342 
SAMUEL RITTEXHOUSE 
stages of progress; and in fig. 21 the entodermal tissue has filled 
the entire cavity. 
According to Metschnikoff, in his description of unipolar in- 
gression or "hypotrope/' the entodermal tissue arises as a rule 
by bodily migration of ectodermal cells into the blastocoele, and 
not by a transverse division of the ectodermal cells, — the inner 
parts going to form entoderm and the outer parts remaining as 
ectodermal cells. In fig. 20, plate 2 Metschnikoff shows a cell 
in the process of transverse division; and in fig. 21 of the same 
plate two cells are so situated that one can easily infer that they 
may have arisen by transverse division of a single ectodermal cell. 
These figures are of Clytia and in his description of the same species 
• he mentions the cell in fig. 20 as the only one that he found in 
which transverse division occurred. This he seems to regard as 
an exception, and claims that as a rule the ectodermal cells in- 
crease by longitudinal division and migrate into the interior. 
My material for studying the formation of the entoderm in 
Stomotoca was scarce and it is not impossible to have misinter- 
preted the phenomena. However, I am inclined to think that the 
entodermal cells arise by a transverse division of the ectodermal 
cells, as Metschnikoff shows in the exceptional case of Clytia 
viridicans. Fig. 18 is drawn from the only section I was able to 
secure from preserved material showing the beginning of the form- 
ation of the entoderm; and that section was cut shghtly oblique, 
causing some doubt. A section of a little older stage and drawn 
with a higher magnification is shown in fig. 19. Here there are 
three cells that appear to have just divided by transverse division. 
Another reason which causes me to think that the entodermal cells 
arise by transverse division of the original ectoderm cells is the 
fact that the ectodermal cells in this region are practically as wide 
as those in the other parts of the blastula. This would not be the 
case if the longitudinal division occurred; for cell division is 
necessarily more rapid in the region where the entoderm is given 
off, and consequently the cells would be narrower. Unfortunately, 
because of scarcity of material, the exact cellular details of the 
formation of the entoderm will have to be left for future study. 
