438 FRANK E. BEDDARD. 
stance which does not stain. So far as I have been able to 
follow the formation of these colloid-looking masses, I am 
inclined to think that they are the product of a fusion 
between smaller droplets which appear in the interior of the 
cells. Fig. 29 illustrates the general aspect of a section through 
the diverticulum with the larger and smaller drops of secre- 
tion, and the remains of the columnar cells which are more 
perfect in parts. The secretion when formed does not seem 
to be evacuated into the interior of the numerous czecal pouches 
which constitute the extremity of the diverticulum, as it is in 
the spermatheca itself, but to remain where it was formed. In 
nearly all the viscous drops (figs. 29d, 30d) were embedded 
bundles of spermatozoa which were always very distinct in 
my preparations owing to the fact that they were deeply 
stained by the colouring reagents used (alum carmine and 
borax carmine). A superficial examination of such a section 
almost conveyed the idea that the spermatozoa were developed 
in the diverticulum, so close is their relation to the epithe- 
lium; there can be little doubt, however, that the function of 
these masses formed by the breaking down of the epithelium 
is to retain the spermatozoa within the pouch until ready for 
transference to another individual. The spermatopores of 
Acanthodrilus are not known, and we have no information 
whatever with regard to the processes of fecundation in that 
genus. It seems likely on a priori grounds that they will 
prove to be different from those of Lumbricus, owing to the 
far greater complication of glandular appendages connected 
with the reproductive ducts. 
In Pericheta intermedia, P. aspergillum, and P. 
Houlleti the diverticula are of a different structure from the 
spermatheca; in all these cases the epithelium lining the 
diverticulum is formed of low quadrangular cells staining 
deeply ; the epithelium of the pouch on the other hand is 
formed of tall columnar cells of which the nuclei only are 
stained. These facts would appear to indicate that the cells 
lining the diverticula are uot of a glandular nature. It is 
worth remarking that the structure of the diverticula in these 
