No. t.]° 7ZHE DEVELOPMENT OF BALANOGLOSSTS. a7 
The second pair of body-cavities (Fig. 31) is also attached 
to the ectoderm, but not so closely. The separation into two 
layers — somatic and splanchnic —is indicated in the prepara- 
tion. Both the second and third body-cavities extend around 
the sides of the larva at their respective levels, but those of 
the right side are separated from those of the left side on the 
mid-dorsal and mid-ventral lines.! 
The first body-cavity, or proboscis-cavity, is seen in cross- 
section in Fig. 26. Its dorsal wall is found to be thickened. 
Surface preparations of the wall of this body-cavity show that 
the number of nuclei is much greater than before and each 
nucleus is smaller. The protoplasm of each cell has elongated 
to a narrow filiament with the nucleus at the centre. Examined 
on the inner surface, the fibrous cells are seen projecting in 
long rows into the interior of the organ. 
The relationship of the proboscis vesicle to the first body- 
cavity has become more intimate and more complex. The 
probosis vesicle is applied quite near to the exit-tube of the left 
horn of the body-cavity. The proboscis vesicle is an elongated 
sac with thin walls; it is applied to the dorsal and dorso- 
posterior wall of the body-cavity. Where the surfaces are in 
contact anteriorly the body-cavity divides into two prolonga- 
tions, one opening to the left of the median line by the ecto- 
dermal exit-tube, the other ending blindly near the surface. 
Looked at from the anterior end, a small portion of the surface 
of the proboscis vesicle is exposed between the prolongations of 
the body-cavity, but the greater part of the anterior wall of the 
proboscis vesicle is covered by the body-cavity just in front 
of its point of division. As looked at from the posterior end, 
the whole of the posterior wall of the proboscis vesicle is 
exposed. 
A side view of the proboscis vesicle and the surrounding 
body-cavity is shown in Fig. 32, taken from a living larva. 
The left prolongation of the body-cavity, with its exit-tube, is 
seen. Lying against this is the small proboscis vesicle. At 
1 My statement in an earlier paper as to the absence of these body-cavities at 
this stage must be withdrawn. Only cross-sections were examined and the body- 
cavities were overlooked. 
