No. 3.] THE CELL LINEAGE OF AMPAITRITE. A471 
which later form the proctodaeum; so that there is, from first 
to last, no uncertainty about the location of the posterior end 
of the animal. 
In order to be sure that the ventral plate is formed by the 
lateral extension and concrescence of the sides of the somatic 
plate, I followed each cell division on the posterior hemisphere 
until the blastopore had nearly closed, and some of the somatic- 
plate cells from either side had actually come together in the 
mid-ventral line. From this time num- 
erous landmarks make it possible to 
follow the cells in groups to a much 
later period. 
The foundation of the trunk region 
is now laid, and the trochophore begins 
to elongate by a rapid growth at the 
posterior end, so that the distance 
between the prototroch and paratroch 
rapidly increases (Fig. V). 
Thus, all the ectoderm of the trunk _ 2 
is formed from the somatoblast a? or X F8¥ (A) 
of the thirty-two-celled stage. 
The points to which I have called attention will stand out 
more clearly if we compare briefly the axial relationships of 
Amphitrite with those of Nereis, as described by Wilson. 
1. In Amphitrite the prospective sagittal plane passes 
through & and D of the four-cell stage. In Nereis it coincides 
with the second cleavage furrow (Fig. I). 
2. In Amphitrite the cells which form the proctodaeum — 
terminal cells—are descendants of the somatoblast X. In 
Nereis they (the pigment cells) ‘are certainly in part the 
offspring of the primary mesoblast.”’ 
3. In Amphitrite those descendants of the somatoblast 
X which from the first lie farthest from the prototroch, main- 
tain this relative position, and form the posterzor end of the 
trunk; while those cells which lie nearest the prototroch at 
first also maintain this relative position, and form the anterior 
part of the trunk. Only a lateral shifting of the regions takes 
and no cells which 
’ 
place. There are no “posterior teloblasts’ 
