1887] Metschnikoff on Germ-Layers. 427 
It is evident enough that the migration of some individuals 
from the surface into the interior, which was probably filled with 
jelly, does not exclude the occurrence of transverse division in 
other individuals. It is also evident that the superficial portion 
of a cell which has thus divided might, when differentiation 
began, retain its original position, while the central portion lived 
on in the interior as a phagocyte. 
The transitional form between the Metazoa and Flagellata has 
been called by me in a previous publication Parenchymella. I 
would like now to change the name to Phagocytella, for the 
reason that the latter suggests a very characteristic peculiarity 
of the form in question. The Phagocytella, as we have already 
learned, was characterized by the possession of two primitive 
organs, a kynoblast and a phagocytoblast, which were not as yet 
so sharply separated from each other as are the germinal layers 
of most Metazoa. Very probably the phagocytoblast continued 
for a long time to receive recruits from the kynoblast, in the 
shape of immigrating cells. As regards the development of 
Phagocytella we feel safe in supposing that the eggs (sexual 
multiplication must already have been acquired) underwent an 
equal segmentation, in which the divisions of the blastomeres 
followed the three dimensions of space; and that a blastoccel 
appeared very early, which was gradually filled up by immigrant 
cells and by the central portions of other cells that had suffered 
transverse division. 
The Phagocytella theory is in harmony with our morphological 
and physiological knowledge of the Sponges. Indeed, it was in 
the study of this group that the theory had its origin. The em- 
bryology of the Sponges exhibits almost more variety than does 
the development of the Medusz; at least, although far from 
being sufficiently studied, it reveals the several processes of 
immigration, delamination, and invagination. In general, then, 
what has been said for the Medusz will describe the various 
ways in which the endoderm is formed among the Sponges. 
The primitive organs of the latter group, however, have pro- 
gressed relatively but a short distance from their original state; 
it,is for this reason that it is difficult to homologize the layers of 
a sponge with the germinal layers of other Metazoa. The “ endo- 
derm” often shades into the “ mesoderm,” the two structures 
being in the most intimate connection (8); I therefore think they © 
