Talk:Book - Manual of Human Embryology 19

Felix von Winiwarter (1852-1931)

Lie JT, Mann RJ & Ludwig J. (1979). The brothers von Winiwarter, Alexander (1848-1917) and Felix (1852-1931), and thromboangiitis obliterans. Mayo Clin. Proc. , 54, 802-7. PMID: 390261

Felix W. The development of the urinogenital organs. In Keibel F. and Mall FP. Manual of Human Embryology II. (1912) J. B. Lippincott Company, Philadelphia. pp 752-979.

XIX.

THE DEVELOPMENT OF THE URINOGENITAL ORGANS.

Bt W. FELIX, Zurich. 1

The urinary and reproductive organs, notwithstanding their diverse functions, have been regarded in systematic anatomy and in embryology as constituting a single system, an arrangement which is undoubtedly correct for two reasons. In the first place both groups of organs open into the enlarged caudal portion of the digestive tract, the cloaca; by a subdivision of this cloaca into a ventral portion, which furnishes the anlage for the urinary bladder, the urethra and the urogenital sinus, and a dorsal portion, the rectum, the digestive tract becomes completely and permanently separated from the urinogenital organs, but, on the other hand, these remain connected by the urogenital sinus. In the second place one of the urinary organs, the mesonephros, unites with the reproductive gland and, giving up its provisional function of manufacturing and excreting urine, enters permanently into the service of the reproductive apparatus as the male efferent duct. The urinary and reproductive organs consequently form an inseparable whole in the adult organism.

The study of the development of the urinogenital apparatus divides itself into four portions. These are: (1) the development of the excretory glands and their ducts; (2) the development of the reproductive glands and their efferent ducts; (3) the development of the urogenital union, i.e., the union between the urinary and reproductive glands; and (4) the development of the sinus urogenitalis and in connection with this the development of the external genitalia.

That it has been possible to base tbis entire chapter on my own observations, I am indebted to the generous assistance of colleagues who have unselfishly placed at my disposal series of sections from their private collections and from institute collections. In this connection I desire to thank Professor Grosser, Geheimrath Professor R. Hertwig, the late Geheimrath Professor His, Professor Hochstetter, Professor Keibel, Professor R. Meyer, the late Geheimrath Professor Pfannenstiel, Professor Stoerk. and the late Hofrath Professor Zuckerkandl. My colleague, Rob. Meyer, in addition to placing at my disposal over sixty series of sections, has given me the benefit of his rich experience. To my colleagues Dr. Baltischwiler, the late Professor U. Kronlein, Professor Wyder and Professor 0. Wyss I am indebted for fresh material and would also express to them my sincere thanks. 752

I. The Development of the Excretory Glands and their Ducts

Introduction.

In most vertebrates the development of the excretory system from its first formation until its completion occupies an interval which is long in comparison with that shown by other organs. In this interval its activity becomes modified in accordance with the growth of the entire organism, and this modification of the organ, still in the process of development, impresses upon the development of the kidney its distinctive feature. We have to deal not with the gradual development of a single organ, which is laid down and brought step by step to its definitive completion, but with a saltatory development; three organs (pronephros, mesonephros, and metanephros) are developed in succession, and each of these organs is apportioned and adapted to a definite period of one entire development. Each new excretory organ supplants its predecessor and the last to develop, the third, the metanephros, becomes the permanent kidney. The pronephros and lnesonephros are merely provisional kidneys, whose activities become superfluous and are supplanted on the formation of a new excretory organ. But when a structure in the animal body becomes functionless it undergoes degeneration; consequently the entire pronephros and the greater portion of the mesonephros disappear. The lesser portion of the mesonephros enters the service of the reproductive apparatus and is permanently retained in this new connection.

The development of the excretory system presents practically the same beginning and the same course throughout the entire series of vertebrates. Only the final results are different; the higher the grade of organization of the individual vertebrate classes, the higher the grade of development of the excretory organ; the number of provisional excretory organs stands in direct relationship to the grade of organization of the animal. In amphioxus and the myxinoids only a single excretory organ (the pronephros) is formed, in the teleosts and ganoids, the selachians, the petromyzonts, dipnoans and amphibia there is one provisional organ, the pronephros, and one permanent organ, the mesonephros. In the amniotes there are two provisional organs, the pronephros and mesonephros, and, as a permanent organ, the metanephros. The succession pronephros, mesonephros, " metanephros, occurs not only in the ontogenetic development of the individual, but also has occurred in phylogenetic development of the kidney within the vertebrate group. The pronephros, which is a provisional organ in the classes mentioned above, corresponds in part to the persistent excretory organ of amphioxus and the myxinoids, just as the mesonephros of the amniotes corresponds to the permanent excretory organ of the teleosts, ganoids, selachians, etc.

A knowledge of the entire developmental history of the kidneys is very necessary in the consideration of human development, in which the pronephros and perhaps also the mesonephros no longer function as excretory organs, but represent only inherited stages of development.

Each excretory organ, whether it be provisional or permanent, represents a gland with its gland substance and duct. The gland substance is formed of individual urinary tubules, which in the primitive condition run transversely to the long axis of the body and may, therefore, in contrast with the longitudinal duct, be termed the transverse tubules. The duct, which is lacking only in amphioxus, opens into the cloaca. The transverse tubules open either directly or indirectly by means of a collecting canal into the duct. This opening into a common duct unites the separate urinary tubules to a single structure, the gland substance; when the duct is lacking each tubule stands by itself and represents an independent unit. The individual tubules are named according to the nature of the entire gland, prouephric, mesonephric, or metanephric tubules. The pronephric and mesonephrie tubules are recognized by their openings being into the original or primary duct, the metanephric tubules by their opening into a secondary duct (the ureter) specially developed for them.

Differentiation of the Mesoderm.

GENERAL.

The excretory system is of mesodermal origin. The story of the development of its parent tissue may begin with the fully formed but as yet undifferentiated mesoderm.

The account that I shall give will not be based on any definite example, but will rather represent a schema based upon the study of the development in the craniotes in general. The mesoderm forms a wedge-shaped plate on either side of the medullary tube (Fig. 520a), the base of the wedge being its medial surface directed toward the medullary tube and the notochord, its edge being the lateral border. Each mesoderm plate differentiates in two rapidly succeeding intervals into three portions: 1, the primitive segment in its narrower sense, 2, the primitive segment stalk, and 3, the lateral plate. In the first interval only the dorsal portions of the wedge and a narrow strip in its immediate proximity are divided into a series of successive and approximately equal portions, the primary primitive segments; the lateral, broader part of the mesoderm plate remains undivided and forms the lateral plate (Fig. 520 b). In the second interval the primary primitive segments become more sharply marked off from the lateral plates by their lateral portions being folded in from all sides and so converted into a stalk, which bears at its extremity the medial, unaltered portion of the primitive segment (Fig. 520 c) ; this medial, larger portion is termed the secondary primitive segment, the lateral, smaller one the primitive segment stalk (connecting cord, intermediate cell-mass, nephrotome, gononephrotome).

The primitive segment stalks furnish the material from which the tubules of all three excretory organs are developed, and it follows from this that the anlage of the excretory system cannot be a unit, but must be composed of as many parts as there are primitive segment stalks entering into its formation. Furthermore it follows that the tubules are formed outside the body cavity enclosed by the lateral plates, i.e., they are from the beginning retroperitoneal.

With the progress of development the secondary primitive segments and the stalks become separated, the opening in the wall of each at the point of separation closing at the same time.

The stalk presents quite different arrangements in the representatives of the various vertebrate classes. It may : 1. preserve the form of an epithelial canal and its connection with the lateral plate, and be directly transformed into a uriniferous tubule; 2, it may become separated from tbe lateral plate and form either an isolated vesicle or an isolated solid mass of cells, and in this form it may be directly transformed into a tubule; 3, it may be gradually taken up into the lateral plate by an extension of the coelomic cavity of the plate, and is then only distinguishable from the lateral plate by its further development; 4, it may be transformed into a solid mass of cells and fuse with the neighboring stalks, similarly transformed, to form a single cord, known as the nephrogenic cord, which may or may not retain its connection with the lateral plates; or it may finally be transformed into mesenchyme tissue by the separation of its cells, this tissue then fusing with the mesenchyme of the sclerotome to form a single mass in which the source of the constituent parts can no longer be determined. Several of these five modes of development may occur in the same embryo, and, since neighboring segment stalks develop similarly, groups of segment stalks may be distinguished. Each primitive segment stalk may develop several tubules in succession. Since it itself passes along a definite path of development, during which its form changes, an opportunity is afforded for differences between the earlier and later formed uriniferous tubules. Pronephrie and mesonephric tubules of the same animal need not therefore show a similar development, if, between their formations there occurs a distinct interval, during which their common source has acquired a new form. Similarly the pronephrie tubules of different animals show profound differences in their development, differences which find an explanation in the time relations of the formation of the tubules. The change of form which accompanies the progressive development of the common source of the organ suffices to explain the difference in the development of a pronephric, mesonephric, and inetanephric tubule.

Fig. 520 a, b, and c. — Diagram illustrating the segmentation of the mesoderm, a shows the unsegmented mesoderm; b shows the first stage of segmentation, the separation of the primary primitive segment from the dorsal part of the mesoderm wedge; c shows the second stage of segmentation, the formation of the primitive segment stalk and the division of the primary primitive segment into the secondary primitive segment and the primitive segment stalk.

Each primitive segment stalk has theoretically the potentiality of producing all three kinds of tubules; a very complicated organ would be produced in this manner. Actually the arrangement is simpler, since the primitive segment stalks of the cranial zone produce only pronephric tubules, those of the middle zone only mesonephric {ubules, and those of the caudal zone only metanephric tubules, stalks, however, which produce two kinds of tubules, occurring in the intervals between the different zones.

The Segmentation of the Mesoderm in Man

In the human embryo the coeloni first appears as the exocoeloin and at an early period separates the mesoderm into intra- and extra-embryonic portions, the boundary between the two being indicated by the adjacent vena umbilicalis (Fig. 521). The intraembryonic mesoderm is at first absolutely solid, and while it is still in this condition its first segmentation takes place; the relations are accordingly as yet somewhat obscure. The formation of the primary primitive segments precedes the formation of a primitive segment plate, that is to say, the primary segments are separated from the lateral plates before they become separated from, each other, and in an embryo with 5 or 6 pairs of primitive segments the primitive segment plate therefore extends almost to the caudal extremity of the embryo. The grooves which delimit from the primitive segment plate the 5 or 6 cranial segments do not separate the primary primitive segments throughout their entire width, but penetrate only to the lateral border of the future secondary segment, the segment stalk therefore remaining at first unseparated. Fig. 521 is taken from an embryo with 5 or 6 primitive segments; on the right side (the left in the figure) the section passes through the sixth primitive segment, which is not yet separated caudally, and on the left side it passes through the primitive segment plate at a level corresponding to about the middle of the future seventh segment. The intra-embryonic mesoderm of either side is not yet cleft. While on the right the primary primitive segment is not delimited from the lateral plate, and there is no indication whatever by which to distinguish the segment stalk from the secondary segment, on the left the arrangement of the nuclei indicates the future triple division of the intraembryonic mesoderm into the secondary primitive segment, the segment stalk and the lateral plate ; two crosses in the figure indicate the limits of the three portions. During the segmentation the formation of the cavity in the intra-embryonic mesoderm occurs. The ccelom may form discontinuously both in the cranio-caudal and frontal directions; the clefts which succeed one another in the cranio-caudal direction do not show a segmental arrangement, but, on the other hand, those following in the frontal direction may correspond to the region of the lateral plates and that of the primitive segment stalks; the lumen of the lateral plate is not at first continuous with that of the exoccelom (Fig. 522). In an embryo with from 8 to 10 pairs of primitive segments the secondary segments are completely delimited and enclose a distinct lumen (the myoccel), but their stalks are not yet separated from the lateral plates (Fig. 522). The first distinctly separated segment plates are shown by the anterior 8 or 9 segments of an embryo with from 12 to 14 pairs of primitive segments; from the tenth segment onward all the stalks are united with the nephrogenic cord. In the same embryo the degeneration of the more anterior segment stalks has begun; they are becoming mesenchyme tissue. In an embryo with 23 pairs of segments isolated segment stalks are to be distinguished only indistinctly, but, on the other hand, a nephrogenic cord extends from the 13th segment to the unsegmented mesoderm.

Fig. 521. — Transverse section of a human embryo of 1.3S mm. greatest length (as determined from the series) and with 5-6 pairs of primitive segments. (Embryo Pfannenstiel-Kromer, from the collection of Professor Pfannenstiel; slide 6, row 3, section 7.) 120. The section passes on the right through the 6th segment and on the left through the primitive segment plate at the level of the future 7th segment. The section shows the ccelom relations. Only the exoccelomis present. On the right there is no apparent differentiation of the intra-embryonic mesoderm; on the left a beginning differentiation of the secondary segment, the segment stalk and the lateral plate is visible and is indicated by two crosses (x). The periintestinal rete and the anlagen of the aorta can be seen. An injury to the embryo accounts for the peculiar bending downward of the endoderm and visceral mesoblast in the left side.

Fig. 522. — Transverse section of a human embryo with 9-10 pairs of primitive segments and 1.73 mm. in length (determined from the series). (Embryo R. Meyer No. 335, from the collection of Professor R. Meyer, Berlin; slide 7, row 2, section 2.) < 135. The embryo is torn in the region of the medullary tube; there is a medullary tube and not a medullary groove. The yolk-sack is greatly folded, so that it appears as if divided into two portions; hence the peculiar appendage on the right side of the figure, which ia the connection with the second portion of the yolk-sack, not represented. The section passes through about the middle of the 7th primitive segment. The lateral plate contains a single wide ccelomic cavity, which is not, however, in communication with the exoccelom. The segment stalk is not distinguishable from the lateral plate. The visceral mesoblast of the yolk-sack is giving rise to the rete peri-intestinale.

Fig. 523 a and b. — Diagram of the development of the pronephros, a. The principal tubules 5, 6, 7, and 8 develop by evagination from the parietal mesoblast of the primitive segment stalk. They grow caudally <3 and„4) and reach and overlap the succeeding tubules passing laterally to them (2). The principal tubules, which are thus brought into apposition, fuse, and a collecting duct (1) is formed. 6. The collecting duct remains in connection with the various segment stalks by means of the principal tubules.

The human embryo forms, therefore, both secondary primitive segments and segment stalks. The latter retain their individuality only in the anterior 8-9 segments, and from the tenth segment outwards they fuse to form a nephrogenic cord or are separated from the unsegmented mesoderm as such a cord. The connection between the nephrogenic cord and the lateral plate is sometimes retained and sometimes lost; but the connecting bridges, when present, show no segmental arrangement.

The Development of the Pronephros

GENERAL.

The pronephros in man no longer functions as an excretory organ and its development has therefore become abbreviated, incomplete, and in consequence obscure. And a further difficulty is introduced by the fact that its appearance and formation occur in very early stages of development, for the study of which good material is only sparingly available ; consequently the account of the development of the human pronephros must be incomplete. I shall therefore present a review of the development of the pronephros within the vertebrate stem and what is known as to its development in man may be readily included in this review.

The first anlage of the pronephros consists in the development of the pronephrie tubules. The parietal layer of each primitive segment stalk forms an evagination directed toward the ectoderm, the principal tubule (Fig. 523 a, tubules 7 and 8). Each principal tubules extends caudally beneath the ectoderm, without, however, uniting with it (Fig. 523 a, tubules 6, 5, 4, and 3). The tubule reaches the next succeeding one, grows backward along its lateral surface (Fig. 523 a, tubule 2), and fuses with it (Fig. 523 a, tubule 1) ; by the fusion of all the principal tubules a longitudinal canal, the collecting duct, is formed and this is connected with each primitive segment stalk by a principal tubule (Fig. 523 b). The principal tubules of the various segments do not appear simultaneously, but are formed in successive groups, cranial ones first and then caudal ones. Once the collecting duct and the principal tubules are formed a differentiation of the primitive segment stalk begins (Fig. 524 a, b, and c). In Fig. 524 a a transverse section through the segment stalk and the pronephric tubule is represented, the segment stalk being hatched and the tubule black. In Fig. 524 b the secondary primitive segment has separated from the segment stalk, the stalk which has thus become free separates the elements of its medial portion to form mesenchyme tissue, indicated in the figure by the three dots, and the lateral portion of the stalk closes the opening thus formed and now appears as a connecting canal between the principal tubule and the lateral plate; we shall term it the supplemental tubule. In Fig. 524 c the principal tubule (black) and the supplemental tubule (hatched) have elongated; the supplemental tubule, by the broadening of the portion immediately adjacent to the principal tubule, has become divided into two parts, a broad medial one, which is invaginated by a vascular glomerulus, and a narrow lateral one, which establishes the connection with the lateral plate. The broad part may be termed the inner pronephric chamber, the narrow one the nephrostome canal. A vascular glomerulus similar to that which invaginates the medial wall of the pronephric chamber, pouches out the visceral layer of the lateral plate close to the opening of the nephrostome canal into the body cavity (Fig. 524 c). Both glomeruli are supplied by branches from the aorta; they may be termed the internal and external glomeruli. The internal one is a portion of the pronephrie canal, the external one is an independent structure and is related to the pronephrie tubule only by its neighboring position. The portion of the body cavity that contains the external glomerulus may separate from the rest of the ccelom and thus constitute an outer pronephrie chamber. These are all the constituent parts of a pronephrie tubule, and together they form a pronephrie segment; this in its fully developed condition consists of the principal tubule, the pronephrie chamber, the nephrostome canal, the internal and external glomeruli and, in some cases, the outer pronephrie chamber.

Aorta Fig. 524 a, b and c. — Diagram of the development of a pronephrie segment. The segment stalk is hatched, the principal tubule is black, a. The principal tubule arises as an evagination of the parietal layer of the primitive segment stalk, its lumen being continuous with that of the segment stalk and through this with that of the lateral plate, b. The segment stalk has lost its connection with the secondary primitive segment, its medial portion has become converted into mesenchyme tissue and its lateral part has united with the principal tubule. The lateral part of the stalk thus becomes a tubule supplemental to the principal tubule, c. The supplemental tubule (hatched) becomes divided into the inner pronephrie chamber and the nephrostome canal. A glomerulus, forming the internal glomerulus and supplied by the aorta, invaginates into the inner pronephrie chamber, the external glomerulus projects into the general body cavity (ccelom of the lateral plate). The pronephrie segment is now complete in all its parts; starting from the collecting duct we have the principal tubule, the pronephrie chamber with the internal glomerulus, the nephrostome canal and the external glomerulus.

Originally each primitive segment stalk throughout the entire length of the body cavity produced a pronephrie tubule. By the union of these tubules an excretory duct, the collecting duct, was formed, extending throughout the entire length of the body cavity, and this, or the last tubule, opened into the cloaca. This arrangement still persists in the myxinoids and the telosts, probably also in the ganoids, petromyzonts, amphibia and dipnoans. In the remaining classes, the selachians, gymnophiones and amniotes, there is an abbreviation of the pronephric anlage, so that pronephrie tubules are formed only in a varying number of the cranial segments. But since the excretion must still be conveyed to the cloaca, one of the most caudal tubules elongates on its own account and becomes the excretory duct. It grows caudally between the ectoderm and mesoderm and thus reaches the cloaca. During its elongation it lies in close apposition to the ectoderm, but probably is not in connection with it. In such a pronephrie excretory duct, which we will term a primary excretory duct, two portions may be distinguished: one, the collecting duct portion, formed by the fusion of the principal tubules, and another, the free terminal portion, produced by the independent backward elongation of a pronephrie tubule. If the pronephros of the selachians, gymnophiones and amniotes is an organ which has undergone retrogression, if its excretory duct has been formed under ccenogenetic conditions, the results of their ontogeny must be received with caution. At all events clear evidence as to the relation between pronephros and mesonephros cannot be obtained from these forms.

The Pronephros in Man

The human pronephros is a quite rudimentary organ and its development is accordingly abbreviated and incomplete. The abbreviation reveals itself in the transitoriness of its occurrence, the incompleteness in the imperfection of the individual pronephrie segments and in the small number of primitive segment stalks that form pronephrie tubules. The abbreviation of the entire anlage necessitates — according to the general description (p. 753 and p. 759) — the division of its development into two portions: first, the development of the glandular portion — the pronephrie tubules and the collecting duct — and, second, the development of the free terminal portion of the primary excretory canal.

The first anlage of the pronephrie glandular portion occurs in an embryo with from 9 to 10 pairs of primitive segments and a greatest length (determined from the series) of 1.73 mm. A stage of greatest development cannot exist, since the cranial canals have already dissolved into mesenchyme tissue while the caudal ones are developing; the complete formation of these last is almost reached in an embryo of 2.5 mm. greatest length, with 23 pairs of primary segments. The degeneration of the pronephros is already well advanced in an embryo of 4.25 mm. vertex-breech length with from 27 to 28 pairs of primitive segments and, accordingly, takes place between the stages with 23 and 28 pairs of segments. The time of disappearance of the pronephros cannot be determined definitely, since it extends into the territory of the mesonephros, and the cranial tubules of this also degenerate immediately after their formation; it is consequently impossible to determine the significance of the remains of a tubule occurring in the pronephrie region; it may represent a pronephrie tubule, it may be a mesonephric tubule in process of degeneration, or, finally, it may be the remains of a primitive segment stalk or of the nephrogenic cord. As the approximate time for the complete disappearance of the pronephros a stage of 4.9 mm. vertex-breech length may be named. But it is to be remembered that one constituent of the pronephros, the external glomerulus, may, on account of its independence, persist far beyond this stage.

The pronephric anlage extends from the 7th to the 14th primitive segment. It is very probable, however, that it may extend further forwards, since an embryo of 2.6 mm. greatest length and with 12-14 pairs of primitive segments showed on one side pronephric rudiments in the 2nd, 4th and 6th segments. The conditions in the other amniotes are also in favor of the possibility of this arrangement. The anlagen probably appear successively in groups. In an embryo with 9-10 pairs of primitive segments the pronephros extended from the 7th to the as yet unseparated 11th segment, and in one with 12-14 pairs of segments it reached from the 7th to the 14th segment.

Fig. 525. — Human embryo with 9-10 pairs of primitive segments and a length of 1.73 mm. (determined from the series). (Embryo R. Meyer 335, in the collection of Professor R. Meyer, Berlin; slide 7, row 3, section 4.) ' 135. The section passes on the right side through the cranial wall of the 8th primitive segment, on the left through the interstitium between the 7th and 8th segments (reversed in the figure). The ccelom is continuous with the exoccelom, the original boundary between them being marked by the vena umbilicalis. The ccelom is greatly enlarged, only the medial portion, corresponding to the primitive segment stalk, remaining narrow. On the left side of the figure the ccelom of the primitive segment stalk and that of the lateral plate are separate, and the parietal layer of the segment stalk is outpouched to form the first pronephric principal tubule.

The pronephric tubules appear as ridge-like thickenings of the parietal layer of the individual primitive segment stalks, and the lumen of the stalk may usually be followed into the thickening. Fig. 525 shows the pronephric tubule of the 7th segment (which is also the first pronephric segment) of an embryo of 1.73 mm. greatest length (determined from the sections) and with 9-10 pairs of primitive segments. The anlage begins in the posterior half of the 7th segment and extends to the anterior half of the 8th segment. The next anlage begins in the posterior half of the 8th segment, the third at the anterior edge of the 9th segment.

Only the most anterior anlagen are independent and completely separated from one another, showing a certain metamerism.

Fig. 526. — Reconstruction of the right pronephric anlage of an embryo of 2.6 mm. greatest length, with 13-14 pairs of primitive segments. (Embryo Pfannenstiel III, from the collection of Professor Pfannenstiel.) The contours of the embryo, the medullary canal and the intestine are outlined, the mesoderm (secondary primitive segment, segment stalk and lateral plate) is shaded, the pronephric anlage is black. The anlage consists of five parts, a rudimentary tubule in the 7th segment, two well-developed tubules in the 8th and 9th segments, a pronephric ridge in the 10th, 11th and 12th segments and another ridge in the 13th and 14th segments. Luminain the principal tubule and in the primitive segmentjstalks are left white.

For the posterior ones one or two continuous longitudinal ridges are formed, which lie in a direct line with the isolated tubules; I shall call these the pronephric ridges to contrast them with the pronephric tubules. The ridges are in some places separated from their parent tissue, in others they remain in connection with it and at each of these connection regions they possess a lumen. In Figs. 526 and 527 I give a reconstruction of the pronephric anlage of an embryo of 2.6 mm. greatest length and with 12-14 pairs of primitive segments ; the pronephros here reaches its greatest extent. The arrangement is somewhat clearer on the right side (Fig. 526) than on the left. The secondary primitive segments, the segment stalks and the lateral plate are shaded and the pronephros is black; the supplemental canals and the connections of the pronephric ridges with the segment stalks or the nephrogenic cord, have been in the figures displaced laterally to the boundary line of the lateral plate, although in reality they lie beneath the nephrogenic cord. Finally, where a lumen occurs, it is represented as white. In the 7th segment there is a rudimentary anlage and the 8th and 9th each contains a completely developed one. There are therefore three separate pronephric tubules. The second and third tubules are united together and with the following pronephric ridge. In the stretch from the 10th to the 12th segment and again in that from the 13th to the 14th there follow two pronephric ridges ; the lumina in the anterior one show, though not strongly, a metameric arrangement. The left side of this embryo shows the rudimentary anlagen in the 2nd, 4th and 6th segments that were previously mentioned.

Fig. 527. — Reconstruction of the left pronephric anlage of the embryo shown in Fig. 526. The pronephros consists of rudiments in the 2nd, 4th, 6th, 7th and 8th primitive segments and of a long pronephric ridge extending from the 9th to the 14th segment. The horizontal lines in both figures indicate the locality of the transverse sections.

This embryo with 12-14 pairs of primitive segments already shows progress in the development of the individual tubules. The anlagen of the 8th and 9th segments have separated from their parent tissue, have extended backwards to the next anlage and have fused with it to form the collecting duct. This further development corresponds in time with the establishment of continuity between the exo- and entoccelom and with the enlargement of the latter. In the entoccelom, however, only the ccelom of the lateral plate is enlarged, that of the segment stalk remains narrow so that the stalk looks like a tubule and gives the false impression of a direct origin of the pronephric anlage from the lateral plate. Fig. 528 passes on the right (left in the figure; the location of the section is shown in Fig. 527) through the cranial end of the third pronephric tubule, which here, notwithstanding its shortness, consists of a principal tubnle and the supplemental tubule formed from the primitive segment stalk. A comparison with the left side (right in the figure), where the section passes through the separated end of the third left pronephric tubule, shows that there the principal tubule and the segment stalk are distinctly separate. To show the arrangement in the pronephric ridge I give in Fig. 529 a second section through this embryo, which on the right (left in the figure) cuts the ridge just where it possesses a distinct lumen. Here the segment stalk together with its pronephric anlage is separated from. the lateral plate, although it lies in close apposition to it. The existence of the two components is to be made out only by comparison with the anterior tubules and from the further development. Finally the question as to the caudal end of the pronephric anlage is yet to be considered. In the embryo with 12-14 pairs of primitive segments, in which the pronephric anlage reaches its greatest caudal extension, it passes insensibly at its caudal end into the unsegmented mesoderm. The section shown in Fig. 530 passes through the 13th or 14th segment ; its location is shown in Figs. 526 and 527. On the left side of the figure there is merely a thickened primitive segment stalk, separated both from the primitive segment and the lateral plate ; it is termed in the figure the nephrogenic cord and pronephric ridge: on the right side the segment stalk is in connection with the lateral plate and the lumen of the latter is continued directly into the stalk. In Fig. 527 the leader ' ' Fig. 530 ' ' shows how far the lumen passes beyond the medial edge of the lateral plate. More caudally the right (left in the figure) thickened segment stalk also becomes connected with the lateral plate and then the thickening fades out so gradually that a definite limit for its extremity cannot be established. The pronephric anlage is, accordingly, as far as the 14th primitive segment purely mesodermal, any participation of the ectoderm in it is excluded.

Fig. 528. — Embryo of 2.6 mm. greatest Jength and with 13-14 pairs of primitive segments. (Embryo Pfannenstiel III, from the collection of Professor Pfannenstiel; slide 9, row 3, section 4.) X 182. The section passes on both sides through the caudal wall of the 9th primitive segment. On the left side the third, right pronephric segment, with the principal and supplemental canal, is cut. On the right side the section passes through the caudal end of the third, left pronephric principal canal and shows its relation to the primitive segment stalk. Between the ectoderm and the visceral mesoblast is the vascular layer, in which are young red blood-corpuscles, distinguished by the abundance of chromatin in their nuclei. The tears in the medullary canal and beside the pronepheric anlagen are artefacts.

Fig. 529. — Human embryo of 2.6 mm. greatest length and with 13-14 pairs of primitive segments. (Embryo Pfannenstiel III, from the collection of Professor Pfannenstiel; slide 10, row 3, section 2.) X 182. The section passes on both sides through the cranial half of the llth primitive segment. On the right Heft in the figure) the primitive segment stalk is cut and also the pronephric ridge and a lumen corresponding to the llth pronephric tubule. The pronephric ridge and the segment stalk are not delimited from each other. On the right side the same arrangement. Between the visceral mesoblast and the digestive canal are the cells of the vascular layer. On the right side a ramus intestinalis passes to the vascular layer.

14th primitive segment

Pronephric ridge and nephrogenic cord

Fig. 530. — Human embryo of 2.6 mm. greatest length and with 13-14 pairs of primitive segments. (Embryo Pfannenstiel III, from the collection of Professor Pfannenstiel; slide 12, row 1, section 5.) X 182. On the right side (left in the figure) the section passes through the forming cranial wall of the 14th primitive segment, on the left through the caudal wall of the 13th segment. On the right (left in the figure) it cuts the nephrogenic cord. A comparison with the right side of Fig. 528 shows at once that the primitive segment stalk is greatly enlarged, the enlargement being due to the pronephric anlage. The anlage and the primitive segment stalk together form the nephrogenic cord. On the left side (right in the figure) the section passes through a point where the nephrogenic cord, delimited otherwise as on the right side, is fused with the lateral plate and receives a cleft of the coelom. The rete peri-intestinale is well developed, since we are here in the region of the art. umbilicalis. Between the aorta and the rete there is on the right side of the figure another ramus intestinalis.

At this point there is a gap in the available material. The next oldest embryo is 2.5 mm. in length and possesses 23 pairs of primitive segments. The pronephros in this is almost completely developed — so far at least as one may speak of its completion. It consists, as is shown in Fig. 531 a and b, of a number of tubules and the primary excretory duct. There are in all 7 tubules present, the most anterior of which is not united with the succeeding tubule, as was the case in the younger embryo. This fact, together with what was found in the embryo with 8-10 pairs of primitive segments, in which the most anterior anlage was in the 7th segment, and with what was found in the embryo with 12-14 pairs of segments, in which degenerating anlagen occurred in the 7th and 8th segments, confirms the conclusion that the cranial end of the pronephros degenerates and that only the tubules of the 10th and succeeding segments undergo a further development. The six pronephric segments that are developed, the 2nd to the 7th, lie in the 10th, 11th and 12th segments and in the interstitium between the 12th and 13th. Whether this concentration of six pronephric tubules within the limits of 3y 2 body segments indicates a primary dysmetamerism or has resulted from the approximation of originally more separated anlagen cannot be determined. Each pronephric segment (Fig. 531 b) shows distinctly a thickened beginning portion, the inner pronephric chamber, and a smaller canallike portion, the principal tubule. The cells in the. inner pronephric chamber are arranged in such a way as to be concentric to a latent lumen (Fig. 532). The tubules 2-5 have fused and so form a collecting duct (Fig. 531 b). The tubules 5, 6 and 7 are not yet united but their union is imminent (Fig. 531 b) ; this finding is conclusive as to the pronephric character of the 6th and 7th and therefore of the preceding tubules. In Fig. 532 the detailed relations of the 6th and 7th tubules are shown : section a cuts the principal tubule of the 6th pronephric segment and the corresponding 6th pronephric chamber is visible as a marginal section ; Fig. b shows below the principal tubule of the 6th segment the primitive segment stalk of the 7th segment ; the arrangement of its cells divides it into two portions, a medial thickened portion with its cells arranged concentrically, the inner pronephric chamber of the 7th tubule, and a lateral smaller portion whose cells are not concentric ; this can only be the portion that I have termed in the general description the nephrostome canal, and section d shows its connection with the lateral plate. Section c shows the 7th inner pronephric chamber with the 7th principal tubule arising from it. Between this and the ectoderm are four additional cells, which represent the pointed end of the 6th principal tubule. Finally section d shows the 7th principal tubule separated from its pronephric chamber ; the 6th principal tubule has disappeared.

Fig. 531 a and 6. — a. Reconstruction of the left pronephric and mesonephric anlagen of a human embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments. (Embryo R. Meyer 300, from the collection of Professor R. Meyer, Berlin.) The figures by Thompson were employed for the reconstruction. The pronephros consists of seven tubules; the most anterior in the 9th segment is in process of degeneration, the 2nd-5th have united to form the collecting duct, the 6th tubule is still free, but is about to unite with the 5th and 7th. The 7th is continued into the free terminal portion of the primary excretory duct and this extends to the unsegmented mesoderm. Beyond the 7th tubule is the beginning of the nephrogenic cord and in the region of the 13th-15th segments mesonephric vesicles are developed within this. b shows the pronephros under a higher magnification.

Fig. 532 a, b, c, and d. — Four sections from a human embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments. (Embryo R. Meyer 300, from the collection of Professor R. Meyer, Berlin; slide 11, row 5, sections 4 and 6; slide 12, row 1, sections 5 and 7.) X 238. a. Section through the 6th pronephric segment; the caudal wall of the pronephric chamber is just touched, the principal tubule is fully cut. Ventral from the pronephric segment is the anlage of the v. cardinalis post. b. Section through the 7th pronephric chamber and the 6th principal tubule; in the visceral mesoblast a genital cell. c. 7th pronephric chamber and 7th principal canal continuous, d. 7th pronephric segment separated into pronephric chamber and principal canal; rudimentary 7th nephrostome canal. The sections are 5/x thick; if a be taken as the first, then b is the third, c the eighth, and d the tenth section.

In the region of the inner pronephric chamber there is never, so far as the material furnishes a definite indication, a formation of an internal glomerulus; but, on the other hand, in several embryos there was a rudimentary external glomerulus in whose interior sections of vessels could be seen. The external glomerulus first becomes formed when the pronephros is in full degeneration.

The human pronephros is by far the best developed within the group of the mammals. It shows its relations more clearly than it does in the birds and is at least as distinct as in the reptiles. With the exception of the internal glomerulus it possesses all the constituent parts that have been given in the general description as characteristic of a fully formed pronephric segment; the principal canals and the collecting duct formed from them, the inner pronephric chamber, rudiments of nephrostome canals and one or several external glomeruli.

Fig. 533 a, b and c. — Three sections through the caudal end of the primary excretory duct. Human embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments. (Embryo R. Meyer 300, from the collection of Professor R. Meyer, Berlin; slide 13, row 4, sections 5, 6 and 7.) X ISO. a. The end of the primary excretory duct is still to be distinguished from the ectoderm, b. Between the end of the primary excretory duct, consisting of a single cell, and the ectoderm there is interposed a cell in the act of division; whether it belongs to the ectoderm or the duct is uncertain, c. Section immediately behind the end of the duct; the ectoderm does not show the slightest alteration. The rete peri-intestinale between the visceral mesoblast and the intestine.

The development of the free terminal portion of the primary excretorv duct cannot be clearly made out from the available material. While the embryo with 12-14 pairs of primitive segments did not possess such a structure, in the embryo with 23 pairs of segments it is developed throughout almost its entire length. It appears in this (Figs. 558 and 559) as a solid, semilunar mass of cells, which is in apposition by its concave surface with the nephrogenic cord or the mesonephric vesicles ; it shows extensive variations in calibre and remarkably rare mitoses, while these are abundant in the neighboring tissues. Towards its caudal end it gradually diminishes in size, until, finally, it consists of only 1-3 cells and ends at the ectoderm. In Figs. 533 a, b and c the two last sections through the secretory duct and that immediately following are shown. One sees in section a the apposition of the duct to the ectoderm, but the two may yet be distinguished to a certain extent. In Fig. b the distinction is no longer possible; a mitosis is to be seen, but it is impossible to say whether it belongs to the excretory duct or to the ectoderm, its spindle being so placed, however, that the daughter cells must lie sagittally one behind the other. Finally, in c every trace of the duct has disappeared and there is not the slightest indication that preparations for its formation are to be met with in the ectoderm. The entire apposition has at most an extent of 50 micra and the sections that precede it sometimes show a sharp delimitation of the duct from the ectoderm and sometimes do not. It may furthermore be noted that the emigration of cells from the parietal mesoderm to form mesenchyme is already taking place and that these emigrated cells become interposed between the parietal mesoderm and the ectoderm, just as is the end of the primary excretory duct, and frequently their delimitation from the ectoderm cannot be made out. This is the only series which, perfectly preserved, shows the excretory duct on its way to the cloaca. I can make out from it neither a connection of the duct with the ectoderm nor a preparation for the formation of the duct on the part of the ectoderm. I cannot settle definitely the question as to the participation of the ectoderm in the formation of the excretory duct, but I am inclined to deny any such participation. As regards the formation of the duct from the 13th segment caudally, where in the embryo with 23 segments it lies free between the ectoderm and mesoderm, a definite conclusion is not possible. It may be recalled that in the embryo with 12-14 pairs of primitive segments the pronephric anlage was in connection with the mesoderm to beyond the 14th segment and that in its anterior and middle portions it was divided into pronephric tubules and collecting duct. It is accordingly quite possible that the excretory duct had also been formed in the 13th and 14th segments by splitting from the pronephric anlage and the possibility that the same process is repeated in the more caudal segments must be given consideration. The variations in the caliber of the excretory duct and the lack of mitoses throughout its entire extent indicate a formation in loco.

The free terminal portion of the excretory duct has reached the cloaca in an embryo with 27-28 pairs of primitive segments and of 4.25 mm. vertex-breech length. It runs in a straight course, close to the ectoderm in the angle between the urogenital fold and the parietal inesoblast, to the level of the caudal surface of the 28th primitive segment, and there it bends ventrally and terminates in the angle between the cloaca and the ectoderm, close to the cloacal membrane (Fig. 534). The actual situation of its contact with the cloaca is caudal to the middle of the cloacal membrane and at the junction of the middle and lower thirds of the cloaca itself. The excretory duct is not evenly developed in this embryo ; from the 9th or 10th to the 14th segment its diameter is small, but from the 14th segment on it increases rapidly in size and remains large to its termination, its blind end being enlarged to almost double the original size. The lumen appears first at two points, at the anterior end in the region of the 10th and 11th segments and in the enlarged posterior extremity. Its further development takes place discontinuously for the most part and the various cavities show no regularity in their arrangement. Frequently the lumen of the cloaca evaginates into the blind end of the excretory duct.

Fig. 534.— Human embryo of 4.25 mm. vertex-breech length and with 27-28 pairs of primitive segments. (Embryo H. M. I., from the collection of the Anatomical Institute, Zurich; slide 399, J. 13,2, row 15, section 5.) X 150. The ends of the primary excretory ducts are on either side of the cloacal membrane.

The perforation of the cloaca by the duct is said to occur in embryos of 4.2 mm. greatest length (Keibel, 1896), but I have first found it in an embryo of 7 mm. greatest length.

In man the inner pronephric chamber, as it appears, is not invaginated by a vascular glomerulus and the external glomerulus is only a transitory formation; it contains a vessel it is true, but the relation of this to the rest of the vascular system has not been discovered. Where the actual pronephric vessels are to be looked for must first be learnt from comparative embryology.

In amphioxus the vessels pass in arches from a longitudinal vessel, situated ventral to the digestive tract, to a dorsal longitudinal vessel above the digestive Longitudinal commissure of the viscero-ventral arch system. Dorsal longitudinal vessel tract; these arches lie between the wall of the digestive tract and the visceral mesoblast (Fig. 535). To distinguish these arches from the dorsal arches which supply the dorsal musculature, the spinal ganglia and the spinal cord, I shall call them the ventral arches, and to further distinguish them from a second ventral arch system which supplies the ventral musculature and corresponds to the intercostal system of arteries, they may be called the viscero-ventral arches, the intercostal arteries being then termed the parieto-ventral arches.

Fig. 535. — Diagram of the viscero-ventral arch system in amphioxus.

Fig. 536. — Diagram of the viscero-ventral vascular arches of a selachian.

In the viscero-ventral arch system of amphioxus there are retia mirabilia which surrounded the excretory ducts and their solenocyte areas, and these are so arranged that they unite successive arches (Fig. 535) ; thus the paired longitudinal commissure of the viscero-ventral arch system is formed.

In the selachians only a greatly reduced viseero-ventral system occurs. It is no longer a continuous system, but is divided into two groups of arches, a cranial one, formed by the aortic arches, and a caudal one, formed by the pronephric arteries (Fig. 536). The pronephric arteries, which, in consequence of the rudimentary development of the selachian pronephros, do not form retia mirabilia, fuse later on the right side to form a simple and unpaired a. vitellina. The arteries on the left side, which from the beginning are merely rudiments, vanish completely. Later the a. vitellina gives off the a. mesenterica and becomes the a. omphalo-mesenteiica. The supply of the digestive tract is, accordingly, from a part of the left viscero-ventral arch system.

In the ganoids (Fig. 537) we have the same division into groups and a still greater limitation of the viscero-ventral system. A cranial group is formed, as in the selachians, by the aortic arches, a caudal one — its relations in early stages of development have not yet been studied — is formed by the roots of the pronephric vessels. This caudal group is, however, incomplete, since neither of the two arches present reaches the longitudinal vessel on the ventral side of the intestine (v. subintestinalis). The glomerulus itself forms a rete mirabile, which is spread out between the arches like a longitudinal commissure. The glomerulus is paired in its cranial portion and unpaired caudally, so that its rete mirabile has two poles cranially and one caudally. All three poles are in connection with longitudinal vessels; the two anterior ones run cranially along the dorsal surface of the intestine, reach the arches of the cranial group and unite with at least the sixth and fifth; the caudal longitudinal vessel passes backward as the a. mesenterica, supplying the entire intestine as far as its terminal portion.

Fig. 537. — Diagram of the viscero-ventral vascular arches of amia calva.

Fig. 538. — Diagram of the viscero-ventral vascular arches of a trout.

There are accordingly in the ganoids two paired commissural vessels, which, like the paired longitudinal commissures of amphioxus, unite the arches of each group and also the groups themselves. In the region of the caudal group the paired commissures unite to form the unpaired a. mesenterica. Consequently in the ganoids also the vessel which supplies the digestive tract arises from the viscero-ventral arch system.

In the teleosts (Fig. 53S) there is again the same grouping of the visceroventral arch system, the cranial group being formed by the aortic arches and the caudal group by the pronephric vessels. The caudal group is again incomplete, although not to the same extent as in the ganoids, for, in the first place, the number of its arches is greater and, secondly, one arch — and that on the right side — extends to the yolk-sack and so represents the a. vitellina. This fact justifies the assignment of the pronephric arteries of the ganoids and teleosts to the viscero-ventral arch system. Between the individual arches of the caudal group the glomerulus is again drawn out to form a longitudinal commissure. In its cranial half it is paired, in its caudal half it is unpaired, and therefore has three poles, as in the ganoids. It is prolonged, caudally only, into a longitudinal vessel that becomes the a. mesenterica. An important fact is still to be noted: the a. mesenterica in the embryo is connected with the aorta by arches not only in the region of the pronephric glomerulus, but also throughout its further course; the last arch occurs at the level of the cloaca. The caudal group of the visceroventral arch system accordingly extends in the teleosts "throughout the entire posterior half of the body.

Among the amphibia the urodeles possess numerous intestinal arteries, the anura only one. The course of this single vessel is, however, such that we may derive the arrangement in the anura from that occurring in the urodeles by supposing the existence of a longitudinal commissural vessel, which united the individual intestinal arteries of the urodeles and thereby furnished a possibility for their intestinal arteries, with one exception, to relinquish their connection with the aorta. The actual development of the arteries in the anura has not yet been studied. But in any case it may be supposed that the urodeles possess both groups of the viscero-ventral arches; the unpaired condition of the posterior group offers, as we have seen, no obstacle to this homology.

In the reptiles the region of the a. omphalo-mesenterica is originally supplied by a large number of aortic branches, which arise for the most part in pairs, but are partly unpaired. With the formation of a mesentery the paired vessels unite together, and one of these unpaired vessels becomes the a. omphalo-mesenterica, while the others degenerate (Hochstetter, 1898). The original paired intestinal arteries may be identified as the caudal group of the viscero-ventral arch system.

In birds and mammals, finally, there are in early stages of development numerous intestinal or vitelline arteries, arising from the as yet ununited aortse and passing over into a vascular network that surrounds the wall of the yolk-sack. There is not the slightest difficulty in the way of regarding these as the representatives of the caudal group of the viscero-ventral arch system.

From what has been said above we arrive at the following plan of development for the intestinal arteries. A viscero-ventral arch system unites a ventral and a dorsal longitudinal vessel ; the individual arches of the system may originally have been arranged metamerically. In each of the various arches a rete mirabile is interposed, surrounding the region of a uriniferous tubule and uniting two successive arches. From the totality of these retia mirabilia a longitudinal commissure is formed. The gills and uriniferous tubules, originally lying in the same segments, migrate into different regions of the body, and both become, reduced in number in the course of the phylogeny.' There thus comes about the separation of the viscero-ventral arches into a cranial and a caudal group. The anterior group enters the service of the branchial apparatus as the aortic arches, the posterior one is at first devoted to the supply of the pronephros and later, on the degeneration of this, enters the service of the digestive system. On the transformation of the caudal group of arches into the intestinal arteries they undergo such changes that their original arch character is completely lost; both the longitudinal commissural vessels play an important part in the transformation.

Fig. 539. — Reconstruction of the arterial system of an embryo of 1.38 mm. greatest length and with 5-6 pairs of primitive segments. (Embryo Pfannenstiel-Kromer, from the collection of Professor Pfannenstiel.) The aorta is a continuous vessel with completely closed walls from the level of the second primitive segment forward, from there backwards it is in anlage. On the posterior part of the yolk-sack and on the end-gut is the anlage of the rete peri-intestinale. The a. umbilicalis arises from this rete.

Our task in connection with the investigation of the human pronephric vessels is now defined. We have to seek for the caudal group of the viscero-ventral arch system and its longitudinal commissure. The study of the development of the intestinal arteries and especially of the a. cceliaco-mesenterica must at some point lead to the viscero-ventral arch system.

The intestinal arteries in man develop from a vascular network which surrounds the intestine and yolk-sack. Fig. 539 shows the reconstruction of the vessels of an embryo with 5-6 pairs of primitive segments. The aorta as far as the second primitive segment is developed as a tube with completely closed walls; from there on it is still in anlage, appearing in section sometimes as a closed ring, sometimes as composed of individual cells. From the future 7th segment to the tip of the tail a vascular network, the rete peri-intestinale, is interposed between the yolk-sac or end-gut on the one hand and the visceral mesoblast on the other (Figs. 521 and 540). In the region of the mid-gut the meshes of this network are already provided with distinct walls (dark red in Fig. 539), but in the end-gut region there are no meshes whatever, but only a simple vascular layer (Fig. 540), which has separated by delamination from the visceral mesoblast and has not yet developed separate vessels (pale red in Fig. 539) ; the aorta becomes lost in this. From this vascular layer, accordingly, the rete periintestinale and the aorta develop. The umbilicalis stands in connection with this vascular network and determines its early formation; a connection between the a. umbilicalis and the aorta does not yet exist on the left side, but is present on the right.

Fig. 540. — Transverse section of an embryo of 1.38 mm. greatest length and with 5-6 pairs of primitive segments, at the level of the cloaca. (Embryo Pfannenstiel-Kromer, from the collection of Professor Pfannenstiel; slide 7, row 3, section 3.) 120. In consequence of a shrinkage the entoderm has separated from the mesoderm, and the formation of vessels, which takes place from the entire visceral mesoblast, is distinctly seen. It occurs by the separation of an entire cell layer from the visceral mesoblast, with which it remains in connection by cell prolongations or cell groups. A distinction between rete peri-intestinale and aorta caunot be made; it is rather a question of the formation of an intestinal blood-sinus, such as we know from the invertebrates (Lang, 1903).

Fig. 541 shows a reconstruction of the vessels of an embryo with 12-14 pairs of primitive segments. The aorta is fully formed up to the region of the end-gut ; a vascular layer is spread out over the entire mid-gut (compare also Figs. 528 and 530) and is bounded dorsally by a sharp line ; it is represented in Fig. 541 by the pale pink tint. In this vascular layer two independent networks have developed, a cranial and a caudal one. The cranial network is in connection with the v. omphalo-mesenterica and need not here concern us further. The caudal one shows an indistinct cranial and a well-developed caudal region. The cause of the different development of the two regions lies in the union with the a. umbilicalis, which, making use of this network wanders gradually caudalwards and only brings to complete development and preserves for a time those portions of the network which directly place it in connection with the aorta. Between the vascular layer and the ventral periphery of the aorta there are stretched partly hollow (dark red in Fig. 541), partly solid (pale red) connecting vessels, the rami intestinales (Fig. 529, left side; Fig. 541). In the entire embryo there are developed 15 such arteries on the left side and 14 on the right side ; on the left they lie in the regions of the 6th to the future 17th segment, that is to say in 12 segments, on the right in the regions of the 6th to the future 18th segment, that is to say in 13 segments ; a comparison of these figures shows the non-metameric arrangement of the arteries. The rami intestinales are almost equally developed on both sides of the embryo, but they are not arranged in a perfectly paired manner ; their difference in number depends on the variability shown by all organs in process of degeneration. The solid rami intestinales lie in the zone in front of the pronephros, in the 6th to the 9th primitive segments; the hollow ones lie in the pronephric zone, in the 9th to the 14th segments. Where the rami must effect the connection between the aorta and the a. umbilicalis they are especially large and at this point a longitudinal commissure between the rami begins to develop.

Fig. 541. — Reconstruction of the trunk arteries of a human embryo of 2.6 mm. greatest length and with 13-14 pairs of primary segments. (Embryo Pfannenstiel III, from the collection of Professor Pfannenstiel.) The rete peri-intestinale is developed over the entire mid-gut and end-gut and is connected with the dorsal aorta by rami intestinales. The rete peri-intestinale and the rami intestinales are strongly developed at the point where the A. umbilicalis arises from them.

Fig. 542.— Reconstruction of the trunk arteries of a human embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments. (Embryo R. Meyer 300, from the collection of Professor R. Meyer, Berlin . The dorsal aorta is completely developed and from it there arise the viscero-ventral and the dorsal systems of arches. Between the arches of each of the two systems a longitudinal commissure is developed; the A. umbilicalis arises from the commissural vessel of the ventral arch system.

In Fig. 542 the development of the vessels shows a great progress. The aorta is developed as far as the tip of the tail and shows a commencing enlargement in the region of the pronephros (10th to 13th segments). In addition to the viscero-ventral arch system the dorsal one is also developed and from this the parietoventral one is beginning to form. With the formation of the yolk stalk the rete peri-intestinale is carried far from the embryo and is no longer shown in the figure. The number of rami intestinales present is 29, and they show no metamerism. The rami in the neighborhood of the 10th to the 12th segments are the best developed, and next those that connect the aorta and the umbilicalis, the latter having in the meantime wandered further caudally. In the 10th to the 12th segments is the pronephros (compare Figs. 531 and 542) and a relation between it and the development of the rami intestinales is therefore possible. Finally, it may be noted, that in this series a distinct longitudinal commissure is formed, extending as a continuous vessel from the tip of the tail to the 19th ramus intestinalis and as a discontinuous one to the 12th ramus. In Fig. 543 I have reproduced a section that has cut a portion of this longitudinal commissure lengthwise; the vessel is naturally developed on both sides. Just as in the reptiles, birds and other mammals, the paired rami intestinales become unpaired, as do also the paired commissural vessels. The development of the intestinal arteries from the unpaired rami intestinales and the unpaired commissural vessel does not fall within the scope of this chapter.

The description of the pronephric vessels may be summed up by stating that in the human embryo also a viscero-ventral arch system is developed, which separates into two groups, a cranial group of aortic arches and a caudal group of originally pronephric arteries. Just as in Amphioxus, the Ganoids and the Teleosts, a longitudinal commissural vessel develops between the arches of the caudal group, the relation of this vessel to the pronephros no longer appearing in the development; it is possible, however, that the vessel of the external glomerulus may be derived from it. Of the veins in the neighborhood of the pronephros, the v. cardinalis posterior, in the act of developing, is shown in Fig. 532 a-d; it arises in loco from elements which wander out from the middle plate of the mesoderm. A relation between the venous system and the pronephros cannot be made out in the limited material.

Fig. 543. — Transverse section of a human embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments. (Embryo R. Meyer 300, from the collection of Professor R. Meyer, Berlin; 6lide 14, row 2, section 6.) 120. The section passes through the tail bend and cuts, above, the 16th and 17th primitive segments, below, the primitive segment plate. The aorta is cut twice and between the two aortse are sections of four rami intestinales, which are united by the longitudinal commissural vessel. Above is the nephrogenic tissue and the free terminal portion of the primary excretory duct, below and on the left is the end of the primary excretory duct at the ectoderm.

Degeneration of the Pronephros

The pronephros undergoes complete degeneration, but for a time remains of the pronephric tubules and of the collecting duct may be preserved. They are to be found in the retro-peritoneum as completely closed vesicles of a spherical or cylindrical form. Their independence of the mesonephros and the ccelomic epithelium permits their displacement either dorsally behind the aorta and the v. cardinalis posterior or towards the ccelom ; in the latter case they may be situated in fungiform folds projecting into the coelom. Similar displacements can also take place in the degenerating tubules of the cranial portion of the mesonephros, so that they constitute no specific characteristic of the pronephric remains.

The Urogenital Fold.

The mesonephros is formed along the posterior wall of the body cavity. Only at the very beginning of its development does it find sufficient space in the retroperitoneum ; as soon as it begins to expand it needs more room and this it finds in the direction of the body cavity, invaginating the coelom wall as a fold into the cavity. This fold later contains also the Mullerian duct and the reproductive gland, in addition to the mesonephros, and is therefore termed the urogenital fold. It constitutes a region within which a series of very important processes occur, and its origin and fate may, therefore, properly receive attention at this point. At first a posterior body wall does not really exist. The parietal and visceral mesoblast of an embryo of 2.6 mm. greatest length do not meet to form a surface but come together in an angle (Fig. 544) ; to this angle the primitive stalk is attached. Then the two layers of the lateral plate form folds one after the other, first the visceral mesoblast along the dorsal boundary of the entoderm, towards the chorda, and then the parietal mesoblast towards the ectoderm ; at the same time the body cavity, hitherto quite narrow, enlarges. The result of these two processes, a posterior body wall, is shown by an embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments (Fig. 546) ; the point of origin of the primitive segment stalk marks as before the original angle of the body cavity. This posterior body wall throughout its entire breadth and almost throughout its entire length is invaginated into the body cavity by the developing mesonephros and thus forms a portion of the urogenital fold (Fig. 547) ; of the contents of the fold the figure shows only the primary excretory duct (black) and the Malpighian corpuscles (white circles). Both are landmarks which afford the possibility for a detailed description of the further processes. The summit of the urogenital fold lies immediately below the Malpighian corpuscle and the excretory duct is not far from its lateral surface. The first changes in the form of the fold are brought about, first, by the entrance into it of the v. cardinalis posterior and, secondly, by the extension of the body cavity between the lateral body wall and the base of the fold. By the entrance of the vein, which at this time is still very large, the fold increases in size and appears almost quadrangular in section (Fig. 548). Two of the angles are formed by its base, a third represents the original summit, just beneath the Malpighian corpuscles, and the fourth is a newly-formed summit lateral to the primary excretory duct. A comparison of Figs. 547 and 548 shows clearly the extension of the ccelom on the lateral surface of the fold, an extension that leads to the narrowing and modification of its base. The fold, accordingly, is partly invaginated into the body cavity and partly carved out of the dorsal body wall. The position of the primary excretory duct and that of the Malpighian corpuscles with reference to each other and to the surface have not undergone any important modifications.

Fig. 544. — Transverse section of a human embryo of 2.6 mm. greatest length and with 13-14 pairs of primitive segments, at the level of the 13th segment. (Embryo Pfannenstiel III, from the collection of Professor Pfannenstiel; slide 11, row 3, section 5.) X 50. The parietal and visceral mesoblasts of the lateral plate meet to form an acute angle; there is only a dorsal body angle and no body wall. The aorta on either side with a ramus intestinalis.

Fig. 545. — Transverse section of a human embryo of 2.6 mm. greatest length and with 13-14 pairs of primitive segments, at the level of the 9th segment. (Embryo Pfannensteil III, from the collection of Professor Pfannenstiel; slide 9, row 2, section 5.) X 50. Near the dorsal body cavity angle the visceral mesoblast is forming a fold towards the dorsal surface of the intestine. The situation of the dorsal body cavity angle is marked by the attachment of the primitive segment stalk. The aorta on either side with a ramus intestinalis.

Fig. 546. — Transverse section of a human embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments, at the level of the 12th segment. (Embryo R. Meyer 300, from the collection of Professor R. Meyer, Berlin; slide 12, row 1, section 4.) X 50. A fold similar to that of the visceral mesoblast (Fig. 545) has also been formed by the parietal mesoblast. The dorsal body wall, middle plate, is formed by the two folds. The former angle is again marked by the attachment of the primitive segment stalk. A strong ramus intestinalis from the aorta on the left side of the figure (right side of the embryo).

The formation of the urogenital fold begins in the fourth cervical segment ; it is limited at first to the cranial portion of the body cavity and later extends gradually and continuously towards the caudal end of the body cavity to about the 4th lumbar segment. A stage of maximum extent from the 4th cervical to the 4th lumbar segment does not occur, since a degeneration at the cranial end accompanies the caudally directed growth ; actual figures showing the caudal growth and the cranial degeneration are given in the following table:

Size of embryo.

Number of trunk segments.

Extent of the urogenital fold.

2.6 2.5 4.25 4.9 5.3 7 9.5 10 11 13 14.75 17 18 19.4 21 22.5 26 28 29 30 35 50

mm. greatest length ct a it " vertex-breech . . " nape-breech . . . " greatest length it a

10-11 (13-14) 1 20 (23) 25 (28) 30-32 (33-35) 33 (36) 36 (39)

head-foot length

4th cervical 5th 6th 6th 4th 8th 2nd thoracic 3rd 3rd 8th 9th 10th 9th 11th 1st 12th 1st 1st 1st 3rd 3rd 4th

lumbar thoracic lumbar

to 1st thoracic " 8th " 2nd lumbar " 3rd " 2nd " 3rd " 4th " 4th " 4th " 4th " 4th " 4th " 4th " 4th " 4th 4th 4th 5th 5th 5th 5th 5th

segment

a tl

it it

tl It

1 The numbers in parentheses are the numbers for the primitive segments.

From this table it may be seen: 1. That even in an embryo of 9.5 mm. the caudal growth of the urogenital fold is completed ; 2. that the degeneration of its cranial portion is complete in embryos of 2.6 mm. greatest length ; in these the fold is limited to the lumbar region. In older embryos the descent of the cranial limit shows a still greater progress, but this is no longer a result of the abbreviation of the fold, but is due to its rate of growth lagging behind that of the vertebra?.

In the course of its development the urogenital fold undergoes a series of important changes. In the first place, it becomes divided throughout its whole length, with the exception of the cranial and caudal ends, into a genital and a mesonephric fold (Figs. 550 and 552). The anlage of the reproductive gland naturally precedes this division, and although the gland is not represented in Fig. 550, its presence is nevertheless indicated by the displacement of the Malpighian corpuscle (compare Figs. 548550) ; the space that is formed between this and the surface of the fold is occupied by the genital anlage. The reproductive gland does not, consequently, grow out into the body cavity, but into the urogenital fold. Its growth is made possible only by the displacement of the mesonephric tubules and this displacement again only by the diminishment of the v. cardinalis posterior. Once the reproductive gland is formed it becomes surrounded by a fosse, deep grooves cutting into it laterally and medially. Fig. 549 shows the lateral groove at its first formation ; in Fig. 550 both the lateral and the medial grooves are already considerably developed. During the division of the urogenital fold it becomes still more cut off from the dorsal body wall on its lateral and, as Fig. 550 shows, also on its medial side. Thus the urogenital fold, which at its first formation has a broad base, becomes stalked (Fig. 548) and, at the same time, the stalk becomes twisted; the fold, therefore, no longer lies sagittally (Figs. 547 and 548), but frontally. The formation of a fosse around the genital fold begins somewhat below the cranial pole of the reproductive gland and thence proceeds in a caudal direction. Fig. 552 shows the model of a dividing urogenital fold; the division is here completed to both the cranial and caudal poles.

Fig. 547. — Transverse section of the middle plate of a human embryo of 4.9 mm. nape length, at the level of the 17th primitive segment. (Embryo 139, G. 31, from the collection of the 2nd Anatominal Institute, Berlin [Professor O. Hertwig]. i 50. The middle plate is evaginated into the body cavity by the development of the mesonephros; it forms the urogenital fold. The fold separates a medial and a lateral bay of the body cavity. In this and the succeeding figures the excretory ducts are black and the Malpighian corpuscles white. These latter lie just below the summit of the urogenital fold, the excretory duct at the ateral surface of its base.

Fig. 548. — Transverse section of the urogenital fold of a human embryo of 7 mm. greatest length at a level between the 13th and 14th trunk segments. (Embryo Chr. l.from the collection of Professor Hochstetter, Vienna; slide 8, row 6, section 7.) 50. By the ingrowth of the v. cardinalis posterior the urogenital fold is enlarged at the expense of the retroperitoneum, the lateral bay of the body cavity (Fig. 547) penetrates deeply into the latter, and the modification of the base of the fold is thus begun. Compare the depths of the lateral bay of the body cavity in Figs. 547 and 54S. The Malpighian corpuscle determines, as in the preceding figure, the position of the summit of the fold. A new summit has formed near the excretory duct.

The two grooves that form the fossae around the reproductive gland anlage never meet, a portion of the urogenital fold persisting between them and giving rise to the stalk of the genital gland and to the mesogenitale (mesovarium, mesorchium) (Fig. 551). The breadth of the stalk varies ; at the cranial pole it is considerable, immediately below this it diminishes so much that the mesogenitale becomes thread-like, and it remains in this condition as far as the caudal pole of the reproductive gland, where it again increases in size so long as the gland is not completely developed in this region. Fig. 553 may serve to show the size of the urogenital fold as compared with the transverse section of the entire embryo and also to make clear its topographic relations ; the fold is here divided into the mesonephric and genital folds, both of which seem very small as compared with the anlagen of the suprarenal body, the enormously developed liver and the large stomach. In the second place the mesonephric fold, separated from the urogenital fold, becomes further subdivided in correspondence with

Fig. 549. — Transverse section of a human embryo of 12.5 mm. greatest length, at a level between the 18th and 19th trunk segments. (Embryo Ma. 1, from the collection of Professor Hochstetter, Vienna; slide 8, row 4, section 5.) X 50. The urogenital fold has completely altered its position; its base is sagittal and the tip in the neighborhood of the excretory duct has become the summit of the fold, so that a dorsal and ventral slope must be recognized. The lateral bay of the body cavity has penetrated to the medial surface of the the V. cardinalis post. Throughout almost the whole region of the ventral slope the reproductive gland has developed and, by its growth dorsally, the Malpighian corpuscle has been forced in that direction. A groove between the excretory duct and the lateral border of the reproductive gland begins to separate the reproductive gland area from the urogenital fold; consequently the summit of the fold containing the excretory duct projects in a tongue-like manner.

Fig. 550. — Transverse section of the urogenital fold of an embryo of 19.4 mm. vertex-breech length, at a level between the 21st and 22nd spinal ganglia. (Embryo Ma. 2, from the collection of Professor Hochstetter, Vienna; slide 59, row 1, section 3.) 50. The medial bay of the body cavity now also penetrates into the retroperitoneum and narrows the base of the urogenital fold to a stalk. Two fossae separate the area of the reproductive gland from the urogenital fold and bring about a division of the latter into the mesonephric and genital folds. The point where these are still connected is the first anlage of the mesogenitale. (Stiel = stalk.) its contents. Proceeding medially from the lateral surface one meets in it (Fig. 551) : the Mullerian duct, the excretory duct and the convolutions of the mesonephric tubules. The fold forms first a portion for the Mullerian duct, the tubar portion, then a common portion for the excretory duct and the mesonephric tubules, the gland portion, and, finally, the thread-like connection with the posterior abdominal wall, the mesentery portion. The tubar and gland portions regularly become separated by a slight furrow that projects in from the lateral surface; the gland portion passes over gradually into the mesentery portion. In the contents of these three portions a sexual difference becomes evident to the extent that while in female embryos the tubar portion contains only the Mullerian duct, in males it contains both the Mullerian and the

The formation of the various portions begins at the cranial pole and proceeds caudally, and the lateral portions are formed earlier than the medial ones. Consequently one sees at first only

Gland portion

Mesentery portion

Tubar portion

Mesogenitale

Fig. 551. — Transverse section of the right urogenital fold of a human embryo of 5 cm. head-foot length. (Embryo R. Meyer 272, from the collection of Professor R. Meyer, Berlin; slide 2, row 1, section 2.) The mesonephric fold is incompletely divided into three portions by a groove in its dorso-lateral surface and by the narrowing of its base; the tube portion with the transverse section of the Miillerian duct, the gland portion with the transverse section of the excretory duct and the mesonephric tubules, and the mesentery portion. The mesonephric fold is beginning to grow around the reproductive gland portion.

the tubar portion and this' only in the cranial region of the fold. The separation of the' tubar portion begins in embryos of 13 mm. greatest length. It is the most freely movable portion of the mesonephric fold and this mobility serves to explain the peculiar wandering of the ostium abdominale which occurs later on.

A third change is in the course of the urogenital folds. Originally both folds lie parallel to the vertebral column, as may be seen from the parallel course of both mesonephroi in an embryo of 9 mm. greatest length (Fig. 565) ; but as soon as new organs appear between them in the middle line, they become displaced. Already the influence of the suprarenal bodies may be seen in an embryo of 9.5 mm. greatest length, the two folds, which originally lay close together, becoming forced apart. What is begun by the suprarenal bodies is continued by the metanephroi ; they displace the two mesonephric folds still more laterally. We have seen above that at first the whole posterior body wall was invaginated into the body cavity as the urogenital folds. The folds can only be displaced laterally by a broadening of the posterior body wall and an increase in the frontal diameter of the body cavity. The increase occurs apparently in the middle line, since the topographic relations between the urogenital fold and the lateral body wall

Pericardium

Right lung

Left lung

Diaphragm

Vena cava inf.

Diaphragmatic band of mesonephros Superior recess of omental sack Cranial sagittal portion of the mesonephric fold Root of mesentery Horizontal portion of the mesonephric fold Caudal sagittal portion of the mesonephric fold

Fig. 552. — Model of the posterior abdominal wall of a human embryo of 19.4 mm. greatest length. (Embryo Ma. 2, from the collection of Professor Hochstetter, Berlin. The model was prepared by my students Massard and Chome.) The visceral layer of peritoneum is shown cut and one sees the superior recess of the omental sack. The urogenital fold is divided as far as its upper and lower ends into the mesonephric and genital folds. The mesonephric fold is bayonet-shaped and an upper sagittal, a horizontal and a lower sagittal portion, and a first and second bend may be distinguished. At the first bend the mesonephric fold is connected with the anterior abdominal wall by the inguinal fold. Between the aa. umbilicales, which still run horizontally, is the bladder plate.

do not alter, as is shown in Figs. 547-551. Caudal to the metanephros the enlarging force ceases, and in this region the frontal diameter does enlarge and the urogenital fold is not displaced. The occurrence of a displacement above and its absence below necessarily produces a bend of the fold. This is a double bend, and the fold acquires by it a bayonet form; there may be distinguished an upper sagittal, a middle transverse and a lower sagittal portion (Fig. 552). The more the suprarenal bodies and the metanephroi grow later on, the more they cause the upper sagittal portion of the fold to assume an oblique position. The mesentery

Fig. 553. — Transverse section of a human embryo of 19.4 mm. greatest length, at the level of the 21st spinal ganglion. (Embryo Ma. 2; slide 5S, row 2, section 1. From the collection of Professor Hochstetter, Vienna.) The section is intended to show the topographic relations of the abdominal viscera. The stomach has begun to twist to the left and, with the spleen, is in contact with the left urogenital fold. The two urogenital folds are still symmetrically placed; they are divided into the mesonephric and the genital folds. Between the two urogenital folds are the adrenal bodies; behind these lie the metanephroi, surrounded by a very loose tissue.

of the genital fold is fully formed only in the region of the upper sagittal portion and this alone obtains from it a great mobility and consequently alone can yield to the pressure of the neighboring organs. The liver, which grows but does not undergo displacement, does not exert the slightest influence on the urogenital fold, but, on the other hand, both the stomach and intestine alter their position and both must therefore influence the position of the urogenital fold. The stomach, and with it the spleen, becomes displaced to the left side and is rotated; the large coil which becomes the large intestine tends for the most part to the left, and, consequently, both it and the stomach will affect the left urogenital fold, the descending colon coming to lie on its medial side, while the ascending colon runs from the beginning on the lateral side of the right fold. The left urogenital fold yields to the pressure of the stomach and intestine and is displaced by it further in the same direction as it was forced by the suprarenal body and metanephros. This displacement may be carried to such an extent that the upper sagittal portion of the left fold comes to lie horizontally and, therefore, can no longer be distinguished from the horizontal portion. This displacement I observed first in an embryo of 29 mm. greatest length ; from that stage onward it was usually to be made out in embryos of both sexes.

The influence of the intestinal tract in displacement may, of course, be stronger and may therefore produce anomalies of development. It will be seen in the following paragraph that the lower sagittal portions of the two urogenital folds unite throughout their whole length in the median line; the extent of the union may be lessened or quite prevented by an abnormal influence on the part of the intestine. This gives a possibility for the explanation of double formations of the uterus. In the same way the pressure of the stomach and intestine may give the reproductive gland a position in which it cannot effect the connections made in normal development; I refer, principally, to the connection of the testis with the inguinal fold, whose failure results in a failure of the testis to enter the inguinal canal.

A fourth change affects the caudal sagittal portion of the urogenital fold. In this region the two folds come together and fuse, and by this fusion a frontal partition is formed, which divides the primitive pelvis into a ventral and a dorsal half; this frontal partition is termed the genital cord (tr actus genitalis) ; it fuses with the floor of the body cavity and the partition is thus a complete one. In Fig. 554 a, b and c three sections through the lower portions of both urogenital folds are shown ; the mesonephric and reproductive folds are separated, merely a thread-like mesogenitale uniting them; the tubar portion and the lateral part of the gland portion are thrown into an angle directed towards the medial part of the gland portion and bend around the ventral surface of the reproductive gland; they are, however, still far from the median line. In section b the mesonephric folds are hearer together and in c their union has occurred. The formation of the genital cord takes place in embryos between 19.4 and 21 mm. greatest length ; the upper edge of the genital cord lies at the level of the lower border of the third lumbar vertebra (embryos of 36 mm. greatest length), but later a passive displacement to the lower border of the fourth or even the fifth lumbar vertebra mav occur.

Fig. 554 a, b, and c. — Three transverse sections through a human embryo of 30 mm. trunk length. (Embryo R. Meyer 273, from the collection of Professor Meyer, Berlin; slide 11, row 4, section 2; slide 13, row 5, section 2; slide 14, row 5, section 2.) The three sections show the formation of the genital cord. The legends have been arranged so that different structures are indicated in each section and the reader should first study the legends in all three figures, o. The mesonephric fold bends between the tubar and gland portions at a right angle and is growing around the reproductive gland, so that the left and right mesonephros come to lie in the same frontal plane. The excretory duct and the tube are shown in the mesonephric fold. The originally lateral tube comes to lie medially, as a result of the bending of the fold.

b. An evagination (the inguinal fold) extends from the mesonephric fold between the bladder and the lateral wall of the body to unite with the posterior surface of the anterior abdominal wall (in Fig. c it is cut throughout its entire length). The right and left mesonephric folds have come nearer together.

The genital cord seems to be formed at once throughout its entire length, since the upper boundary is found at the same place in all embryos of this period.

Sexual differences appear with the formation of the genital cord. Fig. 554 c is taken from a female embryo and in it one sees a distinct excavatio vesico-uterina between the genital cord and the bladder; in male embryos the mesonephric folds, approaching one another in the median line, at once unite with the wall of the bladder and they, therefore, do not possess a vesicouterine pouch (Fig. 555).

c. The two mesonephric folds have united in the middle line to form the genital cord. By this and the two inguinal folds the body cavity is divided into four parts, — the excavatio recto-uterina between the posterior body wall and the genital cord; a space between the genital cord, the posterior wall of the bladder and the inguinal folds, from which the excavat,io vesico-uterina and the medial part of the saccus vaginalis (see section on descensus) are formed; and two lateral spaces between the plica inguinalis and the lateral abdominal wall, which become the lateral part of the saccus vaginalis.

wall. Already in an embryo of 13 mm. greatest length a knobshaped growth, the inguinal fold, forms at the first bend (Fig. 552). In sections it resembles the anlage of the reproductive gland, which lies on the medial side of the urogenital fold, except that it is a true growth and is not, like the reproductive gland, grooved out of the fold. The growing inguinal fold very soon reaches the lateral abdominal wall in its dorsal half and then unites with a ridge of the wall which I shall call the inguinal crest (Fig. 556 a and b). By this union the horizontal portion of the urogenital fold becomes dependent from the lateral abdominal wall; as a result changes in the position of the urogenital folds must ensue.

The anterior abdominal wall is now almost horizontal ; immediately at the oral slope of the genital tubercle the ectoderm of the surface of the body passes directly over into the wall of the umbilical cord. In Fig. 557 a median section of an embryo of 24 mm. vertex-breech length is shown; the outer contour of the body extends from the tip of the tail over the anal and urogenital orifices to the phallus, and at the cranial surface of this it bends around and passes over into the caudal surface of the umbilical cord. Later, on the closure of the wide umbilicus, the anterior abdominal wall is bent upward and becomes horizontal. That is one of the changes. The other is best seen by comparing the position of the m. rectus abdominis in embryos of different ages; as soon as muscle fibres are to be made out, the muscle lies in the middle of the lateral wall of the body, so that to acquire its definitive position it must be pushed along the abdominal wall to the median plane. It may, in brief, be said that the anterior abdominal wall is pushed in two directions, on the one hand cranially and on the other medially. The pushing takes place during the division of the urogenital fold into the reproductive and mesonephric folds and during the development of the different parts of the latter. The inguinal fold thereby arises finally from the tubar portion of the mesonephric fold (Fig. 554 b and c). The wanderings of the anterior abdominal wall will affect the tubar portion and the adjacent part of the gLand portion, and we may theoretically assume a wandering of the tubar portion, such as actually takes place in the formation of the genital cord. If the formation of the anterior abdominal wall is disturbed or delayed an effect upon the formation of the genital cord is possible, and herein is a second possibility for double formations of the uterus.

Fig. 555. — Median section of a human embryo of 50 mm. head-foot length. (Embryo R. Meyer 257, from the collection of Professor R. Meyer, Berlin; slide 16, row 3, section 1.) X 8. On the development of the genital cord the first indication of a sexual difference appears. In the male the two mesonephric folds fuse with the posterior wall of the bladder before they unite to form the genital cord, and this fusion persists also after the union. The urogenital sinus is cut lengthwise; the anterior border of its external orifice reaches already the coronary sulcus of the glans. Between the external orifice of tlfc sinus and the anus the aboral periphery of the phallus and the perineum are beginning to form.

Fig. 556 a and b. — Transverse section through the urogenital fold at the level of the inguinal fold in a human embryo of 22.5 mm. greatest length. (Embryo R. Meyer 303, from the collection of Professor R. Meyer, Berlin; slide 34, row 2, slide 1 and row 4, section 1.) a. Passes through the inguinal crest and the intermuscular part of the chorda gubernaculi. Bl., bladder, b. Shows the union of the inguinal fold with the inguinal crest, by which the urogenital fold becomes united with the lateral abdominal wall.

Fig. 557. — Median section of the lower half of the body of an embryo of 24 mm. vertex-breech length. (Embryo Hal. I, from the collection of the I Anatomical Institute, Vienna [Professor Zuckerkandl]; slide 44, row 1, section 2.) The urethral plate is reconstructed from four sections. An anterior body wall does not yet exist; the umbilicus begins immediately at the oral slope of the phallus.

The Mesonephros

GENERAL.

The mesonephric tubules develop after the pronephric. They are formed from the same parent tissue, the stalks of the primitive segments, in almost the same segments, and they make use of the same efferent canal, the primary excretory duct. Comparative embryology teaches that the mesonephric tubule, like the pronephric one, is a composite structure. A mesonephric segment consists of a principal and a supplemental tubule, the former being an evagination of the parietal layer of the primitive segment stalk, while the latter is the lateral portion of the segment stalk itself. From the supplemental tubule there are formed the mesonephric chamber, in this case termed Bowman's capsule, and the nephrostome canal, the distinction between the two resulting from the enlargement of the medial portion of the supplemental canal to form Bowman's capsule. This is invaginated by a vascular glomerulus, whose efferent stem arises directly from the aorta. A comparison of this account of the development of the mesonephric tubule with that given on p. 759 for the pronephric tubule will show a complete correspondence. The development of the mesonephric tubule might have been illustrated by printing for a second time the plan of the pronephric segment, modification being necessary only in two particulars, the formation of the collecting duct and the external glomerulus. If then the correspondence between the two sets of tubules is so close, how are they distinguished? In three ways : 1, by their appearance before or after the excretory duct ; 2, by the presence or absence of an external glomerulus, and 3, by the point of origin of the principal tubule from the stalk of the primitive segment.

Let us first consider the last point. The principal tubules of the mesonephros arise more medially than do those of the pronephros. This difference does not, however, suffice in all cases for a differential diagnosis of the two kinds of tubules. If a pronephric and a mesonephric tubule are formed in the same segment and if, furthermore, the anlage of the mesonephric tubule appears while the pronephric tubule is still present, then it is possible to distinguish the two by their position, the pronephric tubule lying lateral, nearer to the lateral plate, the mesonephric one medial, nearer to the primitive segment. But whoever has studied the development of the two provisional excretory organs in the vertebrate series must admit that these two conditions are very rarely fulfilled. If in a given segment only one excretory tubule is formed, or if the pronephric tubule is completely degenerated before the mesonephric one appears, then the position of the principal tubule under consideration, with reference to the primitive segment stalk can give no conclusive evidence as to its character, and this for the following reasons : The supplemental tubule, which, according to the account given above, is the same structure in both the pronephric and mesonephric tubules, may with advancing development be partly taken up into the lateral plate or into the body cavity derived from it. If this shortening of the supplemental tubule occurs before the mesonephric tubule is formed, the origin of the principal tubule of the latter may apparently lie at the very spot where that of the pronephric tubule should occur, namely quite laterally and close to the wall of the body cavity, in some cases, indeed, at the wall. As to the second distinction between the two tubules the following may be said : The occurrence of external glomeruli is distinctive for the pronephros; their absence, however, is no evidence that a tubule must be a mesonephric one, since the glomerulus may also be wanting to a pronephric tubule. I have expressly stated above that the external glomerulus is characteristic of the pronephros and not of a pronephric tubule ; it is quite independent of its tubule and is merely invaginated into the body cavity beside it. This independence of the external glomerulus is shown by the fact that it degenerates independently of the other constituents of the pronephric segment; it always degenerates, but frequently only when the pronephric tubule has already vanished and the mesonephric tubule has developed. An external glomerulus may therefore occur beside a mesonephric tubule as well as beside a pronephric one, and its presence is no decisive evidence as to the character of the tubule occurring in the same segment with it. It tells that in a given animal a pronephros was present, but it does not say that the tubule occurring beside it is a pronephric tubule.

There thus remains as a distinctive indication of whether a tubule is mesonephric or pronephric only the first point, its relation to the excretory duct. Whatever tubules are developed before or simultaneously with the excretory duct are pronephric, those formed after it are mesonephric, This difference can only be determined embryologically, it is of no avail in the case of developed tubules. It is not possible, therefore, on the first study of a fully formed organ to determine whether it is pronephric or mesonephric ; but if by earlier studies the limits of the two organs have been determined and we know how far caudally the pronephros and how far cranially the mesonephros extends, then, without any further evidence, we may speak of pronephric or mesonephric tubules in the case of segments in which only one kind occurs; in all other segments the determination must rest upon a study of the development.

In those vertebrates in which the pronephros extends the entire length of the body cavity, it is at once evident that the pronephric tubules form a first and the mesonephric a second generation of excretory tubules. The mesonephric tubules are, accordingly, something new, distinguishable from the pronephric tubules; it will occur to no one to regard the fruit of one year as of the same generation as that of a preceding year. In the vertebrates in which the pronephros is shortened, the matter is not so clear; indeed, it may be so obscure that one is led so far astray as to regard the pronephric and mesonephric tubules as equivalent parts of a single system. All investigations that have led to this estimate of the mesonephric tubules have been made on vertebrates with a shortened pronephros. Whoever examines the development of the pronephros and mesonephros throughout the whole series of vertebrates cannot for a moment doubt that the pronephric and mesonephric tubules are members of two different systems.

So far as man, in particular, is concerned, the conditions are more favorable for a distinction between the pronephric and mesonephric tubules. Man has such a high grade of organization that the excretory organ has a long developmental path to traverse before it reaches its completion. This path is shortened by the pronephros developing only rudimentarily. So far as is yet known, no human pronephric tubule develops an internal glomerulus, and, so far as known, every mesonephric tubule, even the most cranial ones that at once undergo degeneration, develops one. In the presence or absence of an internal glomerulus in the Malpighian corpuscles of the human mesonephros we have, then, a means, important because it is so readily made out, of determining the mesonephric character of a tubule.

THE PARENT TISSUE OF THE MESONEPHRIC TUBULES.

The parent tissue of the mesonephric tubules is furnished by the primitive segment stalks. In describing the development of the pronephros the process of segmentation of the mesoderm has been considered. It presents two phases. The first phase, that of transverse division of the mesoderm, brings about the formation of the primary primitive segments, the second, that of the sagittal division, divides the primary segments into the secondary segments and the segment stalks (Fig. 520 a, b and c). In man the process of segmentation does not take place quite according to rule. In an embryo with 13-14 primitive segments all three mesodermal

DEVELOPMENT OF THE URINOGENITAL ORGANS. 799 structures, the secondary segment, the segment stalk and the lateral plate, are developed in the first ten segments, but from the eleventh on, while the longitudinal divisions occur in the old direction, the transverse ones no longer occur lateral to the region of the secondary primitive segment, so that while there are welldefined secondary segments, there are no distinct primitive segment stalks. The segment stalks, cut out as a whole from the mesoderm, form what is termed the nephrogenic cord. In an embryo with 23 pairs of primitive segments (Fig. 531 a) this cord not only extends throughout the entire segmental region, but also through a portion of the unsegmental one to the region of the cloaca; in other words, the segmentation process in the posterior portion of the embryo passes through two phases, but they are in inverted order. The sagittal divisions, which originally formed the second phase, occur first and separate, at first incompletely, the lateral plate, nephrogenic cord and primitive segment plate; the transverse divisions, originally the first, now succeed as the second phase, but they only divide the primitive segment plate into the individual secondary segments, the nephrogenic cord remaining undivided. In an embryo of 4.25 mm. vertex-breech length and with 28 pairs of primitive segments the nephrogenic cord extends to the 28th segment, in any case as far as the point where the primary excretory duct applies itself to the wall of the cloaca ; here the cord appears somewhat thickened, its caudal end being lost in the as yet undivided mesoderm of the tail. In no embryo is the nephrogenic cord present in its full extent, for while it is growing caudally it is dividing at its cranial end into the anlagen for the individual mesonephric tubules.

The primary excretory duct in the pronephric territory lay lateral to the primitive segment stalk. It retains this position in its growth caudally and accordingly lies lateral to the nephrogenic cord (Figs. 558 a, 559 and 562).

The nephrogenic cord in an embryo of 5.3 mm. greatest length and 4.6 mm. nape length is interrupted in the 26th or 27th segment (3rd or 4th lumbar segment). The compact cord, which is rather sharply marked off from the surrounding mesenchyme, becomes looser in this region, its cells separate from one another and, finally, are no longer to be distinguished from those surrounding it. Thus the cord becomes divided into two unequal portions, a long cranial one, from which the mesonephric tubules develop, and a short caudal one, from which the metanephric tubules arise. Following Schreiner (1902) the two portions may be termed the mesonephro genie and the metanephrogenic cords. Nothing is to be seen of the division of the nephrogenic cord in an embryo of 4.9 mm. nape-length; in one of 5.3 mm. greatest length it is completed.

FORMATION OF THE MESONEPHRIC TUBULES FROM THE MESONEPHROGENIC CORD.

Immediately after its formation the nephrogenic cord becomes divided, at first only at its cranial end, into a series of spherical masses of cells. In an embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments, anlagen of mesonephric tubules are present in the 13th, 14th and 15th segments, i.e., in the 2nd, 3rd and 4th thoracic segments (Fig. 531 a). In this embryo neither has the nephrogenic cord completed its growth caudalry, nor has the excretory duct reached the cloaca. In an embryo of 4.25 mm. vertex-breech length and with 28 pairs of primitive segments, mesonephric anlagen occur as far back as the 26th segment, i.e., as far as the third lumbar segment, and this caudal limit is not exceeded at first. In this embryo the growth of the nephrogenic cord is completed and the excretory duct has reached the cloaca. The formation of the mesonephric tubules is, accordingly, completed in a rather short space of time ; they are formed almost at a stroke. The mesonephros, however, grows not only caudally, but also at first cranially. In an embryo of 2.5 mm. greatest length the first mesonephric tubule lay in the 2nd thoracic segment, in one of 4.25 mm. vertex-breech length and in one of 4.9 mm. nape length the first tubule was in the 7th cervical segment, and, finally, in one of 5.3 mm. greatest length in the 6th cervical segment; cranial to the 6th cervical segment I have found neither anlagen of mesonephric tubules nor tubules in process of degeneration. Details of the development of the mesonephros may be obtained from the appended table. Its anlage extends over 18 segments, from the 6th cervical to the 3rd lumbar, and in these eighteen segments there are developed in maximo 83 tubules; this number I have obtained by estimating from the table and adding together the maximal number of tubules occurring in each segment. The distribution of the tubules among the various segments is unequal, as will be seen from the table, but 28 occur in the last 4 segments. Since the pronephros extends from the 5th cervical to the 3rd thoracic segment, mesonephric tubules occur throughout almost its whole territory. Why cannot these later cranial mesonephric tubules be pronephric? At first they are not connected with the primary excretory duct and might therefore represent degenerating pronephric tubules, similar to those which actually occur in the 6th segment of the embryo of 4.25 mm. nape length and in the 5th and 6th segments of the embryo of 4.9 mm. nape length. The tubules, however, that I have regarded as mesonephric, are all in statu nascendi, they do not differ in any respect from what are undoubtedly mesonephric tubules in the more posterior segments and they also later acquire a connection with the primary excretory duct.

Fig. 558 a and 6. — Transverse section of a human embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments, at the level of the 13th segment. (Embryo R. Meyer 300, from the collection ot Professor R. Meyer; slide 12, row 4, section 5.) < 317. The section passes through the nephrogenic cord in the interstitium between two anlagen of mesonephric tubules. The primary excretory duct lies on the outer surface of the cord as a semilunar structure.

Fig. 559. — Transverse section of a human embryo of 2.5 mm. greatest length and with 23 pairs of primitive 3egments, at the level of the 13th segment. (Embryo R. Meyer 300, from the collection of Professor R. Meyer, Berlin; slide 12, row 4, section 6.) 317. The section passes through the middle of a mesonephric vesicle; the entire nephrogenic cord is not used in the formation of a mesonephric tubule, but a portion of it, situated laterally or ventrally, persists for a time as a rudimentary nephrostome canal. Lateral from the tubule is the semilunar primary excretory duct and the anlage of the V. cardinalis posterior

The anlagen of the mesonephric tubules are from the first dysmetameric, as many as four occurring in a segment ; details as to the numbers are given in the table facing p. 816. This is not remarkable since the parent tissue has already lost all traces of segmentation.

In the anterior segments, which are somewhat larger than the posterior ones, the anlagen appear as spherical vesicles in which no distinct lumen is yet visible; in the posterior segments

23rd primitive segment

Primary excretory duct Mesonephric tubule 21 Mesonephric tubule 20 Mesonephric tubule 19 Ccelom of lateral plate Mi sonephric tubule 17 Mesonephric tubule 16 Mesonephric tubule 15 Primary excretory duct 18th primitive segment they are more cubical on account of mutual pressure. Figs. 5e8 and 559 represent sections through the mesonephric region of an embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments. Fig. 558 passes through the anterior wall and Fig. 559 through the middle of the third mesonephric vesicle. One sees from both sections that the mesonephric anlage does not exhaust the entire mesonephrogenic cord; a lateral portion remains to form a solid union with the lateral plate and it corresponds to a nephrostome canal. In both figures the primary excretory duct lies lateral to the mesonephrogenic cord. The contour of the vesicle projects beyond that of the mesonephrogenic cord, compresses the excretory duct to a semilunar structure, presses it outwards, thins out the adjacent ectoderm and bulges it out. This prominence of the mesonephric vesicle is the commencing development of the principal tubule.

Fig. 560. — Frontal longitudinal section of a human embryo of 4.25 mm. vertex-breech length and with 28 pairs of primitive segments. (Embryo II M 1. from the collection of the Anatomical Institute, Zurich; 399, J. 13, row 4, section 5.) 150. The section shows the positions of the mesonephric tubules relative to one another and to the body segments. On the left side only mesonephric vesicles are represented and on the right side only primitive segments. Xo cells occur between the various mesonephric vesicles.

The section shown in Fig. 560 passes through the 15th to the 21st mesonephric anlagen of an embryo of 4.25 mm. vertex-breech length and with 28 pairs of primitive segments. On the right side of the figure only primitive segments have been cut, and on the left side only mesonephric vesicles. One sees the mutually flattening, almost quadrangular mesonephric anlagen so closely apposed that there is scarcely room between them for a single cell. A comparison of the right and left sides of the figure shows the dysmetamerism of the anlagen.

The mesonephric vesicles now undergo a series of modifications, which succeed one another in a cranio-caudal direction. I give a diagramatic representation of them in Fig. 561. The sphere or cube (Fig. 561 a) first becomes an olive-shaped vesicle by becoming hollow and by the lateral growth of the principal tubule (Fig. 561 b), the free end of this latter unites with the primary excretory duct (Fig. 561 c), and the original mesonephric vesicle enlarges, its wall at the same time becoming thin, to form a structure resting upon the principal canal just as the transverse bar of the letter T rests upon the upright. The transverse bar has a cranio-caudal direction, which is not shown in Fig. 561 c. In the angle between the upright and transverse bars the glomerulus appears, sometimes lying on the cranial and sometimes on the caudal side of the upright; the part of the transverse bar with which the glomerulus is associated later becomes much more strongly developed than the other part. Once the connection with the primary excretory duct is accomplished and the anlage of the Malpighian corpuscle is laid down, the tubule begins to bend (Fig. 561 d). The result is two loops with a limb in common (Fig. 561 e). The lumen of the tubule appears first at its medial end and thence extends through the entire tubule to its opening into the primary excretory duct. Of the three limbs the most ventral one situated toward the summit of the mesonephric fold (Fig. 562) becomes the Malpighian corpuscle, from the common limb there is formed a portion of the tubule which we shall term the secretorv tubule, and from the third, dorsal limb another portion of the tubule, the collecting tubule (Fig. 561 e). Fig. 562 shows the general topography of a. mesonephric tubule; the section passes through the 16th right (on the left in the figure) and the 18th left mesonephric tubules of an embryo of 4.9 mm. nape length. The mesonephric fold is almost completely filled by the tubule, which shows its S-shaped bends and the commencing transformation of the ventral limb into the Malpighian corpuscle. The primary excretory duct lies quite laterally in the fold and beneath it is the forming v. cardinalis posterior.

As to the details of development the following may be noted. In an embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments there were 8 anlagen of mesonephric tubules, all in the vesicle stage.

In an embryo of 4.25 mm. vertex-breech length and with 28 pairs of primitive segments, of the 28 right and 29 left mesonephric anlagen the two first were in the oliveshaped stage (Fig. 561 b), all the rest were round or flattened vesicles. None of the anlagen are connected with the primary excretory duet. In an embryo of 4.9 mm. nape length and with 33-35 pairs of primitive segments there were 36 tubules, and of these the 1st to the 18th have an S-shape (Fig. 561 e), the 19th20th the form shown in Fig. 561 d, the 21st-33rd are olive-shaped, and the 34th-36th have a round vesicular form; the 35th and the 36th are connected by remains of the nephrogenic tissue and the 36th is not yet separated on its caudal side from the nephrogenic cord. The 1st to the 25th tubules are connected with the primary excretory duct, the 26th-34th have grown toward the duct and are in apposition with it. the 35th and 36th are still free. In the 2nd-12th tubules there is a developed Malpighian corpuscle, in the 13th-lSth tubules the corpuscle is forming.

In an embryo of 5.3 mm. greatest length and with 36 pairs of primitive segments there were 32 tubules, and of these the first 25 are Sshaped, the 26th-28th olive-shaped, the 29th31st are round vesicles and the 32nd is not yet fully differentiated. The lst~20th tubules open into the excretory duct, the 21st-24th are in contact with it, the 25th~32nd are still free. Tubules 1-4 are provided with Malpighian

Fig. 561 a, b, c, d and e. — Diagrams showing the development of the individual mesonephric vesicles, a. Solid vesicle, b. Appearance of the lumen and growth of the principal tubule towards the primary excretory duct. c. The principal tubule reaches Corpuscles.

the excretory duct. d. Beginning of forma- In an embryo of 7 mm. greatest length tion of the Malpighian corpuscle and the n ,1 01 l i. i j.i 1 l nrii „ _ o 1 J secreting and collecting tubules. ..Develop- ° f the 31 tubules the lst-25th were S-shaped, ment completed. the 26th-31st olive-shaped. All 31 tubules were connected with the primary excretory duct. Tubules 1-28 showed Malpighian corpuscles (Fig. 565).

In an embryo of 9.5 mm. greatest length all the tubules were S-shaped, all opened into the primary excretory duct and all possessed Malpighian corpuscles.

With the appearance of the typical S-shape the mesonephric tubule has essentially completed its development; what follows is an elongation of the individual limbs and their histological differentiation. Fig. 563 shows all parts of a mesonephric tubule in the same section. The Malpighian corpuscle has become a large vesicle that is invaginated by the vascular glomerulus only at a limited region. The Bowman's capsule would enclose the glomerulus twice over; the smallness of the glomerulus in contrast with the size of the capsule must be taken as a characteristic of the human mesonephros. In the 22nd tubule only the edge of the vascular glomerulus is cut; in the 23rd tubule it is more distinct. The section through the corpuscle of the 24th tubule goes through a point where there is no invagination. All the vesicles are lined on their parietal wall by a flattened epithelium; the visceral epithelium, which covers the glomerulus, is of a low cubical form. At the transition of the capsule into the secretory tubule the lumen of the Malpighian corpuscle diminishes and the flattened epithelium becomes cubical and finally cylindrical. The most characteristic feature of the secretorv tubule, iu addition to the cvlindrical form of its cells, is their .clear appearance. The cell boundaries become distinct, the oval nucleus passes into the basal half of the cell and the other half becomes eosinophilic. Very frequently the secreting tubule enlarges in a spindle-shaped manner, as is shown in Fig. 563 ; on the other hand, the transitional region into the Malpighian corpuscle and the collecting tubule always remain narrow. The epithelial differentiation is already indicated in an embryo of 5.3 mm. greatest length in the anterior tubules, the spindle-shaped enlargement was first seen in an embryo of 8 mm. nape length. The transition of the secreting tubule into the collecting one does not fall within the section. At this point the tubule of older embryos forms a loop directed either cranially or caudally. This is the only change that occurs after the tubule has acquired its S-shape. I have studied 200 models of mesonephric tubules from the most different stages of development, and in none of them have I found any coiling of the tubule. Consequently the reconstruction of a tubule which I show in Fig. 192 of my contribution to Hertwig 's Handbuch der vergleichenden und experimentellen Entwicklungsgeschichte (1905), taken from Kollmann, must represent a very exceptional rarity. The collecting tubule runs almost straight to the primary excretory duct and is lined by a dark, cubical epithelium.

Fig. 562. — Transverse section of a human embryo of 4.7 mm. vertex-breech length, 4.9 mm. nape length, and with 33-35 pairs of primitive segments. (Embryo 137, G. 31, from the collection of the II Anatomical Institute, Berlin, Professor O. Hertwig; slide 12, row 2, section 1.) The'section shows in its upper part the position of the completely developed mesonephric canal in the urogenital fold and the size of both relative to the entire embryo. In its lower part it still shows the nephrogenic cord plainly, on the left it is in process of breaking up. In the cloaca the rectal.portion is beginning to differentiate from the urogenital sinus portion.

Fig. 563. — Transverse section of the urogenital fold of a human embryo of 9.5 mm. greatest length and with 38-39 pairs of primitive segments. (Embryo Ma. 3, from the collection of Professor Hochstetter, Vienna; slide 12, row 1, section 1.) X 175. The section passes through the 10th thoracic segment. In the fold there succeed each other in the dorso-ventral direction the v. cardinalis post., the complete 22nd tubule, a section of the Malpighian corpuscle of the 23rd tubule and a section through a portion of the Bowman's capsule of the 24th tubule, without any vascular glomerulus. From the Bowman's capsule of the^tubule thejsecretory tubule gradually develops, it ascends upwards for a short distance and then bends medianly almost at a right angle, enlarges, and again bends around and runs as the collecting tubule to the primary excretory duct. At the point where it bends around to become the collecting tubule it makes a srnalljjoop and here the tubule appears to be interrupted, the ends of the secretory and collecting tubules appearing.in section. The 14th mesonephric artery arises from the aorta, but it cannot be followed further. The reproductive gland anlage is of great extent, but has not yet begun to proliferate.

Fig. 564 a and b. — Transverse section of the mesonephric^ fold of a human embryo of 4.9 mm. nape length at the level of the 14th primitive segment. (Embryo 137, G. 31, from the collection of the II Anatomical Institute, Berlin, Professor O. Hertwig; slide 13, row 1, section 3.) - 150. The figure shows distinctly a mesonephric ccelomic funnel. It probably is in connection with a rudimentary nephrostome canal, but its lumen is not continued into the lumen'of the tubule.

Later, when the various parts of the tubules increase in thickness they no longer find sufficient room to lie beside each other in the transverse plane and they become displaced either caudally or cranially, the collecting tubule sometimes and sometimes the secretory one coming to lie more cranially.

The connection of the various mesonephric vesicles with the lateral plate, which is shown in Figs. 558 and 559, disappears completely in the anterior tubules, but in the posterior ones it may persist for a time and may be invaginated in a funnel-like manner ; the lumen of the body cavity, however, never reaches that of the mesonephric tubule. Fig. 564 shows one of these ccelomic funnels, that is situated at the point where the primitive segment stalk originally was. A large number of such funnels occur in an embryo of 4.9 mm. nape length, one corresponding to each 21st, 23rd and 28th-32nd tubule on the right and on the left to the 22nd and the 24th-34th.

The union of the tubules to form the mesonephros and the position of the various tubules in this can only be followed in a model. Fig. 565 shows a model of both mesonephroi of a human embryo of 7 mm. greatest length. The mesonephroi are still in the process of development; of the 30-31 tubules present on either side only the 25 cranial ones are fully formed. Both mesonephroi form almost a straight line and are parallel, the Malpighian corpuscles being so placed one above the other that they form a column ; in reality, however, the mesonephroi are not straight but are curved in correspondence with the contour of the body. The Malpighian corpuscle passes over into the secretory tubule always at its lateral side, and the secretory tubule passes over into the collecting one on the medial side of the corpuscle. The corpuscle and the two tubules still lie in the same plane; the coil that they form is almost in contact with the primary excretory duct. The space at the disposal of the various tubules within the mesonephros is still quite sufficient for the degree of development they present, and consequently while the Malpighian corpuscles are to some extent in contact, they do not press upon one another. If one compares the corpuscles of the mesonephros with those of the metanephros, the size of the former is striking, and this size brings it about that the vascular glomerulus never fills the entire cavity of the Bowman's capsule, both its cranial and caudal ends remaining free from vascular loops. Figs. 566 and 567 represent the model of both mesonephroi of an embryo of 9.5 mm. greatest length. The organs are fully developed, each of the 32-34 tubules consisting of a Malpighian corpuscle, a secreting and a collecting tubule. A comparison with Fig. 565 shows that the separation of the two organs transversely has begun. The excretory duct in the model represented in Fig. 565 runs in a straight course to the point where the ureter is given off; the right-angled bend caudal to the 31st right tubule is an illusion of perspective, the excretory duct, which is seen partly from the right side, making a strong bend ventrally.

The duct of the model shown in Fig. 566 is curved and the curvature is especially marked in the lower portion caudal to the 24th right tubule. This curvature of the duct corresponds to the bayonet-shaped bending of the urogenital fold (Fig. 552). The Malpighian corpuscles also form a column. They press upon or overlap one another, however, as a result of their increased size. As the lateral view (Fig. 567) shows, individual corpuscles have already been forced dorsally or ventrally. The collecting and secreting tubules have increased in size and their connecting loops now frequently pass beyond the medial borders of the Malpighian corpuscles, a fact which is of importance for the understanding of the urogenital union. The parts of the various tubules no longer lie in the same transverse plane ; the Malpighian corpuscle has dropped down and frequently lies at the same level as the collecting tubule of the next mesonephric segment. Finally, in Fig. 568 the model of the right mesonephros of an embryo of 19.4 mm. vertex-breech length is shown. The mesonephroi have separated still more (Fig. 552 is taken from the same embryo) ; that the cause of this separation is the suprarenal bodies and the metanephroi has already been shown in the section on the urogenital fold. As a result of it the curvature of the primary excretory duct becomes continually stronger, since it must reach the medially placed bladder, and now it appears to be bent around almost at a right angle. This strong curvature cannot be without influence on the tubules of the region in which it occurs. Even from the 24th tubule downwards the Malpighian corpuscles are pressed together and form in their totality an arch convex caudo-laterally. The corresponding tubules are greatly elongated, the collecting tubules form long passages which converge towards the concave side of the arch of Malpighian corpuscles and then bend around into the secreting tubules; they are thereby greatly pressed together, lying close beside each other through long stretches and may connect with one another. By this new union the collecting duct of one tubule may serve as the efferent for several. A further result of the bending of the excretory duct is the union of the orifices of several of the collecting tubules to form a single one. In this way the number of collecting tubules opening into the excretory duct is diminished and no longer corresponds with that of the Malpighian corpuscles. I shall discuss later in a special section the degeneration phenomena which are clearly shown in Fig. 568. As the result of a strong growth of the collecting tubules the entire coil of a mesonephric segment is further away from the excretory duct, and with further development the interval becomes continually greater until, finally, there occurs the already described subdivision of the mesonephros into two portions, a lateral one with the excretory duct and the lateral portions of the collecting tubules, and a medial one with the Malpighian corpuscles, the secreting tubules, and the medial portions of the collecting ones.

Fig. 565. — Model of the mesonephroi, the metanephroi and the bladder of a human embryo of 7 mm. greatest length. (Embryo Chr. 1, from the collection of Professor Hochstetter, Vienna. The model has been prepared by my students A. Wachter and E. vonWyss.) View from in front. 75. The mesonephroi have a perpendicular position and the individual tubules are at sufficient distances, so that the Malpighian corpuscles lie in a single row. The lowest tubules are still developing, but have already united with the excretory duct (Fig. 561 e). The fully developed canals are S-shaped. Most ventrally are the Malpighian corpuscles, then in the middle is the secreting tubule and most dorsally the collecting tubule. The loop formed by the secreting and collecting tubules usually has not yet reached the medial border of Bowman's capsule. The whole coil of a mesonephric segment (Malpighian corpuscle and the two tubules) lies still in one horizontal plane. The primary excretory duct is narrow in its upper part and only widened in a spindle-shaped manner corresponding to the entrance into it of a collecting tubule; in its lower portion it is remarkably wide and the spindle-like enlargements have disappeared. Towards the bladder the duct narrows again. The ureters are short and almost straight canals terminating in clubshaped enlargements.

Fig. 566. — Model of the mesonephros of a human embryo of 9.5 mm. greatest length. (Embryo Ma. 3, from the collection of Professor Hochstetter, Vienna. IkThe model was prepared by my students J. Klausler and Wydler.) Seen from in front. 75. The two mesonephroi are slightly curved. The individual tubules are pressed together; the Malpighian corpuscles as a result no longer form a single row, but some are beginning to overlap their neighbors. All the tubules are bent into an S-shape. The opening of each tubule into the excretory duct lies about on a level with its Malpighian corpuscle, sometimes slightly more cranially, sometimes slightly more caudally. The various openings are not always the same distance apart. Some tubules are already broken down, on the right the 1st, 3rd and 4th, on the left the 1st, 2nd, 4th and 5th. Some have a common collecting duct, the 15th and 16th, the 19th and 20th. Some Malpighian corpuscles are entirely without tubules, 5a and 27 on the right, 22 on the left. Some tubules are completely formed, but have no opening into the excretory duct, on the right the 16th, on the left the 23rd. The excretory duct is widened in its caudal course from the opening of the 21st tubule, and beyond the opening of the last tubule it narrows again. Seen from in front the bladder appears foreshortened, one looks directly on the cloacal membrane.

Fig. 567. — The same model that was represented in Fig. 566 seen from^the right side. The figure shows how the Malpighian corpuscles as they enlarge press each other out of the row. The primary excretory duc t increases in size in its caudal half, the increase affecting both the dorsoventral and the frontal diameter, the latter to the greater extent^ It again d imini shes before its union with the ureter.f This has already developed the primitive pelvis of the kidney, from which the cranial and caudal pole tubules are beginning to grow out.J. The common terminal part of the ureter and the primary excretory duct still persists, but is greatly enlarged. The cloaca is almost completely subdivided.

The special relations of the caudal tubules make possible a division of the mesonephros into a cranial and a caudal portion; a similar subdivision occurs in the mesonephros of all vertebrates. The upper portion later unites with the reproductive gland and becomes the sexual portion of the mesonephros, the epigenitalis (epididymis in the male, epoophoron in the female). The lower portion in animals without a metanephros becomes the actual functioning kidney and is consequently termed the glandular portion, but in animals in which the metanephros separates from the mesonephros this lower portion also undergoes degeneration and becomes the paragenitalis (paradidymis in the male, paroophoron in the female).

The primary excretory duct is also modified during the development of the mesonephros. It is at first (in embryos up to 4.25 mm. vertex-breech length) solid. Its lumen appears discontinuously, at first at both ends ; at the caudal end the lumen appears to extend into it from the cloaca. In an embryo of 7 mm. greatest length the lumen is fully formed and the excretory duct is in open communication with the cloaca. The course of the duct to its opening is not straight. Attention has already been called to the frontal bend which gives the first impetus to the concentration of the paragenitalis. A second bend in the sagittal direction appears quite early. In an embryo of 4.25 mm. vertex-breech length, the duct extends to the 28th primitive segment, parallel to the vertebral column. Having arrived at this point it bends almost at a right angle and reaches the bladder by a short horizontal piece. This bend is important ; I shall term it the sagittal bend of the excretory duct.

At the openings of the collecting tubules the excretory duct is widened. In the upper portion these widened places are separated by narrow ones, but in the lower part they flow together (Figs. 565, 566). The wide portion of the duct thus formed has at least twice the diameter of the unwidened cranial portion. This widening persists for an extraordinarily long time, and one might be inclined to correlate it with the ampulla of the ductus deferens which is formed later on. That is not possible, however, for the widening completely disappears, as will be seen from what follows. It has already been pointed out that the widened places receive the collecting tubules of those mesonephric tubules which later concentrate to form the paragenitalis. I shall anticipate by stating that the union of several collecting tubules of the paragenitalis to form a common terminal part, already referred to in the description of Fig. 568, continually progresses in the later stages of development, until, finally, only a single efferent duct remains, which, in the adult individual, is the ductulus aberrans Halleri. With the formation of this ductulus it becomes possible to determine the position of the embryonic widening; it corresponds to the lower part of the epididymis. It extends down to the sagittal bend of the excretory duct and at this point the duct becomes smaller; its horizontal piece is narrow. In Fig. 568 a narrowed portion of the excretory duct occurs in the region of the openings of the 24th-29th collecting tubules. The duct does not, however, actually narrow; the appearance is produced by a spiral twisting of the duct through 90°. It is very possible that this twisting is due to the union of the mesonephric fold with the anterior abdominal wall by means of the inguinal fold. In the section on the urogenital fold the displacement of the anterior abdominal wall and the influence this exerts on the entire mesonephric fold have been described. This influence will, naturally, also be felt by the contents of the fold and thus the spiral twisting of the primary excretory duct may be the result of a displacement of the anterior abdominal wall which is carried to the mesonephric fold by the inguinal fold.

Finally the relation of the excretory duet to the a. umbilicalis must be considered. The artery has two root areas, a visceral one from- the rami intestinales and a parietal one from the parietoventral vascular arches of the aorta. The visceral root area begins in the third cervical segment and extends into the lumbar region. It is present throughout its entire length in no embryo ; the individual roots are developed in succession ; at first only the cervical ones are present ; then Fig. 568. — Model of the right mesonephros and Mullerian duct of a human embryo of 19.4 mm., greatest length. (Embryo Ma. 2, from the collection of Professor Hochstetter, Vienna. The model was prepared by my students Massard and E. Chome'.) The mesonephros has undergone extensive degeneration. All the tubules in the thoracic segments and some of those in the lumbar show degeneration. Either the Malpighian corpuscle is wanting or it is rudimentary and separated from its secreting tubule; this is the case in tubules 1-12, that is to say in a full third of those present. All these tubules are also broken, either in the secreting tubule or between this and the collecting one. These latter are rarely hollow, and if so, their lumina are filled with degenerating epithelium. Finally the tubules 33, 34 and 35 consist only of collecting tubules; 33 has still a portion of the secreting tubule and near it are two large Malpighian corpuscles without any connection. Similarly tubule 31 consists only of a collecting tubule; a Malpighian corpuscle with two secreting tubules seems to correspond with it; some of the remaining tubules are broken, 14, 17, 18, 19; some have solid collecting tubules, 13, 14, 15, 16, 17, 19, 21, 23, 25 and 26, and some solid secreting tubules, 22 and 24; the collecting tubules of 28, 29 and 31 form ureter-like structures. Only the remaining tubules 20, 27, 30 and 32 are fully formed. The appearances of degeneration seen in the most caudal tubules are not to be regarded as persistent without further evidence. From older series it is seen that the lower collecting tubules which have separated from the excretory duct may acquire a new opening into a neighboring collecting tubule, just as separated Malpighian corpuscles and secreting tubules may communicate anew with neighboring tubules. By the bending of the primary excretory duct at the end of the mesonephros, the Malpighian corpuscles of the 26th to the 34th tubules are pressed together so as to form a special group, from which, later, the paragenitalis is formed. The Mullerian duct is quite independent of the ccelomic epithelium and the excretory duct; it is still straight, since its down-growing tip has just reached the head of the excretory duct.

the origin of the artery wanders more caudally into the upper thoracic region, and the cervical roots degenerate; after the thoracic roots have functioned for a time the artery wanders into the lower thoracic and the upper thoracic roots become fused, and so on. The parietal root area first appears when the artery has reached the lumbar region. From its mode of origin the visceral root area must lie to the medial side of the excretory duct and the parietal area on its lateral side. Since both root areas exist for a time together, the excretory duct runs in an island surrounded directly by the arteries; in Fig. 569 a section is shown passing through a region where both roots of the a. umbilicalis are present at the same time.

Fig. 569. — Transverse section of an embryo of 5.3 mm. greatest length and 4.6 mm. nape length, at the level of the 24th segment. (Embryo 1420, H. s. 112, from the collection of Professor Keibel, Freiburg i. Br.; slide 12, row 2, slide 1.) X 75. The section passes through the root area of the a. umbilicalis; between the parietal and visceral roots there is an island of tissue in which the primary excretory duct occurs.

The First Degeneration Period of the Mesonephros

The degeneration of the mesonephros occurs in two periods. The first begins even before the organ has reached its full extent and is finished in embryos of 21 mm. greatest length. During its progress the greater part of the mesonephros beginning at the cranial end, is degenerated. The second period comes on so gradually that definite time limits are not possible for it. In it there occurs a new suppression of tubules and the selection of those that will enter the service of the reproductive system and of those that will persist even in the adult as the paragenitalis.

The three models of the mesonephros (Figs. 565, 566 and 568) do not reveal the important part played by the degeneration, that is beginning in the cranial portion ; its influence becomes clear only by a study of the table opposite. In the lowest horizontal column of this table the entire number of mesonephric tubules is given for each embryo in which they were counted. The numbers, as they stand there, vary so little that they do not suggest a necessity for further controls ; we shall further learn from the description of the development of the Mullerian duct that its ostium abdominale lies apparently always at the upper end of the mesonephros in embryos of both sexes throughout the entire development. Both these facts distract the attention of the observer from the really enormous degeneration that the mesonephros undergoes.

The degeneration shows itself as soon as the growth cranially of the mesonephros is completed, that is to say in an embryo of 5.3 mm. greatest length, at a time therefore at which the organ is still growing caudally. It always begins at the most cranial tubule and makes a continuous and regular progress up to the time men tioned above. In the table the number of the tubules occurring in each segment is given, those that are degenerating being given in roman numerals and those that are fullv formed in arabic. An asterisk with an arabic number indicates that the tubule is still developing. If we follow the cranial limits of the various mesonephroi through the table from left to right we see a marked and, what is important for the regularity of the process, a continuous recession of these limits. Beginning with an embryo of 5.3 mm. ureatest length and ending with one of 21 mm. greatest length the line that unites the cranial limits recedes from the 6th cervical to the 1st lumbar segment, that is to say, there is a recession through 15 segments. The relative constancy of the total number of tubules throughout this period justifies the question whether the recession of the cranial limit may not find its explanation in the crowding together of the tubules within a smaller space; the question is all the more justified, because we actually know of the occurrence of this process in the development of other craniota, both in the case of the pronephros and in that of the mesonephros. That there is a crowding together of the tubules during the first period of degeneration is very probable, at all events the table opposite shows an increase in the number of tubules in the segments of the degenerating zone. The process plays, however, only an unimportant part, as is shown by the course of the mesonephric arteries, to be described later on. A crowding together of the tubules of 18 segments into 3 would cause the arteries to assume almost a vertical course, but they remain horizontal throughout the first degeneration period. The principal cause of the caudal recession is the progressive degeneration of the mesonephric tubules. Throughout this period I have uot seen a single mesonephros that did not show degenerating tubules in at least one, usually in two or three of the cranial segments; the regularity of its occurrence speaks for the normality of the process. Reckoned by the segments the upper five-sixths of the mesonephros, the 6th cervical to the 12th thoracic, 15 segments in all, degenerate and only the caudal sixth, the 1st to the 3rd lumbar segments, three segments in all, persist. Estimated by the tubules 57 are degenerated out of a maximum of 83 that develop, that is to say, two-thirds of all the tubules degenerate and 26, or one-third, persist. The number 57 is estimated, not observed; the number 26 was the maximal number found in the enumeration of the tubules of different embryos. Naturally the corresponding portion of the primary excretory duct disappears along with the tubules. Once all the tubules of the cervical and thoracic segments have disappeared, 1 a period of rest supervenes ; from the stage of 21 mm. greatest length onwards, all embryos show a rather constant number of mesonephric tubules in the lumbar segments, but these tubules are almost all broken in one or several places.

The degeneration appears under very different forms: 1. the Malpighian corpuscle may separate from the secreting tubule; 2, the transition of the seci'eting into the collecting tubule may be broken; 3, the collecting tubule may be solid or its epithelium desquamated; 4, the secreting tubule may be broken into several portions; 5, the collecting tubule may not be connected with the primary excretory duct. The variations are arranged according to their frequency of occurrence. The separation of the Malpighian corpuscle from the secreting tubule is almost always to be found in all degenerating tubules and all show a tendency to lose to a greater or less extent their S-shape.

Notwithstanding the degeneration, which occurs in the cranial canals before the beginning of the histological differentiation of their epithelium, an epithelial differentiation between the collecting and secreting tubules occurs very frequently even in the remains of the cranial canals.

Later Formed Tubules

In a 7 mm. embryo the mesonephros has reached its caudal limit and no new anlagen are added in older embryos at the caudal end of the organ, and, furthermore, within the mesonephric area at least two-thirds of the tubules degenerate and yet the number of tubules between the stages of 5.3 mm. and 19.1 mm. greatest lengths increases rather than diminishes. This is only possible if a new formation of tubules occurs. From the table facing p. 816 one may determine this fact, at least for the lower thoracic and lumbar segments, by a comparison of the numbers for the individual embryos.

• Remains of cranial tubules may persist a long time; my colleague, Professor Zuckerkandl. called my attention to such remains in the 3rd thoracic segment. Vol. 11—52

The question as to the origin of these later formed tubules is difficult of solution and an absolutely certain statement cannot be made. In the first place, the sources from which they can not come may be excluded. In most vertebrates the later formed tubules arise from unused remnants of the parent tissue. At the first appearance of the mesonephric tubules (embryo of 2.6 mm.) such remnants of the nephrogenic cord may be seen, but in an embryo of 7 mm. they have completely disappeared; there is no trace of compact remnants of the nephrogenic cord and the only remaining possibility is that scattered cells, indistinguishable from the surrounding mesoderm, later become aggregated to form cell masses from which the later formed tubules arise just as the primary ones did. This process occurs in Teleosts and Amphibia, but the most thorough search fails to reveal any structures that might serve for such a method of new formation in human embryos. New formation from remnants of the nephrogenic cord I regard as excluded ; there remains only the hypothesis that the new formation takes place from the tubules already present by division or budding. For both these methods incomplete evidence occurs. Malpighian corpuscles occur which are twice as large as the rest, others occur that are in process of division, others that are completely isolated and without a tubule, others with tubules which communicate after a short course with fully developed tubules, and, finally, completely developed tubules occur that are not yet in connection with the primary excretory duct. These facts taken by themselves might be interpreted by supposing that a Malpighian corpuscle divides, the one portion remaining in connection with the original tubule, while the other develops a new tubule, which opens either into a neighboring tubule or into the primary excretory duct. This process of development receives plausibility from the fact that the Malpighian corpuscle represents the original parent tissue, that is, the lateral part of the primitive segment stalk.

Further divisions occur in the canals, usually in the secreting tubules, and also diverticula occur, given off either from the transition of the secreting into the collecting tubule or, less frequently, from the transition of the Malpighian corpuscle into the secreting tubule ; corresponding with this, collecting tubules occur with two or three secreting tubules communicating with them. This allows us to assume that a mesonephric tubule in some part of its course, almost always in the secreting tubule, forms a diverticulum which, elongating, gives rise to a new tubule with a Malpighian corpuscle. All the various conditions mentioned above are shown in Figs. 565-568 and are described in the explanations of these figures.

In the Gymnophiones and birds special efferent ducts are developed for the later formed tubules by the primary excretory duct sending out towards them small canals, which may serve as collecting tubules for several of the later formed tubules. These new formed ducts are termed mesonephric ureters, in contrast to the metanephric ureter. Such ducts also occur in the human embryo. In an embryo of 12.5 mm. greatest length I found in the 12th thoracic and 1st lumbar segment six of them certainly and perhaps two others, and in an embryo of 19.4 mm. three of them. All these pieces of evidence are, however, not final, since in addition to the new formation degeneration is also taking place and the degeneration, whose varieties have been mentioned above (p. 817), may present almost the same appearance as the new formation. Only those cases of division and diverticulum formation are of consequence which occur in segments lying far away from the degeneration zone. Since such cases may be found, I assume that later formed tubules arise by the division and budding of those already present.

Rudiments of Pronephric and Mesonephric Tubules

Now that the development and degeneration of the rnesonephros has been considered, we are in a position to discuss the tubule remnants that have been described in the literature as pronephric rudiments. We have established 1. That pronephric tubules occur only in the region of the first 12 primitive segments; the pronephros, accordingly, never extends caudally beyond the 1st thoracic segment.

2. That mesonephric tubules are developed in the region from the 5th cervical to the 3rd lumbar segments; consequently the mesonephros does not extend cranially beyond the 5th cervical segment.

We may say then at once that all remnants of tubules from the 2nd thoracic segment caudally are to be referred to mesonephric tubules and all those in front of the 5th cervical segment are certainly derived from the pronephros. Those that occur between the 5th cervical and the 1st thoracic segment may be either pronephric or mesonephric, a certain diagnosis is impossible. According to the results of the table facing p. 816, however, it is probable that all these uncertain remains come from the mesonephros ; the pronephros, therefore, has already completely degenerated. The statements that occur in the literature to the effect that the occurrence of the remains of a tubule dorsal to the v. cardinalis posterior is a certain indication of its pronephric origin (Tandler 1905, Veit 1909) have no weight.

It is clear that no pronephric remnants can persist in the neighborhood of the epigenitalis, which arises from the caudal sixth of the mesonephros; all accounts of the occurrence of pronephric tubules on the epoophoron are therefore incorrect.

Arteries of the Mesonephros

There are developed on each side in maximo 30 mesouephric arteries, which are distributed throughout the entire mesouephric area. At first they are entirely distributed to the mesonephros, but later they also supply the reproductive glands, the suprarenal bodies, the metanephroi and the diaphragm. These new regions of distribution prevent their complete degeneration when the mesonephros disappears. A variable number of them persist as the phrenic, suprarenal, renal, accessory renal, internal spermatic, and accessory spermatic arteries and as the rami ad lymphoglandulas lumbal es and ad sympathicum.

The first mesonephric arteries are found in embryos of about 5.3 mm. greatest length. Thev arise from the lateral surface of the aorta and pass almost horizontally laterally to the urogenital fold, reaching the Malpighian corpuscles and terminatiug in them with an enlargement which usually assumes a spherical shape (Fig. 570). Later a network of vessels (Fig. 571) occupies the place of the enlargement, and this also is connected with the v. cardiualis posterior. The number of the mesonephric arteries cannot be determined with perfect certainty, because in their case as in that of the mesonephric tubules degeneration follows upon the steps of the progressive development from the cranial end. The degeneration process is, however, less readily followed in the arteries, since manv series are not satisfactorv on account of the emptiness of the vessels: this may be shown by the following statements.

Fig. 570. — Transverse section of the right mesonephros of a human embryo of 5.3 mm. greatest length and 4.6 mm. nape length. (Embryo 1420, H.s. 112. from the collection of Professor Keibel, Freiburg i. Br.; slide 11, row 4, section 1.) A mesonephric artery is cut almost throughout its entire length; the middle portion on account of a curve does not fall within the plane of the section. The artery ends in a spherical enlargement within a Malpighian corpuscle.

In an embryo of 7 mm. greatest length there were 10 mesonephrie arteries in the region of the lst-Gth thoracic segments; an embryo of 9.5 mm. greatest length possessed I5 arteries in the 2nd-12th thoracic segments; in an embryo of 10 mm. greatest length 9 arteries were limited to the 7th thoracie-lst Lumbar segments; one had 11 arteries distributed to the 6th-llth thoracic segments. In an embryo of 13 mm. the degeneration was completed, it still possessed 11 arteries m the region of the 10th thoracic to the 3rd lumbar segment, and, similarly, in an embryo of 18 mm., 9 arteries were counted from the 10th thoracic to the 3rd lumbar segment.

From this statement it appears that with increasing age the arteries continually recede into the lumbar segments, disappearing from the thoracic ones. Consequently the direction of the arteries

Fig. 571. — Transverse section of a human embryo of 7.0 mm. greatest length. (Embryo Chr. 1, from the collection of Professor Hochstetter, Vienna; slide S, row 10, section 6.) The a. mesonephridica (the 8th) is cut almost throughout its entire length. It forms two distinct branches inside the glomerulus. The anlage of the reproductive gland, indicated by the thickening of the coelomic epithelium, is of a fair extent; its growth has not yet begun.

is always horizontal (Fig. 572), so that an extensive compression of the mesonephros, which has already been excluded in considering the degeneration of the mesonephrie tubules, cannot be assumed. If the mesonephros were compressed from a condition in which it extended from the 6th cervical to the 3rd lumbar segment to one in which it extended only from the 1st to the 3rd lumbar, the vessels remaining the same, as they naturally would if all the tubules persisted, then the arteries situated more cramally would run almost parallel to the aorta.

The arteries are not paired, nor are they arranged segmentallv. In the cranial segments one artery may lie in each segment or two in three segments, but in the caudal region the arteries become more numerous, just as the mesonephrie tubules are more numerous in the lumbar segments than in the thoracic ones ; in the lumbar region as many as 4 arteries may lie in a single segment. There are always fewer arteries than tubules, so that one artery must always supply several tubules ; this may easily be seen in the lumbar segments. If, for the purpose of determining the maximal number of mesonephric arteries the highest numbers found in each segment be added together, one obtains 30 arteries for 83 possible tubules.

From the 16th thoracic to the 3rd lumbar segment all the mesonephric arteries may persist; from them are formed the phrenic, suprarenal, renal, accessory renal, internal spermatic and accessory spermatic arteries and also the branches to the lymph nodes and the sympathetic ganglia in the region between the superior and inferior mesenteric arteries. These persistent mesonephric arteries we must consider more in detail. In order to reach the mesonephros they must traverse a rather broad area, in which the suprarenal body lies, this organ in consequence of its rapidly increasing size, coming to lie with its ventral surface in the same frontal plane with the mesonephric fold (Fig. 553). By the suprarenal body the arteries — I am describing them from the reconstruction of the vessels of an embryo of 18 mm. greatest length shown in Fig. 572 — are divided into three groups ; a cranial group, running dorsal to the suprarenal body (1st and 2nd arteries on either side), a middle group, whose vessels pass through the suprarenal body (on the right 3rd and 4th, on the left 3rd-5th arteries), and, finally, a caudal group, whose vessels pass over the ventral side of the suprarenal body (on the right 5th and 6th, on the left 6th and 9th arteries). The 7th and 8th arteries on the right and the 9th and 10th on the left pass caudal to the suprarenal body and therefore are not necessarily influenced by it, so that they are compressed by the development of the metanephros in the same ventral position.

Fig. 572. — Reconstruction of the mesonephric arteries of a human embryo, made from an embryo of 18 mm. greatest length and superposed on the model of an embryo of 19.4 mm. greatest length. Mesonephroi, reproductive glands, and suprarenal bodies are represented with solid, the metanephroi with dotted contours. The circles on the anterior surface of the aorta indicate the origin of the cceliac (which in this embryo is still paired), the superior and inferior mesenteric arteries. The mesonephric arteries are indicated by numbers placed below them. The right 5th, 6th, 8th, and 9th and the left 7th, 8th, and 9th arteries form a network between the reproductive gland and the metanephros, the rete arteriosum urogenitale.

The mesonephric arteries 5-9 or 7-9, and probably more, in the angle between the reproductive gland ventrally, the mesonephros laterally, and the metanephros dorsally, form a network, the rete arteriosum urogenitale (Fig. 572). From this network, which is remarkably coarse-meshed and composed of large vessels, the mesonephros, reproductive gland and metanephros are supplied with arterial branches, these organs being thus independent of single branches for their blood supply; should one or several roots degenerate, neighboring ones can take their places. In Fig. 572, for instance, the second mesonephric artery on the right side has become divided into an ascending and a descending branch; the ascending one supplies the entire upper half of the mesonephros and the reproductive gland, a region that in younger embryos receives its blood from several mesonephric arteries belonging to more cranial segments. The occurrence of this network at once explains why all persisting arteries that arise from the roots of this network show, within certain limits, a variability in the point of their origin from the aorta, Each of the 9-11 remaining mesonephric arteries may become an a. spermatica interna, since all supply the reproductive gland; this explains the frequently observed multiplicity of these arteries and the not infrequent difference in the place of origin of the right and left ones. Usually the artery is formed from one of the arteries of the third group and will therefore lie ventral to the kidney and to the ureter. Rarely an artery of the first group becomes a spermatic artery, the accessory spermatic; in this case the vessel will pass dorsal to the a. renalis, which is formed from an artery of the second or third group, and dorsal to the pelvis of the kidney.

The suprarenal arteries arise from the second, rarely from the first group. Since the inferior phrenics may also form from the first group it is clear that a suprarenal artery may arise from the inferior phrenic or vice versa. The renal arteries are not new formations, as has hitherto been supposed, but each is formed from a mesonephric artery. The kidney climbs upwards to the mesonephric arteries, as if on a ladder : as soon as a sufficient blood supply is assured crania Uy, the caudal branches separate from it. When the kidney has acquired its definitive position it possesses several renal arteries, and of these one becomes greatly enlarged to form the definitive artery, while the others either degenerate or persist as accessory renals. The definitive renal artery is either the last vessel of the second group or the first of the third group. The relations of both groups, of the second to the suprarenal artery and of the third to the internal spermatic, explain the variation in which the renal artery arises from a suprarenal or from an internal spermatic ; in the first case it may be the principal stem, in the latter onlv an accessorv renal. The relations between the urogenital rete and the metanephros show how the accessory renal arteries may develop, and explain their very varied relations to the kidney. Accessory renals from the first group will be branches of the superior suprarenal and must pass over the dorsal surface of the kidney and consequently they first reach the kidney on its dorsal surface and there penetrate its cortex. Those from the second group will be branches of either the middle or inferior suprarenal and will reach either the hilus of the kidney or the medial edge above this. Those from the third group may enter at the hilus, at the medial edge below it or on the ventral surface of the caudal half of the organ.

Should, exceptionally, a caudal branch be retained, it will be drawn upwards by the migration of the kidney, a condition that explains why such an artery may cross the principal stem or another accessory.

Since the mesonephric arteries of the third group also give branches to the lymphatic nodes and to the sympathetic ganglia of this region, even if they do not become renal or internal spermatic arteries, they will be retained at least as far as the point of origin of the small branches. In consequence of the medial position of these structures these mesonephric arteries will run directly ventrally and there will be a displacement of their points of origin from the aorta, so that they seem to be purely ventral branches and might therefore readily be mistaken for persistent rami intestinales. One must, accordingly, be very careful in identifying as a ramus intestinalis any ventral branch occurring between the origins of the superior and inferior mesenteries or at the level of the latter. Only when they arise some distance below the inferior mesenteric may one without hesitation identify ventral branches as rami intestinales.

A displacement of their origins from the surface of the aorta appears very early in the arteries of the various groups. The vessels of the dorsal group shift their origins to the dorsal surface of the aorta, those of the third group to the ventral surface, and only those of the second group preserve their original lateral position.

In embryos of 26 mm. greatest length the definitive arrangement of the arterial apparatus is acquired.

Veins of the Mesonephros

The veins form within the mesonephros a special circulatory system, the mesonephric portal circulation. A schematic description based upon the venous arrangement of all craniotes that possess mesonephroi, will explain what is meant by the above term.

The mesonephric portal circulation is formed by two veins, the v. cardinalis posterior and the v. subcardinaJis. The former courses along the dorsal surface of the mesonephros, somewhat nearer its lateral border; and the latter along the ventral surface and somewhat nearer the medial border (Fig. 573 a). At the caudal border of the organ both veins are connected with the veins of the caudal half of the body, the v. caudalis and the v. ischiadica, and are thereby connected with one another (Fig. 573 a). Very soon numerous branches develop between them, these branches traversing the mesonephros (Fig. 573 b). From the beginning the v. cardinalis posterior opens into the ductus. Cuvieri and through this into the sinus venosus of the heart (Fig. 573 a and b) ; the v. subcardinalis has to acquire a connection with the heart and it does so by means of the hepatic vein. This unpaired hepatic vein, v. hepatica revehens communis (Fig. 573 a and b), arises by three branches, the v. revehens dextra and sinistra and the ductus venosus Arantii; it grows caudally and, dividing, comes into connection with both v. subcardinales; these latter thus become connected with the heart, since the v. hepatica revehens communis opens into the sinus venosus. Accordingly the blood from the posterior half of the body flows toward the heart by the two paths: 1, by the v. cardinalis posterior and the ductus Cuvieri; 2, by the v. subcardinalis and the v. hepatica revehens communis. This changes, however, very soon ; for, in the first place, the v. cardinalis posterior becomes interrupted at the upper pole of the mesonephros and the return of its blood to the heart is thus prevented, and, secondly, the connection between the v. subcardinalis and the w. caudalis and ischiadica breaks, so that the blood from the posterior half of the body is received only by the v. cardinalis posterior (Fig. 573 c). If now the blood is to return from the latter to the heart it must make use of the transverse anastomoses between the v. cardinalis posterior and the v. subcardinalis, and reach its destination through this last vein. The transverse anastomoses, however, are broken up into capillaries between the coils of the mesonephric tubules (Fig. 573 c), so that the venous blood carried by the v. cardinalis posterior must, on its way to the heart, pass through a second set of capillaries, just as does the blood that passes by the vena portarum through the liver to the v. hepatica; hence the employment of the temi mesonephric portal circulation. "When this circulation is established we no longer speak of vv. cardinalis and subcardinalis, but of vv. mesonephridica advehens and revehens.

In the human embryo there also develops, in addition to the v. cardinalis posterior, a subcardinal vein, and the possibility is therefore present for the formation of a mesonephric portal circulation; that it really becomes functional is not proven, but all

826 observations made up to the present indicate that we have to do only with a rudimentary anlage, that exists only for a short time. The posterior cardinal vein is formed in a caudal direction from its opening into the ductus Cuvieri, and always lies in the angle between the primitive segment stalk and the parietal mesoblast. The discontinuity of the anlage and, in addition, the fact that it sometimes appears as a tube, sometimes as a groove, sometimes as composed of only one or two cells, makes probable its development in loco. The cell material used in its formation wanders from the parietal mesoblast and, eventually, from the primitive segment stalk ; I observed cells that were still connected with this germ layer, but whose bases projected beyond the other cells and formed part of the wall of the vein.

Fig. 573 a, b and c. — Plan of the development of the mesonephric portal circulation, a. A vein runs on both sides of the mesonephros, the posterior cardinal laterally and dorsally and the subcardinal medially and ventrally. The posterior cardinal opens cranially into the ductus Cuvieri, while the subcardinal ends blindly. Caudally both veins are connected with the paired v. ischiadica and the unpaired v. caudalis. b. Transverse anastomoses have formed between the posterior cardinal and the subcardinal, traversing the mesonephros. The subcardinal opens into the v. hepatica revehens communis, c. The cardinal has lost its connection with the ductus Cuvieri, but retains that with the vv. ischiadica and caudalis. The subcardinal has retained its opening into the v. hepatica revehens communis, but has lost its connection with the vv. ischiadica and caudalis. Capillary networks have developed in the transverse anastomoses. The blood of the vv. caudalis and ischiadica must now take the following course: 1, v. cardinalis, 2, the transverse anastomoses with capillary networks inside the mesonephros, 3, v. subcardinalis, 4, v. hepatica revehens communis, 5, sinus venosus.

In an embryo of 2.6 mm. greatest length and with 13-14 pairs of primitive segments no posterior cardinal vein was present. In one of 2.5 mm. greatest length and with 23 pairs of primitive segments they were developed from the ductus Cuvieri to the posterior border of the pronephros, that is to say, from the 5th to the 13th primitive segment. In an embryo of 4.9 mm. nape length they had already reached the caudal pole of the mesonephros — i. e., the 3rd lumbar segment. Their connections with the mesonephric glomeruli were first seen in an embryo of 5.3 mm. greatest length and 4.G mm. nape length. A duplication of the cardinal vein may occur in the anterior part of its course.

The subcardinal vein is at first double, a median and a lateral vein may be recognized. Both arise — so far as can be determined from individual embryos — from medial and lateral outgrowths from the posterior cardinal. The lateral outgrowths pass through the intervals that are left between the primary excretory duct and the coils of the mesonephric tubules, the medial ones run past the medial side of the Malpighian corpuscles (Fig. 574). Both sets of branches then develop ascending and descending twigs, which unite together to form a medial and a lateral longitudinal vessel, the subcardinal veins. Fig. 574 represents the arrangement of the mesonephric veins of a human embryo of 9.5 mm. greatest length. The posterior cardinal veins are fully developed and are connected with the v. caudalis and the v. ischiadica (these veins are not shown in the figure). The two subcardinals are in process of development; one sees in the cranial half of the mesonephros simple lateral and medial branches given off from the posterior cardinal, at the middle of the organ the formation of T-shaped branches and, finally, in the lower half the union of the T-shaped branches to large longitudinal pieces. A single complete subcardinal vein does not exist anywhere and it is questionable if such a vessel is ever formed. The two lateral subcardinals still terminate freely posteriorly and so also the left medial vein, but the right medial one is connected with a short but large vessel, which is developed on both sides and opens into the posterior cardinal after it has crossed the primary excretory duct; I term this the impaired terminal part of the subcardinal vein.

I see in the arrangement of these veins the homologue of a portal circulation; but since in an embryo of 12.5 mm. greatest length the subcardinal veins, and with them the transverse branches which connect them with the posterior cardinal, are in the act of degeneration, I do not believe that a portal circulation is actually established. The anlage is accordingly probably only a rudimentary one.

The first outgrowths from the posterior cardinal vein occur in an embryo of 7 mm. greatest length. Corresponding to the cranio-eaudal direction of the growth of the mesonephros these transverse veins develop in batches in the same direction, and similarly in correspondence with the early degeneration of the mesonephi-os the transverse veins in its cranial half degenerate very early, usually before they have reached the T-shaped stage. The unpaired terminal portion of the subcardinal appears in an embryo of 7 mm. greatest length, but at first only on the left side.

Fig. 574. — Mesonephric veins of a human embryo of 9.5 mm. greatest length. (Embryo Ma. 3, from the collection of Professor Hochstetter, Vienna.) The veins have been reconstructed and drawn upon the model of the mesonephros. The posterior cardinal vein develops horizontal branches arranged in two longitudinal rows, a medial and a lateral one. The branches of the medial row pass medially to the Malpighian corpuscles, those of the lateral row between the primary excretory duct and the coils of the tubules. At their ends most of the branches of both rows, especially in the caudal half of the mesonephros, develop ascending and descending branches, and portions of two longitudinal vessels are thus formed on either side, these vessels being the medial and lateral v. subcardinalis. The posterior cardinal, the transverse branches, and the two subcardinals represent a very rudimentary mesonephric portal circulation.

The Second Period of Degeneration of the Mesonephros

It has been shown above that the degeneration of the mesonephros is completed in two periods and the first of these has been described. We have now only to consider the second period. At its commencement, which occurs after a stage of 21 mm. greatest length, the mesonephros consists of only 2(5 tubules in maximo, all situated in the lumbar region and all broken, that is to say in process of degeneration, in some portion of their course, usually at the angle between the secreting and collecting tubules. This remnant of the mesonephros is divided into two portions by the establishment of the urogenital union, an upper genital and a lower so-called gland portion. I shall term the former the epigenitalis and the latter the paragenitalis. The number of tubules belonging to the epigenitalis varies between 5 and 12, and it is a striking fact that in the male embryo the lower number is more nearly approached and in the female the higher one. The number of tubules in the paragenitalis varies, since it is in process of complete degeneration. The epigenital tubules correspond approximately to the 58th-69th mesonephric tubules and the paragenital to the 70th-83rd.

The epigenital tubules in both sexes are all broken at the same point, namely, at the bend of the secreting into the collecting tubule. The secreting tubule and the Malpighian corpuscle of each tubule have either disappeared in the first period of degeneration or else disappear at the beginning of the second period. The blind ends of the persisting collecting tubules enlarge, their walls at the same time thickening; they may also become somewhat convoluted. Their further fate in both sexes will be described in the section on the urogenital union.

The tubules of the paragenitalis suffer at first only interruptions in their course. These breaks allow of a separation of the individual parts of the tubules and thus the formation of two groups of tubules; the one group includes the Malpighian corpuscles and portions of the secretory tubules, the other the primary excretory duct and portions of the collecting tubules. Both tubule groups lie at first rather close together, but with the formation of the anterior abdominal wall they become widely separated. Simultaneously with the formation of the anterior abdominal wall the coiling of the intestines, and with this the enlargement of the abdominal cavity, begins. The urogenital fold, however, is attached to the anterior abdominal wall by the inguinal fold, and since it must, consequently, follow that wall, it becomes drawn out ventrally in the horizontal direction. Where the urogenital fold unites with the anterior abdominal wall, the primary excretory duct makes a double bend (Fig. 552), bending first around the caudal pole of the mesonephros and again as it enters the primitive true pelvis, and between the two bends there is a horizontal portion of the excretory duct; it is this horizontal portion that receives the collecting tubules of the paragenitalis. Along with the urogenital fold, this portion of the primary excretory duct is drawn out lengthwise and forms a loop directed towards the anterior abdominal wall. The formation of the loop, which the collecting tubules must follow, separates these latter from the secreting tubules and the Malpighian corpuscles. Both these structures retain their position immediately caudal to the epigenitalis and not infrequently are divided into two groups, a cranial one, that lies so closely upon the epigenitalis that it seems to consist of epigenital Malpighian corpuscles and secreting tubules, and a somewhat more caudal one. The corpuscles and tubules forming the first group usually vanish completely, but may sometimes persist. In the former case there is but a single paragenitalis, in the latter case two. The collecting tubules that are drawn out longitudinally by the looping of the excretory duct may unite with each other, so that one may become the efferent for several. The number of openings into the excretory duct may thus be reduced, and the reduction may be carried still further by the portions of the wall of the duct between the openings being employed in the elongation of the collecting tubules ; thus it may happen that eventually only one opening remains, and this may persist in the adult condition as the canaliculus aberrans Halleri; a preferable name would be the ductus collectivus paradidymidis.

In the male the organ of Griraldes (paradidymis) develops from the paragenitalis; it may persist throughout life and will then be found between the epididymis and the testis, slightly below the head of the epididymis. In the female the paradidymis becomes the paroophoron; the tubules of this occupy a remarkable position in older fetuses and children, caudal to the lateral half of the ovary (Eielander 1904), somewhat medial to the free edge of the broad ligament. How they acquire this position has not yet been explained. In exceptional cases the primary excretory duct persists in the female and it then runs from the epoophorou medially between the tube and the ovary to the uterus ; the paroophoron, if it retained its original position, should be found medial to the epoophoron and therefore cranial to the medial pole of the ovary. Wakleyer, indeed, believes that he has found remains of tubules in this position, but his observation has not been confirmed by any other author. Apparently the branches of the internal spermatic artery play a part in the displacement of the paroophoron, at least one finds the remains of it always internal to the first branch of this artery. The paroophoron quickly degenerates in extra-uterine life; Eielander (1904) was able to detect remains of it with certainty only up to the 5th year.

The further history of the primary excretory duct in both sexes will be described in the section on the urogenital union.

Development of the Metanephros

INTRODUCTION.

The metanephros may be divided embryologically, and to a certain extent physiologically also, into two portions, the secretory or actual glandular portion and the efferent apparatus. We find therefore in the metanephros the same two parts as in the mesonephros, where there were on the one hand the mesonephric tubules and on the other the primary excretory duct; in the metanephros, however, the two parts enter into much more intimate relations. The efferent apparatus in the last analysis develops from the primary excretory duct, which, by an outgrowth, forms on either side a ureter. This again gives rise to a series of tubules, the primary collecting tubules and each of these to secondary ones, these again to tertiary ones and so on, until finally there is formed an efferent apparatus consisting of thousands on thousands of collecting tubules. We therefore speak of a collecting tubule system, of collecting tubules of the 1st to the 12th order and of terminal collecting tubules.

The excretory apparatus develops from the nephrogenic cord ; just as this gave rise to mesonephric tubules in its anterior threefourths, so in its posterior portion it gives rise to metanephric tubules. The mesonephric tubules open into the primary excretory ducts, the metanephric tubules unite with the terminal collecting tubules. In order to determine accurately the limits of the two portions of the metanephros f omied from different sources, we may briefly trace the course of a fully formed metanephric tubule from the Malpighian corpuscle to the renal pelvis (Fig. 575), and in so doing we shall set forth the nomenclature to be employed in the following description. To the Malpighian corpuscle there succeeds the tubulus contortus (represented by stippling), which may be divided into two parts, a part that is actually convoluted and a pars recta ; this latter passes over into the narrow descending limb of Henle's loop and this again into the ascending limb. The broader limb may be divided into a darker and a lighter part, the darker part is cross-hatched and the lighter is left white. The lighter part of the ascending limb bends over the upper (peripheral) surface of the Malpighian corpuscle (collecting piece) and passes over into the intermediate tubule. Between this and the terminal collecting' tubule there is no definite boundary. From the terminal collecting tubule we pass by the collecting tubules of the 1st to the 12th orders to the renal pelvis and the ureter. From the Bowman's capsule of the Malpighian corpuscle to the intermediate tubule the tubule system develops from the metanephrogenic tissue; the development of the collecting tubule system up to the terminal collecting tubule has already been stated to be from the primary excretory duct. We must therefore look for the point of union of the two systems at the junction of the terminal collecting and intermediate tubules. The two systems therefore grow towards each other from a considerable distance, and the length of the path which they must traverse before meeting forms a complication of the developmental process, and the more complicated such a process is, the greater is the possibility for disturbances of it. The cystic kidney, so familiar to the pathologist, is in many cases due to a failure to find one another on the part of the uriniferous tubules and the terminal collecting tubules.

Fig. 575. — Diagram of a uriniferous tubule from the adult kidney, showing the relative positions of the various parts to one another and to the collecting tubule. The tubulus contortus consists of two portions, a pars convoluta, in which the Malpighian corpuscle lies excentrically, toward the renal pelvis, and a pars recta, which participates in the formation of the descending limb of Henle's loop. The summit of Henle's loop (in short loops), is formed by the wider ascending limb, which consists of a darker and a lighter portion; the lighter portion passes over into the connecting piece, which, crossing the vascular pole of the Malpighian corpuscle, passes over into the intermediate tubule and this opens into the terminal collecting tubule.

Development of the Ureter, the Primitive Renal Pelvis, and the Collecting Tubule System.

It has been stated in the section on the pronephros (p. 772) that the primary excretory duct on its way to the cloaca runs beneath the ectoderm as far as the 28th primitive segment (5th lumbar), and that there it bends to a horizontal direction and reaches the lateral wall of the cloaca. At this sagittal bend the dorsal wall of the duct forms a hemispherical outgrowth, the an] age of the ureter (Fig. 576). Even before the appearance of the ureter bud the duct is enlarged at the bend and slightly thickened throughout its entire circumference. This enlargement affects only the limited region of the bend, the portion of the duct between it and' the cloaca remaining narrow during the earlier

Fig. 576. — Reconstruction of the lower end of the body of a human embryo of 5.3 mm. greatest length, 4.6 mm. nape length, and with 36 pairs of primitive segments. (Embryo 1420, from the collection of Professor Keibel, Freiburg i. Br.) The primary excretory duct has come to lie between the visceral and the parietal roots of the a. umbilicalis. At the level of its opening into the bladder it bends at a right angle and at the bend it pushes out the ureter bud stages of the development of the ureter. The hemispherical ureteric anlage is not exactly in the middle of the dorsal wall of the primary duct, but inclines more and more towards its medial surface, the degree of inclination varying in different embryos. When the anlage later elongates to form a canal it always lies at the middle of the dorsal surface of primary duct and, later on, even on its lateral surface, yet in an embryo of 7.8 mm. greatest length it still opened distinctly on the medial surface.

This original medial position of the ureteric bud is of importance from the comparative standpoint, since it persists throughout life in the marsupials, and in these forms is an obstacle to the union of the two Miillerian ducts (Keibel, 1896).

The time at which the ureteric anlage appears cannot be definitely determined, various observations giving different suits, but it may with some assurance be placed between the stages of 4.5 mm. and 5.3 mm. total length.

An embryo of 4.25 mm. and with 28 pairs of primitive segments, another of 4.9 mm. vertex-breech length and with 33-35 pairs of segments and, finally, one of 5 mm. greatest length, showed no trace of an ureteric anlage, but, on the other hand, an embryo of 5.3 mm. greatest length, 4.6 mm. nape length, and with 36 pairs of segments, showed distinct ureter buds, that on the right having a length of 20 /* and that on the left, 30 M. An embryo of 4.75 mm. greatest length already had a right ureteric anlage SO M in length, and a left one of 50 M. The enlargement at the bend of the primary excretory duct had already appeared in an embryq of 4.25 mm. greatest length and with 2S pairs of primitive segments; in this embryo the primary excretory ducts had reached, but had not yet united with, the walls of the cloaca.

The short ureter bud grows at first almost altogether dorsally toward the vertebral column and the blood-vessels lying in front of it, but later (in embryos of 8.5-9.5 mm. greatest length) it forms a curve, which becomes gradually flatter with advancing age, and the ureter grows cranially. This ascent is due to actual growth of the ureter, since its tip presses against the surrounding structures. Thus the ureter gains the dorsal surface of the mesonephros lying in the retroperitoneum ; although only loose mesenchyme is present in this situation, nevertheless it presents an obstacle to the forward growth of the ureter that can be overcome only so long as the anlage has but a single tip. As soon as the radial outgrowths of the collecting tubules begin, an obstacle will be presented sufficient to prevent any further forward growth of the anlage; consequently the definitive position of the renal pelvis in the fetus (2nd lumbar vertebra) has already been reached in embryos of 9.5 mm.-13 mm. greatest length . An examination of an embryo of 9.5 mm. greatest length (Fig. 578) will give an idea of how slight the forward growth of the ureter really is.

The two ureters do not grow with equal rapidity, but it is sometimes the right and sometimes the left that is in advance, and it cannot be said that either side is the more favorable.

The attainment of its definitive position by the tip of the ureter completes its forward growth, but not its increase in length, for in order to maintain this position during the growth of the body, it must elongate enormously, since it is in the region of the lumbar vertebrae that the body growth principally takes place. The growth in length is the result of growth along the entire length of the ureter, at least mitoses can be found from the opening into the primary excretory duct to the renal pelvis.

At the time when the renal pelvis has acquired its definitive position, neither that of the cranial nor that of the caudal pole of the kidney has been established ; both are vet to be pushed cranially or caudally, as the case may be, by the outgrowing collecting tubules. In this way the cranial pole of the kidney has reached the suprarenal body in an embryo of 18 mm. greatest length and is growing forward dorsal to that organ (Fig. 553). The space behind the suprarenal body and behind the urogenital fold affords ample room for it, and is occupied by a mesenchyme containing numerous large cavities (Fig. 553) and so loose that there can hardly be any question of a hindrance to the ascent of the kidney from the suprarenal bodies, nor can the liver have any effect in this respect ; Fig. 553 shows that the latter does not come into contact with the metanephros. That it influences the postfetal development of the right kidney is not denied.

Fig. 577 a and 6. — Model of the bladder, metanephric anlage and the caudal end of the mesonephros of an embryo of 7 mm. greatest length, seen from the right side. (Embryo Chr. 1, from the collection of Professor Hochstetter, Vienna. The model prepared by my son Willy Felix.) X 52. From the caudal end of the primary excretory duct the ureteric anlage has developed. It is growing apparently directly backwards. A comparison with the orientation figure b (reduced from a drawing by Elze, 1907) shows that the anlage is growing almost horizontally in its segment. It is still elongated and ends in a club-like swelling, the primitive renal pelvis. Fig. 577 6 is represented by mistake from the left instead of from the right side.

The completely developed expansion of the ureter extends through the length of three or four vertebras ; from the upper border of the 12th thoracic to the lower border of the 4th lumbar is its maximal extent. The left kidney is already the higher as a rule in embryos of 22 mm. and onwards.

Even during the first stages of its growth the ureter changes its form; at first a simple hemisphere, it quickly develops a stalk and assumes the shape of a wedge, flattened from right to left (Fig. 577) ; later, the swollen end becomes marked off from the stalk, so that there is a differentiation of the primitive renal pelvis from the ureter in the narrower sense, the latter remaining narrow and having an almost circular transverse section. The primitive renal pelvis is broad in the cranio-caudal direction, narrow in the frontal one, and as seen in the model it has the form of a completely compressed funnel. It lies ventral to the v. cardinalis posterior. While the ureter is still in the period of outgrowth the primitive pelvis elongates in the cranio-caudal direction and thus acquires its poles, both of which begin on their own accounts to grow out in opposite directions and so begin the formation of the collecting tubules of the first order.

There are formed in all four collecting tubules of the first order, but their number may be increased to six or fall to three. The first two are formed by the elongation of the cranial and caudal poles of the primitive pelvis and I shall therefore name them the cranial and caudal pole tubules. Somewhat later horizontal tubules develop, corresponding to the middle of the renal pelvis, and since of these one is directed more ventrally and the other more dorsally, I shall term them the ventral and dorsal central tubules. These also appear as simple blind sacks and form with one another an angle of 45°. Rarely a doubling of the central tubules occurs and very rarely the formation of only a single one. The two pole tubules may show different relations to the v. cardinalis posterior; in an embryo of 12.5 mm. greatest length the pelvis and the caudal pole tubule were ventral to this vein, while the cranial pole tubule ran at first along its medial side and eventually reached its dorsal surface.

Fig. 578. — Model of the bladder, primary excretory duct, and ureter of an embryo of 9.5 mm. greatest length. (Embryo Ma. 3, from the collection of Professor Hochstetter, Vienna. Model prepared by my students J. Klausler and Wydler.) The common portion by which the primary excretory duct and the ureter open into the bladder is enlarged and is more sharply differentiated from the wall of the bladder. The ureter has differentiated into the primitive renal pelvis and the ureter s. s. The two pole tubules are beginning to grow out from the primitive pelvis.

The four tubules of the first order never appear simultaneously; the pole tubules are always the first to form.

As soon as the primary collecting tubules have reached a certain length they become enlarged at their blind ends to the so-called ampulla? (Figs. 579-581) ; the primitive renal pelvis is after all merely a primary ampulla of the ureter. Each ampulla has from the beginning the shape of a sphere flattened towards the future surface of the kidney, and in its further growth it elongates parallel to that surface and forms, according to the number of collecting tubules of a lower order to which it gives rise, either a simple transverse tubule, or a three-sided or a four-sided prism (Fig. 579). From the two to four angles of the ampulla the secondary collecting tubules grow out parallel to the future surface of the kidney, and they also are at first simple blind sacks. When they have reached a certain length, they also develop ampullae, from two to four sides of which tertiary collecting tubules are formed. And so the process goes on up to the formation of the terminal collecting tubules, and the ampullae always persist, so that one can determine in young kidneys from their number, counting from the pelvis to the surface of the kidney, the number of the budding centres, even although one may not see the branches that are given off. All the tubules, from those of the second and third order onwards, are so arranged that they run as well in the dorsal as in the ventral direction, two usually running ventrally and one dorsally. The formation of new collecting tubules ceases in the fifth fetal month, and the terminal collecting tubules are of the 11th to the 13th order; I may state, however, that these numbers are estimated, not observed. The primary collecting tubules are formed in two stages and in the formation of the secondaries still more occur. In Fig. 580 the caudal pole tubule has already developed three secondary tubules; the most medial of these is provided with three buds, representing tertiary tubules, the most lateral has developed an ampulla which shows preparations for the formation of the new generation, and only the middle one still preserves the form of a simple blind sack. Of the four central tubules only one is in the ampulla stage, and the cranial tubule does not yet show further development. Fig. 581 shows already a distinct ureteric tree; here there is only one central tubule, which is just forming an ampulla; the cranial and caudal pole tubules, on the other hand, are already far advanced in their development. The individual tubules are labelled P or Z according as they arise from a pole or a central tubule and the figure gives the number of their generation. The cranial pole tubule (P x ) possesses three secondary tubules (P 2 ), the medial of which is the furthest developed, then the lateral one and least of all the middle one. If the medial secondary tubule be followed further, it will be seen to have developed tertiary tubules (P 3 ), and of these also

Fig. 579 a, b, c, and d. — Diagrams representing the form of the ampullae of collecting tubules giving rise to 2, 3, or 4 tubules of a new generation.

Fig. 5S0. — Ureteric tree of a human embryo of 12.5 mm. greatest length. Reconstruction of transverse sections. (Embryo Ma. 1, from the collection of Professor Hochstetter, Vienna.) X 100. From the primitive renal pelvis six primary collecting tubules have budded out, a cranial and a caudal pole tubule and four central tubules, two ventral and two dorsal. The development of the tubules proceeds in the caudo-cranial direction, the caudal pole tubule has already three secondary tubules, the caudal dorsal central tubule is just dividing, the cranial dorsal one is still pointed, the caudal ventral one is further developed than the cranial ventral one; the cranial pole tubule is still undivided. Of the three secondary tubules from the caudal pole tubule the caudal one has already developed three tertiaries, the cranial is just about to give rise to others, and the middle one has not yet formed an ampulla. Of the primary tubules the cranial pole tubule is a continuation of the ureter or renal pelvis, and the central and caudal pole tubules arise at right angles to the renal pelvis. The cranial secondary tubule is the prolongation of the primary caudal one, the caudal secondary has budded off at right angles. By these different relations of the primary and secondary tubules there is produced the first group-like expansion of the ureteric tree. The dorsal and ventral central tubules form an angle of 45° with one another.

the medial one is the furthest developed, for it has pushed out two quaternary buds and these have already their ampullae so far developed that one may determine from their shape the number of tubules of the fifth order (P 5 ) that are to be formed; the lateral tertiary tubule has formed two quaternaries, only one of which is seen in the figure; the middle one (shaded in the figure) is still a simple blind sack. The lateral secondary collecting tubule has only produced two tertiaries (P 3 ). and the most medial of these is again further developed. The same conditions obtain in the caudal tubules; in all cases the two outer of three tubules of the same generation are more developed than the middle one, and of the two outer ones the medial is more developed than the lateral. The only difference between the caudal and cranial pole tubules is that the former has already formed six generations, while the cranial one is just beginning to form a fifth. The constancy of these results, which are always obtained, leads to the following laws of development for the collecting tubules: 1. The two pole tubules precede the central ones in development. 2. Each of the two pole tubules usually develops three secondary tubules and of these the most medial always precedes the other two in its development and only in its successive branchings is there a formation of four tubules in a generation, the maximum number ; after it comes the lateral branch, which also shows a rich development of branches, and the middle branch, finally, remains behind the others. This process is repeated in the formation of each new generation and must, therefore, lead to a differentiation between the various branches and twigs of the ureteric tree. Strong branches will be found to produce a rich crop of branches, moderate ones a less rich crop and, finally, weak ones only a poorer crop. The reason for this difference of development lies in the adaptation to abundant or scanty space. Since the tubules of the second order are at first developed horizontally, they must grow out from the various primary tubules toward each other. So long as there are only 4 primary tubules with in all 10-12 secondary ones, space is not yet limited, but as soon as the 30-36 tertiary tubules bud out and require room, the space becomes insufficient and this insufficiency will increase with the addition of each generation. Some relief from the insufficiency is afforded by the fact that the entire ureteric tree spreads in a fan-like manner, that is to say, it curves round the renal pelvis as a centre. (In Fig. 581 the direction of the curvature is shown by the arrows.) At first it surrounds only about two-fifths of an oval, but later more, until, finally, the form of the adult kidney is acquired. And now the various stem branches with their branch areas become arranged like the branches of an actual tree, those branches which lie to the outside having room for growth and those at the centre being prevented from growing, so that one finds the branching abundant at the periphery and scanty at the centre. Exactly so is it with the ureteric tree. Those collecting tubules which lie at the sides are favorably situated for development, these are the pole tubules; those situated in the centre are prevented from growing, these are the central tubules. Furthermore the tubules which occur in the centre of a branch area will not only form fewer branches of a younger order, owing to lack of room, but must also wait, before they can develop them, until by growth in length they have reached a region where there is sufficient room; these regions naturally can only occur towards the periphery. These tubules will, therefore, become longer than their favored fellows, but will develop much fewer budding centres, i.e., ampullar, in the same distance.

Fig. 581. — Model of the right ureteric tree of an embryo of 19.4 mm. vertex-breech length, seen from in front. (Embryo Ma. 2, from the collection of Professor Hochstetter, Vienna.) In this ureter there are two pole tubules and only one central tubule. The latter is greatly retarded in its development. The letters P and Z denote the pole and central tubules, the numbers accompanying the letters indicate the order of the tubules concerned. The caudal pole tubule is further developed than the cranial one, having produced tubules of the sixth order, while the cranial one is just producing those of the fifth order. If one traces out the various pole tubules one finds that the advancement of the formation of new tubules varies on the individual branches. If we regard the whole kidney as a half-moon, those branches of the pole tubules are most developed that are nearest the horns. This agrees with the general development; the ureteric tree curves medially around the renal pelvis in the direction of the two arrows. From the position of the tubules of the third and fourth orders one sees that already the growth in thickness in the dorso-ventral direction has begun.

Since there are four primary collecting tubules, a special ureteric tree (primary tree, collecting tubule area of the 1st order) will be formed over each one ; from what has been said the trees over the pole tubules will be large and those over the central tubules small; moreover the primary trees must influence each other and adapt themselves to the space available for the kidney, so that each tree in its entirety will have a conical shape, its tip being towards the kidney pelvis and its base towards the kidney surface. Within the limits of each of the primary trees the principal branches must adapt themselves to this conical shape, and, consequently, the tubules of the 5th, 6th, etc., orders, that were budded off at right angles to the parent tubule, become more and more drawn out and thereby so crowded together that finally they run one close alongside the other as medullary rays radiating to the surface of the kidney. The tubules of the 2nd-4th order are not affected in this way, since they become enlarged and are later taken up into the kidney pelvis (see p. 859, the definitive renal pelvis and the reduction of the collecting tubules). The tubules of the 8th, and later those of the higher orders, retain partly their right-angled origin from the parent tubules, especially those that lie at the periphery of the branching of a tubule of the 3rd or 4th order (collecting tubule area of the 3rd or 4th order). By this right-angled origin the peripheral tubules of the 8th, etc., orders are pressed upon by the neighboring tubule areas of the 3rd or 4th order, so that they grow towards each other and force one another back; thereby the tubules of the 5th order (those of the third and fourth orders do not come into consideration on account of the reduction process mentioned above) hitherto lying massed together, are separated into groups, and each group with its branches forms a secondary tree. The stem of each secondary tree later becomes a secondary Malpighian pyramid. Within the individual secondary trees, at least within the pole pyramids, the formation of tertiary trees may occur.

The tubules of the first or second order, whose growth is prevented and which must grow far towards the surface before they can develop to a tree, must produce cones which, so far as their size is concerned, cannot be distinguished from secondary cones formed by tubules whose growth was not prevented. The branching of the central tubules presents enormous variability, for kidneys occur in which they almost keep pace with the pole tubules in the development of younger generations, naturally always remaining somewhat behind them, and, on the other hand, others occur, as for example that shown in Fig. 581, in which the central tubules have produced no second generation, while the pole tubules are always engaged in the formation of the fifth and sixth.

Detailed statements as to the branching of the individual primary collecting tubules will be possible only when numerous models have been prepared, since these alone allow of a certain count. Since such models are not yet available in sufficient numbers, I shall limit myself to a statement as to the highest number of orders of completed tubules that occurs in the branching of one of the primary tubules, without, however, wishing to imply that all the branches of this tubule give rise to as many orders. An embryo of 10 mm. greatest length had the two pole tubules in the process of formation, one of 11.5 mm. greatest length showed anlagen of the central tubules in addition to rather long pole tubules. One of 12.5 mm. greatest length possessed two pole tubules and four central tubules, the caudal pole tubule with tubules of the third order, the cranial one without branches, and the central ones in the act of forming ampullae. An embryo of 13 mm. had two pole tubules and one central tubule, the caudal pole tubule had quaternary and the cranial tertiary branches, while the central one had developed secondaries. An embryo of 17 mm. had two pole and two central tubules; both pole tubules had developed tubules of the 4th order and the central ones those of the third order. In an embryo of 18 mm. the pole tubules had tubules of the 4th order and the central tubules those of the 3rd order. In an embryo of 19.4 mm. greatest length the cranial pole tubule had tubules of the 5th order, the caudal pole tubule those of the sixth order, while the central tubules were branchless. In an embryo of 21 mm. the pole tubules possessed tubules of the 5th-6th order, the central tubules those of the 3rd order. In an embryo of 30 mm. greatest length (Hauch, 1903), collecting tubules of the 6th order were developed, in an embryo of 60 mm. head-foot length those of the 8th order, and in embryos of 65-70 mm. trunk length (Hauch, 1903), tubules of the 8th-llth orders.

The Metanephrogenic Tissue during the Development of the Collecting Tubules.

In the section on the parent tissue of the mesonephric tubules it was shown that during the segmentation a cord was cut off from the mesoderm between the primitive segments and the lateral plates, this cord representing the combined primitive segment stalks. This cord of nephrogenic tissue extends — this can be determined in all clearness in human embryos — only to the point where the primary excretory duct bends in to the cloaca, that is to say, to the 28th primitive segment (5th lumbar segment) ; throughout its entire length it lies on the medial side of the primary excretory duct. In an embryo of 5.3 mm. greatest length, 4.6 mm. nape length, and with 36 primitive segments it was interrupted in the region of the 26th and 27th primitive segments and thus separated into the long mesonephrogenic and the short metanephrogenic cord. It is only for didactic reasons that one speaks of a metanephrogenic cord in man, it is really only a small roundish heap of cells. The division of the nephrogenic cord coincides with the formation of the ureter bud ; after the division the metanephrogenic mass of cells lies at first on the medial side of the ureter anlage — as is natural from its genesis, — but in an embryo of 7 mm. greatest length it has grown around it and covers it like a closely fitting cap (Fig. 582). On account of this intimate relation with the ureter the further development of the metanephrogenic tissue is closely associated with the fate of the ureter and its branchings. At first the metanephrogenic cap is forced dorsally by the growing ureter bud and thereby is carried out of the line of the mesonephrogenic cord, and it appears from now on as a special structure that has no longer any relation to the mesonephrogenic cord ; nevertheless an embryo of 7 mm. greatest length still shows a denser strip of mesenchyme tissue extending from the cap toward the mesonephros, i.e., cranially and ventrally, and ending at a distance from the last mesonephric tubule equal to about the diameter of two of these tubules. Then the metanephrogenic tissue accompanies the ureter in its ascent, always covering only the primitive renal pelvis. The cap, which it forms, is at first rather thin (Fig. 582), and it becomes still more so on the enlargement of the ureter bud to form the primitive renal pelvis ; when now the budding out of the primary and secondary collecting tubules occurs the cap becomes so stretched that in an embryo of 9.5 mm. greatest length it becomes torn into as many pieces as there are collecting tubules of the first order present. Each of these four to six pieces of the cap shows, however, rapid cell multiplication and again grows completely over the ureteric tree to which it belongs. The various ureteric trees are accordingly covered by metanephrogenic tissue not only on their surfaces but also on the sides turned towards neighboring trees ; each ureteric tree and its metanephrogenic cap may be regarded as a unit and I shall term it a primary Malpighian pyramid. The various pyramids are closely pressed together and so their metanephrogenic caps are apparently united again to form a single investment over the entire kidney. The various pyramids will naturally reproduce the form of the corresponding ureteric trees, and consequently possess a base directed towards the surface of the kidney and correspondingly curved, and an apex directed towards the renal pelvis. The more the ureteric trees develop and curve medially around the renal pelvis (Fig. 581), the more pronounced the lateral convex and the median concave surfaces of the kidney become, and, finally, the bases of all the pyramids (Fig. 583) form a shell with an opening directed medially for the entrance of the ureter; the renal pelvis and the ureteric tree accordingly lie in the space within the shell. The Cranial pole pyramid space within the metanephrogenic shell unoccupied by the various ureteric trees may be termed the primitive renal sinus. The lines along which the metanephrogenic tissue passes down from the surface between the various pyramids are marked by grooves, which limit the area occupied by each pyramid, so that one may speak of a lobation of the kidney surface (Figs. 583 and 584). Beneath each groove and extending as far as the renal sinus small quantities of connective tissue lie between the metanephrogenic investments of adjacent pyramids, and these strips of connective tissue, together with the two accompanying layers of metanephrogenic tissue, form what are termed the columns of Bertini (Fig. 583). There will accordingly be at first four (three to six) pyramids — the two pole pyramids corresponding to the two pole tubules, and two (one to four) central pyramids, corresponding to the two (one to four) central tubules — and three columns of Bertini, a cranial one between the cranial pole pyramid and a central pyramid, one between the two central pyramids, and a caudal one between a central and the caudal pole pyramid (Fig. 583).

Fig. 582. — From a transverse section through a human embryo of the fifth week, after Schreiner (1902). X 120. The section cuts the primary excretory duct transversely; the ureter, which has already separated into the primary renal pelvis and the ureter s. s., is cut lengthwise. The metanephrogenic tissue lies like a cap around the primitive renal pelvis.

Fig. 583. — Frontal section of the kidney of a human fetus of 160 mm., S% months old, after Hauch (1903). X8. Three primary collecting tubule systems are to be seen in the figure, those of the two pole tubules and those of the central one. Around each of these systems is situated a cap of metanephrogenic tissue, each system with its cap forming a renal pyramid. Where the two pole pyramids abut upon the central one the neogenic zones of the two metanephrogenic caps come together to form a column of Bertini. This extends to the renal pelvis.

Fig. 584. — Right kidney of a five months' human fetus seen from the ventral surface. X about 3}4. The relatively large suprarenal body is seated upon the kidney, whose surface is distinctly lobed.

The columns of Bertini appear one after the other in the 8th to the 10th week, and are distinctly formed in the 12th week.

It has already been pointed out that each ureteric tree eventually divides into several parts and the various metanephrogenic caps take part in these divisions. They divide into smaller portions and each of these tertiary pieces covers and encloses a minor ureteric tree just as completely as the secondary caps enclosed the undivided trees; thus secondary pyramids are formed within the primary ones (Fig. 585). Exactly as the metanephrogenic tissue sinks in between the primary pyramids to form a septum, so it also separates the secondary pyramids, and one may, therefore, speak of secondary columns of Bertini (Fig. 585). A finer lobation of the surface corresponds to these secondary pyramids, since each principal lobe is divided into subordinate lobes. Fig. 584 shows the extensive lobation of the surface of the kidney of a five months' fetus. The lobation of the kidney persists until birth and disappears in the early years, although under some circumstances it continues to persist throughout life.

Fig. 585. — Frontal section through the kidney of a human fetus of the nineteenth week, 175 mm. in length, after Hauch (1903). The primary pyramids have been divided into secondary pyramids by the appearance of secondary columns of Bertini.

According to the number of collecting tubules of the second order that bud from those of the first order, from two to four secondary pyramids occur in the interior of each primary one. Within the secondary pyramids, at least within the pole pyramids, other tertiary pyramids, three in number, develop and with them of course the columns of Bertini that separate them. All the tertiary pyramids reach the surface, but the columns separating them penetrate inwards only a short distance. The tertiary pyramids therefore produce a distinct, finer lobation of the surface, but an incomplete separation of the pyramids. The variability that is to be seen in the lobation of the kidney surface — rich or scanty lobation — depends on the variable development of the secondary arid tertiary pyramids.

The Development of the Uriniferous Tubules from the Meta= nephrogenic Tissue.

The metanephrogenic tissue is divided by the progressive outgrowth of the collecting tubule systems into separate portions, corresponding not only to the primary, secondary and tertiary Malpighian pyramids, but also, within all these pyramids, into as many separate masses as there are peripheral collecting tubules. The primary collecting tubules raise the metanephrogenic tissue as a whole away from the renal pelvis. As soon as the tubules of the second order grow out the mantle is split into four pieces, and again the tubules of the second order raise their quarters of the mautle as a whole and push it towards the periphery; when the tubules of the third order form, each mantle quarter is again divided and the portions are again raised up and again separated by the new generation. As a result around the end of each collecting tubule, which, in the stage under consideration, represents the temporary terminal tube, there is a cap of metanephrogenic tissue. When the new collecting tubules grow out the caps are again torn into from two to four parts according to the number of the tubules. This stage is shown in Fig. 586, in which we see a tubule that has grown to the periphery of the kidney and is there forming two new tubules, both of which carry in front of their pointed blind ends the metanephrogenic tissue. This surrounds only a small portion of the young tubule on the side nearest the surface of the kidney, but on the side towards the pelvis it surrounds it throughout its entire length, as far as its origin from the older tubule. In the angle between the old and the young tube the metanephrogenic tissue condenses to form a compressed, spherical cell-mass, whose constituent cells are arranged around a point situated somewhat excentrically in the interior of the mass. A single cell mass is formed for each young tubule, so that around the older tubules usually three cell-masses are arranged. The concentric arrangement of the cells very soon renders the cell-mass independent of the rest of the metanephrogenic tissue and a sphere of cells, a metanephric sphere (Fig. 587, on the right side above), is formed. This is now completely isolated and has no connection whatever with any of the neighboring structures; its position in the angle between the older and younger tubule is typical and is repeated with absolute regularity in all anlagen, whether they are the first or last to form. A lumen situated excentrically appears (Fig. 586) and converts the sphere into a metanephric vesicle, whose walls vary in thickness at different parts owing to the Surface of kidney Young collecting tubule Metanephrogenic cap !

Anlage of uriniferous tubule Old collecting tubule Metanephric vesicle

Fig. 586. — Portion of a section through the kidney of a human embryo of 30 mm. vertex-breech length. (From the collection of Professor Stoerck, Vienna; II, 3, 4.) X 300. An older collecting tubule with two young ones budding out from it is shown. Each young tubule carries in front of it a cap of metanephrogenic tissue, which surrounded the end of the older tubule, and tears it. In the angles between the older and each younger tubule there develops a metanephric vesicle from the metanephrogenic tissue. The left vesicle shows a slight growth directed towards the periphery, the anlage of a unriniferous tubule.

excentricity of the lumen. The wall towards the surface of the kidney — for brevity I shall term it the peripheral wall — is the thickest and from it the entire uriniferous tubule, with the exception of Bowman's capsule, is formed; the wall towards the renal pelvis — I shall term it the central wall — is thinner and it forms Bowman's capsule; the side towards the older collecting tubule I shall call medial and that directed away from the tubule will be spoken of as lateral. The two metanephric vesicles of Fig. 586 are unequally developed. In the left one the upper wall is drawn out to a knob, which is the anlage of the actual uriniferous tubule, and rising perpendicularly upwards it unites with the younger collecting tubule (Fig. 587, upper tubule). By this the uriniferous tubule becomes connected with the collecting tubule system, its lumen is continuous with that of the collecting tubule. The upper tubule of Fig. 587 shows a further progress in development; on its lateral side a slight groove has formed, which is so placed that it separates the former vesicle from the newly-formed uriniferous tubule; it separates, in fact, the anlage of Bowman's capsule from the uriniferous tubule, and its central wall represents the anlage of the parietal layer of the capsule. In order that an idea of the actual appearance of the anlage of the uriniferous tubule may be obtained the left side of Fig. 588 should be studied ; the anlage appears as a sack attached to the young collecting tubule. At the moment when the union of the uriniferous and collecting tubules is established, the former begins to bend, and the various new changes may be followed in Fig. 589 a and b. First it will be observed that the groove that separates the uriniferous tubule and Bowman's capsule and was only indicated in Fig. 587, has now extended to the medial side of the vesicle, and the delimitation of Bowman's capsule is thus completed; its parietal and visceral epithelia are formed, the parietal being simple and already completely flattened, the visceral being two-layered and cubical; the shape of the capsule is that of a bowl with a doubled wall (Fig. 588. right tubule; Fig. 590). Since the groove penetrates to the medial side of the capsule, the passage of the lumen of the latter into that of the tubule is forced entirely to the medial side, whereas formerly it lay at the middle. From this point the tubule makes a triple bend into the shape of an S, the lower limb of the S extending from Bowman's capsule to the single asterisk (Fig. 589 b), the middle piece from this to the double asterisk, and the upper limb from this to the ampulla of the collecting tubule. In these three limbs all parts of the future uriniferous tubules are laid down and we are now in position to determine the portion of the tubule from which, for example, the tubulus contortus or Henle's loop arises. This we are able to do because one topographic relation persists unchanged throughout all the future modifications, namely, the situation of the apex of the middle limb in the concavity of Bowman's capsule; it is retained in this position by the capillary network arising from the vas efferens. In Fig. 591 the future parts are indicated by different kinds of shading; the apex of the middle limb is left unshaded and it becomes the connecting piece of the adult uriniferous tubule (compare Fig. 575 in which the connecting piece is also unshaded). The left part of the middle limb and the upper limb become the intermediate piece and both are simply shaded in Figs. 591 and 575. The right part of the middle limb (cross-hatched) becomes Stoerck's or Henle's loop, and the lower limb (stippled) becomes the tubulus contortus.

Fig. 587. — Portion of a section through the kidney of a human embryo of 30 mm. vertex-breech length. (From the collection of Professor Stoerck, Vienna; II. 1, 7.) X 300. The anlagen of two uriniferous tubules are cut by the section. They are in different stages of development, belong to different generations, and therefore he at different levels. The left upper anlage has developed a uriniferous tubule that has connected with the collecting tubule. It has a club-shaped form, the handle of the club being represented by the uriniferous tubule and its head by the anlage of Bowman's capsule. A slight groove begins to grow in on the left side, marking the limits between the tubule and the capsule. The lower right tubule is just about to form Stoerck's loop. Its parts are not continuous in the section. Below is Bowman's capsule, still shell-shaped; then comes a portion curved like the letter S, and from the upper limb of the S a convoluted tubule arises; the middle limb is the parent tissue for Stoerck's loop and the lower one becomes the connecting piece. Above the S is the intermediate piece cut transversely.

Fig. 58S. — Model of two developing uriniferous tubules of a human kidney, after Stoerck (1904). X400. The left tubule hangs from its collecting tubule as a plump sack; the right has differentiated somewhat further; the S-shaped uriniferous tubule is followed by the shell-shaped Bowman's capsule.

Fig. 5S9 a and b. — Anlage of a uriniferous tubule of a human embryo of 130 mm. (Embryo 217, from the collection of Professor Stoerck, Vienna; I, 6, 7.) The anlage of the tubule has bent to an S-shape, the limbs of the S being shown in the diagrammatic complementary figure 589 b. It is important that the tip of the middle limb comes to lie exactly in the concavity of the shell-shaped Bowman's capsule. It is held in this position later on by the capillaries developed from the vas efferens. The glomerulus is in the very earliest stages of development, but already contains blood-corpuscles.

Fig. 590. — Model of a developing uriniferous tubule, after Stoerck (1904). X400. One sees very well in this view the shell-like form of Bowman's capsule and the tip of the middle limb of the S-shaped tubule situated in its concavity.

Fig. 591. — Diagram of a young uriniferous tubule in which the various portions are distinguished according to their fate. The mode of distinguishing them (stippling, hatching, etc.) is the same as in Fig. 575 a. One sees that from the apex of the middle limb of the S-shaped tubule, which is retained in position by the vessels passing from the vas efferens, the future connecting piece is formed.

The formation of the S-shaped bends and the transformation of the various limbs into definite portions of the adult tubule are almost perfectly regular and are repeated with wearying regularity in thousands of tubules.

The formation of the S immediately succeeds the coiling of the tubule and the lower limb and the medial part of the middle limb of the S take part in the coiling, that is to say, those parts from which the tubulus contortus and Henle's loop are formed.

Fig. 592. — Model of a developing uriniferous tubule from a human kidney, after Stoerck (1904). X 400. One sees Stoerck's loop slipping out of the coil. Only the horizontal limb of the leader points to this.

As soon as the coil is formed the loop slips out of it; since this is formed immediately above the Malpighian capsule (Fig. 591), it must grow out iu au arch arouud this (Fig. 593/ left tubule) ; theu having a free path towards the renal pelvis it grows to a notable length (Fig. 593, right tubule, Fig. 594 a). During the formation of this loop, which is not Henle's loop — I must insist on this here to avoid errors — a histological differentiation of the epithelium occurs in the tubule, affecting only all parts of the tubulus contortus. The size of its lumen increases, its cells become large and the cell boundaries vanish completely in some places, while in others they become at least indistinct. The nuclei are round and pale and have retreated to the basal portions of the cells, the protoplasm is distinctly granular. All this leads to a very great thickening of the tubulus contortus, while the remaining portions remain thin as before. The tubulus contortus, stippled in Fig. 594 b, now forms an ascending, a horizontal and a descending portion. The ascending and horizontal portions later become the convoluted portion (compare Fig. 594 6 and Fig. 595), the descending one forms the straight part and, at this stage of development, almost the whole descending limb of the loop. The histological differentiation and the twisting of the tubulus contortus occur simultaneously in all the tubules of the same generation. Fig. 595 shows two adjacent tubules belonging to different collecting tubule systems and yet showing an almost ridiculous likeness even in the finest details. They do not lie in the same plane, so that in one the Malpighian capsule does not appear in the section,

Fig. 593. — Model of two developing uriniferous tubules from a human kidney, after Stoerck (1904). X 400. The left tubule is just forming Stoerck's loop. One sees its long intermediate piece, that is beginning to coil somewhat. The right tubule already possesses a rather long Stoerck's loop.

Fig. 594 a and 6. — Section through a uriniferous tubule with a Stoerck's loop. (Human embryo of 105 mm., from the collection of Professor Stoerck, Vienna.) X 300. Fig. 594 6 explains the continuity of the various parts, which are again indicated as in Figs. 575 a and 591. Stoerck's loop consists of an ascending and a descending limb. The descending one is formed by the straight part of the tubulus contortus and a portion of the growth zone, that is left unshaded in Fig. 594 b, and from which the descending limb of Henie's loop later develops. The ascending limb of Stoerck's loop is thinner and darker than the descending one.

After birth a considerable increase in length and thickness occurs in the tubulus contortus, and the increase in length leads to an enlargement of the coil. The enlargement may best be shown by determining the number of Malpighian corpuscles in equal areas taken from kidneys of different ages; the more the coils increase in mass the more must the corpuscles be separated from one another. According to the observations of Kiilz (1899) the cortical region of the kidney of a new-born child contains five times as many glomeruli as a corresponding area of the cortex of an adult kidney. As regards the increase in thickness Kiilz (1899) determined that the tnbuli contorti of the adult were twice as great in diameter as those of embryos. The increase in length and thickness of the tubnli contorti is also shown by the formation of the so-called cortex corticis; the young corpuscles during the growth of the kidney lie close beneath the kidney capsule. This position they retain even after birth up to about the tenth day. At that time the tubnli contorti of the last generation are beginning to grow out and thus form the outermost layer of the renal cortex, which no longer contains Malpighian corpuscles.

Fig. 595 a and b. — Section through three developing uriniferous tubules of a human embryo of 45 mm. nape length, from the collection of Professor Stoerck, Vienna; 4.8. On the left side the section cuts two tubules of the same stage of development, but belonging to different collecting tubules; on the right side is a younger stage of development. The section shows the markedly intimate arrangement of the various portions of the tubules that all the tubules of this kidney present. 1, Bowman's capsule; 2, descending limb of Stoerck's loop; 3, horizontal part of the tubulus contortus; 4, ascending part of the tubulus contortus; 5, ascending limb of Stoerck's loop; 6, connecting piece; 7, intermediate piece.

As a result of the extensive thickening of the tubulus contortus part and the formation of at least the proximal half of the descending limb of the loop by the straight part of the tubulus contortus, and further as a result of the absence of change in all the remaining portions of the tubule, that is to say, their retention of their original slenderness, the form of the loop seems to be almost the reverse of what it is in the adult (Fig. 594 a, b and Fig. 575) ; the descending limb is the thick one and the ascending limb the thin one. We have therefore to do with a specific embryonic form of the loop and in order that we may have a definite term for it, I shall name it Stoerck's loop after its illustrious discoverer (Stoerck 1904). This loop is accordingly formed of the disproportionately long pars recta of the tubulus contortus and the actual Henle's loop, which, however, is not yet histologically differentiated in its two limbs. It is difficult to determine exactly how Stoerck's loop becomes converted into Henle's loop. I shall first call attention to the distal portion of the descending limb of Stoerck's loop shown in Fig. 594 a. It possesses a peculiar epithelium that corresponds

Fig. 596. — Short loop of Henle from an adult kidney, resembling exactly a Stoerck's loop as far as the broad ascending limb. Somewhat simplified from Peter (1909.) neither with that of the pars recta of the tubulus contortus nor with that of the ascending limb of the loop; in older stages of development it is frequently in active division and in such cases resembles a proliferation zone. I am, accordingly, inclined, although I cannot make a positive assertion, to regard this region as the embryonic homologue of the descending limb of Henle's loop. I am strengthened in this view by the results obtained from the study of adult Henle's loops of different lengths. Peter (1909) has shown that the length of a Henle loop is chiefly dependent upon the development in length of the descending limb ; in the formation of a long loop both limbs of course take part, the descending one, however, in the proportion of 1: 15 and the ascending one only in the proportion of 1:2; in the short loops the apex is formed by the ascending limb, in the long ones by the descending. Accordingly it is the greater or less growth of the thin descending limb that determines the length of the entire loop. In the short loops of the adult kidney that scarcely reach the outer zone of the medulla we have actual embryonic, that is to say Stoerck's, loops (Fig. 596), the descending limb remaining unaltered and the straight portion of the tubulus contortus forming the main portion of the loop ; the actual descending limb corresponds to the proliferation zone of Stoerck's loop, while the ascending limb has become thicker and has differentiated into its dark and clear portions.

The differentiation of Henle's loops into a thin descending and a thick ascending limb takes place, according to Toldt (1874), in the fourth fetal month; in the 5th month the long loops reach the papilla? in embryos of 140-175 mm. (Hauch, 1903).

At its first development Stoerck's loop is always so placed that the descending limb is the farther away from the collecting tubule and therefore has a medial course, while the ascending limb is nearer to the collecting tubule and therefore runs laterally. In the stage shown in Fig. 595 b one limb becomes twisted around the other in a half spiral, so that each of the limbs in its course is sometimes lateral and sometimes medial. The loop may remain in this position throughout life, but in most cases there is a complete twisting and Peter (1909) is correct in naming the descending limb of the adult kidney the lateral one and the ascending the medial.

The intermediate piece also suffers a change in position. From its mode of development (see Fig. 591) it must be the most lateral of all the portions of the tubule. Originally all portions . lie in one plane, but as soon as the tubulus contortus begins to coil and the loop slips out from it, a twisting of the tubule occurs, in such a way that a plane perpendicular to the surface of the kidney and passing through the intermediate piece is perpendicular to a similar plane passing through the tubulus contortus and Stoerck's loop ; I have endeavored to show this in Fig. 594 b. As a result of this twisting the intermediate piece obtains room for further development and, when it begins to coil, some of its turns may come to lie medial to the Malpighian corpuscle. The growth in length of the tubulus contortus naturally determines also a corresponding growth of the intermediate piece. I shall consider further on the peculiar growth of the intermediate piece produced by the transference of the opening of a collecting tubule belonging to a lower order to one of a higher order.

The further development of the Malpighian corpuscle consists in its original shell-like form (Fig. 590 and 589 a) becoming gradually spherical. Three stages of the development are represented in Fig. 597. In a the corpuscle has its simple shell-like form, in b there appears at the point where the tubulus contortus becomes continuous with Bowman's capsule — at the point marked by a cross in the figure — an evagination which tends to diminish the wide opening into the capsule, and in c this evagination has increased in size and at the same time the lateral wall of the shell has grown toward it, so that the wide entrance into the shell-like concavity has become converted into a narrow opening, the vascular pole ; through it the vas afferens and the vas efferens find ingress or egress. These modifications of Bowman's capsule cause also a displacement of its connection with the tubulus contortus, its urinary pole, and always in such a way that it comes to lie at the lower surface of the capsule ; thus the vascular and urinary poles of the Malpighian corpuscles come to lie opposite each other. It must be possible under certain conditions, that the two poles wander in the opposite direction, at least Peter (1909) has seen, though only rarely, the urinary pole at the peripheral surface of the capsule ; the curving of the renculi may well bring it about that individual tubules acquire unusual relations and so form departures from the usual plan. The table by Kiilz will best show the growth of the Malpighian corpuscles ; since the various uriniferous tubules are arranged in layers one over the other, the more central ones are the younger and the more peripheral the older, and therefore in the table representatives from an older are combined with those from a younger group.

Fig. 597 a, b, and c. — Three diagrams showing the development of a Malpighian corpuscle.

Table showing the Mid-size of the Peripheral and Central Glomeruli in Mikra. (After Kiilz, 1899.)

From this table it will be seen that at birth the central Malpighian corpuscles are almost IV2 times as large as the peripheral ones, and that in the course of the second year of life this inequality disappears by the peripheral corpuscles growing while the central ones remain unchanged. When, at the close of the second year, all the corpuscles have attained the same size, the central ones begin to grow as rapidly as the peripheral and all the corpuscles double their diameter. Finally, it will be seen that the growth of the corpuscles lasts until the period of puberty.

The first inetanephric spheres were found in embryos between 13 and 19.4 mm. greatest length. The first S-shaped tubules were shown by an embryo of 19 mm. greatest length, the first anlage of a Malpighian corpuscle — i. e., of Bowman's capsule and a glomerulus — was seen in an embryo of 19.1 mm. greatest length, while in an embryo of 28 mm. greatest length the first completely developed Malpighian corpuscles were found. The first histological differentiation — i. e., the appearance of secretory epithelium in the tubulus contortus — was shown by an embryo of 24 mm. greatest length, the first Stoerck's loops grow out in embryos of from 24-30 mm. greatest length. In connection with these data it is to be stated that each advance in the further development of the urinary tubules first occurs in the pole pyramids.

The metanephrogenic tissue, the parent tissue of the uriniferous tubules, lies, as has been stated above, as a cortical layer around the collecting tubules that have grown farthest towards the surface of the kidney. Around the ampullae of each collecting tubule the anlagen of the first uriniferous tubules group themselves in correspondence with the outgrowing collecting tubules of a new order, which are usually in threes; all the uriniferous tubule anlagen, therefore, lie close together and the tubules connect with neighboring terminal collecting tubules. In this way there is formed around the entire kidneys a first investment of uriniferous tubules, which is always only one anlage deep. As soon as the tubules have separated from their parent tissue, the metanephrogenic tissue regenerates from what remains and is now raised up from the layer of uriniferous tubules and carried towards the periphery by the outgrowing collecting tubules of the new generation. Here the various pieces of metanephrogenic tissue again unite to form a continuous sheet, which may be called the neo genie zone. It invests the new terminal collecting tubules and the layer of uriniferous tubules. Now the same process is repeated, the neogenic zone again forms over the whole surface of the ureteric tree a second layer of uriniferous tubules — we may speak of them as a second generation — and it itself again regenerates to a new neogenic layer, which is again displaced peripherally by the outgrowing collecting tubules of a new order. Since the process always repeats itself in the same way there is formed by apposition to the outer surface a new cortical layer, and the oldest generation of uriniferous tubules is that nearest the renal pelvis, while the youngest lies immediately below the neogenic zone. As soon as the various Malpighian pyramids of the first and second orders are formed, each pyramid becomes invested by its own cortex of uriniferous tubules and its own neogenic layer. The various layers of tubules are, naturally, in different degrees of development; in a stage with three generations of them, those of the first generation are in the stage with the loops of Stoerck, those of the second generation are in the S-shaped stage with the anlagen of Bowman's capsules, and the third generation is in the cell-sphere or cellvesicle stage.

The formation of the uriniferous tubules is not completed at the close of the fetal period, but extends into the first days of extra-uterine life. It is only after the tenth day that the formation of new tubules ceases, the neogenic zone exhausts itself in the formation of the last generation and vanishes. From its remains a portion of the connective tissue of the surface of the cortex is formed.

The first generation of uriniferous tubules is to be found in embryos of from 13-20 mm., the second follows in embryos of 20-30 mm., the third in those of from 30 mm. trunk length to 75 mm. head-foot length; the fourth generation appears in embryos of from 60-150 mm., the 5th-8th generations in einbryos of 120-150 mm. head-foot length; seven months' embryos possess 8-10 layers of uriniferous tubules (Toldt, 1874; Hauch, 1903); new-born children 10-14 generations, and, finally, children up to the third month 14-18 generations. The formation of the various generations takes place rather evenly within any one pyramid.

If one investigates the stage of development of the ureteric tree at the time when the first uriniferous tubules unite with the collecting tubules one finds that the union takes place in the 5th or 6th generation of collecting tubules. When one remembers that all the tubules of the adult kidney open into the collecting tubules of the medullary rays and that these belong to the ultimate or penultimate generation, the question at once arises : Do the openings of the uriniferous tubules become displaced or is there a degeneration of uriniferous tubules that have formed a fully developed union with the collecting tubule system. I may first answer the second question, since the reply is very simply "No"! I have seen degenerating uriniferous tubules in none of the embryos studied. There remains then only the question as to the displacement of the openings. For the first three generations I can admit the method of displacement, for one finds that in these the newly outgrowing collecting tubules carry with them not only the metanephrogenic tissue but also the openings of earlier uriniferous tubules. In this way the uriniferous tubules of at least the three first generations all come to open into collecting tubules of the same order. This carrying outward of the opening of a uriniferous tubule can only, of course, occur along with a corresponding elongation of the intermediate piece. That the same processes occur in younger generations is possible, but it can only be determined in spite of unusual technical difficulties. One must therefore consider other possibilities, since one finds older uriniferous tubules that are not connected with any collecting tubule. Whoever has prepared models of metanephric tubules knows how difficult it is to prepare satisfactory ones, and on this account I am not yet quite certain that these blindly ending tubules really occur, and I would not mention them had I not made two other observations, namely, in the first place, an interruption of the lumen of a uriniferous tubule that, judging from its form in other respects, must have been functional some time previously, and, secondly, the occurrence of blindly ending diverticula at the transition of the tubulus contortus into the descending limb of Stoerck's loop. Both these observations suggest the possibility of a uriniferous tubule separating from its collecting tubule and opening into one of a higher order.

The Definitive Renal Pelvis and the Reduction of the Collecting Tubules.

We have termed the wedge-shaped enlargement of the ureteric anlage the primitive renal pelvis and from it the four collecting tubules of the first order arise. If we compare this primitive pelvis with the pelvis of the adult kidney we perceive, first, an enormous difference in size and, second, a difference in the number of collecting tubules opening into it; the pelvis of the adult kidney may receive over one hundred collecting tubules. This difference might be produced by a belated — but in that case enormous — new formation of collecting tubules of the first order, just as a belated appearance of them occurs in the two central tubules, but all investigations along this line have yielded negative results. We must therefore seek for another source of the increase and this is found in the so-called reduction of the oldest or older collecting tubules. These first of all become enlarged and are then taken up into the wall of the growing renal pelvis so completely that no traces of them remain. The reduction begins regularly at the centre and proceeds towards the periphery; if it begins in such a manner that the collecting tubules of the first order are taken up into the pelvis, then the so-called undivided form of kidney is produced, one with a single papilla and a single large calyx, represented by the entire pelvis, a form which is possessed by the rabbit, for example ; if the reduction takes place in such a way that the collecting tubules of the first and second order are retained, and the tertiaries and quaternaries are taken up into the enlarged secondaries, then the so-called divided form of kidney results, that is to say, the kidney possesses several papillae, each surrounded by a special calyx. The tubules of the second order then form the calices minores, those of the third order the calices majores. The human kidney develops according to the divided type, the reduction begins in embryos of 22 mm. greatest length by an enlargement of the primitive pelvis and of all the collecting tubules of the first four orders. Their epithelium becomes at the same time very clear and, in spite of the enlargement, high cylindrical. Then in embryos of 60 mm. head length the tubules of the third and fourth order are taken up into those of the second order, whereby the tubules of the fifth order open directly into those of the second.

Fig. 598 a and b. — Casts of the renal pelves of two adult men to show the two types resulting from the simple and the double reduction, a, Ramified pelvis (simple reduction), b, Ampullary pelvis (double reduction). After Hauch (1903). Reduced to f.

Into a tubule of the second order there open three tubules of the third order, nine of the fourth order and twenty-seven of the fifth order, and since 20-30 ductus papillares open upon a pyramid of the adult kidney, the complete reduction of the tubules of the third and fourth orders suffices to bring about the condition found in the adult organ.

The human kidney presents an additional point of interest in that in many kidneys a second process of reduction follows the first. The second reduction again starts from the renal pelvis and leads to the absorption into the pelvis of the tubules of the first order completely and those of the second order almost completely. The remains of those of the second order are then seated upon a plump pelvis as short stumpy calices. The simple and the double reduction produce, of course, quite different pelvic forms, the extremes of which I shall term, following Legueu (1891), the ramified aud the ampullary renal pelvis (Fig. 598 a and b): The form of the ramified pelvis will evidently depend upon the different amounts of growth in length of the tubules of the first and second orders.

It has been mentioned above that secondary and even tertiary columns of Bertini may develop. The secondary columns may be followed to the angle of division of the collecting tubules of the third order, the tertiary ones to the angle of division of the tubules of the correspondingly higher order. If now the tubules of the third and fourth orders are taken up into the renal pelvis, the secondary and tertiary columns will reach the sinus of the kidney and thus secondarily acquire the character that the primary columns possess from the beginning. In this way an increased number of pyramids will occur, their tips, the papillae, projecting into the renal pelvis. An increase in the number of renal papilla? is, accordingly, closely associated with the reduction process. From one primary pyramid as many as nine secondary and tertiary ones may be formed. The definitive number of papillae is probably already acquired by embryos of the third and fourth months (Toldt 1874) ; but it must not be assumed from this statement that the reduction process comes to an end at this time; it becomes retarded, but may be in action even at birth and still draw many a ductus papillaris into the pelvis.

The Formation of Cortex and Medulla — Medullary Rays — Papillae.

The development of the cortex begins with the formation of the first generation of uriniferous tubules, and its increase in thickness depends in the first place upon the new formation of additional tubules and, in the second place, on the growth of those already present. The first generation of tubules produces at once a sharp delimitation of cortex and medulla. The latter is whatever lies in the primitive sinus and this space at first contains only a small number of collecting tubules but an abundance of mesenchyme tissue, whose meshes early arrange themselves transversely to the collecting tubules. One may, however, properly speak of a beginning of the medulla formation only after the formation of the definitive renal pelvis. The collecting tubules of the fifth order become grouped together into bundles and with this the actual development of the medulla begins, the collecting tubules of the fifth to the ninth or eleventh orders taking part in its structure. The medulla increases both in length and thickness, in thickness by an increase in the diameter of the individual collecting tubules and by the down-growth of the loops of Henle, in length by the growth in length of the individual collecting tubules.

A series of measurements has shown that a great difference in the relation of cortex to medulla mav exist in different individuals and in the right and left sides of the same individual. Furthermore in the period between the fifth fetal month and the adult condition a series of growth periods may be detected. During fetal development it is principally the medulla that grows; its transverse diameter enlarges on an average by 100 per cent., and at the same time the thickness of the cortex increases only by 20-25 per cent. The stronger growth of the medulla depends on the one hand on a change in the direction of the collecting tubules; with the growth of the ureteric tree the collecting tubules, which pass off at an angle of 90°, parallel to the surface, gradually become straightened out, only the terminal tubules for the time being retaining the original angle. In the second place the growth is associated with the development of the tips of the pyramids, the papillae. After birth a cessation of growth of the medulla occurs and this lasts until the seventh year ; its development remains as it was or progresses but slightly, but in the same period the diameter of the cortex increases regularly. After the seventh year both portions again grow equally and double their diameter by the time of puberty.

Medullary rays properly exist from the time when at least three generations of uriniferous tubules are arranged in layers one above the other. At least the cortex at this time becomes quite regularly divided into separate portions by collecting tubules. These embryonic medullary rays are, however, produced by the collecting tubules of the sixth or seventh order, while those of the adult kidney are formed by those of the tenth to the twelfth orders, the tubules of the sixth or seventh order coming to lie in the medulla proper. Embryonically, therefore, we must distinguish between the temporary and the persistent medullary rays. The first indications of the persistent rays are found in embryos of the 14th-16th week and with a length of 9-13 cm. (Hauch 1903). Each medullary ray contains at this time 3-4 tubules, which are collecting tubules of the 10th (or 11th or 12th) order that have budded out in threes or fours from the collecting tubules of the 9th (or 10th or 11th) order. To these medullary ray collecting tubules 9-16 terminal collecting tubules belong. The medullary ray tubules must behave differently than the tubules of older orders in respect to their mode of outgrowth. I draw this conclusion from the accurate results of Peter (1909). The collecting tubules of the 5th-9th or 11th order (1) give rise to new collecting tubules only at their blind ends, (2) they carry up with them the uriniferous tubules that open into them or (3) their uriniferous tubules break their connection with them and make a new union with a collecting tubule of a higher order, so that, finally, all the uriniferous tnbules open into the collecting tubules of the medullars ravs. If the medullarv ray tnbules behaved like those of the medulla they would show openings of uriniferous tubules only at their ends, but according to Peter this is not the case; they receive terminal collecting tubules in groups along their entire length. How this arrangement is developed is at present unknown; since, theoretically, different ways of reaching this end may be imagined, it would be idle to discuss them so long as we possess nothing that points in favor of one rather than another of the possible ways.

The papillae arise by the growth in length of collecting tubules belonging principally to the middle orders (Schweigger-Seidel 1865, Riedel 1874). If one compares a longitudinal section of a pyramid of a seven months' embryo with a similar section from the adult kidney one sees that in the embryo the branchings of the collecting tubules occur uniformly throughout the entire length of the pyramid, even into the medullary rays, while in the adult kidneys they take place in two groups ; one group lies near the tip of the papilla, the other near the medullary ray. Both groups are connected by a longer undivided stretch of the collecting tubules. This grouping has correctly led Schweigger-Seidel to the conclusion that the growth in length occurs principally in the tubules of the middle order. It produces at all events a pushing of the tips of the pyramids towards the calyx. The first papillae occur in the human embryo in the third month; at first they are small, and consequently appear to be very long, but the more the loops of Henle grow downwards, the thicker and more distinct they become. That primary papilla? may become divided, during the reduction, into secondary papilla? by secondary columns of Bertini is evident from what was said above concerning the reduction.

The Relation between the Right and Left Kidney.

According to the observations of Hauch( 1903), there is a great correspondence of form in the kidneys of either side during development. The right and the left kidney are always at the same stage of development, the number of lobes is approximately the same and the form of the renal pelvis is very similar. Later on, however, the pressure of the liver causes a lesser development of the right kidney.

Change of Position of the Kidney.

The migration of the kidney cranialwards has already been described in connection with the development of the ureter (p. 834), and it was there shown how slight the actual migration really is. A second change of position of parts of the kidney occurs with the development of the collecting tubule system; the cranial and caudal poles are displaced in different directions around the renal pelvis as a fixed point. In the first half of fetal life the craniocaudal diameter of the kidney corresponds approximately to the first three lumbar vertebras, while in the second half of the same period the cranial pole rises to the level of the eleventh rib and the caudal one descends to the upper border of the fifth lumbar vertebra. In this growth in length, which depends on the broadening of the collecting tubule system, the left kidney is almost always in advance of the right, the difference in favor of the left kidney being the greater, the older the embryo examined. After birth a change in the position of the kidney also occurs. In children less than one year old Algiave (1910) in half the cases examined (16 out of 32) found the caudal pole of the kidney in the iliac fossa, in children between 1 and 2 vears it was at the iliac crest in a minoritv of the cases (3 out of 9), and in those of more than 2 years it was always above the iliac crest. This change is a passive one, depending neither upon a more extensive growth in length of the kidney nor upon a displacement of it, but upon a stronger growth of the posterior abdominal wall and especially of the lumbar region. By this growth the space between the 12th rib and the iliac crest is greatly increased and room is thus made for the reception of the kidney. The growth in length and thickness of the kidney also takes place in periods; both diameters increase greatly during intra-uterine life, from 18-50 mm. in length and from 13-25 mm. in thickness; during the first year of life only a slight increase takes place, from 56-70 mm. in length and from 25-37 mm. in thickness. From the second year until the final completion of growth, especially during the period of puberty, there is again a rapid increase, from 73 to 120 mm. in length and from 36 to 60 mm. in thickness (Kiilz 1899).

During the growth of the kidney it also undergoes a rotation around its long axis. The renal pelvis and its first collecting tubules are at first almost dorsal to the ureter (embryo of 12.5 mm. greatest length), but in embryos of 19.5 mm. greatest length the ureteric tree is practically lateral to the ureter. In older embryos there is again a rotation in the opposite direction, probably a result of the development of the vertebral centra, so that the lateral border assumes the position it possesses in the adult, about midway between a frontal and a sagittal position.

The Vessels of the Kidney.

In the chapter on the mesonephric arteries it was stated that one or several mesonephric arteries may become the persistent metanephric arteries. The mesonephric arteries are transverse branches that arise from the aorta and rim to the mesonephric fold, and their terminal branches form a network, the rete arteriosum urogenitale, lying in the angle between the reproductive gland, the mesonephros and the metanephros. In 18 mm. embryos this network comes into connection with vessels which actually enter the renal sinus. In this way the vessels of the metanephros become connected with the mesonephric arteries and through them with the aorta. The network makes it possible for any of the mesonephric arteries, all of which participate in its formation, to become a metanephric artery; hence the variability in the origin of the latter, their frequent dissimilarity on the right and left sides and, finally, their frequent multiplicity. The mesonephric artery that is destined to become the metanephric artery is distinguished from its fellows by its greater diameter, and this difference can usually be made out in embryos of 21 mm. greatest length. The branches of the a. renalis in the renal pelvis are so arranged that they come to lie both on the dorsal and the ventral side of the pelvis. The first a. arcif ormis was shown by an embryo of 30 mm. greatest length.

The development of the v. renalis is so intimately connected with that of the v. cava inferior, that it is preferably considered with it.

The Capsule of the Kidney.

The actual capsule of the metanephros is first distinctly seen in embryos of 70 mm. head-foot length. Embryonic connective tissue, indeed, surrounds from the beginning the growing ureter and the outgrowing collecting tubules together with the metanephrogenic tissue, and is, moreover, arranged concentrically; it cannot, however, be separated from the surrounding connective tissue, and until this can be done one cannot speak of a kidney capsule as an independent tissue.

The Later Development of the Ureter.

The course of the ureter is always retroperitoneal and always close to the mesentery of the mesonephros. Within the abdominal cavity it is almost straight, but on entering the true pelvis it curves so as to be convex caudally and, again ascending to the dorsal side of the bladder, it passes to its opening. Since the ureter is developed from the primary excretory duct, it will at first open into this; how it acquires its opening into the bladder may preferably be considered along with the development of that organ. Up to embryos of 125 mm. the ureter is almost straight and has a uniform calibre from beginning to end. Later three narrowed and two fusiformly dilated parts appear (Seitz 1908). The first narrow part is at the origin of the ureter from the renal pelvis, the second is where the ureter crosses the linea innominata between the false and the true pelvis, and the third is immediately above its entrance into the bladder. The first enlarged part, the lumbar one, is between the first and second stenosis and therefore in the abdominal portion of the ureter ; the second, the pelvic one, is between the second and third stenosis and is therefore in the pelvic portion. The lumbar enlargement appears in embryos of 125 mm. and simultaneously with it the upper and middle narrow parts ; the pelvic enlargement develops only later (in embryos of 320 mm.) ; it is inconstant, and may be completely wanting after birth (Gerard 1908). The lumbar enlargement is never lacking in embryos more than 125 mm. in length and in children may taper off very gradually above and below, in which case the upper and middle narrow parts are only indistinctly present, and the middle one may even be wanting: if the enlargement is sharply defined the upper and middle narrow parts are distinct, the cranial one being the more sharply pronounced (Gerard 1908). The lower narrowing is produced by the formation of the bladder wall and is, therefore, always present.

In the region of the lumbar enlargement the ureter is spirally twisted and this twisting may be the ultimate cause of the formation of the narrowed and enlarged parts.

The epithelium of the ureter is at first single-layered, sometimes more cubical, sometimes more cylindrical; the narrower the lumen, the higher the epithelium. In embryos of from 40-50 mm. greatest length the epithelium is two-layered at various places, especially in the region of the genital cord. In those places where a single-layered epithelium obtains in these embryos, isolated cubical cells occur among the cylindrical ones and so produce the appearance of gland-like depressions. In embryos of 55 mm. head-foot length the epithelium is three-layered in the upper half of the ureter and five-layered in the lower half, and in embryos of 70 mm. head-foot length, the transitional epithelium begins to form, although in an embryo of 90 mm. head-foot length no syncytium is yet to be discovered at its surface. It is remarkable that a sort of cuticular membrane extends from the bladder over the free surface of the ureteric epithelium; in embryos of 50 mm. head-foot length it is present only in the lower portion of the ureter, but later it extends further and further upwards, reaches the renal pelvis and finally also the collecting tubules, with the exception of those that are at the time terminal.

The mesenchyme forms delicate concentric rings, about twenty in number, around the ureter. Inside these rings fine strands of circular musculature appear in embryos of 70 mm. head-foot length. The musculature appears first at the lower end of the ureter and then gradually extends towards the kidney; in embryos of 150 mm. it is developed throughout the entire length of the ureter.

Malformations of the Kidney

1. The union of the uriniferous tubules with the collecting tubule system is, as has been seen above, a complicated process, and it is conceivable that slight disturbances may prevent the union of a uriniferous with its collecting tubule. Nevertheless, such uriniferous tubules continue their development and secrete; but since there is no exit for the secretion, a cystic degeneration of the tubule eventually results. A certain proportion of cystic kidneys may, accordingly, be assigned with certainty to such disturbances of development.

2. A second malformation is produced by the union of the kidney blastema of the right side with that of the left. This union occurs very early. I found it in an embryo of 30 mm. greatest length. The united parts form a transverse bar caught by the inferior mesenteric artery and are retained by this in their caudal position. Nevertheless, the right and left kidneys may develop further in a winglike manner and eventually almost reach the normal position of ununited kidneys. What the cause of the union of the two blastemata may be is unknown, but even in the transverse bar the constituents of the two kidneys are sharply separated.

3. A third malformation consists of a precocious splitting of the ureter, ureter fissus. The cause of a cleft ureter is to be referred to a non-development of the primitive renal pelvis, the pole tubules developing precociously and then ascending parallel to one another. The cleavage may later extend downwards into the undivided portion of the ureter, so that finally the latter is cleft right to its entrance into the bladder. All so-called double ureters which have a single opening into the bladder should be termed cleft ureters, and not double ureters. In this connection most autopsy protocols are defective; the ureter fissus is regularly termed a double ureter, although it is single.

4. The double or triple ureter (ureter duplex and triplex) represents a regressive variation. I shall have occasion to speak of them in the next chapter, and shall only point out here that they arise as two or three ureteric buds from the primitive excretory ducts, and that these multiple ureteric buds do not develop equally, the most caudal one showing the strongest development, not only as regards the expansion of its ureteric tree, but also as regards the height that it reaches in its development. It represents the normal ureteric bud and always reaches a more cranial level than the other two buds.

The Phylogenetic Development of the Metanephros.

We must define again in all clearness what we mean by the metanephros. It is an excretory organ whose tubules open into special collecting tubules, developed from the primary excretory duct. AYbether these special collecting tubules axe represented by a simple tube or develop to such an expanded ureteric tree as we see in man, the opening of the metanephric tubules into a special efferent duct developed from the primary excretory duct is the characteristic feature.

Starting with this point if we consider, first of all, the development of the mesonephros of the Gymnophiones, we will find that we have to distinguish in it between ventral mesonephric tubules, all of which are in a longitudinal row at the summit of the mesonephric fold, and dorsal mesonephric tubules which lie at the root of the fold and, in order to open into the primary excretory duct, must traverse a much longer path than the ventral ones. These dorsal tubules are developed for the most part later than the others, and may occur in varying number and in different series. The later formed series are always the more dorsal ones, and, finally, their tubules can no longer reach the primary excretory duct by their own efforts. Accordingly the primary duct sends out special collecting tubules for the reception of these canals; according to our definition we would have to regard these dorsal mesonephric tubules with their special collecting tubules as metauephrie structures. In the mesonephros of young Gyinnophiones, then, there is scattered throughout its entire length a series of metanephroi ; only the caudal ones persist, however, in the adult. In birds a similar observation is to be recorded. It must be pointed out, in the first place, that the length of the mesonephros in birds is markedly less than that of the mesonephros of the Gyrnnophiones, and, consequently, the mesonephric tubules are compressed into a smaller space. What the mesonephros loses in length, it gains in thickness, and since a great amount of development is possible only in the caudal portion of the body cavity, the mesonephros increases in man — i. e., develops dorsal tubules, principally in this region. These dorsal tubules are too far away to open directly into the primary excretory duct, and, therefore, special collecting tubules are formed for them by the duct. According to our definition these dorsal mesonephric tubules, together with their collecting tubules, are also metanephroi. Their number, however, for the reason stated, is small; we have only a few metanephroi and a few special collecting tubules.

Finally, we have shown that in the development of the human mesonephros a heaping up of tubules occurs at the caudal end; furthermore, that these mesonephric tubules no longer lie in one series, and, finally, that special collecting tubules are formed for them from the primary excretory duct, but these are few in number in comparison with the condition in the chick.

In brief, then : At first numerous metanephroi are interposed between the mesonephric tubules; the more the space in the body cavity becomes limited as development proceeds, the more also the number of these metanephroi becomes diminished, until, finally, in adult mammals only one is developed. We must, therefore, distinguish between different ureters. Those ureters that are developed for the small metanephroi imbedded in the mesonephroi of the Gymnophiones and birds, and also for those in the caudal pole of the human mesnephros I term mesonephric ureters. The ureter which supplies the persistent metanephros I term a metanephric ureter, and logically we must distinguish in man between provisional metanephroi and definitive metanephros. When in man two or three ureters occur as an anomaly, it is, as I have said above, a reversion, since in such cases one or two of the mesonephric ureters and their tubules persist.

I have finally to consider why the last ureter becomes the metanephric ureter. I would associate this with the preference which the last ureter acquires in reptiles and birds. In these the mass of the mesonephric tubules extends caudally beyond the primary excretory duct and forms the so-called caudal kidney. For the tubules of this the primary excretory duct forms a special strong ureter, which must be the last of all the ureters. By the development of the caudal kidney the last ureter acquires a special importance and therefore a preference, which finally allows it to dominate over the others as the metanephric ureter.

The Function of the Mesonephros

In most anamnia the mesonephros begins its activity while the pronephros is still at its functional height. For a time both organs function together; then the pronephros undergoes degeneration and the mesonephros becomes the only functional excretory organ. This entire complex of processes has without further consideration been transferred to the relation between the mesonephros and the metanephros, and the question whether the mesonephros is actually functional in the amniota has never been seriously considered. Weber (1897) was the first to take it up and he endeavored to answer it in the following manner. He compared the time of degeneration of the mesonephros with that of the development of the metanephros; when he found that the mesonephros degenerated before the metanephros could exercise an excretory function, he has assumed that the mesonephros did not function, for if it had been active and had then degenerated before the inetanephros bad begun its activity, there must have been a certain period of development during which there was no excretion. Let us adopt this same method in considering the special case of the function of the human niesonephros. Already in an embryo of 19.4 mm. greatest length the majority of the mesonephric tubules are so far in process of degeneration that they cannot be regarded as having an excretory function. Of the 35 tubules of this embryo only four are actually still intact. In an embryo of 22 mm. greatest length none of the mesonephric tubules were capable of functioning ; in all the tubulus secretorius had separated from the tubulus collectivus. If one inquires how far the development of the metanephros has progressed at this time, one finds that embryos of 22 mm. have just reached the anlage of the second generation of uriniferous tubules. The first generation, however, has as yet no fully formed Malpighian corpuscles. If, then, the mesonephros had functioned as an excretory organ, there must necessarily have been an interruption of this function on its degeneration. Consequently, I regard the question as to the functioning of the mesonephros as settled ; it does not function as an excretory organ. This does not, of course, imply that it may not have been active in another manner unknown to us.

Cloaca, Bladder, Urethra, and Urogenital Sinus

The rectum, bladder, urethra aud urogenital sinus are formed by a triple division of the cloaca. The first division completely separates the dorsal third of the cloaca or the rectum, and the remains of the cloaca are divided by a second incomplete division into the bladder, urethra and urogenital sinus.

From the physiologico-morphological stand-point these are grouped together under the term " bladder " organs, which show an entirely different development, and are not to be derived from one another phylogenetically. In the first place, it is necessary to distinguish between entodermal and mesodermal bladders; a mesodermal bladder arises in some manner from the primary excretory duct; an entodermal bladder may be formed from the cloaca or from its derivative, the allantois, and we may speak, therefore, of cloaeogenic and allantoidogenic bladders. Finally, it must be pointed out that the bladder derived from the cloaca may be formed either from its dorsal or its ventral wall, and we must, therefore, distinguish between dorso-cloacogenic and ventro-cloacogenic bladders. The Amphibia possess a purely ventro-cloacogenic bladder, a purely allantoidogenic bladder occurs temporarily in some birds, a purely mesodermal bladder is developed in the selachians. All other bladders are of mixed origin. The Teleosts, Ganoids, Petromyzonts, snakes, crocodiles, and mammals have bladders of cloaeogenic and mesodermal origin; the lizards derive their bladder from three sources: dorsocloacogenic, allantoidogenic, and mesodermal ; and, finally, in the turtles the bladder is formed from dorso-cloacogenic, ventro-cloacogenic, allantoidogenic and mesodermal anlagen.

In man a ventro-cloacogenic and mesodermal bladder develops; an allantoidogenic origin, such as was formerly supposed to occur, does not exist.

By cloaca is understood the part of the posterior intestinal bay that lies caudal to the point where the allantois is given off. Into it there open from above the end-gut and the allantois (Fig. 599). It is a blind sack, oval in transverse section, its longest axis standing dorso-ventrally and its ventral surface being compressed almost to an angle (Figs. 530 and 534). By this ventral angle the cloaca comes into contact and fuses with the ectoderm of the surface of the body, the mesoderm being pressed aside. The area of fusion is termed the cloacal membrane (Tourneux 1888, Born 1894, Fig. 534). This consists, accordingly, of two layers of epithelium, a stronger ectodermal and a weaker entodermal one (Fig. 600) ; later the difference in thickness becomes more equalized and in the region of the urethral plate it becomes reversed. Seen from without the membrane lies at the bottom of a shallow rhomboid groove (Fig. 601, Keibel 1896). At first the cloaca is short in the cranio-caudal direction; the more the tail grows, the more the cloaca becomes enlarged and the cloacal membrane keeps pace with it. A comparison of Figs. 599 and 600 shows the increase of both. In both figures the membrane begins immediately caudal to the point where the allantois is given off and extends in an embryo of 2.6 mm. greatest length (Fig. 600) to the caudal end of the cloaca ; in this embryo, accordingly, the membrane forms the entire ventral wall of the cloaca.

Fig. 599. — Reconstruction of the embryo Pfannenstiel-Kromer, 1.38 mm. long and with 5-6 pairs of primitive segments, seen from the right side. X 50. The figure is intended to show the relation of the cloaca and cloacal membrane to the allantoic stalk. The latter arises from the cloaca at a right angle, immediately above the cloacal membrane.

The growth of the cloaca is shown by the following data. Embryo of 1.38 mm. greatest length: The length of the cloaca is 90 m, that of the cloacal membrane 20 M. Embryo of 2.6 mm. greatest length and with 13-14 pairs of primitive segments: length of the cloaca 160 M, that of the cloacal membrane 130 m; the cranial end of the cloaca is in the 14th primitive segment. Embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments : length of the cloaca and cloacal membrane 110 M, the cranial end of the cloaca lies at the level of the future 26th primitive segment. Embryo of 4.25 mm. vertex-breech length : the length of the cloaca 250 M, that of the cloacal membrane 240 M; the cloaca lies in the 28th primitive segment. Embryo of 4.9 mm. nape length : length of cloaca 540 M, that of the cloacal membrane 360 M. Embryo of 11 mm. greatest length : length of the cloacal membrane 470 M.

The first division of the cloaca begins in embryos between 4.9 mm. nape length and 5.3 mm. greatest length; it separates a dorsal third or quarter, the rectum, from the rest of the cloaca, the ventral remains of the cloaca. It takes place by the saddle between the cloacal opening of the intestine and that of the allantois growing as a partition downwards from above into the lumen of the cloaca, parallel to its dorsal wall (Fig. 602) ; this partition is termed the septum urorectale. The lower border of the septum is not straight, but forms a curve, since it grows downward faster at the lateral walls of the cloaca than it does in the middle. One thus gets the impression that there are at the lower end of the septum two lateral folds, the plicae urorectales, which unite in the median line and so form the unpaired septum. The middle line of the septum, however, never shows a raphe, such as would be formed by the union of folds, and therefore the description of the septum urorectale as unpaired is more correct. The line, along which the septum grows downwards in the interior, is marked on the outer surface of the cloaca by a groove or a slight ridge (Keibel 1896). The rectum and the ventral remains of the cloaca are differentiated even before the division by differences in the structure of the epithelium of the cloacal wall. That of the dorsal wall, so far as it forms the future rectum, remains high cylindrical and the portion of the lumen corresponding to it becomes narrow ; the epithelium of the ventral wall, so far as it forms the ventral remains of the cloaca, flattens out, and the portion of the lumen eorresiDonding to it remains wide or even widens. At the line of transition of one portion of the lumen into the other there is a ridge (Fig. 562), which corresponds to the position of the. future urorectal fold. Growing further down the urorectal septum reaches the neighborhood of the cloacal membrane and thus divides the cloaca almost completely into the rectum and the remains of the cloaca, a small connecting passage only remaining between the two immediately above the cloacal membrane (Fig. 603 b). At this stage the development may be inhibited and the communication between the rectum and urethra thus is made permanent. On this account this communication is termed the cloacal duct (Reichel 1893). Normally, however, the urorectal septum grows through this last stretch also, reaches the cloacal membrane and fuses with it (Fig. 604 b). By the fusion the cloacal membrane is divided, and we now speak of an anal membrane that closes the rectum, and a urogenital membrane, that closes the ventral remains of the cloaca. In the later development both membranes are broken through independently, and so the urogenital and the anal openings are formed. The breaking through takes place in embryos of between 13 and 18 mm. greatest length ; the anal opening generally is formed somewhat later than the urogenital.

Fig. 600. — Reconstruction of the caudal end of the body of Embryo Pfannenstiel III, 2.6 mm. greatest length and with 13-14 pairs of primitive segments. X 50. Cloaca and cloacal membrane have grown greatly in comparison with the conditions in the embryo shown in Fig. 599; the allantoic stalk is given off at a right angle from the cloaca, immediately above the cloacal membrane. The ventral wall of the cloaca is formed only by the cloacal membrane.

Fig. 601. — Model of the cloacal membrane of a human embryo of 3 mm. greatest length. (Embryo E. B., from the collection of Professor His, Leipzig; from Keibel, 1896.) The cloacal membrane forms a rhomboidal groove slightly depressed below the surface of the embryo, between the umbilicus and the coccygeal tubercle. At either side is one of the paired cloacal swellings.

Fig. 602. — Model of the cloaca of a human embryo of 7 mm. greatest length (embryo Chr. I, from the collection of Professor Hochstetter, Vienna), in which the gradual separation of the rectum from the ventral remains of the cloaca is shown diagrammatically by the series of dotted lines.

As a mark by which the progressive division of the cloaca may be measured, only the point of opening of the primary excretory duet is available, and this is by no means satisfactory, since it also alters its position. As is shown in Fig. 534, when the point of contact of the primary excretory duct with the cloacal wall "is first established, it lies close to the ventral ridge of the cloaca, and therefore close beside the cloacal membrane. At the beginning of the division of the cloaca this point moves dorsally, so that seen from the side, it comes to lie at the middle of the lateral cloacal wall (Fig. 605). Whether a displacement in the eranial direction accompanies this evident horizontal one cannot be determined with certainty. With this limitation the following data may be accepted. In young embryos up to those of 4.9 mm. greatest length the opening of the excretory duct almost divides equally the cranio-caudal diameter of the cloacal wall. In an embryo of 7.8 mm. the saddle lies between the opening of the intestine and the allantois, somewhat above the opening of the excretory duct. In an embryo of 8.5 mm. greatest length the saddle already lies 50 M beneath this opening — i. e., the opening is already into the divided cloaca. In a 7 mm. embryo the saddle is 190 M below the opening of the duct, in one of 9.5 mm. greatest length it is 260 /" below the opening and in one of 11 mm. greatest length the division of the cloaca is complete.

While the division that separates the rectum from the ventral remains of the cloaca is still progressing, the second division, which divides the ventral remains of the cloaca into the bladder, the urethra and the urogenital sinus, is beginning; the possibility of such a division depends on the curving of the caudal end of the body (see Vol. I, Fig. 272). The cloaca is fastened to the outer wall of the body at three points, to the umbilicus by the allantoic stalk or urachus, along the entire cloacal membrane to the anterior abdominal wall, and loosely by the primary excretory duct and ureter to the posterior wall of the abdomen. The positions of these three fixation points, relative to one another, determine the form and position of the cloaca. As has been pointed out the length of the cloacal membrane represents at first the length of the ventral wall of the cloaca ; the membrane begins immediately below the origin of the umbilical cord or belly stalk and extends to the caudal end of the cloaca (Figs. 600, 605 and 606) ; the dorsal wall of the cloaca in young embryos lies on the ventral surface of the aorta or chorda (Fig. 600). This position changes with the elongation of the embryo; the umbilicus and cloacal membrane become pushed apart, a piece of the anterior abdominal wall is formed between them and, corresponding to this, a new portion of the ventral wall of the cloaca is formed, situated cranial to the cloacal membrane; compare Figs. 600 and 603 b. A little later the anterior abdominal wall from the umbilicus to the caudal end of the cloacal membrane becomes pushed out to form an elevation (Figs. 607 and 608). On this elevation, the cloacal tubercle, the ventral wall of the cloaca is so placed that the cloacal membrane lies on the anal slope of the tubercle, extending from its summit to the caudal periphery of its base, i.e., to the anus, while the newlyformed portion of the cloacal wall extends from the summit of the tubercle to the umbilicus (Fig. 608) ; the newly-formed wall and the membrane therefore form an angle open dorsally. The dorsoventral diameter of the abdominal cavity increases with the curvature of the tail and thereby the primary excretory duct be comes at first elongated ; compare its course in Fig. 606 and in Figs. 603 and 604. As the enlargement of the abdominal cavity increases still more, the primary excretory duct and the ureter, which are fastened to the posterior abdominal wall, draw out dorsally the posterior wall of the ventral remains of the cloaca (Figs. 603 and 604). The form and position of the ventral remains of the cloaca become altered by these modifications, the cavity becoming almost quadrangular in median section ; while in Fig. 603 b we can distinguish only a ventral and a dorsal wall in the remains, in Fig. 604 there is a ventral wall, formed by the urogenital membrane, a cranial one formed by the newly-formed piece of the ventral cloacal wall (in Fig. 604 b , this wall is already depressed caudally), a posterior one and a caudal one formed by a bending of the actual dorsal wall at the level of the primary excretory duct. The cranial wall now becomes curved downwards and the lumen, originally rectangular in median section, becomes divided into three portions (Fig. 604 5). The depression of the cranial wall has taken place in an embryo of 11 mm. greatest length, a model of which is shown in Fig. 604, and the originally simple lumen of the cloacal remains is incompletely divided into a dorsal and a ventral wide portion and a median narrow portion. From the posterior broad lumen the vesico-urethral anlage is formed, the narrow middle portion becomes the pars pelvina and the broad ventral portion the pars phallica of the sinus uro genitalis. Since the urogenital membrane runs along the anal slope of the cloacal tubercle and later of the phallus, by the outgrowth of the phallus the pars phallica of the urogenital membrane becomes enclosed in that organ and hence its name.

Fig. 603 a and 6. — Model of the bladder of a human embryo of 7 mm. greatest length. (Embryo Chr. I, from the collection of Professor Hochstetter, Vienna.) Model prepared by my son, Willy Felix, a. From the right side; b, a median sagittal section. The cloaca has grown somewhat; by the gradual formation of the lower half of the ventral abdominal wall the umbilicus, and with it the allantoic stalk, is carried away from the cloacal membrane, and, as a result, there is formed, above the latter, a newly formed portion of the ventral wall of the cloaca, from which the allantoic stalk arises at a right angle as before. The division of the cloaca into the rectum and the ventral remains of the cloaca is almost complete, the saddle between the two being immediately over the cloacal membrane. Between the two there is still a tubular communication, the cloacal duct of Reichel.

Fig. 604 a and b. — Model of the cloacal region of a human embryo of 11 mm. greatest length. (Embryo P. I., from the collection of Professor Hochstetter, Vienna.) Model prepared by my pupil Frau Gervai. a, From the right side; b, median sagittal section. The rectum is completely separated from the ventral remains of the cloaca, and the cloacal membrane has been divided into the canal and urogenital membranes. The ventral remains of the cloaca has changed its position in that it has been drawn dorsally by the ureters and primary excretory ducts as a result of the enlargement of the dorso-ventral diameter of the body cavity. Thereby its former dorsal wall — compare Fig. 603 — is divided into a ventral and a cranial portion, the former being formed by the urogenital membrane and the latter by what was termed in Fig. 603 the newly formed ventral cloacal wall. The cranial wall is depressed caudally in its middle part. Thus the ventral remains of the cloaca is divided into three portions: into, 1, the vesico-urethral anlage, the dorsal portion, 2, the pars pelvina of the urogenital sinus, the middle narrow portion, and, 3, the pars phallica of the urogenital sinus, the ventral broad portion; the latter is closed externally by the urogenital membrane.

Fig. 605. — Model of the caudal portion of the body of a human embryo of 5.3 mm. greatest length. (Embryo 1420, from the collection of Professor Keibel, Freiburg i. Br.) Position and course of the primary excretory duct, position of its opening with reference to the cloaca and the cloacal membrane, anlage of the ureteric bud.

Fig. 606. — Reconstruction of the caudal end of the body of a human embryo of 4.25 mm. vertexbreech length and with 28 pairs of primitive segments. (Embryo H. M. 1, from the collection of the Anatomical Institute, Zurich.) Course of the primary excretory duct and the position of its orifice relatively to the cloaca and the cloacal membrane.

The first traces of the division of the ventral remains of the cloaca into the vesico-urethral anlage and the sinus urogenitalis are already indicated in an embryo of 7 mm. greatest length (Fig. 603 b). The division is more distinct in an embryo of 9.5 mm. greatest length (Fig. 567) and is complete in one of 11 mm. (Fig. 604).

Fig. 607. — Caudal extremity of the body of an 18 mm. embryo. (From a photograph by Professor R. Meyer, Berlin.) Between the umbilicus and the coccygeal tubercle cranio-caudally and the two lower extremities laterally, the cloacal tubercle has appeared. On it there may be distinguished an oral slope towards the umbilicus and an anal one towards the coccygeal tubercle. On the anal slope the ostium urogenitale and the groove which later becomes the ostium anale are visible.

With further development the vesico-urethral anlage and the pars pelvina of the urogenital sinus become broadened in the frontal direction and compressed in the sagittal one, so that the dorsal and ventral walls are almost brought into contact. The pars phallica of the urogenital sinus remains broad in the sagittal direction, but is very much narrowed in the frontal one, so that finally its left and right walls are in contact in the median line. Between the two walls there always remain one or several clefts ; only the portion that corresponds to the future glans becomes completely solid, and this and this only can be termed the urethral plate.

The irregular form of the wall of the sinus urogenitalis, both in its pelvic and phallic portions, shown by the occurrence of several completely separated lumina, gives opportunity for the formation of diverticula and for duplication of the sinus (Grubenmann, 1911).

The broadening of the vesico-urethral anlage in the frontal direction is principally caused by that portion of the primary excretory duct that intervenes between the opening of the ureter and the bladder being taken up into it. We shall call this portion the terminal piece of the primary excretory duct. It is at first narrow, but during further development it becomes more and more enlarged in a funnel-shaped manner from the opening into the bladder upward, the enlargement forming the so-called cloacal horn (Fig. 604 a). The neighboring portions of the wall of the vesico-urethral anlage also take part, however, in the formation of the cloacal horn, but, finally, the funnel is so completely taken up into the wall of the vesico-urethral anlage that it can no longer be recognized as a special structure. With the absorption of this terminal piece of the primary excretory duct into the wall of the bladder the ureter acquires a direct opening into the latter. At the beginning of its development the ureter lies on the dorsal wall of the primary excretory duct, but during the formation of the cloacal horn it passes to the lateral side of the duct, so that when the direct opening into the bladder develops it lies immediately lateral to the opening of the primary excretory duct. The portion of the wall between the two openings then begins to grow extensively, so that while the opening of the primary excretory duct remains where it was, that of the ureter gradually migrates upwards. This migration takes place not only cranially, but also in a lateral direction. The two excretory duct openings remain close together, and the region in which they open becomes raised up into a hillock and projects into the lumen of the urethra (Fig. 655) ; this hillock we shall term Midler's hillock. The triangle that is marked out by the three points, Miiller's hillock and the openings of the right and left ureters, and bears the name trigonum vesicae in systematic anatomy, is, therefore, developed from the wall of the primary excretory duct, so that with it we have described the development and extent of the mesodermal portion of the bladder.

Fig. 608. — Sagittal section of the lower end of the body of an embryo of 24 mm. vertex-breech length. (Embryo Hal. 1, from the collection of the First Anatomical Institute, Vienna.) The division of the ventral remains of the cloaca is completed. The cloacal tubercle lies between the caudal periphery of the umbilicus and the base of the coccygeal tubercle. The urogenital membrane extends from its tip along its anal slope and forms the ventral boundary of the pars phallica of the sinus urogenitalis.

The absorption of the terminal piece of the primary excretory duct into the wall of the bladder begins in embryos between 5.3 and 7 mm. greatest length; in embryos of 9.5 mm. greatest length the formation of the funnel is completed and thereby the boundary between it and the wall of the bladder disappears. In embryos of from 10 to 14.5 mm. greatest length the ureters and primary excretory ducts open close together into the bladder. In those of 13 mm. greatest length the cranio-lateral migration of the ureters begins, and in embryos of 22.7 mm. greatest leng-th it may be so far completed that the interval between the apex of the bladder and the openings of the ureters and that between the latter and the orifice of the urethra may have the same relative proportions as in the adult. The ureteric openings always lie at the tips of the half-moon which a transeverse section of the bladder presents. As to their lateral displacement I have obtained the following data: Embryo of 11 mm. greatest length, distance between the two ureteric openings 340 M; embryo of 12.5 mm. greatest length, distance 460 m; embryo 19.4 mm. greatest length, distance 500 /*.

Separation of Bladder and Urethra

The separation of the bladder from the urethra takes place quite gradually and consists, firstly, of an enlargement of the bladder portion while the urethral portion remains narrow, and, secondly, of changes in the epithelium of the urethral wall. At first the entire vesico-urethral anlage is lined by a single layer of cylindrical epithelium with a vesicular character, almost all the nuclei lying in the outer portion of the cells. The first changes take place in such a way that the boundary between the two kinds of epithelium is formed at about the openings of the ureters, ventrally at a somewhat more caudal point. Above the openings the epithelium remains one-layered at first, and later becomes at most two- or three-layered, while below it at once becomes twoor three-layered and later four- to five-layered. But in this also the superficial cells always show the vesicular character and have their nuclei in their outer portions. As a result of the rapid celldivision which precedes the formation of new layers, the cells of the urethra become smaller and their nuclei come to lie closer together, so that they give the epithelium a darker appearance than that of the bladder ; finally the superficial layer of the bladder loses its vesicular character and becomes cubical. Only one portion of the epithelium, that of Miiller's hillock, remains exempt from this change; it remains for a long time one-layered and high cylindrical, and its nuclei lie close beneath the free surface.

The first difference in the epithelium — i. p., the formation of a two- or threelayered epithelium in the vesico-urethral anlage — occurs in embryos of 13 mm. greatest length; a four- to five-layered epithelium occurs in the urethra of embryos of 45 mm. greatest length; and a transitional epithelium, similar to that found in the adult urethra, is shown by embryos of 55 mm. head-foot length. The enlargement of the lumen of the bladder is shown in embryos of from 25 to 30 mm. greatest length.

Later Development of the Bladder

The bladder, together with the aa. umbilicales that accompany it, lies at first (embryos of from 10 to 24 mm. greatest length) completely within the anterior body wall. Only on the enlargement of the bladder and the two arteries is the former crowded out of the abdominal wall and, developing a ventral mesentery, projects into the body cavity (Fig. 554 a). A view of the anterior abdominal wall from behind shows that the two umbilical arteries bound a triangular area, the vesical plate, whose apex is of course formed by the umbilicus (Fig. 552). Within this area the bladder lies in such a way that its cranial apex reaches the umbilicus, where it is connected with the remains of the allantoic stalk; the internal orifice of the urethra lies at about the level of the symphysis. This position the bladder retains at birth, except that its apex diminishes more and more and finally becomes the urachus, this structure being, accordingly, a product of the bladder, i.e., of the cloaca and not of the allantoic stalk. Usually the urachus degenerates at an early period to such an extent that its lumen disappears, and later the epithelial cord thus formed becomes divided into separate portions which may completely vanish before birth, but persistent remains of the urachus may for unknown reasons enlarge in a vesicular manner and give opportunities for the formation of urachal cysts. While the epithelium of the urachus completely disappears the surrounding, concentrically arranged connective tissue persists as a cord, which forms the lig. vesicoumbilicale medium. So long as the bladder lies in the triangular area between the two umbilical arteries its mesentery persists, but when it gradually sinks down after birth and comes to lie in the true pelvis, the vesical plate applies itself to the anterior abdominal wall and fuses with it ; its margins, which contain the obliterated umbilical arteries, may, however, especially in fat individuals, remain free throughout life. In postfetal life the bladder gradually descends along the anterior abdominal wall into the true pelvis, this descent taking place differently in the two sexes. In females the process of descent is a continuous one ; one may recognize in it two periods differing in the rate of the descent, it being rapid in the first and second years and quite slow from the 2nd to the 20th year. In males there are also two periods in the descent, but they are separated by a resting period; in the first two years the descent is rapid as in females of like age, then comes a resting period that lasts from the 2nd to the 8th year, and finally then succeeds the slow descent from the 8th to the 20th year. I have taken these data from a table by Disse (1891) ; Disse himself adopts a somewhat different grouping.

The Bladder Epithelium. — As has been pointed out (p. 87S), the bladder epithelium remains unaltered for a long: time. It is two-layered and thicker on the concave dorsal than on the concave ventral surface. The one-layered epithelium occurs at the apex and the many-layered at the fundus. The cells vary in height in different embryos, being sometimes high cylindical, sometimes low cubical. The epithelium first begins to be several-layered in embryos of 50 mm. head-foot length, first of all in the caudal portion. The first traces of the transitional epithelium are shown by an embryo of 60 mm. head-foot- length.

The Bladder Musculature. — A condensation of the loose mesenchyme tissue surrounding the bladder precedes the formation of the musculature, the direction of the young cells indicating their later character as circular or longitudinal fibres. The condensation of the mesenchyme, as well as its later transformation into musculature, proceeds from the apex of the bladder towards the orifice of the urethra. The musculature varies in its development and a portion above and another below the openings of the ureters may be distinguished. We may consider, first, the upper portion. The first muscle layer appears in an embryo of 22.5 mm. as a longitudinal musculature, that extends from the apex almost to the openings of the ureters. This layer is always stronger on the dorsal than on the ventral surface, and, indeed, may be completely wanting ventrally. It surrounds the epithelium in wide curves, so chat a broad sheet of loose mesenchyme persists between the two, and in this sheet a layer of circular musculature develops in embryos of 26 mm. greatest length, again appearing first in the upper portion of the bladder. In embryos of 55 mm. head-foot length a third and last layer — an inner longitudinal musculature — is formed, and both longitudinal layers give off oblique bundles to the intervening circular layer; in embryos of 80 mm. head-foot length, the distinctness of the various layers becomes obscured by the bundles of one curving into the others, so that a felted condition results.

The musculature below the openings of the ureters is always only in a single layer, and this is of circular fibres. It is to be seen first in embryos of 30 mm. and may extend as a longitudinal musculature upon the terminal portion of the ureter. The sphincter vesicas is plainly developed in a male embryo of 90 mm.

The ureter traverses the wall of the bladder in a straight course, and this condition still persists in an embryo of 70 mm. head-foot length. When the oblique course, which is already present in the new-born chiM. is acquired, is unknown.

II. THE DEVELOPMENT OF THE REPRODUCTIVE GLANDS AND THEIR DUCTS

Introduction.

The ovary and testis are up to a certain period of development exactly alike. Every vertebrate embryo forms at first an indifferent reproductive gland from which, by the emphasis of certain characters, the sexually differentiated organ is formed. The sexual differentiation of the reproductive gland is associated with a second differentiation which earlier, simultaneously, or later effects other organs. To the sexual differentiations of the reproductive glands belong : 1. The transformation of special genital cells, differing from all other cells of the body, into ova or spermatozoa.

2. The formation of ovarian follicles or seminal tubules.

3. A series of small differences: (a) The caudal pole of the female reproductive gland reaches the posterior wall of the genital cord, while that of the testis ends just above it.

(b) The cranial half or two- thirds of the ovary is rotated during development through 90° and so comes to lie at right angles to the rest.

(c) The tunica albuginea of the testis appears very early, that of the ovary very late.

(d) The a. spermatica interna has a quite different course in the two sexes.

(e) The caudal pole of the testis is united by the ligamentum testis to the chorda gubernaculi, the caudal pole of the ovary by the ligamentum ovarii proprium to the wall of the uterus.

To the sexual differentiations of other organs belong : 1. The formation of different efferent ducts for the products of the two reproductive glands. It may be noted here that every embryo, whether it later manifests the male or the female type, develops efferent ducts as well for the ova as the spermatozoa ; the duct which remains functionless degenerates.

2. The different formation of the external genital organs. Here also there is an indifferent shape through which both sexes pass before they acquire the special sex characters.

3. A series of small differences: (a) Accessorv tubes are onlv to be found in female embryos.

(b) The anlagen of the fimbriae or accessory ostia are present at an early period in the female embryo, but are wanting in the male.

c) The excavatio vesico-uterina is present in the female embryo, but not in the male.

(d) At the beginning of the third month the clitoris hangs decidedly downwards, while the penis stands out horizontally (Herzog, 1904).

(e) The developing penis rises directly from the cloacal tubercle, while the clitoris is separated from it by a groove.

(f) The praeputium penis is formed by the ingrowth of one glandar lamella, which has the form of a cylindrical mantle. The prseputium clitoridis develops by the ingrowth of three glandar lamellae, an unpaired, middle, cylindrical one and two paired, straight ones, to the right and left of the unpaired one.

The Genital Cells.

The most important cells of the reproductive glands are the genital cells. They are distinguished from all other cells of the body, the soma cells, by the size of their cell-bodies and nuclei; the content of the cell-body is about 27 times as great as that of the nucleus. This latter has usually the form of a round vesicle, frequently of a double vesicle (twin form), and contains a widemeshed delicate chromatin network (noyau leptotene [ratvia thread, /.--roc slender], von Winiwarter, 1900) with several nucleoli ; the cell-body is feebly granular, stains only with difficulty and contains in the young condition yolk granules and granules (mitochondria) of a special form.

Hitherto it has been supposed that the genital cells were specially differentiated coelom cells, derived from that portion of the ccelom wall that forms the reproductive glands. But the more our knowledge of the origin of the genital cells in the vertebrates increases, the more probable does it become that we must modify this original belief. It has now been shown for all the classes of vertebrates with the exception of the mammalia (Rubaschkin, 1909), that the first (?) genital cells have a special origin, probably being derived directly from the segmentation cells. I, therefore, term these cells primary genital cells in contradistinction to those that are differentiated from the epithelial covering of the reproductive glands; these may be termed secondary genital cells.

The proof of the origin of the primary genital cells from the segmentation cells is based on the special form in which the mitrochondria occur in their protoplasm (Lams and Doorme [1907], Meves and Duesberg [1907, 1908], van der Stricht [1900, 1909], Rubaschkin [1910], Tschaschin [1910]). The mitochondria of all the cells of the segmented germ have the form of granules, and the various granules are completely separate from one another. With the formation of the germinal layers a differentiation takes place among the cells in such a way that in the cells of the entoderm of the caudal portion of the body the mitochondria retain their granular form, while in those of the entoderm of the anterior part of the body, and in those of the ectoderm and mesoderm the isolated granules are transformed first of all into chains of granules and finally into rods or filaments.

Subsequently also the majority of the cells of the posterior half of the body show the filament form, so that at the stage of the first segmentation of the mesoderm only a few cells still show the original granular form of the mitochondria; these few cells are the primary genital cells. It is possible that with the aid of mitochondrial staining it will be shown that the secondary genital cells are derived from the primary ones and then all grounds will be removed for contrasting them as special cells with the primary genital cells.

In holoblastic ova the primary genital cells are formed in the posterior wall of the gut, in meroblastic ova in the floor of the segmentation cavity and in the germinal wall. Since the reproductive glands develop in all vertebrate embryos beside the root of the mesentery, we must suppose that a wandering of the primary germinal cells occurs. By noting all the localities in which primary genital cells are found we may reconstruct the path which is followed by the individual cells from their place of origin to the genital fold. It leads from the mesenchyme between the wall of the intestine and the parietal mesoblast into the root of the mesentery, and from there into tbe urogenital fold, at first into its connective tissue and finally into its epithelial covering. We may distinguish, therefore, between regional and extraregional genital cells, understanding by regional cells those that lie among the epithelial ceils of the genital folds; all those that have not reached that position we group together as wandering or extraregional cells. All the primary genital cells do not migrate at the same time; whether all reach their goal is uncertain, but in any case the possibility of strayed genital cells must be admitted. The secondary genital cells occur only in the genital folds and they are, accordingly, regional from the beginning.

Fig. 609 a and 6. — Part of a transverse section through the embryo Pfannenstiel III, 2.5 mm. greatest length and with 13-14 pairs of primitive segments. (From the collection of the late Professor Pfannenstiel, Kiel.) The section passes through the cloacal membrane and its exact position may be seen from the adjacent figure b. The mesoderm is still quite loosely aggregated. Between it and the entoderm there is at the upper edge of the figure a fully-formed portion of the rete peri-intestinale, and further down there are cell chains and masses (to the left of the cloaca) from which new networks of the rete are forming. The section shows at two places, to the right and left of the cloacal membrane, large cells partly multinucleate, partly filled with yolk granules and partly free from them, and lying free between the mesoderm and endoderm. These cells may be termed wandering cells and hypothetically may be interpreted as primary genital cells.

A series of transitional forms may be found between the secondary genital cells and the ordinary coelom cells (cells with noyaux protobroques a and b, von Winiwarter, 1900). I denote these transitional forms as genitaloid cells (cells with noyanx deutobroques, von Winiwarter, 1900). They agree with the genital cells in the structure of their nuclei, which are vesicular with a wide-meshed, delicate chromatin network, with the eoelom cells in the smallness of their cell bodies, which closely surround the nuclei. So soon as genital cells — whether they be primary or secondary — occur in the epithelium of the genital fold we speak of a germinal epithelium (Waldeyer, 1S70) ; this, then, denotes a mingling of ordinary ccelomic and genital cells. Just as a region of eoelom epithelium may become germinal epithelium, so also it may lose its character of germinal epithelium should its genital cells migrate out of it or otherwise disappear. The ordinary eoelom cells react to the immigration of genital cells by increasing in volume, passing from a flattened into a cubical form. But such a modification of flattened eoelom cells must not, without the presence of genital cells, be taken as evidence of the formation of a germinal epithelium. For wherever, as the result of a folding process, two ccelomic epithelial surfaces come into contact, an increase in the height of the flattened epithelium occurs, apparently as the result of a mutual formative stimulus.

Fig. 610. — Transverse section through an embryo of 4.7 mm. vertex-breech length and 4.9 mm. nape length and with 33-35 pairs of primitive segments. (Embryo 137, G. 31, from the collection of the II Anatomical Institute, Berlin, Professor O. Hertwig; slide 9, row 3, section 1.) The section passes through the 11th primitive segment and the 5th mesonephric tubule. The urogenital fold is shown passing on the right into the root of the mesentery, on the left into the lateral wall of the body. A primary genital cell is to be seen in the root of the mesentery.

All the primary genital cells disappear in amniotes ; whether they pass through a latent period to become manifest later as secondary genital cells, though possible, has not been proved.

And now, as regards man it is probable that he also possesses genital cells. The following facts are in favor of such a supposition : in an embryo of 2.6 mm. greatest length and with 13-14 pairs of primitive segments there were in the neighborhood of the cloaca, that is to say, in the region of what was originally the primitive streak, between it and the visceral mesoblast, large, free cells similar to the genital cells of other vertebrates and distinguished from the surrounding cells by the size of their cell-bodies and by possessing yolk granules. In the entire embryo seven such cells were recognizable, all in the immediate vicinity of the cloaca and all between it and the visceral mesoblast, i.e., extraregionai. Id Fig. 609 a two of the primary genital cells of this embryo are shown, one to the right and the other to the left of the cloaca. An embryo of 2.5 mm. greatest length and with 23 pairs of primitive segments had twelve primary genital cells, ail extraregional, situated in the vicinity of the cloaca and the adjoining regions of the body; one of these cells is shown in Fig. 532 6. Finally an embryo of 4.9 mm. nape length and with 33-35 pairs of primitive segments showed typical genital cells in the first to the fifth and in the eleventh to the twelfth body segments, the cells lying partly in and partly below the ccelomic epithelium near the root of the mesentery or in the medial slope of the urogenital fold. I show in Fig. 610 one of these cells in the root of the mesentery.

The smallness of the number of these observations allows of no final conclusion, but they speak in favor of the view that man also possesses primary genital cells.

But even if wandering genital cells are to be recognized in human embryos, there is still the possibility that they may be " strays." Such strayed genital cells do not, perhaps, degenerate, but may develop further, and, above all, divide, forming, perhaps, parent cells for tumors, and especially for teratomata.

What has been said above for the amniotes in general, namely, that all the primary genital cells disappear, holds also for man. In any case, at the time at which the indifferent reproductive glands are formed there are neither extraregional nor regional genital cells.

Development of the Indifferent Reproductive Glands.

The anlagen of the reproductive glands appear within the urogenital folds, whose development and fate have been described above (p. 783-787). A small strip of the epithelium of the genital fold and this alone, forms the parent tissue of the reproductive gland. The epithelium of the urogenital fold usually consists of two layers of cells (Figs. 558, 559, 611) and is spread out uniformly over the entire surface of the fold in embryos up to 4.7 mm. vertex-breech length. In embryos of 5.3 mm. greatest length the epithelium over the summit and the medial slope of the fold, as far as the root of the mesentery, commences to become many layered, that on the lateral slope remaining two-layered or even becoming single-layered (Figs. 563, 612). The region of the manylayered epithelium represents the reproductive gland area, which, as a broad strip, extends throughout the entire length of the medial half of the urogenital fold. Since the thickened epithelium passes over quite gradually into the non-thickened portion, the reproductive gland has no sharp boundaries either medially, laterally, cranially or caudally. To tlio stage of the thickening there follows the stage of the ingrowth of the epithelium into the interior of the urogenital fold, and as the epithelium grows in it compresses the loose mesenchyme tissue of the fold and there is thus formed in the zone of ingrowth a strip of denser mesenchyme tissue, that is everywhere sharply defined from the ingrowing epithelium; in the epithelium itself there are numerous mitoses (Figs. 563, 612). That the epithelium actually grows inwards and not outwards towards the body cavity is shown, in the first place, by the permanent dorsoventral diameter of the urogenital fold and, in the second place, by the displacement of the Malpighian corpuscles of the mesonephros; the ingrowing epithelium pushes the various corpuscles before it ; while the corpuscle in Fig. 548 is still at the surface of the fold, in Fig. 549 it is displaced quite dorsally. The growth of the epithelium of the reproductive gland area forms a solid mass, which has, however, a wavy boundary towards the mesenchyme (Fig. 613). As soon as the growth has reached about one-third of the dorsoventral diameter of the urogenital fold, the formation of the genital fold, described above (p. 785), begins. The lateral and medial grooves, which produce it, press inwards just at the boundary between the mesenchyme and the epithelial growth, and thus the whole genital fold is filled by a perfectly homogeneous mass, composed entirely of derivatives of the coelomic epithelium. We must, consequently, note at this point that everything that is later developed within the genital fold has a common origin from the coelomic epithelium.

Fig. 611 a and 6. — Transverse section of the urogenital fold of a human embryo of 4.9 mm. nape length, at the level of the 14th primitive segment. (Embryo 137, G. 31, from the collection of the II Anatomical Institute, Berlin, Professor O. Hertwig; slide 13, row 1, section 3.) X 150. The urogenital fold is uniformly covered over its entire surface by a one- to two-layered epithelium.

Fig. 612. — Transverse section of the urogenital fold of a human embryo of 7 mm. greatest length. (Embryo Chr. I, from the collection of Professor Hochstetter, Vienna; slide 8, row 10, section 6.) The epithelium on the medial side of the urogenital fold has thickened and forms the anlage of the reproductive gland.

Since the ingrowing epithelial mass displaces the mesonephric tubules, it is from the beginning in intimate relations with these, in the first place, with the medial surfaces of the Bowman's capsules and, in the second place, with the points where the tubuli collectivi bend into the tubuli secretorii ; there is always, however, a sharp boundary between the mesonephric structures and the germinal epithelial mass.

The reproductive gland anlage accompanies the urogenital fold in its gradual growth caudally, but its definitive extent is reached only after sexual differentiation. In its maximal extent it reaches from the sixth thoracic to the second sacral segment. This extent, however, occurs in no one embryo, since degeneration begins at the cranial end before the addition of new reproductive gland elements at the caudal pole is completed. Since an epithelial thickening is present cranial to the sixth thoracic segment in the line of the anlage of the reproductive gland, the anlage may be regarded as reaching almost to the cranial end of the definitive body cavity. But no matter how far cranially we place the cranial end of the reproductive gland area, its cranial pole never reaches the cranial end of the urogenital fold. The caudal pole comports itself differently in the two sexes ; while it reaches the dorsal surface of the genital cord in the female and so comes to lie within the excavatio recto-uterina, in the male it remains above the horizontal portion of the urogenital fold.

In the following table I give determinations of the position of the upper and lower reproductive gland poles in a series of embryos and, when possible, the length of the entire anlage.

Table showing the Growth in Length and the Degeneration of the Reproductive Gland

Embryo.

10 mm.

11 12.5 13 13 14-75 17

18 19.4 21 22 22 22.5 26 28 28 28.5 29 30 30 35 50 60 60 70

9? ' 9 9 ? d 9 9 d d d 9 9 9 9 9 9 9 9 d d

Direction of section.

Right gland.

Left gland.

Cranial. Caudal. Cranial. Caudal.

Transverse

Sagittal Transverse Sagittal Transverse

Sagittal Transverse

Sagittal Transverse

6Th 7 " 8 " v 9 " 10'" 10 " 11 " 1L

12 Th 1L 12/1 Th/L 3L 1" I" 3" 2/3" 3" 4" 3/4" 5" 4"

12 Th 1L 1" 2 3/4" 4" 3" 3" 3/4 " 4/5"

3/4" 3/4" 4" 5" 3" S i( 5" 5" 5" 2Sa 5/1 L/Sa 1/2 Sa 1/2 "

7Th 7 " 8 " 9 " 10 " 9 " 10 11 Th 1011 " 10 11 " 12/1 Th L 2L 12 I Th L i2Th 12 1 Th/L

2L

3L 4/5 " 4/5 " 5 " 5 " 5 "

12 Th 1 L 1 " 2" 3/4" ' 3 a " 3" 3 " 5" 5" 5" 3/4" 3/4" 4" 5 " 3/4"

Length of gland in micra

5 1LSa 1/2 Sa 1/2 "

Right.

Left.

2805

1720 2100 1270

1290 1540

1500 2490 1740

2325

1550 2010 1550

1575

1800 1980

Explanation of Table: The table represents measurements from transverse and sagittal sections. The determinations of the position from sagittal sections are certain and their figures are printed in heavy type; the direction of the transverse sections varied in the various series and according as they were craniodorsal — caudoventral or eaudodorsal— cranioventral they might yield quite different results.

The table shows the gradual extension of the anlage of the reproductive gland; it reaches in maximo from the sixth thoracic to the second sacral segment, that is to say, over fourteen segments, and, eventually, it extends over only three or four, so that it degenerates in ten to eleven segments.

The table also shows the caudally directed growth of the anlage. The caudal pole gradually descends from the 12th thoracic to the 2nd sacral segment (at least in female embryos). The caudal pole — whether the embryo be male or female — lies from the beginning as low as or very frequently lower than in the adult and, accordingly, the so-called internal descent of the testis and ovary vanishes, neither really exists. The cranial pole of the gland does, indeed, change its position, but it changes not because it descends, but because the upper three-fourths of the gland degenerates. What seems to be a descent reveals itself to be a shortening, and we may see from the table that the absolute length of the gland diminishes in spite of its progressive growth in length along with the growth of the entire body. That the ovary eventually becomes rotated so that the cranial pole becomes the lateral and the caudal the medial, this has nothing to do with a descent, for during the rotation the caudal pole does not come to lie at a lower level. As regards the conus inguinalis (see p. 793), the mark for the later abdominal opening of the inguinal canal, the caudal pole of the reproductive gland lies lower than it does in both sexes.

Fig. 613. — Transverse section of the urogenital fold of a human embryo of 11.0 mm. greatest length, 9.0 mm. head-foot length. (Embryo P. 1, from the collection of Professor Hochstetter, Vienna.) X 150. Almost all the parts of a mesonephric tubule are cut. Medial to the tubule is the suprarenal anlage; the mesonephric fold lies in the frontal plane, its summit is marked by the primary excretory duct and a dorsolateral and a ventral surface may be distinguished. At about the middle of the ventral surface is the thickening of the peritoneal epithelium which forms the reproductive gland area. ./This consists solely of ccelomic epithelial cells, which are becoming somewhat loosely arranged. No differentiation whatever of the epithelial mass is to be seen.

The indifferent stage of the reproductive gland lasts only a short time. During it the uniform epithelial mass of the genital fold separates into a superficial epithelium and an epithelial nucleus, a sharp boundary existing between the two. The surface epithelium retains its closed epithelial structure and may consist of from one to at most two layers. The nucleus becomes looser in texture, so that at the end of the indifferent stage its epithelial origin is no longer evident.

The Differentiation of the Reproductive Glands.

In the differentiation of sex we must distinguish between the differentiation of the genital cells into spermatogonia or oogonia and the differentiation of the genital cell mass, the actual reproductive gland, into testis or ovary. This distinction finds its justification, in the first place, in the fact that a reproducive gland may assume the characters of a testis without forming spermatogonia, as is the case with cryptorchid testes, and, in the second place, in the occurrence of malformations, scattered ova in testicular tissue or testicular ampullae in the ovary.

Whether or not a differentiation of genital cells into spermatogonia or oogonia really occurs is yet an open question. It is possible that the future sex of all the genital cells is already determined at fertilization. Every sex-cell, no matter whether it is male or female, possesses a definite force of heredity, which, during the course of development, becomes increased, or perhaps is only made more active, and reaches an optimum stage, after which it perhaps diminishes or become less active. In the optimum stage each sex-cell may possess the power of reproducing the other sex, the ovum males and the spermatozoa females (crossed inheritance). If now the ovum and spermatozoon unite during fertilization, differences in the immaturity, maturity or over-maturity of the two bring about numerous combinations, the stronger partner, that is to say the one which is nearest the stage of maturity, whether it be on the side of immaturity or overmaturity, will dominate in the determination of the sex. And the further possibility is not to be dismissed off hand, namely, that there are sex-cells of different types, male and female ova, etc. All these questions are not as yet ready for settlement, but will play an important part in the immediate future and therefore must at least be mentioned here.

The differentiation of the reproductive gland into testis or ovary is actually to be observed. The opponents of the theory that the sex of the ovum is already determined, overstep the mark when they deny a differentiation of the reproductive gland.

This differentiation consists in the characters of the male gland being developed in embryos of 13 mm. greatest length, at the earliest, while the female reproductive gland of the same age still lingers in the indifferent stage. The two characters of the male gland are: 1. The occurrence of branched, anastomosing epithelial cords, the testis cords. 2. The occurrence of a broad tunica albuginea between the superficial epithelium and the testis cords. It is therefore possible in young stages to identify the male individuals positively, and the females per exclusionem. One may say that the embryo has reached the age when testis cords should be present; they are not present and therefore the embryo must be a female. Such a determination has always chances of error and the doubts of the observer increase the more the embryo reaches the limits of the indifferent stage. Probably no observer will venture to label an embryo of from 13 to 15 mm. greatest length as a female, if he finds in it no testis cords ; the possibility that it is a case of retarded differentiation of a male, can never be excluded. Under these circumstances the small sexual differences in other organs acquire increased importance, because they are positive for the female sex. The formation of accessory funnels on the Miillerian duct seems to occur only in the female sex. Since the principal funnel occurs in embryos of 11 mm. and accessory funnels certainly in those of 13 mm., these latter will supply a valuable means for sex diagnosis, provided that further observations on a larger amount of material show that accessory ostia actually occur only in female embryos and that all female embryos develop them. Accessory tubes are too rare to be of use for sex diagnosis, but, on the other hand again, the presence or absence of the excavatio vesico-uterina is an important sign. The genital cord, by whose fusion with the dorsal wall of the bladder the excavatio is formed, is, indeed, first seen only in embryos of 20 mm., but the fusion of the urogenital folds (by whose union the genital cord is formed) with the posterior wall of the bladder takes place in male embryos before their union. The formation of the excavation in female embryos and its absence in males is a regularly occurring phenomenon.

By employing all these means the beginning of the sexual differentiation may be determined in female embryos at a stage of 18 to 20 mm. in length.

Transformation of the Indifferent Reproductive Gland into the Testis.

The differentiation of the sexless reproductive gland into the testis will be considered first, because in this case the relations are much simpler than in the differentiation into the ovary.

We left the indifferent reproductive gland at the stage of development in which the epithelial mass had separated into a superficial epithelium and the epithelial nucleus. This separation is preserved throughout the eutire developmeut of the testis. Consequently the epithelial nucleus alone is the active layer, it forms the tunica albuginea, the testis tubules and the rete testis; the superficial epithelium remains passive and grows only in proportion with the enlargement of the entire organ, it is for the most part one-layered, several cell layers being present only on the surface opposite the attachment of the mesorchium; the greater portion of its cells are indifferent epithelial cells, among which genitaloid cells are scattered here and there. The superficial epithelium has, therefore, the character of a germinal epithelium formally but never functionally, and since it very early loses the genitaloid cells it is better not to speak of a germinal epithelium, but to adhere to the indefinite name "superficial epithelium." The epithelial nucleus becomes very loose and develops quite suddenly the testis cords, the loosely arranged cells concentrating at certain places, separating incompletely from their surroundings and arranging themselves to form anastomosing cords. It seems as if all the testis cords are formed at once, and the subsequent enlargement of the organ depends entirely on the elongation of the tubules, on their coiling, which is associated with their elongation, and on their increase in diameter. The cords are arranged transversely to the long axis of the testis, their inner ends, being arranged around the point of insertion of the mesorchium — indicating at once, therefore, the future hilus — while the outer ones radiate towards the surface of the testis (Fig. 614). The outer end of each cord is thickened, the inner one is pointed, and all the inner ends unite to form a closely aggregated epithelial mass (Fig. 614), which extends along the insertion of the mesorchium throughout the entire length of the embryonic testis ; this epithelial mass contains the cells that will form the anlage of the rete testis and may therefore be termed the rete blastema. This blastema stands in no relation to the parts of the mesonephros, except for its association with the epididymis tubules. The outer ends of the cords never reach the superficial epithelium, but a layer of at least two or three or usually more cells intervenes, from which the tunica albuginea develops. At first the testis cords have an irregular form on cross section, but later they become distinctly round. Since they all converge to the narrow line of insertion of the mesorchium, they cannot all reach the rete blastema, and therefore they unite by twos or threes in order to attain this connection in common. They are united together by anastomoses which are distinctly arranged in two groups; the one group lies about midway between the periphery and the hilus, the other unites their peripheral ends. These peripheral anastomoses are so constantly arranged in the same direction that they appear like large arches into which the testis tubules open (Fig. 614). The testis cords are so arranged that they leave intervals or intermediate cords between them, of the same width as themselves (Figs. 614 and 615). At the boundaries between the two there is even in an embryo of 21 mm. a distinct formation of connective tissue, consisting of spindle-shaped cells arranged one behind the other and alongside each other, all with their long axes parallel to the long axes of the testis cords (Figs. 614 and 615). Thus there are formed around the individual testis cords and their anastomoses actual connective tissue sheaths; at the hilus all the sheaths unite to form a boundary around the rete blastema, surrounding it in a zig-zag line (Fig. 614) ; this is the anlage. of the mediastinum testis.

Fig. 614. — Transverse section of the testis of a human embryo of 70 mm. head-foot length. (Embryo R. Meyer 267; slide 35, row 3, section 3, from the collection of Professor R. Meyer, Berlin.) X50. Solid testis cords have appeared in the testis and are connected by anastomoses at their middle and at their outer ends. The majority of them have already acquired an investment of young connective tissue. Their inner ends unite at the mesorchium to form the rete blastema and their connective-tissue sheaths form a connective-tissue boundary around the blastema, the first indication of the mediastinum testis. Two distinct layers may be recognized in the tunica albuginea, an outer one of connective tissue and an inner one composed of what are still indifferent epithelial cells. At the boundary between the two are twigs of the ascending branch of the a. spermatica int.

The testis cords consist at first of numerous indifferent epithelial cells with dark, homogeneous nuclei. Between these lie scattered genitaloid cells, which very soon develop to genital cells (Fig. 615). ' These are usually present in embryos of 70 mm. headfoot length, while the genitaloid cells appear simultaneously with the testis cords. The narrow inner ends of the cords contain only indifferent cells ; the tubuli recti are formed from them. The testis cords once formed alter little during the fetal period. The indifferent cells, that at first are altogether without arrangement, gradually acquire one, their nuclei become oval and place themselves radially to the future lumen. Thus there is formed a manylayered epithelium, into which the genital cells enter as spermatogonia (Fig. 616 a). The first lumina appear quite irregularly at the outer ends of the solid testis cords in the seventh fetal month. They arise partly by the migration of the cells from the centre to the periphery, partly by the resorption of the inner cells. From the outer ends the lumen gradually extends towards the hilus, and a second lumen extends from the rete testis "along the tubuli recti towards the first one, the two meet and thus the testis tubules, the tubuli recti and the rete tubules become hollow. All the testis cords are not transformed into tubules at birth. According to the observations of Branca and Basseta (1907) the number of genital cells increases progressively from the fifth month up to birth; after birth all the genital cells disappear and the testis tubules are lined by indifferent cells alone. With the onset of puberty a new generation of genital cells is formed, which then enter upon the formation of spermatozoa. The results of these two authors have been confirmed by Popotf (1909). I have no personal experience on this question and can only point out that in a human embryo of the seventh month obtained by laparotomy the number of genital cells had not diminished and that a well-preserved testis from a newborn child showed a remarkable number of genital cells.

Mesenchyme sheath Fig. 615. — Part of a transverse section through the right testis of a human embryo of 70 mm. headfoot length. (Embryo R. Meyer 267; slide 35, row 1, section 2, from the collection of Professor R. Meyer, Berlin.) The testis cords are shown as they pass into the rete blastema. One sees the structure of the testis cords from indifferent epithelial cells and genital cells, and, further, their passage into tubuli recti. Betweenjthe testis cords are the intermediate cords, filled with cells whose nuclei are vesicular and remarkably poor in chromatin.

The intermediate cords are at first composed of indifferent and genitaloid cells, but later the indifferent cells disappear almost completely; I assume that they are employed in the formation of the connective tissue sheaths. In embryos of 45 mm. greatest length the genitaloid cells are transformed into large pale cells (interstitial cells), which resemble genital cells in shape and size, but their nuclei contain only little or even no chromatin (Fig. 615). These cells gradually disappear after the fifth month, and in fetuses of the seventh month (Fig. 616) they are still present only here and there beside the vessels in the spaces between the testis tubules. After birth an abundant connective tissue develops between the tubules, and, as a result, the number of interstitial cells diminishes still further; after puberty (in the 33rd, 37th and 40th year) they again appear to occur in increased numbers (Popoff 1909).

Fig. 616 a and 6. — Transverse section through the testis of a human embryo of the 7th month. (Embryo obtained by operation. From the collection of the Anatomical Institute, Zurich.) The testis cords have a rounded form; their cells are partly indifferent epithelial cells that are beginning to arrange themselves like an epithelium, and partly genital cells, scattered as spermatogonia among the indifferent cells; in the centre is the remains of an intermediate cord with Leydig cells. The tunica albuginea is already formed of a distinctly fibrous connective tissue.

The septula testis are formed by the thickening of the connective tissue sheaths of the testis cords. After the resorption of the intermediate cords the sheaths of adjacent testis cords come into apposition and then appear as a single structure, which extends radially from the hilus to the periphery. True septula testis are present from the sixth fetal month onwards.

It has been stated that the rete blastema is derived from a cell mass formed by the union of the thin inner ends of the testis cords. It extends throughout the entire length of the testis, leaving only the cranial and caudal poles free. It is certainly a derivative of the epithelial nucleus and is always, in young stages, sharply defined from the mesonephros and its mesenchyme. One sees, indeed, sharply defined mesonephric tubules passing in the region of the epididymis as far as the rete blastema and even producing depressions in it, but one never sees a proliferation arising at their blind ends, from which the rete blastema might be derived. A union between them and the rete occurs later as the urogenital union and will be discussed under this title in a special section. The rete blastema is surrounded by a mantle of young connective tissue on the surface towards the testis (Fig. 614). In the blastema there now occurs at definite points a concentration of the cells and there are formed irregular solid spheres of cells, imperfectly bounded externally. The individual spheres are connected by straight cords and there is thus formed a solid network (rete testis) which gradually becomes more distinct by the remaining cells of the blastema assuming a spindle shape and becoming connective tissue cells (Fig. 615). From these spindle-shaped cells of the rete and from the connective tissue boundary that separates them from the testis tubules the mediastinum testis is developed. Since it is derived from the rete blastema, like this it must extend throughout the entire length of the testis, as it actually does in the testes of older fetuses and in those of new-born children. When, at puberty, an unequal growth of the testis occurs, the mediastinum comes to lie more in the cranial two-thirds of the organ.

The tunica albuginea appears simultaneously with the formation of the testis cords and represents simply the cortical portion of the epithelial nucleus that is not traversed by the cords; it is accordingly formed at first only of scattered, loosely arranged, round cells. Then the cells lying immediately below the superficial epithelium begin to be transformed into connective tissue, whose spindle-shaped nuclei are all parallel to the surface (Fig. 614) . The remains of the epithelial bounding layer that is not converted into connective tissue increases very greatly; thus there is gradually formed an exceedingly thick bounding layer, which step by step from the periphery is transformed into connective tissue. In fetuses of the fifth month there is already a broad connective tissue tunica albuginea.

Malformations of the Testis

Only those malformations that are due to inhibition- of development will be considered, and no mention will lie made of those resulting from pathological processes.

A testis may be lacking, in which case the entire urogenital fold has usually failed to form, and therefore the kidney, ureter, ductus deferens, and suprarenal body of the same side will also be lacking.

The testis may be doubled. Ahlfeld (1880) records only one autopsy in which the duplication was observed; all other cases were determined during life and are therefore not certain. One must guard against errors in diagnosing a duplication; Romanovsky and von Winiwarter (1905) have described a case in which the right and left testes were both situated in the left scrotal sack.

Both testes may come into contact in the median line and may be united together; in such a case they will lie between the bladder and the anterior abdominal wall. In one embryo (R. Meyer, 270) of 60 mm. head-foot length the two testes had their hilus sides turned towards each other, the retia testis being well developed and fused.

The form of the testis may be completely normal macroscopically, while the microscopic examination shows that spermatogonia are completely lacking; the testis was accordingly sterile. This sterility is usually associated with cryptorchism.

The Transformation of the Reproductive Gland into the Ovary.

The indifferent reproductive gland consists of an epithelial nucleus and a superficial epithelium, both sharply separated from one another. But while in the testis this first separation is a permanent one and the. superficial epithelium is reduced to an insignificant investment, this is not the case in the ovary. Between embryos of 70 mm. head-foot length (this limit may perhaps be placed somewhat earlier) and those of 180 mm. head-foot length there is a development period in which the superficial epithelium is sometimes only indistinctly and sometimes not at all marked off from the subjacent epithelial nucleus. Whether this undeniable new union between the two parts is to be regarded as a new impulse by which a second epithelial nucleus is formed from the superficial epithelium, cannot be determined with certainty. The arguments pro and con will be considered later on and it will suffice at this point to indicate the possibility of such an impulse ; for by it the superficial epithelium of the female reproductive gland acquires the significance of a germinal epithelium, and even although it contains no genital cells, yet it has cells which are able to transform themselves into these and into young ova. The superficial epithelium possesses the significance of a germinal epithelium, however, only during the period mentioned above. In embryos of 180 mm. trunk-length, a connective tissue layer, the tunica albuginea, develops between it and the epithelial nucleus, and with this a sharp delimitation of the epithelial nucleus occurs, the tunica forming a partition which excludes any further participation of the germinal epithelium in the formation of young ova ; the germinal epithelium thus again becomes a simple superficial epithelium.

In older fetuses and in new-born children foldings of the surface epithelium occur, but they have nothing to do with the formation of ova, indicating only the completion of the lobed form of the ovary. Solid, hair follicle-like, downwardly growing epithelial cords are formed from the superficial epithelium, but they never extend beyond the limits of the albuginea and again degenerate without effect.

The actual reproductive gland producer is the epithelial nucleus. In embryos of 22 mm. greatest length it is composed of indifferent cells with sparingly scattered genitaloid and genital cells and it shows this composition up to embryos of 50 mm. headfoot length (Fig. 617). The nucleus fills the entire space enclosed by the superficial epithelium and also projects like a knob into the mesovarium. This knob consists of indifferent cells and very few genitaloid cells; it forms the rete ovarii and therefore deserves the name of rete blastema. Between the stages of 50 mm. and 80 mm. head-foot length the epithelial nucleus begins to become looser, starting from the rete blastema which remains unaltered, and one can distinguish a clear medullary zone from a denser cortical zone (Fig. 618). The loosening may occur under most varied forms ; the special example represented in Fig. 618 shows a structureless medullary zone and a cortical zone that is incompletely broken up into anastomosing cords.

Fig. 617. — Transverse section of the ovary of a human embryo of 50 mm. head-foot length. (Embryo R. Meyer 272; slide 2, row 1, section 2, from the collection of Professor R. Meyer, Berlin.) X ca. 75. The section shows the triple division of the mesonephric fold into a tubal portion with the Miillerian duct, a gland portion with the primary excretory duct and mesonephric tubules and, finally, a very thin mesentery portion. In the ovary the superficial epithelium is distinctly separated from the epithelial nucleus. The nucleus is indistinctly dividing into a medullary layer, with numerous genital cells, and a cortical layer, poor in genital cells. Numerous trabecular of young connective tissue occur in the epithelial nucleus. The rete blastema is partly in the mesovarium and contains genitaloid cells.

In the epithelial nucleus three concomitant but independent processes occur: (1) the ingrowth of connective tissue and vessels from the hilus towards the periphery; (2) the conversion of most of the genitaloid and indifferent cells into young ova, and (3) the new formation of the epithelial nucleus at the periphery, that has already been mentioned. While in the epithelial nucleus of the testis active testis cords appear, tins is never the case in the ovary, its epithelial nucleus forming no cords, but becoming split up into portions of the most varied forms by ingrowing connective tissue. The first traces of the connective tissue are seen in embryos of 28 mm. and are at first without regularity (Fig. 617). Then they gradually arrange themselves so that one can recognize a central investment around the rete blastema, a sort of mediastinum ovarii, and bands radiating out from this, the septa ovarii (Fig. 619). The septa traverse the central two-thirds of the transverse section and then begin to develop lateral branches and send out several apical shoots. These branches and shoots unite with those of neighboring septa so that a network is formed, which has wide and coarse meshes in the so-called medullary zone and narrow and fine ones in the so-called cortical zone; the fine cortical network eventually extends to the under surface of the superficial epithelium. The form of the meshes is exceedingly variable, the one extreme — very rare and only observed in one case — shows long rectangular meshes containing long, slender epithelial cords, such as are seen in the ovaries of other mammals, which unite to form a trabecular network like that occurring in the medullary cords of lymph-nodes; the other extreme — very frequent — shows polygonal meshes containing plump masses of the epithelial nucleus of most varied forms and all, of course, in connection. Between the two extremes are all sorts of intermediate forms, one of which is shown in Fig. 620.

Fig. 618. — Transverse section of the left ovary of a human embryo of SO mm. head-foot length. Embryo 266; slide 31, row 1, section 2; from the collection of Professor R.Meyer, Berlin.) X40. The section passes longitudinally through the upper horizontally bent part of the ovary and hence its apparent breadth. The epithelial nucleus has divided into medullary and cortical zones and the rete blastema is distinctly at the attachment of the mesovarium. Resides the Mullerian duct there is an accessory tuba, the portion of the mesonephric fold that contains it has already separated and we have before us the anlage of the appendix vesiculosus. The mesonephros is in complete degeneration; the section cuts it in the region of the paroophoron, and one may see how far the tubuli collectivi have already separated from their tubuli secretorii and Malpighian corpuscles.

Almost simultaneously with the connective tissue the first vessels appear at the hilus— probably in situ — and they grow towards the periphery more slowly than the septa. Since the vessels have a definite diameter and since, also, they produce a stronger development of the surrounding connective tissue, there is for a time a rather sharp distinction in the mesh-work of the connective tissue between a medullary zone with vessels and a cortical zone without them. The boundary between the two may coincide with that between the medulla and cortex, as this is drawn out by the loosening up of the medullary cells (Fig. 618) or, what is most frequently the case, there is no coincidence. It follows that a precocious differentiation into medullary and cortical zones is somewhat doubtful, since different processes — we shall hear of others — cause a separation into different cortical and medullary zones.

The tunica albuginea is also a formation of the septa ovarii. One sees (Fig. 620 a) how prolongations of the septa reach the superficial epithelium and then begin to creep under it. The more prolongations reach the surface the more compact becomes the layer formed by them beneath the superficial epithelium and the more distinct does the tunica become. It is difficult to assign a definite time for the beginning of its development. I have chosen the stage of embryos of 180 mm. trunk length, because in these the connective tissue prolongations first reach the superficial epithe

Fig. 619 a and b. — Transverse section of the ovary of a human embryo of 80 mm. trunk length, 100 mm. head-foot length. (Embryo R. Meyer 151, from the collection of Professor R. Meyer, Berlin.) • 230. In Fig. 619 b the portion drawn is shown in the entire transverse section. The ovary has divided into the superficial epithelium, the neogenic zone, the peripheral and the central portions of the epithelial nucleus. Around the rete ovarii a kind of mediastinum has formed, from which a septum ovarii radiates towards the periphery. The neogenic zone consists of indifferent, genitaloid and genital cells, closely packed. The peripheral zone of the epithelial nucleus is composed of the same kinds of colls as the neogenic zone, but the cells are much more loosely arranged. In the rete blastema there are b still genitaloid cells.

Hum. An actually closed layer is formed only later. Fig. 621 shows the tunica of an embryo of eight months ; one sees the connective tissue septa entering it perpendicularly and then bending in arches to a horizontal direction. It is noteworthy that the folds of the ovarial surface, mentioned above, occur at the places where a perpendicular connective tissue strand enters the tunica.

The transformation of the indifferent and genitaloid cells into genital cells and young ova begins in the immediate vicinity of the rete blastema and proceeds thence towards the periphery. This transformation process enlarges the poorly staining cell bodies and makes the nuclei paler, so that the central portion of the medulla appears very pale in comparison with the peripheral portions and with the cortex. If the transformation of the indifferent cells begins with the development of the connective tissue and vessels and progresses synchronously with these towards the surface, a marked distinction is established between medulla and cortex: the "medulla' is traversed by a broad, vascular connective tissue network, in whose coarse meshes are pale, young ova; the "cortex" is traversed by fine connective tissue trabecule, whose narrow meshes contain dark indifferent cells. This synchronous progress is, however, of rare occurrence. The two processes are usually not only unequally advanced, but one of the processes may show very different degrees of progress at different parts of the same ovary; there is thus brought about so much confusion that the determination of a boundary between the medulla and the cortex becomes impossible.

Fig. 620 a and 6. — Transverse section through the ovary of a human embryo of 180 mm. trunk length, 270 mm. head-foot length. (Embryo R. Meyer 152, from the collection of Professor R. Meyer, Berlin.) X 230. The trabecular of connective tissue have grown throughout the entire ovary and have reached the superficial epithelium. Towards the centre the meshes are wide, towards the periphery narrow. The three zones — neogenic, peripheral, and central — can still be recognized, but the superficial epithelium is indistinctly marked off from the neogenic zone. As a glance at the orientation figure shows, only the outermost layer of the central zone is drawn. In the meshes of the connective-tissue strands are ova spheres, the most of whose ova are degenerating. In the neogenic zone are numerous genitaloid cells, in the central zone already circumscribed primary follicles. In all lumina of blood-vessels a "G " is inscribed.

At first the ovary grows rapidly. A comparison of the transverse sections 619 and 620, both drawn under the same magnification, shows this at once; so much of Pig. 619 as extends from the surface to the transverse line corresponds to the entire Fig. 620. While the length of the body has increased in the proportion of 4:9, the ovary has grown in the proportion of 7 : 22. The growth of the testis is quite regular, the tubules elongating and enlarging and also the intermediate cords and the tunica albuginea. The ovary grows irregularly, the cortical portion of the epithelial nucleus alone undertakes the new formation of indifferent and genitaloid cells, the medullary portion remaining unaltered; a young cortical zone is thus developed over the old epithelial nucleus. Whence this young cortical zone comes I cannot say with certainty ; there are, however, two possibilities which may act singly or together. On the one hand, the indifferent and genitaloid cells of the cortical zone of the epithelial nucleus divide and thus, notwithstanding the progressive transformation of the cells of the nucleus into young ova, a new neogenic cortical zone may be continually reformed, just as the neogenic zone of the growing metanephros continually regenerates, notwithstanding its transformation into uriniferous tubules. In favor of this mode of development are the mitotic figures and the gradual transition of the cortical zone of the epithelial nucleus into the neogenic zone. The second po>-i bility has been referred to above; just as the ccelom epithelium grows in to form the epithelial nucleus at the beginning of the development of the reproductive gland, so the process may repeat itself, the superficial epithelium alone forming the neogenic zone. In favor of this mode is the temporary impossibility of distinguishing the epithelium from the neogenic zone ; against it is the striking absence of mitoses in the epithelium. I assume, therefore, that the principal growth takes place in the cortical layer of the epithelial nucleus and that only a very slight addition, if any, takes place from the superficial epithelium. But no matter how the development of the neogenic zone occurs, cords are never formed, the zone always forms a single mass. I come to the conclusion, therefore, that Pfliiger's cords do not occur in man.

As soon as the neogenic zone has appeared it becomes modified by the two processes described above ; the connective tissue divides it into irregular masses of cells connected with each other and the cells for the most part become transformed into ova. According as the two processes develop quickly or slowly and according as they proceed continuously or are interrupted by pauses, we find a broad or a narrow neogenic zone.

While epithelial material is being newly formed at the periphery, a degeneration of genital cells and young ova takes place at the centre from the third month onwards. In the destruction of both kinds of cells the nucleus is first broken down, its chromatin becomes massed and the nuclear membrane disappears (Fig. 620). The first sight of such a. picture as is shown in Fig. 620, when whole meshes are filled only with degenerating cells, suggests that the process is not a normal one ; it is only when one finds such pictures over and over again and finds no ovary of the last fetal months without degenerating cells that one becomes convinced that the destruction is a normal process. It progresses from within outwards and produces decided differences between the medulla and cortex, since with the destruction of the ova in the medulla there is a formation of the stroma ovarii. When, that is to say, all the epithelial cells in a mesh have been destroyed, the connective tissue grows strongly and fills the entire space, and when adjacent meshes have become transformed into connective tissue there is formed a closed connective tissue nucleus, the stroma ovarii. Gradually the connective tissue consumes in this manner so much of the epithelial nucleus that the old one (medulla and cortex) completely vanishes, as is shown in Figs. 617 and 618 ; layers of the neogenic zone probably undergo the same fate; the cortex of the mature ovary is, accordingly, a product of the neogenic zone alone. In different embryos cortex should not be compared with cortex or medulla with medulla off hand. When statements occur in the literature to the effect that "the medulla degenerates," "the medulla persists in certain cords," "the medulla forms Graafian follicles," they can be confirmed, with an addition, however, to the effect that quite different generations of medulla are under consideration and that in the "medullary portions" cortical portions of younger embryos are included.

All the cell masses and cords in the region of the stroma ovarii are not destroyed *by this tissue: some always persist and deserve the name of "cords," since they are completely closed, elongated, sometimes straight and sometimes curved structures. They are always situated close below the rete and therefore are derived from the central portions of the epithelial nucleus; accordingly they are correctly termed "medullary cords." These eventually degenerate, the last disappearing usually towards the close of fetal life, although scattered cords are frequently still to be found in the first year of life and very rarely in the adult ovary.

The extension of the destruction process to young ova varies, so that one finds ovaries that are rich in ova and others that are poor, the number of ova showing very decided differences.

When the formation of cortical substance ceases and the process of destruction, whose final result is the formation of the stroma ovarii, comes to a standstill, the remaining cortical layer becomes broken up by outgrowing connective tissue. In this process also a great variability occurs. The connective tissue may divide the cortex into individual ova (Fig. 621) or into larger masses of ova. Each ovum surrounds itself with an investment of cells, the follicle cells, which are in no way to be distinguished from the surrounding connective tissue cells. The ovum together with its follicle cells form what is termed a primary follicle.

In the first year a part of the primary follicles begin to grow and to be- converted into Graafian follicles ; scattered Graafian follicles are to be found even in the ninth month. In the second year some of the follicles are fully formed and contain apparently mature ova. In the third year all the characters of the adult ovary are present and from that time onwards there is no further histological differentiation but merely an increase in size (Runge 1906). The question arises as to the fate of these precociously mature follicles. Runge found in a new-born child (in one case only) a true corpus luteum ; it is accordingly possible that a newborn child may extrude sexually mature ova in the normal maimer, but the usual fate of these precociously formed Graafian follicles is that they become transformed into atretic follicles or undergo cystic degeneration.

The rete blastema, whose formation was described above, remains unaltered for a long time. It always lies partly in the ovary and partly in the mesovarium. Genital and genitaloid cells gradually disappear from it completely and it then consists only of closely packed indifferent cells; the last genital cells in the rete were seen in an embryo of 55 mm. head-foot length. While the rest of the epithelial nucleus is divided up by strands of connective tissue, this is not the case with the rete blastema. It is, indeed, traversed by some blood-vessels, but it always remains a remarkably compact mass and is always delimited, though not sharply, from the surrounding tissue. Only towards the medulla of the ovary is it connected with the medullary cords (Fig. 618). In it there arise, in embryos of 60 mm. head-foot length at the earliest, net-like branched areas, in which the nuclei are even more closely packed than in the surrounding tissue; these are the rete cords, which are completely solid and are never sharply defined from the surrounding rete blastema. These cords are connected both with the medullary cords and with the tubules of the epoophoron (see under Urogenital Union). Towards the end of the fetal period distinct lumina appear in the solid rete and tubules lined with a single layered epithelium are formed. These tubules may persist throughout life (von Franque, 1896), but they always show a tendency towards cystic enlargement.

Fig. 621 a and 6. — Transverse section of the ovary of a human embryo of the 8th month. (From the collection of Professor R. Meyer, Berlin.) X 230. The superficial epithelium, tunica albuginea, cortical layer, and medullary layer are all well marked; of the last only the outer third is shown, : as may be seen from the orienting figure. In the cortical layer are isolated primary follicles, whose epithelium is'formed by genitaloid and indifferent cells and, on account of the latter, cannot be separated from the adjacent connective tissue. The superficial epithelium is sharply defined from the tunica albuginea and prevents thickening where connective-tissue strands ascend vertically towards it. In the stroma there are still remains of "medullary cords" and masses of genital cells.

The form of the fully developed ovary is very variable, since it must adapt itself to the space left free by the coils of the intestine. Even more than in the adult the position of the portions of the intestine with reference to each other and to the abdominal wall is very variable in older embryos, and this variability will produce a variability in the space available for the ovaries. The usual form of the ovary may best be compared with a three-sided prism, whose principal surfaces are pointed. The ovaries are sometimes short and thick, sometimes long and slender, sometimes straight, sometimes angled, sometimes twisted spirally. When the ovary has a sagittal direction one can distinguish a dorsal, a ventral, and a medial edge; when it is rotated into a horizontal position the dorsal edge becomes caudal, very rarely cranial, the ventral one cranial, very rarely caudal, the medial one dorsal, very rarely ventral. The three edges are generally sinuous and especially in the case of the medial, later the dorsal ones do not always extend throughout the whole length of the organ. Corresponding to the position of the edges one can recognize a lateral (later dorsal), a dorsal (later caudal) and a ventral (later cranial) surface. In the middle of the lateral surface the mesovarium is attached. Since the dorsal and ventral edges overhang slightly, the lateral surface is sometimes strongly, sometimes weakly concave (Figs. 619 b, 620 b, 621 &). The three edges are sometimes smooth, at other times they are sparingly or frequently notched. The notches are for the most part quite superficial, rarely deep, in which case they may extend for a considerable distance on the surfaces. They are either arranged parallel to one another or may radiate somewhat towards the hilus, but they are never so arranged as to permit of the derivation from them of that form of ovary which, on account of its likeness to the surface of the cerebral hemisphere, is termed an ovarium gyratum. This is not a persistent embryonic form, but one sui generis.

The position of the ovary in older embryos is, like its shape, very variable. The two organs may be quite symmetrical, but they may both be crowded to the same side and assume various positions. In the displacement of both ovaries to the same side portions of the intestine (rectum and sigmoid flexure), excessively filled with meconium, play an important part. The rotation of the ovaries into the horizontal position occurs at very different periods, frequently one still finds the sagittal position in new-born children, frequently the rotation is completed in the fourth fetal month. It is quite possible that each ovary again becomes upright after the rotation and is later again brought into the horizontal position, and these changes may be repeated several times. Until the first year of life both organs lie for the most part in the false pelvis or above the entrance into the

908 true pelvis; usually they are symmetrically placed, the most striking asymmetry being produced by one ovary having descended into the true pelvis, with its long axis placed sagittally; it is usually the left ovary that undergoes this displacement. According to the position of the coils of the small intestine one finds the ovaries at one extreme pressed against the posterior abdominal wall and at the other against the anterior wall. Between the two extremes all intermediate positions occur; usually they lie in the dorsal portion of the false pelvis, as is to be expected from their development.

The relative position of the ovary and tube also varies. If both are still sagittal, the ovary usually lies medial to the tube. As a result of the degeneration of the mesonephros, especially in its cranial portion, the tube, however, acquires a very broad mesentery, and may be pushed by other viscera over the ventral surface of the ovary to its medial side, in which case the ovary lies in a bay of the plica mesonephridica, completely surrounded by the mesosalpinx. If the rotation to the horizontal position is completed, the tube usually lies ventral and somewhat caudal to the ovary. If the tube and ovary are forced strongly ventrally, the ovary may be pushed over the upper edge of the ligamentum latum and come to lie in the excavatio vesico-uterina, instead of in the excavatio rectouterina; the tube then lies dorsal to the ovary.

The following table shows data concerning the growth of the ovary. One may see from it that a slow and continuous growth is maintained throughout the entire embryonic period. After birth the rate of increase seems to be somewhat accelerated, but it diminishes again, to increase a second time at puberty. A difference in growth between the left and right ovaries can hardly be perceived.

Greatest diam. of head.

Right ovary.

Left ovary.

Comparison between Vertexbreech length.

Breadth.

Length.

Breadth.

Length.

Length.

Breadth.

R.

L.

R.

L.

50.0 125.0 138.0 156.0 173.0 190.0 223.0 235.0 260.0 272.0 305.0 347.0 355.0 386.0 402.0 3 weeks 6 « 6 " 10 " 2 months 3 7 " 15 If vears 4 " 5* " 14 " 42.6 123.0 115.0 131.0 163.5 175.0 162.0 190.0 213.0 213.0 238.0 273.0 324.0 301.0

0.9 1.2 1.6 1.9 3.0 2.9 2.9 4.2 3.6 3.0 3.0 3.5 4.0 5.1 5.05 5.0 7.5 7.0 6.6 6.0 5.9 9.0 7.0 10.0 11.1 11.9 1.9 5.9 5.0 7.2 9.0 7.7 10.5 10.0 11.1 10.0 9.9 10.8 14.0 11.5 10.5 17.0 15.0 18.0 14.0 14.5 15.5 15.5 18.0 20.0 27.0 29.0 26.5 0.9 1.5 2.0 2.0 2.1 3.0 3.8 4.0 3.5 3.9 4.9 5.2 3.0 3.0 5.0 7.0 8.0 4.0 5.0 4.5 9.0 8.5 12.7 9.1 12.0 2.5 4.1 5.0 7.1 7.8 9.1 12.0 11.4 9.2 10.9 8.5 9.9 9.9 12.0 14.0 14.7 17.0 16.0 13.0 14.7 18.1 19.5 15.0 23.2 26.1 29.5 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +

Explanation of table. — The measurements are all given in millimeters, breech length is measured along the nape and the back.

Malformations of the Ovary

A complete absence of both ovaries is very rare, and, according to Nagel (1897) and.Gebhard (1899), occurs only in monsters incapable of maintaining life. Menge reports a case during life in which it was readily seen that in an otherwise normal person both ovaries, the uterus and the vagina were wanting. Such cases must, however^ be accepted only with caution, since observations made upon the living body can never give assurance that the case is not one of secondary atrophy. Even an autopsy is not always sufficient to demonstrate this process; I may recall the case recorded by Braun (1896) in which several years after ovarial atrophy not even the slightest trace of scar tissue was to be found at the point of constriction (Menge, 1910).

The absence of one of the ovaries is also very rare, yet it has certainly been observed, usually, but not always, with a concomitant absence of the Mullerian duct of the same side (Kossmann, 1899). Hypoplasias are more frequent and are associated with a hypoplastic condition of other portions of the genital apparatus (Menge, 1910). Supernumerary ovaries occur; von Winckel has described a third ovary in the vesico-uterine fold, together with a third tube.

Divided ovaries — ovaria partita, usually bipartita — are usually secondary formations; they may be derived from the lobed condition.

The Development of the Blood=Vessels of Both Reproductive Glands.

The aorta develops on either side about 30 mesonephric arteries, and of these 9-11 persist on each side (see p. 820). The majority of these persisting arteries form the rete arteriosuin urogenitale in the angle between the metanephros, mesonephros and reproductive gland, and this rete unites with vascular lumina which are formed independently inside the three organs. The first vascular lumina in the reproductive gland occur in an embryo of 18 mm. greatest length. Of the 9-11 persisting mesonephric arteries one, and that the lowest, is destined to become the a. spermatica interna, the rest become obliterated or else one or two persist as aa. spermaticae accessorise. Up to this point the development is exactly alike in the two sexes (see p. 822), but now it begins to differ. In the male (embryo of 60 mm. head-foot length) the artery passes from its origin from the aorta downwards in the retroperitoneum, meets the mesonephric fold a little below the cranial pole of the testis, traverses it obliquely cranio-dorsally to caudo-ventrally, and, finally, reaches the surface of the testis at the level of its lower pole. Here it lies in the tunica albuginea, runs as a rule on the dorsal surface around the caudal pole, reaches the lateral surface and again ascends cranially, lying always in the albuginea. Thus the artery forms an actual loop, between the limbs of which the testis is situated. A medial descending and a lateral ascending limb may be recognized ; with the exception of a small lateral twig to the caudal extremity of the testis, the descending limb is without branches, and it is important to note that it also passes the hilus without branching. The ascending limb divides into several branches, in the present case into two ventral and one dorsal, which ascend respectively on the ventral and dorsal surfaces of the testis. From these three branches the actual nutritive vessels of the testis arise, penetrating between the testis tubules and branching among them as far as the mediastinum. In this they are collected into venous stems, pass out as veins from the hilus into the mesorchium and there open into the v. spermatica interna. The branch given off by the descending limb at the lower pole of the testis behaves differently on the two sides, but it eventually ascends to the middle of the testis and participates in the supply of the testis cords. The three or four ascending arteries run upwards in the tunica albuginea, and mark out a tunica vasculosa (Fig. 614).

The a. spermatica of female embryos has at first the same path as in males, but very early it meets the mesovarium and there gives off branches to the hilus of the ovary until it is exhausted. It never reaches the caudal pole of the ovary.

The very different relations of the aa. spermaticse in the two sexes furnish a means for determining at once the sexual character of a reproductive gland. If the artery consists of merely a descending limb and this runs only along the hilus of the gland, this is an ovary; if the artery consists of a descending and an ascending limb and the latter lies on the surface of the gland, this is a testis.

The development of the vv. spermatica? is so intimately associated with the development of the veins in general that it may better be considered in that chapter.

Comparison of Testis and Ovary.

The testis and ovary agree in that the superficial epithelium, the so-called germinal epithelium, either plays no part or, as in the ovary, only a small, diminishing one. The epithelial nucleus has the principal role in the development of both glands. But while in the male the nucleus is developed quickly and at a single stroke and then enlarges by the growth of all its parts, in the female its formation is slow and by steps; special germinal zones may be recognized in it and their growth is independent. The testis forms from the epithelial nucleus active testis cords, which only secondarily become surrounded by a connective tissue investment; no cords are formed in the ovary, but its epithelium is passively divided by inwandering connective tissue into a plump network of masses and cords of cells, whose individual constituents show very varied forms. The division goes so far that eventually each genital cell, or the young ovum developing from it, forms a unit by itself. In the testis the epithelium, in the ovary the connective tissue determines the form of the epithelial constituents. The testis tubules are formed mainly from indifferent cells, the form of the cord and later of the tubule is determined by these alone, the genital cells or spermatogonia are intrusions which have no influence on the form. The primary follicles of the ovary have their form, determined by the ova, the indifferent cells play, as follicle epithelium, a role which is unimportant so far as the form of the follicle is concerned. In Graafian follicles, it is true, the follicle epithelium has a dominant influence, but we cannot take these structures for comparison, since they are secondary formations, without homologues in the testis. We arrive, then, at this conclusion: testis cord and primary follicle, the foundation stones of the two sexes, arise from the same parent tissue, but follow very different developmental paths. The testis cords have no homologues in the female reproductive gland.

The tunicae albugineae may be regarded as homologous, in spite of the difference in their mode of formation, both are formed ultimately from the outermost layer of the epithelial nucleus. The rete testis and the rete ovarii are completely homologous, both as regards their anlage and their further development.

The Development of the Female Ducts.

THE EARLY DEVELOPMENT OF THE MULLERIAN DUCT.

As an efferent path for the products of the female reproductive gland a round duct, the Miillerian duct, is formed in both sexes on either side of the body ; it attains complete development only in the female and undergoes degeneration in males while they are still in the fetal period. According to its development the duct is divisible into a very short cranial portion (ostium abdominale tubae) and a very long caudal portion (the tube proper). The ostium abdominale is formed by an invagination of the ccelomic epithelium into the summit of the urogenital fold, the tube proper by an independent outgrowth of the blind end of this invagination. From the beginning the Miillerian duct lies in the secondary summit of the urogenital fold (p. 784), lateral to the primary excretory duct. This summit portion of the fold prepares itself for the reception of the Miillerian duct by beginning to separate from the rest of the fold as the tubar portion, and its epithelial covering becomes higher, at least in the region of the thoracic segments. The first anlage of the ostium abdominale is noticeable in embryos of 10 mm. greatest length as a circumscribed thickening of the epithelium at the summit of the urogenital fold on a level with the third thoracic segment; this place is termed the funnel area; in an embryo of 11 mm. greatest length the area has extended into the region of the fourth thoracic segment. The funnel area is situated immediately below the dorsal limb of the pleuro-peritoneal membrane at the opening of the pleuro-peritoneal duct into the abdominal cavity. In an embryo of 11 mm. greatest length a deep groove (Fig. 622) appears in the region of the posterior part of the third and in the whole of the fourth thoracic segment, in the middle of the transverse section of the funnel area, projecting into the subjacent mesenchyme of the urogenital fold; the groove presents, therefore, a dorsal and a ventral lip. The deeper posterior part of the groove closes to form a tube by the ventral lip growing toward and fusing with the dorsal lip. The tube thus formed separates from the epithelium of the urogenital fold; and the anlage of the cranial portion of the Mullerian duct is thus completed. When fully formed it is cornet-shaped, the opening of the cornet corresponding to the cranial portion of the groove which remains open, while its tip corresponds to the closed portion, separated from the coelomic epithelium; the opening of the cornet into the abdominal cavitv is termed the ostium abdominale tubse. The separated apex of the cornet lies in the immediate neighborhood of the primary excretory duct, between it and the coelomic epithelium ; it has, however, no connections with either.

Fig. 622. — Transverse section of the urogenital fold of a human embryo of 11 mm. greatest length and 9 mm. head-foot length. (Embryo P. 1; slide 13, row 3, section 5; from the collection of Professor Hochstetter, Vienna.) The section passes through the middle of the third thoracic foramen intervertebrale. The mesonephros is undergoing degeneration in its cranial portion, and the urogenital fold has therefore diminished in size and appears to be triangular; its covering of coelomic epithelium is greatly split on the ventral surface. At the apex of the triangle the coelomic epithelium of the urogenital fold is greatly thickened (funnel area) and invaginated. This funnel-like invagination is the anlage of the ostium abdominale of the Mullerian duct. The funnel is placed frontally, and one can therefore distinguish a dorsal and a ventral lip. The latter has grown out and is curved into a hook to close the opening of the funnel. Into the primary excretory duct the degenerating tubulus collectivus of the second mesonephric tubule opens, beside it and medially is the tubulus secretorius of the third tubule, then the third Malpighian corpuscle, and, finally, the second Malpighian corpuscle.

In addition to the principal funnel, whose development has just been described, two to four accessory funnels are also formed as solid growths of the funnel area in the neighborhood of the principal funnel. These growths also separate partly from the parent tissue and their ends, which have thus become free, unite with the principal funnel; when this union is completed they become hollow. The accessory funnels are situated both on the dorsal and on the ventral side of the principal funnel; the scattered dentations of the margin form the anlage of the fimbriae. The first distinct fimbriae occur in embryos of from 28 to 30 mm.; in those of 60 mm. head-foot length the fimbria ovarica appears as a grooved projection directly caudally towards the cranial pole of the ovary. These accessory funnels I have so far found only in female embryos, but in these they occur regularly; in the males they are wanting and the male tubes develop no fimbriae. From accessory funnels the accessory tubes are to be distinguished; these occur in female embryos, but not regularly. They are usually situated somewhat more caudally than the accessory funnels and — what is characteristic — they never unite during development with the principal funnel or principal tube; they end blindly after a short course. The portion of the urogenital fold that contains such an accessory tube may be grooved out from the rest of the fold so that it remains in connection with it only by a narrow stalk (Fig. 618) ; in this way the appendix vesdculosus, the hydatid of Morgagni, is formed. The number of accessory tubes in the best example was four; they also appear to occur only in female embryos. All accessory tubes, with the exception of that one which forms the appendix, appear to vanish in the course of development. Concerning the origin of a double tube and concerning the tubar appendages such as have been described in the adult (Kossmann. 1895) I have no personal knowledge.

At the very time when the posterior end of the groove is separating from the epithelium it begins to grow out caudally, and in this process we come to the development of the distal portion of the Miillerian duct; it is formed by the gradual outgrowth of the tip of the cornet. The path that it will follow is already laid out ; it must pass, caudally between the ccelomic epithelium laterally and the primary excretory duct medially, as if between two bars. Its growth results entirely from its own forces, the outgrowing tip being always free and connected neither with the ccelomic epithelium nor .the primary excretory duct ; only poorly preserved or poorly fixed preparations could deceive one on this point. The growth depends on the increase of cells partly along the entire extent of the duct, as is shown by the mitoses, partly at the outgrowing end, which is frequently swollen and presents all the peculiarities of a so-called cone of growth. The reader will find the various stages of growth in the following table .

The lumen extends continuously from the funnel into the caudal portion of the duct, and follows the growth of the duct so closely that all that is ever solid is a small terminal portion. In this way the larger proximal portion of the duct becomes hollow; in its distal portion the lumen arises in loco, and is, accordingly, frequently discontinuous.

When one considers that the duct rapidly grows from the third or fourth thoracic segment to the sacral region it is clear that one has to deal with a decided increase in length; from the third row of the table one sees in fact the rapid increase. In an embryo of 12.5 mm. greatest length the length of the duct is 330 p, in one of 17 mm. it is already 1440 or 1220 ^ In this latter embryo the duct has just reached the level of the second lumbar vertebra. Notwithstanding that the duct has still to grow over four or five vertebrae and these the highest in the entire vertebral column, there is from this time onwards a remarkably slight increase in length. This fact finds an explanation when one compares in the two first columns of the table the position of the ostium in successive stages. It lies at the beginning of its development Table showing the Growth of the Mullerian Duct.

Length of Right.

Left.

Right, absolute length in mikra.

Left, absolute length in mikra.

the embryo (mm.) Beginning.

End.

Beginning.

End.

12.5 13.

13.5 14.75 17.

18.

19.4 21.

22.

28.5 30.

35. 50. 60. 70.

7Th. 9 " 9 " 10/11 " 10 " 11/12 " IL.

12 Th.

2L.

2/3" 2/3" 4 " 4 " ISa.

8Th. 11/12 " 11/12 " 2L.

2 " 3" ISa.

Mailer's tubercle it it ii it it 8/9 Th.

11 •' 10" 10 11 " 10 11 " 1 L.

3 1, 45 " 5 " 5 " 11 Th. IL. 1" 2" 2" ISa, Midler's tubercle it u a 330 990 1440 1420 1785 1560 1035 1220 1270 1740

in the third and fourth thoracic segment and at the close of development it is in the fourth lumbar segment, having wandered downwards through twelve segments. Since the ostium is always the same one, no traces of a new structure being shown, this wandering must be regarded as a true descensus. This is the more peculiar as the similar wandering seen in the cranial part of the mesonephros and of the reproductive gland is, in both these organs, due to the degeneration of the cranial portion. The wandering of the ostium abdominale is, of course, a passive process, and is probably due to three causes: First, the upper portion of the Mullerian duct no longer increases in length, it therefore lags behind in the total growth. Secondly, the ostium is attached to the eras of the diaphragm; since this descends, the ostium must follow it. Thirdly, the cranial part of the urogenital fold degenerates when its principal contents, the mesonephros and the reproductive organ, vanish; the upper portion of the Mullerian duct then hangs by a very loose fold and may therefore curve and bend and thus bring the ostium into a lower position. The end of the tube never quite reaches the cranial pole of the ovary, it always projects beyond it even in the adult; in the new-born child the cranial or lateral pole of the ovary still lies frequently at about the middle of the tube.

The Miillerian duct grows along the primary excretory duct. This makes a double bend in its course (Fig. 552) ; we may distinguish (1) a cranial vertical portion, (2) a horizontal one, and (3) a lower vertical one. The bends of the excretory duct appear before the Miillerian duct in its caudal growth has reached the places where the bends occur; Fig. 558, for example, shows the tip of the tube not yet arrived at the first bend of the excretory duct. As the Miillerian duct now grows onwards it lies ventral to the horizontal portion of the excretory duct, sometimes more cranially, sometimes slightly caudally. As it grows along the horizontal portion of the excretory duct the union of the two urogenital folds to form the genital cord occurs, and the Miillerian ducts now transverse this, situated between the two primary excretory ducts, lying now medial to these and close together. The space between the two primary excretory ducts is, as is shown in Fig. 552, exceedingly small. The two Miillerian ducts, accordingly, grow downwards close together until they reach the Miillerian tubercle. When they have arrived at this level they suddenly bend almost at right angles and run horizontally on the wall of the urogenital sinus. This horizontal portion is usually thickened, and without a lumen ; very frequently it is filled from the beginning with a vesicular epithelium. Miiller's tubercle is at first merely the entire dorsal wall of the vesico-urethral anlage at the level of the orifice, and projects into the lumen of this. The vesico-urethral anlage then grows greatly in breadth by taking up into itself the two cloacal limbs (see p. 878), and thus Miiller's tubercle becomes marked off as a special projection within the dorsal wall of the vesico-urethral anlage.

The left Miillerian duct reaches Miiller's tubercle at the earliest in embryos of 21 mm. greatest length, the right in embryos of 28.5 mm. greatest length. The union of the duct with the sinus epithelium does not occur for some time; in embryos of 45 mm. the blind ends of the ducts first bore into the stratified epithelium of the sinus ; they do not reach its surface, however, but end blindly under the superficial layer of cylindrical epithelium. The actual breaking through takes place first in embryos of 70 mm. head-foot length.

Formation of the Uterovaginal Canal.

In both sexes there is a union of the right and left Miillerian ducts within the genital cord, and the unpaired canal so formed is termed the utero-vaginal canal. The union occurs at the earliest in embryos of 22 mm. and at the latest in those of 28.5 mm. ; the sex seems to have no influence on the earlv or late union. The union takes place first in the second fourth of the later uterovaginal canal and advances thence in both the cranial and caudal direction; frequently, especially in the caudal portion, the union takes place discontinuously. It extends upwards nearly to the upper end of the genital cord, i.e., to the level of the second bend of the urogenital fold (Fig. 552), between the horizontal and lower vertical portions. This bend later on lies almost at the same level as the first bend between the horizontal and upper vertical portions (Fig. 552) at which the inguinal fold arises; but on this account one cannot give this first bend as the cranial limit of the union of the ducts. While the union in male embryos proceeds at once in the horizontal terminal portions of the Miillerian ducts as far as their blind ends, in females these portions remain separate for a long time ; their union first occurs in embryos of 50 mm. headfoot length.

The length of the uterovaginal canal is 600 M in embryos of 26 mm.; at 28 mm. it is 860 /*, at 30 mm. lOOO^to 1550 ^, at 50 mm. 2000 /*, at 60 mm. 2040 m.

When the union of the two Miillerian ducts is completed above we term the portions that remain separate the tubar anlagen. The primitive tube consists of two portions, a sagittal and a horizontal, the sagittal one lying in the upper vertical and the horizontal one in the horizontal portion of the urogenital fold (compare Figs. 552 and 623).

The union of the two ducts is at first only an external one. They are placed together so that they form a single external contour, but in the interior each has still its own medial wall. Later the two medial walls fuse to form a septum, and then this becomes resorbed in the caudo-cranial direction. Since frequently the two Miillerian ducts do not lie in the same frontal plane of the embryo, the utero-vaginal canal is frequently placed obliquely, and since, furthermore, the ducts change their position during their course, sometimes the right one and sometimes the left lying more ventrally, a spiral twisting of the canal may be brought about.

Formation of the Wall of the Uterovaginal Canal.

GENERAL STATEMENT AND EXTERNAL FORM.

After the completion of the utero-vaginal canal the following relations obtain in the genital cord : In the mesonephric fold the primitive tubes run vertically downwards (vertical portion), reach the lateral edge of the genital cord, bend at right angles and run almost horizontally towards the middle line (horizontal portion) and there unite to form the utero-vaginal canal (Fig. 623 a). At the transition of the vertical into the horizontal portion of the tube the plica inguinalis (shown in Fig. 623 as the lig. rotundum) arises from the mesonephric fold.

The first anlage of the supportive tissue portion of the uterine or vaginal wall consists of a thickening of the surrounding mesenchyme ; from the beginning a rather sharp and smooth line of demarcation exists between the connective tissue investment of the uterus or vagina and the loose mesenchyme of the broad ligament, so that the external contour of the uterus is clearly defined (Fig. 623 b). The mesenchymatous wall of the uterus appears in both sexes, but in quite different form. In the female embn'o it surrounds the utero-vaginal canal and the horizontal portion of both tubes (Fig. 623 b) ; in the male embryo, where it develops after the degeneration of the horizontal portion of the tubes, it surrounds only the utero-vaginal canal (Fig. 624).

Fig. 623 a, b, c, and d. — Four diagrams of the development of the external form of the female uterus. The mesonephric folds and the genital cord are shaded obliquely, the mesenchymatous uterine wall vertically. The various parts of the primitive uterine wall and of the utero-vaginal canal are lettered. S vertical portion; h, horizontal portion of the wall of the primitive tube; F, fundus uteri; C, cervix uteri; V, vagina. Diagram a shows the position of the primitive tubes and of the utero-vaginal canal after it is completed. Diagram 6 shows the relation of the mesenchymatous uterine wall to the primitive tubes and to the utero-vaginal canal. It encloses the whole of the horizontal portion of both tubes in the uterine region and as a result the lig. rotundum is brought into relation with the uterine wall. The fundus uteri is bent in at an angle (uterus introrsum arcuatus). Diagram c shows the broadening of the horizontal portion of both tubes to form the fundus uteri, the broadening taking place in such a way as to straighten out the inward bend {uterus plani/undus) . Diagram d: The broadening of the fundus has increased and it is curved outward (uterus foras arcuatus).

lig. rotund, (chorda gubernaculi)

Development of the External Form of the Female Uterus.

In consequence of the horizontal portion of both tubes being taken up into the uterine wall, the primitive tubes may be divided into the definitive tubes and the tubar portion of the uterus. Since the two tubar portions of the uterus give the lumen of the uterus a triangular form from the beginning and produce in the fundus an angular bend towards the lumen (uterus introrsum arcuatus, Fig. 623 b), we may distinguish between the broader corpus (uterine portions of the tubes) and the narrow cervix portion (uterovaginal canal). Then the corpus portion begins to broaden, its angled inward bend being thus straightened out (uterus planifundus, Fig. 623 c) and then curved out cranially (uterus foras arcuatus, Fig. 623 d). How much of the later uterine mucous membrane is formed from the uterine portions of the tubes and how much from the utero-vaginal canal can hardly be determined with certainty; probably the point of entrance of the primary excretory duct into the wall of the uterus forms the limit between the two. This much is certain, that the fundus is formed from the tubar portion, the cervix from the utero-vaginal canal ; the boundary between the two must lie in the isthmus and may coincide with the boundary between the corpus epithelium and the cervix epithelium of the adult. By the extension of the mesenchymatous uterine wall outwards upon the tubes the inguinal fold, or the round ligament derived from it, first comes into relation with the uterine wall (Fig. 623 b-d) ; the ligament has nothing to do with the utero-vaginal canal.

Development of the External Form of the Male Uterus.

On account of the non-participation of the horizontal portions of the tubes in its formation the fundus, and perhaps also the entire corpus, must be wanting in the male uterus ; it represents, accordingly, from the beginning only the cervix and perhaps a part of the isthmus (Fig. 624). We shall learn later that in it more than the half of the utero-vaginal canal is degenerated in the cranio-caudal direction, so that in the male there is no persistence of the uterine portions but only of the vaginal one ; the term uterus masculinus is therefore certainlv incorrect.

Fig. 624. — Diagram of 'the development of the male uterus. Only the utero-vaginal canal becomes enclosed by the mesenchymatous uterine wall, the horizontal portions of both primitive tubes had previously degenerated. Consequently the plica inguinalis, or the chorda gubernaculi which is derived from it, does not come into relation with the uterus.

Formation of the Definite Wall of the Tubes.

In embryos of 50 mm. head-foot length the loose connective tissue of the tubar fold begins to arrange itself in concentric rings around the epithelium of the Miillerian duct, and so forms the anlage of the supportive tissue of the tube (Fig. 618). In fetuses of 80 mm. trunk length the mesenchyme of the tube wall separates into two layers, an inner one composed of round cells, no longer arranged regularly, and an outer one, three times as broad as the inner, formed of curved, spindle-shaped cells in regular layers ; thus the mucosa and muscularis are separated. In embryos of 180 mm. trunk length the first muscle fibres appear, the circular musculature being the first to form and then an outer and an inner longitudinal layer.

The epithelium of the tube is from the beginning singlelayered, but is in several rows. According to the degree of extension it is sometimes high cylindrical, sometimes cubical. In newborn children one finds between stretches with cylindrical epithelium groups of cubical cells which have the appearance of single alveolar glands. As to the time at which the ciliation begins I possess only negative results; according to Popofr" (1893) the epithelium bears cilia in the ninth fetal month.

An embryo of 50 mm. head-foot length already shows folds projecting into the lumen, two ventral and two dorsal (Fig. 617), giving a cross section of the lumen the form of a four-rayed star. The formation of the folds begins in the ostium abdominale and proceeds slowly towards the uterus. At first the folds are produced almost altogether by the difference in the height of the epithelium, but later the connective tissue of the mucosa grows into them and forces the epithelial covering apart. In embryos between 80 and 250 mm. trunk length secondary folds form on the primary ones and in the eighth month in addition to folds (directed towards the lumen), depressions (directed towards the mucosa) occur. The tube of the new-born child already shows almost completely the form of that of the adult. As soon as the folds have reached a certain size the connective tissue in them forms a special scaffolding of parallel bundles of connective tissue fibrils, which are surrounded by looser connective tissue.

Transformation of the Tubar Portion of the Uterus and the Uterovaginal Canal into the Uterus and Vagina.

The utero-vaginal canal is a long tube (Fig. 625). It runs vertically downwards parallel to the posterior surface of the bladder and urethra, bends almost at right angles at the level of Miiller's tubercle, and accordingly at the boundary between the urethra and the urogenital sinus, forming a horizontal portion, and opens into the sinus. The vertical portion is long, the horizontal quite short (Fig. 625). In transverse section the utero-vaginal canal forms a flattened oval, whose longest axis lies in the frontal plane. The epithelium is in general arranged so that it is high on the ventral and dorsal surfaces and low at the angles ; at first it is everywhere one-layered and high cylindrical. The nuclei are in several rows. To this rule only the horizontal portion forms an exception from the beginning, it being completely filled with a vesicular polygonal epithelium. In embryos of 38 mm. greatest length differences between the epithelium of the tubar portion of the uterus and that of the uterovaginal canal appear; in the tubar portion a cylindrical onelayered epithelium, varying in height, occurs; in the uterovaginal canal it is several layered and the superficial layer is sometimes cylindrical, sometimes cubical. This many-layered epithelium appears first in the cranial part and gradually extends into the caudal part of the uterovaginal canal, meeting (in embryos of 70 mm. greatest length) the vesicular epithelium that gradually ascends from the horizontal portion; wherever the vesicular epithelium extends it completely fills the lumen. Towards the middle of the fourth fetal month three regions may be distinguished in the uterine and vaginal canals by means of differences in their epithelium: first, about an upper fourth, hollow and lined with a simple cylindrical epithelium, the future corpus uteri (probably formed from the tubar portion of the uterus) ; second, a second fourth, hollow and lined by a manylayered, cylindrical epithelium, the future cervix uteri ; and, finally, a distal half, filled with a vesicular, polymorphous epithelium, the future vagina ; the cervix uteri and vagina are derived from the utero-vaginal canal. The boundary between the corpus and cervix epithelium corresponds almost with the internal os uteri, and it is also marked externally by a bend or at least a greater distinctness of the ventrally concave curvature. I shall consider the topography of the boundary between the cervix and vagina later on.

Fig. 625. — Primary excretory duct and uterovaginal canal seen from behind and somewhat from the left. Human embryo of 29 mm. nape length. (After Keibel 1896, from Felix-Biihler, Entwicklung der Geschlechtsorgane in Hertwig's Handbuch der Entwicklungsgeschichte, Vol. 3.) The two Miillerian ducts have united to form the utero-vaginal canal. The latter shows a long vertical and a short horizontal portion. Also the two primary excretory ducts are bent and accordingly also possess a horizontal portion at their orifices.

Development of the Vagina.

In the third to the sixth fetal month (the statements of authors vary, apparently on account of a variability in the appearance of the structure) the portio vaginalis is formed by the following processes: first, the wall of the utero-vaginal canal thickens in all its dimensions — even in the third month, and, second, the epithelium forms at the boundary between the cervix and vagina, but always in the vaginal territory, a ventral and a dorsal solid projection (Fig. 626), and these grow out into the mesenchyme in a shovelshaped form; they are the solid anlagen of the fornix anterior and posterior vaginas and therefore bound the anterior and posterior lips of the external os uteri. The projection that represents

Fig. 626. — Median longitudinal section of the utero-vaginal canal at the level'of the portio vaginalis uteri of a human embryo of 260 mm. (After Werth and Grusdew, from Felix-Buhler, Entwicklung der Geschlechtsorgane in Hertwig's Handbuch der Entwicklungsgeschichte, Vol. 3.) The portio vaginalis is beginning to be defined and the supravaginal circular muscle is developed in it. Its lumen extends to its lower end, which is closed by an epithelial plug. The vagina is still altogether solid, but its future lumen is indicated by the solid epithelial cord that traverses it. From this cord there grow into the surrounding mesenchyme, forward and backward at different levels, two solid projections of epithelium, the anlagen of the fornix anterior and posterior.

the anterior fornix is always considerably lower than that representing the posterior fornix and in consequence the anterior lip of the os is lower than the posterior, a relation that may be retained even in the adult. With the anlage of the portio vaginalis or of the two fornices the upper limits of the vagina become sharply determined.

Only after the formation of the fornices does the vagina become hollow. The lumen first appears in embryos of 150 to 200 mm. trunk length (Nagel 1891) in the distal portion of the vagina. It is formed by the breaking down of the central cells and the arrangement of the peripheral ones into a stratified cubical and later a stratified pavement epithelium; starting in the distal portion the breaking down of the central cells proceeds cranially. In this manner the vagina, together with the two fornices, becomes hollow and the protuberant portio vaginalis is grooved out from its surroundings. Since the utero-vaginal canal is bent or strongly curved at the point where the portio vaginalis projects, the angle between the uterus and the vagina is present from the beginning.

In older embryos, new-born children and young girls until puberty, the outer surface of the portio vaginalis of the uterus is marked in the most delicate manner by fine grooves, which appear quite symmetrically. They occur always on the borders of both lips, almost always on the outer surface of the anterior lip and more rarely on the outer surface of the posterior lip. They appear as principal grooves that usually run radially towards the os and are connected by transverse grooves; more rarely transverse grooves are the principal ones and in this case they are connected by very few radial accessory grooves.

The epithelium of the vagina is henceforward a many-layered pavement epithelium, that of the cervix is a stratified cylindrical epithelium; the question arises where the boundary between the two occurs. According to R. Meyer (1910) it always lies at first in the cervical canal. Then "the cervical epithelium differentiating in the mucous epithelium destroys the pavement epithelium as far as the external os uteri and in about one-third of the cases even further." "There is thus formed Fischel's congenital histological ectropion." "Some islands of the basal cell rows of the pavement epithelium are spared and these always regenerate the pavement epithelium of the outer surface of the portio vaginalis uteri, raising up and compressing the mucous epithelium. " " The definitive abolition of the congenital histological ectropion occurs, however, only in childhood." We see then that the cervical and vaginal epithelium are engaged in a mutual struggle. At its close the boundary between the cervical and vaginal epithelium coincides in general with the level of the external os. With a small portio vaginalis, a narrow cervical canal and a narrow external os the boundary lies above the os; with a large portio vaginalis and a broad os externum it lies below the os, that is on the vaginal surface of the portio vaginalis (R. Meyer, 1898 c). Since the boundary struggle does not progress equally over the whole periphery, the boundary line comes to be wavy and, indeed, islands of one kind of epithelium are included within the other ; islands of cervical epithelium on the vaginal surface of the portio vaginalis have been termed physiological erosion.

With the formation of the portio vaginalis uteri and the anlage of the two fornices the cranial limits of the vagina are determined once and for all. The length relations between the vagina and uterus remain the same during the late fetal period and the first year of extra-uterine life, that is to say, the vagina is as long as or longer than the uterus.

The colunmae rugarum are formed by the ingrowth of numerous solid epithelial projections into the subjacent mesenchyme at a time when the lumen of the vagina is not quite filled by epithelium.

Fig. 627 a and b. — External genitalia of a female embryo of 40.2 cm. vertex-breech length (measured over nape and back) and 30.1 cm. greatest head circumference. X \Y>- The labia minora surround a long slit-like opening, the urogenital orifice, and present two distinct portions, an oral high portion (6 mm.) and an anal low one (1-1.5 mm.). Between the oral portions is the shallow urethral groove, between the anal portions the urogenital sinus. The hymen appears between the anal portions of the labia minora.

The lower boundary of the vagina is formed from the hymen, which is formed from Miiller's tubercle. It has been shown above that the distal end of the utero-vaginal canal fuses with the epithelium of the urogenital sinus, and that this point of fusion persists as a solid mass of epithelium even when the rest of the vagina becomes hollow. Immediately above Miiller's tubercle the vagina then forms an ampulla-like enlargement, by which Miiller's tubercle is compressed to a disk. This is lined on one surface by vaginal epithelium and on the other by epithelium from the urogenital sinus ; the two layers of epithelium and the intervening mesenchyme form the hymen. In its centre the hymen has for a long time a cavity closed by a solid knob of epithelium, the remains of the point of penetration of the utero-vaginal canal.

Later the disk-like hymen becomes funnel-shaped; the funnel is compressed from right to left so that the ostium vaginaa no longer is a circle, but a sagittal slit. In embryos of 272 mm. vertexbreech length (measured over nape and back) and 213 mm. in the greatest circumference of the head, there regularly develops between the hymen cleft and the anal periphery of the hymen, usually somewhat to the left of tlie middle line, a sagittal fold, the hymen fold (Figs. 627, 645 and 646). At first small and inconspicuous it may increase in height up to birth to such an extent that it projects out like a partition between the labia minora. Rarely the fold is double, that is to say, a right one may be present in addition to a left one, the two folds then uniting at their anal ends and thus enclosing a groove that extends orally as far as the ostium vaginae (Fig. 646).

Recently it has been suggested that the vagina has a double origin, a pars Miilleriana and a pars adjuneta being distinguished. The pars Miilleriana is supposed to be derived from the utero-vaginal canal and the pars adjuneta from a frontal division of the urogenital sinus (Bolk, 1907). If this were true the hymen could not correspond with Miiller's tubercle, but would be a new formation inside the urogenital sinus. Bolk's suggestion is, however, incorrect. In the first place direct embryological observation shows that the hymen arises from the tubercle, and, in the second place, cases in which the distal end of the primary excretory duet persists show this running to the hymen ; the opening of the primary excretory duct is always at Miiller's tubercle.

Development of the Uterine Wall.

The transformation of the tubar portions of the uterus and of the proximal portion of the utero-vaginal canal into the corpus and cervix uteri will first be considered together. In this connection I would refer to the work of R. Meyer (1898) . The epithelium of both forms in transverse section an oval, whose longest axis is placed in the frontal plane (Fig. 628). In embryos of the fourth month the oval becomes a wavy slit, the number of waves depending on the width of the uterus and of the slit. In the lower part of the uterus one finds, therefore, only one entire wave, in the upper part 1% or 2. The anterior and posterior uterine walls are so close together that the summit of a wave on one wall fits into the depression between two waves of the other and vice versa.

This wavy form of the uterine slit persists as the fundamental form until birth and is often found in children and, in rare cases, even in the adult. In embryos of about 150 mm. secondary folds arise from the depressions between the waves and, indeed, from their deepest parts, these folds appearing as longitudinal folds of the mucous membrane; in the upper part of the uterus, where there are two waves, there are two folds on each of the walls, in the lower part only one. From these folds other smaller accessory folds arise and produce throughout the entire cavity of the uterus a very complicated pattern. But as the accessory folds develop the entire lumen of the corpus uteri elongates and, consequently, the folds again become flattened out, until, finally, the surface of the mucous membrane in the adult corpus uteri appears quite smooth. The folds are retained most distinctlv in the cervix.

Fig. 628. — Transverse section of the utero-vaginal canal at the level of the upper anlage of the prostatic glands of a female embryo of 80 mm. head-foot length. (Embryo R. Meyer 266; slide 69, row 1, section 1, from the collection of R. Meyer, Berlin.) The epithelial cylinder of the utero-vaginal canal is surrounded by an extensive layer of thickened mesenchyme, the primitive utero-vaginal wall. In this a middle zone is beginning to be defined by a closer arrangement of the nuclei; this is the future muscularis. Inwards from this is the future mucosa and outwards the future subserosa and serosa. The urogenital sinus develops on its dorsal surface two solid epithelial projections, the anlagen of the prostatic glands.

Long transverse folds are added to the longitudinal ones in embryos of 125 mm. at the earliest, and these form broad swellings projecting into the lumen. When these folds become so long that there is no longer space for them in the cervix, they ascend obliquely and the uppermost may then separate the corpus from the cervix like a valve. In such cases, if the mucous membrane of the corpus secretes mucus there will necessarily be a retention of the secretion in the uterine cavity. The transverse folds — forerunners of the plicae palmata? — appear in the cervix in such a manner that in embryos of 160 mm., for example, an upper third and also a small marginal zone above the ostium externum remain free from them. Consequently the upper third of the cervix — frequently even more (see the following table showing the growth in length of the uterus) — belongs macroscopically to the cervix but microscopically to the corpus uteri. On this account this portion of the cervix has been termed the isthmus and as a result two different internal ostia have been recognized, the ostium internum anatomicum, between the corpus and isthmus, and the ostium internum histologicum, between the isthmus and cervix.

The mucous glands of the cervix develop in embryos of from 110 mm. (Bosger 1894) to 175 mm. (Tourneux and Legay 1887), and always at the base of a fold. The cylindrical glands of the corpus uteri appear at very different times; one may find them abundant and well developed even in new-born children and fail to find them in young girls of 12 or 14 years. (Wyder 1878). Their appearance and growth in length is, therefore, until puberty, altogether independent of age. The cilia of the uterine epithelium first appear with the approach of puberty (Wyder 1878).

Development of the Musculature of the Vagina and Uterus.

Around the epithelial cylinder of the utero-vaginal canal the mesenchyme thickens to form a closed investment. This primitive utero-vaginal wall extends forward and backward to the surface of the genital cord and, accordingly, fills the cord completely in the sagittal direction, but in the frontal direction a strip occurs right and left along the lateral walls of the primitive true pelvis, which still contains the original loose mesenchyme. In this way the genital cord becomes divided into three parts ; the uterus and the right and left ligamentum latum (Fig. 629). The boundary between the uterus and the ligamenta lata almost corresponds with the course of the primary excretory duct, which is enclosed in the uterine wall so far as this corresponds to the utero-vaginal canal. In the primitive uterine and vaginal walls strands of spindleshaped cells arise in embryos of 60 mm. head-foot length, and are arranged concentrically around the lumen in such a manner that they mark off three zones; first and most internallv, a zone of v 7 v 7 round cells, the future mucosa, then the layer of spindle-shaped cells, the future muscularis, and, finally and outermost, a second layer of round cells, the foundation for the future subserosa and serosa. The same arrangement of layers is also found in the wall of the tube and the tubar layers pass over praeter propter into those of the uterus. The blood-vessels, already present in considerable numbers, still lie in the subserosa. and thereby define this from the serosa. In embryos of 80 mm. head-foot length the primitive muscularis becomes stronger and splits into a series of concentric rings, but it is only in embryos of from 120 to 140 mm. trunk-length that the first muscle cells are distinct.

The musculature of the uterus and vagina may be traced back to three primitive layers: inner and outer longitudinal and middle circular musculature. In the uterus the circular musculature appears first, in the vagina the longitudinal. Since the anlage of the cervix musculature is entirely under the influence of the vaginal musculature, we must distinguish embryologically between the corpus musculature on the one hand and the cervix-vaginal musculature on the other.

In embryos of from 240 to 300 mm. (according to Werth and Grusdew 1898) two superposed muscle rings appear in the muscularis of the uterus, the one, strong and compact in the corpus, extends almost to the cervix, the other, weak and scattered, occurs in the neighborhood of the portio vaginalis. The vagina at the same time forms a longitudinal layer at the boundary between the subserosa and the primitive muscular wall, and this layer extends upwards over the vagina and into the cervix. In embryos of from 340 to 355 mm. the circular muscle layers of the corpus and those of the cervix fuse, and, in addition, the circular musculature of the corpus fuses with that of the tube. The two tubes meet the body of the uterus at an angle, and that produces a certain amount of confusion in the previously simple arrangement of the musculature. The inner strands of the musculature of the tubes pass directly over into the circular laver of the uterus, but the outer strands stream out as longitudinal bundles both on the ventral and on the dorsal surface of the uterus, and thus form an outer longitudinal muscle layer on the body of the uterus. The circular musculature of the cervix, which has hitherto remained weak, now becomes very strong and forms the supravaginal muscle ring. This develops in the region of the longitudinal musculature streaming up from the vagina, and becomes intermingled with it. In the meantime a circular layer has appeared in the vagina itself, in the region of the longitudinal layer which has now become considerably thickened ; it intermingles with the longitudinal layer and a pure longitudinal layer is left only on the outer and inner surface. With this the anlage of the vaginal musculature is completed; there is an outer and an inner longitudinal musculature and an intermediate circular one.

Fig. 629. — Transverse section of a human embryo of 32 mm. vertex-breech length at the level of the utero-vaginal canal. (Embryo 426, I, IS in the collection of the Anatomical Institute, Zurich, IX. 1, 2.) X 15. The section shows the genital cord stretched transversely through the cavity of the true pelvis, and in it are the two primary excretory ducts and the utero-vaginal canal. The mesenchymatous uterine wall, which encloses the two primary excretory ducts, is so situated that it divides the genital cord into three portions, the uterus and the two lig. lata.

The inner longitudinal musculature of the uterus has a double origin; on the one hand longitudinal bundles stream from the orifices of the tubes on the inner surface of the circular musculature, and, secondly, fibres separate from the inner ring of the supravaginal circular muscle, and these also lie on the inner side of the circular musculature of the uterus, passing upwards towards the muscle bundles streaming down from the tubes and fusing with these to form the stratum longitudinale submucosum.

From the seventh fetal month to the end of pregnancy there is at first only an enlargement of the layers already present. In the body of the uterus the circular musculature always constitutes the principal mass. It grows in a three-fold manner : by increase in the thickness of its parts, by the development of new layers at the periphery and between the bundles of the circular muscle. These last, the interstitial bundles, are sometimes oblique and sometimes longitudinal, and produce the felted mass of the adult uterus, whose constituent parts can scarcely be made out. The outer longitudinal bundles that stream down from the tubes become included in this confused mass.

The vessels situated in the subserosa still lie on the outer surface of the musculature, but later they become enclosed by it, bundles of circular muscle fibres forming a coarse-meshed network, the stratum vasculare, around them. External to this network an outermost layer, the stratum supravasculare, is also formed, consisting principally of longitudinal bundles, which are in connection with the muscle bundles in the ligamentum rotundum, the ligamentum ovarii proprium and the plicae recto-uterinse.

Growth of the Uterus in the Postfetal Period.

The following table, compiled from data given by Hegar (1908), will furnish information as to the growth of the uterus. It gives the length of the corpus uteri, isthmus, and cervix and the entire length.

Age.

Fetus of 7 months Child of 5 weeks I year 14 months 2>£ years 3 years 3 Yz years 9 years I I years 13 years 15 years 16 years 17 years 17 years 18 years 19 years 19 years 19 years 20 years 20 years 22 years 28 years 29 years (nulliparous wife) 30 years (virgin)

Length of Length of corpus (mm.) isthmus (mm.)

10 10 8 9-10 6 9 12

41 27 20 36 27 28 24 30 30 35 40 34 38

5 6 5-6 5 4.5 6

12 6 4 5 5 6 8 6 7 5 10 10 7

Length of cervix (mm.)

12 12 10 16 13 19 27 25 22 16 31 28 27 21 16 21 29 28 34 29

Total length i mm. i

22 27 23 27 26 25 27 27 37 56 59 78 55 40 72 60 61 53 52 58 69 78 78 74

The table shows that the entire uterus passes through a resting stage in the early years of life, the uterus of a child of five weeks having the same length as that of a child of nine years. From this time onward an actual increase in length takes place, at first slowly, but more rapidly with the beginning of puberty. The increase is not, however, equally distributed among the various portions of the uterus. The body and neck elongate, but the isthmus practically not at all. Furthermore the body grows decidedly more than the neck.

Degeneration of the Tubes and of the Uterovaginal Canal in Male Embryos.

We have seen that the Miillerian ducts are formed in both sexes and that in both sexes they unite with the lower portion of the utero-vaginal canal; it has also been stated that this union occurs in embryos between 22 mm. and 28.5 mm. greatest length. Almost immediately after the union, that is to say in embryos of 30 mm., the degeneration of the ducts and of the utero-vaginal canal begins in male embryos. It makes its appearance in the middle of the anlage, in the portion of the tube that lies between the lower pole of the testis and the utero-vaginal canal. Both tubes lose their lumina, their epithelial cells show signs of degeneration at the centre and of proliferation at the periphery. In embryos of 60 mm. the tubes have degenerated as far as the portion between the testes and epididymides, and the utero-vaginal canal as far as the lower horizontal portion: in the intervening space only scattered remains occur, as a rule in the neighborhood of the point of crossing of the ureter and the primary excretory duct; these remains regularly show proliferation phenomena. In embryos of 90 mm. remains of the tubes always occur along the upper two-thirds of the testis and the ostia abdominalia are still open; data are wanting as to the time of their closure. Finally, of each entire tube only a very small portion remains, situated between the testis and the head of the epididymis ; it is known in systematic anatomy as the hydatid of Morgagni. It contains a simple, rarely stalked, epithelial canal, frequently without pouchings. Of the utero-vaginal canal only the epithelium degenerates at first, the dense mesenchyme enclosing it persisting for a long time as a sort of primitive uterine wall. A collapse of the epithelial canal regularly precedes its degeneration, so that one frequently sees a small, solid, sagittal portion resting upon the persisting, broad horizontal terminal piece; evidences of nuclear degeneration are, however, distinctly visible, so that one has to do not with a new formation, but with a degeneration. Only the broad horizontal portion persists throughout life, forming the vagina masculina. The portion of its wall surrounding the slit-like opening will represent a male hymen. The vagina masculina frequently shows glandlike growths in older fetuses, but very rarely true open gland sacks (E. Meyer 1909). Miiller's tubercle persists as the colliculus seminalis, the upper and lower cristas urethrales are defined quite early in the embryo.

Inhibitions of the Development of the Uterus and Vagina.

It is almost self-evident that an organ with so complicated a development as that shown by the uterus and vagina would show disturbances of the development. The many forms under which these disturbances present themselves may be arranged in groups, each of which may be brought into relation with a definite stage of development. In the following schema I follow, with slight modifications, the subdivisions proposed by von YVinekel (1S99) ; more important departures for it will be considered later.

Normal Development. Inhibitions of Development.

1. The mesonephric fold is completely developed, but as yet shows no erian ducts, together with complete abtrace of an anlage of the Mullerian sence of the tubes, uterus and vagina duct. (probably never occurs [von Winckel, 1899] or only in association with extensive bodily malformations [Nagel, 1S97]).

la. Complete absence of both Miill

lb. Complete absence of one Mullerian duct occurs only in association with a total absence of the urogenital fold, its organs, and the kidney of the same side (uterus unicornus verus [semiuterus]).

2. Formation of the funnel of the tube in the mesonephrie fold, outgrowth of both blind ends as Mullerian ducts into the urogenital fold along the primary excretory duct.

3. The Mullerian ducts unite, at first in the middle of the later uterovaginal canal, extending thence cranially and caudally. The union is at first only an adhesion, later the septum between the two ducts is resorbed. Around the epithelial utero-vaginal canal the mesenchymatous wall of the uterus is formed like a sheath, and in such a manner that portions of the walls of the primitive tubes enter into the formation of the uterus (see Fig. 623, p. 894), and for this reason the fundus of the uterus appears slightly depressed.

4. The flat fundus uteri gradually becomes convex outwards (uterus foras arcuatus). The form of the uterus fetalis is thus acquired.

5. The uterus infantilis is formed by a strong growth of the cervix and a weak growth of the corpus.

6. The uterus virgineus develops from the uterus infantilis by a stronger growth of the corpus.

7. The sexually mature uterus develops from the uterus virgineus by equal growth.

2a. Complete failure of the fusion of the two Mullerian ducts (uterus didelphys [= uterus duplex separatus], vagina duplex separata.

2b. Combined with a failure of one vagina to communicate with the exterior.

2c. Combined with a failure to develop lumina (uterus didelphys solidus, vagina duplex solida).

2d. Combined with the formation of only a partial lumen in the uterovaginal canal (uterus didelphys partim excavatus, vagina duplex partim excavata).

2e. Combined with a varying amount of inhibition of one-half (uterus didelphys asymmetrieus, uterus unicornis spurius [semiuterus spurius] cum rudimento uteri alterius).

3a. The union of the Mullerian ducts fails completely or partially (uterus bicornis saeptus, subsseptus, simplex; vagina sagpta, subsaepta, simplex).

3b. Uterus introrsum arcuatus saeptus, subsaeptus, simplex; vagina saepta, subsaepta, simplex.

3c. Combined with a failure of one side to open to the exterior.

3d. Combined with incomplete development of lumen.

4a. The fundus of the uterus remains flat (uterus planifundus).

4b. Combined with various degrees of inhibition of resorption of the septum.

4c. Combined with a partial failure of the lumen.

4d. Uterus foras arcuatus.

4e. Combined with various degrees of inhibition of the resorption of the septum, and with partial failure of the lumen.

5. The growth of the uterus does not take place; it remains a uterus fetalis.

6. The stronger development of the body fails, the uterus infantilis persists.

7. The growth takes place unequally or insufficiently and there is formed a uterus virgineus inaequalis or hypoplasia of the uterus.

Various theories have been proposed to account for the origin of malformations of the uterus and vagina. To-day they all have the fault that they are based on a plan of development that needs correction in important points. We have at present — basing our opinion on the facts described above (pp. 916^920) — to distinguish between 1. Disturbances of development in the formation of the mesenchymatous wall, and 2. Disturbances of development in the anlage and in the further development of the epithelial canal.

The epithelial canal is formed in the first place from the utero-vaginal canal, and, secondly, from the uterine portions of the two tubes. Since the tubar portion of the uterus practically forms the corpus uteri, and the utero-vaginal canal the cervix and the vagina, the development of these two portions of the uterus takes place under quite different ' conditions. The difference in their manner of development leads one to expect differences in the causes of their disturbances of development, and one may therefore subdivide the disturbances in the development of the epithelial canal into 2a. Disturbances in the development of the utero-vaginal canal; they affect the cervix uteri and the vagina.

2b. Disturbances in the further development of the tubar portions of the uterus; they affect the corpus uteri.

1. DISTURBANCES OF DEVELOPMENT IN THE FORMATION OF THE MESENCHYMATOUS WALL OF THE UTERUS.

It has been pointed out above that the mesenchymatous wall of the uterus is developed in such a manner that it encloses the utero-vaginal canal as well as the two uterine portions of the tubes (Fig. 623 b). By this inclusion of the horizontal portion of the tubes in the mesenchymatous wall the plica inguinalis, or the ligamentum rotundum that is developed in it, is brought into connection with the uterus. If now one finds in autopsies uteri in which the round ligament is attached to the wall of the tube, but not to the uterine wall, it must either be the result of an inhibition of the development of the mesenchymatous wall of the uterus or a hermaphroditic character. In the fundamental work of Kussmaul (1859) there are two figures, Figs. 17 and 18 (of which Fig. 17 is here reproduced as Fig. 630), which show conditions that are identical so far as our present purpose is concerned. The uterus is T-shaped, the transverse bar of the T being twice as long as the vertical one. and having attached to its ends the two round ligaments and the ligamenta ovarii propria. The point of insertion of both ligaments marks the boundary between the tube proper and the uterine portion.

Fig. 630.— Copy of Fig. 17 by Kussmaul (1859). The normal extent of the mesenchymatous wall of the uterus is indicated by a dotted line on the figure of the uterus rudimentarius solidus.

The normal extent of the mesenehymatons wall of the uterus is shown in Fig. 630 by the dotted line. A comparison of the normal and inhibited uterus shows the remarkable lack of development of the latter. This anomaly was termed a uterus rudimentarius by Kussmaul (1859), but we now describe it more in detail as a uterus with its mesenchymatous wall developed. It can naturally, and it usually does, occur in association with inhibitions in the development of the epithelial canal, complete or partial degeneration of this, obliteration of the lumen, a failure of the union of the two ducts, etc.

In considering a uterus with imperfect mesenchymatous wall one should not fail to notice its great similarity with the male type. We have seen (p. 916) that the formation of the mesenchymatous wall differs in the two sexes. In the male the mesenchyme surrounds only the utero-vaginal canal, the uterine portions of the tubes remaining free from it; furthermore, in the male the epithelial canal of the utero-vaginal wall within the mesenchymatous investment degenerates, though the investment itself persists for a longer time. In all cases of uterus with imperfect development of the mesenchymatous wall, accordingly, one must take into account the possibility of its being a case of hermaphroditism, and eventually consider whether there may not have been an error in the determination of the sex.

2a. DEVELOPMENTAL DISTURBANCES IN THE REGION OF THE EPITHELIAL PORTION OF THE UTERO-VAGINAL CANAL (CERVIX UTERI AND VAGINA).

Here belong all cases of double vagina, double cervix, vagina septa and cervix septa. They arise by the non-union or incomplete union of the two Mullerian ducts or by the non-resorption or incomplete resorption of the septum after the union. What can hinder the union of the two Mullerian ducts to form the uterovaginal canal? I have collected the following suggested causes from the literature :

1. Hydronephrosis.

2. Abnormal distention of the bladder and rectum.

3. Anomalies in the formation of the abdominal wall (hernia umbilicalis, cleft pelvis, abdominal clefts, abnormally short yolk stalk).

4. Fetal peritonitis.

5. Ligamentum recto-vesieale.

6. Shrinkage of the lig. rotunda.

7. Too great breadth of pelvis.

8. Long persistence and too great separation of the mesonephroi or the primary excretory ducts.

9. Congenital tumors.

The first four of these causes can hardly be effective. In the first place, the utero-vaginal canal is formed before the metanephroi become functional, and consequently hydronephrosis and an abnormal distention of the bladder are excluded. Secondly, the formation of meconium begins a considerable time after the completion of the utero-vaginal canal. Meconium was first found in an embryo of 223 mm. vertex-breech length (measured over the back) and 162 mm. greatest head circumference; filled with meconium, but not dilated, was the rectum of an embryo of 235 mm. vertex-breech length (measured over the back) and 190 mm. greatest head circumference; a strongly dilated rectum was first found in an embryo of 347 nun. vertex-breech length (measured over the back). An abnormally dilated rectum cannot, therefore, prevent the union, which has already been completed for a very long time (embryos of 22-28.5 mm.). Thirdly, the normal hernia umbilicalis occurs long after the completion of the uterovaginal canal. If it could prevent the union, non-union would be the rule, union the exception. I shall discuss the disturbance produced by an abdominal cleft later on; what influence a too short yolk-stalk can have upon the union I cannot understand. Finally, in the fourth place, a fetal peritonitis at the time of the development of the utero-vaginal' canal is practically excluded and has never been observed. The ligamentum vesico-rectale — a short falciform fold of peritoneum that extends sagittally from the posterior wall of the bladder to the anterior wall of the rectum, through the true pelvis (Krieger, 1858) — is rather the result of a non-union of the two Mullerian ducts than its cause; it also is frequently lacking in cases of divided uterus (Kehrer, 1899).

According to Thiersch (1852) the non-union of the Mullerian ducts is to be attributed to a strong development and long persistence of the two mesonephroi; Frankl (1902) and Holzbach (1909) have endeavored to establish this theory on a broader basis. The ontogenetic foundations upon which it is built, however, I cannot — at least for man — admit as correct. Firstly, the mesonephros never descends into the region of the utero-vaginal canal and never lies in the true pelvis. Secondly, the union of the two Mullerian ducts is accomplished even although the caudal end of the mesonephros increases considerably in man at the time. If persistence and stronger growth of the mesonephros could prevent the formation of the utero-vaginal canal, its non-formation would be the rule, and its formation the exception. Thirdly, the caudal portion of the mesonephros is actually not degenerated but persists as the epigenitalis and paragenitalis. The results of autopsies also argue against this theory. If an excessively developed mesonephros prevented the union of the Mullerian duct it should, as Kussmaul pointed out, leave behind it a paragenitalis more strongly developed than usual, but this is not the case.

According to R. Meyer (189S) the cause of a non-union of the ducts is frequently an abnormal shortness and strength of the ligamentum rotundum, eventually strengthened by a disproportion between the breadth of the lig. latum and the transverse diameter of the true pelvis. I hold, on the contrary, the shortness of the round ligament to be a result of the non-union of the ducts, but I agree as far as the influence of an excessive breadth of the pelvis, associated with a short lig. latum, is concerned.

Pick (1896, 1S98) endeavored to explain the non-union by intervening detached germinal tissue and by the tumors formed from them. Since he observed a simultaneous occurrence of tumors and deformed or double uteri in 30 cases, his theory cannot be dismissed off-hand. It fails, however, to explain cases of double formation unaccompanied by tumors — and these are the most frequent.

Pick's idea is strengthened by the observation of von Recklinghausen (1896), who, in his search for adenocysts in uterine tumors, structures which he showed to be pathologically altered remains of the mesonephros or paragenitalis, found his best material in deformed uteri of the smooth broad forms and in uteri bicornes.

Von Winckel (1899), who accepts the theories of both R. Meyer and Pick, has called attention to the relation between the Mullerian and the primary excretory ducts. The Mullerian ducts lie at first lateral to the excretory ducts, then they cross over their ventral surfaces and come to lie medial to them, so that a separation of the excretory ducts laterally must produce a lateral pull on the Mullerian ducts. The displacement laterally of the primary excretory ducts may be caused by the mesonephros and the metanephros. Whoever is familiar with the relations which the primary excretory duct holds to the lateral wall of the pelvis at the time when the utero-vaginal canal is formed, will admit that a lateral displacement of the primary excretory duct is possible only with an enlargement of the true pelvis; von Winckel's theory, therefore, agrees closely with that of R. Meyer.

An account of how the mesonephric fold is united to the lateral abdominal wall by the plica inguinalis has been given in the section treating of the urogenital fold (p. 793). At the same place it is furthermore pointed out that on the formation of the ventral abdominal wall all its parts are pushed eranially and medially. With the abdominal wall, the point where the urogenital fold is attached to it and, finally, also, the urogenital fold itself, must alter their positions in the directions named. The best evidence of the nature and extent of the displacement is the varying position of the m. rectus abdominis during the formation of the anterior abdominal wall; it lies at first at the middle of the lateral wall and later immediately beside the ventral median line. This passive displacement of the two urogenital folds towards the median line, however, favors their union. Perhaps it is also the cause of the spiral twisting of the mesonephric folds and thus of the spiral course of the Mullerian ducts around the primary excretory ducts. If the displacement of the anterior abdominal wall takes place early, that is to say, before the enlargement of the transverse diameter of the true pelvis — as actually happens — then a genital cord of normal extent will be formed and the mesonephrie fold will be twisted spirally. If the displacement is retarded, that is, until after the enlargement of the true pelvis, then the formation of the genital cord at least becomes difficult, and the difficulty will increase in proj)ortion as the amount of retardation increases, until finally the union is impossible.

If, however, the fusion of the two mesonephrie folds fails partially or wholly, the pull falls on the plica inguinalis or on the round ligament developed in it, and thereby the ligament becomes shortened. This shortening of the round ligament is in this ease, in my opinion, the result and not the cause of the nonunion of the Mullerian ducts. In this connection I would call attention to an observation by Rudolph (1909), in which the ligamentum rotundum was stated to be completely wanting in a case of uterus didelphys.

I have already called attention to the influence of extensive displacements of various portions of the intestine on the position of the mesonephrie fold. On the left side the influence is so strong that we must accept a right-angled bend of the mesonephrie fold as a regular result of its action (p. 791). If we suppose that the pressure of the intestinal tract on the mesonephrie fold becomes greater by some disturbance in its development, then Ave would have the abnormal pressure demanded by all the authors cited above, but produced by quite a different organ; it would be a second cause of the non-union of the Mullerian duets.

Finally, I must mention the topographical relations of the ascending and descending colons on the one hand and the mesonephrie fold on the other. It must be noted, in the first place, that the ascending colon lies from the beginning lateral to the right fold, and, further, that the descending colon usually runs along the medial surface of the left fold; it is this latter condition, principally, that produces its bend. If we suppose that the descending colon does not come to lie on the lateral side of the mesonephrie fold at the end of the process of readjustment of position, as is normally the case, the outward pressure, already existing, would be active beyond the usual time and so offer an obstacle to the union of the folds.

The same result would occur if the ascending colon came to lie on the medial side of the mesonephrie fold. The colon would then force the fold laterally and so again prevent its union with the fold of the opposite side. That such a thing is possible is shown by a case observed by Descomp (1909), in which the right ovary and tube were covered by the caecum, that is to say. were lateral to it. The influence of abnormalities in the position of the large intestine forms a third cause for the non-union.

All these considerations lead to the conviction that the causes for the occurrence of double formations of the utero-vaginal canal are more manifold than had been supposed, and that in any case several causes may act.

2b. DISTURBANCES IN THE DEVELOPMENT OF THE EPITHELIAL CANAL IN THE REGION OF THE CORPUS UTERI.

Here belong the uterus bicomis in all its different degrees of development, the uterus introrsum arcuatus and the uterus planifundus.

The corpus uteri — it must he continually borne in mind — is not formed by a union of the two Mullerian ducts, but the two uterine portions of the tubes, just as they are and just as they lie, combined with the cranial blind end of the uterovaginal canal, are transformed into the corpus uteri by simple enlargement (Fig. 623 a and b). The two uterine portions of the two tubes lie almost in the same line, and, consequently,— compare Fig. 623 b — there is sufficient room craraally for their former medial, now their cranial, walls to be raised up and so form the fundus uteri (Fig. 623 c and d). On the other hand, if the two portions meet at an acute angle, if they form with the utero-vaginal canal a Y instead of a T, room is lacking for the upgrowth of the fundus, that is to say, the developmental stage of the uterus planifundus cannot be formed and the epithelial canal of the corpus remains two-horned. The more acute the angle at which the tubar portions of the uterus meet, the more the two-homedness of the corpus uteri will be emphasized, and if this form of the epithelial canal persists until the musculature develops, it remains persistent; thus the uterus bicornis is formed. From this mode of development it follows that the forces that produce the uterus bicornis act in exactly the opposite direction from those that prevent the union of the Mullerian ducts to form the utero-vaginal canal. If the urogenital folds are drawn too far laterally, the genital cord is not formed and there is therefore no development of the utero-vaginal canal; if, on the other hand, the urogenital folds are pressed too far medially, although the Mullerian ducts will unite to form a utero-vaginal canal, the formation of a single corpus uteri will be prevented.

The urogenital fold is influenced most strongly and for a very long time by changes in the position of the developing intestine. In different embryos one finds the fold in very different positions: 1, completely rotated outwards, so that its medial surface rests upon the posterior and lateral abdominal walls; 2, placed sagittally so that it projects into the abdominal cavity as a ridge; 3, completely bent over medially, so that its medial surface rests on the posterior abdominal wall and its apex, which contains the Mullerian duct, reaches the root of the mesentery. This third position especially is interesting in connection with the formation of a uterus bicornis; if both Mullerian ducts run downward close to the root of the mesentery a horizontal tubar portion of the uterus cannot be formed (compare Fig. 552). Thus the condition is produced from which we started in the explanation of the uterus bicornis. The size of the angle between the two limbs of the Y will determine the degree of the two-hornedness — if the word is allowable ; the more acute the angle the longer the horns and vice versa.

If the formation of the utero-vaginal canal is prevented there can, of course, be no formation of a single corpus uteri, and a totally divided cervix uteri must always be associated with a divided corpus. On the contrary, even after • the completed union of the utero-vaginal canal, the formation of the corpus uteri may be prevented; the force which makes the formation of the fundus uteri impossible, must even increase the force that unites the utero-vaginal canal. Also in cases of uterus didelphys it is possible — at least in fetuses and young girls in the early years of life — to distinguish between the portion of the Mullerian duct intended for the formation of the utero-vaginal canal and the portion intended for the formation of the tubar portion of the uterus, and this by the relations of the primary excretory duct, which lies only within the mesenchymatous wall of the utero-vaginal canal or the portion homologous with it.

The formation of the utero-vaginal canal throughout its entire length is not necessary for the formation of a single corpus uteri, a union at the cranial end is sufficient and the rest of the canal need not be formed. That such a malformation is possible is shown by the occurrence of a uterus unicorporeus bicollis (that is, with doubled cervix).

The different septa that occur, in the female genital canal have, naturally, a different genesis. The septa in the region of the vagina and cervix are the persistent medial walls of the two Miillerian duets, those of the corpus uteri, on the contrary, are the remains of an inward bending of the fundus uteri.

III. THE UROGENITAL UNION.

The efferent apparatus for the male sexual products is produced in both sexes by a union between the mesonephros and the testis. In describing the second period of degeneration of the mesonephros (p. 829) we have spoken of a division of its remains into the epigenitalis and the paragenitalis. The tubules of the epigenitalis, 5-12 in number and corresponding to the 58th-62nd or 69th mesonephric tubules, degenerate completely as far as the tubuli collectivi. The angle at which the transition of the tubulus secretorius into the tubulus collectivus occurs, and at which the interruption between the two first takes place, of all parts of the mesonephros, projects furthest medially or, what is of especial importance here, furthest towards the reproductive gland area. The blind ends of the tubuli collectivi become surrounded by the epithelial nucleus of the indifferent reproductive gland, so that each comes to lie in a bay within the nucleus. This portion of the epithelial nucleus gives rise, as has been seen, to the rete blastema; from the beginning, accordingly, the tubuli collectivi and the rete tubules lie wall to wall. This proximity of the two has led some observers who studied poorly preserved embryos — and these form the majority — to the erroneous conclusion that the tubules of the male and female rete, and even the tubuli recti of the testis and the medullary cords of the ovary, arose by the ingrowth of mesonephric tubules. There can be no question of such an origin. The rete tubules are developed before their union with the mesonephric tubules occurs, the tubuli collectivi of the latter end with broadened and thickened ends, sharply defined and distinctly marked off from the rete cords. The union of the two, the urogenital union, takes place at very different periods; I saw it for the first time in embryos of 60 mm. head-foot length. Up to this time the development in the two sexes is alike ; a further development now begins in the males, degeneration in the females.

The Further Differentiation in the Male Sex.

As soon as the tubuli collectivi have broken through into the rete tubules, we term them the ductuli efferentes testis. In the fourth to the fifth fetal month (Kolliker 1879) they begin to coil at their ends that are towards the primary excretory ducts, while towards the testis they remain straight. As a result they appear as wedge-shaped structures, whose apices are towards the testis and the bases towards the excretory duct; when, later, they are surrounded with a firm connective tissue membrane, they are known as the coni vasculosi. All the 5-12 tubules of the epigenitalis do not necessarily take part in the urogenital union, nevertheless they frequently persist, appearing as mesonephric remains in the epididymis. They may make secondary connections with neighboring parts, and especially, may break through into the tunica vaginalis propria, that is to say, into a portion of the abdominal cavity. Several coni may unite to form a duct before opening into the rete, the union resulting from the coni being connected in the first place with the rete and, secondly, with each other by means of longitudinal anastomoses, so that the individual tubules become independent of their connections with the rete and may lose them. Those that behave thus are usually the most caudal efferentes. The primary excretory duct becomes the canalis epididymidis and the ductus deferens. At first it is lined by a onelayered cylindrical, non-ciliated epithelium ; in its upper part it coils in the manner described in works on systematic anatomy (canalis epididymidis) and at its lower end it dilates almost to four times its original size and forms the ampulla; the dilatation ends just at the beginning of the horizontal portion (Fig. 625), so that even before the anlage of the vesicula seminalis, the ductus ejaculatorius, which is developed from the horizontal portion, is recognizable.

In fetuses of 50 mm. head-foot length there begins, first in the lower portion of the ductus deferens, a concentric layering of the mesenchyme around the epithelial cylinder, and thence this arrangement extends over the whole of the primary excretory duct. I believe that I was able to recognize the first cilia on the epithelium in embryos of 70 mm. head-foot length. The musculature of the duct must first appear very late, since no muscle fibres were present in a fetus of 240 mm. head-foot length.

The vesicula seminalis arises from- the primary excretory duct, and from the lower portion of it which has dilated to form the ampulla. In this region a simple epithelial tube is gradually constricted off in the cranio-caudal direction. The constriction is first found in embryos of 60 mm. head-foot length; it ends at the beginning of the horizontal portion. This, therefore, from the beginning is the anlage of the later ductus ejaculatorius.

Degeneration of the Urogenital Union and of the Primary Excretory Duct in the Female.

The connection of the epigenitalis tubules with the rete is completed either at birth or even before it. Since the degeneration of the primary excretory duct has already begun some time before this, the epoophoron has become completely free. Notwithstanding its degeneration, it continues to grow and its tubules acquire a ciliated epithelium and subepithelial smooth muscle cells (Biihler 1894). The primary excretory duct is already degenerated in embryos of 30 mm., at the very time, it is interesting to note, at which the degeneration of the Miillerian duct begins in male embryos. The first degeneration is exactly the same as in the case of the Miillerian duct, it affects the portion between the caudal pole of the ovary and the beginning of the utero-vaginal canal. The cranial portion may remain connected with the epoophoron for a longer or shorter time, the distal portion is usually completely degenerated; all the remains of the primary excretory duct are known as Gartner's canal. This may run (H. Meyer 1909) "from the epoophoron through the ligamentum latum, at first parallel with the tube, then more obliquely as the base of a triangle whose opposite angle is formed by the uterine portion of the tube and the corpus uteri." "At about the level of the internal os uteri, the canal first sinks into the substance of the uterus, and running downwards through the muscle layers, it gradually approaches the mucous membrane medially, without reaching it, however, and then passes from the upper part of the portio vaginalis into the lateral fornix of the vagina to run caudally at about the middle of the lateral wall of the vagina." "In its lowest portion the canal lies usually rather far dorsally in the lateral wall and thence passes ventrally in the lateral part of the hymen to reach the free border of that structure." "Its opening to the exterior is immediately at or very close to the free border of the hymen, through the vestibular epithelium ; rarely it is more lateral, in the nymphohymenal sulcus of the vestibule." Gartner's canal is very seldom indeed present throughout its whole extent. Up to the end of the third fetal month one finds, without exception, remains of the canal in the uterus, in the vagina or in the hymen ; from the fourth month onwards the number of cases increases in which remains are no longer present: in new-born infants and in children, one always finds smaller or larger remains of the canal in about onequarter to one-third of all cases, and the proportion is not much less in adults (R. Meyer 1909).

If one compares the course of the primary excretory duct in the embryo with the course of Gartner's canal in the adult, the different relation of the two to the wall of the uterus at once reveals itself. The primary excretory duct of the embryo is completely surrounded by the mesenchymatous wall of the uterus that encloses the utero-vaginal canal; Gartner's canal enters the uterine musculature only at the level of the internal os uteri. The difference between these two arrangements is, however, only apparent; the uterine wall is formed not only by the mesenchyme of the uterovaginal canal, but also by the mesenchyme of the horizontal portion of the tubes (Fig. 623 b, p. 917). The entrance of Gartner's canal marks the point where, in the adult, the portion of the uterus formed from the tube passes over into that formed from the uterovaginal canal. I was, therefore, correct when (p. 918) I placed the boundarv between the two at the internal os.

The persisting Gartner's canal may also form in the fifth month an ampulla and a vesicula seminalis (for details, see R. Meyer 1899 and 1909), and the ampulla then lies at the level of the cervix uteri, of the upper part of the portio vaginalis, and extends to the fornix of the vagina and into the uppermost part of the lateral wall of the vagina.

The Ligaments of the Reproductive Glands.

The urogenital fold divides into a series of portions (pp. 785 and 787). First there is a division into the mesonephric and genital folds (Figs. 631 and 632), and the former is again divided into the tubar, gland and mesenterial portions (Figs. 632). Let us first consider the mesenterial portion. The most cranial portion of it fastens the mesonephros to the diaphragm for a limited period of the development and, therefore, bears the name of diaphragmatic ligament of the mesonephros (Fig. 631). During the development of the mesonephros this ligament is continually degenerating and continually being formed anew. This process depends on the degeneration of the mesonephros and of the mesonephric fold itself. Both undergo an extensive process of degeneration, as we have seen on p. 816, and, since the degeneration of the fold begins at the cranial end and proceeds caudally, it must at once affect the most cranial portion, the diaphragmatic ligament. This will disappear, only to be immediately formed again, since the succeeding portion of the mesonephric fold will now attach the new cranial pole of the mesonephros to the pillar of the diaphragm, which has in the meantime grown downwards, and will so become a new diaphragmatic ligament. This process will be repeated from segment to segment, until an obstacle occurs to end it, and this obstacle is supplied in the vasa spermatica interna. We may say, therefore, that the diaphragmatic ligament of the mesonephros continues to degenerate until it becomes the mesentery of the spermatic vessels; as such it persists throughout life, along with the rest of the mesonephric fold.

Fig. 631. — Model of the posterior abdominal wall of a human embryo of 19.4 mm. greatest length. (Embryo Ma. 2, from the collection of Professor Hochstetter, Vienna.) Model prepared by my pupils Massard and Chome. The urogenital fold is separated by a deep groove into the mesonephric and genital folds. The upper pole of both these folds is attached to the diaphragm by the diaphragmatic ligament of the mesonephros. The mesonephric fold is bent at two places, and we may recognize an upper sagittal, a horizontal, and a lower sagittal portion. At the first bend, between the upper sagittal and horizontal portions, the mesonephric fold is connected with the lateral wall of the abdomen by the inguinal fold.

Fig. 632. — Transverse section of the urogenital fold of a human embryo of 50 mm. head-foot length. (Embryo R. Meyer 272, from the collection of Professor R. Meyer, Berlin; slide 2, row 1, section 1.) The urogenital fold is divided by two deep grooves into the mesonephric and genital folds; both are connected by the mesogenitale. The mesonephric fold is again divided into the tubar, the gland and the mesenterial portions.

The mesenterial portion of the mesonephric fold is connected with the tubar portion and the mesogenitale by means of the gland portion. In female embryos the tubes and ovary become greatly developed, but the mesonephros degenerates and, in consequence, the gland portion collapses and appears to be merely the continuation of the mesenterial portion, which seems to split into the tubar portion and the mesogenitale. We speak, therefore, of a ligamenrinit ovarico-pelvicwm (the mesogenitale) and a lig amentum inf 'undibulo-pelvicum (tubar portion). Both become greatly thickened by acquiring blood-vessels, but they always remain connected by the former gland portion.

• Development of the Ligamentum Ovarii Proprium, the Ligamentum Uteri Rotundum, and the Chorda Qubernaculi.

The genital fold is also a structure that continually alters its form, growing caudally by apposition and shortening at the cranial end by degeneration. The indifferent reproductive gland does not occupy the entire length of the fold, but leaves a portion of it free at either end. One may therefore recognize three portions in the human fold: (1) the progonal portion, the upper end without the reproductive gland; (2) the gonal portion, the largest, middle portion with the reproductive gland, and (3) the epigonal portion, the lower end without the reproductive gland. The degeneration of the genital fold precedes that of the reproductive gland, and the growth of the reproductive gland that of the genital fold. As long, therefore, as growth in length and degeneration prevail in the reproductive gland, so long will gonal portions be transformed into progonal at the upper end and epigonal to gonal at the lower end. In female embryos the caudal pole finally reaches the true pelvis and lies on the posterior surface of the genital cord. The epigonal portion of the genital fold consequently unites the caudal pole of the ovary with the genital cord, but in the genital cord the formation of the primitive wall of the uterus (see p. 916) extends just so far that the point of insertion of the epigonal portion comes to lie on the surface of the uterine wall. Exactly opposite the insertion lies, as we will see shortly, the ligamentum rotundum. In the interior of the epigonal portion the mesenchyme tissue condenses to form a cord, the ligamentum ovarii proprium, the cells of which very quickly become connective tissue and smooth muscle fibres, all of which have a parallel course and stream out into the subserous musculature. From what has been said, it will be clear that the lig. ovarii proprium must arise from the caudal pole of the ovary and pass to the posterior surface of the uterus.

In male embryos the homologue of the lig. ovarii proprium, the lig. testis, is also found, but it has a different form and a different course. The difference in form depends in the first place on the greater length of the epigonal portion (the caudal pole of the testis lies higher than that of the ovary), and, in the second place, in its being constricted off from the rest of the mesonephric fold; the difference in the course depends, firstly, on the position of the caudal pole of the testis, which does not enter the true pelvis, and, secondly, on its non-union with the 'wall of the uterus; consequently the insertion of the epigonal portion, that is to say, the lig. testis, into the mesonephric fold must be at the boundary between the true and the false pelvis. That the points of insertion of the

Fig. 633. — Sagittal section of the lower end of the body of a human embryo of 40 mm. trunk length. (Embryo R. Meyer 259, from the collection of Professor R. Meyer, Berlin; slide 4, row 2, section 2.) X5.5. The section just cuts the right wall of the primitive true pelvis, and shows the foramen obturatum with the m. obturator internus and the levator ani. The testis lies just at the entrance into the true pelvis and is connected with the mesonephric fold, which is drawn out into a loop, by the lig. testis.

lig. testis and lig. ovarii proprium into the mesonephric fold or genital cord are homologous, is evident from the fact that both are opposite the insertion of the chorda gubernaculi or the round ligament (see below).

The urogenital fold in both sexes is united by the inguinal fold (p. 793) with the crista inguinaiis, situated on the posterior surface of the anterior abdominal wall (Fig. 634 a and b). This connection constitutes a bridge between the urogenital fold and the entrance into the inguinal canal, for this is marked quite early by the base of the crista inguinaiis. In the interior of the crista there is from the beginning a cord of compact mesenchyme, the chorda gubernaculi (Fig. 634 a and b), which is evident even before there is any indication of a differentiation of the abdominal musculature. It has a conical shape in transverse section, the apex being towards the crista inguinaiis and the base almost at the integument (Fig. 634). When, later, the abdominal musculature begins to develop, it must grow around the chorda gubernaculi, and thus there is necessarily formed a canal (the inguinal canal, Fig. 635), whose contents were there from the beginning. The developing muscles behave differently with reference to the chorda gubernaculi. The transversus and obliquus internus give off fibres

Fig. 634 a and b. — Two transverse sections of the urogenital fold at the level of the plica and crista inguinalis of a human embryo of 22.5 mm. greatest length. (Embryo R. Meyer 303, from the collection of Professor R. Meyer, Berlin; slide 34, row 2, section 1, and row 4, section 1.) The plica inguinalis arises from the lateral surface of the urogenital fold at the boundary between the false and the true pelvis; opposite the plica on the lateral abdominal wall is the crista inguinalis (a) . The plica and crista grow together (b) .

to its surface, which bend around and accompany it to the crista inguinalis ; by these fibres a muscular mantle is formed around the chorda, and this is known in comparative anatomy as the conus musculosus inguinalis. The obliquus externus presents at first a distinct foramen corresponding to the base of the chorda (Fig. 635), and through this foramen the chorda stands in connection with a second mesenchymatous cord (Fig. 635), which has long been known as the lig. scroti. This extends from the outer open of the inguinal canal to the integument and to the base of the genital tubercle. Later the aponeurosis of the external oblique grows around the lig. scroti, which in the mean time lias completely united with the chorda gubernaculi, and so the aponeurosis forms a sort of thin hernial sack, which becomes the fascia terica; this fascia, accordingly, is not a true fascia, but an aponeurosis. Up to this point the development is the same in both the male and the female embryo. In the male the union of the gilbernaculum with the mesonephric fold is exactly opposite the insertion of the lig. testis (Fig. 635). In the portion of the mesonephric fold between the two insertions there develops another mesenchyniatous cord, which connects both with the lig. testis and with the chorda gubernaculi, and so unites the two. On the completion of this union there is a continuous cord extending from the lower pole of the testis through the inguinal canal and terminating in the integument at the base of the genital tubercle. I shall term this cord the chorda gubernaculi in a broader sense (Fig. 636). Its individual parts are (Fig. 636) : the pars mesorchica, formed by the lig. testis, the pars mesonephridica, the mesenchymatous cord in the mesonephric fold and the inguinal fold, the pars intermuscularis, formerly the chorda gubernaculi in the narrower sense and, finally, the pars scrotalis, formerly the lig. scroti.

Fig. 635. — Portion of a transverse section of a human embryo of 26 mm. greatest length. (Embryo R. Meyer 321, from Tj*.; the collection of Professor R. Meyer, Berlin; slide 49, row 5, section 3.) X ca. 30. The section shows the left mesonephric fold: it has formed a secondary fold, the plica inguinalis. On the posterior surface of the anterior abdominal wall a crista inguinalis has formed, and with it the plica inguinalis has united. The point of union is still distinctly recognizable. Within the crista a cord of compact mesenchyme has appeared, the chorda gubernaculi in the narrower sense. It may be followed through the abdominal wall to an opening in the aponeurosis of the obliquus abdominis externus. From this opening a second cord of compact mesenchyme, the lig. scroti, extends to the integument. The anterior abdominal wall has thickened except at the point of insertion of the crista inguinalis; thus the crista comes to lie in a groove, the anlage of the saccus vaginalis.

Fig. 636.— In a transverse section of the urogenital fold, of the inguinal fold, and of the body wall of a male embryo of 26 mm. greatest length, the various parts of the chorda gubernaculi are represented diagrammatically. The chorda gubernaculi in the broader sense consists of: 1, the lig. testis (pars mesorchica), 2, a mesenchymatous cord in the mesonephric fold and its plica inguinalis (pars mesonephridica), 3, a mesenchymatous cord in the crista inguinalis and between the abdominal muscles (pars intermuscularis), and 4, the lig. scroti (pars scrotalis).

In the female embryo the affair is simpler. The lig. ovarii proprium, which corresponds to the lig. testis, and the chorda gubernaculi in the narrower sense, which in this case is termed the lig. rotundum uteri, are no longer separated by the mesonephric fold, but by the primitive uterine wall. Consequently a union of the two is imposible, but both have an insertion into the wall of the uterus, and it is evident from the development that these insertions must take place in the posterior or anterior wall of the uterus. Consequently there also results in the female embryo a union of the lig. rotundum with the lig. scroti, or if we would be precise, with the lig. labiate. The lig. rotundum extends, acordingly, from the anterior wall of the uterus through the inguinal canal to the base of the genital tubercle; it consists of a pars intermuscular is, the former chorda gubernaculi in the narrower sense, and a pars labialis, the former lig. labiale.

If one would draw homologies, the male chorda gubernaculi in the wider sense corresponds to the following parts in the female : (1) the lig. ovarii proprium, (2) a part of the wall of the uterus, (3) the lig. uteri rotundum. The lig. ovarii corresponds to the former lig. testis or the pars mesorchica of the chorda gubernaculi in the wider sense. The lig. uteri rotundum corresponds by its pars intermuscularis to the similarly named portion of the chorda gubernaculi, by its pars labialis to the pars scrotalis. Finally, one might homologize a portion of the subserous uterine musculature, which is connected on the one hand with the lig. ovarii proprium and on the other with the lig. uteri rotundum, with the pars mesonephridica of the chorda, gubernaculi.

IV. DEVELOPMENT OF THE EXTERNAL GENITALIA.

The parent tissue for the external genitalia is the cloacal tubercle. It has already been pointed out (p. 874) that this appears between the umbilicus, the coccygeal tubercle and the two lower limbs in an embryo of 13 mm. (Figs. 637, 638). We may distinguish on it an oral and an anal, a right and a left slope. Along the anal slope, from the summit to the base, there is, in the interior of the tubercle, the pars phallica of the urogenital sinus (Fig. 638), whose ventral wall, formed by the urogenital membrane, -abuts upon the surface of the cloacal tubercle (Fig. 638). The tubercle is originally a paired structure, lying on either side of the cloacal membrane (Fig. 601), but the paired condition is converted into an unpaired one by the elevation of the entire anterior abdominal wall (Figs. 601, 637).

Fig. 637. — Caudal end of the body of an embryo of IS mm. greatest length. (From a photograph kindly supplied by Professor R. Meyer, Berlin.) Between the umbilicus, the coccygeal tubercle, and the right and left lower limbs is the cloacal tubercle; on its anal slope are the ostium urogenitale and the anal groove.

Upon this cloacal tubercle the sexual organ, the phallus, is seated like a tower on a hill. The base of the phallus thereby comes to lie excentrically on the cloacal tubercle, occupying the whole of its anal slope and a part of the right and left ones (Fig. 639). The tubercle is thus divided into the almost circular base of the phallus and the semilunar genital tubercle (Fig. 639). This latter represents an originally unpaired structure, lying cranial to the phallus and surrounding it on two sides ; from it there are formed later the two genital swellings. The genital tubercle is separated from the phallus in female embryos by a deep groove which is the anlage of the later sulcus nympholabialis, the groove between the labia majora and minora. In male embryos this groove is absent from the beginning.

Fig. 638. — Sagittal section through the caudal end of the body of an embryo of 24 mm. vertex-breech length. (Embryo Hal. 1, from the collection of the I Anatomical Institute, Vienna.) The cloacal tubercle projects between the umbilicus and the coccygeal tubercle. The pars phallica of the sinus urogenitalis runs along its anal slope.

Fig. 639. — Diagrammatic representation ol the division of the cloacal tubercle into the phallus and genital tubercle.

The Indifferent Phallus.

The quickly growing phallus assumes the penial form at an early stage in both sexes, and at the beginning of the third month a transitory sexual difference occurs in the direction of the phallus, which in female embryos bends downwards, while in male embryos it retains its direction more or less perpendicular to the long axis of the body (Herzog 1904).

The pars phallica of the urogenital sinus, which is contained within the phallus, behaves differently in different parts, that part which corresponds to the future glans becoming perfectly solid and forming the urethral plate, while the part corresponding to the rest of the penis remains hollow, and by destruction of the urogenital membrane breaks through to the exterior throughout its entire length; it forms the trough-like orificium urogenitale primitivum. As soon as the coronary sulcus of the glans is developed in embryos of 21 mm., it lies at about the boundary between the anterior circumference of the orificium urogenitale and the urethral plate.

The glans is marked off as a spherical protuberance in embryos of 26 mm.

The Sexual Differentiation.

The beginning of the sexual differentiation can hardly be assigned to a definite period, since it takes place quite gradually and even in advanced stages presents difficulties to diagnosis. We base our diagnosis on the position of the ostium urogenitale relative to the coronary sulcus on the one hand and the anal opening on the other. Male embryos are distinguished by the distal ("distal" has reference here to the base of the phallus) circumference of the ostium urogenitale always lying in the coronary sulcus, while the proximal circumference becomes more and more distant from the anal opening. Female embryos may be recognized by the urogenital opening retreating away from the coronary sulcus by the gradual closing of its distal part, while its proximal part always lies close to the anal opening; the differentiating moment is, accordingly, both positive and negative in each sex. In Fig. 640 are shown the external genitalia of an embryo in which a determination of the sex from the external genitalia is impossible. The urogenital opening is on the anal slope of the phallus, whose projecting summit already shows the anlage of the glans. The distal circumference of the opening reaches the sulcus, the proximal one still lies close in front of the anal opening; the first fact is positive for male embryos, the second positive for female ones, and consequently these genitalia are to be termed indifferent. Fig. 641 shows the female differentiation: the phallus has become the clitoris; the urogenital opening lies close in front of the anal opening, and from its distal circumference a shallow groove, the urethral groove, extends to the coronary sulcus. A wandering away from the sulcus coronarius and a persistence of the relations to the anal opening are both female characters. Fig. 642 shows the male type. The urogenital opening is preserved throughout its entire length, its distal circumference reaches the sulcus coronarius, the proximal one has separated from the anal opening: a wandering away from the anal opening and the preservation of the relation to the coronary sulcus are both male characters.

Fig. 640. — The indifferent external genitalia of an embryo of 28 mm. greatest length. (From a photograph kindly furnished by Professor R. Meyer, Berlin.) The cloacal tubercle is divided into the phallus and the genital tubercles or swellings. The ostium urogenitale and anus are still close together.

Fig. 641. — Female external genitalia of an embryo of 32.5 mm. greatest length. (From a photograph kindly furnished by Professor R. Meyer, Berlin.) The phallus has become the clitoris, the glans and coronary sulcus are recognizable. The ostium urogenitale has preserved its relation to the anal opening, but no longer reaches the sulcus coronarius glandis.

Fig. 642. — Male external genitalia of an embryo of 3J months. (From a photograph kindly furnished by Professor R. Meyer, Berlin.) The phallus has become the glans penis, the scrotal swellings are fully developed and the unpaired scrotal area is indicated. The ostium urogenitale and ostium anale have separated, the ostium urogenitale extends to the sulcus coronarius glandis.

Development of the Penis and Scrotum.

The most striking characteristic of the transformation of the indifferent phallus, into the penis is the great elongation of the male organ. The formation of the penis takes place in such a way that the glans and the distal portions of the shaft of the penis are first formed. It has been pointed out that the sulcus coronarius glandis appears in the indifferent stage in embryos of 21 mm. and that embryos of 26 mm. already possess a distinctly projecting glans. After the glans and distal portions of the shaft are formed, the formation of the rest of the shaft occurs, and these newlyformed portions gradually raise up the already formed glans and distal portions of the shaft. The glans is passive to the extent that it only enlarges in proportion to the entire growth. Along with the glans and the distal portions of the shaft the coronary sulcus and the ostium urogenitale are also raised up. The form, size and position of the latter remains the same from the beginning to the end of the growth of the penis and is in all stages of development a distally broadened slit (Fig. 642), which extends to the sulcus coronarius glandis.

The urogenital sinus, accordingly, appears at once in its whole length ; it does not grow by the gradual closure of a groove along the anal wall of the penis until it reaches the glans, but its growth in length keeps pace with that of the penis. The ostium urogenitale, accordingly, already represents the primitive external orifice of the sinus urogenitalis.

Since in this growth the first-formed parts of the penis are pushed out passively, we must assume the existence of a basal zone of growth lying to the pelvic side of the urogenital opening, that is to say, in the neighborhood of the pars pelvina of the urogenital sinus. The pars pelvina is, accordingly, with reference to the growth, the active portion of the urogenital sinus, and the pars phallica the passive portion. The basal growth in the pars pelvina must produce a new area, interposed between the base of the penis and the anal opening, and the best name for this is the unpaired scrotal area (Fig. 642). In embryos of 60 mm. head-foot length this becomes raised up in toto and forms the unpaired scrotal swelling, into which the two genital swellings, which we now must term the scrotal swellings, extend from above. Since the middle of the area is connected by dense connective tissue with the connective tissue of the pars pelvina, the projecting scrotal swellings gradually draw out this connective tissue to a band and so the septum scroti is formed, it being at all times only a connective tissue structure. To its stretching it offers a certain amount of resistance, and consequently, there is formed at the surface of the scrotal area a median depression (Fig. 643), which, from the beginning, is the only external division of the unpaired scrotal swelling into a right and left half, while the septum scroti effects a complete internal division. As soon as the descensus is complete this unpaired scrotal area alone forms the scrotal sack, the paired scrotal swellings to the right and left of the penis fading out into the surrounding areas.

Fig. 643. — Transverse section of a male embryo of 70 mm. head-foot length (embryo R. Meyer 267, collection of Professor R. Meyer, Berlin; slide 67, row 1, section 3), below the symphysis pubis. X ca. 23. From behind forwards the following parts are cut: rectum, pars pelvina of the urogenital sinus and scrotum. From the pars pelvina Cowper's glands have developed and bet ween A the^ mesenchyme of the urogenital sinus and the epithelium of the scrotum there is stretched the septum scroti.

The primitive glans becomes divided into two portions (Fig. 644 a and b) by an epithelial cylinder that grows in from the epithelium of the external surface; the two parts are an outer cylindrical mantle, the praeputium, and an inner sphere, the glans in the narrower sense. The glandar lamella (Fleischmann 1907), as we shall term this epithelial cylinder, is not a closed tube, but is imperfect on the anal surface ; by this opening the prasputium and the glans in the narrower sense remain in connection, the frenulum prceputii is a persistent portion of the mesenchyme of the primitive glans. The solid glandar lamella becomes hollowed out by the degeneration of its central cells, the cavity breaks through to the exterior and so the space between the prepuce and the glans in the narrower sense is formed. The glandar lamella makes its appearance in embryos of 50 mm. head-foot length.

Fig. 644 a and 6. — Two sections through the glans clitoridis of a human embryo of SO mm. head-foot length. (Embryo R. Meyer 266, from the collection of Professor R. Meyer, Berlin; slide SS, row 2, section 5, and row 2, section 3.) From the surface epithelium a solid lamella, the glandar lamella, grows into the mesenchyme of the glans and separates the glans proper from the prxputium.

The corpora cavernosa appear as four portions, which, enumerated in the order of their appearance, are: the two corpora cavernosa penis, the corpus cavernosum glandis and the corpus cavemosum urethrae. They are formed from the young connective tissue by a dense aggregation of the mesenchyme cells, which are packed so closely together that the anlage has the appearance of being perfectly solid; only after some time greatly scattered vascular lumina occur and these later enlarge; between the first formed vessels new ones are formed and the characteristic structure of the spongy tissue is acquired.

The corpora cavernosa penis are united together at their apices and along their oral surfaces.

The first traces of a corpus cavernosum penis are to be found in embryos of 14.75 mm., the corpus cavernosum glandis occurs in embryos of 22 mm. greatest length. A compact layer of mesenchyme cells occurs around the epithelial lining of the sinus urogenitalis, at an early period, but a definite delimitation of a corpus cavernosum urethra? is first seen in embryos of 70 mm. head-foot length. The first vascular lumina occur in the glans in embryos of 28 mm. greatest length and in the corpora cavernosa penis in embryos of 70 mm. head-foot length.

Development of the Clitoris and of the Labia Majora and Minora.

In the female there is formed from the phallus the clitoris, the frenulum clitoridis and the labia minora. Since the penis develops from the entire phallus it cannot be considered exactly homologous with the clitoris. The clitoris corresponds to the entire glans penis and the oral slope of the penis shaft, the f renula clitoridis and the labia minora correspond to the anal slope of the penis shaft (Fig. 648).

The phallus of the female shows at first the same amount of growth as that of males of the same age, indeed, it may even be longer. The urogenital opening always remains at the base of the phallus and must, therefore, descend anally along the anal surface of the clitoris; a groove, the urethral groove, between it and the sulcus coronarius of the glans shows the former extent of the urogenital opening. We may therefore recognize in the female also an ostium urogenitale primitivum and definitivum. The principal growth in length, which in the male occurs in the neighborhood of the pars pelvina of the urogenital sinus, is entirely wanting in the female and consequently after a short period of growth there is cessation of the growth in length of the clitoris and the urogenital opening remains in the neighborhood of the anal opening ; between the two there develops a short perineum, but never a scrotal area.

The two genital swellings on this account are not prolonged anally to form an unpaired scrotal tubercle and consequently they do not fade out as in the male, but persist as the labia majora. The middle part of the genital tubercle becomes the mons Veneris.

The labia minora or nymphae arise in their oral parts from the margins of the urethral groove, in their anal parts from the margins of the ostium urogenitale definitivum, that is to say, they arise from the entire anal surface of the phallus with the exception of the short portion that forms the anal surface of the clitoris ; they extend from the anal periphery of the ostium urogenitale definitivum to the sulcus coronarius glandis. The oral and anal portions do not arise simultaneously; the time of the first development of the oral portions cannot be given, for they are present as the swollen margins of the urethral groove so soon as the urogenital opening has broken through (Fig. 645) ; the swelling of the margins of the ostium urogenitale definitivum and consequently the anlage of the anal portions of the nymphae begins in embryos of 223 mm. vertexbreech length (measured over the nape and back) and 162 mm. in greatest head circumference; it remains, however, quite insignificant for a long time and it is only in embryos of 347 mm. vertexbreech length (measured over the nape and back) that one can speak of labia minora in the neighborhood of the ostium urogenitale definitivum, but even in this embryo the oral portions of the nymphae are still almost three times as high as the anal portions (Fig. 627). In embryos of 386 mm. vertex-breech length (measured over the nape and back) and 324 mm. in greatest head circumference the form of the adult nymphae is acquired and no difference obtains between the oral and anal portions.

The formation of the glandar lamella proceeds somewhat more complicatedly in the female than in the male. Indeed, we must recognize in the female three glandar lamellae, an unpaired, ringshaped one that penetrates at the tip of the clitoris, and two paired lateral ones, that lie in the oral portions of the nymphae (Fig. 646). All three lamellae together form a narrow band, which encloses the glans clitoris and the later formed frenula clitoridis (Fig. 646). The middle ring-shaped lamella I found macroscopically for the first time in an embryo of 173 mm. vertex-breech length (measured over nape and back) and with a greatest head circumference of 163.5 mm. ; it corresponds to the male glandar lamella, but is less complete than this, forming only somewhat more than the half of a cylindrical mantle (Fig. 644 a and b). The frenulum praeputii is consequently broader than in the male. The formation of the cavity in the glandar lamella takes place practically as in the male and by it the glans clitoridis and the middle part of the praeputium are separated. Whether the two lateral lamellae, like the middle one, are actually formed as lamellae and later form a cavity, or whether they occur as grooves from the beginning I cannot state positively. Both the lateral grooves appear simultaneously in embryos of 347 mm. vertex-breech length (measured over nape and back) ; they begin at the left or right periphery of the middle groove and run obliquely over the outer surfaces of the nymphae to the sulcus nympholabialis (Fig. 646). By them the oral portions of the nymphae are divided lengthwise and from now on we may recognize an undivided portion, the definitive

Fig. 645 aland[b. — External genitalia of a female embryo of 27.2 vertex-breech length and 21.3 mm. in greatest head circumference. X 13^- The phallus is enclosed by two genital swellings (labia majora). Its shorter oral surface is lost in the mons Veneris, the longer anal surface fading out into the posterior commissure of the labia majora. The summit of the phallus is occupied by the slightly projecting glans; a sulcus coronarius glandia cannot be seen. Anal from the glans is the cleft-like ostium urogenitale, that leads in its oral two-thirds into a shallow groove, the urethral groove, and in its anal third into the sinus urogenitalis. Into this sinus the hymen fold projects, but it does not reach the margin of the ostium urogenitale.

nymphse and a genital portion, whose lateral parts give rise to the lateral parts of the praeputium, while its medial part becomes the frenulum clitoridis. Both frenula may be followed orally to the clitoris ; they are incompletely separated from the frenulum praeputii by the ends of the circular glandar lamella (Fig. 646). Although actually separated, the three glandar lamellae give the impression of a single structure; between them and the sulcus prasputic-labialis (Fig. 646) lies the praeputial area (Fig. 646), which is composed of the greater part of the anal and the right and left surfaces of the phallus. This preputial area Incomes raised up like a wall along the boundary formed by the three glandar lamellae, and so forms a rather deep groove (sulcus praeputio-clitoridis) between itself and the glans and frenula clitoridis. When the wall has reached a certain height it advances over the sulcus prEeputio-clitoridis, the clitoris and its frenula. The extent of this advancement varies individually; when it is slight the clitoris remains free, when it is greater the clitoris may be completely covered. What is termed the praepntrum in systematic anatomy is formed from this fold of the preputial area and consequently the prnaputium of the adult is a single structure, if, however, it is drawn backwards one may see the sulcus praeputio-clitoridis and at the bottom of this the three grooves formed by the splitting of the glandar lamellae. The adult praeputium is therefore a pseudopraBputium, the true praeputium lies at the bottom of the sulcus praaputio-clitoridis ; we may preferably speak of a clitoris sack. From what has been said it follows that the praeputium penis cannot be directly homologized with the praeputium clitoridis ; the praeputium penis is equivalent only to the small, middle, ring-shaped, true praeputium clitoridis.

Fig. 646 a and 6. — External genitalia of a female embryo of 34.7 cm. vertex-breech length (measured over nape and back). XIJ^. On the oral surface of the phallus and in the oral portions of the two nymphcc three glandar lamella? are developed, a middle unpaired one and two lateral paired ones. The three lamellae bound an area oralwards, the praeputial area, whose oral limit forms a groove (sulcus praeputio-labialis i between itself and the labia majora; from the praeputial area the clitoris sack develops. Between the two nymphs the hymen fold projects; it has in this case the unusual spoon-like form.

The formation of the corpora cavernosa takes place in the following order: corpus cavernosum clitoridis, corpus cavernosum glandis ; a corpus cavernosum urethrae does not occur, since the growth in length of the pars pelvina of the urogenital sinus is omitted. The formation of the corpora takes place as in the male.

The corpus cavernosum clitoridis appears in embryos of 19.4 mm., the corpus cavernosum glandis in embryos of 28 mm. greatest length and probably earlier.

Homologies of the Male and Female External Genitalia.

I shall now give a resume of the entire development of the external genitalia and shall endeavor to represent it in four diagrams. (1) The cloacal tubercle forms between the umbilicus and the oral periphery of the coccygeal tubercle (Fig. 647 a). (2) The phallus is placed excentrically on the cloacal tubercle, as one erects a tower on a hill. Thereby the cloacal tubercle is divided into the base of the phallus and the semilunar tuberculum genitale (Figs. 639 and 647 b). The indifferent condition lasts up to this period. (3) In the female the phallus becomes separated from the tuberculum genitale by a groove, sulcus praeputio-labialis and nympholabialis. The latter surrounds the phallus and so forms a circular swelling from which are formed: (1) the mons veneris from the oral portion, (2) and (3) the right and left labia majora from the right and left portions, and (4) the posterior commissura labiorum, from the anal portion (Fig. 648 a). On the phallus the glans becomes separated from the. shaft by a groove, the sulcus coronarius glandis. From this sulcus the urethral groove runs along the middle of the anal surface to the base of the phallus, where it is continued into the ostium urogenitale definitivum (Fig. 645). On both sides of the urethral groove and of the ostium urogenitale the anal surface of the phallus becomes enlarged and forms the labia minora (stippled in Fig. 648 a). The rest of the phallus becomes the clitoris and clitoris sack. (4) In the male the tuberculum genitale fades out in all its parts and in its place there is formed at the anal periphery of the phallus, from the unpaired scrotal area, a new swelling, the unpaired scrotal sack. The entire phallus becomes the penis, in which, as in the female, the glans becomes marked out by the sulcus coronarius glandis.

According to this account the clitoris corresponds only to the glans and the neighboring oral portions of the penis shaft white

DEVELOPMENT OF THE UKINOGENITAL ORGANS. 959

Umbilical cord

Cloacal tubercle

Umbilical cord

Genital tubercle Phallus

Cloacal tubercle

Anus

Fig. 647 a and 6.

Mons Veneris Sulcus praeputio-labialis Clitoris Frenulum clitoridis Labium minus (nympha) Sulcus nympho-Iabialis

Homologue of labium minus

Penis

Labium majus Omimissura posterior Anus

Scrotal sack

Fig. (548 a and b.

in Fig. 648 b) ; a portion that corresponds to about the basal threequarters of the penis shaft is not formed in the case of the clitoris. The primitive ostium urogenitale of the male is not equivalent to that of the female, since in the male the ostium retains its place, while in the female it is displaced basally, or at least is shortened along the urethral groove. A homologue of the labia major a exists in the male only during development in the paired scrotal swellings. No structure corresponding to the unpaired scrotal swelling is formed in the female; we must take the commissura posterior to be its equivalent. The homologue of the labia minora of the female is not formed in the male, the anal surface of the penis between the sulcus coronarius glandis and the base is the corresponding region (stippled in Fig. 648 b).

Further Development of the Sinus Urogenitalis.

In the preceding sections of this chapter it has been impossible to avoid occasional allusions to the development of the urogenital sinus ; I shall now give an account of the entire development. The sinus is formed by a division of the cloaca into three parts in the frontal direction. The first division separates off the rectum (Figs. 649 and 650), the second divides the remains of the cloaca into the vesico-urethral anlage and the pars pelvina of the sinus urogenitalis on the one hand, and the pars phallica of the sinus urogenitalis on the other (Fig. 651). The sinus begins at Miiller's tubercle and ends at the tip of the phallus (Fig. 652). It is laid down at the beginning throughout its entire length and from the beginning is divided into its two portions, the pars pelvina and the pars phallica. The lumen of the pars phallica, which was formerly the lumen of the cloaca, and is present from the beginning, persists only in the basal portion of the cloaeal tubercle, vanishing in the distal portion by the right and left walls of the pars phallica being pressed into contact and finally fusing (embryos of from 24 to 29 mm. greatest length). The hollow portion of the pars phallica breaks through to the outside by the resorption of its ventral wall throughout its entire extent and so the ostium urogenital primitivum is formed. The solid part, which occurs only in the region of the glans, is termed the urethral plate. . From its mode of origin the urethral plate must extend to the summit of the cloacal tubercle and must be fused with the anal slope of the tubercle throughout its entire extent; the region of fusion corresponds to the cranial portion of the former cloacal membrane. Seen in a transverse section of the phallus the urethral plate extends from the anal periphery only to about the centre and, accordingly, it divides into two parts only the anal portion of the phallus (Fig. 653 a) ; at the summit the division is incomplete (Fig. 653 b),

Fig. 649. — The cloaca is divided into the rectum and the ventral remains of the cloaca by the gutterlike saddle sinking down between the end-gut and the allantois stalk.

Fig. 650. — The division of the cloaca into the rectum and the ventral remains of the cloaca is almost completed, between the two there remains for a time a cloaeal duct, as an undivided portion of the cloaca (see also the explanation to Fig. 603).

Fig. 651. — The ventral remains of the cloaca is drawn out in the ventro-dorsal direction and we maydistinguish in it four walls, — cranial, ventral, caudal, and dorsal. The cranial wall is depressed so that it approaches the caudal one. Thereby the ventral remains of the cloaca is incompletely divided into three parts, a dorsal broad part, the vesico-ure.hral anlage, a middle narrow part, the pars pelvina, and a ventral broad part, the pars phallica.

Fig. 652. — The division of the ventral remains of the cloaca is completed. The vesico-urethral anlage and the pars pelvina of the sinus urogenitalis are narrowed in their dorso-ventral diameters, broadened in their frontal ones. The pars phallica remains broad in the sagittal diameter, but is narrowed in the frontal one, so that the lumen is abolished in the region of what will later be the glans; from the two walls an epithelial plate, the urethral^plate, is formed. The lumen is retained, corresponding to the extent of the ostium urogenitale (not cut in the section figured).

Fig. 653 a and b. — Two sections through the clitoris of an embryo of 60 mm. head-foot length. (Embryo R. Meyer 268, from the collection of Professor R. Meyer, Berlin; slide 66, row 2, section 1 and slide 67, row 1, section 2.) The urethral plate in section a extends from the anal surface of the clitoris only to the centre. In section 6, which passes through the tip, the clitoris is completely divided by the urethral plate.

the urethral plate here standing in connection with the so-called epithelial horn, a meaningless epithelial growth (Fig. 652).

In both sexes the ostium urogenitale primitivum is broadened out in a trough-like manner at the tip of the phallus (Fig. 642 and 654), and at the point where it breaks through there is a sharp distinction between the endodermal epithelium of the sinus and the ectodermal epithelium of the surface.

With the growth of the phallus and penis described above there is also, of course, growth of the sinus urogenitalis. In Fig. 654 one sees almost the entire length, perhaps it would be better to say shortness, of the sinus urogenitalis. It begins above with the crista urethralis, which runs downwards from Miiller's tubercle, and ends in the urethral plate. The pars pelvina and the pars phallica were originally of the same length, but during the growth in length of the entire region the pars phallica remains unchanged and the pars pelvina becomes enormously enlarged. The ostium urogenitale primitivum always extends to the sulcus coronarius glandis, its sagittal diameter remains always the same. The entire length of the sinus depends solely on the enlargement of the pars pelvina, which carries before it the pars phallica or the ostium urogenitale primitivum and the urethral plate. It has thus been shown that the urogenital sinus extends from Muller's tubercle to the primary ostium urogenitale and is entirely of entodermal origin.

Fig. 654.— Longitudinal section of the clitoris of an embryo of 60 mm. head-foot length. (Embryo H. Meyer 268, from the collection of Professor R. Meyer, Berlin; slide 64, row 2, section 1.) X 50. The section cuts the sinus urogenitalis lengthwise from the crista urethralis int. to the ostium urogenitale. The lumen of the sinus is enlarged at its end to a triangular basin, on whose distal wall is the urethral plate. To the right and left are the genital swellings, the anlagen of the labia majora, which are becoming separated from the clitoris.

The ostium urogenitale primitivum is the definitive one in neither of the sexes ; we know from macroscopic anatomy that the sinus of the male opens at the summit of the glans. This elongation is brought about by the closure of the ostium primitivum and the development of the glandar portion of the sinus from the urethral plate. I have not observed these processes myself and therefore follow the account given by Herzog (1904). In an embryo of 68 mm. greatest length the trough-like groove begins to close from the basal side, the closure taking place not at the margins of the opening where the ectoderm and endoderm meet, but at the middle of the groove; the wide groove is thus divided into a closed tube, the prolongation of the pars phallica of the shaft, and a shallow groove. This mode of closure is important in that it shows that the formation of the pars glandaris of the sinus takes place entirely in the region of the endoderm. In embryos of 105 mm. a splitting of the urethral plate begins, whereby the plate is converted into the urethral groove. This is then gradually closed again from behind forwards and is so transformed into a tube. During this process the ostium primitivum is completely closed and the new ostium wanders forward with the closure of the urethral groove, along the anal surface of the glans to its tip. In embryos of 105 mm. the ostium is already on the surface of the glans and in those of 180-190 mm. the position at the summit of the glans has been reached.

The male sinus urogenitalis is accordingly formed in three stages: (1) The formation of the sinus from Muller's tubercle to the sulcus coronarius glandis, the opening of the ostium primitivum immediately behind the sulcus; (2) closure of the ostium primitivum; (3) formation of the glandar portion and of the ostium urogenitale definitivum.

In the female the wandering of the ostium urogenitale is probably partly active and partly passive. It is active in that the part that reaches the sulcus coronarius degenerates and persists only as a groove, it is passive in that the shaft of the clitoris enlarges between the distal periphery of the ostium and the sulcus coronarius glandis. Growth of the pars pelvina does not take place and hence the shortness of the female sinus.

The epithelium of the sinus urogenitalis is two-layered from the beginning. The more the sinus grows the more numerous do the layers become in the basal portion, where as many as five may be seen superposed. The epithelium is a stratified cubical one up to embryos of 90 mm. head-foot length ; when the superficial layer becomes columnar is unknown.

Malformations of the Sinus Urogenitalis.

From the account of the sinus urogenitalis given above an entirely new view of hypospadias results. Under this name is understood the opening of the sinus urogenitalis at any point whatsoever of the anal surface of the penis. In hypospadias of the glans the sinus opens into the sulcus coronarius; this opening is nothing else than the persistent ostium urogenitale primitivum, and this form of hypospadias is a simple inhibition of development, the closure of the ostium primitivum and the formation of the glandar portion of the sinus and of the ostium definitivum are omitted. Hypospadias on the free shaft of the penis or in the region of the scrotal sack is not an inhibition, and, therefore, not true hypospadias, but an hermaphroditic phenomenon, a further development of the sinus urogenitalis in the female sense.

Epispadias, the opening of the urogenital sinus on the oral surface of the penis, can only be explained by assuming a displacement of the pars phallica towards the oral surface of the cloaeal tubercle.

The Glands in the Sinus Urogenitalis.

DEVELOPMENT OF THE PROSTATE.

A prostate is developed in both sexes ; at its first appearance it consists of several independent glands, which grow out into the surrounding denser mesenchyme. By the formation of smooth muscle fibres and strands of denser connective tissue this mesenchyme later becomes marked off from the surrounding tissue in the male and encloses the various glands to form the prostate; in

Fig. 655. — Transverse section of the urogenital sinus of a male embryo of 70 mm. head-foot length. (Embryo R. Meyer 267, from the collection of Professor R. Meyer, Berlin; slide 60, row 2, section 3.) X47. The section shows Mailer's tubercle projecting into the lumen of the sinus and around the sinus the anlagen of the prostatic glands.

the female the development of smooth muscle fibres and of a more compact connective tissue does not take place and the prostatic tubules remain isolated structures.

In its development the prostate belongs to the urethra as well as to the sinus urogenitalis, since the glands develop both above and below the opening of the primary excretory duct.

I have found the first anlage of the prostatic glands in a male embryo of 55 mm., and in a female embryo of 50 mm. It consists of solid epithelial buds that grow out into the surrounding thickened mesenchyme from the epithelium of the urogenital sinus (Figs. 628 and 655). These epithelial buds may arise around the whole periphery of the sinus, but they are most numerous on the dorsal surface, less so on the lateral surfaces and rare ventrally. Simultaneously with their appearance those of the dorsal and lateral surfaces branch, while those of the ventral surface remain unbranched and for the most part degenerate. The maximal number of glands that appear is twenty-six; a bilateral symmetry is unmistakable in the anlage. The direction of growth of all the glands is cranial, only the lowermost anlagen of the lateral surfaces grow caudally ; at feast two-thirds of the glands arise caudal to the openings of the primary excretory ducts.

The first muscle cells make their appearance in male embryos of 60 mm. head-foot length ; in the upper part of the prostate they are principally developed on the ventral surface, at the middle they form a closed ring and towards the lower border they are limited to the dorsal surface. Since the prostatic glands grow into this muscle mass, they force the muscle fibres apart and so produce the characteristic trabecular arrangement of the musculature. All the prostate glands are not united in the prostate, accessory glands occur in the interval between the prostate and Cowper's glands.

An embryo of 55 mm. head-foot length had four anlagen in the right side and none in the left; of those on the right 1 was above the primary excretoryduet and 3 below it. An embryo of 60 mm. head-foot length had 7 anlagen on the right side, 4 of which were dorsal and 3 lateral ; on the left there were 7 anlagen, of which 4 were dorsal, 2 lateral and 1 ventral. Of these anlagen 6 were above and 18 below the opening of the primary excretory duct. Another embryo of 60 mm. head-foot length had 11 anlagen on the right side, 7 of which were dorsal, 3 lateral and 1 ventral; on the left side there were 13 anlagen, 9 of which were dorsal, 3 lateral and 1 ventral. An embryo of 70 mm. head-foot length had 13 anlagen on the right side, 5 of which were dorsal, 5 lateral and 3 ventral; on the left there were 13 anlagen, 6 of which were dorsal, 4 lateral and 3 ventral ; 8 were above and 18 below the primary excretory duets.

In the female embryo few glands are formed, I have found as a maximal number only three ; they may undergo development and are then known as the paraurethral or Skene's ducts.

Development of the Bulbourethral (Cowper's) and the Vestibular (Bartholin's) Glands.

Cowper's glands are formed from the pars pelvina of the urogenital sinus and are therefore of endodermal origin. Their first appearance has not been established. Lichtenberg (1906) found them already advanced in development in embryos of 48 mm. They arise as paired solid epithelial buds from the pars pelvina of the urogenital sinus. Frequently there are two on each side, and frequently one on one side and two on the other. The solid epithelial buds grow upwards almost parallel to the urogenital sinus, and lie from the beginning (Fig. 643) in the compact mesenchyme which forms the corpus cavernosum urethral The glands grow through this mantle, and only when they have reached the looser mesenchyme between the rectum and the sinus are they able to enlarge. Each gland will therefore possess an intrabulbar portion with few, short and slightly developed lateral branches, and an extrabulbar portion with numerous longer and richly branched lateral branches. The lateral branches appear immediately after the anlagen, the lumina develop in the main stem and in the primary branches in embryos of 65 mm., in the smaller later branches in those of 105 mm. (Lichtenberg 1906). In embryos of 120 mm. the transformation of the ordinary epithelium into a glandular epithelium takes place. The epithelium of the main stem also shows the character of glandular epithelium, and Lichtenberg (1906) is therefore right when he speaks, not of an efferent duct, but of a glandular stem tubule.

Bartholin's glands are the exact homologues of Cowper's glands, and, like them, arise as solid buds from the dorsal wall of the pars pelvina of the urogenital sinus. I first found them in an embryo of 36 mm. greatest length, their first division in one of 80 mm. head-foot length. The first appearance of a lumen in the efferent duct was observed by Spuler (1910) in an embryo of 82 mm. vertex-breech length, the first terminal vesicles in one of 120 mm., and the first formation of a secretion in embryos of from 150 to 160 mm. vertex-breech length. Up to and after birth there is a slow increase in the number of the terminal vesicles (Spuler 1910), and after puberty there is another more rapid growth (Huguier 1849). At the climacterium these glands undergo degeneration and may be entirely wanting in old age (Tiedemann 1840).

Development of the Small Sinus Glands.

These glands also are of endodermal origin; only in the case of those situated in the glans penis is it impossible to exclude a. participation of the ectoderm. The glands appear in succession around the entire periphery of the sinus, first on the oral surface in embryos of 60 mm., on the right and left in embryos of 70 mm. and, finally, on the anal surface in embryos of 120 mm. (Lichtenberg 1906). These glands possess the same character as Cowper's glands, except that they do not attain to the same extensive development.

Descensus Testiculorum.

It has been accepted by all authors as an established fact that the mesonephros, ovary and testis undergo an active displacement in the abdominal cavity. All three organs are supposed to gradually glide downwards from their original positions into the true pelvis, or, in the case of the testes, as far as the true pelvis. After what I have said as to the development of the mesonephros (p. 815 et seq.) and of the indifferent reproductive glands (p. 885 et seq.), it is clear that I cannot agree with this view. The descent of the mesonephros or the ovary and the internal descent of the testis to the inguinal canal does not take place, but on the other hand the testis does wander from the abdominal cavity into the scrotal sack. The anlagen of the mesonephros and of the reproductive glands extend from the diaphragm to or into the true pelvis. When the growth in length of the anlagen is completed their caudal ends become stationary, those of the mesonephros and testes at the boundary between the abdominal cavity and the primitive true pelvis, those of the ovaries in the true pelvis at the posterior surface of the genital cord. What seems to be a descent of the organs is caused, on the one hand, by their degeneration in their cranial portion, which produces a shortening of the entire organ and a downward displacement of its cranial end, and, in the second place, by the fact that the degeneration of the formed parts is proceeding while the anlagen themselves are still progressing caudally. We have to consider, then, only the external descent of the testis.

Between the testis, lying in the abdominal cavity, and the floor of the scrotal sack there is formed a band, the chorda gubernaculi.

Ureter Mesonephric fold Testis

Pars mesorchica (lig. testis) Pars mesonephridica Pars inguinalis Pars intermuseularis (chorda gubernaculi s. s.)

Pars scrotalis ' (lig. scroti)

Fig. 656. — The various parts of the*chorda gubernaculi are shown diagrammatically in a transverse section of the urogenital fold and body wall of a male embryo of 26 mm. greatest length. The chorda gubernaculi in the broader sense consists of 1, the lig. testis (pars mesorchica), 2, a mesenchymatous cord in the mesonephric fold and its plica inguinalis (pars mesonephridica), 3, a mesenchymatous cord in the crista inguinalis and between the abdominal muscles (pars intermuseularis), and 4, the lig. scroti (pars scrotalis).

whose development has already been discussed at two different places (p. 943), and may therefore be recapitulated here. Between the mesonephric fold and the lateral or anterior abdominal wall a connection is effected; the mesonephric fold sends out from its lateral surface a secondary projection, the inguinal fold; covered by a thickened epithelium and filled with a somewhat more compact mesenchyme; the lateral or anterior abdominal wall, on its part, forms an almost sagittal, ridge-like fold, the inguinal crest. The upper third of the crest lies opposite the fold, and in this upper third there develops in the crest a thickening of the mesenchyme, and spindle-shaped cells appear and form bundles that run in a horizontal direction from the crest towards the integument (Fig. 634). The fold and the upper third of the crest fuse together (Fig. 635), and thereby the connection between the mesonephric fold and the anterior abdominal wall is accomplished. Since the testis is united to the mesonephric fold by its mesorchium and by the epigonal portion of the genital fold, it too, along with the mesonephric fold, is attached to the anterior abdominal wall. The epigonal portion of the genital fold lies on the medial or dorsal side of the mesonephric fold, just where the plica inguinalis is given off from the ventral or anterior surface of the mesonephric fold (Fig. 635). We may, accordingly, recapitulate thus: The testis is connected with the anterior abdominal wall (1) by its mesorchium or the epigonal portion of the genital fold; (2) by the plica mesonephridica; (3) by the plica inguinalis, and, finally, (4) by the crista inguinalis (Fig. 656). Within the fold so constituted there is formed, at first in separate parts, a cord of closely compact, spindle-shaped cells, the chorda gubernaculi. In Fig. 656 the chorda has been inserted and its various parts are indicated by interruptions of the lines representing it; in the mesorchium and the epigonal portion of the genital fold the pars mesorchica develops, in the mesonephric and inguinal folds the pars mesonephridica, in the inguinal crest and the abdominal wall, so far as this is formed of the abdominal muscles, the pars intermuscularis, and, finally, in the integument, the pars scrotalis. This last streams out into the subcutaneous connective tissue of the genital swellings at- first, and into that of the unpaired scrotal area, when this is developed. In this way the caudal pole of the testis is connected with the testis sack and the chorda marks the path which the testis must traverse in its descensus.

The Formation of the Saccus Vaginalis.

The saccus vaginalis is a part of the general body cavity that becomes completely cut off from the rest of the cavity by the formation of the anterior abdominal wall.

We may start the account of its development from the conditions seen in the body cavity of an embryo of 19.4 mm. greatest length (Fig. 631). The cavity may be divided into an upper broad portion and a lower narrow one, the primitive true pelvis. In the latter lie the rectum, projecting from the posterior wall, the urogenital fold projecting from the right and left walls, and, finally, the vesico-urethral anlage with the two art. umbilicales projecting from the anterior wall. At the boundary of the true pelvis and still lying in the primitive false pelvis is the union between the urogenital fold and the anterior abdominal wall (Fig. 631). Two processes bring about a change in these simple relations: (1) The bringing into the upright position of the anterior abdominal wall which in Fig. 631 is still horizontal, and (2) the taking into the body cavity of the loops of intestine that lie in the exoccelom. We may consider first the effect of this latter process. Space must be made in the body cavity for the loops of the intestines, which consist of almost the entire small intestine and a part of the large. This is accomplished by the enlargement of the sagittal diameter. The anterior abdominal wall thus becomes separated from the posterior one; but the mesonephric fold is attached to the anterior abdominal wall by the gubernaculum and the testis is attached to the mesonephric fold by the mesorchium and the epigonal portion of the genital fold. Both the mesonephric fold and the testis must therefore follow the anterior abdominal wall and become separated from the posterior one ; the caudal pole of the testis therebybecomes dircted ventrally and the long axrs changes from a vertical to a more horizontal position. The mesonephric fold always lies in front of the pole of the testis ; since it is attached not only to the posterior abdominal wall by its mesenterial portion, but also by its transition into the genital cord to the posterior wall of the vesicourethral anlage, it forms a loop around the testis, and the gubernaculum is attached at the summit of the loop. The formation of the loop naturally results in a stretching of the fold and its contents, and it is the medial limb of the fold that is most stretched, the lateral one, which forms the attachment to the posterior abdominal wall, being hardly stretched at all. The posterior abdominal wall follows the testis, an extensive layer of loose mesenchyme tissue becoming interposed between the ccelomic epithelium and the actual wall (Fig. 657). Thus the lower portion of the abdominal cavity, which in position corresponds to the fossa iliaca, is narrowed in its sagittal diameter to a third of its original size (Fig. 657), and appears to be a sack-like evagination of the undiminished portion of the abdominal cavity ; this sack-like portion is the anlage of the saccus vaginalis. What affects the mesonephric fold affects also its contents, but the primary excretory duct alone concerns us here. It also is thrown into a loop, whose summit corresponds to the first bend of the mesonephric fold, where the upper sagittal portion (Fig. 631) passes into the horizontal portion, and at this summit the tubuli collectivi of the paradidymis and, later, the ductus aberrans Halleri open (Fig. 657). It thus becomes certain that the entire ductus deferens is formed only from the medial limb of the loop of the excretory duct.

The bending of the anterior abdominal wall into the upright position will, of course, affect all the organs attached to its posterior surface; these are the bladder and the umbilical arteries. The bladder is simply brought from its bent position into an upright one, but it is different with the arteries. Since they arise from the aorta they are attached to both the anterior and posterior abdominal walls. From the aorta they run at first almost horizontally along the boundary between the false and true pelves (Fig. 631) to the anterior abdominal wall, and produce in this situation low peritoneal folds. Then they bend, and ascend towards the umbilicus along the bladder. When the arteries are drawn upwards by the pull exerted by the anterior abdominal wall, the horizontal portions are of course also raised, and by this elevation the peritoneal folds already present are materially increased in size.

Fig. 657. — Part of a transverse section through a human embryo of 60 mm. head-foot length. (Embryo R. Meyer 264, from the collection of Professor R. Meyer, Berlin; slide 44, row 2, section 2.) By the union of the mesonephric fold and testis to the anterior abdominal wall, formed by the chorda gubernaculi, both have been carried quite away from the posterior abdominal wall. Between the urogenital fold and the posterior wall a layer of loose mesenchyme has formed. Thereby the portion of the abdominal cavity which contains the urogenital fold is separated off from the rest of the cavity as the saccus vaginalis. The separation between the two is completed by a peritoneal fold formed by the a. umbilicalis.

Rising from below like a septum, they form a partition between the primitive true pelvis and the anlage of the saccus vaginalis (Fig. 657), and the latter becomes marked off from the true pelvis and now appears to be a sack-like evagination of the body cavity, well defined on all sides, although actually, as we have seen, it is not an evagination, but a portion cut off from the general body cavity by the encroachment of the neighboring parts. In the region of the saccus vaginalis lie the caudal pole of the testis, the loop of the mesonephric fold and the gubernaculum. All three structures are at first so placed that they run through the centre of the saccus vaginalis, but later the lateral surfaces of the urogenital fold and gubernaculum fuse with the lateral wall of the. saccus, and the lumen of the saccus thus disappears on the lateral side and persists only on the medial one. The saccus vaginalis is so large that it extends both over and under the point of insertion of the gubernaculum into the anterior abdominal wall, that is to say, above and below the opening of the inguinal canal.

The so-called invagination of the saccus vaginalis into the abdominal wall is certainly at first not an invagination. The anterior abdominal wall thickens greatly, but the thickening does not affect the point at which the gubernaculum passes through the wall. Consequently the abdominal wall grows around the saccus and a groove appears in the wall for its reception (Fig. 635).

After this point there is a considerable hiatus in the history of the descent. We know that the saccus vaginalis eventually penetrates through the musculature, and, passing on below the former genital swelling, enters the scrotal sack. How this outgrowth takes place has not yet been determined. In the seventh month the testis wanders down through the inguinal canal and retains its original relation to the saccus, i.e., it is, as before, invaginated into it. The definitive position of the testis in the testis sack is acquired in the eighth month or at the latest before birth. The retention of a testis in the abdominal cavity or the inguinal canal is due to an inhibition of development, and constitutes what is known as cryptorchism.

After the completion of the descensus the saccus vaginalis remains connected with the abdominal cavity by a narrow canal. A short time after birth this becomes at first solid and is then resorbed. The saccus vaginalis, now completely cut off from the abdominal cavity, has become the tunica vaginalis propria testis. All possible variations in the extent of the degeneration of the narrow canal between the abdominal cavity and the tunica vaginalis propria may occur; parts of it may remain unclosed. If the opening of the canal into the abdominal cavity remains unclosed its persistence may predispose to an oblique inguinal hernia; a persistence of the entire vaginal canal may lead to the occurrence of a congenital inguinal hernia.

The tunica vaginalis communis develops from the fascia transversa. The spermatic fascia, as was pointed out above, is not a fascia but is formed from the hernia-like protrusion of the aponeurosis of the muse, obliquus abdominis externus.

In the female embryo there is an incomplete formation of a saccus vaginalis, which is frequently persistent as the diverticulum of Nuck, but otherwise it vanishes completely. The ovary cannot undergo a descensus similar to that of the testis because: (1) it descends lower than the testis; (2) it comes to lie in a different portion of the body cavity (the testis lies in the saccus vaginalis, the ovary in the recto-uterine excavation) ; (3) the lig. ovarii pro.prium is attached to the wall of the uterus, i.e., at the middle line.

Diagrammatic Representation of the Fate of the Mesonephros, Primary Excretory Duct, and Mullerian Duct in Both Sexes.

In conclusion I present in three diagrammatic figures (Fig. 658 a, b and c) a resume of the changes in position undergone by the organs of the urogenital fold in the two sexes. Those structures that are retained are represented by solid lines, and those that degenerate are outlined by broken lines ; the structures that change their position are shown in grey, and those that have attained their final position are in black. The diagrams show that no internal descent of the testis occurs, nor is there any downward migration of the ovary or mesonephros. The mesonephros and the reproductive gland are laid down throughout the entire length of the abdominal cavity, and their lower poles lie from the beginning at or in the true pelvis ; on the other hand, in both organs a cranial portion, which may include more than half of the entire organ, degenerates, and this degeneration may simulate a descent of the organs.

Fig. 658 a, b, and c. — Diagrammatic representation of the entire renal organ, reproductive gland, and the efferent ducts. What persists is represented by solid, and what degenerates by dotted lines; whatever changes its position is shown in grey, whatever has reached its final position is in black, a. On the left is the male and on the right the female reproductive gland. Before the migration. The numbers are the serial numbers of the tubules.

Fig. 658 b. — After the migration in the male embryo. The testis wanders out of the body cavity into the scrotal sack (descensus). The Mullerian duct degenerates throughout the greater part of its extent, the closed ostium abdominale persisting as the appendix testis (hydatid of the testis) , and the lowest portion of the utero-vaginal canal as the vagina masculina. All the tubules of the epididymis are not employed in the urogenital union: the first tubule remains free and becomes the appendix epididymis (hydatid of the epididymis) and the other unemployed tubules persist as appendices retis. The paradidymis (paragenitalis) is separated into its individual parts, some of the ductuli collectivi persist as the organ of Giraldes and the terminal portions of the ductuli, united to form a canal, persist as the ductus collectivus paradidymidis (ductus aberrans Halleri).

Fig. 658 c. — After the migration in the female. The ovary remains in the body cavity, but rotates through 90° in such a way that its cranial pole comes to lie laterally, while the caudal pole, through which the axis of rotation passes, remains where it was and becomes the medial pole. The urogenital fold is rotated at the same time around the Miillerian duct, and as a result the tube comes to lie cranial to the ovary. The mesonephros and primary excretory duct degenerate ; of the mesonephros the epigenitalis persists as the epoophoron and parts of the paragenitalis as the paroophoron. Of the primary excretory duct only the portion that receives the tubules of the epoophoron is retained, though portions of the rest may persist as Gartner's canal.

The diagrams show furthermore that the tubules of the epigenitalis and paragenitalis — the persisting tubules of the mesonephros — are the most caudal tubules.

The testis undergoes an external descensus, during which no alteration of the relative position of the parts occurs. The ovary rotates through 90° and comes to lie with its long axis horizontal. Simultaneously with this rotation the genital fold rotates around its longitudinal axis, the Mullerian duct forming the axis. In consequence of this rotation, which is through almost 180°, the Mullerian duct comes to lie more cranially and the ovary more caudally.

The ostium abdominale of the Mullerian duct always lies in the neighborhood of the cranial pole of the ovary, notwithstanding that this wanders downwards as a result of the degeneration of the cranial portion of the ovary; this is due to the fact that its growth ceases for a considerable time, during which all the other organs increase materially in dimension.

Literature

(In this list only the works cited in the text are included. The reader will find the remaining literature in the bibliography appended to the chapter on the Urogenital System in Hertwig's Large Handbuch, Vol. 3, or in the Normentafel of Keibel and Elze.) 1 Since my manuscript was sent to the publisher, more than a year has elapsed. The literature has been brought up to July, 1911. although, with a few exceptions, only that which had appeared up to the summer of 1910 could be used in the text.

Ahlfeld, F. : Die Missbildungen des Menschen, F. W. Grunow, Leipzig, 18S0.

Alglave, P. : Note sur la situation du rein chez le jeune enfant par rapport a la crete iliaque et reflexions sur l'ectopie renale, Bull, et mem. Soc. Anat., vol. 85, p. 595-599, with 2 text-figs. Paris, 1910. Biex, G. : Ueber Furchenbildung an der Oberflache des menschlichen Ovarium, Alonatsschr. f. Geburtsh. u. Gynakol., vol. 32, p. 175-179, with 1 plate.

1910. Born, G. : Entwicklung der Ableitungswege des Urogenitalapparates und des Dammes bei Saugetieren, Anat. Hefte. Ergebnisse, vol. 3, 1894. Boveri. T. : Die Niei'enkanalehen des Amphioxus. ein Beitrag zur Phylogenie des Urogenitalsystems der Wirbeltiere, Zool. Jahrb.. Abt. f. Ontogenie, vol. 5, p. 429-507, 1802.

Branca, Q. and Basseta, A. : Sur le developpement du testicule hmnaine, Arch.

general de Chirurgie, vol. 1, p. 116-124, with 1 text-fig., 1907. Braun : Ueber Abschniirung der Ovarien, Dissert, ined. Giessen, 1896. Broek, van der, A. J. P. : Ueber den Schliessungsvorgang und den Bau des Urogenitalkanales (Urethra) beim menschlichen Embryo, Anat. Anz., vol.

37, p. 106-120, with 2 plates and 4 text-figs., 1910. Buehler, A. : Beitrage zur Kenntnis der Eibildung beim Kaninehen und der Markstrange beim Fuchs und Menschen, Zeit. f. Wiss. ZooL, vol. 58, 1894. Dandy, W. E. : A Human Embryo with Seven Pairs of Somites, measuring about 2 mm. in length, Amer. Jour. Anat., vol. 10, p. 85-102, with 6 plates. Delmas, J. and P. : Sur les anomalies ureterales, Ann. des malad. des Organes genitourin., vol. 28, 1, p. 865-896 and 984-1009. Descomps: Anomalie des organes genitaux de la fern me. Arret de developpement.

Uterus bicornis, ectopie de la trompe et de Fovaire, Bull, et Mem. Soc.

Anat., Paris, vol. 84, p. 460-^61, with 1 text-fig., 1909. DlSSE, J. : Untersuchungen fiber die Lage der menschlichen Harnblase und ibre Veranderungen im Laufe des Wachstums, Anat. Hefte, vol. 1, with 8 plates. 1891. Elze, C. : Beschreibung eines menschlichen Embryo von ca. 7 mm. gr. L., Anat.

Hefte, No. 106, with 7 plates and 32 text-figs., 1907. Evatt, E. J. : A Contribution to the Development of the Prostata in Man, Jour.

of Anat. and Phys., vol. 43, p. 314-321, 1909. Felix, W. : Die Entwicklung der Harnorgane, Hertwig's Handbuch der vergleich.

u. experiment. Entwicklungsgesch., vol. 3, 1905. Fleischmann, A. : Die Stilcharaktere am Urodaeum und Phallus, Morph. Jahrb., vol. 36, 1907. Frankl, O.: Ueber Missbildungen der Gebarniutter und Tumoren der uterusliga mente im Lichte embryologischer Erkenntnisse, Volkmann's Sammlung klin ischer Vortrage, No. 363. 1902. von Franque, O. : Ueber Umierenreste im Ovarium, zugleich ein Beitrag zur Genese der cystoiden Gebilde in der Umgebung der Tube, Zeit. f. Geburtsh.

u. Gynak., vol. 49, p. 499-524, 1896. Gebhard: Pathologisehe Anatomie der weiblicben Sexualorgane, Leipzig, 1899. Gerard, L. : La forme de Furetere chez le fetus et le nouveau-ne, Medical Thesis, Paris, 1908. Grubenmann, I. : Eine sagittale Verdoppelung der weiblicben Urethra, Inaug.

Dissert. Zurich, with 2 plates and 5 text-figs., 1911. Halban, J. : Die Grossenzunahme der Eier und Neugeborenen mit dem fortsch reitenden Alter der Mutter. Arch. f. Entwieklungsmech., vol. 29, p. 439 455, 1910. Hamburger, O.: Ueber die Entwicklung der Saugetierniere, Arch. f. Anat. u.

Phys., Supplement, 1890. Hart, D. B. : The Nature and Cause of the Physiological Descent of the Testes, Jour, of Anat. and Phys., vol. 42, p. 244-265, vol. 44, p. 4h26, with 28 text-figs., 1909. Hart, D. B. : The Physiological Descent of Ovaries in the Human Fetus, Journ. of Anat. and Phys., vol. 44. p. 27-34. 1909. Hauch, E.: Ueber die Anatomie und die Entwicklung der Nieren, Anat. Hefte.

vol. 22, p. 153-248, 1903. Hegar, K. : Anatornische Untersuchungen am nulliparen Uterus mit besonderer Berucksichtigimg der Isthmus, Beitraae zur Geburtsh. u. Gynak., vol.

13, 1908. Herzog, F. : Beitrage zur Entwicklungsgeschichte und Histologie der miinnlichen Harnrohre, Arch. f. mikr. Anat. u. Entw., vol. 63, 1904. HochsteTter, F. : Ueber die Arterien des Darmkanales der Saurier, Morph.

Jahrb., vol. 26.

Holzbach, E. : Die Hernmungsbildungen der MUller'schen Gange im Lichte der vergleiehenden Anatomie und Entwicklungsgeschichte, Beitriige zur Geburtsh. u. Gynak., vol. 14, p. 3-57, with 6 text-figs., 1909. Huguier : Memoire sur les appareils secreteurs des organes genitaux externes chez la femme et chez les aniruaux, Ann. des Sci. nat., Series 3, vol. 13, 1849. Ingalls, N. W. : Beschreibung eines rnensehlichen Embryo von 4.9 mm. L., Arch.

f. mikr. Anat. u. Entw., vol. 70, p. 506-570, with 3 plates and 28 textfigs., 1907. Jixsabato, L. : Sul connetivo nell' utero fetale con particolare riguardo alia sua istogenesi, Monit. Zool. Ital., vol. 19, p. 281-285, 1908. Kehrer, E. : Das Nebenhorn des doppelten Uterus, Inaug. Dis., Heidelb., 1899. Keibel, F. : Zur Entwicklungsgeschichte des mensehlichen Urogenitalapparates, Arch. f. Anat. u. Entw., 1896. Kolliker, A. : Lehrbuch der Entwicklungsgeschichte, 2nd ed., Engelmann, Leipzig, 1879. Kossmanst, R. : Missbildungen und Lageanomalien, in Martin, Krankheiten der Eileiter, Leipzig, 1895. Kosshann, R. : Mangel, Unvollkommenheit, Ueberzahl, Verlagerung der Eierstocke, in Martin, Krankheiten der Eierstocke, Leipzig, 1899. Krieger: Monatsschr. f. Geburtsk., vol. 12, p. 179, 1858 (cited by Kussmaui, 1859). Kuelz, L. : Untersuchungen iiber das postfetale Wachstum der mensehlichen Niere.

Inaug. Dissert. Kiel, 1899. Kussmaul, A. : Von dem Mangel, der Verkiimmerung und Verdoppelung der Gebarrnutter, Wiirzburg, 1859. Lams and Doorme : Nouvelles recherches sur la maturation et la f eeondation de l'oeuf des mammiferes, Arch, de Biologie, 1907. Legueu, F. : L'anatomie chirurgicale du bassinet et l'exploration interne du rein, Ann. de Gynec, 1891. Leydig, F. : Lehrbuch der Histologie des Menschen und der Tiere, Frankfurt a.

M., 1857. Lichtenberg, A. : Beit rage zur Histologie, mikroskopischen Anatomie und Entwicklungsgeschichte des Urogenitalkanales des Mannes und seiner Driisen, Anat. Hefte, vol. 31, 1906. Menge, C. : Bildungsfehler der weiblichen Genitalien, in Veit's Handbuch der Gyniikologie, J. F. Bergmann, Wiesbaden, 1910. Meves and Duesberg : Ueber Mitochondrien bzw. Chondrioconten in den Zellen junger Ernbryonen, Anat. Anz., vol. 31, 1907. Meves and Duesberg : Die Chondriosomen als Trager erblicher Anlagen, Arch.

f. mikr. Anat. u. Entw., 1908. Meyer, R. : Ueber die fetale Uterusschleimhaut, Zeit. f . Geburtsh. u. Gynak., vol.

38, 1898. Meyer, R. : Zur Entstehung des doppelten Uterus, Zeit. f . Geburtsh. u. Gynak., vol. 38, 18986. Meyer, R. : Ueber epitheliale Gebilde im Myometrium des fetalen und kindlichen Uterus einsehl. des Gartner'schen Ganges, Berlin, 1899. Meyer, R. : Ueber sogenannte Vornierenreste und das nephrogene Zwischen blastem bei mensehlichen Embryonen und ihre eventuelle pathologische Per sistenz, Charite-Annalen, vol. 33, p. 649-656, with 2 text-figs., 1909. Meyer, R. : Zur Kenntnis des Gartner'schen Ganges besonders in der Vagina und dem Hymen des Menschen, Arch. f. mikr. Anat., vol. 73, 1909. Meyer, R. : Zur Entwicklungsgeschichte und Anatomie des utrieulus prostaticus beim Menschen, Arch. f. mikr. Anat. u. Entw., vol. 74, 1909b. Meyer, R. : Ueber erosio portionis uteri, Deut. patholog. Gesellsch. Erlangen, 1910. Meyer, R.: Die Erosion und Pseudoerosion des Erwachsenen, Arch. f. Gynak..

vol. 91, part. 3, p. 1-35 with 1 plate and 3 text-figs., 1910. Meyer, R. : Die Epithelentwicklung der cervix und portio vaginalis und die pseudoerosio congenita (Kongenital. histol. Ectropium), Arch. f. Gynak., vol. 90, part 3, p. 1-20 with 1 plate, 1910. Meyer, R. : Beitrag zur Frage nach der Genese der im Urogenitalgebiet vorkom menden Mischgeschwiilste und Teratome, Reprint from the Charite An nalen, vol. 34, p. 1-23 with 10 text-figs., 1910. Nagel, W. : Ueber die Entwicklung des Uterus und der Vagina beim Menscken, Arch. f. mikr. Anat., vol. 36, 1891. Nagel, W. : Entwicklung und Entwicklungsfehler der weiblichen Genitalien, in Veit's Handbuch der Gynak., J. F. Bergmann, Wiesbaden, 1897. Ogushj, K. : Zur Frage des mensehlichen Eidotters, Anat. Anz., vol. 37, p. 83-86 with 1 fig. Papin, E. : Contribution a, Petude des anomalies de Puretere, duplicite et bifidite des ureteres (1. mem.), Revue de Gynecol., vol. 15, p. 108-132 with 26 figs., 1909. Perna, G. : Sullo sviluppo e sul significato delP uretra cavernosa delP uomo, Soc.

med. chir. di Bologna Adunauze, 1909. Perna, G. : Sullo sviluppo e sul significato delP uretra nelP uomo, Arch. Ital. Anat.

e EmbrioL, vol. 8, p. 145-154, 1909. Peter, K. : Untersuchungen fiber Bau und Entwicklung der Niere. I. Die Nieren kanalchen des Menschen und einiger Saugetiere, G. Fischer, Jena, 1909. Pick, L. : Gebarmutterverdoppelung und Geschwiilstbildung unter Berucksich tigung ihres atiologischen Zusamrnenhanges, Arch. f. Gynak., vol. 52, 1896. Pick, L. : Zur Anatomie und Genese der doppelten Gebarmutter, Arch, f : Gynak., vol. 57, 1898. Piquand, G. : Les uterus doubles. Anatomie et developpement, Revue de Gynecol., vol. xv, p. 401-466, 1910. Popoff, N. : Zur Morphologie und Histologie der Tnben und des Parvoariums beim Menschen wahrend des intra- und extrauterinen Lebens bis zur Pubertat, Arch, f . Gynak., vol. 44, 1893. Popoff, N. : L' ovule male et le tissu interstitiel du testicule chez les animaux et ehez Phomme, Arch, de Biologie, vol. 24, p. 433-500 with 3 plates, 1909. von Recklinghausen: Die Adenomyome und Cystadenome des Uterus und der Tubenwandungen, ihre Abkunft von Resten des Wolffschen Korpers, Berlin, 1896. Reichel, P. : Die Entwicklung der Harnblase und der Harnrohre, Verb. phys.

med. Gesellsch. Wiirzburg, 1893. Riedel: Entwicklung der Saugetierniere, Unters. anat. Inst. Rostock, 1874. Rielander, A.: Das Paroophoron (vergl. anat. u. path. -anat. Studie), Marburg, 1904. Roesger., P. : Zur fetalen Entwicklung des mensehlichen Uterus, insbesondere seiner Muskulatur, Festschr. z. 50 jahr. Jubil. d. Gesellisch. Geburtsh. u. Gynak., Berlin, 1894. Romanowsky, R. and von Winiwarter, J. : Dystopia testis transversa, Anat. Anz., vol. 26, 1905. Rubaschkin, W. : Ueber die Urgeschlechtszellen bei Saugetieren, Anat. Hefte, No. 119 (vol. 39), p. 603-652 with 4 plates and 6 text-figs., 1909. Rubaschkin, W. : Chondriosomen und Differenzierangs-prozesse bei Saugetier embryonen, Anat. Hefte, No. 125 (vol. 41), with 4 plates and 15 textfigs., 1910. Rudolph : Eine Hemmungsbildung weiblicher Geschlechtsorgane, Med. Dissert., Bonn, 1909. Schoener, O. : Bestimmung des Geschlechtes am mensehlichen Ei vor der Befruch tung und wahrend der Schwangerschaft, Beitrag z. Geburtsh. u. Gynak., vol.

14, p. 454-475, 1909.

Schreiner, K. E.: Ueber die Entwieklung der Amniotenniere, Zeit. f. wiss.

Zool., vol. 71, p. 1-188, 1902. Schweigger-Seidel : Die Nieren des Menschen und der Siiugetiere in ihrem feineren Ban, Halle, 1865. Seitz, L. : Ueber die Form der Ureteren, speziell bei Feten nnd Nengeborenen, Beitr. z. Geburtsbilfe u. Gyniik., vol. 13, with 5 figs., 190S. Spuler, A. : Ueber die normale Entwieklnng des weiblichen Genitalapparates, Veit's Handbuch der Gyniik., vol. 5, J. F. Bergmann, 1910. Stoerk, 0. : Beitrag zur Kenntnis des Aufbaues der menschlichen Niere, Anat.

Hefte, No. 72, 1904. Stricht, van der : La structure de l'oeuf des mammif eres. II Partie, Bull, de l'Acad. roy. de med. de Belgique, 1900. Stricht, van der: La structure de l'oeuf des mammif eres. Ill Partie, Mem.

publ. par la classe des Sciences de l'Acad. roy. Belgique, 1909. Tandler, J. : Ueber Vornierenrudimente beim menscblicben Embryo, Anat. Hefte, 1905. Tandler, J. : Ueber den Einfluss der innersekretoriscken Anteile der Geschlechts driisen auf die aussere Erscbeinung des Menschen, Klin. Wochenschr., vol.

23, No. 13, p. 1-24, Vienna, 1910. Tausig, F. J. : Die Entwieklung des Hymen, Monatsscbr. f . Geburtsh. u. Gyniik., vol. 30, p. 696^705 with 13 text-figs., 1909. Thiersch, C. : Bildungsfehler der Ham- und Geschlechtswerkzeuge des Mamies, Ulustr. Med. Zeitschr., vol. 2, 1S52. Tiedehann, Fr. : Von den Duverney'schen, Bartholini'schen oder Cowper'schen Driisen des Weibes, Heidelberg and Leipzig, 1840. Toldt, C. : Untersuchungen liber das Waehstum der Nieren des Menschen und der Siiugetiere, Sitzungsber. Akad. Wissensch. YVien, Abt. Ill, vol. 69, 1874. Tourneux, J. : Sur les premiers developpements du tubercle genitale etc. chez le mouton, Comptes rendus Soc. biol., ser. 8, vol. 5, 1888. Tourneux, J. and Legay : Developpement de l'uterus et du vagin depuis la fusion des conduits de Midler a la naissance, Comptes Rendus Congr. internat.

sciences med., Copenhagen, 1887. Tschaschin, S. : Ueber die Chondriosomen der Urgeschlecbtszellen bei Vogel embryonen, Anat. Anz., vol. 37, with 8 text-figs., 1910. Veit, O. : Ueber das Vorkommen von Urnierenrudimente und ihre Beziehungen zur Urniere beim Menschen, Marburger Sitzungsber., 1909. Versari, R.: La morfogenesi della guaina dell' uretere umano, Atti R. Acad. Sci.

Med., 1909; Palermo, 1910. Waldeyer, W. : Eierstock und Ei, Engelmann, Leipzig, 1870. Weber. S. : Zur Entwicklungsgeschichte des uropoetischen Apparates bei Saugern, mit besondei-er Beriicksichtigung der Urniere zur Zeit des Auftretens der bleibenden Niere, Dissert, med., Freiburg i. Br.. 1892. Werth and Grusdew : Ueber die Entwieklung der menschlichen Uterusmuskula tur. Arch. f. Gyniik., vol. 55, 1898. Wimmer, H. : Doppelbildungen an den Nieren und ein Versuch ihrer entwick lungsgescliichtlichen Deutung, Virchow's Arch., vol. 200, p. 487-522 with 14 figs., also as Dissert, med. Munich. 1910. von "Winckel : Ueber die Einteihrag, Entstehung und Benennung der Bildiin?s hemmungen der weiblichen Sexualorgane, Volkmann's- Sammlung klin.

Vortrage, 251-252, Leipzig, 1899. von Winiwarter, II.: Recherches sur l'ovogenese el l'organogenese de l'ovaire des mammif eres (Lapin et Homme), Arch, de Biol., vol. 18, p. 33-199 with 6 plates, 1900. Wyder, Th. : Beitriige zur normalen und pathologischen Histologie der menschlichen Uterusschleimhaut, Arch. f. Gvniik.. vol. 13, 1S7S.