Talk:Book - Manual of Human Embryology 17

XVII.

THE DEVELOPMENT OF THE INTESTINAL TRACT AND RESPIRATORY ORGANS.

By OTTO GROSSER, Prague; FREDERIC T. LEWIS, Harvard University; and J. PLAYFAIR Mc^IURRICH, University of Toronto.

INTRODUCTION.

By FREDERIC T. LEWIS.

Formation of the Intestines from the Umbilical Vesicle. — In the youngest human embryos which have yet been obtained, the entoderm forms the lining of a more or less spherical sac, which the early anatomists named the umbilical vesicle (vesicula wnbilicalis). The vesicle enlarges with the growth of the embryo, and during the second month it is a conspicuous object. At birth it is still present.

It was observed at birth by Hoboken, in 1675, as a granule of oval shape, white, about the size of a hemp-seed, with indurated contents. It was probably the umbilical vesicle which Diemerbroeck found in an embryo of the sixth week, and described in 1672 as a sac, the size of a small hazel-nut, filled with clear fluid. But the first satisfactory description of this structure in a human embryo is credited to Albinus, who published an excellent drawing of it in 1754, and referred to it as the vesicula ad umbilicum parvuli embryonis. The vesicle was lodged between the amnion and chorion near the distal end of the umbilical cord, and a slender thread-like prolongation extended from it, through the cord, to the body of the embryo. Beyond this point Albinus did not follow it for fear of damaging his specimen.

At the beginning of the nineteenth century the umbilical vesicle was recognized as a constant structure in young human embryos and its significance was being discussed (Lobstein, 1802). Wrisberg had shown that blood-vessels passed from it into the mesentery of the embryo. Oken believed that the embryo was nourished through the umbilical vesicle, and in 1806 he published a notable treatise, in which he declared that the following propositions would be proved with absolute certainty: (1) The intestines of embryos originally do not lie in the abdominal cavity, but arise from a vesicle, situated outside of the amnion, called the vesicula umbilicalis in man, and the tunica erythroides in other animals.

(2) The intestines do not lie in the vesicle as in a sac. but they are a prolongation of it, as the duodenum is a prolongation of the stomach. The prolongation splits into an anterior and a posterior intestine, both of which pass through the umbilical cord into the abdominal cavity, one part going to the anus, the other to the stomach.

291

292 (3) The stalk of the vesicle, between the splitting of the intestine and the vesicle, becomes obliterated after some weeks, closing and becoming cut off like an umbilical artery; it appears at first as the caecum, and later also as the vermiform process, so that at this place there is no continuity in the intestines but an angular splicing- with a valve.

(4) The intestines now begin to draw back toward the umbilicus and finally enter the abdominal cavity, so that all embryos necessarily have the so-called umbilical hernia.

These propositions were defended by Kieser in 1810, who published an interesting figure of the intestines of a three months' human embryo, a reduced copy of which is shown in Fig. 223. He found that the coils of the intestine were lodged in the umbilical cord and not in the abdominal cavity. He was uncertain whether the intestinal tube continued across the insertion of the stalk of the umbilical vesicle, but, influenced by Oken, he wrote, " It appears as if the ends

Fig. 223. — Kieser's figure, reduced and re-lettered, showing the umbilical cord of a human embryo of three months, the intestines within it, the cord of the umbilical vesicle (a), and the vesicle itself (6); c, lower part of stomach; d, coil of the gastric portion of the intestine which has a blunt end at e; f, anal part of the intestine, also with a blunt end; g, "cord of the umbilical vesicle as it surrounds the two ends of the intestine like a funnel; " h, vena omphalo-mesenterica; i, arteria omphalo-mesenterica.

of the two parts of the intestine were here still divided." The gastric portion is represented as ending in a knob-like expansion, which is in contact with the blunt end of the anal half of the intestine and with the cord of the umbiHcal vesicle, at a place where later the caecum and vermiform process will appear. The cord of the vesicle is a prolongation of the mesentery, in which both the blood-vessels and the tube connecting the vesicle with the intestine have become obliterated.

In Oken's chapter entitled " Proof that all mammals possess an intestinal vesicle and that the intestines arise from it," he says, " I could easily extend this proof over the classes of egg-laying animals." The human umbilical vesicle had already been compared with the yolk-sac of birds, and in 1768 Caspar Friedrich Wolff had published his fundamental studies upon the development of the intestine in the chick. He found that the primitive intestinal cavity is in the dorsal part of the yolk. From this cavity he saw a slender prolongation, which admitted a fine needle, grow forward to make the stomach and oesophagus. This prolongation is

INTESTINAL TRACT AND RESPIRATORY ORGANS. 293 called the fore-gut. Somewhat later in the development of the chick, Wolff saw a similar prolongation grow backward to make the rectum, and this is the hind-gut. Between the two is the mid-gut, open below toward the yolk, and becoming relatively small as the fore-gut and hind-gut lengthen, partly at its expense. Thus Wolff saw in the chick what Oken later conjectured for man, and what has since been actually observed, namely that the intestine arises by the outgrowth of foregut and hind-gut respectively, from the dorsal part of the cavity of the yolk-sac.

Separation of the Intestines from the Yolk-sac. — The term yolk-sac, sacculus vitetlinus, since it is applicable both to lower animals and to man, has largely replaced the term umbilical vesicle. The name mid-gut, although in common use, may well be abandoned. The fore-gut can then be sharply denned as that portion of the intestine anterior to the attachment of the yolk-sac, and the hind-gut as the part which is posterior. The attachment of the yolk-sac, broad at first, becomes reduced to a slender stalk, the base of which may remain as a diverticulum of the intestine.

Oken wrongly supposed that a portion of the yolk-stalk persists as the vermiform process. This error was corrected by Meckel (1812) in a most thorough manner. An out-pocketing of the human small intestine, usually about an inch in length but sometimes several times as long, had frequently been observed. It was generally found opposite the mesenteric attachment, about three feet from the beginning of the large intestine. Sometimes it was turned toward the mesentery. Its walls included all of the layers which enter into the formation of the intestinal tube, with which its lumen was in free communication. Meckel regarded the opinion of Fabricius, that such diverticula arise from the pressure of substances within the intestinal canal, as improbable. He saw the diverticulum several times in children at birth, once in an embryo of six months and twice at three months. Since it is a congenital structure, essentially constant in position, Meckel sought to explain it through the normal development of the intestinal tract, and concluded as follows : " Even into the third month of embryonic life a small elevation remains in the lower part of the small intestine as a trace of the former connection (with the yolk-sac), and if this is retained beyond this time it appears as a blind appendage." Meckel found one abnormal case in which it remained as an open duct extending from the intestine to the umbilicus, accompanied by its vessels which were still pervious. He saw cases also in which the obliterated vessels formed cords extending from the diverticulum of the intestine across the abdominal cavity to the umbilicus. Such cords have frequently been observed, and, as Meckel recorded, they may leadt-to adhesions and intestinal obstruction. The diverticulum was found not only in man but in other mammals. Cuvier had seen it in birds, and Meckel concluded that it was a constant structure in ducks and geese, in which, moreover, its genesis from the yolk-sac could be clearly demonstrated. Thus the true embryonic interpretation of the diverticulum ilei was clearly established by Meckel. He found also that " the vermiform process appears first as a little knob which gradually enlarges considerably." " I saw it arise thus in the human embryo, as Wolff had seen the caeca in the chick, where previously no trace of them could be identified.' - ' This conclusion is in accord with later observations.

The Allantois. — Several human embryos have been obtained which are so young that neither the fore-gut nor the hind-gut has begun to grow out from the cavity of the yolk-sac. In most of these, however, the tubular entodermal outgrowth known as the

294 HUMAN EMBRYOLOGY.

allantois is present. The allantois grows out from the posterior portion of the yolk-sac near its dorsal surface. When the hind-gut pushes out, the allantois is carried with it, so that then it empties into the terminal part of the hind-gut, which is called the cloaca.

In the horse, cow, and pig the distal portion of the allantois dilates enormously, forming a somewhat cylindrical vesicle, so attached to its stalk that the entire allantois is T-shaped. At certain stages the terminal vesicle is many times the size of the embryo. Thus, with an embryo goat of eighteen days Haller found an allantoic sac two feet long, whereas the embryo itself measured less than two inches (twenty lines). The allantoic sac is found between the amnion and the chorion, to which it may be adherent.

The part of the allantois near the intestine, which develops through subdivision of the cloaca, is commonly called the allantoic stalk. Its formation will be fully considered in the chapter on the urogenital tract. 1 A part of the allantoic stalk expands to form the bladder. Between the apex of the bladder and the allantoic sac, the allantois remains slender and is known as the urachus. The urachus extends from the bladder into the umbilical cord.

The allantois has been known for centuries. According to Fabricius ab Aquapendente (1600'), " The membrane is called aXkavroeuS^g because it is similar to aXkag, that is, sausage. But it must not be understood that it resembles anything filled with chopped meat, with which sausage skins are usually filled (for it contains only urine), but because its form seems similar to a sort of intestine from which sausages are generally made; hence it is called allantoic, that is, intestinal, by Galen and early writers. According to Suidas, a/ldq seems generally to be used in the sense of svrepov, although this too cannot be denied, that in such membranes, along with urine, particles like chopped meat or sausage are sometimes found." The solid particles referred to may be the hippomanes found in the allantois of the horse, and said to be known to Aristotle (Bonnet, 1907, p. 194).

Many attempts were made to find an allantois in human embryos, but at the beginning of the nineteenth century no agreement had been reached. Hale in 1701 had announced " The Human Allantois discover'd," but. according to Oken and Velpeau, it was probably an amnion which he described. Lobstein declared that the human yolk-sac was the allantois. " Who will not be entirely in doubt," Oken wrote in 1806, " when one finds that writers have described as allantois the most heterogeneous things which were ever seen ? " It soon became established, however, that in human embryos a slender urachus extends from the bladder into the umbilical cord, accompanied by the umbilical arteries. Velpeau, in 1834, thought that after the urachus had passed the whole length of the cord it became lost in a porous tissue between the amnion and the chorion, and that this tissue represented the allantoic sac. Von Baer, in 1837, declared that what was found between the amnion and the chorion had been somewhat rashly interpreted as the allantois. " The true allantois it certainly is not." Allantoic vesicles continued to be reported for many years, until finally the fundamental relations of the human allantois were established by His. He wrote as follows (1885, p. 222) : " I designate as body-stalk [pedunculus abdominalis] that thick cord which in very young embryos forms a connection between the 1 See also page 322.

INTESTINAL TRACT AND RESPIRATORY ORGANS. 295 embryo and the chorion. . . . The main portion of the body-stalk is loose connective tissue with a few smooth muscle-cells; its dorsal surface has an ectodermal covering, and the ventral half surrounds the allantoic duct and the two umbilical arteries running with it. ... A vesicular or even only a free allantois has never been found in human embryos, and the slender duet in the body-stalk, the allantoic duct as I have formerly named it, is indeed only a very rudimentary representative of the structure which is so large in many mammals." Although the human allantois may be described as rudimentary because of its small size, it is nevertheless differentiated very early. Keibel and Elze (1908, p. 152) have recorded that it appears in man and the apes before any segments have formed.

It arises somewhat later in Tarsius, but still before there are any segments. It first appears in pigs of four or five pairs of segments, in rabbits of about eleven pairs, and in chicks of more than twenty pairs.

In the first human embryo which is now to be described, the body-stalk will be seen connecting the yolk-sac with the chorion. Into this stalk in the second specimen the allantois has grown out, thus forming the first subdivision of the entodermal tract.

THE EARLY DEVELOPMENT OF THE ENTODERMAL TRACT AND THE FORMATION OF ITS SUBDIVISIONS.

By FREDERIC T. LEWIS.

Peters 's Embryo. Yolk-sac. — It has been inferred from comparative studies that the human entodermal tract arises as a solid mass of cells, and diagrams of this hypothetical stage have been published by Keibel, Schlater, and others. In the youngest embryos which have been observed, however, the entodermal cells surround a cavity. This is the condition in an embryo obtained by Peters (1899) at the autopsy of a suicide who had taken caustic potash one month after her last catamenia. Bryce and Teacher (1908) estimate the age of this embryo as 13 % to 14% days. It is generally conceded to be the youngest properly preserved human embryo yet described. The embryo extended through nineteen 10 /a sections, and was cut "obliquely to the longitudinal axis." A .drawing of only one section was published, and this is reproduced in Fig. 224. The cavity of the yolk-sac contains round masses of coagulum. It is bounded by a layer of entodermal cells which are not everywhere distinct. The entoderm in this section appears to be completely surrounded by mesoderm, which forms the outer layer of the wall of the yolk-sac. A strand of mesoderm

296 extends from the yolk-sac to the chorion, bounding a space designated Sp in the figure. 2 Dorsal to the yolk-sac is the amniotic cavity, bounded above by a thin layer of amniotic ectoderm and below by the very thick embryonic shield, also ectoderm. According to von Spee, who studied Peters 's specimen (Peters, 1899), "It is impossible to speak of an isolated body-stalk which connects the embryo with the chorion, because almost the whole embryonic formation seems imbedded in a thickening of the chorionic mesoderm. Whether the first small beginning of an entodermal diverticulum (allantoic duct) has already started to grow out from the caudal end, and

Mes. eh.

Cav. am.

Mes.

Fig. 224. — Obliquely longitudinal section of Peters's embryo. (After Peters.) Cav. am. (cavum amnii), amniotic cavity; S. v. (sacculus vitellinus), yolk-sac; Ent., entoderm, and Mes., mesoderm of the yolk-sac; Mes. eh., mesoderm of the chorion; Sp., "cleft, in the exoccelom"(?).

appears in the form of a ring of epithelioid cells arranged about a lumen (in section 11, etc.), remains to me entirely uncertain." Keibel has modelled Peters's embryo from outline drawings made upon wax plates by Selenka. He failed to find an allantois, but records that the outer surface of the yolk-sac is uneven as if blood and vessels had begun to develop in its mesoderm (Keibel and Elze, 1908). It will be noted that von Spee does not state definitely that Peters's specimen has no allantois. In describing another very young embryo he had recorded that "as compared with the embryonic shield, the allantois is remarkably long, and ought therefore to appear very early" (1896, p. 9).

2 Grosser, who has examined the specimen, thinks that this space may be the beginning of the cavity of the chorion, and that elsewhere the chorion is filled with loose tissue. Zentralblatt fiir Physiol., Bd. 22, Nr. 1.

INTESTINAL TRACT AND RESPIRATORY ORGANS. 297 Two other embryos, both removed from the uterus by curetting within a month after the last catamenia, may have no allantois. In one of these "an allantoic duct of the yolk-sac does not stand out clearly" (Beneke, 1904), and in the other "the body-stalk consists only of mesodermal cells" (Jung, 1908). Neither of these accounts, however, is convincing in regard to the absence of the allantois, for in the first case the statement is indefinite, and in the second the allantois is not specifically mentioned. .

Herzog's Embryo. Yolk-sac and Allantois. — Herzog (1909) has recently described an embryo obtained at the autopsy of a woman who had been struck over the heart by the shaft of a swiftly

Mes.ch.

Mcs.

Fig. 225. — Longitudinal section of Herzog's embryo. X 215 diam. (After Herzog.'! Mes.ch., mesoderm of the chorion; P. a. (pedunculus abdominalis), body-stalk; All., allantois ; Cam., amniotic cavity; Ect., ectoderm of the embryonic shield; M. pr., membrana prima, to which some cells are adherent; S. v., yolk-sac; Mes., mesoderm.

moving carriage and almost instantly killed. The specimen is unquestionably normal, and is well preserved histologically, but it has suffered considerable mechanical injury, partly after being mounted. After Herzog had published and described accurate figures of twenty-two successive sections (7/* thick), including all of the embryo except a portion of the yolk-sac, he deposited the specimen in the Harvard Collection. For the privilege of studying further and modelling this embryo, the writer is under great obligation to Dr. Herzog.

The plane of section is nearly longitudinal. In the section shown in Fig. 225 the allantois is found extending through the

298 HUMAN EMBRYOLOGY.

body-stalk toward the chorion. The entoderm of the yolk-sac is seen below, extending toward the allantois, but the connection between the two has been destroyed. It presumably occurred in this section. The ectoderm bounding the amniotic cavity consists of a thin layer above, for the most part broken away in this section, and the thick ectoderm of the embryonic shield below, which is broken into two pieces. The shield is bent upon itself, and the depression shown in the figure is transverse to the axis of the embryo. Between the ectoderm of the shield and the entoderm of the yolk-sac there is a structureless membrane such as Hensen (1875) observed in the rabbit and called the membrana prima. It is the detached basement membrane of the ectoderm of the embryonic shield, and was noted by von Spee in Peters 's specimen. A layer of mesoderm passes over the ventral surface of the yolk-sac, and

Ect.

Fig. 226. — Longitudinal section of Herzog's embryo, separated by twelve "/"• sections from Fig. 3. X215 diam. (After Herzog.) Ect., a mass of ectoderm bounding the amniotic cavity on the left side of the embryo. Ent., entoderm, and Mes., mesoderm of the yolk-sac S. v.

anteriorly (to the left of the figure) it is in relation with the ectoderm of the shield. It does not extend between the ectoderm of the shield and the yolk-sac, and in this respect Herzog 's embryo differs from Peters 's specimen as described by von Spee. In Fig. 225 the yolk-sac is cut tangentially, but it is evident that toward the allantois its cells are cuboidal. The allantois consists of similar cells and contains a lumen. Over the greater part of the yolk-sac, however, the entoderm forms a very thin layer resembling endothelium, precisely as recorded by Beneke. In the most ventral portion there are occasional cuboidal cells with large round nuclei and protoplasm which projects above the general level into the cavity of the yolk-sac. The mesoderm of the yolk-sac is also very thin, and shows neither vessels nor distinct blood islands (see Fig. 226).

INTESTINAL TRACT AND RESPIRATORY ORGANS. 299 A reconstruction of Herzog's embryo, cut through at right angles with the plane of section, is shown in Fig. 227. The model will be more readily understood in comparison with a similar view of a somewhat older embryo, Fig. 228. In both cases the amniotic cavity sends a prolongation, torn in Herzog's specimen, toward

Fig. 227. — Wax reconstruction of Herzog's embryo. The plane of section is transverse to the axis of the embryo. X100 cliam. AIL, allantois; Cam., amniotic cavity; Ch., chorion; M.pr., membrana prima; P. a., body-stalk; S. v., yolk-sac; V.ch.\ (villi choriales), chorionic villi.

Fig. 228. — Wax reconstruction of Minot's embryo, showing the part corresponding with that drawn in Fig. 227. X50 diam. L.pr. (linea primitiva), primitive streak; Mes., mesoderm. Other abbreviations as in Fig. 227.

the chorion; and in both the amniotic cavity is asymmetrical, extending farther toward the left of the embryo (which is on the right of the figure). In Herzog's specimen the prolongation of the amniotic cavity toward the left is at first tubular; it is then reduced to a solid clump of cells, lodged between the entoderm

300 HUMAN EMBRYOLOGY.

and mesoderm of the yolk-sac, and closely applied to the latter (see Fig. 226).

The body-stalk contains the allantois, which is apparently disintegrated near its tip. The terminal sections, however, are well preserved, and the extremity of the allantois is probably recurved. The body-stalk contains also rings of cells, some of them very near the surface. It is not certain that these represent blood-vessels — which would be the first to appear in the embryo — as Herzog interpreted them, and yet it is clear from later stages that the blood-vessels in the body-stalk arise very early. Jung has carefully described similar rings in the body-stalk of his specimen. There are occasional clefts in the mesoderm of the chorion in Herzog 's embryo, but they are of doubtful significance. As seen in the reconstruction, a strand of mesoderm extends between the yolk-sac and chorion much as in Peters 's specimen.

Von Spee's Embryo "v.H." — In 1896 von Spee published a notable contribution to human embryology, to which reference has already been made. In it he describes an embryo, designated "v. H.," which has heretofore been placed next to Peters 's specimen, but which seems older than Herzog 's for the following reasons. The axis of the embryo is said to be represented by a primitive groove (not well defined, however) which is absent in Herzog 's embryo; the mesoderm extends between the ectoderm and the yolk-sac, reaching the median line; the entoderm of the yolk-sac is not flat, but is cuboidal throughout; the mesoderm of the ventral portion of the yolk-sac is thrown into elevations by the blood islands within it. The age of "v. H.," obtained through abortion following influenza five weeks after the end of the last catamenia, has been estimated as 17 to 18 days.

Minot's Embryo. The Primitive Knot. — Hensen (1875), in describing the primitive streak of the rabbit, stated that anteriorly it developed a disk-shaped termination, which he designated as a "knot." He found that the layer of entoderm could be stripped from the embryo except at the knot, where it tore. There the ectoderm and entoderm are intimately blended. This primitive knot (often called Hensen 's knot) is shown in the reconstruction of a human embryo in Fig. 229. It is possible that it is represented in one section of Herzog 's embryo (Fig. 11 of his publication), but von Spee did not find it in "v. H." Toward the allantois from the primitive knot, as seen in Fig. 229, the primitive groove is found, along which the mesoderm fuses with the ectoderm, as shown in the cross section, Fig. 228. Anterior to the primitive knot the thick ectoderm forms the medullary plate, but the medullary groove has not yet appeared. Beneath the medullary plate the mesoderm extends from side to side across the

INTESTINAL TRACT AND RESPIRATORY ORGANS. 301 median line. These conditions are noted as determining the stage of development of this specimen. 3 The entodermal tract in this embryo still consists of only two parts, the yolk-sac and allantois. As compared with Herzog's

L.med.

Fig. 229. — Wax reconstruction of Minot's embryo, showing a median sagittal section. X50 diam All., allantois; Cam., amniotic cavity; Ch., chorion; L.med. (lamina medullaris), medullary plate; N.pr. (nodulus primitivus), primitive knot; P.a., body-stalk; S.pr. (sulcus primitivus), primitive groove; yolk-sac; V.ch.. chorionic villi.

specimen, the allantois has increased in length from about 0.12 3 The specimen, in the form of an intact chorionic vesicle already hardened and in alcohol, was placed at the writer's disposal by Professor Minot, and it may be referred to as Minot's embryo. A full account of it is in preparation, justified by its superb preservation.

302 to 0.20 mm. It is slightly expanded distally but is not recurved. There is a minute lumen, and the structure connects with the yolksac by a small funnel-shaped enlargement. The yolk-sac has grown more rapidly than the allantois, its transverse diameter having increased approximately from 0.25 to 0.75 mm. In the ventral portion of the yolk-sac the entoderm now consists of cuboidal cells, and the mesoderm has been curiously transformed, as shown in Fig. 230. The cells are large and extensively vacuolated, so that the protoplasm in places is reduced to strands. The nuclei are large, round, and pale, each containing a very delicate chromatic reticulum and often a single conspicuous knot of chro

Ent.

Mes.

End.

c - • ... £ T\ v V

i ^

/

V

viffav a :

Coag.

Fig. 230. — A portion of the ventral wall of the yolk-sac of Minot's embryo (Harvard Collection, Series 825, Section 18). X280 diam. Co., coagulum in the yolk-sac; Coag., coagulum in the chorionic cavity; End., endothelium lining a blood-vessel containing five blood-corpuscles, one of which shows a vesicular nucleus; Ent., entoderm, and Mes., mesoderm of the yolk-sac.

matin. Among these mesodermal cells blood-vessels have appeared, lined by true endothelium. They contain blood-corpuscles, characterized by finely reticular protoplasm, ill-defined cell-membranes, and nuclei which may be round, with distinct chromatin granules, or irregularly shrunken and deeply stained. Sometimes a corpuscle is closely applied to the endothelium as if arising from it.

In a very few places the entoderm of the ventral surface of the yolk-sac sends a prolongation into the mesoderm. In one case the outgrowth is solid, in another it contains a cavity in its outer part, and in a third a detached entodermal cyst is found near the surface of the mesoderm. These appear to be chance irregularities in the expansion of the yolk-sac.

There is granular coagulum within the yolk-sac, and also in the chorionic cavity, but there are no globular formations as in

INTESTINAL TRACT AND RESPIRATORY ORGANS. 303 Peters 's specimen. Eternod examined the yolk-sac of a young human embryo (the age is not stated) removed by operation and still living. He states (1906, p. 256), "The very transparent liquid, which fully distended the yolk-sac, had a beautiful goldenyellow color comparable with that of the yolk in the eggs of salmon or trout. Under the influence of light, in a few moments, the color clouded and faded, becoming opalescent." In sections of very young human embryos he found entodermal cells projecting into the yolk-sac or detached and floating within it. In the Minot specimen it is very difficult to find a floating cell, but in embryos less well preserved they occur frequently. To what extent the contents of the human yolk-sac has a nutritive function is wholly undetermined.

In the Minot embryo the lining of the yolk-sac is a simple layer throughout (it is obliquely cut in Fig. 230). In the dorsal half of the sac the entodermal cells are quite flat. The mesoderm also becomes a thin layer, and the blood-vessels are very small. Apparently those in the yolk-sac do not pass into the body-stalk, which, however, contains numerous vessels. There are also many spaces in the chorion, especially near its lower surface, which are probably true vessels. Frequently these contain strands of darkly staining cells, suggesting collapsed endothelium. Similarly in the slightly older Frassi embryo there are "vessels on the yolksac, in the body-stalk, and the adjacent chorion; no vessels in the embryo proper." From the study of the specimens thus far considered it appears that the yolk-sac is not the only source of bloodvessels, but that they arise also in the body-stalk and chorion. Recently Dandy has reached this conclusion from the stud}" of an older embryo.

Von Spee's "Gle." Neurenteric Canal. Chordal Plate. Beginning of the Fore-gut. — At a slightly later stage than that just described, a canal develops through the primitive knot, by which the cavity of the amnion communicates with that of the yolksac. This neurenteric canal has not formed in Minot 's embryo, and in the Frassi embryo an aperture could not be demonstrated. Beneke recorded a neurenteric canal in his younger specimen, but his account is not convincing.

Von Spee, however, found a very large canal in an embryo designated "Gle" (1889). The specimen was obtained by spontaneous abortion five weeks after the end of the last catamenia, and Bryce and Teacher estimate its age as 19-20 days. A diagrammatic median section of the embryo is shown in Fig. 231, A. Before the embryo was sectioned it was made transparent with turpentine. In the position of the primitive knot (that is, between the anterior end of the primitive groove and the posterior end of the medullary groove) a ring-shaped elevation was seen, 0.13 mm.

304 in diameter, pierced by a central aperture 0.02 mm. wide. In the series this opening was found in four sections, one of which is shown in Fig. 231, B. It will be seen that at the neurenteric canal the ectoderm is continuous with the entoderm. The mesoderm does not form any part of its wall.

Eternod (1899) has recorded two other cases of open neurenteric canals, one in an embryo very much like "Gle," measuring 1.3 mm., the other in an older specimen, measuring 2.11 mm.

The entoderm lining the yolk-sac in "Gle" resembles that in the Minot embryo, except that just beneath the medullary groove

Cam.

Fig. 231. — Sections of von Spee's embryo "Gle." A, median sagittal section from a model; B, transverse section through the neurenteric canal, C. n.; C, portion of a transverse section showing the chordal plate, L.ch. (After von Spee.) F.g., "fore-gut; " S. med. (sulcus medullaris), medullary groove. Other abbreviations as in preceding figures.

it exhibits a plate of low columnar cells (Fig. 231, C). This chordal plate gives rise to the notochord and perhaps to a portion of the intestinal epithelium. It begins at the anterior margin of the primitive knot, with which it is continuous. It extends forward as far as the yolk-sac is in contact with the medullary plate. In the anterior part of the embryo there is a slight forward prolongation of the yolk-sac, which is the beginning of the fore-gut (Fig. 231, A). The heart is developing in the fold of mesoderm just beneath it. At the opposite end of the embryo the allantois takes a somewhat zigzag course in the body-stalk. Certain portions of

INTESTINAL TRACT AND RESPIRATORY ORGANS. 305 the mesoderm of the body-stalk, as stated by von Spee, are very rich in spaces, some of which have a smooth lining of flat cells

Fig. 232. — Wax reconstruction of Mall's Series 391, showing a median sagittal section of the embryo. X50 diam. All., allantois; A.um., arteria umbilioalis; C. am., amniotic cavity; Ch., chorion; Cor, heart; F.g., " fore-gut; " H .g., "hind-gut; " L.ch., chordal plate; L.pr., primitive streak; N. pr., primitive knot; S. med. (sulcus medullaris), medullary groove; St., stomodeeum; S. v., yolk-sac; V. urn., vena umbilicalis.

like those of embryonic endothelium. The hind-gut has not yet appeared.

Mall's Series 391. Formation of the Hind-gut. — In the Mall collection there is an embryo with seven pairs of somites, measuring about 2 mm. in length, which has recently been described by Dandy. It was obtained through abortion, mechanically induced, and its age is estimated at about 24 days. Fig. 232 is from a model of the embryo, and shows a median longitudinal section. 4 The back bends sharply downward toward the cavity of the yolksac. Such flexures occur frequently, but not invariably, in embryos of about this age, and are probably due to imperfect preservation. "The fore-gut is present in thirty-two sections representing a length of 320 microns." It ends blindly in front, and, according to Dandy, it is separated by mesoderm from the ectodermal depression (or stomodaeum) which gives rise to the mouth. The presence of intervening mesoderm is, however, difficult to make out, since the tissues are somewhat fragmented in this region. The fore S. med.

Fig. 233. — Sections showing the separation of the notochord from the digestive tract. A, from Mall's Series 391, X360 diam.; B, section in the region of the first pair of somites, and C, in the caudal region, of Low's embryo. (B and C after Low.) Ch. (chorda), notochord; Ent., entoderm of the digestive tract; L. ch., chordal plate; S. med., medullary groove.

gut shows a lateral expansion, which is the first pharyngeal pouch. Beneath the fore-gut the heart is well developed.

"The hind-gut is a blind pouch 120 microns in length by sections, but on account of the dorsal kink of the embryo the actual length is somewhat greater." It does not come in contact with the ectoderm so as to form a cloacal membrane (where the anus will later appear), but the bend in the embryo may possibly have caused the separation of the layers. In the decidedly younger Frassi specimen the beginning of the cloacal membrane is said to be present. In Mall's embryo the allantois arises from the ventral surface of the hind-gut and passes into the body-stalk, accompanied by very large umbilical vessels. The allantois has a knob-like branch. The distal end of the allantois is apparently detached. In both portions a narrow lumen is found.

The chordal plate (Fig. 233, A), extending along the middorsal line of the yolk-sac, is more sharply defined than in "Gle." 4 Through the kindness of Professor Mall, the writer has been permitted to study this embryo and prepare a figure to correspond with that of the Minot specimen. This work has been greatly facilitated by Mr. Dandy's publication.

INTESTINAL TRACT AND RESPIRATORY ORGANS.

307

In an embryo of 13-14 somites, described by Low (1908), the chordal plate is still a portion of the wall of the yolk-sac anteriorly (Fig. 233, B), bnt it lias completely separated from it posteriorly (Fig. 233, C. See also Kollmann, 1890). In older embryos it is detached throughout, and the further history of the notochord, or chorda dorsalis, will be found in the chapter on the development of the skeletal system.

The Cloacal Membrane, Caudal Intestine, and the Later History of the Primitive Knot. — In the Mall specimen the chordal

vied.

"N. pr ~ ? ^Sy C - N. pr.

L.pr.

D Fig. 234. — Median sagittal sections through the region of the primitive knot. All X75 diam. A, Eternod's embryo with 8 pairs of somites (after Eternod); B, sketch of the "Kroemer-Pfannenstiel" embryo, based upon sections published by Keibel and Elze; C, Bremer's 4 mm. embryo; D, distal portion of the tail of a 7.5 mm. specimen. All., allantois; Can. med. (canalis medullaris), medullary tube; C. ch., chordal canal; C. n., neurenteric canal; Ch., chorda; Int. can., intestinum caudale; L. ch., chordal plate; L. pr., primitive streak; Mem. cl., membrana cloacalis; Neu. p., posterior neuropore, the last part of the medullary groove to close; N. pr., primitive knot; S. med., medullary groove; x, extension of the primitive streak (?) beyond the cloacal membrane.

plate ends posteriorly in a rounded knot of tissue in connection with the ectoderm. This is clearly the primitive knot, posterior to which is the primitive streak. There is no neurenteric canal. This region in a similar embryo (2.1 mm. long, with eight pairs of somites) has been figured by Eternod (1906 2 ), as seen in Fig. 234, A. The neurenteric canal is still present and leads anteriorly into a chordal canal, the floor of which is formed by detached cells, and the roof of which is the chordal plate. Such a chordal canal

308 HUMAN EMBRYOLOGY.

in human embryos has apparently not been found by other observers. Posterior to the neurenteric canal the primitive streak extends to the cloacal membrane, which is "composed of a mass of epithelial cells." In an embryo of 5 to 6 pairs of somites, of which Keibel and Elze have published a series of sections, it is possible that the relations are as shown in Fig. 234, B. The dorsal and ventral openings of the neurenteric canal can be found, but the middle part is not pervious. The primitive streak extends to the cloacal membrane, which is described as "the thickened ectoderm applied to the thickened entoderm." Thus the primitive streak passes around from the dorsal to the ventral side of the embryo.

Keibel and Elze believe that the primitive streak extends beyond the cloacal membrane along the body-stalk, for in several sections the ectoderm covering the body-stalk shows a local thickening which is nearly in contact with the allantoic duct. 5 The occasional occurrence of a bladder opening freely along the ventral body wall (exstrophia vesicae) may be connected with this relation. It seems probable, however, that the thickened epithelium along the body-stalk is due to a prolongation of the cloacal membrane in the urogenital area, and that it is not a part of the true primitive streak. The primitive streak is formed by a fusion of ectoderm and mesoderm, but the cloacal membrane is a fusion of ectoderm and entoderm. According to Keibel, however (1896), the cloacal membrane should be regarded as a modified part of the primitive streak, and exstrophia vesicae represents a persistent portion of the blastopore.

In an older embryo, measuring 4 mm. (Fig. 234, C), the position of the primitive knot can still be located. There the notochord ends and the primitive streak begins. It will be observed that the hind-gut has extended beyond the cloacal membrane into the tail. This prolongation is named the caudal (or postanal) intestine.

The tip of the tail of a 7.5 mm. embryo is shown in Fig. 234, D. It is still possible to recognize the primitive knot, which at this stage is commonly called the "tail-bud." The notochord terminates in this bud; the caudal intestine fuses with it ventrally, the extremity of the medullary tube dorsally, and the mesoderm laterally. In the 2.11 mm. specimen described by Eternod (Fig. 234, A ) there is a short prolongation of the chordal canal beyond the neurenteric canal, but there is apparently no other evidence that the notochord ever extends beyond the primitive knot. In a 2.1 mm. specimen described by Mall the obliterated neurenteric 6 It may be noted that His described the medullary groove as extending along the body-stalk. Anatomie menschlicher Embryonen, iii, p. 224.

INTESTINAL TRACT AND RESPIRATORY ORGANS. 309 canal, represented by a solid cord of cells, is said to communicate with the medullary tube (1897, p. 419). "The location is opposite the twelfth muscle plate, or in the neighborhood of what will later be the position of the first rib." But the location is also at the posterior end of the notochord which will later be near the tip of the tail. It seems probable that if a neurenteric canal should persist it would be found opening externally beyond the limit of the spinal cord and its filum terminate, in the coccygeal region. 6 Marwedel (1901) has described a ease which may be interpreted as a neurenteric canal leading into the detached end of the caudal intestine. A child thirteen days old was found to have a sac, 6 cm. long, lined with mucous membrane similar to that of the large intestine and surrounded by muscle coats, opening to the surface between what were " evidently the cornua sacralia of the lowest sacral vertebra." The sac had no connection with the rectum or anus.

It should be noted that congenital cysts and sinuses in the coccygeal region are frequent (Mallory, 1892), but they are ectodermal structures, and the persistence of a neurenteric canal has not yet been satisfactorily demonstrated.

The Pharyngeal Membrane and the Prce-oral Intestine. — In describing Mall's specimen with seven pairs of somites, it was

— Ch. '—Ph.

Mem. ph.

Fig. 235. — The pharyngeal membrane as figured by His. X37 diam. A, embryo "Lg," 2.15 mm.; B, embryo "BB," 3.2 mm. Ch., chorda; C.med., medullary tube; Mem. ph., membrana pharyngea; Ph., pharynx ("fore-gut"); R-, Rathke's pocket (anterior lobe of the hypophysis); S., Seessel's pocket; St., stomodseum.

stated that the anterior end of the fore-gut comes in contact with an ectodermal pouch called the stomodseum or buccal sinus. There the ectoderm unites with the entoderm to form the pharyngeal (or buccopharyngeal) membrane. This membrane is clearly shown in an embryo 2.15 mm. long, as figured by His (Fig. 235, A). The digestive tract at this stage has no anterior opening. Just in front of the pharyngeal membrane the ectoderm forms a pocket extending toward the base of the brain. Although this pocket is now generally called the hypophysis, or more precisely the anterior lobe of the hypophysis, it is often referred to embryologically as Rathke's pocket.

In 1838 Rathke described it as follows : " For a long time I have observed in several animals ... a small irregularly rounded depression which belongs

Professor Mall now regards this embryo of 2.1 mm. as pathological.

310 HUMAN EMBRYOLOGY.

to the mucous membrane of the mouth, of which it is clearly a thin-walled outpocketing. . . . Finally I saw that this depression represents the first step in the formation of the pituitary gland" (p. 482). The animals studied included sheep and pig embryos.

On the entodermal side of the pharyngeal membrane a much smaller pocket bulges toward the brain.

This was discovered by Seessel in the chick (1877). He wrote: "Shortly after the hypophyseal pocket has become distinctly formed, on about the fourth day, near and under it a second pocket-like outgrowth of the intestinal layer is seen. ... Its length compared with that of the hypophysis is as 1 : 5." Although His considered that both pockets were represented in the 2.15 mm. embryo, they are better denned in a specimen measuring 3.2 mm. (Fig. 235, B). The pharyngeal membrane has largely disappeared. "As the remains of it, there is only the prominence inserted between Ratlike 's pocket and Seessel 's accessory pocket." In sheep embryos von Kupffer (1894) found a solid entodermal outgrowth extending forward from Seessel's pocket, closely connected with the notochord. Later this mass of cells becomes detached and appears as an appendage of the notochord. It was interpreted as a rudimentary prae-oral intestine. Bonnet (1901) identified a similar structure, but with a lumen, in a dog embryo with sixteen pairs of segments, and Zimmermann (1899) found three sharply defined little cavities near Rathke's pocket in a human embryo of 3.5 mm. These cavities may have been derived from a prae-oral intestine; they are at present the only evidence of such a structure in human embryos.

Thompson's Embryo. Early Stages of the Thyreoid Gland, Lungs, and Liver. — An embryo 2.5 mm. long, with 23 pairs of somites, has been modelled by Thompson (1907). The entodermal tract is shown in Fig. 236, in which the embryo is arbitrarily placed in an upright position, with its ventral surface toward the left of the figure. The yolk-sac has been cut away. It was connected with the intestine by a somewhat constricted neck, called the vitelline duct, a part of which is shown in the figure. In ventral view the connection between the duct and the intestine would appear as an elongated opening near the middle of the straight intestinal tube.

The pharynx is separated from the mouth by the pharyngeal membrane, which is already perforated. There is no trace of Rathke's pocket. Four pharyngeal pouches are present, but they are not indicated in the figure. In the median line, connected with the floor of the pharynx, there is a small, hollow, rounded diverticulum, which is the beginning of the thyreoid gland. In earlier stages it has a less constricted neck, as found by Low (1908) in a specimen with 13 or 14 pairs of somites. Posterior to the pharyngeal pouches the entodermal tube suddenly narrows. It becomes compressed laterally so that it has a cleft-like lumen. In this

INTESTINAL TRACT AND RESPIRATORY ORGANS.

311

portion of the entodermal tract, a short distance beyond the fourth pharyngeal pouches, the lungs are indicated by a pair of lateral outgrowths (Thompson), and perhaps by the ventral swelling which is shown in Thompson's figure but not labelled. 7 Beginning with the region where the lung outgrowths are found, and extending backward as far as the liver-bud, the epithelium is markedly thickened.

The liver-bud is a median ventral knob-like outgrowth of the

Ph. /—

Mem. ph.— 01. th.~.

Fiq. 236. — Graphic reconstruction of an embryo with 23 paired somites, , 'showing a median sagittal section of the digestive tract. X40diam. (After Thompson.) All., allantois; CL, cloaca; C. per., cavum pericardii; Div. hep. (diverticulum hepaticum), liver bud; D.v., ductus vitellinus; Gl.th., glandula thyreoidea; Mem. cl., membrana cloacalis; Mem. ph., pharyngeal membrane; PA.,lpharynx; Sept. tr., septum trans versum.

digestive tube, extending into the septum transversum, which is the layer of mesoderm between the pericardial cavity and the vitelline duct. The hepatic bud contains a cavity which communicates freely with the alimentary canal. There is no trace of the pancreas.

7 According to Grosser, the pulmonary area in this embryo has been erroneously regarded as the gastric portion of the intestine by Thompson and also by Keibel and Elze (Normentafel, Embryo Nr. 7). In Thompson's embryo the stomach has not yet developed. Compare with Grosser^ description of the development of the lungs at the end of this chapter.

312 Beyond the vitelline duct is the hind-gut, with a rounded lumen. It expands at the cloaca, where it joins the allantois, and extends a short distance into the tail. The allantois is a small tubular structure with two marked dilatations and, at its distal end, ' ' a small swelling bent upon itself. ' ' Separation of the (Esophagus from the Trachea. — In an embryo which has the general shape of Thompson's specimen, but which is somewhat more advanced (Bremer, 1906), the lungs and trachea form a pear-shaped mass attached to the ventral border of the oesophagus. The lower portion of the mass, which bulges toward either side, represents the division of the trachea into the bronchi. Its cavity is still in free communication with that of the oesophagus. The trachea will become separated from the cesopha

Sept.

Pul.

Sul.

Fig. 237. — Wax model from Bremer's 4 mm. embryo, showing the "lung-bud," Pul., and the adjacent part of the oesophagus. X175 diam. Sept., tracheo-cesophageal septum; Sul., lateral oesophageal groove.

Oes.

Fig. 238. — Abnormal communication between the oesophagus, Oes., and the trachea, Tr. The upper portion of the oesophagus, Oe., ends blindly below. (After Keith.)

gus by the down-growth of the lung-bud and the upward extension of the notch between the lung-bud and the oesophagus. The notch extends upward following a fusion of the lateral walls of the fore-gut, which begins from below (His, 1885, p. 17-18). The approximation of the lateral walls to form the tracheo-oesophageal septum is seen at the top of Fig. 237.

In the most common anomaly of the oesophagus, which must arise at the stage under consideration, the oesophagus is transversely divided into two parts. The upper portion ends blindly below, and the lower portion arises from the trachea near its bifurcation (Fig. 238), or even from a bronchus, into which its lumen opens.

This malformation is detected soon after birth, since milk cannot enter the Stomach. Happich (1905) has tabulated the records of 59 cases. Sometimes the

INTESTINAL TRACT AND RESPIRATORY ORGANS. 313 two portions of the oesophagus are connected by a strand containing smooth muscle-fibres, but " it is entirely unknown whether there is any remnant of epithelium in the interval." No epithelial connection has been recorded. Although this complex anomaly is common, a simple communication between cesophagus and trachea, when these are otherwise normal, is extremely rare. Forssner (1907) and Giffhorn (1908) have discussed the origin of the divided cesophagus and explained it with diagrams.

To produce the common form of the anomaly the lower portion of the tracheocesophageal septum must fail to develop, thus leaving the cesophagus in communication with the lower part of the trachea. In the model shown in Fig. 237 there is externally, on either side of the cesophagus, an oblique depression in the epithelium with a corresponding internal elevation. It is so situated that if the walls of the cesophagus should coalesce along this groove a ventral portion would be cut off, communicating freely with the trachea near its bifurcation. The groove does not reach the dorsal border of the cesophagus, but extends downward toward the liver. However, the part of the cesophagus dorsal to this groove has a narrower lumen than the ventral part; to produce the anomaly, this portion must become occluded. Apparently the lateral oesophageal groove, which seems correlated with the shape of the adjacent body-cavity, has not been previously described.

It has been thought that the closure of the cesophagus in the anomaly is due to the pressure of neighboring arteries, particularly the right dorsal aorta, and several cases have been found associated with the low origin of the right subclavian artery. Keith (1906) reported four cases, in three of which this abnormal artery was present and crossed to the right side between the two parts of the cesophagus. In the 4 mm. embryo, however, which is close to the stage in which the anomaly must arise, the arteries are not near this portion of the cesophagus. In the embryo of 4.9 mm. shown in Fig. 239 the separation of the cesophagus and trachea has proceeded so far that the anomaly could hardly develop.

The further history of the anterior portion of the digestive tract, including the mouth, pharynx, trachea, and lungs, will be presented by McMurrich and Grosser in separate sections of this chapter.

Ingalls's Embryo. The Formation of the Stomach and Pancreas. — Ingalls (1907) described an embryo measuring 4.9 mm., the digestive tract of which is shown in Fig. 239. Rathke's pocket is present; nothing is said of Seessel's pocket. The pharyngeal membrane has entirely disappeared. The thyreoid gland is still connected with the pharynx, as in Thompson's specimen. The trachea is quite separate from the cesophagus.

"The cesophagus is a tube, circular in cross section, which has thinner walls and is much narrower than the ventrally placed trachea. In the region of the fourth cervical segment it becomes gradually larger, its walls thicken, and at the same time it becomes flattened laterally. Thus it forms the stomach, and the digestive tube then assumes an oblique position, with its ventral border turned somewhat to the right and its dorsal border correspondingly to the left. ... At its caudal end, where it passes over into the duodenum, the stomach again becomes very narrow." (P. 549-550.)

314 At this stage, as shown in the figure, the stomach is a welldefined spindle-shaped enlargement of the fore-gat.

The liver has become very large. From the knob-like diverticulum, such as was seen in Thompson's specimen, a great mass of anastomosing cords of cells has grown out, invading the septum transversum. In a cross section of the embryo this mass is

Div. hep.

"--. Pane. v.

Fig. 239. — The digestive tract of an embryo of 4.9 mm., shown in median sagittal section. X24 diam. (After Ingalls.) All., allantois; Div. hep., diverticulum hepaticum; D. v., ductus vitellinus; D. W. (ductus Wolffi), Wolffian duct; Ga. (gaster, ventriculus), stomach; Gl. th., glandula thyreoidea; Hep., medial surface of the right lobe of the liver (hepar); Int. cau., intestinum caudale; Mem. cl., membrana cloacalis; Pane. d.,?pancreas dorsale; Pane, v., pancreas ventrale; R., Rathke's pocket; Tr., trachea.

U-shaped, and the stomach and duodenum are lodged in the hollow of the U. In sagittal section, as in the figure, the part of the mass which crosses the median line has been cut through. It is shown in section. The inner surface of the right lobe of the liver has also been drawn.

The pancreas of the adult arises in the embryo as two separate organs, namely the dorsal pancreas and the ventral pancreas. In Ingalls's specimen, between the stomach and the common bile-duct,

INTESTINAL TRACT AND RESPIRATORY ORGANS. 315 the dorsal surface of the duodenum presents a thick-walled outpocketing, which is the beginning of the dorsal pancreas. The ventral pancreas grows downward from the lower side of the hepatic diverticulum, at its junction with the intestine. It is adherent to the wall of the intestine. It contains a minute lumen, not shown in the figure, which appears to communicate with the hepatic diverticulum. (A more detailed account of the development of the liver and pancreas will be found in subsequent sections of this chapter.) The connection between the yolk-sac and the intestine in Ingalls's specimen is a slender vitelline duct, lined with a single layer of large cuboidal or cylindrical cells. The intestine bends ventrally toward its junction with the duct, and beyond this point is becomes much smaller. It then enlarges toward the cloaca and continues with a relatively large lumen into the tail. Toward the tip of the tail its wall becomes irregular in thickness.

The cloacal membrane does not consist of thickened ectoderm applied to thickened entoderm, as in a younger specimen already described. On the contrary, the ectodermal layer is here so thin that in places it can scarcely be recognized. The entodermal layer is also thinner than in the lateral walls of the cloaca. Keibel (1896) found the ectoderm of the cloacal membrane thinner than the entoderm in a 3 mm. specimen. At 4.2 mm. the layers were indistinguishable, and in discussing later stages he wrote, "The question, how much ectoderm and how much entoderm take part in the formation of the cloacal plate, must remain undecided. ' ' In Ingalls 's specimen the allantois shows two small expansions near its distal end. Proximally the allantois joins the cloaca, which is being gradually subdivided, in the cranio-caudal direction, by the growth of the cloacal septum. The Wolffian ducts empty into the ventral part of the cloaca, one on either side, and, although they are not of entodermal origin, they have been included in the accompanying figures.

Embryo of 7.5 mm. Detachment of the Yolk-sac. Origin of the Caecum and Vermiform Process. — The entodermal tract in an embryo measuring 7.5 mm. is shown in Fig. 240. Rathke's pocket still has a broad connection with the oral cavity, but the thyreoid gland has become detached. The oesophagus is much longer than in Ingalls's specimen. It becomes gradually smaller toward the stomach and then enlarges, but there is no definite boundary between stomach and oesophagus. The epithelial portion of the stomach is flattened laterally, and is so placed that its left side faces somewhat ventrally and its right side dorsally. The stomach passes gradually into the duodenum, the diameter of which is considerably greater than that of the distal part of the small intestine.

316 The liver consists of a large mass of anastomosing cords of cells, connected with the hepatic diverticulum by a short thick stem which represents the hepatic duct. Distal to the hepatic duct the diverticulum gives rise to the gall-bladder and cystic duct. Proximal to the hepatic duct it forms the common bile-duct (ductus choledochus) , and it connects with the ventral pancreas just before

Pane. d.

- Pane. v.

Pr. ver.

Int. can. Mem. el.

Fig. 240. — The digestive tract of an embryo of 7.5 mm. (Harvard Collection, Series 256). X 16 diam. In addition to the structures lettered as in Fig. 239, the following are shown. Oes., oesophagus; Pr. ver., a dilatation of the lower limb of the intestinal loop, which gives rise to the processus vermiformis and the csecum, and which marks the boundary between the small intestine above and the large intestine below; Vea. fel. (vesica fellea), gall-bladder.

joining the duodenum. The dorsal pancreas is more sharply denned than in Ingalls's specimen.

The slight bend of the intestinal tube toward the vitelline duct, which is seen at 4.9 mm., has increased and now forms the important primary intestinal loop. The vitelline duct has become detached from the bend of this loop, and is separated from it by a considerable interval. It lies in a prolongation of the mesentery

INTESTINAL TRACT AND RESPIRATORY ORGANS. 317 which, together with its contents, is called the yolk-stalk. The fused vitelline veins lie in a portion of the stalk which has become separated from the rest, forming the upper subdivision shown in the figure.

The detachment of the vitelline duct usually occurs at about this stage. A similar condition has been figured by Elze (1907) in an embryo of "about 7 mm." and by Mall (1891) in a 7 mm. specimen. But sometimes the vitelline duct, reduced to a strand of epithelial cells, retains its connection with the intestine much longer. Keibel and Elze (1908) recorded its presence in an embryo of 12.4 mm. and Thyng has found it in a specimen measuring 13.6 mm. In about 2 per cent, of adults, according to several tabulations, a persistent pouch of the intestine, 3 to 9 cm. long, marks the place where the vitelline duct formerly opened into it. This diverticulum ilei of Meckel has already been described, and the pathological importance of persistent remnants of the yolk-stalk has been noted (p. 293). The further history of the detached vitelline duct, which extends through the umbilical cord, and of the yolk-sac, which is lodged between the amnion and chorion at the distal end of the cord, may be found in vol. i, p. 173-174. The account of the yolk-sac as a jDortion of the digestive tract may be concluded with the following note concerning its entodermal layer. In Bremer's 4 mm. specimen and in a 4 mm. embryo figured by Keibel and Elze, the entoderm presents numerous solid outgrowths and hollow outpocketings (Fig. 241, A). Toward the cavity of the yolk-sac the entoderm has a wavy outline. The gland-like structures were described by von Spee as follows (1896 2 ) : "Such glands arise as little outpocketings of the entodermal lining of the distal pole of the yolk-sac, but they rapidly become elongated sacs, which branch dichotomously, and soon develop expanded club-like or vesicular end-pieces. They extend through almost the entire thickness of the wall of the yolk-sac, but their blind ends remain separated from the body-cavity by a single layer of the mesoderm. The glands are lined by a single layer of prismatic entodermal cells, the protoplasm of which contains many fine vacuoles, and, especially in later stages, an increasing quantity of fat drops blackening with osmium." In an embryo measuring 9.4 mm. (Fig. 241, B) these glandlike structures are found all over the yolk-sac. Most of them are closed cysts with walls of varying thickness. Occasionally distinct branching is seen. Von Spee believed that the yolk-sac in its glandular stage is an active organ comparable physiologically with the liver. Other investigators have questioned the glandular nature of the yolk-sac tubules, and Meyer (1904) states that, in spite of the large amount of material at his disposal, he is unable to reach any satisfactory conclusion as to the meaning of these

318 HUMAN EMBRYOLOGY.

tubules. The entodermal cells of the volk-sac have been found to contain granules which yield a typical mucin reaction (Jordan, 1907), and bundles of filaments which stain with iron haematoxylin (Branca, 1908). According to Branca, the superficial cells are provided with terminal bars, and, except within the glands, some of them show distinct cilia or brush borders. Jordan (1910) failed to find terminal bars or cilia. They do not appear in the specimens shown in Fig. 241, but the appearance which Branca described as a brush border is sometimes clearly seen. In later stages the epithelium lining the yolk-sac becomes stratified, and the diverticula and intra-epithelial cysts disappear. Thus, in an embryo of 23 mm. (Fig. 241, C) the epithelium consists of vacuolated degenerating cells. Subsequently these are lost, and the wall of the yolk-sac is then merely ' ' a dense wavy layer of fibrous connective tissue. ' ' In this condition it is found at birth.

In the posterior half of the primary intestinal loop in the 7.5 mm. embryo (Fig. 240), there is an abrupt enlargement of the entodermal tube, which marks the boundary between spaall and large intestine. The expansion does not affect the dorsal border of the intestine, but is wholly a ventral bulging. The cavity of the large intestine extends slightly forward into the ventral swelling, so that in one section at this point there is a double lumen: the enlargement is therefore already a shallow pouch. This structure is generally considered to be the beginning of the ccecum. Keibel and Elze (1908) have noted that the caecum is indicated in embryos measuring from 6.25 to 7.0 mm. But Tarenetzky (1881) described this enlargement as the processus verrniformis. "At this stage ... a caecum is not present." He found that an actual caecum first appeared in an embryo of 65 mm. A distinction between the caecum and the vermiform process in young human embryos is not easily made, and Toldt (1894) is probably correct in referring to the primary enlargement as the common origin of both. This question is further discussed on page 328.

In the 7.5 mm. specimen the large intestine curves forward to join the allantois at the cloaca. There is still no external opening. The caudal intestine, which in Ingalls 's specimen had a wide lumen, is reduced to a strand of cells. Toward the tip of the tail a minute lumen is distinctly seen, ending in a terminal expansion (Fig. 234, D). In an embryo of 9.4 mm. (Fig. 242) the caudal intestine has disappeared, except for an isolated nodule of epithelium. Keibel and Elze have shown that it usually disappears at about this stage, but they found remnants of it in one specimen measuring 11.5 mm.

Embryo of 9 A mm. Torsion of the Primary Loop of Intestine.-^-The intestinal tube in the 9.4 mm. embryo (Fig. 242) differs from that of the preceding stage chiefly through the disappearance

INTESTINAL TRACT AND RESPIRATORY ORGANS.

319

of the yolk-stalk and the torsion of the intestinal loop. (The considerable changes in the course of the bile-ducts will be described in a separate section.) The torsion of the intestine occurs, in a general way, as follows. The large intestine at first forms part of the posterior half of the intestinal loop, and the loop is in the median plane. Then the loop becomes rotated so that its plane is transverse. The anterior half is then on the right, and the posterior half on the left. Further rotation causes the posterior half to become anterior. In side view the large intestine then crosses the small intestine, as seen in Fig. 242. At a considerably later stage the torsion is completed by the migration of the caecum

B

Fig. 241. — Sections of the wall of the yolk-sac. X115 diam. A, embryo of 4 mm. (Harvard Collection, Series 714); B, embryo of 9.4 mm. (Harvard Collection, Series 529); C, embryo of 23.0 mm. (Harvard Collection, Series 192). Developing "glands," Gl., are shown in A. They have become cystic in B. In C the yolk-sac is degenerating, and the "glands" have disappeared.

to the right side of the body and down toward the pelvis. When this has occurred, the large intestine passes from right to left, ventral to the upper part of the small intestine ; it then descends to the rectum on the left of the small intestine.

In an interesting case of " imperfect torsion of the intestinal loop " reported by Reid (190S), the embryonic twisting evidently did not occur. In a man over sixty years of age he found that the caecum was within the pelvis, a little to the right of the median line. The ascending colon passed gradualty to the left, and most of it, together with all of the transverse and descending colon, was on the left side of the body. The small intestine was wholly on the right side, and was not crossed by the large intestine. Similar eases have been described by Faraboeuf

320 (1885), Descomps (1909), and others. A reversal of the intestinal torsion, which would cause the colon to pass under the small intestine, has apparently never been seen.

Embryo of 22.8 mm. The Normal Umbilical Hernia. — Since 1817, when Meckel published an account of the " formation of the intestinal canal of mammals and particularly of man," it has been well known that, at a certain stage, " the greatest part of the intestinal canal is found within

Gi. th.

Oes.

Int. cau Mem. cl

Pane.

- Pr. ver.

- D.W.

Fig. 242. — The digestive tract of an embryo of 9.4 mm. (Harvard Collection, Series 1005). X13 diam.

The lettering is the same as in Figs. 239 and 240.

the umbilical cord." It had been discovered previously. Meckel observed it in the goat, sheep, cow, pig, rabbit, and man. He says, " Through a very pleasant coincidence Oken and I, at the same time and quite independently of one another, expressed the opinion that in a very early embryonic stage this position of the intestine is normal." According to His (1885), "the underlying cause of the ventral extension of the intestine is doubtless to be sought in its connection with the yolk-sac. ... As long as the yolk-stalk is present it is attached to the end of the loop extending through the umbilicus." But Mall (1898) states that " before the intestine begins to enter the cord its connection with the duct is

i

INTESTINAL TRACT AND RESPIRATORY ORGANS.

321

severed." Since the liver grows downward and crowds upon the rapidly elongating intestine, Mall considers that " the intestine must escape if it has a chance, and the coelomic space within the cord naturally receives it." The loop of intestine begins to enter the cord in embryos of about 10 mm.; at 22.8 mm. (Fig. 243) the umbilical hernia is well

Hy.—

Pr. ver.*

Hern. umb.-^~.'_

Or. u.-g.Mem an.

Fig. 243. — The digestive tract of an embryo of 22.S mm. (Harvard Collection, Series 871). X6 diam. (Drawn by F. P. Johnson.) In addition to the structures lettered as in previous figures the following are shown. Fund., fundus of the stomach; Hern, umb., coils of intestine within the umbilical cord, forming the umbilical hernia; Hy., anterior lobe of the hypophysis, — the detached end of Rathke's pocket; Int. rec, intestinum rectum; Mem. an., membrana analis; Or. ti.-g., orificium urogenitale.

developed. 8 It will be seen that the large intestine presents no

8 The reconstruction of the 22.8 mm. embryo (Fig. 243), which has not been previously published, is the work of Mr. F. P. Johnson, and the writer is much indebted to him for permission to use it. Vol. II.— 21

322 HUMAN EMBRYOLOGY.

well-marked convolutions, but that there are several bends in the course of the small intestine. This is due to the relatively rapid growth of the anterior half of the original loop. According to Mall (1898), there are six primary loops of the small intestine, first indicated in embryos of about 17 mm., and recognizable, in spite of secondary coils, even in the adult. The first loop encircles the head of the pancreas. The third loop is concave below and occurs where the small intestine passes through the narrow umbilical outlet. It is clearly shown in Fig. 243. Loop 2 is between 1 and 3, forming together with 3 an S-shaped curve. The other loops are not so easily defined, but all coils between the caecum and the place of attachment of the yolk-stalk are included in loop 6.

Mall considers that these coils in the umbilical cord are so fixed that it is not difficult to recognize the various loops after their return to the abdominal cavity. In the adult he finds that loops 2 and 3 make two distinct groups of coils in the left hypochondrium, loop 2 communicating with the duodenum. " After this the intestine passes through the umbilical region to the right side of the body (loop 4). Then the intestine recrosses the median line to make a few convolutions in the left iliac fossa (5), after which it fills the pelvis and lower part of the abdominal cavity between the psoas muscles (6)." He concludes that "the various loops of the adult intestine, as well as their position, are already marked in embryos of five weeks, and the position of the convolutions in the adult is as definite as the convolutions of the brain." Separation of the Intestine from the Allantois. — In embryos of the stage of Ingalls's specimen (Fig. 239) the cloaca is elongated anteroposteriorly, and the Wolffian ducts empty into its ventral portion. Later this ventral portion is split off from the dorsal part, apparently by the down-growth of the connective tissue between the allantois and rectum. The portion of the allantois 9 below the Wolffian ducts, since both the urinary and genital passages open into it, is then called the urogenital sinus.

In the 9.4 mm. embryo (Fig. 242) the urogenital sinus has been formed, but the cloaca still remains as a broad connection between the allantoic and intestinal tracts. By further downgrowth of the connective tissue, this connection becomes reduced to a slender passage called the cloacal duet. Finally the walls of the duct coalesce and the communication between the intestine and the allantois is obliterated. The connective tissue between the rectum and the urogenital sinus, where it reaches the ectoderm, constitutes the primitive perineum.

In Keibel and Elze's tables, a cloacal duct is recorded in embryos from 11 to 12.5 mm. At 15 mm. the "cloaca is still not 9 It is clearly a matter of definition whether the portion which is added to the allantois by the subdivision of the cloaca should thereafter be called allantois (see Chap. XIX).

INTESTINAL TRACT AND RESPIRATORY ORGANS.

323

fully divided." At 15.5 mm. the "cloaca is just divided, bul tin* epithelia of the urogenital sinus and of the rectum still connect; a mesodermal perineum is not yet formed." Fig. 244, A, from a specimen measuring 18.1 nun., presents this condition. The cloaca! membrane is now subdivided into the urogenital membrane ventrally and the anal mem Inane dorsal ly. The mesodermal primitive perineum is about to form, but the urogenital sinus and the rectum are apparently still connected by entoderm. At 22.8 mm. (Fig. 243) the primitive perineum is well developed. It is found at the bottom of a median sagittal ectodermal groove, known as the ectodermal cloaca. The anal membrane is also at the bottom of an ectodermal depression which may be regarded as a part of the ectodermal cloaca, but which is termed the proctodeum or anal pit. When the edges of the ectodermal cloaca coalesce in the perineal region, so as to form a raphe, the permanent perineum is produced.

Rec.

Mem.u-g. Mem

— Muse.

Am.rec.

Mem. an.

Sph.ex.

A

B

An.

Sph. in. *P' iex - Am.rec. C

Fig. 244. — Median sagittal sections to show the separation of the rectum from the urogenital sinus. A, embryo of 18.1 mm. (Harvard Collection, Series 1129), X30 diam.; B, embryo of 22 mm. (Harvard Collection, Series Sol), X25 diam.; C, embryo of 32 mm. (Harvard Collection. Series 292), X 12 diam. Am. rec, ampulla recti; -4/).. anus; Mem. an., membrana analis; Mem. u.-g., membrana urogenitals; Muse., tunica muscularis (including inner circular and outer longitudinal layers); Or. u.-g., orificium urogenitale; Per., perineum; Pr., proctodeum; Rec., rectum; Sph. ex., M. sphincter ani externus; Sph, int., M. sphincter ani internus; S. u.-g., sinus urogenitalis; x, terminal bulbous enlargement of the rectum.

The urogenital tract acquires ai^ external opening before the anal membrane is perforated, and this has occurred in the 22.8 nun. specimen. The further history of the allantois and urogenital tract will be found in Chapter XIX.

The Formation of the Anns. — A sagittal section through the rectum and proctodeum of a 22 mm. embryo is shown in Fig. 244, B. Just before the rectum reaches the anal membrane it forms a bulbous enlargement, seen also in Fig. 243 and in embryos of 17.5 and 18.5 mm. drawn by Keibel. Iu these specimens the circular and longitudinal muscle layers of the rectum are easily recognized. The terminal swelling of the rectum extends beyond the muscle layers, as recorded by Keibel ( L896) in a beautifully

324 HUMAN EMBRYOLOGY.

illustrated and fundamental description of the development of the human urogenital tract. In a 29 mm. specimen he described the musculature as follows (cf. Fig. 244, B and C) : "The circular muscle layer ends very abruptly at the level of the little caudal swelling of the intestine, and already it may be referred to as the beginning of the M . sphincter ani internus.

"The outer longitudinal layer of the intestinal musculature is arranged differently. It ends at the same level, but no sharp caudal limit can be recognized. The M. sphincter ani externus is clearly indicated and is relatively quite large. The cranial border of this muscle is found where the musculature of the intestine ends, therefore at the level of the little caudal entodermal enlargement. The M. sphincter ani externus is separated from the epithelium of the intestine by a rather thin layer of connective tissue, which is continuous with the connective tissue surrounding the intestine further cranially, and also with the longitudinal layer of the muscularis, from which strands of cells may be followed into it." In a 32 mm. specimen (Fig. 244, C) the anal membrane has disappeared. Along the dorsal wall of the anal canal there is a slight indication of the terminal bulbous enlargement, but it seems clear that it is a transient structure. It is probable that the elongated swelling above it gives rise to the rectal ampulla of the adult.

Otis (1905) has studied the external configuration of the embryonic anus, and has found that the development of the external sphincter produces characteristic elevations. In embryos of 21-23 mm. there is a pair of external elevations, one on either side of the anal pit. At 26 mm. these have united dorsal to the anus, thus forming a single crescentic mound. The horns of the crescent grow forward toward the perineum and finally meet, so that the mound encircles the anus.

Malformations of the Anus. — The perforation of the anal membrane normally takes place in embryos of about 30 mm. It is accompanied by the formation of degenerative material staining intensely with eosine, which blocks the outlet and makes the determination of an aperture somewhat difficult. In the Harvard Collection there are embryos measuring 22 mm., 22.8 mm., and 29 mm. in which perforation has occurred, and specimens of 22.8 mm. and 30 mm. in which the anus seems still impervious. Keibel and Elze's series of seven embryos measuring from 22 to 26 mm. includes only one (22.5 mm.) in which the anus is open. A persistence of the anal membrane until birth has been assumed to account for cases of atresia ani, in which the rectum ends blindly below, and the anus is represented merely by a slight depression in the skin. In these cases, however, the epithelial connection between

INTESTINAL TRACT AND RESPIRATORY ORGANS.

325

the rectum and anal pit has been lost, so that an invasion of the anal membrane by connective tissue must have taken place. In other cases, known as atresia recti, the anal canal is present but it leads into a blind sac of intestine.

In this connection the obliteration of the lumen of the rectum, observed by Keibel in an embryo measuring 11.5 mm., is of interest. He found that " the ej)ithelium of the lower portion of the intestine blocked the lumen at two small places," but since similar conditions were not observed in other specimens he concluded that " this may well be only a chance and meaningless adhesion " (1896, p. 79). Although this observation in human embryos remains unique, Lewis (1903) has recorded a similar condition in pig and rabbit embryos, and it occurs more extensively in birds (Minot, 1900). It has been suggested that its function is to prevent the passage of the excretion of the "Wolffian bodies back into the intestine. After the cloacal duct has become obliterated, it is not needed for this purpose, and hi these later stages it is not found. It is possible that atresia recti may be due to the persistence of such an occlusion, with invasion by connective tissue.

Less common than the simple imperforate anus or imperforate rectum are cases in which the cloacal duct persists, forming

Am.rec.

La.mi

Fig. 245. — Abnormal position of the anus in a child. V. natural size. (After Mackenzie.) Am. rec., ampulla recti; La.mi. labium minus ; Ur., urethra; Va., vagina.

a slender passage from the rectum to the raphe of the perineum, scrotum, or under side of the penis, or to the prostatic urethra or bladder. In the female such a fistula may open along the perineal raphe or into the vestibule of the vagina.

An interesting series of diagrams of these cases was published by Stieda (1903). The fistula may exist with a normal anus, but more frequently it is associated with an imperforate condition. Of the many cases reported two examples may be cited.

Reichel (1888) received a patient 25 years of age who complained of involuntary discharge of fseces through the vagina, beginning after her marriage three years before. The anal canal was found to be normal ; the perineum was extremely short and the vestibule strikingly deep; the labia were normal. A canal, lined with mucous membrane, was found leading from the rectum to the vestibule directly below the hymen. After discussing the possibility of mechanical injury, etc., Reichel concludes that the abnormal communication between the rectum and the

326 HUMAN EMBRYOLOGY.

vestibule was present as a slender fistula from birth, and that it was dilated following coitus. Such a condition would arise in an embryo of about 12 mm. provided that the perineal tissue should encircle the cloacal duct instead of obliterating it.

In 1906 Mackenzie reported the case shown in Fig. 245. The patient died at the age of 1 year and 11 months, having suffered from alternating attacks of diarrhoea and constipation since birth. Robinson described the condition as follows : " The anal passage runs, not in the normal direction, downwards and backwards, but downwards and forwards, and the anal orifice opens into a chamber common to it, the vagina, and the urethra; that is, the anal passage opens, not on the surface behind the genito-urinary chamber, but into a cloaca." It may be considered that the anal canal in this case, although provided with well-marked anal columns (of Morgagni), corresponds with the slender fistula in Reichel's case, and that the normal anal outlet is not represented. This is Robinson's interpretation. He says, " The entodermal cloacal chamber has never been separated into two parts : it has opened into the anterior part of the external cloacal depression, and the posterior part of that depression, if it existed, has disappeared, no trace of a proetodeal opening being discoverable." It seems possible, however, that the entodermal cloaca has been completely divided but that the primitive perineum has persisted. Thus by an imperfect development of the perineum the normal anal opening is displaced forward. Robinson rejects this interpretation.

Embryo of 42 mm. Return of the Intestines to the Abdominal Cavity.— According to Mall (1895), the return of the intestines from the umbilical cord into the body must take place very rapidly, for in embryos of 40 mm. the intestine is either in the cord or in the abdominal cavity. He found no intermediate stages. Mall was unable to determine the cause of the return, but he showed that the abdominal walls do not bulge forward so as to include the cavity of the cord within the abdomen. The intestines slip back through a rather small aperture, and the cavity in the cord is then obliterated. From a study of pig embryos Mall suggested that the increase of loops within the abdominal cavity, and their rotation, may draw upon the loops in the cord. The enlargement of the umbilical arteries on the under side of the hernia may also exert a favorable pressure. As seen in the reconstruction of a 42 mm. embryo (Fig. 246), the return has taken place, but the abdominal cavity still extends into the cord. 10 Development of the Ccecum and Vermiform Process. — The first appearance of the intestinal enlargement which is to produce the vermiform process has already been described. In the embryo of 7.5 mm. it is an entodermal swelling on the lower side of the caudal limb of the intestinal loop. After the torsion of the loop it may still project from the lower side of the intestine, as shown in the 9.4 mm. specimen (Fig. 242) and in a 13.8 mm. embryo

10 In a " Supplementary Note on the Development of the Human Intestine," Mlall (1899) described a human embryo of 32 mm. "in which the intestine is in the act of returning from the coelom of the cord to the peritoneal cavity." The intestine, is said to be " sucked back " to fill the space made by the enlargement of the abdominal cavity.

INTESTINAL TRACT AND RESPIRATORY ORGANS.

327

figured by His, But in a specimen measuring 12.5 mm. His has drawn it as projecting from the upper side of the intestine, and

Ghih.

Oes.

Fund.

Hep.

All

Am. rec.

Fig. 246. — The digestive tract of an embryo of 42 mm. (Harvard Collection Series 838). . X4 diam. The lettering is like that in previous figures with the addition of Ves. ur. (vesica urinaria), bladder.

it has been similarly figured by Keibel and Elze in an embryo of 14 mm. Mall found it projecting laterally in a 17 mm. specimen.

328 In all these cases, however, the apex of the projection is directed ventrally (that is, toward the small intestine). With the formation of the umbilical hernia, the vermiform process enters the cavity of the cord (Fig. 243). Later it is withdrawn into the abdomen and comes to lie against the under side of the liver.

Four stages in the development of the vermiform process are shown in Fig. 247, two of which are drawn from models and two from dissections. All are viewed from the median side. Fig. 247, A, shows the simple arrangement at 9.4 mm. To produce the condition shown in B the tip of the vermiform process must be brought toward the large intestine. Thus a U-shaped bend would result, and this U should then be twisted upon the small intestine

Val.co.

Fig. 247. — Models (A and B) and dissections (C and D) to show the development of the vermiform process. A, embryo of 9.4 mm. (Harvard Collection, Series 1005), X50 diam.; B, embryo of 42 mm. (Harvard Collection, Series 838), X20 diam.; C, embryo of 95 mm., X3.5 diam.; D, embryo of 218 mm., X3.5 diam. Cae., caecum; Co., colon; II., ileum (small intestine); Ale., mesenteriolum; Mes„ mesentery; Pr. ver., processus vermiformis; Val. co., valvula coli, represented by two slight vertical swellines between which is the outlet of the ileum.

so that its extremities extend dorsally, with the vermiform process on the right side of the colon. In Fig. 247, B, from the 42 mm. embryo, a window has been cut in the round bend made by the vermiform process and the colon, so that the outlet of the ileum is exposed. The ileum empties into the large intestine in the concavity of the bend.

Tarenetzky (1881) has described very similar relations in a 33 mm. embryo as follows: " The processus vermiformis has assumed an elongated form ; it is no longer parallel with the ileum but forms a right angle with it. It has taken a position toward the right and obliquely above and in front of the terminal part of the ileum, so that its tip is already directed somewhat toward the colon. In this manner it forms also a right angle with the colon. The knee-shaped bend at the passage of the vermiform process into the colon is not expanded, so that at this stage no true caecum is present. The tip of the vermiform process is completely free. Along its base and middle piece there is attached a well-defined peritoneal

INTESTINAL TRACT AND RESPIRATORY ORGANS. 329 fold, which arises from the adjacent ventral right plate of the mesenterium communeThis fold is new, and represents the mesenteriolum of the vermiform process, the chief vessels to which are contained in it.' ; The mesenteriolum is shown in Fig. 247, B, and, as Tarenetzky recorded, a dilatation to indicate the caecum is not well denned. It has already been noted that Tarenetzky first recognized the caecum in embryos of 65 mm. Toldt (1894) describes a clearer separation between the caecum and vermiform process in a "7 weeks" embryo than his figures indicate. In his drawing of a 50 mm. specimen the caecum can scarcely be distinguished. The demarcation evidently forms very gradually and at a late stage. According to Toldt, the taeniae of the caecum are present at birth, and the haustra of the caecum develop in the first half year, the smallest of them, situated nearest the vermiform process, appearing first; but the caecum does not acquire the characteristic adult form until the third or fourth year.

It was shown by Toldt that the bending of the vermiform process and caecum upon the colon gives rise to the valves of the colon. Until this bend occurs there is no indication of the valves. As a result of the bend, the end of the small intestine, where it is caught in the angle, becomes flattened by the adjacent walls of the colon and caecum respectively. At 42 mm. (Fig. 247, B) the aperture is still nearly round, but as the U-shaped bend becomes angular the flattening will result. This explanation accounts for the two lips of the valve, the labium inferius being toward the caecum, and the labium superius toward the colon. In the last fetal months and especially after birth, the relatively great expansion of the large intestine, as compared with the ileum, causes the vailves to increase in size. In this process the bulging colon and caecum still further invest the end of the ileum and adhere to it. In case the embryonic bend is not highly developed, imperfect valves may arise by the expansion of the large intestine. Toldt recorded several such cases.

Recently Parsons (1907) has reported the case of an elderly man in whom the caecum formed a straight continuation of the colon, and there was no valve whatever. He considers that the U-shaped bend had never formed in that individual. Smith (1903) described a case in which a vermiform process was present but there was a " complete absence of a properly constituted caecum " and no trace of the valvula coli.

The stage of the bend represented in Fig. 247, B, is therefore a critical one. In a 95 mm. embryo, Fig. 247, C, the vermiform process is still in contact with the liver, and the U-shaped bend is well marked, but the descent of the caecum toward the pelvis has begun. At 218 mm. (Fig. 247, D) the vermiform process has taken its final position. In this case it is coiled so as to make l 1 /? revolutions.

330 HUMAN EMBRYOLOGY.

Form and Position of the Stomach. — At first the stomach lies approximately in the median sagittal plane. It is then a flattened expansion of the digestive tube, with dorsal and ventral borders and right and left surfaces. Gradually it rotates so that its left side becomes ventral and its right side correspondingly dorsal. At the same time the dorsal border is turned to the left and the ventral border to the right. The upper portion of the stomach is displaced to the left side of the body, and the borders thus become curvatures, which are concave toward the right. The original dorsal border forms the greater curvature, and the ventral border becomes the lesser curvature. These changes in position are usually described in connection with the development of the mesenteries. They are best shown in ventral views of the embryo. It may be noted, however, that in the 7.5 mm. embryo the rotation of the stomach has partly occurred, so that the original dorsoventral axis forms an angle of 20° with the median plane; in the 22.8 mm. specimen the angle has increased to 55° ; and in the 45 mm. embryo it is 75° in the pyloric half of the stomach. The cardiac end has not rotated so much, and at 45 mm. its angle is 40°. Thus the pyloric part of the stomach is twisted across the body from left to right. .

The descent of the stomach has been described by Jackson (1909) as follows: In the 11 mm. embryo the cardia lies opposite the 3d or 4th thoracic segment, and the pylorus opposite the 7th or 8th. In the 17 mm. embryo the two ends of the stomach seem to have reached approximately their permanent positions, the cardia opposite the 10th thoracic vertebra and the pylorus opposite the 1st or 2d lumbar vertebra.

The descent is accompanied by a great elongation of the oesophagus. In a 9.4 mm. specimen the oesophagus measures 1.8 mm. At this proportion it should measure 4.3 mm. in an embryo of 22.8 mm., but its actual length is found to be 8 mm. In Jackson's paper the relations of the stomach to the adjacent viscera in early embryos have been considered.

The most notable external feature in the early development of the stomach is the formation of the fundus, which occurs in the manner described by Keith and Jones (1902). According to these authors, the fundus of the human stomach is developed, not as a general expansion of the gastric part of the fore-gut, but in the form of a localized outgrowth or diverticulum at the cardiac end of the greater curvature (dorsal border). In its manner of origin it has much in common with the caecum and vermiform process. They find that the outgrowth is best marked in embryos of the third and fourth month. After these months the diverticulum is not so well defined, since it expands and merges with the body of the stomach. The gradual development of the fundus

INTESTINAL TRACT AND RESPIRATORY ORGANS. 331 as a conical diverticulum is shown in the embryos of 9.4, 22.8, and 42 mm. (Figs. 242, 243, and 246).

A very .similar diverticulum was observed in pig embryos of 12 nun. by Lewis, wbo considered that it was characteristic of the pig and gave rise to the wellmarked pouch attached to the fundus of the adult. Strecker (1908) has recently called attention to a human stomach described by Luschka "in which the transition from the oesophagus to the stomach took place gradually, and beyond this the fundus possessed a conical appendage directed upward and backward, thereby, to a certain extent, resembling the form of a pig's stomach." In addition to the dilated corpus and the conical fundus, the embryonic stomach presents a third subdivision, — the tubular pars pylorica. As seen in three models of the stomach, from embryos of 16, 19, and 19.3 mm. respectively, this pyloric portion extends toward the right and slightly apward, to the pylorus. In every case the position of the pylorus is indicated by a local dilatation of the epithelial tube, such as is shown in Thyng's model from a 13.6 mm. specimen (Fig. 285, A, p. 392). The junction between the corpus and the pars pylorica, measured along the lesser curvature, occurs midway between the pylorus and the cardia. At the place of junction there is an angular bend in the lesser curvature (incisura angularis) and an abrupt change in the diameter of the tube.

Another subdivision of the human stomach is that which Luschka (1863) described as the cardiac antrum. Sometimes in the adult a bulbous enlargement is found at the junction of the oesophagus and stomach, and this is the region of the special form of glands known as cardiac glands. Strecker has studied this area, and concludes that sometimes a "Vormagen" can be recognized, but in other cases it is totally absent.

In the 22.8 mm. specimen, as seen in Fig. 243, such a subdivision is suggested, but the formation of a distinct cardiac antrum in early human embryos has never been demonstrated.

The further development of the oral cavity and its organs, and of the oesophagus, stomach, and intestine, with their folds and glands, will be considered in the following sections.

The embryology of the branchial region and respiratory system will form the concluding part of the chapter.

LITERATURE.

Albinus, B. S.: Academicarum annotationum. Liber 1. See Tab. 1, Fig. 12.

Leidae 1754. von Baer, K. E.: tjeber Entwickelungsgeschichte der Thiere. Zweiter Theil.

S. 1-315. Konigsberg 1837. Beneke: Ein sehr juns:es raensohliches Ei. Monatssehr. f. Geb. u. Ovn. Bd. 19.

S. 771-773. 1904.

332 HUMAN EMBRYOLOGY.

Bonnet, R. : Beitrage zur Enibryologie des Hundes. Anat. Hefte. Bd. 16, S.

233-332. 1901. Lehrbuch der Entwicklungsgeschichte. S. 1-467. Berlin 1907. Branca, A. : Recherches sur la vesieule ombilicale de l'bomme. Ann. de Gyn. et d'Obstet. Annee 35, p. 577-607. 1908. Bremer, J. L. : Description of a 4 mm. Human Embryo. Amer. Joum. of Anat.

Vol. 5, p. 459-480. 1906. Bryce, T. H., and Teacher, J. H. : Contributions to the Study of the Early Development and Imbedding of the Human Ovum. P. 1-93. Glasgow 1908. Dandy, W. E. : A Human Embryo with Seven Pairs of Somites. Amer. Journ. of Anat. Vol. 10, p. 85-108. 1910. Descomps, P. : Anomalie de la torsion intestinale. Journ. de l'Anat. et de la Phys.

Annee 45, p. 616-631. 1909. Diemerbroeck : Anatome corporis humani. P. 303-304. Ultrajecti 1672. Opera omnia, p. 263—265 (as cited by Lobstein). 1687. Elze, C. : Beschreibung eines menschlichen Embryo von zirka 7 mm. grosster Lange. Anat. Hefte. Bd. 35, S. 411^92. 1907. Eternod, A. C. F. : II y a un canal notochordal dans l'embryon humain. Anat Anz. Bd. 16, S. 131-143. 1899. II y a un leeithophore dans l'embryon humain. Bibliogr. anat. Tome 15, p 247-258. 1906. La gastrule dans la serie animale. Bull. soc. Vaud. Sc. nat. 42, p. 197-224 1906. Fabricius ab Aquapendente : De formato foetu. 1600. (Opera omnia. Lugduni Batavorum 1738.) Faraboeuf, M. : Arret devolution de l'intestin. Le progres medical. Annee 13, p. 411-413. 1885. Forssner, H. : Die angeborenen Darm- und Oesophagusatresien. Anat. Hefte.

Abt. I. Bd. 34, S. 1-163. 1907. Frassi, L. : Ueber ein junges menschliches Ei in situ. Arch, f . mikr. Anat.

Bd. 70, S. 492-505. 1907. Giffhorn, H. : Beitrag zur Aetiologie der kongenitalen Atresie des oesophagus mit Oesophagotrachealfistel. Arch. f. path. Anat. Bd. 192, S. 112-126.

1908. Happich, C. : Ueber Oesophagusmissbildungen. S. 1-65. Inaug.-Diss. Marburg 1905. Hensen, V. : Beobachtungen iiber die Befruchtung und Entwicklung des Kanin chens und Meersehweinchens. Zeitschr. f . Anat. u. Entw. S. 353-423. 1875. Herzog, M. : A Contribution to Our Knowledge of the Earliest Known Stages of Placentation and Embryonic Development in Man. Amer. Journ. of Anat.

Vol. 9, p.. 361^00. 1909. His, W. : Anatomie menschlicher Embryonen. I. Embryonen des ersten Monats.

S. 1-184. Leipzig 1880. — II. Gestalt- und Grossenentwicklung bis zum Schluss des 2 Monats. S. 1-104. Leipzig 1882. — III. Zur Geschichte der Organe. S. 1-260. Leipzig 1885. Hoboken, 1ST. : Anatomia secundinae humanae repetita. P. 37 and 217. Ultrajecti 1675. (Cited by Lonnberg.) Ingalls, N. W. : Beschreibung eines menschlichen Embryos von 4.9 mm. Arch, f .

mikr. Anat. Bd. 70, S. 506-576. 1907. Jackson, C. M. : On the Developmental Topography of the Thoracic and Abdominal Viscera. Anat. Record. Vol. 3, p. 361-396. 1909. Jordan, H. E. : The Histology of the Yolk-sac of a 9.2 mm. Human Embryo. Anat.

Anz. Bd. 31, S. 291-303. 1907. A Microscopic Study of the Umbilical Vesicle of a 13 mm. Human Embryo.

Anat. Anz. Bd. 37, S. 12-32, 56-66. 1910.

INTESTINAL TRACT AND RESPIRATORY ORGANS. 333 Jung, P. : Beitrage zur friihesten Ei-Einbettung beiin menschlichen Weibe. Berlin 1908. Keibel, F. : Zur Entwickelungsgeschiehte des menschlichen Urogenitalapparates.

Arch. f. Anat. u. Entw. S. 55-156. 1896. Keibel F., and Elze, C. : Normentafeln zur Entwicklungsgeschichte der Wirbel tiere. Heft 8, S. 1-314. Jena 190S. Keith, A. : Three Cases of Malformation of the Tracheo-cesophageal Septum.

Journ. of Anat. and Phys. Vol. 41, p. 52-55. 1906. Keith, A., and Jones, A. W. : A Note on the Development of the Fundus of the Human Stomach. Proc. of the Anat. Soc. of Great Britain and Ireland, p. 34-3S, in the Journ. of Anat. and Phys. Vol. 36. 1902. Kieser, D. G. : Der Ursprung des Darmkanals aus der vesicula umbilicalis, dar gestellt ini menschlichen Embryo. S. 1-31. Gottingen 1810. Kollmann, J. : Die Entwiekelung der chorda dorsalis bei dem Menschen. Anat.

Anz. Bd. 5, S. 308-321. 1890. vox Kupffer, C. : Die Deutung des Hirnanhanges. Sitz.-Ber. der Gesellseh. f .

Morph. und Phys. in Miinchen. S. 59-87. 1894. Lewis, F. T.: The Gross Anatomy of a 12 mm. Pig. Amer. Journ. of Anat.

Vol. 2, p. 211-225. 1903. Lobstein, J. F. : Essai sur la nutrition du foetus. Strassburg 1802. Ueber die Ernahrung des Fetus. Aus dem Franzosischen ubersetzt von T. F. A. Kestner.

S. 1-214. Halle 1804. Lonnberg, J.: Studien iiber das Nabelblaschen an der Nachgeburt des ausge tragenen Kindes. S. 1-118. Stockholm 1901. Low, A. : Description of a Human Embryo of 13-14 Mesodermic Somites. Journ.

of Anat. and Phys. Vol. 42, p. 237-251. 1908. Luschka, H. : Die Anatomie des Menschen. Bd. 2, S. 179. Tubingen 1863. Mackenzie. F. S. : On a Specimen of the Hind-gut Opening into a Cloacal Chamber in a Child. Journ. of Anat. and Phys. Vol. 40, p. 409-411. 1906. Mall, F. P. : A Human Embryo Twenty-six Days Old. Journ. of Morph. Vol. 5, p. 459^80. 1891. Development of the Human Coelom. Journ. of Morph. Vol. 12, p. 395-453.

1897. Ueber die Entwiekelung des menschlichen Darms und seiner Lage beirn Er wachsenen. Arch. f. Anat. und Entw. Suppl.-Bd., S. 403-434. 1897. Development of the Human Intestine and its Position in the Adult. Bull, of the Johns Hopkins Hosp. Vol. 9, p. 197-208. 1898. Supplementary Note on the Development of the Human Intestine. Anat. Anz.

Bd. 16, S. 492-495. 1899. Mallory, F. B. : Sacro-coccygeal Dimples, Sinuses, and Cysts. Amer. Journ. of the Med. Sci. Vol. 103, p. 263-277. 1892. Marwedel, G. : Ein Fall von persistierendem Urmund beim Menschen. Beitr. z.

klin. Chir. Bd. 29, S. 317-326. 1901. Meckel, J. F. : Handbuch der pathologisehen Anatomie. Bd. 1, S. 553-597.

Leipzig 1812. Bildungsgeschichte des Darmkanals der Saugthiere und namentlich des Menschen. Deutsches Arch. f. d. Phys. Bd. 3, S. 1-84. 1817. Meyer, A. W. : On the Structure of the Human Umbilical Vesicle. Amer. Journ.

of Anat. Vol. 3, p. 155-166. 1904. Minot, C. S. : On the Solid Stage of the Large Intestine in the Chick. Journ. of the Boston Soc. of Med. Sci. Vol. 4, p. 153-164. 1900. Oken and Kieser: Beitrage zur vergleichenden Zoologie, Anatomie und Pliysi ologie. Bamberg und Wiirzburg. Heft 1. 1806. Heft. 2. 1S07. Otis, W. J.: Die Morphogenese und Histogenese des Analhockers. Anat. Hefte.

Bd. 30, S. 201-258. 1905.

334 HUMAN EMBRYOLOGY.

Parsons, F. G. : On the Form of the Caecum. Journ. of Anat. and Phys. Vol. 42, p. 30-39. 1907. Peters, H. : Ueber die Einbettung des menschlichen Eies. S. 1-143. Leipzig und Wien 1S99. Rathke, H. : Ueber die Entstehung der glandula pituitaria. Arch, f . Anat,, Phys.

und wiss. Med. S. 482^85. 1838. Reichel, P. : Die Entwickelung des Dammes und ihre Bedeutung fiir die Entstehung gewisser Missbildungen. Zeitschr. f. Geb. und Gyn. Bd. 14, S. 82 94. 1888. Reid, D. G. : Imperfect Torsion of the Intestinal Loop. Journ. of Anat. and Phys.

Vol. 42, p. 320-325. 1908. Schlater, G. : Zur Frage vom Ursprung der Chordaten nebst einigen Bemerkungen zu den friihesten Stadien der Primaten-Embryogenese. Anat. Anz. Bd. 34, p. 33-48, 65-81. 1909. Seessel, A. : Zur Entwicklungsgeschiehte des Vorderdanns. Arch, f . Anat. und Entw. S. 449-467. 1877. Smith, G. E. : Note on an Abnormal Colon. Journ. of Anat. and Phys. Vol. 38, p. 32-33. 1903. vox Spee, F. : Beobachtungen an einer menschlichen Keimscheibe mit offener Medullarrinne und eanalis neurentericus. Arch. f. Anat. und Entw. S.

159-176. 1889. Neue Beobachtungen iiber sehr friihe Entwickelungsstufen des menschlichen Eies. Arch. f. Anat. und Entw. S. 1-30. 1896. Zur. Demonstration iiber die Entwickelung der Driisen des menschlichen Dotter sacks. Anat. Anz. Bd. 12, S. 76-79. 1896. Stieda, A. : Ueber atresia ani congenita und die damit verbuudenen Missbildungen.

Arch. f. klin. Chir. Bd. 70, S. 555-583. 1903. Strecker, F. : Der Vormagen des Menschen. Arch. f. Anat. und Entw. S. 119 188. 1908. Tarenetzky, A. : Beitrage zur Anatomie des Darmkanals. Mem. de l'Acad. imp.

des sciences de St. Petersbourg. Ser. 7, Tome 28, No. 9, p. 1-55. 1881. Thompson, P. : Description of a Human Embryo of Twenty-three Paired Somites.

Journ. of Anat. and Phys. Vol. 41, p. 159-171. 1907. Toldt, C. : Die Formbildung des menschlichen Blinddarmes und die valvula coli.

Sitz.-Ber. der math.-naturw. Classe der k. Akad. der Wissensch. S. 41-71.

Wien 1894. Velpeau, A. A. L. M. : Embryologie on ovologie humaine. P. 1-104. Paris 1833. Die Embrvologie und Ovolog-ie des Menschen. Uebersetzt von C. Schwabe.

S. 1-84. Ilmenau 1834. Wolff, C. F. : De formatione intestinorum. In Nov. Comment. Acad. Sc. J.

Petrop. 12, 1768, and 13, 1769. Ueber die Bildung des Darmkanals im bebriiteten Hiihnchen. Uebersetzt von J. F. Meckel. S. 1-263. Halle 1812. Zimmermann, K. W. : Ueber Kopfhohlenrudimente beim Menschen. Arch. f.

mikr. Anat. Bd. 53. S. 481-4S4. 1S99.

THE MOUTH AND ITS ORGANS.

335

THE MOUTH AND ITS ORGANS.

By J. PLAYFAIR McMURRICH, Toronto.

The Mouth. — The examination of a human embryo a little over 2 mm. in length will reveal upon the ventral surface immediately in front of the yolk-sac a rounded elevation, the heart, and in front of this a somewhat pentagonal depression, the oral sinus, the anterior boundary of which is formed by the projecting frontal extremity of the brain, while the remaining sides are formed by the maxillary and mandibular processes of the first branchial arch (Fig. 248). The mandibular processes separate the sinus from the anterior surface of the pericardium, and their union in the middle line is

Os H. -I

Fig. 248. — Ventral view of the anterior portion of an embryo of 2.15 mm., from a reconstruction. (His.) Fr., frontal process; H ., heart; Mx., maxillary process; .}fan., mandibular process; Os., oral sinus.

Fig. 249. — Median longitudinal section of a rabbit embryo. (After Keibel.) Ch, chorda; Ekt, ectoderm; En, endoderm; H, heart; Rp, Rathke's pouch; Ph, pharynx; Pm, pharyngeal membrane; v.H, forebrain.

marked by a groove, which forms the posterior angle of the sinus. The remaining angles are paired, the posterior ones being the angles between the maxillary and mandibular processes of either side, while the anterior ones are formed by the ventral ends of grooves which separate the maxillary process of either side from the frontal process. The floor of the sinus is formed by a thin pharyngeal membrane (Fig. 249, Pm), which separates it from the pharyngeal cavity and is lined upon its outer surface by ectoderm and upon its inner surface by endoderm, the two layers, indeed, being in contact throughout the entire extent of the membrane, no mesoderm intervening. Immediately anterior to the membrane a pocket-like evagination of the ectoderm toward the base of the brain occurs, forming what is known as Rathke's pouch

336 (Fig. 249, Rp), destined to form the anterior lobe of the hypophysis cerebri.

The oral sinus, however, does not correspond to the definitive month, which includes also a portion of the embryonic pharynx. Shortly after the stage just described (2.15 mm.), the pharyngeal membrane ruptures and disappears, with the exception of a part of its anterior border, which persists for a time as a transverse ridge upon the roof of the mouth, immediately posterior to Eathke 's pouch. Behind this ridge an evagination of the endoderm toward the base of the brain takes place, forming what is known,

Fig. 250. — Median longitudinal section through the mouth region of an embryo chick of 5 days. (After Seessel.) Ch, chorda; H, Rathke's pouch; G, brain; M, mouth; N, Seessel's pouch; P, pharynx; In, infundibulum; Mn, mandibular process.

from its discoverer (Seessel, 1877), as Seessel's pouch (Fig. 250), a structure whose significance is uncertain.

Seessel's pouch has been described (Nussbamn, 1896) as elongating in embryos of the dog until it came into contact with the hypophyseal downgrowth of the brain, whereupon its lumen disappeared and it eventually fragmented into a number of portions, the uppermost of which remained in connection with the hypophysis and became part of it.

By the disappearance of the pharyngeal membrane the oral sinus becomes continuous with the embryonic pharynx, and the anterior part of the digestive tract is placed in communication with the exterior ; the grooves, however, which separate the medial ends of the maxillary processes from the frontal process still maintain the open communication between the mouth cavity and the nasal pits. Later, the region of the frontal lobe which forms the anterior boundary of the oral sinus becomes a flat or slightly concave surface, whose lateral, thickened margins form the medial walls of the nasal fossae and terminate posteriorly in rounded elevations, the processus globulares (Fig. 251). In embryos of 8 mm. the nasal fossae are still open to the mouth, but later the posterior

THE MOUTH AND ITS ORGANS. 337 border of the lateral wall of each fossa approaches the corresponding processus globularis and eventually unites with it, the fossae thus becoming converted into pits completely shut off from the mouth (Hochstetter, 1892). Still later the medial ends of the maxillary processes also unite with the processus globulares, these gradually approach each other until they meet in the median line, and the anterior boundary of the mouth is completed, consisting of the processus globulares medially, and laterally of the right and left maxillary processes. At this stage the floor of the nasal pits is separated from the mouth cavity merely by a thin membrane, the bucconasal membrane, formed of the nasal epithelium in contact with that of the roof of the mouth cavity, and in embryos of 15. 5 mm. this membrane breaks through and the nasal and oral

np.

mxp. 'ffP^^A m mp.

Fig. 251. — Face of human embryo of 8 mm. Fig. 252. — Roof of the mouth of human em (AfterHis.) mxp., maxillary process; np., nasal pit; bryo showing the formation of the primary labial pg., processus globularis; os., oral sinus. grooves. (After His.) lg., primary labial groove; mp.. maxillary process; pg., processus globularis.

cavities are again in communication, but the communications, the primitive choance, are now behind the maxillary processes.

Several abnormalities may arise in connection with the development of the oral sinus, the most frequent of which is harelip, consisting 1 in a cleft extending through the upper lip slightly lateral to the middle line on either one or both sides and placing the vestibulum oris in communication with nasal pits. This finds its morphological explanation in a failure of the maxillary processes to unite with the processus globulares, whereby the original connection of the nasal pits with the oral sinus is retained. Other abnormalities of less frequent occurrence are dependent upon the imperfect or excessive development of the connection between the maxillary and mandibular processes from which the cheeks are developed, imperfection in this respect producing an abnormal broadening of the rima oris (macrostomia), and excess to its abnormal diminution (microstomia) or even to its suppression (astomia). Inhibition of the development of the mandibular processes may also occur, the two failing to meet in the median line and thus producing a more or less pronounced defect of the lower portion of the face (aprosopia).

The Formation of the Lips and Cheeks. — Shortly after the fusion of the maxillary processes with the processus globulares is effected, a slight groove makes its appearance on the free border of the frontal process and eventually extends laterally on the Vol. II.— 22

338 HUMAN EMBRYOLOGY.

maxillary processes. These are the primary labial grooves (Fig. 252), and are due to a downgrowth into the subjacent mesoderm of the epithelium covering the structures concerned. In embryos about 4 cm. in length a disintegration of the central cells of the downgrowths occurs, the result being a deepening of the grooves to form the secondary labial grooves (Bild, 1902), which separate the lips from the alveolar portions of the various processes (Fig. 254) and themselves form the vestibulum oris. The portion of the upper labial groove which forms on the processus globulares is at first partly interrupted in the median line by an anteroposterior furrow, which corresponds to the line of union of the two processus globulares. In some mammals, as for instance the rabbit, this furrow persists throughout life as a deep median slit in the upper lip, but in man it becomes almost obliterated and is represented in the adult only by the philtrum. The labial groove, however, does not extend as deeply into the tissue of the frontal lobe in the region of the furrow as it does more laterally, and there is consequently formed in the median line a slight fold lying in the sagittal plane and extending between the lip and the alveolar portion of the jaw, the frenulum labii superioris. Its development is associated with the occurrence of the intermaxillary suture, and a similar frenulum labii inferioris is formed opposite the intermandibular suture.

At the angle of the mouth the upper and lower primary labial grooves become continuous, and the epithelial downgrowths are here directed laterally and dorsally. By the disintegration of the central cells of the downgrowths in these regions the buccal cavities are formed, separating the alveolar portions of the maxillary and mandibular processes from the cheeks. The buccal cavities are thus merely lateral extensions of the labial grooves and the structure of the lips and cheeks is identical, both being formed in these early stages of an external epidermal layer and an internal mucous layer, these two meeting at the angles and margins of the mouth, an<J between the two a layer of mesenchyme. It is not until after the beginning of the second month of development that muscular tissue begins to make its appearance in the mesenchyme layer, wandering into it from the region of the second branchial arch and bringing with it branches of the facial nerve (see Vol. I, p. 513).

The condition in which the epidermal and mucous layers meet at the margins of the mouth does not, however, persist, but the mucous layer becomes everted to form the red portions of the lips, its meeting with the epidermal layer being some distance away from the actual rima oris. At birth, as was first pointed out by Luschka (1863), the red portion of the lip consists of two parts, an external pars glabra, whose surface is quite smooth, and a more

THE MOUTH AND ITS ORGANS.

339

proximal pars villosa, covered with numerous minute villosities which contain blood-vessels and may in some cases reach a length of 1 mm. (Fig. 253, A). These villosities, which also occur upon the mucosa of each cheek along a band extending from the angle of the mouth almost to the back of the buccal cavity (Fig. 253, B), make their appearance during the fourth month and are fully developed at the seventh month (Stieda, 1889) ; they disappear during the first weeks of extra-uterine life, but even in the adult the area occupied by the pars villosa can be distinguished by the

A

MMI

B Fia. 253. — A, the lips of a new-born child, showing the villosities and the tubercle; B, the distribution of the villosities upon the vestibulum oris. (After Ramm.) a, opening of the parotid duct.

papillae of its corium being more scattered and more irregular in height than those of the region representing the pars glabra. In addition to this differentiation of the red portions of the lips, there is in the upper lip, from the third month until shortly after birth, a well-marked tubercle, situated in the median line below the philtrum, from which it is separated by a portion of the pars glabra about 1 mm. in breadth (Fig. 253, B). At birth the tubercle is from 5 to 6 mm. broad and has a height of about 4 mm., and it bears along its median line a whitish raphe, continuous below with the frenulum. It is formed entirely from the pars villosa.

340 HUMAN EMBRYOLOGY.

As stated, both the tubercle and the villosities of the pars villosa visually disappear during the first few weeks after birth, but indications of the tubercle are frequently to be seen even in the adult, and occasionally the pars villosa remains distinctly recognizable apart from its histological peculiarities, as a roughened projecting area of the red of the lip, separated from the pars glabra by a distinct groove. Such a condition forms what is termed a double lip.

The Formation of the Palate. — The mouth cavity formed as described in the preceding pages does not, however, correspond with that of the adult, its roof being formed by the base of the skull, so that it includes portions of what will later be the nasal cavity as well as the mouth cavity proper. The separation of these two cavities is brought about by the formation of the palate, which takes place as follows : At the time of the formation of the labial grooves the maxillary processes have a triangular shape in transverse section, one of the angles being directed medially: As development proceeds, this angle enlarges to form a plate-like fold (Fig. 254, p), which projects downward toward the floor of the mouth, between the lateral surface of the tongue and the alveolar

Fig. 254. — Roof of the mouth of a human embryo showing the formation of the palate. (After His.) dr., dental ridge; lg., secondary labial groove; mp., maxillary' process; p., palate; pg., processus globularis; ph., pharynx; Rp., Rathke's pouch.

portion of the maxillary process. In these early stages the long axis of the tongue is directed almost vertically and its dorsal surface is still in contact with the base of the skull, but later, with the enlargement of the arch formed by the two mandibular processes, the tongue sinks down between these processes, its tip at the same time becoming bent downwards. By these changes the tongue is withdrawn from between the two palatal plates, and they gradually bend upward so that their free borders are directed medially instead of toward the floor of the mouth. Exactly how this change is effected has been a matter of discussion. His (1901) believed the withdrawal of the tongue from between the two palatal plates to be due to muscular action, and, with Dursy (1869), supposed the palatal processes were simply bent upward to their final horizontal position. Polzl (1905), however, opposed both these ideas, maintaining that the withdrawal of the tongue was due to changes in the proportions of the parts entering into the formation of the face and that the change of the palatal processes was due to a change

THE MOUTH AND ITS ORGANS.

341

in the direction of their growth and not a mere process of bending upward, basing this latter conclusion on the fact that the palatine nerve, which can be traced into the processes while they still have a vertical position, does not change its course in later stages. Schorr (1908) dissents from this conclusion, and finds that the change is really due to a bending up of the processes, a lively proliferation of the tissue on the oral surface of the processes taking place in the angle between them and the alveolar portion of the jaw (Fig. 255, A and B), so that this angle becomes gradually increased. He finds that the palatine nerve lies lateral to the region in which the proliferation takes place, a fact which explains its retention of its original vertical course ; its branches, however, which are directed medially, approach more nearly a horizontal direction as older embryos are examined.

The palatal processes are entirely confined to the maxillary processes, not extending upon the processus globulares (Fig. 254),

.^- '

3\ .:• :;••.'• ••."•7 /^vV^; : :/^.v £'-:..

^^^\-'l' \.'\. ;'-': •" '••':' : 'io':'.

AW.

Fig. 255. — Frontal section through the palatine process of a pig embryo of 24 mm. (A) and of 25 mm. (B). (After Schorr.) a., art. palatina; x., proliferating mesenchyme over AW., the angle between the palatine and alveolar processes; xx., proliferating mesenchyme over the dental ridge.

and when they have assumed their horizontal position their free, borders are closer together anteriorly than posteriorly, owing to the curvature of the maxillary processes. As the palatal processes increase in breadth in the further course of development, their free borders gradually approach each other and eventually unite, at first anteriorly, the fusion later extending backwards. The mouth cavity proper thus becomes separated from the nasal cavities, these latter now opening posteriorly into the pharynx by the secondary or definitive choanae, which thus owe their existence to the development of the palate. Furthermore the development of the palate brings about the delimitation of the mouth cavity from the definitive pharynx by the formation of the arcus pharyngopalatini, these being the backward prolongation of the palatal processes upon the lateral walls of the pharynx. They pass down

342 ward and backward upon the pharyngeal wall almost in the line of the third branchial arches, the arcus glossopalatini and the sinus tonsillar es being formed respectively by the second branchial arch and the second branchial grooves.

The palatal processes are thns derived entirely from the maxillary processes, and anteriorly, in the median line, there projects backward between them the lower border of the frontal process. With this the palatal processes eventually fnse, but at the meeting point in the median line there remains upon the oral surface a depression known as the incisive fossa. Furthermore, the fusion is not perfectly complete; the epithelia on both surfaces become perfectly continuous, but the intervening mesenchyme does not, a strip of epithelium extending through it from the fossa incisiva, upwards, backwards, and somewhat laterally, in the line of fusion

Fig. 256. — Palate of (A) a human embryo of 5.5 cm. and of (B) a new-born child showing the palata ridges. (After Gegenbaur.) ±a, alveolar process; p., incisive papilla; pro, velum palati; -., median raphe: u, uvula.

of each palatal process with the corresponding processus globularis. The cells of the epithelial strip break down so that a lumen is formed in it, placing the mouth and nasal cavities again in communication anteriorly by what are known as the incisive canals (canals of Stenson). These usually become obliterated during further development, but occasionally they persist until adult life. Toward the close of the second month of development ossification begins to extend into the mesenchyme of the palatal processes from the alveolar processes, and the hard palate becomes readily distinguishable from the velum palati ; but for a considerable time, up to at least about the middle of the third month, the palate continues to show a distinct median raphe (Fig. 256, A), this indicating the line of fusion of the two palatal processes and terminating anteriorly in the incisive papilla. The uvula also remains distinctly notched for a considerable period (Fig. 256, A), an indication that it is really a bilateral structure and not, as it appears

THE MOUTH AND ITS ORGANS. 343 in the adult, a median unpaired organ. On either side of the median raphe on the hard palate from five to seven almost transverse ridges appear (Fig. 256, A), which represent the palatal ridges occurring in the lower mammals.* These ridges later develop minute fringe-like processes on their posterior borders, but at the same time they begin to undergo degeneration, the posterior ones breaking up into rows of papillae, while the regularity of the anterior ones is disturbed by the formation of cross branches. At birth (Fig. 256, B) the fringe-like processes have almost disappeared, as has also the posterior ridge, and the anterior ones have become very irregular. In this condition they persist throughout childhood, but in adult life they become still more reduced and may eventually disappear altogether.

Inhibition of the development of the palatal processes occasionally occurs, resulting in a failure of their fusion in the middle line, the defect constituting what is known as cleft palate. This may vary considerably in its extent, being limited in some cases to the velum palati, in others appearing as a perforation of the palate in the median line, and in others again involving the hard palate as well as the velum palati. In these last cases the cleft cannot continue forward in the median line beyond the anterior extremities of the maxillary processes, since it there meets with the unpaired frontal process, but it may be continued along the line of union of the maxillary and frontal processes on one or both sides, in which case it will usually be associated with harelip.

The Tongue. — The tongue in the amniote vertebrates consists of two portions, quite distinct in their origin and represented in the adult by the body of the tongue anteriorly and the root posteriorly, the two being separated by a V-shaped groove, the sulcus terminalis. An examination of the floor of the mouth of an embryo of 5 mm. (Fig. 257) shows a rhomboidal depression in the median line between the ventral ends of the first and second branchial arches, and from this there projects dorsally a rounded tubercle, the tuberculum impar of His. Immediately behind this is a deep evagination of the epithelium, which is the median thyreoid evagination, and behind this again is a transverse elevation formed by the ventral ends of the second and third branchial arches, the copula. Since the apex of the V-shaped sulcus terminalis corresponds with the foramen caecum, and since this is the remains of the median thyreoid evagination, it would seem that the anterior portion of the tongue is formed from the region between the first and second branchial arches. It was held by His that it was formed by the enlargement of the tuberculum impar, although as early as 1869 Dursy had described the body of the tongue as having a paired origin, a condition more recently described as occurring in the pig by Born and in man by Kallius and by Hammar (1901). In embryos of 7.5 mm. a swelling appears in the anterior part of the mouth on each side of the median line (Fig. 258, t), and the

344 two increase in size until they occupy the greater part of the interval between the lower ends of the first and second branchial arches, becoming separated from the former by an alveoloUngual groove. The two swellings eventually meet in the median line to form the main mass of the body of the tongue, the amount to which the tuberculum impar participates being probably small, Hammar, indeed, maintaining that it is merely a transitory structure and takes no part at all in the formation of the tongue.

Immediately behind the median thyreoid evagination the lower ends of the second and third branchial arches join to form a median elevation, the copula (Fig. 258, cop), on the floor of the mouth, and from the anterior portion of this, together with the neighboring portions of the second arch, the root of the tongue develops. His

Mn.

m "

'•jBr~~ u

Ti -z*r-r

ity ni '

F — - IV m I- — y cop.

' H-fe

\

Fig. 257. — Floor of the mouth and pharynx of an embryo of 5 mm. (After His.) F., furcula; Mn., mandibular arch; Ti., tuberculum impar; II-V, the branchial arches. The thyreoid evagination is indicated by a dotted line.

Fig. 258. — Floor of the mouth and pharynx of an embryo of 7.5 mm. (From a reconstruction.) cop., copula; F., furcula; t., anlage of the body of the tongue; Ti., tuberculum impar; I-III , branchial arches.

believed that the third arch also took part, but recent observers either limit extensively the participation of this arch or exclude it altogether.

It must be remembered, however, that the tongue is a complex of mucous membrane and muscle tissue, and the statements given above indicate only the regions from which the mucosa is derived in the earlier stages of development. The origin of the musculature has not yet been thoroughly studied in the human embryo, but the fact that it is for the most part innervated by the hypoglossus is indication of its derivation from postbranchial myotomes. When first identifiable the various muscles have already reached the branchial region, so that His assigned the hyoglossus to the third arch, but it is probable that it had already undergone a considerable forward migration before it became recognizable. Indeed, even after it is distinctly differentiated its distribution in the tongue is materially extended, and there is reason for supposing that practically the whole of the tongue musculature under

THE MOUTH AND ITS ORGANS.

345

goes an extensive migration from the postbranchial region, pushing forward beneath the mucons membrane of the floor of the pharynx and mouth until it occupies the elevations on the floor of the mouth mentioned above. During this migration it invades in succession the territories of the various branchial arches, and consequently the mucosa of the fully developed tongue is supplied by the nerves corresponding to these arches, that is to say, by the trigeminus and facialis anteriorly and by the glossopharyngeus and vagus posteriorly (Fig. 258).

If this view of the development of the tongue musculature be correct, it would seem that a more extensive area of the oropharyngeal mucosa is involved in the formation of the tongue than that associated with the elevations usually regarded as its origins. These undoubtedly represent the first indications of the organ, but with the later elaboration and increase in bulk of its musculature other portions of the mucosa become involved, the complicated innervation being thus produced. It is hardly accurate, therefore, to regard the mucosa of the tongue as being the product of the first and second branchial arches alone, even though its first indications are confined to them. Phylogenetically the copular portion of the tongue seems to be the older, being the only part present in the fishes. In the amphibia also it constitutes the main mass of the tongue, but anterior to it a glandular fold of the mucosa is formed on the floor of the mouth immediately behind the mandibular arch, and in later larval stages this unites with the copular portion representing the body of the amniote tongue. This in its origin is, therefore, essentially a glandular portion of the tongue, with which muscle-fibres, separated from the geniohyoideus, become associated ; but in the higher f orms its glandular character becomes subordinated and its muscle-fibres increase in number to form the genioglossus. The hyoglossus is probably a derivative of the geniohyoideus, and the intrinsic musculature is apparently derived from these two primary muscles, the transversus linguae and the vertical fibres from the genioglossus (Fig. 261, G-Grl) and the longitudinalis from the hyoglossus (Kallius).

Fig. 259. — Diagram of the distribution of the sensory nerves of the tongue. (After Zander.) The area supplied by the fifth (and seventh) nerve is indicated by the transverse lines, that supplied by the ninth nerve by the oblique lines, and that supplied by the tenth nerve by the small circles.

346 HUMAN EMBRYOLOGY.

The fungiform papillae have become evident in embryos of 50 mm. as elevations of the submucous tissue which project upward into the epithelium, this frequently undergoing proliferation over the elevations so as to produce finger-like papillae, which are, however, merely temporary structures. The filiform papillae are at first indistinguishable from the fungiform, only becoming recognizable in embryos of 64 mm. In embryos of 100 mm. tastebuds begin to make their appearance upon the fungiform papillae, and somewhat later, about the beginning of the fifth month, both varieties begin to project above the general surface of the tongue, owing to the degeneration of the superficial layers of the epithelium in the intervals between the papillae. The development of the taste-buds on the fungiformes continues during the later fetal months, and at birth, as well as for some time after, the buds are greatly in excess of the number present in the adult, their subse A B C Fig. 260. — Diagrams illustrating the development of the vallate papillae. (After Graberg.) quent reduction in number being associated with the, change in the nature of the food of the child occurring at the time of weaning (Stahr).

The vallate papillae are represented in embryos of 90 mm. by two epithelial ridges, situated toward the posterior portion of the tongue and inclined toward one another in a V-shaped manner, the apex of the V practically corresponding with the mouth of the median thyreoid evagination. From these ridges downgrowths of the epithelium take place into the subjacent submucosa, each downgrowth having the form of a hollow truncated cone whose base is continuous with the mucosa and whose centre is occupied by a portion of the submucosa, which thus becomes surrounded by a solid wall of epithelial cells (Fig. 260, A). During the fourth month lateral outgrowths take place from the deeper edges of the wall, and at about the same time clefts begin to appear in its substance (Fig. 260, B) ; these increase in size and eventually open to the surface a trench, lined by epithelium, thus surrounding a papilla (Fig. 260, C). The lateral outgrowths from the deeper edges of the downgrowths also become hollow by the degeneration of their central cells and form the glands of Ebner, and during the development taste-buds differentiate from the basal layers of the epithelium. These make their appearance quite early in the

THE MOUTH AND ITS ORGANS. 347 development of the papillae, being recognizable even in a fetus of three months (Graberg), and increase in number as development proceeds, being formed not only on the sides of the papillae but also on their free horizontal surfaces, those in the latter situations, however, for the most part disappearing after birth. The development of the individual papillae is subject to considerable variation both in number and time, and, as a rule, is not completed until after birth. The foliate papillae appear much later than the other varieties, being indistinguishable in embryos of four and a half and five months although quite distinct in those of seven months (Tuckerman).

Anomalies of the tongue which may be assigned to interferences with the normal processes of development are not of frequent occurrence. A condition of diglossia has, however, been described, in which the anterior portion of the organ is divided throughout the greater or lesser portion of its extent, producing what might be spoken of as a forked tongue. This is especially interesting as indicating the development of the body of the tongue mainly from two primary anlagen, rather than from a single median structure such as the tuberculum impar.

The Salivary Glands. — Of the glands of the mouth the most important are the salivary glands, — that is to say, the parotid, the submaxillary, the sublingual, and the alveololingual. The first three of these are individual glands, formed from a single epithelial outgrowth and having in the adult condition a single duct opening into the mouth cavity, that for the sublingual being known in anatomy as the ductus sublingualis major (duct of Bartholin). The alveololingual glands, on the contrary, consist of a group of glands each of which is provided with its own duct, and they are generally associated with the sublingual gland proper as the glandula sublingualis, a structure which, however, is not comparable morphologically to one of the other salivary glands, but rather to a group of them.

The first of the salivary glands to appear in the embryo is the parotid, which has been detected in an embryo of 8 mm. as a furrow in the floor of the alveolobuccal groove in the neighborhood of the angle of the mouth (Hammar). At first quite small, the furrow gradually elongates, and before the embryo has reached a length of 17 mm. it separates from the epithelium and forms a tubular structure lying beneath the epithelium of the alveolobuccal groove and opening into the mouth cavity at a point which corresponds with the anterior end of the original furrow. Mesenchymatous tissue gradually forces its way between the tube and the alveolobuccal epithelium, and the tube, increasing in length, pushes its way back over the masseter muscle to the neighborhood of the external ear. As it comes into this region the tube or duct, as it may be called, begins to branch at its posterior extremity, the

348 HUMAN EMBRYOLOGY.

branches being at first solid outgrowths from the wall of the duct, and, as these increase in number and size and become surrounded by a mesenchymatous capsule, the gland assumes the position and general form of the adult structure. The accessory parotid gland arises as an outgrowth from the duct as it passes over the masseter muscle, and its further development is similar to that of the main gland.

The account given above of the origin of the gland is based on the recent observations of Hammar, and these differ in some respects from those of earlier investigators (His, Chievitz), who first perceived the gland in embryos of seven and a half or eight weeks' development and describe it as formed from a solid outgrowth from the alveolobuccal epithelium. It is worthy of note that the gland is primarily lateral to the internal carotid artery, the posterior facial vein, and the facial nerve, and although these structures may eventually become more or less surrounded by its alveoli, yet their position is always medial to the principal ducts of the gland.

In embryos of the twelfth week Chievitz observed a branch arising from the parotid duct just where it crossed the anterior border of the masseter muscle and passing deeply to come into relation with the internal pterygoid, where it ended blindly in a small enlargement. The same structure was also observed in an embryo of ten weeks, but in this case it had lost its connection with the parotid duct, and the same condition may be observed in embryos of nine weeks or even earlier. What its significance may be is at present uncertain, but the possibility of its being the origin of cystic growths in the cheek is perhaps worthy of mention.

The submaxillary gland appears in embryos of the sixth week (13.2 mm.) as a ridge-like thickening of the epithelium of the alveololingual groove, the anterior end of the thickening lying some distance behind the frenulum linguae. The ridge later separates from the epithelium from behind forward, and the solid cord so formed grows downward and backward toward the submaxillary region, its enlarged terminal portion branching to form the gland proper, while the remainder of the ridge becomes the duct (Fig. 261, SMx) and gradually shifts its anterior connection with the epithelium forward until it reaches the adult position. During this development the duct acquires its lumen, although the buds which form the alveoli of the gland remain solid until a much later period.

The sublingual and alveololingual glands develop in a manner very similar to the submaxillary. They appear as solid downgrowths of the epithelium of the alveololingual groove (Fig. 261, SL), the sublingual beginning to form at about the eighth week immediately lateral to the anterior termination of the submaxillary duct, and the alveololingual s somewhat later and posterior to the larger sublinguals. Frequently, however, the sublinguals do not appear (Chievitz), the so-called sublingual gland of the adult then being formed entirely by the alveololinguals, and these also seem to be variable in number, Chievitz finding from 11 to 13 on the two sides in an embrvo of the twelfth week, while in one of 40 mm.

THE MOUTH AND ITS ORGANS.

349

I have found 11 on the left side and 9 on the right, the left side also possessing a sublingual gland although it was absent on the right side.

The histogenetic development of the salivary glands is not completed until some time after birth, probably not until after the child is weaned. The canalization of the solid anlagen of the glands proceeds peripherally, and so long as the terminal branches remain solid they have the power of producing additional buds. When, however, the lumen is formed in a bud and it becomes an alveolus, its power of budding is lost, and the further increase in the size of the gland is due to the development of the investing connective tissue and to an increase in the size of the alveoli already present. The specific characters of the cells also become

I.Al.

— Man.

Fig. 261. — Transverse section of the lower jaw and tongue of an embryo of about 20 mm. D, digastricus; G.GL, genioglossus; GH., geniohyoideus; I.Al., inferior alveolar nerve; Man., mandibular ossification; Mk., Meckel's cartilage; My., mylohyoideus; SL., sublingual gland; S.Mx., submaxillary duct; T., tongue.

evident only after the canalization of the alveoli, mucin cells becoming distinguishable in the alveololingual glands of embryos of the 16th week and acinus cells in the parotids of those of five months. The demilune cells of Gianuzzi are developed from the cells lining the alveoli and are only secondarily overgrown by the mucous cells.

Anomalies of the salivary glands are of rather infrequent occurrence; a case of inhibition of the growth of the parotid has, however, been described, the gland being entirely confined to the buccal region, no trace of it occurring behind the masseter muscle.

The Teeth. — At about the time when the primary labial grooves are formed — that is to say, in embryos of about 11 mm. —

350 a ridge-like thickening of the epithelium appears upon what will be the alveolar portions of the maxillary and mandibular processes and also extends upon the portion of the upper jaw formed from the processus globular es (Fig. 253). These ridges are parallel with and immediately posterior (medial) to the labial grooves, and in later stages they penetrate more deeply into the mesenchyme in a somewhat oblique direction, so that they seem almost to be derivatives of the epithelium of the labial grooves (Fig. 262, A). From the deeper surface of each of these dental ridges a series of papillae project more deeply into the mesenchyme, and in embryos of 40 mm. the deeper surface of each papilla has become concave and the concavity is occupied by a mass of condensed mesenchyme, the mesenchyme papilla, the epithelial and mesenchyme papillae together constituting a dental papilla (Fig. 262, B). The number of papillae so formed is normally ten in each jaw, one

LF.

LFL

ZL.

Fig. 262. — Section through the dental ridge of the lower jaw of embryos of (A) 17 ram. and (B) 40 mm. (After Rose.) LF. and LFL., labial groove; Pp., dental papilla; UK., lower jaw; uL., lower lip; ZL. dental ridge.

corresponding to each tooth of the milk dentition, and as they proceed in their development they gradually separate from the dental ridge, which, on its part, becomes prolonged backward in the mesenchyme beyond the point at which the papilla for the second molar of the milk dentition is formed. Three additional papillae appear on each side on these prolongations of the ridges, representing the permanent molars, that for the second molar forming, however, only in the sixth week after birth and that for the third molar not until the fifth year. As the papillae for the milk dentition separate from the dental ridges these begin to degenerate, becoming converted into a network of epithelial trabeculae (Fig. 263), except along their lingual border, where a continuous cord persists ; from this a second series of papillae arises, from which the permanent teeth, which replace the milk dentition, are formed. As the papillae for these teeth separate from the cord, it finally undergoes degeneration and, with the other remains of the dental ridges, eventually disappears, except for fragments

THE MOUTH AND ITS ORGANS.

id

of either the cord or the trabeculae which may persist imbedded the surrounding mesenchyme and are known as epithelial pearls.

In each dental papilla two different structures are concerned, a mesenchyme papilla, from which the tooth pulp and dentine are formed, and an epithelial papilla, which invests the mesenchyme papilla like a cap and gives origin to the enamel, whence it is spoken of, in its later stages, as the enamel-organ. Nerves and blood-vessels make their way into the mesenchyme papilla, and certain of its cells arrange themselves in a single continuous layer over its surface and assume a columnar form, constituting the odontoblasts (Fig. 264, Od) by which the dentine is manufactured. This material appears to be formed by the transformation of a

Fig. 263. — Reconstruction of the dental ridge and a papilla of an embryo of 30 cm. (After Rose.) D, dentine; S, enamel ; Zl, cord-like remnant of the dental ridge which gives rise to the papillae of the permanent teeth ; Ms, oral mucous membrane; ET, epithelial trabeculae representing the original dental ridge; Sp, enamel pulp.

portion of the protoplasm of the odontoblasts into a gelatinous substance, which later becomes fibrillar and in which lime salts are eventually deposited. These deposits are at first in the form of spherical concretions, but later the interstices become filled up, numerous minute dentinal tubules, branching at their outer ends, traversing the matrix from within outwards and containing slender prolongations of the unaltered protoplasm of the odontoblasts, whose growth during the active period of dentine formation compensates for the loss of substance entailed in the formation of the matrix.

This account of the formation of the dentine follows essentially the results of von Ebner. Recently von Korff has maintained that the dentine has a double origin, the first indication of it being bundles of connective-tissue fibrils, which are formed by the pulp cells of the mesenchyme papilla and extend outward between the odontoblasts. These latter structures produce the interfibrillar substance of the dentine and secrete the lime salts which are deposited in this. While

XX UlVX^ilN XH±YXX>X\ X yj±J\J\J X

4 results of von Korff bear the stamp of probability, it seems advisable to »u their further confirmation before adopting them in their entirety.

The dentine is formed from the outer ends of the odontoblasts and therefore lies immediately internal to the products of the enamel-organ. This differentiates (Fig. 264) into an outer epithelial layer consisting of more or less flattened cells, beneath which is a mass of tissue composed of stellate cells widely separated by the distention of the intercellular spaces, so that the tissue has a spongy appearance. This is the enamel-pulp, and internal to it is a single layer of large columnar cells, the ameloblasts, which are the active elements in the production of the enamel. The inner cells of the enamel-pulp are usually more closely aggregated than the rest, and form an epithelial-like layer external to the ameloblasts, which is termed the intermediate layer. As in the case of the odontoblasts the ameloblasts persist throughout the entire formation of the enamel, each cell producing one of the enamel-prisms. The first indication of the enamel is a delicate cuticular membrane covering the inner extremities of the ameloblasts, and to this succeeds the formation of a series of homogeneous columns, one corresponding to each ameloblast. Later the homogeneous material differentiates into bundles of fibrils, the enamel processes {processes of Tomes), imbedded in a homogeneous matrix, and, finally, the calcification of the columns ensues, this process being, according to some observers a calcification of the enamel processes, while others hold it to be a deposit of lime salts in the matrix surrounding the fibres. Even before the formation of the enamel is completed the degeneration of the enamel-organ begins, blood capillaries making their way through the outer epithelial layer into the enamel-pulp, which gradually becomes indistinguishable from the surrounding mesenchyme, and finally the layer of ameloblasts

Fig. 2G4. — Section through a developing molar tooth of Didelphys. (After Rose.) C, connective tissue; D, calcified, and D\, uncalcified dentine; K, wall of dental alveolus; Od, odontoblasts; P, pulp 'ells; S, enamel; SEa, outer epithelial layer of the enamelorgan; S.E.i, ameloblasts; S.P., enamel-pulp; Str.i, intermediate layer of enamel-organ; T, Tomes's processes of the ameloblasts.

THK -MOl TJ1 AM) ITS OHliANS.

.;..:;

breaks up into fragments, some of which are usually to be found around the roots of the teeth even in the adult.

The cement which covers the dentine of the roots of the teeth is formed from the surrounding mesenchyme by a process identical with that by which membrane bone is formed.

As the teeth increase in size they gradually approach the surface of the alveolar processes and eventually break through the gum, not, however, at a point in the line of the original downgrowth of the dental ridge, but posterior to this, the first teeth to erupt being usually the median incisors, which make their appearance during the last half of the first year after birth. The remain

mp. II -'

Fig. 265. — Skull of a 5-year-old child showing the milk and permanent dentitions. (After Sobotta.) cp., permanent canine; ip., permanent incisor; mm., milk molars; mp.I, first permanent molar; mp.II, second permanent molar; pmp., permanent premolar.

ing teeth of the milk dentition appear in succession up to about the middle of the third year. Shortly after their appearance, however, these teeth begin to undergo absorption, this being associated with the continued growth of the permanent teeth (Fig. 265). The milk-teeth lose their shiny appearance, their pulp dies, and an absorption of their roots occurs, beginning at the side in contact with the corresponding permanent tooth and being associated with the appearance of osteoclasts similar to those producing the absorption of ordinary bone. Their alveoli also undergo absorption, and finally their attachments become so feeble that the teeth are readily pulled or broken away.

The exact period of eruption of the various teeth varies con Vol. TT.— 23

354 HUMAN EMBRYOLOGY.

siderably according to racial, climatic, and nutritive conditions, but the usual sequence is somewhat as follows : THE MILK DENTITION".

Median incisors 6th to 8th month Lateral incisors 8th to 12th month First molars 12th to 16th month Canines 17th to 20th month Second molars 20th to 24th month THE PERMANENT DENTITION.

First molars 7th year Median incisors 8th year Lateral incisors 9th year First premolars 10th year Second premolars 11th year Ca 11 ™© 8 \ 13th to 14th year Second molars J Third molars 17th to 40th year Anomalies are not infrequent in connection with the development of the teeth, leading sometimes to a diminution and sometimes to an excess of the normal number. A case of total congenital absence of the teeth has been observed, and also cases in which there had apparently been a defect of the enamel-organ leading to the development of rudimentary teeth lacking enamel. The fusion of two neighboring tooth germs may also occur, as well as the reverse, that is to say, a splitting of a tooth germ so that an accessory tooth or indeed a number of small teeth may be present in the place of one of the normal teeth. More remarkable are the instances of heterotopy which occur, due apparently to the existence of aberrant processes of the dental ridges extending into regions beyond the alveolar processes. Thus, incisor teeth have been observed to form in the nasal cavity, in the maxillary sinus, and even in the orbit, and molars have developed upon the hard palate. Numerous cases of supernumerary dentitions have also been recorded, one or more teeth being replaced more than once. Many of these cases have been supposed to be really the belated development of the normal permanent tooth, but some do not seem referable to this condition, and must be regarded as due to the persistence in an active condition of portions of the dental ridge or to the awakening to functional activity of some of the epithelial pearls which are remnants of it.

LITERATURE.

Born, G. : Ueber die Derivate der embryonalen Schlundbogen und Schlundspalten bei Saugethieren. Arch, fur mikr. Anat. Bd. 22. 1883. Chievitz, J. H. : Beitrage zur Entwicklungsgeschichte der Speicheldriisen. Arch.

fur Anat. und Phys. Anat. Abth. 1885. Dursy, E. : Zur Entwicklungsgeschichte des Kopfes des Menschen und der hoheren Wirbelthiere. Tubingen 1869. Gegenbaur, C. : Die Gaumenf alten des Menschen. Morph. Jahrb. Bd. 4. 1878.

Zur Phylogenese der Zunge. Morph. Jahrb. Bd. 21. 1894. Graberg, J. : Beitrage zur Genese des Geschmacksknospen des Menschen. Morph.

Arbeiten. Bd. 8. 1898. Hammar, J. A.: Notiz iiber die Entwieklung der Zunge und der Mundspeichel driisen beim Menschen. Anat. Anzeiger. Bd. 19. 1901. His, W. : Anatomie menschlicher Embryonen. Heft 3. 1885.

Die Entwieklung der menschlichen und tierischen Physiognomien. Arch, fur Anat. und Phys. Anat. Abth. 1892.

THE MOUTH AND ITS ORGANS. 355 Hochstetter, F. : Ueber die Bildung der primitiven Choanen beim Menscben.

Verbandl. Anat. Gesellscb. 1892. Kallius, E.: Beitrage zur Entwicklung der Zunge. Verbandl. Anat. Gesellscb.

1901. Keibel, F. : Zur Entwicklungsgescbicbte der Chorda bei Saugern (Meerscbweincben und Kanincben). Arcb. fiir Anat. und Phys. Anat. Abtb. 1889. Zur Entwicklungsgescbicbte und vergleicbenden Anatomie der Nase und des oberen Mundrandes (Oberlippe) bei Vertebraten. Anat. Anzeiger. Bd. 8.

1893. von Korpf, K. : Die Analogie in der Entwicklung der Knochen- und Zahnbein grundsubstanz der Saugetiere, nebst kritiscben Bemerkungen iiber die Osteo blasten- und Odontoblastentbeorie. Arcb. fiir mikr. Anat. Bd. 69. 1907. Neustatter, 0. : Ueber den Lippensaum beim Menscben, seinen Bau, seine Entwicklung und seine Bedeutung. Jenaische Zeitscbr. Bd. 29. 1895. Nusbaum, J. : Einige neue Tbatsachen zur Entwicklungsgescbicbte der Hypophysis cerebri bei Saugethieren. Anat. Anzeiger. Bd. 12. 1896. Polzl, A.: Zur Entwicklungsgeschichte des menschlichen Gaumens. Anat. Hefte.

Bd. 27. 1905. Ramm, M. : Ueber die Zotten der Mundlippen und der "Wangenschleimhaut beim Neugeborenen. Anat. Hefte. Bd. 29, 1905. Rose, C. : Ueber die Entwicklung der Zahne des Menschen. Arch, fiir mikr. Anat.

Bd. 38. 1891. Ueber die erste Anlage der Zahnleiste beim Menscben. Anat. Anzeiger. Bd.

8. 1893. Schorr, G. : Zur Entwicklungsgeschichte des secundaren Gaumens bei einigen Saugethieren und beim Menschen. Anat. Hefte. Bd. 36 (Heft 108). 1908. Seessel, A.: Zur Entwicklungsgeschichte des Vorderdarms. Arch, fiir Anat. und Phys. Anat. Abth. 1877. Stahr, H. : Ueber die papillae f ungif ormes der Kinderzunge und ihre Bedeutung als Geschmacksorgan. Zeitschr. fiir Morph. und Anthropol. Bd. 4. 1901. Stieda, A. : Ueber das Tuberculum labii superioris und die Zotten der Lippen scbleimhaut des Neugeborenen. Anat. Hefte. Bd. 13. 1899. Tuckerman, F. : On the Development of the Taste Organs of Man. Journ. of Anat. and Phys. Vol. 23. 1889. Vol. 29. 1890.

THE DEVELOPMENT OF THE OESOPHAGUS.

By FREDERIC T. LEWIS.

Early Development. — The oesophagus in the 4.0 mm. Bremer embryo (Fig. 266) is an epithelial tube which is greatly flattened laterally. Its lumen is a well-defined dorso-ventral cleft. In most places the epithelium shows two rows of somewhat elongated nuclei, and the row next the lumen exhibits numerous mitotic figures. In the upper part of the oesophagus, at the place where its lateral walls meet dorsally, the epithelium has only one row of nuclei, but the ventral border is expanded and has three or four rows. This thickened portion, however, belongs with the respiratory tract, which has not yet been separated from the oesophagus. The mesenchyma around the oesophagus is an undifferentiated layer with crowded nuclei and many mitotic figures. Below the

356 HUMAN EMBRYOLOGY.

lung-bud, on either side, the mesenchyma is closely connected with the adjacent ccelomic epithelium, from which it is being produced. There is no histological demarcation between the oesophagus and pharynx above or the oesophagus and stomach below.

Sept.

Pul

Sul.

Fig. 266. — Wax model from Bremer's 4 mm. embryo, showing tbe "lung-bud," Pul., and the adjacent part of the oesophagus. ' 175 diam. Sept., tracheo-cesophageal septum; Sul., lateral oesophageal groove.

The Epithelial Tube. — In older embryos, as Forssner recorded ^1907), the oesophagus becomes "not only relatively but absolutely smaller in cross section, and the lumen is reduced to a fraction of its former size. " In an embryo of 7.5 mm. the most slender portion of the oesophagus has a cross section about one-third as large as in the 4 mm. embryo, and a lumen one-twentieth as large, yet the length of the oesophagus has increased from less than 0.5 mm. to 1.5 mm. The lower portion of the oesophagus remains flattened laterally, but the upper part has become a round tube which is entirely separate from the trachea. It tapers from the larynx downward, and the lumen becomes minute. The epithelium has 3-4 rows of nuclei above, and is quite like the lining of the pharynx. In the narrowest part of the oesophagus there are but two rows. Mitotic figures are seen almost exclusively in the layer of cells bordering upon the lumen.

In four embryos measuring from 8.4 to 16 mm. the epithelial tube of the oesophagus is shaped as follows : At its laryngeal end it is crescentic, with the concavity of the crescent directed toward the trachea. On its way to the stomach it first becomes round and then transversely elliptical. Near the level of the bifurcation of the trachea it is again round and finally it becomes dorso-ventrally elliptical. In this shape it merges with the stomach. In all of these specimens the lumen is pervious throughout, but it contains a reticular coagulum. The epithelium shows from two to four rows of nuclei. Schridde (1907 and 1908) states that in embryos meas

DEVELOPMENT OF THE (ESOPHAGUS. 357 Tiring from 4 to 35 nmi. the oesophageal epithelium has only two layers, although in thick sections (7-10 /*) the number may appear greater. In a 13 mm. embryo he finds that "in a striking manner, the nuclei of both layers are in the upper ends of the cells, toward the lumen," and this condition was figured by Schaffer in 1904. Jahrmaerker (1906) has described two 8 mm. embryos in which the oesophageal epithelium is composed of two layers of tall cells, with a narrow zone free from nuclei along both the basal and free borders. He finds similar conditions in embryos of 14 and 16 mm. A basal zone free from nuclei is shown in Fig. 267, A, but in this section the nuclei are crowded toward the free surface, forming a darkly staining band. Jahrmaerker, in describing embryos of 17 and 18 mm., states that the superficial layer of the epithelium is more deeply stained than the basal layer, but he does not attribute this to a crowding of the nuclei. At 10 mm., as in the smaller embryos, mitotic figures were frequent in the inner layer. In the older specimens no figures were preserved.

Vacuoles in the Epithelium. — In human embryos of about 20 mm., large vacuoles occur in the oesophageal epithelium, so that in cross section the oesophagus may appear to have two or three lumina. This was noted by O. Schultze in 1897. Kreuter (1905) studied the vacuoles, and concluded that they were associated with an epithelial proliferation which led to a temporary occlusion of the oesophagus. He had previously studied the solid oesophagus of various vertebrates, following Balfour and others. Forssner (1907) showed by means of a model that the main lumen of the human oesophagus is not obliterated.

In an embryo of 22.7 mm. Forssner found " an uninterrupted open central lumen with a mass of cavities on either side of it, some of which communicate with the mam lumen and others do not; some of them are much smaller and others considerably larger than the lumen itself. These formations may be found scattered along the entire oesophagus (22.7 mm.) ; sometimes only below (20 mm.), sometimes only above (30.5 mm.). In the 31 mm. embryo the wall of the oesophagus, as compared with the main lumen, is considerably thinner than before. The epithelium is several layered; it shows none of the cavity formations just described and the lumen is everywhere undivided. That this process in the esophagus has the result of enlarging the lumen appears probable." Schridde (1908) likewise failed to find an occlusion at any stage.

He denies the pi-esence of vacuoles in the following conclusion : " All these facts go to show that vacuoles never occur in the oesophagus. On the contrary, epithelial bridges are clearly present, having arisen by epithelial proliferation in circumscribed places." The structures in cmestion arc shown in Fig. 2(57, B. Here the central lumen of the oesophagus is bounded by a compact dark zone of nuclei, thus differing from the accessory cavities. C and

358 D in Fig. 267 represent successive sections from a 22.8 mm. embryo. In C there are two accessory cavities with compact linings, and the one on the left communicates with the general lumen in D. The oesophagus of these embryos has been modelled by F. P. Johnson in a study of the development of the intestinal mucous membrane. 11 His work affords an independent confirmation of Forssner's conclusions, and shows that Schridde was in error in denying the presence of vacuoles.

Fig. 267. — Transverse sections of the epithelial tube of the oesophagus. X 160 diani. A, embryo of 16 mm. (Harvard Collection, Series 1322). B, 19 mm. (Harvard Collection, Series 819). C and D, successive sections from an embryo of 22.8 mm. (Harvard Collection, Series 871).

Vacuoles have been found in embryos of 14.5 mm. (Keibel and Elze), but they are sometimes absent in those of 18.5 mm. (Forssner). They acquire a maximum development in specimens of about 20 mm. In a 30 mm. embryo there are occasional vacuoles in the upper part of the oesophagus. In a 42 mm. specimen some small intercellular cavities are found, but there are no characteristic vacuoles.

In discussing the origin of the vacuoles, Kreuter has said correctly that "we have no ground for believing that there is a degeneration of cells, but must conclude that it involves throughout only vital processes. ... A degeneration of cells followed by resorption is nowhere demonstrable." Forssner suggests that

11 The work of Mr. Johnson, which was undertaken in connection with this chapter, has recently heen published in the Amer. Journ. of Anat., vol. 10, p. 521561, 1910.

DEVELOPMENT OF THE (ESOPHAGUS.

359

there are two sorts of vacuoles, those due to the accumulation of intercellular fluid and those due to an active moving apart of the cells. The cavities in the oesophagus seem to belong to the latter class. It is possible that their formation is associated with the transfer of mitotic activity from the inner to the outer row of cells. A centre of mitosis in the outer layer would account for the local bulging of the epithelium. It is clear that a transfer of mitotic activity from the inner to the outer layer must take place in the embryo, but at what stage this happens is not known. It is generally agreed that the result of the vacuole formation is the enlargement of the lumen.

A B Fig. 268. — Models showing the epithelial tube of the oesophagus cut longitudinally. X 120 diam . (After F. P. Johnson.) A, embryo of 19 mm. (Harvard Collection, Series 819). B, 22.8 mm. (Harvard Collection, Series 871).

Folds. — In cross sections of the oesophagus in embryos of about 10 mm., the lumen presents a clear-cut, round or elliptical outline. In older embryos, owing to the formation of broad folds, in which the mesenchymal layer takes part, the lumen becomes irregularly crescentic, tri-radiate, or shaped like a "Greek cross" (Fig. 269). The fusion of the vacuoles with the central lumen contributes to the irregularity of the shapes presented. Notwithstanding the secondary folds, however, the early form of the oesophageal tube may be recognized even in 30 mm. specimens. The long axis, which is transverse above, becomes dorso-ventral below. This arrangement suggested to Kreuter that the lower part of the oesophagus shared in the rotation of the stomach, but he concluded that

360 "mechanical considerations are against this idea." Johnson has described a dorsal and a ventral fold in the middle part of the oesophagus of a 42 mm. embryo, which become left and right respectively as the stomach is neared. The main trunks of the vagus nerves, which are lateral in the upper part of the oesophagus, become dorsal and ventral below, where, however, they are involved in a coarse plexus. This relation lends support to the idea that the epithelial tube may rotate, but to demonstrate this a more

Fig. 269. — Models showing the development of the epithelial folds in the middle portion of the oesophagus. X 90 diam. (After F. P. Johnson.) A, embryo of 37 mm. (Harvard Collection Series 820). B, 42 mm. (Harvard Collection, Series 838). C, 120 mm.

critical study is required. The primary folds in the oesophagus appear to be definitely situated, but those which come later vary in different embryos.

Ciliated Cells.- — In 1876 Neumann recorded that in embryos of from 18 to 32 weeks the oesophagus is lined with stratified ciliated epithelium, which, however, is interrupted in many places by nonciliated areas. He states that by isolating the cells through maceration in Miiller's fluid, he obtained all sorts of transition

DEVELOPMENT OF THE (ESOPHAGUS.

361

forms between ciliated columnar epithelium and flat epithelium. In a later publication (1897) he has figured the isolated cells, and has shown that the cilia are associated with distinct basal bodies. The smallest embryo in which the cilia have been found measures 44 mm., and its age is estimated at " about 69-70 days" by Jahrmaerker, and at "10-11 weeks" by Schridde, both of whom described this specimen. Schaffer failed to find cilia in a twelve weeks' embryo. They are apparently absent in a specimen of 42 mm. in the Harvard Collection, but are abundant at 55 mm. Cilia are still present at birth according to several observers, but in the specimens examined by Jahrmaerker and Schridde none were found. Fig. 270 is from the oesophagus of a negro child at birth, in which ciliated cells are abundant.

m

Fig. 270. — Section of the oesophageal epithelium at birth. X 000 diam.

The ciliated cells arise simultaneously in various parts of the oesophagus at a time when the epithelium is two-layered. They appear to belong with the superficial layer, but Schaffer (1904) has found that some of them may be traced through the entire epithelium to the basement membrane. Jahrmaerker nevertheless considers that the ciliated cells belong with the outer layer, in which some cells become ciliated and others do not. He finds that both forms of cells have finely granular, darkly stained protoplasm.

In the 44 mm. embryo, according to Jahrmaerker, the free surface of many non-ciliated columnar cells, generally in small groups or bordering upon the ciliated areas, shows a distinct dark border, which seems to indicate a transition to the ciliated form.

Schridde (1907), by using Unna's Wasserblau-Orcein, found that the ciliated cells have a dark-blue, finely granular protoplasm, and stand out distinctly from the clear columnar cells. He wi-ites : "The discovery of ciliated cells extending to the basement membrane seems to me to be of special significance. In my opinion it is therefore certain that the ciliated cells are not derivatives of the upper layer. . . . TVe must rather consider that these elements are formed from the basal cells." However, Schridde has neither figured nor described any darkly stained cell which has not reached the free surface, such as would be expected if certain basal cells were pushing outward.

362 HUMAN EMBRYOLOGY.

For a time the number of ciliated cells increases. Thus the ciliated areas in a 99 mm. embryo are more extensive than at 55 mm., as shown in models made by Johnson. The epithelium becomes 3-5 layered, but even in five-layered epithelium, according to Schridde, ciliated cells may sometimes be traced to the basement membrane. Ultimately their basal processes are lost and the ciliated cells appear crowded between adjacent vesicular cells. They are more deeply stained than before, which has been attributed both to compression and to degeneration. It is agreed by Schaffer, Jahrmaerker, and Schridde that the ciliated cells are desquamated, together with the outer non-ciliated cells, and in wellpreserved specimens they may be found free in the lumen of the oesophagus. It appears improbable that they lose their cilia and become vesicular cells, as Neumann originally maintained. He seems to have observed various shapes of ciliated cells, rather than transition forms.

Non-ciliated Cells. — Schridde has described the differentiation of the non-ciliated cells as follows: In a 100 mm. specimen (16 weeks) the epithelium appears 4-5 layered, and is composed of clear, polyhedral cells. The lowest layer likewise consists of clear cells, which almost throughout are cuboidal or low columnar in form. In a slightly older specimen, under low magnification, the basal layer appears darkly stained. With an immersion lens, the protoplasm of the dark cells is seen to contain interlacing fibrils. The "fibre-cells" are pushed outward, gradually displacing the clear cells. In embryos between 195 and 240 mm. they are found in all of the layers, but some of the earlier generation of clear cells are retained at birth, and they were seen in a child of three days. Intercellular spaces bridged by fibrils were first found in a child at birth. Keratohyalin granules do not appear in the superficial cells until some time after birth.

Schridde finds the number of layers in the epithelium to be 8-10 in a thirtysix weeks' embryo, 9-10 at birth, and 12-15 three days after birth. In the specimen from which Fig. 270 was drawn, the number of layers is from 3 to 7, and this accords with Riickert's statement (1904) that the flat epithelial covering of the oesophagus at birth is very thin, sometimes consisting of only two layers. The outermost cells, moreover, are not greatly flattened.

In the lower part of the oesophagus at birth, Strecker (1908 1 ) found numerous irregular clefts in the epithelium, so disposed that sometimes the intervening cells appeared as pointed epithelial papilla?. In the oesophagus of a child of 13 months he reports true epithelial papillae with connective-tissue cores. " These are occasionally pointed, but generally they are conical, suggesting in their shape the papillce fungiformes of the tongue." The portion of the oesophagus in which they occur, he regards as belonging with the cardiac antrum or " Vormagen." Glands. — Small groups of secreting cells, which represent the earliest gland formations in the oesophagus, may be found in embryos of about 78 mm. (3 months). An imperfect series of such a specimen in the Harvard Collection is sufficient to show that these areas are present both at the upper and lower ends of the oesophagus.

Schaffer (1904) described such cells in a 4 months' embryo as follows: " With low magnification a well-defined, small, lighter group of cells was seen

DEVELOPMENT OF THE (ESOPHAGUS.

inj

in the epithelium of the lateral pocket of the oesophagus, at the level of the thira or fourth tracheal cartilage. With higher magnification I found the typical, several-rowed ciliated epithelium . . . interrupted by a group of clear, remark ably tall columnar cells, arranged in a single layer which bulged slightly ab.-> V e the epithelial surface. The number of these cells, in the cross section, was about ten. Their nuclei, placed well toward the base, formed a row which bulged somewhat toward the underlying tissue. In their finer structure the cells accorded fully with the account which d'Hardivillier (1897) has given of the prismatic gland-cells in a 7 months' embryo. The cells appeared as if empty; only their walls stood out clearly. Their uppe~ ends lacked not only the cilia but the border of basal bodies." Schridde (1907) found a similar group of five very tall columnar cells in the lateral pocket of an embryo of 105-110 mm. (16-17 weeks). They were at the level of the cricoid cartilage. In regard to their structure he states: "The upper end of these cells was filled with an elongated oval plug, distinctly red-stained, and presenting a well-defined honey-comb structure. That these plugs were of mucus was shown by Unna's stain, which I am convinced offers a good reaction for mucus, and also by staining with mucicarmin."

•e Hi

'» is

Fig. 271. — Section through a group of mucous cells near the cardiac end of the oesophagus of an embryo of 240 mm. X 600 diam.

At the lower end of the oesophagus, as seen in older erubryos (120 mm. and 240 mm.), such cells are very abundant. Some of them occur in small groups, such as Schaffer and Schridde described (Fig. 271). The secretion, as indicated by the vacuolated protoplasm, nearly fills the cells, so that the nuclei at the basal ends appear compressed. Terminal bars, or intercellular cement lines, are seen at the free surface. These groups of cells are usually, but not invariably, bounded by ciliated epithelium. In the 240 mm. specimen the secreting cells often cover considerable areas which have been evaginated so as to form branching glands (Fig. 272). Usually several short tubules open into a broad cavity, whi;^ in turn connects with the central lumen of the oesophagus. The cavities are lined in part with stratified epithelium, and in part with the simple glandular epithelium which may form a portion of the lining of the oesophagus around the outlet of the gland.

A longitudinal section through the junction of the oesophagus and stomach at 120 mm. shows that the irregular clumps of secreting tubules gradually give place to a succession of quite uniform pits. As the distance from the oesophageal epithelium increases, the tubules become less and less branched (Bensley, 1902). The irregular forms, which occur both in the oesophagus and the cardiac end of the stomach, are the cardiac glands. The simple tubes, occurring further within the stomach, are gastric pits.

At birth the upper group of cardiac glands in the oesophagus may have the simple character which has been described, but as found in the adult they have undergone further development.

362 HUMAN EMBRYOLOGY.

in f They were present in 70 per cent, of the cases examined by Schaffer, being found in the lateral folds of the oesophagus between the cricoid and fifth tracheal ear tilages, frequently on both sides. They may appear macroscopically as erosions a b-mt 1 mm. in diameter. (The largest area which Sehridde obseiwed was ?3.5 x 9 mm.) The glands discharge through a dilated duct lined with simple columnar epithelium, which is said to open at the top of a connective-tissue papilla.

Tubules of a new soil have grown out from the gland; they consist of cells with round nuclei, and may produce a serous secretion. Certain of the tubules are provided with parietal cells and chief or zymogenic cells, so that the glands resemble those of the stomach. Schridde in 1904 described such areas as " islands of gastric mucosa," and considered that they were remnants of entoderm isolated by the downgrowth of the ectodermal layer (stratified epithelium) from the mouth, — an error which led to prolonged discussion. E. Schwalbe (1905) has found resemblances between the epithelium of the cardiac glands and that of the intestine, even in the production of cells resembling Paneth's cells.

'i^e

ii-P

Fig. 272. — Model of a superficial gland from the cardiac end of the oesophagus at 240 mm. X120 diam. (After F. P. Johnson.) The extent of the glandular epithelium is indicated by the ruled surface; the unruled area is occupied by squamous epithelium.

The lower group of cardiac glands of the oesophagus is usually limited to a /one from 1 to 4 mm. wide, situated at the entrance to the stomach. The glands vary in their development. Those which Strecker fig-ured from a twelve weeks' child are simpler in form than the one shown in Fig. 272, from an embryo of 240 mm. They consist of tall glandular cells forming a simple epithelium. Later, as in the upper group, new tubules develop which may contain chief and parietal cells. The ducts are usually distended and cystic. Between the upper and lower groups cardiac glands are rarely found, but Eberth (1897) has recorded a small area in the beginning of the lower half of the oesophagus in a man 25 years old.

The cardiac glands of the oesophagus have been named by Hewlett (1901) the superficial glands (glandulse oesophageal superficiales). They do not extend through the muscnlaris mucosae. The deep glands, which have their secreting portion in the submucosa, arise later.

They are apparently indicated in the 240 mm. embryo by short rounded downgrowths of stratified epithelium. At this stage there is no evidence of secretory activity. At birth, as shown in Johnson's model (Fig. 273), these glands are somewhat tortuous tubes. Some of them show expanded terminal portions and others have begun to branch. Occasionally, as on the right of Fig. 273, a gland is found in which the terminal secretory portion has not yet developed. The lower portion of the ducts is lined generally with low two-layered epithelium, but in some places only a single layer is found. As the duct approaches the surface, its outer cells become somewhat elongated and they are seen to be continuous with the

DEVELOPMENT OF THE (ESOPHAGI'S 3 basal layer of the stratified surface epithelium. The secreting portion consist? typical mucous cells. They are not as slender as those in the cardiac glands, the part occupied by secretion is more homogeneous. They yield the sti reactions for mucus more readily than the cells of the cardiac glands, the diffei being so great that the mucous nature of the latter has been questioned. In the adult the deep glands are said to open between connective-tissue papillae, whereas the cardiac glands open at their summits. This distinction s» ems arbitrary, especially since the papilla 3 arise after the glands are present.

Fig. 273. — Model showing three deep oesophageal glands at birth. X 90 diam. (After F. P. Johnson.) The Outer Layers. — In the oesophagus, as elsewhere in the digestive tube, it is well known that the circular muscle layer is the first of the outer coats to be differentiated.

At 10 mm. it is represented by a concentric layer of myoblasts, separated from the epithelium by a broad band of undifferentiated mesenchyma. The circular muscle is so far outside of the epithelium that it is undisturbed by the epithelial folds and pockets which arise in later stages. In the 10 mm. embryo there are numerous branches of the vagus nerves, some of them associated with groups of cells with crowded nuclei, found just outside of the circular muscle. These represent the myenteric plexus. Occasionally at this stage similar groups of cells appear along the inner border of the musculaiis, and these give rise to the plexus submucosus.

At 12.5 mm. Keibel and Elze note that the oesophagus shows a circular, but no longitudinal, muscle layer. At 17 mm. they find a strong circular layer, with the longitudinal layer only indicated. Kreuter finds that the circular muscle is already differentiated in the fifth week (9 mm.), but the longitudinal muscle first appears in the eighth week. Happich states that, although the circular muscle in a four months' embryo has attained a considerable strength, the longitudinal muscle is indicated only by very' weak fibres. Schridde, on the contrary, finds that both layers are clearly marked at 12.4 mm., and at 21 mm. the longitudinal musculature is everywhere well developed. It is possible that Schridde mistook the conspicuous layer of nerves, found just outside of the circular muscle at 12 mm., for the longitudinal muscle. These nerves, with the undifferentiated ganglioncells, form a nearly continuous layer.

The longitudinal muscle is perhaps indicated at 30 mm., but at 42 mm. it is thinner and less conspicuous than the layer of nerves which separates it from the circular muscle. At 55 mm. it is present as a definite layer.

3fc6 HUMAN EMBRYOLOGY.

The rnuscularis mucosas is not found in the 55 mm. embryo.

78 mm. it is not distinct at the upper end of the oesophagus, but it SB very definite below. At 91 mm. it is a well-developed layer of longitudinal fibres equalling the tunica propria in breadth, and thrown into folds corresponding with those of the epithelium.

The development of the striated muscle of the human oesophagus has not been satisfactorily studied. The oesophageal smooth muscle layers at first extend to the larynx, where they contrast sharply with the striated fibres of the inferior pharyngeal constrictor. There is no evidence of a downgrowth of these fibres upon the oesophagus.

In pig embryos, according to McGill (1910), the smooth and striated musclefibres of the oesophagus have a common origin in the mesenchymal syncytium. " Until the cross striations appear in the fibrillar of the striated muscle, both developing tissues look precisely alike." Cross striations were first observed in pigs of 13 mm., but " only a few fibrillse become striated before the embryo reaches a length of 30 mm." In cross sections of the upper part of the human oesophagus at 7 8 mm. the longitudinal fibres are triangular or polygonal, with peripheral nuclei, and they show coarse myofibrils, but the circular fibres do not appear to be striated. Striated circular fibres are distinct at 120 mm. It is probable that these are "a further differentiation of smooth muscle" (McGill).

i The musculature of the upper half of the oesophagus in the adult consists chiefly of striated fibres, but Klein (1868) has concluded that smooth muscle in the longitudinal layer begins in the upper quarter. Once in an adult he found that the circular layer, 1 cm. below the upper end of the oesophagus, consisted chiefly of smooth muscle. In another case he found that the ventral part of the longitudinal layer at the upper end of the second quarter consisted chiefly of smooth fibres, but that further down the striated fibres increased so that the relation was reversed. He found no striated fibres in the lower half of the oesophagus. Coakley, however (1892), has described striated fibres intermingled with the non-striated in both coats of the diaphragmatic portion of the oesophagus. The majority were in the inner circular layer. He considers that the pillars of the diaphragm are the source of these fibres.

The layer of mesenchyma between the circular muscle and the epithelium in the 10 mm. embryo is quite free from bloodvessels. Vessels have entered it at 14.5 mm., and at 16 mm. they form a distinct plexus. Beginning at about 30 mm. the inner portion of the mesenchymal layer becomes gradually denser, due ( to an abundance of nuclei. Thus the tunica propria, consisting of s reticular tissue, is slowly differentiated from the fibrous conneca tive tissue of the submucosa. At birth the propria contains "" abundant blood-vessels, and apparently lymphatic vessels are present also. No lymph nodules were seen in the sections examined. In the oesophagus of a child Klein (1868) found that

DEVELOPMENT OF THE CESOPHAGUS. 367 the reticulum contained "more or less numerous round cells similar to lymphocytes," but he speaks of nodules only in the adult. The nodules of the oesophagus apparently develop later than those of the stomach and intestine.

As already noted, papillae of the tunica propria are absent at birth, but in cross sections the basal border of the epithelium presents a slightly wavy outline. Since Strecker finds that in longitudinal sections the basal line is usually straight, he considers that the elevations are ridges and not papillae. He finds that the oesophagus passes through three stages of development: 1, in which the tunica propria has a smooth contour ; 2, in which it has formed ridges; 3, in which there are conical papillae upon the ridges. At birth the human oesophagus is in the second stage. In a child of 12 months all the later characteristics are present.

Anomalies of the (Esophagus. — In a previous section the anomaly of the oesophagus in which the upper segment ends blindly below and the lower segment arises from the trachea has been discussed (p. 312). It was stated that it must originate in embryos of about 4 mm. This has been confirmed by finding the anomaly well developed in an embryo of 18.1 mm. in the Harvard Collection. It> this specimen there is no trace of epithelial connection between the two parts of the oesophagus. Ribbert (1902) has interpreted traction diverticula as a modification, or partial development, of this anomaly. In these cases the ventral wall of the oesophagus, near the level of the bifurcation of the trachea, presents a funnelshaped diverticulum with its apex directed obliquely upward toward the trachej. The epithelial pocket may penetrate the muscle coat, and from its apex a strand of vascular connective tissue generally extends toward the wall of the trachea. The inflammatory conditions which are often found associated with the pocket are regarded by Ribbert as secondary. Although he states that traction diverticula occur chiefly in older people, he believes that in the great majority of cases they have an embryological origin. He considers that there is a defective development of the oesophageal wall at the place where in more radical cases the tracheooesophageal fistula occurs.

An examination of the embryos in the Harvard Collection fails to show such a defect. However, Happich (1905) has recorded that in embryos from 8 or 9 mm. to 3 or 4 months, the entire musculature on the ventral side of the oesophagus is thinner than on the dorsal side, as far down as the bifurcation of the trachea. Below the trachea this distinction is wholly lacking. At birth the ventral musculature is slightly weaker than the dorsal, but the difference is almost imperceptible. It is clear, however, that such a thinning cannot account for the anomaly in question, since it extends the whole length of the trachea and disappears at birth. In addition to the general thinning, Happich has found that the circular muscle, in embryos of 3 or 4 months, is completely interrupted in small areas extending through one or two sections. " These places can readily be distinguished from those through which a vessel penetrates the wall." Schridde (1908) found a larger defect, extending through five sections, in that part of the longitudinal muscle layer which is toward the trachea. This, however, was in a 13 mm. embryo, which is a stage when the longitudinal muscle is not ordinarily recognizable.

Riebold (1903 and 1908) believes that the embryological interpretation of traction diverticula is not justified, and he adheres to the older idea that they are pathological. He cites the literature to show that Ribbert's theory has not met with general acceptance, and states that " up to the present time not a single case of traction diverticulum has been found at birth." Lymphadenitis, with

368 HUMAN EMBRYOLOGY.

adhesions of the gland to the trachea and a spread of the inflammation along vessels to the oesophagus, is believed to produce a dense scar which draws upon the oesophageal tube, and, as a result, of the motions in swallowing, the diverticulum is drawn out. Diverticula may occur wherever a vessel penetrates the muscle, and therefore below the trachea. They may be multiple, and they are not always ventral.

Embryologically it is probable that if the trachea and oesophagus have separated normally at 4^5 mm., the muscle layers which arise at 9-12 mm. will show no local defect at the place of the former separation. Unless the diverticula are primarily epithelial, they are presumably not congenital.

Pulsion diverticula occur on the dorsal wall of the oesophagus, at its junction with the pharynx, where the tube is narrowest and the muscle coat thinnest (Riebold). They apparently have no embryological significance. Diverticula occur also in other parts of the oesophagus. Some of these are evidently of inflammatory origin. D'Hardivillier has asked whether the islands of simple epithelium do not offer places of lesser resistance which would lead to diverticula, and it has been pointed out that pulsion diverticula and these thin areas both occur at the upper end of the oesophagus. Apparently, however, there is no relation between them.

The irregularities in the oesophageal epithelium in embryos of 18-22 mm. have been supposed to give rise to the cases of atresia and stenosis, and possibly to diverticula, but direct evidence is lacking. Atresia of the oesophagus is abnormal in embryos of all stages. Many records of oesophageal anomalies have been gathered by Happich, Kreuter, and Forssner, who have discussed them embryolosrically.

THE DEVELOPMENT OF THE STOMACH.

By FREDERIC T. LEWIS.

Early Development. — Remak (1855) described the intestinal wall of vertebrates as composed primarily of two layers, — the gland-layer (Darmdrusenblatt) and the fibre-layer (Darmfaserplatte). The former gives rise to the epithelium and glands, and the latter produces the remaining layers. Schenk (1868), from a study of chick embryos, concluded that Remak had overlooked a third layer, which develops downward from the mesodermic somites and extends between the gland-layer and the fibre-layer. He found that this third layer was clearly connected with the somites, but was separate from the adjacent layers. Therefore he concluded that Remak's fibre-layer produced only the lining of the peritoneal cavity. Schenk's interpretation was rejected by Kolliker (1879, p. 850) and by Maurer (1906). Maurer finds that in all vertebrates the embryonic intestinal wall (including that of the stomach) consists at first of two layers, — the entodermal epithelium and the mesodermal epithelium. ' The latter, in the Amniota, becomes stratified, and for some time it may exceed the delicate entoderm in thickness. It produces mesenchymal cells, which form a third layer situated between the two primary epithelia.

The mesodermal epithelium covering the digestive tube is called the splanchnopleure by Maurer, but, as pointed out by Minot (1901), this usage is incorrect, since the term was introduced by Foster to designate the entire intestinal wall. His (1865) proposed the name endothelium in the following passage (here somewhat abbreviated) : " We are accustomed to designate the layers of cells which cover the serous and vascular cavities as epithelia. But all the layers of cells which line the cavities within the middle germ layer have so much in common, and from the time of their first appearance differ so materially from those derived from the two peripheral

DEVELOPMENT OP THE STOMACH. 360 germ layers, that it would be well to distinguish them by a special term, — -either to contrast them, as false epithelia, with the true, or to name them endothelia, thus expressing their relation to the inner surfaces of the body." The temi endothelium, as proposed by His. is itself too extensive, since it includes both the epithelium lining the vessels and that which lines the body cavities. These epithelia, although similar in the adult, are very distinct embryologically. Accordingly Minot (1892) uses the term mesothelium for the mesodermal cells bounding the body cavities, and applies endothelium to the vascular system. Thus the nomenclature lias become complex. The layer covering the intestine is perhaps best referred to as the ccelomic or peritoneal epithelium.

The two-layered stage of the stomach is seen in the 4 mm. Bremer embryo. Here the fore-gut presents a dorso-ventral cleftlike lumen, both in the oesophageal and gastric regions. The thick ccelomic epithelium is in direct relation with the ventral part of the sides of the fore-gut, as far anteriorly as the lung-bud. Thus laterally the gastric region is in the primary two-layered stage, but dorsally and ventrally, and to some extent on the sides, the entodermal epithelium is bounded by mesenchyma. The mesenchyma appears to be derived chiefly from the ccelomic epithelium, yet it is possible that some has grown down from the somites. In this specimen there is no difference between the oesophageal and gastric epithelium.

In a 10 mm. embryo the gastric epithelium is distinctly thicker than that of the oesophagus, and its nuclei are more elongated. In the sections examined, the nuclei form four or five overlapping rows, but the true number of cell layers is probably less. Jahrmaerker finds that at 8 mm. the gastric epithelium is 2-3 layered, with tall columnar basal cells, whereas both the oesophageal and intestinal epithelia have only two layers. In the 12 mm. embryo he attributes the greater thickness of the gastric epithelium, as compared with that of the 'oesophagus or intestine, to the tall basal cells which are found in the stomach.

Vessels and Nerves. — The general relations of the stomach in the 10 mm. embryo are shown in Fig. 274. At the oesophageal end, the vagus nerves occupy dorsal and ventral positions. Their bundles of fibres are associated with small clumps of cells with crowded nuclei. The dense layer of mesenchyma, indicating the circular muscle, which is distinct along the oesophagus, gradually disappears at the cardia.

In this region a vessel leaves the stomach and passes through the lesser omentum to enter the ductus venosus (Fig. 274, A). This vein was first described by Broman (1903) as follows: " In human embryos 5-16 mm. long, there are always one, two, or several branches of the ductus venosus passing through the lesser omentum to the mesodermal wall of the stomach, where they form a thick plexus. The branches of the cceliac artery connecting with this plexus appear to be relatively insignificant, at least in the earlier stages. In older embryos I have sought in vain for the branches of the ductus venosus, and may therefore believe that they have degenerated." Vol. II.— 24

370 The coeliac artery is seen leaving the aorta in Fig. 274, C. Its branch, the left gastric artery, lies at the root of the great omentum, along which it ascends to the cardiac end of the stomach. The hepatic branch of the coeliac artery is seen beside the portal vein. Subsequently branches of the portal vein and hepatic artery extend to the pylorus and along the greater curvature, thus forming the right gastro-epiploic vessels. In the 10 mm. embryo these appear to be indicated by minute twigs. In a 22.8 mm.

Ao. N.sym.

N.sym.

Ao. N.sym.

v *o % - to

%ZCW O.ma.

D.ch.

• >^A. hep. V.p.

B

C

Fig. 274. — Sections of the stomach of a 10 mm. embryo (Harvard Collection, Series 1000). A, through the cardia. B, through the fundus. C, through the pylorus. A.coel., coeliac artery ; A.g.s., left gastric artery ; A. hep., hepatic artery ; Alien., splenic artery ; Ao., aorta ; B.om., omental bursa ; C.W., Wolffian body ; D.ch., common bile-duct ; D.v., ductus venosus ; F.ep., foramen epiploicum ; N.sym., sympathetic nerve ; O.ma., greater omentum ; O.mi., lesser omentum ; Pul., lung ; Va., vagus nerve ; V.p., portal vein ; V.8., left suprarenal vein.

specimen the portal vein communicates with the left suprarenal vein by a vessel which receives branches from the stomach, following the course of the left gastric artery. This communicating vein corresponds with the coronary vein of the adult, by forming an anastomosis between the portal and cardinal systems along the lesser curvature of the stomach.

In the 10 mm. embryo the dorsal and ventral trunks of the vagus nerves unite to form a large ganglionated plexus on the right side of the stomach, nearly in the median plane of the body (Fig. 274, B). A similar arrangement was found in embryos of 9.4 and 14 mm. There are no distinct nerves along the greater curvature, and there is no indication of the muscle layer. In older embryos (14.5 mm.) the sympathetic nerves communicate with this ganglionic mass, but in the 10 mm. embryo the connection could not be demonstrated. The sympathetic nerves are seen extending forward on either side of the aorta, ventral to which they form a coeliac plexus. From the latter, in older embryos, bundles of fibres extend to the stomach along the dorsal mesentery, following the path shown in Fig. 274, B.

DEVELOPMENT OF THE STOMACH. 371 Kuntz (1909) has recently published a similar description of the nerves in pig embryos. In 12 mm. specimens he found a vagus plexus around the oesophagus, and " vagus fibres "which are still accompanied by numerous cells may now be traced along the lesser curvature of the stomach." There are still no fibrous connections between the cceliac plexus and the plexuses in the digestive tube. In 16 mm. embryos fibrous connections have become established.

There is another path by which sympathetic fibres may enter the stomach. They may extend from the cceliac ganglion to the pylorus, following the gastric branches of the hepatic artery (Fig. 274, C). His, jun. (1897), has figured a section of a 9.1 mm. embryo which shows sympathetic nerves passing to the stomach along this course. In describing a 10.2 mm. embryo he speaks of a branch of the cceliac plexus which is lost in the mesoderm of the pylorus, and, as shown in his reconstruction, it does not anastomose with the vagus. In the specimens in the Harvard Collection the pyloric branches of the sympathetic cannot be identified at such an early stage. It appears rather as if the gastric plexus first extends downward to the pylorus and duodenum, and is then joined by such sympathetic branches as His described. These are distinct in a 30 mm. embryo.

Epithelium and Gastric Glands.— In the 10 mm. embryo the free surface of the epithelium is somewhat wavy, whereas the basal surface is nearly smooth. In 16 and 19 mm. specimens the epithelium exhibits occasional vacuoles and a few scattered pits. The vacuoles are small, and, like the pits, they do not cause the basement membrane to bulge. Sometimes the pits expand laterally within the epithelium so that they are flask- shaped. These structures bear a certain resemblance to the oesophageal vacuoles and the intestinal diverticula to be described later.

Elze (1909) has noted that in the stomach of ape embryos {Nasalis larvatus) there are several epithelial buds and diverticula which have the same appearance as the early stages of those found in the intestine.

At 22.8 mm. a few vacuoles are still present. The intraepithelial pits have become numerous. As seen in Fig. 275, A, they are produced by the varying height and characteristic arrangement of cells in an epithelium which has nearly smooth surfaces. In places the epithelium is clearly simple, but elsewhere it may show several rows of nuclei and is perhaps stratified. In a 42 mm. embryo the epithelium is more definitely simple, and the pits form rounded swellings along its mesenchymal surface.

This characteristic stage has been figured by Toldt (1881) in an embryo of the tenth week. It was not seen by Laskowsky (1868), who considered that the gastric glands were produced by the growth of the mesenchymal layer, rather than by epithelial proliferation. His view was accepted by Schenk (1874, p. 117) and others, but Toldt, who considered the pits to be a part of the glands, correctly concluded that "the first formation of the glands is a process which takes place exclusively in the epithelial layer."

372 At 55 mm. the pits still project but slightly below the general level of the basement membrane. The epithelial cells between adjacent pits, in positions corresponding with x in Fig. 275, A, have become greatly compressed below, so that the basal portions of a group of these cells resemble a clump of connective-tissue fibres. In all later stages the epithelial cells along the free surface and the adjacent portions of the sides of the pits may exhibit

Fig. 275.

-Sections of the gastric epithelium. Collection, Series 871).

X 330 diam. .4, from an embryo of 22.8 mm. (Harvard B, from an embryo of 120 mm.

slender basal prolongations ; they have been described by Baginsky (1882) in a 7 months' embryo, and by Fischl (1891) at birth. In the 55 mm. embryo the outer portions of these cells are clear, suggesting a mucous transformation. This is true of the cells on the sides of the pits, but at the bottom of the pits the protoplasm toward the lumen is coarsely granular.

At 99 mm. there are distinct mesenchymal elevations between the pits. At the bottom of the pits there are small, nearly solid buds of granular cells, which represent the beginning of the glands proper.

DEVELOPMENT OF THE STOMACH. 37'J The conditions at 120 mm. are shown in Fig. 275, B. The glands at the base of the pits already exhibit two sorts of cells, differing from one another in their affinity for eosin. The eosinophilic cells occur chiefly at the blind ends of the glands. Very generally they border upon the lumen. In later stages the eosinophilic cells are peripheral in position and are called parietal cells (delomorphous cells). The non-eosinophilic cells represent the chief or adelomorphous cells of later stages. Between the gland and the pit there may be a constriction, as seen in Fig. 275, B. The surface epithelium and that lining the pits is a simple columnar layer, containing mucous cells in various stages of development, but apparently all covered by distinct top plates. Generally the nuclei are elliptical, but occasionally a cell is seen with its nucleus flattened in the basal protoplasm. The basal protoplasm is sometimes eosinophilic, and groups of cells of the parietal type may be found in direct relation with the surface epithelium (as on the right of Fig. 275, B). These seem to represent new gland buds. There are also non-eosinophilic basal cells — the Ersatzzellen of Ebstein (1870) — which presumably develop into new columnar cells. In young embryos Toldt found these basal cells so abundant that in poorly preserved specimens they may easily give the impression of a stratified epithelium, whereas at birth they are relatively infrequent.

The form of the pits and glands in the 120 mm. embryo is shown in a model made by Johnson, the upper and under surfaces of which are shown in Figs. 276 and 277 respectively. The gastric pits are seen to be clefts rather than tubules, and the intervening tissue may be considered to form imperfectly separated villi. The pits are separated from one another below by irregular ridges of mesenchyma.

Brand (1877) states that in embryos of two and three months the stomach contains numerous villi, and Kolliker (1879) regards the mesenchymal projections between the pits as " villi." Of their later development he says : " In the fourth month the formation of glands has begun in the mucosa, while between the mesodermal villi, which have become longer, low inter-villi and ridges have grown up, marking out spaces like a honeycomb, into which the epithelium sends hollow cylindrical processes." Sewall (1879) found that in the sheep "from the first the mesodermal outgrowths are not papilliform, but take place along continuous lines of greater or less extent, giving rise to ridges which intersect in all directions." Toldt (1881 ) likewise found, in cat embryos, ridge-like elevations of mesoderm, but he states that it is not to be questioned that in stomachs of human embryos from the third to the fifth month, especially in the pyloric region, villus-like elevations occur, and even true elongated villi. Baginsky (1882) states that " the surface of the gastric fundus in a 4 months' embryo has an exquisite villous appearance." In later stages he finds that the surface becomes gradually smoother as the villus-like elevations disappear. Strecker (1908 1 ) describes an exceptional stomach at birth (?) showing typical villi in the cardiac region.

374 A mesodermal origin for certain epithelial and gland cells has been considered possible by several investigators. Thus, Ebstein thought that the basal cells may proceed directly from the blood-vessels, and Toldt recorded certain appearances suggesting

Fig. 276. — Model of the gastric epithelium at 120 mm., showing the free surface. X 120 diam. (After F. P. Johnson.) that in young stages mesodermal cells wander into the epithelium. Sewall (1879) and more recently Strecker (1908 2 ) have described the gastric glands as mesodermal.

Sewall (1879) concluded that in sheep embryos only the early generations of chief and parietal cells are formed from the primitive gland cells, and that the later generations arise in the mesenchyma. The parietal cells appear first in the

Fig. 277. — Model of the gastric epithelium at 120 mm., showing the basal surface.

F. P. Johnson.)

X 120 diam. (After

deep parts of the gland (in embryos of about 140 mm.). In later stages he concluded that new parietal cells were produced by the differentiation of the surrounding " mesoblast corpuscles," and that, from the parietal cells so formed, new chief cells developed to replace those broken down in the process of secretion. Physiologically he found that extracts of the stomach of the sheep, " even some time before term, showed a considerable proteolytic power." This function appears to coincide with the specialization of the chief cells. The fluid in the embryonic

DEVELOPMENT OF THE STOMACH. 375 stomach was found to be neutral, even after the differentiation of the parietal cells. It yielded an abundant precipitate of mucus.

Toldt (1881) rejected Sewall's conclusion concerning the mesodermal origin of parietal cells, and described the development of the human gastric glands as follows : " In the fourth and fifth months and also in the beginning of the sixth, parietal cells and their developmental stages are found only at the blind ends of the glands. Beginning with the middle of the sixth month they increase considerably in number and are found everywhere along the sides of the glands, yet they are still in the row of chief cells and therefore border upon the gland lumen. Not earlier than about the middle of the eighth month could I find regularly a considerable number of parietal cells situated on the outer side of the chief cells. At birth and in the first weeks following, this is almost always the case along the sides of the glands, but near and at the base the lumen is still bounded largely by parietal cells which are not fully developed. In children of four or five years all transitions from chief to parietal cells are constantly present in abundance, but later, when the growth of the glands takes place only very slowly, they are seldom found." According to Toldt the chief cells also develop from those which form the walls of the primitive gland. " These cells, differing from the later characteristic chief cells by their cuboidal or polygonal form, their delicate outline, their affinity for eosin, and the strikingly large size of their nuclei, gradually assume the typical form. ... In man this transformation is completed toward the end of the fifth and in the beginning of the sixth month." Chemical tests showed that pepsin was present in the gastric mucosa in the last half of the sixth month, " long before it passed over into the secretion." Strecker (1908 2 ) examined the stomach at birth, and found conditions which have generally been ascribed to post-mortem disintegration, such as the absence of columnar epithelium on the free surface, the presence of detached gland cells in cavities bounded by the tunica propria, and even a superficial layer of fibrin. All these he regards as normal, and states that " unquestionably the large gland cells appear distributed more or less irregularly in the tissue without any typical arrangement. They seem to be lodged in a well-marked reticular tissue, the meshes of which they fill. . . ." He described the embryonic development of these glands as follows: The primitive glands are purely epithelial, but in embryos of 100 mm. another sort of gland formation is seen taking place in the tunica propria. " The propria at this stage is not a connective-tissue layer, but an epithelioid organ." It contains many free nuclei (Bildungskerne) , which produce protoplasmic bodies and form groups of cells, thus giving rise to glands. " Both the chief and parietal cells arise from the same source, namely the Bildungskerne." The Bildungskerne form autogeneously in the original mesenchymal plate of the intestine, and Strecker names them " mesenchymal-epithelioid corpuscles." Not only are free nuclei found in the propria, but there are also non-nucleated masses of protoplasm. Nuclei wander into these, thus giving rise to giant cells. The multi-nucleate cells are generally found at the base of the glands. Portions of them become split off, so that they produce cell material for the gland tube. Strecker states that a true mitotic division in embryological preparations of the gastric glands has never been found by any investigator, but Salvioli (1891) has recorded abundant mitotic figures in rabbit embryos and has made a special study of their location.

From the fact that Strecker found the non-epithelial origin of glands beginning in 100 mm. specimens, it is probable that the purely epithelial glands" are the gastric pits, and those arising in the propria are the glands proper. Although, owing to tan

376 gential sections, parietal cells often appear isolated in the tunica propria, the conclusion of Sewall and Strecker concerning their mesodermal origin may be confidently rejected. The glands arise as further downgrowths of the pits. In the stomach, as in both small and large intestine, there are at first irregular coarse depressions (pits and intervillous spaces), from the bottom of which glands extend downward. The cells of the pits and villi are characteristically clear, whereas those at the depths of the glands are granular and deeply staining. The transition between the two is not abrupt, as shown in Fig. 275, B.

Fig. 278. — Models of the gastric pits and glands, A, at 240 mm.; B, at birth. X 80 diam. (After F. P.

Johnson.)

As compared with the pits, the glands steadily increase in length. In a 240 mm. embryo they occupy the basal third of the mucosa ; at birth they form nearly half of this layer, and therefore nearly equal the pits. They have branched repeatedly and have increased greatly in number. Their form is shown in Fig. 278, A and B, from models made by Johnson.

DEVELOPMENT OF THE STOMACH. 377 Their multiplication has been described by Toldt. He estimated that the total number of gland outlets in the stomach of an eight months' embryo is 128,912; at birth, 268,770; and at ten years, 2,828,560. In the three stomachs referred to, the number of outlets per square millimeter is nearly constant, averaging 56, 51, and 56 respectively. In studying the way in which the glands multiplied, Toldt failed to find primitive stages of gland development among the differentiated glands, but gastric pits were often observed to be partly divided, and he " sees no objection to regarding these divided pits as forerunners of the complete division of the glands." This method of multiplication would cause a reduction in the number of glands opening into each pit, and some reduction was found to occur. The average number of glands emptying into a pit in the last months of embryonic development is 7; at ten years, 6; at fifteen years, 5; and in the adult, 3. During this period Toldt found, however, that the number of gland tubules in the stomach had increased from 930,000 to 25,179,000, which means that many new tubules have been formed. These arise through lateral sprouts of glands already present. Toldt says that " It may be noted that these hollow sprouts are generally seen to develop at places along the gland wall where one or more parietal cells are situated, and that these pass over into the new gland body." Epithelium and Glands at Birth. — Fischl describes the gastric epithelium at birth as a "moderately high columnar epithelium with basal nuclei, appearing somewhat lower on the ridges than in the pits; moreover the nuclei in these two places differ, since they appear more elliptical and deeply stained on the ridges, but in the pits they are rounded and decidedly paler." Except that the cells on the ridges seem taller than in the pits, these observations have been verified. The cells exhibit distinct terminal bars. Those lining the pits are producing and discharging mucus, which fills the lumen and spreads over the free surface. The cells bordering upon the free surface contain a more granular protoplasm, and according to Toldt they sometimes give no indication of the formation of mucus.

Disse (1905), by the use of a mucin stain, found that "the true surface epithelium contains only here and there an isolated mucous cell, but chiefly consists of cells containing no trace of mucus." He concludes that, although in some places the mucous layer is well developed in embryos at term, there are other places in the same stomach where mucus is wholly lacking or forms an interrupted layer. Reyher (1904) and Von der Leyen (1905) have found that the mucous layer is continuous. It is possible that the surface cells with granular protoplasm are those which have previously discharged mucus (see Fig. 275, B).

Fischl was unable to find mitotic figures among the epithelial cells, but Ascoli (1900) has declared that at birth they may be found in large numbers, in cells containing mucus.

Neumann (1876) repeatedly found well-developed ciliated cells among the epithelial cells of the embryonic stomach. (The age of the embryos is not definitely stated.) Baginsky (1882) described the gastric contents of a 7 months' ernbryo as alkaline and containing, together with epidermal cells which were probably swallowed with the amniotic fluid, small ciliated cells, generally isolated. In the specimens in the Harvard Collection no ciliated cells were found.

378 HUMAN EMBRYOLOGY.

The glands at birth appear distinctly broader, shorter, and more widely separated than in the adult, as noted by Fischl. In seven cases, all from the first half of the first year, he found the parietal cells only partially differentiated, and represented by rounded, rather small cells, often situated near the gland lumen, and never pushing out into the tunica propria. At the end of the second year, he states that they show no essential difference in staining, form, and arrangement from those of the adult, although they are less abundant.

Fischl's difficulty in demonstrating the parietal cells at birth has not been shared by others, Kalopothakes (1894) having reported them as " perfect " in a six months' embryo; but it is doubtless true that neither they nor the chief cells are fully differentiated until after birth.

Cardiac and Pyloric Glands.

The early writers grouped the cardiac and pyloric glands together and named them the mucous glands of the stomach. Thus, Toldt (1881) states that these glands are " quite alike in form and structure," and Strecker has recently noted the " striking similarity" between them. The cardiac glands have apparently been more thoroughly studied than the pyloric, and the literature concerning them has been reviewed by Bensley (1902) and Strecker (1908 1 ).

Embryologically the pyloric region very early differs from the remainder of the stomach. In a 42 mm. embryo the pits are deeper and more irregular toward the pylorus, where there is an abrupt transition to the characteristic villi of the duodenum, and this is true of all later stages. At 240 mm. the epithelium of the duodenum is a darkly staining granular layer frequently interrupted by clear globular goblet-cells. The pyloric epithelium is a uniform layer of clear columnar cells filled with secretion, thus resembling the epithelium which lines the gastric pits. In the pyloric region the pits are very deep and they coalesce with one another laterally so that the intervening tissue forms long irregular villi. These have been described by Toldt, Baginsky, and others, but apparently they have not been modelled.

In early stages the entire lining of the pyloric glands is like the surface epithelium. This condition is found in an embryo of 120 mm., and Baginsky has recorded it at 4 months. At 7 months, however, he found, in addition to such glands, others which showed, toward their bases, darker, finely granular cells staining clearly with eosin. In an embryo of 240 mm. there are occasional basal eosinophilic cells which resemble parietal cells. Toldt, however, in twenty stomachs from older embryos and children under five years, failed to find any parietal cells associated with the pyloric glands.

The cardiac glands of the oesophagus have already been described (p. 362). They are groups of short tubules lined with a columnar epithelium resembling that of the pyloric glands.

DEVELOPMENT OF THE STOMACH. 379 Passing from the oesophagus into the stomach, the cardiac glands become more elongated and more compactly arranged. Their epithelium gradually blends with that of the gastric pits. In the transition the cells become somewhat shorter and stain less brightly with orange G. At the same time gastric glands appear at the base of the pits, and the number of their parietal cells increases. Toldt considered that there is a sharp distinction between the cardiac and the gastric glands, inasmuch as the cells from which they arise are of very different sorts, but there is undoubtedly a gradual transition between them. In man, however, there is no embryological evidence in favor of Bensley's conclusion that "the cardiac glands are decadent or retrogressive structures derived from the fundus glands by the disappearance of their more highly specialized constituents." On the contrary, the cardiac glands are differentiated very early. They can be recognized in a 91 mm. embryo, in which there are still no chief or parietal cells. 12 The Outer Layers. — In 10 mm. embryos the gastric wall consists of three layers, — entodermal epithelium, mesenchyma, and peritoneal epithelium. At 16 mm. there is a condensed zone of mesenchyma indicating the circular layer of muscle. It is best denned along the lesser curvature, but it can be identified over the greater portion of the stomach. A uniform layer of mesenchyma extends between the muscle layer and the entoderm. It already contains a plexus of blood-vessels. The nerves and ganglia have spread from the lesser to the greater curvature. Thev are chiefly outside of the muscularis. At 22.8 mm. there is a slight condensation of the mesenchyma toward the entodermal epithelium, indicating the beginning of the tunica propria. The circular muscle layer is complete, and shows a slight thickening toward the pylorus. A prolonged gradual thickening of this layer, followed by an abrupt thinning at the duodenum, is distinct at 37 mm. and in all later stages. At 37 mm. large lymphatic vessels are seen in the mesentery along the lesser curvature, but apparently they do not penetrate the muscularis. This is true also at 42 mm. At 55 mm. there is a dense tunica propria ; no muscularis mucosae; a submucosa containing blood-vessels and occasional nerves toward the muscularis; a single layer of circular muscle, 12 The histogenesis of the gastric glands in the pig has recently been studied by Kirk (1910). He finds that the parietal cells arise very early as epithelial cells staining intensely with eosin, situated in the deeper parts of the glands. Kirk confirms Toldt regarding the absence of these parietal cells from the pyloric glands. He finds a very gradual transition between the gastric and cardiac glands, and considers that the latter are retarded or regressive glands, following Bensley. But he states that mucous differentiation occurs slightly earlier in the cardia than in the fundus.

380 HUMAN EMBRYOLOGY.

outside of which nerves are numerous; and a relatively wide serosa.

At 91 mm. the muscularis at the cardiac end of the stomach shows a few inner longitudinal bundles. These are seen also at 120 mm. At this stage the outer longitudinal layer of the oesophagus may be followed a short distance over the cardia, external to the circular layer. The greater portion of the stomach has only the circular layer. At 240 mm. an outer longitudinal layer is distinct in the pyloric part of the stomach and it becomes thicker toward the duodenum. Some of its bundles are continuous with the longitudinal layer of the duodenum, but others turn into the thick circular layer near the pylorus, forming the dilator pylori (cf. Cunningham, 1908). At birth the thin outer longitudinal layer, according to Fischl, is entirely absent in places, especially along the greater curvature.

The muscularis mucosae is indicated at 120 mm. At birth Fischl finds it clearly divisible into an inner circular and an outer longitudinal layer.

Lymphatic vessels appear in the submucosa in embryos of 214 and 240 mm. Lymph-nodules were found at birth in a considerable percentage of the cases examined by Fischl. They were observed in all parts of the stomach ; sometimes they were at the base of the glands, and did not extend upward between the tubules.

The longitudinal folds of the stomach, which are often found in preserved specimens, appear to be quite irregular. Toldt has seen them formed by muscular contraction in freshly opened embryonic stomachs of cats, and does not consider them to be "specific formations of the mucosa." Kolliker (1879, p. 854) has recorded the number of such longitudinal folds of the mucosa found in human embryonic stomachs of different ages.

Anomalies of the Stomach. — Congenital pyloric stenosis is essentially an excessive development of the circular musculature of the pylorus. The other layers in this region, especially the longitudinal layer, may be more or less hypertrophied, and the folds of the mucous membrane are sometimes so highly developed that they appear to obstruct the lumen.

There has been considerable discussion concerning the nature of this condition, the literature of which has been analyzed by Ibraham (1905) and Torkel (1905). It appears to be established that the stenosis is not due to spastic contraction of a normal pylorus, since the muscle layer is actually thickened. A thickening through excessive physiological activity before birth has been suggested. More probably the unknown conditions which normally induce the formation of the sphincter muscle have, in these cases, led to an excessive development. Thus, as Cunningham has recorded, the extremity of the pyloric canal protrudes into the commencement of the duodenum, presenting a striking resemblance to the portio vaginalis of the cervix uteri. In the full-term fetus the protrusion is more marked than in the adult, and in cases of pyloric stenosis it is in all probability still more pronounced. A similar explanation is advocated by Ibraham as according with the relative frequency and remarkable uniformity of the cases observed and the favorable clinical course which they often follow.

DEVELOPMENT OF THE SMALL INTESTINE. 381 Diverticula of the stomach are rare.

Kiiss (1905) has recorded a case in a man 61 years of age. Along the greater curvature, li or 7 cm. from the pylorus, there was a small cavity lined with normal mucous membrane, which penetrated the muscle layer's, pushing a few strands before it. Kiiss states that we are forced to accept a congenital origin for this diverticulum, *' perhaps at the expense of an aberrant outgrowth comparable with the evaginations'of the duodenum which form biliary and pancreatic ducts." Gardiner (1907) has reported a case of accessory pancreas in relation with gastric diverticula, and he refers to Weichselbaum's similar case in which a gastric diverticulum ended in a nodule of pancreatic tissue. According to Orr (1907), W. F. Hamilton has described a stomach with a diverticulum 2 cm. broad and 3 cm. deep, situated on the posterior wall of the cardiac end, near the oesophagus.

The possible embryonic origin of such diverticula will be discussed under anomalies of the small intestine.

It has long been known that stomachs in the adult are occasionally divided more or less completely into two chambers (hourglass stomach), and it was generally believed that some of these cases were congenital. It is now admitted, however, that the great majority of them are due to local physiological contraction of the gastric musculature. Delamare and Dieulafe (1906) described a stomach at birth, which was bilocular, owing to a constriction in the middle part of its corpus. They found that the circular muscle was abnormally thick at the place of constriction, and they attributed this to hypertrophy and not to contraction. But Cunningham (1908) concludes that the hour-glass stomach never arises as a congenital deformity.

Quite distinct from the cases of physiological contraction are those in which a large pars pylorica is separated by a permanent constriction from the corpus. Gardiner (1907) described such a stomach from a three-months child, in which there was a welldeveloped accessory pancreas at the place of constriction. A very similar condition is seen in a 19 mm. embryo in the Harvard Collection. Such cases of "hour-glass stomach" must be distinguished from those which are phases of functional activity.

THE DEVELOPMENT OF THE SMALL INTESTINE. By FREDERIC T. LEWIS.

The early stages in the histogenesis of the small intestine are like those of the stomach, which have already been described. The further differentiation of the epithelial tube proceeds as follows : Vacuoles in the Duodenal Epithelium. — In embryos of 6.5 and 7 mm. the duodenum usually presents a well-defined round lumen, bounded by a 2-3 layered epithelium. In slightly older embryos

382 the epithelium proliferates, and vacuoles are formed within it, especially on the dorsal and right sides of the tube. Later the proliferating epithelium bridges and subdivides the original lumen, as seen in the section of a 10 mm. embryo, Fig. 279, A. Of the three cavities found in this section, the upper one is a vacuole, and the two lower ones are parts of the original lumen. In this embryo there is still a continuous passage from the stomach to the jejunum. The outer surface of the epithelial tube is generally smooth, but occasionally at this stage — niore frequently in somewhat older embryos — the masses of cells surrounding the vacuoles produce local bulgings of the basement membrane. At 22.8 mm. (Fig. 279, B) the outpocketings are so numerous that the epithelium appears folded, and mesenchyma has begun to extend

. p.

>e *z. *.- o ^

i

Fig. 279. — Cross sections of the duodenal epithelium. X 130 diam. A, at 10 mm. (Harvard Collection, Series 1000). B, at 22.8 mm. (Harvard Collection, Series 871). C, at 30 mm. (Harvard Collection, Series 913).

inward between the pockets or folds. In sections the vacuoles cannot be distinguished from the main lumen. A model of the duodenum of this embryo, made by F. P. Johnson, shows that the passage from the stomach to the jejunum is completely blocked by epithelial septa (Fig. 280). At 30 mm. (Fig. 279, C) the vacuoles begin to become confluent so that a central lumen is re-established. The projections between the vacuoles remain as the foundations of villi.

Tandler (1900) was the first to recognize that the duodenal lumen, in embryos from " 30 to 60 days," is normally "more or less completely " obliterated. In an 8.5 mm. specimen he recorded a complete obliteration between the outlets of the duct of the dorsal pancreas and the common bile-duet. At 14.5 mm., when the proliferation is at its maximum, he found that the bile and pancreatic ducts emptied into closed cavities, and that below them the duodenal epithelium formed a solid cord of cells. Forssner (1907) likewise found that, in places, the lumen was completely obliterated in embryos of 11.7, 14, and 22.7 mm. ; and at 30.5 mm.

DEVELOPMENT OF THE SMALL INTESTINE.

383

he described transverse septa dividing the lumen into compartments. Other embryos, of 18.5, 21, and 31 mm. respectively, showed no epithelial vacuoles or occlusions. Schridde (1908) failed to find a solid stage.

Tandler considered that the cause of the occlusion was the resistance exerted upon the expanding epithelium by the surrounding mesenchyma. He found that the diameter of the mesodermal tube of the duodenum increased very slowly in embryos from 7 to 15 mm., whereas from 15 to 20 mm. the increase is rapid. Forssner has confirmed this observation, and thinks it " not improbable that purely mechanical factors play a part in producing occlusions." Both Tandler and Forssner have compared the vacuolization in the duodenum with* that in the oesophagus.

Vacuoles {Diverticula) in the Jejunum and Ileum. — The lower portion of the small intestine never presents a subdivided lumen such as is found in the duodenum, but its epithelium contains scattered vacuoles, which develop in a very characteristic manner. These vacuoles occur chiefly along the portion of the intestine found within the umbilical cord, and they are situated along the convex surface of the intestinal coils, opposite the mesenteric attachment.

Fig. 280. — Model of the duodenum of a 22.8 mm. embryo (Harvard Collection, Series 871), seen in longitudinal section. X 120 diam. (After F. P. Johnson.) In an embryo of 14.5 mm. there are three of these structures, all of which are near the bend of the primary loop of intestine. In a 16 mm. specimen seven were counted, and at 22.8 mm. thirty-two were present.

The intestinal vacuoles are first indicated by a concentric arrangement of the basal nuclei, and in this stage they have been described as "buds" or " pearls.' 1 In the centre of such a bud a small cavity can often be detected (Fig. 281, A). In later stages the cavity communicates with the intestinal lumen, and the bud forms a knob-like basal projection (Fig. 281, B). These projections often have a somewhat constricted neck, and the overhanging portion may become asymmetrical, extending aborally along the intestine. Thus Fig. 281, C, is an aboral section of the diverticulum shown in B. Four of the thirty-two diverticula in the 22.8 mm. embryo project aborally. One diverticulum, longer than any of the others, extends laterally so that its tip penetrates the dense mesenchyma of the muscularis (Fig. 281, D). Usually they are in

384 close relation with the epithelial layer, and they cause no disturbance in the course of the circular muscle fibres. In older embryos (Fig. 281, E and F) the folded appearance of the epithelium renders the detection of the diverticula more difficult. It is probable that, by the enlargement of their necks, some of them are incorporated in the general epithelial layer. Others, however, retain their identity. One of these was found and modelled by F. P. Johnson in an embryo of 134 mm., — a stage when the villi are well developed and the intestinal glands are being formed (Fig. 282). Some of the glands open into the base of the diver

^M

Fig. 281. — Cross sections of the epithelial tube of the intestine, showing the development of diverticula . X 130 diam. A-D, from an embryo of 22.8 mm. (Harvard Collection, Series 871). E and F, from an embryo of 30 mm. (Harvard Collection, Series 913).

ticulnm. Around it the mesenchyma is dense and suggests the formation of lymphoid tissue. This is apparently the oldest embryo in which such a structure has been found, and they are not known to occur after birth.

The intestinal diverticula were described independently by Keibel (1905) and Lewis and Thyng (1908). Keibel noted and figured the two stages in their development (buds and diverticula) and recorded their presence in several mammals, including man. Lewis and Thyng described similar structures, but included with them certain more compact buds which occur chiefly on the dorsal wall of the intestine in the lower duodenal region. These were found frequently in the pig. In an 18.1 mm. human ernbryo there are two buds of this sort situated on the dorsal wall of the intestine as it turns forward to enter the umbilical cord. Lewis and Thyng compared the diverticula with somewhat similar structures found along various epithelial tubes, such as the mammalian bile-ducts and the large intestine in amphibia. They appear to be localized centres of cell proliferation, which either arise in the outer layers of the intestine or are due to the outward displacement of mitotic cells from the innermost layer. Thus mitotic

DEVELOPMENT OF THE SMALL INTESTINE.

385

figures appear to be limited to the inner laj^er and the diverticula, but their distribution requires further study. Elze (1909) has stated that a sharp distinction should be made between the dorsal diverticula of the upper intestine and the ventral diverticula which arise later lower down. He was the first to record the typical aboral growth of the latter. It is probable that the vacuoles of the cesophagus, stomach, duodenum, and intestine are comparable structures.

TJte Formation of Villi. — The development of villi begins in the upper part of the small intestine and extends downward. In the duodenum their formation is complicated by the presence of

~Mk

Fig. 282. — Model of the intestinal epithelium from an embryo of 134 mm., showing villi, glands,, and, in the centre, a "flask-shaped gland." X 80 diam. (After F. P. Johnson.) the epithelial proliferations described in the preceding section. There, as seen in Fig. 279, B and C, the epithelial tube expands by producing irregular outpocketings. Forssner (1907) agrees with Tandler that the epithelium is invaded by mesenchymal papillae, but the apparent invasion is probably due to irregularities in the expansion of the epithelium. Such elevations as are seen in Fig. 279, C, have been described both as folds and as villi.

According to Meckel (1817), the first elevations are longitudinal folds which become gradually indented along their crests, and are thus broken apart into villi. This interpretation has been defended by Berry (1900), who found folds, but no villi, in a human embryo of 24 mm. At 28 mm. the folds, as seen in his reconstructions, show indications of transverse furrows, as if they were about to break up into blocks or villi, and in later stages he found that villi had replaced the folds. Forssner (1907) agrees with Meckel and Berry. Kolliker (1861), on the contrary, states that the villi arise in the beginning of the third month as wart-like outgrowths of the mesenchymal layer, which push the epithelium before them and become cylindrical. This was confirmed by Barth (1868). Brand (1877) found scattered villi at one and a half months. Voigt (1899), by means of reconstructions of pig embryos, found that depressions and furrows develop on the free surface of the epithelium, marking out areas of greater diameter than the future villi. These apparent epithelial elevations are due to the downgrowth of the surrounding furrows. They are described by Voigt as the bases of the future villi.

Vol. II.— 25

386 Johnson (1910) states that villi begin to develop in 19 mm. embryos. At 22.8 mm. he describes isolated rounded elevations occurring between the pylorus and the duodenal occlusion, and .also in the upper part of the jejunum. In a model (Fig. 283, A) he has shown the transition from the villous portion of the jejunum to the smooth part, and has found that the villi in this region arise independently and not as subdivided folds. In the corresponding portion of the intestine of a 24 mm. embryo, the villi

Fig. 283. — Models showing the development of villi in the upper portion of the jejunum. X 110 diam. (After F. P. Johnson.) A. from an embryo of 22.8 mm. (Harvard Collection, Series 871). B, from an embryo of 24 mm. (Harvard Collection, Series 24).

are more numerous (Fig. 283, B). . Although they are arranged in five more or less definite longitudinal rows they do not appear as subdivided folds. At 30 mm. villi are found throughout the upper half of the intestine, but there are none in the ileum. The latter, in cross section, generally shows a trifoliate or four-lobed lumen, due to longitudinal folds of variable length. As this portion of the intestine expands, these folds seem to be obliterated, but villi arise at that time and it is possible that the villi in the ileum are remnants of the folds. The definite relation between theni described by Meckel and Berry is not shown in Johnson's models. At 42

DEVELOPMENT OF THE SMALL INTESTINE. 387 mm. there are still a few distal coils of the ileum which are without villi. According to Berry they do not extend to the colon at 80 mm., but are found throughout the small intestine at 130 mm.

During these and later stages the villi increase greatly in length, but their diameter remains nearly constant. Many new villi develop among the old ones, and the way in which they are formed is shown in Fig. 279, C. At the bottom of an outpocketing a secondary elevation appears, which increases in height with the expansion of the epithelial tube. By relatively rapid growth these elevations attain a length equal to that of the older villi.

Another explanation for the uniform height of the villi is given by Fusari (1904). He finds that the distal ends of the older villi degenerate and are cast off simultaneously, forming, with the mucus, a sort of membrane. This process " is certainly repeated at least twice." These observations, however, have not been confirmed, and the appearances are perhaps due to post-mortem degeneration.

At 55 mm. approximately 12 villi are seen in a cross section of the middle portion of the intestine, and at 99 mm. there are 25. Berry has estimated that in an 80 mm. embryo there are 50,000 villi in the entire intestine, and at 130 mm. the number has increased to 330,000. He finds that fully developed villi and young villi exist in the growing intestine side by side, and this conclusion is well established by the reconstructions of Voigt, Berry, and Johnson.

The epithelium covering the elevations in the 30 mm. embryo is thinner and more definitely simple than that in the depressions. At 55 mm. the epithelial cells of the villi are columnar, with conspicuous cell walls and bulging top plates. The rounded nuclei are somewhat below the middle of the cells, and the protoplasm of the outer part is remarkably clear. In the hollows between the villi the nuclei are more oval and the cells are more crowded. The protoplasm is granular. Altogether the epithelium of the depressions appears much darker than that of the villi. In both regions, however, there are occasional dense triangular or saucershaped nuclei, apparently belonging with goblet-cells. Sometimes nuclei are seen displaced outward, but these do not resemble the wandering cells of later stages. Baginsky (1882) contrasted the clear cells of the villi with the dense cells in the hollows between them, as seen in the jejunum at 4 months, and he described the depressions as the first stage in gland formation.

The Formation of the Intestinal Glands. — The intestinal glands (of Lieberkuhn) develop gradually among the deeply staining cells in the hollows between the villi, appearing first in the duodenum. As the villi increase in number, the rounded hollows between them give place to narrow clefts, along the base of which knobs and short tubules extend downward. Glands in the form of short tubules are present, near the pylorus, at 78 mm.

388 HUMAN EMBRYOLOGY.

At 91 mm. they occur in the middle part of the duodenum, but below this, in the sections examined, they are still absent. They are found in the middle portion of the small intestine at 120 mm., and their size at 134 mm. is shown in the model, Fig. 282.

Brand (1877) found no trace of the glands at 3 months, and states that they first appear in the upper part of the small intestine in embryos of 3 1-2 months (110 mm.?). He considered that they are epithelial pits due to the partial fusion of the bases of adjacent villi. Barth (1868) had previously stated that they are produced by the upward growth of the surrounding mesenchyma, but Kolliker (1861) had described them as tubular downgrowths of the epithelium. Voigt (1899), Hilton (1902), and Johnson (1910) agree with Kolliker.

New glands arise at first as independent buds at the base of the villi, but the older glands branch dichotomously, as observed by Baginsky in a 7 months ' embryo. Branched glands are frequent at birth, and doubtless the branches subsequently become independent glands. Thus the number of tubules increases through bifurcation, as in the stomach.

Although the epithelium of the glands is darker and in early stages taller than that of the villi, the transition is gradual. The relation between them is similar to that which obtains, in the stomach, between the gastric pits and glands. But in the stomach the epithelium of the glands becomes more sharply differentiated from that of the pits, whereas in the intestine the difference gradually disappears. At 240 mm. it is less marked than at 134 mm. Goblet-cells are then found near the bottom of the glands, but often the fundus is composed of darker, granular cells. This is the condition at birth, when the glands have become approximately one-fifth as long as the villi. It is possible that the dark granular cells represent the cells of Paneth.

The Duodenal Glands. — According to Brand, the duodenal glands (of Brunner) develop from the intestinal glands, beginning in embryos of Sy 2 months, but Baginsky failed to find them at 4 months. In a 78 mm. embryo, near the pylorus, some of the intestinal glands appear to be more tortuous than others and occasionally show lateral bulgings near their blind ends. At 120 mm., which is before the appearance of the muscularis mucosae, certain of them have grown almost to the circular muscle layer, where they terminate in tubules composed of clear cells, entirely unlike the dark cells at the fundus of the adjacent intestinal glands. A longitudinal section through the stomach and duodenum at this stage shows that these duodenal glands are quite close together near the pylorus, but further on in the duodenum they occur at considerable intervals. At 240 mm., as shown in Johnson's model (Fig. 284), the older glands have branched repeatedly. Certain of the bifurcating intestinal glands in this model probably represent the young stages of the duodenal glands.

DEVELOPMENT OF THE SMALL INTESTINE.

389

The secretory cells of the duodenal glands stain a bright yellow with orange G, and exhibit a delicate reticular structure. Thus they resemble the cells of the pyloric glands, which develop at about the same time, and of the cardiac glands, which arise somewhat earlier. The duodenal glands have been regarded as an

Fig. 284. — Model showing developing duodenal glands in an embryo of 240 mm.

F. P. Johnson.)

X 160 diam. (After

extension downward of the pyloric glands, but the considerable morphological differences between them in early stages are against this opinion. In the adult, parietal cells have been found in relation with both the pyloric and duodenal glands (Kaufmann, 1906), but, as already noted, they have been found in the cardiac glands of the oesophagus. They have not been seen in the duodenum of the embrvo.

Outer Layers. — As elsewhere in the digestive tract, the circular muscle layer is the first product of the surrounding mesen

390 HUMAN EMBRYOLOGY.

chyma. In a 10 mm. embryo, in which this layer is distinct in the oesophagus, but has not yet appeared in the stomach, it may be identified in the duodenal region. Tandler, however, states that it arises at 12.5 mm. In later stages it spreads down the small intestine, and at 22.8 mm. it is present at the junction with the colon. The mesenchyma within the muscle layer contains numerous branches of the superior mesenteric vessels, but no lymphatics. In the duodenal region ganglia are present, and they are found, almost entirely, just outside of the muscle layer. They appear to connect with sympathetic trunks which pass on the right side of the pancreas and also below it. In these specimens it is impossible to determine the lower limit of the vagus plexus, which, according to Kuntz (1909), may invade the small intestine. The nerves to the lower part of the small intestine appear somewhat later. At 42 mm. the ganglia are conspicuous, especially along the line of mesenteric attachment.

The longitudinal muscle layer becomes distinct at about 75 mm. At 134 mm. no musculo ris mucosa was seen, but it is present at 187 mm. Apparently this layer appears first in the oesophagus, then in the stomach, and later in the small intestine. Mall (1897 and 1898) inferred that peristalsis occurs in 130 mm. embryos, since in several embryos of this stage he found that the meconium had been propelled downward toward the csecum.

The tunica propria becomes gradually differentiated before the muscularis mucosae has appeared. It is a dense layer of mesenchyma at 99 mm. The lymphatic vessels, which in earlier stages were present in the mesentery, are now found in the submucosa, but they cannot be seen in the propria. At 240 mm. both solitary and aggregate nodules of lymphoid tissue have appeared in the tunica propria. Their relation to the lymphatic vessels could not be determined in the specimens studied. According to the early observers, the lymphoid tissue arises from the epithelium, aud Retterer (1895 and 1897) has more recently defended this interpretation. It was rightly rejected by Stohr (1889), who concluded that "the lymph-nodules of the intestine arise in the tunica propria, or in the adjacent parts of the submucosa, through mitotic division of the round cells (leucocytes) which are found there." Similarly Czermack (1893) has maintained that the lymphoid tissue develops as a "condensation of the mesenchyma." Czermack is probably correct in concluding that the lymphocytes arise in genetic connection with the reticular tissue. The presence of aggregate nodules (Peyer's patches) in the human intestine at six months and later has been recorded by Kolliker (1861). At seven months Baginsky (1882) recognized very distinctly the central lymphatic vessels within the villi of the duodenum.

The development of the circular folds (valvules conniventes) requires further study. In the middle portion of the intestine at

DEVELOPMENT OF THE SMALL INTESTINE. 391 78 mm. (3 months?), as seen in longitudinal section, there are frequent slight elevations of the submucosa, in which the muscularis is not involved. These are so small that they displace upward (in longitudinal sections) only five villi. In similar sections at 240 mm. there are about ten villi on either side of a fold.

Meckel (1817) stated that there was no trace of the circular folds until the seventh month, when they appeared as slight elevations readily obliterated on stretching the intestine. At birth he found them still poorly developed. Delamare (1903), on the contrary, states that at birth they are as numreous and relatively as high as in the adult. According to Hilton, these folds are not found in apes, but are peculiar to the human intestine.

Fischl (1903) has studied the elastic tissue of the intestine. He finds that at birth there is no elastic tissue in the walls of the intestine or stomach, except in connection with blood-vessels. It begins to develop in the first weeks after birth.

Anomalies of the Small Intestine. — In addition to the imperfect torsion of the intestinal loop and the presence of Meckel's diverticulum ilei, which have been described in a previous section, the congenital anomalies of the small intestine include atresia and stenosis, diverticula, and cysts.

Tandler (1900) concluded that the embryological atresia of the duodenmn may sometimes persist and become congenital. He considered this a rare occurrence, since only two cases of intestinal occlusion were found among 111,541 children in Vienna, and nine cases among 150,000 in St. Petersburg. Altogether more than a hundred cases of stenosis or atresia of the small intestine have been described, and rather more than a third of these were found in the duodenum'. Thus they are far more abundant in the duodenum than in any other equal length of the intestinal tract. Kuliga (1903), who reviewed the literature, could not decide between inflammatory and developmental causes for these conditions, but Kreuter (1905) and Forssner (1907) both advocate the ernbryological origin.

The cases vary greatly in degree, and include perforate iris-like folds or valves, complete membranes, and more or less extensive strictures and obliterations of the epithelial tube. Sometimes the muscularis passes smoothly around the blind ends of the divided intestine without extending from one to the other. Cases like that of Preisich (1903), in which, in a boy 6 days old, two valve-like folds were found in relation with the bile and pancreatic ducts, strikingly suggest the conditions in embryos between 15 and 25 mm., and certain of the congenital atresias and stenoses presumably arise at that stage. 13 In other cases, discussed by Forssner, meconium has been found below a complete atresia. This indicates a late origin, possibly through the adhesion of valve-like folds. Moreover, atresia is found in portions of the small intestine where obliteration of the lumen does not normally occur. Such cases may represent the persistence of an abnormal embryological condition. Forssner thinks it probable that exceptionally an epithelial occlusion may be found in all parts of the embryonic intestine.

13 The 19 mm. embryo in the Harvard Collection, which has an abnormally shaped stomach with an accessory pancreas, shows also a distinct local constriction of the duodenal epithelium. There is an actual stenosis of the descending part of the duodenal loop.

392 Diverticula of the duodenum, especially near the outlets of the pancreatic ducts, are relatively common. According to Jach (1899), who found but one case in 200 bodies, Schiippel found seven instances in 45 bodies. They are generally round sacs, opening into the duodenum by clear-cut, circular orifices. Since they are not covered by the muscularis, but push their way through it, they have been described as hernias of the mucous membrane, and as false diverticula, in distinction from the true Meckel's diverticulum. The latter is covered by the muscular coats. Jach believes that they are generally pulsion diverticula, produced by the distention of the upper part of the duodenum following an obstruction lower down. The obstruction may be a cicatricial contraction, or the pressure from a displaced transverse colon. Their occurrence about the outlets of the bile and pancreatic ducts has been attributed to a deficiency in the muscularis where the ducts penetrate it. But Letulle (1899), who has described two cases, concludes that they are undoubtedly of early embryonic origin. Lewis and Thyng (1908) have stated that the diverticula observed in the embryo may possibly give rise to those in the adult. In Fig. 285 their drawing of a model of a duodenal divertie

D. parted:

D. chol.

Div.

A

Fig. 285. — Diverticula of the duodenum. .4, in an embryo of 13.6 mm. (Harvard Collection, Series 839). X 55 diam. (From a model by F. W. Thyng.) B, in an adult. (After C. M. Jackson.) In B the outline of the pancreas is dotted. D. chol., common bile-duct; D. pane, d., duct of the dorsal pancreas; Div., diverticulum ; St., stomach.

ulum from a 13.6 mm. embiyo is placed beside Jackson's sketch of a large diverticulum, 3.5 cm. deep, found in a man of 50 (Jackson, 1908), and the correspondence in location is striking. It is possible that some of the duodenal diverticula are congenital, although apparently no case has yet been recorded at birth (Fischer, 1901).

Occasionally a single diverticulum has been found in the jejunum or in the ileum, but more often there are multiple diverticula. They occur usually in old people, and differ from those of the embryo in their greater relative size and larger number, in occurring only near the mesenteric attachment, and in their distribution which includes the colon. Like the diverticula of the oesophagus, they have been found in relation with the blood-vessels, and have been attributed both to traction by the vessels and to pulsion along the path of the veins as they cross the musculature to enter the mesentery. Hansemann (1896) has produced them experimentally by distending the intestine with water. It is improbable, as stated by Elze (1909), that there is any genetic connection between such structures and the diverticula of the embryo.

Cysts derived from the intestine may be found at birth. Usually they are correctly ascribed to a detached Meckel's diverticulum, even when found within the mesentery (Hennig, 1880; Roth, 1881; Dittricb, 1885). These cysts are occa

DEVELOPMENT OF THE LARGE INTESTINE. 393 sionally very large (22 cm. long). Their walls include all the layers of the intestine and may contain aggregate nodules. The epithelium is sometimes smooth and ciliated, but it may exhibit more or less perfect glands and villi. In one of Roth's cases, there were two cysts in the abdomen and one in the thorax, and it is stated that they may have arisen in loco from the duodenum and oesophagus. In a pig embryo of 20 mm., Lewis and Thyng have figured a mesenteric cyst which had become detached from the intestine in the lower duodenal region, and it is possible that certain of the congenital intra-mesenteric cysts have a similar origin. The relation of intestinal diverticula and cysts to an accessory pancreas will be considered with the pancreas.

Several small oval cysts have been found by F. P. Johnson among the duodenal villi of a 7 months' embryo. They appear to be distended with mucus, derived from the small group of glands emptying into their basal portions. The epithelium which lines the cysts is separated from the surface epithelium by a thin layer of connective tissue. These structures resemble the cystic glands which Stohr (1898) has figured in the vermiform process of a 5 months' embryo. The closure of the neck of the flask-shaped gland shown in Fig. 282 would apparently produce a similar structure.

THE DEVELOPMENT OF THE LARGE INTESTINE.

By FREDERIC T. LEWIS.

Early Development. — In a 10 mm. embryo the large intestine consists of an epithelial tube, an undifferentiated layer of mesenchyma, and, except along the mesenteric attachment and near the pelvic termination, a layer of peritoneal epithelium. The entodermal epithelium shows nsnally two rows of nuclei. Mitotic figures are limited to the row next the lumen. The lumen is circular, and, though minute in places, it is continuous throughout. There is a marked local dilatation of the lumen in the region of the caecum, and a gradual enlargement downward in the rectal portion. The diameter of the colon is about equal to that of the lower portion of the small intestine, but it is less than that of the duodenum, the epithelium of which has begun to proliferate.

At 14.5 and 16 mm. the large intestine is still a round tube with a well-defined lumen, gradually enlarging from the colon toward the cloaca. At 22.8 mm. the lumen of the lower part of the rectum has become elliptical and the long axis is transverse. Passing upward the lumen gradually becomes circular. Near the caecum the inner surface of the epithelium shows some irregularities due to the varying height of the cells. Here, in cross section, the lumen is stellate or polygonal, but the circumference of the epithelium is circular. Patzelt (1883) has described the same condition in cat embryos of 25 and 33 mm.

In a 37 mm. embryo the lumen of the greater part of the ascending colon is pentagonal, but the circumference is circular. Toward the right (or hepatic) flexure the lumen becomes round, and so continues into the descending colon. Then it assumes a

394 three-lobed form (in places four-lobed), but the circumference still remains nearly circular. After reaching the part of the rectum with a long transverse axis, the lumen shows additional lobes and the circumference becomes indented. Thus, in cross section, the lower part of the rectum shows five or six folds.

The transverse colon at 42 mm. is shown in Fig. 286, A. The cells at the bottom of the three outpocketings are lower than the others, and show a characteristic pearl or bud-like arrangement, suggesting the diverticula of the small intestine. They do not produce local pockets, however, but elongated ridges. Toward the rectum the epithelial irregularities are more marked. They are shown in longitudinal section in Fig. 286, B and in cross section in Fig. 286, C. Between the sheaf-like bundles of tall cells with superficial nuclei, there are concentric groups of short ones with

Fig. 286. — The epithelium of the large intestine in a 42 mm. embryo (Harvard Collection, Series 838). X[130 diam. A, cross section of the transverse colon. B, longitudinal section of the sigmoid colon. C, cross section through the upper part of the rectum.

basal nuclei. The latter appear to be growing outward, and by the shortening of the intervening cells the epithelium becomes folded. Thus in C there are three folds, but the places where two others will arise are clearly indicated, and lower down in the rectum five or six folds are found.

There is no solid stage in the development of the large intestine. The atresia described by Kreuter (1905) at the beginning of the fifth week has not been found by Forssner (1907). There are no diverticula. Small vacuoles are found at 37 mm. in connection with the nests of low cells, and similar vacuoles have been observed in the stomach, but these are never conspicuous as in the oesophagus. The expansion of the epithelial tube occurs slowly, so that at 37 and 42 mm. the large intestine is very much smaller than the adjacent coils of the jejunum and ileum, in which villi are conspicuously present.

Villi and Glands. — With the continued growth of the large intestine the outpocketings bifurcate, and, as seen in cross sections,

DEVELOPMENT OF THE LARGE INTESTINE.

395

the number of elevations projecting toward the lumen increases. In a section of the transverse colon at 55 mm. there are five or six projections; at 73 mm. there are about ten; and at 99 mm. as many as twenty. A model of the transverse colon at 99 mm. shows that these elevations are true villi (Fig. 287). At their bases there are irregular epithelial clefts and pockets which give rise to the glands. The glands continue to grow downward, and they multiply through bifurcation. The cells at the bottom of the depressions are granular and dark, but the villi are covered with clear elon

Fig. 287. — Model showing the villi of the transverse colon in an embryo of 99 mm. X 120 diam. (After F. P. Johnson.)

gated cells, apparently containing mucus. Patzelt found occasional goblet-cells in a 75 mm. embryo, and he notes that the cuticular border is distinct. The relation between the dark cells below and the clear cells above is similar to that seen in the stomach and small intestine.

Meckel in 1817 described the villi of the large intestine at 3 months as much lower than those of the small intestine, but quite as numerous. In the fourth month he found that they are not only considerably smaller, but less club-shaped and more scattered. By the eighth month the villi have gradually given place to very low, slightly indented longitudinal folds, which produce irregularities on the inner surface of the intestine. Kolliker (1861) stated that at the fifth month the villi begin to fuse from their bases upward, around the gland outlets, and that this process is completed in the seventh month. Brand (1877) described the development of septa between the villi, beginning below and extending upward, transforming the spaces between the villi into prolongations of the glands. At the sixth month the tips of the villi are reached by the septa. Patzelt (1883) states that he can confirm the observations of Kolliker and Brand regarding the disap

396 HUMAN EMBRYOLOGY.

pearanee of the villi. But Hilton (1902) has found that the villi shorten and disappear as the intestine enlarges, without forming outer portions of the glands. This accords with Meckel's original description and is apparently correct. The gradual disappearance, whereby the villi become slight elevations, probably accounts for the discrepancy in determining the time of their extinction (Meckel, during the eighth month; Kolliker, in the seventh month; Brand, at the sixth month). Even at 240 mm. the villi are quite low.

The Outer Layers. — In embryos of 14.5 and 16 mm. the mesenchyma of the large intestine contains branches of the inferior mesenteric vessels. The circular muscle layer, which is present at this stage in the ileum, is still absent from the large intestine. It appears first in the rectal region and spreads upward. Thus at 22.8 mm. it is present only in the lower part of the large intestine, where it is in relation with branches of the pelvic sympathetic ganglia. At -42 mm. it is found throughout the colon. The longitudinal layer appears as a crescentic condensation along the mesenteric attachment of the transverse colon at 75 mm. In the transverse colon at 99 mm. the mesenteric taenia is still the most prominent part of the longitudinal muscle, but the other two taeniae are indicated. There is probably a thin layer of longitudinal muscle in the intervals between the taeniae. In the rectum at this stage the longitudinal layer is well defined. At 187 mm. the muscularis mucosae, is distinct in the rectum, but is apparently absent from the transverse colon. It is present there at 240 mm.

Lymphatic vessels are abundant along the mesenteric attachment of the rectum in early stages (37 mm.). At 120 mm. the lymphatic vessels in this region have extended into the submucosa, and lymph-glands are developing outside of the muscularis. At 187 mm. lymphoid tissue is present in the propria of the rectum, forming nodules and extending into the submucosa. Baginsky found developing nodules in the submucosa of the colon at 4 months. At birth the nodules are abundant.

The transverse folds of the rectum are seen in longitudinal sections at 120 mm. The haustra or sacculations of the colon, according to Meckel, are not present until the end of the fifth month and they first appear in the transverse colon. In this region they are distinct at 240 mm. Corresponding with the external indentation, which bounds the sacculation, there is a prominent internal fold of the mucosa. The appendices epiploiccc were found by Baginsky at 4 months. Meckel had said that they are distinct in the fifth month, although they contain quite as little fat as the omentum.

The Vermiform Process. — At 16 mm. the vermiform process is an epithelial tube surrounded by undifferentiated mesenchyma. Its lumen is clear-cut and round. At 55 mm. the liunen has become lobed, resembling that of the colon in earlier stages. The circular muscle is present as a well-defined layer, but the longitudinal

DEVELOPMENT OF THE LARGE INTESTINE. 397 muscle has not yet appeared. At 120 mm. (about 4 months) villi are present, and between them there are bifurcating glands or pits. Brand (1877) found only villi at 3>< months, but at ± l / 2 months the glands had appeared. Some of them showed "lateral outpocketings" which Brand thought were perhaps concerned with the later increase in the number of glands. At 120 mm. the tunica propria, because of its crowded nuclei, is quite distinct from the underlying submucosa, and the contrast between them is greater than in the adjacent colon. At 140 mm. Stohr (1898) found small "groups of leucocytes" close beside blood-vessels, not only in the deep connective-tissue layer, but also within the villi. In the tunica propria they form compact masses of cells which are the beginnings of the nodules. At this early stage scattered leucocytes wander into the epithelium which covers the tips of the villi.

In the fifth month (170 mm.?) Stohr has found that the muscularis mucosa is indicated in the caecum but is absent from the vermiform process. Villi are still present. The glands of the vermiform process vary greatly in diameter and length, and some of them extend almost to the circular muscle. At this stage Stohr has described a degeneration of some of the glands, beginning with a characteristic thickening of the connective tissue which surrounds them. The goblet-cells at the neck of a degenerating gland first become flattened, and then give place to a solid epithelial strand. Later the strand ruptures, and the detached basal portion, after becoming cystic, undergoes involution. The last remnants of such a gland are small groups of epithelial cells, surrounded by thick connective-tissue capsules. Stohr finds that degenerating glands are relatively less numerous at six months, and he infers that the degeneration may be limited to embryonic stages.

At 240 mm. both the muscularis mucosae and the longitudinal muscle layer are present. The glands are branched and some of them are long enough to cause a local bulging of the muscularis mucosae. Occasionally they appear to penetrate it. In this specimen detached glands were not found. The propria contains diffuse lymphoid tissue and five nodules, of which the two most highly developed are near the distal end of the vermiform process.

At birth the glands are still branched. The number of lymphnodules has increased, and five were seen in a single longitudinal section of the distal third of a vermiform process 35 mm. long. Berry and Lock (1906) have stated that the lymph-nodules increase rapidly after birth, but these authors are in error in denying the presence of lymph-nodules at term.

Anomalies of the Large Intestine. — It is generally agreed that the sigmoid colon is relatively long and freely movable at birth.

398 HUMAN EMBRYOLOGY.

Bourcart (1863) has determined its course and form in 150 cases. Frequently in the adult the sigmoid colon is excessively long, forming two or three very large coils. According to Frommer (1902), Curschmann found an elongated colon in 15 of the 233 bodies which he examined, and he regarded it as a persistence of the infantile condition. Concetti (1899) recognizes three types of congenital cases, which cause habitual constipation : 1, those with simple elongation of the descending and sigmoid colon; 2, cases in which, in addition to the elongation, the colon is dilated and its walls hypertrophied ; 3, cases in which there is local dilatation of the lowest part of the colon, above which there is apt to be a region of compensatory hypertrophy. In the third group Concetti describes a case at 2 1-2 years, in which the longitudinal muscle was completely absent from the portion of the colon just above the rectum, and the transverse muscle was thinner than the muscularis mucosa?. The embryological factors which control the length of the colon are unknown.

Stenosis and atresia are less frequent in the colon than in the small intestine, but they present the same forms. They occur both as membranes and as complete interruptions of the intestinal tube, with blind ends more or less widely separated (Forssner, 1907). In the large intestine of the embryo there is ordinarily no occlusion such as occurs in the duodenum, and these cases are essentially abnormal. The same is true of the doubling of the cascum and colon described by Lockwood (1882).

Diverticula of the colon are frequent, but apparently they are acquired late in life. Hansemann (1896) finds that they are not limited to the mesenteric attachments, but may occur on the convex side of the intestine, sometimes projecting into the appendices epiploicse. Unlike the false diverticula of the small intestine with which they are sometimes associated, he states that the outpocketings of the colon often push the atrophic muscle before them and are "true dilatation diverticula." Hedinger (1904) has described a case of congenital diverticula of the vermiform process. In the distal third of the vermiform process of a new-born child there were numerous outpocketings, which either only reached the muscularis or extended nearly through it so as to produce an uneven serous surface. Hedinger considers that in his case alone, the congenital origin of a diverticulum of the digestive tube has been established.

Contents of the Intestine. — When the glands of the digestive tract begin to secrete, their secretions together with desquamated and disintegrating entodermal cells are found in the intestinal tube. These become mixed with amniotic fluid, containing lanugo hairs and fatty material from the vernix caseosa, which has been swallowed. In early stages the fluid is yellowish in color, and at birth it is still light colored in the upper part of the intestine. Toward the rectum it gradually becomes dark brown or dark green, and is known as meconium (a Greek term for the juice of the poppy and for sepia). The color is due to bile pigments. Schenk (1896) states that in embryos of four months the meconium

DEVELOPMENT OF THE ALIMENTARY (ANAL.

appears as a bright yellow or pale greenish fluid, and that it tills the entire large intestine in embryos of five months. In later stages it becomes dark brown. But, according to Tourneux ( 1909) . it does not pass the valve of the colon in embryos of five and months; it is only from the seventh to the ninth month that it passes into the large intestine, where it Incomes greenish brown. Bacteria are absent and there is no gas in the embryonic intestine.

LITERATURE. (On the development of the oesophagus, stomach, and intestine.) Ascoli, C. : Ueber die histologische Entwicklung der mensch lichen Magenschleim haut. Verb., der anat. Gesellcli., Vers. 14. Erg.-Heft zu Anat. Anz. Bd. 18, S. 149-150. 1900. Baginsky, A. : Untersuchungen iiber den Dannkanal des menschlichen Kindes.

Yirchow's Arch. Bd. 89, S. 64-94. 1882. Barth: Beitrag zur Entwicklung der Darmwand. Sitzb. der Akad. der Wiss.

Wien. Bd. 58, S. 129-136. 1868. Bensley, R. R. : The Cardiac Glands of Mammals. Amer. Journ. of Anat. Vol.

2, p. 105-156. 1902. Berry, J. M. : On the Development of the Villi of the Human Intestine. Anat.

Anz. Bd. 17, S. 242-249. 1900. Berry, R. J. A., and Lack, L. A. H. : The Vermiform Appendix of Man and the Structural Changes therein coincident with Age. Journ. of Anat. and Phv*.

Vol. 40, p. 247-256. 1906. Brand, E. : Beitrage zur Entwicklung der Magen- und Darmwand. Yejrh. der phys.-med. Gesellschaft in Wiirzburg. Bd. 11, S. 243-255. 1877. Broman, I. : Ueber die Existenz eines bisher unbekannten Kreislaufes in einbryon alen Magen. Anat. Anz. Bd. 23, S. 390-391. 1903. Bourcart : De la situation de l'S iliaque chez les nouveau-nes dans ses rapports avec l'etablissement de l'anus artificiel. These de Paris. 1S63. Coakley, C. G. : The Arrangement of the Muscular Fibres of the (Esophagus.

Researches of the Loomis Laboratory. Univ. of the City of New York.

Vol. 2, p. 113-114. 1892. Concetti, L. : Ueber einige angeborene, bei Bandera die habit uelle Yerstopfung hervorrufende Missbildungen des colon. Arch, fiir Kinderheilk. Bd. 27.

S. 319-366. 1899. Cunningham, D. J. : The Varying Form of the Stomach in Man and the Anthropoid Ape. Trans, of the Roy. Soe. of Edinburgh. Vol. 15, p. 9-17. 1908. Czermack, N. : Einige Ergebnisse iiber die Entwicklung, Zusammensetzung und Funktion der Lymphkndtchen der Darmwand. Arch f. mikr. Anat. Bd. 42.

5. 581-632. 1893.

Delamare, G. : Recherches sur la structure de l'intestin grele du nouveau-ne.

C. R. Soc. bid. Vol. 55, p. 1151-1152. 1903. Delamare, G., and Dieulafe: Estomac de nouveau-ne a tendance biloculaire.

Journ. de l'Anat. et de la Phys. Vol. 42, p. 624-629. 1906. Disse : TVeitere Mitteilungen iiber das Verhalten des Schleims in Magen von menschlichen Embryonen und von Xeugeborenen. Beitrage zur Klinik der Tuberkulose. Bd. 4, S. 227-238. 1905. Dittrich, P.: Ueber zwei seltenere, auf mangelhafte Involution des ductus omphalomesaraicus zu beziehende Darmbefunde. Zeitschr. fiir Heilk. Bd.

6, S. 277-286. 1 400 HUMAN EMBRYOLOGY.

Eberth, C. J. : Verirrtes Magen-Epithel in der Speiserbhre. Fortschritte der Median. Bd. 15, S. 251-254. 1897. Ebstein, W. : Beitrage zur Lehre vom Bau und den physiologischen Funktionen der sogenannten Magenschleimdriisen. Arch. f. rnikr. Anat. Bd. 6, S. 515 539. 1870. Elze, C. : Beitrag zur Histologic des embryonalen Saugetierdannes. S. 1-18.

Inaug. Diss., Freiburg. 1909. Fischer, M. H. : False Diverticula of the Intestine. Journ. of Exp. Med. Vol. 5, p. 332-352. 1901. Fischl, R. : Beitrage zur normalen und pathologischen Histologic des Sauglings magens. Zeitschr. fur Heilk. Bd. 12, S. 395-446. 1891. Ueber das Elastingewebe des Sauglingsdarmes. Jahrb. fiir Kinderheilkunde.

Bd. 57, S. 439-443. 1903. Forssner, H. : Die angeborenen Dann- und Oesophagustresien. Anat. Hefte. Bd.

34, S. 1-163. 1907. Frommer, A. : Zur Casuistik der Anomalien des Diekdarmes. Arch, fiir klin.

Chir. Bd. 67, S. 27-58. 1902. Fusari, R. : Sur les phenomenes que l'on observe dans la muqueuse du canal digestif durant le developpenient du foetus humain. Arch. ital. de biol.

Vol. 42, p. 205-212. 1904. Gardiner, J. P. : A Case of Congenital Hour-glass Stomach with Accessory Pancreas. Journ. of the Amer. Med. Assoc. Vol. 49, p. 1598-1600. 1907. Hansemann, D. : Ueber die Entstehung falscher Darmdivertikel. Virchow's Arch. Bd. 144, S. 400^05. 1896. Happich, C. : Ueber Oesophagusmissbildungen. S. 1-65. Inaug.-Diss. Marburg 1905. d'Hardivillier : Sur l'existence d'un epithelium prismatique simple dans la partie superieure de Foesophage du foetus humain. L'Echo medical du nord. Vol.

1, p. 521-522. 1897. Hedinger, E. : Kongenitale Divertikelbildung im processus vermiform is. Virchow's Arch. Bd. 178, S. 25^3. 1904. Hennig, C. : Cystis intestinalis, cystis citra oesophaguni bei einem Neugeborenen.

Zentralbl. fur Gyn. Bd. 4, S. 398-399. 1880. Hewlett, A. W. : The Superficial Glands of the Oesophagus. Journ. of Exp.

Med. Vol. 5, p. 319-331. 1901. Hilton, W. A. : The Morphology and Development of Intestinal Folds and Villi in Vertebrates. Amer. Journ. of Anat. Vol. 1, p. 459-504. 1902. His, W.: Die Haute und Hohlen des Korpers. S. 1-34. Basel 1865. His, jun., W. : Ueber die Entwicklung des Bauchsympathicus beim Hiihnchen und Menschen. Arch, fiir Anat. und Entw. Supplementband. S. 137-170.

1897. Ibraham, J. : Die angeborene Pylorusstenose im Sauglingsalter. S. 1-120.

Berlin 1905. Jach, E. : Ueber Duodenaldivertikel. S. 1-15. Inaug.-Diss. Kiel 1899. Jackson, C. M. : An Unusual Duodenal Diverticulum. Journ. of Anat. and Phys.

Vol. 42, p. 219-220. 1908. Jahrmaerker, E. : Ueber die Entwicklung des Speiserohrenepithels beim Menschen. S. 1-43. Inaug.-Diss. Marburg 1906. Johnson, F. P.: The Development of the Mucous Membrane of the OEsophagus, Stomach, and Small Intestine in the Human Embryo. Amer. Journ. of Anat. Vol. 10, p. 521-561. 1910. Kalopothakes, M. : Contribution a l'etude de la structure normale de Pestomac chez le foetus et le nouveau-ne. Bull, de la Soc. anat. de Paris. Annee 69, p. 685-696. 1894.

DEVELOPMENT OF THE ALIMENTARY CANAL. 401 Kaufmann, M. : Ueber das Vorkoruruen von Belegzellen im pylorus und duodenum des Menschen. Anat. Anz. Bd. 28, S. 465^174. 1906. Keibel, F. : Zur Embryologie des Menschen, der Alien und der Halbaffen. Verh.

der anat. Gesellschaft. Erg.-Heft zu Anat. Anz. Bd. 27, S. 39-50. 1905. Keibel, F., u. Elze, C. : Normentafeln zur Entwicklungsgeschichte der Wirbel tiere. Heft 8, S. 1-314. Jena 1908. Kirke, E. G. : On the Histogenesis of Gastric Glands. Aruer. Journ. of Anat.

Vol. 10, p. 473-520. 1910. Klein, E.: Ueber die Verteilung der Muskehi des oesophagus. Sitz.-Ber. der kais Akad. Wiss. Wien. Bd. 57, S. 1111-4.121. 1868. Kolliker, A. Entwicklungsgeschichte des Menschen. S. 1-468. Leipzig 1861. Entwicklungsgeschichte des Menschen und der hoheren Tiere. S. 1-1033.

2 Aufl. Leipzig 1879. Kreuter, E. : Die angeborenen Verschliessungen und Verengerungen des Darm kanals im Liehte der Entwicklungsgeschichte. Deutsche Zeitschr. fiir Chir.

Bd. 79, S. 1-89. 1905. Kuess, M. G. : Estoniac a cellule. Bull, et mem. de la soc. anat. de Paris. Annee SO, p. 561-567. 1905. Kultga, P. : Zur Genese der kongenitalen Dunndarrustenosen und Atresien. Beitr.

zur path. Anat. und zur allg. Path. Bd. 33, S. 481-584. 1903. Kuntz, A. : The Role of the Vagi in the Development of the Sympathetic Nervous Svstem. Anat. Anz. Bd. 35, S. 381-390. 1909. Laskowsky : Ueber die Entwieklung der Magenwand. Sitz.-Ber. der kais. Akad.

der Wiss. Wien. Bd. 58, S. 137-143. 1868. Letulle, M. : Malformations duodenales. La presse medicale. Annee 7, p. 13.

1899. Lewis, F. T., and Thyng, F. W. : The Regular Occurrence of Intestinal Diverticula in Embryos of the Pig, Rabbit, and Man. Amer. Journ. of Anat.

Vol. 7, p. 505-519. 1908. Leyen, E. von der: Ueber die Schleinxzone des menschlichen Magen- und Darm epithels vor und nach der Geburt. Virchow's Arch. Bd. 180, S. 99-107.

1905. Lockwood, C. B. : On Abnormalities of the Caecum and Colon with Reference to Development. Brit. Med. Journ. Vol. 2, p. 574-576. 1882. Mall, F. P. : Ueber die Entwieklung des menschlichen Darmes und seiner Lage beim Erwachsenen. Arch, fiir Anat. unt Entw. Supplementband. S. 403 434. 1897. Development of the Human Intestine and its Position in the Adult. Bull, of the Johns Hopkins Hosp. Vol. 9, p. 197-208. 1898. Maurer, F. : Die Entwieklung des Darmsystems. In Hertwig's Handbuch der vergleichenden und experimentellen Entwicklungslehre der Wirbeltiere. Bd.

2, S. 109-252. 1906. Erschien 1902. McGill, C. : The Early Histogenesis of Striated Muscle in the (Esophagus of the Pig and the Dogfish. Anat. Record. Vol. 4, p. 23-47. 1910. Meckel, J. F. : Bildungsgeschichte des Darmkanals der Saugetiere und namentlich des Menschen. Deutsehes Arch, fiir die Phys. Bd. 3, S. 1-84. 1817. Minot, C. S. : Human Embryology. S. 1-815. New York 1892.

Sollen die Bezeichnungen " Somatopleura " und " Splanchnopleura " in ihrem urspriinglichen rightigen oder in dem in Deutschland gebrauchlich geword enen Shine verwendet werden? Anat. Anz. Bd. 19, S. 203-205. 1901. Neumann, E.: Flimmerepithel im oesophagus menschlicher Embryonen. Arch.

f. mikr. Anatomie. Bd. 12, p. 570-574. 1876. Die Metaplasie des fetalen Oesophagusepithels. Fortschritte der Medizin.

Bd. 15. S. 366-369. 1897.

Vol. II.— 26

402 HUMAN EMBRYOLOGY.

Orb, A. E.: Hour-glass Stomach, journ. of Anat. and Phys. Vol. 41, p. 49-50.

1907. Patzelt, V. : Ueber die Entwicklung der Diekdannschleirnhaut. Sitz.-Ber. der kais.

Akad. der Wiss. Wien. Bd. 86, S. 145-172. 1883. Preisich, K. : Angeborener doppelter Klappenverschluss des duodenum. Jahrb.

fiir Kinderheilk. Bd. 57, S. 346-349. 1903. Remak, R. : Untersuchungen iiber die Entwieklung der Wirbeltiere. S. 1-194.

Berlin. 1855. Retterer, E. : Sur Forigine des follieules clos du tube digestif. Verb, der anat.

Gesellschaft. Vers. 9. Erg.-Heft z. Anat. Anz. Bd. 10, S. 31-39. 1895. Histogenese du tissu reticule aux depens de l'epitnelium. Verb, der anat.

Gesellschaft. Vers. 11. Erg.-Heft z. Anat. Anz. Bd. 13, S. 25-37. 1897. Reyher, P. : Ueber die Ausdebnung der Sehleimbildung in den Magenepitbelien des Menscben vor und nach der Geburt. Jahrb. fiir Kinderheilk. Bd. 60, S. 16-28. 1904. Riebert, H. : Zur Kenntnis der Traktionsdivertikel des oesophagus. Virchow's Arch. Bd. 167, S. 16-29. 1902. Ribbert, H. : Noch einmal das Traktionsdivertikel des oesophagus. Virchow's Arch. Bd. 184, S. 403-413. 1906. Riebold, G. : Ein Beitrag zur Lehre von den Oesophagusdivertikeln. Virchow's Arch. Bel. 173, S. 395-466. 1903. Weitere Untersuchungen iiber die Pathogenese der Traktionsdivertikel des oesophagus. Virchow's Arch. Bd. 192, S. 126-226. 1908. Roth, M. : Ueber Missbildungen im Bereicb des ductus omphalo-mesentericus.

Virchow's Arch. Bd. 86, S. 371-390. 1881. Ruckert, A. : Ueber die sogenannten oberen Cardiadriisen des oesophagus.

Virchow's Arch. Bd. 175. S. 16-32. 1904. Salvioli, I. : Quelques observations sur le mode de formation et d'accroissement des glandes de l'estomac. Arch. ital. de biol. T. 14, p. 71-80. 1891. Schaffer, J. : Die oberen cardialen Oesophagusdriisen und ihre Entstehnng.

Virchow's Arch. Bd. 177, S. 181-205. 1904. Schenk, S. L. : Beitrag zur Lehre von den Organanlagen im motorischen Keim blatte. Sitz.-Ber. der kais. Akad. der Wiss. Wien. Bd. 57, Abt. 2, S. 189 202. 1S68. Lehrbnch der vergleichenden Embrvologie der Wirbeltiere. S. 1-198. Wien 1874. Lehrbnch der Embryologie des Menscben und der Wirbeltiere. S. 1-688.

Wien und Leipzig. 1896. Schridde, H. : Die Entwicklungsgescbichte des menschlichen Speiserohrenepithels.

S. 1-101. Wiesbaden. 1907. Ueber die Epithelproliferation in der embryonalen menschlichen Speiserohre.

Virchow's Arch. Bd. 191, S. 178-192. 1908. Schultze, O. : Grundriss cler Entwicklungsgescbichte. S. 1-468. Leipzig 1897. Schwalbe, K. : Ueber die Schaffer'sehen Magenschleimhautinseln der Speiserohre.

Virchow's Arch. Bd. 179, S. 60-76. 1905. Sewall, H. : The Development and Regeneration of the Gastric Glandular Epithelium during Fetal Life and after Birth. Journ. of Phys. Vol. 1, p. 321 33S. 1879." Stohr. P. : Ueber die Lymphknotchen des Dartnes. Arch. f. mikr. Anat. Bd. 33, S. 255-2S3. 1SS9~ Ueber die Entwieklung von Darmlymphknotehen und iiber die Ruckbildung von Darmdriisen. Arch. fiir. mikr. Anat. Bd. 51, S. 1-55. 1898. Strecker, F. : Der Vormagen des Menscben. Arch, fiir Anat. und Entw. S.

119-188. 1908.

DEVELOPMENT OF THE LIVER. 403 Neue Anschauungen iiber Eutstehung und Waehstum von Magendriisen beim Mensehen. Arch, fur Anat. and Entw. S. 189-236. 1908. Taxdler, J.: Zur Entwieklungsgesehickte des inenscblicbeii duodenum in friiben Embryonalstadien. Morpb. Jahrb. Bd. 29, S. 1S7-216. 1900. Toldt, C. : Die Entwicklung und Ausbildung der Driisen des Magens. Sitz.-Ber.

der kais. Akad. der Wiss. Wien. Bd. 82, S. 57-128. 1881. Torkel, Die sogenannte kongenitale Pylorusbyperplasie eine Entwieklungsstorung.

Vircbow's Arch. Bd. 180, S. 316-333. * 1905. Tourxeux, F. : Precis d'embryologie hurnaine. 2 e Edit. p. 1-5S9. Paris. 1909. Yoigt. J.: Beitrag zur Entwicklung der Darmschleimhaut. Anat. Hefte. Abt. 1.

Bd. 12, S. 49-70. 1S99.

THE DEVELOPMENT OF THE LIVER. By FREDERIC T. LEWIS.

Early Development of the Entodermal Portion. — In the embryo of 2.5 mm. which Thompson described (Fig. 236, p. 311), the liver is in a very early stage of development. It is a median ventral outgrowth of the entodermal tube, with thick walls which inclose a cavity in wide communication with the alimentary canal (Thompson, 1908). The latter presents a groove-like ventral border, but the liver appears as a well-defined cul-de-sac, the external form of which is shown in Fig. 288, A and B. Brachet (1896) and Swaen (1897) have described the liver as arising from a longitudinal groove.

In the slightly older Bremer embryo (4 mm.), the median hepatic diverticulum is still present, but there has been an extensive proliferation of the cells in its anterior and ventral walls. The proliferating cells have formed irregular^ masses and anastomosing cords. In places the nuclei are peripheral, so that there is a lightly staining protoplasmic core, but no lumen is present in these outgrowths. Ventral and lateral views of a model of the liver in this embryo are shown in Fig. 288, C and D. (A small nodular outgrowth of the digestive tube, shown in the figure, and a similar structure beyond the lower limit of the model, are not connected with the liver.) The 4.9 mm. embryo described by Ingalls (Fig. 239, p. 314) is considerably older. The anastomising cords of hepatic cells have formed a large crescentic mass, Fig. 289, A, with wings extending backward on either side of the intestinal tube. This mass connects with an outpocketing of the intestine, which corresponds with the original hepatic diverticulum. Distally the diverticulum shows a rounded subdivision, or outgrowth, which gives rise to the gall-bladder and cystic duct. Toward the intestine, in the angle between the diverticulum and the duodenum, the ventral pancreas has developed.

404 In an embryo of 7.5 mm. the liver is much larger (Fig. 240, p. 316). The crescentic mass of anastomosing trabecular joins the elongated diverticulum by a short solid stem, the hepatic duct (Fig. 289, B). The diverticulum has become tubular, thus giving rise to the common bile-duct (ductus choledochus) . Distally the gall-bladder and cystic duct are represented by a cylindrical prolongation of the diverticulum, which is nearly solid.

Div. hep

Div. hep.

B

Div. hep.

Fig. 288. — A and B, ventral and lateral views of the liver of an embryo with 23 pairs of somites (2.5 mm.), from a model by Peter Thompson. X 55 (?) diam. C and D, ventral and lateral views of the liver of a 4 mm. embryo, from a model by J. L. Bremer. X 135 diam. Div. hep., hepatic diverticulum; Tr., hepatic trabecular.

Before mammalian embryos had been satisfactorily studied, it was known that in the chick the liver arises from two outgrowths of the intestine, and Bischoff (1845), Remak (1855), and Kolliker (1861) believed that this would prove generally true for mammals. Although His (1SS5) found only single outgrowths in human embryos of 2-3 mm., Felix (1892) concluded that two are present. In addition to the " cranial hepatic duct " or diverticulum he found, in a single specimen, a " caudal groove " which he believed to be analogous with the posterior hepatic outgrowth in the chick. The embryo which he studied had been injured in the hepatic region. According to Felix, the caudal groove in man produces the gall-bladder and cystic duct, together with some of the hepatic trabeculae, and it fuses with the cranial portion. Swaen (1897), in an embryo of 3.8 mm., found

DEVELOPMENT OF THE LIVER.

405

the liver represented by a longitudinal groove, which fomis a cul-de-sac anteriorly and becomes gradually lower until it disappears posteriorly. Other writers have described the cul-de-sac, or hepatic diverticulum, as consisting of an anterior pars hepatica, which gives rise to the trabecular, and a posterior pars cystica, which produces the gall-bladder (cf. Hammar, 1897). Thompson, following Maurer (1906), applies these terms to portions of the diverticulum shown in Fig. 288, A and B. Geraudel (1907), without describing young embryos, concludes that the bile-ducts and the hepatic trabeculae are genetically independent; the former arise from the entoderm, and the latter from the mesoderm. This has been discussed and denied by Debeyre (1910). A bilaterally paired origin of the liver has been described in some vertebrates, but has not been found in man.

Certain investigators, following His (1885), recognize an early compact stage in the development of the liver, in which there are epithelial masses instead of anastomosing trabecular. It has been discussed whether the trabecular arise by epithelial outgrowths or by the breaking up of the solid masses through the invasion of mesenchymal tissue. The liver of the Bremer embryo is in the compact stage.

Ves. fel.

Pane. v.

Pane.

Div. hep.

Ves. fel.

D. cysl

Pane. d.

B

Fig. 289. — A, lateral view of the liver and pancreas of a 4.9 mm. embryo, from a model by N. W. Ingalls. X 65 diam. B, similar view of a model in which the hepatic trabecular are not included, from a 7.5 mm. embryo. Modelled by F. W. Thyng. X 50 diam. D. chol., common bile-duct ; D. cyst., cystic duct ; D. hep., hepatic duct ; Div. hep., hepatic diverticulum ; Ga., stomach ; Pane, d., dorsal pancreas ; Pane, v., ventral pancreas ; Tr., trabecular ; Ves. fel. (vesica fellea), gall-bladder ; x, aberrant duct.

Relation between the Entodermal Portion and the Bloodvessels. — Von Baer (1828) described the hepatic outgrowths in the chick as arising in close relation with the veins. Similarly Janosik (1887) observed that "in birds, through constant ramification, new outgrowths form, which grow into the lumen of the omphalomesenteric vein, pushing the endothelium before them." But in a young human embryo he failed to find such an intimate relation between the liver and the veins. Swaen, however (in 1897), states that in human embryos "the cavities of these veins have probably been invaded by the hepatic trabeculae and transformed into vascular ramifications and capillary networks." Bremer (1906) described the liver shown in Fig. 288, C and D, as follows: "The liver cords are found to be growing into mesenchyma at a level ventral to the vitelline (or omphalomesen

406 teric) veins; in this same niesenchyma, however, we find branches of the veins ramifying and forming plexuses, and in certain places these plexuses come into intimate relation with the liver cords." With further growth the places of contact rapidly increase. As the trabecule ramify, new branches of the venous plexus extend between and around them, so that the cords of liver-cells are closely invested with endothelium. The process, therefore, is not a simple invasion of the lumen of the veins by the trabecular As the right and left wings of the liver extend dorsally, they encounter respectively the main trunks of the right and left omphalomesenteric veins, which are passing from the intestine to the heart. The hepatic trabecular surround these veins. The left omphalomesenteric vein soon loses its identity within the hepatic plexus, but the right vein remains as a distinct channel. These

a» --. — i-v. u

H

vo-m.d.

A B Fig. 290. — Semi-diagrammatic reconstructions of the veins of the liver (ventral views). (After Mall. A, embryo of 4.5 mm.; B, 6.5 mm.; C, 7 mm. d. v., ductus venosus; H, liver; Int., intestine; r. u., recessus umbilicalis; v. h., hepatic vein; v. o-m. d., v. o-m. s., right and left omphalomesenteric veins; v. p., portal vein; v. u., umbilical vein; v. u. d., v. u. s., right and left umbilical veins.

relations are shown in the diagram, Fig. 290, A, from an embryo similar to Ingalls' specimen. (For a reconstruction of these veins, see Ingalls, 1908.) The further development of the large hepatic vessels may be considered briefly, since details are supplied in Chapter XVIII. In an embryo of 6.5 mm., Fig. 290, B, the two omphalomesenteric veins have produced a single vessel, which winds behind the intestine to enter the liver, the entering portion being a persistent part of the right vein. This is the portal vein of later stages. Within the liver the right omphalomesenteric vein can be followed continuously, and the left through a plexiform network, to the superior surface. Here a single vessel, the hepatic vein, conveys the blood to the heart. The hepatic vein is essentially the persistent outlet of the right omphalomesenteric vein.

Thus it will be seen that by intercrescence with the hepatic trabecule, the original omphalomesenteric veins have been resolved -into an afferent portal vein, which empties into a network of branches, and an efferent hepatic vein, which drains these

DEVELOPMENT OF THE LIVER. 407 branches. This purely venous type of circulation has long been described as a portal circulation. (In the liver it is the hepatic portal system; in the Wolffian body, the renal portal, etc.) It has also been described as a sinusoidal circulation (Minot, 1900; Lewis, 1904).

In addition to the blood from the intestines, received through the portal vein, the liver very early receives blood from the placenta, through the left umbilical vein. The pair of umbilical veins at first pass to the heart without entering the liver, but in some 4 mm. embryos the left umbilical vein already sends out branches which join the hepatic plexus. In the 6.5 mm. embryo, Fig. 290, B, one of these branches has become the chief outlet for the placental blood. At first the umbilical vein merges in the general plexus, but later it forms a large channel across the inferior portion of the liver (Fig. 290, C). Although it is a left vein, it gradually moves toward the median line, and the gall-bladder, which is morphologically median, is found on the right.

After birth, the portion of the umbilical vein extending from the umbilicus to the liver becomes reduced to a fibrous cord, — the round ligament (lig. teres hepatis). In the adult, a large branch of the portal vein within the liver extends toward this ligament and ends blindly. Rex (1888) has described this blind ending, as an appendage of the left branch of the portal vein, and named it the recessus umbilicalis. Mall (1906) has applied this term to a portion of the embryonic vessels, as indicated in Fig. 290, C. The vessel which appears as a continuation of the umbilical vein, passing from the portal to the hepatic (in later stages to the vena cava inferior), is the ductus venosus. Since the ductus venosus and the umbilical vein appear on the under surface of the liver, they will be further described with the surface markings.

As stated by Toldt and Zuckerkandl (1875), the capillaries of the liver in early embryonic stages are considerably wider than later, both absolutely and as compared with the glandular parts which they surround. Minot (1892), in describing the embryonic mammalian liver, noted that the "blood-channels are very large," and Schenk (1896) referred to the "lacunar vascularization" in the liver of batrachians. In 1900 Minot proposed the term sinusoids, in distinction from capillaries, for wide endothelial tubes fitted closely against the cells of the organ in which they are developed; those in the adult liver, which have become narrower, he distinguished as capUliform sinusoids. The intimate relation between the hepatic cells and the endothelium has lon<>' been known. Thus Hering (1871) recorded that "the secreting cells of the liver exhibit a peculiar arrangement, whereby there exists a much closer and more extensive contact between them and the capillaries than in other glands." In describing this relation in the embryo, the

408 HUMAN EMBRYOLOGY.

perivascular tissue will be considered first, and then the cells which occur within it.

Perivascular Tissue.— The gross relation between the bloodvessels and the hepatic trabeculae in a 9.4 mm. embryo is shown in the model, Fig. 308, p. 432. Along the upper surface of the model the liver is seen in transverse section. The histological features of such a section are shown in Fig. 291. In many places the endothelium has become separated from the hepatic trabeculae, thus producing a perivascular space. The space is bridged by slender protoplasmic processes of the endothelial cells. These processes, together with the peripheral protoplasm of the endothelial cells, are directly transformed into connective-tissue fibres of a peculiar sort. Kon (1908) observed the transformation in embryos of four and five months, but Mollier (1909) declares that

iff %4 } * s &°:" Q-pCorp.

-Bl.

Fig. 291. — Section of the liver of a 7.5 mm. embryo (Harvard Collection, Series 256). X 120 diam. .BZ.,Lblood-forming cells; C. hep., cells of the hepatic trabeculae; Corp., nucleated blood-corpuscles; Endo., endothelium; V., branch of the portal vein.

it begins much earlier, since the fibres are clearly present at 30 mm. 15 As compared with the adult, Maresch (1905) finds that the fibres in embryonic livers are poorly developed, and that "not until birth can an abundant supporting tissue be demonstrated." The nature of the delicate felt-work of perivascular fibres found in the adult has been discussed by Mall (1896) as follows: " Kupffer considers them to be elastic, while Ewald and Kiihne consider them white fibrous. The fact that they are digested by pancreatin and yield but little gelatine when boiled excludes both views; and, since they seem to be identical with the reticulum of lymphatic glands, spleen, and mucous membrane, I shall retain for them the name reticulum." Usually they are regarded as delicate strands of white fibrous tissue. Apart from the nuclei of the endothelium, no nuclei are found in relation with the fibres, as noted by His (1860) and abundantly confirmed.

" Mall found the network of fibrils in a pig embryo of 20 mm. Amer. Journ. of Anal., vol i, p. 354, 1902.

DEVELOPMENT OF THE LIVER. 409 Perivascular spaces, bounded on one side by the capillary wall and on the other directly by hepatic cells, with here and there a connective-tissue fibre stretched across, were described by Biesiadecki in 1867. He concluded that these spaces were preformed and not due to imperfect fixation. Hering (1871) recognized that, although the spaces are increased by shrinkage, they indicate a structural peculiarity. MacGillavry (1864) had injected these spaces through the lymphatic vessels, and also through the bileducts, but they do not normally open into either. They are interfibrillar tissue spaces, and are quite distinct from lymphatic vessels.

Blood-forming Cells.- — In the liver of the 7.5 mm. embryo, in addition to the flattened nuclei of the endothelium and the round, coarsely granular nuclei of the hepatic trabecular, there is a third group consisting of small, darkly staining, and densely granular forms, occasionally somewhat indented, situated between the endothelium and the trabecular (Fig. 291). Frequently they occur in groups. The protoplasm surrounding them, in this specimen, is scarcely perceptible.

Toldt and Zuekerkandl (1875) designated these as round cells, in contrast with the cuboidal cells of the trabecule, and described them as follows : " These cells are distinctly round, variable in size, sharply outlined, very finery granular, and clear; they never contain fat droplets, even when such are present in considerable quantity in the cuboidal cells, nor do pigment granules occur in them. The relatively large nucleus is generally distinctly outlined, yet not with so refractive a contour as in the cuboidal cells. It is of quite homogeneous consistency. Nucleoli are almost always visible, even two or three in a nucleus." These cells were observed in a 10 weeks embryo and in all the older ones examined. " Toward the end of embryonic life the number of the round cells, as compared with the cuboidal cells, strikingly decreases, but they are still present at birth, situated either singly in the wall of the gland tubes, or in groups by themselves, surrounded only by a capillary mesh." " After birth their number diminishes very rapidly, and even in the first weeks they seem to disappear entirely." Toldt and Zuekerkandl mistook the round cells for young hepatic cells. The fact that they were not dislodged from the liver after the veins had been thoroughly washed out with salt solution confirmed their opinion that they were not the developing blood-corpuscles described by Kolliker. Kolliker (1846), from a study of one human and several sheep embryos, had concluded that ' ' as the liver develops, the multiplication of corpuscles ceases elsewhere in the blood, and in its place, probably because all of the blood of the umbilical vein now flows into the liver, an aotive formation of blood-cells occurs in the hepatic vessels." Kolliker 's communication is followed by a letter from E. H. Weber, who states that in the liver of the frog the corpuscles develop outside of the vessels, to which they may gain entrance by local absorption of the vessel wall. In 1874 Neumann described the corpuscles as occur

410 HIM AX EMBRYOLOGY.

ring in nests between the hepatic cells in human embryos of 8-9 months, and he thought that they arose endogenously in the protoplasm of certain elongated cells (presumably endothelial). Schmidt (1892) states that they arise through karyokinetic division of the endothelial cells and multiply by mitosis. Van der Stricht (1892) concluded that the cells described by Toldt and Zuckerkandl were young red blood-corpuscles. He observed that the capillary wall becomes discontinuous at the places where these cells appear to be imbedded in the trabecule, and therefore he considered that they came from the circulating blood.

In the perivascular spaces of the embryonic liver, not only are red corpuscles produced, but also leucocytes and giant cells such as are characteristic of red bone marrow. They are described in connection with the blood (Chapter XVIII). In early stages these cells are abundant, but, according to Nattan-Larrier (1904), after the fifth month giant cells and basophilic myelocytes are very rare, and at birth the nucleated red corpuscles remain almost exclusively. Lobenhoffer (1908) states that as blood formation in the liver diminishes, the capillary recesses become fewer and smaller, disappearing in the eighth month, but at birth, in almost every field, he found one or two blood-forming groups between the hepatic cells. He agrees with Schmidt that "the cells of the capillaries are capable of forming blood elements." It is probable that the primitive blood-cells seen in the liver of the 7.5 mm. embryo are derived from the endothelium, but their possible origin from cells of the circulating blood must be considered (see Chapter XVIII).

In the adult, as shown by Kupffer (1876 and 1899), the endothelium is so perforated that its cells have become stellate. Mollier (1909) considers that the stellate condition is associated with blood formation and is most highly developed in the embryo. He believes that the endothelial cells and the blood-corpuscles both arise from a reticular syncytium, and, as the formation of corpuscles ceases, the syncytium becomes a closed endothelium with perivascular fibres. This change occurs first along the capillaries which are to become the main branches of the portal vein.

The Gall-bladder, Ductus Cysticus and Ductus Choledochus. — The solid stage of the gall-bladder, which occurs regularly in young human embryos, is presumably acquired with the elongation of the round diverticulum, such as is found at 4.9 mm. (Fig. 289, A). Both the gall-bladder and the common bile-duct have been recorded as solid in an embryo of 6.8 mm. (Piper, 1900) and in another of 6.75 mm. (Keibel and Elze, 1908). These are the youngest stages in which the solid condition has been observed. At 7.5 mm. (Fig. 289, B) there is a lumen in the common bile-duct, but the gall-bladder is impervious. Near the hepatic duct the lumen is subdivded, appearing in cross sections as two or three

DEVELOPMENT OF THE DIVER.

411

minute pores. At 16 mm. there are irregular subdivisions near the hepatic duct, and the distal part of the gall-bladder is solid, but in both the cystic duct and the common bile duct the lumen is single and well defined. Occasionally the common bile-duct has a double lumen in the midst of its course, as was noted at 14.5 and at 22.8 mm. (Fig. 292, B). In these cases the two cavities unite in a single lumen both proximally and distally.

As the gall-bladder expands it may present ' ' intra-epithelial cvsts," as recorded by Keibel and Elze at 18 mm., or the lumen may be bridged by epithelial strands, as in Fig. 292, A (29 mm.). At this stage the greater part of the gall-bladder has a clear-cut round lumen. Its wall shows two rows of oval nuclei, with mitotic figures in the inner row (22.8 mm.). In a 42 mm. embryo the tapering proximal part of the gall-bladder presents several rounded outpocketings, resembling the intestinal diverticula, some

m

A

B

Fig. 292. — A, section of the gall-bladder of a 29 mm. embryo (Harvard Collection, Series 914). X 180 diam. B section of the common bile-duct of a 22.8 mm. embryo (Harvard Collection, Series S71). X 180 diam. C, epithelium of the gall-bladder, two weeks after birth. X 580 diam.

times sectioned so as to appear detached from the main tube. Similar pockets are apparently more definite and abundant in the sheep and pig than in man. The lining of the gall-bladder in a 78 mm. specimen shows numerous well-defined folds, such as are characteristic of all later stages. The development of the folds in the cystic duct, constituting the spiral valve, has apparently not been studied.

The epithelium of the gall-bladder of a child two weeks old, born prematurely at the seventh month, is simple and columnar, with distinct cell walls ending in terminal bars (Fig. 292. C). A broad clear border, or top plate, with radial striation, such as Virchow (1857) described in the gall-bladder of adults and children, could not be detected. The cells contain oval. pale, vesicular nuclei, together with more elongated and darkly staining forms, apparently due to compression. The dark nuclei may be scattered among the others or may form considerable gronps. Occasionally

412 HUMAN EMBRYOLOGY.

at the bottom of the depressions between the folds, a pearl-like group of cells is seen (Fig. 292, C), suggesting the buds of the intestine.

The developnmet of the " glands " of the bile-ducts has not been studied einbryoiogically. These structures are generally considered to be epithelial pockets rather than true mucous glands. They formerly attracted much attention, culminating in a thorough study of them by Riess in 1863, for it had been supposed by Henle (1861) that these glands were the source of bile, and that the hepatic trabecule produced sugar. Riess states that they are most numerous in the hepatic duct, less numerous in the upper part of the common bile-duct and lower part of the cystic duct, and entirely absent from the lower part of the common bile-duct and upper part of the cystic duet; probably there are none in the gall-bladder. The largest of them are branched tubes with rounded terminations; the small ones are simple pockets, which give place, in the smaller branches of the hepatic duct, to rounded diverticula and swellings. Riess has noted that the glandular appendages of the bile-ducts are much less developed in children than in adults, and " in the earlier embryonic life they are perhaps wholly lacking." The outer coats of the gall-bladder and cystic duct develop as follows : At 7.5 mm. the epithelium is surrounded by a layer of mesenchyma, and the entire structure is so imbedded in the under surface of the liver that it causes only a slight swelling of the peritoneal surface. Above and on the sides the mesenchyma is in direct relation with the hepatic trabecular, and it receives a few prolongations of the venous capillaries. Below it is covered by the peritoneal epithelium except on the left, where that layer is reflected to the abdominal walls in connection with the falciform ligament. In later stages the gall-bladder is separated from the hepatic trabecular on either side, and is attached to the liver only along its upper surface.

At 16 mm. the mesenchyma surrounding the gall-bladder is still undifferentiated, but at 22.8 mm.it forms two broad concentric zones, of which the inner is darker and more compact than the outer. At 29 mm. certain cells in the peripheral part of the dark zone form a third layer, which is thin and somewhat interrupted. As seen in later stages these cells are myoblasts, so that at 29 mm. all three layers of the adult gall-bladder are indicated. These are the mucosa, muscularis, and serosa. The layers become gradually less distinct toward the hepatic duct.

The vessels and nerves of the gall-bladder are branches of those seen at 10 mm. near the pyloric end of the stomach (Fig. 274, C, p. 370). Of these the hepatic artery is of special interest.

The Hepatic Artery. — At 10 mm. the hepatic branch of the cceliac artery can be followed to the hepatic duct. Later it extends along the hepatic and cystic ducts, but as the cystic branches develop first, the hepatic artery appears primarily as the artery of the gall-bladder. Thus, at 22.8 mm., the main stem lies in a wing-like fold of the tunica serosa of the gall-bladder, and other

DEVELOPMENT OF THE LIVER, 413 branches follow the attached border of the gall-bladder, lying close to the hepatic trabecular They connect with a capillary plexus in the mesenchyina, which empties at various points into the venous network among the adjacent trabecular The cystic vein of the adult, which conveys the blood from the gall-bladder to the main trunk of the portal vein, is a later formation. Therefore three stages may be recognized in the development of the blood-vessels of the gall-bladder: in the first, the capillaries from the portal network extend into the mesenchyma around the gall-bladder; in the second, they are joined by the arterial capillaries and become efferent vessels; in the third, a single efferent vein, extending along the cystic duct and emptying into the portal trunk, is developed from the capillary system.

After the cystic branch of the hepatic artery has become established, mesenchyma develops around the hepatic duct and its ramifications, and branches of the hepatic artery appear in this mesenchyma. They form capillary plexuses, especially around the branches of the hepatic duct, and the blood passes from these capillaries into the subdivisions of the portal vein found among the adjacent trabecular Veins comparable with the cystic vein, which collect the blood from the arterial capillaries and convey it to the main branches of the portal vein, have been described within the liver of the adult, but according to Mall (1906) they do not exist.

Certain branches of the hepatic artery reach the surface of the liver and ramify in the capsule. They either empty into the portal network beneath the capsule, or are drained by " branches of the hepatic vein which come to the surface of the liver and spread out between the meshes of the arterial plexus " (Mall).

The Hepatic Duct. — At 9.4 mm. the hepatic duct is a short stem connecting the great mass of hepatic trabecular with the common bile-duct (Fig. 308, p. 432). It is solid, or nearly so, in this specimen, but in a 10 mm. embryo it contains a lumen. The nuclei are crowded so that the duct stains deeply and contrasts sharply with the trabecular Where it joins the trabecular the transition is so abrupt that it has led to the erroneous opinion that the two tissues are of different origin.

Although the hepatic duct in man is a single stem, there are certain mammals in which there are several ducts which pass from the trabecular to the cystic duct, or, in some species, to the gallbladder (cf. Rex, 1888). Rudimentary additional ducts are common in human embryos. They may join the hepatic, cystic, or common bile-duct, but usually they occur very near the junction of the three. Thus, at 7.5 mm. (Fig. 289, B), on either side at this junction, there is a solid knob which does not quite reach the trabecular In the same position there is a single outgrowth in

414 HUMAN EMBRYOLOGY.

the 9.4 mm. embryo and also in the 10 mm. specimen. An aberrant duct with a lumen empties into the proximal end of the cystic duct at 14.5 mm., and the same embryo shows a detached nodule of epithelium beside the common bile-duct. A detached nodule containing a lumen was found at 16 mm. These structures may represent outgrowths of the original diverticulum which have contributed to the formation of the mass of trabecule and are now degenerating, or they may be abortive secondary ducts which have never reached the trabecular.

Within the liver of 9-10 mm. embryos the branches of the hepatic duct usually cannot be traced far, but there is marked variation in this respect. Frequently among the hepatic trabecular one or more very short ducts may be found which certainly do not connect with other ducts. They may blend with the hepatic trabecular at one or both ends. These detached ducts are lined with regular cuboidal or columnar epithelium and may show a clear-cut lumen. Such ducts were noted in embryos measuring 8, 9.4, and 10.2 mm., and, according to Elze (1907), in those of 7 and 11 mm. Lewis (1903) found similar detached cysts in the liver of a 12 mm. pig, and considered them to be cut off from the secondary hepatic ducts. Whether they are detached portions of the hepatic ducts is questionable. They may arise in situ by a transformation of the cells of the hepatic trabecular The Periportal Ducts.— In an embryo of 22.8 mm. (Fig. 293, A) the spread of the bile-ducts along the main branches of the portal vein has begun. The trabecular form cords extending along the surface of the periportal mesenchyma, and in them a lumen is formed. In places the cells on the mesenchymal side of the lumen are distinctly flatter than those toward the portal capillaries. As seen in the figure, the trabecular connect freely with these ducts. In a later stage (29 mm., Fig. 293, B) the mesenchyma has increased, so that it surrounds the ducts which were seen forming along its surface. Their epithelium has become regularly cuboidal or columnar. On the upper side of the vein in Fig. 293, B, the ducts are in the earlier stage of development.

The periportal ducts clearly form a plexus. The larger ducts, which have become surrounded by mesenchyma, are also plexiform, although with the enlargement of the liver their anastomoses become less numerous. However, the plexiform arrangement of the main branches of the hepatic duct, which was clearly seen in a single frontal section at 29 mm., persists throughout life, as has long been known.

In the adult, Kiernan (1833) found that an injection of the left branch of the hepatic duct returns by the right duct. "From this experiment ... it appears that the right and left ducts anastomose with each other." Hering (1871) stated that there is

DEVELOPMENT OF T1IK LIVER.

41.-)

an anastomosis of the branches of the large ducts, and that the smallest ones sometimes appear to anastomose around the vein which they accompany (as stated by Riess), "but this requires further investigation." He described the connection between these ducts and the trabecule? as formed by canals "bounded on one side by small epithelial cells and on the other by large hepatic cells." These transitional ducts, sometimes called the "canals of Hering," were observed "in a child of 3 months.

Toldt and Zuekerkandl (1875) described them in an embryo of the tenth week as follows: Those hepatic trabecular which are found in the immediate vicinity of the relatively very large portal branches, almost without exception are

& y**m *\at

"* : w 

9 m

•

V.p.

$ «%

%0 Q •"«"• o?

if

»- 5

D.peri-p

\ Fig. 293. — Sections showing the formation of the periportal ducts (D.peri-p.) around a branch of the portal vein (V.p.). A, from an embryo of 22.8 mm. (Harvard Collection, Series 871 ); B, from an embryo of 32 mm. (Harvard Collection, Series 913). X 185 diam.

perpendicular to the latter, so that in cross sections of a portal branch they show a radial arrangement, and in longitudinal sections they appear in parallel rows. They open into the ducts almost at right angles, and their cuboidal cells are inserted directly into the flat epithelium of the ducts." The Bile-capillaries. — In an embryo of the fourth week Kolliker (1861) found the liver composed of solid trabecular Eemak (1855) had observed a similar condition in chick embryos, and had named the solid cords of cells "hepatic cylinders." J'liisalix(1888),in describing a 10 mm. embryo, states that he agrees with Kolliker that there is no lumen in the primitive hepatic cylinders. This view is generally accepted. But Toldt and Zuekerkandl (1875) described the liver of an embryo of the fourth week, in which the tubular structure appears most distinctly. "That we have to do with tubes and not with solid cords of cells can be shown both in cross sections and frequently in longitudinal sections; the lumen is always bounded by so sharp an outline that

416 HUMAN . MBRYO^OGY.

there is no question of an artificial separation of the cells. ' ' They find that in the slender tubes the lumen is bounded by three or four cells, in the larger tubes by still more.

Although in a 10 mm. embryo most of the trabecular are solid, there are some, scattered irregularly through the liver, which show a very distinct lumen (Fig. 294, A). The lumen is larger than that of the future bile-capillary and lacks the characteristic cuticular border. It is bounded usually by five or six cells and ends blindly in the adjoining sections. At 29 mm. the trabecular are extensively broken up by the nests of blood-cells, but those which are least disturbed often show a lumen. At 37 mm. the

C.hep

Fig. 294. — -4, hepatic trabecula containing a lumen. From a 10 mm. embryo (Harvard Collection, Series 1000). X 1065 diam. B, bile-capillaries in a 44.3 mm. embryo (Harvard Collection, Series 1611). X 1065 diam. Bl., blood-corpuscle; C.hep., hepatic cell; Endo., endothelium ; Lum., lumen of a bilecapillary.

lumen is more distinctly outlined, but it may still be bounded by as many as six cells. In a 44 mm. embryo, bile-capillaries with cuticular borders are abundant (Fig. 294, B). They occur in all parts of the liver, but appear to be most numerous near the periportal ducts, into which some of them empty. They extend axially through the trabecula?, and where the latter branch, the capillaries branch- also. Usually they are separated from the venous endothelium by an entire hepatic cell, but, as seen in the figure, nests of blood-cells sometimes approach very close to them.

Hendrickson (1898) studied the bile-capillaries with Golgi's method, and found them extensively developed at 50 mm. (Fig. 295). In his preparations they are somewhat more abundant

DEVELOP o ,NT OF THE LIVER. 417 around the branches of the portal vein, but it could not be shown that they develop peripherally from the periportal ducts. They may arise within the trabecule as blind tubes, which later anastomose and join the ducts. In a Golgi preparation from an embryo of 100 mm., Hendrickson found numerous polygona] meshes made by the bile-capillaries. These may encircle a single cell within a thick trabecula, or they may run through trabecular which anastomose at both ends. Short lateral branches are seen in Hendrickson's preparations, but whether they represent axial capillaries, either ending blindly or passing out of the plane of section, or whether they are intercellular branches radiating from the axial capillary toward the venous endothelium, was not determined.

Fig. 295. — Golgi preparation of the bile-capillaries in an embryo of 50 mm. X 53 diain.

(After Hendrickson.) The number of hepatic cells bounding a bile-capillary is greater in embryos than in the adult. The capillary of the adult is usually a minute canal in the midst of the boundary between two adjacent cells, but occasionally it is surrounded by three or four cells ; and von Biesiadecki, the discoverer of the human bile-capillaries, declared that five is the usual number, — rarely four. Von Biesiadecki studied only pathological specimens in which the bilecapillaries were distended. Hering (1871) was unable to find any capillaries bounded by five or more cells, either in the adult or at birth. At birth, however, in contrast with the adult, he found that the lumen of a bile-capillary is often surrounded by three or four hepatic cells, thus resembling the condition in certain amphibia. Toldt and Zuckerkandl (1875), who found three or four cells bounding the capillaries in an embryo of the fourth week, stated that in embryos of four to seven months the number Vol. II.— 27

418 HUMAN EMBRYOlS^Cr was "four to six or even more," but they included in their count the blood-cells lodged within the hepatic trabecular Toward the end of fetal life they found usually three or four cells ; and in the fourth and fifth years, "generally only two, but occasionally three and even four." The reduction from three or four to two cells, occurring shortly after birth, is accompanied, according to Toldt and Zuckerkandl, by a "stretching" of the trabeculae, whereby they become longer and more slender, with their cells arranged in rows. They recognize the occasional persistence of three- and four-celled tubes, even after twenty years.

The Hepatic Cells.— In all stages the hepatic cells are characterized by large, very round nuclei, containing a coarse chromatic network, and surrounded by abundant, densely granular protoplasm. According to Toldt and Zuckerkandl, the diameter of the nucleus at birth is generally about 9.6 /* and in the adult about 8 //<, so that the nuclei of the adult are distinctly smaller than at birth. As a whole, however, the cells apparently increase in size. They multiply by mitosis. In a 10 mm. embryo the mitotic figures are abundant throughout the trabeculae, but in later stages, according to Mall (1906), they are particularly numerous around the terminal (periportal) bile-ducts. Frequently the cell division is incomplete, so that a single cell may contain two nuclei. In early stages cell membranes are entirely lacking, and at 10 mm., although the cells readily separate from one another, the membranes are indistinct. Toldt and Zuckerkandl have noted that at five and six months the cells do not show the sharp outlines observed at birth. They find that at birth a portion of the cells, when isolated in salt solution, are irregularly cuboidal, but that most of them are somewhat elongated. The isolated cells of the adult vary in shape, but are more nearly cuboidal than at birth.

The hepatic cells always stand in close relation to the blood which contains absorbed nutriment. First the blood from the yolk-sac, then that from the placenta, and after birth the blood from the intestine passes directly to the liver and flows through its vascular network. Kolliker (1861) noted that the liver in the embryo is physiologically a very important organ, but that its significance is rather in producing chemical and morphological changes in the blood than in secreting bile. Fat appears in the hepatic cells before it is present in the subcutaneous tissue, as shown by Chipman (1902) for the rabbit. In the liver of human embryos Toldt and Zuckerkandl found fat droplets as early as the third or fourth month. Nattan-Larrier (1903) found that fat was forming in the fourth month, and that at birth certain hepatic cells were filled with large fat droplets, separated from one another by thin layers of protoplasm. In the liver of the rabbit, glycogen appears in the 22d day of gestation, six days after the formation

DEVELOPMENT OF THE LIVER. 419 of fat, and it increases steadily and rapidly until birth ( ( Jhipman). Apparently the time of its appearance in human embryos has not been determined.

Kolliker found that the secretion of bile begins as early as the third month. He states that "from the third to the fifth month, material like bile is found in the small intestine, and in the second half of pregnancy it occurs also in the large intestine. . . . Until the sixth month the gall-bladder contains only mucus, but after that it contains bile." Zweifel" (1875) recorded that the intestinal contents of embryos of three months respond to the ordinary tests for bile acids and pigments. Toldt and Zuckerkandl observed yellow pigment granules within the embryonic liver beginning with the fourth or fifth mouth, but the granules were limited to the epithelial cells of the ducts, together with the adjacent hepatic cells. Even at birth they found that pigment granules are infrequent, and that the cells are clearer and more transparent than those of the adult.

Nerves and Lymphatics. — It has already been shown that the common bile-duct, cystic duct, and hepatic duct are very early surrounded by mesenchyma. Later the mesenchyma spreads along the ramifications of the portal vein, into the substance of the liver. Thus, in an embryo of the tenth week Toldt and Zuckerkandl found that the portal branches are surrounded by a considerable mass of connective tissue, and so can easily be distinguished from the thin-walled branches of the hepatic vein which are closely surrounded by the liver-cells. In the third and fourth month the difference becomes very striking. Ducts have formed at the periphery of the periportal tissue, and branches of the hepatic artery together with nerves and lymphatics have extended into it. The path by which the vessels and nerves reach the liver is shown in Fig. 274, B and C (p. 370) ; they enter at the transverse fissure or porta of the liver.

Little is known regarding the development of the nerves. Sympathetic fibres may readily be found at the entrance of the liver in embryos of 20-40 mm., but in the specimens at hand they can be traced no further than the primary division of the hepatic duct. They are associated with scattered clumps of nuclei, apparently ganglionic. In the adult the nerves form plexuses around the branches of the portal vein and hepatic ducts, and especially around the branches of the hepatic artery. In addition to the sympathetic nerves, there are fibres from the vagus, presumably entering from the pylorus.

The lymphatic vessels, which extend to the porta in a 42 mm. embryo, drain into the lower part of the thoracic duct. Later they grow into the periportal tissue, in which at birth they are conspicuously large. In the adult they extend as far as the smallest

420 HUMAN EMBRYOLOGY.

ramifications of the hepatic ducts, and terminate in the connective tissue. Herring and Simpson (1906) find that the lymphatic vessels accompany the hepatic artery and its branches, forming networks around these vessels, as well as around the branches of the portal vein and bile-ducts. There are no lymphatics among the trabecular Mall (1906) concludes that the great amount of lymph which flows from the liver is derived directly from the blood-plasma. It passes out between the stellate and endothelial cells, and flows through the perivascular reticulum to the periportal tissue, where it enters the lymphatic vessels.

There is another system of lymphatic vessels in the liver, which has not been studied embryologically. This includes the vessels which extend downward from the diaphragm, through the ligaments of the liver, to ramify in the capsule.

Lobules. — The great mass of hepatic trabecular is arranged in more or less definite lobules, which were discovered by Wepfer in the liver of the pig in 1664, and which have been familiar to anatomists since the time of Malpighi (see Kiernan, 1833). The liver was compared with a bunch of grapes. According to some anatomists its lobules were appended to the extremities of the portal vein, but Kiernan agreed with those who made the hepatic vein the axial structure. In the centre of each lobule he recognized a terminal branch of the hepatic vein. Between the lobules there are intervals, which Kiernan named portal canals, filled with connective tissue containing branches of the portal vein, hepatic artery, bile-ducts, nerves, and lymphatics. Three portal canals nia}^ be expected at the periphery of a single lobule. In the pig the connective tissue filling these canals spreads around the lobules, investing them with capsules, and thus making them conspicuous. In the human adult there are normally only indications of such capsules, and at birth they are wholly lacking. The portal canals then stand as isolated "boundary stones." Kiernan recognized certain objections to describing the liver on the basis of these lobules, for he wrote, "The essential part of a gland is undoubtedly its duct; vessels it possesses in common with every other organ ; and it may be thought that in the above description too much importance is attached to the hepatic veins." Recognizing this, Brissaud and Sabourin (1888) proposed the term biliary or portal lobule for the group of trabecule centred about a portal canal, leaving hepatic lobule for the structures described by Kiernan. Others also have considered that the portal lobule is morphologically the true unit of the liver.

As stated by Mall, " In all other glands we make the duet the centre of the structural unit. From this centre often the artery and the framework radiate. In the liver everything radiates from the so-called interlobular space, — arterial and portal blood-vessels, bile-duct, lymphatics, nerves, and connective tissue.

DEVELOPMENT OF THE LIVER. 421 . . . Throughout ruy description I shnll use (he term portal unit, structural unit, or unit for the clump of tissue which surrounds each terminal branch of the portal vein. In order to avoid confusion, I shall use the term lobule in its old sense, — as the hepatic lobule, — for after much discussion carried on during two centuries, it has become well established." Sclienk (1874) stated that the embryonic liver, in a certain stage, represents a single lobule of the adult organ. This view was adopted by Toldt and Zuckerkandl, and defined by Mall (1906). Mall states that in a 4 mm. embryo "the single lobule is perfect; it is composed of a complete capillary network without an anastomosing vein through it." Mall believes that the further development of the hepatic vessels is in accordance with the laws established by Thoma (1S93). These are: (1) An acceleration of the current leads to an enlargement of the lumen of a vessel, and a slowing of the current leads to its narrowing and final disappearance.

(2) An increase in the blood-pressure is the cause for new formation of capillaries.

(3) The growth in thickness of the vessel wall depends on the tension of the wall, which in turn is dependent upon the blood-pressure and the diameter of the vessel.

These laws apply to the liver, as shown in Fig. 290, B and C. In C a new and direct channel, the ductus venosus, has been formed across the liver, apparently by the enlargement of capillaries in which the current has been accelerated. At the same time the current becomes slower in the circuitous right omphalomesenteric vein, shown in B, which is reduced to capillaries in C. The liver then consists of two lobules, right and left respectively. The blood enters them from below, and is drained by the hepatic branches above. In an embryo of 11 mm. (Fig. 296) Mall finds that six lobules are indicated. These are obscure in the figure, since many of the enlarging capillary vessels have been drawn. It will be observed that the portal branches tend to alternate with the hepatic branches. "They are beginning to dovetail with each other." Thus the formation of a great number of lobules is suggested. The way in which this is accomplished is shown in Mall's diagrams, Figs. 297, 298, and 299. The single branches d and a in Fig. 297 become the main branches d and a in Fig. 299, and the successive orders of new branches e, f and b, c have arisen by the enlargement of capillary vessels. The lobules b, indicated by dotted outlines in Fig. 298, have become clusters of lobules in Fig. 299, and some new simple lobules, c 2 , have appeared between them.

In this way the 480,000 lobules, which according to Mall's estimate are found in the liver of an adult dog, are produced from a single lobule. Regulated by Thoma 's laws the main vascular stems develop in such a way that all the lobules are equally favored. "If fluids of different consistency are injected either

422

111 MAX EMBRYOLOGY.

into the portal or hepatic vein, all of the terminal veins fill simultaneously. Moreover, the final branches of the portal and hepatic veins are always as far from one another as possible. "At all times this distance is half the diameter of a lobule, and since this is in the neighborhood of one millimetre, the distance is abont half a millimetre, the normal length of a capillary vessel. ' ; The lobules at birth, according to Toldt and Zuckerkandl, differ from those of the adult, as follows : " In the child there are indeed vascular territories which show a certain independence. However, since they are drained by a group of terminal branches and not by a single venous root, they are not comparable with

Fig. 296. — Ventral view of a reconstruction of the hepatic vessels in an embryo of 11 mm. X 25 diam. (After Mall.) The principal veins are — d. v., ductus venosus; p., portal vein; r.m. and r. s., middle and left rami of the hepatic vein; r. »., recessus umbilicalis; u. v., umbilical vein; v. c, vena cava; v. o. m., omphalomesenteric vein.

the lobules of the adult. They correspond rather with a combination of the latter, and to a certain extent represent lobules of a higher order, from which gradually single portions will be detached." The multiplication of lobules continues long after birth, and partly divided, compound forms were recognized in the adult by Kiernan.

Ligaments and Lobes. — The relation of the liver of a 4 mm. embryo to the body cavity is shown in dorsal view in Fig. 300. The model shown in the figure has three vertical cut surfaces, — the body wall on either side, and the mesentery, prolonged upward

DEVELOPMENT OF THE LIVER.

423

as the mediastinal septum, in the centre. The ventral portion of the mesentery bulges laterally below, where the liver is growing into it, and then joins the septum transversum. The latter is the plate of tissue which forms the ventral surface of the model. Its position is so nearly vertical that it separates the pair of pleuroperitoneal cavities behind from the pericardial cavity in front. The pleuroperitoneal cavities are shown in the figure. Anteriorly they turn ventrally over the free margin of the septum transversum and empty into the median pericardial cavity.

Most embryologists. following His, state that the liver develops in the septum transversum. and as seen in median sagittal sections (Fig. 236, p. 311) this appeal's to be correct. The septum transversum. however, is early divisible into two parts, related to one another like the arms of a T. The ti'ansverse portion forms a part

ff

d a Fig. 297.

Fig. 299.

Figs. 297, 298, and 299. — Diagrams of three successive stages in the formation of lobules. (After Mall.) d, branch of the portal vein ; a, branch of the hepatic vein.

of the' diaphragm. The median sagittal portion is the ventral mesentery, and it is in this subdivision of the septum transversum that the liver develops. The relatively very broad attachment of the mesentery to the diaphragm forms the falciform ligament of the liver, and the lateral bulgings indicate respectively the right and left hepatic lobes.

His (1880) recognized that in a -1 mm. embryo the tissue in which the liver develops is more or less independent of the septum transversum, and he named it the " Vorleber." Hertwig (190G), who states — following His — that the liver grows into the septum transversum. writes also that the liver develops in the ventral mesentery.

In the 4 mm. embryo (Fig. 300), each of the lateral lobes of the liver is prolonged upward by an irregular mass of tissue, which nearlv fills the body cavity. Each mass is attached alone,' its ventral border to the septum transversum. and thus it separates the medial pleural part of the body cavity from the lateral peritoneal part. But superiorly these cavities connect with one another and open into the pericardial cavity, as already noted. The

424 HUMAN EMBRYOLOGY.

upward prolongations of the liver may be called the right and left coronary appendages. They are the anterior portions of the "Vorleber" of His, which were described in a 4 mm. embryo as containing a plexus of blood-vessels but no network of trabecular 16 They are certainly the ventral pillars bounding the pleuroperitoneal opening, first described by Uskow (1883) and later, for human embryos, by Swaen (1897). Their relation to the dorsal pillars, which have been called the suspensory ligaments of the Wolffian body, have been discussed in Chapter XIII.

In an embryo of 9.4 mm. (Fig. 301) it is seen that the coronary appendages have fused with the septum transversum and the lateral body wall, thus shutting off the superior lateral recess of the peritoneal cavity (Swaen). The liver now presents a crescentic transverse attachment to the diaphragm, passing from one coronary appendage to the other; this attachment is the coronary ligament. Within its concavity, on either side, are the pleural cavities which communicate below with the peritoneal cavity.

A fundamental feature of the 9.4 mm. embryo is the presence of the plica vence cavce, of Ravn (1889). This is essentially an attachment of the right lobe of the liver to the dorsal body wall, and it has developed downward from the right ala pulmonalis. 17 Through this attachment the right subcardinal vein anastomoses with the veins of the liver, thus giving rise to the vena cava inferior. The portion of the liver between the plica venae cavae and the ventral mesentery, or omentum minus, is the caudate lobe (of Spigelius). The caudate lobe joins the right lobe across the foramen epiploicum (of Winslow). Below the foramen, the portal vein and bile-duct are seen in section. In the lower part of the model the place where the left umbilical vein enters the liver is indicated by a fold. The gall-bladder is on the right of it.

In an embryo of 5 months (Fig. 302) the diaphragm has been completed in the way described in Chapter XIII. The oesophagus, not included in the preceding drawings, is seen passing through it. The thin lateral extensions of each coronary ligament {lig amenta triangularia) mark the position of the former appendages, and filling their dorsal concavity is the portion of the diaphragm which formed last, and which completes the separation of pleural and peritoneal cavities. The vena cava inferior now fills its plica, which has become broad. The lesser omentum is very thin except 10 They were probably included by Lieberktihn (1876) among 1 the "villi" which occur where the omphalomesenteric veins enter the heart, and which were said to be so related to the developing liver that they contained the first bloodvessels of that organ.

" Ala pulmonalis is the term introduced by Ravn for the developing mesodermal portion of each lung. The ala pulmonalis appears on either side of the fore-gut as a wing-like fold, which is flattened dorsoventrally and which lias a free lateral margin. The caudal portion of each ala becomes a pulmonary ligament (lig. pulmonale).

DEVELOPMENT OF THE LIVER,

425

ap. cor. s. c. pi.

ap. cor. d. c. per.

s. tr. I. hep. s. mes. I. hep. d. s. tr. Fig. 300.

ap. cor. s. c pi

ap. cor. d.

. hep. s.

I. hep. d.

om. m

v.c.

v. um. ves. fel.

Fig. 301.

for. ep

I. cor.

v. p. v. um. d. ch. ves. fel. for. ep. Fig. 302.

Figs. 300, 301, and 302. — Dorsal views of the hepatic region. Fig. 300, model from a 4 mm. embryo (Harvard Collection, Series 714), X 66 diam.; Fig. 301, model from a 9.4 mm. embryo (Harvard Collection, Series 1005), X 32 diam. ; Fig. 302, dissection of a 5 months' embryo, 220 mm. in length, X l'i diam. ap.cor. d., ap. cor. s., right and left coronary appendages; c. pi., pleural part of the pleuroperitoneal cavity; c. per., peritoneal part of the pleuroperitoneal cavity; d. ch., common bile-duct; dia., diaphragm; for. > p., foramen epiploicum; I. cor., coronary ligament; 1. hep, d., I. hep. «., right and left hepatic lobes; mes., mesentery; oes., oesophagus; om. m., lesser omentum; pl. v. c, plica venae cavae; r. s., superior lateral recess of the peritoneal cavity; s.tr. septum transversum; v.c, vena cava; ves. fel., gall-bladder; v. p., portal vein; v. um., umbilical vein.

426 HUMAN EMBRYOLOGY.

at the transverse fissure or porta; together with the gall-bladder it marks the true median plane of the liver. The umbilical vein is seen in the ventral abdominal wall, from which it passes to the liver along the free margin of the falciform ligament. It then lies in a deep groove on the under surface of the liver, and with the porta and the gall-bladder it bounds the quadrate lobe. Its extension, the ductus venosus, passes toward the vena cava at the bottom of the fissure of the lesser omentum.

In the preceding description the embryonic liver has been divided into right and left lobes, separated by the falciform ligament. Rex, after careful comparative studies of adult livers, declares that the only subdivision of the human liver which is a true lobe is the omental or caudate, but he admits that the recognition of the right and left lobes is justified by the distribution of the branches of the portal vein. Some embryologists have found this convenient. The absence of deep clefts, such as mark off the dorsolateral lobes in the embryonic liver of pigs and rabbits, is notable in the human liver. Nevertheless Swaen (1897) considers that corresponding lobes should be recognized, and accordingly he describes the human liver as composed of three lobes, — one median and two lateral. Mall (1906) states that each of the six primary lobules which he finds in an embryo of 11 mm. is to expand into a whole lobe, and Bradley (1908) shows the relation of six lobes to the three which he considers fundamental. Thompson (1899) has described the fissures and clefts which frequently appear on the under surface of the liver, especially of the right lobe. Certain of these were found with considerable regularity. The two most frequently met with, occurring respectively in 83 and 50 per cent, of the cases examined, were believed to form partial boundaries of a lobe which is well defined in the gorilla. The almost entire absence of lobes in the human liver has been emphasized by Rex.

The Liver as a Whole. — Except for a temporary decrease at birth, associated with the closure of the umbilical vein, the weight of the liver steadily increases. At the end of the second fetal month it weighs .2 gin. ; at birth, 75 gm. ; and in the adult 1500 gm. (Mall). But the volume of the liver, as compared with that of the body, reaches a maximum in a 31 mm. embryo, as determined by Jackson (1909). He found that in an 11 mm. embryo the liver is 4.85 per cent, of the total body volume, or approximately the same as at birth ; in a 17 mm. embryo it is 6.9 per cent. ; and at 31 mm. it is 10.56 per cent. (Pigs. 303 and 305). In a 65 mm. embryo (Figs. 304 and 306) it has apparently decreased to about 5 or 6 per cent., which is the average for the remainder of the fetal period. In this specimen the liver, as indicated by its relation to the ribs, has aco^ired approximately its final position.

During its development, certain portions of the liver atrophy, while other parts increase. The most extensive degeneration is in the peripheral part of the left lobe. In the 31 mm. embryo the two lobes are still nearly symmetrical, and the left lobe extends between the spleen and the body wall (Fig. 305). At 65 mm. "the liver has partly retracted, so that it covers only the anterior portion of the external splenic surface " (Fig. 306). In the adult

DEVELOPMENT OF THE LIVER.

4l'7

any contact between the liver and spleen is exceptional. The decrease in the size of the left lobe is generally ascribed to pressure from adjacent organs. Pougnault (1905) notes that in cases of

Fig. 303.

Fig. 304.

— gl. s.

Fig. 305.

Figs. 303 and 305. — Ventral and lateral views of a model of the viscera from a :->l nun. embryo. X 4'._» diam. Figs. 304 and 306. — Similar views of a model of the viscera from a 65 mm. embryo. X2 iliam. iAfter C. M. Jackson.) 2, 8, q, 12, ribs; at. d., <it. 8., right and left atria: co., colon; nL s., left suprarenal gland; h., liver; int., small intestine; lien, spleen; /. £., /. m., I. 8., inferior, middle, and superior lobes of the lung; p. v., vermiform process and cecum; r., kidney; v. 8., left ventricle; v. u., umbilical vein.

umbilical hernia the symmetry may be retained, and Jackson's models indicate that the decrease occurs when the intestines enter the abdomen. It may also be associated with the expansion of the gastric fundus. As a result of this degeneration, the left portion

428 HUMAN EMBRYOLOGY.

of the coronary ligament (the appendix fibrosa) contains a network of anastomosing ducts, discovered by Ferrein and described by Kiernan as a " rudimental liver. ' ' Both of these anatomists recognized similar tissue around the vena cava. Usually this vein occupies a fissure on the dorsal surface of the liver, but Kiernan states that the fissure is frequently converted into a canal, either by hepatic parenchyma or by a ligamentous band containing ducts and blood-vessels. He found that similarly the umbilical vein may be completely surrounded by hepatic tissue, or bridged by a band of the same structure as the lig amentum venae cavce. Hepatic trabecular may also invade the diaphragm, and at birth they have been reported as extending into the falciform ligament as far as the umbilicus. In all of these situations, and also near the expanding gall-bladder, the hepatic cells may degenerate, leaving aberrant ducts and blood-vessels. These have been studied through injections by Toldt and Zuckerkandl (1875). The marked variations in the form of the fetal liver have been tabulated by Ruge (1907).

Anomalies of the Liver. — The total absence of the gall-bladder, according to Meckel (1812), is not very unusual. In these cases the hepatic diverticulum has presumably developed normally, but has failed to produce the secondary subdivision which gives rise to the gall-bladder. Sometimes, in addition to the absence of the gall-bladder, there is no trace of the hepatic, cystic, and common bile-ducts. Kirmisson and Hebert have reported such a case in a child of one month, and they found two similar instances in the literature. These are probably due to obliterative processes which begin after the extra-hepatic bile-ducts have developed. Two gallbladders may be present, perhaps produced by a double outpocketing of the diverticulum. Sometimes when the gall-bladder is single there are two cystic ducts, as in a case reported by Dreesman. The two ducts arose from the gall-bladder 1 cm. apart, and united before entering the common bile-duct. Fig. 292, B, indicates how such an anomaly may develop. Beneke (1907) has studied congenital atresia of the bile-ducts. Multiple hepatic ducts have been recorded, sometimes opening separately into the duodenum.

Congenital cysts of the liver generally arise from the ducts in the connective tissue, but they may occur within the hepatic parenchyma (Moschcowitz, 1906). Sometimes they attain very large size (Sanger and Klopp, 1880). The subdivision of the liver into multiple lobes is quite common, and the occurrence of accessory livers, more or less isolated from the central mass, is well known (Toldt and Zuckerkandl, 1875). An excessive atrophy of the left lobe, leading to its "entire absence," has been recorded by Kantor (1903).

DEVELOPMENT OF THE PANCREAS. 429 DEVELOPMENT OF THE PANCREAS.

By FREDERIC T. LEWIS.

Historical Xote. — The main duct of the human pancreas, figured by \\ irsung in 1642, opens into the duodenum in common with the bile-duct. The regular occurrence of an independent accessory duct, opening into the duodenum somewhat nearer the pylorus, was recognized by Santorini in 1775. Meckel (1817) observed this accessory duct in several embryos. It was situated above and to the left of the bile-duct. Meckel mistook it for the only duct of the pancreas, and concluded, therefore, that " the bile and pancreatic ducts at first are quite separate from one another, but gradually they come together and unite." Kolliker (1879), in describing a rabbit embryo, stated that the pancreas is divisible into " two distinct glands which perhaps should be interpreted as an upper and a lowed' pancreas, such as are found in the chick." Not until 1888 was the similar condition observed in a human embryo. Phisalix then recorded that in a 10 mm. specimen the pancreas is represented by two separate outgrowths,—" one, superior and larger, the duct of which will become the accessory duct; the other, inferior and smaller, which corresponds with the canal of Wirsung." The upper gland is now known as the dorsal pancreas and the lower one as the ventral pancreas.

Early Development. — The two pancreases arise almost simultaneously shortly after the formation of the hepatic diverticulum, but, from the first, the dorsal pancreas is the larger. Both have been found in embryos of 3 and 4 mm. The failure of Fol (1884) to record a ventral pancreas at 5.6 mm., Mall (1891) at 7 mm., and Janosik (1909) at 6.1 mm., must be attributed to imperfect description or to abnormal embryos. But the fact that Volker (1903) in a 3 mm. specimen, and Keibel and Elze (1908) in an embryo of 4 mm., describe only a dorsal pancreas, may indicate that the dorsal pancreas arises first. Bremer (1906), however, found only a ventral pancreas at 4 mm., represented by two knobs of intestinal epithelium, one immediately below the hepatic diverticulum (Fig. 288, D), and the other nearer the yolk-stalk. It is doubtful whether these intestinal outgrowths represent a nQrmal stage in pancreatic development.

The dorsal pancreas is at first a "stomach-like" enlargement of the digestive tube. It is somewhat flattened laterally, and has a convex dorsal border which merges anteriorly with that of the intestine. Posteriorly the transition is more abrupt. The posterior part of the dorsal pancreas at 4.9 mm. is shown in Fig. 307, A. At 7.5 mm. (Fig. 307, B) the dorsal pancreas is separated from the intestine by a slight constriction, and the notch on the lower side is characteristically deeper than on the upper side. At 9.4 mm. (Fig. 308) the constricted part is prolonged into a short duct. The distal portion has also elongated, and its surface presents nodular swellings, which are the beginnings of branches.

The ventral pancreas in its early stages is lodged in the inferior angle formed by the hepatic diverticulum and the intes

430 tine. It is a small epithelial mass, continuous above with the diverticulum, and uniting dorsally with the intestinal epithelium. This condition is seen in a series of four models made by Keibel and Elze, descriptions of which have not been published. They are from embryos of 4, ca. 4, 5.3, and 6.75 mm. respectively. The same condition is shown in Fig. 307, in embryos of 4.9 and 7.5 mm. In the latter the lower part of the ventral pancreas has become free from the duodenum. Subsequently it becomes entirely separate from the intestine, as in the 6.8 mm. embryo modelled by Piper (1900) and in the 8 mm. specimen modelled by Felix (1892). With the elongation of the bile-duct it becomes widely separated from the duodenum, as in the 9.4 mm. embryo (Fig. 308).

Hamburger (1892) found that the ventral pancreas in an embryo of 4 weeks; had an independent opening into the duodenum, and concluded that its common outlet with the bile-duct is formed later. This has not been confirmed. However,

Ves. fel

Pane, v

Pane. d.

Div. hep.

Ves. fel.

Ga.

D. cyst

Pane. d.

Fig. 307. — -A, lateral view of the liver and pancreas of a 4.9 mm. embryo, from a model by N. W. Ingalls. X 65 diam. B, similar view of a model in which the hepatic trabeculse are not included, from a 7.5 mm. embryo. Modelled by F. W. Thyng. X 50 diam. D. chol., common bile-duct ; D. cyst., cystic duct ; D. hep., hepatic duct ; Div. hep\, hepatic diverticulum ; Ga., stomach ; Pane, d., dorsal pancreas ; Pane, v., ventral pancreas ; TV., trabecular ; Ves. fel. (vesica fellea), gall-bladder ; x, aberrant duct.

in the Bremer embiwo the ventral pancreas may arise directly from the intestine, and in a 3 mm. specimen, according to Keibel and Elze, the ventral pancreas is an outpocketing found " just caudal to the bile-duct." A pair of ventral pancreases are found in many vertebrates, and have been reported in a human embiwo of 4.5 mm. (Debeyre, 1909). Felix (1S92) recorded that in a section of the upper part of the pancreas at 8 mm., the lumen was toward the right side of the epithelial mass. He considered that the lumen belonged with a right ventral pancreas which had fused with a left ventral pancreas, and that the latter was rep resented by the solid left portion of the section. Jankelowitz (1S95), in a 4.9 mm. embiwo, found a single lumen below, which bifurcated above, sending its branches respectively to the right and left sides of the hepatic diverticulum. He considered that this indicated a fusion of right and left constituents. Ingalls (1907) studied the same specimen, and states that he is inclined to agree with Jankelowitz, although there is " only a suggestion of the paired condition." Keibel and Elze (1908) again examined this specimen, and they state that "it is very questionable whether two outgrowths are present; to us there appears to be only one."

DEVELOPMENT OF THE PANCREAS. 431 A double lumen is often seen in the ventral pancreas of older embryos (6.8 mm., Piper; 7.5 mm., etc.), but this condition, as noted by Helly, may be observed also in the impaired gall-bladder. It is associated with the formation of a lumen in a solid cord of cells.

Helly (1901 j and Kollmann (1907) have figured a pair of ventral pancreases which have not fused. In Kollmann's 7.5 mm. embryo they are cranial and caudal hi position, with the common hepatic duct between them. In Helly's 11 mm. embryo they are right and left. The left is much smaller and contains no distinct lumen. Helly believes that it degenerates without fusing with the righl pancreas, and that " in embryos scarcely older than four weeks it has wholly disappeared." In other embryos between 7.5 and 11 mm. in length, including seven specimens in I lie Harvard Collection, the ventral pancreas appears as a single outgrowth. Moreover, a well-defined pair is not recorded in Keibel and Elze's extensive series. Therefore the specimens described by Helly and Kollmann are properly regarded as exceptional.

A paired dorsal pancreas, such as Stoss described in sheep ernbryos, has been sought for in man, but has not been found (Felix, Helly).

Relative Position of the Dorsal and Ventral Outgrowths. — Although the dorsal pancreas of most mammals enters the duodenum on the distal side of the bile-duct, that of man is normally on the proximal side, toward the pylorus, at all stages of development. In early stages the caudal border of the duct of the dorsal pancreas may be at a lower level than the cranial border of the hepatic diverticulum, as seen in Fig. 307 and in Keibel and Elze's models of the pancreas at 4 mm. Later there is an interval between them which varies in extent. Thus, from Keibel and Elze's model of a 5.3 mm. specimen the distance is found to be only 0.05 mm., whereas in a 5 mm. embryo figured by Tandler (1903) it is approximately 0.2 mm. The distance has increased to 0.5 mm. in a 22.8 mm. embryo (0.7 mm. in a 15 mm. specimen, Swaen), and in the adult, according to Letulle and Nattan-Larrier (1898), who examined 21 cases, it varies from 10 to 35 mm. (20 to 40 mm., Charpy, 1898).

Migrations of the dorsal pancreas in relation to the bile-duct have been described. His (1885) figured the dorsal pancreas on the pyloric side of the bileduct in embryos of 5.7 and 10 mm., but at 11.5 mm. he placed it opposite the bile-duct, and at 1*2.5 and 13.S mm. it is shown on the caudal side. This would necessitate a return to the pyloric side in subsequent stages. Thyng (1908) examined 18 embryos from 7.5 to 24 mm. in length, and failed to rind a single) instance of the caudal position figured by His. Janosik (1805 and 1909) and Volker (1902 and 1903) have held that the dorsal pancreas arises on the distal side of the bile-duct, and that it migrates anteriorly. This is denied by Helly (1904). Janosik's reconstructions (1909) begin with an embryo of 6.1 nun., in which the dorsal pancreas connects with the intestine " a little more distally than the bile-duct." This embryo, however, must be considered abnormal, since it shows "no trace of a ventral pancreas." In the next stage figured (8.7 mm.) the ducts are opposite. Doubtless the duct of the dorsal pancreas may occasionally open into the intestine caudal to the common bile-duct, as is the ease in an embryo of 11.5 mm. in the Harvard Collection. Here, however, thei'e is an abnormal persistence of the adjacent portion of the right omphalomesenteric vein.

432 Union of the Dorsal and Ventral Pancreases. — With the elongation of the bile-duct, which bends dorsally on the right side of the intestine, the ventral pancreas is brought into close relation with

Tr.

„_,.

m

• • Jim

MP?m

^P.v.

"P.d.

D.

Fig. 308. — Model of a part of the liver and the pancreas of a 9.4 mm. embryo (Harvard Collection, Series 1005). X 50 diam. D., duodenum; D.c, common bile-duct; D.cyst., cystic duct; D.h. hepatic duct; P. d., dorsal pancreas; P. v., ventral pancreas; Tr., hepatic trabeculae; V. /., gall-bladder.

the dorsal pancreas (Fig. 308). Subsequently the ramifications of the two pancreases interlock, as shown in Fig. 309, from an embryo of 22.8 mm. In the model represented in the figure, the horizontal body and tail of the pancreas have been cut away at x. The

Pane, d

Pane. v.

D. pane. v.

D. chol. — —

Duo.

D. pane. d.

Fig. 309. — Model of the head of the pancreas of a 22.8 mm. embryo (Harvard Collection, Series 871). X 50 diam. D. chol., common bile-duct; D. pane, d., duct of the dorsal pancreas; D. pane, v., duct of the ventral pancreas; Duo. duodenum; Pane, d., dorsal pancreas; Pane. v. ventral pancreas; x and y are explained in the text.

pancreas at that point bends downward, forming the head of the organ, which in the adult terminates below in the uncinate process. The ventral pancreas forms a part of the head and more or less of

DEVELOPMENT OF THE PANCREAS. 433 the uncinate process; the dorsal pancreas forms the remainder of these parts, together with the entire body and tail.

At 22.8 mm. the duct of the dorsal pancreas is a round stein, which passes into a flattened, plate-like duct, strongly curved upon itself. The cleft leading into its concavity is shown at y in Fig. 309. The convex surface of the flattened portion of the duct is beset with nodular branches, radiating in all directions. Distally the main duct again becomes round, and it may be followed as an axial structure through the tail of the gland. The duct of the ventral pancreas arises from the common bile-duct at some distance from the duodenum. It passes to the centre of a group of ramifications which nearly equal it in diameter. In this embryo, and in specimens of 14.5 and 16 mm., no connection could be found between the two pancreases. Moreover, the branches of either pancreas rarely anastomose among themselves.

D. pane, d

D. pane. v.

1 — D. pane. d.

Fig. 310. — Corrosion preparation of the pancreatic ducts of an adult. Prepared by Dr. S. T. Mixter. X 1/^diam. D. pane, d., duct of the dorsal pancreas; D. pane, v., duct of the ventral pancreas; x, anastomosis between the pancreatic ducts.

In the adult the normal relations of the two glands are shown in the corrosion preparation, Fig. 310. In comparing this with Fig. 309, it will be noted that the duct of the dorsal pancreas appears to open into the duodenum at a higher level in Fig. 310 than in Fig. 309. This is associated with a shifting of the duodenum; in both cases the dorsal pancreas opens nearer the stomach than the ventral. In the adult the duct of the dorsal pancreas, shortly before entering the duodenum, receives a large branch which passes upward from the uncinate process. This branch is in front of the duct of the ventral pancreas. The latter, as in the embryo, lies at a deeper level, and is on the right side of the axis of the dorsal pancreas. The duct of the ventral pancreas forms a single large anastomosis with the duct of the dorsal pancreas, which is shown at x in Fig. 310. The continuous channel formed by the distal part of the duct of the dorsal pancreas, the anastomosis, and the duct of the ventral pancreas constitutes the "pan Vol. II.— 28

434 HUMAN EMBRYOLOGY.

creatic duct" of the adult; the proximal part of the duct of the dorsal pancreas is the "accessory duct." According to Hamburger (1892) the anastomosis between the dorsal and ventral pancreases has formed in a "six weeks' embryo." In a reconstruction of this specimen he shows that the distal end of an unbranched ventral pancreas has fused with the dorsal pancreas, which has a nodular surface but no branches. In a 14 mm. embryo Keibel and Elze (1908) found the pancreases ' ' close together but not yet united. ' ' This is the largest specimen in their series in which the pancreases are separate, and an embryo of 12.4 mm. is the smallest in which they have united. At 14 and 15 mm. they are generally described as "fused." In such embryos the tubules interlock, and it requires a careful study of drawings of successive sections to determine whether there is a passage between the ventral and dorsal ducts. Ordinarily only a single anastomosis is produced, but Bernard (1856), in an abnormal adult specimen, has shown two connections. Charpy (1898) has found that the duct of the ventral pancreas may enter the dorsal duct at any point in its wall, — that is, on its superior, inferior, anterior, or posterior surface. Usually it appears to enter on the inferior surface. In one of the specimens figured by Charpy, the main duct draining the uncinate process passes upward to join the duct of the dorsal pancreas behind the duct of the ventral pancreas, instead of in front of it as in Fig. 310. This arrangement is abnormal and difficult to explain. Hasse (1908) has shown the normal relation of these ducts in the adult, but his inferences regarding their development are incorrect.

Vessels and Nerves. — The dorsal pancreas in early stages is lodged between the right and left omphalomesenteric veins. These vessels form a transverse anastomosis immediately caudal to the pancreas, and branch abundantly around it (see Ingalls, 1908, pi. 2). Later, as described in Chapter XVIII, portions of these veins give rise to the portal vein. The vena ported reaches the inferior border of the pancreas in the notch between the body and head; it then passes behind the dorsal pancreas and curves forward, with the bile-duct, to enter the liver. In a 10 mm. embryo (Phisalix) the dorsal and ventral pancreases are completely separated by the portal vein; at 16 mm. (Fig. 311) they have come together and have partly surrounded the vein. The splenic branch of the portal vein develops early, and may be recognized in a 9.4 mm. embryo. It passes along the dorsal surface of the tail of the pancreas, which it drains. Some of its branches, and its opening into the portal vein, are indicated in Fig. 311.

Although the pancreas is at first in close relation with the portal vein, it does not give rise to a portal or sinusoidal circulation, and thus it differs strikingly from the adjacent liver. Its

DEVELOPMENT OF THE PANCREAS.

435

afferent blood supply is from the splenic and hepatic branches of the cceliac artery, and from small branches of the superior mesenteric artery as it accompanies the portal vein across the inferior border of the pancreas (Fig. 311). At 37 and 42 mm. the pancreatico-duodenal arteries form a loop, which connects the hepatic and superior mesenteric arteries and supplies the head of the pancreas.

In a 42 mm. embryo lymphatic vessels are abundant in the connective tissue around the pancreas, but they do not extend among the tubules. Lymph-glands have not developed. They are present in close relation with the pancreas in a 99 mm. embryo, but they may arise in some much younger stage.

Tr. hep.

D.panc.v

D.cyst.—^Sr

'.

. ?#^ 

Fig. 311. — Section through the stomach, pancreas, and a part of the liver, from an embryo of 16 mm. (Harvard Collection, Series 1322). y>X 40 diam. ; ,A. mes. sup., superior mesenteric artery; tB. oment., omental bursa ;'_Gaster, stomach; Lien, spleen; V. p., portal vein. (Other labels as in preceding figures.)

At 42 mm. the coeliac plexus of nerves sends branches toward the head of the pancreas, and some of them extend among the pancreatic tubules. Within the pancreas there are a few conspicuous groups of nuclei which, from their association with nerve fibres, are presumably ganglionic.

The Outlets of the Ducts. — In entering the duodenum the common bile-duct passes obliquely through the duodenal musculature, and is directed caudally. In its transit across the muscle, in embryos between 20 and 40 mm., it is joined by the duct of the ventral pancreas, and the pancreatic duct is always on its lower or caudal side (Figs. 309 and 311). At 22.8 mm. there is still no duodenal papilla at the outlet of the bile-duet, but Helly (1900) states that an elevation is present at 28.5 mm. In this embryo he

436 HUMAN EMBRYOLOGY.

finds that sphincter muscles have developed around the bile and pancreatic ducts, but in specimens of 44 mm., in the Harvard Collection, the ducts are surrounded only by concentric mesencliyma and by the duodenal muscle through which they pass.

In the adult, according to Letulle and Nattan-Larrier (1898), the pancreatic duct may empty into the bile-duct, in the same way as in the embryo. More often the two ducts reach the duodenal surface independently, either at the base of an ampulla or at the summit of a nipple-like projection.

The duct of the dorsal pancreas, in embryos between 20 and 40 mm., has a longer course within the duodenal wall than the duct of the ventral pancreas. Consequently, although the branches of the ventral pancreas extend close to the outlet of its duct, they are almost entirelv outside the duodenal musculature, whereas +' 7 branches of the dorsal pancreas are regularly found in the submucosa. Helly (1900) states that an outpocketing of the dorsal duct within the wall of the duodenum is present in embryos of 12.5 and 14.5 mm. At 37 and 42 mm. distinct knobs and diverticula are present. Some of them branch and give rise to pancreatic tissue. According to Helly, this explains why true pancreatic tissue is so often found in the papilla of the dorsal duct {papilla minor) of the adult, but almost never occurs in the papilla of the ventral duct {papilla major).

The large pancreatic ducts, both within the duodenal wall and outside of it, give rise to diverticula and mucous glands. Helly has determined that they occur in both papillae of an 80 mm. specimen, as very small outpocketings. In a 90 mm. embryo, and in all later stages, he finds that the mucous glands are easily recognized.

The Development of the Alveoli. — Kolliker (1861) described the pancreas of a four-weeks embryo as consisting of a simple wide and hollow duct, with branches, each of which has a lumen in its more slender proximal part, but terminates in a solid, pearshaped bud. Similar buds arise in later stages, not only in the terminal branches, but also along the sides of the main ducts. The duct from a 42 mm. embryo shown in Fig. 312, A, presents early stages in their development. At a there is a group of cells with crowded nuclei and darkly staining basal protoplasm. At b a similar group is seen at the bottom of an outpocketing of the lumen, and at c there is a larger mass which causes a basal bulging. These structures apparently give rise to the darkly staining knobs which are abundant in the 42 mm. embryo and in younger stages. Three of them, from a 55 mm. embryo, are shown in Fig. 312, B. Often they appear to be solid, but sometimes a slender lumen may be found within them, as shown at e. Considered as terminal parts of the gland, these buds may be called alveoli. They contain central cells which apparently persist as the central cells of the

DEVELOPMENT OF THE PANCREAS. 437 adult. The stalks of the alveoli become elongated, forming branches of the duct, and the alveoli subdivide. Thus in the adult, as seen in the model by Maziarsky (1902), the pyriform alveoli may be cleft nearly in two ; some of them show lateral buds. The extension of the lumen between and into the secreting cells of the alveoli, which has been shown by the Golgi method to occur in the adult (Dogiel, 1893), has not been studied embryologically.

The Development of the Islands. — The youngest human embryo in which the islands of the pancreas have been observed is a specimen of 54 mm. (Pearce, 1903). None are present in embryos of 42 and 44 mm. in the Harvard Collection, nor in the head of a pancreas at 55 mm. Weichselbaum and Kyrle (1909) find none at 50 mm. They appear first in the distal part of the pancreas. Thus, Pearce found none in the head at 90 mm., and in an embryo "believed to be of the third month . . . numerous

-a

m %

.

A B Fig. 312. — Sections of pancreatic tubules; A, from an embryo of 42 mm. (Harvard Collection, Series 838) ; B, from an embryo of 55 mm. X 350 diam. a-e , early stages in the formation of alveoli.

islands are scattered through the tail and body, while for the first time a few are seen in the head." Krister (1904) found them larger and more numerous in the splenic end in an embryo of the 17th week, and this accords with Opie's conclusion that in the adult the islands are almost twice as numerous in sections from the tail as in those from other parts.

The islands in an embryo of 99 mm. (Fig. 313, b) already resemble those of the adult. In sections stained with haematoxylin and eosin, they appear as pale areas, composed of anastomosing solid cords or rows of cells. Capillary blood-vessels extend among the cords, and their endothelium comes into close relation with the cells of the island. The presence of epithelial stalks connecting the islands with the ducts, as shown on the right of Fig. 313, has been observed by Pearce, Kiister. and Weichselbaum and Kyrle.

The general structure of the islands, and of the glomeruli which they contain, is well shown in a model prepared by Miss Dewitt. from the pancreas of an adult. Miss Dewitt (1906) failed to find any arteries connecting with the vessels of the islands, contrary to Laguesse (1900) and others. She regards the blood-vessels of

438 HUMAN EMBRYOLOGY.

the islands as venous, " with abundant capillary connections with the surrounding interalveolar capillary plexus," and describes them as sinusoids. In their development they are quite different from the portal sinusoids of the liver, but they resemble them histologically. Laguesse has described groups of red corpuscles as occurring normally between the endothelium and the cells of the islands.

An earlier stage of the islands than that shown in Fig. 313 has been described by Pearce. In a 54 mm. embryo he found them represented by small groups of from ten to fifteen cells, directly connected with the sides of the ducts. He described them as having round and lightly staining nuclei, with relatively abundant protoplasm which stains deeply with eosin. At this stage the islands are not penetrated by blood-vessels. Weichselbaum and Kyrle describe the islands in an 80 mm. embryo as solid buds, directly connected with the ducts, but composed of paler cells.

<$..

? ©/€)/

"^^k 9

• -X?

( As ' PM>$m&^- '

Vy' Si V

SI Hi fillip* Fig. 313. — Section of the pancreas from a 99 mm. embryo. X 350 diam. a, pancreatic tubules; b, islands; c, blood-vessel; d, alveolar bud.

They are partly surrounded by capillaries. The difficulty of distinguishing these developing islands from the alveolar buds is apparent in sections and in the published figures. Laguesse admits this difficulty, but he concludes that certain formations in sheep embryos, apparently comparable with those seen in Fig. 312, are the first stages in the development of the islands.

In later stages the islands become detached from the epithelial tubes. Some of them have separated in an embryo of 130 mm. (Weichselbaum and Kyrle). Von Hansemann (1910) found them all detached at 210 mm., and concluded that the islands arise from mesenchyma ; but in another embryo of the same length, and also at birth, Weichselbaum and Kyrle found that some islands are still in connection with the ducts. They believe that new islands may arise throughout life by budding from the ducts. The detached stalks of the older islands may be recognized in the small dark cells which have been described at the periphery of

DEVELOPMENT OF THE PANCREAS. 439 certain islands. Occasionally they show a lumen, and pathologically they may give rise to retention cysts (Weichselbaum and Kyrle). Kiister, in embryos of 24 and 32 weeks, has found stalks ending blindly in the islands.

Usually the islands are considered to possess " an anatomical identity as definite as that of the glomeruli of the kidney" (Opie), but some believe that alveoli may be transformed into islands and islands into alveoli. A review of the literature of this subject is presented by Laguesse (1906). Their embryological development does not accord with the idea that they represent a phase of glandular activity, and the presence of mitotic figures indicates that they are not degenerative structures.

The Pancreas at Birth. — In sections of embryos from 270 to 320 mm. in length, Weichselbaum and Kyrle find that the groups of alveoli are not only more compact than in earlier stages, but there is no longer such abundant connective tissue between them (cf. Fig. 313). Nevertheless the pancreatic connective tissue at birth, as compared with that in the adult, is relatively very abundant. It extends around individual alveoli, and forms broad septa between the clusters which are connected with the terminal ramifications of the ducts. These groups of alveoli, bounded by connective-tissue septa, become compact in the adult and constitute the lobules, which are ill-defined and may show secondary subdivisions.

"Islands are more numerous, as pointed out by Kasahara, in the pancreatic tissue of the fetus and of very young children than in that of the adult. . . . The organ being much smaller in the fetus, the same number of islands, though themselves smaller, are closer together and therefore appear more numerous in sections" (Opie). Kiister likewise found that, in relation to the alveoli, the islands are decidedly more numerous at birth than in the adult. His measurements show no difference in the size of the islands, but Miss Dewitt finds that they are smaller, on the average, at birth than in the adult.

Concerning the position of the islands, Opie states that, "though an island is often situated in the centre of a more or less clearly defined lobule, no constancy of position is discoverable." Pearce considers that the islands at first lie free in the connective tissue, but that later, in the fifth and sixth months, "glandular elements surround and inclose the island, and it then occupies the centre of the lobule." But, as noted by Weichselbaum and Kyrle, here and there, at birth, an island occurs at the periphery of a lobule, or in the interlobular connective tissue near a duct. Moreover several may be found within a single lobule.

The cells of the islands at birth lack distinct outlines; thev are crowded with fine granules which do not react to osmic acid (Stangl, 1901). Stangl states that fat appears in the cells of the

440 HUMAN EMBRYOLOGY.

islands at the end of the first year, but this has been denied by Symmers (1909). The cells of the alveoli at birth show the zones characteristic of the adult. In the outer zone they sometimes contain small scattered fat drops (Stangl). Histologically no differences have been established between the dorsal and ventral pancreases, at birth or in preceding stages.

Anomalies and Variations. — The obliteration of the proximal end of the duct of the dorsal pancreas is probably not infrequent.

Charpy (1898) found the papilla minor closed in three-fourths of the thirty eases which he examined. Letulle and Nattan-Larrier (1898) state that the accessory duct is permeable throughout its extent, including the papilla minor, in only three out of twenty-one cases examined, and that usually it appears as a branch of the pancreatic duct. But Helly (1898) concludes that an open accessoiy duct is "by far the rule" (compare with Fig. 311), and Hamburger found it present in all of the fifty cases which he examined. In an embryo of 55 mm. Helly (1900) found, in place of a single papilla minor, two papillae of nearly equal size,, each of which contained a pancreatic duct. Letulle and Nattan-Larrier have recorded a similar case in the adult. The failure of the dorsal and ventral pancreases to unite, so that the duct of the dorsal pancreas persists as the main duct, opening at a normally situated papilla minor, has been figured by Charpy, and reported by Helly, Baldwin (1907), and others. In one of Helly's cases "an independent duct of Wirsung was not to be found." In a specimen figured by Bernard the duct of the dorsal pancreas persists as the main duct, although it has two small anastomoses with the duct of the ventral pancreas. As recorded in a previous section, the dorsal pancreas in an abnormal embryo of 11.5 mm. opens into the intestine lower down than the bile-duct. Charpy has figured an adult pancreas which shows this relation ; in this case the. ducts have not anastomosed.

Pancreatic tissue sometimes surrounds such adjacent structures as the portal vein, the bile-duct, and the intestine.

The encircling of the portal vein by a process of the dorsal pancreas has apparently not been observed in man, though characteristic of the rabbit and pig (Thyng). The common bile-duct occupies a groove in the head of the pancreas which is frequently converted into a canal of pancreatic tissue (see Helly, 1898). An annular pancreas, encircling the intestine, has been recorded by Ecker and by Symington (as cited by Thyng) and a case has been reported by Baldwin. An abnormal condition observed in a pig embryo of 12 mm. suggests that this anomaly may arise early in development, by an extension of the ventral pancreas dorsally on either side of the intestine.

Accessory pancreases are of frequent occurrence.

Among 150 autopsies, Symmers found three cases in which there were accessory glands of considerable size. Sometimes two occur in a single case (Opie, 1903). Gardiner (1907), who has reviewed the literature, finds that in nearly a third of the cases reported, the accessory pancreases are connected with the stomach. They occur also in the duodenum, jejunum, and ileum, and have been frequently found at the apex of a " true " diverticulum. These diverticula were at first considered to be remnants of the vitelline duct (Meckel's diverticula), but Neumann (1870) questioned this interpretation. Nauwerck (1893) reported a diverticulum 9 cm. long, tipped with an accessory pancreas, and situated 2.3 metres above the valve of the colon. In the same case he found another diverticulum, 3 cm. long, situated 80 cm. above the valve of the colon, and he regarded the latter

DEVELOPMENT OF THE LIVER AND PANCREAS. 441 as Meckel's diverticulum. Hanau (Brunner, 1S99) reports a duodenal diverticulum tipped with an accessory pancreas, and Weichselbaum (Gardiner, 1907) has described a similar pancreatic diverticulum of the stomach. It is evident that these diverticula are not vitelline remains, and yet it is not impossible that an accessory pancreas may be associated with a true Meckel's diverticulum. Wright (1901) has reported a case in which pancreatic tissue was excised from the umbilicus of a child of 12 years, who had an umbilical fistula since birth. The fistulous tract had apparently become separated from the intestine.

Accessory pancreases generally penetrate the muscularis, but they may be limited to the submucosa. The larger ones show lobules composed of typical pancreatic alveoli. Islands have been reported in numerous cases, including that of Wright. Sometimes, however, the islands are lacking, and the tubules may be duct-like rather than glandular.

The accessory pancreases develop from elongated epithelial buds, as observed in the wall of the stomach of a 19 mm. entbryo. Lewis and Thyng (1908) have frequently found similar buds along the intestine of pig embryos of 10-20 mm., but not in human embryos. They usually become detached and degenerate. It is possible that accessory pancreases sometimes develop in relation with the embryonic intestinal diverticula described in a previous section.

LITERATURE. (On the development of the liver and pancreas.) Von Baer, K. E.: Ueber Entwicklungsgeschichte der Tiere. T. 1, S. 1-269.

Konigsberg. 1S28. Baldwin, W. M. : The Ductus pancreaticus accessorius in Man. Anat. Rec. Vol.

1, p. 66. 1907. A Specimen of Annular Pancreas. Anat. Rec. Vol. 4, p. 299-304, 1910. Beneke, R. : Die Entstehung der kongenitalen Atresie der grossen Gallengange.

Marburger Lniversitatsprogramm. 1907. Bernard, C. : Memoire sur le pancreas. P. 1-190. Paris. 1856. v. Biesiadecki, A. : Untersuchungen ueber die Gallen- and Lymphgef asse der Menschenleber in pathologischen Zustanden. Sitz.-Ber. der kais. Akad. der Wiss. Wien. Bd. 55, Abt. 1, S. 655-666. 1867. Brachet: Recherches sur le developpement du pancreas et du foie. Joum. de l'Anat. et de la Phys. Vol. 25, p. 620-696. 1891. Bradley, O. C. : A Conti-ibution to the Morphology and Development of the Mammalian Liver. Journ. of Anat. and Phys. Vol. 43, p. 1-42. 1908. Bremer, J. L. : Description of a 4 mm. Human Embryo. Amer. Journ. of Anat.

Vol. 5, p. 459^80. 1906. Brunner, C. : Ein Beitrag zur Chirurgie und pathologischen Anatomie der Darm invagination. Beitrage zur klin. Chir. Bd. 25, S. 251-360. 1899. Bischoff, T. L. W. : Entwicklungsgeschichte des Hunde-Eies. S. 1-134. Braunschweig 1845. Brissaud and Sabourin : Sur la constitution lobulaire du foie et les voies de la circulation sanguine intra-hepatique. Comptes rendus de la Soc de Biol.

Vol. 40, p. 757-762. 1888. Charpy, A. : Varietes et anomalies des canaux pancreatiques. Journ. de l'Anat.

Annee 34, p. 720-734. 189S. Chipmann, W. W. : Observations on the Placenta of the Rabbit, with Special Reference to the Presence of Glycogen, Fat and Iron. Studies from the Royal Victoria Hospital. Vol. 1, No. 4, p. 1-261. Montreal. 1902. Debeyre, A. : Les premieres ebauches du pancreas chez l'embryon humain.

Bibliogr. anat. Vol. 18, p. 249-256. 1909. Le foie est il d'origine endo dermique ou mesodermique ? Bibliog. anat. Vol. 19. p. 202-211. 1910.

442 HUMAN EMBRYOLOGY.

Dewitt, L. M. : Morphology and Physiology of Areas of Langerhans in some Vertebrates. Journ. of Exp. Med. Vol. 8, p. 193-239. 1906. Dogiel, A. S. : Zur Frage ueber die Ausf iihrungsgange des Pankreas des Menschen.

Arch, f . Anat. und Entw. S. 117-122. 1893. Dreesmann : Beitrag zur Kenntniss der kongenitalen Anomalien der Gallenwege.

Deutsche Zeitschr. f. Chir. Bd. 92, S. 401-412. 1908. Elze, C. : Besehreibung eines menschlichen Embryo von ca. 7 mm. grosster Lange.

Anat. Hefte. S. 111-492. 1907. Ferrein : " Mem. de lAcad. des Sciences 1733." , Felix, W. : Zur Leber- und Pankreasentwieklung. Arch. f. Anat. und Entw.

S. 281-323. 1892. Fol, H. : Description d'un embryon humain de cinq millimetres et six dixiemes.

Recueil zool. suisse. T. 1, p. 357-401. 1884. Gardiner, J. P.: A Case of Congenital Hour-glass Stomach with Accessory Pancreas. Journ. of the Amer. Med. Assoc. Vol. 49, p. 1598-1600. 1907. Geraudel, E. : Le parenchyme hepatique et les voies biliaires sont deux formations genetiquement independantes. Journ. de lAnat. et de la Phys. Vol. 43, p. 410-432. 1907. v. Hansemann : Ueber die Struktur und das Wesen der Gef assinseln des Pankreas.

Verh. der deutschen path. Gesellseh. S. 187-196. Hamburg 1901. Hamburger, O.: Zur Entwicklung der Bauchspeicheldriise des Menschen. Anat.

Anz. Bd. 7, S. 707-711. 1892. Hammar, J. A. : Ueber einige Hauptziige der ersten embiyonalen Leberentwick lung. Anat. Anz. Bd. 13, S. 233-247. 1897. Hasse, C. : Die Ausfiihrwege der menschlichen Bauchspeieheldruse. Anat. Anz.

Bd. 32, S. 417-420. 1908. Helly, K. K. : Beitrag zur Anatomie des Pankreas und seiner Ausf iihrungsgange.

Arch. f. mikr. Anat. Bd. 52, S. 773-793. 1898. Zur Entwicklungsgeschichte der Pankreasanlagen und Duodenalpapillen des Menschen. Arch. f. mikr. Anat. Bd. 56, S. 291-308. 1900. Zur Pankreasentwieklung der Saugetiere. Arch, f . mikr. Anat. Bd. 57, S. 271 335. 1901. Zur Frage der primaren Lagebeziehungen beider Pankreasanlagen des Menschen. Arch. f. mikr. Anat. Bd. 63, S. 631-635. 1904. Hendrickson, W. F. : The Development of the Bile Capillaries as revealed by Golgi's Method. Johns Hopkins Hosp. Bull. Vol. 9, p. 220-221. 1898. Henle: Zur Physiologie der Leber. Nachr. von der Georg- Augusts Univ. zu Gottingen. S. 338-352. 1861. Hering, E. : Von der Leber. Kap. 18 in Strieker's Handbuch der Lehre von den Geweben. S. 429-452. Leipzig 1871. Herring, P. T., and Simpson, S. : On the Relation of the Liver Cells to the Bloodvessels and Lymphatics. Proc. of the Roy. Soc. London. Vol. 78, p. 455 497. 1906. Hertwig, O. : Lehrbuch der Entwicklungsgeschichte des Menschen und der Wirbel tiere. S. 1-706. 8. Aufl. Jena 1906. His, Vf. : Beitrage zur Kenntniss der zum Lymphsysteni gehorigen Driisen.

Zeitschr. f . wiss. Zool. Bd. 10, S. 333-357. * 1860. Anatomie menschlicher Enibryonen. I. Embryonen des ersten Monats. S. 1 184. Leipzig 1888. III. Zur Geschichte der Organe. S. 1-260. Leipzig 1885. Ingalls. N. W. : Besehreibung eines menschlichen Embryos von 4.9 mm. Arch, f .

mikr. Anat. Bd. 70, S. 506-576. 1907. A Contribution to the Embryology of the Liver and Vascular System in Man.

Anat. Record. Vol. 2, p. 338-334. 1908.

DEVELOPMENT OF THE LIVER AND PANCREAS. 443 Jackson, C. M. : On the Developmental Topography of the Thoracic and Abdominal Viscera. Anat. Rec. Vol. 3, p. 361-396. 1909. Jankelowitz, A. : Ein junger menschlicher Embryo und die Entwicklung des Pankreas bei demselben. Arch. f. mikr. Anat. Bd. 46, S. 702-708. 1895. Janosik, J.: Zwei junge menschliche Embryonen. Arch. f. mikr. Anat. Bd. 30 S. 559-595. 1887. Le pancreas et la rate. Bibl. anat. T. 3, p. 68-73. 1895. Sur les rapports du conduit ckoledoque et des conduits pancreatiques chez l'homme. Bull, internat. de l'Acad. des Sci. de Boheme. P. 1-11. 1909. Kantor, H. : Zwei Fiille von Lebermissbildung. Virchow's Arch. Bd. 174, S. 571 576. 1903. ELeibel, F., and Elze, C. : Normentafel zur Entwicklungsgeschichte des Menschen.

S. 1-314. Jena 1908. Kiernan, F. : The Anatomy and Physiology of the Liver. Trans, of the Roy. Soc.

of London. P. 711-770. 1833. KIermisson et Hebert: Absence congenitale des voies biliaires extra-hepatiques.

Bull, et mem. de la Soc. anat. de Paris. Annee 77, p. 317-320. 1903. Koelliker, A. : Ueber die Blutkorperchen eines menschliehen Embryo und die Entwicklung der Blutkorperchen bei Saugetieren. Zeitschr. f. rat. Med.

Bd. 4, S. 112-159. 1846. Entwicklungsgeschichte des Menschen. S. 1-468. Leipzig 1861. Entwicklungsgeschichte des Menschen und der hdheren Tiere. S. 1-1033. 2.

Aufl. Leipzig 1879. Kollmann, J. : Handatlas der Entwicklungsgeschichte des Menschen. Bd. 2.

Jena 1907. Kon, J.: Das Gitterfasergeriist der Leber unter normalen und pathologischen Verhaltnissen. Arch. f. Entw.-mech. der Organismen. Bd. 25, S. 492-522.

1908. Kuester, H. : Zur Entwicklungsgeschichte der Langerhans'schen Inseln hn Pankreas beim menschliehen Embryo. Arch. f. mikr. Anat. Bd. 64, S. 158-172.

1904. v. Kupffer, C. : Ueber Sternzellen der Leber. Arch, f . mikr. Anat. Bd. 12, S. 353-358. 1876. Ueber die sogenannten Sternzellen der Saugetierleber. Arch. f. mikr. Anat.

Bd. 54, S. 254-288. 1899. Laguesse, E. : Le pancreas. (Deuxieme partie.) Revue gen. d'hist. T. 2, p. 1 288. 1906. Letulle, M., and Nattan-Larrier, L. : Region vaterienne du duodenum et ampoule de Vater. Bull, de la Soc. anat. de Paris. Annee 78, p. 491-506.

1898. Lewis, F. T. : The Gross Anatomy of a 12 mm. Pig. Amer. Journ. of Anat. Vol.

2, p. 211-225. 1903. The Question of Sinusoids. Anat. Anz. Bd. 25, S. 261-279. 1904. Lewis, F. T., and Thyng, F. W. : The Regular Occurrence of Intestinal Diverticula in Embryos of the Pig, Rabbit, and Man. Amer. Journ. of Anat.

Vol. 7, p. 505-519. 1908. Lieberkuehn, N. : Ueber die Allantois und die Nieren von Saugetierembryonen.

Sitz.-Ber. der Gesellsch. zur Bef. der ges. Naturwiss. zu Marburg. S. 1-11.

1876. Lobenhoffer, W. : Ueber extravaskulare Erythropoese in der Leber unter pathologischen und normalen Verhaltnissen. Beitr. zur. path. Anat. und zur allg. Path. Bd. 43, S. 124-146. 1908. MacGillavry, T. H. : Zur Anatomie der Leber. Sitz.-Ber. der kais. Akad. der Wiss. Wien. Bd. 50, Abt. 2, S. 207-224. 1864.

444 HUMAN EMBRYOLOGY.

Mall, F. P. : A Human Embryo Twenty-six Days Old. Journ. of Morph. Vol. 5, p. 459-480. 1891. Reticulated Tissue and its Relation to the Connective-tissue Fibrils. Johns Hopkins Hosp. Rep. Vol. 1, p. 171-208. 1896. A Study of the Structural Unit of the Liver. Amer. Journ. of Anat. Vol. 5, p. 227-308. 1906. Maresch, R. : Leber Gitterfasern der Leber. Centralbl. f. allg. Path. Bd. 16, S. 641-649. 1905. Maureb, P. : Die Entwicklung des Damisysteins. Handb. der vergl. und exp.

Entw. der Wirbeltiere. Herausg. von O. Hertwig. Bd. 2, T. 1, S. 109-252.

1906. Maziarsky, S. : Ueber den Bau und die Einteilung der Driisen. Anat. Hef te.

Abt. 1. Bd. 18, S. 171-237. 1902. Meckel, J. F. : Handbuch der pathojogischen Anatomic Leipzig 1812.

Bildungsgeschichte des Darmkanals der Saugetiere und namentlich des Menschen. Deutsches Arch, f, die Phys. Bd. 3, S. 1-84. 1817. Minot, C. S. : Human Embryology. P. 1-815. New York 1892.

On a Hitherto Unrecognized Form of Blood Circulation without Capillaries in the Organs of Vertebrata. Proc. of the Boston Soc. of Nat. Hist. Vol. 29, p. 185-215. 1900. Mollier, S.: Die Blutbildung in der embryonalen Leber des Menschen und der Saugetiere. Arch. f. mikr. Anat. Bd. 74, S. 474-524. 1909. Moschcowitz, E. Non-parasitic Cysts (Congenital) of the Liver. Amer. Journ. of the Med. Sci. Vol. 131, p. 674-699. 1906. Nattan-Larrier, L. : Formation de la graisse dans le f oie du foetus. Comptes rendus de la Soc. Biol. Paris. T. 55, p. 1602-1603. 1903. Le tissu hyeloide du foie foetal. Arch, de med. exper. Vol. 16, p. 641-654.

1904. Nauwerck, C. : Ein Nebenpankreas. Beitr. zur path. Anat. Bd. 12, S. 29-32.

1893. Neumann, E.: Nebenpankreas und Darmdivertikel. Arch, der Heilk. Bd. 11, S. 200-201. 1870. Neue Beitrage zur Kenntnis der Blutbildung. Arch, der Heilk. Bd. 15, S. 441-476. 1874. Opie, E. L. : On the Histology of the Islands of Langerhans. Johns Hopkins Hosp. Bull. Vol. 11, p. 205-209. 1900. Anatomy of the Pancreas. Johns Hopkins Hosp. Bull. Vol. 14, p. 229-232.

1903. Pearce, R. M. : The Development of the Islands of Langerhans in the Human Embryo. Amer. Journ. of Anat. Vol. 2, p. 445-455. 1903. Phisalis, C. : Etude d'un embryon humain de 10 millimetres. Arch, de Zool.

Exper. et Gen. Ser. 2. T. 6, p. 279-350. 1888. Piper, H. : Ein menschlicher Embryo von 6.8 mm. Nackenlinie. Arch. f. Anat.

und Entw. S. 95-132. 1900. Pougnatjlt, C. : Le developpement du lobe gauche du foie. These, Paris 1905. Ravn, E. : Ueber die Bildung der Scheidewand zwischen Brust- und Bauchhbhle in Saugetierembryonen. Arch. f. Anat. und Entw. S. 123-154. 1889. Remak, R. : Untersuchungen ueber die Entwicklung der Wirbeltiere. S. 1^194.

Berlin 1855. Rex, H. : Beitrage zur Morphologie der Saugerleber. Morph. Jahrb. Bd. 14, S. 517-617. 1888. Riess, L. : Beitrage zur Struktur der Gailengange der menschlichen Leber. Arch, f .

Anat., Phys. und wiss. Med. S. 473-501. 1863. Ruge, G. : Die ausseren Formverhaltnisse der Leber bei den Primaten. VI.

Morph. Jahrb. Bd. 37, S. 397^87, 614-660. 1907.

DEVELOPMENT OF THE LIVER AND PANCREAS. £45 Sanger, M., u. Klopp, A.: Zur anatomischen Kenntniss der angeborenen Bauch cysten. Arch. f. Gyn. Bd. 16, S. 415-435. 18S0. Santorini, J. D. : Septemdecim tabula?, quas nunc primum edit, etc. Michael Girardi. Tabulae XII et XIII. Parmae 1775. Schenk, S. L. : Lehrbuch der vergleichenden Embryologie der Wirbeltiere. S. 1 198. Wien 1874. Lehrbuch der Embryologie des Menschen und der Wirbeltiere. S. 1-688.

Wien und Leipzig 1896. Schmidt, M. B. : Ueber Blutzellenbildung in Leber und Milz unter normalen und pathologischen Verhaltnissen. Beitr. zur patb. Anat. und allg. Path. Bd. 11, S. 199-233. 1892. Stangl, E. : Zur Histologic des Pankreas. Wiener klin. Wocbenschr. Bd. 14, S. 964-968. 1901. Van der Stricht, 0. : Nouvelles recherches sur la genese des globules rouges et des globules blancs du sang. Arch, de Biol. T. 12, S. 199-344. 1892. Swaen, A. : Recherches sur le developpement du f oie, du tube digestif, de Parriere cavite du peritoine et du mesentere. Journ. de PAnat. et de la Phys. Vol. 33, p. 32-99, 222-258, 525-585. 1897. Symmers, D. : The Occurrence of Fat in the Islands of Langerhans. Areh. of Int.

Med. Vol. 3, p. 279-285. 1909. Tandler, J. : Zur Entwicklungsgeschichte der menschlichen Darrnarterien. Anat.

Hefte. Bd. 23, S. 187-210. 1903. Thoma, R. : Untersuchungen ueber die Histogenese und Histomechanik des Gef asssystems. S. 1-91. Stuttgart 1893. Thompson, A. : The Morphological Significance of Certain Fissures in the Human Liver. Journ. of Anat. and Phys. Vol. 33, p. 546-564. 1899. Thompson, P. : A Note on the Development of the Septum Transversum and the Liver. Journ. of Anat. and Phys. Vol. 42, p. 170-175. 1908. Thvng, F. W. : Models of the Pancreas in Embryos of the Pig, Rabbit, Cat, and Man. Amer. Journ. of Anat. Vol. 7, p. 489-503. 1908.' Toldt, C, and Zuckerkandl, E. : Ueber die Form und Texturveranderungen der menschlichen Leber wahrend des Waehstums. Sitz.-Ber. der kais. Akad. der Wiss. Wien. Bd. 72, Abt, 3, S. 241-295. 1875. Uskow, N. : Ueber die Entwicklung des Zwerchfells, des Perieardiums und des Coloms. Arch. f. mikr. Anat. Bd. 22, S. 143-219. 1883. Virchow, R. : Ueber das Epithel der Gallenblase und ueber einen intermediaren Stoffwechsel des Fettes. Virchow's Arch. Bd. 11, S. 574-578. 1857. Voelker, O.: Beitrage zur Entwicklung des pancreas bei den Amnioten. Arch.

f. mikr. Anat. Bd. 59, S. 62-93. 1902. Ueber die Verlagerung des dorsalen pancreas beim Menschen. Arch. f. mikr.

Anat. Bd. 62, S. 727-733. 1903. Weichselbaum, A., u. Kyrle, J. : Ueber das Verhalten der Langerhans'schen Inseln des menschlichen pancreas im fetalen und postfetalen Leben. Arch.

f. mikr. Anat. Bd. 74, S. 223-258. 1909. Wirsung. G. : Figura ductus cujusdam cum multiplicibus suis ramulis noviter in pancreate inventis in diversis corporibus humanis. Padua 1642. Wright, J. H. : Aberrant Pancreas in the Region of the Umbilicus. Journ. of the Boston Soc. of Med. Sei. Vol. 5, p. 497-498. 1901. Zweifel: Untersuchungen ueber das Meconium. Arch. f. Gyn. Bd. 7. S. 474-490.

1875.

THE DEVELOPMENT OF THE PHARYNX AND OF THE ORGANS OF RESPHtATION.

By OTTO GROSSER, of Prague.

A. THE PHARYNX /AND ITS DERIVATIVES.

The formation of the anterior part of the digestive tract has already been described at the beginning of this chapter, and it will be necessary to consider here only its further differentiation.

At first it is merely a short, tubular outgrowth of the yolksack, somewhat flattened dorsoventrally ; the oral portion of its ventral wall, in the region of the pharyngeal membrane, rests on the ectoderm, and its rostral end projects somewhat beyond this membrane as Seessel's pouch (Figs. 314-316; see also the section of this chapter dealing with the development of the mouth cavity). Whether it presents further differentiations at the time of its first formation cannot be stated with certainty, but in the youngest human embryos that have been studied and in which it is already present the anlage of the first pharyngeal pouch is apparent (embryos of Kromer-Pfannenstiel and Dandy, Fig. 314).

In all Craniota there are formed bilaterally symmetrical lateral diverticula of the anterior portion of the digestive tract, which, pressing aside the lateral mesoderm of the head, come into apposition with corresponding invaginations of the ectoderm; the endodermal diverticula are termed pharyngeal pouches (also pharyngeal grooves, or inner branchial grooves or pouches), while the ectodermal invaginations are known simply as branchial grooves, or as ectodermal or outer branchial grooves (outer pharyngeal grooves, Hammar). By pressing aside the mesoderm the ectoderm and endoderm for a time come into contact and fuse, forming the epithelial closing membrane, which breaks through in all forms that have a branchial respiration; the branchial grooves and pharyngeal pouches thus become continuous and together form the branchial clefts. The formation of open branchial clefts occurs also in reptiles and birds, but not, under normal conditions, in mammals (see below). The pharyngeal pouches and branchial grooves are later again separated by the ingrowth of mesoderm or (sinus cervicalis) by the constriction of the branchial grooves from the surface and the subsequent modification of their epithelium. The number of pharyngeal pouches that are formed in succession on either side varies in the gnathostomatous Craniota from nine to five, the number in general diminishing with an increasing degree of organization.

Between the branchial clefts — that is to say, between the pharyngeal pouches and branchial grooves — are the branchial or visceral arches, each of which contains a skeletal rod, the cartilaginous branchial arch, its musculature, an aortic arch, and a nerve-trunk. The branchial arches are named in succession the mandibular, hyoid, and branchial arches proper, these last being numbered in succession from before backward. Behind the last branchial cleft lies the last branchial arch, the number of arches being one more than that of the clefts, an arrangement determined by the formation of branchial leaflets on both walls of the clefts. The first branchial cleft is also known as the hyomandibular cleft.

PHARYNX AND ORGANS OF RESPIRATION.

447

I. General Morphology of the Pharyngeal Pouches.

The anlagen of the pharyngeal pouches appear in succession and are formed earlier than the corresponding external branchial grooves, as was observed by Eiickert and Piersol. In man the external grooves first become evident when the pharyngeal pouches have come into contact with the ectoderm, — that is to say, when the lateral cranial mesoderm has been pressed aside and the closing

Rh.

Fig. 314.

Fig. 315.

Fig. 314. — Pharynx of the embryo Klb (Kromer-Pfannenstiel ; Normentafel, No. 3; 5-6 primitive segments, length, determined from the number of sections, 1.38 mm.). Ds., yolk-sack; Rh., pharyngeal membrane; 1., 2., etc., St., first, second, etc., pharyngeal pouch; Thyr. or Th., thyreoid; V. S., ventral pharyngeal groove. X 150. In all the models the epithelial lining of the cavity is represented, not the cavity itself.

Fig. 315. — Pharynx of the embryo Rob. Meyer No. 335 (9-10 pairs of primitive segments, length, determined from the number of sections, 1.70 mm.). Ka., doubtful branchial anlage. The remaining lettering as in Fig. 314. X 150.

membrane* formed (Fig. 316). At about the time of the formation of the two first membranes the closure of the neural canal in the brain region occurs, and there is an increase in the amount of the cranial mesoderm, whereby for the first time the transverse diameter of the skull notably surpasses that of the pharyngeal tube and opportunity is afforded for the formation of the external grooves (developmental period between the formation of the 10th and the 15th pairs of primitive segments).

448 The pharyngeal pouches grow out from the anterior part of the digestive tract not only directly laterally but also somewhat dorsally; moreover, a groove, known as the ventral pharyngeal groove, extends along the ventral wall of the pharynx from each pouch, and is recognizable even at its first formation (Figs. 314 to 317). While the surfaces of contact with the ectoderm, i.e., the closing membranes, have at first a somewhat circular outline (Fig.

l.St.

S. St.

1

Fig. 316. — Pharynx of the embryo Hah in the collection of the First Anatomical Institute, Vienna (about 15 pairs of primitive segments, length about 3 mm.). Ect., ectoderm; Ka., doubtful branchial anlage; Rh., pharyngeal membrane, broken through in two spots. The remaining lettering aa in Fig. 314. X 150.

316), they later elongate to a long, narrow strip, almost perpendicular to the axis of the pharynx (Figs. 317 and 318). By an increase in depth — that is to say, a lateral extension of the pouches (a process that depends upon an increase in the thickness of the intervening branchial arches)— these structures become sharply marked off from the principal lumen of the pharynx (compare Figs. 316 and 317). In a pouch which has thus become relatively

PHARYNX AND ORGANS OF RESPIRATION.

449

narrow and deep (Figs. 317 and 318) there is to be distinguished a cranial, a caudal, and an indistinctly delimited dorsal surface, a lateral edge, a dorsal and a ventral angle, and a ventral pharyngeal groove extending toward the mid-ventral line. By the formation of these grooves the branchial arches become marked

Ar. mes

- 3. St.

4. St.

Fig. 317.— Pharynx of the embryo Rob. Meyer No. 300 (Normentafel Xo. 7, 23 pairs of primitive segments, 2.5 mm. vertex-breech length) seen from the ventral surface. The closing membranes of the gillclefts are outlined Jin black. At. mes., the somewhat depressed oral end of the area mesobranchialis ; Hw., heart swelling; Lar., laryngotracheal groove; V.V., ventral prolongations of the pharyngeal 'pouches. Other lettering as^in Fig. 314. X 150.

out upon the ventral wall of the pharynx as elevations projecting into its lumen. The grooves (with, perhaps, the exception of the first) do not, however, at first reach the mid-line, and accordingly leave an area in that situation, known as the area mesobranchialis (His, 1885) (Figs. 317 and 319). In the caudal part of this lies the heart swelling (Fig. 317), but slightly marked in the human embryo. In the three anterior pouches, and later on also in the fourth, the dorsal angle grows dorsally a little beyond the region Vol. IT.— 29

450 of Ihe closing membrane and so forms the dorsal prolongation (dorsal diverticulum) of the pouch (Born, Piersol, Hammar, 1902; Ta i idler, 1909 ). 18 Piersol terms the ventral angle, together with the lateral edge and the ventral groove, the wing of the pouch. By the continued deepening of the lateral parts of the ventral grooves a ventral prolongation of the pouch is formed, most distinct in the

Mw.

L.

Fia. 318. — The model shown in Fig. 317 from the lateral surface. Ka., doubtful branchial anlage; L., liver; Lar., laryngotracheal groove; Lu., lung anlage; Mw., angle of mouth. Other ^lettering as in Fig. 314. X 150.

second and third pouches (Fig. 319; indicated in Fig. 317) ; this is the ventral prolongation of Hammar (1902) and the ventral diverticulum of Fox (1908).

As a result of the elongation of the closing membranes and the formation of the dorsal and ventral diverticula, the dorso

18 H. Rabl (1909) does not recognize a dorsal diverticulum as of general occurrence in mammals, and ascribes to it, in any event, no special significance (in the formation of the epithelial bodies).

PHARYNX AND ORGANS OF RESPIRATION.

451

ventral diameter of the pharyngeal pouches increases much more rapidly in their lateral than in their medial portions, and a distinct delimitation of the pouches medially now becomes possible (Figs. 319 and 320). One may, with H. Rabl (1907 and 1909), term the at first uniformly broad pouches the primary pouches and the later stages, with the lateral portions broadened, the secondary pouches; the latter are connected with the pharynx by narrow connecting portions, the ductus pharyngo-branchiales. These ducts later become verv distinct in the caudal pouches (third to the fifth) (Fig. 325).

Fia. 319. — Pharyngeal pouches of the embryo Hah of the First Anatomical Institute, Vienna (5.2 mm., about the same stage as No. 16 of the Normentafel). The closing membranes of the left side are outlined in black. The thyreoid has been omitted. (From a model in the Institute.) Lor., anlage of larynx. Other lettering>s in Fig. 314. X 50.

Just as the anlagen of the pouches appear in a cranio-caudal series, sc. too, their enlargement takes place in succession in the same direction. Associated with this there is a dorsoventral flattening and a rather considerable lateral broadening of the pharyngeal lumen (Figs. 317 and 319), in such a way that the maximum broadening occurs opposite the first pouch, and caudal to the last pouch there is at first a diminution of the gut. The region in which the pouches occur practically represents the anlage of the pharynx (including the floor of the mouth). At the height of the development of the pouches the embryonic pharynx is strongly flattened dorsoventrally and convex dorsally and has a somewhat triangular outline, the base being directed orally aud the apex at the point of union of the air- and food-passages.

452

HUMAX EMBRYOLOGY.

Altogether five pharyngeal pouches are formed in the human embryo (Hammar, 1904; in 1889 His speaks of a rudimentary fifth pouch), and all of these reach the ectoderm.

According to Tourneux and Soulie (1907), a sixth pouch also occurs; this discovery has not, however, been thoroughly described. Tandler (1909) has described a diverticulum caudal to .the fifth pouch (embryo shown in Fig. 320; indicated by the letters Div.), but he has left it open whether or not it is to be regarded as a rudimentary sixth pouch. Opposed to such an identification is the Hyp.

l.St.

1

Ep.'IV

ub. K.

Div.

V.c. Thyr. D. br. IV

Fig. 320. — Pharyngeal region of the embryo BR (Normentafel No. 42, 9.75 mm. vertex-breech measurement), from the ventral surface. (From a model in the First Anatomical Institute, Vienna.) D. br. II and IV, ductus branchialis from the second and fourth branchial grooves; D. c, ductus cervicalis; Div., diverticulum (see text); Ect., ectoderm; /., 2. St., first and second pharyngeal pouch; Thym. Ill, thymus anlage of the third pouch; Thyr., thyreoid; ub. K., ultimobranchial body; V. c, vesicula cervicalis. The approximate limit between ectoderm and entoderm is shown by a broken line. X 40.

fact that the ultimo-branchial bodies are associated with the fifth pouch, whereas in general they belong to the last formed pouch (the sixth in birds, according to H. Rabl, 1907). The diverticulum seen in Fig. 320 appears to be identical with that figured by Soulie and Bardier (1907) for an embryo of 6 mm. and designated as the fifth pouch, but it is certainly not identical with the fifth pouch of other authors. 1 * 19 Such diverticula are probably without significance and very transitory. H. Rabl (1907) has shown, without description, something similar in the duck in Fig. 6 of his paper. See also p. 454.

PHARYNX AND ORGANS OF RESPIRATION. 453 The available data regarding the appearance of the pouches are as follows: The first pouch appears shortly after or simultaneously with the separation of the anterior part of the digestive tract from the yolk-sac (Fig. 314, p. 447) ; it reaches the ectoderm at a stage in which there are ten primitive segments. In the embryo from which Fig. 315 was constructed the mesoderm has been pressed laterally by the pouch, but the ectoderm and endoderm have not fused to form a closing membrane. The second pouch, which is indicated quite early (Figs. 314 and 315), has formed its closing membrane at a stage with thirteen or fourteen primitive segments (compare also Fig. 316). At this time the third pouch is formed; it has reached the ectoderm in an embryo with twenty-three primitive segments (Figs. 317 and 318). Such an embryo also shows the anlage of the fourth pouch; this shows a certain amount of variability in its development, but it reaches the ectoderm in embryos of about 4 mm. in greatest length (thirty-five primitive segments). The rudimentary fifth pouch is perhaps already formed in the stage shown hi Figs. 317 and 318 (p. 461) ; a closing membrane for the pouch is twice shown in the Normentafel in embryos of 5 mm. The pouch appears to be an appendage of the fourth (for details see later on). — In correspondence with the number of pouches, six branchial arches are to be recognized in the human embryo, of which, however, only four are visible from the surface. The fifth, in correspondence with the incomplete development of the fifth pouch, is at first very indistinct; its aortic arch, the fifth, is rudimentary and its nerve can only be distinguished transitorily (Tandler, Elze, Grosser). Its skeletal portion, which is included in the thyreoid cartilage, is, however, later of considerable size. The sixth arch is not bounded caudally by a pouch (see p. 446) ; its participation in the formation of the larynx is doubtful (see later).

Almost contemporaneously with the formation of the first pharyngeal pouch or only a little later there appears the anlage of the thyreoid gland, usually termed the anlage of the median thyreoid; the word median, however, now seems to be superfluous, since it probably represents the only anlage of the thyreoid tissue (see later, p. 468). The anlage is recognizable before the first pharyngeal pouch has come into contact with the ectoderm, as a prominence in the ventral wall of the pharynx (Figs. 314 and 315) ; it appears, therefore, much earlier than is shown in the Normentafel. It then becomes constricted to form a stalked vesicle (Figs. 317 and 318), and its stalk, whose lumen becomes obliterated, persists for some time as an epithelial cord. The thyreoid anlage belongs primarily to the medial region between the first two ventral pharyngeal grooves, that is to say, to the oral portion of what is later the area mesobranchialis. It is at first anterior to, not in, the region of the second branchial arch. The hollow stalk of the vesicle is the thyreoglossal duct (His).

Even before the obliteration of this duct the first ventral pharyngeal groove becomes prolonged medially and divides into two limbs, which unite in the median line with the corresponding ones of the other side and so enclose a median elevation, the tuberculum impar (Fig. 317). The opening of the thyreoglossal duct is situated at first upon the summit of the tubercle, but later it becomes shifted into the posterior boundary furrow or, according

454 HUMAN EMBRYOLOGY.

to Ingalls (1907), in an embryo of 4.9 mm., into "the region of the second arch, immediately aboral to the tuberculum impar." Since the tuberculum belongs to the medial region (His, 1885), it is evidently not a derivative of a branchial arch. 20 When the arches are formed they are connected behind the tuberculum impar by a transverse elevation, a copula. This and the tubercle are for the most part taken into the anlage of the tongue, whose formation has been described in the portion of this chapter dealing with the development of the mouth. Behind the copula the mesobranchial area remains for some time but little altered; only a median groove, which replaces the heart swelling, becomes more marked (Fig. 319), and later the area is employed in the formation of the larynx (p. 476).

In addition to the embryos figured above, which have been kindly contributed by their owners, there are certain others that have been described which have important bearings on the early stages of the pharynx; these are the embryo Dandy (1910) with seven pairs of primitive segments, the embryo Pfannenstiel III (Normentafel, No. 6), which has been described by Low (1908), with thirteen or fourteen segments, and the embryo XII of Mall's collection, which has been described by Sudler (1901-02) and has fourteen pairs of segments. The embryos Klb and R. Meyer 300 have already been reconstructed by Kroemer (1903) and by Thompson (1907), but on a small scale and with very cursory descriptions. The embryo Dandy corresponds almost exactly with the embryo Klb, and the Pfannenstiel III and Mall XII embryos are almost exactly equivalent to the embryo HaL. In this the closing membrane of the first branchial cleft of the left side is not yet complete, but is divided into two parts by a strip in which there is no mesoderm, but where the two epithelia have not yet fused (Fig. 316). A similar condition occurs in the embryo Pfannenstiel III. Low has assigned the two portions of the closing membrane to two successive pharyngeal pouches, which is clearly an error and has led him to describe the first pouch as lying originally dorsal to the second, when, as the later development shows, he was describing merely the two angles of the first pouch. — Occasionally small irregular evaginations occur in connection with the pharyngeal pouches and the branchial grooves, as in the embryo shown in Fig. 319, on the dorsal side; they have also been observed and figured by Ingalls. Such are perhaps comparable to the embryonic intestinal diverticula described by F. T. Lewis and Thyng (Amer. Journ. Anat., vol. 7, 1907-8).

A remarkable observation has been made by the author in all young embryos with the first pharyngeal pouches well developed; these are the embryos R. Meyer 335, HaL, Pfannenstiel III (loaned for this purpose), R. Meyer 330, and also a somewhat pathological, young embryo from the collection of R. Meyer. In the region of the first pouch there projects ventrally (Figs. 315 and 316) or caudally (Fig. 318) from the closing membrane into the pharyngeal lumen an irregularly knobbed process filled with mesoderm. That it is an accidental structure or due to post-mortem changes seems to be excluded by the regularity of its occurrence (Low has figured, but not described it). It disappears quite early (in the oldest 20 The account given above differs in many respects from that recently given by Kallius (Anat. Hefte, vol. 41, 1910) for the pig. Observations on a larger amount of human material than is at present available may show a necessity for some modifications of the statements made. Compare especially the development of the larvrix described below.

PHARYNX AND ORGANS OF RESPIRATION. 455 embryo examined, Fig. 318, it is present only on the left side and is greatly reduced in size; in embryos of 4.25, 5.0, and 5.8 mm. and in those still older, it is wanting), and may perhaps be interpreted as a rudimentary internal gill. It would not be the first instance of a very ancient rudiment well developed in the human embryo. Similar structures have not yet been observed in other amniote embryos.

The thyreoid anlage in human embryos is at first exceptionally large, but seems to be subject to a certain amount of variation in form. Dandy describes a ventro-median pouch projecting from the union of the two first pharyngeal pouches — evidently the thyreoid anlage (compare Fig. 314), although he denies the occurrence of such a structure. In Fig. 316 it has a similar appearance, in Fig. 315 and also in the embryos Pfannenstiel III and Mall XII it has a much more distinct delimitation. The delimitation starts on the rostral side and appears later caudally; the separation from the obliterated thyreoglossal duct takes place, according to the Normentafel, in embryos of about 6 mm., occasionally, however, earlier or later. At this time the lumen of the anlage, which has usually become bilobed, has disappeared. For an account of the differentiation of the anlage see p. 468.

The closing membrane in the human embryo, as in those of mammals generally, remains imperforated (His) ; open branchial clefts do not occur. Perforation has, however, been frequently observed, most frequently in the case of the second pouch (Kolliker, Tettenhamer, and Hammar), which possesses the longest closing membrane (Hammar). 21 In this case perforation may be regarded as within the limits of variation, but in other pouches it is, as a rule, due to injury from handling, and such injuries assuredly also increase the percentage of cases of perforation of the second pouch.

After the closing membrane has become converted into an elongated strip (p. 448) the mesoderm again penetrates between the two epithelia of the membrane and the pouch once more becomes separated from the surface of the embryo. This process may be followed in its simplest form in the first pouch; in the more posterior ones it is combined with the formation and constriction off of the sinus cervicalis (C. Rabl, 1886 to 1887; sinus prcecervicalis, His, 1885; compare vol. I, p. 69, et seq.) 22 This is formed by the mandibular and hyoid arches growing more rapidly than the other arches in all dimensions, but especially the transverse, while the growth of the branchial arches proper lags behind that of their surroundings, so that they come to lie in the floor of a depression, the sinus cervicalis, which is open laterally. The • 11 The closing membrane of the first pouch is, however, in early stages by no means so short as Born and Hammar have imagined (compare Fig. 318). But the degeneration of the ventral part of the membrane takes place very early in this pouch.

22 Rabl has altered the name proposed by His, on the ground that the term " praecervicalis " implies a structure situated anterior to the neck region ; His intended it to denote the ventral position of the sinus. Both terms are employed synonymously in the literature.

456 HUMAN EMBRYOLOGY.

caudal edge of the hyoid arch later grows backward over the month of the sinus, forming an indistinctly delimited operculum, which is less developed in man (Hammar) than in other mammals ; the sinus retains its connection with the exterior for a short time by means of the ductus cervicalis {prcecervicalis) , but finally becomes completely shut off so as to form the vesicula cervicalis {prcecervicalis) , 23 In man there persists at the surface only a shallow groove, the sulcus cervicalis {sulcus prcecervicalis, Hammar; cervical groove, H. Eabl), which at first marks the boundary line between the head and the thorax. The vesicula cervicalis lies lateral to the third pharyngeal pouch (Figs. 320, 321, 325, and 326), and is connected by diverticula, the former external branchial grooves, with the second and fourth (and for a short time also with the fifth) pouches; these diverticula become drawn out into long canals, the ductus branchiales (II and IV) (Figs 320 and 321). The vesicula and the ductus exist only for a short time ; their lumina vanish and the epithelial structure of the organs disappears.

With the sinus cervicalis are associated three cranial nerve placodes (branchial cleft organs of Froriep ) ; they consist of intimate connections of the epithelium with the ganglia of the glossopharyngeus and vagus (see the chapter on the Nervous System). These placodes are also recognizable only for a short time and disappear with the other derivatives of the sinus.

From the time of His (1885) up to the present a number of authors have agreed in deriving the anlage of the thymus, either in whole or in part, from the epithelium of the sinus cervicalis. These results are hardly reconcilable, however, at least so far as man is concerned, with those of other authors (compare Hammar, 1910). — The fate of the sinus vesicle is, moreover, different in different mammals. While it disappears in man and in the cat at an early period and in the rabbit somewhat later, in the pig (Kastschenko 1887, Fox 1908) and in the sheep (Prenant) it persists for some time as a structure of considerable size, which Kastschenko has termed the thymus superficialis, but whose eventual fate is not yet certainly known. In the mole, according to the recent definite results of H. Rabl (1909), the thymus superficialis is actually formed from the vesicula cervicalis. — The third pharyngeal pouch seems also to vary in different species as to its lateral extension ventral to the sinus vesicle. — The duct-like remains of the second external branchial groove, Hammar (1903 and 1904), following C. Rabl (1886-7), has termed the ductus branchialis^ while he names the corresponding 23 According to H. Rabl (1909) the term vesicula cervicalis is to be applied to the entire complex, including the two ductus branchiales (see above) ; Hammar uses the term vesicula prcecervicalis only for the vesicular portion that is associated with the third pharyngeal pouch, this portion being approximately identical with the fundus prcecervicalis (cervicalis) of His and H. Rabl, as well as with the vesicula thymica of Kastschenko and the sinus vesicle of Zuekerkandl.

24 Fox (1908) has not been able to find this branchial duct in the pig, but, on the other hand, demonstrates the occurrence of a long pouch-like diverticulum of the second pouch. Differences apparently occur in different species in this respect also. The diverticulum in the pig may correspond with the thymus anlage of the second pouch described by Piersol (1888) and others in the rabbit; this structure has not yet been observed in the human embryo.

PHARYNX AND ORGANS OF RESPIRATION. 457 duct of the fourth pouch the dtictus thyreo-cervicalis; since, however, this latter structure, in my opinion, has nothing to do with the actual thyreoid anlage, the term ductus branchialis has been used above for both duets, the numbers II and IV indicating the branchial grooves with which each corresponds.

The time of obliteration of the lumina of the ductus cervicalis and the ductus branchiales is subject to some variation (compare Figs. 320 and 321) ; according tc the Normentafel it occurs in embryos of about 9 mm., and in those of 11-14 mm. the epithelial cord formed from the ductus cervicalis has disappeared, as has also the vesicula cervicalis a little later. — The first pouch has separated from the ectoderm in an embryo of 14 mm. (Normentafel, No. 54).

II. The Differentiation of the Pharyngeal Pouches; the Second Pharyngeal Pouch and the Tonsils.

Only for a short time do all the pharyngeal pouches have a relatively similar structure, differing essentially from one another only in that they diminish in size caudally (Fig. 319) ; they separate first into two groups, one of which consists of the first two pouches and the other of the remaining ones (Figs. 320 and 321). At the level of the first two pouches the pharynx rapidly increases in width, an increase that stands in relation to the development of the first arches as already described (p. 455). The succeeding pouches remain much less extensive, and from their epithelium a number of glandular organs develop.

By the increase in breadth of the pharynx the first and second pouches acquire a common pharyngeal opening (Fig. 320). The broadening occurs even before the separation of the pouches from the ectoderm; it practically corresponds to the primary tympanic cavity of Kastschenko or the pliaryngo-tympanic cleft (lateral pharyngeal enlargement) of Piersol. Nevertheless, as is shown by the thorough study of the later development by Hammar (1902), the entire complex is not concerned in the formation of the anlage of the middle ear. Only the first pouch becomes transformed into the primitive tympanic cavity ; its further development is described in another chapter. The second pouch gradually ceases to be an independent outpouching of the pharynx, its walls being taken up into the walls of that cavity, its dorsal angle only persisting as a slight evagination, which becomes pushed forward toward the point of connection of the first pouch with the pharynx, that is to say, toward the root of the primary tympanic cavity (Hammar) (Fig. 322). Between the derivatives of the first two pouches there now become interposed the palatine ridges, and they are thus definitely separated (Fig. 323), so that the fate of each may be readily determined. The dorsal angle of the second pouch becomes transformed into the palatine tonsil, and may, accordingly, at an early period, be termed the simis tonsillaris (Hammar).

458 The development of the tonsil, which has been thoroughly studied by Hammar (1903), is associated with the appearance of a small elevation, the tuberculum tonsillare, which is situated on the ventral wall of the pharynx and projects into the sinus tonsillaris, lying practically opposite it (Fig. 323). Both structures lie on the lateral edge of the pharynx, and their derivatives are therefore to be found later side by side on the lateral wall of the pharynx. After the appearance of the tuberculum the palatine arches become evident, and the arcus palatoglossi cannot, therefore, be derived directly from the hyoid arches, as His (1885) thought, even although they lie oral to the tonsils. The tuberculum quickly flattens to form a fold, which surrounds the sinus tonsillaris

l.St.

D. br. II Sulc. c.

ub. K.

Fig. 321 — Branchial derivatives of an embryo of 11.7 mm. (nape-length), somewhat simplified. (After Hammar, 1903.) Sulc. c, sulcus cervicalis (pracervicalis) . The remaining lettering as in Fig. 320. X 21.

anteriorly and inferiorly and corresponds to the plica triangularis of the B.N.A. (His). The sinus itself becomes for a time divided by a plica intratonsillaris into two superposed recesses, and from the wall of the sinus tonsillaris epithelial processes grow out into the connective tissue of the mucous membrane; these processes are at first solid, but later become hollow by the degeneration of the central cells. Portions of the processes may be separated off, but these undergo degeneration. Around these epithelial processes there is formed, accompanied by abundant cell division, a lymphoid tissue, from which leucocytes penetrate into the epithelium. The plica triangularis, situated in front of the tonsil, is originally high, but it undergoes a progressive reduction which is continued even after birth and frequently results in the complete disappearance of the fold. The plica retrotonsillaris, which occasionally occurs in adults, belongs, according to Hammar, to relatively later devel

PHARYNX AND ORGANS OF RESPIRATION.

451)

opmental stages (fetuses of 190 mm. and upward). The fossa supratonsillaris (His) is formed from the upper recess of the sinus tonsillaris, with the assistance of the folds which surround the fossa.

Hammar has not confirmed the view that the reeessus lateralis pharyngis (Roseninulleri) is derived from the second pharyngeal pouch (His, C. Rabl, Kastschenko). The recess appears relatively late, but definite observations upon it are wanting. — The reduction of the ventral part of the second pouch begins in embryos with a length of somewhat over 8 mm. (Hammar; compare also Figs. 319 and 320), and in an embryo of 17 mm. only the dorsal angle of the pouch, lying near the entrance into the primary tympanic cavity, is to be found. The same embryo also has a recognizable tuberculum tonsillare. In one of 24.4 mm. the plica intratonsillaris can be seen, in one of 70 mm. the budding out of the epithelial processes

Tu.-ty. R.

Pkh.—

Hyp.

Mh.

PI. itt.

Dors. II Fi ;. 322.

Tr. Oe. Fig. 323.

Fig. 322. — Dorsal view of the left half of the pharynx of an embryo of 21 mm. (nape-length). (After Hammar, 1902.) Dors. II, dorsal diverticulum of the second pharyngeal pouch; Hyp., stalk of the hypophysis (cut); Pkh., tympanic cavity. X 21.

Fig. 323. — Pharynx of an embryo of 24.4 mm. (nape-length), from the left side, somewhat simplified. (After Hammar, 1903.) Gl., palatine ridge; Mh., mouth cavity; Oe., oesophagus; PI. itt., plica intratoneillaris; Tr., trachea; Tu.-ty. R., tubo-tympanic space; Tub. tons., tuberculum tonsillare. X 12.

is beginning, and in one of 110 mm. the accumulation of cells in the connective tissue. Lymphocytes are first recognizable in embiyos of 140 mm. and secondary nodules in one of 235 mm. By this time the tonsil has acquired its characteristic features. — The plicaB triangularis and intratonsillaris are rudimentary structures in man, but may play a part in the manifold modifications of the tonsils which occur in the mammalian series. — As is the case with other adenoid organs, so in that of the tonsils all observers are not agreed as to the origin of the leucocytes; yet their derivation from the epithelium (Retterer) has been, probably correctly, opposed by Stohr, Kollmann, and Hammar. Stohr's view that they migrate into the tonsillar tissue from the blood-vessels is replaced by Hammar by the assumption that they are autochthonous structures of the mesoderm. — Griinwald (1910) derives the tonsil from the ventral portion of the second pouch and regards it as equivalent to a thymus metamere. He finds in fetuses cartilaginous outgrowths from the second and third branchial arches included in the tonsillar anlage and serving it for support. The adenoid tissue is of mesodermal origin.

460 HUMAN EMBRYOLOGY.

III. The Third to the Fifth Pharyngeal Pouches; the Branchiogenic Organs.

The importance of the three caudal pharyngeal pouches in the amniota lies mainly in the fact that their epithelium gives rise to a series of ductless glands; these are the thymus, the epithelial bodies or parathyreoids, and the ultimobranchial body (the so-called lateral thyreoid). Of these the first two are derived from the third and fourth pouches; the question of the development of the last is intimately connected with that as to the occurrence of a fifth pouch.

The thymus and epithelial bodies are formed in the lower vertebrates (with numerous pouches) from a series of pouches in succession; they are, accordingly, metameric (branchiomeric) organs, forming one of the characteristic products of a pharyngeal pouch and (together with the ultimo-branchial bodies) are collectively known as pharyngeal pouch or branchial cleft organs or as branchiogenic organs. Yet it must be left an open question, at least for the thymus (compare Hammar, 1910), whether this does not represent an originally unsegmented epibranchial organ. Epithelial bodies occur first in the tetrapodous vertebrates. Furthermore, they occur on all the pouches in none of the vertebrates, the anlagen being usually suppressed on certain of the pouches, namely the oral ones; in man they have not yet been described in the first two. According to my own observations, a circumscribed epithelial thickening may occur transitorily (embryo Hah of the First Anatomical Institute, Vienna, with about 15 primitive segments) in the region of the dorsal wall of the first pharyngeal pouch, opposite the previously described invagination of the ventral wall (p. 454) ; the significance of this hitherto unobserved structure is, however, still in doubt. , A discussion of the historical development of our knowledge regarding the origin of these organs would lead us too far; one may consult on this subject the accounts given by Kohn (1900) and Hammar (1910). The actual explanation of the developmental processes has been principally due to the work of Kohn (1896) and Groschuff (1896). The very complicated nomenclature of the parts bears witness to our knowledge of the actual relationships having been acquired step by step; the embryonic anlagen were known in part much earlier (Remak, 1855) than the corresponding definitive organs, and they therefore received at first for the most part erroneous interpretations. For instance, the glomus caroticum has repeatedly been included in the series of branchiogenic organs and derived from the third pouch; it is, however, a derivative of the chromaffin system, a paraganglion (A. Kohn; see the chapter on that system), and the anlage seen by various authors was that of an epithelial* body: Even in 1908 Fox, on historical grounds, termed the epithelial body of the third pouch the glandula carotica, although it has no other relation to the carotid than a transitory topographical one. 26 Such temporary topographic relations have also brought it about that the epithelial bodies derived from the third or fourth pharyngeal pouches have been termed glandules thymiques (Prenant) and glandules thyreoidiennes (Gley), as is usually done at the present time by French authors. The terms are historically intelligible, 26 In some mammals an epithelial body is actually situated at the bifurcation of the carotid, as, for instance, in Echidna (Maurer), the sheep (Prenant) and Didelphys azara (Zuckerkandl) ; in the last the glomus caroticum is also recognizable.

PHARYNX AND ORGANS OF RESPIRATION.

461

but are unfortunate, since, apart from the possibility of confusion with the main glands and disregarding the histological and physiological similarity of all the epithelial bodies, the topographical relations are characteristic only for a definite stage of development, and the epithelial bodies in a long series of mammals, including man, are, on the one hand (when in normal position), all attached to the thyreoid and, on the other, are throughout genetically connected with thymus anlagen and also are frequently accompanied by small thymus lobes, so that both names might be applicable to each body. A similar criticism applies to the names parathymus for the derivative of the third pouch and parathyreoid for that of the fourth, employed by Groschuff (1896) ; the same author in 1900 names the bodies parathymus and distinguishes them according to the pouch from which they are formed as parathymus III and parathymus IV, names which have embryologically

Ep. Ill

Ep.IV

Tons.

D. ph.-br. Ill

D. ph.-br. comm.

Thym.

Fig. 324. — Schema of the branchiogenic derivatives in man, adapted from the schemata of Groschuff and Kohn. D. ph.-br. Ill, ductus pharyngobranchialis of the third pharyngeal pouch; D. ph.-br. comm., ductus pharyngobranchialis communis of the fourth and fifth pouches (of the caudal pharyngeal pouch complex) ; D. th.-gl., ductus thyreoglossus; Ep. Ill and IV, epithelial bodies of the third and fourth pouches; Thym. Ill and IV, thymus anlagen of the third and fourth pouches; Tons., tonsil; ub. K., ultimobranchial body.

a greater justification than the expression parathyreoid. Yet, at least for the adult condition in man, the name parathyreoid, proposed by the discoverer of the organs, Sandstrom, is quite characteristic, if one does not prefer the general name epithelial bodies, proposed by Kohn and borrowed from Maurer's description of the relations in amphibia. The very general term " glandules branchiales," used by Herrmann and Verdun (branchial glands, H. Rabl 1907), would also be strictly applicable to the other branchiogenic organs. The further classification of the epithelial bodies may probably be most satisfactorily based on the pouches from which they are formed, just as other metameric organs are grouped under a common designation and distinguished by numbers.

First of all, the development of the fifth pharyngeal pouch, which was long overlooked and whose existence was even denied,

462 HUMAN EMBRYOLOGY.

must receive some further consideration (see above, p. 453). A thorough exposition of the question has been given by Tandler (1909), who, however, had for study no material showing the pouch with an epithelial closing membrane (compare in this respect Harmnar in the Normentafel). The form of the pouch, as it makes its appearance, is extensively modified and complicated by the anlage of the ultimobranchial body. First of all (so far as our present knowledge extends) there appears on the fourth pouch, soon after its differentiation (embryo of about 3 mm., Hammar, 1902, Normentafel No. 11; compare also Ingalls, 1907, embryo of 4.9 mm., Normentafel No. 14), a process, directed ventrally and caudally, 26 which is longer than the ventral diverticulum of the third pouch of the same stage, but which, however, might readily be mistaken for such a diverticulum, as has apparently been done by earlier investigators and quite recently by Fox (1908). This process later becomes more distinctly separated from the fourth pouch, which then acquires a dorsal and a ventral diverticulum, the latter varying in extent (compare Tandler and Fig. 319). A lateral evagination of the caudoventral process reaches the ectoderm as a fifth pouch and forms a closing membrane ; this, however, is perhaps not always formed and is at all events very transitory. A dorsal diverticulum is apparently not formed, 27 and after the degeneration of the actual pharyngeal pouch, which is directed toward the ectoderm, the caudoventral process becomes converted into the ultimobranchial body. The fourth and fifth pouches are, accordingly, intimately associated genetically, and the ultimobranchial body is the sole derivative of the latter. Both pouches possess only a common communication with the pharynx, a ductus pharyngobranchialis communis (compare p. 451), and the intimate connection of these last two pouches may be expressed by uniting them in the term caudal pharyngeal pouch complex (Figs. 319, 320, and 324).

H. Rabl (1909) finds in the mole a common anlage for the two pouches and names it the caudal pharyngeal diverticulum. In this case the fourth poueh is more rudimentary than the fifth. — That the ultimobranchial body does not make its appearance behind the series of pharyngeal pouches (as a postbranchial body according to Maurer), but is derived throughout the whole vertebrate series from what is in each case the last pouch, which has become rudimentary, was insisted upon by Geil, who is responsible for the term here employed for the structure. The body is identical with the suprabranchial body of van Bemmelen. The expression telobranchial body, which has been frequently employed recently, has been rejected by H. Rabl, since it does not express the relation of the structure to the last pharyngeal poueh.

M The structure in Figs. 317, 318, and 331 marked as a doubtful fifth pouch is perhaps merely an analogue of the diverticulum described on p. 452.

27 Compare, however, p. 471 (Getzowa) as regards the occurrence of a corresponding epithelial body.

PHARYNX AND ORGANS OF RESPIRATION.

463

In the third branchial pouch the formation of the thymus is preceded by an elongation of the ventral diverticulum, which extends ventrally and medially (Figs. 319 and 320), and whose epithelium, consisting of closely packed cells, increases in height on the aboral wall of the diverticulum (Fig. 326). This thickening of the epithelium extends also upon the aboral and dorsal

Ep. Ill

Thym.III Ch.

S.c. D.ph.-br.III \ I I

•*?

S.c.

y

D.c.

"I y"

Fig. 320. Thyr. Thym. Ill

JT^Hy.

Figs. 325 and 326. — Two sections through the embryo BR (compare Fig. 320). Fig. 325 is through the point of communication of the third pouch with the pharynx, Fig. 326 through the sinus and ductus cervicalis. Ao.-B., aortic arch; Ao. d. and Ao. v., aorta dorsalis and ventralis; Aw., arytenoid swelling; Ch., chorda dorsalis; D. br. II, IV, ductus branchialis of the second and fourth external branchial grooves; D. c, ductus cervicalis; D. ph.-br. Ill, ductus pharyngobranchialis of the third pharyngeal pouch; D. ph.-br. c, ductus pharyngobranchialis communis; Ep. Ill, IV, epithelial bodies of the third and fourth pharyngeal pouches; Hy., hyoid arch (operculum); Lar., anlage of the larynx; Per., pericardial cavity; S. c, sinus cervicalis; Sy., sympathicus; Thym. Ill, thymus anlage of the third pharyngeal pouch; Thyr., thyreoid; Tr., trachea; ub. K., ultimobranchial body; V.j., vena jugularis; X., vagus; XII. hypoglossus. X 40.

464 walls of the pouch itself (Figs. 324 and 325), and simultaneously there begins on the oral and lateral walls of the dorsal diverticulum and of the pouch itself (Figs. 324 and 325) a proliferation of the epithelium, which very early shows itself, by its histological differentiation, to be the anlage of an epithelial body. The cells appear to be vacuolated, their plasma reticular and refractory to stains (chromatophobe). The cell boundaries are at first indistinct (compare also Maximow, 1909, p. 538).

Ep.IV Sy. D.ph.-br.c.

Ao. d.

V.

D. br. IV

/

XII '

4Sr ^D.br.IV

Lar. Thyr. F ; s. 327.

\/

Per.

ub.K.

Ao. v. Tr.

Fig. 328.

Figs. 327 and 328. — Two sections through the embryo BR (compare Fig. 320). Lettering as in Figs.

325 and 326. X 40.

At the same time the medial portion of the pouch narrows to become the ductus pharyngobranchialis HI (connecting piece or duct, ductus thymopharyngeus of Hammar), while the lateral portion becomes the secondary pouch of H. Eabl or, if the anlagen of the thymus and epithelial body be disregarded, the remains of the pharyngeal pouch. The formation of a large vesicle from the secondary pouch does not occur in the human embryo. The ductus pharyngobranchialis soon atrophies completely, so that the derivatives of the pouch become free. The thymus anlage becomes first

PHARYNX AND ORGANS OF RESPIRATION.

465

of all a thick-walled cylinder, at whose cranial end the cavity of the remains of the pharyngeal pouch is visible, while the sinus vesicle, which was in relation to the pouch laterodorsally, vanishes. Soon, however, the lumen of the thymus anlage disappears 28 and the thymus cord is formed. This thickens at its caudal end and so forms the body of the thymus (thoracic portion), while the uppermost portions become gradually thinner (cornu of the thymus, GroschufT; cervical portion) and come into connection with the epithelial body (Fig. 329). The entire anlage migrates caudally, the body more rapidly than the cranial end and the epithelial body, so that the cervical portion becomes more and more drawn out and finally vanishes. The migration usually takes place in front of (ventral to) the vena anonyma sinistra, frequently, however, behind it (Tourneux and Verdun). The epithelial

Thym.-Cy. V. scl.

Thum.

Fig. 329. — The branchiogenic organs of an embryo of 26 mm., somewhat simplified. (After Verdun, 1898.) Lob. pyr., lobus pyramidalis; V. scl., vena subclavia. The remaining lettering as in Figs. 325 and 330. X 20.

body, which as a rule becomes quite separated from the thymus cord, normally halts in its caudal regression at the lower pole of the thyreoid. If the cranial end of the thymus cord does not disappear completely, there occurs beside the epithelial body an accessory thymus lobe (A. Kohn). Frequently in older embryos there is associated with the epithelial body a cyst, probably derived from the remains of the pharyngeal pouch (Kiirsteiner, 1899).

The thymus is from the beginning a paired organ and it remains so permanently ; neither fusion of the two anlagen by transverse connections nor a complete division into several lobes occurs, according to Hammar (1910). — The thymus is first represented in the Normentafel as a short cylinder in an embryo of 5 mm. ; the epithelial body, according to Tourneux and Verdun (1S97) and Tandler (1909), appears as an epithelial thickening in embryos of S mm., and Tandler finds cell differentiation in the einbryo represented in Fig. 325, Hammar, however, finding it earlier in an embryo of 8.3 mm. Nevertheless it is in some instances stated expressly in the Normentafel that the epithelial body is wanting in older embryos, 28 The thymus canal, described by older authors as occurring in later stages, does not exist.

Vol. II.— 30

466 HUMAN EMBRYOLOGY.

up to 11 nun. According to the Nomientafel, the ductus pharyngobranchialis has disappeared in an embryo of 14.0 mm., but Hamrnar (1904) finds that this happens sometimes earlier and sometimes later (in embryos between 11 and 19 mm.). In an embryo of 12.5 mm. in the Normentafel the thymus has reached the pericardial cavity, and in one of 15 mm. the lumen is limited to the uppermost portion. The remains of the pharyngeal pouch are still to be seen, according to Hamrnar (1904), in an embryo of 24.4 mm.

According to Hamrnar (1911), there is a definite thickening of the wall of the thymus diverticulum while it is still cylindrical, situated dorsally in the more proximal portion and laterally in the more distal portion (compare Figs. 325 and 326) ; the thickening does not extend, however, to the tip of the diverticulum and accordingly cannot be regarded as the sole anlage of the thymus. The ends of the anlagen remain at first in close topographic relationship to the aortic arches, and their elongation corresponds to the increase in the distance between the third and fourth aortic arches and to the formation of the anterior surface of the neck. The elongation depends mainly on the stretching and not on the growth of the anlagen. During the elongation the anlagen, which have at first a more transverse position, come to occupy a position more nearly parallel to that of the body axis (as may be seen from Figs. 320 and 329). By the resulting rotation the dorsolateral anlagen of the epithelial bodies are brought to the ventral side of the remains of the pharyngeal pouches. At the junction of the cervical and thoracic portions the aperture bend of the thymus anlage is formed, and, in addition, a heart bend and an aortic bend may also be distinguished.

The epithelial character of the thymus anlage continues plainly evident for a considerable time. In fetuses of about 50 mm. vertex-breech length one finds at the cranial end of the thymus anlagen and also in the cornna numerous closed vesicles with a high cubical epithelium (compare also Kursteiner, 1899) and transitions from these to epithelial cords without a lumen. Cells arranged in an epithelium-like manner are generally distributed over the surfaces of the lobes of the body of the thymus. In fetuses of about the same stage of development (from 42 mm. onward), the centre of the anlage is beginning to appear clearer in sections — the differentiation of cortex and medulla is beginning. From now on the organ gradually assumes its characteristic appearance of being formed of small ("lymphoid") cells. Hassall's corpuscles make their appearance in it in fetuses of from 60 to 70 mm. (Hamrnar, 1910). — The thymus grows not only throughout the entire period of embryonic life, but also through childhood, until puberty, and only at that time does its involution begin, a process which goes on but slowly in individuals with perfect health ; during illnesses, however, an accidental involution may supervene and this explains the widely divergent statements of various authors concerning the condition of the postfetal thymus (Hamrnar, 1910).

According to Hamrnar (1911), the beginning of the "leucocyte infiltration" may be seen in fetuses of from 30 to 40 mm., and the formation of the medulla in those of 50 mm. The medulla appears at first in the central portions in the form of a longitudinal cord, extending later into the superficial boss-like anlagen of the follicles, which have appeared in the meantime ; the formation of the

PHARYNX AND ORGAN'S OF RESPIRATION. 467 medulla is, accordingly, a continuous process, and for the most part it continues to be so later. In this manner the tract us centralis is formed, in whose composition, however, the cortical substance takes part later on. Since the boundary between the cervical and medullary portions is subject to functional modifications, the tractus centralis should not be termed the medullary cord, but preferably the parenchyme cord.

The histogenesis of the thymus is at present a very contentious question (compare Maximow, 1909; Hamiuar, 1910; and Stohr, 1910), and cannot therefore be described in detail. Authors are agreed only as to the epithelial origin of the cellular reticulum and of the Hassall corpuscles. The chief elements of the organ, however, the small thymus cells, have been regarded either as immigrated cells and therefore as thymus lymphocytes (more recently especially by Maximow and Hammar) or as modified epithelial elements (among recent authors chiefly by Stohr). According to Sehaffer (1909), they must be regarded as lymphocytes, even if their epithelial origin be admitted. From the histological conditions and from their development in mammals, conclusive evidence is not at present to be drawn (compare Stohr, 1910), but comparison and theoretical considerations undoubtedly speak in favor of their epithelial nature.

Little can be said concerning the histogenesis of the epithelial bodies. Even in the stage when they are epithelial thickenings their cells, as has been already mentioned, are characteristically differentiated ; then the anlagen become split up by the epithelium growing out in the form of cords and probably also by the penetration of connective tissue between the cells, and the cell boundaries become distinct. The formation of different kinds of cells (compare the summary by Getzowa, 1907) occurs ouly in postfetal life. Lumina are not distinguishable in the epithelial bodies during embryonic life, but they may appear later. They may then become filled with a secretion, which shows the staining reactions of colloid, and such structures have hitherto continually led authors astray by causing the epithelial bodies to be confused with thyreoid tissue and to be regarded as young stages of it. Yet the acidophilous nature of a secretion is no indication of its colloid uature (Kohn, 1896) ; according the Erdheim (1904), any albuminous secretion, contained within a closed cavity, may assume the appearance of colloid.

According to H. Rabl (1000). (he whole of the secondary pouch, except so much of it as is employed for the formation of the thymus anlage, is transformed into the epithelial body, even the wall opposite the original epithelial proliferation, since the connective tissue also penetrates the lumen of the pouch; at least this is the case in the mole. The other authors, who limit the extent of the anlage of the epithelial body to a greater degree than is done in the text, namely to the dorsal diverticulum or to the dorsal angle of the pouch, believe that the wall of the pouch itself degenerates; precise observations are. however, wanting.

468 HUMAN EMBRYOLOGY.

The development of the fourth pharyngeal pouch follows essentially the same lines as that of the third ; the ventral diverticulum is, however, much more feebly developed (Tandler, 1909), and it only occasionally undergoes a development into thymus tissue ; when this tissue is formed, it remains in the neighborhood of the epithelial body, which is formed from the dorsal and lateral portion of the pouch, and, accordingly, again is not limited to the dorsal diverticulum. The development of the epithelial body resembles that of the third pouch; no difference in the structure of the two bodies is discernible. The anlage of the ultimobranchial body is a derivative of the fifth 'pouch, which projects caudoventrally from the fourth pouch in the form of a thick-walled cylinder (Figs. 320, 324, and 328). The common communication of the two pouches with the pharynx, the ductus pharyngobranchialis communis, diminishes in size and becomes constricted off from the caudal pharyngeal pouch complex, just as does the corresponding part of the third pouch. The complex then separates from the pharynx, the epithelial body becomes independent by the disappearance of the thymus anlage and of the remains of the fourth pouch, and the ultimobranchial body becomes an elongated vesicle with thick walls. The two structures (epithelial body and ultimobranchial body) usually remain close together, however, and migrate somewhat ventrally and caudally, thus coming into relation with the thyreoid anlage.

The thymus ruetarnere of the fourth pouch was discovered by Groschuff; according to Erdheiru (1904), its persistence is to be regarded as a rarity. — The ductus pharyngobranchialis communis, which has also been termed the lateral thyreoid anlage, has been named, with reference to its relation to the ultimobranchial body, the ductus Oiyreopharyngeus. This name must be rejected, for the reasons given with reference to the ductus thyreocervicalis (see p. 457) ; in addition it may be mentioned that an occasionally persistent thymus anlage is also present in connection with this duct, as well as with the ductus pharyngobranchialis III, the. ductus thymopharyngeus of Hammar.

The thyreoid anlage (the middle or anterior thyreoid anlage of various authors), before its separation from the pharynx (p. 453), becomes bilobed with a divided lumen; at about the time when the thyreoglossal duct becomes broken it loses its lumen, and, undergoing a continuous displacement caudally, it develops into a broad structure composed of irregular cords of cells, disposed for the most part transversely. The derivatives of the caudal pharyngeal pouch complex apply themselves to the somewhat dorsally bent lateral portions of the anlage and become partly enclosed by it. This is the case with the ultimobranchial bodies, which then lose their lumina, but further than this they apparently do not always behave in the same manner. While in some cases they appear as compact bodies, in others they separate into an irregular group of small

PHARYNX AND ORGANS OF RESPIRATION.

469

cells with strongly staining nuclei (Grosser; see Fig. 330). In man, however, in normal development, no cell formations that can be referred to the ultimobranchial bodies are to be distinguished after a time (see also p. 471) ; up to the present no evidence has been advanced in favor of the widely accepted view that the bodies become converted into thyreoid tissue, and such a transformation is rendered highly improbable by the results of comparative investigation (see p. 471). The name lateral thyreoid anlage, which

Fig. 330. — Section through the laryngeal region of the embryo Nat2 of the First Anatomical Institute, Vienna (19.75 mm. vertex-breech length). Cr., cricoid; Ep. Ill, IV, epithelial bodies; Sy., sympathicus; Thym., thymus; Thym.-Cy., small cyst of the thymus; Thyr., thyreoid; ub. K., ultimobranchial body; Vag., vagus; W., vertebra. X 60.

has been applied to the ultimobranchial bodies, is therefore to be rejected (Verdun).

Toward the centre of the lateral lobes of the thyreoid there is to be found for some time (in fetuses of about 50 mm. vertexbreech length), as an expression of the more rapid growth of the lobes, a closer grouping of the thyreoid cords with a slighter development of the connective tissue; the differentiation of the cords takes place principally at the periphery. In this region there occurs in the stage mentioned the formation of lumina in the cell cords, which consequently appear beaded, and then the cords

470 HUMAN EMBRYOLOGY.

become divided up into individual groups of cells, the anlagen of follicles, which in part possess a lumen before becoming constricted off, although for the most part the lumina appear later and successively, even forming to some extent in the first years of childhood.

According to the Normentafel, the (middle) anlage of the thyreoid shows indications of a bilobed condition in embryos of about 5 mm. ; the occurrence of a lumen is variable. The thyreoglossal duet loses its lumen at a slightly earlier stage. It becomes drawn out to a long solid cord, which is broken in embryos between 6 and 7 mm.; it is occasionally distinguishable in later stages and remains of it may be found in embryos of 14 mm. In those of 8 mm. the anlage begins to separate into cords. — The ductus thyreoglossus, or its remains, occurs ventral to the hyoid bone and therefore between the derivatives of the first and second arches (His). The ultimobranchial body makes its appearance, according to Hammar, in embryos of 5 mm. as a cylindrical transformation product of the fifth pouch, or, it might be said, as an appendage of the fourth; the epithelial bodies IV, as well as the thymus anlagen, are defined in embryos of 8 mm. (Tandler, 1909), occasionally perhaps not until somewhat later (see p. 465). The caudal pharyngeal pouch complex separates from the pharynx in embryos of about 14 mm. (frequently only later, according to Hammar, 1904, in embryos over 18.5 mm. in length), and applies itself directly to the thyreoid. A little later, in embryos of somewhat over 15 mm., the lumen of the ultimobranchial body disappears. The small-celled proliferation of the ultimobranchial body, mentioned and figured above, can be perceived in two embryos in the collection of the First Anatomical Institute, Vienna (Nat. 1, with a length of 19.75 mm., and T. 1, with a length of 23 mm.) ; it seems also to have been observed by Tourneux and Verdun (1897) in an embryo of 19 mm. The denser grouping of the embryonic thyreoid cells in the lateral lobes of somewhat later stages (see p. 469), which these authors have also noticed, is not to be referred to the ultimobranchial bodies, as they have supposed. The (unpaired, middle) thyreoid anlage has been known since Rathke's time. That a derivative of the pharyngeal pouch region becomes associated with the thyreoid anlage in mammals was first observed by Wolfler (1880), and firmly established by Stieda (1881) and Born (1883) ; from the latter comes also the term lateral or posterior thyreoid anlage, which has been applied to the untimobranchial body, but which is rejected in the account given above.

In man the formation of the branchial derivatives is less completely and less easily followed than in many other mammals, as, for instance, the cat, but, on the other hand, more completely than in such a form as the rat. The differences which have been found in different species are partly to blame for the confusion which has long prevailed with regard to the development of these structures. Even in man the development of the individual organs has never yet been systematically followed throughout.

In correspondence with the extensive dislocation of the thymus, the epithelial body of the third pouch undergoes a much greater migration than that of the fourth, passing beyond it to come to rest at the lower border of the thyreoid. Consequently it appears as the inferior epithelial body, in contrast to the superior body of the fourth pouch situated at about the middle of the posterior (dorsal) surface of the thyreoid. This latter, in correspondence with the fusion that occurs between the ultimobranchial body and the thyreoid, in certain animals (rabbit, cat) regularly

PHARYNX AND ORGANS OF RESPIRATION. 471 and in man frequently becomes more or less enclosed within the thyreoid, and then appears as an internal epithelial body, with which the thymus IV may be associated as an internal thymus lobe, in contrast to the external one arising from the third pouch. Nevertheless all these conditions are very variable, and striking anomalies of position, as well as diminution and increase of number of the epithelial bodies, occur. Thus, that of the third pouch may remain even in man near its place of origin, not far from the division of the carotid (p. 460), or, on the other hand, it may descend into the thoracic cavity with the thymus. A diminution in the number of epithelial bodies is very difficult to demonstrate, on account of the possible occurrence of anomalies in position ; increase, probably by division of the anlagen, was first observed by Kursteiner and has since been repeatedly seen; Zuckerkandl has described a case in which there were eight, and Erdheim one with eight and one with twelve. The epithelial body III seems especially subject to division.

An internal epithelial body completely surrounded by the thyreoid is very rare in man, according to Getzowa (1907). According to the same authoress, cell cords of typical epithelial body tissue may occur in the interior of the thyreoid even when an external epithelial body IV is present. She is inclined to ascribe these cords to epithelial bodies of the fifth pouches, but embryological confirmation of this idea is as yet wanting. — The retrogression and atrophy of the cranial end of the thymus occasionally fails to take place, especially in connection with certain variations of the cervical nerves; a thymus lobe then occurs high up in the neck (Bien 1906 and 1907, Hammar 1910).

The ultimobranchial body in all vertebrates below the mammals is an independent structure which assumes a glandular character, produces alveoli and cell cords, but develops no colloid. In Echidna, according to Maurer, who has thoroughly studied the whole question, the body is also independent, but develops alveoli with colloid; nevertheless this material has been identified with that of the thyreoid only on the basis of its staining reactions. In all the higher mammals the body fuses with the middle thyreoid anlage, and its further history cannot then be followed with certainty. In many mammals a cyst can be found situated beside the internal epithelial body, surrounded by thyreoid tissue, and frequently finished with a ciliated epithelium and possessing mucous glands in its wall (central canal of the thyreoid of Prenant, vesicule postbranchiale of Herrmann and Verdun) ; furthermore there may be cell cords which extend into the interior of the thyreoid and vesicles, which are not always to be distinguished from undeveloped thyreoid tissue (Herrmann and Verdun 1899, Sehaffer 1909). Similar rudiments occasionally occur in older human fetuses, of 55 to 65 mm. vertex-breech length, according to Herrmann and Verdun (1899). The cysts and glands have been derived by most authors from the ultimobranchial bodies themselves, yet some of them at least may represent the remains of the fourth and even of the fifth pharyngeal pouch (Groschuff, 1896). If this be the case, then only the cell cords and the small vesicles can be ascribed to the ultimobranchial bodies, these, however, occurring distinctly only in a few species (dromedary, sheep, cow, hedgehog, mole), as well as cysts whose epithelium is in a state of proliferation and is producing the cell cords (Herrmann and Verdun 1900). These authors also describe a case of an ultimobranchial body remaining independent in a camel one year of age ; its structure was that of a gland, which was quite different from

472 HUMAN EMBRYOLOGY.

the thyreoid but rudimentary, and consisted of coiled tubules, closed vesicles, massive cell columns, and cell spheres.

In thyreo-aplasia, the defect of the (middle) thyreoid anlage, the ultimobranchial body gives rise to no thyreoid tissue (Maresch 1898, Peucker 1899, Erdheim 1904). In such cases one finds in addition to the epithelial body IV larger and smaller cysts, partly with contents which stain bright red with eosin; beside the cysts lie some lobes of serous or mucous gland tissue, and in one case Erdheim found immediately beside the cyst " a thin layer of small epithelial cells with dark nuclei." In any event these observations are opposed to the formation of thyreoid tissue from the ultimobranchial bodies; their significance is in harmony with the view stated above. The bodies are essentially rudiments, and one need not assume, as Erdheim has done, that only in thyreo-aplasia " the lateral thyreoid anlagen are also aplastic." In the atrophic thyreoids of cretins and idiots Getzowa (1907) observed cell masses and cords which likewise point to an occasional persistence of the ultimobranchial bodies in man. They correspond histologically with no other glandular tissue of the region, and are composed of large polyhedric cells rich in protoplasm and with large nuclei moderately rich in chromatin. In addition there were also small cysts which were not formed of thyreoid tissue. According to the same authoress, struma? may arise from the parathyreoids or from the ultimobranchial bodies as well as from the thyreoid, whence the form variability of these tumors.

In general the so variable behavior of the ultimobranchial bodies throughout the whole mammalian series may be explained on the supposition that in the mammalia they constitute functionless rudiments (Herrmann and Verdun). Groschuff (1896) rightly sees, in the union of the bodies with the thyreoid, a condition that is confined to the mammalia, a process which essentially corresponds to the formation of an internal epithelial body or thymus lobe, but does not justify the derivation of the thyreoid from different anlagen.

The thymus of the mammalia is not directly homologous with that of the lower vertebrates, since in the latter it owes its origin to dorsal and in the former to ventral diverticula of the pharyngeal pouches. A harmonizing of the relations seems to Maurer to be made possible by the conditions in Lacerta, in which a transitory slender ventral process of the third pharyngeal pouch is formed, which occupies the same position as the thymus anlage of the mammalia, but is not transformed into thymus tissue, but atrophies. Further the extension of the thymus anlage upor the dorsal wall of the pharyngeal pouch itself may be mentioned in this connection. — The sometimes occurring differentiation of the ventral diverticulum of the fourth pouch into thymus tissue has never been followed directly, but has been assumed on account of the occasional occurrence of thymus lobes on the epithelial body IV. — In the rabbit the second pouch also gives rise to a transitory ventral diverticulum, that is to say, to a thymus anlage (compare p. 456, footnote).

The persistence of portions of the branchial system of cavities may give rise to branchiogenic fistulae, cysts, and tumors (see especially Hammar, 1904). The region that corresponds to the outer opening of the ductus branchialis II and the ductus cervicalis is to be found at the anterior border of the m. sternocleidomastoideus. The second pharyngeal pouch corresponds to the tonsillar depression; the openings of the ductus pharyngobranchialis III and pharyngobranchialis communis are to be looked for near the larynx, about in the region of the sinus pyriformis, whence the n. laryngeus superior, as a branch of the fourth branchial arch, must pass between them. A fistula of the second cleft must lie, if the development of the vessels be normal, between the external and internal carotids and ventral to the glossopharyngeus and vagus; a fistula of the third cleft, between the common carotid and vagus, as well as between the glossopharyngeus and laryngeus superior; while a fistula of the fourth cleft must bend around the subclavian on the right and the arch of the aorta on the left, since these are derivatives of the

PHARYNX AND ORGANS OF RESPIRATION. 47°

•->

fourth aortic arch. The occurrence of these fistulse is, therefore, very unlikely; the fistula of the second cleft is the only one that has hitherto been recognized with perfect certainty. — As to the persistence of the ductus tbyreoglossus and the formation of median cervical fistuhe from it, the resume of Erdheim (1904) may be consulted.

B. THE DEVELOPMENT OF THE RESPIRATORY APPARATUS I. The Earliest Anlage.

The first anlage of the respiratory apparatus appears caudal to the pharyngeal pouches as a median ventral groove; the oral portion of this forms a ridge on the outer surface of the epithelial tube, but its caudal end is more rounded and hemispherical (Figs. 317, 318, and 331). In the region of the groove the epithelium is thickened. The ridge-like portion is the laryngotracheal groove and the rounded end the unpaired anlage of the lungs. These structures make their appearance very early, simultaneously with the last pharyngeal pouches and before the formation of the last two closing membranes, and show at first no connection whatever with the pharyngeal pouch region, except that the anterior end of the laryngotracheal groove extends just to the aboral part of the mesobranchial area. During the further development of the anlage the lungs grow much more rapidly than the remaining parts and form an unpaired, almost spherical vesicle (Figs. 332 and 333), which is in connection with the digestive tract dorsally and passes over into the tracheal groove orally.

The question as to whether the mammalian lung anlage is paired or unpaired has been answered in the latter sense almost unanimously by authors who have written since Kolliker's time, and may be regarded as definitely established for the human embryo by the observations of Blisnianskaja (1904) and Broman (1904) and the models figured here. 29 The view of most authors (compare Narath 1901 and Flint 1906), that the anlage is from the beginning asymmetrical, is not borne out by the models, since the asymmetry shown in Figs. 318 and 331 and limited to the laryngotracheal groove is produced by a torsion of the digestive tube at the boundary between the head and trunk, and is practically wanting in Fig. 332.

The unpaired character of the lung anlage marks the great difference that exists between the lungs and the gills. The unpaired anlage of the mammalia 80 is, 29 Thompson (1907) ascribes a paired anlage to the embryo from which Fig. 331 is taken, but does not figure it, and this statement has been transferred to the Normentafel. That he has, however, made a mistake as to the position of the lung anlage is apparent from his own description and from a figure published later (1908) ; he identifies it in 1908 as the stomach anlage, while in 1907 he transfers the lung anlage to the region of the diverticulum doubtfully identified in Fig. 331 as a fifth pharyngeal pouch. The embryo has, however, no stomach anlage; probably also his model was made on too small a scale. Fig. 145 of Broman (1904) agrees essentially with my Fig. 331.

80 A. Weber and his coworker Buvignier in several papers declare themselves in favor of a paired anlage for the mammals and for the homology of the lungs with the gills.

474 HUMAN EMBRYOLOGY.

it is true, probably a secondary condition, for in other lung-breathing animals, and among these in the lowest Tetrapoda, the amphibia, the anlage is paired (Remak, G-oette). Nevertheless even in these forms it is not to be homologized with a final (sixth or seventh) pharyngeal pouch, but is to be derived from a swim -bladder ; probably this structure was originally generally paired, and has, as a rule, become permanent only unilaterally (Greil, 1905).

Two longitudinal grooves (boundary grooves) on the lateral walls of the anterior part of the digestive tract mark out at an early period a ventral respiratory from a dorsal digestive zone (Kolliker, Flint; compare Fig. 333 and the transverse section of the region in Fig. 317, where the left groove is already indicated).

Fia. 331. — Anlage of the respiratory tract of an embryo of 23 primitive segments (Rob. Meyer No. 300, 2.5 mm.; compare Figs. 317 and 318). L., liver; Lar„ laryngeal groove; Tr., tracheal groove; Lu., lung anlage; 5. St. f, doubtful anlage of the fifth pharyngeal pouch. X 150.

The unpaired anlage of the lung has, however, only a short existence; from it there develop laterally and caudally the two pulmonary sacks. The boundary furrows of the respiratory anlage at the same time begin to be more sharply defined, and first the lung anlage and then the tracheal groove become separated from the oesophagus by a septum that grows forward from below. Orally, however, the laryngeal portion of the groove encroaches more upon the mesobranchial area (p. 449) until it lies between the medial ends of the fourth branchial arches and later between those of the third. The respiratory apparatus now consists of the cleft-shape entrance of the larynx lying between the caudal

PHARYNX AND ORGANS OF RESPIRATION. 475 pharyngeal pouch complexes, of the relatively long and slender laryngotracheal tube, 31 and of the two pulmonary sacks.

II. The Trachea.

No striking modifications occur later in the tracheal tube. Its lumen is at first cylindrical (Fig. 341), but later, with the development of the membranous dorsal wall, it becomes heart- or horseshoe-shaped in transverse section (Fig. 330), and in older embryos (more than 30 mm. in length) the dorsal wall is always thrown into longitudinal folds. The epithelium undergoes no marked changes, except that it develops cilia. The glands appear at the close of the fourth month, almost simultaneously with the elastic

Fig. 332. Fig. 333.

Fig. 332. — Lung anlage of an embryo of 4.25 mm. vertex-breech measurement, from the ventral side. Embryo R. Meyer No. 399 of the Zurich Anatomical Institute (Stage I of Blisnianskaja). X 150.

Fig. 333. — The same model seen from the left side. The outlines of the lumen shown by the broken line. X 150.

fibres of the mucous membrane, and very soon become hollow; the formation of glands appears to continue throughout pregnancy. The tracheal cartilages make their appearance in the places where they are finally found; their anlagen are to be recognized as condensations of the tissue in embryos of 17 mm., and cartilage appears in embryos of 20 mm. The differentiation is always more advanced in the neighborhood of the larynx (Philip, Kolliker, cited by Merkel, 1902). Musculature occurs in the dorsal wall of the trachea before the rings become cartilaginous. (For further details see Merkel, 1902.) 81 The occurrence of the cesophagotracheal septum and the anomalies that are occasionally associated with its formation have already been briefly discussed by F. T. Lewis in a preceding section of this chapter.

476 HUMAN EMBRYOLOGY.

No evidence is furnished by human embryos nor yet by those of the Plaeentalia in general in favor of a derivation of the tracheal skeleton from that of the larynx, that is to say from the cartilago lateralis of the amphibia. In Echidna Goppert (1901) has found a union of the prechondral rings by paired prsechondral cords and consequently their differentiation from a common anlage.

III. The Larynx.

The skeleton and musculature of the larynx have already been considered in the first volume of this Handbook ; the differentiation of the entrance of the larynx and of the laryngeal cavity remains to be considered here.

The oral end of the laryngotracheal groove shortly after its formation becomes embraced by two swellings, the arytenoid swellings (Kallius; crista terminalis, His). They indicate also the caudal limits of the branchial portion of the intestine, characterized by its lateral widening. The caudal pharyngeal pouch complex lies at first craniolaterally to the swellings, which are at first actually only the somewhat more pronounced borders of

-3. St. O. c. ph.-tr.- ^H ^ig i 'J^^ — i JH . / c/ Fig. 334. — Entrance to the larynx of an embryo of 8 mm. (After Soulie and Bardier, 1907.) Aw., arytenoid swelling; Ep., epiglottis; O. c. ph.-tr., orifice of the canalis pharyngotrachealis; PI. ar.-ep., plica ary-epiglottica; 8., 4- St., third and fourth pharyngeal pouches (their boundaries indicated schematically by dotted lines); T.ar., tuberculum aryteuoideum. X 30.

the oral end of the laryngotracheal groove (Kolliker, Soulie and Bardier). After the formation of the oesophago tracheal septum the swellings persist as the boundary of the laryngeal groove, and between them the groove deepens, its margins come into apposition, and its epithelium fuses, producing an obliteration of the cavity of the larynx (see below). At the time when the swellings lie almost parallel with one another, there may be distinguished, according to Soulie and Bardier (1907), at about their middle a thickening (arytenoid tubercle, bourrelet arytenoidien) and orally a narrower part, the later plica ary-epiglottica. By these the region of the laryngeal entrance becomes early delimited from that of the later interarytenoid notch. Anteriorly the swellings at first pass into the mesobranchial area and later they bend in this area in an arch-like manner to form a transverse swelling lying in front of the laryngotracheal groove. This, the furcula of His, is perhaps to be interpreted as the copula region of the branchial arches (see p. 454 and Kallius, 1910). This separates into the root of the tongue anteriorly and the anlage of the

PHARYNX AND ORGANS OF RESPIRATION.

477

epiglottis posteriorly (Haimnar, 1902). In the median line the boundary furrow between these two structures is less deep than it is laterally, and there is thus formed the anlage of the plica glosso-epiglottica media, which, however, becomes more sharply denned only after birth (Kallius, 1897; Soulie and Bardier). The epiglottis lies at first between the fourth branchial arches; to what extent the third arches are concerned in its formation is still in dispute. A temporary median furrow would appear to indicate that it has a paired anlage, but this disappears very soon (Soulie and Bardier).

Whether the epiglQttis is formed from a growth of the arches into the mesobranchial area (the view of the majority of authors; compare Soulie and Bardier), or as an elevation of this area itself, it being independent of the arches (His, F.c.

T. corn.

T. cun. i

T. corn.

Fig. 335.

Fig. 336.

Fig. 335. — Laryngeal entrance of an embryo of 28 to 29 days (8-9 mm.). (After Kallius, 1897.) Ep., epiglottis; F. c, foramen caecum; 8., 4- "St., third and fourth pharyngeal pouches; T . corn., tuberculum corniculatum. X 33.

Fig. 336. — Median section of the larynx shown in Fig. 335. (After Kallius, 1S97.) Can. ph.-tr., canalia pharyngotrachealis. The remaining lettering as in Fig. 335. X 33.

Hammar), is not yet definitely determined, and, furthermore, the first developmental processes of the entrance to the human larynx are not yet thoroughly known. — Kallius (1897), with His and others, derives the arytenoid swellings from the last (sixth) branchial arches (he names them the fifth, since the fifth pouch and the fifth aortic arch were at that time scarcely known). The manner of their formation contradicts this, however. The opinion of Kohlbrugge, cited by Kallius, that the ventriculus laryngis is a branchial pouch and therefore the caudal boundary of the arch mentioned, is untenable in view of the late appearance of the ventricle. At all events the material of the sixth arch, which is undoubtedly present (on account of its aortic arch), must pass continuously into that of the arytenoid swelling, since a separating sixth pharyngeal pouch is to be seen (see above, p. 446, 452).

Frazer (1910) comes to results which are in general quite similar, but, owing to the employment of a peculiar nomenclature, they are not easily compared with those of others. He also lays special stress on the identity of the arytenoid swellings with the last branchial arches, but later on allows also the ventral ends of the fourth arches to participate in the formation of the swellings. The epiglottis he derives principally from the third arch.

478 HUMAN EMBRYOLOGY.

After the formation of the arytenoid swellings and the epiglottis the evolution of the larynx can be followed more thoroughly. The arytenoid swellings gradually become folded in the middle almost to the extent of a right angle, so that the caudal portions become parallel while the oral ones diverge more and more (Figs. 334 and 335). While this process is going on, they move forward and their oral portions apply themselves to the aboral surface of the epiglottis, with which the folds are connected by the plicae ary-epiglotticse. The aditus laryngis has thus assumed the form of a T-shaped cleft (Figs. 334 to 338 and 325) ; the horizontal limb of the T lies between the arytenoid swellings and the epiglottis, the vertical one between the aboral portions of the two swellings (interarytenoid notch). At this time, however, the cleft ends blindly, since the epithelium of the laryngeal entrance has fused (Figs. 326 and 327). At the points where each arytenoid F.c.

s. Rw.

-Pl.ep.l. PI. ar.-ep.

Fig. 337. — The entrance of the larynx in an embryo of 40-42 days (15-16 mm.). (After Kallius, 1897.) PL ar.-ep., plica ary-epiglottica; PL ep. L, plica epiglottica lateralis; s. Rw., lateral pharyngeal swelling. The remaining lettering as in Fig. 335. X 15.

swelling is folded there is a tubercle, the tuberculum corniculatum of Kallius or the tub. arytcenoideum of Soulie and Bardier. Lateral to this a second tubercle, the tuberculum cuneiforme, appears, according to Kallius, in embryos of 8-9 mm., but according to Soulie and Bardier, only much later, in fetuses of about 40 mm. (Compare Figs. 334—339.) Kallius finds also about this time the plicce epiglottic^ laterales, extending from the anlage of the epiglottis toward two lateral folds of the mucous membrane of the pharynx (lateral phamygeal swellings) (Fig. 337) ; Soulie and Bardier do not, however, perceive these folds (Fig. 338). In fetuses of about 40 mm. the fusion of the laryngeal walls becomes dissolved, the tubercles of the arytenoid swellings withdraw from contact with the caudal surface of the epiglottis, and the entrance of the larynx becomes oval (Fig. 339). Later, according to Kallius, the lateral plicae epiglotticse unite with the lateral pharyngeal swellings to form the plicce pharyngo-epiglotticce, probably as a result of the descent of the larynx; at all

PHARYNX AND ORGANS OF RESPIRATION.

479

events, these folds become very distinct later on (Fig. 340) and are even much higher in the new-born child than in the adult. The plicce glosso-epiglotticce laterales separate from these, according to Soulie and Bardier, in fetuses of more than 29/43 cm.

According to Frazer (1910), the cuneiform tubercle corresponds to the medial end of the fourth arch, the corniculate tubercle to that of the fifth. The transverse limb of the T-shaped laryngeal cleft represents a portion of the pharyngeal cavity, whose caudal boundary in the adult would be represented by the free edge of the true vocal cord and a line drawn from one tip of the processus vocalis to the apex of the arytenoid.

Kallius finds in the lateral plicae epiglotticae the similarly named, skeletonless portion of the epiglottis observed by Groppert in the lower mammals; the folds remain recognizable throughout life in the majority of the mammals, oral to the plicae arj-epiglotticae. The lateral pharyngeal swellings may be phylogenetic representatives of the plicae palatopharyngeal of the mammals (Goppert).

Ep.

T. ar.

F. ia.-—\

Rw.

m

r-B. I.

y-PZ. ar.-ep. f- — Aw.

• — O. c. ph.-tr.

I

m

Fig. 338. — Entrance of the larynx of an embryo of 30 mm. From a dissection. (After Soulid and Bardier, 1907.) Aw., arytenoid swelling; B. I., base of the tongue; Ep , epiglottis; F. ia., fissura interarytaenoidea; O. c. ph.-tr., orifice of thepharyngotracheal canal; PL ar.-ep., plica ary-epiglottica; Rw., wall of pharynx. X 20.

The shape of the cavity of the larynx changes considerably during development. The cleft-shaped lumen of the laryngeal groove becomes obliterated, as has been stated, by the arytenoid swellings coming into apposition and by the fusion of their epithelium (compare Figs. 325-327). Nevertheless, this epithelial fusion, first described by Roth (1880), is in the beginning, at least, by no means complete (Fig. 336) : on the one hand, there remains orally, between the arytenoid swellings and the epiglottis, a funnelshaped cavity which usually ends blindly ventro-caudally ; on the other hand, a fine canal persists in the epithelium along the posterior wall of the larynx, beginning at the interarytenoid notch and passing, with a gradual enlargement, into the tracheal lumen (embryo of 8-9 mm., according to Kallius; canalis pharyngotrachealis of Soulie and Bardier; Figs 334—338). Frequently, however, even in this stage and also later, complete fusion occurs (compare the thorough account by Fein, 1903). The fusion extends

480 HUMAN EMBRYOLOGY.

caudally beyond the region of the glottis, — that is to say, to tne region of the vocal cords, and its lower boundary may correspond vvith the linea arcuata inferior, described by Eeinke and occasionally perceptible even in the adult (Kallius). According to Soulie and Bardier, however, the fusion for a time extends downward as far as the region of the cricoid cartilage (embryos of 19 mm.). The fusion gradually dissolves in embryos between 17 and 40 mm. (Fein; ; indeed it perhaps begins somewhat earlier (Kallius), the solution showing itself at first as small spaces in the line of fusion. It results probably from a breaking down of cells. (Fein). Kallius mentions the incisura interarytamoidea as one of the places where the fusion persists for a long time, but Fein contradicts this statement.

PI. ph.-ep.-i

Ad. I.

T. can...

/.' B . ia. .-4. -I

« 

Fig. 339. — Entrance of the larynx of an embryo of 16/23 cm., male. From a dissection. (After Soulie" and Bardier, 1907.,) Ad. I., aditus laryngis; PI. ph.-ep., plica pharyngo-epiglottica. The remaining lettering aa in Fig3. 335 and 338. X 6.

A satisfactory explanation of the epithelial fusion in the larynx has not yet been given. A difference from the epithelial fusions in other portions of the body exists in this case, in that an epithelial proliferation does not precede it; the epithelium is simply compressed between the mesodermal arytenoid swellings, its nuclei are arranged parallel to the mesodermie surface (Kallius). On account of its transitiveness it cannot be regarded as a protective phenomenon. Compare also V. Schmidt (1910).

The vocal cords are recognizable when the fusion is completely dissolved; the anlage of the ventriculus laryngis indicates their position. According to Soulie and Bardier, the anlage of the ventricle appears in embryos of 24 mm. as a solid epithelial bud, which acquires an independent lumen at the beginning of the third month; this later unites with the lumen of the larynx by its

PHARYNX AND ORGANS OF RESPIRATION. 481 epithelial stalk becoming hollow. Accordingly the ventricle has for a time the form of a spherical vesicle with a cylindrical stalk ; but in a fetus of 44/57 mm. the typical form occurs. The ventricle makes its appearance earlier than the date Kallius assigns to it (middle of the fourth month) ; Hansemann (1899) finds it in an embryo of 27 mm. as a blind sack. The distinct delimitation of the vocal cords first occurs, however, in the middle of the third month (fetus of 37 mm., according to Soulie and Bardier) ; their epithelium differs from that of the surrounding regions, after stages of 45-50 mm., by the absence of cilia. Elastic fibres and a distinct musculature first appear at about the middle of pregnancy; yet at birth the free edge of the vocal cord is rounded and only assumes y—B. i.

%l -^ft-Pl, ph.-ep.

YBg V PI. ar.-ep.

T . • ™ J /-i t j - * ^f""~ , w~ &• P O. c. ph.-lr.~-' '

Fig. 340. — Entrance of the larynx of an embryo of 29/43 cm., male. From a dissection. (After Soulie and Bardier, 1907.) S. p., sinus piriformis. The remaining lettering as in Figs. 335, 338, and 339. X 3.

its definitive form within the first six months of extra-uterine life. According to Frazer (1910), a prechondral "node" occurs imbedded in the ventral ends of each cord during the second fetal month; these disappear later. — The plicce ventricular es form at first, after the formation of the ventricle, roundish elevations, in which glands appear in the fourth month. The ciliated epithelium with which they are covered is replaced by a squamous epithelium in the course of the first year.

The epithelium of the larynx caudal to the region of fusion rests at first close upon the cricoid cartilage (embryos of about 20 mm.; compare Fig. 330); later a rather thick layer of loose connective tissue becomes interposed between the epithelium and the cartilage (fourth month, according to Kallius). The cricoid cartilage at this time consequently grows more rapidly than its epithelial lining, but later again is equalled by it.

Vol. n.— 31

482 HUMAN EMBRYOLOGY.

The entire larynx in embryos from about 8 mm. onward is proportionately large and only acquires its proper dimensions after birth (Kallius, Merkel). It is much higher in embryos and fetuses than in the adult, and in the fifth month projects into the pharyngonasal cavity, whereby the epiglottis rests upon the dorsal surface of the soft palate as in most mammals. At the time of birth the descent of the larynx is not yet completed ; the glottis in the new-born child is at about the level of the disk between the second and third cervical vertebra, while in the adult it is in front of the fifth vertebra. This relation, however, is subject to some slight individual variation (compare Merkel, 1902).

IV. The Lungs.

After the lung anlage has become paired two pulmonary sacks or vesicles are to be distinguished, and at first these appear to be symmetrical (compare Fig. 157 of Broman, 1904). Very soon, however, they become unsymmetrical, the right one becomes larger and bends caudally and dorsally, while the left at first has an almost transverse position (in embryos of 5 mm.; compare Fig. 341, and also Fig. 2 of Blisnianskaja and Fig. 170 of Broman). Each pulmonary sack ends in a swollen flask-shaped stem bud.

Our morphological knowledge of the arrangement of the bronchial system does not date further back than Aeby (1880), whose results, apart from the establishment of a special " eparterial " bronchial system, have been confirmed by later investigations. The account that follows is based especially upon the unsurpassed work of Narath (1901). Unfortunately, the number of human embryos studied by this author was very small, and the figures of Blisnianskaja (1904) that have since been published are rather incomplete. More recent investigations and figures of the development of the human lungs, especially in later stages, are wanting.

A short statement as to the nomenclature used in describing the branchings of the bronchi may be given. The stem bud is the anlage of the stem bronchus, which traverses the entire lung and from which branches or lateral bronchi are given off. These extend out in the four principal directions and are either direct or indirect lateral bronchi (bronchi of the first or second order; the latter also known as accessory bronchi) ; they are repeated at approximately regular intervals (Aeby). Each group of lateral bronchi given off, according to the principal directions at approximately the same level, constitutes one of the stories or tiers of the lung; they are perhaps genetically related (Narath). The most important and strongest lateral bronchi are those termed ventral by Aeby ; they arise laterally and extend at first laterally, but later supply the ventral region of the lung and also in its oral portions pass more or less to the ventral side; they lie ventral to the main stem of the pulmonary artery. On this account Narath has retained Aeby's name for them, while other authors (His, Robinson, d'Hardivillier, Flint) term them lateral or external bronchi. Toward the end of the stem bronchus they pass more and more toward its dorsal surface; a line connecting their origins would therefore have a spiral course, an arrangement that is repeated in the other bronchi and in the course of the arterial stem. The second most important group is that of the dorsal bronchi, which arise orally to the corresponding ventral

PHARYNX AND ORGANS OF RESPIRATION. 483 bronchi from the dorsal surface of the stem bronchus, and are not infrequently represented by several branches in each lung tier. The apices of the lungs are supplied by apical bronchi (Narath) ; their significance is still disputed. The right apical is Aeby's eparterial bronchus, and in that author's opinion is a special element not represented elsewhere in the lung, while Narath regards it as the first dorsal bronchus (see below). In addition there occur ventral and dorsal accessory bronchi (lateral bronchi of the second order), whose importance is small and whose formation is quite irregular. The ventral ones have been regarded as direct outgrowths from the stem bronchus and have been termed by His and Flint, for example, simply ventral bronchi, in contrast to the lateral ones (Aeby's ventral). Among them one, the infracardial bronchus, is especially well developed and worthy of mention. Flint (and also d'Hardivillier) regard the dorsal accessory bronchi also as direct medial branches of the bronchus.

Fig. 341. — Epithelial lung anlage of the embryo Rob. Meyer No. 338 (Normentafel No. 18. 5 mm.), from the ventral surface. M, stomach. X 100.

The first lateral bronchus formed is the first ventral (lateral) one of the right side (Narath) ; yet the stage of the human embryo in which it is alone present has not yet been described. 32 Shortly after this laterally directed anlage and proximal to it there appears the smaller right apical bronchus and, in the left lung, the first ventral bronchus (Narath, embryo of 7 mm., Fig. 342). At the point of origin of the anlagen (buds) of the ventral bronchi the stem bronchus is distinctly bent medially. The anlage of the (right) apical bronchus is well separated from that of the first ventral bronchus in the stage represented in Fig. 342, 33 but it

"Perhaps Fig. 2 of Blisnianskaja (1904) represents such a stage. The text lacks a definite statement.

"Blisnianskaja describes a stage in which the right apical bud is seated on the ventral one.

484 HUMAN EMBRYOLOGY.

flattens out toward the trachea. The right stem bronchus is markedly longer than the left. This stage corresponds approximately to the youngest figured by His (1887), but differs in form and proportion somewhat from His's figure.

An embryo of 11 mm. (Normentafel No. 45), whose bronchial tree almost exactly agrees with that shown in Figs. 343 and 344, had, according to Narath, on the right side the apical bronchus (Ap.), extending dorsolaterally and with three buds, the purely lateral first ventral bronchus (V\) with two buds, the purely dorsal second dorsal bronchus {D 2 ) with two buds, the infracardial bronchus (Jc) between D 2 and V 2 , directed ventromedial^ and with indications of buds, and then V 2 and V 3 , both undivided, as is also the stem bud; on the left there is the first ventral bronchus, passing laterodorsally and having a strong

Fi

Fig. 342. — Epithelial lung anlage of the embryo Chr. 1 (Normentafe No. 28, 7 mm.;. (After Narath, 1901.) Apparently the plates have been somewhat displaced in preparing the model; the right stem bronchus should descend more directly than the left (Narath). Ap., apical bud; Vi, first ventral bud. X 100.

dorsal branch, the left apical bronchus; at the origin of this V x bends sharply ventrally. The ends of both show lateral buds. Then follow D 2 passing dorsally, V 2 passing laterally and having at its root what is perhaps the anlage of a left infracardial bronchus, and Y 3 , as well as the stem bud, which does not extend quite so far caudally as that of the right side.

In an embrvo of 15.5 mm. Narath found in the right lung Ap.—V—D—Jc—V 2 —V ?i , and in the left V 1 ~D 2 —Jc—V 2 — V z — D 4 — F 4 and also a small bud which was perhaps an accessory bronchus from V 4 . This stage, though slightly older than that shown in Fig. 345, is very similar to it. The left lung is at first decidedly more advanced in the development of its deeper parts, as His has pointed out. V 1 on the left side bears the strong, dorsally directed apical bronchus. The left infracardial bronchus, which has become interposed in the series, arises from the ventral side of the stem bronchus close to V 2 and already possesses three

PHARYNX AND ORGANS OF RESPIRATION.

485

buds. D z is wanting on both sides. — Altogether Narath finds in each lung from four to five ventral bronchi, a sixth rarely forming in the left lung; the dorsal bronchi are usually fewer, frequently only two on each stem bronchus. An infracardial bronchus occurs on the right side as a rule, but is rare on the left, and its presence there in the embryo just described must be regarded as an anomaly. Its suppression on the left side appears to be due to the position of the heart and pulmonary vein on the left side (Flint). Other ventral accessory bronchi appear only here and there. The infracardial bronchus belongs almost without exception to the second lung tier ; that it may occur in the left lung has been shown by Ewart and Schaffner, while the bronchus thus described by Hasse is identical with the distal part of the main stem of the left V x (Narath). The apical bronchus of the left

Fig. 343.

Fig. 344.

Figs. 343 and 344. — Reconstruction of the lungs of an embryo at the beginning of the fifth week, ventral and dorsal views. (After Merkel, 1902.) Ap., apical bronchus; D\, Dz, etc., dorsal; Vi, Fj, etc. , ventral bronchi; Jc, infracardial bronchus.

side, leaving variations out of consideration, arises even from the beginning from V t , but in its branching it behaves throughout like the right apical, notwithstanding its smaller calibre (Narath).

The point of bifurcation of the trachea and the entire lung anlage migrates caudally during the course of development. In the embryo described by Ingalls (1904) (Normentafel No. 14, 4.9 mm.) the lung anlage lies at the level of the third cervical segment, but in one of about one month it is already at the level of the first thoracic vertebra, according to Blisnianskaja. From that time onward the recession proceeds more slowly, the level of the fourth thoracic vertebra being reached at birth. The bifurcation angle of the trachea at first diminishes (compare Figs. 341 and 342), but increases again later (for numerical data see Blisnianskaja).

The pulmonary arteries in the youngest stages that have been studied (7 mm.) arise, according to Narath, quite symmetrically

±86 HUMAN EMBRYOLOGY.

from the last aortic arch at the level of the larynx anlage, and course downward along the trachea, diverging somewhat caudally ?.nd dorsally, the left being a little more dorsal than the right. In the region of the stem bud the left artery lies laterodorsally, the right laterally. Later (embryo of 11 mm.) the right vessel bends ventrally to avoid the apical bud, distal to this again lying lateral and then laterodorsal to the stem bronchus, while the left lies at first laterodorsal and then dorsal (see also Fig. 345). Still later the recession of the heart influences the course of the arteries; these no longer run caudally alongside the trachea, but approach the lungs more and more from the ventral side, and accordingly become bent around the bronchial tree until its branches prevent a continuation of the process. The first of these branches is on both sides the first ventral bronchus, the right apical bronchus, being always situated dorsal to the artery, having no such effect upon it.

The pulmonary veins form at first a single stem opening into the atrium (see the chapter on the development of the heart). Narath observed it coming out of the angle of bifurcation of the trachea in an embryo of 7 mm. (Fig. 348). In an embryo of 11 mm. there was a main vein on each side, situated ventromedially to the stem bronchus, and opening into it a transverse vein from the first lung tier (compare also Fig. 345, from His). By the common stem being taken up into the wall of the left atrium, all four principal veins finally open directly into the atrium. The course of the veins is also influenced by the recession of the heart; the vein from the upper lung tier is forced to descend, the main stems become partly bent around the stem bronchus and their proximal portions pass transversely to the heart.

The situation of the pulmonary veins ventromedially to the stem bronchus is determined by the heart, according to Flint, and their position again explains the dorsolateral course of the arteries. Flint, however, ascribes to the arteries no importance in determining the arrangement of the bronchial branching. Ontogenetically they certainly have no such influence, since the first lateral bronchi are formed before the arterial stems appear (Narath).

A number of important questions cannot be settled from the study of human embryos, partly because the necessary material is not yet available and partly because the questions are not to be settled by what takes place iu the highly modified human lung alone. The modification is due principally to the shortening of the trunk (G-. Ruge) ; the human lung is exceptionally short and broad, and consequently the stem bronchus is so overshadowed in the adult lung by the very long and strong branches, especially by the ventral bronchi, that it was for a long time believed that the branching was of the dichotomous type. The available

PHARYNX AND ORGANS OF RESPIRATION.

481

embryonic material suffices to demonstrate the surpassing role of the stem bronchus during development; it also shows that the principal branches are formed not dichotomously but monopodially. The stem bronchus is throughout a continuous structure, whose undivided tip keeps on growing, while the lateral bronchi appear at some distance from it. Nevertheless, according to Narath, the stem bud always takes part in the formation of the lateral branches; the lateral bud always arises in its territory, and this is true not only for the stem bud, but also for the terminal buds of each lateral branch, and therefore for the entire branching. However, Narath did not study older stages with more than the fourth order of branches; for the later ones the

.4. p. d.

A. p. s.

Fia. 345. — Anlage of the lung of embryo N (10.5 mm.), seen from in front with arteries and veins. (After His, 1887.) A. p. d. and A. p. s., right and left pulmonary artery; V. p., pulmonary vein. The remaining lettering as in Fig. 343. X 50.

occurrence of equal or unequal dichotomous divisions or of division even into three has been generally accepted (Kolliker, Merkel, Flint). These smaller branches, however, are so greatly under the influence of their surroundings that probably no very great importance is to be assigned to these departures from the main type. Indeed, such twigs, formed dichotomously, do not remain symmetrical (Flint). From Narath's conception of the branching certain further consequences relative to the significance of the stem bronchus follow. For only in the formation of the ventral bronchi does the stem bud participate, the dorsal bronchi and the ventral accessory ones, including the infracardial bronchi, arise from the stem bronchus frequently only after the formation of the corresponding ventral ones, or from the roots of the latter,

488 HUMAN EMBRYOLOGY.

and, at least in some cases, proximal "to these, so. that they are separated by them from the stem bud. Narath assumes, on the basis of comparative observations, that all these bronchi are primarily branches of the ventral ones and that they have secondarily become displaced on to the stem bronchus. This view is accepted by Blisnianskaja, while d'Hardivillier and Flint, for example, advocate the equivalency of all branches passing off in the principal directions and deny a migration of them. According to this the stem bronchus as well as the stem bud possesses a capability for branch formation. Narath 's conception of monopodial branching is, accordingly, somewhat different from that of the remaining authors; it is monopodial with acropetal development of the lateral buds. A final statement on this question can hardly be given here; indeed, it cannot be given on the basis of development alone.

As already stated, Aeby assigns a special significance to the right apical bronchus, which he terms the eparterial bronchus. Its development and comparative anatomy show, however, that it is the first dorsal bronchus; the "crossing" of the stem bronchus by the artery distal to it does not occur in the lower mammals and occurs late ontogenetically, being dependent on the degree of recession of the heart (Narath). 34 More difficult of explanation are the relations on the left side. Most authors (most recently Pensa, 1909) assume either a lack of the corresponding bronchus of the left side or (d'Hardivillier) its degeneration. According to Narath, who derives the dorsal from the ventral bronchi, the left apical bronchus is possibly a true dorsal bronchus which, from some cause or other, perhaps the course taken by the arteries (the aortic arches and the ductus Botalli, which, indeed, are generally made responsible for the asymmetry of the first lung tier), has not been able to migrate on to the stem bronchus and, on account of the course of the left pulmonary artery, then arises from the ventral bronchus far from its origin. With this explana tion Blisnianskaja agrees. If this be the case, the upper lobe of the left lung is equivalent to the upper and middle lobes of the right. At all events, the idea of a special "eparterial" group of bronchi cannot be maintained.

In the embryonic lung the abundance of connective tissue is very striking; the intervals between the relatively widely sepa s * Flint (1906) identifies the right apical bronchus, which arises from the trachea in the pig, as the first lateral bronchus (a ventral bronchus according to the nomenclature vised here), and sees in it the sole representative of the first lung tier, which is completely wanting on the left side. However, the course of the arteries is not sufficiently established and, furthermore, the question is not to be settled by what occurs in the pig, which in this respect is certainly not a primitive form.

PHARYNX AND ORGANS OF RESPIRATION.

48!)

rated branches of the bronchi are filled with loose mesodermal tissue. This in the immediate vicinity of the epithelial tubes arranges itself concentrically around them and is here somewhat richer in cells than in the middle region between the bronchial rami (Fig. 346). With the progress of the bronchial branching the end buds become gradually smaller ; in the fourth month they have, according to Kolliker (quoted by Merkel, 1902), a diameter of 0.18 to 0.27 mm., in the beginning of the fifth month they measure only 0.09 to 0.13 mm., with a maximum of 0.15 mm. At about this time the lobules appear as the result of an increase of the embryonic connective tissue in the intervals between the areas of the bronchioles; the lobules in a fetus of 20 weeks have an

Fig. 346. — Section through the lower lobe of the right lung of a fetus of 100 mm. vertex-breech length taken at right angles to the dorsal surface. Br., bronchus; A., artery. X 20.

average diameter of 0.25 mm., according to Merkel (1902). At this stage of development the transverse sections of the larger branches are stellate, "the largest ones are lined by a ciliated epithelium. The cartilage plates in their walls do not extend beyond the first branches, nor do the gland anlagen. The muscles are distinctly recognizable from the surrounding mesodermic tissue. ' ' At the end of the sixth month the ends of the finer bronchi have a diameter of only 0.056 to 0.067 mm. and are very closely packed; they may already be termed alveoli (Kolliker). Their epithelium is low and the connective tissue in their immediate vicinity is still very rich in nuclei, although its quantity is on the whole greatly reduced. The cartilage formation extends in the sixth month far into the bronchial twigs, but the glands are still confined to the largest trunks (Merkel).

490 According to Kolliker (quoted by Merkel), the further development takes place in the following manner : ' ' The formation of the air-cells and smallest lobes, beginning in the sixth month, is completed only in the last months of pregnancy, for while the aircells of the mature fetus are scarcely greater than in the sixth month, and measure only from 68 to 135 w, even in the lungs of new-born children who have already breathed the lobes themselves increase very markedly in size, so that the secondary lobes, which have a diameter of only 0.65 to 2.23 mm. in an embryo of six months, measure 4.5 to 9.0 mm. and over in the new-born child." The formation of new branches is, however, according to Merkel, scarcely to be followed in the later months, on account of the abundant foldings of the alveolar walls.

O. E. —

i i Un. P. rug. Mg.v.

Fig. 347. — The mesodermal anlage of the lungs of an embryo of 5 mm. (Normentafel No. 20). (After Broman, 1904.) The epithelial anlage, according to Broman, is almost identical with that shown in Fig. 341, except that it is more asymmetrical and the left lung is directed even a little cranially. t. L., I. L., right and left lung; Mc, posterior mesocardium; Mg., stomach; Mg. v., ventral mesogastrium; Ng., accessory (mesolateral) mesentery; O. E., upper limb; P. mg., plica mesogastrica; Un., mesonephros. X 35.

Elastic tissue is relatively late in making its appearance in the lung, according to Linser (1900), whose results, according to Merkel, agree essentially with those of Lenzi (1898). Already recognizable in the vessels in the third month, it appears at the beginning of the fourth month in the largest bronchi and increases slowly in amount ; in the middle of the fifth month its fibres first appear in the alveoli, and in the seventh they occur free in the stroma. The tissue does not stain as deeply as it does later on, and is to be regarded as young, immature elastic tissue, which becomes mature a few weeks after birth under the influence of use. At this time, too, an enormous increase in the number of fibres takes place. — The pulmonary arteries, which possess the typical structure before birth, afterwards come to resemble the pulmonary veins, owing to a reduction of their tunica media, the elastic tissue in the walls of the veins becoming increased in amount.

PHARYNX AND ORGANS OF RESPIRATION. 491 The formation of the lobes of the lung was first considered by Aeby and was first worked out by Narath. Their development in the human lung has been described principally by Blisnianskaja, although some important figures of very young stages have been given by Broman (1904).

Aeby defines a lung lobe as follows : "A true lobe is never supported by more than a single lateral bronchus and therefore includes no portion of the stem bronchus." The "lower lobe" of the lung which contains the stem bronchus and most of the branches, without showing any corresponding markings on the surface, is termed by Aeby the "lung stem." Soon after the formation of the first lateral bud in the embryo, each bud becomes marked out upon the surface of the mesodermal anlage of the lung. This is a thick growth of mesoderm which projects on each

Ap.-,

Ki

Fig. 348. — Mesodermal anlage of the lung of the embryo Chr. 1 (7 mm.; compare Fig. 342). (After Narath, 1901.) Ven. p., pulmonary vein. The remaining lettering as in Fig. 342. X 100.

side into the ccelom (see Vol. I, Chapter 13) and surrounds the epithelial pulmonary sack, surpassing it in volume, however, many times. At first, before the development of the lateral bronchi, the surface of this anlage is smooth and rounded (Fig. 347; for still younger stages see Broman, Figs. 144 and 146 for a 3.4 mm. embryo, and Figs. 156, 158, 160, and 162 for a somewhat further developed embryo of 3 mm.) ; later it becomes almost mulberrylike, since not only the buds of the first lung tier (Fig. 348), but soon also those of the following tiers produce elevations on the surface (Figs. 349 and 350). Only the youngest part of the lung appears for a time as the lung stem in Aeby's sense. With the development of the buds the mesoderm over their tips gradually diminishes in quantity, although the branches continue to be imbedded in an abundant mesoderm. A series of lateral and a series of dorsal elevations are especially marked, and after these

492 the infracardial elevation. "Each primary elevation then becomes again divided into several secondary elevations 35 by the budding of the bronchial bud which it contains, and so the process goes on until the pulmonary wings become covered with fine granules. . . . With the growth of the lung these gradually disappear and the surface usually becomes smoother" (Narath). Only the first-formed furrows persist (Figs. 351 and 352), probably on account of the rapid and extensive growth of the first bronchi. Consequently in man only the first lung tier finally takes part in the formation of the lobes.

Oes.

Fig. 350 Figs. 349 and 350. — Mesodermal anlage of an embryo of about 13 mm. seen from the ventral and the dorsal surface. (After Blisnianskaja, 1904.) Oes., oesophagus. The remaining lettering (added by the present author) as in Fig. 342; in the right apical lobe (di) a subdivision into a dorsal and a ventral (d and v) portion is indicated. X 25.

Oes. Oes.

Fig. 351. Fig. 352.

Figs. 351 and 352. — Lungs of an embryo of about 17.5 mm. seen from the right and from the left. (After Blisnianskaja, 1904.) Oes., oesophagus. X 12.

The arrangement of the lobes in animals and also a number of human variations, as well as the embryonic arrangement of the lobes, can be referred to a schema given by Narath (Fig. 353), in which the boundaries of the lung tiers (principal grooves) and those between the dorsal and ventral zones (accessory grooves) are shown as boundaries of the lobes. The most frequent variety is probably, however, the occurrence of a right infracardial lobe, which owes its existence to a similar process taking place on the medial surface and base of the lung. The lobes, as well as the bronchi, are, however, greatly reduced in man, in correspondence with the shortening of the trunk and the appoximation of the heart to the diaphragm (Ruge). As regards the corresponding pleural space see Vol. I, p. 547. The separation of a lobe from the right apex by the vena azygos is a

To be seen in the upper lobe in Fig. 350.

PHARYNX AND ORGANS OF RESPIRATION. 493 variation that has nothing to do with the formation of the bronchial tree, but is rather to be explained as an adaptation of the lung to the space at its disposal (Narath, Bluntschli, 1905). The complete separation of portions of the lung, occasionally observed, is to be referred to disturbances in the early embryonic stages of development (Hammar, 1904).

Blisnianskaja has given some data regarding the evolution of the form of the lung as a whole. The form of the embryonic lungs is especially influenced by the great size of the heart; what are later the medial surfaces are first directed ventrally, and the

Fia. 353. — Schema of the lobation of the lung. (After Narath, 1901.) D and V, dorsal and ventral lobes.

lateral ones dorsally. The lungs are at first relatively more developed in what is later the dorsoventral (in the embryo approximately transverse) diameter than in the transverse (in the embryo almost sagittal) one, a condition that recalls what occurs in lower forms (apes). The same holds for the position of the base of the lung, which at first is much more sloped than it is later (compare Figs. 351 and 352). The lungs are drawn out almost to a point, which forms the lower and posterior pole. In the third fetal month they gradually assume the proportions seen in the adult.

LITERATURE.

(The Normentafel referred to in the text is the Normentafel des Menschen, Jena, 1908, by Keibel and Elze.) , A. Literature on the Development of the Pharynx.

Bien, G. : Ueber accessorische Thymuslappen imj trigonum caroticum, Anat. Anz., vol xxix, 1906. Bien, G. : Ueber accessorische Thymuslappen im trigonum caroticum bei einem Embryo von 17 mm. grosster Lange, Anat. Anz., vol. xxxi, 1907. Born, G. : Ueber die Derivate der embryonalen Schlundbogen und Schlundspalten bei Saugetieren, Arch, fiir mikr. Anat., vol. xxii, 1883. Dandy, W. E.: A Human Embryo with Seven Pairs of Somites, measuring about 2 mm. in length, Amer. Journ. of Anat. vol. x, 1910.

494 HUMAN EMBRYOLOGY.

Elze, C. : Beschreibung eines raenschlichen Embryo von ea. 7 mm. grosster Lange, etc., Anat. Hefte, vol. xxxv, 1907. Erdheim, J.: I. Ueber Sebilddriisenaplasie. II. Geschwiilste des ductus thyreo glossus. III. Ueber einige menschliche Kiemenderivate. Beitr. zur path.

Anat. und allg. Path., vol. xxxv, 1904. Fox, H. : The Pharyngeal Pouches and their Derivatives in the Mammalia, Amer.

Journ. Anat., vol. viii, 1908. Getzowa, S. : Ueber die glandula paratbyreoidea, intrathyreoidale Zellhaufen derselben und Reste des postbranchialen Korpers, Arch, fur path. Anat., vol. clxxxviii, 1907. Greil, A. : 1905. See p. 496. Groschupp, K. : Bemerkungen zu der vorlaufigen Mitteilung von Jacoby : Ueber die Entwicklung der Nebendrusen der Sehilddruse und der Carotidendriise, Anat. Anz., vol. xii, 1896. Groschupp, K. : Ueber das Vorkommen eines Tbymussegmentes der vierten Kiementasche beim Menschen, Anat. Anz., vol. xvii, 1900. Grosser, O. : Zur Kenntnis des ultimobrancbialen Korpers beim Menchen, Anat.

Anz., vol. xxxvii, 1910. Grosser, O. : Der Nerv des funften Visceralbogens beim Menschen, Anat. Anz., vol. xxxvii, 1910. Grosser, O. : Zur Entwicklung des Vorderdarmes menscblicher Embryonen bis 5 mm. grosster Lange. Sitzber. R. Akad. Wiss. Wien, vol. cxx, 1911. Gruenwald, L. : Ein Beitrag zur Entstehung und Bedeutung der Gaumenmandeln, Anat. Anz., vol. xxxvii, 1910. Hammar, J. A.: Studien iiber die Entwicklung des Vorderdarmes und einiger angrenzender Organe. I. Abth. Allgemeine Morphologie der Schlundspalten beim Menschen. Entwicklung des Mittelohrraumes und des ausseren Gehor ganges, Arch, fur mikr. Anat., vol. lix, 1902. — II. Abth. Das Schicksal der zweiten Schlundspalte. Zur vergleichenden Embryologie und Morphologie der Tonsille, Arch, fiir mikr. Anat, vol. Ixi, 1903. Hammar, J. A. : Ein beachtenswerter Fall von kongenitaler Halskiemenfistel nebst einer Uebersicht iiber die in der normalen Ontogenese des Menschen existierenden Vorbedingungen solcber Missbildungen, Beitr. zur path. Anat.

und allg. Path., vol. xxxvi, 1904. Hammar, J. A.: Ueber die Natur der kleinen Thymuszellen, Arch, fiir Anat. u.

Phys. Anat. Abth. 1907. Hammar, J. A. : Fiinfzig Jahre Thymusf orschung. Kritische Uebersicht der normalen Morphologie, Ergebn. der Anat. u. Entwicklungsgesch., vol. xix, 1910. Hammar, J. A.: Zur groberen Morphologie und Morphogenie der Menschen thymus. Anat. Hefte, vol. xliii, 1911. Herrmann, G., and Verdun, P. : Notes sur l'anatomie des eorps post-branchiaux.

Miseellanees biologiques dediees au professeur Alfred Giard. Paris 1899. Herrmann, G., and Verdun, P. : Persistance des corps post-branchiaux chez l'homme. Remarques sur l'anatomie comparee des corps post-branchiaux.

Comptes Rend. Soc. Biol. Paris, 1899. Herrmann, G., and Verdun, P. : Note sur las corps post-branchiaux des Cameliens.

— Les corps post-branchiaux et la thyroide; vestiges kystiques. Comptes Rend. Soc. Biol. Paris, 1900. His, W. : Anatomie menschlicher Embryonen, Heft 3, Leipzig, 1885. His, W. : 1. Ueber den sinus praecervicalis und iiber die Thymusanlage. 2. Nach trag zu vorstehender Abhandlung. Arch, fiir Anat. u. Phys. Anat. Abth.

1886. His, W. : Schlundspalten und Thymusanlage, Arch, fiir Anat. u. Phys. Anat. Abth.

1889.

PHARYNX AND ORGANS OF RESPIRATION. 495 Ingalls, N. W. : Beschreibung eines menschlichen Embryo von 4.9 mm. Lange, Areb. fiir mikr. Anat., vol. lxx, 1907. Kastschenko, N. : Das Scbicksal der embryonalen Scblundspalten bei Saugetieren, Arcb. fiir mikr. Anat., vol. xxx, 1887. Kohn, A.: Studien iiber die Scbilddriise, I and II, Arch, fiir mikr. Anat., vols.

xliv and xlviii, 1895 and 1896. Kohn, A.: Die Epithelkorperchen, Ergebn. der Anat. u. Entwicklungsgescb., vol.

ix, 1899. KLroemer: Wacbsmodell eines jimgen menschlichen Embryo, Verb. d. Deutsch.

Ges. fiir Gynak., 1903. Kuersteiner, W. : Die Epithelkorperchen des Menschen in ibrer Beziehung zur thyreoidea und thymus, Anat. Hefte, vol. xi, 1898. Low, A. : Description of a Human Embryo of 13-14 Mesodermic Somites, Journ.

Anat. and Phys., vol. xlii, 1908. Mabesch, R. : Kongenitaler Defekt der Schilddriise bei einem elf jahrigen Madchen mit vorhandenen " Epitbelkorpercben," Zeit. fiir Heilk., vol. xix, 1898. Maurer, F. : Die Schilddriise, thymus und andere Schlundspaltenderivate bei der Eidechse, Morph. Jahrb., vol. xxvii, 1899. Maurer, F. : Die Entwicklung des Darmsystems, Hertwig's Handb. der vergl. u.

exp. Entwicklungslehre, vol. ii, 1902. Maximow, A. : Untersuchungen iiber Blut und Bindegewebe. II. Ueber die Histogenese der thymus bei Saugetieren, Arch, fiir mikr. Anat., vol. lxxiv, 1909. Meyer, R. : Ueber Bildung des Recessus pharyngeus medius s. Bursa pharyngea im Zusammenhang mit der Chorda dorsalis bei menschlichen Embryonen.

Anat. Anzeiger. Bd. 37. 1910. Peuker, H. : Ueber einen neuen Fall von kongenitalem Defekt der Schilddruse mit vorhandenen " Epithelkorperchen," Zeitschr. f . Heilk., vol. xx, 1899. Piersol, G. A. : Ueber die Entwicklung der embryonalen Scblundspalten und ihre Derivate bei Saugetieren, Zeit. fiir wiss. Zool., vol. xlvii, 1888. Prenant, A. : Contribution a l'etude organique et histologique du thymus, de la glande thyroide et de la glande carotidienne, La Cellule, vol. x, 1894. Prenant, A. : Sur le developpement des glandes accessoires de la glande thyroide et celui de la glande carotidienne, Anat. Anz., vol. xii, 1896. Rabl, C. : Zur Bildungsgeschichte des Halses, Prager ined. Wochenschr., vols, xi and xii, 1886 and 1887. Rabl, H. : Ueber die Anlage der ultimobranchialen Korper bei den Vogeln, Arch.

fiir mikr. Anat., vol. lxx, 1907. Schaffer, J., and Rabl, H. : Das thyreothymische System des Maulwurfs und der Spitzmaus. I. Morphologie und Histologic by J. Schaffer. II. Die Entwicklung des thyreo-thymischen Systems beim Maulwurf by H. Rabl. Sitzber.

kais. Akad. Wiss. Wien, vols. 117 and 118, 1908 and 1909. Simon, Ch. : Thyroide laterale et glandule thyroidienne chez les mammif eres, Nancy, 1896. Stteda, L. : Untersuchungen iiber die Entwicklung der glandula thymus, glandula thyreoidea und glandula carotica. Leipzig, 1881. Stoehr, P. : Die Entwicklung des adenoiden Gewebes, der Zungenbalge und der Mandeln des Menschen, Festschr. z. fiinfzig jahrigen Doktorjubilaum von Nageli und Kolliker, 1891. (Author's abstract in Anat. Anz., vol. vi, 1891 1892.) Stoehr, P. : Ueber die Natur der Thymuselemente, Anat. Hefte, vol. xxxi, 1906. Stoehr, P. : Ueber die Abstammung der kleinen Thymusrindenzellen, Anat. Hefte, vol. xii, 1910. Sudler, M. T. : The Development of the Nose and of the Pharynx and its Derivatives in Man, Amer. Journ. Anat., vol. i, 1901-1902.

496 HUMAN EMBRYOLOGY.

Tandler, J. : Ueber die Entwicklung des V. Aorteubogens und der V. Schlund tasche beirn Menschen, Anat. Hefte, vol. xxxviii, 1909. Tettenhamer, E. : Ueber das Vorkommen offener Seblundspalten bei einem mensch lieben Embryo, Miincbener med. Abhandl., Series 7, part 2, 1892. Thompson, P. : Description of a Human Embryo of Twenty-three Paired Somites, Joum. of Anat. and Pbys., vol. xli, 1907. Thompson, P. : A Note on the Development of the Septum Transversum and the Liver, Journ. of Anat. and Pbys., vol. xlii, 1908. Tourneux, F., and Soulie, A.: Sur l'existence d'une V. et d'une VI. poche endodermique ehez l'embryon humain, Comptes Rendus Soe. Biol. Paris, .1907. Tourneux, F., and Verdun, P. : Sur les premiers developpements de la thyroide, du thymus et des glandules thyroidiennes, Journ. de l'Anat. et de la Phys., vol. xxxiii, 1897. Verdun, P. : Derives branchiaux chez les Vertebres superieurs, These, Toulouse, 1898. Woeleler, A. : Ueber die Entwicklung und den Bau der Schilddruse mit Riick sicht auf die Entwicklung der Kropfe. Berlin, 1881. Zuckere:andl, E. : Die Epithelkorperchen von Didelphys azara nebst Bemerkungen iiber die Epithelkorperchen des Menscben, Anat. Hefte, vol. xix, 1902. Zuckerkandl, E. : Die Entwicklung der Schilddruse und der thymus bei der Ratte, Anat. Hefte, vol. xxi, 1903.

B. Literature op the Development oe the Respiratory Tract.

Aeby, C. : Der Bronchialbamn der Saugetiere und des Menscben, Leipzig, 1880. Blisnianskaja, G. : Zur Entwicklungsgesehichte der menschlichen Lungen : Bronchialbaum, Lungenform, Dissert., Zurich, 1901. Bluntschli, H. : Bemerkungen iiber einen abnormen Verlauf der vena azygos in einer den Oberlappen der rechten Lunge durchsetzenden Pleurafalte, Morph. Jahrb., vol. xxxiii, 1905. Broman, J. : Die Entwicklungsgesckiehte der bursa omentalis, Wiesbaden, 1904. Fein, J.: Die Verklebungen im Bereiche des ernbryonalen Kehlkopfes, Arch, fiir Laryngol., vol. xv, 1903. Flint, J. M. : The Development of the Lungs, Amer. Journ. Anat., vol. vi, 1906 1907. Frazer, J. E. : The Development, of the Larynx, Journ. of Anat. and Phys., vol.

xliv, 1910. Goeppert, E. : Die Entwicklung des Mundes, etc., die Entwicklung der Schwimm blase, der Lunge und des Kehlkopfes der Wirbeltiere, Hex-twig's Handb. der vergl. und exp. Entwicklungslehre, vol. ii, 1902. Greil, A.: Ueber die Anlage der Lungen, sowie der ultimobranchialen (post branchialen, suprapericardialen) Korper bei anuren Amphibien, Anat. Hefte, vol. xxix, 1905. Greil, A. : Bemerkungen zur Frage nach dem Ursprung der Lungen, Anat. Anz., vol. xxxvii, 1910. Grosser, O. : 1910 and 1911. See p. 491. Hammar, J. A.: 1902. See p. 494. Hammar, J. A. : Ein Fall von Nebenlunge bei einem Menscben fetus von 11.7 mm.

Nackenlange, Beitr. zur path. Anat. und allg. Path., vol. xxxvi, 1904. d^Hardhtllier, A. D. : Developpement et homologation des bronches principales chez les mammiferes, Thesis, Nancy, 1897. His, W. : 1885. See p. 494. His, W. : Zur Bildungsgeschiehte der Lungen beim menschlichen Embryo, Arch.

fiir Anat. und Phys., Anat. Abth., 1887.

PHARYNX AND ORGANS OF RESPIRATION. 497 Kallius, E. : Beitrage zur Entwicklungsgeschichte des Kehlkopf es, Anat. Hef te, vol. ix, 1897. Kallius, E.: Beitrage zur Entwicklung der Zunge. III. Teil: Siiugetiere. 1. Sus scrofa dom., Anat. Hefte, vol. xli, 1910. Linser, P. : Ueber den Ban und die Entwicklung des elastiscken Gewebes in der Lunge, Anat. Hefte, vol. xiii, 1900. Merkel, P. : Atmungsorgane, in Bardeleben's Handb. der Anat des Menschen, vol. vi, 1902. Moser, F. : Beitrage zur vergleichenden Entwicklungsgeschichte der Wirbeltier lunge, Arcb. fiir mikr. Anat,, vol. lx, 1902. Narath, A.: Der Broncbialbaum der Saugetiere und des Menschen, Bibliotbeca medica, Abtb. A., Anat., Pt. Ill, Stuttgart, 1901. Pensa, A. : Considerazioni intorno alio sviluppo dell' albero broncbiale nell' Uomo e in Bos taurus. Boll. Soc. Med. Chir., Pavia, vol. xxiii, 1909. Robinson, A. : Observations on tbe Earlier Stages in the Development of the Lungs of Rats and Mice, Journ. of Anat. and Phys., 1889. Schmidt, V. : Zur Entwicklung des Kehlkopfes und der Luf trohre bei den Wirbeltieren, Anat. Anz., vol. xxxv, 1910. Soulee, A., and Bardier, E. : Recherches sur le developpement du larynx chez l'homme, Journ. de l'Anat. et de la Phys., vol. xlii, 1907. Thompson, P. : 1907 and 1908. See p. 496. Weber, A., and Buvtgnier, A. : L'origine des ebauches pulmonaires chez quelques vertebres superieurs, Bibliogr. Anat., vol. xii, 1903.