THE EVOLUTION OF MAN
Volume I



CHAPTER X

THE CŒLOM THEORY

The two “primary germinal layers” which the gastræa theory has shown to be the first foundation in the construction of the body are found in this simplest form throughout life only in animals of the lowest grade—in the gastræads, olynthus (the stem-form of the sponges), hydra, and similar very simple animals. In all the other animals new strata of cells are formed subsequently between these two primary body-layers, and these are generally comprehended under the title of the middle layer, or mesoderm. As a rule, the various products of this middle layer afterwards constitute the great bulk of the animal frame, while the original entoderm, or internal germinal layer, is restricted to the clothing of the alimentary canal and its glandular appendages; and, on the other hand, the ectoderm, or external germinal layer, furnishes the outer clothing of the body, the skin and nervous system.

In some large groups of the lower animals, such as the sponges, corals, and flat-worms, the middle germinal layer



[ 91 ]

 

remains a single connected mass, and most of the body is developed from it; these have been called the three-layered metazoa, in opposition to the two-layered animals described. Like the two-layered animals, they have no body-cavity—that is to say, no cavity distinct from the alimentary system. On the other hand, all the higher animals have this real body-cavity (cœloma), and so are called cœlomaria. In all these we can distinguish four secondary germinal layers, which develop from the two primary layers. To the same class belong all true vermalia (excepting the platodes), and also the higher typical animal stems that have been evolved from them—molluscs, echinoderms, articulates, tunicates, and vertebrates.


Diagram of the four secondary terminal layers.

Figs. 74 and 75Diagram of the four secondary germinal layers, transverse section through the metazoic embryo: Fig. 74 of an annelid, Fig. 75 of a vermalian. a primitive gut, dd ventral glandular layer, df ventral fibre-layer, hm skin-fibre-layer, hs skin-sense-layer, u beginning of the rudimentary kidneys, n beginning of the nerve-plates.

The body-cavity (cœloma) is therefore a new acquisition of the animal body, much younger than the alimentary system, and of great importance. I first pointed out this fundamental significance of the cœlom in my Monograph on the Sponges (1872), in the section which draws a distinction between the body-cavity and the gut-cavity, and which follows immediately on the germ-layer theory and the ancestral tree of the animal kingdom (the first sketch of the gastræa theory). Up to that time these two principal cavities of the animal body had been confused, or very imperfectly distinguished; chiefly because Leuckart, the founder of the cœlenterata group (1848), has attributed a body-cavity, but not a gut-cavity, to these lowest metazoa. In reality, the truth is just the other way about.

The ventral cavity, the original organ of nutrition in the multicellular animal-body, is the oldest and most important organ of all the metazoa, and, together with the primitive mouth, is formed in every case in the gastrula as the primitive gut; it is only at a much later stage that the body-cavity, which is entirely wanting in the cœlenterata, is developed in some of the metazoa between the ventral and the body wall. The two cavities are entirely different in content and purport. The alimentary cavity (enteron) serves the purpose of digestion; it contains water and food taken from without, as well as the pulp (chymus) formed from this by digestion. On the other hand, the body-cavity, quite distinct from the gut and closed externally, has nothing to do with digestion; it encloses the gut itself and its glandular appendages, and also contains the sexual products and a certain amount of blood or lymph, a fluid that is transuded through the ventral wall.

As soon as the body-cavity appears, the ventral wall is found to be separated from the enclosing body-wall, but the two continue to be directly connected at various points. We can also then always distinguish a number of different layers of tissue in both walls—at least two in each. These tissue-layers are formed originally from four different simple cell-layers, which are the much-discussed four secondary germinal layers. The outermost of these, the skin-sense-layer (Figs. 74, 75 hs), and the innermost, the gut-gland-layer (dd), remain at first simple epithelia or covering-layers. The one covers the outer surface of the body, the other the inner



[ 92 ]

 

surface of the ventral wall; hence they are called confining or limiting layers. Between them are the two middle-layers, or mesoblasts, which enclose the body-cavity.

Coelumula of sagitta. Fig. 76Cœlomula of sagitta (gastrula with a couple of cœlom-pouches. (From Kowalevsky.) bl.p primitive mouth, al primitive gut, pv cœlom-folds, m permanent mouth.

The four secondary germinal layers are so distributed in the structure of the body in all the cœlomaria (or all metazoa that have a body-cavity) that the outer two, joined fast together, constitute the body-wall, and the inner two the ventral wall; the two walls are separated by the cavity of the cœlom. Each of the walls is made up of a limiting layer and a middle layer. The two limiting layers chiefly give rise to epithelia, or covering-tissues, and glands and nerves, while the middle layers form the great bulk of the fibrous tissue, muscles, and connective matter. Hence the latter have also been called fibrous or muscular layers. The outer middle layer, which lies on the inner side of the skin-sense-layer, is the skin fibre-layer; the inner middle layer, which attaches from without to the ventral glandular layer, is the ventral fibre layer. The former is usually called briefly the parietal, and the latter the visceral layer or mesoderm. Of the many different names that have been given to the four secondary germinal layers, the following are those most in use to-day:—


1. Skin-sense-layer
    (outer limiting layer).
I. Neural layer
    (neuroblast).
The two secondary germinal
layers of the body-wall:
I. Epithelial.
II. Fibrous.
2. Skin-fibre-layer
    (outer middle layer).
II. Parietal layer
      (myoblast).
3. Gut-fibre-layer
    (inner middle layer).
III. Visceral layer
        (genoblast).
The two secondary germinal
layers of the gut-wall:
III. Fibrous.
IV. Epithelial.
4. Gut-gland-layer
    (inner limiting layer).
IV. Enteral layer
      (enteroblast)

The first scientist to recognise and clearly distinguish the four secondary germinal layers was Baer. It is true that he was not quite clear as to their origin and further significance, and made several mistakes in detail in explaining them. But, on the whole, their great importance did not escape him. However, in later years his view had to be given up in consequence of more accurate observations. Remak then propounded a three-layer theory, which was generally accepted. These theories of cleavage, however, began to give way thirty years ago, when Kowalevsky (1871) showed that in the case of Sagitta (a very clear and typical subject of gastrulation) the two middle germinal layers and the two limiting layers arise not by cleavage, but by folding—by a secondary invagination of the primary inner germ-layer. This invagination or folding proceeds from the primitive mouth, at the two sides of which (right and left) a couple of pouches are formed. As these cœlom-pouches or cœlom-sacs detach themselves from the primitive gut, a double body-cavity is formed (Figs. 74–76).

Coelomula of sagitta, in section. Fig. 77Cœlomula of sagitta, in section. (From Hertwig.) D dorsal side, V ventral side, ik inner germinal layer, mv visceral mesoblast, lh body-cavity, mp parietal mesoblast, ak outer germinal layer.

The same kind of cœlom-formation as in sagitta was afterwards found by Kowalevsky in brachiopods and other invertebrates, and in the lowest vertebrate—the amphioxus. Further instances were discovered by two English embryologists, to whom we owe very considerable advance in ontogeny—E. Ray-Lankester and F. Balfour. On the strength of these and other studies, as well as most extensive research of their own, the brothers Oscar and Richard Hertwig constructed in 1881



[ 93 ]

 

the Cœlom Theory. In order to appreciate fully the great merit of this illuminating and helpful theory, one must remember what a chaos of contradictory views was then represented by the “problem of the mesoderm,” or the much-disputed “question of the origin of the middle germinal layer.” The cœlom theory brought some light and order into this infinite confusion by establishing the following points: 1. The body-cavity originates in the great majority of animals (especially in all the vertebrates) in the same way as in sagitta: a couple of pouches or sacs are formed by folding inwards at the primitive mouth, between the two primary germinal layers; as these pouches detach from the primitive gut, a pair of cœlom-sacs (right and left) are formed; the coalescence of these produces a simple body-cavity. 2. When these cœlom-embryos develop, not as a pair of hollow pouches, but as solid layers of cells (in the shape of a pair of mesodermal streaks)—as happens in the higher vertebrates—we have a secondary (cenogenetic) modification of the primary (palingenetic) structure; the two walls of the pouches, inner and outer, have been pressed together by the expansion of the large food-yelk. 3. Hence the mesoderm consists from the first of two genetically distinct layers, which do not originate by the cleavage of a primary simple middle layer (as Remak supposed). 4. These two middle layers have, in all vertebrates, and the great majority of the invertebrates, the same radical significance for the construction of the animal body; the inner middle layer, or the visceral mesoderm, (gut-fibre layer), attaches itself to the original entoderm, and forms the fibrous, muscular, and connective part of the visceral wall; the outer middle layer, or the parietal mesoderm (skin-fibre-layer), attaches itself to the original ectoderm and forms the fibrous, muscular, and connective part of the body-wall. 5. It is only at the point of origination, the primitive mouth and its vicinity, that the four secondary germinal layers are directly connected; from this point the two middle layers advance forward separately between the two primary germinal layers, to which they severally attach themselves. 6. The further separation or differentiation of the four secondary germinal layers and their division into the various tissues and organs take place especially in the later fore-part or head of the embryo, and extend backwards from there towards the primitive mouth.

Section of a young sagitta. Fig. 78Section of a young sagitta. (From Hertwig.) dh visceral cavity, ik and ak inner and outer limiting layers, mv and mp inner and outer middle layers, lk body-cavity, dm and vm dorsal and visceral mesentery.

All animals in which the body-cavity demonstrably arises in this way from the primitive gut (vertebrates, tunicates, echinoderms, articulates, and a part of the vermalia) were comprised by the Hertwigs under the title of enterocœla, and were contrasted with the other groups of the pseudocœla (with false body-cavity) and the cœlenterata (with no body-cavity). However, this radical distinction and the views as to classification which it occasioned have been shown to be untenable. Further, the absolute differences in tissue-formation which the Hertwigs set up between the enterocœla and pseudocœla cannot be sustained in this connection. For these and other reasons their cœlom-theory has been much criticised and partly abandoned. Nevertheless, it has rendered a great and lasting service in the solution of the difficult problem of the mesoderm, and a material part of it will certainly be retained. I consider it an especial merit of the theory that it has established the identity of the development of the two middle layers in all the vertebrates, and has traced them as cenogenetic modifications back to the original palingenetic form of development that we still find in the amphioxus. Carl Rabl comes to the same conclusion in his able Theory of the Mesoderm, and so do Ray-Lankester, Rauber, Kupffer, Ruckert, Selenka, Hatschek, and others. There is a general agreement in these and many other recent writers that all the different forms of cœlom-construction, like those of gastrulation, follow one and the same strict hereditary law in the vast vertebrate stem; in spite of their apparent differences, they



[ 94 ]

 

are all only cenogenetic modifications of one palingenetic type, and this original type has been preserved for us down to the present day by the invaluable amphioxus.

But before we go into the regular cœlomation of the amphioxus, we will glance at that of the arrow-worm (Sagitta), a remarkable deep-sea worm that is interesting in many ways for comparative anatomy and ontogeny. On the one hand, the transparency of the body and the embryo, and, on the other hand, the typical simplicity of its embryonic development, make the sagitta a most instructive object in connection with various problems. The class of the chætogatha, which is only represented by the cognate genera of Sagitta and Spadella, is in another respect also a most remarkable branch of the extensive vermalia stem. It was therefore very gratifying that Oscar Hertwig (1880) fully explained the anatomy, classification, and evolution of the chætognatha in his careful monograph.


Transverse section of amphioxus-larvae.
Figs. 79 and 80.Transverse section of amphioxus-larvæ. (From Hatschek.) Fig. 79 at the commencement of cœlom formation (still without segments), Fig. 80 at the stage with four primitive segments. ak, ik, mk outer, inner, and middle germinal layer, hp horn plate, mp medullary plate, ch chorda, * and * disposition of the cœlom-pouches, lh body-cavity.)

The spherical blastula that arises from the impregnated ovum of the sagitta is converted by a folding at one pole into a typical archigastrula, entirely similar to that of the Monoxenia which I described (Chapter VIII, Fig. 29). This oval, uni-axial cup-larva (circular in section) becomes bilateral (or tri-axial) by the growth of a couple of cœlom-pouches from the primitive gut (Figs. 76, 77). To the right and left a sac-shaped fold appears towards the top pole (where the permanent mouth, m, afterwards arises). The two sacs are at first separated by a couple of folds of the entoderm (Fig. 76 pv), and are still connected with the primitive gut by wide apertures; they also communicate for a short time with the dorsal side (Fig. 77 d). Soon, however, the cœlom-pouches completely separate from each other and from the primitive gut; at the same time they enlarge so much that they close round the primitive gut (Fig. 78). But in the middle line of the dorsal and ventral sides the pouches remain separated, their approaching walls joining here to form a thin vertical partition, the mesentery (dm and vm). Thus Sagitta has throughout life a double body-cavity (Fig. 78 lk), and the gut is fastened to the body-wall both above and below by a mesentery—below by the ventral mesentery (vm), and above by the dorsal mesentery (dm). The inner layer of the two cœlom-pouches (mv) attaches itself to the entoderm (ik), and forms with it the visceral wall. The outer layer (mp) attaches itself to the ectoderm (ak), and forms with it the outer body-wall. Thus we have in Sagitta a perfectly clear and simple illustration of the original cœlomation of the enterocœla. This palingenetic fact is the more important, as the greater part of the two body-cavities in Sagitta changes afterwards into sexual glands—the fore or female part into a pair of ovaries, and the hind or male part into a pair of testicles.

Cœlomation takes place with equal clearness and transparency in the case of



[ 95 ]

 

the amphioxus, the lowest vertebrate, and its nearest relatives, the invertebrate tunicates, the sea-squirts. However, in these two stems, which we class together as Chordonia, this important process is more complex, as two other processes are associated with it—the development of the chorda from the entoderm and the separation of the medullary plate or nervous centre from the ectoderm. Here again the skulless amphioxus has preserved to our own time by tenacious heredity the chief phenomena in their original form, while it has been more or less modified by embryonic adaptation in all the other vertebrates (with skulls). Hence we must once more thoroughly understand the palingenetic embryonic features of the lancelet before we go on to consider the cenogenetic forms of the craniota.


Transverse section of amphioxus embryo.
Figs. 81 and 82.Transverse section of amphioxus embryo. Fig. 81 at the stage with five somites, Fig. 82 at the stage with eleven somites. (From Hatschek.) ak outer germinal layer, mp medullary plate, n nerve-tube, ik inner germinal layer, dh visceral cavity, lh body-cavity, mk middle germinal layer (mk1 parietal, mk2 visceral), us primitive segment, ch chorda.

The cœlomation of the amphioxus, which was first observed by Kowalevsky in 1867, has been very carefully studied since by Hatschek (1881). According to him, there are first formed on the bilateral gastrula we have already considered (Figs. 36, 37) three parallel longitudinal folds—one single ectodermal fold in the central line of the dorsal surface, and a pair of entodermic folds at the two sides of the former. The broad ectodermal fold that first appears in the middle line of the flattened dorsal surface, and forms a shallow longitudinal groove, is the beginning of the central nervous system, the medullary tube. Thus the primary outer germinal layer divides into two parts, the middle medullary plate (Fig. 81 mp) and the horny-plate (ak), the beginning of the outer skin or epidermis. As the parallel borders of the concave medullary plate fold towards each other and grow underneath the horny-plate, a cylindrical tube is formed, the medullary tube (Fig. 82 n); this quickly detaches itself altogether from the horny-plate. At each side of the medullary tube, between it and the alimentary tube (Figs. 79–82 dh), the two parallel longitudinal folds grow out of the dorsal wall of the alimentary tube, and these form the two cœlom-pouches (Figs. 80, 81 lh). This part of the entoderm, which thus represents the first structure of the middle germinal layer, is shown darker than the rest of the inner germinal layer in Figs. 79–82. The edges of the folds meet, and thus form closed tubes (Fig. 81 in section).

During this interesting process the outline of a third very important organ, the chorda or axial rod, is being formed between the two cœlom-pouches. This first foundation of the skeleton, a solid cylindrical cartilaginous rod, is formed in the middle line of the dorsal primitive gut-wall, from the entodermal cell-streak that remains here between the two cœlom-pouches (Figs. 79–82 ch). The chorda appears at first in the shape of a flat longitudinal fold or a shallow groove (Figs. 80, 81); it does not become a solid cylindrical cord until after separation from the primitive gut (Fig. 82). Hence we might say that the dorsal wall of the primitive gut forms three parallel longitudinal folds at this important period—one single fold and a pair of folds. The single middle fold becomes the chorda, and lies immediately below the groove of the ectoderm, which becomes the medullary



[ 96 ]

 

tube; the pair of folds to the right and left lie at the sides between the former and the latter, and form the cœlom-pouches. The part of the primitive gut that remains after the cutting off of these three dorsal primitive organs is the permanent gut; its entoderm is the gut-gland-layer or enteric layer.


Chordula of the amphioxus.
Figs. 83 and 84Chordula of the amphioxus. Fig. 83 median longitudinal section (seen from the left). Fig. 84 transverse section. (From Hatschek.) In Fig. 83 the cœlom-pouches are omitted, in order to show the chordula more clearly. Fig. 84 is rather diagrammatic. h horny-plate, m medullary tube, n wall of same (n' dorsal, n" ventral), ch chorda, np neuroporus, ne canalis neurentericus, d gut-cavity, r gut dorsal wall, b gut ventral wall, z yelk-cells in the latter, u primitive mouth, o mouth-pit, p promesoblasts (primitive or polar cells of the mesoderm), w parietal layer, v visceral layer of the mesoderm, c cœlom, f rest of the segmentation-cavity.

Chordula of the amphibia (the ringed adder).
Figs. 85 and 86Chordula of the amphibia (the ringed adder). (From Goette.) Fig. 85 median longitudinal section (seen from the left), Fig. 86 transverse section (slightly diagrammatic). Lettering as in Figs. 83 and 84.

I give the name of chordula or chorda-larva to the embryonic stage of the vertebrate organism which is represented by the amphioxus larva at this period (Figs. 83, 84, in the third period of development according to Hatschek). (Strabo and Plinius give the name of cordula or cordyla to young fish larvæ.) I ascribe the utmost phylogenetic significance to it, as it is found in all the chorda-animals (tunicates as well as vertebrates) in essentially the same form. Although the accumulation of food-yelk greatly modifies the form of the chordula in the higher vertebrates, it remains the same in its main features throughout. In all



[ 97 ]

 

Diagrammatic vertical section of coelomula-embryos of vertebrates.
Figs. 87 and 88Diagrammatic vertical section of cœlomula-embryos of vertebrates. (From Hertwig.) Fig. 87, vertical section through the primitive mouth, Fig. 88, vertical section before the primitive mouth. u primitive mouth, ud primitive gut. d yelk, dk yelk-nuclei, dh gut-cavity, lh body-cavity, mp medullary plate, ch chorda plate, ak and ik outer and inner germinal layers, pb parietal and vb visceral mesoblast.

Transverse section of coelomula embryos of triton.
Figs. 89 and 90Transverse section of cœlomula embryos of triton. (From Hertwig.) Fig. 89, section through the primitive mouth. Fig. 90, section in front of the primitive mouth, u primitive mouth. dh gut-cavity, dz yelk-cells, dp yelk-stopper, ak outer and ik inner germinal layer, pb parietal and vb visceral middle layer, m medullary plate, ch chorda.

cases the nerve-tube (m) lies on the dorsal side of the bilateral, worm-like body, the gut-tube (d) on the ventral side, the chorda (ch) between the two, on the long axis, and the cœlom pouches (c) at each side. In every case these primitive organs develop in the same way from the germinal layers, and the same organs always arise from them in the mature chorda-animal. Hence we may conclude, according to the laws of the theory of descent, that all these chordonia or chordata (tunicates and vertebrates) descend from an ancient common ancestral form, which we may call Chordæa. We should regard this long-extinct Chordæa, if it were still in existence, as a special class of unarticulated worm (chordaria). It is especially noteworthy that neither the dorsal nerve-tube nor the ventral gut-tube, nor even the chorda that lies between them, shows any trace of articulation or segmentation; even the two cœlom-sacs are not segmented at first (though in the amphioxus they quickly divide into a series of parts by transverse



[ 98 ]

 

folding). These ontogenetic facts are of the greatest importance for the purpose of learning those ancestral forms of the vertebrates which we have to seek in the group of the unarticulated vermalia. The cœlom-pouches were originally sexual glands in these ancient chordonia.


Vertical section of the dorsal part of three triton-embryos. Fig. 91. A, B, C.Vertical section of the dorsal part of three triton-embryos. (From Hertwig.) In Fig. A the medullary swellings (the parallel borders of the medullary plate) begin to rise; in Fig. B they grow towards each other; in Fig. C they join and form the medullary tube. mp medullary plate, mf medullary folds, n nerve-tube, ch chorda, lh body-cavity, mk1 and mk2 parietal and visceral mesoblasts, uv primitive-segment cavities, ak ectoderm, ik entoderm, dz yelk-cells, dh gut-cavity.

From the evolutionary point of view the cœlom-pouches are, in any case, older than the chorda; since they also develop in the same way as in the chordonia in a number of invertebrates which have no chorda (for instance, Sagitta, Figs. 76–78). Moreover, in the amphioxus the first outline of the chorda appears later than that of the cœlom-sacs. Hence we must, according to the biogenetic law, postulate a special intermediate form between the gastrula and the chordula, which we will call cœlomula, an unarticulated, worm-like body with primitive gut, primitive mouth, and a double body-cavity, but no chorda. This embryonic form, the bilateral cœlomula (Fig. 81), may in turn be regarded as the ontogenetic reproduction (maintained by heredity) of an ancient ancestral form of the cœlomaria, the Cœlomæa (cf. Chapter XX).

In Sagitta and other worm-like animals the two cœlom-pouches (presumably gonads or sex-glands) are separated by a complete median partition, the dorsal and ventral mesentery (Fig. 78 dm, vm); but in the vertebrates only the upper part of this vertical partition is maintained, and forms the dorsal mesentery. This mesentery afterwards takes the form of a thin membrane, which fastens the visceral tube to the chorda (or the vertebral column). At the under side of the visceral tube the cœlom-sacs blend together, their inner or median walls breaking down and disappearing. The body-cavity then forms a single simple hollow, in which the gut is quite free, or only attached to the dorsal wall by means of the mesentery.

The development of the body-cavity and the formation of the chordula in the higher vertebrates is, like that of the gastrula, chiefly modified by the pressure of the food-yelk on the embryonic structures, which forces its hinder part into



[ 99 ]

 

a discoid expansion. These cenogenetic modifications seem to be so great that until twenty years ago these important processes were totally misunderstood. It was generally believed that the body-cavity in man and the higher vertebrates was due to the division of a simple middle layer, and that the latter arose by cleavage from one or both of the primary germinal layers. The truth was brought to light at last by the comparative embryological research of the Hertwigs. They showed in their Cœlom Theory (1881) that all vertebrates are true enterocœla, and that in every case a pair of cœlom-pouches are developed from the primitive gut by folding. The cenogenetic chordula-forms of the craniotes must therefore be derived from the palingenetic embryology of the amphioxus in the same way as I had previously proved for their gastrula-forms.

The chief difference between the cœlomation of the acrania (amphioxus) and the other vertebrates (with skulls—craniotes) is that the two cœlom-folds of the primitive gut in the former are from the first hollow vesicles, filled with fluid, but in the latter are empty pouches, the layers of which (inner and outer) close with each other. In common parlance we still call a pouch or pocket by that name, whether it is full or empty. It is different in ontogeny; in some of our embryological literature ordinary logic does not count for very much. In many of the manuals and large treatises on this science it is proved that vesicles, pouches, or sacs deserve that name only when they are inflated and filled with a clear fluid. When they are not so filled (for instance, when the primitive gut of the gastrula is filled with yelk, or when the walls of the empty cœlom-pouches are pressed together), these vesicles must not be cavities any longer, but “solid structures.”

The accumulation of food-yelk in the ventral wall of the primitive gut (Figs. 85, 86) is the simple cause that converts the sac-shaped cœlom-pouches of the acrania into the leaf-shaped cœlom-streaks of the craniotes. To convince ourselves of this we need only compare, with Hertwig, the palingenetic cœlomula of the amphioxus (Figs. 80, 81) with the corresponding cenogenetic form of the amphibia (Figs. 89–90), and construct the simple diagram that connects the two (Figs. 87, 88). If we imagine the ventral half of the primitive gut-wall in the amphioxus embryo (Figs. 79–84) distended with food-yelk, the vesicular cœlom-pouches (lh) must be pressed together by this, and forced to extend in the shape of a thin double plate between the gut-wall and body-wall (Figs. 86, 87). This expansion follows a downward and forward direction. They are not directly connected with these two walls. The real unbroken connection between the two middle layers and the primary germ-layers is found right at the back, in the region of the primitive mouth (Fig. 87 u). At this important spot we have the source of embryonic development (blastocrene), or “zone of growth,” from which the cœlomation (and also the gastrulation) originally proceeds.

Transverse section of the chordula-embryo of a bird (from a hen's egg at the close of the first day of incubation). Fig. 92Transverse section of the chordula-embryo of a bird (from a hen’s egg at the close of the first day of incubation). (From Kölliker.) h horn-plate (ectoderm), m medullary plate, Rf dorsal folds of same, Pv medullary furrow, ch chorda, uwp median (inner) part of the middle layer (median wall of the cœlom-pouches), sp lateral (outer) part of same, or lateral plates, uwh structure of the body-cavity, dd gut-gland-layer.

Hertwig even succeeded in showing, in the cœlomula-embryo of the water salamander (Triton), between the first structures of the two middle layers, the relic of the body-cavity, which is represented in the diagrammatic transitional form (Figs. 87, 88). In sections both through the primitive mouth itself (Fig. 89) and in front of it (Fig. 90) the two middle layers (pb and vb) diverge from each other, and disclose the two body-cavities as narrow clefts. At the primitive-mouth itself (Fig. 90 u) we can penetrate into them from without. It is only here at the border of the primitive mouth that we can show the direct transition of the two middle layers into the two limiting layers or primary germinal layers.

The structure of the chorda also shows the same features in these cœlomula-embryos of the amphibia (Fig. 91) as in the amphioxus (Figs. 79–82). It arises from the entodermic cell-streak, which forms the middle dorsal-line of the primitive gut, and occupies the space between the flat cœlom-pouches (Fig. 91 A).



[ 100 ]

 

While the nervous centre is formed here in the middle line of the back and separated from the ectoderm as “medullary tube,” there takes place at the same time, directly underneath, the severance of the chorda from the entoderm (Fig. 91 A, B, C). Under the chorda is formed (out of the ventral entodermic half of the gastrula) the permanent gut or visceral cavity (enteron) (Fig. 91 B, dh). This is done by the coalescence, under the chorda in the median line, of the two dorsal side-borders of the gut-gland-layer (ik), which were previously separated by the chorda-plate (Fig. 91 A, ch); these now alone form the clothing of the visceral cavity (dh) (enteroderm, Fig. 91 C). All these important modifications take place at first in the fore or head-part of the embryo, and spread backwards from there; here at the hinder end, the region of the primitive mouth, the important border of the mouth (or properistoma) remains for a long time the source of development or the zone of fresh construction, in the further building-up of the organism. One has only to compare carefully the illustrations given (Figs. 85–91) to see that, as a fact, the cenogenetic cœlomation of the amphibia can be deduced directly from the palingenetic form of the acrania (Figs. 79–84).


Transverse section of the vertebrate-embryo of a bird (from a hen's egg on the second day of incubation).

Fig. 93Transverse section of the vertebrate-embryo of a bird (from a hen’s egg on the second day of incubation). (From Kölliker.) h horn-plate, mr medullary tube, ch chorda, uw primitive segments, uwh primitive-segment cavity (median relic of the cœlom), sp lateral cœlom-cleft, hpl skin-fibre-layer, df gut-fibre-layer, ung primitive-kidney passage, ao primitive aorta, dd gut-gland-layer.

The same principle holds good for the amniotes, the reptiles, birds, and mammals, although in this case the processes of cœlomation are more modified and more difficult to identify on account of the colossal accumulation of food-yelk and the corresponding notable flattening of the germinal disk. However, as the whole group of the amniotes has been developed at a comparatively late date from the class of the amphibia, their cœlomation must also be directly traceable to that of the latter. This is really possible as a matter of fact; even the older illustrations showed an essential identity of features. Thus forty years ago Kölliker gave, in the first edition of his Human Embryology (1861), some sections of the chicken-embryo, the features of which could at once be reduced to those already described and explained in the sense of Hertwig’s cœlom-theory. A section through the embryo in the hatched hen’s egg towards the close of the first day of incubation shows in the middle of the dorsal surface a broad ectodermic medullary groove (Fig. 92 Rf), and underneath the middle of the chorda (ch) and at each side of it a couple of broad mesodermic layers (sp). These enclose a narrow space or cleft (uwh), which is nothing else than the structure of the body-cavity. The two layers that enclose it—the upper parietal layer (hpl) and the lower visceral layer (df)—are pressed together from without, but clearly distinguishable. This is even clearer a little later, when the medullary furrow is closed into the nerve-tube (Fig. 93 mr).

Special importance attaches to the fact that here again the four secondary germinal layers are already sharply distinct, and easily separated from each other. There is only one very restricted area in which they are connected, and actually pass into each other; this is the region of the primitive mouth, which is contracted in the amniotes into a dorsal longitudinal cleft, the primitive groove. Its two lateral lip-borders form the primitive streak, which has long been recognised as the most important embryonic source and starting-point of further processes. Sections through this primitive streak (Figs. 94 and 95) show that the two primary germinal layers grow at an early stage (in the discoid gastrula of the chick, a few hours after incubation) into the primitive



[ 101 ]

 

streak (x), and that the two middle layers extend outward from this thickened axial plate (y) to the right and left between the former. The plates of the cœlom-layers, the parietal skin-fibre-layer (m) and the visceral gut-fibre-layer (f), are seen to be still pressed close together, and only diverge later to form the body-cavity. Between the inner borders of the two flat cœlom-pouches lies the chorda (Fig. 95 x), which here again develops from the middle line of the dorsal wall of the primitive gut.


Transverse section of the primitive streak (primitive mouth) of the chick.

Figs. 94 and 95Transverse section of the primitive-streak (primitive mouth) of the chick. Fig. 94 a few hours after the commencement of incubation, Fig. 95 a little later. (From Waldeyer.) h horn-plate, n nerve-plate, m skin-fibre-layer, f gut-fibre-layer, d gut-gland-layer, y primitive streak or axial plate, in which all four germinal layers meet, x structure of the chorda, u region of the later primitive kidneys.

Cœlomation takes place in the vertebrates in just the same way as in the birds and reptiles. This was to be expected, as the characteristic gastrulation of the mammal has descended from that of the reptiles. In both cases a discoid gastrula with primitive streak arises from the segmented ovum, a two-layered germinal disk with long and small hinder primitive mouth. Here again the two primary germinal layers are only directly connected (Fig. 96 pr) along the primitive streak (at the folding-point of the blastula), and from this spot (the border of the primitive mouth) the middle germinal layers (mk) grow out to right and left between the preceding. In the fine illustration of the cœlomula of the rabbit which Van Beneden has given us (Fig. 96) one can clearly see that each of the four secondary germinal layers consists of a single stratum of cells.

Finally, we must point out, as a fact of the utmost importance for our anthropogeny and of great general interest, that the four-layered cœlomula of man has just the same construction as that of the rabbit (Fig. 96). A vertical section that Count Spee made through the primitive mouth or streak of a very young human germinal disk (Fig. 97) clearly shows that here again the four secondary germ-layers are inseparably connected only at the primitive streak, and that here also the two flattened cœlom-pouches (mk) extend outwards to right and left from the primitive mouth between the outer and inner germinal layers. In this case, too, the middle germinal layer consists from the first of two separate strata of cells, the parietal (mp) and visceral (mv) mesoblasts.

These concordant results of the best recent investigations (which have been confirmed by the observations of a number of scientists I have not enumerated) prove the unity of the vertebrate-stem in point of cœlomation, no less than of gastrulation. In both respects the invaluable amphioxus—the sole survivor of the acrania—is found to be the original model that has preserved for us in palingenetic form by a tenacious heredity these



[ 102 ]

 

most important embryonic processes. From this primary model of construction we can cenogenetically deduce all the embryonic forms of the other vertebrates, the craniota, by secondary modifications. My thesis of the universal formation of the gastrula by folding of the blastula has now been clearly proved for all the vertebrates; so also has been Hertwig’s thesis of the origin of the middle germinal layers by the folding of a couple of cœlom-pouches which appear at the border of the primitive mouth. Just as the gastræa-theory explains the origin and identity of the two primary layers, so the cœlom-theory explains those of the four secondary layers. The point of origin is always the properistoma, the border of the original primitive mouth of the gastrula, at which the two primary layers pass directly into each other.


Transverse section of the primitive groove (or primitive mouth) of a rabbit.

Fig. 96Transverse section of the primitive groove (or primitive mouth) of a rabbit. (From Van Beneden.) pr primitive mouth, ul lips of same (primitive lips), ak and ik outer and inner germinal layers, mk middle germinal layer, mp parietal layer, mv visceral layer of the mesoderm.

Transverse section of the primitive mouth (or groove) of a human embryo (at the coelomula stage).

Fig. 97Transverse section of the primitive mouth (or groove) of a human embryo (at the cœlomula stage). (From Count Spee.) pr primitive mouth, ul lips of same (primitive folds), ak and ik outer and inner germinal layers, mk middle layer, mp parietal layer, mv visceral layer of the mesoblasts.

Moreover, the cœlomula is important as the immediate source of the chordula, the embryonic reproduction of the ancient, typical, unarticulated, worm-like form, which has an axial chorda between the dorsal nerve-tube and the ventral gut-tube. This instructive chordula (Figs. 83–86) provides a valuable support of our phylogeny; it indicates the important moment in our stem-history at which the stem of the chordonia (tunicates and vertebrates) parted for ever from the divergent stems of the other metazoa (articulates, echinoderms, and molluscs).

I may express here my opinion, in the form of a chordæa-theory, that the characteristic chordula-larva of the chordonia has in reality this great significance—it is the typical reproduction (preserved by heredity) of the ancient common stem-form of all the vertebrates and tunicates, the long-extinct Chordæa. We will return in Chapter XX to these worm-like ancestors, which stand out as luminous points in the obscure stem-history of the invertebrate ancestors of our race.



Title and Contents
Glossary
Chapter IX
Chapter XI
Figs. 1–209
Figs. 210–408