~> 



12 THE VOYAGE OF H.M.S. CHALLENGER, 



Membranipora dumerilii, Audouin. 



A specimen, marked Membranipora crassimarginata, var. incrustans, has the zooecia 

 irregularly oval-shaped, with a small triangular avicularium above each zooecium, 

 resembling the figure I gave of Membranipora dumerilii,^ but with the ovicells some- 

 what narrower. There do not seem to be any vicarious avicularia. 



Habitat. — Station 135a, ofi" Inaccessible Island, Tristan da Cunha, 75 fathoms. 



Membranipora umbonata, Busk, var. nov. {PI. I. fig. 21). 



There is a specimen from Station 320, 600 fathoms, growing on Ascidia meridionalis, 

 which, in the shape of the zooecium, of the peculiar erect avicularium, and the two spines, 

 corresponds with Membranipora umbonata ; but the calcareous granulated surface and 

 the small ovicell difi"er from the typical Membranipora umbonata. 



A small specimen of Membranipora, named Membranipora galeata, var. furcata, 

 from off Marion Island, does not seem in any particular to differ from Membranipora 

 cervicornis. 



Electra cylindracea, Busk (PI. I. figs. 13, 14 ; PI. III. fig. 23). 



Eledra cylindracea, Busk, Zool. Chall. Exp., part xxx. p. 78, pi. xxxiii. fig. 2. 



I was unable to understand this species from the description and figure, and it is clear 

 that the characters were not quite correctly appreciated. The large avicularia at the base 

 of the zooecia form a wide tubular chamber, from the walls of which grow hollow calcareous 

 cervicorne spines. To some avicularia there are as many as twelve such spines, and Mr. 

 Busk mistook these for " furcate chitinous processes," which he thought grew from the 

 base of the avicularium. The " clavate curved spines or horns " are curved in two planes, 

 so that it would be impossible to give a satisfactory figure. The same is true (as already 

 expressed by Mr. Busk) for the surface of the zooecia, which is very much involved in 

 consequence of the numerous raised spines at various levels. I have nevertheless tried to 

 give a figure, which, in supplementing those already published, may enable the structure 

 to be understood. It is, however, not characteristic in one particular, as the zooecia are 

 not usually immediately above those of the inferior row, but alternate or rather verticillate. 



There are rosette plates on the opercular side of the distal wall. 



It does not now fully correspond with the description of the family Electrinidse, 

 whereas it certainly seems to fall into the family Membraniporidse, whether we retain a 

 genus Electra or not ; but as it is now seen to belong to the same group of the Membrani- 



> Quart. Journ. Geol. Soc, vol. xliii. pi. vi. fig. 4. 



