132 s. H ATTA. 



iviipffer' maiiitnin.s tlud tlie process of invag-ination takes place, at the 

 same time as tlie ditterentia.tion of the epiblast. lie states further : 

 " EJie der Vonjawj der Blaslodermhildang d<e Region der grossfen Zellen an 

 Gegenpo^ erreich' liât, erschein hereita der Blasioporm. ])as Blaslidasia- 

 dium is liier also ver],Urz(, die Bildunij der Blislospliiire coincidirt mil 

 dem Beginn der Ga^trulatim.^' 



Although I can not be positive ssboiit this point, my observations 

 tend to conhum the views of Kuptfe?-. Amorjg a niimljer of sections 

 belonoing to these stages, I can not iind a single one which lias the 

 roof of the segmentation cavity consisting of ti single layer of cells. I 

 have, however, one series in which I think the process of invagination 

 is just ready to l)egin. T have figured a section of this series in fig. 2. 

 11. Xlir. It will be seen that in this egg a portion of the side wall of the 

 seo-mentation cavitv is thinned out into one laver of cells wliich haAe 

 assumed a regular columnar sha])e. This is probably the point where 

 invao-ination 1)egins, and possibly corresponds witli the differentiated 

 region in fig. 10 of Knpffer's paper. The rest of tlie roof of the seg- 

 mentation cavity is still two to three cells thick. I have also some 

 efro-s, in which, while the gastrulation is going on, the roof of tlie 

 segmentation cavity is in some places one cell and in other ])laces two 

 cells thick. In such cases, wherever the roof consists of a single 

 laver, the cells assume a colamnar shape, different from the irregular 

 polygonal shapes of earlier stages. 



As to the manner in which the roof of tl\e segmentation cavity 

 thins out, previous writers seem to consider that it is brought about 

 l)y the inner cells of the roof passing round to the sides and to the 

 floor of the segmentation cavit}^, while the outermost layer of cells 

 becomes the definitive epiblast. If the upper cells passed really to the 

 sides aud to the floor of the segmentation cavity, they ought to be 



1. C. Knpffer: — Die Entwicklung von Petromyzon Planeri. Arch. f. Miki-os. Amt. Bd. 35. 



