OF THE HEAD IN THE BATRACHIA URODELA. 



115 



otiier staofe does the first metaotic somite produce iimscle- 

 sub.stance. 



Text-fic 49. 



mt 



Right side view of a reconstruction of an embryo 3 mm. long. The first metaotic 

 somite appears below the developing anditory sac and the glossopharyngeal 

 epidermal thickening. Thi 2nd somite is closely applied to the 3rd. 



Conclusions. 



From the foregoing account it will be understood that in 

 almost every particular my observations confirm the conclusion 

 reached by Miss Piatt in her study of Necturus. In Ambly- 

 stoma as in Necturus there are three occipital segments. The 

 first metaotic somite, hovi'ever, disappears very early. Probably 

 the examination of an insufficiently complete series of stages 

 misled Sewertzoff (9) into the belief that there are only two 

 metaotic somites ; the first being either missed or later confused 

 with the second. Gaupp, in his excellent review of the develop- 

 ment of the skull (3), seems rather to favour Sewertzoflf's inter- 

 pretation. But if the diagram given by the lattez^, and repro- 

 duced by Gaupp, were correct, the second and third branchial 

 rami of the vagus would lie outside the head area in the first two 

 trunk-segments ; and the third bra.nchial ramus, passing behind 

 the last gill-slit, would then belong to the second trunk-segment 

 already provided with a complete spinal nerve possessing two 

 roots and a ganglion. This is obviously not the case. The 

 results of Sewertzofi" and myself are compared in text-fig. 50. 

 Only in unimportant details do my own observations difier from 

 Miss Piatt's. For instance, I find a ventral nerve-root in seg- 

 ment 3 and a dorsal root in segment 5. Moreover, the ganglion 



8* ^ 



