FORMATION OF GERMINAL LAYERS IN CHELONIA. 39 



1. Under the primitive streak the entoblast is differentiated^ 

 so far as it has not done so already. 



2. In the middle of the primitive streak the canalis neuren- 

 tericus is sunk, at first perpendicularly below and then hori- 

 zontally forward. 



3. In the region of the primitive streak the ectoblast differen- 

 tiates from the mesoblast. This differs in the regions before 

 and behind the neurenteric canal. In front of the canal the 

 whole mass is differentiated into the ectoblast and the meso- 

 blast (i. e. mesoblast according to his views : we would call it 

 the chorda- entoblast). In the region behind the canal, only 

 the epidermal layer of the ectoblast is differentiated, the 

 differentiation of the remaining cells into the structures for 

 which they are destined : viz. the extreme end of the medullary 

 canal, of the chorda, &c., takes place at a much later date. 



As we stated before, we did not succeed in obtaining the 

 stages earlier than fig. 1. We will try, however, to reason back 

 from our earliest stages and to deduce what processes have given 

 rise to such a form. Of course, such a priori reasoning is 

 liable to mistakes, and we offer the following remarks merely as 

 suggestions which need verification by future investigations. 

 If the blastoporic passage really commences as an epiblastic 

 invagination, it seems to us that Kupffer is quite right in con- 

 sidering the invaginated sac as the gastrula cavity much reduced 

 in size (No. 5, p. 2). But apart from the inherent improba- 

 bility that the bottom of the archenteron should after- 

 ward give way and the archenteron should become connected 

 with some cavity beyond itself, we think we have another 

 sufficient reason in rejecting the view of an epiblastic invagi- 

 nation in this fact that directly beliind the passage, when 

 it is established, there is an area which is not covered by the 

 ectoblast, i. e. the yolk-plug. We think then that what really 

 takes place must be very much as Weldon and Strahl describe 

 it, for these two writers differ after all, when we leave out 

 minor points, only in this, that the former thinks the passage 

 arises at the front end, and the latter at the middle of the 

 primitive streak. Our views, then, on these earliest stages 

 are as follows : 



