Paper - The non-specificity of the germ-layers

From Embryology
Embryology - 24 Aug 2019    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

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

Oppenheimer JM. The non-specificity of the germ-layers. (1940) Q Rev Biol 15:98–124.

Online Editor  
Mark Hill.jpg
This historic 1940 paper by Oppenheimer describes historic embryology research in gastrulation, including a survey of the key historic research and researchers in establishing this developmental concept.

Modern Notes: gastrulation | ectoderm | mesoderm | endoderm

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

The Non-Specificity of the Germ-Layers

The Quarterly Review Of Biology March, 1940

By Jane M. Oppenheimer

Department of Zoology, University of Rochester, Rochester, N. Y., and Department of Biology, Bryn Mawr College, Bryn Mawr, Penna.

Eigentlich bcginnt in jeder dicser drei Schichten cine eigene Metamorphosc, und jcde cilt ihrem Ziele entgcgen; allein es ist jede noch nicht selbststéindig genug, um allcin das darzustellen, Wozu sie bcstimmt ist; sie bedarf noch der Hulfe ihrer Gefahrtinnen, und daher wirken alle drey, obglcich schon zu verschiedenen Zvvecken bcstimmt, dennoch, bis jede eine bcstimmte Hohe erreicht hat, gemeinschaftlich zusammen. . . .
Chr. Pander.
A poor translation  
Actually, in each of these three layers, there is a separate metamorphosis, and there is no escape from its goal; but each is not yet self-sufficient enough to represent all that it is for; it still requires the help of its other companions, and therefore all three, though already in different parts, nevertheless cooperate together, until each one has reached a certain height....

Chr. Pander.


Probably the single set of facts that biologists who specialize in other branches of their science than embryology carry away in their store of general information is that describing the histological accomplish ments of the germ-layers. The embryol ogists themselves as pedagogues are prob ably more dogmatic concerning these facts than concerning any other [the recent edition of the Brachct (1935) textbook is a Welcome exception]. The fixed and simple concept as expressed in the germ-layer doctrine is easy to remember and accordingly has been gratefully retained for its usefulness as a rule of thumb.

Time and time again contradictions have found their Way into the specialized literature of embryology but they have only rarely penetrated into general consciousness. Since the recent work in embryology has had important bearings on the problem, it seems Well to review the field once more. The historical approach to the problem has been chosen simply because it is so interesting to trace through the dogged attempt of the human mind to cling to a fixed idea.

Early History

Up to the middle of the eighteenth century, the existence of the process of development was recognized, but its constituent mechanisms were quite unknown. Aristotle, Fabricius, Harvey and many others less illustrious had seen cmbryos or foetuses and had even watched them develop; they had seen in them the forerunners of adult structures, and had noted the appearance of some of these. How the structures were formed—what their source and what their material—vvas a question unaskablc for many reasons.

In the earlier days of embryology, before the Renaissance, unquestioning belief in a rigidly Aristotelian philosophy limited the explanation of all biological processes, development included, to the basis of a concoction of four humours. Even after the Renaissance, when this explanation had been abandoned in anatomy, it sufficed for embryology because no one could detect, with unaided eye, those developmental processes whose very existence defied suspicion until the invention of the microscope.

For some reason embryologists delayed using this instrument until a century after its introduction, and then, ironically enough, the results of its first employment denied the process of development. Malpighi, in perhaps the most famous error of biology, thought that he discovered the principal adult organs present in the unincubated egg. It was in the refutation of this error, that the possibility that there might be such a thing as a germlayer was called into existence. Caspar Friedrich Wolff (1758; 1812) looked at the unincubated chick blastoderm and found that the organ rudiments were not yet present in it. He further found that later the gut and probably also the nervous system were formed by a process of folding of a layer of the stuff of which the blastoderm is composed.

Wolff, however, simply saw that the adult organs were not necessarily preformed in the unincubated blastoderm, but that some of them were formed later from layers of the blastoderm. It was Pander, however, student of Dollinger, who really elucidated what these layers were.

First Description of the Layers

Christian Pander, in his two papers published in Wurzburg in 1817, first described the trilaminar structure of the incubated chick blastoderm. (It was Pander who originally coined the word “blastoderm”: “Because the embryo chooses this as its seat and its domicile, contributing much to its configuration out of its own substance, therefore in the future we shall call it bZ4.rtodemz" (1817b, p. 2.1).) His own description of the three layers (1817a, pp. 5, 11-12) is better translated than paraphrased:

At the twelfth hour the blastoderm consists of two entirely separate layers, an inner one, thicker, granular and opaque, and an outer one, thinner, smooth and transparent. The latter, because of its development and for the sake of greater accuracy of description, we may call the serous layer and the former the mucous layer. . . .There arises between the two layers of the blastoderm a third middle one in which the blood vessels are formed, which we therefore call the vessel-layer; from its origin events of the greatest importance subsequently occur. . . . Actually there begins in each of these three layers a particular metamorphosis, and each one strives to achieve its goal; only each is not yet sufficiently independent by itself to produce that for which it is destined. Each one still needs the help of its companions; and therefore all three, until each has reached a specific level, work mutually together although destined for different ends.

Pander not only recognized the layers once they were formed, but he also realized clearly that one layer was formed at the expense of another. He wrote elsewhere (1817b, pp. 26-27):

What merits most attention is the composition of the blastoderm out of two layers. For before incubation this membrane consists of a single layer, made up of granules which cohere to each other by their own viscosity. As incubation progresses, however, there originates from this another layer, more delicate but firmer in structure, so that at a specific time the blastoderm can be divided by a fairly long maceration into two layers.

Elaboration of the Germ-Layer Concept

Karl Ernst Von Baer
Karl Ernst Von Baer (1791-1876)

Pander crystallized the germ-layer concept for the chick embryo. His friend and colleague, Karl Ernst von Baer, also a student of Dollinger’s, extended it to encompass all of vertebrate development, thereby laying down the fundamental bases for the study of comparative embryology.

Von Baer (1837, Bd. 7., S. 68) recognized the value of Pander’s contribution; only in a way he reorganized it perhaps a trifle too assiduously, since he unjustifiably stretched Pander’s three layers to four, breaking up the middle layer into two layers—roughly the equivalent of what we know as somatopleure and splanchnopleure:

We can speak of an upper and a lower layer; the former we call the skin layer and the latter the mucous layer. The material that lies between the two clings partly to the upper layer and partly to the lower. In this way there gradually develop two inner layers, an upper and a lower. In the lower of the inner layers the granules become clear and dissolve into vesicles, and finally part of the contents of this layer begins to flow. It becomes a vessellayer. In the upper the granules become darker; this becomes a flesh- or muscle-layer.

It seems hardly necessary to emphasize that von Baer’s most significant contribution, so far as the germ-layers are concerned, was the recognition of the fact that Pander’s discovery for the chick was valid for all the rest of vertebrate development. This has been the basis of embryology from von Baer’s time until today. The question often arises as to what the status of embryology would be without the contribution of von Baer. It might conceivably be little different than it is, because of the insight of another investigator who had the misfortune, from posterity’s point of view, to be a contemporary of von Baer. In 187.5, several years before the publication of von Baer’s treatise, Martin Heinrich Rathke, who had read Pander’s paper, applied this author’s observations on the germ—layers to the development of an invertebrate, Aymcm, describing the splitting of the blastoderm into a serous and mucous layer which fit one inside the other “like the coats of an onion” (quoted in E. S. Russell, 1916, p. 7.08). As a matter of fact Rathke seems to have applied Pander’s figurative vocabulary as well as his concept to the invertebrates, since he wrote, without anywhere in his paper referring to Pander: “After the blastoderm has divided into two particular layers, each of these layers proceeds by itself towards its final goal” (translated from Rathke, 189.5; pp. I094-5). So much was stated in the preliminary note. The final paper, which was published in 187.9, described the layers more explicitly:

One of them clings closely to the yolk and corresponds to the mucous layer of vertebrates, and is subsequently used for the production of the intestine and of a special yolk-sac. The other, on the other hand, is essentially comparable to the serous layer of the vertebrate, insofar as it . . . forms the body wall of the embryo, from which the different appendages as well as the central part of the central nervous system take their origin. A special and separate vessel—layer is never perceptible. . . [there is] more the idea of it than its actual presence. (Translated from Braem, 1895, S. 495.)

These words show clearly that Rathke saw the implication of Pander’s discovery as well as did von Baer. He was able to transfer the analogy even to the Invertebrates. His work was less well known than von Baer’s, probably principally because his generalizations were more on the embryological and less on the transcendental side. So far as their actual content is concerned, it would have made as solid a groundwork on which to build the science of embryology as the more celebrated Scholia of von Baer.

The significance of von Baer’s concept was immediately recognized. In one way, however, it seemed almost a culmination rather than a new point of departure, and for many years the concept was accepted with only slight amplification and refinement.

The refining, very nearly completed by Robert Remak .between 1850 and 1855, consisted of a double process: first, the interpretation of the germ—layers as composed of cells which were derived from the single cell of the original egg, second, the essentially correct demonstration that each of the germ—layers has a specific histological future. Remak recognized two primary germ—layers: (1) the upper, or sensory layer, subdivided into medullary plate and its derivatives,. and the epidermic plate, and (2.) the under layer subdivided into (2a) the trophic layer which gives rise to the alimentary canal and its derivatives, and (Lb) the motorgerminative layer which furnishes peripheral nerves, muscle, blood vessels, connective tissue, sex glands, etc. Furthermore, this middle motorgerminative layer is separated into dorsal and ventral somite plates by the pleuroperitoneal cavity, which is the precise equivalent of what we know as the coelome. It is obvious that these are the precise facts of the germ—layer concept as we recognize them today, with the exception of the one major error concerning the origin of the peripheral nerves.

Germ-Layers in the Vertebrate Embryo and the Adult Coelenterate

Even before Remak had published his book, one investigator was beginning to realize that the two so-called primary germ-layers formed as fundamental a part in the architecture of the adult coelenterate as in the molding of the vertebrate embryo, and he laid the foundation of what was subsequently to become the whole superstructure of phylogenetic and ontogenetic studies so extravagantly elaborated by the adherents of the evolution concept.

Huxley is credited with this discovery. He wrote, in a matter-of—fact manner, in a short paper “On the Anatomy and Affinities of the Family of the Medusae” (1849, 1313-414, 425):

I would lay particular stress upon the composition of this (stomach) and other organs of the Medusae out of two diytinct membrmm, as I believe that this is one of the essential peculiarities of their structure, and that a knowledge of the fact is of great importance in investigating their homologies. I will call these two membranes as such, and independent of any modification into particular organs, ‘foundation membranes’. . . . A complete identity of structure connects the ‘foundation membranes’ of the Medusae with the corresponding organs in the rest of the series; and it is curious to remark, that throughout the outer and the inner membranes appear to bear the same physiological relation to one another as do the serous and the mucous layer of the germ; the outer becoming developed into the muscular system and giving rise to the organs of offense and defense; the inner, on the other hand, appearing to be more closely subservient to the purposes of nutrition and‘ generation.

In a way, it is almost surprising that this discovery had not been made earlier, in view of the fact that Rathke had so early noted that Invertebrates as well as Vertebrates were bilaminar in early development. Also, Cuvier, as Huxley knew, had remarked on the bilaminar structure of the Coelenterates. In speaking of the Tubularians, Cuvier wrote (1846, p. 557):

Here the polyps do not form simple aggregations in which the individuals are distinct: but they are intimately united in such a way that they compose a more or less complicated individual which we call a compound polyp.
Whatever way the compound polyp is composed, the alimentary or digestive cavity of each polyp opens into a common nutritive tube, into which flows, or is secreted, the nutrient fluid produced by the digestive processes of each polyp.
The walls of the nutritive tube are formed by a double membrane, always intimately fused in this part of the compound polyp; the external corresponds to the skin; the internal is a continuation of the digestive portion of the alimentary cavity (of the individual polyps).
The former, in the compound polyp, secretes from its external surface a tube or sheath, thin like parchment, or horny in nature....

As a matter of fact, however, the investigator who came closest to anticipating Huxley’s brilliant generalization was none other than von Baer himself. Von Baer (I837, Bd. 2., S. 67) early compared the primary germ-layers with the walls of the coelenterate, in the following statement:

Yet originally there are not two distinct or even separable layers, it is rather the two surfaces of the embryo which show this difference, just as polyps show the same contrast between their internal digestive and external surface. In between the two layers there is in our embryo as in the polyp an indifferent mass.

The primary germ-layers of the Coelenterates were given their definitive names shortly after the publication of Huxley’s paper. In a paper “On the Anatomy and Physiology of Coelenterate” Allman wrote in 1853 (p. 368): “All the hydroid zoophytes can be proved to consist essentially of two distinct membranes; to the external of these membranes I shall give the name of ectoderm, and to the internal that of endoderm.”

In spite of its importance, the implications of Huxley’s brilliant observation remained unnoticed for almost twenty years. Haeckel’s Gmerelle Morp/aologie, for instance, published in 1866, makes no mention of the germ—layers. The dearth of progress is perhaps nowhere better shown than by a study of Huxley’s own Introduction to the Clzzuification of Ammaly, published in 1869, exactly 7_o years after his first statement concerning the Medusae. In this book, the only statements concerning the germ—layers are: (I) that the Hydrozoa are separated into two layers of tissue, the ectoderm and the endoderm, (7.) that the Actinozoa are likewise constructed of two membranes, ectoderm and endoderm, (3) that the author can confirm Remak’s statement that the brain and spinal cord of Vertebrates are a result of the modification of the serous layer of the germ, and (4) that the serous layer of the germ helps to form the amnion in the chick embryo while the allantois is formed from neither mucous nor serous layer but from the intermediate stratum. In no case here does he mention any relationship between coelenterate ectoderm and endoderm on the one hand and embryonic serous and mucous layers on the other. In other words, although Huxley had appreciated the fundamental relationship between the body-layers of Invertebrates and the embryonic layers of Vertebrates, yet twenty years later he and all other investigators were still waiting to utilize the generalization in any way, even for pedagogic reasons.

Even at this early date, before the word mesoderm had even been coined, and before the obvious generalization had been made, the germ-layer concept became subject to distortion. In 1865 William His formulated his ‘archiblast-parablast” theory. This theory, based on a study of the extremely specialized teleostean development, claimed that the archiblast, composed of the three classical germlayers, gives rise to all the embryo except the blood vessels and connective tissue which are furnished by the parablast. His’s theory bore little lasting effect on the development of embryology, and fortunately was ultimately abandoned even by its author. But it gives perhaps the first example of the way in which the germ-layer theory has been distorted in the course of its development.

Evolutionary Significance of the Germ Layers

In the same years when Huxley was issuing lectures on comparative anatomy that included the words ectoderm and endoderm only in discussion of the Coelenterates, and when His was worrying about the archiblast and parablast, a Russian investigator was making the observations that were most instrumental —partly because of their own inherent worth and brilliance, partly because of the fortunate time at which they were published-in effecting the bond between embryology and anatomy, and between the study of ontogeny and phylogeny.

Alexander Kowalewski, in the years 1867-71, found that all the invertebrate embryos he studied, and these were of many types, were formed of the same primary layers as the vertebrate embryos, and, furthermore, that in all of them alike the layers arise in the same fashion, the inner layer being produced from the outer by a process of invagination. Kowalewski’s words can speak for himself more convincingly than we can speak for him (I867a, p. 3):

The first change in the embryo (Psolinus, a holothurian) consists of an insignificant invagination which becomes visible at one pole of the egg, and whereby the whole embryo takes on a somewhat conical form. The invagination progresses gradually farther and after a few hours forms a deep sack. . . . A similar division of the cells of the embryo into two layers, an outer and a central one, I have also observed in many other animals, and especially clearly in the eggs of Psolinus.

Nineteen days later, when he presented his paper on “Amp/aioxiif’ (I867b, pp. 3, 5), the generalization had broadened considerably:

The embryo now consists of two sheets or germlayers, the outer and the inner; we can therefore compare it with the embryonic anlage of the bird, mammal and turtle-egg, when these still consist of two layers. If we compare figure 15 of Reichert’s paper on the development of the guinea-pig with our figures 8 and 9, the similarity between these two forms of development immediately strikes us. . . . The embryo quite agrees, even in the most insignificant details, with the embryo of the corresponding stages of Pboronir, of Limnaezir, of Artemcantbion berylimi: Ag., of Opbizim and of Ecbinm, according to my own still unpublished observations; and if we leave the cilia out of consideration, our larva agrees also with the corresponding stage of Sagitta, of the Ascidians (Ar. intestinczlir and Pizczlzuia mcmzmillata); if we consider that the segmentation cavity is filled with yolk, it resembles also the larva of Ercboltzia, of Cerium and of Sepiola. In all of the embryos mentioned here the formation of the two laminae or layers proceeds in exactly the same way. . . . Thus the first formation of the embryo would be quite in agreement for all these different animals; only in the further changes do we see appear the differences which characterize the individual type.

And in 1869 (p. 7.9), he wrote in a paper on worms and arthropods:

Now if we compare the development of the worms we have described with that of other animals, the analogy of the germ-layers of the worms with those of the Vertebrates, even in the details, astonishes us. The same two primitive layers which play the leading roles in the development of the worm appear also in the ,Vertebrates; as in the one group so in the other the middle layer appears only later. The destinies of the layers and of the organ anlagen are in very great agreement even down to the individual processes.

Rathke’s comparison of the embryonic germ-layers of Vertebrates and Invertebrates remained buried. Huxley’s recog nition of the relationship of the germlayers in the adult coelenterate and the embryonic vertebrate, so strangely anticipated by von Baer, scarcely received comment for twe decades. But the researches of Kowalewski bore immediate fruit. By 1870 the scientific world was flaming with the debate on evolution that was kindled by the publication of Darwin’s Origin of Speciex. The publication of Kowalewski’s observations on the universality of the germ-layers and on their comparable origin in a multitude of forms made it possible to consider the evolution of the individual and the evolution of the race in the same light. The decade of the I870’s saw embryology adduced as a complete confirmation of the evolution-hypothesis, and the evolution of the race as an explanation sine qua non of the course of evolution or development of the individual.

The first significant attempt to relate phylogeny with ontogeny was that of Kleinenberg, who published in 1877. his monograph on the histology and development of Hyiim, a paper dedicated, by the way, to Ernst Haeckel. In this the author found the coelenterate the perfect organism to represent the transition form from coelenterate to vertebrate embryo, and to represent the fundamental type on which all other forms are patterned and from which they are necessarily derived:

The low position of the Coelenterates in the system is perfectly understandable from their developmental history. Their type is determined by their maintenance of the fundamental spatial relationship of the germ-layers, and of their different layers in turn, to each other and to the external world. . . . The resultant great simplicity and uniformity of the whole body structure distinguishes the Coelenterates from all other animal groups: in the latter the definitive body arises through far-reaching histological segregations, but principally through manifold transformations and interminglings of the germ-layers, with the result that these are scarcely recognizable at all in the completed organs, and only in vague outlines in the body as a whole. But if we follow the developmental history of these complicated organizations backwards, we arrive finally, in the Vertebrates and probably in all animal groups, to forms which correspond essentially to those of the Coelenterates. Now since these forms are necessary, but transitory, developmental stages upon which the specific type is built, while on the other hand among the Coelenterates the same forms, maintained unchanged, portray the type, so the conclusion is apparent that not only the developmental processes in all animals are identical up to a certain stage, but that even in individual development the transition of one type into another occurs, since the constant type of the coelenterate is passed through as a developmental stage by all higher animals. The sim ple type of the coelenterate is the common ground form to which all the infinitely rich and manifold configurations of the animal body can be directly or indirectly referred. (Kleinenberg, 87-88.) 1872., pp.

In 1869 the terms ectoderm and endoderm had not yet been applied to describe the germ-layers of the embryo: in 1872. the terms were being used as they are now by Haeckel in Germany and in 1873 by Lankester in England. The term mesoderm was introduced by Huxley in 1871 (pp. 10—11 of the 1872. edition) in his Mamzml of Anatomy, and it was used by Haeckel (1872) in his monograph Die K41/eycfzwdmme. Lankester’s paper, published in May 1873, represents, according to its author, “part of a course of lectures commenced in the University Museum, Oxford, during Michaelmas term 1872.” Haeckel’s Die K41/zxcfzwaimme appeared in 1872 after Lankester’s paper was drawn up but before it was published; whether he adopted Haeckel’s terminology, or whether Lankester adopted his, or whether both independently used the same terminology, is not apparent. In 1873 Balfour and Lankester both were using the terms epiblast, mesoblast and hypoblast in place of the other terms. The history of terminology may be a futile study, but in the present case it is interesting since the men we have mentioned as changing the use of words simultaneously deflected the course of thought.

Lankester published in 1873 the preliminary and in 1877 the final communication in which he created a classification of animals based on their constitution into layers: All animals are homoblastic, diploblastic or triploblastic; the triplo— blastic forms are derived from the diploblastic through the Vermes; the planula, or the larva of the coelenterate, is the parent, phylogenetically speaking, of all diploblastic and triploblastic forms.

Ernst Heinrich Haeckel (1834-1919) (Photo. by Tokuzo Kimura)

Haeckel (1872), in Germany, published simultaneously a similar theory destined to become of greater influence than Lankester’s because of Haeckel’s genius of expression. Haeckel’s concept, in a way, was influenced heavily by the English school, partly because of Haeckel’s blind acceptance of the evolution-doctrine and partly because of his deep personal affection for Huxley. The parent of all forms, in Haeckel’s theory, is a two-layered sac similar to the bilaminar stage of all embryos described so skilfully by Kowalewski and known to us by Haeckel’s term “gastrula.” Haeckel wrote in 1872 (Bd. 1, S. 466-467) in the general part of his monograph on the Calcispongiae:

In all of these representatives of the most varied animal groups the gastrula has exactly the same structure. In each case its simple monaxial oval body encloses a simple central cavity (gastral cavity) which opens through a mouth at one pole; in each case the thin wall of the cavity consists of tWO celllayers, an inner layer of larger darker cells (entoderm, gastral layer, inner, trophic or vegetative germlayer) and an outer layer of smaller generally ciliated lighter cells (exoderm, dermal layer, outer, sensory or animal germ-layer). I conclude from this identity of the gastrula in representatives of the most diversified animal groups, from the Sponges to the Vertebrates, according to the fundamental biogenetic law, that the animal phyla have a common descent from a single unknown stock form, gastraea, which is constructed essentially like the gastrula.

According to this theory, the gastrula, Kowalewski’s bilaminar sac, produces all embryos; a similar Urmzztter is therefore the necessary progenitor of all multicellular forms. Finally, the course of evolution of the embryo is step by step explained and maxed by the evolution of the race to which the developing individual belongs. The development of the gill-slits in a mammalian embryo, to take a familiar instance, would be caused according to Haeckel necessarily and solely by the fact that in the course of evolution the ancestor of the mammal possessed gill-slits. One wonders how the promulgator of such a distorted doctrine of cause and effect could have been championed by the same Huxley who wrote: “Fact I know and Law I know: but what is this Necessity save an empty Shadow of my own mind’s throwing?”

Haeckel was probably the transcendentalist par excellence of all biology. Equipped naturally with those rarely combined virtues, an aesthetic appreciation of a high degree and a passion for methodical terminology and organization, he produced in the gastraea theory a scheme of ideas as intricate and symmetrical as the figures of Radiolarians which he loved to portray with his pen. The most perfect example imaginable of fitting the facts to the theory, a beautiful intellectual feat, totally devoid of scientific value, his gastraea theory was the culmination of the early work on the germlayers. Some of his contemporaries realized that phylogeny could not explain away ontogeny, and His (1874), for instance, vainly suggested and even attempted a study of the mechanical causes of development. But the beautiful unity of Haeckel’s scheme was too seductive. Huxley and the English embryologists spent their days apotheosizing its author and looking at embryos only for the purpose of fitting the facts of ontogeny into the ideal of phylogeny. Kleinenberg (1886, p. 2), who had nominated Hydra and the Coelenterates for the throne usurped by Haeckel’s “gastraea,” gave a succinct and vivid critique of the gastraea theory in his paper on the development of L0jmd0r/aymm which this time bore no dedication:

The good in it (the gastraea theory) belongs to Huxley; what Haeckel has done to it is false, or perverted, or meaningless. It is false to trace all kinds of endoderm-formation back to the invagination of the blastoderm. It is perverted to substitute a problematical gastraea in the place of the coelenterate type. The value of Huxley’s idea lay for the greater part in that it brought the early developmental stages of the higher Metazoa into immediate alliance with the completed forms of countless living Coelenterates. These latter are very diverse among themselves and still remain Coelenterates; that greater differences must exist between an adult coral and the larval form of an annelid is understandable, and will hinder no one from seeing the essential similarity of both organizations. In any case it was unnecessary to leave the Present and to descend into the Laurentian night to call forth as lean an animal spectre as the gastraea. It is, incidentally, obvious that the gastraea is not able by itself to create the slightest hypothetical conception of the unknown origin of the Coelenterates, because the gastraea z'.r nothing more than the coelenterate type schematized. Courageous hypotheses-—daring conclusions-—these almost always are of service to Science. But Schemata injure her if they bring existing knowledge into an empty and warped pattern, and claim thereby to give deeper understanding. Unfortunately the gastraea was not fertile, but it was strongly infectious; it has propogated itself as Neuraea, Nephridaea, etc. and is guilty of all the Original-animals, the Trochosphaera, the Trochophora, the Originalinsect, and I know not what besides.

Meaningless is the homologising of the gut-cavity of the higher Metazoa with the endoderm-cavity of the gastraea; a hole is a hole anywhere in the world. If once the equivalence of the walls is established, people will not need to worry their heads about the empty spaces inside.

Early Objections to the Germ-Layer Doctrine

Rudolf Albert Von Kölliker
Rudolf Albert Von Kölliker (1817-1905)
Oscar Hertwig
Oscar Hertwig (1849-1922)(Photo; by Tokuzo Kimura)
Richard Hertwig
Richard Hertwig (1850-1937) (Photo. by Tokuzo Kimura)

The first voices were raised against the germ-layer doctrine during the 1870’s. Since those of Kolliker and the Hertwigs were loudest, and since the precepts of these authors epitomized those of their contemporaries, they may best be chosen as an example of the mode of reasoning of the opponents of the doctrine.

Kolliker (1879, ’84, ’89) questioned the validity of the doctrine principally from the histologist’s point of view. While some of his reasoning now seems quaint, and some has since been invalidated by modification of the doctrine, some is still cogent. He claimed that the outer layer gives rise to many diverse types of cellsto epithelium, nervous cells, neuroglia, the pigment epithelium in the eye, etc. He added, however, basing his statement on his own observations and those of Leydig and Ranvier, that the outer germlayer could give rise to smooth musculature in the case of the sweat-glands.

So far as the middle germ-layer is concerned, Kolliker claimed that this also gave rise to too many diversified types of cell to have any meaning as a single entity. He emphasized that chorda is in some cases derived from the mesoderm and in others from endoderm — a fact without significance to us, who know that chorda and mesoderm each stem from a different group of cells, but which was at one time one of the most disputed facts of the whole doctrine. He adds further that so far as the hindmost part of the embryo is concerned, in a certain sense even the nervous system is formed from mesoderm, since the blastema from which all the hindmost structures are formed is predominantly mesodermal in origin.

So far as the endoderm is concerned, he erroneously claims that in Ampbioxu: it goes so far as to form somites and muscle and connective tissue, and that in many lower forms it produces chorda.

When Kolliker turns to a study of the Coelenterates, he claims, on the basis of his own work and that of other authors, that muscles and germ-cells, and in some cases even nervous tissue, are formed sometimes from ectoderm and sometimes from endoderm. His evidence so far as the germ—cells are concerned is invalidated by the later discovery that these are derived from neither germ-layer but from cells which were probably segregated during early development. However, his conclusions concerning the muscle are still valid. He concluded:

In consequence of all these considerations the conviction is irresistably striking that the significance of the germ-layers is not histo-physiological but morphological. If we proceed from the fact that originally all the cells of the embryo, as they are produced by cleavage, are equivalent, so we may assert the proposition that all three germ-layers possess the potency and the capacity also for trans formation into all tissues, but because of their specific morphological configurations they cannot everywhere manifest this power. (1879, S. 389.)

While many of the facts on which Kolliker based his claims have been disputed, many of them still hold true, and as a matter of fact his reasoning does not suffer even when the evidence has been invalidated. Many of Kolliker’s conclusions were admittedly based on evidence derived from the Hertwigs studies on the Coelenterates. In 1878 the Hertwigs raised their first questions about the application of the germ-layer theory in a small monograph dealing with the histology of the Medusae. Inquiring, as had Kleinenberg, into the relationships of ectoderm and endoderm to mesoderm, they concluded that what they consider mesoderm in the Medusae is simply a product of the histological differentiation of ectoderm and endoderm. In their monograph on the Actinians, published a year later (1879) as the first of their definitive “Studies on the Germ-layer Theory,” they continue their discussion, questioning the precise relationship of the two layers of the Coelenterates to the three of higher forms. On evidence that in some coelenterate groups germ-cells or musculature are derived from ectoderm and in others from endoderm, they conclude that “within particular animal groups the germ-layers have dififerentiated organologically inequivalently” (1879, p. 205). Furthermore they extend their generalizations, supporting themselves by evidence from other authors similar to that of Kolliker’s outlined above, to include the other animal groups as well as the Coelenterates:

The germ-layers are neither organological nor histological entities. It is not possible, if one knows the origin of an organ in one animal group, to carry over the result to all other animal groups. . . . Just as the capacity for transformation of individual cells, so is that of a germ-layer highly manifold, and it can express itself in the most different ways in the production of organs and tissues. (1879, pp. 216, 217.)

Here the Hertwigs have put their fingers on the whole solution of the germlayer problem; but unfortunately they were not satisfied to stop here with a constructive contribution. Instead they chose to supplant the gastraea theory, which they had destroyed, with another theory which was not only unnecessary but even more far-fetched, if possible, than its predecessor.

This they achieved by continuing the discussion of the difficulty arising from the attempts to homologize similar tissues developing from dififerent germ-layers in diploblastic and triploblastic forms, and to find any uniformity whatsoever in the mesoderm which originates and develops so widely divergently in the various animal forms. Instead of following out their original suggestion that the germlayers have wider capacities for differentiation than is usually recognized, they preferred to force the widely differing behaviour of mesoderm in different forms into a common pattern by their coelome theory (1881). According to this, the mesoderm (the mesoderm is in this paper first subdivided into mesenchyme and mesoblast; the latter term had been in use to describe the whole middle layer for many years) necessarily always arises from the endoderm, enclosing within its two layers part of the alimentary cavity as the “coelome” (Haeckel’s name for the pleuro-peritoneal cavity of Remak). According to the Hertwigs (1881, p. 112):

Ectoblast and entoblast are the primary germlayers which originate by invagination of the blastula; they are therefore always the first formed and they can be referred back to a simple stem-form, the gastraea. . . . Parietal and visceral mesoblast, or the middle germ-layers, always originate later, and arise through a pouching or folding of the entoblast; . . . They bound a new cavity, the enterocoel, which may be considered a pinched-off diverticulum of the archenteron. As the two-layered animals are derivable from the gastraea, so are the four-layered from a coelome-form.

The whole theory, as an explanation of development, can probably never be better described than it has been by Braem (1895, p. 468) who wrote: “So the coelome theory, with all of its consequence, presents one of the most glaring inconsequences to which the morphological conception of the germ-layers can lead.”

This nice statement of Braem’s formed part of a series of papers published in 1895 and entitled “What is a Germlayer?”, in which the author scrutinized the Whole germ-layer doctrine from many angles. He raised the same doubts as to its validity as had the other investigators whom we have quoted; the only solution of the problem that he could offer was that the germ-layer theory was based purely on topography, while the homologies, or rather the analogies, of the layers could be comprehended only on a physiological basis.

The Germ-Layers in Regeneration and Budding

Wilhelm His
Wilhelm His (1831-1904)

To be sure, this particular period saw the beginnings of the first attempt since His’s to deal with the problem from a physiological point of view, or at least from an experimental rather than a descriptive point of‘ view. Perhaps the word experimental is dangerously used in this connection, since in one of the first important cases the experiment was performed by nature and simply observed and interpreted by the investigator. Hjort (1894a, b), in his study of bud-formation in the Ascidians, made the accurate and cogent observation that in Botryllu: organs are not formed from the same germ-layers in egg-development and in budding. In development from the egg, for instance, the atrial chamber and the ganglion are each formed from the ectoderm in Botrylus the bud they are each derived not from the outer but from the inner layer. Hjort concluded that the layers of the bud are not germ-layers in the ordinary sense, but were composed of still indifferent material.

The import of this discovery was at once appreciated as jeopardizing the validity of the doctrine. Heider (1897), in a paper that like Braem’s was entitled with a rhetorical question never answered, “Is the Germ-layer Doctrine Shattered?”, insisted that findings like those of Hjort were irrelevant so far as the germ-layer doctrine in embryology was concerned.

The doctrine, according to Heider, still holds for embryology, where it belongs, whether or not it holds true for the cases of budding and regeneration which are problems separate from those of embryOlogy.

A more convincing argument, however, had been promulgated the year before by E. B. Wilson (1896) in a brilliant lecture at Woods Hole on “The Embryological Criterion of Homology." Wilson maintained that the conditions in bud-formation and regeneration were vitally significant for a comprehension of the processes in the embryo. He wrote (pp. 111—113):

It may be urged that in regeneration and agamogenesis development is condensed and abbreviated so as no longer to repeat the phyletic development, and this is no doubt true. This explanation contains, however, a fatal admission; for if secondary modification may go so far as completely to destroy the typical relationships between the germ-layers and the parts of the adult, then those relationships are not of an essential or necessary character, and we cannot assume that the germ-layers have any fixed morphological value, even in the gastrula.

First Experimental Attack on the Problem

While philosophically speaking it may have seemed difficult to their contemporaries to choose between the interpretations of Heider and Wilson, even before their papers had been published the first experiment had been performed along the lines of the later ones which were finally to throw the balance to the side of Wilson. During 1892-93 Herbst subjected echinoderm eggs to treatment with a large variety of salts. He found that lithium had a specific effect on their development, namely the production of exogastrulae and entogastrulae in which the amount of endoderm in the embryo is greatly increased at the expense of the ectoderm.

In figures 11, 6, 12-14, there thus occurs a gradual increase of endoderm, and hand in hand with it a successive reduction of ectoderm. In figure 13 the latter is present only as the small button labelled g4, and in figure 14 it is no longer present at all; here the whole blastula wall has been transformed to endoderm (1893, p. 144).

Herbst appreciated the implication of his results; not so did his contemporaries who did not wish to. Heider, for instance, in the paper referred to above, mentioned the lithium-embryos only as a possible explanation of how in the production of endoderm in Coelenterates the processes of multipolar migration and delamination might have been derived from polar migration.

The Pathologists and the Germ-Layer Doctrine

Indeed the main body of embryologists (cf. Sedgwick, 1910) went their way, promulgating the germ-layer doctrine and attempting always to support and strengthen it. One other group of scientific investigators discussed the problem avidly: the pathologists, who were seeking panaceas to solve the atypical growth problem. Some of them, perhaps most notably Marchand (1899-1900), unsuccessfully sought the aid of the germlayer doctrine. They contributed more to the embryologists, however, than they received from them, since they could furnish evidence both pro and con. Not the most influential, but perhaps one of the most interesting results of the cooperation between pathology and embryology, was the prophecy of the experiment, and its results, which have been most instrumental in abolishing the notion of the fixity of the layers. C. S. Minot in a paper on the embryological basis of pathology wrote, in 1901 (p. 485): “It seems quite probable to me that the cells of the germ-layers are at first quite indifferent, so that if it were possible to graft a young mesodermal cell on to the ectoderm or endoderm, it would become a true ectodermal or endodermal cell, as the case may be.” A similar experiment, performed by Mangold a quarter of a century later, has been considered one of the crucial experiments in the demonstration of the non-specificity of the germ-layers.

Modern Experimental Work

1. Vertebrates

Hans Spemann
Hans Spemann (1869-1941)

The turn of the century saw the embryologists change their method of attack from observation to actual operative manipulation. The capacity of the germlayers for differentiation could now be tested as well as inferred. We may first discuss in this connection the results of the work on vertebrate embryos.

The interpretation of the experiments on the germ-layers of the Vertebrates Was facilitated, in fact made possible, by the background work of Vogt (197.5) who charted out on the amphibian blastula, by means of local vital staining, the precise locations of the areas later to become chorda, mesoderm, gut, epidermis and nervous system. Once the position of these cell-groups before gastrulation was known, their behavior in unusual positions, or in isolation, could be appreciated.

Perhaps the most far-reaching results on the activity of the germ-layers, established experimentally, were those which demonstrated the influence of the lower invaginated layers of the amphibian embryo on the differentiation of the overlying ectoderm. Spemann had, in 1918, as is well known, noted the power of transplanted dorsal lip of the blastopore of the amphibian gastrula to induce the formation of a new embryo from presumptive epidermis; in 1924, in an investigation carried out with the collaboration of Hilde Mangold, similar experiments were performed using hosts and donors of different species whose tissues were sharply distinguishable from each other in sectioned material. From the results of this experiment it became apparent that the medullary plate in the induced embryo was formed by the host, while the underlying gut, chorda and mesoderm were furnished in part by the grafted dorsal lip. This result, and others of Spemann’s (1918) from experiments in which presumptive epidermis was exchanged with presumptive medullary plate, or in which areas of medullary plate and mesoderm of the neurula were rotated through 180° (1912), suggested that the lower layers were somehow responsible for the differentiation of the upper. This was crucially and definitely demonstrated by Marx, who found, in 197.5, that a piece of already invaginated archenteron roof, implanted into the blastocoele cavity, induced the formation of medullary plate from presumptive epidermis. Subsequently Bautzmann (1916, '18)was able to show by similar transplantation experiments that both presumptive chorda and presumptive mesoderm display the power of inducing the overlying epidermis to differentiate medullary plate.

All of these transplantation experiments suggest immediately the interpretation that the underlying layers, particularly mesoderm and chorda, as the inducing system (to use a term later introduced), are somehow inherently different in their capacities than the responding system of the overlaying ectoderm. This interpretation has been in one way borne out by results of explantation experiments, and by the production of amphibian exogastrulae. When amphibian embryos, especially axolotls, are treated with appropriate salt solutions, the mesoderm, endoderm and chorda roll outwards instead of inwards, with the result that the ectoderm remains simply an empty bag. In such exogastrulae, the mesoderm and endoderm and chorda self-differentiate to form somites, gut and notochord histologically quite typical of the normal embryo; the ectoderm, on the other hand, deprived of the proximity of these layers, forms only epidermis and never nervous system (Holtfreter, 1933).

Similarly in explantation experiments, in which isolated parts of the amphibian gastrula are cultured in salt solutions, as shown by Holtfreter’s (1938 a, b) masterly work, isolated ectoderm fails to differentiate nervous tissue; it forms only epidermis when deprived of influence of cells of the other layers. The presumptive endoderm, whose role has been least clearly analyzed in the transplantation experiments, when isolated in salt solution self-differentiates only endodermal structures.

Cells of the presumptive chorda and mesoderm regions, in contrast to those of the other two germ-layers, readily overstep the classical bounds of the germlayers, and when isolated in salt—solution can self—differentiate, or induce from their own cells, according to interpretation, medullary plate and epidermis on the one hand, and gut on the other, in addition to forming the usual mesodermal structures.

If the experiments are carried out, however, by placing the isolates in viva rather than in vitro, the ectoderm and endoderm can accomplish far more than in salt solution. This has been demonstrated by Kusche (1929) and Bautzmann (1929), who placed the tissues into the empty orbital cavity of older larvae, and Holtfreter (1929) who implanted them into the abdominal cavity of amphibian larvae. In these cases, presumptive ectoderm formed not only medullary tube, but also notochord and muscle, and presumptive endoderm was able to differentiate notochord. Here where the cells were subjected to influences of highly complex organic nature, even ectoderm and endoderm were able to differentiate structures normally formed by the other germ-layers. The nature of these external influences and of their action is unknown. But their eflect is sufficient to demonstrate that the differentiating cells themselves have a wider capacity for diversification than the other experiments had suggested.

These are by no means the only clearcut experiments demonstrating the variety of potencies expressible by the cells of the amphibian gastrula. Mangold had in 1923 performed the experiment postulated twenty years before by Minot and thus dealt the germ-layer doctrine one of its most mortal blows. He found that presumptive ectoderm formed somites, chorda, pronephros, etc. when grafted into appropriate regions. Furthermore, Bruns (1935) showed that when large defects were made in the presumptive ectoderm of amphibian gastrulae, the medullary plate could be formed from presumptive mesoderm. Lopaschov (1935) also showed that when several explants of presumptive mesoderm fused they frequently produced medullary plate. Lehmann (1937) has shown that lithium has a specific meso dermizing efiect on presumptive notochord material in the amphibian gastrula.

Similar demonstrations have been made on other forms than amphibians. Hunt (1937 a and b) has shown for the chick that after removal of the presumptive endoderm the mesoderm can form gut, and indeed that even normally it makes a large contribution to the formation of this structure. It also has been shown for the fishes (Oppenheimer, 1938) that presumptive mesoderm can differentiate nervous structures under certain conditions.

The perfectly valid objection can be raised that in all the cases enumerated the cells whose accomplishments are being studied have been observed acting under highly abnormal conditions, and though it does not necessarily invalidate the results of the experiments, this is to a large extent true. There are, however, some striking cases in vertebrate development where, even in the intact embryo, cells of one germ-layer form structures usually contributed by the others. One of these, touched on by Kolliker (1884), is in tail-formation in the Vertebrates. Kolliker had postulated that all the structures in the tail of a vertebrate embryo were formed from a blastema that was primarily mesodermal in origin. Such is not precisely the case in the amphibian embryo, but the actual conditions as they are support K6lliker’s conclusion very strongly. The notochord of the amphibian embryo’s tail is formed as a prolongation of that of the trunk. The tail somites, however, as demonstrated conclusively by the vital staining experiments of Bijtel (1931), are formed by the posterior portion of the medullary plate itself. Here is a clearcut case in normal development where typically mesodermal structures are formed by cells ectodermal in origin.

Another such case is shown by the behaviour of the cells of the neural crest. The history of the work on this problem has been fully reviewed by Harrison and therefore need only be summarily discussed here. [All references concerning the neural crest cited here may be obtained The neural crest, as every one knows, is clearly an ectodermal derivative in the sense of the germ-layer doctrine. However, it was demonstrated even in the last century that it furnished mesenchyme (Kastschenko and Goronowitsch), and, in 1894, it was given, together with the branchial sensory placodes which make a similar contribution, the name of mesectoderm, by von Kupffer. In 1897 Miss Platt showed, on morphological grounds, that the branchial skeleton was derived from mesectoderm, The recent experimental evidence has demonstrated that the neural crest unquestionably forms the Schwann sheath-cells, the spinal ganglia, part of the cranial ganglia, the branchial skeleton, mesenchyme, melanophores and Xanthophores, possibly the ganglia of the sympathetic nervous system, and probably the pia—arachnoid membranes.

2. Invertebrates

Curt Herbst
Curt Herbst (1866—1946) (Photo. by Tokuzo Kimura)

Such varied accomplishments of the germ-layers are characteristic not only of the vertebrates, but of invertebrates as well. Probably as many single isolated cases could be enumerated for the Invertebrates as have been for the Vertebrates, but here we shall confine our remarks to two groups.

One of the most clear-cut cases imaginable of the transformability of the germlayers has been presented by Penners (197.6, 37a and b, '38) in his beautiful studies of the development of Tzebifex. The Tzebifex egg, as described by Penners (199.4) exhibits the spiral cleavage characteristic of the annelid egg. It is further characterized by the presence at its animal and ventral poles of a special "pole-plasm” which passes during cleavage into the cells 7_d and 4d which give rise respectively to the ectodermal and mesodermal germbands of the embryo. If the pole-plasm is eliminated or divided the cells 9_d and 4d and subsequently the germ-bands fail to form or are doubled, as the case may be. Penners (199.6) showed that if 2_d or 4d were excluded from development, ectodermal or mesodermal germ-bands respectively failed to form; each type of germ-band, however, could dififerentiate apparently normally in the absence of the other (i.e. when 7_d was removed and 4d left intact or vice versa). Later (1937a) he performed similar experiments, allowing the worms to develop to late stages, and the results here were extraordinarily interesting. In the absence of the ectodermal germ-bands, the mesoderm can later form all the organs usually formed by the ectodermal—central nervous system, circular musculature, lateral line, and the ectodermal portion of the seta-sacs. If, however, the source of the mesodermal germ-bands is removed and the ectodermal left intact (Penners, 1937b) the organs normally formed by the mesoderm are not replaced. It is nevertheless extremely interesting and important that in a form characterized by a relatively highly mosaic development the organs formed normally by one germ-layer can be formed by another.

The modern work on the Echinoderms has shown that in this form also the germ layers have great adaptability. This was first suggested by Herbst’s (1892-93) chemical experiments; it has been demonstrated repeatedly in defect- and transplantation experiments. The Riinnstroms (1918-19) showed, for instance, that animal halves of a holothurian egg, containing only presumptive ectoderm, could differentiate the coelome Which is normally formed by mesenchyme, and subsequently Horstadius (197.8) has shown that coelome may be formed in regeneration by ectoderm, endoderm or mesoderm. Recently the chemical and defect- and transplantation experiments have been extended, and they have been amplified and reinterpreted in the light of physiological studies on the developing egg and its parts in such a Way that our picture of the developing echinoderm egg is as complete as any We have in embryology.

In the early defect-experiments, Driesch (1891—97, 1892-93, 1900) erroneously supposed all the parts of the developing echinoderm to be totipotent. This problem Was clarified by Horstadius (197.8, '35) Who showed that isolated animal halves of eggs, Which contain only presumptive ectoderm, cannot gastrulate or form endoderm, While isolated vegetal halves can gastrulate and form plutei; sometimes the plutei have an overabundance of endoderm, and sometimes the isolated vegetal halves form exogastrulae With very large guts. The results of these experiments, and of similar ones involving smaller parts of the egg, gained significance With the publication of von Ubisch’s (1933) paper Which demonstrated on the basis of vital staining experiments the precise normal roles of the various portions of the egg. The animal half of the egg, consisting at the 39.-cell stage of a dorsal group of 8 cells (anl) and a ventral group of 8 (an2), forms the ectoderm of the dorsal surface of the pluteus. The vegetal half is divided at the 64-cell stage into a dorsal ring of 8 cells Cvegl) Which forms aboral ectoderm, a more ventral ring of 8 cells (veg?) Which forms the endoderm, and at its most ventral pole the micromeres Which form the mesenchyme. Driesch (1893) kneW that the micromeres formed the mesenchyme, and all the workers previous to von Ubisch appreciated that the animal pole represented presumptive ectoderm. The main important point demonstrated by von Ubisch Was that the upper part of the vegetal half also contained presumptive ectoderm.

The transformability of the germ layers has been fully demonstrated by Horstadius (1935) in his ingenious experiments of separating and recombining the cells Whose normal behaviour is Well knoWn. He has shoWn, for instance, that When veg2, Which comprises the presumptive endoderm and normally forms gut, is eliminated from development, a normal pluteus results Whose gut is formed by veg1 Which is composed of the presumptive ectoderm for the aboral surface. If both veg1 and veg2 are eliminated, again a normal pluteus forms Whose gut is made by ectoderm of the animal half of the egg. In both of these cases the micromeres are considered to induce the ectoderm to become endoderm; but no matter What the mechanism, the result is that presumptive ectoderm forms endodermal structures. Similarly, micromeres implanted into the animal pole of Whole eggs induce the formation of accessory gut from presumptive ectoderm. Entodermizing of presumptive ectodermal material occurs also When blastomeres are separated and recombined in such a Way that the proportion of presumptive ectoderm to presumptive endoderm is far greater than usual; for instance, When a meridional half of an egg is combined With an animal half.

The transformability of the presumptive endodermal cells has also been shoWn by Horstadius (1928). An isolated veg2 group, for instance, is able to form ectoderm, though to be sure such ectoderm is slightly abnormal, as shoWn by the fact that it forms no stomodaeum nor ciliated band, and the skeleton, Whose arrangement depends on an interaction between ectoderm and mesenchyme, is somewhat abnormal. Driesch (1893) had known that eggs deprived of their micromeres could form normal plutei. Horstadius demonstrated that in such eggs the secondary mesenchyme, which is derived from veg2, is formed earlier than normal and serves to form the skeleton.

Fortunately in the case of the echinoderm egg the physiological basis for the transformability of the germ-layers is being studied. Herbst (1892, '93) had shown that in whole eggs presumptive ectoderm could be caused to differentiate endoderm by the action of lithium. Von Ubisch (1919) showed that isolated animal halves, consisting only of presumptive ectoderm, which normally form no endoderm and fail to gastrulate, could accomplish both these tasks after treatment with lithium. Horstadius (1936) has shown, by varying the length of treatment and by using parts of eggs isolated for varying lengths of time, that the action of the lithium on the presumptive ectoderm is strikingly similar to that of implanted micromeres.

This work, beautiful in itself, has gained considerably in significance through the work of Lindahl and his co-workers (Lindahl, I936; Lindahl and Stordal, 1937; Lindahl and Ohman, 1938). Lindahl suggested originally that in the whole egg the animal half, the presumptive ectoderm, exhibits a higher respiratory rate than the vegetal. Similarly, eggs “animalized” by chemical treatment have a higher respiratory rate than isolated vegetal halves. The action of lithium inhibits respiration, so that the presumptive ectoderm which is made to form endoderm simultaneously decreases its respiratory rate. Conversely, the presence of NaSCN or the absence of sulphate ion stimulates respiration and ectodermalizes presumptive endoderm. There are probably two systems of respiration involved at the two poles of the egg which act in a way synergistically, and the system at the animal pole is probably concerned with carbohydrate metabolism.

Lindahl and Holter (1938, unpublished) have been unable to find confirmatory evidence of Lindahl’s original statement that the respiratory rate is higher in the animal than in the vegetal portion of the egg. In any case, no matter what the precise nature of the respiratory systems, the transformation of ectoderm to endoderm, and vice versa, has been shown definitely to have a metabolic as well as a morphological basis.


The only conclusion that can be maintained, as a result of all the experiments that have been enumerated, is that the doctrine of the absolute specificity of the germ-layers as enunciated in the last century must be abandoned. There are no doubt countless cases where in a specific animal form, the cells of one germ-layer cannot alone perform the functions characteristic of another, as for instance is the case with the ectodermal germ-bands of Tzzbifex (Penners, 1937b). There are however so many contrary accomplishments that even in the classical cases the differentiation of the cells must be based on other factors than their derivation from a specific layer. The nature of such factors probably varies in every instance. In some cases, such as the Tzzbifex egg, the constitution of the cytoplasm before cleavage plays an important role. In other cases, as with the echinoderm egg, the special metabolism of parts of the egg is of decisive importance. The precise topographical position of the cells is often significant, as in the case of amphibian development. The interactions of various cells one with another are of vital importance in controlling their differentiation in the vertebrates.

If so many factors other than the origin of the cells from particular germ-layers are of such vital importance, the question arises:— What is the significance of the germ-layers, if any? No matter what the precise factors involved, it seems certain that the precise location of a cell during gastrulation in many forms, or the precise origin of its cytoplasm from the egg in others, is in many cases correlated with the type of its later activity; therefore in a certain sense the germ-layers are of topographic significance, since the cells pass through them in their orderly progression of movements. In a teleological sense, formation of germ-layers seems to be the embryo’s method of sorting out its constituent parts. The essential point is, however, that this method is not the only method that the embryo can call upon to attain a specific end, and here as in many other cases in development the embryo can, when necessary, modify or abandon one method in favor of another. This point, anticipated by Kolliker (1879), as we have already shown, has been well stated in the Brachet textbook (p. 296): “In reality, the germ-layers, like the blastomeres, have an actual potentiality and a total potentiality; the former is what they normally become; the latter what they are capable of forming in addition under diverse natural or experimental influences.”

The task of the student of the germlayers then must become more than an attempt to discern how the embryo sorts its cells into one layer or another, it must become an elucidation of how the wide potencies of the germ-layers become subject to limitation to their normal accomplishment. Pander, who was first to describe the germ-layers, was fortunate and wise in emphasizing the interactions of the layers with each other. We should do well to emulate him, for only when we can more appreciate the manner and mechanisms of such interactions shall we understand the true significance of the germ-layers themselves.

  • The portraits of Karl Ernst von Baer and Rudolf Albert von Kolliker are reproduced from Nordenskiold’s History of Biology, 1928, with the permission of the publisher Alfred A. Knopf, New York.

List of Literature

Papers with comprehensive bibliographies on the germ-layer doctrine are marked with an asterisk.

ALLMAN, G. 1853. On the anatomy and physiology of Cord)/loplvordz A contribution to our knowledge of Tubularian zoophytes. Plail. Trnny. Roy. Soc. London, vol. 143, pp. 367-384.

VON BAER, K. E. 187.8-37. Ueber Entwicklungs— geschichte der Thiere. Beobachtung und Reflexion. Konigrberg.

BALFOUR, F. M. 1873. The development and growth of the layers of the blastoderm. Quart. ]. Micr. Sci., vol. 13, pp. 7.66-7.76.

BAUTZMANN, H. 197.6. Experimentelle Untersuchungen zur Abgrenzung des Organisationszen— trums bei Triton tnenintns, mit einem Anhang: Ueber Induktion durch Blastulamaterial. Arc/9. f. Entw.-nzeclx, Bd. 1o8, S. 7.83—37.1.

1928. Experimentelle Untersuchungen fiber die Induktionsféihigkeit von Chorda und Mesoderm bei Triton. Arc/9. f. Entw.—rnec/9., Bd. 114, S. 177—7.7.5.

1929. Ueber bedeutungsfremde Selbstdifferenzierung aus Teilstiicken des Amphibienkeimes. Nntitrwimx, Jahrg. 17, S. 818-87.7.

BIJTEL, 1931. Ueber die Entwicklung des Schwanzes bei Amphibien. Arc/9. Entw.—rnec/9., Bd. 17.5, S. 448-486.

BRACHET, A. 1935. Traité d’embryologie des vertebrés. Seconde édition revue et completée par A. Dalcq et P. Gerard. Paris.

BRAEM, F. 1895. Was ist ein Keimblatt? Biol. Centr4ZbZ., Bd. 15, S. 47.7—443, 466-476, 491—5o6.

BRUNS, E. 1931. Experimente fiber das Regulationsvermogen der Blastula von Triton tnenintm und Bornbinntor pnclvypiu. Arc/9. Eifill/.-7728617., Bd. 17.3, S. 687.-718.

CUVIER, G. 1846. Legons d’anatomie comparée. Tome huitieme contenant les organes de la génération et des sécretions, avec une legon complémentaire des organes de relations; par G. Cuvier et G.—L. Duvernoy. 7.ieme édition, corrigée et augmentée. Paris.

DRIESCH, H. 1891-92. Entwicklungsmechanische Studien. I. Der Werth der beiden ersten Furchungszellen in der Echinodermenentwick— lung. Experimentelle Erzeugung von Theil—und Doppelbildung. II. Ueber die Beziehungen des Lichtes zur ersten Etappe der thierischen Formbildung. Zeit. win. Zool., Bd. 53, S. 160-189.

1897.-93. Entwicklungsmechanische Studien. III. Die Verminderung des Furchungsmaterials und ihre Folgen (Weiteres ueber Theilbildungen). IV. Experimentelle Veranderungen des Typus der Furchung und ihre Folgen (Wirkungen von Warmezufuhr und von Druck). V. Von der Furchung doppelbefruchteter Eier. VI. Ueber einige allgemeine Fragen der theoretischen Morphologie. Zeit. win. ZooZ., Bd. 55, S. 1-67..
1893. Entwicklungsmechanische Studien. VII. Exogastrula und Anenteria (iiber die Wirkung von Warmezufuhr auf die Larvenentwick— lung der Echiniden). VIII. Ueber Variation der Mikromerenbildung (Wirkung von Verdunnung des Mcerwassers). IX. Ueber die Vertretbarkeit der “An1agen" von Ektoderm und Entoderm. X. Ueber allgemeine entwicklungsmechanische Ergebnisse. Mitt. zool. Stat. Neapel, Bd. 11, S. 2.2.1-2.54.

Dnriesch, H. 1900. Die isolirten Blastomeren des Echinidenkeimes. Eine Nachpriifung und Erweiterung friiherer Untersuchungen. Arab. f. Entw.-meclJ., Bd. 10, S. 361-410.

HAECKEL, E. 1866. Generelle Morphologie. Berlin.

1872.. Die Kalkschwéimme. Eine Mono graphie. Berlin.

Harrison, R. G. 1938. Die Neuralleiste. Erganzheft. zum Anat. Anz., Bd. 85, S. 3-30.

HEIDER, K. 1897. Ist die Keimblattlehre erschiittert? Sclmberg: Zool. Centralbl., Jahrg. 4, S. 77-5-737

Herbst, C. 1892.. Experimentelle Untersuchungen iiber den Einfluss der veranderten chemischen Zusammensetzung des umgehenden Mediums auf die Entwicklung der Thiere. I. Theil. Vcrsuche an Seeigeleiern. Zeit. win. Zool., Bd. 55, S. 446-518.

1893. Experimentelle Untersuchungen iiber den Einfluss der veranderten chemischen Zusammensetzung des umgebenden Mediums auf die Entwicklung der Thiere. II. Theil. Weiteres iiber die morphologische Wirkung der Lithiumsalze und ihre theoretische Bedeutung. Mitt. Zool. Stat. Neapel, Bd. 11, S. 136-220.

HERTWIG, O. 1881. Die Entwicklung des mittleren Keimblattes der Wirbelthiere. Jena.

1906. “Die Lehre von den Keimblattern.” In 0. HERTWIGs Handbuch der vergleichenden und experimentellen Entwickelungslehre der Wirbeltiere. Bd. 1, Theil 1, S. 699-967. ]ena.
, AND R. 1878. Der Organismus der Medusen und seine Stellung zur Keimblattertheorie. jena.
, AND ———-. 1879. Studien zur Bléittcrtheorie. Heft I. Die Aktinien anatomisch und histologisch mit besonderer Beriicksichtigung des Nervenmuskelsystems untersucht. jena.
, AND ——. 188o. Studien zur Bliittertheorie. Heft II. Die Chaetognathen. Ihre Anatomie, Systematik und Entwickelungsgeschichte. jena.
, AND . 1881. Studien zur Blattertheorie. Heft IV. Die Coelomtheorie. Versuch einer Erklarung des mittleren Keimblattes. jena.

HERTWIG, R. 1880. Studien zur Blattertheorie. Heft III. Ueber den Bau der Ctenophoren. Jena.

His, W. 1865. Die Haute und H6hlen des Korpers. Basel.

1874. Unsere Korperform und das physio logische Problem ihrer Enstehung. Leipzig.

HJORT, J. 1893. Ueber den Entwicklungscyclus der zusammengesetzten Ascidien. Mitt. zool. Stat. Neapel, Bd. 10, S. 584-618.

1895. Beitrag zur Keimblatterlehre und Entwickelungsmechanik der Ascidienknospung. Anat. Anz., Bd. 1o, S. 2.15-2.2.9.

HORSTADI, S. 1928. Ueber die Determination des Keimes bei Echinodermen. Acta Zool., vol. 9, pp. 1-191.

1935. Ueber die Determination im Verlaufe der Eiachse bei Seeigeln. Pabbl. Staz. Zool. Nap., vol. 14, pp. 253-429.
1936. Ueber die zeitliche Determination im Keim von Paracentrota: livida: Lk. Arab. Entw.-meclJ., Bd. 135, S. 1-39.

HOLTFRETER,J. 192.9. Ueber die Aufzucht isolierter Teile des Amphibienkeimes. I. Methode eine Gewebezuchtung in vivo. Arab. f. Entw.-mecl2., Bd. 117, S. 422-510.

1933. Die totale Exogastrulation, eine Selbstablosung des Ektoderms vom Entomesoderm. Entwicklun g und funktionelles Verhalten nervenloser Organe. Arab. f. Entw.-mecl2., Bd. 12.9, S. 670-793.
1938a. Dilferenzierungspotenzen isolierter Teile der Urodelengastrula. Arch. f. Entw.-mecl2., Bd. 138, S. 522-656.
1938b. Dilferenzierungspotenzen isolierter Teile der Anurengastrula. Arch. f. Entw.-meel2., Bd. 138, S. 657-738.

HUNT, T. E. 1937a. The development of gut and its derivatives from the mesectoderm and mesentoderm of early chick blastoderms. Anat. Rec., vol. 68, pp. 349-369.

1937b. The origin of entodermal cells from the primitive streak of the chick embryo. Anat. Rea, vol. 68, pp. 449-459.

HUXLEY, T. H. 1868-96. “The physical basis of life. In Method and Results, New York, 1896. (Lecture delivered in 1868.)

1849. On the anatomy and affinities of the family of the Medusae. Phil. Tram. Roy. Soc. London, vol. 139, pp. 413-434.
1869. An Introduction to the Classification of Animals. London.
1871-72. A Manual of the Anatomy of Vertebrated Animals. New York, 1872. (London, 1871).

KLEINENBERG, N. G, N. I87 . Hydra. Eine Monographie. Leipzig. 1886. Die Entstehung des Annelids aus der Larve von Lopadorbyncm. Nebst Bemerkungen iiber die Entwicklung anderer Polychaeten. Leipzig. 1872.

VON KDLLIKER, A. 1879. Entwickelungsgeschichte des Menschen und der hoheren Thiere. Zweite ganz umgearbeitete Auflage. Leipzig.

1884. Die embryonalen Keimbliitter und die Gevvebe. Zeiz‘. wire. Zool., Bd. 40, S. 179-113.
1889. Handbuch der Gewebelehre des Menschen. 6. umgearbeitete Auflage. Erster Band: Die allgemeine Gewebelehre und die Systeme der Haut, Knochen und Muskeln. Leipzig.

KOWALEWSKI, A. 1867a. Entwickelungsgeschichte des Arnpbioxnr lanceolatnr. Mern. ele l’acad. de St. Piterrboarg, VIIe Série, T. 11, No. 4.

1867b. Beitréige zur Entwicl<e1ungsgeschichte der Holothurien. Mem. ele l'acael. ole St. Pétersbonrg, VIIC Série, T. 11, N0. 6.
1869-71. Embryologische Studien an Wiirmern und Arthropoden. Mem. de l’aead. de St. Pirerrbozerg, VIIe Série, T. 16, No. 11.

KUSCHE, W. 1919. Interplantation umschriebener Zellbezirke aus der Blastula und Gastrula von Amphibien. Arcb. f. Enrw.-rnecb., Bd. 110, S. 191-171.

LANKESTER, E. R. 1873. On the primitive celllayers of the embryo as the basis of genealogical classification of animals, and on the origin of vascular and lymph systems. Ann. and Mag. Nat. Hist, Series 4, vol. 11, pp. 311-338.

1877. Notes on the embryology and classification of the animal kingdom: comprising a revision of speculations relative to the origin and significance of the germ-layers. Quart. J. Micr. Sci., vol. 17, pp. 399-454.

LEHMANN, F. E. 1937. Mesodermisierung der priisumptiven Chordamaterials durch Einvvirkung von Lithiumchlorid auf die Gastrula von Triton alperrtrir. Arab. Entw.-n2ecb., Bd. 136, S. 111-146.

LINDAHL, P. 1936. Zur Kenntnis der physiologischen Grundlagen der Determination im Seeigelkeim. Acta Zool., vol. 17.

, AND L. O. Dhmann. 1938. Weitere Studien fiber Stoffwechsel und Determination im Seeigelkeim. Biol. Zentralbl., Bd. 58, S. 179-118.
, AND AXE STORDAL. 1937. Zur Kenntnis des vegetativen Stoffvvechsels im Seeigelei. Arcb. f. Entw.-rnecb., Bd. 136, S. 44-63.

LOPASCHOV, G. 1935. Die Entvvicklungsleistungen des Gastrulamesoderms in Abhiingigkeit von Veréinderungen seiner Masse. Biol. Zentralbl., Bd. 55, S. 606-615.

MANGoLD, O. 1913. Transplantationsversuche zur Frage der Spezifitéit und der Bildung der Keimbléitter in der Entvvicklung. Arcb. f. mikr. Anat. nnol Ent.-ge.rcb., Bd. 100, S. 198-301.

MARCHAND, F. 1899. Ueber die Beziehungen der pathologischen Anatomic zur Entwickelungsgeschichte, besonders der Keimblattlehre. Verb. el. dentrcb. patbolog. Gerellrcb. Mzincben, Jahrg. 1, S. 38-107.

MARX, A. 1915. Experimentelle Untersuchungen zur Frage der Determination der Medullarplatte. Arcb. f. Entw.-rnecb., Bd. 105, S. 1o-44.

MINOT, C. S. 1901. The embryological basis of pathology. Sci., N. S., vol. 13, pp. 481-498.

OPPENHEIMER, J. M. 1938. Potencies for differentiation in the teleostean germ-ring. ]. Exp. Zaol., vol. 79, pp. 185-111.

PANDER, C. 1817a. Beytriige zur Entwickelungs geschichte des Hiihnchens im Eye. Wzirzbnrg.

1817b. Dissertatio inauguralis, sistens historiam metamorphoseos, quam ovum incubatum prioribus quinque diebus subit. Wiirzbzerg.

PENNERS, A. 1914. Experimentelle Untersuchungen zum Determinations-problem am Keim von Tnbifex rioielornm Lam. I. Die duplicitas cruciata und organbildenden Substanzen. Arcb. f. mikr. Anat. ze. Entw.-ge.rcb., Bd. 101, S. 51-100.

1916. Experimentelle Untersuchungen zum Determinations-problem am Keim von Tabifex rionlornrn Lam. II. Die Entwicklung teilweise abgetiiteter Keime. Zeit. wire. Zool., Bd. 117, S. I-I40.
1937a. Regulation am Keim von Tzebifex rivnlornm Lam. nach Ausschaltung des ekto dermalen Keimstreifs. Zeit. wire. Zool., Bd. 149, 3. 86-130.
1937b. Abhéingigkeit der Formbildung vom Mesoderm in Tubifex-Embryo. Zeit. win. Zool., Bd. 150, S. 305-357.

RATHKE, M. H. 1815. Flusskrebs. Isis von Oken, Jahrg. 1815, Theil 1, S. 1o93-1100.

1819. Untersuchungen fiber die Bildung und Entvvickelung des Flusskrebses. Leipzig.

REMAK, R. 1850-5 5. Untersuchungen fiber die Entwickelung der Wirbelthiere. Berlin.

RRUNNSTROM, J., AND S. 1918-19. Ueber die Entvvicklung von Ceecnrnaria fronelorn: Gunnerus und Proln: pbantajm: Strussenfelt. Bergen: Mm. Aarbok. Narnro. Raekke, 5; 9.

RUSSELL, E. S. 1916. Form and Function. A Contribution to the History of Animal Morphology. London.

SEDGWICK, A. 1910. “Embryology.” In Encyclopedia Brittanica, 11th edition, vol. 9, pp. 314-319. Cambridge.

SPEMANN, H. 1911. Ueber die Entvvicklung umgedrehter Hirnteile bei Amphibienembryonen. Zool. ]abrb., Suppl. 15, pp. 1-48.

SPEMANN, H. 1918. Ueber die Determination der ersten Organanlagen des Amphibienembryo I—VI. Arab. Entw.-maab., Bd. 43, S. 448-555.

, AND H. MANGOLD. 192.4. Ueber Induktion von Embryonalanlagen durch Implantation artfrcmdcr Organisatoren. Arab. f. mikr. Amt. und Entw.-ga.rab., Bd. Ioo, S. 599-638.

VON UBISCH, L. 192.9. Ueber die Determination der larvalen Organe und der Imaginalanlage bei dem Seeigelkeim. Arab. 1‘. Entw.-meab., Bd. 117, S. 80-127..

1933. Formbildungsanalyse an Seeigellarven. Naturwi.r.r., Jahrg. 2.1, S. 183-186.

VOGT, W. 192.5. Gestaltungsanalyse am Amphibienkeim mit ortlicher Vitalfarbung. Vorwort fiber Wege und Ziele. I. Methode und Wir kungsweise der ortlichen Vitalfarbung mit Agar als Farbtralger. Arch. 1‘. Entw.-maab., Bd. 1o6, S. 542-610.

Wolff, C. F. 1759. Theoria generationis. Hallo. 1817.. Ueber die Bildung des Darmkanals im bebriiteten Hiihnchen. Uebersetzt und mit einer einleitenden Abhandlung und Anmerkungen von Johann Friedrich Meckel. Halla. Originally published by Wolff as follows: De formatione intestinorum praecipue, tum et de amnio spurio, aliisque partibus embryonis Gallinacei nondum visis. Novi Comment. Acad. Sci. Impt. Petropol. vols. 12. and 13, 1768-69.

W1LsoN, E. B. 1894-96. The Embryological Criterion of Homology. Biological Lectures delivered at the Marine Biological Laboratory in the summer session of 1894. Boston.

Cite this page: Hill, M.A. (2019, August 24) Embryology Paper - The non-specificity of the germ-layers. Retrieved from

What Links Here?
© Dr Mark Hill 2019, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G