﻿212 



PROCEEDINGS OP THE 



[April, 1857. 



arate the diverticula, while that of the less tentaculum is decidedly 

 shorter, fig. 4. The tentacula are rather stiff, commonly car- 

 ried slightly bent downward. They have the marked transversely 

 striate structure, which is so conspicuous in .ZEgenidae. 



I cannot doubt, from the identity of general form, of the number 

 of lobes and alternating tentacula, that this species is the same 

 as that found parasitic in the bell of Turritopsis, and which from 

 the analogy of Tubularia I was at first induced to look upon as 

 the young of T. nutricula. The principal difference between the 

 oldest form figured in pi. 7, and that here described pi. 12, fig. 4-5, 

 is that in this more mature form the diverticula are quadrate and 

 deeper, and the muscular lobes, instead of a single capsule and 

 tubercle, each, are provided with three capsules and three 

 tubercles each. This, I think, merely a difference in stage of 

 growth. If this be true, Cunina octonaria is probably first a free 

 swimming planule which seeks the bell of Turritopsis nutricula, 

 developes two tentacula, and a tubular mouth, which it uses to 

 draw food from the stomach of its foster-parent. It developes there 

 four, and at last eight tentacula, beneath which grows a medusa- 

 disk with concretionary corpuscles, then assumes the form of a 

 proboscidian medusa, becomes free, and at last by the shrinking of 

 the oral tube, becomes a flat Aeginoid medusa. 



The specimen figured is the only one I have taken at this state 

 of maturity. Its motion in swimming, though not very swift, is 

 very lively, being effected by many rapidly repeated strokes, and 

 has more resemblance to that of an Ephyra than of any Hybroid 

 medusa with which I am acquainted. It was taken in the 

 beginning of August. The larvae were observed in the" bell of 

 Turritopsis from the early part of July to the middle of September. 



Concluding Remarks. — At the close of these descriptions and in 

 review of the characters assigned to the various groups of genera, 

 the single remark should be made, that it is probable the greater 

 and more inclusive groups, viz: Corynidse, Velellidse, Tubularidse 

 and Siphonophorse, among Endostomata, with Campanularidae, 

 Sertularidse, among Exostomata, will, as our knowledge increases, 

 appear to be founded on less palpable distinctions than at present 

 seem to divide them from each other. These distinctions are 

 drawn principally from characters of the larvae, and in certain in- 

 stances we may already see that these groups of embryonic forms 

 exhibit a tendency to pass into each other by insensible degrees. 



