FORMATION OF GERMINAL LAYERS IN CHELONIA. 281 



front of the ventral opening of the blastoporic passage, which is still 

 visible as a groove in the median line. The darm-entoblast [end)^ 

 Avhich is distinct laterally, does not reach the chorda-entoblast, but 

 passes into a zone from which the mesoblastic sheet spreads away 

 laterally and which in its turn becomes continuous with the chorda- 

 entoblast. This corresponds to fig. 25 or fig. 29 of the anterior 

 region, or to fig. 20 of the previous stage. 



In fig. 31 the chorda-entoblast is beginning to be marked off 

 from the mesoblast, which is, however, still united with the darm- 

 entoblast, at least on the left side. This corresponds to fig. 21 of 

 the previous stage. 



In fig. 32 the mesoblast has become entirely separated from 

 both the chorda-and the darm-entoblast (excepting a little spot on 

 the left). The chorda-entoblast is now a compact mass by itself, 

 against the sides of which the darm-entoblast is applied. This is 

 more clearly shown on the right side than on the left. This corres- 

 ponds to fig. 26 of the anterior region, and to fig. 22 of the previous 

 stage. 



In fig. 33 the darm-entoblast has passed some way under the 

 chorda which has almost the appearance of the finished structure. 

 Thie corresponds to fig. 27. 



In fig. 34 the darm-entoblast has passed completely under the 

 chorda and forms a continuous sheet across, and the formation of 

 the notochord is finished. This corresponds to fig. 28. 



The formation of the chorda at the anterior region comes to an 

 end much earlier than in the posterior region, where it is continued 

 on until considerably later, and where the growth in length of the 

 embryo seems mainly to take place. 



It remains now to state the fate of the l)lastoporic passage. In 

 an embryo taken out tAVo days later than that given in figs. 4 a and h, 



