278 Part III. — Eleventh Annual Report 



has been enclosed by the splanchnic layers (fig. 9), and there is now a 

 myocardium as well as endocardium. The endocardium is only one cell- 

 layer in thickness, while 'the myocardium is one or more layers thick. 

 The myocardial walls push the endocardium inwards, from above down- 

 wards till the dorsal and ventral layers are almost in apposition (fig. 

 9 en.). In embryos a day older (fig. 11, en.), the two layers of endocardium 

 are in close contact, as seen in median longitudinal section. In this 

 section the endocardium is apparently not connected with the myocar- 

 dium, but in neighbouring sections of the same embryo the endocardium is 

 connected with the myocardium, as in fig. 9. Till nearly twelve hours 

 afterwards the endocardium is only one cell-layer thick, but in fig. 1 2 an 

 increase is seen at the ventricular end of the heart, but this apparent 

 thickening may be due to the division between the dorsal and ventral 

 layers not being clearly discernible on the left side of the figure. 

 This is more likely to be the case, as older embryos of the eleventh day 

 (fig. 13) exhibit the endocardium as only a single layer in thickness at the 

 ventricular end of the heart. It is interesting to notice the relation of the 

 myocardium and endocardium from the end of the eighth till the end of the 

 eleventh day. At first the endocardial tissue is made to approximate dorsally 

 and ventrally, till it appears to lie as a solid mass in the myocardial tube, 

 and then its dorsal and ventral layers separate, and they form an endocardial 

 tube within a myocardial envelope. This latter separation or change takes 

 place simultaneously, with the indication of the division of the heart in bo 

 two chambers (figs. 12 and 13, v. and a). The heart at first does not 

 project ventrally, but lies in the angle (fig. 10) between the medulla and 

 the ventral portion of the mid-brain; as the development of the embryo 

 however advances, the protuberance becomes very distinct. This seems 

 to be the stage at which Kupffer first noticed the heart in Gasterosteus, 

 and it is the first stage generally figured by " those who have contented 

 themselves with observations on the living embryos. 



The pericardium is first seen as double, being in fact the lumen between 

 the dorsal and ventral layers of the mesoblast bounding the endocardium 

 laterally. I have been unable to trace the pericardium at so early a stage 

 as Zeigler shows in fig. 30, of an embryo salmon of thirteen days, but I 

 can corroborate his observations on embryos a little more advanced in 

 regard to the pericardial cavities, or what becomes the pericardium. 

 From embryos of the eighth day onwards, it is possible to trace in serial 

 section the ultimate destination of the two cavities enclosed within the 

 splanchnic layers of either side, and the pericardium is figured as 

 double in fig. 8, pc, and single in figs. 9 and 11. The walls of the peri- 

 cardium from the first are quite distinct, and the heart occupies the larger 

 portion of its space. 



The heart is suspended in this chamber, and on the ninth and tenth 

 days seems to be connected anteriorly with the walls of the pericardium. 

 Ventrally, the space between the heart and the pericardial wall is lessened, 

 and the heart is seen to press on the latter. 



The heart is continued into the sinus venosus (fig. 13, sv.) posteriorly. 

 This vessel is first noticed about the time when the first indication of the 

 division of the heart by a slight transverse constriction into auricle and 

 ventricle is shown. In fig. 12, at the right hand (auricular) end of the 

 heart is seen a proliferation, which may represent the incipient sinus 

 venosus ; certainly, a few hours afterwards, the connection of auricle and 

 sinus venosus is complete. Simultaneously the myocardium changes, 

 the posterior portion remains single-layered, while the anterior and ventri- 

 cular portion increases by a proliferation of cells from its wall. The 

 single layer of the posterior portion is continuous with the very much 

 thinner walls of the sinus venosus. 



