FURTHER STUDIES UN THE (TERMINAL LAYERS ETC. 45 



gut-diverticiilum in Fig. 2o, and th:it this part becomes eventually in- 

 corporated in the mesoblast. This obliges us to come to the conclu- 

 sion that the median part of Figs. 9 and 11, wliich shows a regular 

 epithelial arrangement — the part wliicii T liavc; hitherto called the 

 chor(hi-]iypol)]ast — does not give rise to tiu; chorda alone but has 

 potentially the elements 'of the mesoblast in it. I have tried to show 

 what I mean by shading in Fig. 11. The media. n unshaded part is 

 the part tliat gives rise to the chorda ; the lightly shaded stretch, im- 

 mediately adjacent to it on eacli side, which corresponds to the 

 similarly shaded part of Figs. 13-17, gives rise to the gastral meso- 

 blast and becomes finally incorporated in the mesoblast. The section 

 represented in Fig. 11 is therefore strictly comparalile to the well- 

 known section of Amphioxus reproduced in Fig. o9. According 

 to this view, the apparently insignificant features : the bulging down- 

 wards of the chorda-a/î/a^^3, and the slight bay on each side of this 

 median projection receive their explanation. 



I think, I have now sufficiently demonstrated the proposition I 

 started with, viz: In Chelonia, at least, only the median 

 part of what lias hitherto been called the chorda-hypo- 

 blast gives rise to the chorda dorsalis while the lateral 

 p a r t s of it b e c (3 tu e eve n t u a 1 1 y i n c o r p o r a t e d i n t li e meso- 

 blast. 



r will now rapidly go over different stages and give a brief but 

 connected history of the above occurrences, as well as of those points 

 not yet touched upon. 



I have already described in full the stage represented in Figs. 1 

 and la. It will be remembered that this embryo consists almost en- 

 tirely of two layers, except in the region behind the blastoporic passage, 

 where the peristomal mesoblast is beginning to form. This mesoblast 



