Talk:Paper - Experimental studies on the origin of monsters 2 (1917)

407

author's abstract of this paper issued by the bibliographic service november 10.

EXPERIMENTAL STUDIES ON THE ORIGIN OF MONSTERS 1

II. REGARDING THE MORPHOGENESIS OF DUPLICITIES 2 E. I. WERBER

From the Osborn Zoological Laboratory, Yale University

TWENTY-SEVEN FIGURES

INTRODUCTION

In a recent publication (Werber, '16 b) I have attempted an etiology of monstrous development and an analysis of the morphogenetic factors underlying it. The observations there recorded and the theoretical conclusions based on them pertained largely to terata of the head, but some other experimentally produced forms of pathological development have also been sufficiently considered.*

The .present paper is a continuation of this work. Its primary purpose is to put on record some duplicities which have resulted from the employment of a chemical method. The attempt is also made to account for their morphogenesis although I am fully aware of certain possible inadequacies of our

1 This contribution is a part of the work carried on with the aid of a grant from the Bache Fund of the National Academy of Sciences in 1915.

2 The term 'duplicity' has, unfortunately, lost its original meaning and come to be employed almost exclusively in its metaphoric sense. In view of the lack of an other suitable English term I propose henceforth in dealing with monozygotic double embryos to employ it in its original meaning which I believe to be proper and precise, being the equivalent of the Latin 'duplicitas' and the German 'Doppelbildung.' The commonly employed term 'double monsters' ('Doppelmissbildung') can be employed only for deformed conjoined double embryos.

3 In a very recent paper Newman (Biological Bulletin, vol 32, No. 5, 1917) makes the statement that I failed to account for such teratomata as the 'isolated eye' (or the 'solitary eye'). In reply I refer the reader to pp. 541-552 of my 1916 b paper where these teratomata, for the first time experimentally produced by myself, are described and their morphogenesis fully accounted for.

409

410 E. I. WERBER

interpretation which may be largely due to the insufficiency of data at command at the present time.

I also appreciate the justification of an objection often raised against such attempts, namely that deductions regarding the morphogenesis of monsters are uncertain, if based only on the analysis of the end-products of such atypical development. 4 It is, of course, obvious that certainty could here be gained only by direct observation in vivo from the beginning of development through successive stages. However, this direct method presents great difficulties in view of which it seems necessary to take into account the inferential clues often furnished by the analysis of already developed monsters. 5 The indirect evidence thus gained is in accord with data established by direct observation of successive stages during atypical development in many experiments in the lower animals.

BLASTOLYSIS AS A MORPHOGENETIC FACTOR IN THE DEVELOPMENT OF DOUBLE MONSTERS

"While attempting to account for the morphogenesis of 'monstra per defectum' I (I.e.) have shown that in many of the embryos of which a microscopical study was made there was unquestionable evidence of either dissociation of tissues or of dislocation of organs or of both. From these observations 6 the conclusion was drawn that we are here, manifestly, confronted with some action (or actions) that tended to disintegrate and to dissociate parts of the earliest embryonic primordium. As the chief components of this complex process (blastolysis)

4 These objections were pointed out already by Rauber ('79-'80).

5 Every teratological experiment in which, like the present, a chemical modification of the environment is employed, must be carried out on a great many eggs, since their mortality is very high. This greatly increases the difficulty of •continuous observation of the still surviving eggs of which probably no two develop alike. Besides, the chemical processes and the alterations they are responsible for, elude direct observation. This is particularly true for the translucent egg of Fundulus.

6 Similar observations were made by Mall ('08) who, however, dealing with material not coming from experiments, did not attempt to account for the factors underlying this shifting and dissociation of tissues which he correctly pronounced as 'histolysis.'

ORIGIN OF MONSTERS. II 411

I regard destruction of parts of the primordium by chemical action and fragmentation resulting from the latter as well as from differences between the osmotic pressure in the eggs and that of their surrounding, experimentally modified, medium.

These conclusions can, I think, be extended to the morphogenesis of 'monstra per excessum.' For Driesch ('93) has demonstrated for the sea-urchin and Wilson ('93) for Branchiostoma that the development of various 'monozygotic' 7 double or multiple embryos can be induced by the employment during the first or the second cleavage of a dissociating force such as raising the temperature or shaking the eggs, which would separate the blastomeres. Similar results were obtained also by Fischel ('98) by mechanical pressure exerted on the eggs of the ctenophore Beroe ovata. It was further shown by Loeb ('95) that like results can be obtained in sea-urchin eggs transferred soon after fertilization to a medium of lowered osmotic pressure.

Of particular interest for our consideration is also the work of Bataillon ('01) who produced duplicities experimentally in lower vertebrates (Petromyzon and the teleost Leuciscus rutilus) by increasing the osmotic pressure of the surrounding medium. Employing this method Bataillon was able to ascertain by direct observation in Petromyzon that the development of monozygotic twins and various other duplicities resulted from the more or less complete separation of the blastomeres of the two-celled stage. He also states that in Petromyzon at least one-third of the egg seems to be necessary for the development of a whole embryo.

This equi- and totipotency of the first blastomeres was demonstrated also in the amphibians. Thus O. Schultze ('94) and Wetzel ('95 and '96) have succeeded in producing conjoined double embryos in Rana fusca by inverting for some length of time eggs in the two-celled stage, previously compressed between two glass plates.

Even complete monozygotic twins (of half the normal size may be obtained in amphibians by separating the blastomeres, as shown by the ingenious experiments of Herlitzka ('95 and

7 This very precise term is adopted from Newman ('17).

412 E. I. WERBER

'97) on the eggs of Triton cristatus. These interesting results were subsequently supplemented by the important experiments of Spemann ('00-04) who produced various duplicities in Triton taeniatus by incomplete constriction of the eggs along the first cleavage plane or along the corresponding plane at any stage up to the gastrula.

Bearing in mind these data as well as also Morgan's much earlier ('93) discovery of the totipotency of the blastomeres after the first cleavage in the teleost egg (Fundulus) it would seem perfectly safe to assume that the duplicities recorded in our experimeuts (on the eggs of the same species) have resulted from (primarily osmotic) dissociation ('blastotomy' — Bataillon,

I.e.)- 8

The origin of the double embryonic anlage can, in this way, be accounted for without difficulty — at least in anamniotes. So far as this point alone is concerned it would seem no longer necessary to resort to such assumptions as ' dichotomous growth' at the anterior end of the embryonic anlage (Gerlach, '82), or the oft-assumed binucleate eggs and coincidental polyspermy as the agent causing duplication, or the 'radiation'-theory of Rauber ('79-80) and its modified offshoot, the 'double gastrulation'-theory of 0. Hertwig ('92, '06).

The inadequacies of all these hypotheses have been clearly pointed out by Fischel ('02) and in full agreement with the latter also by Schwalbe ('07). To Fischel we also owe what may be regarded as the most rational analysis of the manner in which the various combinations of duplicities in teleosts may be formed.

Fischel's considerations are based primarily on the assumption of a duplication of the germinal anlage soon after fertilization probably by abnormal osmotic pressure, as demonstrated by Bataillon (I.e.). Starting thus with two embryonic primordia and bearing in mind the mode of formation of the embryo as established (particularly for the teleosts) by several authors and notably by Kopsch ('96) Fischel, by employing the dia 8 This assumption might reasonably be extended to the double embryos often recorded in fish hatcheries (notably in Salmonidae), where an abnormal osmotic pressure may be due to temporary impurities of the water.

ORIGIN OF MONSTERS. II 413

grams of the latter demonstrates with almost mathematical precision how the development of any diplopagus may result from the variation in the degree of approximation and the angle of incline of two embryonic primordia. 9 It is not practicable at this place adequately to present Fischel's views. They have, however, been fully discussed by Schwalbe (I.e.) to whom as well as to Fischel's paper the reader must be referred. Here it may suffice to say that his analysis accounts successfully for the morphogenesis of all double monsters, some of which (like the diprosopos, the dicephalos, the catadidymus and the mesocatadidymus) have eluded all other attempts at interpretation.

However, it must be admitted that absolute proof for all of Fischel's postulates is wanting for vertebrates higher than Petromyzon, since Bataillon. unfortunately, made no observations regarding the morphogenesis of the double embryos he produced by osmotic pressure in Leuciscus. Nor is such proof furnished by the experiments on amphibians (referred to above), as they demonstrate only how certain duplicities may result from the separation of equipotent parts of the egg, but give us no clue regarding the formation of various other duplicities (in other vertebrates) by secondary fusion of duplicated primordia, as assumed by Fischel.

It may also be mentioned, in passing, that Fischel's theory (blastotomy by osmotic pressure and subsequent chance configuration of the duplicated primordia) does not account for the malformations often exhibited by duplicities. The point is not an unimportant one, for as I shall attempt to show, these deformities may often be syngenetic with the doubling of the primordium. Besides, as will be pointed out later, the deformities of one or both components of a duplicity give us an important clue to the morphogenesis of the 'parasitic double monsters.'

9 The various possib.le combinations in man were portrayed very suggestively in the well known diagrams of Wilder ('04) who, however, in his more recent worjv ('08) has abandoned his former views on diplogenesis in favor of his present 'theory of Cosmobia.'

The deformities observed in duplicities are of the same nature as those of other monsters, i.e., they are, usually, due to some defects and, sometimes, 10 to an inhibition in development.

It is relatively easy to account for the genesis of defects as the result of lesions produced by osmotic pressure. 11 The latter, however, will not account for an inhibition of development such as may be found in some monsters and, occasionally, in one or both components of a duplicity. As I have pointed out elsewhere (Werber, '15 and 16 b) for this form of deviation from the ontogenetic norm we must assume chemical alteration as responsible for the decrease of the germ's inherent capacity for development and differentiation. This assumption will account for such deformities observed in duplicities as may properly be regarded as due to an inhibition. 12

Moreover, chemical action may play a much greater part in contributing to the genesis of defects by either directly destroying certain parts of the germ or by lowering their resistance to the action of osmotic pressure. The combined action of these factors (osmotic pressure and chemical alteration) may result in any conceivable monstrosity depending entirely upon its degree and upon the part of the germ which has suffered most from it. 13

Keeping this in mind we shall be able to understand how not only (variously malformed) duplicities may result from the

10 Rarely.

11 This assumption is justified in view of the effects of osmotic pressure directly observed by Loeb (I. c.) in the sea-urchin. It should also be noted thatLoeb's method — decrease of external osmotic pressure by diluting the sea-water — excludes altogether the possibility of chemical action.

12 In previous papers I ('15 and '16 b) have defined inhibition as a decrease of the germ's originally inherent 'chemomorphic' potentiality. This conception is based on the exceedingly suggestive hypothesis advanced in the last two or three decades by a number of biochemical authors and most recently presented in a very succinct and attractive form by Reichert ('14), according to which the development of an organism is a complex series of reactions in a stereochemical system. Reichert's paper came to my attention through the kindness of Prof. H. V. Wilson after the completion of the manuscript for the present paper.

13 The theory of blastolysis was presented rather exhaustively in a former paper (Werber '16b). Owing, however, to misrepresentations which it suffered at the hands of certain authors the reader will; I hope, find that a repetition of its salient points in this paper is justified.

ORIGIN OF MONSTERS. II 415

separation of blastomeres, but also how such monsters may come about through dissociation at a later stage (up to the time of gastrulation) as in the experiments of Spemann ('00'04) where such dissociation was effected by mechanical constriction along the potential embryo's longitudinal axis. It is only necessary to imagine that a chemical lesion of a moderate degree is sustained more or less along the germ's chief axis. Along this chemically altered area a rupture may take place owing to least resistance to osmotic pressure. Depending upon the extent of this passive 'fission/ a diprosopos, dicephalos or any diplopagus may result from the further development of the, thus partially, doubled primordium. The assumption of this chemical action, made already by Bataillon (I.e.), 14 does not necessarily apply to all duplicities of 'Nature' and not even to all double embryos in teleosts, for as I have already mentioned, the duplication of the embryonic anlage at any stage before the conclusion of gastrulation might be accomplished by the physical action of osmotic pressure only. And even the malformations of the components of the resulting duplicities might well be due to this factor alone. 15

In my experiments, however, substances have been added to the sea water which, owing to their chemical properties are toxic (i.e., injurious to the most important life processes), and which also, owing to their molecular weights have changed the tonicity of the sea-water. It is therefore not surprising

14 Bataillon (I.e.) also assumes that in his experiments chemical alteration may have played a considerable part in the genesis of defects.

15 As an example I am inclined to regard the frequently observed Salmonid double monsters. In a considerable number of trout double embryos in my possession (collected in a fish hatchery in the vicinity of Freiburg i. B.) examination of sections fails to disclose anything that would suggest chemical alteration as underlying the duplication of the germ and the accompanying deformities of the resulting double embryos. The nature of the alteration of the water in which these embryos have developed is, of course, unknown and the factors responsible for their developmental anomalies are, undoubtedly, left altogether to speculation. It seems not improbable, however, that in such cases the osmotic pressure of the water may be increased by the presence in it of some (otherwise perhaps not injurious) metallic salts, or possibly blastotomy here results from increased temperature or "old age" of the eggs (cf. pp. 319-321 and 327).

416 E. I. WERBER

that on examination of sections of some double monsters resulting from these experiments conditions are found which suggest that the duplication of the embryonic anlage was due to dissociation by the combined action of both chemical alteration and osmotic pressure (blastolysis) .

The following cases may serve as illustrative examples of the apparent effects of such action.

In figure 11 is presented a 'parasitic' duplicity which in toto made the impression of a greatly deformed craniothoracopagus. Examination of sections practically confirms this diagnosis, except for the incompleteness in the duplication of the head.

The transverse sections (6 m in thickness) being somewhat oblique, those farthest anteriorly contain more of the right than of the left side of the duplicity. A large lens is noted in them, an olfactory pit and a distorted, unilobed fore-brain. The lens (but no eye on this side) was observed already in toto and was mistaken for a 'free lens.' In further sections, however, there comes into view in apposition to it a very small optic cup (fig. 14). A very small, lens-like body (I) can also be observed in several sections at this level between the ill-differentiated optic cup and the large lens. On following the sections posteriorly the eye of the left side gradually comes into view (figs. 14 and 15), while the optic cup of the right side becomes more and more elongate and is noted still to be in direct connection with the brain, no optic stalk having been formed. In sections still further posteriorly this eye is very plainly seen to be broken up (dissociated) into several parts — one of which at the base of the brain with which it is connected, has made a feeble attempt towards differentiation of the retinal layers,

Fig. 1 (a and b) Anterior and posterior views of slightly conjoined twins, A the larger, B the smaller one, From acetone solution (40 cc. gram-molecular to 50 cc. of sea-water), 19 days old, h., heart, pc, pericardial vesicle.

Fig. 2 (a and b) Two views of slightly conjoined double embryo, A the larger, B the smaller one. From butyric acid solution (10 cc. of a 1/16 gram-molecular solution added to 50 cc. of sea-water), 13 days old, pc, pericardial vesicle.

Fig. 3 From acetone solution (35 cc. gram-molecular tooOcc. of sea-water), 14 days old.

Fig. 4 From acetone solution (35 cc. gram-molecular to 50 cc. of sea-water), 14 days old.

ORIGIN OF MONSTERS. II

417

418 E. I. WERBER

rods and cones being discernible (fig. 15). At this level the brain has increased in size and presents a very distorted mass of nervous tissue in which cellular areas predominate. Still further posteriorly the buried optic cup of the right side is observed to gradually 'reconstitute' itself. It is very large at this level and its structure still betrays evidence of dissociation (ox. fig. 16). The brain increases in size with every successive section (fig. 17). Some parts of it give the impression of retinal tissue, but no certainty can be felt regarding the latter point.

The duplication of the brain becomes distinct at a level where the (dissociated) part of the larger component is on one side intimately fused with the medulla of the smaller component ('parasite' — fig. 18). Two notochords and two (incomplete) alimentary tracts are also noted on the monster's right and left side respectively (figs. 17 and 18). Sections passing through the larger component's tail show distinctly (fig. 19) that the partial doubling of the central nervous system has come about through a splitting of one anlage by dissociation.

A clue to the genesis of the described conditions is found in anterior sections. Here the epithelium is dissociated (d.e., fig. 14) in the region of the mouth and in a part of the blastoderm. The single oral cavity of the double monster is not continuous, but anteriorly partly occluded by the dissociated epithelium of the mouth (d.m., fig. 15) and posteriorly by the dissociated posterior part of the optic cup, while posteriorly to the latter the lumen of the mouth appears as a narrow slit, being 'plugged' by dissociated epithelium. The latter as well as all of the epithelium of the mouth consists of large, vesicular cells.

Turning, now, to a consideration of the mode of formation of this monster, it would appear improbable that it has arisen from separated blastomeres and subsequent fusion during the formation of the embryonic bodies. The condition of the nervous system described above would seem to indicate clearly that at first one embryonic anlage has existed until probably about the time of formation of the embryonic body when the greater part has apparently been doubled by dissociation.

ORIGIN OF MONSTERS. II 419

As an other case where the duplication has apparently occurred at a late stage (during the formation of the embryonic body) may be regarded the embryo (mesocatadidymus?) presented in figure 13.

This is a monster to which I attach much significance, because it most strikingly suggests the syngenesis of 'defects' with 'excess.' For its anterior part is highly deformed and defective while the greater part of the trunk and the tail — the former only partly — are duplicated, the spinal cord and notochord being multiple.

Examination of sections shows these conditions in a very convincing manner.

With the description of the defects of the head we shall not concern ourselves here, as it has a special bearing on the origin and differentiation of the lens and is therefore reserved for another paper. Suffice it to state, however, that the central nervous system is very defective throughout for about the anterior third of the body, i.e., to about the level of a plane passing transversely through the posterior part of the (only) pectoral fin. Just about from this level onwards posteriorly the spinal cord, in successive transverse sections, is observed at first to flatten and then to spread out more and more from side to side. It gradually increases in size and in many consecutive sections appears to be vacuolated and to exhibit evidence of cytolytic degeneration (fig. 20). The latter now becomes more and more conspicuous with each consecutive section (fig. 21), cellular elements disappearing more and more, until in some sections not much more than the framework of supporting tissue is left. The nuclear debris of the chromatolyzed cells can be observed at quite a distance from this degenerated part of the cord.

This has then, evidently, been the part which suffered the greatest injury from chemical alteration and subsequently presented a point of least resistance to the action of osmotic pressure. For from this level on the spinal cord divides at first into two and then into more parts. The notochord is also multiplied, while in sections at a still more posterior level the

THE JOURNAL OF EXPERIMENTAL ZOOLOG1 , VOL. 24, NO. 2

420 E. I. WERBER

body of the embryo is observed also externally to be partly duplicated (figs. 21 and 22). The musculature is not doubled; yet there is a decided increase of it in bulk. The alimentary tract is also single and even incomplete, some of its parts lacking altogether. In sections through about the posterior fourth of the embryo there appear just about the level where the intestine ends, two large cavities lined with endothelium whose genesis and morphological significance I am unable to interpret.

The genesis of the chief morphological deviations of this embryo might be imagined in the following manner:

The part of the embryo which was to form the head has sustained the greatest destructive alterations from the sojourn in the toxic solution. Hence the extreme malformation and the defects of this part of the body. The bilateral parts of the germring have apparently — owing to osmotic pressure — failed at first to fuse completely, this fusion occurring only later when some anlagen have, in this way, been doubled. The increase in osmotic pressure has also most likely contributed to the fragmentation and dispersion of some anlagen, as is evidenced by the condition of the notochord and the spinal cord. 16

While the embryo is not a double monster in the strict sense of the word, it well illustrates the manner in which — according to Fischel' s suggestion — certain double embryos may arise during the formation of the embryonic body. It would also seem to offer strong support to our contention that the same

Fig. 5 From acetone solution (35 cc. gram-molecular to 50 cc. of sea-water), 14 days old.

Fig. 6 (a and b) From acetone solution (35 cc. gram-molecular to 50 cc. of sea-water), 16 days old; i.e., 'isolated eye.'

Fig. 7 From acetone solution (35 cc. gram-molecular to 50 cc. of sea-water), 14 days old.

Fig. 8 From acetone solution (35 cc. gram-molecular to 50cc.of sea-water), 14 days old, pc, pericardial vesicle.

Fig. 9 From acetone solution (35 cc. gram-molecular to 50 cc. of sea-water), one of the 'twins' amorphous, 14 days old, pc, pericardial vesicle.

16 Neither verbal description nor the accompanying figures (of necessity limited in number) are adequate to give a complete picture of these conditions. To obtain an adequate idea of them and the dissociation that underlies them and is apparent in the sections, it is necessary to examine the latter in the series.

ORIGIN OF MONSTERS. II

421

422 E. I. WERBER

morphogenetic factor — blastolysis — is apparently responsible for both the 'monstra per defectum' and the double monsters and other 'monstra per excessum.'

The next example which we have chosen for the demon stration of the apparent syngenesis of both these deviations from the norm may now follow.

In figure 12 is presented an egg with a large 'pericardial' vesicle, dense vascularization of the yolk-sac and a very curiously misshapen embryonic mass which in toto was with some doubt interpreted as a dwarfed and highly malformed double monster. Microscopic examination of sections fully confirms this interpretation.

The first sections pass through the anterior part of the larger component only. They show (fig. 23) a dissociated tissue of an indifferent character in which are embedded two lenses, I and li. Further sections show that this tissue mass increases in density posteriorly and that it is the anterior, greatly dissociated, part of the single optic cup which 'eventually comes fully into view surrounding on all sides the larger one of the two lenses (fig. 24). The brain of this component is unilobed, very irregular in shape and unusually small, while the eye is very large. The unusual shape of the latter (fig. 25) as well as also the fact that the ill-differentiated layer of rods and cones does not (as it should) appear throughout the entire outer margin of the optic cup wall, but is observed for a considerable stretch to merge into the other layers of the retina, suggests that the eye has been formed out of a dissociated anlage. The above mentioned dissociated optic tissue mass anterior to the eye adds considerable weight to this evidence of blastolytic action.

The smaller component appears first in the twelfth transverse section. It is also possessed of one eye only which is rather small anteriorly and very irregular in shape. In more posterior sections, however, this optic cup gradually increases to an unusually large size, (o.c, fig. 26) it being almost as large as the brain of the component. The latter is unilobed, very irregular in shape and relatively very large.

ORIGIN OF MONSTERS. II 423

On following the sections further the brain of the larger component is observed to increase to a very large size (figs. 26 and 27). Cellular areas predominate in it and are strangely intermingled with fibrous areas. It is difficult to escape the impression that distortions of this nature can have resulted from anything but a disorganization of parts of the primordium and subsequent processes of regulation.

At this level there can be observed in the larger component also a very small, vestigial ear vesicle (figs. 26 and 27). More posteriorly the notochord of this component is doubled, thus indicating dissociation of its anlage at an earlier stage of development. The (rudimentary) tail of this component comes into view in the last sections. A rudimentary tail is also observed in sections passing through the last part of the smaller component (fig. 27).

The internal organs of both components are almost entirely obliterated.

A point of interest is presented also by the epidermis of both components on the side on which they are conjoined. A cluster of large cells (d.e. figs. 25, 26 and 27) is noted here to come off directly from the epidermis. The large size of the cells would seem to indicate that having through dissociation lost their correlation with the epithelium of which they originally formed a part, there was no restriction to their expansive growth. In some sections the nuclei of these cells exhibit evidence of chromatolysis, thus suggesting that chemical alteration may have partly been responsible for dissociation (chemical blastolysis) .

Turning, now, to the morphogenesis of this monster, we are justified, I believe, in assuming that it is a product of the combined action of, both osmotic and chemical, blastolysis. The former ('blastotomy') may by separating blastomeres (of the two-celled stage) have severed their mutual correlation, thus allowing each to develop independently. In the course of further development, however, the two, apparently not widely separated, embryonic primordia came into contact along their lateral surfaces where they coalesced.

424

E. I. WEBBER

The deformities and great defects (cyclopia) exhibited by both components are due to loss of embryonic substance by the double primordia owing largely to destruction by chemical alteration. The parts surviving this destruction have after regulation developed into two dwarfed, highly defective bodies.

Fig. 10 'Parasitic' double monster from acetone solution (35 cc. of grammolecular to 50 cc. of sea water), 14 days old. (Note the protruding, dissociated eyes of the 'autosite'.)

Fig. 11 Parasitic double monster from acetone solution (35 cc. gram-molecular to 50 cc. of sea-water), 14 days old.

Fig. 12 Dwarfed double monster from acetone solution (25 cc. gram-molecular to 50 cc. of sea-water), 15 days old, pc, pericardial vesicle.

Fig. 13 Mesocatadidymus (?) from acetone solution (30 cc. gram-molecular to 50 cc. of sea-water), 26 days old.

ORIGIN OF MONSTERS. II 425

A number of other duplicities have been recorded which also are monstra et per defectum et per excessum (figs. 1 to 10). Their genesis is, in all probability, due to the same factors that have been assumed for the monsters above described. Thus in the monster illustrated in figure 9, presenting double embryos of strikingly unequal size and greatly deformed, one of the ' twins' being amorphous, on microscopic examination indications are found of chemical alteration, while in the duplicity of figure 6 in which an isolated head fragment with a well differentiated eye (ascertained by microscopic examination) is observed at a considerable distance from either of the 'twins', action of osmotic pressure is apparent already on examination in toto, if conclusions from analogous observations in invertebrates (cf. for instance, Loeb, I.e.) be permitted.

The smaller one of the double embryos of figure 8 when examined in sections is observed to lack the anterior part of the head, the otic capsules appearing in the anterior-most sections. This is one of the most conspicuous defects of this component and, in my opinion, no error will be committed by assuming that the factor which brought about the duplication of the originally single embryonic primordium (osmotic pressure) may have contributed to the genesis of this and other defects by splitting off the chemically incapacitated parts. In the other double embryos recorded in our experiments which obviously are, at the same time also monstra per defectum microscopic examination of sections would, undoubtedly, furnish indications of the combined action of chemical alteration and osmotic pressure as underlying the doubling of primordia and the origin of the defects.

THEORETICAL REMARKS AND CONCLUSIONS

As pointed out repeatedly in the preceding pages no direct evidence is available for this combined action of osmotic pressure and chemical alteration. It may, therefore, seem desirable to inquire in how far their assumption is justified.

Nearly all duplicities of our experiments have resulted from the addition to sea-water (50 cc.) of a varied quantity (30 cc.

426 E. I. WERBER

or 35 cc. or 40 cc.) of a gram-molecular solution (in distilled water) of acetone, only a very few resulting from the employment of butyric acid. Considering the molecular weights of sea-water on the one hand and of acetone on the other hand, it is at once evident that by the addition of the acetone solution to sea-water the latter was quite appreciably diluted. In other words, as far at least as this physical condition is concerned, our experiments with acetone are in principle similar to the experiments performed by Loeb (I.e.) on sea-urchin eggs by diluting the sea-water with fresh water. In both instances the osmotic pressure of the medium surrounding the eggs was lowered, although less so in my experiments than in those of Loeb. Since, in this way, the difference between the internal osmotic pressure (of the eggs) and the external osmotic pressure (of the surrounding medium) was modified in experiments which besides other monsters yielded a certain number of duplicities, there is no apparent reason to doubt that the origin of the latter is directly traceable to this modification.

It might perhaps be wondered why relatively few double embryos have resulted from the above described treatment of the eggs. This, however, can, at least partly, be accounted for. The egg of Fundulus heteroclitus, well known for its hardiness, 17 is relatively little susceptible to physical insults, and it is possible that the dilution of sea-water by the addition to it of such quantities of acetone as I have employed, is not sufficient to cause osmotic blastolysis ('blastotomy' — Bataillon, I.e.) except in a very few eggs. Greater dilution of the sea-water by the addition of quantities of acetone greater than those employed in these experiments is, however, impracticable, for, in that case the degree of chemical action would be higher, and owing to it the mortality would, according to my experience, increase greatly and the few surviving eggs, having sustained so much chemical alteration, would result only in dwarfed, shapeless embryonic masses and fragments. The paucity of

17 In a number of experiments performed in 1915 in which eggs of Fundulus heteroclitus were subjected to a strong centrifugal force, no alteration of development or any injury resulted.

ORIGIN OF MONSTERS. II 427

double monsters in our experiments can, accordingly, not serve as an argument against the justification of our assumption of osmotic pressure as an important factor underlying their morphogenesis.

The second component factor of blastolysis, chemical alteration — is also largely inferred. However, this inference would seem to be well justified, if the action of acetone, partly solvent and partly as a precipitant of lecithine, is considered. Owing to the effect of this action some groups of cells may be entirely destroyed, while others may be only, more or less, chemically modified* and thus lose much of their inherent capacity for differentiation (inhibition) as well as of the power to resist the action of osmotic pressure.

Besides the two factors mentioned above attention may be called to another factor which may play a contributory part to the effect we term blastolysis insofar as the latter may be facilitated by it. This is the age of the egg, counting from the time of its maturation.

It has been found by several observers that the viability of the eggs and their ability to develop in a typical manner decreases gradually after their maturation.

Thus Hertwig ('92) noted that unfertilized frog eggs retained (for some time after the animal had been killed) in the uterus (on ice), showed with each consecutive day a greater tendency to abnormal development, which in many of them resulted in 'spina bifida.' Like results have been obtained by him also if female frogs were separated from males long enough for the eggs to become over-mature.

Conklin's ('97) observations on the eggs of Crepidula are very striking and may therefore be best presented in his own words (p. 30):

. . . . when the adult Crepidulas are brought to the laboratory, and kept in the best possible conditions, the percentage of these abnormalities increases, and when the egg capsules are removed from the mantle cavity of the mother and. kept in dishes of sea-water, the monstrosities increase to such an extent that after a few days not a single normally developing egg or embryo can be found.

428 E. I. WERBER

Bataillon (I.e.) has observed a very marked tendency to fragmentation and formation of double embryos in eggs of Petromyzon which had been retained in the females for several days after stripping. Since no modification of the environment whatsoever was employed in this case, Bataillon speaks of 'blastotomie spontaneV as the factor responsible for duplication.

Of the causes underlying this apparent deterioration of 'stale eggs' Goldfarb ('16, '17 a and b) has made a systematic study. He finds that it is apparently due to a progressive increase in their permeability and to the thus resulting rise in the rate of their oxidation, as postulated long ago by J. Loeb.

I am inclined to attribute considerable importance to this factor in our experiments in which, usually, eggs of several females were fertilized together before they were subjected to the action of the environmental modification. In other words, the material of our experiments was of a varying viability and varying susceptibility, the 'younger' eggs being less susceptible than the 'older' ones. This variability cannot well be obviated in any experiments of an explorative nature. For a single female of Fundulus heteroclitus will not yield eggs enough for a coherent series of experiments in which the same environmental modification is employed, but differing in each experiment in degree or in the stage at which the eggs were subjected to it. Besides, it is well known that not all eggs spawned by, or stripped from, a female at the same time are of the same 'maturation age.' It is a common experience to find mature and immature eggs in the same female during the spawning season, a fact which well indicates that the mature eggs have not all matured at the same time.

To this variation in degree of susceptibility of the eggs may be due the enormous range in variation in the end-products of development — from apparently perfectly normal embryos to monstrosities of a most bewildering kind. The duplicities which were recorded in our experiments are an incident in this almost endless 'series.' They result primarily from duplication of the embryonic primordium by blastolysis. And they as well as all other monstrosities are expressions of the variation in the

ORIGIN OF MONSTERS. II 429

degree of this action, for they, too, range from apparently symmetrical, well formed double embryos through various degrees of deformation to highly defective and grossly malformed double monsters (figs. 1 to 13).

Among some other points which would also seem to call for explanation is the often recorded unequal size of the components of a duplicity where two whole slightly conjoined or entirely separate embryos have developed (figs. 1, 2, 3, 4, 6, 7, 8, 9, 10, ll). 18 In my opinion the only explanation possible for this phenomenon is that apparently even less than one-half of the entire embryonic primordium is still totipotent in teleosts just as Eataillon (I.e.) has found this to be the case in Petromyzon. In the case of our double embryos I would further suggest that after the splitting of the germ into two parts one of them has apparently sustained further blastolytic lesion which, however, while dwarfing it in size, has not diminished its totipotency.

In this connection it may be well to consider also the sometimes very striking differences in degree of malformation between the components of a duplicity (figs. 2, 4, 8, 9, 10 and 11). In accordance with our interpretation of the genesis of malformations (Werber, '1G) it will have to be assumed that the more deformed component was subject to blastolytic action in a higher degree than the less deformed one. The above mentioned difference in size between components of the same duplicity is just another effect of this simple principle. I am, however, unable to account for these differences in degree of action on duplicates of the same germ.

Yet these differences (both in size and degree of malformation) between components of the same duplicity I regard as important insofar as they would seem to furnish a clue to the morphogenesis of the so-called 'parasitic double monsters'. For, granting an - early and intimate secondary fusion of primordia duplicated by blastolysis, and that one of these double

18 Similar observations on very young double embryos have been made also by Klaussner ('90) in the chick and in the lizard Lacerta viridis and by Dareste ('91) in the chick, while one such double embryo of the chick is in my possession.

430 E. I. WERBER

germs has sustained much greater injury than the other, the end-product of further cfevelopment will be a more or less well formed (sometimes even much deformed) embryo (the 'autosite') with a sort of 'parasitic appendage'-- the 'parasite.' 19 This conception of the manner of formation of the 'parasitic' duplicities differs essentially from Wilder's ('08, pp. 363-364, ff.), for no resort is made to the improbable secondary degeneration of one (whole) of the already formed components of the diplopagus owing to — what briefly may be called — physiological dominance of the other component. 20

It may now be questioned in how far our conclusions regarding the origin of the duplicities in our experiments can be extended to those occurring in nature in other vertebrates and particularly in mammals. "While I am, of course, disinclined to do so unconditionally, I think that with some restrictions this extension may be regarded as permissible. For the oftassumed possibility, if not probability, that in nature, too, the primary causal agent of atypical development is a chemical one (probably due to some disturbances of metabolism during which substances are produced by the body otherwise foreign to it) can not be denied. The human organism — and possibly that may apply to other mammals and to Sauropsidae as well — is subject to disturbances of metabolism of both a serious and a relatively harmless nature. To the former may be counted nephritis, diabetes, jaundice and other diseases of metabolism, while as a relatively harmless, temporary disturbance of metabolism may be viewed the presence in the organism of certain substances of fatigue after muscular exertion (for instance, lactic acid or monopotassium phosphate) and perhaps some other products of transitory abnormal conditions. While the substances of the former owing to their continued more or less severe

19 Fischel (I.e., p. 290) also suggests that these 'parasitische Anhange' develop from primarily rudimentary anlagen.

20 The differences between the 'parasitic appendages' and the foetal inclusions ('teratomata', 'embryomata', etc.) I regard, with Fischel, (I.e., pp. 290299 ff) as secondary only. A discussion of this subject is deferred to a future publication.

ORIGIN OF MONSTERS. II 431

action 21 on the ovum of a female suffering from any of these diseases may cause either its death and subsequent expulsion or any conceivable monstrosity, the latter, if altering the tonicity only of the blood, may by (osmotic) blastotomy produce slight defects or duplication of the embryonic primordium and thus in the latter case be responsible for the development of monozygotic twins, or any monozygotic, more or less well formed, 'symmetrical' duplicity.

It might further be questioned whether the analysis of the morphogenesis of our experimental fish duplicities as attempted in the preceeding pages on the basis of FischePs theory is applicable to the higher vertebrates and mammals.

Sobotta ('14) has already pointed out that while it is very doubtful whether the first two blastomeres are equipotent and totipotent in mammals it may be regarded as hazardous to extend to the latter the conclusions reached regarding the morphogenesis of duplicities in lower animals. However, occasioned primarily by the recent researches of Fernandez ('09) and Newman and Patterson ('10) and Patterson ('13) he suggests hypothetically a stage (the ' embryonic blastomere' of the fourcelled stage) whose parts after the next cleavage might be totipotent and thus capable of giving rise to monochorionic double embryos if separated. 22 Undoubtedly, in view of the meager data available for the earliest development in mammals, generalizations must be made with extreme caution. Yet, whatever the mammalian embryo-forming equivalent of the ovum of amphibia (or other animals in which the first blastomeres are totipotent) may be, there is, as Fischel (I.e.) has already pointed out, no apparent reason to expect that its prospective potency should be lower than that of the latter.

21 Depending partly upon the 'maturation age' of the ovum at the time of its fertilization.

22 For Tatusia novemcincta, on which Patterson's ('13) investigations were carried out, Sobotta assumes the separation of the four cells resulting from the second cleavage of the hypothetical 'embryonic blastomere.' Pending further investigations on the earliest stages of development in armadillo I should regard as questionable the analogy between the polyembryony in the latter and the occasional instances of 'twinning' in other mammals.

432 E. I. WERBER

Granting, however, the possibility that up to the time of gastrulation the mammalian (and Sauropsid) ovum is 'capable of being divided into two or possibly more equi- and totipotent parts, it would remain to be shown that the parts thus separated may, as in Petromyzon and teleosts, be able to form any kind of duplicity, depending upon the degree of separation, the distance between the parts separated and the angle of their respective convergence. , There is reason to believe that no differences should exist in this regard between the teleost ovum on the one hand and the Sauropsid and mammalian ovum on the other hand. For, as Fischel sets forth (I.e., p. 274), the processes leading to the formation of the embryonic body (separate anlagen for the head and trunk, the latter formed by concrescence of two lateral parts) being in principle very much the same in all vertebrates the morphogenesis of double monsters in teleosts may — at least with certain restrictions of secondary significance — be extended to amniotes and even to man.

These conclusions are, obviously, tentative and — in default of experimental tests in mammals — they will probably remain so for an indefinitely long time. Their advantage, however, besides offering a plausible explanation, is that they may help to keep alive further inquiries into the, so far inapproachable, problem of diplogenesis ('twinning') in amniotes.

It is for this reason mainly of methodological advantage that, — in default of evidence to the contrary— I am inclined to hold to the views above expressed in preference to Wilder's speculations- 3 on the subject. For, while I can find no fault with this writer's reasoning, I consider that his separation of the double monsters (sensu stricto) from well formed, symmetrical duplicities ('orderly beings — cosmobia') and the assumed mysterious origin of the latter by a sort of mutation due to ' germinal variation' are both unnecessary and arbitrary. The lack of deformities in 'Cosmobia' as well as in viable monozygotic twins would rather seem to indicate that while they may be developmental products of primordia duplicated by blastotomy, they

23 A view similar to that of Wilder's seems to have been advanced also by Bolk ('06), whose paper is, unfortunately, inaccessible to me.

ORIGIN OF MONSTERS. II 433

have apparently sustained no destruction due to this or any other factor. The symmetry in anatomical details observed in 'Cosmobia' may, I venture to say, be only an incident — symmetrical blastotomy and the subsequent undisturbed development of the primordia which havej thus incidentally, remained equipotent. Conversely we may assume that where (osmotic) blastotomy or blastolysis (both osmotic and chemical) causes destruction of some areas of the germ, any monster and — if the germ has also been split into two more or less equivalent parts — any deformed duplicity may result from further development. The distinction between these two kinds of duplicities is, accordingly, mainly one of degree; it is a quantitative difference rather than a qualitative one and it may depend upon the specific action (either chemical or physical or both chemical and physical) of the (probably chemical) agent primarily responsible for the deviation from the typical course of development.

Moreover, this quantitative difference may, as pointed out above, depend also upon the 'age' of the egg, the 'younger' eggs being hardier and more capable of surviving blastotomy without sustaining fatal lesions, while the 'older' ones with a tendency to cytolytic fragmentation may apparently be (in a varying degree) more subject to such lesions produced by blastotomy or blastolysis.

The 'old age' of the egg is, undoubtedly, a germinal variation, but this 'variation' is in the nature of a pathological change and the causes underlying the latter may well be within the bounds of discovery, as is suggested by the recent researches of Goldfarb (I.e.). The germinal variation, however, assumed by Wilder is not a pathological condition and being nothing else that could intelligibly be defined, it automatically puts a stop to further inquiry into the primary causes underlying diplogenesis in nature.

While I am not at all inclined to underestimate the almost insuperable difficulties presented by the problem, I cannot on the other hand subscribe to Wilder's skepticism voiced in the following characteristic remark ('08, p. 428) :

434 E. I. WERBER

. . . . even though the monsters resulting from the present or future experiments, in which the cause is applied after the formation of the germ and is therefore secondary, be found upon careful examination to be definite Cosmobia, it will prove only that such monsters can be thus produced and not that Nature does produce them in the same way.

It is, obviously, difficult to deny the justifiability of this argument, but on the other hand — I think Professor Wilder will admit — it practically amounts to questioning the value of many explorative experiments in biology.

May 30, 1917.

LITERATURE CITED

(For a more complete list of the extensive literature on the subject the reader is referred to the chapter on double monsters in Schwalbe's textbook and to Hiibner's article.)

Bataillon, E. 1901 La pression osmotique et les grands problemes de la

Biologie. Arch. f. Entw.-Mech., vol. 11. Bolk, L. 1906 Dubbelmonstra, hun Classificatie en Ontstaan. Genees kundige Bladen uit Klinik en Laboratorium. 12. Reeks., Nr. 9 en

10. Haarlem. (Quoted from Hiibner.) Conklin, E. G. 1897 The embryology of Crepidula. Jour. Morph., vol. 13. Driesch, H. 1893 Entwicklungsmechanische Studien. III-VI. Zeitschr. f.

Wiss. Zool., vol. 55. Fernandez, M. 1909 Beitriige zur Embryologie der Giirteltiere. Zur Keim blattinversion und spezifischen Polyembryonie der Melita (Tatusia

hybrida Dem.). Morpholog. Jahrb., vol. 39. Fischel, A. 1898 Experimentelle Untersuchungen am Ctenophorenei, I-IV,

Arch.' f. Entw.-Mech., vols. 6 and 7.

1902 tjber den gegenwartigen Stand der experimentellen Teratologic

Verhandlg. d. deutschen patholog. Gesellsch., Berlin, 1903. S. 255-357.

Referat gehalten auf der 5. Tagung der deutschen patholog. Gesellsch., Karlsbad, 22-26 Sept., ^902. Gerlach, L. 1882 Die Entstehungsweise der Doppelmissbildungen bei den

hoheren Wirbeltieren. Stuttgart. Goldfarb, A. J. 1916 Experimental studies upon stale germinal products.

Carnegie Institution of Washington. Year Book No. 14, for the year

1915.

1917 a Experimental studies upon the aging and death of germ-cells.

Carnegie Institution of Washington. Year Book No. 15, for the year

1916.

1917 b The effects of ageing upon germ cells and their development.

Proc. of the Society for Exp. Biol, and Medicine. Eighty-first meeting, February 10.

ORIGIN OF MONSTERS. II 435

Herlitzka, A. 1895 Contributo alio studio dello capacita evolutiva dei due primi blastomeri noil' uovo di tritone. Arch. f. Entw.-Mech., vol. 2. 1897 Sullo svilupo di embryoni completi da blastomeri isolati di uova di tritone (Molge cristata). Arch. f. Entw.-Mech., vol. 4.

Hertwig, O. 1892 Urmund und Spina bifida. Arch. f. mikr. Anat., vol. 39.

1903 Missbildungen und Mehrfachbildungen, the dutch Stoning der ersten Entwicklungsprocesse hervorgerufen werden. Handbuch der vergleichenden und oxperimentellen Entwiclungslehre. Jena. pp. 967-989.

Hubner, H. 1911 Die Doppelbildungen des Menschen und der Tiere. Lu barsch und Ostertag's Ergebnisse der allgemeinen Pathologic und

pathologischen Anatomie der Tiere. Klaussner, F. 1890 Mehrfachbildungen bei Wirbeltieren. Eine teratolo gische Studie. Mtinchen. Loeb, J. 1895 Beitrage zur Entwicklungsmechanik der aus einem Ei ent standenen Doppelbildungen. Arch. f. Entw.-Mech., vol. 1. Mall, F. P. 1908 A study of the causes underlying the origin of human monsters. Jour. Morph., vol. 19. Morgan, T. H. 1893 Experimental studies on teleost eggs. Anat. Anzeiger,

vol. 8. Newman, H. H. and Patterson, J. T. 1910 The development of the ninebanded armadillo. Jour. Morph., vol. 21. Newman, H. H. 1917 The biology of twins. The University of Chicago Press. Patterson, J. T. 1913 Polyembryonic development in Tatusia novemcincta.

Jour. Morph., vol. 24. Rauber, A. 1879-1880 Formbildung und Formstbrung in der Entwicklung

von Wirbeltieren. Morpholog. Jahrb., vol. 5 and 6. Reichert, E. T. 1914 The germplasm as a stereochemic system. Science,

N. S., vol. 40, No. 1036, November 6. Schtjltze, O. 1894 Die kiinstliche Erzeugung von Doppelbildungen bei

Froschlarven mit Hilfc abnormer Gravitationswirkung. Arch. f.

Entw.-Mech., Bd. 1. Schwalbe, E. 1907 Die Doppelbildungen. In: Die Morphologie der Miss bildungen des Menschen und der Tiere. II. Teil. Sobotta, I. 1914 Eineiige Zwillinge und Doppelmissbildungen des Menschen

im Lichte neuerer Forschungsergebnisse der Siiugetierembryologie.

Studien zur Pathologic der Entwicklung, herausgegeben von R. Mayer

und E. Schwalbe, vol. 1. Spemann, H. 1900 Experimentelle Erzeugung zweikopfiger Embryonen. Sitz ber. d. phys.-med. Ges., Wiirzburg.

1901 Experimented crzeugte Doppelbildungen. Verb.. V. Intern.

Zool. Kongress. Berlin.

1901, 1902, 1903 Entwicklungsphysiologische Studien am Tritonei.

I, II u. III. Arch. f. Entw.-Mech., vol. 12, 15 and 16.

1904 tJber experimented crzeugte Doppelmissbildungen mit zyklopischem Defekt. Zool. Jahrb., Suppl., Bd. VII. Festschr. f. ^'eissmann.

THE JOVRXAL OF EXPERIMENTAL ZOOLOGY, VOL. 24, XO. 2

436 E. I. WERBER

Werber, E. I. 1915 Experimental studies aiming at the control of defective

and monstrous development. A survey of recorded monstrosities

with special attention to the ophthalmic defects. Anat. Rec, vol. 9.

1916 a Blastolysis as a morphogenetic factor in the development of

monsters. Proc. Amer. Ass. Anat. in Anat. Rec, vol. 10.

1916 b Experimental studies aiming at the origin of monsters. I.

An etiology and an analysis of the morphogenesis of monsters. Jour.

Exp. Zool., vol. 21. Wetzel, G. 1895 Uber die Bedeutung der zirkularen Furche in der Entwick lung der Schultze'schen Doppelmissbildungen von Rana fusca. Arch.

f. mikrosk. Anat., Bd. 46.

1896 Beitrag zum Studium der kiinstlichen Doppelmissbildungen

von Rana fusca. Inaug.-Diss. Wilder, H. H. 1904 Duplicate twins and double monsters. Am. Jour. Anat.,

vol. 3.

1908 The morphology of Cosmobia. Speculations concerning the

significance of certain types of monsters. Am. Jour. Anat., vol. 8. Wilson, E. B. 1893 Amphioxus and the mosaic theory of development. Jour.

Morph., vol. 8.

PLATE 1

EXPLANATION OF FIGURES

14, 15 and 16 Transverse sections through 'parasitic double monster' of figure 11. L., lens.; I., small lentoid body; b., brain; e., anterior-most part of eye; d.e., dissociated epidermis; o.c, optic cup; d.o., dissociated optic cup; d.m., dissociated epithelium of mouth; y., yolk. X 105.

ORIGIN OF MONSTERS. II

E. I. WERBER

PLATE 1

O.C:

437

PLATE 2

EXPLANATION OF FIGUKES

17, 18 and 19 More posterior sections through the embryo of figure 11. b., brain; n., notochord; i., intestine; d.m., dissociated epithelium of mouth; m.o., medulla oblongata of 'parasite'; s.c. spinal cord of 'autosite'; s.c. t , spinal cord of 'parasite'; y., yolk. X 105.

438

ORIGIN OF MONSTERS. II

E. I. WERBER

PLATE 2

439

PLATE 3

EXPLANATION OF FIGURES

20, 21 and 22 Transverse sections through embryo of figure 13, n., notochord; s.c, spinal cord; d.s.c, dissociated and partly 'degenerated' spinal cord; s.c.i, s.c.t and s.e. 3 , dissociated, multiple spinal cord, cavities lined with endothelium. X 105.

440

ORIGIN OF MONSTERS. II

E. I. TVERBES

PLATE 3

Ad.s.c.

411

PLATE 4

EXPLANATION" OF FIGURES

23, 24, 25, 26 and 27 Transverse sections through dwarfed double embryo of figure 12. 6 1( brain of the larger component; b 2 . brain of the smaller component; h, and h lenses in anterior-most, greatly dissociated part of the eye of the larger component; I., lens (smaller component); e., eye; d.e., dissociated epidermis; 7n.o., medulla oblongata of smaller component; e.v., rudimentary ear vesicle; o.c, posterior-most, greatly dissociated part of the optic cup of the smaller component; p., plasma in pericardial vesicle; /., tail of smaller component. X 105.

442

ORIGIN OF MONSTERS. II

E. I. WEBBER

PLATE 4

No

443

author's abstract of this paper issued by the bibliographic service, november 3