60 : HENRY BURY. 
convex, and extending quickly forward along the ventral sur- 
face of the stomach nearly in the middle line. Although the 
concavity of its posterior margin is not so marked as this 
anterior convexity, yet it is for a short time sufficiently 
evident to enable us to distinguish an ascending portion 
running up on the left side from the junction with the 
stomach, and a descending portion, continuous with the 
remainder of the intestine, on the right. (It is curious that 
Semon in all his figures represents the ascending portion as 
being on the right side.) 
The forward growth of the convex margin of the intestine 
reaches its limits in the young pupa, in which it extends 
nearly as far forward as the water-vascular ring, the polian 
vesicle being just internal to it (fig. 15). The combined width 
of the ascending and descending portions of the intestine, 
just where it bends over, very nearly equals that of the 
stomach. 
After this the importance of this part rapidly declines ; and 
before the ciliated bands of the pupa have wholly disappeared, 
the intestine is almost completely straightened out (see 21, 
pl. iii, fig. 23). 
Catom.—The question of the existence of an anterior 
enteroccl has already been discussed; and we have only here 
to consider the behaviour of the posterior body-cavities, which 
we left in Auricularia just meeting (but not uniting) on the 
ventral surface of the stomach, the left one having a tubular 
prolongation forward on the ventral side of the hydroccel. 
This is as far as the cavities can be traced in the living 
animal, the collapse of the stomach and increased opacity of 
the tissues rendering further observations, except by means 
of sections, almost impossible. Even in sections the changes 
are sufficiently hard to trace, owing to the rapidity with which 
they occur and to the excessive delicacy of the tissues. All 
through stage A there should not be a greater difference than 
three hours between the ages of the larve examined, while 
to avoid shrinkage the greatest care must be taken in every 
stage of preservation ; embedding is best done first in celloidin 
