672 MR. A. W. WATERS ON MEMBEANIPORIDJE. 



Flabellaeis roborata (ffincks). (PI. 48. figa. 10-11 ; 

 PI. 49. figs. 7-10.) 



This was called Membranipora by Hincks, but MacGrillivray 

 subsequently created the genus Craspedozoum, on account of the 

 bundle of chitinous tubes at the border o£ the zoarium, though 

 as bundles of tubes occur in such different forms as Diporula 

 marqinata, several Flustra and Euthyris, Caberea, Flustramorpha, 

 &c., I was unable to accept the genus and left it provisionally 

 under Membranipora (Ann. & Mag. Nat. Hist. ser. 5, vol. xx. 

 p. 183, 1887), and Hincks on the same grounds in 1892 (op. cit. 

 vol. ix. p. 331) also left it with Membranipora. 



MacGillivray places it in the family Flustridae, and Busk, in 

 his ' Challenger ' Eeport, did not recognize that it had been 

 described, and called it Flustra membraniporides. The internal 

 shape of the zooecia does not correspond with that of any Flustra, 

 nor do I now think that, on the ground of internal structure, it 

 can be left with Membranipora. 



The shape of the zocecia and the character of the rosette- 

 plates is the same as that of Menipea triseriata, Busk ; but as I. 

 have pointed out in this Journal (vol. xxvi. p. 2), Menipea tri- 

 seriata, fiahellmn, ternata, &c. must be removed from Menipea. 

 M. roborata is sub-articulated and the others are distinctly arti- 

 culated. I have provisionally called the group Flabellaris, as 

 M.Jlabellum is a characteristic member of it. 



If the name OeZ/wZarm had not been so bandied about it might 

 have been retained, but now this would cause confusion. 



The ovaria are near the distal end, close to the basal wall, but 

 situated about the same distance from the two side walls, and 

 consist of a large number of ova round the border of the ovarium. 

 The position is unusual. Near the ovarium there are small 

 groups of nucleated cells, connected by the general parenchym. 



As already said, the zooecia dovetail into one another (PI. 48. 

 fig. 11) so that the lower part of the younger zocecia comes 

 under the aperture of the older ones. The rosette-plate on the 

 distal wall spreads out at each end and nearly meets in the 

 middle (PI. 49. fig. 9). Perhaps it is derived from a form which 

 had two rosette-plates. There are internal denticles (PL 48. 

 fig. 10) which do not seem to have any knob or head, and some- 

 times look as if they were the continuation of the distal wall. 

 There is no bar across the avicularium. 



In the variety F. ligulata, MacGr., there is one avicularium 



