408 WILSON. [Vot. III. 
of dissepiments (Fig. 80). Into this cavity projects the foun- 
dation of the cerebral ganglia, which is covered inwardly by a 
layer of mesoblast. 
In passing from above downward in a series of horizontal 
sections the prostomial cavity first appears (Figs. 55 and 56), and 
then a little further down the anterior pair of coelomic cavities 
come into view at the sides of the stomodzum. In slightly 
later stages all three cavities may be seen in one section, as 
shown in Fig. 58. I find it difficult to understand Kleinenberg’s 
account of the matter, and am forced to conclude that the two 
lateral cavities by the fusion of which he supposed the pro- 
stomial cavity to arise, are simply the anterior pair of trunk- 
cavities, and that he overlooked the partition separating them 
from the median cavity. Their true relations will be made 
clear by comparison of Figs. 50, 57, and 58. 
3. Blood-Vessels.— The first blood-vessel to appear is the 
ventral or sub-intestinal. It makes its appearance in the median 
ventral line, shortly after the fusion of the mesoblastic plates in 
this region, as aspace lying between the wall of the archenteron 
and the mesoblast (Fig. 72). At first it has no proper wall, 
being bounded below by the mesoblast and above by the ento- 
blast, so that it would seem to represent a part of the original 
cleavage cavity. Here and there along its course, however, a 
single cell can be seen on its dorsal side applied to the ento- 
blast. In later stages— or what amounts to the same thing, 
further forwards —these cells increase in number, so that the 
vessel becomes enclosed in a distinct wall of its own and appears 
to lie in the splanchnic mesoblast as figured in Fig. 25 a, Pl. XI. 
of Kleinenberg’s paper. The most careful examination has 
failed to satisfy me as to the precise origin of the walls of this 
vessel. But without being able to give absolute proof, I believe 
them to arise from cells which migrate out of the mesoblast. 
In any case the cavity of the vessel would seem to be of the 
same character as the lacunar cavities between the cells of the 
migratory mesoblast, which are for a time at least in connection 
with the head-cavity. 
As to the development of the dorsal vessel, I can do little 
more than confirm Kowalevsky’s interesting discovery (after- 
wards extended by Vejdovsky to Criodrilus, No. 43) of its double 
origin. Shortly after the appearance of the ventral vessel two 
