CHAPTER XXI.

THE BODY CAVITY, THE VASCULAR SYSTEM, AND THE VASCULAR GLANDS.

_The Body cavity._

In the Coelenterata no body cavity as distinct from the alimentary cavity is present; but in the remaining Invertebrata the body cavity may (1) take the form of a wide space separating the wall of the gut from the body wall, or (2) may be present in a more or less reduced form as a number of serous spaces, or (3) only be represented by irregular channels between the muscular and connective-tissue cells filling up the interior of the body. The body cavity, in whatever form it presents itself, is probably filled with fluid, and the fluid in it may contain special cellular elements. A well developed body cavity may coexist with an independent system of serous spaces, as in the Vertebrata and the Echinodermata; the perihæmal section of the body cavity of the latter probably representing the system of serous spaces.

In several of the types with a well developed body cavity it has been established that this cavity originates in the embryo from a pair of alimentary diverticula, and the cavities resulting from the formation of these diverticula may remain distinct, the adjacent walls of the two cavities fusing to form a dorsal and a ventral mesentery.

It is fairly certain that some groups, _e.g._ the Tracheata, with imperfectly developed body cavities are descended from ancestors which were provided with well developed body cavities, but how far this is universally the case cannot as yet be definitely decided, and for additional information on this subject the reader is referred to pp. 355-360 and to the literature there referred to.

[FIG. 350. LONGITUDINAL SECTION THROUGH AN EMBRYO OF AGELINA LABYRINTHICA.

The section is taken slightly to one side of the middle line so as to shew the relation of the mesoblastic somites to the limbs. In the interior are seen the yolk segments and their nuclei.

1-16. the segments; _pr.l._ procephalic lobe; _do._ dorsal integument.]

In the Chætopoda and the Tracheata the body cavity arises as a series of paired compartments in the somites of mesoblast (fig. 350) which have at first a very restricted extension on the ventral side of the body, but eventually extend dorsalwards and ventralwards till each cavity is a half circle investing the alimentary tract; on the dorsal side the walls separating the two half cavities usually remain as the dorsal mesentery, while ventrally they are in most instances absorbed. The transverse walls, separating the successive compartments of the body cavity, generally become more or less perforated.

Chordata. In the Chordata the primitive body cavity is either directly formed from a pair of alimentary diverticula (Cephalochorda) (fig. 3) or as a pair of spaces in the mesoblastic plates of the two sides of the body (fig. 20).

As already explained (pp. 294-300) the walls of the dorsal sections of the primitive body cavity soon become separated from those of the ventral, and becoming segmented constitute the muscle plates, while the cavity within them becomes obliterated: they are dealt with in a separate chapter. The ventral part of the primitive cavity alone constitutes the permanent body cavity.

The primitive body cavity in the lower Vertebrata is at first continued forwards into the region of the head, but on the formation of the visceral clefts the cephalic section of the body cavity becomes divided into a series of separate compartments. Subsequently these sections of the body cavity become obliterated; and, since their walls give rise to muscles, they may probably be looked upon as equivalent to the dorsal sections of the body cavity in the trunk, and will be treated of in connection with the muscular system.

As a result of its mode of origin the body cavity in the trunk is at first divided into two lateral halves; and part of the mesoblast lining it soon becomes distinguished as a special layer of epithelium, known as the peritoneal epithelium, of which the part bounding the outer wall forms the somatic layer, and that bounding the inner wall the splanchnic layer. Between the two splanchnic layers is placed the gut. On the ventral side, in the region of the permanent gut, the two halves of the body cavity soon coalesce, the septum between them becoming absorbed, and the splanchnic layers of epithelium of the two sides uniting at the ventral side of the gut, and the somatic layers at the median ventral line of the body wall (fig. 351).

[FIG. 351. SECTION THROUGH THE TRUNK OF A SCYLLIUM EMBRYO SLIGHTLY YOUNGER THAN 28 F.

_sp.c._ spinal canal; _W._ white matter of spinal cord; _pr._ posterior nerve-roots; _ch._ notochord; _x._ subnotochordal rod; _ao._ aorta; _mp._ muscle-plate; _mp´._ inner layer of muscle-plate already converted into muscles; _Vr._ rudiment of vertebral body; _st._ segmental tube; _sd._ segmental duct; _sp.v._ spiral valve; _v._ subintestinal vein; _p.o._ primitive generative cells.]

In the lower Vertebrata the body cavity is originally present even in the postanal region of the trunk, but usually atrophies early, frequently before the two halves coalesce.

On the dorsal side of the gut the two halves of the body cavity never coalesce, but eventually the splanchnic layers of epithelium of the two sides, together with a thin layer of interposed mesoblast, form a delicate membrane, known as the mesentery, which suspends the gut from the dorsal wall of the body (figs. 119 and 351). On the dorsal side the epithelium lining of the body cavity is usually more columnar than elsewhere (fig. 351), and its cells partly form a covering for the generative organs, and partly give rise to the primitive germinal cells. This part of the epithelium is often known as the germinal epithelium.

Over the greater part of the body cavity the lining epithelium becomes in the adult intimately united with a layer of the subjacent connective tissue, and constitutes with it a special lining membrane for the body cavity, known as the peritoneal membrane.

Abdominal pores. In the Cyclostomata, the majority of the Elasmobranchii, the Ganoidei, a few Teleostei, the Dipnoi, and some Sauropsida (Chelonia and Crocodilia) the body cavity is in communication with the exterior by a pair of pores, known as abdominal pores, the external openings of which are usually situated in the cloaca[219].

[219] For a full account of these structures the reader is referred to T. W. Bridge, "Pori Abdominales of Vertebrata." _Journal of Anat. and Physiol._, Vol. XIV., 1879.

The ontogeny of these pores has as yet been but very slightly investigated. In the Lamprey they are formed as apertures leading from the body cavity into the excretory section of the primitive cloaca. This section would appear from Scott's (No. 87) observations to be derived from part of the hypoblastic cloacal section of the alimentary tract.

In all other cases they are formed in a region which appears to belong to the epiblastic region of the cloaca; and from my observations on Elasmobranchs it may be certainly concluded that they are formed there in this group. They may appear as perforations (1) at the apices of papilliform prolongations of the body cavity, or (2) at the ends of cloacal pits directed from the exterior towards the body cavity, or (3) as simple slit-like openings.

Considering the difference in development between the abdominal pores of most types, and those of the Cyclostomata, it is open to doubt whether these two types of pores are strictly homologous.

In the Cyclostomata they serve for the passage outwards of the generative products, and they also have this function in some of the few Teleostei in which they are found; and Gegenbaur and Bridge hold that the primitive mode of exit of the generative products, prior to the development of the Müllerian ducts, was probably by means of these pores. I have elsewhere suggested that the abdominal pores are perhaps remnants of the openings of segmental tubes; there does not however appear to be any definite evidence in favour of this view, and it is more probable that they may have arisen as simple perforations of the body wall.

Pericardial cavity, pleural cavities, and diaphragm. In all Vertebrata the heart is at first placed in the body cavity (fig. 353 A), but the part of the body cavity containing it afterwards becomes separated as a distinct cavity known as the pericardial cavity. In Elasmobranchii, Acipenser, etc. a passage is however left between the pericardial cavity and the body cavity; and in the Lamprey a separation between the two cavities does not occur during the Ammocoete stage.

[FIG. 352. SECTION THROUGH THE TRUNK OF A SCYLLIUM EMBRYO SLIGHTLY YOUNGER THAN 28 F.

The figure shews the separation of the body cavity from the pericardial cavity by a horizontal septum in which runs the ductus Cuvieri; on the left side is seen the narrow passage which remains connecting the two cavities.

_sp.c._ spinal canal; _w._ white matter of spinal cord; _pr._ commissure connecting the posterior nerve-roots; _ch._ notochord; _x._ subnotochordal rod; _ao._ aorta; _sv._ sinus venosus; _cav._ cardinal vein; _ht._ heart; _pp._ body cavity; _pc._ pericardial cavity; _oes._ solid oesophagus; _l._ liver; _mp._ muscle-plate.]

In Elasmobranchii the pericardial cavity becomes established as a distinct space in front of the body cavity in the following way. When the two ductus Cuvieri, leading transversely from the sinus venosus to the cardinal veins, become developed, a horizontal septum, shewn on the right side in fig. 352, is formed to support them, stretching across from the splanchnic to the somatic side of the body cavity, and dividing the body cavity (fig. 352) in this part into (1) a dorsal section formed of a right and left division constituting the true body cavity (_pp_), and (2) a ventral part the pericardial cavity (_pc_). The septum is at first of a very small longitudinal extent, so that both in front and behind it (fig. 352 on the left side) the dorsal and ventral sections of the body cavity are in free communication. The septum soon however becomes prolonged, and ceasing to be quite horizontal, is directed obliquely upwards and forwards till it meets the dorsal wall of the body. Anteriorly all communication is thus early shut off between the body cavity and the pericardial cavity, but the two cavities still open freely into each other behind.

The front part of the body cavity, lying dorsal to the pericardial cavity, becomes gradually narrowed, and is wholly obliterated long before the close of embryonic life, so that in adult Elasmobranch Fishes there is no section of the body cavity dorsal to the pericardial cavity. The septum dividing the body cavity from the pericardial cavity is prolonged backwards, till it meets the ventral wall of the body at the point where the liver is attached by its ventral mesentery (falciform ligament). In this way the pericardial cavity becomes completely shut off from the body cavity, except, it would seem, for the narrow communications found in the adult. The origin of these communications has not however been satisfactorily worked out.

The septum between the pericardial cavity and the body cavity is attached on its dorsal aspect to the liver. It is at first nearly horizontal, but gradually assumes a more vertical position, and then, owing to the obliteration of the primitive anterior part of the body cavity, appears to mark the front boundary of the body cavity. The above description of the mode of formation of the pericardial cavity, and the explanation of its relations to the body cavity, probably holds true for Fishes generally.

In the higher types the earlier changes are precisely the same as those in Elasmobranch Fishes. The heart is at first placed within the body cavity attached to the ventral wall of the gut by a mesocardium (fig. 353 A). A horizontal septum is then formed, in which the ductus Cuvieri are placed, dividing the body cavity for a short distance into a dorsal (_p.p_) and ventral (_p.c_) section (fig. 353 B). In Birds and Mammals, and probably also in Reptilia, the ventral and dorsal parts of the body cavity are at first in free communication both in front of and behind this septum. This is shewn for the Chick in fig. 353 A and B, which are sections of the same chick, A being a little in front of B. The septum is soon continued forwards so as completely to separate the ventral pericardial and the dorsal body cavity in front, the pericardial cavity extending at this period considerably further forwards than the body cavity.

Since the horizontal septum, by its mode of origin, is necessarily attached to the ventral side of the gut, the dorsal part of the primitive body space is divided into two halves by a median vertical septum formed of the gut and its mesentery (fig. 353 B). Posteriorly the horizontal septum grows in a slightly ventral direction along the under surface of the liver (fig. 354), till it meets the abdominal wall of the body at the insertion of the falciform ligament, and thus completely shuts off the pericardial cavity from the body cavity. The horizontal septum forms, as is obvious from the above description, the dorsal wall of the pericardial cavity[220].

[220] Kölliker's account of this septum, which he calls the mesocardium laterale (No. 298, p. 295), would seem to imply that in Mammals it is completed posteriorly even before the formation of the liver. I doubt whether this takes place quite so early as he implies, but have not yet determined its exact period by my own observations.

[FIG. 353. TRANSVERSE SECTIONS THROUGH A CHICK EMBRYO WITH TWENTY-ONE MESOBLASTIC SOMITES TO SHEW THE FORMATION OF THE PERICARDIAL CAVITY, A. BEING THE ANTERIOR SECTION.

_p.p._ body cavity; _p.c._ pericardial cavity; _al._ alimentary cavity; _au._ auricle; _v._ ventricle; _s.v._ sinus venosus; _d.c._ ductus Cuvieri; _ao._ aorta; _mp._ muscle-plate; _mc._ medullary cord.]

With the complete separation of the pericardial cavity from the body cavity, the first period in the development of these parts is completed, and the relations of the body cavity to the pericardial cavity become precisely those found in the embryos of Elasmobranchii. The later changes are however very different. Whereas in Fishes the right and left sections of the body cavity dorsal to the pericardial cavity soon atrophy, in the higher types, in correlation with the relatively backward situation of the heart, they rapidly become larger, and receive the lungs which soon sprout out from the throat.

The diverticula which form the lungs grow out into the splanchnic mesoblast, in front of the body cavity; but as they grow, they extend into the two anterior compartments of the body cavity, each attached by its mesentery to the mesentery of the gut (fig. 354, _lg_). They soon moreover extend beyond the region of the pericardium into the undivided body cavity behind. This holds not only for the embryos of the Amphibia and Sauropsida, but also for those of Mammalia.

[FIG. 354. SECTION THROUGH THE CARDIAC REGION OF AN EMBRYO OF LACERTA MURALIS OF 9 MM. TO SHEW THE MODE OF FORMATION OF THE PERICARDIAL CAVITY.

_ht._ heart; _pc._ pericardial cavity; _al._ alimentary tract; _lg._ lung; _l._ liver; _pp._ body cavity; _md._ open end of Müllerian duct; _wd._ Wolffian duct; _vc._ vena cava inferior; _ao._ aorta; _ch._ notochord; _mc._ medullary cord.]

To understand the further changes in the pericardial cavity it is necessary to bear in mind its relations to the adjoining parts. It lies at this period completely ventral to the two anterior prolongations of the body cavity containing the lungs (fig. 354). Its dorsal wall is attached to the gut, and is continuous with the mesentery of the gut passing to the dorsal abdominal wall, forming the posterior mediastinum of human anatomy.

The changes which next ensue consist essentially in the enlargement of the sections of the body cavity dorsal to the pericardial cavity. This enlargement takes place partly by the elongation of the posterior mediastinum, but still more by the two divisions of the body cavity which contain the lungs extending themselves ventrally round the outside of the pericardial cavity. This process is illustrated by fig. 355, taken from an embryo Rabbit. The two dorsal sections of the body cavity (_pl.p_) finally extend so as completely to envelope the pericardial cavity (_pc_), remaining however separated from each other below by a lamina extending from the ventral wall of the pericardial cavity to the body wall, which forms the anterior mediastinum of human anatomy.

[FIG. 355. SECTION THROUGH AN ADVANCED EMBRYO OF A RABBIT TO SHEW HOW THE PERICARDIAL CAVITY BECOMES SURROUNDED BY THE PLEURAL CAVITIES.

_ht._ heart; _pc._ pericardial cavity; _pl.p_ pleural cavity; _lg._ lung; _al._ alimentary tract; _ao._ dorsal aorta; _ch._ notochord; _rp._ rib; _st._ sternum; _sp.c._ spinal cord.]

By these changes the pericardial cavity is converted into a closed bag, completely surrounded at its sides by the two lateral halves of the body cavity, which were primitively placed dorsally to it. These two sections of the body cavity, which in Amphibia and Sauropsida remain in free communication with the undivided peritoneal cavity behind, may, from the fact of their containing the lungs, be called the pleural cavities.

In Mammalia a further change takes place, in that, by the formation of a vertical partition across the body cavity, known as the diaphragm, the pleural cavities, containing the lungs, become isolated from the remainder of the body or peritoneal cavity. As shewn by their development the so-called pleuræ or pleural sacks are simply the peritoneal linings of the anterior divisions of the body cavity, shut off from the remainder of the body cavity by the diaphragm.

The exact mode of formation of the diaphragm is not fully made out; the account of it recently given by Cadiat (No. 491) not being in my opinion completely satisfactory.

BIBLIOGRAPHY.

(491) M. Cadiat. "Du développement de la partie céphalothoracique de l'embryon, de la formation du diaphragme, des pleures, du péricarde, du pharynx et de l'oesophage." _Journal de l'Anatomie et de la Physiologie_, Vol. XIV. 1878.

_Vascular System._

The actual observations bearing on the origin of the vascular system, using the term to include the lymphatic system, are very scanty. It seems probable, mainly it must be admitted on _à priori_ grounds, that vascular and lymphatic systems have originated from the conversion of indefinite spaces, primitively situated in the general connective tissue, into definite channels. It is quite certain that vascular systems have arisen independently in many types; a very striking case of the kind being the development in certain parasitic Copepoda of a closed system of vessels with a red non-corpusculated blood (E. van Beneden, Heider), not found in any other Crustacea. Parts of vascular systems appear to have arisen in some cases by a canalization of cells.

The blood systems may either be closed or communicate with the body cavity. In cases where the primitive body cavity is atrophied or partially broken up into separate compartments (Insecta, Mollusca, Discophora, etc.) a free communication between the vascular system and the body cavity is usually present; but in these cases the communication is no doubt secondary. On the whole it would seem probable that the vascular system has in most instances arisen independently of the body cavity, at least in types where the body cavity is present in a well-developed condition. As pointed out by the Hertwigs, a vascular system is always absent where there is not a considerable development of connective tissue.

As to the ontogeny of the vascular channels there is still much to be made out both in Vertebrates and Invertebrates.

The smaller channels often rise by a canalization of cells. This process has been satisfactorily studied by Lankester in the Leech[221], and may easily be observed in the blastoderm of the Chick or in the epiploon of a newly born Rabbit (Schäfer, Ranvier). In either case the vessels arise from a network of cells, the superficial protoplasm and part of the nuclei giving rise to the walls, and the blood-corpuscles being derived either from nucleated masses set free within the vessels (the Chick) or from blood-corpuscles directly differentiated in the axes of the cells (Mammals).

[221] "Connective and vasifactive tissues of the Leech." _Quart. J. of Micr. Science_, Vol. XX. 1880.

Larger vessels would seem to be formed from solid cords of cells, the central cells becoming converted into the corpuscles, and the peripheral cells constituting the walls. This mode of formation has been observed by myself in the case of the Spider's heart, and by other observers in other Invertebrata. In the Vertebrata a more or less similar mode of formation appears to hold good for the larger vessels, but further investigations are still required on this subject. Götte finds that in the Frog the larger vessels are formed as longitudinal spaces, and that the walls are derived from the indifferent cells bounding these spaces, which become flattened and united into a continuous layer.

The early formation of vessels in the Vertebrata takes place in the splanchnic mesoblast; but this appears due to the fact that the circulation is at first mainly confined to the vitelline region, which is covered by splanchnic mesoblast.

_The Heart._

The heart is essentially formed as a tubular cavity in the splanchnic mesoblast, on the ventral side of the throat, immediately behind the region of the visceral clefts. The walls of this cavity are formed of two layers, an outer thicker layer, which has at first only the form of a half tube, being incomplete on its dorsal side; and an inner lamina formed of delicate flattened cells. The latter is the epithelioid lining of the heart, and the cavity it contains the true cavity of the heart. The outer layer gives rise to the muscular wall and peritoneal covering of the heart. Though at first it has only the form of a half tube (fig. 356), it soon becomes folded in on the dorsal side so as to form for the heart a complete muscular wall. Its two sides, after thus meeting to complete the tube of the heart, remain at first continuous with the splanchnic mesoblast surrounding the throat, and form a provisional mesentery--the mesocardium--which attaches the heart to the ventral wall of the throat. The superficial stratum of the wall of the heart differentiates itself as the peritoneal covering. The inner epithelioid tube takes its origin at the time when the general cavity of the heart is being formed by the separation of the splanchnic mesoblast from the hypoblast. During this process (fig. 357) a layer of mesoblast remains close to the hypoblast, but connected with the main mass of the mesoblast by protoplasmic processes. A second layer next becomes split from the splanchnic mesoblast, connected with the first layer by the above-mentioned protoplasmic processes. These two layers form together the epithelioid lining of the heart; between them is the cavity of the heart, which soon loses the protoplasmic trabeculæ which at first traverse it. The cavity of the heart may thus be described as being formed by a hollowing out of the splanchnic mesoblast, and resembles in its mode of origin that of other large vascular trunks.

[FIG. 356. SECTION THROUGH THE DEVELOPING HEART OF AN EMBRYO OF AN ELASMOBRANCH (Pristiurus).

_al._ alimentary tract; _sp._ splanchnic mesoblast; _so._ somatic mesoblast; _ht._ heart.]

[FIG. 357. TRANSVERSE SECTION THROUGH THE POSTERIOR PART OF THE HEAD OF AN EMBRYO CHICK OF THIRTY HOURS.

_hb._ hind-brain; _vg._ vagus nerve; _ep._ epiblast; _ch._ notochord; _x._ thickening of hypoblast (possibly a rudiment of the subnotochordal rod); _al._ throat; _ht._ heart; _pp._ body cavity; _so._ somatic mesoblast; _sf._ splanchnic mesoblast; _hy._ hypoblast.]

[FIG. 358. TRANSVERSE SECTION THROUGH THE HEAD OF A RABBIT OF THE SAME AGE AS FIG. 144 B. (From Kölliker.)

B is a more highly magnified representation of part of A.

_rf._ medullary groove; _mp._ medullary plate; _rw._ medullary fold; _h._ epiblast; _dd._ hypoblast; _dd´._ notochordal thickening of hypoblast; _sp._ undivided mesoblast; _hp._ somatic mesoblast; _dfp._ splanchnic mesoblast; _ph._ pericardial section of body cavity; _ahh._ muscular wall of heart; _ihh._ epithelioid layer of heart; _mes._ lateral undivided mesoblast; _sw._ part of the hypoblast which will form the ventral wall of the pharynx.]

The above description applies only to the development of the heart in those types in which it is formed at a period _after_ the throat has become a closed tube (Elasmobranchii, Amphibia, Cyclostomata, Ganoids (?)). In a number of other cases, in which the heart is formed before the conversion of the throat into a closed tube, of which the most notable is that of Mammals (Hensen, Götte, Kölliker), the heart arises as two independent tubes (fig. 358), which eventually coalesce into an unpaired structure.

In Mammals the two tubes out of which the heart is formed appear at the sides of the cephalic plates, opposite the region of the mid- and hind-brain (fig. 358). They arise at a time when the lateral folds which form the ventral wall of the throat are only just becoming visible. Each half of the heart originates in the same way as the whole heart in Elasmobranchii, etc.; and the layer of the splanchnic mesoblast, which forms the muscular wall for each part (_ahh_), has at first the form of a half tube open below to the hypoblast.

[FIG. 359. TWO DIAGRAMMATIC SECTIONS THROUGH THE REGION OF THE HIND-BRAIN OF AN EMBRYO CHICK OF ABOUT 36 HOURS ILLUSTRATING THE FORMATION OF THE HEART.

_hb._ hind-brain; _nc._ notochord; _E._ epiblast; _so._ somatopleure; _sp._ splanchnopleure; _d._ alimentary tract; _hy._ hypoblast; _hz._ heart; _of._ vitelline veins.]

On the formation of the lateral folds of the splanchnic walls, the two halves of the heart become carried inwards and downwards, and eventually meet on the ventral side of the throat. For a short time they here remain distinct, but soon coalesce into a single tube.

In Birds, as in Mammals, the heart makes its appearance as two tubes, but arises at a period when the formation of the throat is very much more advanced than in the case of Mammals. The heart arises immediately behind the point up to which the ventral wall of the throat is established and thus has at first a Lambda-shaped form. At the apex of the Lambda, which forms the anterior end of the heart, the two halves are in contact (fig. 357), though they have not coalesced; while behind they diverge to be continued as the vitelline veins. As the folding in of the throat is continued backwards the two limbs of the heart are brought together and soon coalesce from before backwards into a single structure. Fig. 359 A and B shews the heart during this process. The two halves have coalesced anteriorly (A) but are still widely separated behind (B). In Teleostei the heart is formed as in Birds and Mammals by the coalescence of two tubes, and it arises before the formation of the throat.

The fact that the heart arises in so many instances as a double tube might lead to the supposition that the ancestral Vertebrate had two tubes in the place of the present unpaired heart.

The following considerations appear to me to prove that this conclusion cannot be accepted. If the folding in of the splanchnopleure to form the throat were deferred relatively to the formation of the heart, it is clear that a modification in the development of the heart would occur, in that the two halves of the heart would necessarily be formed widely apart, and only eventually united on the folding in of the wall of the throat. It is therefore possible to explain the double formation of the heart without having recourse to the above hypothesis of an ancestral Vertebrate with two hearts. If the explanation just suggested is the true one the heart should only be formed as two tubes when it arises prior to the formation of the throat, and as a single tube when formed after the formation of the throat. Since this is invariably found to be so, it may be safely concluded _that the formation of the heart as two cavities is a secondary mode of development, which has been brought about by variations in the period of the closing in of the wall of the throat_.

The heart arises continuously with the sinus venosus, which in the Amniotic Vertebrata is directly continued into the vitelline veins. Though at first it ends blindly in front, it is very soon connected with the foremost aortic arches.

The simple tubular heart, connected as above described, grows more rapidly than the chamber in which it is contained, and is soon doubled upon itself, acquiring in this way an S-shaped curvature, the posterior portion being placed dorsally, and the anterior ventrally. A constriction soon appears between the dorsal and ventral portions.

The dorsal section becomes partially divided off behind from the sinus venosus, and constitutes the relatively thin-walled auricular section of the heart; while the ventral portion, after becoming distinct anteriorly from a portion continued forwards from it to the origin of the branchial arteries, which may be called the truncus arteriosus, acquires very thick spongy muscular walls, and becomes the ventricular division of the heart.

The further changes in the heart are but slight in the case of the Pisces. A pair of simple membranous valves becomes established at the auriculo-ventricular orifice, and further changes take place in the truncus arteriosus. This part becomes divided in Elasmobranchii, Ganoidei, and Dipnoi into a posterior section, called the conus arteriosus, provided with a series of transverse rows of valves, and an anterior section, called the bulbus arteriosus, not provided with valves, and leading into the branchial arteries. In most Teleostei (except Butirinus and a few other forms) the conus arteriosus is all but obliterated, and the anterior row of its valves alone preserved; and the bulbus is very much enlarged[222].

[222] _Vide_ Gegenbaur, "Zur vergleich. Anat. d. Herzens." _Jenaische Zeit._, Vol. II. 1866, and for recent important observations, J. E. V. Boas, "Ueb. Herz u. Arterienbogen bei Ceratoden u. Protopterus," and "Ueber d. Conus arter. b. Butirinus, etc.," _Morphol. Jahrb._, Vol. VI. 1880.

In the Dipnoi important changes in the heart are effected, as compared with other Fishes, by the development of true lungs. Both the auricular and ventricular chamber may be imperfectly divided into two, and in the conus a partial longitudinal septum is developed in connection with a longitudinal row of valves[223].

[223] Boas holds that the longitudinal septum is formed by the coalescence of a row of longitudinal valves, but this is opposed to Lankester's statements, "On the hearts of Ceratodus, Protopterus and Chimæra, etc." _Zool. Trans._ Vol. X. 1879.

In Amphibia the heart is in many respects similar to that of the Dipnoi. Its curvature is rather that of a screw than of a simple S. The truncus arteriosus lies to the left, and is continued into the ventricle which lies ventrally and more to the right, and this again into the dorsally placed auricular section.

After the heart has reached the piscine stage, the auricular section (Bombinator) becomes prolonged into a right and left auricular appendage. A septum next grows from the roof of the auricular portion of the heart obliquely backwards and towards the left, and divides it in two chambers; the right one of which remains continuous with the sinus venosus, while the left one is completely shut off from the sinus, though it soon enters into communication with the newly established pulmonary veins. The truncus arteriosus[224] is divided into a posterior _conus arteriosus_ (pylangium) and an anterior _bulbus_ (synangium). The former is provided with a proximal row of valves at its ventricular end, and a distal row at its anterior end near the bulbus. It is also provided with a longitudinal septum, which is no doubt homologous with the septum in the conus arteriosus of the Dipnoi. The bulbus is well developed in many Urodela, but hardly exists in the Anura.

[224] For a good description of the adult heart _vide_ Huxley, Article "Amphibia," in the _Encyclopædia Britannica_.

In the Amniota further changes take place in the heart, resulting in the abortion of the distal rows of valves of the conus arteriosus[225], and in the splitting up of the whole truncus arteriosus into three vessels in Reptilia, and two in Birds and Mammals, each opening into the ventricular section of the heart, and provided with a special set of valves at its commencement. In Birds and Mammals the ventricle becomes moreover completely divided into two chambers, each communicating with one of the divisions of the primitive truncus, known in the higher types as the systemic and pulmonary aortæ. The character of the development of the heart in the Amniota will be best understood from a description of what takes place in the Chick.

[225] It is just possible that the reverse may be true, _vide_ note on p. 640. If however, as is most probable, the statement in the text is correct, the valves at the mouth of the ventricle in Teleostei are not homologous with those of the Amniota; the former being the distal row of the valves of the conus, the latter the proximal.

In Birds the originally straight heart (fig. 109) soon becomes doubled up upon itself. The ventricular portion becomes placed on the ventral and right side, while the auricular section is dorsal and to the left. The two parts are separated from each other by a slight constriction known as the canalis auricularis. Anteriorly the ventricular cavity is continued into the truncus, and the venous or auricular portion of the heart is similarly connected behind with the sinus venosus. The auricular appendages grow out from the auricle at a very early period. The general appearance of the heart, as seen from the ventral side on the fourth day, is shewn in fig. 360. Although the external divisions of the heart are well marked even before this stage, it is not till the end of the third day that the internal partitions become apparent; and, contrary to what might have been anticipated from the evolution of these parts in the lower types, the ventricular septum is the first to be established.

It commences on the third day as a crescentic ridge or fold springing from the convex or ventral side of the rounded ventricular portion of the heart, and on the fourth day grows rapidly across the ventricular cavity towards the concave or dorsal side. It thus forms an incomplete longitudinal partition, extending from the canalis auricularis to the commencement of the truncus arteriosus, and dividing the twisted ventricular tube into two somewhat curved canals, one more to the left and above, the other to the right and below. These communicate with each other, above the free edge of the partition, along its whole length.

[FIG. 360. HEART OF A CHICK ON THE FOURTH DAY OF INCUBATION VIEWED FROM THE VENTRAL SURFACE.

_l.a._ left auricular appendage; _C.A._ canalis auricularis; _v._ ventricle; _b._ truncus arteriosus.]

Externally the ventricular portion as yet shews no division into two parts.

By the fifth day the venous end of the heart, though still lying somewhat to the left and above, is placed as far forwards as the arterial end, the whole organ appearing to be drawn together. The ventricular septum is complete.

The apex of the ventricles becomes more and more pointed. In the auricular portion a small longitudinal fold appears as the rudiment of the auricular septum, while in the canalis auricularis, which is now at its greatest length, there is also to be seen a commencement of the valvular structures tending to separate the cavity of the auricles from those of the ventricles.

About the 106th hour, a septum begins to make its appearance in the truncus arteriosus in the form of a longitudinal fold, which according to Tonge (No. 495) starts at the end of the truncus furthest removed from the heart. It takes origin from the wall of the truncus between the fourth and fifth pairs of arches, and grows downwards in such a manner as to divide the truncus into two channels, one of which leads from the heart to the third and fourth pairs of arches, and the other to the fifth pair. Its course downwards is not straight but spiral, and thus the two channels into which it divides the truncus arteriosus wind spirally the one round the other.

At the time when the septum is first formed, the opening of the truncus arteriosus into the ventricles is narrow or slit-like, apparently in order to prevent the flow of the blood back into the heart. Soon after the appearance of the septum, however, semilunar valves (Tonge, No. 495) are developed from the wall of that portion of the truncus which lies between the free edge of the septum and the cavity of the ventricles[226].

[226] If Tonge is correct in his statement that the semilunar valves develop at some distance from the mouth of the ventricle, it would seem possible that the portion of the truncus between them and the ventricle ought to be regarded as the embryonic conus arteriosus, and that the distal row of valves of the conus (and not the proximal as suggested above, p. 639) has been preserved in the higher types.

The ventral and the dorsal pairs of valves are the first to appear: the former as two small solid prominences separated from each other by a narrow groove; the latter as a single ridge, in the centre of which is a prominence indicating the point where the ridge will subsequently become divided into two. The outer valves appear opposite each other, at a considerably later period.

[FIG. 361. TWO VIEWS OF THE HEART OF A CHICK UPON THE FIFTH DAY OF INCUBATION.

A. from the ventral, B. from the dorsal side.

_l.a._ left auricular appendage; _r.a._ right auricular appendage; _r.v._ right ventricle; _l.v._ left ventricle; _b._ truncus arteriosus.]

As the septum grows downwards towards the heart, it finally reaches the position of these valves. One of its edges then passes between the two ventral valves, and the other unites with the prominence on the dorsal valve-ridge. At the same time the growth of all the parts causes the valves to appear to approach the heart, and thus to be placed quite at the top of the ventricular cavities. The free edge of the septum of the truncus now fuses with the ventricular septum, and thus the division of the truncus into two separate channels, each provided with three valves, and each communicating with a separate side of the heart, is complete; the position of the valves not being very different from that in the adult heart.

That division of the truncus which opens into the fifth pair of arches is the one which communicates with the right ventricle, while that which opens into the third and fourth pairs communicates with the left ventricle. The former becomes the pulmonary artery, the latter the commencement of the systemic aorta.

The external constriction actually dividing the truncus into two vessels does not begin to appear till the septum has extended some way back towards the heart.

The semilunar valves become pocketed at a period considerably later than their first formation (from the 147th to the 165th hour) in the order of their appearance.

At the end of the sixth day, and even on the fifth day (figs. 361 and 362), the appearance of the heart itself, without reference to the vessels which come from it, is not very dissimilar from that of the adult. The original protuberance to the right now forms the apex of the ventricles, and the two auricular appendages are placed at the anterior extremity of the heart. The most noticeable difference (in the ventral view) is the still externally undivided condition of the truncus arteriosus.

[FIG. 362. HEART OF A CHICK UPON THE SIXTH DAY OF INCUBATION, FROM THE VENTRAL SURFACE.

_l.a._ left auricular appendage; _r.a._ right auricular appendage; _r.v._ right ventricle; _l.v._ left ventricle; _b._ truncus arteriosus.]

The subsequent changes which the heart undergoes are concerned more with its internal structure than with its external shape. Indeed, during the next three days, viz. the eighth, ninth, and tenth, the external form of the heart remains nearly unaltered.

In the auricular portion, however, the septum which commenced on the fifth day becomes now more conspicuous. It is placed vertically, and arises from the ventral wall; commencing at the canalis auricularis and proceeding towards the opening into the sinus venosus.

This latter structure gradually becomes reduced so as to become a special appendage of the right auricle. The inferior vena cava enters the sinus obliquely from the right, so that its blood has a tendency to flow towards the left auricle of the heart, which is at this time the larger of the two.

The valves between the ventricles and auricles are now well developed, and it is about this time that the division of the truncus arteriosus into the aorta and pulmonary artery becomes visible from the exterior.

By the eleventh to the thirteenth day the right auricle has become as large as the left, and the auricular septum much more complete, though there is still a small opening, the _foramen ovale_, by which the two cavities communicate with each other.

The most important feature in which the development of the Reptilian heart differs from that of Birds is the division of the truncus into three vessels, instead of two. The three vessels remain bound up in a common sheath, and appear externally as a single trunk. The vessel not represented in Birds is that which is continued into the left aortic arch.

In Mammals the early stages in the development of the heart present no important points of difference from those of Aves. The septa in the truncus, in the ventricular, and in the auricular cavities are formed, so far as is known, in the same way and at the same relative periods in both groups. In the embryo Man, the Rabbit, and other Mammals the division of the ventricles is made apparent externally by a deep cleft, which, though evanescent in these forms, is permanent in the Dugong.

The attachment of the auriculo-ventricular valves to the wall of the ventricle, and the similar attachment of the left auriculo-ventricular valves in Birds, have been especially studied by Gegenbaur and Bernays (No. 492), and deserve to be noticed. In the primitive state the ventricular walls have throughout a spongy character; and the auriculo-ventricular valves are simple membranous projections like the auriculo-ventricular valves of Fishes. Soon however the spongy muscular tissue of both the ventricular and auricular walls, which at first pass uninterruptedly the one into the other, grows into the bases of the valves, which thus become in the main muscular projections of the walls of the heart. As the wall of the ventricle thickens, the muscular trabeculæ, connected at one end with the valves, remain at the other end united with the ventricular wall, and form special bands passing between the two. The valves on the other hand lose their muscular attachment to the auricular walls. This is the condition permanent in Ornithorhynchus. In higher Mammalia the ends of the muscular bands inserted into the valves become fibrous, from the development of intermuscular connective tissue, and the atrophy of the muscular elements. The fibrous parts now form the chordæ tendineæ, and the muscular the musculi papillares.

The sinus venosus in Mammals becomes completely merged into the right auricle, and the systemic division of the truncus arteriosus is apparently not homologous with that in Birds.

In the embryos of all the Craniata the heart is situated very far forwards in the region of the head. This position is retained in Pisces. In Amphibia the heart is moved further back, while in all the Amniota it gradually shifts its position first of all into the region of the neck and finally passes completely within the thoracic cavity. The steps in the change of position may be gathered from figs. 109, 111, and 118.

BIBLIOGRAPHY _of the Heart_.

(492) A. C. Bernays. "Entwicklungsgeschichte d. Atrioventricularklappen." _Morphol. Jahrbuch_, Vol. II. 1876.

(493) E. Gasser. "Ueber d. Entstehung d. Herzens beim Hühn." _Archiv f. mikr. Anat._, Vol. XIV.

(494) A. Thomson. "On the development of the vascular system of the foetus of Vertebrated Animals." _Edinb. New Phil. Journal_, Vol. IX. 1830 and 1831.

(495) M. Tonge. "Observations on the development of the semilunar valves of the aorta and pulmonary artery of the heart of the Chick." _Phil. Trans._ CLIX. 1869.

_Vide_ also Von Baer (291), Rathke (300), Hensen (182), Kölliker (298), Götte (296), and Balfour (292).

_Arterial System._

In the embryos of Vertebrata the arterial system consists of a forward continuation of the truncus arteriosus, on the ventral side of the throat (figs. 363, _abr_, and 364, _a_), which, with a few exceptions to be noticed below, divides into as many branches on each side as there are visceral arches. These branches, after traversing the visceral arches, unite on the dorsal side of the throat into a common trunk on each side. This trunk (figs. 363 and 364) after giving off one (or more) vessels to the head (_c´_ and _c_) turns backwards, and bends in towards the middle line, close to its fellow, immediately below the notochord (figs. 21 and 116) and runs backwards in this situation towards the end of the tail. The two parallel trunks below the notochord fuse very early into a single trunk, the dorsal aorta (figs. 363, _ad_, and 364, _a´´_). There is given off from each collecting trunk from the visceral arches, or from the commencement of the dorsal aorta, a subclavian artery to each of the anterior limbs; from near the anterior end of the dorsal aorta a vitelline artery (or before the dorsal aortæ have united a pair of arteries fig. 125, R _of_ A and L _of_ A) to the yolk-sack, which subsequently becomes the main visceral artery[227]; and from the dorsal aorta opposite the hind limbs one (or two) arteries on each side--the iliac arteries--to the hind limbs; from these arteries the allantoic arteries are given off in the higher types, which remain as the hypogastric arteries after the disappearance of the allantois.

[227] In Mammalia the superior mesenteric artery arises from the vitelline artery, which may probably be regarded as a primitive cæliaco-mesenteric artery.

[FIG. 363. DIAGRAMMATIC VIEW OF THE HEAD OF AN EMBRYO TELEOSTEAN, WITH THE PRIMITIVE VASCULAR TRUNKS. (From Gegenbaur.)

_a._ auricle; _v._ ventricle; _abr._ branchial artery; _c´._ carotid; _ad._ dorsal aorta; _s._ branchial clefts; _sv._ sinus venosus; _dc._ ductus Cuvieri; _n._ nasal pit.]

The primitive arrangement of the arterial trunks is with a few modifications retained in Fishes. With the development of the gills the vessels to the arches become divided into two parts connected by a capillary system in the gill folds, viz. into the branchial arteries bringing the blood to the gills from the truncus arteriosus, and the branchial veins transporting it to the dorsal aorta. The branchial vessels to those arches which do not bear gills, either wholly or partially atrophy; thus in Elasmobranchii the mandibular trunk, which is fully developed in the embryo (fig. 193, 1_av_), atrophies, except for a small remnant bringing blood to the rudimentary gill of the spiracle from the branchial vein of the hyoid arch. In Ganoids the mandibular artery atrophies, but the hyoid is usually preserved. In Teleostei both mandibular[228] and hyoid arteries are absent in the adult, except that there is usually left a rudiment of the hyoid, supplying the pseudobranch, which is similar to the rudiment of the mandibular artery in Elasmobranchii. In Dipnoi the mandibular artery atrophies, but the hyoid is sometimes preserved (Protopterus), and sometimes lost.

[228] The mandibular artery is stated by Götte never to be developed in Teleostei, but is distinctly figured in Lereboullet (No. 71).

In Fishes provided with a well developed air-bladder this organ receives arteries, which arise sometimes from the dorsal aorta, sometimes from the cæliac arteries, and sometimes from the dorsal section of the last (fourth) branchial trunk. The latter origin is found in Polypterus and Amia, and seems to have been inherited by the Dipnoi where the air-bladder forms a true lung.

The pulmonary artery of all the air-breathing Vertebrata is derived from the pulmonary artery of the Dipnoi.

In all the types above Fishes considerable changes are effected in the primitive arrangement of the arteries in the visceral arches.

In Amphibia the piscine condition is most nearly retained[229]. The mandibular artery is never developed, and the hyoid artery is imperfect, being only connected with the cephalic vessels and never directly joining the dorsal aorta. It is moreover developed later than the arteries of the true branchial arches behind. The subclavian arteries spring from the common trunks which unite to form the dorsal aorta.

[229] In my account of the Amphibia, Götte (No. 296) has been followed.

In the Urodela there are developed, in addition to the hyoid, four branchial arteries. The three foremost of these at first supply gills, and in the Perennibranchiate forms continue to do so through life. The fourth does not supply a gill, and very early gives off, as in the Dipnoi, a pulmonary branch.

The hyoid artery soon sends forward a lingual artery from its ventral end, and is at first continued to the carotid which grows forward from the dorsal part of the first branchial vessel.

In the Caducibranchiata, where the gills atrophy, the following changes take place. The remnant of the hyoid is continued entirely into the lingual artery. The first branchial is mainly continued into the carotid and other cephalic branches, but a narrow remnant of the trunk, which originally connected it with the dorsal aorta, remains, forming what is known as a ductus Botalli. A rete mirabile on its course is the remnant of the original gill.

The second and third branchial arches are continued as simple trunks into the dorsal aorta, and the blood from the fourth arch mainly passes to the lungs, but a narrow ductus Botalli still connects this arch with the dorsal aorta.

In the Anura the same number of arches is present in the embryo as in the Urodela, all four branchial arteries supplying branchiæ, but the arrangement of the two posterior trunks is different from that in the Urodela. The third arch becomes at a very early period continued into a pulmonary vessel, a relatively narrow branch connecting it with the second arch. The fourth arch joins the pulmonary branch of the third. At the metamorphosis the hyoid artery loses its connection with the carotid, and the only part of it which persists is the root of the lingual artery. The first branchial artery ceases to join the dorsal aorta, and forms the root of the carotid: the so-called carotid gland placed on its course is the remnant of the gill supplied by it before the metamorphosis.

The second artery forms a root of the dorsal aorta. The third, as in all the Amniota, now supplies the lungs, and also sends off a cutaneous branch. The fourth disappears. The connection of the pulmonary artery with both the third and fourth branchial arches in the embryo appears to me clearly to indicate that this artery was primitively derived from the _fourth arch_ as in the Urodela, and that its permanent connection with the third arch in the Anura and in all the Amniota is secondary.

[FIG. 364. DIAGRAM OF THE ARRANGEMENT OF THE ARTERIAL ARCHES IN AN EMBRYO OF ONE OF THE AMNIOTA. (From Gegenbaur; after RATHKE.)

_a._ ventral aorta; _a´´._ dorsal aorta; 1, 2, 3, 4, 5. arterial arches; _c._ carotid artery.]

In the Amniota the metamorphosis of the arteries is in all cases very similar. Five arches, viz. the mandibular, hyoid, and three branchial arches are always developed (fig. 364), but, owing to the absence of branchiæ, never function as branchial arteries. Of these the main parts of the first two, connecting the truncus arteriosus with the collecting trunk into which the arterial arches fall, always disappear, usually before the complete development of the arteries in the posterior arches.

The anterior part of the collecting trunk into which these vessels fall is not obliterated when they disappear, but is on the contrary continued forwards as a vessel supplying the brain, homologous with that found in Fishes. It constitutes the internal carotid. Similarly the anterior part of the trunk from which the mandibular and hyoid arteries sprang is continued forwards as a small vessel[230], which at first passes to the oral region and constitutes in Reptiles the lingual artery, homologous with the lingual artery of the Amphibia; but in Birds and Mammals becomes more important, and is then known as the external carotid (fig. 125). By these changes the roots of the external and internal carotids spring respectively from the ventral and dorsal ends of the primitive third artery, _i.e._ the artery of the first branchial arch (fig. 365, _c_ and _c´_); and thus this arterial arch _persists in all types_ as the common carotid, and the basal part of the internal carotid. The trunk connecting the third arterial arch with the system of the dorsal aorta persists in some Reptiles (Lacertilia, fig. 366 A) as a ductus Botalli, but is lost in the remaining Reptiles and in Birds and Mammals (fig. 366 B, C, D). It disappears earliest in Mammals (fig. 365 C), later in Birds (fig. 365 B), and still later in the majority of Reptiles.

[230] His (No. 232) describes in Man two ventral continuations of the truncus arteriosus, one derived from the mandibular artery, forming the external maxillary artery, and one from the hyoid artery, forming the lingual artery. The vessel from which they spring is the external carotid. These observations of His will very probably be found to hold true for other types.

The fourth arch always continues to give rise, as in the Anura, to the system of the dorsal aorta.

[FIG. 365. DEVELOPMENT OF THE GREAT ARTERIAL TRUNKS IN THE EMBRYOS OF A. A LIZARD; B. THE COMMON FOWL; C. THE PIG. (From Gegenbaur; after Rathke.)

The first two arches have disappeared in all three. In A and B the last three are still complete, but in C the last two are alone complete.

_p._ pulmonary artery springing from the fifth arch, but still connected with the system of the dorsal aorta by a ductus Botalli; _c._ external carotid; _c´._ internal carotid; _ad._ dorsal aorta; _a._ auricle; _v._ ventricle; _n._ nasal pit; _m._ rudiment of fore-limb.]

In all Reptiles it persists on both sides (fig. 366 A and B), but with the division of the truncus arteriosus into three vessels one of these, _i.e._ that opening furthest to the left side of the ventricle (_e_ and _d_), is continuous with the _right_ fourth arch, and also with the common carotid arteries (_c_); while a second springing from the right side of the ventricle is continuous with the _left_ fourth arch (_h_ and _f_). The right and left divisions of the fourth arch meet however on the dorsal side of the oesophagus to give origin to the dorsal aorta (_g_).

In Birds (fig. 366 C) the _left_ fourth arch (_h_) loses its connection with the dorsal aorta, though the ventral part remains as the root of the left subclavian. The truncus arteriosus is moreover only divided into two parts, one of which is continuous with all the systemic arteries. Thus it comes about that in Birds the right fourth arch (_e_) alone gives rise to the dorsal aorta.

In Mammals (fig. 366 D) the truncus arteriosus is only divided into two, but the _left fourth arch_ (_e_), instead of the right, is that continuous with the dorsal aorta, and the right fourth arch (_i_) is only continued into the right vertebral and right subclavian arteries.

The fifth arch always gives origin to the pulmonary artery (fig. 365, _p_) and is continuous with one of the divisions of the truncus arteriosus. In Lizards (fig. 366 A, _i_), Chelonians and Birds (fig. 366 C, _i_) and probably in Crocodilia, the right and left pulmonary arteries spring respectively from the right and left fifth arches, and during the greater part of embryonic life the parts of the fifth arches between the origins of the pulmonary arteries and the system of the dorsal aorta are preserved as ductus Botalli. These ductus Botalli persist for life in the Chelonia. In Ophidia (fig. 366 B, _h_) and Mammalia (fig. 366 D, _m_) only one of the fifth arches gives origin to the two pulmonary arteries, viz. that on the right side in Ophidia, and the left in Mammalia.

The ductus Botalli of the fifth arch (known in Man as the ductus arteriosus) of the side on which the pulmonary arteries are formed, may remain (_e.g._ in Man) as a solid cord connecting the common stern of the pulmonary aorta with the systemic aorta.

The main history of the arterial arches in the Amniota has been sufficiently dealt with, and the diagram, fig. 366, copied from Rathke, shews at a glance the character of the metamorphosis these arches undergo in the different types. It merely remains for me to say a few words about the subclavian and vertebral arteries.

The subclavian arteries in Fishes usually spring from the trunks connecting the branchial veins with the dorsal aorta. This origin, which is also found in Amphibia, is typically found in the embryos of the Amniota. In the Lizards this origin persists through life, but both subclavians spring from the right side. In most other types the origin of the subclavians is carried upwards, so that they usually spring from a trunk common to them and the carotids (arteria anonyma) (Birds and some Mammals); or the left one, as in Man and some other Mammals, arises from the systemic aorta just beyond the carotids. Various further modifications in the origin of the subclavians of the same general nature are found in Mammalia, but they need not be specified in detail. The vertebral arteries usually arise in close connection with the subclavians, but in Birds they arise from the common carotids.

[FIG. 366. DIAGRAMS ILLUSTRATING THE METAMORPHOSIS OF THE ARTERIAL ARCHES IN A LIZARD A, A SNAKE B, A BIRD C AND A MAMMAL D. (From Mivart; after Rathke.)

A. _a._ internal carotid; _b._ external carotid; _c._ common carotid; _d._ ductus Botalli between the third and fourth arches; _e._ right aortic trunk; _f._ subclavian; _g._ dorsal aorta; _h._ left aortic trunk; _i._ pulmonary artery; _k._ rudiment of ductus Botalli between the pulmonary artery and the system of the dorsal aorta.

B. _a._ internal carotid; _b._ external carotid; _c._ common carotid; _d._ right aortic trunk; _e._ vertebral artery; _f._ left aortic trunk of dorsal aorta; _h._ pulmonary artery; _i._ ductus Botalli of pulmonary artery.

C. _a._ internal carotid; _b._ external carotid; _c._ common carotid; _d._ systemic aorta; _e._ fourth arch of right side (root of dorsal aorta); _f._ right subclavian; _g._ dorsal aorta; _h._ left subclavian (fourth arch of left side); _i._ pulmonary artery; _k._ and _l._ right and left ductus Botalli of pulmonary arteries.

D. _a._ internal carotid; _b._ external carotid; _c._ common carotid; _d._ systemic aorta; _e._ fourth arch of left side (root of dorsal aorta); _f._ dorsal aorta; _g._ left vertebral artery; _h._ left subclavian artery; _i._ right subclavian (fourth arch of right side); _k._ right vertebral; _l._ continuation of right subclavian; _m._ pulmonary artery; _n._ ductus Botalli of pulmonary artery.]

BIBLIOGRAPHY _of the Arterial System_.

(496) H. Rathke. "Ueb. d. Entwick. d. Arterien w. bei d. Säugethiere von d. Bogen d. Aorta ausgehen." Müller's _Archiv_, 1843.

(497) H. Rathke. "Untersuchungen üb. d. Aortenwurzeln d. Saurier." _Denkschriften d. k. Akad. Wien_, Vol. XIII. 1857.

_Vide_ also His (No. 232) and general works on Vertebrate Embryology.

_The Venous System._

The venous system, as it is found in the embryos of Fishes, consists in its earliest condition of a single large trunk, which traverses the splanchnic mesoblast investing the part of the alimentary tract behind the heart. This trunk is directly continuous in front with the heart, and underlies the alimentary canal through both its præanal and postanal sections. It is shown in section in fig. 367, _v_, and may be called the subintestinal vein. This vein has been found in the embryos of Teleostei, Ganoidei, Elasmobranchii and Cyclostomata, and runs parallel to the dorsal aorta above, into which it is sometimes continued behind (Teleostei, Ganoidei, etc.).

In Elasmobranch embryos the subintestinal vein terminates, as may be gathered from sections (fig. 368, _v.cau_), shortly before the end of the tail. The same series of sections also shews that at the cloaca, where the gut enlarges and comes in contact with the skin, this vein bifurcates, the two branches uniting into a single vein both in front of and behind the cloaca.

In most Fishes the anterior part of this vein atrophies, the caudal section alone remaining, but the anterior section of it persists in the fold of the intestine in Petromyzon, and also remains in the spiral valve of some Elasmobranchii. In Amphioxus, moreover, it forms, as in the embryos of higher types, the main venous trunk, though even here it is usually broken up into two or three parallel vessels.

It no doubt represents one of the primitive longitudinal trunks of the vermiform ancestors of the Chordata. The heart and the branchial artery constitute a specially modified anterior continuation of this vein. The dilated portal sinus of Myxine is probably also part of it; and if this is really rhythmically contractile[231] the fact would be interesting as shewing that this quality, which is now localised in the heart, was once probably common to the subintestinal vessel for its whole length.

[231] J. Müller holds that this sack is not rhythmically contractile.

[FIG. 367. SECTION THROUGH THE TRUNK OF A SCYLLIUM EMBRYO SLIGHTLY YOUNGER THAN 28 F.

_sp.c._ spinal canal; _W._ white matter of spinal cord; _pr._ posterior nerve-roots; _ch._ notochord; _x._ subnotochordal rod; _ao._ aorta; _mp._ muscle plate; _mp´._ inner layer of muscle-plate already converted into muscles; _Vr._ rudiment of vertebral body; _st._ segmental tube; _sd._ segmental duct; _sp.v._ spiral valve; _v._ subintestinal vein; _p.o._ primitive generative cells.]

On the development of the cardinal veins (to be described below) considerable changes are effected in the subintestinal vein. Its postanal section, which is known in the adult as the caudal vein, unites with the cardinal veins. On this junction being effected retrogressive changes take place in the præanal section of the original subintestinal vessel. It breaks up in front into a number of smaller vessels, the most important of which is a special vein, which lies in the fold of the spiral valve, and which is more conspicuous in some Elasmobranchii than in Scyllium, in which the development of the vessel has been mainly studied. The lesser of the two branches connecting it round the cloaca with the caudal vein first vanishes, and then the larger; and the two posterior cardinals are left as the sole forward continuations of the caudal vein. The latter then becomes prolonged forwards, so that the two cardinals open into it some little distance in front of the hind end of the kidneys. By these changes, and by the disappearance of the postanal section of the gut, the caudal vein is made to appear as a supraintestinal and not, as it really is, a _subintestinal vessel_.

From the subintestinal vein there is given off a branch which supplies the yolk-sack. This leaves the subintestinal vein close to the liver. The liver, on its development, embraces the subintestinal vein, which then breaks up into a capillary system in the liver, the main part of its blood coming at this period from the yolk-sack.

The portal system is thus established from the subintestinal vein; but is eventually joined by the various visceral, and sometimes by the genital, veins as they become successively developed.

The blood from the liver is brought back to the sinus venosus by veins known as the hepatic veins, which, like the hepatic capillary system, are derivatives of the subintestinal vessel.

There join the portal system in Myxinoids and many Teleostei a number of veins from the anterior abdominal walls, representing a commencement of the anterior abdominal or epigastric vein of higher types[232].

[232] Stannius, _Vergleich. Anat._, p. 251.

[FIG. 368. FOUR SECTIONS THROUGH THE POSTANAL PART OF THE TAIL OF AN EMBRYO OF THE SAME AGE AS FIG. 28 F.

A. is the posterior section.

_nc._ neural canal; _al._ postanal gut; _alv._ caudal vesicle of postanal gut; _x._ subnotochordal rod; _mp._ muscle-plate; _ch._ notochord; _cl.al._ cloaca; _ao._ aorta; _v.cau._ caudal vein.]

In the higher Vertebrates the original subintestinal vessel never attains a full development, even in the embryo. It is represented by (1) the ductus venosus, which, like the true subintestinal vein, gives origin (in the Amniota) to the vitelline veins to the yolk-sack, and (2) by the caudal vein. Whether the partial atrophy of the subintestinal vessel was primitively caused by the development of the cardinal veins, or for some other reason, it is at any rate a fact that in all existing Fishes the cardinal veins form the main venous channels of the trunk.

Their later development than the subintestinal vessel as well as their absence in Amphioxus, probably indicate that they became evolved, at any rate in their present form, within the Vertebrate phylum.

The embryonic condition of the venous system, with a single large subintestinal vein is, as has been stated, always modified by the development of a paired system of vessels, known as the cardinal veins, which bring to the heart the greater part of the blood from the trunk.

[FIG. 369. DIAGRAM OF THE PAIRED VENOUS SYSTEM OF A FISH. (From Gegenbaur.)

_j._ jugular vein (anterior cardinal vein); _c._ posterior cardinal vein; _h._ hepatic veins; _sv._ sinus venosus; _dc._ ductus Cuvieri.]

The cardinal veins appear in Fishes as four paired longitudinal trunks (figs. 363 and 369), two anterior (_j_) and two posterior (_c_). They unite into two transverse trunks on either side, known as the ductus Cuvieri (_dc_), which fall into the sinus venosus, passing from the body wall to the sinus by a lateral mesentery of the heart already spoken of (p. 627, fig. 352). The anterior pair, known as the anterior cardinal or jugular veins, bring to the heart the blood from the head and neck. They are placed one on each side above the level of the branchial arches (fig. 299, _a.cv_). The posterior cardinal veins lie immediately dorsal to the mesonephros (Wolffian body), and are mainly supplied by the blood from this organ and from the walls of the body (fig. 275, _c.a.v_). In many forms (Cyclostomata, Elasmobranchii and many Teleostei) they unite posteriorly with the caudal veins in the manner already described, and in a large number of instances the connecting branch between the two systems, in its passage through the mesonephros, breaks up into a capillary network, and so gives rise to a renal portal system.

The vein from the anterior pair of fins (subclavian) usually unites with the anterior jugular vein.

The venous system of the Amphibia and Amniota always differs from that of Fishes in the presence of a new vessel, the vena cava inferior, which replaces the posterior cardinal veins; the latter only being present, in their piscine form, during embryonic life. It further differs from that of all Fishes, except the Dipnoi, in the presence of pulmonary veins bringing back the blood directly from the lungs.

In the embryos of all the higher forms the general characters of the venous system are at first the same as in Fishes, but with the development of the vena cava inferior the front sections of the posterior cardinal veins atrophy, and the ductus Cuvieri, remaining solely connected with the anterior cardinals and their derivatives, constitute the superior venæ cavæ. The inferior cava receives the hepatic veins.

Apart from the non-development of the subintestinal vein the visceral section of the venous system is very similar to that in Fishes.

The further changes in the venous system must be dealt with separately for each group.

Amphibia. In Amphibia (Götte, No. 296) the anterior and posterior cardinal veins arise as in Pisces. From the former the internal jugular vein arises as a branch; the external jugular constituting the main stem. The subclavian with its large cutaneous branch also springs from the system of the anterior cardinal. The common trunk formed by the junction of these three veins falls into the ductus Cuvieri.

The posterior cardinal veins occupy the same position as in Pisces, and unite behind with the caudal veins, which Götte has shewn to be originally situated below the postanal gut. The iliac veins unite with the posterior cardinal veins, where the latter fall into the caudal vein. The original piscine condition of the veins is not long retained. It is first of all disturbed by the development of the _anterior_ part of the important unpaired venous trunk which forms in the adult the vena cava inferior. This is developed independently, but unites behind with the right posterior cardinal. From this point backwards the two cardinal veins coalesce for some distance, to give rise to the _posterior_ section of the vena cava inferior, situated between the kidneys[233]. The anterior sections of the cardinal veins subsequently atrophy. The posterior part of the cardinal veins, from their junction with the vena cava inferior to the caudal veins, forms a rhomboidal figure. The iliac vein joins the outer angle of this figure, and is thus in direct communication with the inferior vena cava, but it is also connected with a longitudinal vessel on the outer border of the kidneys, which receives transverse vertebral veins and transmits their blood to the kidneys, thus forming a renal portal system. The anterior limbs of the rhomboid formed by the cardinal veins soon atrophy, so that the blood from the hind limbs can only pass to the inferior vena cava through the renal portal system. The posterior parts of the two cardinal veins (uniting in the Urodela directly with the unpaired caudal vein) still persist. The iliac veins also become directly connected with a new vein, the anterior abdominal vein, which has meanwhile become developed. Thus the iliac veins become united with the system of the vena cava inferior through the vena renalis advehens on the outer border of the kidney, and with the anterior abdominal veins by the epigastric veins.

[233] This statement of Götte's is opposed to that of Rathke for the Amniota, and cannot be considered as completely established.

The visceral venous system begins with the development of two vitelline veins, which at first join the sinus venosus directly. They soon become enveloped in the liver, where they break up into a capillary system, which is also joined by the other veins from the viscera. The hepatic system has in fact the same relations as in Fishes. Into this system the anterior abdominal vein also pours itself in the adult. This vein is originally formed of two vessels, which at first fall directly into the sinus venosus, uniting close to their opening into the sinus with a vein from the truncus arteriosus. They become prolonged backwards, and after receiving the epigastric veins above mentioned from the iliac veins, and also veins from the allantoic bladder, unite behind into a single vessel. Anteriorly the right vein atrophies and the left continues forward the unpaired posterior section.

A secondary connection becomes established between the anterior abdominal vein and the portal system; so that the blood originally transported by the former vein to the heart becomes diverted so as to fall into the liver. A remnant of the primitive connection is still retained in the adult in the form of a small vein, the so-called vena bulbi posterior, which brings the blood from the walls of the truncus arteriosus directly into the anterior abdominal vein.

The pulmonary veins grow directly from the heart to the lungs.

For our knowledge of the development of the venous system of the Amniota we are mainly indebted to Rathke.

Reptilia. As an example of the Reptilia the Snake may be selected, its venous system having been fully worked out by Rathke in his important memoir on its development (No. 300).

The anterior (external jugular) and posterior cardinal veins are formed in the embryo as in all other types (fig. 370, _vj_ and _vc_); and the anterior cardinal, after giving rise to the anterior vertebral and to the cephalic veins, persists with but slight modifications in the adult; while the two ductus Cuvieri constitute the superior venæ cavæ.

The two posterior cardinals unite behind with the caudal veins. They are placed in the usual situation on the dorsal and outer border of the kidneys.

With the development of the vena cava inferior, to be described below, the blood from the kidneys becomes mainly transported by this vessel to the heart; and the section of the posterior cardinals opening into the ductus Cuvieri gradually atrophies, their posterior parts remaining however on the outer border of the kidneys as the venæ renales advehentes[234].

[234] Rathke's account of the vena renalis advehens is thus entirely opposed to that which Götte gives for the Frog, but my own observations on the Lizard incline me to accept Rathke's statements, for the Amniota at any rate.

[FIG. 370. ANTERIOR PORTION OF THE VENOUS SYSTEM OF AN EMBRYONIC SNAKE. (From Gegenbaur; after Rathke.)

_vc._ posterior cardinal vein; _vj._ jugular vein; _DC._ ductus Cuvieri; _vu._ allantoic vein; _v._ ventricle; _ba._ truncus arteriosus; _a._ visceral clefts; _l._ auditory vesicle.]

While the front part of the posterior cardinal veins is undergoing atrophy, the intercostal veins, which originally poured their blood into the posterior cardinal veins, become also connected with two longitudinal veins--the posterior vertebral veins--which are homologous with the azygos and hemiazygos veins of Man; and bear the same relation to the anterior vertebral veins that the anterior and posterior cardinals do to each other.

These veins are at first connected by transverse anastomoses with the posterior cardinals, but, on the disappearance of the front part of the latter, the whole of the blood from the intercostal veins falls into the posterior vertebral veins. They are united in front with the anterior vertebral veins, and the common trunk of the two veins on each side falls into the jugular vein.

The posterior vertebral veins are at first symmetrical, but after becoming connected by transverse anastomoses, the right becomes the more important of the two.

The vena cava inferior, though considerably later in its development than the cardinals, arises fairly early. It constitutes in front an unpaired trunk, at first very small, _opening into the right allantoic vein_, close to the heart. Posteriorly it is continuous with two veins placed on the inner border of the kidneys[235].

[235] The vena cava inferior does not according to Rathke's account unite behind with the posterior cardinal veins, as it is stated by Götte to do in the Anura. Götte questions the accuracy of Rathke's statements on this head, but my own observations are entirely in favour of Rathke's observations, and lend no support whatever to Götte's views.

The vena cava inferior passes through the dorsal part of the liver, and in doing so receives the hepatic veins.

The portal system is at first constituted by the vitelline vein, which is directly continuous with the venous end of the heart, and at first receives the two ductus Cuvieri, but at a later period unites with the left ductus. It soon receives a mesenteric vein bringing the blood from the viscera, which is small at first but rapidly increases in importance.

The common trunk of the vitelline and mesenteric veins, which may be called the portal vein, becomes early enveloped by the liver, and gives off branches to this organ, the blood from which passes by the hepatic veins to the vena cava inferior. As the branches in the liver become more important, less and less blood is directly transported to the heart, and finally the part of the original vitelline vein in front of the liver is absorbed, and the whole of the blood from the portal system passes from the liver into the vena cava inferior.

The last section of the venous system to be dealt with is that of the anterior abdominal vein. There are originally, as in the Anura, two veins belonging to this system, which owing to the precocious development of the bladder to form the allantois, constitute the allantoic veins (fig. 370, _vu_).

These veins, running along the anterior abdominal wall, are formed somewhat later than the vitelline vein, and fall into the two ductus Cuvieri. They unite with two epigastric veins (homologous with those in the Anura), which connect them with the system of the posterior cardinal veins. The left of the two eventually atrophies, so that there is formed an unpaired allantoic vein. This vein at first receives the vena cava inferior close to the heart, but eventually the junction of the two takes place in the region of the liver, and finally the anterior abdominal vein (as it comes to be after the atrophy of the allantois) joins the portal system and breaks up into capillaries in the liver[236].

[236] The junction between the portal system and the anterior abdominal vein is apparently denied by Rathke (No. 300, p. 173), but this must be an error on his part.

In Lizards the iliac veins join the posterior cardinals, and so pour part of their blood into the kidneys; they also become connected by the epigastric veins with the system of the anterior abdominal or allantoic vein. The subclavian veins join the system of the superior venæ cavæ.

The venous system of Birds and Mammals differs in two important points from that of Reptilia and Amphibia. Firstly the anterior abdominal vein is only a foetal vessel, forming during foetal life the allantoic vein; and secondly a direct connection is established between the vena cava inferior and the veins of the hind limbs and posterior parts of the cardinal veins, so that there is no renal portal system.

Aves. The Chick may be taken to illustrate the development of the venous system in Birds.

On the third day, nearly the whole of the venous blood from the body of the embryo is carried back to the heart by two main venous trunks, the anterior (fig. 125, _S.Ca.V_) and posterior (_V.Ca_) cardinal veins, joining on each side to form the short transverse ductus Cuvieri (_DC_), both of which unite with the sinus venosus close to the heart. As the head and neck continue to enlarge, and the wings become developed, the single anterior cardinal or jugular vein (fig. 371, _J_), of each side, is joined by two new veins: the vertebral vein, bringing back blood from the head and neck, and the subclavian vein from the wing (_W_).

On the third day the posterior cardinal veins are the only veins which return the blood from the hinder part of the body of the embryo.

[FIG. 371. DIAGRAM OF THE VENOUS CIRCULATION IN THE CHICK AT THE COMMENCEMENT OF THE FIFTH DAY.

_H._ heart; _d.c._ ductus Cuvieri. Into the ductus Cuvieri of each side fall _J._ the jugular vein, _W._ the vein from the wing, and _c._ the inferior cardinal vein; _S.V._ sinus venosus; _Of._ vitelline vein; _U._ allantoic vein, which at this stage gives off branches to the body-walls; _V.C.I._ inferior vena cava; _l._ liver.]

About the fourth or fifth day, however, the vena cava inferior (fig. 371, _V.C.I._) makes its appearance. This, starting from the sinus venosus not far from the heart, is on the fifth day a short trunk running backward in the middle line below the aorta, and speedily losing itself in the tissues of the Wolffian bodies. When the true kidneys are formed it also receives blood from them, and thenceforward enlarging rapidly becomes the channel by which the greater part of the blood from the hinder part of the body finds its way to the heart. In proportion as the vena cava inferior increases in size, the posterior cardinal veins diminish.

The blood originally coming to them from the posterior part of the spinal cord and trunk is transported into two posterior vertebral veins, similar to those in Reptilia, which are however placed dorsally to the heads of the ribs, and join the anterior vertebral veins. With their appearance the anterior parts of the posterior cardinals disappear. The blood from the hind limbs becomes transported directly through the kidney into the vena cava inferior, without forming a renal portal system[237].

[237] The mode in which this is effected requires further investigation.

On the third day the course of the vessels from the yolk-sack is very simple. The two vitelline veins, of which the right is already the smaller, form the ductus venosus, from which, as it passes through the liver on its way to the heart, are given off the two sets of _venæ advehentes_ and _venæ revehentes_ (fig. 371).

With the appearance of the allantois on the fourth day, a new feature is introduced. From the ductus venosus there is given off a vein which quickly divides into two branches. These, running along the ventral walls of the body from which they receive some amount of blood, pass to the allantois. They are the _allantoic_ veins (fig. 371, _U_) homologous with the anterior abdominal vein of the lower types. They unite in front to form a single vein, which becomes, by reason of the rapid growth of the allantois, very long. The right branch soon diminishes in size and finally disappears. Meanwhile the left on reaching the allantois bifurcates; and, its two branches becoming large and conspicuous, there still appear to be two main allantoic veins. At its first appearance the allantoic vein seems to be but a small branch of the vitelline, but as the allantois grows rapidly, and the yolk-sack dwindles, this state of things is reversed, and the less conspicuous vitelline appears as a branch of the larger allantoic vein.

[FIG. 372. DIAGRAM OF THE VENOUS CIRCULATION IN THE CHICK DURING THE LATER DAYS OF INCUBATION.

_H._ heart; _V.S.R._ right vena cava superior; _V.S.L._ left vena cava superior. The two venæ cavæ superiores are the original 'ductus Cuvieri,' they open into the sinus venosus. _J._ jugular vein; _Su.V._ anterior vertebral vein; _In.V._ inferior vertebral vein; _W._ subclavian; _V.C.I._ vena cava inferior; _D.V._ ductus venosus; _P.V._ portal vein; _M._ mesenteric vein bringing blood from the intestines into the portal vein; _O.f._ vitelline vein; _U._ allantoic vein. The three last mentioned veins unite together to form the portal vein; _l._ liver.]

On the third day the blood returning from the walls of the intestine is insignificant in amount. As however the intestine becomes more and more developed, it acquires a distinct venous system, and its blood is returned by veins which form a trunk, the _mesenteric vein_ (fig. 372, _M_) falling into the vitelline vein at its junction with the allantoic vein.

These three great veins, in fact, form a large common trunk, which enters at once into the liver, and which we may now call the _portal vein_ (fig. 372, _P.V_). This, at its entrance into the liver, partly breaks up into the _venæ advehentes_, and partly continues as the ductus venosus (_D.V_) straight through the liver, emerging from which it joins the vena cava inferior. Before the establishment of the vena cava inferior, the venæ revehentes, carrying back the blood which circulates through the hepatic capillaries, join the ductus venosus close to its exit from the liver. By the time however that the vena cava has become a large and important vessel it is found that the venæ revehentes, or as we may now call them the _hepatic veins_, have shifted their embouchment, and now fall directly into that vein, the ductus venosus making a separate junction rather higher up (fig. 372).

This state of things continues with but slight changes till near the end of incubation, when the chick begins to breathe the air in the air-chamber of the shell, and respiration is no longer carried on by the allantois. Blood then ceases to flow along the allantoic vessels; they become obliterated. The vitelline vein, which as the yolk becomes gradually absorbed proportionately diminishes in size and importance, comes to appear as a mere branch of the portal vein. The ductus venosus becomes obliterated; and hence the whole of the blood coming through the portal vein flows into the substance of the liver, and so by the hepatic veins into the vena cava.

Although the allantoic (anterior abdominal) vein is obliterated in the adult, there is nevertheless established an anastomosis between the portal system and the veins bringing the blood from the limbs to the vena cava inferior, in that the caudal vein and posterior pelvic veins open into a vessel, known as the coccygeo-mesenteric vein, which joins the portal vein; while at the same time the posterior pelvic veins are connected with the common iliac veins by a vessel which unites with them close to their junction with the coccygeo-mesenteric vein.

Mammalia. In Mammals the same venous trunks are developed in the embryo as in other types (fig. 373 A). The anterior cardinals or external jugulars form the primitive veins of the anterior part of the body, and the internal jugulars and anterior vertebrals are subsequently formed. The subclavians (fig. 373 A, _s_), developed on the formation of the anterior limbs, also pour their blood into these primitive trunks. In the lower Mammalia (Monotremata, Marsupialia, Insectivora, some Rodentia, etc.), the two ductus Cuvieri remain as the two superior venæ cavæ, but more usually an anastomosis arises between the right and left innominate veins, and eventually the whole of the blood of the left superior cava is carried to the right side, and there is left only a single superior cava (fig. 373 B and C). A small rudiment of the left superior cava remains however as the sinus coronarius and receives the coronary vein from the heart (figs. 373 C, _cor_ and 374, _cs_).

[FIG. 373. DIAGRAM OF THE DEVELOPMENT OF THE PAIRED VENOUS SYSTEM OF MAMMALS (MAN). (From Gegenbaur.)

_j._ jugular vein; _cs._ vena cava superior; _s._ subclavian veins; _c._ posterior cardinal vein; _v._ vertebral vein; _az._ azygos vein; _cor._ coronary vein.

A. Stage in which the cardinal veins have already disappeared. Their position is indicated by dotted lines. B. Later stage when the blood from the left jugular vein is carried into the right to form the single vena cava superior; a remnant of the left superior cava being however still left. C. Stage after the left vertebral vein has disappeared; the right vertebral remaining as the azygos vein. The coronary vein remains as the last remnant of the left superior vena cava.]

The posterior cardinal veins form at first the only veins receiving the blood from the posterior part of the trunk and kidneys; and on the development of the hind limbs receive the blood from them also.

As in the types already described an unpaired vena cava inferior becomes eventually developed, and gradually carries off a larger and larger portion of the blood originally returned by the posterior cardinals. It unites with the common stem of the allantoic and vitelline veins in front of the liver.

[FIG. 374. DIAGRAM OF THE CHIEF VENOUS TRUNKS OF MAN. (From Gegenbaur.)

_cs._ vena cava superior; _s._ subclavian vein; _ji._ internal jugular; _je._ external jugular; _az._ azygos vein; _ha._ hemiazygos vein; _c._ dotted line shewing previous position of cardinal veins; _ci._ vena cava inferior; _r._ renal veins; _il._ iliac; _hy._ hypogastric veins; _h._ hepatic veins.

The dotted lines shew the position of embryonic vessels aborted in the adult.]

At a later period a pair of trunks is established bringing the blood from the posterior part of the cardinal veins and the crural veins directly into the vena cava inferior (fig. 374, _il_). These vessels, whose development has not been adequately investigated, form the common iliac veins, while the posterior ends of the cardinal veins which join them become the hypogastric veins (fig. 374, _hy_). Owing to the development of the common iliac veins there is no renal portal system like that of the Reptilia and Amphibia.

Posterior vertebral veins, similar to those of Reptilia and Birds, are established in connection with the intercostal and lumbar veins, and unite anteriorly with the front part of the posterior cardinal veins (fig. 373 A)[238].

[238] Rathke, as mentioned above, holds that in the Snake the front part of the posterior cardinals completely aborts. Further investigations are required to shew whether there really is a difference between Mammalia and Reptilia in this matter.

On the formation of the posterior vertebral veins, and as the inferior vena cava becomes more important, the middle part of the posterior cardinals becomes completely aborted (fig. 374, _c_), the anterior and posterior parts still persisting, the former as the continuations of the posterior vertebrals into the anterior vena cava (_az_), the latter as the hypogastric veins (_hy_).

Though in a few Mammalia both the posterior vertebrals persist, a transverse connection is usually established between them, and the one (the right) becoming the more important constitutes the azygos vein (fig. 374, _az_), the persisting part of the left forming the hemiazygos vein (_ha_).

The remainder of the venous system is formed in the embryo of the vitelline and allantoic veins, the former being eventually joined by the mesenteric vein so as to constitute the portal vein.

The vitelline vein is the first part of this system established, and divides near the heart into two veins bringing back the blood from the yolk-sack (umbilical vesicle). The right vein soon however aborts.

The allantoic (anterior abdominal) veins are originally paired. They are developed very early, and at first course along the still widely open somatic walls of the body, and fall into the single vitelline trunk in front. The right allantoic vein disappears before long, and the common trunk formed by the junction of the vitelline and allantoic veins becomes considerably elongated. This trunk is soon enveloped by the liver.

The succeeding changes have been somewhat differently described by Kölliker and Rathke. According to the former the common trunk of the allantoic and vitelline veins in its passage through the liver gives off branches to the liver, and also receives branches from this organ near its anterior exit. The main trunk is however never completely aborted, as in the embryos of other types, but remains as the ductus venosus Arantii.

With the development of the placenta the allantoic vein becomes the main source of the ductus venosus, and the vitelline or portal vein, as it may perhaps be now conveniently called, ceases to join it directly, but falls into one of its branches in the liver.

The vena cava inferior joins the continuation of the ductus venosus in front of the liver, and, as it becomes more important, it receives directly the hepatic veins which originally brought back blood into the ductus venosus. The ductus venosus becomes moreover merely a small branch of the vena cava.

At the close of foetal life the allantoic vein becomes obliterated up to its place of entrance into the liver; the ductus venosus becomes a solid cord--the so-called round ligament--and the whole of the venous blood is brought to the liver by the portal vein[239].

[239] According to Rathke the original trunk connecting the allantoic vein directly with the heart through the liver is aborted, and the ductus venosus Arantii is a secondary connection established in the latter part of foetal life.

Owing to the allantoic (anterior abdominal) vein having merely a foetal existence an anastomosis between the iliac veins and the portal system by means of the anterior abdominal vein is not established.

BIBLIOGRAPHY _of the Venous System_.

(498) J. Marshall. "On the development of the great anterior veins." _Phil. Trans._, 1859.

(499) H. Rathke. "Ueb. d. Bildung d. Pfortader u. d. Lebervenen b. Säugethieren." _Meckel's Archiv_, 1830.

(500) H. Rathke. "Ueb. d. Bau u. d. Entwick. d. Venensystems d. Wirbelthiere." _Bericht. üb. d. naturh. Seminar. d. Univ. Königsberg_, 1838.

_Vide_ also Von Baer (No. 291), Götte (No. 296), Kölliker (No. 298), and Rathke (Nos. 299, 300, and 301).

_Lymphatic System._

The lymphatic system arises from spaces in the general parenchyma of the body, independent in their origin of the true body cavity, though communicating both with this cavity and with the vascular system.

In all the true Vertebrata certain parts of the system form definite trunks communicating with the venous system; and in the higher types the walls of the main lymphatic trunks become quite distinct.

But little is known with reference to the ontogeny of the lymphatic vessels, but they originate late in larval life, and have at first the form of simple intercellular spaces.

The lymphatic glands appear to originate from lymphatic plexuses, the cells of which produce lymph corpuscles. It is only in Birds and Mammals, and especially in the latter, that the lymphatic glands form definite structures.

_The Spleen._ The spleen, from its structure, must be classed with the lymphatic glands, though it has definite relations to the vascular system. It is developed in the mesoblast of the mesogastrium, usually about the same time and in close connection with the pancreas.

According to Müller and Peremeschko the mass of mesoblast which forms the spleen becomes early separated by a groove on the one side from the pancreas and on the other from the mesentery. Some of its cells become elongated, and send out processes which uniting with like processes from other cells form the trabecular system. From the remainder of the tissue are derived the cells of the spleen pulp, which frequently contain more than one nucleus. Especial accumulations of these cells take place at a later period to form the so-called Malpighian corpuscles of the spleen.

BIBLIOGRAPHY _of Spleen_.

(501) W. Müller. "The Spleen." _Stricker's Histology._

(502) Peremeschko. "Ueb. d. Entwick. d. Milz." _Sitz. d. Wien. Akad. Wiss._, Vol. LVI. 1867.

_Suprarenal bodies._

In Elasmobranch Fishes two distinct sets of structures are found, both of which have been called suprarenal bodies. As shewn in the sequel both of these structures probably unite in the higher types to form the suprarenal bodies.

One of them consists of a series of paired bodies, situated on the branches of the dorsal aorta, segmentally arranged, and forming a chain extending from close behind the heart to the hinder end of the body cavity. Each body is formed of a series of lobes, and exhibits a well-marked distinction into a cortical layer of columnar cells, and a medullary substance formed of irregular polygonal cells. As first shewn by Leydig, they are closely connected with the sympathetic ganglia, and usually contain numerous ganglion cells distributed amongst the proper cells of the body.

The second body consists of an unpaired column of cells placed between the dorsal aorta and unpaired caudal vein, and bounded on each side by the posterior parts of the kidney. I propose to call it the interrenal body. In front it overlaps the paired suprarenal bodies, but does not unite with them. It is formed of a series of well-marked lobules, etc. In the fresh state Leydig (No. 506) finds that "fat molecules form the chief mass of the body, and one finds freely imbedded in them clear vesicular nuclei." As may easily be made out from hardened specimens it is invested by a tunica propria, which gives off septa dividing it into well-marked areas filled with polygonal cells. These cells constitute the true parenchyma of the body. By the ordinary methods of hardening, the oil globules, with which they are filled in the fresh state, completely disappear.

The paired suprarenal bodies (Balfour, No. 292, pp. 242-244) are developed from the sympathetic ganglia. These ganglia, shewn in an early stage in fig. 380, _sy.g_, become gradually divided into a ganglionic part and a glandular part. The former constitutes the sympathetic ganglia of the adult; the latter the true paired suprarenal bodies. The interrenal body is however developed (Balfour, No. 292, pp. 245-247) from indifferent mesoblast cells between the two kidneys, in the same situation as in the adult.

The development of the suprarenal bodies in the Amniota has been most fully studied by Braun (No. 503) in the Reptilia.

In Lacertilia they consist of a pair of elongated yellowish bodies, placed between the vena renalis revehens and the generative glands.

They are formed of two constituents, viz. (1) masses of brown cells placed on the dorsal side of the organ, which stain deeply with chromic acid, like certain of the cells of the suprarenals of Mammalia, and (2) irregular cords, in part provided with a lumen, filled with fat-like globules[240], amongst which are nuclei. On treatment with chromic acid the fat globules disappear, and the cords break up into bodies resembling columnar cells.

[240] These globules are not formed of a true fatty substance, and this is also probably true for the similar globules of the interrenal bodies of Elasmobranchii.

The dorsal masses of brown cells are developed from the sympathetic ganglia in the same way as the paired suprarenal bodies of the Elasmobranchii, while the cords filled with fat-like globules are formed of indifferent mesoblast cells as a thickening in the lateral walls of the inferior vena cava, and the cardinal veins continuous with it. The observations of Brunn (No. 504) on the Chick, and Kölliker (No. 298, pp. 953-955) on the Mammal, add but little to those of Braun. They shew that the greater part of the gland (the cortical substance) in these two types is derived from the mesoblast, and that the glands are closely connected with sympathetic ganglia; while Kölliker also states that the posterior part of the organ is unpaired in the embryo rabbit of 16 or 17 days.

The structure and development of what I have called the interrenal body in Elasmobranchii so closely correspond with that of the mesoblastic part of the suprarenal bodies of the Reptilia, that I have very little hesitation in regarding them as homologous[241]; while the paired bodies in Elasmobranchii, derived from the sympathetic ganglia, clearly correspond with the part of the suprarenals of Reptilia having a similar origin; although the anterior parts of the paired suprarenal bodies of Fishes have clearly become aborted in the higher types.

[241] The fact of the organ being unpaired in Elasmobranchii and paired in the Amniota is of no importance, as is shewn by the fact that part of the organ is unpaired in the Rabbit.

In Elasmobranch Fishes we thus have (1) a series of paired bodies, derived from the sympathetic ganglia, and (2) an unpaired body of mesoblastic origin. In the Amniota these bodies unite to form the compound suprarenal bodies, the two constituents of which remain, however, distinct in their development. The mesoblastic constituent appears to form the cortical part of the adult suprarenal body, and the nervous constituent the medullary part.

BIBLIOGRAPHY _of the Suprarenal bodies_.

(503) M. Braun. "Bau u. Entwick. d. Nebennieren bei Reptilien." _Arbeit. a. d. zool.-zoot. Institut Würzburg_, Vol. V. 1879.

(504) A. v. Brunn. "Ein Beitrag z. Kenntniss d. feinern Baues u. d. Entwick. d. Nebennieren." _Archiv f. mikr. Anat._, Vol. VIII. 1872.

(505) Fr. Leydig. _Untersuch. üb. Fische u. Reptilien._ Berlin, 1853.

(506) Fr. Leydig. _Rochen u. Haie._ Leipzig, 1852.

_Vide_ also F. M. Balfour (No. 292), Kölliker (No. 298), Remak (No. 302), etc.