REPORT OX THE PTEROPODA. 
29 
The most primitive form among the living Thecosomata has scarcely been sought for 
in any special manner, and opinions differ very greatly on this subject ; some think that 
it is to be found among the Cavoliniiclse {Clio, subgenus Creseis), others are of opinion 
that it belongs to the family Limacinidse. 
In order to arrive at a positive result on this question we shall study the mutual 
relations of the three groups of Thecosomata (Limacinicke, Cavoliniidse, Cymbuliidse), 
comparing their organisation. 
These relations are not very easy to explain, taking into consideration the great 
apparent differences presented by the three above-named families — differences which have 
not been sufficiently considered hitherto in the relationships which the three groups bear 
to each other. 
1. If in the first place we consider the Cymbuliidse, we find that their affinities are 
very obscure. Boas 1 seems to regard them as specialised Cavoliniidse, and for my own 
part, before I had had the opportunity of studying the organisation of the genus Peraclis, 
I was in a state of the most complete uncertainty regarding their relationships. The 
knowledge of this genus, however, has thrown some light upon their affinities. 
Apart from the presence of the “ cartilaginous shell,” which has no homology with 
the calcareous shells of other Thecosomata, we may see that the Cymbuliidse differ from 
the Cavoliniiclse by very definite characters, particularly in the shape of the head, which, 
in the former, is very distinct and quite symmetrical as regards the tentacles, which have 
no sheath, and in the arrangement of the central nervous system, which has three visceral 
ganglia instead of two closely placed as in the Cavoliniiclse. 
On the other hand, we have seen that among the Limacinidse, Peraclis, which, in all 
other respects resembles the Cymbuliidse quite as much as does any of the Cavoliniiclse, 
has a distinct head agreeing with that of the older larvse of the Cymbuliidse in its general 
form and also in the symmetry of its tentacles, which are further devoid of a sheath ; 
besides this the nervous system is constructed on the same type and has three visceral 
ganglionic masses arranged in the same manner. 
Of all other Thecosomata, then, Peraclis is the one which most closely resembles the 
Cymbuliidse, and with which this family has the closest affinity. 
2. On the other hand, the Cavoliniiclse, as well as the Cymbuliidse, have undoubted 
affinities to the Limacinidse, but these are with the genus Limacina, which presents 
numerous resemblances to the subgenus Creseis of Clio. The head is indistinct and has 
the same asymmetrical arrangement as regards the tentacles and penis ; besides which 
the fins in Creseis present the small tentacular lobe which is found in many Limacinse. 
If, however, the Cavoliniiclse and Cymbuliidse, which are not directly connected with 
each other, have each of them close affinities with the Limacinidse ( Limacina or 
Peraclis), these relations between the straight Thecosomata (Cavoliniidse and Cymbuliidse), 
1 Spolia atlantica, p. 188. 
