40 K. MITSUKURI AND C. ISHIKAWA. 



At the end of the segmentation the blastoderm becomes 

 divided into two primary layers, the ectoblast above consist- 

 ing of columnar cells, and the entoblast below consisting 

 of irregularly shaped cells without any definite arrangement. 

 At the region of the future blastopore and primitive streak, 

 this process of diflPerentiation is somewhat modified from what 

 takes place elsewhere. When the differentiation of the ectoblast 

 has proceeded backward and come to the future dorsal lip of the 

 blastopore, it does not extend further in the median line over the 

 blastodermic surface, but becomes reflected downward and con- 

 tinuous with the axial strip of the lower layer cells which acquire 

 the columnar character from this point forward in the median line 

 of the future embryo, and arranged themselves into the chorda- 

 entobiast. This process has proceeded to the front end of 

 the embryonic shield in the embryo represented in fig. 1. 

 Whether there is any actual invagination of cells from the 

 dorsal lip of the blastopore we cannot tell, but this is of no 

 moment so long as the ectoblast becomes continuous with the 

 axial strip of the entoblast at the dorsal lip, and the arrange- 

 ment of the lower layer cells into the chorda-entoblast pro- 

 ceeds from here towards the front. We can conceive the blas- 

 toporic passage itself arising in this way. As the cells arrange 

 themselves into the chorda-entoblast, these columnar cells 

 separate from the cells directly behind them and thus a fissure 

 or canal is produced just at the same rate as the cells arrange 

 themselves into the chorda-entoblast. The posterior wall of 

 this canal would thus be composed of undiflPerentiated cells, as 

 it actually is. 



While the diflferentiation of the ectoblast thus stops, in the 

 median line, at the dorsal lip of the blastopore, and the above- 

 mentioned changes leading to the formation of the blastoporic 

 passage are going on, we can suppose that the differentiation of 

 the ectoblast is at the same time proceeding actively in the more 

 lateral parts and is extending backwards and meeting in the 

 median line again slightly behind the blastopore (see fig. 6) . 

 There would thus be left behind the blastopore a small space not 

 covered by the ectoblast. This is the yolk- plug, which is of course 



