1887] _ Metschnikoff on Germ-Layers. 349 
homologous with it. If therefore the latter represents a placula, 
so must the former. But in this case the later delamination of 
the Geryonidz, by means of transverse division of the blasto- 
sphere cells, could no longer be explained as the expression of 
a placula stage, because this stage would already have been 
passed. 
Suppose, however, the assumption be correct that a placula 
stage occurs in the Metazoa with an invaginate gastrula. Then, 
since the placula theory can only explain the formation of 
endoderm, which takes place by means of a transverse division 
of the cells of a proplacula, all those cases would remain inex- 
plicable where—the cell-division being exclusively longitudinal— 
the endoderm is formed by multipolar or local immigration. 
According to Biutschli’s theory, the blastosphere in animals 
that suffer delamination should be formed, as in Volvox, from a 
plate-like proplacula stage. He thinks, indeed (p. 423), the 
statements of Fol justify him in assuming the occurrence of such 
a stage in Geryonia proboscidalis. But the assumption is unwar- 
ranted, for the sixteen-celled embryo of the Geryonide is in 
itself a typical blastula, produced from an eight-celled stage pre- 
cisely as in Medusze which form their endoderm by a totally 
different method. The eight-celled embryo has likewise been 
produced by just such an equatorial division as occurs in other 
Medusze and in most Metazoa with an equal segmentation. 
Among the latter Sycandra raphanus, according, to F. E. 
Schulze’s account, most nearly resembles Volvox as regards the 
stages preceding the blastosphere. In this sponge is founda 
plate-like eight-celled stage, which, however, can be of no value 
to the placula theory, since later on in the development there 
occurs an invaginate gastrula. 
The morphology of the interesting Trichoplax adherens F. E. 
Sch. (32) and its relations to the placula cannot be seriously 
discussed at the present time, since it is impossible to decide, 
with even a show of truth, what significance must be attached to 
the several layers of this animal. From the histological differ- 
, ence between the epithelial coverings of the two surfaces of the 
body, we cannot infer that different germinal layers were involved 
in the formation of these coverings, any more than in the Sponges 
where the same layer, the endoderm, appears in the chambers as’ 
‘flagellate epithelium and in the central cavity as flat epithelium 
