EEPOET OX THE PTEEOPODA. 
21 
organs, ancl of the embryonic shell, show also that these two species are most nearly allied 
to Clio, which is a more ancient genus than Cavolinia. Finally, the presence of a liver 
divided into two separate lobes distinguishes them from the other typical species of 
Cavolinia. 
These differences, upon which I did not lay sufficient stress in my systematic Report 
on the Thecosomata, lead me to regard the group A as a subgenus of Cavolinia, and the 
name Diacria, Gray, 1842, created for the species Cavolinia trispinosa , appears to me 
suitable for it. 
Family III. Cymbuliid^;. 
The animals of this family differ greatly, in appearance at least, from those contained 
in the two preceding families. The disposition of the various parts of the body, as com- 
pared with other Thecosomata, has already been explained. 1 
When a member of the Family Cymbuliidse and another form, one of the Cavoliniidas 
for example, are placed in corresponding positions, it is easy to see that their organisation 
is similar in all essential respects. 
The three genera of this family will now be examined in succession. 
1 . Cymbulia. 
The Head, as we have had occasion to indicate when speaking of Peraclis, differs 
from that of the above-mentioned Thecosomata, in the fact that it is distinct, situated at 
the dorsal side of the fin, and flattened down upon this latter without being free, as in 
Gleba. 
It is further characterised by its two symmetrical tentacles, of equal size and with no 
sheath at their base, as well as by the position of the orifice of the copulatory organ, 
which is in the middle line of the dorsal surface of the head, a little behind the 
tentacles. 
As regards the latter, Gegenbaur 2 throws doubt upon the existence of a nerve in the 
interior of these sensory organs. As we shall see, however, a nerve is distributed there, 
and terminates in a little ganglionic enlargement. 
The Foot is in the form of a large undivided natatory disc, extending ventrally as far 
as the head, which is bent backwards. 
The ventral lash-like appendage is not homologous with the posterior pedal lobe of 
the Cavoliniidse and Limacinidse. This latter, which also bears the operculum in the 
Limacinidse, corresponds to the posterior operculigerous part of the foot of the Gastro- 
poda. On the contrary, the filiform appendage of Cymbulia and of the larva of Gleba is 
1 ZooL Chall. Exp., part lxv. pp. 96, 97. 
2 Untersuchungen iiber Pteropoden und Heteropoden, p. 45. 
