SEGMENTATION OF THE OVUM OF THE SHEEP. 247 



more advanced stage. If this were so^ the greater part of the 

 ectoplacenta would be of primitive streak origin, and not tro- 

 phoblastic, which would be adverse to the hypothesis I have 

 been tempted to put forward as the result of my observations 

 on the segmentation of the sheep^s ovum. 



From a renewed examination I am quite convinced that the 

 outskirts of the mesoblast surrounding the hinder end of the 

 embryonal area of the rabbit about the time the primitive 

 streak attains its maximum elongation, are in part made up of 

 cells of hypoblastic origin, though whether to such a large 

 extent as Bonnet describes for the sheep, and Hubrecht for 

 Sorex, I cannot say. 



In the sheep Bonnet says the first formed mesoblast cells 

 are those of hypoblastic origin, and in this case it is clear that 

 the primitive streak mesoblast does not extend far beyond the 

 embryonic area ; whereas in the rabbit the first formed cells 

 are undoubtedly tiiose of epiblastic origin, and if these are 

 very quickly removed from the immediate region of the primi- 

 tive streak, as seems probable, it is impossible at present to 

 determine the actual boundary between the two sets of cells. 



If the greater part of mesoblast which underlies the ecto- 

 placenta is of hypoblastic origin, and if the true primitive 

 streak mesoblast does not really extend beyond the inner 

 limits of the ectoplacenta — i. e. not further than the circle 4 

 or 3 in fig. 44 of the paper referred to above (3), — there would 

 then be no reason for supposing the ectoplacenta to be any- 

 thing but trophoblast, and the rabbit would agree with other 

 mammals in this respect. 



Dr. Robinson tells me that in the ferret the ectoplacenta is 

 formed, and attachment to the uterus thereby effected, before 

 the mesoblast has appeared beneath the primitive streak. In 

 this case there is, therefore, still less reason for doubting a 

 trophoblastic origin for the ectoplacenta. 



In my diagrams on PI. 18, I have shown how the true 

 amnion is in most cases formed entirely of true epiblast ; that 

 is to say, it can be shown to have been derived from the 

 embryonal area, and not the trophoblast. 



