BEYOZOA FROM ZANZIBAR. 509 



Hiiicksin his ' Brit. Mar. Polyzoa' says that the type of Cellejyora 

 WRSpmnicosa, but this is not the case, for as Levinsen points out in 

 his hii'ge work, Linnpeus refers to C. ramulosa as the first species 

 of Cellepora. Hincks incorrectly considered it the genus of 

 Fabricius. Levinsen is, however, mistaken in supposing that 

 C ramulosa is schizostoraous, as it belongs to the holostomatous 

 group, though not to Holoporella *. I have not as yet had any 

 oppoitunity of cutting sections of C. ramulosa, but the operculum 

 has the proximal border but slightly curved, and the small 

 muscular attachments are close to the thicker part of the border. 

 In most lioloporellce the attachment is a little neaier to the 

 edge, on the other hand it is much closer to the distal edge than 

 is usual in the Schizostomoiis groups. 



The ovicell of ravnuJosa is cap-like as in Holoporella etc., and 

 although Hincks says perforate or imperforate, none of the speci- 

 mens in my collection, nor any in the British Museum general 

 collections, including Busk's and Hincks's,' have a perforate 

 ovicell ; also in the Norman collection a few from each locality 

 were examined without finding any perforate. There are two or 

 three species externally corresponding with C. ramulosa, so that 

 a mistake is easily made, and in three cases fiiends have sent me 

 specimens so marked, of which only a part were ramulosa. 



By taking (even if provisionally) C. ramulosa as definitely 

 described by Hincks, and perhaps by some before him, as the 

 type of Cellepora, we get out of a difficulty, for when Schismopora 

 was created by MacGillivray Cellepora remained for the holo- 

 stomous division. C. ramulosa is the first of Linnaeus' species, 

 the others being spongites, ptimicosa, ciliata, hyalina. Linnaeus' 

 description of ramulosa would do for several species of branching 

 forms, and we are doubtful what the other species were meant 

 for. In Linnseus' copy of the 12th edition of Syst. Naturse, 

 there is in his small writing, under C. ramulosa, a reference 

 " nidros pi. i. fig. 6," which was hieroglyphic to me until 

 l)r. Daydon Jackson kindly explained that it refei'red to Det 

 Kongelige Norske Videnskabets Selskabs Skriften, 4th part, 

 1768-1774, in which there is a paper by Gunnerus, who supplied 

 Linnfeus with both C, ramulosa and C. jnomicosa. In this work 

 (pi. i. fig. 6) is a figure of a Cellepora, which, jvidging from the 

 locality, " Oceano Norvegico," is probably f 0. incrassata Sm., 

 although C. coronopus S. Woods, an entirely different species 

 from the Mediterranean, corresponds equally well with this 

 figure, which shoves nothing but zoarial shape. It is, however, 

 what LinufBus described as C. ramulosa. On the same plate 

 the figure 7 shows a similar growth, though with smaller 

 branches, and to this under C. j^umicosa Linnseus refers by a 



* Tlune ave various species of holostomous Bryozoa which do not belong to 

 Holoporella Waters, as, for instance, C. sardonica Waters, which will fall into 

 Holoporellidse. 



t Waters, " Hry. from Franz Josef Land," Journ. Linn. Soc, Zool. vol. xxviii. 

 p. 94 (1900). 



