DEVKLOPMENr OF THE HEART IN MARSUPIALS. 



491 



closure inwards, towards the axial line, of the fold of splanchno- 

 pleure limiting the gut. 



It is necessary now to consider the early development of the 

 heart in relation to the mode of closure of the gut described 

 above. It is evident, as Rouviere points out, that if gut-closure 

 be effected by the backward movement of the cardiac fold, no 

 ventral mesocardium will be formed at any stage in forms, 

 such as mammals, in which a continuous pericai-dial cavity is 

 present prior to head-fold formation. It remains to be considered 

 how the lateral primordia of the heart reach their position in 

 the dorsal wall of the median pericardium. Various authors, e. g. 

 Robinson (13), have assumed that as the head-fold forms, the 

 pericardium undergoes a complete reversal, so that its anterior 



Text-figure 24. 



Transverse sections tlirougli embr^-os oiDasynrus viverrinus,{^) Stage I. (7'5 mm.)i 

 (B) Stage II. (S'omm.), to show tlie direction of extension of the pericardial 

 cavitj". 



D.A. Dorsal aorta. ^,t!i. Endothelium. M.P. Medullary plate. 

 P.p.C. Pleuro-pericardial canal. 



wall becomes posterior and its ventral wall, dorsal. Of such a 

 process of revei'sal, the longitudinal sections figured here (text- 

 figs. 4, 5 & 12) give no evidence. Moreover, in the anterior 

 region of the pericaruium, the primordium of the heart on each 

 side of the embryo lies at or near the inner, medial margin of 

 its pleuro-pericardial canal, so that a reversal which affected the 

 anterior limb of the pericardium would indeed bring the heart 

 primordia into position ventral to the gut, but would carry them 

 also to the lateral mai-gins of the gut, a position which they do 

 not occupy. 



On tlie other hand, if we take into consideration the fact that 

 Proc. Zool. Soc— 1915, No. XXXIV. 34 



