182 
sperm-ropes are moulded in the mouth of the pouch makes 
_ it probable that the bulk of the cementing matrix is in- 
troduced into them with the spermatozoa from the male organs 
of another worm. A very large gland, adapted in every 
respect to this function, is situated on the male efferent 
duct in both Tubifex and Limnodrilus, and its function has 
hitherto been in doubt. Claparéde regarded it as a seminal 
vesicle, though he admitted that he had never seen any 
spermatozoa in it, nor had he grounds for considering it 
vesicular. The gland may be called a ‘‘ cement-gland,” and 
its large size in Tubsfex and Limnodrilus is in correlation 
with the abundant and complete character of their sperma- 
tophors. In €litellio, which presents sperm-ropes, but has 
not this gland, the glandular portion of the male efferent duct 
itself is largely developed, and probabiy supplies its place. 
In Nais, where the sperm-ropes are very simple and small as 
compared with the Senuride (see this Journal, April, 1870), 
there is no glandular development at all in connection with 
the male efferent duct, which is of a perfectly simple mem- 
branous character, and very short. In Nas, therefore, it is 
probable that the secretion of the copulatory pouch alone forms 
the cementing matrix of the spermatozoa. 
It is not unusual to find (especially in the summer) very 
loose aggregations of spermatozoa in the copulatory pouches— 
such as that drawn in fig. 13—which are apparently ill-formed 
spermatophors. They are wanting in the compact, sharply- 
outlined character which the well-formed spermatophor 
presents, and this may be attributed to a deficiency in the 
secretion of the cement-gland, or to their not having been 
properly “‘ set ” in the neck of the copulatory pouch. They, 
and others less incomplete, are sometimes observed to adhere 
more or less to the wall of the copulatory pouch, which seems 
to show that a secretion from that wall enters into their forma- 
tion. This adherence to the wall of the pouch is especially 
noticeable in Nais, where the spermatophors seem sometimes 
to be actually continuous with the lining membrane of the 
pouch. Those of the sperm-ropes of the Senuride, which 
are well formed, do not exhibit this adherence. 
A difference is noticeable among the sperm-ropes of 
Tubifex rivulorum in their refractive power, and their colour 
(some being of a brownish tint), which is jin all probability 
due to the amount and quality of the secretion from the 
cement-gland, and perhaps, too, to their age. 
In most of the sperm-ropes it is easy to observe a striated 
structure, which is clearer in some specimens than others. 
Further observation shows that this striated structure is due 
