﻿112 



PROCEEDINGS OF THE 



[April, 1857. 



inent in them than in the other; there are, at the same time, a 

 larger number of genera among them, which apparently never 

 have free Medusae. Now, on structural grounds, I think I may 

 distinguish these two general groups of Medusas, as follows: 



First, as to general form. Among the Sarsiadae and the correlated 

 groups, we find that the general form is nearly always a deep bell; 

 while in the Campanularian Medusas, and their relatives, the gen- 

 eral form of the disk has a constant tendency to being more or 

 less shallow; the few deep-belled species which exist are the 

 extraordinary forms. Both shallow and deep-belled species are 

 found in each group, but in one of them the deep-bell is the pre- 

 vailing form; in the other, the broad, shallow, cymbal-like or 

 watch-glass shape predominates. 



They differ also in the general character of important organs. 

 The digestive cavity in the Sertularian group (if we include the 

 Aeginidas, for which I shall attempt to show there is reason,) va- 

 ries through Aequorea and Staurophora, from the form of a de- 

 pressed polygonal chamber, which may be said to be imbedded in 

 the disk, to that of a flower-shaped organ, pendent from the vertex 

 of the disk concavity. But in this group the prevailing charac- 

 teristic of the organ is that it is comparatively shallow, and we 

 see this still hold good, even where, as in Liriope and Tima, it is 

 placed at the extremity of a long and habitually exserted pedun- 

 cle or proboscis. Very different is the case in the Tubularian 

 group. There, the prevailing character of the digestive cavity is 

 that of a more or less long cylinder or tube, never imbedded in 

 the disk, but always pendent from it. There are several forms, as 

 Sarsia, Slabberia, and Dipurena,* with long exsertile peduncle, but 

 here it is formed entirely of the digestive organ usually surrounded 

 by the generative glands, and never is traversed by radiate tubes 

 to the extent seen among Geryonidae and in Tima.* Next, with 

 regard to the radiate tubes, we find that in the genera allied to the 

 Campanularian Medusae, there is rather a tendency to a vegetative 

 repetition of them, as in Aequorea, Berenice, &c; while in the 

 other group they appear to be more nearly limited to a small and 

 definite number. We also observe that in the Campanularian 

 group, there is a strong tendency to the formation of sinuses in the 

 radiate tubes, while the junction of these with the marginal or 



*In Turritopsis, however, as I have shown already, there is an arrangement 

 somewhat analogous to that of Tima. Indeed, the Oceanidee, as a gro ip, are an- 

 alogous to Thaumantiads. 



