SEGMENTATION OF THE OVUM OF THE SHEEP. 227 



the pink mass instead of between the pink and green. In tliis 

 case the amnion formation of Erinaceus would resemble more 

 closely that of Pteropus or Cavia ; whereas, as I have interpreted 

 it, it is more like Vespertilio. 



Such is briefly the hypothesis to which I venture to draw 

 the attention of those who are interested in the problems of 

 mammalian embryology. I cannot deny that there are diffi- 

 culties, but these do not seem to me to be formidable. 



Let it be noted that we have now to face the fact, based 

 upon actual sections, that there is in certain mammals a clear 

 separation of segments at an early stage into two groups, one 

 of which eventually completely surrounds the other. Until 

 the last few years this has not been so. It is true that van 

 Beneden in 1880 described this process in the rabbit. His 

 interpretation, however, led to many difficulties, which were 

 only partially removed by his subsequent modification of it. 

 I showed in 189^ that van Beneden's description, derived from 

 surface views and optical sections only, was not supported by 

 my actual sections, and urged that the phenomena of the 

 development of the rabbit's ovum were open to a more strictly 

 epigenetic explanation. Since then Duval's work on the bat 

 and Hubrecht's on Tupaia, and my own observations on the 

 sheep, have convinced me that such a process does occur in 

 certain mammals. 



What does this phenomenon mean? It must surely have some 

 most profound significance. Van Beneden and Duval maintain 

 that it is the division into epiblast and hypoblast, the former 

 being the external layer. Hubrecht considers it is a division 

 into epiblastic trophoblast and embryo. Sobotta (54) connected 

 the lighter appearance of certain cells in the early stages of 

 the mouse's development with the state of division of the cells. 

 It is, I think, quite impossible to explain the differences in my 

 specimens in this manner. The present hypothesis supposes 

 that in cases where this differentiation does clearly occur it is 

 a division into epiblast and hypoblast, the latter being the 

 external layer ; and that if this is or at some time has been 

 the mode of development in all placental mammals, it is then 



