()\ THE (iA«Tliri.ATl()N l\ TK IKo.M Ï/OX. 19 



mass lying close to (he Ixmiulary groove has slipi)ecl a liltlc into 

 the segmentation cavity, as may be gathered by comparing the 

 points marked with | in Figs. 14 and lö. From these facts it 

 is not dillicult to inter that the com})onent cells in this part, 

 especially in the inner zone, have been nndergoing repeated divi- 

 sion. This increase in bulk of the inner mass must be one of 

 the causes of the upheaval of the conical eminence. 



Notwithstanding that there is detected as yet no trace of 

 an indentation which represents the commencing de2)ression of the 

 blastopore, the flattening of the macromeric hemisphere can safely 

 be regarded as the first step of gastrula invagination (p. 3). 



The relations of these parts are made still clearer in the 

 section through the median sagittal plane of a little older ovum 

 {Fi(j. 16) in which the blastopore had come into view in the 

 form of a nail-mark (compare with Figs, öa and ob). The 

 flattened area of the vegetative hemisphere is converted into a 

 large notch {bp.) ; this notch represents evidently the earliest 

 trace of the visible blastopore. I wish to explain in the following 

 lines the mode by which the notch is brought about, the posi- 

 tion where it ajipears and the eflects which are imparted by the 

 formation of it to other parts. In the first place there is little 

 room for doubt that the notch is formed not passively by down- 

 ward pushing of the conical eminence (c.em.), but by a gradual 

 infolding of the flattened field, which infolding is carried on by 

 its own activity and is the continuation of the j^i'ocess that caused 

 the flattening. The pushing in of the macromeric mass of this part 

 forms a large fold raised into the segmentation cavity ; and the 

 component cells which in younger stages formed the superficial 

 row of the flat field now take a radial arrangement. 



It follows, in the second place, that the blastoporic notch 



