EAELT DEVELOPMENT OE JULU3 TBEEESTEIS, 225 



one hand with one another, and on the other with the cells in 

 the interior of the yolk by means of fine processes of proto- 

 plasm. The cells in the interior of the yolk are the direct 

 descendants of the first segmentation masses. They constitute 

 the endoderm. Their fate is various. Some of them are em- 

 ployed in the formation of the keel, which T am about to describe 

 in the next section ; that is, in the formation of the splanchnic 

 and somatic layers of the mesoderm. Another part is employed 

 in the formation of the endodermal lining of the mesenteron, 

 while a third part remains in the yolk after the mesenteron 

 is formed, and gives rise to mesoderm cells, which are em- 

 ployed in the formation of various muscles and of the circula- 

 tory system. These cells will be mentioned again in the last 

 part of this paper. 



The flat surface cells enclosing the yolk constitute, as already 

 stated, the ectoderm, and give rise to the usual ectodermal 

 derivatives. 



With regard to the retention of the primitive connection of 

 the cells of the ovum until this stage, nothing of the sort has, 

 I believe, been described before, except by Sedgwick in Peri- 

 patus (17). The most important part is, it seems to me, not 

 the connection of cell to cell, but the connection of layer to 

 layer by means of processes of the cells. 



Formation of the Mesoderm. 



About the middle of the fourth day several of the stellate 

 endoderm cells approach the ectoderm, in the middle line of 

 what will eventually be the ventral surface of the embryo. Such 

 cells are shown in figs. 7, 8, 19. Fig. 7 is an earlier stage 

 than that shown in fig. 8. That the cells are really endodermal, 

 and are not divided off from the ectoderm, is, I think, con- 

 clusively proved by the shape of the cells which at this period 

 compose the ectoderm. They are flat and thin and the nucleus 

 is long and oval, and lies in the direction of the long axis of 

 the cell. I cannot believe that they would divide in the direc- 



