Paper - Common Traits in Development and Structure of the Organs Originating from the Coelomic Wall
|Embryology - 13 May 2021 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)
|Historic Disclaimer - information about historic embryology pages|
|Embryology History | Historic Embryology Papers)|
Common Traits in Development and Structure of the Organs Originating from the Coelomic Wall
Department of Anatomy, The Chicago Medical School, Illinois
Four Plates (Twenty-Six Figures)
It has long been known that the lining of the coelomic cavities differs from other epithelia during early periods of development by its close structural and genetic relations to the underlying mesenchyme. A persistence of the respective potencies permanently distinguishes the coelomic epithelium from other tissues of similar structure, as pointed out in Maximow and Bloom’s text-book, and in more detail by Maximow (727 a,b). It is the principal purpose of the present report to demonstrate that the tissues originating from this mesothelium in the embryo retain, permanently or for a limited period of time, one important character of their parent tissue: a combination of epithelial and mesenchymal properties and potencies. Changes from one of these structural types to the other can be produced in many tissues with varying case under experimental conditions, such as those of tissue culture. The mesothelium and its derivatives, however, undergo such changes more readily than most other tissues, even under normal or only slightly altered conditions. This accounts for many peculiarities in their development, structure, and pathology, some of which will be discussed in this report.
It is difficult if not impossible to give a clear-cut definition of epithelial arrangement of cells. Most text-books of histology stress the close connection of the cells, with but little cement substance present between them. Consequently, the occurrence of lattice fiber networks surrounding individual eclls will be considered as indicating non-epithelial arrangement. In addition, it will be seen that the presence of lattice fibers in cell groups resembling epithelia marks a transition either from or into distinctly mesenchymal aggregations.
A brief discussion of the structure of the early coelomic wall may serve as a starting point for the following deseriptions and considerations. Many investigations have established the fact that the lining of the early embryonic coelomic walls consists of a layer of cells which resembles an epithelium only on its distal side. In bounding the tissue against the cavity, the cells necessarily form a continuous layer with an even surface; this is the epithelial side of the lining. ‘Toward the solid tissuc, however, these same cells are provided with tapering processes identical with those of the underlying mesenchyme (fig. 1). No boundary separates the superficial and deep layers, and daughter cells of the superficial cells contribute to the growth of the mesenchyme. Fischel (713) and Politzer (’386) pointed out that migration of cells from the superficial layer into the depth may be so abundant in certain areas of the dorsal coelomic wall that it resembles the early primordia of the adrenal cortex. Thus the superficial layer may in these early stages be regarded as a germinal layer of the adjacent mesenchyme. It is, in this respect as well as structurally, somewhat similar to the ependyma of the embryonic neural tube; however, the mesenchyme contributes more to its own growth than does the mantle layer of the neural primordium. The condition just described for the coelomic lining gradually changes and a layer with all morphological characteristics of an epithelium develops. In human embryos of about 8 mm. and mammal embryos of corresponding stages, only limited areas still show the original structure (fig. 2).
This peculiar condition of the coelomic lining has been mentioned by various investigators in connection with the early development of its derivatives, particularly the spleen (see below). The principal objective of these considerations ORGAN FORMATION FROM COELOMIC WALL ood
was to point out that mesenchyme and mesothelium are derived from one souree, and to determine whether the organ in question forms before or after the separation of these layers is completed. However, very little attention was given to the impheations of the fact that in the early condition the superficial layer contributes to the growth of the mesenchyme and that both together form a common unit, particularly with regard to tissue growth. Observations of limb bud formation in amphibians (Filatow, ’33) suggest that this early condition ean be retained or reactivated if special developmental processes require a rapid localized multiplication of the mesenchyme of the body wall. It will be pointed out soon that a very similar transformation of the higher differentiated mesothelium and mesenchyme back into a common unit takes place in the initial stages of the development of various organs from the coelomic wall. However, the peculiar combination of epithelial and mesenchymal characteristics and the readiness to change from one type to the other are not limited to these early stages as will be shown by various examples.
As far as the gonads are concerned, many observations and considerations related to the present problem were recently published in a separate article (’42); this publication will be referred to on numerous occasions throughout the present report in order to avoid repetitions.
Material and Methods
The importance of an adequate staining technic in investigations of the type to be reported here was emphasized previously (’42). The use of methods showing collagenous and lattice fibers is essential whenever relations of mesenchymatous and epithelial tissues are examined. Many misconceptions of development and structure of mesodermal organs, particularly gonads and adrenal cortex, could have been avoided by the use of such methods instead of the common hematoxvlin stains which mav be perfectly adequate in similar investigations of most organs with ectodermal or entodermal epithelial components. As in the previous investiga306 PETER GRUENWALD
tion, azan stain and silver impregnation were employed, except in a few instances when older material was used. These methods were supplemented recently by a very useful combination of Gomori’s (’39) silver impregnation with azocarmine or acid fuchsin stain by which nuclei and a few other cellular details could be shown in the otherwise empty spaces between the impregnated fibers. The stains were applied after completion of the impregnation.
Because of the large amount of disparate material used in the present work, no enumeration of the specimens will be given. Most of the complete embryos and gonads used are listed in a previous report (’42). The present investigation is based mainly on human and mammalian material. Chick embryos will be used only to demonstrate conditions not so clearly visible in mammals, as is the case in the changes in the miillerian duct and tubal ridge described later in this report.
In line with the aim of the present work, only such conditions and changes will be described here in detail as are useful to demonstrate the peculiar common characteristics of mesothelial derivatives. These traits will be found to be most outspoken and permanently persisting in the gonads and the adrenal cortex; the miillerian ducts and the spleen show similar behavior only during early developmental stages. As a supplement, the formative activity of the coelomic lining in late stages will be demonstrated by means of several examples.
The adrenal cortex
A. review of the literature on adrenal development shows a fair agreement as to the statement that the epithelial cell cords of the cortex grow into the retroperitoneal mesenchyme from the coelomic epithelium. The investigation of Politzer (736), based on well preserved human material, is in line with this coneeption. Only Uotila (’40) recently admitted that ‘‘the formation of distinet cell buds has not been observed’’. Many investigators probably desired to trace epithelium back to epithelium; however, the actual conditions to be seen in sections offer very little in support of the description advocated by the majority of related reports. When trying to refrain from any prejudice as to the epithelial nature of the adrenal cortex and its parent tissue, one can, according to the conditions seen in human and mammalian embryos, give the following description of early adrenal development.
At the time and location thoroughly described by Politzer for human embryos, cell masses appear in the retroperitoneal mesenchyme, reaching from the peritoneal surface dorsad (fig. 3). The arrangement of their cells indicates that the superficial layer is active in their production, but neither the lining of the coelomic cavity in this area, nor the cells apparently moving dorsad from it, show an epithelial arrangement. Investigation of the sections themselves does not reveal more epithelial characteristics than does the photograph of our figure 3 or similar figures in Politzer’s and Uotila’s reports. Soon after the stage just described, the primordium of the adrenal cortex is separated from the coelomic lining by loose mesenchyme, and its cells begin to differ from the surroundings by their round, large nuclei and more darkly staining cytoplasm. An indistinct arrangement in cords can also be noticed, but it is not epithelial. Proper staining methods, as silver impregnation, show reticulum fibers surrounding each individual cell and separating it, from its neighbor in the same cord (fig. 4). This condition persists long and in great parts of the adrenal cortex even throughout life. There is considerable variation as to the appearance of these reticulum fibrils in the adult adrenal, among different mammalian species as well as among individuals of the same species. Figures 6 to 12 may serve to illustrate this. It was found that in general the zona glomerulosa loses most completely the reticulum separating individual cells, and assumes the structure of epithelial cords, whereas the zona reticularis and the adjacent part of the zona fasciculata are always found to retain this reticulum. The outer zona fasciculata shows the largest variability as to the presence of reticulum fibrils within its cords; an outer portion of varying extent may be free of them.
The reticulum of the adrenal cortex was the subject of an investigation by Corner (’20). He found that there are no connective-tissue cells between the cortico-adrenal cords to account for the rich reticulum. Assuming that the adrenal cells proper do not form lattice fibers, Corner designated the cells of the capillary endothelium as the only possible source of the reticulum. Knowing the cortico-adrenal cells as descendants of cells of the coelomic wall with potencies of embryoni¢e connective tissue, we cannot decline a priori to consider them as possible producers of reticulum. According to Plenk (’27), many derivatives of the embryonic connective tissue, such as endothelial and fat cells and muscle fibers, smooth as well as striated, may form argyrophil fibers. This may very well hold for cortico-adrenal cells as well. It is not surprising to find these cells enveloped by lattice fibers during early stages of embryonic development, when they still resemble their parent tissue, that is, the coclomic wall (fig. 3). A persistence of this condition accounts for part of the reticulum, particularly that within the cords and not in contact with capillary walls. It is interesting to notice that Corner found a very similar condition in the corpus Inteum which, being also a derivative of the coelomic wall, is very similar in its origin to the adrenal cortex. In other organs investigated by Corner (hypophysis, thyroid, kidney), the conditions are different since there are always at least a small number of connective tissue eclls that can be held responsible for the production of the reticulum.
Plenk (’27) and Bachmann (’37 and ’39) also investigated the argyrophil connective tissue of the adrenal cortex. Both authors consider its gradual increase in the deeper layers as an expression of a decrease in function and vitality of the cortico-adrenal cells;.Bachman tried to corroborate this by a comparison with the corpus luteum where the lattice fibers inerease with its age. Recent work of Bennet (’40) corroborates this statement as far as the age of the .cells is concerned. As to the relation of appearance of the reticulum to function, Bachmann’s suggestion is not confirmed by Bennet who localizes the zone of maximal function of the cortical cords to an area central to the outer cortex devoid of reticulum.
Tn spite of these embryological and histological observations, the adrenal cortex is still widely believed to consist of epithelial cell cords. This is expected in analogy with most of the other endocrine organs; the cord-like arrangement of the cells so rich in cytoplasm also seems to point to an epithelial structure. Most of the stains ordinarily employed either do not show reticulum separating the eclls, or make it appear as unusually well-marked cell boundaries. Our own observations, as well as previous reports of Plenk and Bachmann, however, show that the cords deserve to be called epithelial only in a smaller outer portion comprising the zona elomerulosa and a part of varying height of the zona fasciculata (figs. 6 to 12). There is a gradual transition to non-epithelial structure as we follow the cords toward the center of the organ, and in early embryonic stages is there no epithelial zone at all (fig. 4).
With this fact and our characterization of the mesothelium in mind, we are able to arrive at a more satisfactory conception of development and structure of the adrenal cortex. In its carly stages, large numbers of cells develop from the coelomic wall, and mesenchyme as well as mesothelium actively participate in this process as parts of one uniform tissue comparable to the carly embryonic coelomic wall. The blastema thus formed has a dense framework of argyrophil fibrils like the surrounding mesenchyme; as its cells swell and form cords, this reticulum remains between them and the changes in its distribution are only due to the widening of the meshes by the growth of the cell bodies (fig. 4). Only in the periphery of the organ do the reticulum fibrils eventually disappear within the cords, thus allowing the cells to assume an epithelial arrangement. This behavior of the reticulum was described here in detail because it is a good indicator of the nature of the cortico-adrenal cells: they are swollen cells of the coelomic wall with the combination of epithelial and non-epithelial properties characteristic of their mother tissue.
The close developmental relations of the adrenal cortex to the surrounding mesenchyme of the coelomic wall are not limited to early embryonic stages. Zwemer, Wotton and Norkus (’38) called attention to cell types in the capsule of the adrenal, showing all transitional stages between connective tissue cells and typical cortico-adrenal cells, and they concluded that the capsule participates in replacing cortical cells worn out by use. This was recently confirmed by Salmon and ZAwemer (’41) when they traced cells laden with trypan blue from the capsule to the cortical cords. It was also recognized that presence of the capsule is essential for adrenal grafts to survive; most of the transplanted differentiated cortical cells become necrotic and a new organ is formed from the capsule (Higgins and Ingle, ’88; Ingle and Higgins, 389; Baker and Baillif, ’39). The author’s own observations fully confirm the report of Zwemer and his co-workers. Particularly in adrenals of man and the rhesus monkey, but also in other mammals, tapering cortical cell cords reach into the capsule, assuming there a course parallel to the surface of the adrenal (figs. 13 to 15). Their appearance may very well be the expression of a new-formation of cortico-adrenal cells from the capsule. It is, however, difficult to imagine how such new-formation should take place in the many species with a more distinctly differentiated zona glomerulosa. In these cases, the cell groups of the adrenal cortex are most sharply bounded and epithelium-like just where they face the capsule (figs. 9 to 12). With the few observations at hand it is diffienlt to reconcile this with the possibility of physiological replacement of cells from the capsule. However, even in adrenals with a very distinct zona glomerulosa, occasional small cortical cell groups can be encountered in the inner part of the capsule, sometimes counected with cords of the zona glomerulosa, so that even in these organs a participation of the capsule in regeneration eannot be excluded. Bachmann (’39 b) rejects the conclusions of Zwemer and his coworkers and traces the regeneration of cortico-adrenal cells to a sub-capsular blastema. This seems to be a dispute about terminology rather than facts; both authors probably saw the same histological pictures. The layer peripheral to the zona glomerulosa, not consisting of compact cortical tissue, was called a blastema by Bachmann, and a loosely woven inner portion of the capsule by Zwemer and his coworkers,
Another indication of a special activity of the capsule of the adrenal was found by H. Popper (personal communication). When investigating adrenals by means of fluorescence microscopy, Popper found in the eapsule lipoid droplets containing vitamin A, as they occur in cortico-adrenal cells, but not in ordinary connective tissue. This, too, suggests that the capsule of the adrenal is more than just a protecting layer of connective tissue and should be considered in all investigations on the adrenal cortex.
It is commonly agreed upon that mitosis of cortico-adrenal cells, particularly in the outer portion of the zona fasciculata, contributes to the production of new cortical cells. Tt is unknown to which extent this process and the before-mentioned transformation of capsular cells cooperate in the regeneration of the organ, and whether there are conditions causing the oue or the other process to prevail.
The conditions found in early gonad primordia, although similar to the initial stages of the adrenal cortex, have caused a great deal of controversy. There is hardly any theoretically possible mode of early gonad development that has not been suggested by several investigators. The majority of authors hold that epithelial cell cords growing into the deeper lavers from the covering mesothelium give rise to the primary sex cords; in the female, these were said to be supplemented later by secondary cords which also originate from the surface epithelium. Other authors (Van Vloten, ’27; Higuchi, ’32) trace the primary sex cords back to a blastema composed of a mixture of epithelial and mesenchymal cells. Fischel (730) disputes any participation of the mesothelium in the development of the sex cords. According to his view the primary cords originate from the mesenchyme of the early gonad primordium without any contribution from the surface epithelium, and no new cords are formed during later stages. The author’s own results were recently published in detail (°42). It was pointed out that gonad development starts after the lining of the respective part of the coelomic cavity has been differentiated as an epithelium for a short period of time. Both the epithelium and the underlying mesenchyme participate in the following differentiations, but as a common umt rather than separately. The newly formed basal membrane of the mesothelium disappears (fig. 5; for more figures, see °42) and, just as in early adrenal development, the uniform condition of the coelomic wall reappears to form the gonad blastema. Here, however, the newly formed organ does not sever its relations with the coelomic lining as soon as does the adrenal, but considerable differentiation goes on before a truly epithelial lining is re-established on the surface of the organ. When the cells of the gonad blastema first group themselves in radial rows to form the primary sex cords, the zone of differentiation is sharply bounded neither from the later surface epithelium nor from the mesenchyme at the basis of the organ. Only considerably later, in human embryos of over 20 mm., is the mesothelium restored as a typical epithelium, retaining connections with the sex cords to an extent varying according to sex and species. It follows from this description that the primary sex cords are derivatives of neither the mesothelium nor the mesenchyme alone; they are furnished by a blastema in which these two parts of the coelomic wall cannot be distinguished, just as is generally the case in the early embryonic coeloma. Closest to the mode of development outlined here comes the conception of origin of the sex cords
The immigrated primordial sex cells are not included in these considerations ; the question whether or not they give risc to the sex eclls of the mature gonads will not be discussed here, Vey
from a mixture of epithelial and mesenchymal cells. This theory, however, suggests the presence of morphologically and potentially different mesodermal cells in the gonad blastema, an assumption which is not justified according to the conception outlined here. As to the secondary sex cords, the author agrees in general with earlier investigators on their origin from the surface epithelium of the gonad. For a detailed description of this process the reader is referred to the previous report (’42).
While the primary sex cords are developing, the rete differentiates in a similar manner near the hilus of the organ, and interstitial cells appear, at least in the testis. Most authors take it for granted that the interstitial cells differentiate from the mesenchyme. It would then appear as if a permanent separation of epithelial and mesenchymal tissues were gradually established within the gonad, with the epithelial line represented by sex cords, rete, and surface lining, and the mesenchymal portion by the interstitial tissue and stroma of the organ. There are, however, indications of close developmental similarities of sex cords and the interstitial tissue; the latter develops in many species in the form of cell cords from the gonad blastema much like the primary sex cords, and many other analogies as well as transitions between both tissues are on record (see Gruenwald, ’42). The most impressive example is the regular and abundant transformation of sex cords into interstitial cells in the embryonic horse ovary, as described by Kohn (’26) and Petten (’33). These relations of the epithelial sex cords and the mesenchymal interstitial tissue again recall the peculiar character of the derivatives of the coelomic wall. Further evidence is the fate of the granulose cells after rupture of the follicle: these epithelial cells cooperate with the mesenchymal theca cells in the formation of the corpus luteum to such an extent that both types cannot be distinguished any longer. These changes of the follicle cells during corpus luteum formation have been compared with the succession of forms from the periphery toward the center of the adrenal cortex (Bach364 PETER GRUENWALD
mann, ’37). In both instances fluent transitions exist between epithelial and mesenchymal types of tissue.
The miillerian ducts
The miillerian duets develop from the coclomic wall at about the same time as the gonads and adrenal cortex, but without disturbing the distinctly epithelial structure of the mesothelium. The ducts themselves grow caudad from thickened portions of the mesothelium, and closely follow the wolffan ducts in their course. Details of this process were previously described (Gruenwald, ’41); it was pointed out that the growing end of the miillerian duct is not separated from the wolffian duct by either mesenchyme or a basal membrane. It grows toward the cloaca as a wedge between the cells and the basal membrane of the wolffian duct, and somewhat later this peculiar process is followed by separation of the ducts by basal membranes and mesenchyme. Neither this previous investigation, nor any other related reports in the literature showed any evidence of such transitions between epithelial and mesenchymal structure as were just described in the gonad and adrenal. However, the following observation will demonstrate that the miillerian duct normally forms mesenchyme, at least in its caudal portion. This process will be briefly described here as it oceurs in chick embryos, but there are indications of similar changes occuring to a varying extent in mammal embryos as well. A detailed investigation of this subject will be reported in the near future.
Figures 16 to 18 show cross sections of wolffian and miillerian ducts of a 6-day, 38-hour chick embryo. In the region in which the solid growing end of the miillerian duct is found within the basal membrane of the wolffian duct (fig. 16), the mass of miillerian cells is almost homogeneous; only a slight difference exists between a cell group in the center and the rest of the mass, and even this difference is absent at the very end of the miillerian wedge. The wolffian duct can be distinguished fairly well from the miillerian cells although no basal membrane separates the two tissues at the most caudal level. Figure 17 shows a section 110 » cranial to the one just deseribed; here the development is slightly farther advanced and lattice fibers have begun to form in the arca enveloped in the common basal membrane of the two ducts. These fibers gradually complete the basal membrane of the wolffian duct, thus separating it from the miillerian cell mass, and at the same time appear within that cell mass, thus dividing it into two portions. One part, in the center of this mass, remains free of lattice fibers and ean now be seen to form the miillerian duct proper; the rest of the cell mass, reaching between the two duets from both sides in the form of wedges (fig. 17), develops an increasing amount of fibers. This latter portion, although it is a part of the original miillerian primordiun, is gradually transformed into typical embryonic connective tissue, and cannot be distinguished from the surrounding mesenchyme when the original common basal membrane disappears. Figure 18, 280 u cranial in the series to figure 17, shows the beginnmg dissolution of this membrane. This peculiar type of mesenchyme formation from the miillerian primordium was not observed in the cranial portion of the miillerian duct. A condition similar to these findings was seen in a 22-mm. cat embryo (fig. 19).
The present observation shows that the growing end of the miillerian duct forms, im the caudal part of the urogenital ridge, considerable amounts of mesenchyme in addition to the duct proper. This mesenchyme soon after its formation can not be distinguished from the surrounding tissue of the ridge. We do not know at the present time whether the miillerian duct and its epithelial derivatives permanently retain their potency to form cells morphologically identical with the surrounding mesenchyme, nor is there any indication on record of miilerian potencies of the embryonic connective tissue formed from the miillerian primordium. However, it is safe to assume that in the stages described here the cells of the miillerian ducts show a similar readiness to change from epithelial to mesenchymal arrangement as those of the mesothelium itself and the gonads and adrenal cortex.
Early spleen development has repeatedly been reviewed with regard to its relation to the mesothelium. Thiel and Downey (’21), Bergel and Gut (734), and Holyoke (736) agree that the mesothelium participates in spleen formation only indirectly insofar as it forms part of the mesenchyme from which the spleen is to originate. The author’s own observations fully confirm these reports. At no time after the appearance of the spleen is there any indication of considerable cell migration from the mesothelium that might form more than an occasional cell of the capsule. This is indicated by the position of cells and lattice fibers even in places where the epithelium is not bounded by a definite basal membrane. Figure 20 was taken from the spleen of a 12-mm. human embryo, at a stage when massive proliferation from the surface lining has ceased. It shows the presence of a basal membrane of the mesothelium along part of the surface, and in the remaining portion an irregular, but fairly complete boundary of lattice fibers at the basis of the superficial cells. In this latter area, a few cells were probably being given off by the mesothelium. Soon after this stage more and more of the surface epithelium has a complete basal membrane. In later periods only occasional areas may be seen devoid of a distinct basal membrane (fig. 21). These areas are usually located in places where trabeculae are connected with the capsule, and this indicates that the few cells that may leave the mesothelium on its basal border, are developing into fibroblasts rather than specific elements of the spleen.
It follows from these observations and corresponding reports in the literature that the mesothelium does not to any considerable extent contribute to the splenic tissue proper after the appearance of the primordium of the spleen. Oceasional cells given off in later embryonic periods probably become part of the capsule or adjacent trabeculae.
The coelomic walls in later stages
The mesothelial lining of the serous cavities retains a remarkable activity for a long time after acquiring a distinctly epithelial form. It was mentioned earlier that on the surface of the gonad, particularly the ovary, the mesothelium forms sex cords during the greater part of embryonic life. In the testis there is far less of this activity, although it is not entirely missing. Hett (’27, ’30, ’32) published detailed descriptions of proliferations of the testicular mesothelium in various mammals, but failed to distinguish those related to sex-cord formation from others which, according to the author’s earlier results (’37), have nothing to do with sex cords, but follow in their distribution so closely the arrangement of the large blood vessels in the tunica albuginea that a connection of some kind has to be assumed. Even after birth, the surface epithelium. of the gonad is not reduced to an inactive lining. There are several reports of new formation of follicles from proliferations of the surface lining in the ovary of various adult mammals. For details and references see Swezy (’33) and Duke (’41). Experiments on mice revealed that after x-ray irradiation the epithelium lining the surface of the ovary proliferates abundantly, as if attempting to regenerate the parenchyma destroyed by the experiment (Parkes, ’26; Geller, ’30). No ova were found in these newly formed cell groups, but luteinization was reported. In the dog ovary, similar proliferations can be observed under normal conditions. The ‘‘anovular follicles’? resulting from them were thoroughly described by Jonckheere (’80).-These formations may be considered as the expression of an exceptionally great potency of sex-cord production in this species, causing the mesothelium to proliferate even in the absence.of sex cells. Groups of mushroom-shaped projections were seen on the surface of the ovaries of adult rabbits. In this case, however, is there no reason to bring the outward directed proliferation in any connection with sex-cord formation.
Considerable activity of the embryonic mesothelium can also be noticed in areas adjoining the gonad, particularly those covering the mesonephros. Hett (’30) found the remarkable thickenings of the mesothelium which he indiscriminately considered as evidence of persistent germinal epithelium, extending from the testis onto the mesenephros. Another type of proliferations of the mesonephric mesothelium can be seen in many mammal embryos as, e.g., in human embryos of about 14 to 21mm. The lining of the dorso-lateral surface of the mesonephros is thrown into folds perfectly resembling external glomeruli on section (fig. 22). It must be noticed that this was found.so regularly and so far caudally that remnants of the pronephros can be excluded. No evidence of nephrostomes near these ‘‘glomeruli’’ could be obtained, and their blood supply is, as far as study of sections allows such conclusions, not large enough to indicate an excretory function of considerable extent. In figure 22, one of these structures may be compared with a typical mesonephric glomerulus. In pig embryos, ridges separated by mesothelium-lined clefts were also found on the lateral surface of the mesonephros. These, however, bear little resemblance to glomeruli. Similar uncharacteristie clefts and villi were observed in embryos of other mammalian species as well. The difference in the appearance of proliferations on the surface of the mesonephros is important to notice. It shows that the simuarity with glomeruli as found particularly in human embryos, is probably incidental and is in no way sufficient to identify these structures as functional or vestigial excretory organs.
Another area of great activity of the mesothelial covering of the mesonephros is the tubal ridge. This strip of high mesothelium develops as a caudal continuation of the area of the later ostium tubae, along the entire course of the wolffian and miillerian ducts in the mesonephric ridge. It loses its characteristic appearance shortly after the millerian duct has completed its caudal growth. The continuity of the tubal ride with the ostium of the miillerian duct, and its close proximity to this duct suggest that it might have to do with its formation. Careful investigation of serial sections, however, revealed nothing to indicate a contribution of cells of the tubal ridge to the developing miillerian duct. Whereas the ridge may be indistinctly bounded, the basal membrane of the duct is always intact and separates the miillerian epithelium from the coelomic lining in all places except the ostium. Even in the absence of the miillerian duct in consequence of experimental destruction of the wolffian duct does the tubal ridge not regenerate a new miillerian duet (Gruenwald, ’87). The basal boundary of this strip of high mesothelium is indistinct in places, and the basal membrane interrupted. The histological pictures in these areas indicate new-formation of mesenchymal cells from the proximal layers of the mesothelium.
The clearest evidence of extensive mesenchyme formation from the tubal ridge was gained in chick embryos of the seventh and eighth day. Two sections of this area from a 7-day, 17-hour embryo are shown in figures 23 and 24. In the more caudal one (fig. 24), the tall epithelium of the tubal ridge is bounded proximally by a distinct basal membrane. The greater proximal portion of the epithelium shows early stages of lattice fiber formation between its cells. A view of a later stage of this same process may be obtained in a more cranial section of the same structures (fig. 23). A dense framework of lattice fibers has now developed in the tubal ridge, sparing only its most distal part. The old basal membrane can hardly be recognized between the lattice fibers of the tubal ridge and those of the underlying mesenchyme; it is apparently becoming part of this framework. In still later stages of this transformation a new basal membrane forms underneath the distal portion of the former tubal ridge, and only this portion turns out to remain epithclial whereas the greater basal part of the original epithelium of the tubal ridge is now completely transformed into mesenchyme surrounding the miillerian duet. (It will be remembered that another portion of this mesenchyme is furnished by the miillerian duct itself, see above.) In human embryos the epithelium of the tubal ridge never reaches the height of that in the chick embryo, and consequently mesenchyme formation is not so exteusive and obvious. In a 13-mm. hedge hog embryo, however, stages of transformation of the tubal ridge were seen well comparable to those in the chick. In one portion, the basal boundary of the original tall epithelium is still visible, but much of the tissue distal to it 1s obviously being transformed into mesenchyme; only a thin mesothelium remains on the surface, and a new membrane is formed at its base.
Another area of quite close relations of mesothelimn and adjacent mesenchyme may be seen in chick embryos at the caudal end of the abdominal cavity, near the points where the wolffian ducts join the cloaca. There is an area on either side of the body with large groups of buds, with or without lumen, extending from the mesothelium into the body wall (fig. 25). Near the abdominal cavity, these structures are epithelial and have distinct basal membranes; however, with increasing distance from the coelomic cavity the basal boundaries become indistinct, the lumina disappear, and the entire mass of these buds gradually passes over into the mesenchyme. The significance of these structures is unknown. The only reference to their existence known to the author, was found in Lillie’s book (719) as a legend to a figure, describing them as follows: ‘‘Posterior angle of the body cavity; the epithelium is invaginated and folded so as to simulate a glandular strueture.”’ Tt is possible that these peculiar buds represent vestigal caudal continuations of the coelomic cavities, or that they are related to some change, increase or decrease, in the volume of the body cavity. The only observation in mammals faintly similar to these findings was made in a 15-mm. dog embryo (fig. 26). In a large area on the dorsal wall of the urogenital sinus the basal membrane of the mesothelium is lacking, and the lattice fiber structure suggests moderate mesenchyme formation by the mesothelium. In two areas near the midplane, however, grooves extend into the tissue and there the mesothelium is continuous with the deeper layers without any trace of a boundary. Delicate lattice fibers radiate into this tissue from the deeper layers. Here, too, gradual transitions exist between mesothelium and mesenchyme.
The proliferating capacity of the mesothelium is not limited to the peritoneum. Patches of thickened mesothelium with formation of mesenchyme from their basis were described in the parietal pleura of newborns by Kampmeier (’28). They are, just like the above mentioned thickenings of the testicular epithelium, associated with blood vessels. Kampmeier is certainly correct in saying about these patches: ‘‘We must look upon that phenomenon as a continuance into postnatal life of the primitive activity of the mesoderm lining the body cavities —the generation of mesenchymal and many other cells with all the potencies which they possess.’’ Another observation of Kampmeier (37) concerns fold-like projections of the pericardium to both sides of the aorta. I can, from my own observations, confirm Kampmeier’s statement that these structures are regularly present in human embryos of about 20 to 830mm. Their significance is unknown.
A number of observations were compiled in this chapter concerning proliferations of the mesothelium in distal or proximal direction, ranging from true sex-cord formation contributing essentially to the development of an organ to apparently insignificant folds and villi. These observations not only indicate a great morphogenctie activity of the mesothelium, but also show that we have to be extremely careful in interpreting such findings. In view of so many proliferations of the mesothelium not related to the development of any organ, we are not entitled to consider any thickening of the mesothelium of the gonad as evidence of a potency of sex-cord formation, nor can we conclude from the presence of projections similar to glomeruli that we are dealing with a vestigial excretory organ. Other characteristics have to be found in addition, to justify such conclusions. The present observations are evidence of the fact that the mesothelium, apart from participating in the formation of organs in early embryonic periods, has a great capacity of nonspecific proliferation, forming epithelial projections as well as mesenchyme.
Discussion and Conclusions
The present investigation revealed a remarkable conformity of the derivatives of the coelomic wall and their parent tissue in one important respect. The capacity of the cells of the coclomic wall to change from epithelial to mesenchymal arrangement or vice versa, was found in the tissues of the gonads, adrenal cortex, and miillerian ducts as well. During early development of the gonads and the adrenal cortex, the coelomic wall again assumes its early embryonic structure with the surface lining acting as a germinal layer. This has repeatedly been mistaken for proliferations of a coelomic epithelium; proper staining methods show that no epithelium exists in these areas during the period of rapid proliferation. At this time the superficial layer has essentially the same structure as is generally the case in carlier stages: its cells have on their basal side the same structure as the underlying mesenchyme, and no natural boundary exists between the two. These findings, along with similar observations concerning early development of the extremities in amphibians (Filatow), suggest that the capacity of reversion to the primitive condition is a general property of the embryonic coelomic wall, activated whenever organ development necessitates abundant cell proliferation.
The above mentioned capacity of structural transformation does not end in the primordia of gonads and adrenal cortex when the coelomic wall returns to its resting condition with a well defined surface epithelium. In the gonads, developmental relationships as well as transitions between the epithelial sex cords and the mesenchymal interstitial tissue may be observed for a long time (see Gruenwald, *42), and the formation of the corpus luteum from granulosa and theca cells recalls this peculiarity even in the adult. In the adrenal cortex, gradual transitions from epithelial to non-cpithelial structure are permanently found in all cell cords within the outer portion of the zona fasciculata, and stages of formation of epithelial cortico-adrenal cell cords from mesenchymal cells may be scen in the capsule of the organ.
Not quite so obvious are corresponding findings in the miillerian ducts. The material of these ducts is not laid down simultaneously as a large blastema as happens in the gonads and adrenal cortex; these ducts rather develop from small initial primordia, and without further contribution by the coelomic wall. Accordingly, no radical structural changes of the coelomic wall seem to be necessary and we find the miillerian ducts arising from a distinctly epithelial part of the coelomic lining. The ducts remain purely epithelial in their cranial parts. Caudally, however, their primordia differentiate so that only the central portions develop into continuations of the ducts, whereas the rest forms typical embryonie connective tissue. This process can be traced distinctly in certain species (chick, cat), and is indistinct or entirely absent in others. There are, wp to the present time, no indications of similar mesenchyme formation in later embryonic or postnatal stages.
In addition to these observations it was shown that during later embryonic periods the mesothelium itself participates more actively in mesenchyme formation than is usually believed. The area of greatest activity is the tubal ridge where the greater part of the stratified mesothelium is transformed into embryonic connective tissue. On the surface of the spleen of older embryos, the mesothelium may occasionally show pictures indicating contribution of a few cells to the connective tissue capsule.
The classical conception of histology held that a definite separation of epithelial and mesenchymal tissues is established soon after the earliest periods of embryonic development. Processes such as differentiation of kidney epithelia from a mesenchymal blastema were considered as rather exceptional and in that case, too, the epithelial differentiation, once established, was thought to be irrevocable. During the past decades experimental histology, particularly tissue culture, revealed that various factors entered into the determination of epithelial or mesenchymal structure, and that experimental changes of these factors may be followed by structural changes not in aceord with the above mentioned rigid conception. With these results in mind, the question must be raised whether the derivatives of the coelomic wall are essentially, or gradually, or not at all different from other tissues as far as their above described structural lability is concerned.
In a summarizing review of the results of tissue culture regarding differentiation, Bloom (’37) quotes reports demonstrating the effect of consistency of the medium and similar physical factors on the arrangement of cells in vitro, and eoncludes that ‘‘.... There seems to be no fundamental difference between epithelium and connective tissue’’. These results and conclusions are doubtless correct; they concern, however, only one of the many groups of factors involved in determining tissue differentiation. Many well known developmental processes show that other, probably intrinsic factors are sufficient to produce the changes in question in the absence of such environmental changes as active in experiments in vitro. A number of examples may be found in the present report. Neither the formation of epithelial sex cords from the mesenchymal gonad blastema, nor the transformation of the greater part of the epithelial tubal ridge into embryonic connective tissue, are preceded by extrinsic changes of consistency, surface conditions, or similar factors in the environment. It is evident from these considerations that the results of tissue culture and those of the present investigation concern different factors affecting tissue differentiation; each of these groups of factors may alone or combined with others affect a tissue sufficiently to produce a change in its differentiation.
Perhaps more important than the investigation of the environmental conditions themselves in the above mentioned in vitro experiments, is the fact that the number of potencies left in the body cells after the early periods of development is much larger than was originally believed. Most or all of the body cells seem to have the capacity to form epithelial as well as mesenchymal aggregations, and in this respect the derivatives of the coelomic wall are probably not essentially different from other tissues. The difference to be pointed out here is a quantitative one. In ectodermal, entodermal, and probably certain mesodermal tissues, changes from epithelial to mesenchymal structure or vice versa are rare under normal conditions. One example was given above when it was mentioned that the early coelomic wall bears certain structural parallels to the developing neural tube; the nervous tissue, however, is so peculiar in its structure that it is usually not included in the system worked out for the majority of the tissues. Other examples are the enamel pulp and possibly the thymus; these, however, do not compare with the variety and duration of transitions recorded in the present report. One is, therefore, justified to assume that, although all cells of the body may have the necessary potencies, those of the coelomic derivatives show the greatest readiness to undergo changes from epithelial to mesenchymal structure or vice versa. Transformations comparable to those normally oceurring in the coelomic derivatives, could be produced only by radical experimental procedures in most other tissues.
The present observations were gathered from a limited selection of material, mostly embryonic and comprising only a few species. Further work of histologists and embryologists will doubtless reveal many more examples to illustrate the peculiar behavior of the tissues derived from the coelomic wall. The structural changeability of the coelomic derivatives will also have to be considered in explaining pathological growth and differentiation of these tissues. It is certainly not incidental that so many diffieult problems have arisen for the pathologist from the study of abnormal growth in the gonads, or tumors related to the adrenal cortex. It will have to be remembered that the clear-cut separation of epithelial and mesenchymal tumors cannot be kept up in these instances and that the tissues arising from the coelomic wall can produce an unusually great variety of pathological growth types, due to their peculiar wide range of differentiation. 376 PETER GRUENWALD
- The capacity of the cells of the early embryonic coelomic wall to change from epithelial to mesenchymal arrangement or vice versa, was traced through the development of their derivatives, particularly gonads, adrenal cortex and miillerian ducts, as well as through later stages of the mesothelium itself.
- When large masses of cells are to proliferate for the formation of new organs, the coelomic wall meets the demand by reverting to the early embryonic condition in which the superficial layer takes essential part in producing mesenchymal cells. This process, characterized by abolishment of the basal boundary of the mesothelium, occurs during the earliest stages of gonad and adrenal formation. No true epithelial proliferations occur in either instance.
- In the gonads, the thus formed blastema differentiates into the epithelial primary sex cords and the separating mesenchyme. Transitional forms are represented by the interstitial tissue which may develop in the embryo from epithelial sex cords as well as from the unspecific mesenchyme ; this is parallelled in the adult by corpus luteum formation from both granulosa and theca cells.
- The adrenal cortex also develops from a mesenchymal blastema, and remains entirely mesenchymal for a considerable period of embryonic life. Only its most peripheral layers then acquire epithelial structure, and gradual transitions exist between these and the permanently non-epithelial central portion. Reports of new-formation of cortico-adrenal cell cords in the capsule of the organ are confirmed.
- In certain species the epithelial primordia of the miillerian ducts form, in their candal parts, considerable amounts of embryonic connective tissue in addition to the duets proper.
- Several examples of the proliferating activity of the coelomic epithelium in later embryonic stages are described. In some of these, the boundary of epithelium and mesenchyme remains intact, whereas in others the epithelium adds cells to the underlying mesenchyme. This can best be demonstrated in the tubal ridge, but it also happens in other places as, for instance, on the surface of the spleen.
- The present results are compared with those of tissue culture indicating that all or most of the bodv cells are capable of assuming epithelial as well as mesenchymal arrangement. The conclusion is reached that there is a gradual difference between the tissues developing from the coelomic wall and those of other parts of the body, to the effect that the former have under normal conditions a far greater tendency toward changes between epithelial and mesenchymal structure than the latter.
- It is to be expected that the elimination of the prejudice as to the stability of epithelial or mesenchymal structure in the organs discussed here, will help the understanding of their normal and pathological changes.
BacumMany, R. 1937 Ueber die Bedeutung des argyrophilen Bindegewebes (Gitterfasern) in der Nebennierenrinde und im Corpus luteum. Zeitsehr, f. mikr.-anat. Forsch., vol, 41, pp. 483-446.
- 1939 a Ueber die Nebennierenrinde des Meerschweinchens wihrend der Tragzeit. Zeitschr. f. mikr.-anat. Forsch., vol. 45, pp. 157-178.
- 1939 b Zur Frage der Zona germinativa der Nebennierenrinde, Klin.
Wochenschr., val. 18, pp. 783-784,
Baker, D. D., AND R. N. Barturr 1939 Role of capsule in suprarenal regeneration studied with aid of colchicine. Proc. Soe, Exper. Biol. and Med, vol. 40, pp. 117-121.
Bexnety, H. §. 1940 The life history and secretion of the adrenal cortex of the cat. Am. J. Anat., vol. 67, pp, 151-227.
BerceL, A., AND H. Gur 1934 Zur Friihentwicklung der Milz beim Menselen. Zeitschr, f. Anat. u. Entwicklungsgesch., vol. 103, pp. 20-29.
Broom, W. 1937 Cellular differentiation and tissue culture. Physiol. Rev., vol. 17, pp. 589-617,
CoRNER, G. W. 1920 On the widespread occurrence cf reticular fibrils produced by capillary endothelium. Contr. to Embryol., vol. 9, pp. 85-93.
Duke, K. L, 1941 The germ cells of the rabbit ovary from sex differentiation to maturity. J. Morphol., vol. 69, pp. 51-82.
Finarow, D, 1933 Ueber die Bildung deg Anfangestadiums bei der Extremitiitenentwicklung. Arch, f. Entwicklungsmech., vol. 127, pp. 776-802. 378 PETER GRUENWALD
FiscueL, A. 1913 Zur Entwicklungsgeschichte des visecralen Bindegewebes und der Zwischenniere. Anat. Hefte, vol. 48, pp. 153-164.
1930 Ueber die Entwicklung der Keimdriisen des Menschen. Zeitschr. f. Anat. u. Entwicklungsgesch., vol. 92, pp. 34-72.
CGRLLER, F.C. 1930 Zellveriinderungen im Hierstock der geschlechtsreifen weissen Maus nach Réntgenbestrahlung. Arch. f, Gyniikol., vol. 141, pp. 61-75.
Gomori, G. 1939 The effect of certain factors on the results of silver impregnation for reticulum fibers, Am, J. Pathol., vol. 15, pp. 493-495.
GRUENWALpD, P. 1934 Ueber Beziehungen zwischen der Beschaffenheit des Hodenepithels und den darunter gelegenen Blutgefaissen. Zeitschr. f. Anat. u. Entwicklungsgesch., vol. 102, pp. 424-433.
- 1937 Zur Entwicklungsmechanik des Urogenitalsystems beim Huhn. Arch, f. Entwieklungsmceeh., vol. 136, pp. 786-813.
- 1941 The relation of the growing miilleriau duct to the wolffian duct and its importance for the genesis of malformations. Anat. Rec., vol. 81, pp. 1-19. ,
- 1942 The development of the sex cords in the gonads of man and mammals. Am. J. Anat., vol. 70. (In press.)
Wer, J. 1927 Ueber das Keimepithel des Hodens. Zeitschr. f. mikr.-anat. Forsch., vol. 8, pp. 477-488.
- 1930 Vergleichende Untersuchungen tiber das persistierende Keim epithel des Hodens einiger Saiugcr. Zeitschr. f. mikr.-anat. Forsch.,
vol, 20, pp. 185-252.
1932 Vergleichende Untersuchungen iiber das persistierende Keiimepithel des Hodens ciniger Sauger. IT. Zeitschr. f. mikr.-anat. Forsch., vol, 28, pp. 529-564.
Hieeixs, G. M., ano D. J. Incue 1938 Functional homeoplastie grafts of the adrenal gland of newborn rats. Anat. Ree., vol. 70, pp. 145-154.
Hievcil, K. 1932 Ueber die erste Anlage der menschlichen Keimdriise und ihre geschlechtliche Differenzierung. Arch. f. Gyniikol., vol. 149, pp. 144-172,
Hotyoxer, E, A, 1936 The role of the primitive mesothelium in the development of the mammalian spleen. Anat. Rec., vol. 65, pp. 333-349,
Ivete, D. J., anp G. M. Higgins 19388 Autotransplantation and regeneration of the adrenal gland. Endocrinology, vol. 22, pp. 458-464.
- 1989 The extent of regeneration of the enucleated adrenal gland in the rat as influenced by the amount of capsule left at operation. Endocrinology, vol. 24, pp. 379-382.
JONCKHEERE, I. 1930 Contribution 4 l’histogénése de l’ovaire des Mammiféres. L’ovaire de Canis familiaris, Arch. de biol., vol. 40, pp. 357-436.
Kampmeicr, O. F, 1928 Concerning certain mesothelial thickenings and vascular plexuses of the mediastinal pleura, associated with histiocyte and fatcell production, in the human newborn. Anal. Ree., vol. 39, pp. 201-214.
1937 Origin and development of the mediastinal and aortic thyroids and the periaortic fat bodies. Illinois Med. and Dent. Monogr., vol. 2, pp. 1-82.
Kot, A. 1926 Ueber den Bau des embryonalen Pferdceierstockes. Zeitschr.
f. Anat. u. Entwicklungsgesch., vol. 79, pp. 366-390. ORGAN FORMATION FROM COELOMIC WALL 379
LILLIE, F. R. 1919 The developmeut of the chick, Holt, New York.
MaximMow, A. 1927a Ueber das Mesothel (Deckzellen der serésen Hiiute) und die Zellen der serésen Exsudate. Untersuchungen an entziindlichem Gewebe und an Gewebskulturen. Arch. f. exper. Zellforsch., vol. 4, pp. 1-42.
- 1927b Bindegewebe und blutbildende Gewebe. Handbuch der mikroskopischen Anatomie des Menschen, vol. II/1, pp. 232-583.
Maximow, A., AND W. BLoom 1938 ‘'ext-book of histology, Saunders, Philadelphia and London,
Parses, A. 8. 1926 On the occurrence of the oestrus cycle after x-ray stcrilization. Part I. Irradiation of mice at three weeks old. Proe. Royal Svc. London, Series B, vol. 100, pp. 172-199.
Perren, J. L. 1933 Beitrag zur Kenntnis der Entwicklung des Pferdeovariums. Zeitschr. f. Anat. u. Entwicklungsgesch., vol. 99, pp. 838-383.
PuLexkK, H. 1927 Ueber argyrophile Fasern (Gitterfasern) und ihre Bildungszellen. Erg. d. Anat., vol. 27, pp. 302-412.
PotrrzEr, G. 1936 Ueber die Frithentwicklung der Nebennierenrinde beim Menschen, Zeitschr. f. Anat. u. Entwicklungsgesch., vol. 106, pp. 40-48.
Sanmon, T. N., anp R. L. ZwemMer 1941 A study of the life history of eorticoadrenal gland cells of the rat by means of trypan blue injections. Anat. Ree., vol. 80, pp. 421-429,
Swuzy, O. 1938 The changing concept of ovarian rhythmus, Quart. Rev. Biol., vol. 8, pp. 423-433.
Tug, G. A., anD H. Downey 1921 The development of the mammalian spleen, with special reference to its hematopoietic activity. Am. J. Anat., vol. 28, pp. 279-339.
Uotita, U: U. 1940 The early embryological development of the fetal and permanent adrenal cortex in man. Anat. Ree., vol. 76, pp. 183-203.
Van Vuoten, J. G.C. 1927 De ontwikkeling van den testikel en de urogenitaalverbinding bij het rund. Inaug.-diss. Utrecht.
Zwemer, R. L., R. M. Worron anp M. G. Norkus 1938 A study of corticoadrenal cells. Anat. Ree., vol. 72, pp. 249-263.
EXPLANATION OF FIGURES
1 Coelomic wall of a human embryo of the third week. The cells of the surface lining have, on their proximal sides, the same structure as those of the underlying mesenchyme. Hemalum-eosin stain.
2 Coelomic wall at the lateral border of a mesonephros of a 7.5-mm. human embryo. Part of the surface lining still shows the same structure as in the earlier embryo (fig. 1). Azan stain.
3 Primordium of the adrenal cortex of a 9-mm. human embryo. Azan stain.
4 Part of the adrenal cortex (lower portion of fig.) and adjacent tissues of a human embryo of 48mm. CRL. Each cortico-adrenal cell is surrounded by a network of lattice fibers. Gomori’s silver impregnation.
5 Gonad region of a 9-mm. rabbit embryo. The basal membrane of the mesothelium is interrupted in the area of gonad formation (arrows). Azan stain,
EXPLANATION OF FIGURES
Lattice fiber structure in the postnatal adrenal cortex
6, 7, 8 Iluman adrenals. Gomori’s silver impregnation.
Dog adrenal. Gomori’s silver impregnation.
9 Same specimen and technic as shown in figure 9, at highcr magnification.
10 11 Horse adrenal, Gomori’s silver impreguation. 12 Same specimen and techuie as shown in figure 11, at higher magnification.
PLATE 3 EXPLANATION OF FIGURES
18 Adult guinea pig, zona glomerulosa (below) and capsule of the adrenal. Groups of cortico-adrenal cells are present in the capsule. Azan stain.
14 Adult human adrenal. Cell cords of the cortex radiate into the capsule. Azan stain.
15° Adult human adrenal. The outlines of cortical cell cords radiating into the capsule are shown by Gomori’s silver impregnation.
16 Miillerian duct and adjacent structures of a chick embryo of 6 days and 3 hours, M, miillerian duct; W, wolffian duct. Gomori’s silver impregnation.
17 Section of tho same series, 110 cranial to that shown in figure 16. Arrows point through the basal membrane of the original miillerian primordium to the portions being transformed into mesenchyme.
18 Section of the same scrics, 280 4% eranial to that shown in figure 17, The basal mombrane of the original miillerian primordium is disappearing (lower side of fig.), and the miillerian duct acquires a new basal membrane,
19 Miillerian duct and adjacent structures of a 22-mm. eat embryo, in a stage of transformation comparable to figure 17. Azan stain.
PLATE 2 3 PLATE 4
20 Spleen of a 12-mm, human embryo. The surface lining is distinetly separated from the underlying mesenchyme only along part of the surface. Gomori’s silver impregnation.
21 Spleen of a 160-mm. human cmbryo. No distinct basal membrane separates the mesothelium from the capsule. In the center of the figure a trabecula connects with the eapsule, and there lattice fibers scem to reach into the mesothelium. Gomori’s silver impregnation.
22 Mesonephros of a 14.5-mm. human embryo, A structure resembling an external glomerulus is attached to the dorsal surface of the organ.
23 and 24 Tubal ridge of a chick embryo of 7 days and J7 hours. M, miillerian duct; W, wolffian duct. Lattice fibers uppear in the tall mesothelium eovering the miillerian duct (fig. 24). 6804 cranial to this level (fig. 23), the greater proximal part of the tubal ridge is transformed into mesenchyme coating the miilllerian duct. The old basal membrane can no longer be distinguished. Gomori’s silver impregnation. .
25 Ducts connected with the caudal end of the coelomic eavity of a 5-day chick embryo. No definite boundary between cpithelium and mesenchyme ean be established. Gomori’s silver impregnation.
26 Posterior wall of the urogenital sinus of a 15-mm. dog ombryo. The mesothelium lines symmetrical invaginations. At their bottom no boundary between epithelium and mesenchyme ean be determined. Gomori’s silver impregnation.