VIVIPAROUS PISHES OF THE PACIFIC COAST. 
447 
relation the cephalic end of the embryo bears to the blastopore as I was formerly, but 
I am inclined to think that in my paper on sex cells I mistook the primitive streak for 
the head. I am sure that in the embryo with three protovertebrse thus described I 
mistook the head for the tail. This misconstruction does not affect the results as to 
the early segregation of the sex cells, but changes their place of origin. The needed 
corrections to be made in the account of the sex cells will be made in the chapter 
bearing on this subject. At the closing of the blastopore the embryo usually consists 
of two layers — the ectoderm and the primitive entoderm. Each of these is several 
cells deep, and they both extend over the entire yolk, merging into each other at the 
blastopore. Just before the closing of the blastopore the entoderm is but one cell 
deep in all places but the axial line. Immediately after the blastopore closes cells 
are heaped up, probably in the region of the closed blastopore, certainly in the caudal 
region of the body. With this heaping up of cells the lower and upper layers become 
merged into a solid mass (figs. 55 to 59). Shortly afterward the mesoderm is split off 
from the entoderm over the whole of the egg except along this region (figs. 58 to 59). 
In some eggs (fig. 43) the mesoderm is not definitely separated from the entoderm 
after the closing of the blastopore. In others, on the other hand, the mesoderm is at 
least partly separated even before the closing of the blastopore. In all the sections 
the mesoderm is seen to be intimately associated with the entoderm, so there can be 
no doubt as to the source from which it is derived, although the ectoderm at this 
time is quite thick. Two eggs cut in nearly the same planes and representing the 
earliest and latest stages found between the closing of the blastopore and the three- 
protovertebrse stage may be described in some detail. The sections are at right angles 
to the median plane of the embryo and the first in each of the two embryos is tangential 
to the primitive streak or thickened caudal mass. 
Fig. 55 is the sixth section of the early stage and the first that cuts the yolk. 
The ectoderm is here several cells deep, being cut obliquely. The outer layer of cells 
is much lighter than the deeper layer — a fact due to the absorption of the surrounding 
ovarian fluid. The entoderm is also several cells deep, and at this place no distinction 
between entoderm and mesoderm can be made out. At the axis the entoderm is 
much thicker and so intimately joined to the ectoderm that but a very faint line of 
division is perceptible. The ectoderm at this point is reduced in thickness. 
Fig. 56 is the eighth section of the series. The relation of the parts to each other 
is very much as in the preceding figure. The axial entoderm forms, however, a more 
rounded mass. 
Fig. 57 is the tenth section of the series. This section, being more median, differ- 
entiates the parts much better. Over the ventral half the outer layer of ectodermal 
cells is enlarged and much less densely stained ; over the dorsal half no such differ- 
entiation is seen. Laterally the entoderm is thinner than ventrally and, in places at 
least, a distinct dividing line can now be seen between mesoderm and entoderm. The 
axial primitive entoderm is still as important a structure as in the preceding section. 
Fig. 58, which represents the twelfth section, differs little from the tenth. The 
primitive entoderm is still thinner laterally and the distinction between entoderm and 
ectoderm is still harder to make. 
Fig. 59 represents the fifteenth section. All the sections following this are dam- 
aged along the axial line. In this section some of the entoderm cells have evidently 
been misplaced at the embryonic axis, where it is impossible to detect any line sepa- 
