﻿April, 1857.] 



ELLIOTT SOCIETY. 



141 



is individually far more perfect than its analogue, and just as we 

 see it in Hydractinia, the medusa-buds are borne on the side-walls 

 of these untentaculated individuals, where they may be seen some- 

 times as dark specks, even with the unassisted eye. Imper- 

 fect otherwise, these individuals are provided with digestive 

 cavity and mouth, which perform their functions independently 

 of the great central polyp. The hepatic cells which exist in 

 these animals are probably the homologues of the dark-colored 

 cells which line the digestive cavity of many fixed hydroids. 



I have placed the Velellidee between the Corynidse and the 

 Tubularidse, not because I consider them as realty an interme- 

 diate group between these extremes, but because they have con- 

 nections with genera yet included in these groups, and very little 

 connection indeed with the Siphonophoras, with which they are 

 usually associated. The latter are more nearly related, I think, to 

 Tubularia and its immediate allies than to Velellidse, which, 

 indeed, constitute a very distinct family. We should specially 

 compare this group with Hydractinia.* 



There have been, hitherto, but two genera in this group. They 

 are founded on the differences of the hydroid entirely, and are 

 due to Lamarck. It is, however, probable that more genera will 

 he formed, so soon as we become acquainted with the medusa- 

 forms of more species of the Hydroid. That of Velella spirans, 

 from the Mediterranean, has been described above from the 

 description of Gegenbaur. 



The characteristics of Velella are a quadrangular shield, which 

 supports a diagonally-placed pointed crest. Its marginal tentaculi- 

 forra individuals are without buttons of thread-cells, but have rows 

 of such organs disposed longitudinally upon them. 



The characters of Rataria, a genus defined by Eschscholtz, are 

 a contractile, not a fixed crest, a disk or shell disposed longitu- 

 dinally, not diagonally with regard to the elliptical outline of the 

 fleshy border, and supporting the crest which consequently has 

 also this longitudinal position. Lastly, an absence of the mar- 

 ginal tentaculiform individuals of Velella and Porpita. The spe- 

 cies hitherto known of this genus are all small, none of them ex- 

 ceeding three lines in diameter. In view of this, it is not un- 

 likely that, as de Blainville supposed, they should prove to be 

 the young of Velella. 



* Prof. Agassiz, in his Nat. Hist, of 77. S. vol. 1st, speaks of a relation between 

 Hydractinia and the Siphonophorae. Vol. I, p. 72. 



