EMBRYOLOGY OF THE SEA BASS. 
210 
thinuer center and a thicker peripliery ; then, curving round in the direction of arrows 
h and c, it gave rise to a marginal wiilst (see cut, Fig. 3). 
Still couliniug our attention to the embryonic pole of the 
blastoderm, we see, Fig. 41, PI. xciii (autero-posterior section 
through a stage slightly older than Fig. 40) that a tongue of 
cells grows out from the randwulst towards the center. The 
tongue {pr. h.), several cells deei), can be recognized with ease 
before there is any ingrowth at all round the rest of the blas- 
toderm edge. (Compare opposite halves of Fig. 41.) 
Round the rest of the edge the ingrowth is likewise, at 
least in most places, iireceded by the formation of a rand- 
wnlst, which, however, is inconspicuous. In Fig. 46, PI. 
xciv, a longitudinal section through the anterior pole of a 
blastoderm similar to Fig. 33, PI. xcii, the periphery of the 
section is seen to be slightly thicker than the inner part. 
From the thickened periphery the ingrowth of cells (v. mes.) 
has already started, though only the apical cell is as yet 
clearly separated from the upper layer. In the transverse sec- 
tion, Fig. 47, PI. XCIV (through a-h of Fig. 33), a randwulst 
can not be said to exist, the edge of the blastoderm from 
which the ingrowth (r. mes.) starts being actually thinner 
than the more central portion. 
The early condition of the ingrowing layer at the cardinal points of the blastoderm, 
posterior pole, anterior pole, and lateral i)oles, is shown in the three sections. Fig. 41, 
PI. XCIII (right half). Fig. 46, PI. XCIV, and Fig. 47, PI. xciv, and a surface view of the 
Flf 
3. Diaf^ramraatic section to 
illustrate lormation of rand- 
waist— r. 71;., randwulst ; a., api- 
cal lino. 
blastoderm at this stage is given in Fig. 33. 
The ingrowing 
under layer is known as the 
germ ring. Before going further I must dissent from a point in the description which 
Agassiz and Whitman (/. c.) give of the formation of the ring. In their Fig. 6, corre- 
sponding with my Fig. 47, the cells of the ring are represented as forming a differenti- 
ated unicellular layer extending quite out to the epidermic stratum. In their paper, 
which is a preliminary one, the transverse section is the only one given, and they do 
not describe the condition at the embryonic pole. I have not found that the under 
layer cells are differentiated from the rest of the blastoderm at the extreme edge. 
The perii)heral part of the blastoderm, both where there is a large randwulst. Figs. 41 
and 46, and none at all. Fig. 47, is an undifferentiated area, and the germ ring con- 
sequently starts at some little distance from the extreme edge of the blastoderm. 
I have not as yet mentioned the behavior of the epidermic stratum during the 
period of invagination. Like most authors I have found that this stratum does not 
share in the invagination. It is at no point continuous with the ingrowing layer, as 
all the figures show. The peripheral epidermic cells at the embryonic pole, however, 
act in a way which at least suggests the persistence of a tendency in them to take 
part in the invagination, though it is more probalfie that their behavior is due to some 
much less significant cause. The peripheral cells (m. ep.c. Figs. 40 and 41, PI. xoiii. Fig. 
45, PI. xciv), in this region during the early stages of invagination, are larger than else- 
where, and they project into the narrow ring-like space leffbetween the periblast and 
the incurving surface of the randwulst. Fig. 41 shows the ordinary amount of this 
lu'ojection, but blastoderms are met with in which the epidermic layer follows the sur- 
