SEGMENTATION OF THE OVUM OF THE SHEEP. 221 



&c.), but rather later in some (hedgehogs bats). According to 

 my hypothesis the epiblast never grows round the blastocyst 

 in Monodelphic Mammals. From Robinson I should differ in 

 supposing that the whole of the subzonal epithelium is hypo- 

 blastic instead of only the " greater part " of it, and further in 

 denying that the epiblast ever extends round a previously 

 existing hypoblastic vesicle. Both layers of the extra- 

 embryonic part of the wall of the didermic vesicle are on 

 this hypothesis hypoblastic. 



The accompanying diagram (p. 222) will render clear what I 

 believe to be a perfectly legitimate inter|)retation of the facts 

 derived from the study of the segmentation of the ovum and 

 formation of the embryonal area in the sheep and pig. 



Such a conception of the early embryo of the mammal is very 

 closely in accord with the meroblastic eg§ of the Sauropsida. 



If we regard the cavity of the blastocyst as the segmentation 

 cavity, it is then impossible to compare the blastocyst with the 

 egg of the Sauropsida. But if the cavity is regarded as the 

 equivalent of the subgerminal cavity the comparison is com- 

 plete. The origin and the fate of the subgerminal cavity of 

 the bird and reptile are exactly the same as the origin and fate 

 of the cavity of the blastocyst of mammals. 



In each case it begins as a split amongst hypoblast cells, 

 and becomes enormously enlarged and distended by fluid, and 

 provides ample space for the development of the embryo free 

 from pressure. A portion of it in each case becomes the gut 

 cavity of the embryo. The diagrams A, B, C, on PI. 18 repre- 

 sent the bird's egg during the first few hours of development 

 according to DuvaPs description. 



Fig. A represents an early stage of segmentation. The 

 segmentation cavity is seen as a narrow slit between the epi- 

 blast, which is coloured pink, and the hypoblast, which is dis- 

 tinguished by a green tint. The segmentation cavity does not 

 occur in the placental mammals, except perhaps in the case of 

 the rat (R/obinson) and sheep ; otherwise figs. A and C are 

 exactly comparable with the first and last of the diagrams upon 

 the following page. 



VOL. 41, PART 2. NEW SKRIES. Q 



