12 J. C4RAHA1M KEKR. 



wliole cavity is slit-like in form and is eminently suggestive 

 of having been formed by a process of splitting amongst the 

 large yolk-cells, after the manner described by Robinson and 

 Assheton in the case of the frog. Further evidence is 

 afforded in this direction by the fact that many sections 

 show the archenteric slit to end in a perfectly sharp edge 

 (fig. 9), which may even appear to be prolonged by division 

 planes along which the cells have not yet separated. Had I 

 had to rely upon a small amount of material, I should almost 

 certainly have described the archenteric formation during 

 this stage as being carried on by a process of splitting, I 

 have, however, examined now a very large number of series 

 of sections, and I am disposed to think that the process is by 

 no means one of simple splitting. In the first place, by 

 looking through complete series of sections, one as a rule 

 finds that, in certain sections, the archenteron terminates in 

 a clear rounded end (fig. 10). It appears impossible to me to 

 imagine that this can occur if the cavity is only extending by 

 a splitting process. Further, it is usual to find that, round 

 the tip of the archenteron, the cells have assumed a triangular 

 shape in section, with their tips towards the archenteron, 

 which strongly suggests the existence of a compressing force 

 acting round the tip of the cavity, and of such a nature as 

 would be caused by growth of either roof or floor of the 

 cavity. On the whole, I conclude that increase of the ai-ch- 

 enteric cavity does take place mainly by true invagination 

 during this stage aho. The slit-like appearance in many 

 sections may conceivably be an artefact due to the roof of 

 the archenteron being squeezed down against its floor by the 

 action of the fixing agent, or possibly the process of invagina- 

 tion may be aided by one of splitting. There seems nothing 

 improbable and indeed little of importance in this, notwith- 

 standing how much has been written on the subject. If it 

 does occur it is only another example of a very common 

 phenomenon in yolky eggs, — the formation by splitting of a 

 cavity elsewhere formed by invagination. 



As regards the probable cause of the invagination — beyond 



