Talk:Paper - The fusion of the cardiac anlages and the formation of the cardiac loop in the cat (1916)

From Embryology

The fusion of the cardiac anlages and the formation of the cardiac loop in the cat (felis domestica)

H. VON W. SCHULTE

FroDi Ihc Anatomical Laboratory of Cohonhia University


SIXTEEN FIGURES


111 the trausfonnation of the bilateral anlages of the heart into a single median organ, two processes are to be distinguished, the fusion of the myoepicardial mantles and the fusion of the endothelial tubes. For while the union of the mantles is a necessary condition of the coalescence of the tubes and its character determines the general features of the endothehal fusion, yet the latter process is relative^ much retarded and evinces a considerable degree of independence in many details. For these reasons it will be convenient to follow the two processes separately.

This study is based upon the now numerous earlj^ embryos of the cat in the Columbia Embryological Collection. These were cut in transverse serial sections of 13.3 fx and variously stained. In selected embryos reconstructions by the Born method were made of the myoepicardium, of the endothelium and mesenchyme, and finally of the lumina of the heart tubes and angiocysts. By a comparison of the last two it is possible to ascertain the precise limits of the solid and hollow parts of the anlages.

The processes of approximation, of fusion, and of loop-formation take place in a short period of development as measured by the rate of formation of the mesodermic somites, the whole duration of these changes in the cat falling between the stages 8 and 21 somites. At 8 somites (fig. 1) ,the foregut in the cardiac region is widely open and the splanchnopleure is spread out


46


H. VON W. SCHULTE


flat : the cardiac aiilages lie in the same plane as the neural tube and are separated by a wide interval. In the embryo of 11 somites (fig. 3) they are ventral to the closed foregut and their proximal or bulbar ends are close together. At 12 somites



Fig. 1. Reconstruction of compact mesoderm of cut embr3-o of eight pairs of somites. Columbia Collection, No. 588. X 180. Reduced one-half. Dorsal view the parietal mesoderm has been exsected to show topography of parietal cavity. 1, Parietal cavity; 2, pericephalic cavity; .3, parietal recess; 4, myoepicardial mantle; 5, oblique groove dividing mantle into mesal and lateral protions; 6, retrocardiac plate; 7, jirecardiac plate.


FUSION OF CARDIAC ANLAGES IN THE CAT 47 fusion of the mantles is complete (fig. 4) ; at 14 the formation of the loop is initiated (fig. 7); and at 16 it is completed. The fusion of the endothelial tubes begins in the embryo of 12 somites, in that of 14 it is all but complete, and the embryo of 20 somites is the oldest that shows a remnant of septum between the endothelial tubes (figs. 11, 13, and 16).

THE MYOEPICARDIAL MANTLES

There are many brief statements and illustrations of the myoepicardial mantles in the literature recording conditions in these structures immediately antecedent to and during fusion, from which nevertheless it would be impossible to compile a history of the process. The observations of Kolliker, His, Tiirstig, Schultze, Spee, Bonnet, Fleischmann, Selenka and Heape, are as familiar as they are important and have been adequately summarized by Mollier,i who has also included the less well known and less accessible work of Martin, ^ which is valuable for its excellent illustrations. Strahl and Carius^ and Parker," alone have given continuous and detailed accounts of the processes here considered. The former have illustrated by a series of diagrammatic cross sections two different types of fusion, one in which the formation of the ventral mesocardium antecedes that of the dorsal, the other in which the converse obtains. Two factors here come into play, first the width of the foregut and second the primitive position of the bilateral cardiac anlages. The area of visceral mesoderm which intervenes between the myoepicardial mantle and the mesal angle of the parietal cavity, they designate the retrocardiac plate, similarly that between the mantle and the lateral angle the precardiac plate

' MoUier, 8. Die crste AnLige ties Herzen bei den Wirbeltieren in Hertwig's Haudbuch der vergl. und experiment. Entwickel. der Wirbeltiere, Jena, 1906, Bd. 1, which see for literature.

2 Martin. Lehrbuch der Anatomie der Hausteire. Bd. 1, Stuttgart, 1902.

  • Strahl and Carius. Beitrage zur Entwickelungsgeschichte des Herzens und

der Korperhohlen. Arch. f. Anat. und Physiol., Bd. 15, 1899.

  • Parker, Katherine M. The early development of the heart and anterior

vessels in Marsupials, with special reference to Perameles. Proc. Zool. Soc, London, 1915, Pt. III.


48 H. VON ^y. schulte

(fig. 1). If the retrocardiac plate is narrow relatively to the foregut, the myocardial mantles will be widely separated upon the closure of the gut and the ventral mesocardium will precede the dorsal in its formation. This type obtains in the rabbit, and to it the cat conforms with some peculiarities in detail which are recorded below. If, however, the con\'erse obtains, if the retrocardiac plate is broad relative to the width of the foregut, the formation of the dorsal mesocardium will be accelerated. Of this modification of the process the guinea-pig is an example. This in brief is the analysis of Strahl and Carius. Mollier comments justly that is it 'not quite correct to say that the closure of the foregut is the cause of the fusion of the anlages of the heart, because this occurs independently and later, a point which KoUiker had already made.' The possibility of a third type of approximation of the mantles must be admitted, such for example as obtains in the chick, where the retro- and pre-cardiac plates are of such proportions that ventral and dorsal mesocardiac are formed at approximately the same time. Further it is not necessary to assume that the retrocardiac plate is of the same breadth in its whole extent, and a difference in this respect might reach such a degree as to produce a mixed type of mesocardial formation.

In marsupials Parker describes a developmental process resembling but not identical with the conditions present in the cat. Like that species and the rabbit there is no ventral mesocardium. On closure of the foregut the mantles are widely separated, the intervening remnants of the precardiac plates forming a median plate between them. Associated with this middle cardiac plate are mesenchyme cells, later capillaries. Subsequently the middle cardiac plate is inconspicuous in marsupials not forming a thicked ridge as in the cat (Cf. Parker, fig. 17, with fig. 12 of this paper), nor are the angiocysts of the region assigned a role in effecting the fusion between the endothelial anlage. These first fuse in the region of the bulb. In embryos of 15-16 somites a constriction between bulb and ventricle is present only on the right side, which as a whole exceeds the left side in size. Caudad the ^'entricle is limited by a constriction,


FUSION OF CAHDIAC AXLACJES IN THE CAT 49

and the following dilatiition where endothelium and myocardium laterale are closely approximated is interpreted as auricle. The actual stages of fusion other than in the bulbs and the early stages of the loops are not represented in Miss Parker's material.

In the cat from the first the myoepicardial mantle has an obliquely sagittal direction in consequence of which the retrocardiac plate broadens somewhat as it is followed caudad, and tapers in the opposite direction, (embryos of 4-7 somites). The topography of the parietal cavity in an embryo of 8 somites is shown in figure 1, which agrees closely with Fleischmann's figure of a total view of a cat embryo of this stage. ^

Before the infolding of the splanchnopleure is begun, the topographj^ of the mantles is such as would seem to entail their earlier approximation caudad where the retrocardiac plates are broadest. The direct contrary is the case; cephalad where these plates are narrow the bulbar segments are quickly brought into apposition, while the ventricular portions diverge and are widely separated caudad, nor is there sufficient difference in the width of the gut opposite these two segments of the heart to account for their difference in position. The problem thus presented is not easy of solution and its difficulty is increased by the rapidity of the process, for the heart passes in the period required for the development of three somites from the position shown in figure 1 to that in figure 3. Thus in the interval between the appearance of the eighth and eleventh somites the formation of the foregut is completed as far as the atrial extremity of the heart, and the anlages of that organ have been moved through a dorso-ventral arc of m arly 180° and become approximated ventral to the pharynx.

It is not possible on the basis of the material in hand to attempt a complete solution of this problem for which several processes extrinsic as well as intrinsic require minute investigation. Primarily there are the changes incident at this period in the general shape of the region, notably the shortening, associated with the beginning ventral flexion of the forebrain, which

  • Fleischmann, A. EmhryoloKische Untersufhunficii. I, 1889.

THE AMERICAN JOURNAL OF ANATOMY, VOL. 29, NO. 1


50 H. VON W. SCHULTE

together result in such a remodelling as is hardly to be attributed to unequal growth. In this shortening is to be found an explanation of the kinking of the heart tubes prior to their fusion. In the second place the acceleration of infolding of the splanchnopleure cephalad tends to bring the cardiac anlages into an oblique position with their arterial ends near together, their venous extremities divergent. The supposition of accelerated growth in the cephalic portion of the retrocardiac plate is not supported by a conspicuous increase of mitotic figures, but the perceptible diminution of its thickness and its rectilinear position at the end of the process suggest the action of a moulding force. Finally MoUier's comment must be borne in mind — the closure of the gut is not the cause of fusion between the cardiac anlages. It determines their approximation, but when this is complete the mantles are separated by an interval, which is bridged by a plate of mesoderm derived from the precardiac plates. By compression and ultimate absorption of this into the mantles fusion is accomphshed, and the process as a whole is evidently one of remodelling which cannot be explained simply as a cessation of growth in this district. Similar forces may therefore come into play in effecting the early and close approximation of the arterial ends of the mantles notwithstanding the narrowness of the cephalic portion of the retrocardiac plates.

In an embryo of 9 pairs of somites the cardiac anlages have a position intermediate between those shown in figures I and 3. The mantles are obliquely placed, their arterial ends close together though separated by a deep cleft, their venous ends widely divergent. Each is indented in its lateral contour by an angular incisure which marks the junction of their approximated and divergent portions. The differentiation of the heart tube into segments j^rior to fusion was observed by His and the portions have been variously designated bulbar and atrial or ventricular and atrial. Their subsequent history in the cat show the first segment to be the bulb continuing forward into the truncus, the second the ventricle later expanding at its caudal end into the atrium, but this only after the loop is initiated and fusion well under way. The angulation is therefore the bulbo


FrSIOX OF CARDIAC ANLAGES IN THE CAT


51


veiitricLihir sulcus. Caudad the mantles diverge and pass into the lateral mesocardia with a gentle curve. A distinct angulation at this point is not apparent until the stage of 11 somites.

On account of the obliquity of the myoepicardial mantles, a triangular interval with its base at the anterior intestinal portal is left between them. This is bridged by the residue of the



Fig. 2 Photomicroo;raph of a cat emliryo of nine pairs of somites. Colunilna Collection, Xo. 532. X 125. Reduced one-half. /, Ventricular segment of mantle; 2, middle cardiac plate.


precardiac plates after the closure of the foregut and the disappearance of the ventral mesocardium. It may be designated the middle cardiac plate.

Between the bulbar segments it is narrow in this differing from the rabbit (Strahl and Carius), but caudad attains a considerable width (fig. 2). It is seen to agree in thickness and in the character of its cells with the myoepicardial mantles and like them to project entadin numerous longitudinal ridges, here shown in cross-section. These ridges as elsewhere in the splanchnopleure are intimately related to the formation of mesenchyme


52


H. VON W. SCHULTE


and differ in no important character here in the middle cardiac plate and in the myoepicardial mantles from those observed in the splanchnopleure at large. I shall recur to this field of mesenchyme and endothelium in connection with the description of the endocardium. The middle plate is demarcated from the mantle on each side by a sulcus which gradually becomes shallow and is effaced as it is followed caudad.


'i



Fig. 3 Myoei)icar(lial mantle of a cat eml)ryo of eleven pairs of mesodermic somites. Columbia Collection, No. 534. X 300. Reduced one-half. 1, Middle, cardiac plate; 2, bulbus; .3, ventricle; 4, bulbo-ventricular sulcus; .5, atriovenous angle.


The mantles in their ventricular segments are dorso-ventrallj' flattened and of great transverse extent. Between them in ventral view the middle field forms a flat depression. The bulbar segments are close together but separated by a deep cleft, the roof of which is continuous with the middle plate. They are much narrower and less flattened than the ventricles.


FUSION OF CARDIAC ANLAGES IN THE CAT 53

Embryos of 11 pairs of somites show some variation in development. Some lia\'o hardly progressed beyond the one of 9 somites, Init in the most advanced of this stage the mantles are approximated in the greater part of their length and the middle plate is considerablv reduced. Fusion has not yet occurred (fig. 3).

The mantles present two angulations in thier ectal contour. The first bend is the bulbo-ventricular sulcus already described, which becomes accentuated on the left side in the later stages as the flexion of the tube develops. In this embryo it is nearly rectangular and is approximately symmetrical on the two sides. I take the deep incisure in Martin's cat embryo of 4 mm. to be its equivalent. It is also to be recognized in Duval's figure of a chick of 8 somites, though obscured by the outline of the amnion. The second bend is a more gradual change of direction at the junction of the heart tube with the omphalomesenteric vein and may accordingly be designated the cardiovenous angle. It is not to be distinguished in Martin's figure, but in the heart of the chick as shown by Duval it is a deep cleft. These angulations are important in that they determine the earliest points of fusion in later stages between the endothelial tubes.

The middle plate is much reduced. Its cephalic portion is concealed in the deep cleft between the bulbs; caudad it forms a convex triangle exposed in its whole extent. Throughout the mantles are markedly dorso-ventrally flattened.

The fusion between the mantles and the formation of a dorsal mesocardium is effected in embryos of 12 and 13 pairs of somites. The dorsal mesocardium is very short, set off by a sulcus from the mesoderm covering the foregut and ventrally by a deeper sulcus from the mantles. Cephalad its leaves separate to give passage to the forming aortic roots and caudad it expands and is continuous with the lateral mesocardia.

^ Both these cuts are given by Mollier. Op. cit. figures 713 and 715. Allen Thompson illustrates this form of the heart in the chick in his article 'On the development of the vascular system in the foetus of vertebrated animals.' Edinburgh new philosophical journal, 1830, PI. 2, fig. 12.


54


H. VON W. SCHULTE


The mantles are clearly demarcated from the lateral mesocardia by deep sulci so that the junction between omphalomesenteric vein and heart tube is deeply indented on each side. From this point the mantles gradually expand again to l^e in



Fig. 4 Myoeijit-ardial mantle of a cat embryo of twelve pairs of somites. Cokimbia Collection, No. 547. X 300. Reduced one-half. 1, Middle cardiac plate; 2, y)ull)ns; .?, ventricle; 4. iMilbo-ventricular sulcus; 5, atrio-venous angle; 6, shoulder of mantle.


dented by the deep oblique bulbo-ventricular clefts. Beyond them the truncus has the form of a cylinder, slightly flattened dorso-ventrally, emerging between the high shoulders of the mantles (fig. 4), In this stage the heart is approximately symmetrical and the bulbo-ventricular sulci of the two sides are in


FUSION OF CARDIAC ANLAGES IN THE CAT 55

all important respects identical. The (Mital projections which thej occasion are shown in figure 5.

The mid-region of the fused mantles requires some comment. Here the middle-plate has been compressed to a longitudinal ridge bounded b}^ well defined sulci. These are expressed entally l)y ridges, which in an embryo of 13 somites extend far into the bulbus. In the heart of this 12 somite embryo they are reduced in this segment and the middle region of the bulbus is slightly concave. The fusion is here complete. As the bulbo


Fig. 5 Cephalic portion of same model as figure 4, ental view. 1, Middle cardiac plate; £, fundus of right bulbo-vcntricular sulcus; 3, fundus of left bulhoventricular sulcus; 4, interior of bulbus; 5, dorsal mesocardium; 6, right ventricle; 7, left ventricle.


ventricidar fissures are approached the ridges begin as low elevations which increase in height to about the middle of the ventricular segment fading out towards the terminal constriction of the heart. Between the ridges projecting from the middle plate entally are occasional small processes of mesoderm, evidently remnants of the more numerous projections of earlier stages.

As compared with the heart of the 11 somite embryo this heart has lengthened somewhat, but its striking changes in contour are associated with the deepening of the two pairs of incisures at its lateral margin. The bulbar segment has diminished absolutely in breath and so in less degree has the ventricular especially at its caudal end where it joins the lateral meso


56


H. VON W. SCHULTE


cardia and septum transversum. The dorso-ventral increase in diameter is also marked (fig. 12).

Upon this condition follows so rapidly the development of the cardiac loop that in only one embryo of 14 somites was an



fig. 6 Myocardium of a cat embryo of fourteen pairs of somites. Columbia Collection, No. 188. X 300. Reduced one-half, ventral view. L Interventricular sulcus; 2, shoulder of left mantle; 3, left bulbo-ventricular sulcus; 4, right bulbo-ventricular sulcus; 5, lateral mesocardia.


intermediate stage observed. The model of this myocardium is shown in figures 6 to 8. In figure 6 in which the reconstruction is viewed from in front it presents resemblances to Mall's^

' Mall, F. P. On the development of the human heart. Am. Jour. Anat., vol. 13, 1912, fig. 1. Cf. Dandy, W. E. A human embryo with seven pairs of somites, measuring about 2 mm. in length. Id., vol. 10, 1910. Also Evans, Keibel and Mall. Manual human embryology, vol. 2, figs. 408-9, and Mall, Ibid., vol. 1, figs. 382-G.


FUSION OF CARDIAC ANLAGES IN THE CAT


57


freely-treated model of the heart of a human embryo of 7 to 8 pairs of somites. In both there is a bulbo-ventricular cleft on the left and there is little in this view of either model to suggest a mode of formation of the loop different from that given by jNIall, which is simply the deepening of this sulcus between bulb



Fig. 7 The same model as in figure 6 viewed somewhat from the right and above. 1, Bulbus; 2, shoulder of left mantle; 3, shoulder of right mantle; 4, bulbo-ventricular sulcus ; 5, right bulbo-ventricular sulcus ; 6, right lateral mesocardium.


and ventricle. The heart of the cat is rather more plump, its contour more convex, but this may well be due to a greater distension with blood. Two details small in size but not in morphologic significance are present in the cat, which are not shown in the figure of the human heart. The right margin has a small indentation which is the remnant of the right bulbo-ventrical


58 H. VON W. SCHULTE

cleft as is readily seen in the view from the right and above shown in figure 7, and caudad there is a slight nearly transverse depression furrowing the apex of the cardiac loop, associated with an ental ridge which marks the beginning of the septum ventriculorum.



Fig. 8 Cephalic portion of same model as in figures 6 and 7, ental view. /, Ridge representing middle cardiac plate; 2, fundus of right bulbo-ventricular sulcus; 3, fundus of left Indbo-ventricular sulcus; 4, interior of bulbus; 5, atrial region; 6, dorsal mesocardiiun.


The changes that have supervened to transform this heart in the period between appearance of the twelfth and fourteen pairs of somites are easily appreciated on the comparison of the figures (figs. 4 and 7). The most striking changes affect the shoulders of the mantles, that of the left side is greatly elevated, that of the right correspondingly depressed. This entails a deepening of the left bulbo-ventricular sulcus and an opening out on


FUSION OF CARDIAC ANLAGES IN THE CAT 59

the part- of the i-ight. There is also axial rotation. The shoulder of the left mantle thrusts ventrad displacing the bulbils to the right, and the accompanying rotation displaces the depressed shoulder of the right mantle dorsad so that it is concealed in ventral view. The loop is directed to the right, ventrad and slightly caudad, and the left lateral mesocardium comes to occupy a slightly more cephalic position than the right. As a whole, the dorso-ventral depth of the myocardium is increased at the expense of its transverse breadth (fig. 15) and its extremities are a little approximated, the sagittal distance between the end of the bulbus and the junction of the mantles with the lateral mesocardia being absolutely diminished. The dorsal mesocardium is retained unbroken in its whole extent.

The middle plate is undergoing reduction by being absorbed into the mantles. It is represented by a ridge projecting entad and at the sides gradually diminishing in thickness as it fades into the mantles. It can be followed into the beginning of the bulb where it lies opposite the partition between the endothelial tubes (fig. 14). It then runs along the convexity of the loop occupying the same position relative to the fusing endocardial anlages as in the bulb (fig. 15). On reaching the caudal contour of the loop it becomes continuous with the ental elevation produced by the interventricular sulcus. Thus by the ridge and interventricular sulcus the primitive median line of the heart is represented, a conclusion which is borne out by their location in their whole course opposite the line of fusion, as yet incomplete, between the endothelial tubes.

In the heart of an embryo of 16 pairs of somites the loop has increased and projects more strongly. The left bulbo-ventricular sulcus is nearly horizontal and the right has been reduced, appearing only as a slight furrow ectally and a slight angle within the myocardium. The middle plate is again represented by a low ridge extending from the end of the bulb to the beginning of the interventricular septum. This now has an obliquely transverse direction. In an embryo of 18 to 19 pairs of somites the ridge of the middle plate disappears. The interventric


60 H. VON W. SCHULTE

iilar sulcus is still oblique. It is only as the venous end of the heart begins to move towards the right that the septum assumes a dorso-ventral direction. Its transverse direction in early stages is rendered possible by the primary displacement of this extremity to the left as will be demonstrated in the consideration of the endothelial analges.

To summarize, the fusion of the myoepicardial mantles is accomplished with the aid of a middle cardiac plate, which subsequently becomes reduced to a ridge marking the primitive median line during the formation of the loop. It is continued caudad as the interventricular septum, which thus forms in the line of original fusion of the heart anlages and by its appearance separates again, so far as the ventricles are concerned, the primitive bilateral anlages. The myocardium in the early stages of fusion is bilaterally symmetrical with a well marked bulbo-ventricular sulcus on each side. In the formation of the loop the left sulcus deepens and the right opens up and gradually is obliterated. It is possible that the middle plate now reduced to a ridge and located at the convexity of the forming loop, is less plastic than the thinner portions of the mantles and failing to lengthen to a sufficient degree exerts a traction which occasions the appearance of the interventricular sulcus.

THE ENDOTHELIAL TUBES

The origin of the endocardium differs in nowise from the origin of endothehum elsewhere in the cat. It develops from mesenchyme which is formed in loco, first by migration of cells from the compact visceral mesoderm, second by delamination of groups of cells from the same source. In this process ridges and projections are formed by the mesoderm from which the mesenchyme loosens itself. A means of migration is afforded the amoeboid cells by the presence of interdermal cytodesmata, delicate protoplasmic bridges stretching between the mesoderm and entoderm. The early mesenchyme consists of single cells and scattered groups which are arranged in plates or even cords. Within these groups vacuoles appear and enlarging flat


FUSION OF CARDIAC ANLAGES IN THE CAT 61

ten the containing cells to endotheliiiin.'* Th(^ rosnlting vesicles Bremer^ has termed angiocysts.

In the heart the conformation of the myoepicardial mantle confines the mesenchyme and angiocysts lodged within its concavity and entails their transformation into a longitudinal channel. In the embryo of 4 pairs of somites the projecting ridges of the mantle still intervene between the angiocysts and delay the formation of a continuous lumen. In embryos, of 7 to 8 somites this is nearly complete, and is so from omphalomesenteric vein to ventral aorta in the embryo of 9 somites. In all of these however, and to a less degree in still older embryos there are present unannexed angiocysts and abundant mesenchyme about and especially between the endothelial tubes. The mesenchyme is so gradually transformed into endothelium that it is not easy to define the limit between the two stages, but the endothelium is the dominant tissue by the time fusion begins. The cat thus conforms to MoUier's^*^ observation of the late fusion relative to condition of tissue in amniotes, the fusion occurring when the heart is mesenchymatous in sauropsids, when it has become endothelial in mammals. His recognition of a stage of solid cords antecedent to the mesenchymatous stage, in which the accelerated fusion of anamniotes occurs, is theoretically and terminologically not altogether fortunate, for the undifferented cells first moving into the mesostroma and subsequently gi\ing rise to a variety of products are properly termed mesenchyme on grounds of morphology. The term does not necessarily connote any theoretical prepossessions. It simply designates a position and arrangement of cells other than that obtaining in the three germ-layers. In immediately subsequent stages the descendants of some of these cells retain this character, while other become flattened in response to the

^ Cf. for interdermal cytodesmata, v. Szily. Anat. Anz., Bd. 24, 1903, p. 417; and Studnicka, Ibid., Bd. 40, 1910, p. 33. For origin of endothelium in the cat. Fleischmann, A. Embryologische Untersiichungen I, 18S9; Schulte, Mem. Wistar Inst. Anat. and Biol., no. 3, 1914. 9 Am. Jour. Anat., vol. 13, 1912.

'"Mollier, Op cit., p. 1051.


62


H. VON W. SCHULTE


collection of fluid and are modified to endothelium. I should prefer, therefore, to term MoUier's first stage of solid cords, so far as the mammal is concerned, simply mesenchyme. His second 'mesenchymatous' stage is really a mixed condition of mesenchyme and endothelial vesicles; it might be termed the stage



Fig. 9 Reconstruction of endothelial heart tubes and adjacent mesenchyme of a cat embryo of eleven pairs of somites. Columbia Collection, No. 534. X 300. Reduced one-h;iIf. 7, Bulb; 2, isthmus, corresponding to bulbo-ventricular sulcus; 3, ventricle: 4. umphalo-mesenteric vein; 5, atrio-venous angle; 6, mesenchyme between tul)es; 7, anterior intestinal portal; 8, oral plate.


FUSION OF CARDIAC ANLAGES IN THE CAT 63

of angiocysts. The tliircl is well designated as that of the endothehal tube with a continuous lumen but in this stage there may also be separate angiocysts and mesenchyme adjacent to the tube.

The heart of the embryo of 9 pairs of somites is an example of this last mentioned state. In it the endothelial tube on each side has a continuous lumen not of equal diameter throughout, it is true, for it has t\\() constrictions and its walls still show traces of their component angiocysts. The constrictions are located at the bulbo-ventricular sulcus and at the cardio-venous angle. The former reduces the lumen to a narrow dorso-^'entral cleft, the latter produces a smaller but quite perceptible diminution of the lumen. The bulb and ventricle are on the contrary dilated, and markedly so in their transverse diameter.

Between the bulbs corresponding to the narrow middle cardiac plate there is room for but little mesenchyme. Between the divergent ventricles, however, it is more abundant and stretches across between the endothelial tubes in plates and anastomosing strands, among which are scattered small angiocysts (fig. 2).

Conditions in the embryo of 11 somites (figs. 9 and 10) differ only in that the tubes are approximated and nearly parallel in their whole length. The slight asymmetry of the two sides is due mainly to the collapse of the shoulder of the left ventricle and in the cast of the lumen to a similar collapse of the right omphalo-mesenteric vein, which in other embryos of about this stage is actually a little larger than the left.

The bulbs are dilated and merge into the ventral aortic roots and the plexus forming about the foregut. As yet fusion has taken place across the median line only at one point situated well forward towards the oral plate. Elsewhere between the bulbs are strands of mesenchyme in which are two small angiocysts.

Corresponding to the bulbo-ventricular sulcus on each side is a narrow isthmus compressed laterally.

The ventricles are wide, prolonged into the shoulders of the mantles and diminishing caudad to their junction with the oinphalo-mesenteric veins. The right in four places shows rem


64


H. VON W. SCHULTE


nants of the partitions separating its component angiocysts. On the left but one minute one is present. The separated lumina at the shoulder of the left ventricle are due to collapse of the tube. Between the two ventricles is a net work of mesenchyme strands which have for the most part a longitudinal direction. In the model of the lumina but two angiocysts arejound in this area, one median in position, one close to the leffendo


Fig. 10 Lumina of endothelial tubes of eml)ryo shown in figure 9, same scale. 1, Bulb; 2, isthmus; 3, ventricle; 4, omphalo-mesenteric vein; 5, atrio-venous angle; 6, angiocysts.


thelial tube. The irregular contours of both tubes mesad suggest the addition of imperfectly assimilated angiocysts.

In embryos of 12 and 13 somites the mesenchyme of the middle cardiac plate is condensed into a single strand which occupies the concavity of the plate between the ridges which demarcate it from the mantles (figs. 11 and 12). In the model of the endothelium (not illustrated) this strand extends from the constricted regions or isthmi between bulbs and ventricles as far as


FUSION OF CARDIAC ANLAGES IN THE CAT 05

the anastomosis between the oniphalo-mesentei-ic veins. In it are three angiocysts. Two are very small; the more caudal of these is in process of annexation to the right ventricle, the other is attached by its wall to the left ventricle but does not communicate with its lumen. The largest angiocyst is elongated and has important connections. It communicates cephalad with the isthmian region of each bulbus and so establishes the first



^


<^^


r


i^


Fig. II Reconstruction of the luniina of the endothelial heart tubes and adjacent vessels of a cat embyro of twelve pairs of somites. Columbia Collection, Xo. 547. X 300. Reduced one-half. ^ Median angiocyst; ^, bulb; 3, ventricle; 4, atrio-venous angle; 5, omphalo-mesenteric vein; 6, forming aortic arches.

continuity of lumen between the endothelial tubes. In addition it has a small connection with the right ventricle. Thus the median angiocysts play a role in the coalescence of the endothelial anlages analogous to that of the middle plate in the fusion of the myoepicardial mantles. For the rest the changes accomplished in this stage are easily appreciated on comparison with the heart of the embryo of 11 somites (fig. 10). The cardio THE AMEKKAX JOrRNAI, OF ANATOMY, VOL. 20, NO 1


66 H. VON W. SCHULTE

venous angles are greatly deepened and the omphalo-mesenteric veins are brought close together. Between their approximated portions a small communication has formed. As a whole the right tube is larger, perhaps more distended than the left. The right omphalo-mesenteric vein is certainly larger than that of the opposite side.



Fig. 12 Photomicrograph of section of heart shown in figure 11. X 125. Reduced one-half. 1, Middle cardiac plate; 2, median angiocyst; 3, ventricle.


In the heart of the embryo of 14 somites as marked changes have supervened in the endothelial tubes as in the myocardium. The loop is well formed but in its entire length is double, being composed of two parallel tubes as yet only at the beginning of fusion. The general configuration is shown in figure 13, in which the model is seen from behind and slightly from above, bringing its caudal aspect prominently into view and to a less


FUSION OF CARDIAC ANLAGES IN THE CAT


()7


degree its vontml Mirfaco. 'Vhv drawing is greatly foreshortened. The bulbar segments are approximated but their lumina are still completely separated by an endothelial partition. The right bulb has a slightly greater cross-section than the left (fig. 14) and in its terminal segment the left is slightly irregular in contour as can be seen in the illustration of the model. These changes, though small, foreshadow the reduction of the left bulb



Fig. 13 Reconstruction of the lumen of the heart in a cat embrj-oof fourteen pairs^of somites. Columbia Collection, No. 188. X 300. Reduced one-half. 1, Right bulb; 2, left bulb; 3, left ventricle; 4, right ventricle; 5, fusion between isthmi; 6, fusion between ventricle; 7, fusion at cardio-venous angles; 8, left omphalo-mesenteric vein; 9, light omphalo-mescnteric vein.

in later stages. The isthmus of the left side is strongly kinked by the deepening left bulbo-ventricular sulcus, while the opening out of the sulcus of the right side places that bulb under more favorable conditions as regards flow. The area of fusion between the isthmi has greatly increased.

The ^'entricles are pyramidal in form, tapering toward the isthmus and caudad and prolonged in strong angular projec


68


H. VON W. SCHULTE


tions into tlie shoulders of the mantles. The projection on the left side is long and pointed, that of the right blunt and short conformably to the alteration in shape of the mantles at this



Fig. 14 Photomicrograph of section through the bulbs of embryo shown in figure 13. X 125. Reduced one-half. 1, right bulb; 2, left bulb; 3, shoulder of left ventricle myocardium.


period. Coalescence has been effected at about the middle of the ventricles. Immediately ^-entrad of this area is a region where the endothelium of the tubes is not yet in contact and a perforation extends through the heart from the convexity to the


FUSION OF CARDIAC ANLAGES IN THE CAT


()9


concavity of the loop (fig. 15). Elsewhere only a partition of endothelium separates the ventricles.



Fig. 15 Section through ventricles of the same embyro. X 125. Reduced one-half. 1, Area of coalescence of ventricles; 2, interval between endothelial tubes; 3, middle cardiac plate; 4, atrium.


Caudad the area of union between the omphalo-mesenteric veins has increased and extends upon the terminal portion of


'0


H. VON W. SCHULTE


the heart tubes, which here are dilated. This enh^rgement represents the atrium. It is displaced well to the left of the median line. This preliminary excursion in a direction opposite to that occurring in later stages probably depends on the larger size of the right omphalo-mesenteric vein and seems capable of play



Fig. 16 Reconstruction of the lumen of the heart in a cat embryo of sixteen pairs of somites. Columbia Collection, No. 551. X 300. Reduced one-half. /, Right bulb; 2, remnants of left bulb; 3, left ventricle; 4, right ventricle; 5, interruption of right tube; 6, atrium; 7, left omphalo-mesenteric vein; 8, right omphalo-mesenteric vein.


ing a decisive role in determining the direction which the cardiac loop will take. For the movement of the venous confluence to the left favors flow into the left ventricle as against the right. This becoming engorged thrusts itself strongl}^ against the shoulder of the left mantle, and the direction of the flow be


FXTSION OF ('AUl)IA(^ ANLAGES IN THE CAT 71

iiig ventrjiil -is well as ccpluihitl teiuls to tlirow the left mantle ventrad and so initiates the axial rotation begun in this stage. These changes of position on the part of the left heart would seem to entail as consequences the observed displacements of the right, the opening out of the right bulbo-ventricular sulcus, the reduction of the shoulder of the right mantle, and its rotation dorsad.

But there is an additional factor to be considered. The accentuation of the left bulbo-ventricular sulcus increases the compression of the isthmus of that side and in so far impedes circulation through it. This condition favors the engorgement of the left ventricle and so participates in producing the effects enumerated above. However, prior to these events a connection has been formed between the two isthmi close to their junction with the ventricles. When the left isthmus is compressed this communication serves as a collateral channel leading the blood stream into the right bulb which from now on exceeds the left in development. The distension of the right ventricle is maintained by the interventricular communication already described. There is need of some arrangement of this sort for the communication of the atrium with the right ventricle is of smaller caliber than on the left side. This also depends upon the shift of the venous end of the heart to the left with a consequent marked accentuation of the right cardio-venous angle and a diminution of angularity on the left.

The excess of the right omphalo-mesenteric vein over the left seems then the efficient cause of the displacement of the atrium to the left and this joined with the configuration of the tubes and mantles at the beginning of the process is capable of affording a mechanical explanation of the formation and direction of the loop.

Conditions in the embryo of 16 pairs of somites are corroborative of the findings in the embryo just described. The lumen of the heart is shown in figure 16. The right omphalo-mesenteric is the larger, the atrium is strongly displaced to the left and joins the atrio-ventricular canal almost at right angles. Here an important change has been effected for the right tube, atten


72 H. VON W. SCHULTE

uated in the embryo of 14 somites is now interrupted and represented in the model of the Uimen only by a pointed projection of the atrium. In the model of the endothelium, which was made of this embryo, there was also a solution of continuity at this point and in addition to the atrial protrusion shown by the lumen there was also a small projection of collapsed endothelium from the ventricle. The ends of the two processes were separated by a small but perfectly definite gap.

The fusion between the ventricles has progressed, especially caudad, and the two gaps in the line of luminal coalescence are tilled with epithelium. There is no longer a foramen leading between the ventricles as in the 14 somite stage. At the isthmus also the communication between the two tubes is greatly increased, and from this point on the functional bulb is that of the right side. The left is interrupted in its continuity and represented only by irregular projection of the lumen and patches of cells adherent to the wall of the right tube.

In later stages there is no trace either of the left bulb or the right atrio-ventricular canal and nothing in these slightly older embryos as far as they have come under observation is indicative of the history of these regions. The remnants of the interventricular partition persist awhile but are entirely effaced in an embryo of 21 somites.

The coalescence of the endothelial tubes in the cat is thus seen to be delayed until the cardiac loop has been completed. It is begun just prior to the initiation of the loop at points where the tubes are bent mesad. In the actual coalescence angiocysts developed in relation to the middle cardiac plate are involved. Two segments of the primitive tube are in part sacrificed — the left bulb and the right atrio-ventricular canal. The factor seemingly responsible for the direction of the loop is the larger size of the right omphalo-mesenteric vein with the consequent displacement to the left of the region of venous confluence.