and is unable to develop or express an independent system of symmetry like that of a normal embryo. In this case the lagging component might be thought of as sharing with its more vigorous mate in a single system of symmetry, that is, the symmetry of the monster taken as a unit. The result would then be a mirror-imaged condition of the viscera. A suggestion of inferiority in one component was noted in the human monster described above where the transposed set of organs were found to be slightly smaller and more irregular in shape than those of the opposite side. In the fish monsters, however, no such in- equality between the two sets of organs was observed. Further- more, the assumption that a retardation of development in one component predisposes to transposition of viscera is rendered improbable by the conditions found in monsters of the autosite- and-parasite type. Here it may be fairly assumed that the parasite tends to be weaker than the autosite, and in fact is often defective; still whatever asymmetry exists in the parasite is that of a normal fish and never reversed. In other words, the parasite, even in its failing struggle for existence, retains the power to develop its plan of asymmetry as a separate individual. It is extremely difficult to formulate a theory which will satisfactorily account for a condition so casual in its appear- 280 Cc. V. MORRILL ance as mirror imaging in the viscera of monsters, and further work on the early developmental stages of these forms is neces- sary before any definite conclusions can be drawn. The solution will, of course, involve the more fundamental problem of what determines the normal asymmetry of unpaired organs and why single individuals occasionally appear with transposed organs.?° Very little progress has been made in this direction. The most suggestive observations in the field are those of Pressler (’11), on experimentally produced situs inversus in Bombinator. The material was obtained from Spemann who performed the fol- lowing experiment: In the neurula stage, a four-sided piece of the medullary plate together with a portion of the roof of the primitive gut lying under it was cut out and replaced in reversed position, so that the anterior extremity of the piece was directed posteriorly, the posterior extremity, anteriorly. From these experimental embryos, tadpoles were reared which showed in many cases a complete situs inversus viscerum. It has some- times been assumed, as stated above, that the asymmetrical growth of the liver bud normally toward the right influences the position of the remaining organs. The question then arises, what determines the direction of growth of the liver bud? Spe- mann and Pressler’s work seems to indicate that the factors controlling asymmetry are located in the primitive gut and prob- ably arranged in such a fashion as to cause the gut in normal development to bend first toward the left, thus forcing the liver bud to grow toward the right. We may suppose that when the arrangement of these factors is reversed, as in the experiment, 10 Bateson (’94, p. 560) points out that cases of this kind cannot be explained on the ground that one member of duplicate twins has died or failed to develop, since it has been shown that in duplicate twins neither member has transposed viscera. Conversely, Kiichenmeister (l.c.) collected 152 cases of transposition, of which only one could be shown to have been a twin. A somewhat similar suggestion has been made to account for cases of situs inversus in single individuals, namely, that this condition results from com- plete reduction of one component of a monster (autosite- and-parasite) in which mirror imaging occurred. The autosite in this instance must necessarily present the reversed asymmetry. In some cases it is thought that the parasite is taken into the body of the autosite during development, and gives rise to certain kinds of tumors. As far as I am aware, there is no evidence recorded that individuals with complete situs inversus have possessed tumors of this sort. SYMMETRY-REVERSAL IN MONSTERS 281 transposition is produced. Pressler’s observations are, it seems to me, very important and indicate the direction along which further experiments should be made to determine the cause of asymmetry. They do not, however, throw any light on the cause of transposition in integumentary structures as found by Newman and Wilder. A very interesting suggestion as to the cause of asymmetry in the viscera is based upon the fact, first pointed out by Crampton (94), that in certain gasteropods the position assumed by the adult organs is correlated with the early segmentation pattern. In these snails the more usual type of asymmetry with dextral shell is associated with a right-handed spiral cleavage. Some forms, however, such as Physa (Crampton) and Ancylus rivu- larius (Holmes), have normally sinistral shells and reversed asymmetry in the viscera; this condition was found to be asso- ciated with a reversal of cleavage. These. observations very naturally led to the view that in gasteropods there is a causal relation between cleavage pattern and the type of asymmetry found in the adult." It is very questionable, I think, whether this conception of the primary cause of asymmetry can be ap- plied to vertebrates. For in monsters, as has been shown, two sets of organs may develop as mirror images of each other, one with normal, the other with reversed asymmetry, though ob- viously both have arisen at the same period of development from a single blastoderm. It is difficult to imagine how changes in _ early cleavage pattern, if such occur in higher forms, could bring about the development of two types of asymmetry in the same embryo, as in the case just cited. From the evidence at hand, it seems probable that the primary cause of visceral asymmetry in vertebrates is to be sought for at the completion of cleavage rather than in the period of cleavage itself. 11 This view, first expressed tentatively by Crampton (’94), was later more fully developed by Conklin (’97, Jour. Morph., vol. 13) and by Holmes (’99, Amer. Nat.; 00, Jour. Morph., vol. 16) on the basis of additional evidence. Conklin more recently (Heredity and Environment, 2nd ed., 1917, p. 177) has expressed the opinion that the correlation between inversion of cleavage and in- version of symmetry observed in certain snails, will be found ‘‘probably in all animals showing inverse symmetry’’ (italics mine). I do not believe this latter generalization is warranted for the reasons given in the discussion (see beyond). 282 Cc. V. MORRILL LITERATURE CITED Bateson, W. 1894 Materials for the study of variation. London (Macmillan & Co.). 1916 Problems of genetics. New Haven (Yale Univ. Press). Bisuop, M. 1908 Heart and anterior arteries in monsters of the dicephalus group; a comparative study of cosmobia. Am. Jour. Anat., vol. 8. Carey, E. 1917 The anatomy of a double pig, Syncephalus thoracopagus, with especial consideration of the genetic significance of the circulatory apparatus. Anat. Rec., vol. 12. Crampton, H. E. 1894 Reversal of cleavage in a sinistral gasteropod. Ann. New York Acad. Sci., vol. 8. Fisuer, G. J. 1866 Diploteratology. Trans. Med. Soc. State of N. Y. GremmitL, J. F. 1901 The anatomy of symmetrical double monstrosities in the trout. Proc. Roy. Soc. London, vol. 68, no, 444. 1902 An ischiopagus tripus (human), with special reference to the anatomy of the composite limb. Jour. Anat. and Phys., vol. 36. 1912 The teratology of fishes. Glasgow (James Maclehose & Sons). Hirst AND Piersot 1893 Human monstrosities, 4 vols. Philadelphia. KarstNer, A. 1907 Doppelbildungen an Vogelkeimscheiben. Arch. f. Anat. u. Phys. Korscu, Fr. 1899 Die Organisation der Hemididymi und Anadidymi der Knochenfische und ihre Bedeutung fiir die Theorien itiber Bildung und Wachstum des Knochenfischembryos. Internat. Monatsschr. f. Anat. u. Phys., Bd. 16. McIntosH 1868 Notes on the structure of a monstrous kitten. Jour. Anat. and Phys., no. 2. Newman, H. H. 1916 Heredity and organic symmetry in armadillo quadrup- lets. II. Mode of inheritance of double scutes and a discussion of organic symmetry. Biol. Bull., vol. 30. 1917 The biology of twins. Univ. of Chicago Press. Pressier, K. 1911 Beobachtungen und Versuche iiber den normalen und inversen Situs viscerum et cordis bei Anurenlarven. Arch. f. Entw.- Mech., Bd. 32. . Reese, A.M. 1911 The anatomy of a double cat. Anat. Rec., vol. 5. Witver, H. H. 1904 Duplicate twins and double monsters. Am. Jour. Anat., vol. 3. 1908 The morphology of cosmobia; speculations concerning the sig- nificance of certain types of monsters. Ibid., voi. 8. 1916 Palm and sole studies, part II. Biol. Bull., vol. 30. Winpte, B. C. A. 1894 Report on ‘Radica-Doodica.’ Jour. Anat. and Phys., vol. 28. 1895 On double malformations amongst fishes. Proc. Zool. Soc. London, pt. 3. ' ty > Pi - ~ . ¢ » PLATE 1 EXPLANATION OF FIGURES 1 and 2 Specimens of monstrous trout showing complete mirror imaging in the abdominal viscera, ventrolateral view. S.B., swim bladder; L., liver; the stomach and intestine are not labeled. The position of the viscera in one com- ponent is the reverse of that in the other. 3 Specimen in which doubling is less extensive than in the foregoing (1 and 2). There are two stomachs and two swim bladders (S.B.) one of which is almost concealed by the compound liver (L.) The intestines unite immediately beyond the stomachs into a common enlargement (C.J.). Apparently no mirror imaging is present in this case. The two pear-shaped bodies anterior to the abdominal cavity are the hearts. SYMMETRY-REVERSAL IN MONSTERS PLATE 1 Cc. V. MORRILL if H. Murayama, del PLATE 2 EXPLANATION OF FIGURES 4 Specimen showing complete mirror imaging similar to figures 1 and 2 except that the two sets of organs are closer together, the livers (L.) almost in contact; ventrolateral view. 5 Specimen showing the position of the viscera in the majority of monsters in this stage of doubling. The normal situs is present in both components; ventrolateral view. Compare with figure 4. 6 Twins from the same egg. These two specimens lay on opposite sides of a single yolk mass. The position of the viscera is the same in both, the liver on the right, the stomach bulging toward the left as in normal fish (i.e., normal situs in both). 286 PLATE 2 SYMMETRY-REVERSAL IN MONSTERS Cc. V. MORRILL H. Murayama, del. PLATE 3 EXPLANATION OF FIGURES 7 Diagram of the viscera in the human monster shown in figure 8. The compound liver is omitted. The apices of the two hearts are widely separated to show the medial surfaces which were in close contact. A., aorta; P.A., pul- monary artery (origin); S.V.C., superior vena cava; Vv.h., hepatic veins; Sp. spleen; Pan., pancreas; G.B., gall-bladder; A.C., ascending colon; Cae., caecum 288 PLATE 3 SYMMETRY-REVERSAL IN MONSTERS c. V. MORRILL H. Murayama, del. 289 PLATE 4 EXPLANATION OF FIGURE 8 Photograph of the human dicephalus tribrachius dipus described in the present paper. 290 SYMMETRY-REVERSAL IN MONSTERS PLATE 4 Cc. V. MORRILL 291 pigurd 1: > Ob aioe Bevis €O! io oun weno 2a : ‘pt B ohsiei SH Somos botysin off turnane ste « A 7 , . - ond TP ERA | : : f : , an . Pao, ry : fe oh UDINE : ; i? «oe Shed ares: iy f [ oA A 52 ir 1 4 R pt : : > eT e ial det if ie Le td > and - r i os os icine sis es ee aun Be wit : i oR ns fi iS te Bh A e At : erie!) MG : ; Mae, +! : ; Bo war. 2 j 7 iat y 2 ‘ > = dp is atds 4 ss ' Banonn Ci 5 t ” \ isé > Resumen por el autor, Edward Phelps Allis, Jr. Mentone, Francia. La inervacién de los miisculos intermandibulares y geniohioideos de los peces 6seos. El miusculo geniohioideo, llamado asi con propiedad, esta inervado exclusivamente por el trigémino en los diversos peces 6seos examinados. En ciertos teleésteos une parte del Ilamado geniohioideo en las descripciones corrientes, estd inervada por el nervio facial y de aqui ha nacido la suposicién de que este miisculo tiene una doble inervacién, representando un miusculo, derivado de un segmento del cuerpo, que esté adquiriendo inervacién por el nervio de otro segmento. Esta suposicién es, sin embargo, errénea, porque la parte del geniohioideo inervada por el facial se ha diferenciado del mitsculo hiohioideo y en los adultos de Esox, Scomber, Gadus y Silurus se encuentran estruc- turas que representan estados sucesivos de su desarrollo. El llamado geniohioideo en estos peces se forma por consiguiente a expensas de dos musculos claramente diferentes, uno derivado del miotomo mandibular y el otro del miotomo hial y las inerva- ciones presentes actualmente son primarias y no secundarias. La parte derivada del miotomo hial esta colocada exteriormente a los radios branquidstegos y, a causa de esta posicién y su deri- vacion del hiohioideo, puede Ilamarse hiohioideo superficial. A veces forma un mitisculo continuo con el geniohioideo superior pero en todo caso, en los ejemplares examinados, esta separado de este por una fascia membranosa o una aponeurosis transversa. Translation by José F. Nonidez Columbia University AUTHOR’S ABSTRACT OF THIS PAPER ISSUED BY THE BIRLIOGRAPHIC SERVICE, JUNE 30 THE INNERVATION OF THE INTERMANDIBULARIS AND GENIOHYOIDEUS MUSCLES OF THE BONY FISHES EDWARD PHELPS ALLIS, JR. Menton, France ONE FIGURE The so-called musculi intermandibularis and geniohyoideus of Vetter’s (’78) descriptions of the bony fishes are among those muscles of vertebrates that are frequently said to vary greatly in the manner of their innervation, Holmqvist even saying (711, _ p. 68), in a work relating especially to them, that their innerva- tion is so variable that it has no morphological significance what- ever. Work that I have under way on the cranial anatomy of Polypterus having led me to doubt this statement, I have had the innervation of these muscles carefully traced not only in this | fish, but also in several of the Teleostei. The work on the adult Polypterus was done by my assistant Mr. Jujiro Nomura, the work on the other fishes by Mr. John Henry. In Polypterus there is no muscle that corresponds, topographi- cally, to the intermandibularis of Vetter’s descriptions of other fishes, but there are two muscles that correspond to the superior and inferior divisions of the geniohyoideus of that. author’s de- scriptions of those fishes, and that strikingly resemble, in their topographical relations, the two so-named muscles of my de- scriptions of Amia (Allis, 97). These two muscles of Polypterus were first described by Pollard (’92), and were called by him the intermaxillaris anterior and intermaxillaris posterior, some fib- ers of the latter muscle being said by him to be continued onward, as intrinsic muscles, into the mantle flap. Another and separate muscle, which corresponds to the hyohyoideus inferior of Vetter’s descriptions of other fishes, is called by Pollard both the mantle 293 294 EDWARD PHELPS ALLIS, JR. muscle and the muscle of the jugular plate, and it also is said to send some fibers into the mantle. The intermaxillaris anterior (geniohyoideus inferior) is said by Pollard to be innervated by a branch of the ramus mandibularis trigemini, the intermaxillaris posterior (geniohyoideus superior) and the mantle muscle (hyo- hyoideus inferior) both being innervated by branches of the ramus hyoideus facialis. Holmavist, in 1910, briefly describes these muscles of Polyp- terus, but he calls the intermaxillaris anterior of Pollard’s de- scriptions the intermandibularis, and the intermaxillaris posterior the protractor hyoidei. The fibers of this latter muscle are said to run insensibly into those of the hyohyoideus, without apparent limiting boundary, this latter muscle being the mantle muscle of Pollard’s descriptions. The innervation of these muscles of Polypterus, individually, is not given by Holmqvist, but it is said that, in the bony fishes in general, the intermandibularis and the anterior portion of the protractor hyoidei are innervated by the trigeminus, and the posterior portion of the latter muscle by branches of the nervus facialis, and it is evident that this is in- tended to apply to Polypterus as well as to the Holostei and Teleostei. That part of the protractor hyoidei of all these fishes that is innervated by the facialis is said to be its most primitive (alteste) portion, the part innervated by the trigeminus being a later acquisition (ein spiterer Erwerb). In a later work, Holmqvist (’11) says that the intermandibu- laris and protractor hyoidei of all of the bony fishes are derived, respectively, from the mandibular and hyal portions of the primi- tive musculus constrictor ventralis. The primitive condition of the intermandibularis is said to have been that of a muscle ex- tending transversely from one ramus of the mandible to the other, and this condition is said to be actually found in the Selachii, in Lepidosteus, and in certain of the more primitive Teleostei. In the remainder of the Teleostei, and in Amia, the muscle is said to have undergone a vertical cleavage into two parts, one of which is called the intermandibularis I and the other the inter- mandibularis IT, these two muscles of Amia being the intermandib- ularis and geniohyoideus inferior of my descriptions of that fish. INNERVATION OF MUSCLES OF BONY FISHES 295 Fig. 1 Ventral view of the head of Polypterus, the skin removed so as to show the geniohyoideus and hyohyoideus muscles. Ghi, musculus geniohyoideus in- ferior; Ghs, musculus geniohyoideus superior; hi, musculus hyohyoideus inferior; nt, branch of nervus trigeminus; nf, branch of nervus facialis. 296 EDWARD PHELPS ALLIS, JR. These two so-called intermandibularis muscles are said by Holm- qvist to both be innervated, either wholly or in part, by a branch of the trigeminus which he calls the ramus mylohyoideus. The protractor hyoidei is said to be innervated by the ramus hyoideus facialis, and to be derived from that deeper layer of the constrictor ventralis of the Selachii that has its insertion on the ceratohyal. Holmaqvist then says (l.c., p. 70) that these muscles of the Ganoi- dei and Teleostei have, like the corresponding muscles of the Selachii, a double innervation, by the nervi trigeminus and facialis, and that the extent to which they are innervated by one or the other of these two nerves varies greatly, and he attributes this variation, not to the abortion of any portion of the muscle fibers concerned, but either to the abortion of certain fibers of one of the two nerves, and the secondary innervation of the mus- cle fibers that they are primarily innervated by fibers of the other nerve, or to a change of course of certain motor fibers of the facialis, those fibers abandoning their normal course to fol- low the path of the trigeminus, and then gradually forcing the fibers of the latter nerve away from the muscle fibers that they primarily innervated, and there supplanting them. Edgeworth (711) also discusses these muscles of Polypterus, and he introduces still another name for one of them, calling the intermaxillaris anterior of Pollard’s descriptions the interman- dibularis, and the intermaxillaris posterior the hyomaxillaris. Like Holmqvist, whose works he evidently had not seen, he says that the intermandibularis (geniohyoideus inferior) is derived from the mandibular myotome, and although its innervation in Polypterus is not particularly given, he doubtless considered it to be by the nervus trigeminus, for he says that this muscle is usually, in the numerous vertebrates considered by him, inner- vated by that nerve. The hyomaxillaris is said to be the homo- logue of the geniohyoideus superior of my descriptions of Amia, to have been developed from the hyal myotome, and to have been primarily innervated by the nervus facialis; and, in a foot- note (1.c., p. 210), it is said that the term geniohyoideus is avoided because it “‘is generally used to denote the anterior element of the hypobranchial spinal muscles.’ It is said that, in the adults INNERVATION OF MUSCLES OF BONY FISHES 297 of the many vertebrates considered by him, the innervation of the intermandibularis and the hyomaxillaris varies considerably, the intermandibularis sometimes being wholly or in part inner- vated by the facialis, and the hyomaxillaris sometimes in part by the trigeminus. Luther (713) also refers to these muscles of Polypterus, and he adopts the term intermandibularis for the intermaxillaris an- terior of Pollard’s descriptions, but calls the intermaxillaris pos- terior the musculus C.vh. In a figure of the adult Calamoich- thys he shows the former muscle innervated by a branch of the trigeminus and the latter muscle by a branch of the facialis, this thus doubtless being the innervation that he ascribes to the muscles of Polypterus. In an adult specimen of Polypterus bichir I find these muscles as shown in the accompanying figure, and it seems to me best, for the present, to still retain for them the time-honored names given by Vetter to the corresponding muscles in other fishes, | notwithstanding that the term geniohyoideus is evidently inap- propriate. The geniohyoideus inferior (intermaxillaris anterior, intermandibularis) is as described by Pollard and Holmavist, but it cannot be said to be greatly reduced (sehr reducirt. Holm- qvist). The geniohyoideus superior (intermaxillaris posterior protractor hyoidei) is not continuous, posteriorly, with the hyo- hyoideus inferior, as both Pollard and Holmqvist say that it is, but the mesial edge of the posterior portion of the geniohyoideus is contiguous with the lateral edge of the hyohyoideus, the two muscles there forming a continuous sheet, so that it is difficult to tell exactly where one ends and the other begins. Anteriorly the two muscles are wholly distinct, the geniohyoideus superior lying superficial (ventral) to the hyohyoideus inferior and separated from it by a fold of the dermal tissues. The geniohyoideus su- perior arises from the proximal (posterior) end of the ceratohyal, and its fibers run anteromesially and diverge somewhat. The anterolateral fibers pass internal (dorsal) to the geniohyoideus inferior and are inserted on a dorsal extension of the median raphe of the latter muscle, the posteriomesial fibers being inserted on a posterior continuation of the same raphe. The anterior 298 EDWARD PHELPS ALLIS, JR. portion of the mantle muscle of Pollard’s descriptions is the hyo- hyoideus inferior. The so-called ramus mylohyoideus trigemini of Holmqvist’s descriptions issues from the ramus of the mandible onto the ven- tral surface of the geniohyoideus inferior, and there separates into anterior and posterior portions, both of which send branches into the muscle to innervate it. The posterior branch then contin- ues onward beyond the hind edge of the geniohyoideus inferior, onto the ventral surface of the geniohyoideus superior, and sends branches into the latter muscle, the terminal branch of the nerve either lying close to, and parallel to, a branch of the ramus hyoi- deus facialis, or anastomosing completely with that branch so that it appears to run directly into it; these two conditions being found, one on either side of the head, in the single specimen used for the accompanying figure. The dissections of this adult specimen thus show that the greater part, at least, of the geniohyoideus superior must be in- nervated by the nervus trigeminus, and that if any part of it is innervated by the nervus facialis it is only a few posteromesial fibers. In a 75-mm. specimen of Polypterus senagalus, examined in serial transverse sections, the two branches here under considera- tion of the nervi trigeminus and facialis do not anastomose with each other on either side of the head, and the branch of the tri- geminus sends branches into both divisions of the geniohyoideus, unquestionably innervating them. The main branch of the hyoi- deus facialis runs anteriorly between the geniohyoideus superior and the hyohyoideus inferior, and goes wholly to the latter mus- cle and to tissues of the region. A small branch perforates the posteromesial edge of the geniohyoideus superior, and, passing through it, without sending any perceptible branches to it, goes to tissues that lie internal to it, this branch evidently being that branch of the hyoideus facialis of the adult that runs anteriorly onto the ventral surface of the geniohyoideus superior and there, on one side of the head of the adult specimen above described, anastomoses with the terminal branch of the nervus trigeminus. INNERVATION OF MUSCLES OF BONY FISHES 299 The conditions in this embryo thus show, even more positively than those in the adult, that the two divisions of the geniohyoideus are both innervated by the nervus trigeminus, and by that nerve alone. The posterior division of the muscle could not then have been derived from the hyal myotome, as Holmqvist and Edge- worth both maintain, unless it had, in stages younger than that represented in my 75-mm. specimen, lost its primitive innerva- tion by the nervus facialis and secondarily acquired innervation by the trigeminus. This I greatly doubted, but it evidently would be confirmed if, as is frequently stated, these muscles are, in certain of the Teleostei, in part innervated by the nervus fa- cialis. I accordingly had certain of these latter fishes examined, and was surprised to find that in certain of them the posterior portion of the so-called geniohyoideus superior is, in fact, inner- vated by the facialis. This definitely established, the muscles themselves were carefully examined and compared, and it was — found that, in every case of such innervation, the fibers so in- nervated belonged to the hyohyoideus and not to the geniohyoi- deus. ‘This was first recognized when considering the muscles in the Siluridae, and it will be best to first consider the condi- tions in those fishes. In both Ameiurus and Silurus there are two muscles which MeMurrich (’84), Juge (99), Herrick (’01), and Jacquet (’01) all describe or refer to as the intermandibularis and geniohyoideus. Holmqvist (11) calls these two muscles the intermandibulares I and II, and hence considers them to be the homologues of the muscles similarly designated by him in Amia, which are the mus- cles called by me (Allis, ’97) the intermandibularis and genio- hyoideus inferior. Both of these muscles are said by all these authors to be innervated by the nervus trigeminus. The musculus hyohyoideus of these fishes is said by both Mc- Murrich and Juge to have two distinctly different portions, and MeMurrich calls them its anterior and posterior portions, and Juge its inferior and superior portions. Holmaqvist calls the an- terior one of these two portions of the muscle the protractor hyoidei, says that it contains no part of the hyohyoideus of other S00 EDWARD PHELPS ALLIS, JR. fishes, and homologizes it definitely with the geniohyoideus su- perior of current descriptions of those other fishes. The so-called posterior or superior, portion of this muscle lies internal to the branchiostegal rays, but is continued forward beyond the rays, the muscle thus corresponding topographically to both the su- perior and inferior divisions of the hyohyoideus of Amia. The so-called anterior, or inferior, portion of the muscle is a thick fleshy muscle which, as shown in the figures given by authors, lies external to, and directly upon, the basal portions of the six or seven anterior (ventral) branchiostegal rays, and external, also, to that part of the hyohyoideus that corresponds topographically to the hyohyoideus inferior of other fishes. Both portions of the muscle are said by all these authors to be innervated by the nervus facialis. In a 30-mm. specimen of Ameiurus nebulosus, examined in serial sections, the branches of the trigeminus and facialis here under consideration anastomose with each other as they do on one side of the head of my adult specimen of Polypterus. In six other specimens, varying in length from 11 mm. to 55 mm., there was no anastomosis of these nerves, and all branches of the tri- geminus and facialis nerves that penetrated the muscles were distributed, respectively, to the geniohyoideus and hyohyoideus, and to those muscles only. There is thus here no possible ques- tion of a double innervation of either of these muscles, this con- firming the statements made by earlier authors. That portion of the hyohyoideus that lies external to the branchiostegal rays is therefore a hyal muscle, and, because of its relations to those rays, it will hereafter be referred to as the hyohyoideus super- ficialis. The two other portions of the muscle will be called, as they are in other fishes, the hyohyoideus inferior and hyohyoi- deus superior. In the one adult specimen of Silurus that was examined, the anterior (ventral) branchiostegal ray on one side of the head was bent outward at its base and there perforated the hyohyoideus superficialis close to its external surface, thus practically lying external to that muscle and hence not coming into any relations whatever with the ceratohyal. INNERVATION OF MUSCLES OF BONY FISHES 301 The innervation of these muscles of Ameiurus and Silurus thus showing that the hyohyoideus superficialis must be a muscle de- rived from the hyal myotome, there is evidently question as to how it could have acquired a position external to the branchio- stegal rays, for those rays, being dermal structures similar in origin and character to the opercular bones, must have lain, primarily, external (morphologically anterior) to all muscles de- rived from the hyal myotome, and external also to all motor fibers of the nervus facialis. Those sensory branches of the latter nerve that were distributed to the dermal tissues on the anterior sur- face of the hyal arch would naturally, when the branchiostegal rays developed, have been left running outward between adja- cent rays, but no motor fibers could have acquired such a course unless they had been dragged outward, out of their normal course, by migrating muscle fibers. The hyohyoideus superfi- cialis of these fishes must then have acquired its actual position, © external to the branchiostegal rays, either by passing outward be- tween certain of those rays or by growing upward, external to the rays, from that part of the primitive muscle that lay anterior to the anterior ray. That the primitive hyohyoideus had sepa- rated into superficial and deeper portions before the branchioster- gal rays were developed, and that the rays had then grown in- ward between those two portions and so acquired insertion on the ceratohyal, internal to the superficial portion of the muscle, seems in itself, a most improbable assumption, and, furthermore, it is not in accord with the fact above referred to that the anterior ray of my specimen of Silurus lies practically superficial to the superficial muscle. If, then, the facile assumption of a secondary innervation of the hyohyoideus superfacialis be excluded from consideration, the re- lations to the branchiostegal rays of the nerve or nerves that in- nervate this muscle of the adult should definitely show from what part of the primitive constrictor of the hyal arch it has been de- veloped. In all my specimens of Ameiurus the superficialis mus- cle is innervated by a single branch of the ramus hyoideus fa- cialis, and in the one adult specimen of Silurus this branch ran outward between the seventh and eighth branchiostegal rays, 302 EDWARD PHELPS ALLIS, JR. counting upward from the ventral end of the arch. Certain fib- ers of the primitive constrictor must then have passed outward between these two rays, pulling with them, the nerve fibers that innervated them, and then have there developed into the large muscle actually found. If several sections of the constrictor had thus passed outward between several pairs of adjacent rays, there would have been a corresponding number of nerves innervating the superficialis muscle, and no such nerves are found. And if the superficialis muscle had grown upward from that part of the entire constrictor that lay anterior (ventral) to the most anterior (ventral) branchiostegal ray, the nerve that innervates the su- perficialis muscle would have first run forward (ventraily) inter- nal to all the branchiostegal rays and then upward external to the ventral ray, and no such nerve is found. It is thus quite certain that but a single section of the primitive constrictor of the hyal arch passed outward between the branchi- ostegal rays to form the hyohyoideus superficialis, and the point where it passed outward lies approximately between certain dorsal ones of the branchiostegal rays that have their attachments to the external (anterior) surface of the ceratohyal, and ventral ones that have their attachments to its internal (posterior) sur- face. This, then, at once suggested that whatever it may have” been that gave rise to this arrangement of the rays had permitted, or perhaps induced, the differentiation of the hyohyoideus super- ficialis, and as this arrangement of the rays is not peculiar to the Siluridae, I at once reexamined my material to see if there were in any of the several specimens of the Holostei and Teleostei any indications of the differentiation of this muscle, and hence an explanation of the fact, already established, that the so-called geniohyoideus superior is in certain of these fishes in part inner- vated by the nervus trigeminus and in part by the nervus facialis In Amia the branchiostegal rays are all attached to the external (anterior) surface of the ceratohyal (Allis, 97), and there is no musculus hyohyoideus superficialis. In Lepidosteus there is also no hyohyoideus superficialis, but the conditions in this fish differ from those in Amia in that, in the single specimen examined, there were but three branchiostegal rays, and they apparently cor- responded to the dorsal ones found in Amia and the Telostei. INNERVATION OF MUSCLES OF BONY FISHES 303 In 37-mm. specimens of Cottus aspera and Clinocottus analis the branchiostegal rays are in part attached to the external surface of the ceratohyal and in part to the internal surface of that ele- ment, as they are in the Siluridae, but there is no musculus hyo- hyoideus superficialis. A branch of the ramus hyoideus facialis runs outward approximately between the rays that are attached to the external and internal surfaces of the ceratohyal, but it is not connected with the nervus trigeminus by anastomosis. The musculi intermandibularis and geniohyoideus are both innervated by the nerves trigeminus, and by that nerve alone. An artery, which arises from the so-called mandibular artery in the ramus of the mandible, accompanies this branch of the trigeminus and then the related branch of the ramus hyoideus facialis, and falls into an artery that runs upward in the hyal arch with the ramus hyoideus facialis, this artery being found in all of the Teleostei that were examined in serial sections. What the significance of — this artery may be could not be determined, but it suggests the commissural vessels that, in the Selachii, connect the anterior and posterior efferent arteries of the branchial arches. In Porichthys notatus there is no musculus hyohyoideus super- ficialis. In an 18-mm. specimen of this fish the two branches of the trigeminus and facialis here under consideration are not con- nected by anastomosis. In a 25-mm. specimen they are con- nected by a delicate anastomosing branch, but this branch cer- tainly contains no motor fibers. In an adult specimen of Exos lucius there is a small hyohyoideus superficialis, but its innervation could not be determined in the dissections. The branches of the trigeminus and facialis here under consideration anastomose with each other, as they do in Amia and on one side of the head of my adult specimen of Polyp- terus; but when the so-formed continuous nerve was treated with weak nitric acid and examined under the microscope, it was seen that there was no interchange of fibers between its two components and that it was the trigeminus nerve, alone, that sent branches into the geniohyoideus. In Scomber scomber there is a fairly large hyohyoideus super- ficialis. In an earlier work (Allis, ’03) I said that the genihyoi- 304 EDWARD PHELPS ALLIS, JR. deus superior of this fish arises by two distinctly different heads, one of which is entirely muscular and the other entirely tendi- nous. ‘The muscular head was said to become tendinous slightly anterior to its surface of origin on the ceratohyal, and to there form a broad flat tendon from which the anterior and wholly muscular part of the muscle arose. I now find that this so-called tendinous part of this head of the muscle is a membranous fascia which lies external to the posterior (dorsal) portion of the muscle and internal to the anterior (ventral) portion, thus completely separating these two so-called portions of the muscle, the one from the other. In a 65-mm. specimen I find these two portions of the muscle innervated on one side of the head by anastomosing branches of the trigeminus and facialis that are strictly similar to those described by me in the adult. On the other side of the head there is no anastomosis between these two nerves, the branch of the facialis ending in the posterior portion of the muscle and the branch of the trigeminus ending in the fascia that separates that portion of the muscle from the anterior portion. ‘The pos- terior portion of the muscle is thus a hyohyoideus superficialis, innervated wholly by the facialis, and the anterior portion a geniohyoideus, innervated wholly by the trigeminus. In a 76-mm. specimen of Caranx caranx the conditions are similar to those in Scomber, but the hyohyoideus superficialis is but slightly developed. In an adult specimen of Pelamys (Scomber) sarda the condi- tions are as in Scomber scomber, but the hyohyoideus superfi- cialis has here pushed forward until it nearly reaches the level of the hind end of the geniohyoideus superior, but is there separated from that muscle by the covering fascia. In an adult specimen of Gadus morrhua, I find the protractor hyoidei of Holmqvist’s (’11) descriptions crossed by two aponeu- roses, one lying not far from the surface of origin of the muscle on the ceratohyal and the other slightly anterior to the point where the muscles on opposite sides meet in the median line. The anterior aponeurosis is described by Holmqvist in Gadus callar- ius, but the posterior one is not mentioned by him. The anterior aponeurosis apparently separates the inferior and superior divi- INNERVATION OF MUSCLES OF BONY FISHES 305 sions of the geniohyoideus, and these two parts of the muscle are both innervated entirely by the trigeminus. The posterior aponeurosis quite unquestionably corresponds to the membran- ous fascia above described in Scomber, and it separates the geniohyoideus superior from a hyoideus superficialis. The tri- geminus and facialis nerves break up into small branches as they approach this aponeurosis, and the branches perforate the apo- neurosis and anastomose with each other, but the size of the fibers, the general conditions, and, more particularly, comparison with the fishes above described, all indicate that the anastomosing branches are not motor ones. This is thus in accord with Holm- qvist’s statement that the posterior portion of his protractor hyoi- dei, which is my hyohyoideus superficialis, is innervated by the facialis and that the anterior portion of that muscle, my geniohyoi- deus superior, is innervated by the trigeminus. Herrick (’00) says that the intermandibularis and geniohyoideus of his fish are both innervated by the trigeminus and the hyohyoideus by the facialis; and the hyohyoideus as referred to by him certainly does not include my hyohyoideus superficialis. The conditions in these several fishes thus show, in my opinion conclusively, that there has been no change whatever in the in- nervation of any of these muscles, the muscles derived from the mandibular myotome all being innervated by the trigeminus, and those derived from the hyal myotome all innervated by the facialis. The disposition of the muscles innervated by the tri- geminus then shows that the ventral portion of the primitive constrictor of the mandibular arch underwent, in all these fishes, a more or less complete longitudinal cleavage, from its dorsal end downward, the dorsal end of one of these two parts acquiring in- sertion on the mandible and the dorsal end of the other part in- sertion on the ceratohyal. That part of the muscle that acquired insertion on the ceratohyal would then naturally there lie exter- nal (morphologically anterior) to the branchiostegal rays of the hyal arch, and that is, as is well known, its actual relation to those rays. The ventral end of this entire constrictor muscle must have been primarily attached either to the ventral end of the branchial bar of the mandibular arch or, and more probably, 306 EDWARD PHELPS ALLIS, JR. to its fellow of the opposite side, in the median line, the fibers of the two muscles there either interdigitating with each other or being inserted in a common median aponeurosis. That part of the muscle that was inserted at its dorsal end on the mandible then frequently separated into two parts, the intermandibularis and geniohyoideus inferior, and the insertions on the mandible of both of these parts tended to shift forward (morphologically ventrally) toward the symphysis. The geniohyoideus inferior thus apparently acquired, in certain fishes, an insertion so close to the symphysis that its course was actually reversed, and it then ran from the ventral end of the arch toward its dorsal end, or, as actually found, from in front directly posteriorly. This muscle then formed with the geniohyoideus superior a single con- tinuous muscle, the median portion of which was the primarily ventral end of the entire muscle. Frequently, however, certain fibers of the geniohyoideus superior did not have this insertion with the fibers of the geniohyoideus inferior, but passed onward internal to the latter muscle and acquired independent insertion on the mandible. Reductions or abortions of one or the other of these three muscles, or of certain parts of them, and a slight shift- ing of origins or insertions then gave rise to all the various ar- rangements actually found. The intermandibularis and geniohyoideus of these fishes would seem to correspond, respectively, to the mylohyoideus and the anterior digastricus of mammals, and some part of the hyohyoi- deus to the posterior digastricus; and this, if correct, should be taken into account in any change of nomenclature that may be proposed. Palais de Carnolés, Menton, France January 6, 1919 INNERVATION OF MUSCLES OF BONY FISHES 307 LITERATURE CITED Auuis, E. P., Jk. 1897 The cranial muscles and cranial and first spinal nerves in Amia calva. Jour. Morph., vol. 12. 1903. The skull and the cranial and first spinal muscles and nerves in Scomber scomber. Jour. Morph., vol. 18. EpcewortH, F.H. 1911 On the morphology of the cranial muscles in some ver- tebrates. Quart. Journ. Microsc. Sci., vol. 56. Herrick, C. J. 1900 A contribution upon the cranial nerves of the cod fish. Jour. Comp. Neur., vol. 10. 1901. The cranial nerves and cutaneous sense organs of the North Americal siluroid fishes. Jour. Comp. Neur., vol. 11. Hotmevist, O. 1910 Der Musculus Protractor Hyoidei und der Senkungsmech- anismus des Unterkiefers bei den Knochenfischen. Lunds Universit. Arsskrift. N. F. Afd. 2, Bd. 6, Nr. 6. 1911 Studien in der von den NN. Trigeminus und Facialis innervier- ten Muskulatur der Knochenfische. Lunds Universitets Arsskrift N. F. Afd. 2, Bd. 7, Nr. 7. Jaquet, M. 1901 Recherches sur l’anatomie et l’histologie du Silurus glanis L. Bull. de la Société des Sciences de Bucarest, An. 10, No. 5. Juce, M. 1899 Recherches sur les nerfs Cérébraux et la musculature cépha- ~ lique du Silurus glanis L. Revue suisse de zoologie, T. 6, Fase. 1. Luruer, A. 1913 Uber die vom N. trigeminus versorgte Muskulatur der Ga- noiden und Dipneusten. Acta. Soc. Sc. Fenn., Tome 41, No. 9. McMorricx, J. P. 1884 The myology of Amiurus catus. Proc. Canadian Insti- tute, vol. 2, fase. 3. PoutiarpD, H. B. 1892 On the anatomy and phylogenetic position of Polypterus. Zool. Jahrb. Abtheil. f. Anat. u. Ontog., Bd. 5. VeTrer, B. 1878 Untersuchungen zur vergleichenden Anatomie der Kiemen- und Kiefer-musculatur der Fische. Jenaische Zeitschr., Bd. 12. THE ANATOMICAL RECORD, VOL. 16, No.5 Resumen por el autor, Francis Marsh Baldwin. Colegio del Estado de Iowa. Variaciones de las arterias carétidas del conejo. El presente estudio fué emprendido para averiguar la presencia y extensién de la variacién de las arterias carétidas del conejo, habiéndose empleado 114 ejemplares, veinte y tres de los cuales (cerca del 20 por ciento) difieren de las condiciones normales descritas en los libros de texto. El autor discute las principales variaciones observadas bajo los siguientes epigrafes: Diferencias encontradas en las arterias carétido-occipitales internas, en las arterias maxilo-linguales externas, maxilo-superficiales temporales internas, temporales-occipitales superficiales y otras diferencias generales. Las relaciones de tamano, orden de secuencia y posiciones comparadas de las diversas ramas se anotan también y discuten. En varios casos se forman interesantes arterias in- nominadas, mientras que en otros las ramas terminales forman a modo de tenedores con tres, cuatro o cinco ptias. El trabajo esta ilustrado con una limina y doce figuras. Translation by José F. Nonidez Columbia University AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, JUNE 23 VARIATIONS IN THE CAROTID ARTERIES OF THE RABBIT FRANCIS MARSH BALDWIN Iowa State College, Department of Zoology, Ames, Iowa ONE PLATE (TWELVE FIGURES) That the blood-vessels of any group of mammals in general are subject to great variations is well known. Such minor varia- tions as have been observed within any group have usually been ignored or at most, resolved to conform to the type. Using the rabbit as a basis of study in mammalian anatomy during the past two years, the writer has had an opportunity to make some in- teresting observations on the variations of the carotid arteries. Of one hundred and fourteen specimens dissected in the labora- tory, twenty-three, or about 20 per cent, were found to differ from the usual condition described in the texts. Of these, eleven individuals possessed marked differences, as shown in the accom- panying figures. In the majority of cases, the common carotid artery (fig. 1, C.C.) passes forward from the superior thoracic aperture along the side of the trachea. Its branches include the superior thyre- oid artery supplying the thyreoid gland, and the superior laryn- geal artery. The latter arises at the level of the thyreoid plate (larynx) and passes to the sternohyoid and sternothyreoid mus- cles. A short distance cephalad the common carotid artery gives off a very small internal carotid artery which passes dor- sad, and disappears beneath the auditory bulla. From this point forward the vessel is the external carotid artery, which gives off successively the occipital, the lingual, the external maxillary, the superficial temporal (one of the terminal branches), and the internal maxillary (the other terminal branch) arteries in the order named. 309 310 FRANCIS MARSH BALDWIN The occipital artery passes to the posterior portion of the head from the dorsal wall of the external carotid artery at a point just cephalad to the internal carotid. The lingual artery arises from the ventral wall of the external carotid artery at a point about at the same level as the occipital, and passes forward into the tongue. The external maxillary artery is given off just cephalad of the lingual branch and passes to the medial surface of the ventral border of the mandible. It gives branches to the submaxillary gland and the muscles of mastication. The internal maxillary and the superficial temporal arteries form the two terminal branches of the external carotid artery, the former passes in the direction of the orbit and gives off the inferior alveolar branch to the mandible, the latter passes to the temporal region and gives off the transverse facial artery to the cheek and face. To simplify the presentation of differences found, it is con- venient to use the following captions: The internal carotid-occipital differences. A common difference in the relationships just noted is a condition where the internal carotid artery and the occipital branch arise from the common carotid as a single trunk, an innominate (figs. 2, 3, 7, and 12, IN.), which subsequently divides. Interesting gradations in re- spect to the division are found, from the condition (fig. 8) where the two arteries arise separately from the common carotid artery, and where there is no innominate formed, to that where a long innominate is formed as shown in figure 7. The order of the division is of interest also, since in some eases (figs. 2 and 3) the occipital branch is morphologically the most posterior, in others (figs. 7 and 12) the internal carotid artery occupies such a posi- tion. In the first condition there is no crossing of thetwo, the occipital passes dorsad to the muscles of the head and neck, and the internal carotid artery passes directly mesad under the audi- tory bulla. In the second condition there is a crossing, the occip- ital branch usually passing laterad of the deeper lying internal carotid trunk, although here again there seems to be some vari- ability, since in one instance (fig. 4) the opposite is the case. CAROTID ARTERIES OF THE RABBIT SA The external mazillary-lingual differences. While there has been noted no case where the sequential order varies in which these two arteries are given off from the external carotid artery, the rela- tive differences in distances from one another in their origin is worthy of study. In two individuals (figs. 6 and 7) the interval between the two is very considerable, being nearly a centimeter apart. From this extreme, gradations occur, the distance be- tween their points of origin on the carotid gradually approximat- ing one another, as is shown in figures 3, 2, and 5, respectively. Finally, there is formed in some cases a common trunk, an in- nominate, before the division takes place, as shown in figures 4, 9, and 12. It is apparent that this approximation may take place in either direction; that is, the lingual may move cephalad to effect the junction with the maxillary, as shown in figure 9, or the maxillary may move caudad, as represented in figure 4. In the condition as shown in figure 5, both arteries have been slightly displaced from the usual position of either. In one case (fig. 6) it is interesting to note that the external maxillary is given off as a branch of the internal maxillary artery some dis- tance cephalad of the latter’s junction with the superficial tem- poral. In this case it occupies the relative position of the in- ferior alveolar branch, and might easily have been taken for the latter on Superficial examination. The inferior alveolar branch in this case being somewhat more cephalad than usual. In two cases, however, figures 4 and 5, where the maxillary and lingual branches are closely approximated, the inferior alveolar branch is considerably more caudad than is ordinarily the case. The internal maxillary-superficial temporal differences. In some instances these two vessels differ conspicuously in size, and where this condition is most marked, one may be considered a branch of the other. In conditions shown in figures 2 and 3 the super- ficial temporal is a small side branch passing dorsad, while the larger internal maxillary artery continues forward. In other cases (fig. 9) the opposite is true, the smaller internal maxillary artery is a branch of the larger superficial temporal trunk, and in this case its point of origin from the temporal is well cephalad. 312 FRANCIS MARSH BALDWIN In two cases represented by figure 7 the relationships of these terminal branches of the external carotid artery are of interest, since they together with the external maxillary artery form a three-parted fork; the external maxillary artery turns abruptly ventrad, the superficial temporal passes dorsad, and the internal maxillary bends mesad. In size there is very little difference be- tween the three vessels, any one of which could be considered a terminal branch of the external carotid artery. The superficial temporal-occipital differences. In three cases the occipital artery originates as a branch of the superficial tem- poral. In one individual (fig. 9) it passes dorsad from what may be considered the base of the superficial temporal or its innomi- nate. In figure 5 it is but a little more cephalad, while in figure 10 it is shown passing away from the temporal well cephalad to the latter’s junction with the other arteries. Other differences. In one case shown in figure 11 all the ar- teries pass forward away from the common trunk in such a way as to form a sort of corona radiata. In such a condition the ex- ternal carotid artery is practically eliminated, since the common carotid artery is broken up immediately into five terminal branches. In the condition shown in figure 12, the common carotid artery can be considered as terminating in three innomi- nate trunks; one giving rise to the internal carotid and occipital branches, one forming the external maxillary-lingual branches, and the third, the internal maxillary-superficial temporal branches. In‘the condition shown in figure 10 the external carotid artery is very short, terminating in four branches, one of which is an in- nominate which forms the superficial temporal and occipital branches. In two cases the inferior alveolar artery which normally is con- sidered a branch of the internal maxillary, shows a tendency to branch well down on the external carotid trunk. - This condition is indicated in figures 4 and 5. On the other hand, the external. maxillary artery which normally is a branch of the external caro- tid, in two individuals (figs. 3 and 6), branches well cephalad from the internal maxillary artery. CAROTID ARTERIES OF THE RABBIT 313 SUMMARY Variations in the relative positions and points of origin of the several vessels along the common carotid artery results in the ‘formation of several innominate arteries. Those of especial in- terest are the occipital-internal carotid, the external maxillary- lingual, the internal maxillary-superficial temporal, and the super- ficial temporal-occipital arteries, represented in figures 2, 4, 9, and 10, respectively. After the common carotid artery gives rise to the internal caro- tid or to the innominate (internal carotid-occipital), the remain- ing trunk, the external carotid artery, may terminate in a num- ber of ways; it may end as a single trunk (either the internal maxillary or the superficial temporal); it may be bi-parted (as normally) or by two innominates, as in figures 9 and 12; it may be three-parted formed by the two maxillaries and the temporal, as in figure 7; it may be four-parted, formed by the two maxil- laries, the temporal and the lingual, as in figure 8, or by the lin- gual, the two maxillaries and the innominate, as in figure 10; or it may be five-parted, formed by the two maxillaries, the lingual and the temporal and the occipital. The inferior alveolar artery, which normally is a branch of the internal maxillary, is in two cases (figs. 4 and 5) well down on the external carotid artery. The external maxillary artery in one case ( Gp 6) is given off as a branch of the internal maxillary some distance cephalad to the latter’s junction with the temporal. In several cases both the lingual and the external maxillary arteries pass out of the external carotid artery at the level of the internal maxillary and superficial temporal arteries, as in figures 8, 10, and 11. The occipital artery varies considerably in its origin. It may be a branch from the internal carotid artery, as in figures 2, 3, 7, and 12; it may branch as an independent twig from the external carotid artery as in the normal condition, as shown in figures 4, 6, 8, and 11, or it may be a branch from the temporal, as in figures 5, 9, and 10. In one case, figure 3, the lingual and the external maxillary arteries may be regarded as branches from the internal maxillary, since their points of origin are well cephalad to the point at which the temporal branch is given off. PLATE 1 EXPLANATION OF FIGURES 1 Branches of the common carotid artery (left side) showing usual order and distribution of its various branches. 2 Variation in which the occipital and internal carotid arteries leave the com- mon carotid artery together as an innominate artery. Thesuperficialtemporal . is much smaller than the internal maxillary and can be regarded as a branch of the latter. 3 The innominate of the internal carotid and occipital, and the superficial temporal are close to each other, while the lingual, external maxillary, and inferior alveolar may be regarded as branches from the internal maxillary. 4 The external maxillary and lingual arteries arise as aninnominate from the common carotid. The occipital and the internal carotid arteries are in the re- verse sequence from the condition shown in figures 2 and 3, and the former passes mesad to the latter. The inferior alveolar is a branch from the external carotid. 5 The occipital leaves the superficial temporal close to the latter’s base; the external maxillary and the lingual arteries arise close together from the external carotid, while the inferior alveolar is at the base of the internal maxillary. 6 The external carotid artery in this case terminates in three branches, the lingual, superficial temporal, and internal maxillary, the external maxillary and inferior alveolar being branches of the latter. 7 The external carotid artery terminates in three branches, the external max- illary, the internal maxillary, and the superficial temporal. Note the compara- tively long innominate which divides to form the occipital and internal carotid arteries. 8 The lingual and external maxillary arteries originate close to the junction of the internal maxillary and superficial temporal arteries. 9 The external carotid terminates in two innominate arteries; one giving rise to the lingual and external maxillary arteries, the other forming the internal max- illary and superficial temporal arteries. The occipital branch is small and comes off at a point of junction of the two innominates. ' 10 The occipital artery is a branch of the superficial temporal which leaves the latter well cephalad. 11 All the branches of the common carotid are close to one another, forming a sort of corona radiata as its termination. 12 The common carotid artery breaks in this case into three innominate ar- teries, the external maxillary-lingual, the internal maxillary-superficial temporal, and the internal carotid-occipital arteries. ABBREVIATIONS C.C., common carotid artery IN., innominate artery E.C., external carotid artery L., lingual artery © E.M., external maxillary artery O., occipital artery I.A., inferior alveolar artery S.T., superficial temporal artery JI.C., internal carotid artery T.F., transverse facial artery I.M., tternal maxillary artery 314 PLATE 1 CAROTID ARTERIES OF THE RABBIT FRANCIS MARSH BALDWIN (Ge SMM) = Pry so 315 Resumen por el autor, James Frederick Rogers. Escuela Normal de Gimnasia de New Haven. El pié como palanca. Ha habido mucha discusién entre los anatémicos sobre la naturaleza de la palanca formada por el pié cuando el individuo se pone de puntillas. Algunos han considerado al pié como una palanea de primer grado, cuyo fulero seria la articulacién del tobillo, mientras que otros autores le comparan con una palanca de segundo grado, con el fulcro situado en el talén. La serie de palancas, peso y tensién de muelle ilustrados en el texto parecen probar de un modo indudable que la palanea es de primer grado cuando se usa el pié de este modo. Translation by José F. Nonidez Columbia University AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, JUNE 30 THE LEVERAGE OF THE FOOT JAMES FREDERICK ROGERS New Haven Normal School of Gymnastics, New Haven, Connecticut ONE FIGURE In most works on anatomy and physiology the bones of the foot working about the ankle-joint are, when the person owning the foot lifts himself on the ball, considered as behaving as a lever of the second class. Some authors state that there is a difference of opinion as to whether it becomes a lever of the second or of the first class, while a few assert positively that, no matter how used, this is always a lever of the first class, the ful- - crum being at the ankle-joint. They explain that, when rising ’ on the balls of the feet, we should consider ourselves, mechani- cally, as standing on our heads and pushing the earth from us with a force equal to the weight of the body. The new American . edition of Gray makes this positive statement, but an excellent English work of about the same date of issue considers the lever in the more usual way. Where there is so much antagonism of opinion, it would seem as if no one had taken the trouble to work out the problem ex- perimentally, and we have never seen any suggestion of such an attempt. The device pictured herewith seems to solve the prob- lem. Two pieces of lath, d and e, which represent the bones of the leg and foot, respectively, are hinged at a. The lower end of e (ball of foot) rests on a table directly or can be hinged in a block, c. A spring balance is attached in position to represent the calf muscles. A known weight, which represents the weight of the thigh and body above, is balanced on the upper end of d. If a weight of 5 pounds is used, a spring balance with a capacity of 25 pounds is most suitable. When the weight is in place and the machine held as in the drawing, the spring balance represent- 317 318 LEVERAGE OF THE FOOT ing the pull of the calf muscles is read. This is always much greater than the amount of the weight and (with friction reduced to a minimum) in proportion to the relative length of the power and weight arms of a lever of the first class, the fulerum being at a. Were this behaving as a lever of the second class, the power arm would, of course, be longer than the weight arm, and the power needed to extend the foot would be less than the weight resting on the foot. Instead of using a fixed weight, c can be placed on a spring scale and pressure made downward upon the upper end of d. The reading of the scale will represent the weight of the body and that of the balance the pull of the muscles. aad } j — * ‘ : . =. 37% ; oy ' Pitues ff 44 a b . 4 mea ofits Ehe d iatinjsid.« yaiill z entaziert des ; oO vi j ‘ ri bent aire: PeeonriyTad'in) (ferae i p a. 7 dint: Z ; d 44 ae 4 si do _s { : - > : - 7 See ie anal “3 a! mat ods #) vw 2 | Aes " . 1), a 2. ee 7 id - - - “4 , Z ‘ ss = - * ee} : - ' j e - = , a 7s a bad * , . “~ , _ y ¢ Resumen por el autor, Robert Wesley Henderson. El sistema linfatico del adulto de la ardilla terrestre rayada (Sper- mophilus tridecemlineatus Mitchell. El presente estudio se llev6 a cabo como preliminar de un es- tudio sobre el desarrollo del sistema linfaitico del embri6n, y la presente descripcién es lo mas préxima posible a la de la forma tipo de sistema linfico en este animal, incluyéndose en ella la disposicion anatomica, posicién y relacién de los vasos valvulas, nédulos y sacos linfdticos, dreas de drenaje y orificios veno-lin- faticos. Los puntos especiales de mas interés son las deserip- ciones de los orificios linfaticos renales y post-cavos, la de un saco-linfatico yugulo-subclavio en el adulto y la de una formula para la inyeccién de una masa especial ideada para vencer los obstdculos inesperados encontrados en este animal, siendo la mayor dificultaad para conseguir buenos preparados el extra- ordinario poder decolorante de los fluidos de los tejidos sobre el azul de Prusia. La nomenclatura empleada se aproxima lo mas posible a la usada en la anatomia humana. La relacién y vari- acion de estructura estdn ilustradas por seis figuras. Translation by José F. Nonidez Columbia University AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, JUNE 30 THE ADULT LYMPHATIC SYSTEM OF THE STRIPED . GROUND-SQUIRREL (SPERMOPHILUS TRI- DECEMILINEATUS MITCHILL) ROBERT W. HENDERSON Laboratories of Animal Biology, State University of Iowa SIX FIGURES The original purpose of this study of the gross adult anatomy of the lymphatic system of the striped ground-squirrel (Spermo- philus tridecemlineatus) was that it should be merely a prelimi- nary to a study of the development of the lymphatic system in the embryo. However, so much difficulty was encountered in the securing of an injection mass which would give good differentia- tion and yet remain permanent that the work practically resolved itself into a search for an injection mass which would fulfill these requirements in this species of animal.* When this mass was finally made, the work of dissecting out and describing the adult lymphatic system was a very simple matter. No work has as yet been done on the embryology. Forty-eight specimens were successfully injected in some re- gion. So this description is as near as possible that of the type form of lymphatic system in this animal, including the more unusual variations. India-ink gelatin mass was used as the injecting media. For- mula for making: 1. Finely powder stick India ink in a mortar. 2. Add 4 to 5 cc. saturate solution NH,OH to the powder in the mortar. Stir with pestle till thick and syrupy. 3. Allow to stand till coarse particles of ink settle. 4. Decant. *Berlin blue gelatin mass exactly filled all the requirments of an injection mass except that the tissue fluid of this animal would completely decolorize it in a few minutes. 319 320 ROBERT W. HENDERSON 5. Soften plate gelatin in distilled water and liquefy over warm water-bath, using no more water than absolutely necessary. 6. Thoroughly stir India-ink-ammonia solution into gelatin. 7. Strain through several layers of cheese-cloth and allow to stand in warm place for several days. 8. As a preservant add a small crystal each of thymol and potassium iodide. 9. Keep in tightly stoppered bottle. The injecting apparatus consisted of a 5-cc. glass syringe and a steel hypodermic needle. Ether, chloroform, and illuminating gas were the killing agents. The best results, however, were obtained from those specimens killed with illuminating gas. Injections were made in most specimens while the animal was still warm and as soon as possible after death, though good in- jections were secured from nodes after the animal had been sev- eral days dead and preserved in a 5 per cent formalin solution. The best cutaneous injections were obtained in females heavy with milk. Cutaneous injections were made in the soles of the feet, between the toes, in the heels, in the back, at the base of the tail, in the lips of the rectum, vulva and mouth, in the sides and abdomen, at the bases of the ears, and the tip of the nose. Injections were also made in the tip of the tongue, walls of the intestines, the stomach, the testes, in the spleen, the anterior inguinal nodes, the intestinal nodes, the cervical and axillary nodes, and one in- jection was made in the right thoracic node. In making cutaneous injections the best results were obtained from superficial subcutaneous injections, since the mass spread out from the point of injection as a fine anastomosing network which finally formed one or more large capillaries. Injections in the superficial fascia merely formed large lakes, as a rule, except in the feet and the head where the lymph systems were nearly always injected. The intestinal lymphatics were easily demonstrated by starv- ing the animal for a day or so and then feeding it suet. If the animal was killed two or three hours after such feeding the intes- LYMPHATIC SYSTEM—STRIPED GROUND-SQUIRREL 321 tinal lymphatics would be completely demonstrated to the mi- nutest detail by the presence of chyle in the intestinal walls and the mesenteries. The intestinal lymphatics may also be dem- onstrated by injection if care is taken not to break the mucosa. Not a single successful injection of the spleen and the walls of the stomach was made, though it was tried repeatedly. Injection of the lymph nodes was nearly always successful if the point of the needle was completely included in the node. LYMPHATIC SYSTEM The lymphatic system of the common ground-squirel, exclu- sive of the digestive tract, is essentially bilaterally symmetrical and consists of three elements, lymph vessels, lymph nodes, and lymph sacs. The lymph vessels are transparent, thin-walled tubes of very small caliber and lie in the connective tissue surrounding blood- vessels, in the skin, and in some muscles. The thoracic duct, which is the largest vessel in the lymphatic system, when dis- tended with an injection mass has an average diameter of less than1 mm. Many other lymph vessels even when filled with a colored injection mass can be detected only with the aid of a microscope. These vessels transfer the lymph either from tissues to nodes, from node to node, from nodes to sacs, or from nodes or sacs to empty it into blood vessels. Their arrangement will be described in detail, with the description of the nodes, as afferent vessels, those carrying lymph to the nodes, and efferent vessels, those carrying lymph away from the nodes. Some vessels, since they carry lymph from one node or group of nodes to another, will be described under both heads. The lymph nodes will be denominated according to the loca- tion in, or near, which they lie and as nearly as possible in ac- cordance with the nomenclature used in human anatomy. Not all nodes, however, will be given individual names, as certain nodes vary greatly in their frequency of appearance and location, so, for the sake of simplicity, they will be treated in groups when- ever convenient. According to location, the nodes may be divided into the cervical, axillary, thoracic, appendicular, inguinal, lumbar, cisternal, and intestinal groups. THE ANATOMICAL RECORD, VOL. 16, No. 5 322 ROBERT W. HENDERSON CERVICAL NODES AND VESSELS This group, which will be treated as superficial cervical and deep cervical nodes, is difficult to identify from the other glands of the neck unless it is injected. The superficial cervical nodes lie along the course of the exter- nal jugular vein and its branches, and are usually four in number, the anterior parotid node, the posterior parotid node, the isi ficial thyroid node, and the submaxillary node. The anterior parotid node (fig. 1, Ant.Par.N.) lies on the an- terior surface of the parotid gland at the junction of the anterior auricular and posterior facial veins. Its afferents come from the anterior portion of the ear and the dorsal part of the head, and its efferents pass posteriorly along the posterior facial vein to the posterior parotid node, or to the level of the internal carotid artery along which they may pass directly mesial to the common carotid and then posteriorly to the deep cervical nodes. Size: approximately, 0.5 x 0.5 x 0.5 mm. The posterior parotid node (fig. 1, P.Par.N.) is located on the posterior surface of the parotid said near the junction of the posterior auricular and external jugular veins. Afferent vessels come from the posterior portion of the ear and the dorsal side of the head and neck. The efferent vessels may pass ventrally and posteriorly along the external jugular vein to the superficial thy- roid node or may go forward along the posterior facial vein to pass directly to the deep cervical nodes along with those from the anterior parotid node. Size: 3x 2x 1mm. The superficial thyroid node (fig. 1, Sup.Thy.N.) is located un- der the platysma on the lateral surface of, and immediately in contact with, the thyroid gland in the angle formed by the junc- tion of the anterior facial vein and the external jugular vein. It receives afferent vessels from the skin of the chest, which pass an- teriorly over the external jugular vein, and usually a vessel from the posterior and anterior parotid lymph nodes. Size: approxi- mately, 2.6 x 1.5 x 1 mm. The submaxillary nodes (fig. 1, Sub.Maz,N.) usually two in number, are located at the anterior margin of the submaxillary LYMPHATIC SYSTEM—STRIPED GROUND-SQUIRREL 323 gland on either side of the anterior facial vein, at the junction of the internal and external maxillary veins. They are just ventral to the tendon of the digastric muscle, lateral to the sternothy- roid muscle and covered by the platysma and skin. They re- ceive afferent vessels from the lower lip which parallel the exter- nal maxillary vein or pass over the surface of the anterior belly of the digastric muscle; from the upper lip which parallel the an- gular vein; from the tongue which parallel the internal maxillary vein or pass directly through the musculature of the tongue, and from the lateral thyroid node along the anterior facial vein. They send efferent vessels to the lymphatic plexus surrounding the larynx. This plexus connects with the corresponding nodes of the opposite side. They also send vessels which pass over the internal maxillary vein mesially to a point near the external carotid artery and then follow posteriorly along the external caro- tid and common carotid arteries to the deep cervical nodes. - Size: approximately, 2.5 x 2.5 x 0.5 mm. The deep cervical nodes (fig. 1, (D.Cer.N.) are usually two in number and are located in the angle between the sternocleido- mastoid and sternohyoid muscles, ventral to the common caro- tid artery and the internal jugular vein, and lateral to the trachea. One of these nodes is usually considerably larger than the other. They receive afferent vessels directly from the tongue (fig. 1, T.), from the lymphatic plexus over the larynx, from the submaxillary lymph nodes, and sometimes from the anterior and posterior parotid nodes. In fact, all the lymph from the super- ficial cervical nodes and other lymphatics of the head and neck eventually passes through these nodes. The efferent vessels, usually two in number, pass posteriorly, the larger along the common carotid artery and the other along the internal jugular vein. On the right side, very close to the junction of the internal and external jugular veins, these lymph vessels unite and enter the veins near the jugulosubclavian Junc- tions. Size: 4x3x2mm.and2x1x0.5 mm. 324 ROBERT W. HENDERSON T Saag Lab.V. \ SS \ \ C.Lab.V. P.F.V. ——inf.0.V. _P.Oro.Y. Ex.Max.V. A.AU.V. Per.V. S.T.V. D.T.V. ee D.Cer.N Jug-jug.-T- fs P.Au.V. Int.Jug.V. -Jug.V. wae ceph.V. J.S.T. L.Sub.V. Cir.V. Ax.V. “oro Pewee one ® L.Thor.vV. sup -Th.N L.Pre.V. A-Ing-N- -L.V.or L.I-1.V. In.Fud.V Pop.br. — Fig. 1 Venous and lymphatic systems (diagrammatic) LYMPHATIC SYSTEM—STRIPED GROUND-SQUIRREL 325 ABBREVIATIONS ao., aorta Ang.V., angular vein A.Au.V., anterior auricular vein Ant.Fac.V., anterior facial vein Ant.Par.N., anterior parotid node A.Ing.N., anterior inguinal nodes Az.N., axillary node Az.V., axillary vein In.A., innominate artery B.W., body wall B.I.-L.V., branch of iliolumbar vein Cis.C., cisterna chyli Ceph.V., cephalic vein Cir.V., circumflex vein Caud.V., caudal vein Com.Lab.V., common labial vein C.C.A., common carotid artery C.A., celiac axis D.T.V., deep temporal vein D.Thor.V., dorsal thoracic vein D.Cerv.N., deep cervical nodes Ex.Jug.V., external jugular vein. E.C.A., external carotid artery E.Maz.A., external maxillary artery Epi.V., epigastric vein E.Max.V., external maxillary vein Fem.V., femoral vein GI.V., gluteal vein Hep.V., hepatic vein Int.V., intercostal vein Inf.O.V., inferior orbital vein Hyp.V., hypogastric vein In.Pud.V., internal pudic vein Int.Jug.V., internal jugular vein I.C.A., internal carotid artery Inf.Lab.V., inferior labial vein In.Maz.A., internal maxillary artery In.Pud.V., internal pudic vein In.Maz.V., internal maxillary vein Inf.Mes.V., inferior mesenteric vein Jug-jug.T., jugulojugular tap Kid., kidney L.Sub.V., left subclavian vein Lum.N., lumbar node L.R.V., left renal vein L.I.V., left iliac vein L.I-1.V., left ilio-lumbar vein L.E.Jug.V., left external jugular vein L.P.C.V., left precava L.R.T., left renal tap L.Thor.V., long thoraic vein L.Jug.-sub.T., left jugulo-subclavian tap Lar., larynx P.V., portal vein P.C.V., postecava P.S., portal system P.Orb.V., postorbital vein Pop.V., popliteal vein Pop.B., popliteal branch P.Au.V., posterior auricular vein. Par.V., parotid vein P.Co.A., posterior colic artery P.Par.N., posterior parotid node Post.T., postcaval tap P.Ing.N., posterior inguinal node P.F.V., posterior facial vein Ph.V., phrenic vein R.P.C.V., right precava R.Jug.-sub.T., right jugulosubclavian tap R.Sub.V., right subclavian vein St.V., sternal vein Sc.V., sciatic vein Sp.V., spermatic vein Sub.Maz.N., submaxillary nodes Subm.V., submaxillary vein Sup.Thy.N., superficial thyroid node S.T.V., superficial temporal vein Sup.Br.N., superficial brachial node Sup.Th.N., superficial thoracic node Sup.Lab.V., superior labial vein Th.D., thoracic duct Th.N., thoracic node 326 ROBERT W. HENDERSON AXILLARY NODES AND VESSELS In the axilla there is usually found but a single large axillary node (fig. 1, Ax.N.), though occasionally a small node is also pres- ent. These nodes lie close to the long thoracic vein immediately posterior to its junction with the axillary vessels. It is embedded in the anterior lateral side of a large gland. It receives afferent vessels from the skin of the thorax which pass to it anteriorly along the branches of the long thoracic vein; from a small node, the superficial thoracic node (fig. 1, Suwp.Th.N.), which lies just behind the shoulder close to a branch of the long thoracic (this node either is not always present or else cannot always be found) ; from the front leg and the sole of the foot, which vessels parallel the brachial artery and its branches; from the skin over the side of the front leg, which vessels pass obliquely across the muscula- ture at about the middle of the humerus to terminate at the an- terior end of the axillary gland; from the superficial brachial node, which vessels pass through the musculature of the leg with the circumflex vein. The efferent vessels extend from the anterior end of the axil- lary node along the subclavian vein and artery to empty, on the right side, into the veins between the external jugular and sub-. clavian veins at their junction. On the left side it unites with the thoracic duct or the jugular lymph sac if present. Size: approximately, 7 x 2x 1mm. THORACIC NODES AND VESSELS The thoracic nodes (fig. 1, T7h.N.), are two in number, one on each side, and lie in the fatty tissue near the walls of blood-ves- sels. The node on the left side is located posterolateral to, and on the wall of, the subclavian artery at its anterior end. Itisa rather large node, measuring 6x 2x 1mm. _ Its afferent vessels have not been determined, but its efferent vessel is short and empties into the thoracic duct or into the efferent from the left axillary node near its termination. The right thoracic node is somewhat smaller and is located more ventrally along the mesial wall of the right precava near its LYMPHATIC SYSTEM—STRIPED GROUND-SQUIRREL 327 anterior end. Afferent vessels come from the thymus, the an- terior walls of the trachea, and the region of the heart and lungs. The distribution of these vessels has not been exactly determined, as only one specimen has yielded to injection and that only par- tially. Theafferent vessels pass anteriorly along the precava and around it dorsally to unite with the efferent vessel from the right axillary node just before it makes connection with the veins. APPENDICULAR NODES AND VESSELS The superficial brachial node (fig. 1, Sup.Br.N.) lies subcu- taneously between the triceps and deltoid muscles posterior to the cephalic vein at its junction with the circumflex vein. It re- ceives afferent vessels from the dorsal side of the foot and foreleg and from the skin, and also vessels from the shoulder which par- allel the upper portion of the cephalic vein. Its efferents lead to the axillary node over the circumflex vein. Size: approximately 2x2x0.5 mm. The popliteal node (fig. 1, Pop.N.) lies back of the knee in the hind leg, in the popliteal space, and is buried in fat. It isa small gland approximately 2x 1.5x0.5mm._ It receives afferent vessels from the sole of the foot and the skin which pass proxi- mally through the skin on the posterior surface of the gastroene- mius to a point opposite the popliteal node. Then they pass be- tween the biceps femoris and semimembranosus muscles in the fat to the lymph gland. It also receives afferent vessels from the mesial and posterior surface of the leg and from the region around the anus. The efferent vessels pass proximally along a branch of the pop- liteal vein and the popliteal vein to the femoral vein where it unites with other lymph vessels. INGUINAL NODES AND VESSELS The inguinal nodes may be divided into anterior inguinal nodes (figs. 1 and 4, A.Ing.N.) and the posterior inguinal nodes (fig. 1, P.Ing.N.). The posterior inguinal nodes are single, small, approximately 2x 1.5x 1 mm., and are very difficult to find unless they are in- 328 ROBERT W. HENDERSON jected as they lie buried in the superficial fascia on the posterior abdominal wall, just lateral to the ventral midline and along a branch of the epigastric artery. Afferent vessels come from the mesial side of the hind led, the region of the tail and anus, and the posterior abdominal wall. Efferent vessels pass along the branch of the epigastric, near which the node lies, to the epigas- tric and iliac veins, where they unite with other lymphatic vessels from the leg. The anterior inguinal nodes may be single or number as many as five. If single, they are large, approximately 7 x 2 x 2 mm., and if more than one they may be assmallas2x1x1mm. The usual number is two or three with one more node on the left side than on the right. These nodes lie along the iliolumbar vein and receive afferents from the skin over the front and lateral side of the hind leg even as far posterior as the tail and anus, and from the skin over the sides and ventral abdominal wall. Their ef- ferents pass ventrally along the iliolumber vein to the abdominal wall where they join lymph vessels from the musculature of the abdominal walls. LUMBAR NODES AND VESSELS The lumbar nodes (fig. 1, Lwm.N.), are arranged around the posterior end of the aorta where the spermatic (ovarian), ilio- lumbar, inferior mesenteric, and iliac arteries are given off. The simple type form of their arrangement is a group of five nodes; one on each side of the aorta just anterior to the iliolumbar arter- ies; one on each side of the aorta just posterior to the iliolumbar arteries and just anterior to the right and left iliac arteries, re- spectively, and one caudal node, which lies between the iliac arteries at the bifurcation of the aorta. It is very seldom, how- ever, that this simple type occurs, it being found only once in all the specimens dissected, and in that case the nodes on the right of the aorta were so closely connected across the ventral side of the right iliolumbar artery and vein that it was doubtful whether they were in the same or in separate capsules. If, how- ever, the nodes lateral to the aorta are completely separate an- teroposteriorly, the caudal node varies and has been found to con- LYMPHATIC SYSTEM—STRIPED GROUND-SQUIRREL 329 sist of as many as two single nodes and a double node. In other specimens the caudal node has been found to be entirely lacking, in which case the other nodes are of larger size. In the majority of specimens the caudal node is present and single, and, on one side or the other, one pair of the lateral lumbar nodes are united with each other anteroposteriorly across the iliolumbar artery and vein on the ventral side to form a single or a double node. The size of these nodes is extremely variable, 14 mm. being the length of the longest node found, and 2.5 mm. the greatest width, while the smallest node was approximately 1x 1x1lmm. These nodes are all connected by a very rich plexus of lymph vessels, so that if an injection from any point fills any of the nodes it will usually fill all. Occasionally, however, injections from the an- terior inguinal nodes would fail to inject the caudal node, which result was never the case if the injection was made from the hind feet, tail, or posterior inguinal nodes. The afferent vessels to these nodes are very numerous since they drain the whole posterior region of the body. Several ves- sels come from the hind legs along the iliac veins and receive branches accompanying the saphenous vein from the dorsum of - the hindfoot and branches from the popliteal node and the pos- terior inguinal node. A lymph vessel comes from the tail accom- panying the caudal vein and artery. From the anterior inguinal nodes usually two vessels parallel the iliolumbar artery and vein to the body wall, where a junction is made with other lymph ves- sels coming from the musculature of the body wall over branches of the iliolumbar blood-vessels. From the body wall to the lum- bar nodes the lymph channels vary from one single vessel to as many as three, in which case a richly anastomosing plexus of ves- sels is formed surrounding the iliolumbar vein and artery. A lymph vessel comes from each testis along the spermatic vein and artery and another passes through the mesentery from each testis to follow a branch of the hypogastric vein, hypogastric and iliac veins to the lumbar nodes. Still another lymph vessel follows the inferior mesenteric artery and vein from the large intestine to the lumbar nodes. 330 ROBERT W. HENDERSON Of the efferent vessels from the lumbar nodes only one is con- stant and lies between the aorta and the postcava. A second which lies on the left wall of the aorta is constant*in the most of the specimens, though it sometimes passes over the ventral side of the aorta to unite with the more constant. vessel. A third is occasionally present which passes from the lumbar nodes along the right side of the posteava. These vessels may vary from the almost simple condition as described to a very rich lymph plexus continuous from the lumbar nodes to the cisternal nodes and cis- terna chyli over the surface of the aorta and sometimes over both the aorta and the postcaval vein. CISTERNAL NODES AND VESSELS The cisternal nodes (figs. 3, 4, 5, and 6, Cis.gr.) are small nodes not larger than 5 x 1.5 x 0.5 mm. and vary in number from one to three. One, which is located just anterior to the left re- nal artery (figs. 3, 4, 5, and 6), is always present and always in this position. Of the others any or all may be present or all may be absent, and when present are exceedingly variable in position. One may be found posterior to the left renal artery (fig. 5), another posterior to the right renal artery (fig. 5) or an- — terior to the right renal artery (fig. 4), in which case there are sometimes two nodes present. The only afferent vessels of these nodes which have been demonstrated are those coming from the lumbar nodes, and their efferents all lead to the cisterna chyli. INTESTINAL NODES AND VESSELS The intestinal lymphatics comprise the superior mesenteric, gastric, and inferior mesenteric groups of nodes and their allied lymph vessels. The superior mesenteric nodes (fig. 2, Sup.Mes.N.) are a large group of nodes near the cecum, lying mainly on the left side of the mesenteric veins and arteries. These nodes are subject to great variation and may consist of a large number of small nodes closely grouped together or may consist of what appears to be a single large node having several parts, usually three or four. In any case the most distal end of the group lies along the cecal branch LYMPHATIC SYSTEM—STRIPED GROUND-SQUIRREL \ Mes.A. A.ides.L.V Loans R.I-1.A. a 331 Sup.Meg.N, P.Co.A. S.Mes.A. Fig. 2 Lymphatics of the digestive tract (diagrammatic) Ao., aorta A.Mes.L.V., afferent mesenteric lymph vessel A.Gas.N., anterior gastric node A.Co.A., anterior colic artery Colon, C.A., celiac axis Cor.V., coronary vein Cis.C., cisterna chyli Cae., caecum Ca.A., caudal artery Duo.A., duodenal artery Inf.Mes.N., inferior mesenteric node I.Mes.A., inferior mesenteric artery Mes.A., mesenteric artery P.V., portal vein P.Gas.N., posterior gastric node P.C.A., posterior colic artery R.I-l.A., right iliolumbar artery Rec., rectum Sp., spermatic artery S.Int., small intestine Sup.Mes.N., superior mesetnteric node S.Mes.A., superior mesenteric artery S.Mes.V., superior mesenteric vein Stom., stomach Sp.V., splenic vein 332 ROBERT W. HENDERSON of the superior mesenteric artery and vein, its distal end in con- tact with the cecum at the junction of the cecum and small in- testine. At the junction of the cecal blood-vessels with the rest of the mesenteric blood-vessels this portion of the node is continu- ous with a greatly expanded portion of the group which overlies all of the blood-vessels and continues as a gradually diminishing node to approximately a point where the superior mesenteric artery and vein become parallel with each other in the mesen- tery. This latter portion may consist of several small nodes which lie along the mesenteric blood-vessels at their proximal terminations, but which are more or less continuous with the main portion of the group. At the proximal end of the group, between the angle formed by the divergence of the superior mes- enteric vein and artery from their parallel position in the mesen- tery, is frequently a small node, approximately 3 x 1 x 0.5 mm. Almost continuous with it and lying along the mesenteric blood- vessels to the posterior end of the duodenum is a slightly larger node, 5x 2x 1 mm., which is always present. The afferent vessels to the superior mesenteric group of nodes pass through the mesentery from the walls of the small intestine, in general parallel with the mesenteric blood-vessels, though not necessarily close to them. In the walls of the intestine and ce- cum these afferent mesenteric lymph vessels are continuous with very fine plexuses of lymph vessels. However, there has never been any evidence of such plexuses in the mesentery. Appar- ently the afferent mesenteric lymph vessels, in the mesentery, are merely conducting vessels and not collecting vessels. The efferent mesenteric lymph vessels pass, usually as a plexus, to the cisterna chyli over the surface of the superior mesenteric artery. Anterior to the superior mesenteric group of nodes is a group of two nodes which, from their position, have been designated the gastric nodes, anterior and posterior, respectively. The posterior gastric node (fig. 2, P.Gas. N.) is approximately 4x 2x 1mm. in size and lies in the mesentery near the celiac axis in the angle formed by the splenic and coronary veins. A short distance anterior to the posterior gastric node, on the wall of the LYMPHATIC SYSTEM—STRIPED GROUND-SQUIRREL 333 coronary vein and near the cardiac wall of the stomach, is the anterior gastric node (fig. 2, A.Gas.N.) which is sometimes larger and sometimes smaller than the posterior gastric node. The efferent vessels from the posterior gastric node pass over the surface of the celiac axis to the cisterna chyli and those from the anterior gastric node pass along the coronary vein to the posterior gastric node or to its efferents. The afferent vessels of the anterior gastric node have never been demonstrated, but from its position it is inferred that they come from the stomach. The inferior mesenteric nodes (fig. 2, Inf. Mes. N.) when pres- ent, are usually two in number and lie in the mesentery on either _ side of the bifurcation of the inferior mesenteric artery and vein dorsal to the colon. The larger is usually 2 x 1 x 0.5 mm. in size, and the other, when present, is about half as big. The larger is the most posterior and receives an afferent vessel along the dorsal side of the rectum which closely parallels the rectum and receives - branches along its course from the rectal walls. The more an- terior node receives an afferent vessel which parallels the anterior branch of the inferior mesenteric artery and may connect anteriorly with a small node, sometimes present, in the mesentery near the superior mesenteric nodes which connects with the superior mes- enteric nodes. All along its course it receives branches from the intestine. The efferent vessels from the inferior mesenteric nodes pass dorsally on either side of the inferior mesenteric artery for a short distance, when they unite to form one vessel which enters the lumbar group. In some specimens the inferior mesenteric nodes are not pres- ent, in which case the lymph from the rectum and posterior colon drains into the lumbar nodes through several channels which pass dorsally in the mesentery. CISTERNA CHYLI The cisterna chyli (figs. 1, 3, 4, 5, and 6, Cis.C.) is the only lymph sac which may be considered as constantly present in the grounds-quirrel and is a sacculation of the lymph vessels which lie, in the greatest number of cases, on the ventral and lateral “SMOIA [CIJUDA PUB [RSIOP SMOYS G AINSI “wWI94sAs o1yeydursy] [vayUed ayy Jo SUOTyBIAvA eMl0s 9g pur ‘G ‘fF ‘Ee ‘sBT 9 g MOTA v § MSTA TRUqUOA . TEA qua, Tssu0g MOTA TBIZUSA MOTA Teaquon lin Be or °w°l"u a0 " "WTEC OBs "WI°l “39° oN | / a3°T\ RA) *W'TeI'y dated Z (e) Nn oa] a *12°ST9O : syne gj *9°8TO iq P | “T° ° ei heey f eae . * Si oh ae +4 a°ULe” on] ‘ ! Po ! on 1 ! = { ms 1 “dy a - 1 ata uy) /! ‘ | “A'2y ! | ait peNW | NA) "WITS, Ni qTa puzsy N *Yetry r iS sctgd | sta 3 : =H my uN “ATaNs* Tangy oF *a‘ans*7 Myeaey “ans’BeeT— Ain ‘ oO ay ‘ NS *a‘qns* * QO , "a" dag* Png ~ABto \ *ABTO *A‘Bar*g*7 *A*Bnp eg ; A*ane gy "Ate gy “atne art LYMPHATIC SYSTEM—STRIPED GROUND-SQUIRREL aa sides of the aorta between the superior mesenteric artery and the celiac axis. It is a rather variable structure and may be merely a rich plexus of lymph vessels or it may be a well-defined sac al- most completely surrounding the aorta, or it may be divided into two saes, one in the usual location and the other a little for- ward on the dorsa] surface of the aorta. Its afferent vessels are the vessels from the cisterna, lumbar, and intestinal nodes and its efferent is the thoracic duct. THORACIC DUCT The thoracic duct (figs. 1, 2, 3, 4, 5, and 6, Th.D.) is the largest lymph duct in the body and extends from the cisterna chyli along the right dorsolateral side of the aorta forward to the aortic arch and then along the dorsal side of the left subclavian artery an- terior to the first rib, where it turns ventrally to tap the venous system at the junction of the external jugular and left subclavian - veins. The thoracic duct is not always a simple vessel (fig. 5, R. and L.Th.D.) in fact, in most of the cases the posterior half is a more or less complex plexus of lymphatic vessels and in some cases a distinct left thoracic duct is evident. In one case (fig. 6, L.Th.D.) ABBREVIATIONS Ao., aorta L.I.V., left iliac vein Az.V., azygos vein L.I-l.A., left iliolumbar artery A.Ing.N., anterior inguinal node L.I-l.V., left iliolumbar vein B.C.A., brachiocephalic artery L.Jug.Sub.T., left jugular-subclavian B.W., body wall tap Clav., clavicle ' L.Th.D., left thoracic duct Cis.C., cisterna chyli L.E.Jug.V., left external jugular vein C.A., coeliac axis L.P.C.V., left precaval vein Car.A., carotid arteries P.C.V., postcaval vein Dia., diaphragm R.I-1.V., right iliolumbar vein I.Mes.A., inferior mesenteric arteries R.JI-1.A., right iliolumbar artery Kid., kidney R.Th.D., right thoracic duct L.R.V., left renal vein S.Mes.A., superior mesenteric artery L.Sub.A., left subclavian artery Sac., sacculation L.I.A., left iliac artery Sp., spermatic (ovarian) arteries L.Sub.V., left subclavian vein S.Int.A., superior intercostal artery 336 ROBERT W. HENDERSON the left thoracic duct, instead of passing dorsally around the aorta to unite with the thoracic duct proper, as is usually its termination, extended forward to a point opposite the seventh rib, then swung to the left and extended diagonally lateral and forward to the posterior end of the left superior intercostal ar- tery, where it could be traced no farther. VENO-LYMPHATIC TAPS The exact location at which the lymphatics tap the venous sys- tem is constant for only two points. On the right side the effer- ent lymph vessels from the axillary node tap the veins in, and a little to the dorsal side of, the angle formed by the junction of the right external jugular and right subclavian veins (fig. 1, R.Jug.-sub. T.), while the efferent vessels from the deep cervical nodes make a similar tap in the angle formed by the junction of the internal and external jugular veins (fig. 1, Jug.-jug.T.). A third tap which is constant in most specimens is the jugulo- subclavian tap (figs. 1, 3, 4, and 5, L.Jug.-sub.T.) on the left side. This tap is located in most cases at the same relative point as that on the right side and usually receives the lymph from the thoracic duct and the efferents from the left deep cervical and the left axillary nodes, which vessels unite, just at the point of junction, with the veins. Some rather interesting exceptions, however, have been found. In one specimen the lymph vessels from the left deep cervical nodes connected with the veins in the angle formed by the left internal and external jugular veins, while the thoracic duct and efferent vessels from the left axillary region made the usual tap in the jugulosubclavian junction. In another specimen the efferent vessel from the left deep cer- vical nodes, which follows the internal jugular vein, tapped the veins in the jugulojugular junction, while the efferent vessel from the left deep cervicals which follows the common carotid artery made the usual jugulosubclavian junction with the thoracic duct and left axillary efferent vessel. In two other specimens the thoracic duct, left deep cervical vessels and the left axillary vessels united to form a small but LYMPHATIC SYSTEM—STRIPED GROUND-SQUIRREL 337 perfectly distinct lymph sac dorsal to the left external jugular vein at its posterior end. From this lymph sac a short lymph vessel passed ventral to tap the left external jugular vein on its dorsal surface. Two other taps were found in the posterior end of the body which are unusual, at least direct evidence of their presence is unusual. Each tap was demonstrated but once and both oc- curred in the same specimen. One was in the postcava (fig. 1, P.C.T.) and the other was in the left renal vein (fig. 1, L.R.T.). The injection which demonstrated the presence of the renal and postcaval tap was made in the right inguinal node. The mass was seen to pass into the lumbar nodes and start forward toward the cisternal nodes. At almost the same instant the mass appeared in the postcava. As the mass filled the lumbar efferent vessels and cisternal nodes, it passed through a lymph vessel, on the posterior side of the left renal vein, in a diagonal direction, lateral and anterior, over the lumbar musculature. At a point near the left renal vein the mass disappeared under some fatty tissue, and almost simultaneously near the hilus of the kidney was seen to appear in the renal vein. Later an attempt was made to prove the presence of these taps by dissection in the following manner. The animal was har- dened in formalin solution and the veins opened longitudinally on the ventral side. The hardened blood was then removed piece by piece, the work being done with needles under a high- power binocular microscope. After the blood was all removed the pattern of the injected lymphatic vessels, which formed a plexus on the dorsal side of the vein, was plainly visible. From this plexus a vessel extended posteriorly for some distance parallel with the direction of the postcava and gradually passed through the wall of the postcava at a very small angle, to form a tap on the dorsal side, a little in front of the anterior level of the lumbar nodes. The point of the tap was easily seen as the hard injection mass projected out into the vein and was continuous with that in the lymphatic vessel. Contrary to expectations, the direction of the vessel approaching the tap and the direction of the mass pro- jecting from the tap was opposite to that of the flow of the blood in the vein. THE ANATOMICAL RECORD, VOL. 16, NO. 5 338 ROBERT W. HENDERSON In the dissection of the renal vein the results were not so good, as none of the mass had remained in the lymph vessel approach- ing the tap and the point of entrance could not be determined. At two other points in a number of specimens the injection mass was observed to appear in the veins. One was in the por- tal vein and the other was in the right iliolumbar vein near the body wall. While it is probable that venolymphatic taps occur at these points, still no such absolutely definite evidence of their presence was obtained as in the case of the postcaval and renal taps. For instance, in the case of the portal vein, very fre- quently the mass passed into it when the injection was made from the superior mesentric nodes, yet while almost complete injec- tions were often obtained in the superior mesenteric nodes by the mass backing up into them from the cisternal region during in- jection from the anterior inguinal nodes, still never in the case of this sort of injection did the mass pass into the portal vein, or if it did, it was in such small amounts as not to be noticed. It is felt that more work is necessary on these points before a definite statement can be made in regard to their presence. LYMPH VALVES Lymph valves are very numerous in all main lymph vessels, as from the cervical nodes to their venous taps, or in the thoracic duct or from the lumbar nodes to the cisterna chyli, but they do not seem to be present at all from the cisterna chyli into the in- testinal nodes. Due to this fact, practically complete injec- tions of the intestinal nodes have been obtained from injections made in the inguinal nodes which resulted from the mass backing up into the intestinal nodes from the cisterna chyli. The presence of a valve is evidenced by a constriction in the lymph vessel, so that if there are many valves present the vessel, when filled, has the appearance of a series of bulges. LYMPHATIC SYSTEM—STRIPED GROUND-SQUIRREL 339 CONCLUSION Two points at least ought to be emphasized which are unusual in this animal. They are the jugular lymph sac and the renal and postcaval venolymphatic taps. The presence of the jugular lymph sac is an unusual structure in adult anatomy and perhaps is but a remnant of the embryonic condition. The embrylogy of this form has not as yet been worked out, however, so this structure is interesting merely from the speculative standpoint. The renal and postcaval taps may be more common than their demonstration frequency would indicate, though that is largely conjecture. However, the appearance of the postcaval tap was a total surprise, while that of the renal tap was rather expected, since in a number of specimens the injection mass passed out through the lymphatics toward the renal vein as described, though in only one case did it actually appear in the renal vein. The fact of the presence of these taps even in so small a per- centage of the animals as this study would indicate is additional proof that the venolymphatic taps of the jugulosubclavian region are not always the only points through which lymph is returned into the blood. Occasion is taken here to express my appreciation for the kindly interest and assistance rendered by Doctor Stromsten, under whom these studies were made. t q j ; Right ada ndiiracie ls 1% ealey eal 0! Oi 2] al, bby O1on) s Prilayis!: fA ‘5 os De HOS: sett ¥ re id tig dail aaa walang , i OO tthe teies a> Renn EL) my | yy ~-# ANA ‘ Be SI a iY Bah ul Be BS Aneel eer: Bipewirai> | vicloa #-AGt ised) A ARTEL iri! oan niet » ga f Amini i A 2g a HOt bit ‘NJ a f 77 * + fee * ys i ‘ 4 At jek ” - * t4 “rear ‘i . te : . : 5 a ‘ Resumen por el autor, Otto F. Kampmeier, Universidad de Illinois. Sumario de una monografia sobre la morfologia del sistema linfatico de los anfibios anuros, con especial mencién de su origen y desarrollo. La presente monografia, brevemente resumida, consta de las siguientes secciones: 1). Observaciones sobre el sistema linfdtico de los individuos adultos; 2) El desarrollo de las venas sistémicas y su relacién con el sistema linfadtico; 3) Los linfdticos de un embrién de sapo de 15 mm., con descripciones del orfgen y desarrollo de: 4) El seno maxilar primario; 5) El seno yugular linfatico; 6) Los corazones linfaticos anteriores; 7) Los linfaticos laterales del troneco; 8) Los linfdticos subvertebrales; 9) Los corazones linfaticos posteriores; 10) Los linfaticos dorsales, laterales y ventrales dela cola; 11) Los capilares linfdticos; 12) Los sacos linfaticos definitivos. La falta de espacio permite tan solo el dar a conocer aqui algunas conclusiones generales. Los grandes canales linfaticos se forman a expensas de esbozos pequefios y discontinuos, que estan colocados en contacto con las venas o son independientes y se hallan en el mesenquima. Su continuidad se establece por su fusién por los extremos. Vasos linfdticos mds pequefos y capilares se producen por proliferacién del endotelio de los canales mayores. El corazén linfatico anterior se produce en un periodo poco avanzado del desarrollo y se diferencia de un plexo venoso-linfatico que forma parte del plan circulatorio primario; el coraz6n posterior aparece relativamente tarde y se diferencia de un plexo linfdtico, que en este estado esta claramente separado de la organizacién hemal. En la formaci6n de los sacos linfdticos definitivos (secundarios) pueden reconocerse tres métodos distintos: 1) Conversién de un plexo de capilares linfdticos pre-existente con la posible par- ticipacién del mesenquima que le rodea; 2) La simple adaptacién de un seno embrionario a las condiciones del estado larvario avanzado y las del adulto; 3) La distensién marcada de un conducto. Translation by José F. Nonidez Carnegie Institution of W shington AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, AUGUST t A SUMMARY OF A MONOGRAPH ON THE MOR- PHOLOGY OF THE LYMPHATIC SYSTEM IN THE ANURAN AMPHIBIA, WITH ESPECIAL REFERENCE TO ITS ORIGIN AND DEVELOPMENT" OTTO F. KAMPMEIER Department of Anatomy, College of Medicine, University of Illinois, Chicago, Ill. The writer has devoted much time during the past five years to a study of the lymphatic system in Amphibia and has brought together the results in a monograph, illustrated by many figures and plates. The first half was ready for publication in August, 1917, but on account of the war and its attendant difficulties and unavoidable delays, it was not to appear until the coming March (1919). Meanwhile, the second half has been completed, and since it would be more logical and satisfactory to publish the two parts together, the first was withdrawn and has been com- bined with the second part in a single treatise. Present circum- stances will not permit its early appearance in print, and because a great many of the data were gained several years ago, he deems it expedient to furnish a summary of it now for the benefit of other investigators. Larval and adult specimens of Bufo vulgaris, B. lentiginosus and Rana pipiens constituted the material used in the investi- gation. The facts which the author’s researches revealed and the conclusions which have been reached are briefly outlined as follows: 1. ON THE LYMPHATIC SYSTEM IN FULLY FORMED INDIVIDUALS 1. In the neighborhood of the lymph hearts, valves guard the openings between adjacent lymph sinuses. 2. The anterior lymph heart, paired, is situated dorsal to the transverse process of the third vertebra, and it communicates 1 Ready for publication, January 1, 1919. 341 342 OTTO F. KAMPMEIER by means of a valve with the anterior vertebral vein, a short tributary of the internal jugular vein. In a young specimen of Bufo lentiginosus, soon after its metamorphosis, there were five valvular apertures between the lymph heart and the bordering lymph sac (sinus subscapularis). 3. The single pair of posterior lymph hearts in a young indi- vidual of Bufo lentiginosis is slightly smaller than the anterior (anterior heart, 0.45 mm.; posterior, 0.42 mm.); each joins the posterior vertebral vein of the corresponding side, a branch of the ischiadic vein, and each possesses four afferent pores. 4. In an adult Rana pipiens two posterior lymph hearts were present on the right side and three on the left, the third or most caudal one being very small and partly fused with the next one in front; each one of these hearts, the vestigeal one included, has its own ostium venosum. The larger and functional hearts have from twenty to thirty afferent portals. 5. The wall of a lymph heart is composed of three coats: 1), a tunica interna, formed of typical endothelial cells; 2), a tunica media, which contains muscle bundles interlaced in a complex manner; 3), a tunica externa or adventitia, which binds the lymph heart to the surrounding tissues. 6. Sometimes a trabeculum is found stretching across the cavity of the lymph heart. 2. ON THE MODIFICATION OF THE VENOUS SYSTEM DURING DEVELOPMENT 1. The primary venous ground-plan of anuran embryos is similar to that of other vertebrate embryos. 2. The réle played by the subcardinal veins in the formation of the posteava corresponds closely to that in other vertebrates. 3. By the degeneration of the anterior segment of the post- cardinal vein, the pronephric sinus (the plexus of venous sinu- soids encompassing the pronephros and originally constituting the junction of the pre- and postcardinals and Cuvierian duct) dwindles in size and becomes the proximal segment of the pre- cardinal or internal jugular vein. DEVELOPMENT OF LYMPHATIC SYSTEM—ANURA 343 4. The primitively symmetrical system of intersegmental veins undergoes radical modifications, at first producing the vein of the lateral line by longitudinal anastomoses, and subsequently dif- ferentiating into the anterior and posterior vertebral and lumbar- dorsal veins. 5. The anterior and posterior lymph hearts of the toad (Bufo vulgaris and B. lentiginosus) develop in relation to the third and eleventh intersegmental veins, which drain into the postcardinal. These tributaries ultimately become parts of the anterior and posterior vertebral veins, which, in consequence of the reduction of certain segments of the embryonic venous plan and the marked shifting of relations that occurs during development, come to be branches of the internal jugular and ischiadic veins, respectively. 3. ON THE COMPONENTS OF THE LYMPHATIC SYSTEM IN 15-MM. TADPOLES (BUFO) 1. A large sinus, the primary maxillary lymph sinus (sinus lymphaticus primarius maxillaris), which may be divided into circumoral, mandibular, temporal, and paracardial divisions, is situated in the ventral and lateral regions of the head. 2. A jugular lymphatic (lymphatica jugularis),? one on either side, runs parallel to the proximal portion of the internal jugular vein and connects the primary maxillary sinus and the anterior lymph heart. 2 At the present time a few investigators, the writer included, are using the adjective ‘lymphatic’ (Latin, lymphaticus-a-um) also as a noun to designate the larger lymph channels. The lymphatic system, constituting the third major subdivision of the vascular system, the term ‘lymphatic,’ used instead of ‘lymph duct,’ seems more specific and to accord better with the terms ‘vein’ and ‘artery.’ In his present work the writer has also carried this idea over into the scientific terminology. The expressions ‘vas lymphaticum’ and ‘truncus lymphaticus’ have been generally used to designate the important lymph vessels. The writer, on the other hand, has dropped the word ‘vas,’ signifying vessel, and has taken over the Latin adjective in its feminine form, ‘Lymphatica,’ as a substantive sub- stituting it as the generic name for the larger lymph channels or ducts, because it conforms with the terms ‘vena’ and ‘arteria.’ The Latin word ‘lymphatus,’ meaning practically the same as ‘lymphaticus,’ might be employed as ‘lymphata,’ which being shorter than ‘lymphatica’ would be preferable, but the latter term, being more familiar, is more acceptable. 344 OTTO F. KAMPMEIER 3. The pair of anterior lymph hearts (corda lymphatica an- teriora) lie lateral to the third and fourth myotomes at the level of the third spinal ganglia and join the rudiments of the anterior vertebral veins at the junction of the latter with the pronephric venous sinuses. On the opposite or dorsal side of each heart a single afferent gateway is present. 4. The lateral lymphatic of the trunk (lymphatica lateralis corporis) courses along the myotomes between the anterior and posterior lymph hearts. 5. The paired subvertebral lymphatic (lymphatica subverte- bralis) passes from the anterior lymph hearts medially, thence along the aorta to the caudal end of the trunk, where they bend lateralward to unite with the lateral lymphatic in the vicinity of the posterior lymph heart. 6. A dorsal lymphatic (lymphatica dorsalis corporis et cau- dalis) exists in the midline above the neural tube and stretches from the anterior region of the trunk to the tip of the tail; it con- nects with the lateral lymphatics by transverse branches. 7. The posterior pair of lymph heart anlagen lie at the level of the eleventh spinal ganglia against the posterior vertebral veins, but in 15-mm. embryos have not yet established a com- munication with these veins. 8. The lateral lymphatic of the tail (lymphatica lateralis cau- dalis) is a direct continuation of the lateral-line lymphatic of the trunk. 9. The ventral caudal lymphatic (lymphatica ventralis cau- dalis) begins as a paired vessel at the point of confluence of the subvertebral and lateral lymphatics andis prolonged into the tail, converging and fusing with its fellow further back. 10. The lymphatic extensions ot the fore and hind extremities (lymphaticae brachialis, iliacis et femoralis) appear later. 11. Lymph capillary plexuses springing from tributaries of the large deep lymph ducts, named above, traverse the subcutaneous tissue. DEVELOPMENT OF LYMPHATIC SYSTEM—ANURA 345 4. ON THE ORIGIN AND DEVELOPMENT OF THE PRIMARY MAXILLARY LYMPH SINUS 1. The anlagen of the primary maxillary lymph sinus first be- come visible in 5-mm. embryos (Bufo vulgaris) as small, solid and discontinuous cell masses frequently attached to the wall of a bilateral haemal channel, the potential inferior jugular vein which is at this time a component of the primitive vascular plexus ventral to the oral and pharyngeal cavities. 2. The writer is somewhat in doubt regarding the significance of the adherence of some of the maxillary sinus anlagen to the potential inferior jugular vein during their early genetic stage. Are they derivatives of the haemal endothelium, or are they analogous to the extra-intimal lymphatic spaces of mammalian embryos? (Aside from the fact that the view of the discontinu- ous mesenchymal origin of all vascular anlagen, irrespective of whether haemal or lymphatic, has been shown to be fairly con- clusive in recent investigations, the blood vascular plexus, just referred to is still indifferent, that is, has not clearly differenti- ated into arteries and veins, at the time when the lymphatic an- lagen adhere to it. It is conceivable how at such an early de- velopmental stage, the actively proliferating endothelium of the primitive vascular plexus may also assist in the production of the third set of vascular channels, the lymphatics.) 3. The discrete lymphatic anlagen of the primary maxillary lymph sinus lose their contact with the veins, acquire lumina, elongate and enlarge, and proliferate extensions, some of which coalesce with one another of the same side, and others project ventromedially to connect with those from the other side, thereby creating an intricate network of lymph vessels just below the external jugulars and carotids. 4. This lymphatic network, representing the pars mandibu- laris or principal portion of the latent sinus, gives rise to the other divisions, the circumoral, the temporal, and the paracardial, by outgrowths forward, lateralward, and backward. 5. During these developmental stages the endothelium of the vascular channels and anlagen, both haemal and lymphatic, can be distinguished plainly from the surrounding mesenchymal cells by its possession of a greater number of yolk globules. 346 OTTO F. KAMPMEIER 6. The lymphatic plexus gradually becomes converted into a spacious and uninterrupted lymph reservoir by the expansion and fusion of its numerous components. In this genetic process, the interstices of the cireumjacent mesenchyme do not partici- pate, for the outlines of the sinus remain sharply defined during its multilocular condition. 7. During the period of its active formation the primary maxil- lary lymph sinus is nowhere in open connection with other vascu- lar channels, neither with blood nor with lymph vessels. Its appearance of great distention is doubtlessly due to the lymph, which, permeating the tissues, accumulates within it in rapidly increasing volume, there being no outlet. Blood cells are not present in its cavity at any time. 8. A communication is established between the temporal divi- sions of the primary maxillary sinus and the jugular lymphatics, and thus indirectly with the anterior lymph hearts, in approxi- mately 8- to 9-mm. embryos (Bufo vulgaris). 5. ON THE ORIGIN AND DEVELOPMENT OF THE JUGULAR LYMPHATICS AND THE ANTERIOR LYMPH HEARTS 1. The jugular lymphatic and the anterior lymph heart have their origin in a venolymphatic plexus, derived from the first three intersegmental veins which are an integral part of the early vas- cular system of the embryo. 2. All connections between the veins (pronephric venous sinus) and the plexus are lost, except those of a demarcated and more close-meshed portion of the plexus which represents the anlage of the lymph heart. 3. The first definite indications of the heart anlage can be ob- served in 4- to 5-mm. embryos (Bufo vulgaris). 4. The simple globular chamber of the anterior lymph heart is produced by the dilation and confluence of the channels of its antecedent plexiform phase and by the further distention of the single cavity so formed. 5. The remainder of the venolymphatice plexus becomes a lym- phatic one by its detachment from the vein; in it a conduit be- comes dominant which extends forward to the pars temporalis of the primary maxillary lymph sinus as the jugular lymphatic. DEVELOPMENT OF LYMPHATIC SYSTEM—ANURA 347 6. The developing lymph heart temporarily breaks away from the surrounding lymphatic plexus. 7. A union is secondarily reestablished between the lymph heart and the common segment of the jugular and lateral-line lymphaties. This is effected by the coincident expansion of the caliber of both structures which brings about an intimate contact between them and so makes possible the development of a portal from a rupture in the intervening partition. At this point the afferent valve is formed by a thickening of the endothelial cells; later, more such portals arise in the manner indicated. 8. The efferent valve is developed from oppositely placed en- dothelial cushions in the lumen of the lymphaticovenous junction. 9. Blood cells are found in the heart until the valves have been laid down and its pulsations have commenced. 10. The characteristics of certain cells during the transforma- tion of the plexus into the single-chambered lymph heart suggest a concomitant haemopoietic process. . 11. The muscle coat of the lymph heart is derived from the bordering mesenchymal cells, which become fusiform and modi- fied into contractile fibers and gradually acquire a parallel ar- rangement in a distinct layer. 6. ON THE ORIGIN AND DEVELOPMENT OF THE LATERAL LYMPHATICS OF THE TRUNK 1. A double series of discontinuous lymphatic anlagen appears in the lateral axial region of the trunk throughout its entire length, one row situated along the longitudinal anastomosis (lat- eral-line vein) of the intersegmental veins, and the other along the postcardinal vein. 2. When the lymphatic anlagen can first be clearly recognized as such, they lie as small definitely walled spaces, either against the intima of the venous channel or in close proximity to it. 3. By their elongation and fusion two lymph channels result, the lymphatic of the lateral line (lymphatica lateralis lineae cor- poris) and the juxtacardinal one (lymphatica juxtacardinalis). 4. The two lymph conduits, at first independent of one another, become joined as they sprout numerous branches and produce a luxuriant plexus lateral to the myotomes. 348 OTTO F. KAMPMEIER 5. This lymphatic plexus establishes continuity with the lym- phaties in the territory of the anterior lymph heart; in fact, the extreme anterior portion of the plexus is derived from the veno- lymphatic one from which the jugular lymphatic and the lymph heart are developed. 6. In the lateral lymphatic plexus, a vessel becomes paramount by distention and the incorporation of collateral vessels, and it consequently transmits more and more of the lymphatic stream; this vessel has been termed the lateral lymphatic of the trunk lymphatica lateralis corporis). 7. ON THE ORIGIN AND DEVELOPMENT OF THE SUBVERTEBRAL LYMPHATICS (THORACIC DUCTS) 1. The rudiments of the paired subvertebral lymphatic, when first distinguishable as such, exist as consecutive spindle-shaped spaces, located lateral to the aorta and joined to slender medial extensions of the juxtacardinal lymphatics. 2. Acquiring continuity by the end-to-end union of its rudi- ments, the lymph duct is at first very much attenuated, but pro- gressively becomes more conspicuous in section by the marked expansion of its lumen. 3. Haemopoiesis occurs in the axial mesenchyme; this is es- pecially true in the iliac region where blood cells in different stages of development are scattered throughout the mesenchyme and make their way into lymph vessels, thus demonstrating the accessory haemophoric function of the embryonic lymphatics. 8. ON THE ORIGIN AND DEVELOPMENT OF THE POSTERIOR LYMPH HEARTS 1. The first sign of the origin of the posterior lymph heart man- ifests itself in 10- to 1l-mm. embryos (Bufo vulgaris) at the level of the eleventh spinal ganglia as an accumulation of mesenchymal cells in a small circumscribed area encompassing certain channels of the lateral lymphatic plexus which lie against the potential posterior vertebral vein. 2. The lymph channels within the zone of the mesenchymal condensation become even more plexiform, after which by their DEVELOPMENT OF LYMPHATIC SYSTEM—ANURA 349 widening and complete fusion along their apposed surfaces they produce a single unbroken cavity, the chamber of the lymph heart. 3. During the foregoing genetic changes the anlage of the lymph heart is nowhere in direct and open communication with the contiguous vein. 4. The original connections (approximately seven in number) between the lateral lymphatics and the lymph-heart anlage be- come constricted and break away. The isolation of the heart cavity is only a transitory one, for one or two junctions are soon reestablished where lymph heart and the nearest lymph vessels come into juxtaposition as a result of their expansion. 5. The valves of the afferent portals of the heart chamber ' originate in the same way as do those of the anterior lymph heart. The ostium venosum, or lymphatico-venous tap, on the contrary, is formed by the thinning out of the wall between lymph heart and vein, followed by the growth into the venous lumen of an endo- — thelial projection, which becomes hollow, thereby producing the aperture as well as the margins of the teat-like valve. 6. During its development, the posterior lymph heart is an haemopoietic focus; blood cells are formed within it, and still others pass through it on their way to the blood stream. 7. The muscular tissue of the heart wall is derived from the original aggregation of mesenchymal cells. 8. Comparing the formation of the anterior lymph heart with that of the posterior lymph heart, a considerable difference is evident. Whereas the former arises early in embryonic devel- opment and is differentiated from a venolymphatic plexus, which is part and parcel of the primitive circulatory plan, the latter ap- pears relatively late and is differentiated from a lymphatic plexus at that time clearly sequestered from the haemal organization. 9. In an advanced larva of Rana pipiens there were three pos- terior lymph hearts on both sides, corresponding to the eleventh, twelfth, and thirteenth intersegments; another larva, only slightly older, possessed two on each side, thus demonstrating that a dif- ference in the number of such hearts may exist in the same species. 350 OTTO F. KAMPMEIER 10. Variations may also occur in the same individual, as evi- denced by the finding of a normally sized posterior lymph heart on one side and a small one on the opposite side in a tadpole of Bufo lentiginosus (also compare section 1, 4). 9. ON THE ORIGIN AND DEVELOPMENT OF THE DORSAL, LATERAL, AND VENTRAL LYMPHATICS OF THE TAIL 1. The dorsal lymphatic has a bilateral origin in the trunk, beginning as a double series of spindle-shaped anlagen located dorsal to the neural tube. 2. When the successive anlagen have become continuous, the two channels combine into a single one; its posterior end then grows caudally into the tip of the tail by the proliferation of its endothelium. 3. The lateral caudal lymphatic represents a caudal prolonga- tion of the similarly situated duct in the trunk. 4. The ventral caudal lymphatic develops as a growth back- ward from the confluence of the subvertebral and lateral ducts of the trunk; at first paired, it fuses into a single structure except at its proximal end. 10. ON THE FORMATION OF THE LYMPHATIC CAPILLARIES 1. After the larger lymph ducts are laid down by the coales- cence of originally separate anlagen, they elongate by the prolif- eration of their intimal cells and bud branches, which grow centrif- ugally and form the lymph capillary plexuses. 2. The observation of the sprouting and centrifugal growth of the lymphatic capillaries cannot be used as an argument against the theory of the in situ and discontinuous origin of the larger systemic lymph conduits. 11. ON THE TRANSFORMATION OF THE LYMPHATIC VESSELS OF THE TADPOLE INTO THE LYMPH SACS AND SINUSES OF THE ADULT 1. Three variations in method can be recognized in the forma- tion of the definitive lymph sacs: 1), the conversion of an ante- cedent lymph capillary plexus with the possible participation of DEVELOPMENT OF LYMPHATIC SYSTEM—ANURA abt the circumjacent mesenchyme; 2), the simple adaptation of an embryonic sinus to late larval and adult conditions; 3), the marked distention of a duct. 2. All of the deep lymph sinuses of the mature animal are de- rivatives of the large lymphatic channels of the embryo; all of the superficial sacs, with the exception of the submaxillary and temporal, have their origin in the most terminal or subcutaneous branches of these channels. 3. The submaxillary and temporal lymph sacs are directly de- rived from the mandibular and temporal divisions of the primary maxillary lymph sinus, the sublingual, hyoidal, pulmonary and sternal sinuses originate as diverticuli of the paracardial divi- sions, and the large unpaired basilar sinus develops from a bi- lateral extension forward of the junctions of the paracardial and temporal divisions of the primary maxillary sinus and the jugular lymphatics. 4. The paired subscapular sinus of the adult is in part derived from the posterior portions of the paracardial and temporal divi- sions of the primary maxillary sinus and in part from the entire jugular lymphatic, which explains its intimate relation to the anterior lymph heart. During development, it becomes divided from the temporal lymph sac by the formation of the scapula and its musculature. 5. The two subvertebral lymphatics, having come in apposi- tion by their dilation, combine into a single channel, which by further expansion becomes an extensive sinus. In this genetic process, it absorbs the lateral lymphatics and their deep tributaries. 6. The dorsal lymphatic of the trunk undergoes regression dur- ing the later larval period, but a small vessel probably persists to drain the neural canal and its contents. 7. The mesenteric lymphatics and the deep sinuses of the pos- terior portion of the abdominal cavity, such as the pelvic, pubic, periproctal, etc., arise as extensions and evaginations of the sub- vertebral sinus and insert themselves between the viscera. 8. The iliac sinus, adjacent to the posterior lymph heart in the mature individual, corresponds to the influence of the sub- vertebral, lateral, and ventral caudal lymphatics in the larva. 352 OTTO F. KAMPMEIER 9. All of the superficial lymph sacs (except the submaxillary and temporal) of the head and trunk, such as the supra-orbital, craniodorsal, lateral, abdominal, and pectoral sacs, as well as those of the extremities, have their beginning in the subcutane- ous lymph capillary plexus, derived from tributaries of the dor- sal and lateral lymph ducts of the trunk. 12. ON THE HOMOLOGY OF THE CHIEF COMPONENTS OF THE LYMPHATIC GROUND-PLAN IN THE DIFFERENT GROUPS OF VERTEBRATES 1. Among vertebrate animals three regions of lymphatico- venous communications can be recognized: an anterior, a middle, and a posterior region. The anterior or jugular one is most con- stant, being present in all classes and orders of vertebrates. In the tailed amphibians (Urodeles and Gymnophionia) the numer- ous metamerically arranged lymphaticovenous connections of the lateral line, extending from the jugular to the caudal regions, do not permit such a definite division into regions. The middle group of taps have been clearly demonstrated in Mammals (Pri- mates, Marsupials, Rodents). The posterior or caudal lymphati- covenous junctions are present in Fishes, Amphibia, Reptiles, Birds, and perhaps in the embryos of Mammals. 2. The multiple posterior lymph ,hearts of certain Anura must be considered as a persistence of a number of the segmental lymph hearts found in the more primitive Amphibia, the Uro- deles and Gymnophionia. 3. The writer regards the caudal hearts, both venous and lymphatic, of Fishes, the posterior lymph hearts of Amphibia, Reptiles, and Birds, and the vestigeal posterior or iliac sacs of mammalian embryos as homologous structures. 4. The author believes he has morphological evidence to show that the posterior portions of the jugular lymph sacs of Mammals, Birds, and Reptiles are identical with the anterior lymph hearts of Anura. 3 References to the literature will be cited in the monograph. DEVELOPMENT OF LYMPHATIC SYSTEM—ANURA 353 5. Topographical relations and genetic data show that the primary maxillary lymph sinus of anuran tadpoles corresponds to the subocular lymph sinus of Fishes. 6. The paired subvertebral lymphatic is represented in all classes of vertebrates, but different names are applied to it, such as, thoracic ducts, periaortic sinus, prevertebral lymphatics, ab- dominal sinus, etc. 7. The bilateral jugular lymphatic (cephalic duct, truncus jugularis, lateral pharyngeal lymphatic) is present in all groups of vertebrates. 8. The lymphatic of the lateral line evidently is of constant occurrence in all Fishes, Gymnophionia, and Urodeles, in the tadpoles of Anura, in the embryos of Reptiles and Birds, and the writer believes vestiges will eventually be found in mammalian embryos. 9. The dorsal lymphatic is a common feature of Anamia; the . embryos of Amniotes have not been sufficiently studied in this respect. 10. A number of the superficial lymph channels in Fishes and in the Amphibia. can be readily homologized, but this is impossi- ble in the higher vertebrates. Resumen por el autor, George Washington Tannreuther. Universidad de Missouri. Duplicacién parcial y completa en los embriones de gallina. Los embriones que se describen en el presente trabajo han sido seleccionados en una coleccién hecha durante un periodo de diez anos. A. El estado més jéven observado, en el cual existen dos o més embriones, est& representado por un blastodermo con cuatro lineas primordiales, dos de las cuales se extienden ante- riormente y las otras dos posterior y lateralmente encontrdndose sus extremos posteriores en una regién comin. Las lineas primitivas no presentan desviacién alguna de la estructura normal. Un estado ulterior de duplicacién completa esta representado en un blastodermo con dos embriones normales completamente independientes, correspondiente al comienzo del segundo dia de incubacién. Los dos embriones diferen algo en su estado de desarrollo. B. Duplicaci6én parcial: 1). Un blastodermo préximamente a las 21 horas de incuba- cidn presenta un embrién con una sola linea primitiva, la cual termina anteriormente en dos nudos de Hensen, dos pro- cesos cefdlicos y la formacién temprana de los pliegues neural y cefalico. 2). En otro caso un embrién estd representado por una sola cabeza y regién del cuello, que termina posteriormente en dos troncos distintos. 3). Un caso de anormalidad extrema esta representado en un blastodermo en el cual dos embriones poseen regiones del troneo normales las cuales son continuas en sus extremos anteriores y presentan las regiones cefdlica y del cuello comprimidas anormalmente. 4). Un embrién prdéxima- mente a las 68 horas de incubacién presenta dos corazones normales claramente separados. El autor incluye en el texto diversas figuras que demuestran las anormalidades descritas. Translation by José F. Nonidez Carnegie Institution of Washington AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, JULY 7 PARTIAL AND COMPLETE DUPLICITY IN CHICK EMBRYOS GEORGE W. TANNREUTHER Zoological Laboratory, University of Missouri SIX FIGURES It is not an unusual thing to find abnormalities of varying degrees in chick embryos. Especially is this true in the case of poorly regulated incubators. Development may begin in a per- fect normal way, but a low or a high temperature often causes abnormal growth in later stages. The question, how a single blastoderm possesses the potency to produce one or several embryos, is as yet an unsolved problem. It has been proved experimentally, in some animals at least, that two or more em- bryos can be produced from the parts of one ovum. In no instance, however, has it been demonstrated that the number of individuals resulting from a single blastoderm can be controlled. Until some plausible explanation is given for the splitting up of the blastoderm into several equipotent regions, we can merely give a description of these unusual departures as they are found from time to time in the embryology of the different vertebrates. In the armadillo, normally, four or more embryos result from a single fertilized egg. The separation of the embryonic rudi- ments becomes visible after the formation of the two primary germ layers. In the chick, normally, but a single embryonic rudiment appears on the blastoderm. In the case of the chick, partial duplication of parts is frequent, but the formation of two or more complete embryos on a single blastoderm is a rare occurrence. The following paper is a brief description of a few very un- usual chick embryos, which may be considered of considerable value or importance, especially for the embryologist, who is more directly interested in the occurrence of polyembryony as it may be found in the different groups of vertebrates. 305 THE ANATOMICAL RECORD, VOL. 16, NO. 6 356 GEORGE W. TANNREUTHER The embryos were fixed in picrosulphuric killing reagent, stained in alum-cochineal, cleared in xylol, and mounted in damar. The embryos were sufficiently transparent that the various structures at different levels could be followed with a considerable degree of accuracy. The embryos were collected and preserved during a period of ten years. Those with minor or slight abnormalities were discarded, while those of unusual formations were mounted and preserved. Figure 1 represents the earliest stage in which abnormalities were observed. The area opaca and area pellucida do not vary much from that of the normal. The four primitive streaks fuse at a common point (C.F.Pr.St.), which from all appearances represents the posterior ends of the four structures. The primi- tive streaks 1 and 2 are perfectly normal and correspond to an embryo of about seventeen hours. In either case the primitive folds and groove are well marked and show a distinct primitive pit at their anterior ends. Hensen’s node is rather faint. The head process and the head fold in either case are distinctly nor- mal and show no departure from the usual condition. The formation of the three germ layers compares to that of the typ- ical embryo at the same stage of development. The primitive streaks 3 and 4 are much larger than 1 and 2 and more widely separated. The primitive folds, grooves, and pits are better developed than the two smaller streaks. The head process in either case is very faint and almost indistinguishable. The head folds are well developed. The primitive streaks 3 and 4 near their common point of fusion are poorly developed and do not show the presence of folds at all. The mesoderm in’the region equidistant from either streak is well developed and gives the appearance of a distinctly drawn-out shaded area (Mes.). The germ layers have kept equal pace with the streaks in their devel- opment. The primitive streaks 3 and 4 might be taken as the posterior extensions of 1 and 2, but when we take into consider- ation their well-developed structures corresponding to the con- ditions found in the normal primitive streak, we can undoubtedly regard them as distinct primitive streaks. Figure 1 in all probability represents a blastoderm which has Me, ee ’. sen, Ae, Astin ie, cent RE NMOS ALIANT tase ep peciaicnt tite ~are alec oN Te mete? Ate “ 2* pe tect erate Fig. 1 All figures were made with the aid of a camera lucida, at table level with a 16-mm. objective and a no. 6.4 eye-piece. removed. All drawings are about the same magnification and reduced four- sevenths. ABREVIATIONS Ao.A., aortic arch Ab.Br., abnormal brain A.C.S., anterior cerebral suture A.C.V., amnio cardiac vesicle A.I.P., anterior intestinal portal Al., allantois Am., amnion A.Op., area opaca A.P., area pellucida A.Vas., area vasculosa Au.Ves., auditory vesicle B.Is., blood islands C.F.Pr.St., common fusion of prim- itive streaks C.Pr,St., common primitive streak Ec., ectoderm Ent., entoderm F.B., fore brain F.G., fore gut F.L., fore limb G.C., gill cleft H.B., hind brain H.F., head fold H.G., hind gut H.N., Hensen’s node H.Pr., head process Ht., heart I.Gr., intestinal groove Mes., mesoderm Mes.So., mesoblastic somite M.B., mid brain Nch., notochord N.F., neural fold Op.Ves., optic vesicle Pr.Am., pro amnion Pr.Gr., primitive groove Pr.Pl., primitive plate Pr.St., primitive streak S.T., sinus terminalis T.F., tail fold Tr.Ars., truncus arteriosus V.Ao., ventral aorta Vi.Ar., vitelline artery Vt.V., vitelline vein 307 eae event em seri vahe die NYAS Leen eseoryy Bien The front lens of the objective was 358 GEORGE W. TANNREUTHER become divided into four equipotent regions. The question arises, if the blastoderm with the four primitive streaks con- tinued to develop, would the four resulting embryos be inde- pendent or united into a single monster. Judging from the conditions found in figure 3, four distinctly independent embryos would have resulted. Figure 2 represents an embryo of about twenty-one hours’ incubation. The area opaca, pellucida, and vasculosa are more elongated than in the normal blastoderm. The early formation of the vascular area with the sinus terminalis does not show any departure from the normal course of development. The blas- toderm shows a single primitive streak, terminating at its an- terior end with two Hensen’s nodes and head processes. The embryo anterior to the primitive streak is represented by an almost complete duplication of parts. The development of the head folds, the neural tubes, and the notochords correspond to the typical conditions. The head regions are distinctly inde- DUPLICITY IN CHICK EMBRYOS 359 pendent, but a fusion occurs in the somite region, where the middle paraxial mesoblast with its somites is common to both of the notochords. In figure 2, if development had continued until the end of the incubation period, no doubt an embryo with two distinct heads and neck regions with a common trunk would have resulted. ‘The blastoderm in figure 3 shows a well-developed vascular area, with a distinct sinus terminalis. The anterior vitelline veln, corresponding to either embryo, has begun its early devel- opment. With the exception of a very small part of the pos- terior end of the primitive streak, there is a complete duplication of structures on a common blastoderm. There is no means of 360 GEORGE W. TANNREUTHER determining from a study of the two embryos why one has reached a further stage in its development than the other. The formations in either embryo do not show any departure from the normal course of growth. The head folds in either case are distinct and show the fore gut well developed. The amnion and tail folds of either embryo have begun their early stages of development. In the early period of incubation the two embryos began as two primitive streaks with their posterior ends fused. The persistence of the fusion is well marked, as indicated in figure 3 (C.Pr.St.). ‘Judging from the independent formation of the two tail folds, as shown in the figure, two distinctly independent chick embryos would have resulted at the end of the incubation period. The size of the yolk and egg from which the blastoderm was taken corresponded to that of the average in the lot. If the progress of development in figure 3 can be taken as a safe criterion, we could readily foretell the resultant develop- ment in figure 1, where the posterior ends of the four primitive. streaks show a common connection. As in figure 3, an inde- pendent embryo would have resulted from each of the primitive streaks. Thus producing four individuals on a single blastoderm. The embryo represented in figure 4, no doubt began its devel- opment as two independent primitive streaks, with a later con- nection or fusion of the anterior ends of the two head processes. The embryo shows a complete duplication of the trunk region and the vitelline veins. The development of the head region is normal. It shows the differentiation of the anterior end of the neural tube into the primary brain vesicles and the early for- mation of the optic vesicles. The head fold is well developed and represents the demarcation of the head region from the blastoderm. The amnion shows the usual course in its early development. The heart shows the characteristic development of a thirty-two-hour embryo. The ventral aortae extend to the anterior end of the fore gut, where they continue as the aortic arches to the dorsal side of the gut and extend posteriorly, be- coming the two dorsal aortae. The ventral aortae, first pair of aortic arches, and the dorsal aortae are unusually well developed DUPLICITY IN CHICK EMBRYOS 361 and easily followed. There was no indication of the early devel- opment of a second pair of aortic arches. The two dorsal aortae are widely separated on the dorsal side of the fore gut, and on reaching the double trunk region, either aorta continues poste- riorly on its corresponding side. Thus either trunk region has but a single dorsal aorta. There were no indications of either dorsal aorta dividing in the trunk region. The vitelline veins are in their early stages of development. The anterior vitelline veins are present. There are four lateral vitelline veins. The anterior pair, right and left, are of the usual type as found in the normal embryo. They are well developed, as indicated in the figure. The accessory pair of vitelline veins extend anteriorly on the blastoderm between the two trunk regions and enter the sinus venosus on the median ventroposterior end. Thus, either 362 GEORGE W. TANNREUTHER trunk region has two vitelline veins, the normal number. The circulatory system taken in its entirety, with the exception of the vitelline veins and arteries, does not show any unusual developments. The fore gut, as shown in the figure, is a single closed tube anterior to the somite region, but at its posterior end the gut becomes divided into a right and a left posterior extension. This division is due to the presence of the median splanchnic folds. Two anterior intestinal portals instead of one is formed, as indicated in figure 4 (A.J.P.). Thus there is formed a gut cay- ity corresponding to either trunk region. A slightly later stage undoubtedly would show the median splanchnic folds much better. The notochord is duplicated in the trunk region, but continues anteriorly as a single structure. The primitive streaks and Hensen’s nodes are well developed. The anterior end of the neutral tube corresponds to the con- dition found in the normal embryo, but at the point where the division into a right and a left extension occurs, it is abnor- mally wide in a transverse plane. The accessory neural folds of the trunk region, which form the inner half of either neural tube, are continuous at the point where they form the posterior limit of the undivided part of the tube. The early stages of development in the tail-fold regions are well marked. The dotted outline shows the limit of the area pellucida. The em- bryo began as two independent primitive streaks, but complete development would have resulted in an individual with a single head, a single neck, and a double trunk. Figure 5 represents a very unusual abnormality. It would be rather difficult to conjecture the conditions as they occurred in the first stages of development. In all probability, the em- bryos began as two independent primitive streaks, with the anterior ends of the head processes continuous or in immediate contact. With the exception of the shape of the pellucida area, the blastoderm shows no departure from that found in the nor- mal individual. The sinus terminalis was well developed and showed its termination anteriorly into the right and left anterior vitelline veins, which were continuous with heart number (2). DUPLICITY IN CHICK EMBRYOS 363 The anterior ends of the neural tubes are condensed into a very small region. The brain (Ab.Br.) of either embryo is com- posed of a series of indefinite folds and rudimentary vesicles. There are no structures present that could be considered as optic vesicles or as any definitely marked formations. The neural tube in either trunk region is well developed and corre- sponds to the normal embryo of about forty-two hour’s incuba- tion. There is a distinct notochord in either trunk region. A common fore gut is present beneath the abnormal brain and is limited laterally by the splanchnic folds on either side of the head region (Hnt.Fo.). The fore gut opens in either direction and the anterior intestinal portal of either embryo is situated near the anterior end of either somite region. Thus the anterior end of either fore gut is a common continuous structure. The tail folds and hind gut in either case have just begun their devel- opment and agree with the conditions in the normal developing chick. 364 GEORGE W. TANNREUTHER The area vasculosa is unusually well developed. The upper side of the figure corresponds to the anterior end of the extra- embryonic blood system. The hearts (1) and (2) are normal in every respect, and possess a distinct well-developed truncus arteriosus (7’r.Ars.). Either truncus forks right and left and give rise to the ventral aortae (V.Ao.), which continue dorsally around the fore gut in either embryo and give rise to the dorsal aortae. The two dorsal aortae of either embryo are independ- ent, except at their anterior ends. The two hearts, when the blastoderm was removed from the egg, did not pulsate simul- taneously, but in a regular alternate order. The blood passed directly from the ventral aortae on either side through the aortic arches to the corresponding dorsal aortae, which continue independently in the trunk region. The two vitelline arteries in either embryo are distinct and show no unusual variations. Two well-developed vitelline veins are present on either side, which unite anteriorly to form the sinus venosus of the corre- sponding heart. Thus the vitelline veins of either heart pass out in the splanchnic folds of different anterior intestinal portals. Either heart, as represented in figure 5, no doubt begun as a distinctly independent formation, two endocardial tubes being present in the early development of either structure. This belief is further substantiated by the fact that two distinct vitelline veins are present in either case. A distinct common amnion (Am.) has begun its development, which partially covers heart (2). In all probability the amnion would have continued as a common structure for both abnormal embryos. One of the trunk regions is slightly larger than the other with a few extra somites. Figure 5 shows a complete duplication of the heart, vitelline veins, vitelline arteries, and trunk regions. End development of the embryos in figure 5 would have produced a very unusual chick monster. Figure 6 represents an embryo of sixty-eight hours’ incuba- tion. The blastoderm with its well-developed vascular area does not show any departure from that of the normal developing individual. The sinus terminalis with its two anterior vitelline veins is not represented. The brain shows considerable ab- DUPLICITY IN CHICK EMBRYOS 365 normal development. The primary brain vesicles are present, but they are very much distorted. The optic vesicles are unus- ually small and poorly developed. The crystalline lenses, if present, cannot be recognized in the whole mount. The audi- Spats alt STO tory vesicles are present and show the normal structures found in a sixty-eight-hour embryo. The various structures in the entire trunk region correspond to the normal condition“and need no further description. The intestinal groove (J.Gr.) is unusu- ally well developed and can readily be distinguished. 366 GEORGE W. TANNREUTHER There are two distinct hearts present, (1) and (2), as indicated in the figure. Judging from the single well-developed vitelline vein of either heart, the two hearts began as two parallel endo- cardial tubes, which failed to unite into a common tube, and later either independent tube resulted in a distinctly normal heart. Either heart is normal in every respect. The embryo, as in figure 5, showed an alternate pulsation of the two inde- pendent hearts. The ventral aortae and aortic arches on either side are distinctly independent and do not show any crossing over. The aortic arches extend dorsally around the fore gut to the dorsal aortae, which unite into a single vessel posteriorly. The aortic arches in heart (1) are of the usual type, and extend dorsally through the gill arches. Two gill slits, corresponding to the first and second are present. The aortic arches of heart (2) differ from those in (1). No gill slits are present. The vitelline veins of either heart continue forward from the blasto- derm in their corresponding splanchnic fold. Either heart projects unusually far into the extra coelom. There is no indication that the hearts will later be drawn into the body cavity. This condition is partially due to the poorly developed somatopleuric folds in the head region. The fore gut is abnormally wide and shows but two gill slits present on the right side. The hind gut and allantois are normal. Columbia, Mo. February 13, 1919 DUPLICITY IN CHICK EMBRYOS 367 LITERATURE CITED Autsop, F. M. 1919 Abnormal temperatures on chick embryos. Anat. Rec., vol, 15, no. 6. Gasser, Orro 1913 On the origin of double-yolked eggs. Biol. Bull., vol. 24, no. 3. ; MrrcHeLt 1890-91 On a double-chick embryo. Journ. of Anat. and Physiol., vol. 25, pp. 316-324. O’Donocuuz, C.H. 1910 Three examples of duplicity in chick embryos with a case of ovum in ovo. Anat. Anz., Jena, Bd. 37. Wurman, C.D. A rare form of the blastoderm of the chick and its bearing on the question of the formation of the vertebrate embryo. Jour. Micr. Soc., vol. 23. Resumen por el autor, Leo Carl Massopust. Universidad Marquette, Milwaukee. Un método simple para preparar vidrios de luz diurna para trabajos microscépicos. El vidrio empleado por el autor es blanco, esmerilado en una de sus superficies; el color un tubo de “azul permanente” al dleo. Se coloca un poco de color en una vasija, se afiaden unas cuantas gotas de trementina y se mezcla bien. Se toma un pedazo de gasa fina, y con él se envuelve un poco de algodén formando una mufiequilla. Se impregna esta con el color y se frota ligeramente sobre la superficie esmerilada del vidrio teni- endo cuidado de mantener el mismo tono. Cuando se ha aplicado el color la superficie del vidrio adquiere un aspecto granuloso o punteado, que produce la luz blanca. Por este método se obtiene una luz blanca brillante, aun mds blanca que la que se obtiene usando los vidrios productores de luz diurna que se encuentran en el comercio. Translation by José F. Nonidez Carnegie Institution of Washington AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, JULY 7 A SIMPLE METHOD OF PREPARING DAYLIGHT GLASS LEO C. MASSOPUST Marquette University, School of Medicine Owing to the almost universal use of daylight glass, a simple and inexpensive method of preparing daylight glass for micro- scopic work may prove of interest to the laboratory worker, particularly at the present time, when it is difficult to procure the commercial article. The glass used is white glass ground on one side; this may be procured from any glass dealer and cut to the required size. For the color a tube of permanent blue oil color can be obtained at any artist’s supply house. (We obtained the best results by | using pigment manufactured by the Devoe & Raynolds Co.) Some of the pigment is placed in a small dish, a few drops of turpentine added, and mixed so that the consistency is that of a soft paste. A small piece of finely meshed gauze is taken, a ~ small wad of cotton is placed on the gauze and gathered up at the ends so as to make a dauber (pounce). The dauber is dipped into the preparation described above, and dappled (not rubbed) on the ground surface of the glass. Care must be taken to main- tain an even tone over the entire surface. When this even tone has been produced, the surface will have a stippled effect, which effect accounts for the white light. If this preparation is applied as a flat tone, no matter how light the tone, the field will be blue. Therefore, it is very necessary to have some of the yellow light come through with the blue. A microscope and microscope lamp may be kept in readiness while preparing the glass to ascertain whether the field is too blue or too yellow. Should there be too much color on the glass, it may be removed by pressing a clean portion of the dauber firmly on the glass and then lifting it off in a vertical direction without any rubbing whatever. If there is not sufficient color another light coat of 369 370 DAYLIGHT GLASS the preparation may be applied until the proper density is obtained. The color is then allowed to dry and harden thor- oughly. We have obtained by this method a bright white light, even whiter than that obtained by the use of the com- mercial daylight glass. The glass so prepared has been used in our laboratories for several months with very satisfactory results. Examination of this glass by means of the comparison spectroscope shows little difference between glass prepared in the laboratory according to this method and the daylight glass obtained from dealers. AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, JULY 7 BIRTH OF TWO UNEQUALLY DEVELOPED CAT FETUSES (FELIS DOMESTICA) HARRISON R. HUNT Department of Zoology, West Virginia University TWO FIGURES In April; 1918, while conductmg some experiments: with: the domestic cat on the effects of inhaling chlorine gas, Dr. William H. Schultz, at that time head of the Department of Pharma- cology of the Medical School of West Virginia University, observed an interesting case. He very kindly presented the material and data to the writer, who was assisting him in the investigation. Shortly after 11 a.m. on April 6, 1918, a pregnant cat was placed in a closed container where it breathed chlorinated air for twelve minutes. It was then removed and put back in its cage, looking very sick. With Doctor Schultz’s permission, the following is quoted from his note book. Apr. 7, 1918. 12.00. At about this time I observed the cat clean- ing the vulva and eating a small bladder. Close examination revealed a small embryo. The gassing had caused violent contraetion of the uterus. 3 P.M. This cat delivered at least one full term kitten with hair, which I collected along with the placenta. Whether the cat ate other Praia or not I can not say. Etherized it and placed in Bouin’s uid. The smaller fetus was fixed in the same way. The facts certainly suggest that the inhalation of the chlorine had caused the cat to abort, though the larger fetus was so mature that it might have been born on the 7th in the absence of the chlorine treatment. According to the writer’s recollection, the fetal membranes of the smaller fetus were much torn, probably by the mother cat’s 371 372 HARRISON R. HUNT teeth. Unfortunately, they were not saved for histological study. Figure 1 is an accurate outline of the fetuses made from a photograph taken by Dr. A. M. Reése. The outline shows accurately the relative sizes of the two. The crown-rump measurements were 105 mm. and 14.1 mm., so that the larger fetus was about seven and a half times as long as the smaller. The former was beyond doubt practically a full-term kitten. Figure 2 shows the smaller fetus in greater detail. The mouth, eyes, tail, external ears, and limbs were well developed, and the olfactory pits were visible. The digits did not show externally. The smaller fetus was compared with two normal cat fetuses, each 17.5 mm. long, the only ones of comparable age that could be secured. In these normal fetuses the digits were plainly visible, the nasal region had begun to protrude, and the eye and external ear were somewhat more developed than in the aborted fetus. Otherwise, the normal ones resembled the aborted fetus rather closely, except that the latter’s mouth was open, showing the tongue, and the chin did not touch the breast as in the nor- mal fetuses. Probably the mother’s attempt to eat the fetus pulled the head backward and forced the mouth open. In ex- ternal appearance this fetus certainly appeared to be normal. TWO UNEQUALLY DEVELOPED CAT FETUSES 373 It was impossible to determine accurately the age of the smaller fetus, for illustrations of cat embryos of different ages were not available. When compared with His’ figures of normal human embryos (Keibel and Mall, vol. 1, p. 61), the smaller fetus was seen to correspond approximately, judging from its external appearance, with a forty-day human fetus. Such a fetus has passed through about one-seventh of its period of development. Since the period of gestation of the cat is about eight weeks, it would seem (though this estimate is necessarily a rough one) that the smaller cat fetus was in the second week of development. A study of serial sections of the smaller fetus showed the following conditions. The cerebral hemispheres and body cavity had collapsed. The linings of the hemispheres and stomach were badly fragmented. In the lumbar and sacral regions a longitudinal ragged crack on the dorsal side of the body extended down into the spinal cord, disorganizing it ex- tensively. In the neck region the cord appeared to have been slightly twisted on its long axis. The auricular walls of the heart were much cracked and fragmented. Many of the kid- ney tubules were normal. The walls of other tubules were © more or less broken up. The outlines of the organs could be clearly made out. There was evidence of nuclear fragmenta- tion, especially near the surface of the body, but for the most part the nuclei were normal in appearance. Considering that the mother attempted to eat the fetus, it is not surprising that its delicate tissue showed mechanical injury. Probably the mother seized the ovum in such a way that pres- sure was brought to bear on the sides of the body, causing the coelom, auricles, cerebral hemispheres, and stomach to collapse. The resulting impact of the opposite walls of these cavities probably fragmented the walls as described above. The time which elapsed between the birth of the fetus and its immersion in the fixing fluid is not known. Possibly the nuclear fragmen- tation and the disintegration of some of the kidney tubules were due to postmortem autolytic processes following its expulsion from the uterus. 374 HARRISON R. HUNT The difference in the degree of development of the two fetuses must have been due to one of the four following causes: 1. Both ova may, conceivably, have been fertilized at about the same time, one dying and remaining in the uterus for five or six weeks while the other continued its normal development. Dead human fetuses have been known to remain for days in the uterus without undergoing extensive changes (De Lee, ’15). But frequently they undergo liquefaction, maceration, saponi- fication, mummification, putrefaction, ete. (De Lee, ’15; Amer- ican Text-book of Obstetrics, 96; Edgar, 03). Stockard and Papanicolaou (718) have observed degenerative changes in guinea-pig embryos. It may not be possible to demonstrate beyond all doubt that the difference between the two fetuses was due to some cause other than the death of the smaller fetus. But it seems probable that a dead fetus would have shown some gross external evidence of dissolution within five or six weeks after death. This fetus certainly showed no such changes. 2. Another possibility is that pathological processes appeared in one of the fetuses during the first week or so, retarding its development. Of the one hundred pathological human embryos studied by Mall (08), representing stages of development from two weeks on, only 8 per cent, or one-twelfth, had a normal externalappearance. If pathological processes retarded the devel- opment of the smaller cat fetus, they must have appeared early, possibly in the first or second week. Eleven out of twelve of Mall’s embryos were abnormal in external appearance. There- fore, it is highly improbable that pathological changes extensive enough to retard development so decidedly in the smaller fetus could have acted for several weeks without causing gross ex- ternal abnormalities. 3. It is conceivable that the smaller fetus developed more ~ slowly than its litter mate as a result of malnutrition. Unfor- tunately, the blood supply of the uterus was not available for study. However, the observations of Mall show that malnu- trition is one of the chief causes of pathological changes in the developing ovum. ‘‘It is no longer necessary,” says Mall, ‘‘for us to seek for mechanical obstructions which may compress the TWO UNEQUALLY DEVELOPED CAT FETUSES 375 umbilical cord, such as amniotic bands, for it is now clear that the impairment of nutrition which naturally follows faulty implantation, or the various poisons which may be in a diseased. uterus, can do the whole mischief” (Mall, ’10, p. 240). Since no pathological changes were observed in this fetus, the hy- pothesis of malnutrition is certainly not supported by the facts. 4. Probably the most satisfactory explanation is that the eggs from which the two fetuses developed were fertilized several weeks apart. They may have belonged to the same period of ovulation, but were fertilized by spermatozoa introduced by separate coitions (superfecundation); or ovulation, followed by fertilization, may have occurred during pregnancy (superfeta- tion). In the following discussion these definitions of the terms are used. Cristopher (’86) believes that the cat can ovulate during pregnancy. Jepson (’83) reports the case of a cat which gave birth to an immature fetus in the same litter with two full-term kittens—possibly an instance of superfetation. Harman (717) has reported a cat whose uterus contained three fetuses which “were developed near to term, and one apparently was much smaller, and showed a much less degree of development.’’ She applies the term superfetation to this condition. Harman (18) describes also a probable case of superfetation in the cow. King (713) concludes that superfetation and superfecundation occasionally occur in the albino rat. Both of these phenomena have been observed in man (Scott, 717). Sumner (’16) believes that mouse eggs ovulated during preg- nancy may be fertilized, one period of gestation being thereby imposed upon another. The writer is unable to decide between superfetation and superfecundation as the true explanation of this case, though probably one or the other is the real explanation. If it be the latter, the two ova belonged to the same period of ovulation, but one was soon fertilized, while the other was fertilized several weeks after ovulation. Whether cat ova can retain their vitality this long is an open question, though Sumner’s observations (16) led him to believe that the spermatozoa of mice may retain their fertilizing power for days or weeks. 376 HARRISON R. HUNT In the cat reported by Harman (’17) one of the more advanced fetuses lay between the smaller fetus and the ovary. A nor- mally implanted ovum like the former might offer an impassable barrier to an egg ovulated during pregnancy. She suggests, therefore, that all the ova belonged to the same period of ovula- tion, but that one of them was fertilized later than the others. In the case under discussion, however, the observed facts do not rule out the possibility of superfetation. Possibly the older fetus occupied but one horn of the uterus, the spermatozoa thereby having access to eggs ovulated during pregnancy. Longley (’10, ’11) observed in the cat that ovulation and maturation of the egg depend upon copulation. This suggests that coition during pregnancy may have led to the liberation (Cristopher, ’86) and fertilization of the ovum from which the smaller fetus developed. j The probabilities certainly are that the case described is either one of superfecundation, in which the fertilization of the egg was delayed for weeks, or a case of superfetation. oN) ~I a | TWO UNEQUALLY DEVELOPED CAT FETUSES LITERATURE CITED CurisTorHeR, W. S. 1886 Ovulation during pregnancy. Amer. Jour. of Ob- stetrics, vol. 19, pp. 457-467. De Lez, J.B. 1915 The principles and practice of obstetrics. W. B. Saunders Co. Epcar, J. C. 1903 The practice of obstetrics. P. Blakiston’s Son & Co. Harman, M.T. 1917 A case of superfetation in the cat. Anat. Rec., vol. 13, no. 3, pp. 145-153. 1918. A probable case of superfetation in the cow. Anat. Rec., vol. 14, no. 5, p. 335. ° Jepson, 8. L. 1883 A case of superfetation in a cat. Am. Jour. of Obstetrics, vol. 16, p. 1056. -Kerset anp Matt 1910 Manual of human embryology. J. B. Lippincott Co. Kine, H. D. 1913 Some anomalies in the gestation of the albino rat (Mus norvegicus albinus). Biol. Bull., vol. 24, no. 6, pp. 377-391. Lonetey, W. H. 1910 Factors which influence the maturation of the egg and ovulation in the domestic cat. Science, N.S., vol. 31, p. 465. 1911 The maturation of the egg and ovulation in the domestic cat. Am. Jour. Anat., vol. 12, no. 2, pp. 139-172. Matt, F. P. 1908 A study of the causes underlying the origin of human mon- sters. (Third contribution to the study of the pathology of human embryos.) Jour. Morph., vol. 19, no. 1, pp. 3-367. Norris, R.C. 1896 An American text-book of obstetrics. W.B. Saunders Co. Scorr, R. J. E. 1917 Superfetation. Reference handbook of the medical sciences, vol. 8. Wm. Wood Co. Srocxarp, C. R., anp Papaniconaov, G. N. 1918 Further studies on the modification of the germ-cells in mammals. The effect of alcohol on’ treated guinea-pigs and their descendants. Jour. Exp. Zoél., vol. 26, no. 1, pp. 119-226. Scuner, F. B. 1916 Notes on superfetation and deferred fertilization among mice. Biol Bull., vol. 30, no. 4, pp. 271-285. Resumen por el autor, Edgar F. Cyriax. Londres. Nota sobre la clavicula ‘‘flotante.”’ El nombre de clavicula “‘flotante”’ se aplica a una estructura no rara, aparentemente adquirida, en la cual el extremo interno de Ja clavicula en vez de reposar sobre la faceta articular del esternon formando con ella una articulaci6én, est’ completamente libre y es movil. La amplitud de movimiento en diversas direcciones es, en general, préximamente de un cuarto de pulgada. Esta estructura se ha observado particularmente en el lado izquierdo. No se descubrieron sintomas subjetivos. Translation by José F. Nonidez Carnegie Institution of Washington AUTHOR'S ABSTRACT OF THIS PAPER ISSUED BY THE BIBLIOGRAPHIC SERVICE, JOLY 7 A BRIEF NOTE ON “FLOATING” CLAVICLE EDGAR F. CYRIAX London I have ventured to apply the name of ‘‘floating” clavicle to a condition which as far as I know has up to the present not been described, namely, one in which the inner end of the clavicle instead of constantly impinging upon and forming a joint with the articular facet on the sternum, lies quite free and moveable. I am inclined to think that the condition is not so uncommon, and that the reason why it passes unnoticed is simply because it is not looked for, inasmuch as it apparently produces no subjective symptoms; in the cases I have seen the patient himself was not even aware of the abnormality until it was pointed out to him. My attention was first drawn to the condition of floating clavicle about twelve months ago, when I discovered it while examining the joints in a case of articular rheumatism of the left arm; the. left sterno-clavicular joint was the one that was involved. Six months later I found another floating clavicle, also on the left side, in a girl of ten who stammered, and recently I found a third case again also on the left side, in an aviator who in- formed me that he had once had phthisis pulmonum some years previously. In none of these cases were there any subjective symptoms whatever. They all showed practically the same objective ones, as follows: The inner end of the clavicle was free and was not in contact with the sternum; when the arm was dependent it oc- cupied a position about one quarter of an inch above its fellow on the opposite side. On grasping the sternal end of the clavicle between the forefinger and thumb it was found that it could perfectly easily be moved downwards onto the articular facet on the sternum, and from there with equal facility upwards, inwards, outwards, backwards and forwards. The range of all 379 380 EDGAR F. CYRIAX these movements was about the same, namely approximately one quarter inch; the final stage of each movement was somewhat abruptly limited, suggesting that this was done by means of ligaments rather than muscles. The only one of these move- ments that had any result was the backwards one which caused some irritation of the larynx; none of the others caused even a passing inconvenience, although the inwards movement seemed to exercise a fair amount of pressure on the sterno-mastoid, causing it to bulge locally. On moving the clavicle away from the sternum, palpation of both articular surfaces could be done with fair accuracy; as far as could be judged both were normal as regards surface and outline. This leads me to suppose that the condition is ac- quired rather than congenital. In none of the subjects examined did the condition of the clavi- cle seem to have any effect on the strength of the muscles or the range of the movements of the shoulder girdle. On endeavour- ing to replace the clavicle into its joint, this as stated above could readily be done, but displacement occurred as soon as the patient attempted movements which threw any strain on the joint, even when these movements were executed only through. a very small range. In two of the cases mentioned medico- gymnastic treatment was applied in order to try and effect a per- manent reposition of the joint, but in both cases no improvement at all resulted. In none of the three cases was I able to find any evidence that the floating clavicle was responsible for any of the symptoms, excepting perhaps in the patient who stammered. Her chief impediment was spasmodic action of the diaphragm, and it is not impossible that the free movements of the sternal end of the clavicle may have acted as a continued source of irritation to the phrenic nerve. In her case however pressure of the bone backwards did not aggravate the condition. 47 ne > - crf oe, «6 ty “? ‘ - ‘ a) i 4 . “ 7 é i ’ hy ' K ; ey, : e t s ~~ A =% . : aor oo err 1a AE Goh PEE cere QL The Anatomical record Biologicat : & Medical Serial ~ PLEASE DO NOT REMOVE CARDS OR SLIPS FROM THIS POCKET EE a ee UNIVERSITY OF TORONTO LIBRARY 2 a STC: ‘ eyelet) - ~~ Std ey Tee ig sey hn ore trate Upee bet eee rei ee seprene sera > Laat vey ty “¥ ith wyitesee a: nat } Fara a hears ns yur Hatt, ire ites is Ze " = od ite ot t + 2 Be a fis EE has ie oH piticme yh ies Aries a et ’ er . seers CAR) : ty piace eae ans ae “¢ 5 re ” a4 if « 4 exe Dae The ~ 2 ie est Sees 3 ew. ppt nee: ao a Suet oe si ao pian AR hie A ily ti te i eras iW My vee o| ‘tae ! Hat te toby Baty ett phe Leahey tat aia Lae fit as ifeeeteth yet Burge i ak mili rt ) Hf tii set re aia HU Tae mie Pipe atin sit ee arenes on aii La ye ge HACER MAM ie in 7 aay ct ie tie ys \ yyy igh wy bibs Prete’ Nee Liters \ ai titi ry ' py ‘ HUN HE Torn Dautvaty: Hal REI or : att i tie baieatt ae eieyegey ‘ y a tpalt ad wharh ¥ Pte aeeteacenttars deers trie] ves Kt Betbear ie i bat ie ae aya it + i ; veer ke! hr Anite ft feat hey rake ratseee Pesacrs domwoets trent 1 ve yO Ew Tehgn ir 4 a lity ite Ns 7 tye a ae es is ctal : +