Talk:Paper - Development and growth of the metanephros or permanent kidney in chick embryos eight to ten days incubation (1922)

From Embryology

Development And Growth Of The Metanephros Or Permanent Kidney In Chick Embryos (Eight To Ten Days' Incubation)

Rienhoff WF. Development and growth of the metanephros or permanent kidney in chick embryos (eight to ten days' incubation). (1922) Johns Hopkins Hospital Bulletin 33: 392-405.

By William Francis Rienhoff, Jr., M. D.

From, the Department of Surgery of the Johns Hopkins University, and the Department of Embryology, Carnegie Institution of Washington.

The story of the development of the metanephros, the organ which forms the permanent kidney in the higher mammals, occupies a volumiuous literature and is based on the work of many competent obsenei*s extending over a long period of time. Herring (1900), Schreiner (1902), Stoerk (1904), Huber (1905), and Felix (1914), in their classical contributions to the subject, have given a very complete resume of the work of previous observers whose investigations, made by means of fixed sections and reconstructions, were continued until this method could yield no further results. However, with the perfection of the tissue-culture method it became evident that by this means of approach new results might be obtained in this field. Carrel and Burrows (1910) were the fli-st to make cultures of the kidney. They observed, after five or six days, that tubes had grown into the plasma for a short distance and that there seemed to be a lumen limited by epithelial like cells. These tubes had, they thought, the appearance of renal tubules. Champy, in 1914, also cultivated renal tissue, but he claimed that a dedifferentiation of the entire renal anlage into an indifferent epithelium took place. In 1920 Atterbun- grafted the metanephric anlage of chick embn'os (6 to 7 days' incubation) in the allantois, with the result that the already formed epithelial tubules proliferated and the less differentiated nephrogenic tissue acquired an epithelial arrangement. Hematopoiesis also occurred and glomeruli developed.

In the hope of clearing up some of the controversial points in the development of the kidney, a study of the metanephros by means of cultures, as well as by fixed serial sections and spreads of living tissue, was undertaken. The results obtained in regard to the earliest stages of development of the metanephros, the anlage and early developmental stages of the renal evagination and nephrogenic tissue, and the later development of the iiriniferous tubules, confirmed those of Schreiuer and Huber and therefore need uo further discussion. Only the observations that differ materially from those of previous observers, such as the growth of the collecting tubules, differentiation of the anlage of the uriniferous tubules, and development of the glomerulus, together with the formation of the blood-vascular system, will be discussed in this paper.

I am indebted to Dr. and Mrs. Warren H. Lewis, not only for advice and assistance, but also for the technique employed, which made possible the cultivation and observation of the renal tissue.

Material and Methods

Cultures were made by explantiug small pieces of chick embrj'os after from 8 to 10 days' incubation. Tlie pieces to be explauted were chosen with the aid of the dissecting microscope from the most inferior medial pole of the mesouephros and the superior medial pole of the metanephros, because, as is commonly known, the differentiation of the mesonephrovS proceeds antero-posteriorly, whereas in the metanephros it is postero-anteriorly. For this reason the degree of differentiation at any age depends to some extent on the part of the renal body from which the specimen is taken. The areas chosen were the least differentiated embryologically and the most isolated anatomically. The tissue was cut up into pieces as thin as possible and placed on sterile coverslips, in a small drop of Locke- Lewis solution (85 c.c. Locke's solution plus 15 c.c. chicken bouillon plus 0.5% dextrose). The coverslip was then inverted so as to form a hanging drop, sealed onto a ring of vaseline, and incubated at 39° C. The cultures were observed at varying intervals in a warm box at 39° C. Some of them, after varying periods of growth, were fixed either by means of iodine vapor or of Zenker's fluid and stained. Serial paraffin sections 5 micra in thickness were cut, stained with hematoxylin-eosin, and mounted separately. The entire metanephric body of both injected and non-injected chicks was studied at different ages in fresh spreads.


The urino-genital system of chick embryos from 6 to 10 days' incubation can be dissected with considerable ease without the microscope. When the abdominal cavity is opened and the stomach and intestines have been removed, the mesonephric bodies appear as two greenish-red structures filling up a large part of the remaining space in the abdomen. They are situated one on each side of the vertebral column and extend from just below the attachment of the liver to a point immediately above the bifurcation of the abdominal aorta. They are more or less bilaterally symmetrical and seem to be connected across the mid-line by only a thin sheet of mesothelium. Their dimensions vary according to the age of the embi-j'o; between 6 and 10 days' incubation they average about 1 mm. in diameter and about 3 mm. in length. They are somewhat wedgeshaped pyramidal bodies, diminishing in width toward the caudal pole. The color is characteristic and is in marked contrast to that of the metanephroi and other abdominal viscera. Over the glistening surface can be seen many pin-point red spots, the glomeruli. At the upper (cephalic) poles are the gonads, two opague white tubular-like bodies, which, starting above and behind the pointed upper poles of the mesonephric bodies, continue across the uppennost anterior surfaces and course caudalward along the antero-medial margins, ending in a point corresponding to one-half the length of the mesonephric bodies. From the dorsal surface, and continuing below the lower pole of each mesonephric body, is the mesonephric or Wolffian duct, which serves as the excretory duct for the mesonephric body or embryonic kidney. These small white ducts approximate each other in a downward course and end in the cloaca. The branching connections of the Wolffian duct with the mesouephros, the mesonephric ureters, could be distinctly seen by elevating the lateral margins of the mesonephric bodies. Along the vertebral column, dorsal and also immediately posterior to the mesonephric bodies, lie two bilaterally symmetrical translucent structures, the metanephroi. The genitourinary apparatus is entirely covered by a thin layer of cubical epithelium which is known as mesothelium or coelomic epithelium. In the youngest ages, the metanephros is completely obscured by the overlying mesouephros, which reaches its greatest volume at about the eighth day. From this time on, however, the metanephros increases rapidly in size, with the result that in a 10-day embryo the mesouephros appears to be a relatively small body situated on the anterior surface of the metanephros. The supply of blood to the metanephric body between these ages is very small, while the blood-flow through the mesouephros is quite abundant, a fact which accounts for the difference in the color of the two bodies. The shape of the metanephros changes constantly during the growth and development of the embryo, so that no definite shape can be described as typical for all ages. However, in general it may be said that the shape approximates that of a dumb-bell, being broad at both poles and Harrow in the middle. This is well shown for older embiyos (18 days) by Minoura (1921, Fig. 8). At the age of 6 days the metanephric body has a smooth exterior surface, with no lobulations. A little later, however, the bodies begin to show definite lobulations, which by the tenth day have become discrete lobes completely covered by coelomic epithelium.

The entire metanephros was dissected out and placed on a slide, thus making it possible to study the fresh tissue at different ages with the aid of the microscope.

In the 6-day embryo it was found that the metanephric tubule, which sprouts off the Wolffian duct, had already grown into the metanephric body, completely traversing the undifferentiated tissue from the posterior to the anterior pole. Along the entire length of the metanephric tubule the primary collecting tubules had budded off in three main planes, lateral, dorsal, and ventral, and there seemed to be an enlargement of the lumen of the metanephric tubule at the anterior and posterior poles, at which places the metanephric body is broader. Each of the primary collecting tubules, although they had grown only a slight distance from the metanephric tubule, had just divided dichotomously into two secondary collecting tubules. Xo further division, that is, beyond the secondary tubules, was ever observed at this age. Immediately about the primary- and secondai-y collecting tubules was a sharply defined covering, consisting of a single layer of endothelial cells. This structure will be described in detail later. Between the bases of the primarj collecting tubules, near the posterior pole of the metanephric body, sinuses had differentiated out of the tissue surrounding the ba.scs of these primary collecting tubules. These spaces or sinuses were lined with flat endothelial cells which gave rise .to blood elements. This was the fii-st appearance of sinus formation observed in the metanephrogenic body.

At 7 days the branches of the primary collecting tubules, namely, the secondary tubules, had again divided dichotomously into tertiaiy tubules. The branching appeared always to occur in a forked manner, the tubules growing toward the peripherj-. There was a marked increase in the length of the primary and secondary collecting tubules during the sprouting of the tertiary tubules. The apparent dilatation of the lumen of the metanephric tubule at the anterior and posterior poles had disappeared by this time, but in these areas there was a much more rapid growth of the collecting tubules than in the intermediate region. This growth was perhaps somewhat more pronounced at the posterior than at the anterior pole, for, as is well known, the metanephros differentiates poster o-anteriorly. In the body of the metanephros the formation of sinuses proceeded hand-inhand with the elaboration of the collecting tubule tree, replacing the undifferentiated tissue about the bases of the primary and secondary collecting tubules. Towards the peripheiT of the metanephric body, strands of mesenchymal cells, arranged in fe.stoons, divided the cortex into a .series of lobes which corresponded roughly to the secondary collecting tubules. The definition of this lobulation was more pronounced in the 8-day embiyo, constituting complete separation of the metanephrogenic lobules into distinct lobes surrounded entirely by a layer of cubical cells and arranged always in relation to the secondary collecting tubules. The length of the primary, secondaiy, and tertiary tubules increased greatly with this lobulation, the tertiaiy tubules being entirely intralobar, while the primary, as well as the secondarj' tubules, were extralobar. In the anterior and posterior poles of the metanephros the tubules grew to a greater length and gave off more sprouts, resiilting in a gi-eater number of secondaiw and tertiary branches in these regions. As has been stated above, the lobes were establi.shed in the 8-day embryo, but an intralobar lobulation was obsen'ed to begin immediately following the lobe fonuation. This subdivision of the lobe into smaller lobules persisted until the elaboration of the collecting tubule tree was complete. In the 9-day embryo it corresponded to the branching of the quarternaiw tubules or the first intralobar division of the tertiar}^ tubules. Up to this time no sign of the secreting or convoluted tubule had appeared. About these quarternary tubules, however, differentiation of the future convoluted tubule had occurred and from now on, with each succeeding division of collecting tubules, there, was a simultaneous formation of convoluted tubules (including glomeruli, etc.) in the angles of the branching collecting tubules. Eventually, each branch of the collecting-tubule tree, with its corresponding convoluted tubules, was surrounded by mesenchymal tissue which formed a sort of capsule about it as a unit. In the kidney, therefore, as in the lung, liver, etc., there is a definitive lobule, which is the elementarj- unit of the entire excretory apparatus. This unit con.sists of the terminal branch of the collecting tubule, the convoluteil tubules, including the glomenili, and the surrounding endothelial and mesenchymal tissues. Owing to the lobulation, the growth and division of the collecting-tubule tree inside the lobe was much more elaborate in the 10day than in the 9-day embrj-o. The distribution of the lobes in the 10-day embryo corresponded exactly to the distribution of the secondaiy collecting tubules. As a result, the greatest number of lobes were found composing the caudal pole, while the next greatest number constituted the cephalic pole. The lobes making up the intermediate region were evenly and regularly distributed in all planes, so that all the lobes at any given level were practically at the same stage of development regardless of their distance from the original metanephric tubule. Differentiation of the lobe always proceeded from the base towards the periphery. The most developed convoluted tubules were associated with the quarternary branches of the metanephric tubule. From this base or center the cycle of development was repeated for each generation of the renal units as we approach the periphery of the lobe towards the cortex and also towards the future columns of Bertini, not only in the longitudinal diameter but also in the transverse diameter. The development of the systems of sinuses obeyed the same general principle, that is to say, they were formed first at the base of the lobe and later at the periphery.

In conclusion, it may be stated that, although the posterior portion of the nietanephric body grew and developed more rapidly than the anterior portion, nevertheless differentiation was always from the nietanephric duct towards the periphery, not only in the lobes and lobules but also in the blood-vascular system and in the supporting tissue; so that even in the older embryos the most mature elements were located near the base of the lobe, while the immature structures occupied the cortical regions.

Growth and Development Within the Explant

In cultures of chick embryos of from 8 to 10 days' incubation the evolution of all the elements making up the excretory unit of the permanent kidney could be followed in detail throughout their differentiation and development. Usually the explauts contained at least one lobe with its lobules, in different stages of development according to the age of the embryo from which the piece was taken. Observations on the inesonephros were made for a comparison, but these will be described only where they differ in some fundamental way from those made upon the metanephros. The component parts of the excretoi-y unit in cultures of different ages will be described in the order of their differentiation and development : i. e., collecting tubules, nephrogenous tissue (secreting or convoluted tubules and glomeruli), and endothelium, including the blood-vascular system. The growth and development within the explant, as well as the growth extending out from the explant, were studied.

Eight-day Embryos

The living explant was slightly yellowish and so transparent that the structure of the metanephros could be readily observed. The most striking characteristic of the explant was the collecting tubules, which at this age have undergone extensive growth from the nietanephric ureter into the undifferentiated metanephrogenic tissue and become branched like a tree, as was mentioned when speaking of the morphology. It is in the growth of this element of the excretory unit that the inesonephros and metanephros differ most \^^del3^ In the inesonephros the collecting tubule, from the beginning of its growth from the primary excretoiy duct, is a single uubranched tubule, twisted upon itself, opening individually into the excretory duct, and destined to become connected with one convoluted tubule. It is coiled from the beginning, and becomes more so as it ajiproximates maturity. The characteristic tree like formation observed in the metanephros was never found in the inesonephros.

In the explauts from 8-day einbiyos masses of undifferentiated metanephrogenic tissue surrounded the ends of the collecting tubules and, as there were no other differentiated elements of the excretory unit, the collecting

tubule tree (primary, secondai-y, and tertiary tubules) stood out strikingly because of the marked contrast between the differentiated and undifferentiated tissue (Fig. 1) . After the culture had grown for 20 hours, the explant became flattened out and attached to the cover-slip, owing to the migration of connective-tissue cells, neiwe filaments, endothelium, and tubules. Coincident with the migration was a division of these cells, thus forming an extensive area of new marginal growth (Figs. 3, 13, and 14). The migration of cells from within the explant to form the marginal growth left the collecting tubule intact and sharply defined (Figs. 2 and 3). The definition was aided by a single layer of highly refractive endothelial cells which completely enshrouded each collecting tubule and its branches, even covering the terminal ends of the tube (Figs. 2, 6, and 18). The endothelial cells formed a vei-y thin sheet, through which the bases of the columnar cells forming the tubule could be seen as triangular or prismatic cellular bodies. The collecting tubules grew in a comparatively straight line from the tertiary tubules and branched out in a characteristic and regular manner, i. c, always from the base towards the periphei-y of the lobe (Figs. 1, 2, and 8). There may be a slight, gentle curvature to the course of the growth of either the main trunk or its branches, but never was there seen the bending and intertwining observed in the tubules of the niesonephros or later in the convoluted tubles. There was a regular gradation in diameter from the tubules making up the central trunk to those constituting the peripheral arborization. The distance between the branches along the main stem was the same, although the branches came off in any plane (Fig. 2). The most common type of branch formation was dichotomous or trichotomous (Fig. 1), although there was occasionally a branching into fours. The branches always came off from the main stem at angles greater than ninety degrees, and grew in a lateral or peripheral direction. However, it must be stated that, while there was no constant direct trunk or axis of the tree growing through the center of the lobe with the branches coining off laterally, yet there was always one main branch, which maintained a general peripheralward course of growth and remained approximately the axis of the lobe (Fig. 2). The lumen of the collecting tubule was established at this age and patency was maintained by a fluid-like medium in which granules and cellular detritus were seen floating about. The lumen was bordered by a high, non-ciliated, columnar epithelium (Figs. 12 and 24) . About the ends of the tubules the undifferentiated metanephrogenic tissue was observed in the foi-m of a cap of oval cells arranged radially. These were packed together quite closely, forming a layer about four cells thick, which was, as a general rule, more dense and thick immediately over the ends of the tubules and tapered around all sides of the tubules for a short distance. Peripheral to this cap was another layer of undif ferentiated metanephrogenic tissue, the cells of which were less densely packed, contained fewer granules, and had no characteristic arrangement. These two regions were first described by Schreiner (1902) as the inner and outer zones, respectively, of the metanephrogenic tissue. Beyond this was undifferentiated mesenchyma, which followed the same arrangement as the cap and, by dipping down between the ends of the branches of the collecting tubules, marked out what would eventually be the lobulation of the permanent kidney (Fig. 2) .

In cultures 48 to 144 hours old, the explant had become so transparent that the actual growth of the collecting tubule was easily followed. This growth took place not only by mitotic division of the cells all along the length of the tubule, but also from the ends of the tubule, the latter being the more active of the two and the more effective in increasing the length of the tubule. In the end of the tubule in which growth and branching were about to take place, there occurred a symmetrical bulbous or ampullar formation, due to a general proliferation and change in shape of the cells making up the the ends of the tubule. In areas of active proliferation and rapid growth, there seemed to be a metamorphosis always to the spherical type of cells with a subsequent reversion to the adult type. As a result of this proliferation, there occui'red, coincideutally with the bulbous expansion, bud like projections from the blind end of the tubule. Usually these projections were situated at opposite poles of the tubule ; occasionally there were three or even four of them. Their rate of gi-owth was approximately the same. At first they appeared to be growing out horizontally to the cortical surface of the lobule; they soon altered their course, however, one growing almost perpendicularty and continuing in a general way the peripheralward growth of the tree, while the other or others tended to grow out more horizontally, making up the surrounding arborization. These bud-like projections were at first solid sprouts, but by continued proliferation and reshaping of the spherical cells into a columnar type the bud increased in length, breadth, and thickness and by a rearrangement of the cells a lumen was formed, which in a very short space of time extended almost the entire length of the sprout ; thereafter, the lumen formation progressed handiu-hand with the growth of the tubule. This phenomenon was of a different type from that described by Dr. Sabin (1920) in the formation of the lumen of the blood-vessels, for in the collecting tubule no vacuolization of the cells took place; on the other hand, a small but gradually widening cleft within the sprout, was formed by a rearrangement of the cells and a transformation in shape to that of the columnar type. The new collecting tubule branches continued to grow with a simultaneous formation of a lumen, until a certain length had been attained, when a new branching occurred. The endothelial sheet covering the tubule proliferated and grew simultaneously

with the growth of the branches, and the collecting-tubule tree was thus covered throughout with a shroud of endothelium. The undifferentiated metanephrogenic tissue was plainly visible in the cultures and its growth and development were closely associated with that of the collecting-tubule tree. Felix (1912) described this as forming a circumferential sheet which surrounded the ends of the ingrowing collecting tubules and was pushed peripheralward by the growth of the tubules after each succeeding layer of newly formed convoluted tubules and glomeruli had been formed. This, however, was not found to occur in the chick. Here the undifferentiated metanephi-ogenic tissue surrounded each individual collecting tubule that grew into the lobe; as the tubule branched, this metanephrogenic tissue was split up, some of it being carried forward by the growing branches, the rest remaining behind in the angle.s between the branches. The portion of the metanephrogenic tissue carried forward over the end of the tubule extended to cover the new branches of the tubule, partly by proliferation and partly by a peripheralward migration, as was evidenced by a slight distortion of the cellular mass.

The deposition of the undifferentiated tissue in the angles of the branching tubules was found not only in the cultures and spreads but also in the sections of the metanephros (Figs. 10, 13 and 16), and it was from this tissue that the future convoluted tubules and glomeruli always developed. This constant relation of the undifferentiated metanephrogenic tissue to the ingrowing collecting tubule was especially striking in the cultures. No matter to what extent the collecting tubule may have grown inside the explant or out into the margin, as will be shown later, there was alwa,vs the same relation maintained between it and the undifferentiated metanephrogenic tissue (Figs. 8 and 10). Since the convoluted tubules always developed in the same position with relation to the collecting tubules, and always from the metanephrogenic tissue of the inner zone and never elsewhere in the mass of undifferentiated tissue, it may, I think, be deducted that the metanephrogenic tissue of the inner zone is made up of cells that are predestined to become the anlage of these convoluted tubules. This anlage always appeared as a mass of undifferentiated tissue, quite distinct from the surrounding nephrogenous tissue of the outer zone and the collecting tubule (Figs. 10 and 15). The cells constituting the inner zone, which formed a cap-like covering over the blind end of the ingrowing collecting tubule, were at first quite similar in form. However, within 24 hours, in cultures of the 8-day embryos, morphological changes, and also changes in the general arrangement of the cells, were observed. The cap became sharply defined by three and often four layers of cells which completely ensheathed it (Figs. 15 and 18). The cells making up these layei-s were rather long and narrow, approximating the endothelial cell in shape, and stood out in marked contrast to the adjacent tissue. Between the two outermost layers appeared a space, which likewise encircled the anlage and served to define it still more sharply (Figs. 16, 18, and 24). Simultaneously with the differentiation of this endothelial sheath, there occurred not only a change in the morphology of the cells making up the cap but also a change in the form of the cap itself. The cells proliferated rapidly, becoming more spherical in shape and somewhat less granular. There was also an increase in the size of the individual cells. Thus there was foniied, as it were, a solid central core of spherical cells approximating the epithelial type (Figs. 15, 16, 18, and 19) . The difference in the polarity of these two distinct types of cells was quite striking. In the spherical cells there was a tendency toward a radial arrangement about a central core. The axes of these cells were perpendicular to the axis of the central core, while the axes of the endotheliallike cells about the periphery of the central core were at right angles to the axes of the spherical cells' (Figs. 16, and 18). In twenty-four hours' growth, before any suggestion of tubule formation, there was thus differentiated in situ two distinct types of cells, endothelial and epithelial, from one common mass of undifferentiated tissue. Furthermore, the entire anlage had become surrounded by a space lined with endothelium (Fig. 18).

Concomitantly with the differentiation of these two types of cells, the inner cell mass, composed of spherical epithelial-like cells, was observed to become shaped like an inverted comma (Figs. 19 and 20). (This comma shape was first mentioned by Ribbert in 1899. ) The head of the comma was always to the side of the ingrowing collecting tubule, while the tail covered its blind end (Figs. 16 and 19). By continued growth the comma-shaped mass increased in size and underwent a gradual transition into a solid S-shaped tubule or core. This transition required about 24 hours. At first its differentiation and growth were uniform throughout, that is, the tail end of the comma appeared quite as early and was as far advanced as the head-pole (Fig. 19). However, during the transition the most rapid growth was seen to occur in the head pole of the comma, or what I shall call the distal pole, where the spherical cells became arranged in what appeared to be a solid, sphere-like dilatation (Fig. 21). Immediately above this sphere the growth was almost as rapid, causing an angulation which resulted in the formation of the first curve of the future S-shaped tubule (Fig. 21).

Differentiation beyond this stage was not observed in the metanephrogenic tissue of the inner zone in the 8-day embryos. In the outer zone, however, growth and differentiation did occur. The future connective-tissue septa appeared at the periphery of this outer zone of the metanephrogenic tissue and with their appearance there were laid down the interlobular septa constituting the microscopic lobulation. Throughout the growth of the collecting tubule tree, lobulation was observed to keep pace with the formation of branches of the tree.

Nothing has jet been said concerning the region about the main stems and branching collecting tubules. Concomitant with the growth and development of the lobule, there was constantly left behind, in the path of differentiation and peripheralward growth, a cellular mass of tissue wTiich surrounded the trunk of the collecting-tubule tree for that lobule. This tissue constituted a supporting framework containing the blood capillaries, for the developing excretoi-j' unit. Simultaneously with the differentiation of the excretory unit, and well advanced before the appearance of the convoluted tubules, there was formed progressively a system of sinuses. As stated under the description of the morphology, the process of differentiation and development of endothelial sinuses was first observed in the undifferentiated metanephrogenic tissue between the bases of the primary collecting tubules in the 6-day embryo. The presence of a sinusoidal circulation in the mesonephroa was first mentioned by Minot (1892), but nothing was said about the existence of a sinusoidal circulation in the metanephros or the method of development of these sinuses. In cultures of the 8-day embryo all stages of development of the sinuses could be seen. The most fully developed were near the base of the lobe, while the least developed ones were found about the anlage of the excretoiy unit, toward the periphery. This formation of sinuses followed, so to speak, in the wake of the growing collecting-tubule tree (Figs. 1, 13, 14) and, as in the differentiation and development of all the other elements of the lobe, the progression was always from base to periphery. In the cultures the sinuses were sharply defined from the surrounding tissue and they also were readily seen in fixed sections, although much shrunken and distorted (Fig. 16). About the base of the lobes the sinuses were quite large and filled with a fluid medium, in which blood elements were commonly found (Figs. 5, 13, and 17). The walls of the sinuses consisted of a single layer of endothelial cells, which at times were in apposition to the endothelial layers covering the tubules, and at other times were connected with these layers of endothelium by endothelial sprigs or offshoots (Figs. 12, 16, 17, 24). The larger sinuses connected freely with each other and also with a capillary network of the lobe, so that in an embryo of this age there were two distinct blood-vascular systems, capillary and sinusoidal. Blood-islands were commonly seen attached to the inside of the walls of the sinuses and from these hemoglobin-containing cells, as well as white bloodcells of lymphocyte type, were differentiated. Sinuses also were differentiated in situ from the undifferentiated tissue making up the bulk of the metanephric body. These sinuses had their origin in the vacuolization of cells quite similar to angioblasts. These cells were rather conspicuous, clue to the fact that there were always two and sometimes three in juxta-position, making them appear as deeply pigmented multinucleated giant cells, joined together by very minute protoplasmic processes (Fig. C). In each cell there appeared a vacuole, which constantly enlarged until there remained only a rim, of cytoplasm and an eccentrically placed nucleus. The vacuolization of these cells was complete in 24 hours and fusion of the two or three original cells occurred, resulting in a small space surrounded bj' flattened out cytoplasm and eccentrically placed nuclei. This space increased in size by multiplication of the flattened-out cells and was held patent by a fluid medium. These sinuses were of different shapes and sizes and were connected with each other, as well as with the capillary network, and with the endothelial spaces about the anlage of the convoluted tubule, by means of direct sprouting of the larger sinuses. Throughout the entire life of the culture differentiation of capillaries in situ and direct growth after differentiation by sprouting was observed. Thus, the sinusoidal and the capillary systems develop in situ at the same time from a totally undifferentiated mass of tissue making up the bulk of the metanephric body. The two types of circulation were verj' similar in their differentiation and development.

In old cultures the macrophages appeared to ingest and digest tissue left behind in the wake of the developing lobe (Fig. 1-4) as well as injured tissue. Inside these macrophages red blood-cells and other forms of cellular detritus were observed. The macrophages were much more numerous in the base of the lobe.

Nine- and Ten-day Embryos In the explants of the metanephros, made from 9- and 10-day embryos, one had, of course, a more completely developed excretory unit to begin with than in the explants from 8-day embryos. Notwithstanding the fact that there Avere growth and differentiation in cultures from the latter, these were always much retarded as compared with an embryo of corresponding age. In the explant of the metanephros from 9- and lOday embi^os, the collecting-tubule tree had become much more branched and complex throughoiit the lobe. The main trunk had increased not only in length but also in diameter, and its primary branches had already divided into secondary, tertiary, and quarternary branches. The dichotomous branching of these tubules contiiiued with a lobulation or division of the lobe into smaller lobules by the rearrangement of the mesenchymal septa, corresponding to the branching tubule tree. The sinuses and blood-vascular elements formed and grew just as in cultures from the 8-day embryo, so that the essential difference between the growth of the explants in the two series lay in the differentiation and development of the secreting tubules. All stages, from completely differentiated to entirely undifferentiated metanephrogenic tissue, could be studied in the explants of 9-day embryos, as there was constantly occurring a repetition of the cycle of differentiation and growth of the complete excretory unit towards and in the future cortical portion of the lobules. Between the 9th and 10th day of incubation seemed to be the optimum age for a study of the development of the metanephros. Whether this was due to the fact that metanephrogenic tissue was of a greater vitality than at 7 and 8 days, or whether the undifferentiated cells were more nearly on the verge of differentiation and therefore less affected by the change in enviionment, I cannot say; in any event, within twenty-four hours all stages of differentiation from a comma-.shaped body to an £-shaped tubule could be observed. The main areas of growth during the transition were the sphere like group of cells at the distal pole and, immediately superior to this, the portion which \5'as to form the first knuckle of the future 3-shaped tubule (Fig. 21). At these two points proliferation was most active. The continued growth of the sphere at the distal pole, immediately under the rapidly proliferating first curve, caused a slight flattening of the sphere with a subsequent bulging about the sides and to some extent the ends of the overhanging knuckle of the first-formed cun'e. The growth of the sphere and that of the large knuckle (first curve), being parallel and in the same direction, caused an acute flexion of the solid core immediately posterior to the sphere, thus forming the second curve of the S (Fig. 22). Owing partly to the relatively slow growth of the cells making up this second curve, and partly to the pull of the rapidly growing sphere and first curve, the second curve became very small in diameter and acutely Hexed (Figs. 6, 23). The solid S-shaped mass, therefore, developed in situ, and not, as heretofore thought, from a sphere into an S-shaped tubule. The entire solid core developed from cells already present, and assumed an 2-shape because of more rapid growth in some regions than in others.

The endothelial sheath, together with the sinus formation, progressed hand-in-hand with the development of the S-shaped core, i. e.. the S-shaped core was completely covered by a layer of endothelial cells and in addition to this the entire anlage was completely surrounded and delimited by a space or sinus lined with endothelium (Figs. 12, 16, 24). The core was thus sharply outlineil. A marked difference in the refractivity of the endothelial and epithelial cells further aided in the definition of the tubule so that it was easy to distinguish changes taking place within the tubule from those in the tissue without. The close approximation of the glomerular pole to the under surface of the first curve of the S might give the false impression of a cleft formation, as stated by Ribbert (1899) and Huber (1905) instead of a solid core doubled upon itself (Figs. 21, 22). The potential space between the distal pole and the under surface of the first cuire contained the endothelial layers covering the solid tubule and those fomiiug the wall of the sinus (Figs. 17, 21, 24), so that the bud of an endothelial sac was thus placed between the glomerular end of the tubule and the under surface of the first curve. This slight indentation, however, approximated in no way the cup formation described by other investigators, notably Huber, Schreiner, and Felix.

After the comma-pattern had developed into an 3shaped tubule, there was a continued growth throughout the entire tubule. Proliferation of the cells in the glomerular end occurred more rapidly than elsewhere in the S-shaped tubule until that end became dilated and bulbous, forming a tuft of epithelial cells (Figs. 2, 6, 11, 22). At this stage no capillaries or blood elements were ever observed in this tuft, all the cellular proliferation being inside the endothelial sheath, and therefore within the tubule. The cells causing the dilatation of the glomerular end were at first all more or less round ; however, a, differentiation of these cells took place in situ along with their continued proliferation, resulting in the fonnation of flat, endothelial-like cells, and spherical epithelial cells, thus constituting the pre-glomerular tuft (Figs. 17, 23 ) . The endothelial cells, because of their shape and greater refractivity, could be traced winding about amongst the epithelial cells (Figs. 23, 24). The latter, however, formed the greater part of the cell mass producing the dilatation of the glomemlar end of the tubule. By 72 to 96 hours, collap.sed endothelial spaces could be seen winding through the epithelial tuft and everywhere covered with epithelial cells (Fig. 24). Thus the presence of endothelial cells intimately intermingled with epithelial cells wa.s observed in the bulbous glomerular end while the future secreting tubule was still in a solid state and as yet no connection with any type of circulatory system could be seen. No evidence of a growth from the endothelium covering the exterior of the tubule down into the glomerular end was observed. The cellular proliferation in the glomerular end proceeded rapidly, the dilated bulbous end being completely filled and much distended in comparison -nith the remainder of the tubule. The curve of the tubule immediately proximal to the glomerular end became reduced to a very small tube whose walls were formed by a rather low cuboidal epithelium one cell in thickness, covered by an endothelial sheath. In the loop of the £ immediately proximal to the glomerular end the formation of a lumen was first observed (Fig. 23 1. This followed the same course as in the collecting tubules, i. e., the lumen formed by a rearrangement and separation of the cells and not by vacuolization, as was observed in the vascular system.

The cells abutting against the lumen were the first ones to become cuboidal in shape. The rearrangement of the cells about the lumen progressed distally and proximally with about equal rapidity. The lumen formation began about the inferior half of the spherical tuft of cells, then progressed to the superior half, until the central tuft was completely separated from the walls of the glomerular end of the convoluted tubule, except at one pole, usually distal to the point of transfonuation of the secreting tubule into the glomerular end (Figs. 6, 23, 24).

This lumen formation in the glomerular end was the anlage of the capsule of Bowman, the cells bordering the lumen constituting the layers of the capsule (Fig. 24). The flat epithelial cells formed the parietal layer, while the cuboidal cells made up the visceral layer. Immediately before the fornuition of the lumen the spherical tuft of cells attained its maximum size, decreasing from that time on until the glomerulus began to function, when it again enlarged. With the completion of the lumen formation and the changes in shape of the bordering cells forming the anlage of Bowman's capsule, the appearance of an invagination of the remainder of the spherical tuft was obtained. However, the etched-like outline of endothelial cells enabled one always to separate that which occurred inside the tubule from that which occurred outside. Any invagination or cup formation that involved the entire glomerular end of the secreting tubule would of necessity cause a disturbance of the contour outlined by the ensheathing endothelial covering. If sections of merely the cuboidal cell layer of Bowman's capsule constituting the most distal margins of the glomerular end of the tubule are reconstructed, then one obtains an incorrect impression of a tubular invagination.

A connection was now established between the newly formed space or sinus surrounding, the secreting tubule and the larger sinuses already foi-med. This connection was by means of direct sprouting oft" of endothelial strands from the larger sinuses which became confluent with the sinuses differentiated in situ about the anlage of the secreting tubule. Thus the endothelium-lined space between the superior surface of the glomerular end and the second curve of the 2-shaped tubule was in direct communication with the larger sinuses. A continuation of endothelium was theu established from the endothelial elements within the glomerular epithelial tuft through the sinus about the secreting tubule by way of endothelial sprouts into the larger sinuses about the collecting tubules (Figs. 2, 24). The endothelial sprouts connected the glomerulus with the sinuses always through the attachment of the glomerular tuft of ceHs to the wall of the tubule ; that is, at the point where the capsule of Bowman was reflected, which usually lay distal to the urinary pole. At this point the visceral and parietal layers of the capsule were confluent. In the older cultures it was found that the sinuses became connected with the capillaries that developed in their neighborhood.

Hematopoiesis occurred in the spherical tuft of the glomerular end with the formation of red blood-cells, plasma, and white blood-cells having the appearance of

IvDiphocytes (Figs. 17, 23, 24). The forruatiou of the blood-elements was usually first noted in that portion of the tuft adjoining the secreting tubule and progressed from this point throughout the glomerular portion. It occurred at about the same time the connection with the sinuses took place. It must be borne in mind, however, that in the cultures no circulation was present. At no other point in the entire excretory tubule did hematopoiesis occur, a fact which also indicates the endothelial nature of the cells inside the spherical tuft Lumen formation by thinning out of the cells, characteristic of the blood-vascular .system elsewhere, was found to obtain in the glomerular tufts. The blood elements made the definition of the endothelial channels much more pronounced and the latter could be traced with ease, winding among the epithelial cells and covered by a single layer of these cells.

After the differentiation, the convoluted tubule con tinned to grow in exactly the same manner as the collecting tubule, that is, by proliferation of the cells along the entire tubule or at the end nearest the collecting tubule. The convoluted tubule grew always toward the collecting tubule, until finally the proximal end of the former abutted against the side wall of the latter. This contact usually, though not always, was in the immediate region of a newly foi-med branch of the collecting tubule. The anastomosis of the convoluted and collecting tubules could be observed in cultures of the 9-day embryos after 24 hours' growth. In general, it required about that length of time to effect complete continuity of the convoluted and collecting tubule lumina. The continued growth of the secreting tubule against the wall of the collecting tubule exerted sufficient pressure to indent the latter and cause a distortion of its constituent cells (Fig. 23). The secreting tubule grew into the collecting tubule at an angle of about forty-five degrees. The cells in the wall of the collecting tubule formed a sort of arciform arrangement about the ingrowing blind end of the convoluted tubule. There was a rapid proliferation of the cells constituting the latter and the cells reverted to a spherical type, as was observed in the growth of the col lecting tubule. As a result of this proliferation, the end of the convoluted tubule protruded like a wedge through the wall of the collecting tubule into the lumen. This was not due merely to pressure from the outside against the wall of the collecting tube; there was an actual growth of the convoluted tube inside the wall of the collecting tube, the cells dividing and making room for the increased number. An interruption in the continuity of the wall of the collecting tubule was thus effected (Fig. 24).

Lumen formation in the secreting tubule followed upon the proliferation of the proximal or blind end, so that about two hours after the secreting tubule had grown through the wall of the collecting tubule the lumina of

the two had become continuous (Fig. 24). This continuity was proved by the coursing of fluid containing granules back and forth from one tubule into the other. After the anastomosis of the convoluted and collecting tubules, there was continued growth of the former throughout its entire length, as evidenced by mitotic figures. The greatest growth activity, however, occurred in the convolutefl portion next to the collecting tubule, which in the adult kidney would correspond to the distal pars convoluta.

Developjient in Marginal Outgrowths The outgrowths from explants of the metanephros (of embryos of 8 to 10 days inclusive) were made up chiefly of four types of tissue — mesenchyme, renal epithelium, endothelium, and nervous tissue. The marginal growth began immediately after implantation. In from two to four hours after explanting, proliferation of mesenchymal cells about the margin and growth of nerve-fibers into the marginal zone could be made out. Growth of the four elements of the marginal zone increased steadily throughout the life of the culture up to 5 days. In the beginning it consisted of only a thin sheet about one cell in thickness. The most advanced edge of the outgrowing margin always remained quite thin and attenuated, but as the growth increased in width, so did it increase in depth. The different elements composing the marginal outgrowth proliferated and grew with different degrees of rapidity, but there was little difference ultimately in the extent to which they grew.

Usually the first tissue to grow was the mesenchyme; it was also the most hardy and prolific of all the tissues (Fig. 5). In appearance, these cells were typical fibroblasts connected with one another by long branching, which were direct extensions of the cytoplasm (Fig. 7). At first there was no definite arrangement of the mesenchyme cells; they simply grew out in a flat spreading sheet with a centripetal tendency (Figs. 4, 14). However, after 48 hours, when the g:i'o\\i:h near the explant had become thicker in different areas, definite patterns and arrangements of cells could be seen, although tlie attempt of the cells to assume their normal formation was often abortive and resulted in bizarre disturbances of organization (Fig. 14). This interference with the tissue in its effort to acquire its predestinetl form was due, no doubt, to surface tension and adhesion to the coverslip.

The renal epithelium grew out readily, but like the mesenchyme, in different forms in the various cultures. Proliferation and growth began immediately at the margin. Arising from the excretory unit (the collecting tubules, convoluted tubules, glomeruli, and undifferentiated nephrogenous tissue) growth proceeded in three fundamentally different ways: (1) with no particular formation, as in a flat sheet (Fig. 5) ; (2) as a perfectly organized element, carrying with it undifferentiated tissue, which later differentiated in the marginal outgrowth (Figs. 8, 9) ; (3) out into the margin where an abortive attempt at formation or organization was made (Figs. 3, 4, 14). Wherever there was an end of a collecting tubule or convoluted tubule near or immediately at the margin of the culture, there was always an outgrowth of renal tissue, which appeared quite as early and grew as rapidly as the mesenchymal tissue. When growing out hand-in-hand, as it were, with other tissue of the margin, the renal tissue always appeared as a flat sheet of cubical cells closely associated with one another. This sheet was usually one cell thick and often bounded on all sides, except towards the periphery of the growth, by outgrowing mesenchjnual and endothelial cells. In these flat outgrowths there was never any attempt at organization or difl^erentiation ; the outgro^rth consisted simply of a multiplication of already differentiated cells.

In the event that a well-advanced marginal growth had preceded the outgrowth of renal tissue by a few hours so that there was some depth of cells into which the renal tissue could grow, then the outgrowth and differentiation occurred exactly as they did inside the explant. In other words, the collecting tubule grew into the marginal zone as a tubule intact, just as the collecting tubule grew inside the explant, carrying with it undifferentiated nephrogenic tissue which went on to complete differentiation in the margin (Figs. 8, 9, 11). The excretory tubule maintained the same relation to the collecting tubule, with regard to position and growth, that it maintained inside the explant: when it differentiated, it always did so in situ, i. €., in the region or at the site of the angle formed by the branching of the collecting tubule, afterward becoming S-shaped by direct growth of the already differentiated cells. There was never a long, straight, direct outgrowth of the convoluted tubule such as there was in the collecting tubule.

When the marginal gi'owth of cells was not thick, the attempts at organization were abortive, although differentiation occurred. In all marginal growths there was a certain point beyond which the entire margin grew out flat and "went wild," that is, all formation or organization was lost and the individual elements broadened out into a flat sheet. This transition could be followed in most of the cultures that sunived long enough.

The third element in the marginal growth was the endothelium. As was the case with the renal tissue, the endothelium grew out in different ways. In the first place, there was a sheet-like growth, with greater intervals between the individual cells, however, than occurred in the mesenchyme or renal tissue (Fig. 3). In fixed preparations the endothelial cells were fibrillated and connected by long branching processes. No formation of blood-vascular elements was seen in this type of outgrowth. On the other hand, capillaries grew out in arciform loops from other capillaries already formed in the explant. The growth of could be watched in detail.

The development in situ of sinuses, blood-islands and capillaries occurred in the marginal outgrowth wherever the necessary undifferentiated tissue had been carried out from the explant. Among the tissue of the explant and marginal outgrowth, at all ages, long nerve-fibers, also ganglia, were sometimes observed. In the marginal growths, therefore, there occurred differentiation and organization of the renal parenchyma, as well as of the other basic tissue elements making up the renal body.

Differentiation, followed by a marked distortion due to the attempt of the cells to assume their usual organized positions, afforded an excellent opportunity for study. One could find places in which epithelial cells had differentiated in small flattened circular areas. Among these were highly refractive endothelial cells, intimately associated with the epithelial cells, being interwoven among them in exactly the manner noted in the development of the glomerulus. As this association of endothelium and epithelium was not observed anywhere else in the marginal growth, although there were many outgrowths of renal epithelial cells, the logical inference is that these small translucent circular areas were probably made up of cells destined to develop into the glomerulus.


Most observers agree on the growth and division of the collecting-tubule tree and the coincident splitting up of the nephrogenous ti.ssue, thus affording a cap or covering for each branch of the collecting tubules. As far as I have been able to find, however, none of the previous observers have described the endothelial covering of these tubules, although several figures given by Herring show it.

In regard to the nephrogenic tissue, all observers agree that an S-shaped tubule is formed from a renal vesicle situated in the angle made by the collecting tubules. Ribbert (1899), was the first to call attention to the fact that a comma-shaped body preceded the formation of the tubule. Huber (1905, p. 17, Fig. 2) presents a drawing of a .section of a human embryo, showing a collecting tubule, on each side of which a comma-shaped body is clearly defined. In the inferior pole of each of these bodies can be seen a renal vesicle, which Huber states differentiates out in situ. He does not specifically refer to the main body or the tail of the comma except to say that the renal vesicle becomes separated from the nephrogenous tissue and becomes bordered by columnar cells. The S-shaped tubule is then formed by constant growth and elongation of the renal vesicle produced by active proliferation and mitotic division of the cells. In the figure referred to above the arrangement of the cells will be found to coincide exactly with the description of j my observations, the cells making up the tail and body of the comma being arranged with their axes perpendicular to the axis of the central core. Huber mentions only the sphere of cells at the lower pole and disregards

the remaintler of the comma. He, as well as all other previous observers, admits that the renal vesicle develops in situ, but though they show in all their figures that the remainder of the comma also undergoes a definite rearrangement of cells and difi'erentiation, still nothing but the spherical end of the nephrogenous tissue has been heretofore considered. In my observations the entire comma-shaped mass was seen to differentiate in situ from the undifferentiated cap of metanephric tissue and to form the S-shaped tubule.

In the formation of the convoluted tubule, from one type of tissue there are differentiated, therefore, cells of very diverse biological activity. . These inherent differences are evidenced in the embiyo by a difference in the growtli rapidity in the several regions of the tubule anlage, and foreshadow the different functions of these respective areas in the adult.

The modus operandi of the union between the secretory and collecting tubules has never been fully described, although Schreiner hypothesizes that it occurs in given regions of the collecting tubule which are recognizable by the presence of mitotic figures. This theory is not surprising, in view of the fact that previous observations have been made on sections. In the study of the living tissue, it was possible actually tO' watch the uniting of the convoluted and collecting tubule, and it was seen that the convoluted tubule definitely grows into the collecting tubule.

Another very interesting fact, brought out in the cultures, was that, no matter to what extent undifferentiated tissue was displaced from the explant, it assumed within certain limits its predestined shape and activity. Nephrogenic tissue developed into secreting tubules and alwaj's became convoluted, while the reverse was true of the collecting tubules, which alwaj's grew in a comparatively straight line, regardless of the absence of pressure from surrounding structures or other physical factors that occur nornuilly in the embryo. This, it would seem, shows that this tissue has been given a stimulus to assume a cei-tain form and activity which will be accomj)lished, unless the tissue reverts, by a state of extreme activity of growth, to a spherical type of cell. This was shown in the development in the exi^lauts and also in the growth comprising the margin.

Bowman's capsule is formed from the end of the secreting tubule, which previous investigators claim invaginates to receive the coincident ingrowth of capillaries to form the glomerulus. While it might be possible to arrive at such a conclusion from an examination of sections, the study of the living tissues, either in spi-eads or tissue-cultures, shows that it is not the correct one. In the first place, the endothelium that differentiates in situ around the convoluted tubule, coincidentally with the growth and differentiation of the latter shows, from the earliest appearance of the bulbous enlargement of the glomerular end, that the wall adjacent to the overhanging first curve of the S-shaped tubule, remains intact throughout the development of the glomerulus. In the second place, in the earliest stages the dilated bulbous end of the glomerulus contains no lumen, but consists of a solid mass of spherical cells, which in appearance resemble epithelial cells. By a rearrangement of these cells there is a separation which results in the formation of a lumen within this cell mass, which leaves the epithelial cells on one side and the mass of epithelial and endothelial cells on the other.

The epithelial cells forming the peripheral wall of the tubule, however, become quite flat and very much elongated, approxinuiting somewhat the endothelial type of cells and thus may have been mistaken for endothelial cells by other observers. According to previous writers the spherical end of the convoluted tubule increases steadily in size until the invaginating cup reaches its maximum, whereas in my observations the spherical tuft reached its maximum size before the completion of the lumen. Immediately upon completion of the lumen, the visceral and parietal layers of the anlage of Bowman's capsule are formed in situ within the tubule fi'om undifferentiated spherical cells. In fixed sections, these layers of Bowman's capsule show up very clearly and it is easily seen how other investigators, considering these walls to be the boundaries of the tubule, would obtain the impression of an invaginating cup. In the fixed sections, however, interrupted stages only can be observed, whereas, in the living, the entire process of lumen development can be followed as a whole.

Concomitant with the formation of the lumen and differentiation of the cells adjacent thereto, there is difl"erentiation of the cells within the spherical tuft, resulting in the development of endothelial cells in the most intimate association with the epithelial cells. These endothelial cells later form channel-like tracts which are covered by a single layer of epithelial cells. Hematopoiesis in the spherical tuft occurred in the cultures before any demonstrable connection with the sinuses coubl be seen, and it must be borne in mind that there was no circulation of blood elements whatever within the explant. At a later stage the connection of endothelial channels with the surrounding sinus was always observed to take place through the small neck of tissue connecting the spherical tuft of cells with the wall of the tubule, a communication with the ingi-owing branches from the aorta being established still later.

The idea that endothelium and blood develop in situ was first suggested as a speculation by Herring. To him it seemed possible that tlie capillaries might develop in situ and not as ingrowths from the dorsal aorta during the formation of the glomerulus. Schreiner and Huber, on the other hand, claimed that the glomerular tuft is formed entirely by an ingrowth of pre-existing capillaries.

The later work of Sabin in blood-vessel formation, in which she states that not only the endothelium, but also the blood of the early vascular system arises in situ, suggests the possibility that, in other regions where primitive tissue exists, the vascular arrangement may also develop in situ. Certainly the observations on the development of the glomerulus in tissue cultures, where the mass of primitive mesenchyme from which this organ develops is entirely isolated from any circulation, show that not only epithelial cells, but also endothelial cells and bloodcells, differentiate in situ from the mass of undifferen tiated spherical cells within the glomerular end of the convoluted tubule which goes to form the glomerulus. The intimate relation of the glomerular capillaries and the epithelial covering is thus explained, for it is quite difficult to figtire out how this epithelial covering of eacli capillary could be acquired by the mere ingrowth of the capillaries into an epithelial sac. In addition to this, the sinuses alsoi develop in- situ, as could be seen not only from a study of spreads and tissue cultures but also in fixed sections. This development of the blood spaces in situ might have been surn^ised by Colberg (1863) from injections. He injected the blood-vessels of the kidney and found that some of the glomeruli were injected while others were not. The latter he called pseudo-glomeriili. These were undoubtedly glomeruli at an early stage of development before connection with the peripheral circulation was established. The same opinion was expressed by Herring in 1!)00. In 1911 Jeidell studied three injected pig embryos and found that a rich capillary network was formed in the mesencliyme surrounding the renal anlage from the inferior mesenteric and middle sacral arteries, the drainage occurring into the post-cardinal vein, in ferior mesenteric veins, and capillaries of the Wolffian body. None of these injections, however, so far as I can make out from her paper and figures, penetrated farther than the primitive capsule. She concludes that the renal anlage is supplied by this means but does not state how the blood enters from the surrounding mesenchyme. In a study of human embryos of different ages, Kelly and Burnam (1911) claimed from injections that the development of the permanent blood supply of the kidney occurs when that organ reaches its definitive position and that, previous to this time, the circulation is maintained by a communication of the capillaries of the kidney with capillaries of the surrounding mesenchyme. Thus, in the ascent of the kidney, the capillaries which are left behind atrophy, while new ones fonn about the superior pole. At the end of seven weeks a connection with the aorta is established. Before this, no injection of tlie kidney could be made. Somewhat the same phenomenon occurred in the chick. Previous to complete development of the secreting tubule, I was unable to reach the sinuses or glomeruli of the metanephros by way of the peripheral circulation by injections of India ink into the heart, whereas the liver and mesonephros became so heavily injected as to appear entirely black. I think it can be concluded, therefore, that in the incompletely developed metanei>hros there is no continuity of or direct connection with the peripheral circulation ; that there is a sinusoidal circulation, which by continued growth and constant sprouting is diminished and transformed into a capillary circulation, has been shown above. The presence of blood elements and a fluid medium inside these sinuses insures a vehicle to carry oxygen or nutritives to the developing renal body. The nourishment of the kidney in its early stages of development is therefore somewhat analogous to the nourishment of a foetus by way of the placenta.

Sabin (1920), from studies of the blood-vascular system of living chick embryos up to four days, concludes that the various white blood-cells have different origins. The lymphocytes and monocytes arise from the endothelium lining the capillaries, while the granulocytes arise from the mesenchyme cells and migrate into the lumen of the blood-vessels. A particular study was not made of the exact origin of each type of blood-cell, but it was seen that not only the capillaries, but also the sinuses, often were filled with Ijniphocytes which had arisen from the endothelial walls, and in many instances, granulocytes were observed scattered among the mesenchyme cells, as though they had differentiated in that region.


1. In the above study of the development and growth of the metanephros of the chick embrj-o it is found that a continuous sheet of endothelial cells entirely surrounds the collecting and convoluted tubules.

2. The convoluted tubules differentiate in situ from a comma-shaped mass of undifferentiated nephrogenic tissue. The S-shaped tubule is developed out of the commashaped mass by different degrees of growth in different areas of the mass, and not by elongation of a renal vesicle.

3. The glomerulus, including its capillaries and blood elements, differentiates in. situ from an undifferentiated cellular mass which completely fills the distal end of the convoluted tubule. There is no cup formation with subsequent ingrowth of capillaries.

•1. The blood system surrounding the metanephric tubules is at first principally sinusoidal, as in the pronephros and mesonephros, being later almost entirely replaced by a capillary system. These sinuses are not at first in direct continuity with the capillaries of the peripheral circulation, but circulation takes place by diffusion between them and the capillaries.

5. The convoluted tubules grow directly Into the intact collecting tubules.

6. In tissue cultures of the metanephros there does not occur a dedifferentiation of tissue but an active growth and differentiation takes place not only in the explant but also, within certain limits, in the marginal outgrowth.


Atterbury, R. R., 1920. Potentialities of the different parts in the l^idney anlage. Anat. Rec, vol. 18-19, p. 219.

Carrel and Burrows, 1910. Cultivation of adult tissues and organs outside the body. Jour. Amer. Med. Assn., vol. 5.

Champy, C, 191314. Quelques resultats de la methode de culture des tissues; generalit^s. (Note pr^liminaire.) Arch, de zool. exper. et gen., t. 3, pp. 42-51.

Idem, 1914. Quelques resultats de la methode des cultures de tissus; le rein. Arch, de zool. exper. 'et gen., t. 4, pp. 307-38G.

Colberg, A., 1863. Zur Anatomic der Niere. Centralbl. med. Wissensch., Bd. 1.

Disse, J., 1902. Harn und Geschlectsorgane. Von Bardelebens Handbuch der Anatomie des Menschen, Bd. 7, Jena.

Felix, W., 1912. The development of the urinogenital organs. Human Embryology, Keibel and Mall, vol. 2.

Herring, P. T., 1900. The development of the Malpighian bodies of the kidney and its relation to pathological changes which occur in them. J. Path, and Bact., vol. (J.

Huber, C. G., 1905. On the development and shape of urlniferous tubules of certain of the higher mammals. Amer. J. Anat., vol. 4.

Jeidell, H., 1911. A note on the source and the character of the early blood vessels of the kidney. Anat. Rec, vol. 5.

Kelly, H. A., and C. Burnani, 1914. Diseases of Kidney Ureters and Bladder: Development of the renal vascularization.

Lillie, F. R., 1919. The Development of the Chick.

Minot, C. S., 1910. A Laboratory Text-book of Embryology.

Minoura, T., 1921. A study of testis and ovary grafts on the hen's egg and their effects on the embryo. J. Exper. Zool., vol. 33.

Ribbert, H., 1899. Ueber die Bntwicklung der bleibenden Niere und uber die Enstehung der Cystenniere. Verhandl. der deutsch. path. Gesellsch., p. 18.

Sabin, P. R., 1920. Studies on the origin of the blood-vessels and of the red blood-corpuscles as seen in the living blastoderm of chicks during the second day of incubation. Contributions to Embryology, vol. 9; Carnegie Inst. Wash., Pub. 272.

Schreiner, K. E., 1902. Ueber die Entwicklung der Amniotenniere. Zeitschr. f. wiss. Zool., Bd. 71.

Stoerk, O., 1904. Beitrag zur Kenntnis des Aufbaus der menschlichen Niere. Anat. Hefte, Bd. 1, 23.


Fig. 1. — 16 mm. lens. Seven days' growth of metanephros of 8-day chick embryo. Extensive growth of the collecting-tubule tree not only within the explant but also out into the marginal growth. The collecting tubules have grown into the mairgin as intact tubules, branching in the characteristic dichotomous manner, the line of growth being a relatively straight one. In this figure the branches are all intralobar, the main stem to the left of the figure being a quarternary branch. The large clear areas between the basic branches are the sinuses; smaller ones are seen toward the periphery. Throughout the culture there are several displaced collections of undifferentiated nephrogenous tissue. The margin of the growth appears in the lower right hand corner of the figure.

Fig. 2. — 16 mm. lens. Seven days' growth of the metamephros of a 10-day chick embryo. Entire excretory unit completely differentiated inside the explant. Union between the convoluted and collecting tubules has occurred in most instances. The main stem of the collecting tubule here is made up of two tertiary tubules. In the upper portion of the figure there are some convoluted tubules which have grown more nearly in a straight line than usual. All of the convolutions show typicad dilation of

the glomerular ends. The glomeruli here are quite small. Numerous macrophages can be seen throughout the culture clearing up the remaining undifferentiated tissue by phagocytosis.

Fig. 3. — 16 mm. lens. Seven days' growth of metanephros of 8-day chick embryo. Marginal outgrowth of collecting tubules on the right. The tubules grew out intact within the thicker region but became flattened out into a sheet-like growth in the more attenuated portion of the margin. At different points in the marginal outgrowth can be seen endothelium composed of cells larger and flatter than the mesenchyme cells, which make up the greater part of the outgrowth. Owing to the heavy staining, the structures inside the explant cannot be made out. Formed sinuses in the explant at the left are well shown about the base of the collecting-tubule outgrowth.

Plate 2.

Fig. 4. — 16 mm. lens. Six days' growth of metanephros of 8-day embryo. Extensive outgrowth of formed collecting tubule into margin, flattening in the attenuated edge of the culture.

Fig. 5. — 16 mm. lens. Seven days' growth of metanephros of 8-day embryo. Extensive outgrowth of renal epithelium from collecting tubule in the explant. There is a lumen in this for a short distance in the marginal growth. The endothelium can be seen a/bout the tubule and is quite characteristic in this mode of outgrowth. This endothelial growth origmates from the endothelial covering about the tubules and presents the same type of outgrowth that occurs from endothelium elsewhere.

Fig. 6. — 4 mm. lens. Four days' growth of metanephros of 9day chick embryo. Iodine vapor. Comparatively straight collecting tubule into which the convoluted tubule has grown. The point of union is shown in the lower left hand corner of the figure, where the convoluted tubule appears to be constricted. The tubules are sharply defined by the characteristic etchedlike endothelium. The lumen formation in the glomerulus has been completed and the capsule of Bowman is well shown. The tuft of epithelial-like cells, among which endothelial are intermingled, shows as yet no evidence of vascularization. This figure demonstrates clearly the ease with which the terminal walls of the convoluted tubule cam be seen and also how these intratubular changes can be distinguished from those taking place outside the tubule. About the tubules many white blood-cells and phagocytes are evident. Vacuolization of cells, exactly as occurred in the formation of the sinuses, is well shown in the lower right quadrant of the figure.

Fig. 7. — Seven days' growth of metanephros of 8-day chick embryo, illustrating the same points as figiire 5.

Plate 3.

Pig. 8. — Five days' growth of metanephros of 9-day chick embryo; 16 mm. lens. Collecting tubules have grown straight out into the marginal growth, carrying with them the undifferentiated nephrogenic tissue which went on to complete differentiation in their final position. They have grown out as intact tubules. The characteristic type of growth for the collecting and convoluted tubules is well shown, the former growing in a straight line and the latter being coiled. At the uppermost point in the figure there is a very small glomerulus in which red bloodcells can be seen. Union of the convoluted and collecting tubules has occurred.

Fig. 9. — Higher power picture of Figure 8; 4 mm. lens, showing extreme upper pole of collecting tubule, convoluted tubule, and glomerulus. The characteristic convolutions of the secreting tubule are well shown, despite the fact that the growth is in the margin where the physical factors attending it were entirely different from those obtaining in the embryo. The glomerulus appears as a small sphere in which red blood-cells are seen as black dots. The defining endothelial cells surrounding the tubules show plainly.

Fig. 10. — 4 mm. lens. Three days' growth of metanephros of 9-day chick embryo. Small explant containing collecting tubule, which presents a very interesting growth. The forward or normal growth is accompanied by the usual dichotomous type of branching, while the posterior growth is in a straight line, with no tendency to branch. On either side of the collecting tubule, in the angles, there are twoS-shaped tubules which differentiated in their characteristic position as regards the collecting tubule. These S-shaped tubules differentiated in the culture. The right side is at a more advanced stage than the left side. Certainly the tendency of the convoluted tubules to be coiled in this culture may be attributed to an inherent tendency of the cells. The collecting tubule, for the same reason, has grown in a straight line.

Fig. 11. — 4 mm. lens. Four days' growth of metanephros of 9-day chick embryo. Immediately at the beginning of the marginal outgrowth there are two convoluted tubules with their glomeruli in the course of development. The one to the right side is not as far advanced as the one to the left. Lumen formation has occurred in the tubule to the left. The tubule to the right is a solid core of cells with a dilated bulbous end in which the formation of a lumen has not as yet occurred. Tbe characteristic pinched-off appearance of the neck of the glomerulus, which results from the relatively slow growth of the cells composing this region, is shown.

Figs. 12 and 13. — Five days' growth of metanephros of 9-day chick embryo. Sinus formation well shown about the convoluted tubules. The white blood-cells show up as dark spots, more pronounced in Figure 12.

Fig. 14. — Same as Figures 12 and 13, except that the sinuses are located more in the denser region about the convoluted tubules.

Plate 4.

Fig. 15. — Metanephros, 8-day embryo. Section 5 micra. x 750. Collecting tubule formation with inverted comma-shaped mass of nephrogenic tissue about end. The cells making up the body and tail of comma are spherical in type with their axes perpendicular to the central core. In the head of the comma there is marked proliferation evidenced by mitotic figures. About the periphery of the comma-shaped mass the endothelial cells can be seen with their axes parallel to the central core. These endothelial cells are just in a stage of transformation from the round spherical type to the long slender endothelial type. The head of the comma Is turned somewhat towards the observer and at the inner angle about the middle of the mass (at X) is an endothelium-lined space which approximates the sphere and, when several sections are studied, proves to communicate with an endothelium-lined sinus, which is somewhat poorly showoi in this figure, just beneath the head of the comma. This is the portion of the sinus surrounding the comma^shaped mass that will later be included in the first curve of the S-shaped tubule. TTils figure shows the comma just prior to the development of the lumen and demonstrates very clearly that the tail and body of the comma, as well as the head, are utilized in the formation of the 2-shaped tubule.

Fig. 16. — Metanephros, 9-day embryo. Section 5 micra. x 750. A "Y"-shaped collecting tubule with two comma-shaped bodies along its inner sides and about its ends. Between the commas there is an endothelium-lined sinus which surrounds the entire anlage. The endothelial walls of the sinus are in opposition along the lateral side of each of the collecting tubules, but the connection with the large sinus at the lower portion of the

figure can be traced even in the photograph. The tubule is here shown in a later stage of development than in Fig. 1. The endothelial covering is evident, as well as the endothelium lining the sinus, the two layers being in apposition. Lumen formation has just begun and is somewhat more advanced on the right side than on the left. The characteristic polarity of the endothelial and epithelial cells, as well as the more rapid proliferation of the spherical head pole of the tubule, is well shown.

Fig. 17. — Metanephros, 9-day embryo. Section 5 micra. x 1500. Cellular tuft in glomerular end of convoluted tubule shown immediately below a loop of the tubule. The lower loop of tubule belongs to another unit. The capsule of Bowman is partially formed about the spherical tuft of cells and the tuft is bounded by a row of radially arranged epithelial-like cells. The flattened epithelial cells making up the parietal layer of Bowman's capsule are evident. The point of junction of the glomerular end with the convoluted tubule occurs just to the right in the figure. A double line of flattened endothelial cells running across the superior surface of the glomerular tuft is evident. In the distal end of the tuft a black spot is seen which is a newly formed red blood-cell. On each side of this cell there are endothelial cells which form a channel in the middle of the epithelial cells making up the major portion of the tuft. This is better shown in Plate 5. A mitotic figure can be seen towards the proximal end of the tuft. The clear areas in the figure, in which red blood-cells appear as black spots, are the sinuses in the immediate neighborhood of the glomerulus.

Plate 5.

Fig. 18. — Collecting tubule covered by cap of metanephrogenlc tissue. The collecting tubule contains granules and is lined with columnar epithelium. Three layers of endothelium can be seen about the cap and one layer surrounding the collecting tubule. The spherical epithelial cells composing the cap are radially arranged about a central core corresponding to the single line shown. Blood cells are shown in the sinus existing between the two outermost endothelial layers.

Fig. 19. — Collecting tubule buds increased in length. The metanephrogenlc cap has become transformed into an inverted comma-shaped mass. The cells making up the tail of the comma show the same arrangement as in Fig. 18. The head of the comma is the seat of most marked proliferation causing a dilatation of the head pole.

Fig. 20. — Collecting tubule only partly shown. Growth of bud extending peripheralward carrying small portion of metanephrogenlc tissue. The comma-shajped mass has approximated its definitive position in regard to the collecting tubule. At points 15 and IG the most rapid proliferation of cells takes place.

Fig. 21. — Transition of comma-shaped mass into that of an ©shaped tubule has occurred, angulation having been brought about by proliferation and more rapid gi-owth of cells at points 17 and 19, point 17 being the first curve of the ©-shaped tubule formed. A sac-like protrusion of the sinus (No. 11 in Fig. 24) has been included between the superior surface of the bulbous glomerular end (19) and the inferior surface of the first curve (17). Blood cells are shown in this sac. Metanephrogenlc cap shown at 14.

Fig. 22. — Collecting tubule and metanephrogenlc cap remain about the same as in Fig. 21. ©-shaped tubule more angulated. Second curve No. 18. First curve increased in size. Glomerular end a dilated bulb of spherical epithelial cells.

Fig. 23. — ©-shaped tubule growing in the collecting tubule at point 13. Lumen formation occurring in second curve with abutting cells becoming cuboidal in shape. Heavy black lines in tuft of epithelial cells composing glomerular end are the endothelial cells. No. 8 in Fig. 24.

Fig. 24. — S-shaped tubule completely joined with collecting tubule. Lumen complete with the formation of Bowman's capsule (5). Note that this formation is entirely inside the endothelial covering layers 1. 2 and 3, 4 constituting the parietal layer of Bowman's capsule and 6 the visceral; 7, spherical epithelial cells composing the tuft surrounded by Bowman's capsule. Intertwining endothelial cells and channels (8) among the epithelial

cells. Blood elements are also shown forming in situ. The connection of the endothelium inside the tuft with the surrounding sinus is shown at 10, and the included endothelial sac or sinus at 11; 9, a connection between the surrounding endothelial sinus and sinuses in other parts of the metanephros. The lumen of the collecting tubule is shown at 12.