1884.] 110 [Hyatt. 



Nat. Sci. Moscow, vol. x, pi. 2, f. 11). This stage is preceded by 

 one in which the larva presents an elongation of the body and a 

 narrowing of the aperture into a semblance of a " hypostorne," in 

 apparently close accord with this theory. 



Even the facts we have cited above in favor of the homology of 

 the fore and hind neuropores with the fore and hind openings of 

 the actinostome seem to have a similar meaning, since the simulta- 

 neous occurrence of other invaginations in Petromyzon shows that 

 some azygos openings must have arisen independently in the Ver- 

 tebra ta, and are not derivatives of the fore and hind neuropores 

 or of similar openings in the body of a coelenterate ancestor. 

 This is one view, but still another may be taken based on facts 

 tending to show that the neural infolding is itself notthehomo- 

 logue of an actinostome. The archenteric mouth of Peripatus 

 capensis, and the neural invagination of Vertebrata and its appear- 

 ance after the blastopore, and anterior to the primitive streak, 

 appear to favor this view; but on the other hand the medullary 

 plate and fold in Ascidia and Amphioxus appear on what must be 

 considered as the dorsal side of the planula, and the blastopore is 

 inclosed and transformed into the posterior neurenteric canal. 1 It 

 may be said by objectors, therefore, that the neural groove must 

 have occurred primitively on the dorsal side, and that the neural 

 groove is not an actinostome at all, but a dorsal fold closed off 

 from the mesenteron by the external wall of the body ; and this is 

 why the blastopore is situated posteriorly. 2 The presence of a fore 

 neurenteric canal may be assumed as has been done by Balfour ; 

 but it must be confessed that the moment one takes this position 

 the whole homology of the neural fold and actinostome assumes a 

 doubtful aspect, and it is at once apparent that an imaginary fore 

 neurenteric canal is a necessity of the theory, since otherwise there 

 is not exact correspondence between the parts of the neural invag- 

 ination and those of the actinostome. It can be claimed with 

 reason that no such canal exists in" the embryo, and cannot be 



1 Hatscheck (Arb. Zool. Inst Wien, vol. iv) distinctly calls the flattened side the 

 dorsum, and gives very nearly the same account as Kowalevsky of the transformations 

 of the embryo of Amphioxus; and also fully supports Kowalevsky's statement of the 

 mode of inclosure of the blastopore. 



2 It is interesting in this connection to note that Kupffer in his article Die Gastrula- 

 tion etc (Arch. Anat.et Physiol., 1884) seems to show in the clearest manner by his 

 sections the solitary origin and exclusively posterior location of the blastopore. 



