BRYOZOA FROM ZANZIBAR. 495 



CbLLARIA GRACILIS, var. TESSELLATA, nov. (PI, LXVII. fig. 7.) 



For synonyms of Cellaria gracilis see Miss Jelly's Catalogue 

 and add : — 



Meissner, " Lists der von Herren Pi-of . Simon bei Amboina und 

 Thursday Island ges. Bry.," Jena Denkschr. vol. viii. p. 730. 



A specimen from Ras Osowamembe growing on Hydrozoa 

 seems to be a variety of C. (jracilis. In this specimen the trabe- 

 culte mentioned by Busk enclose an ai'ea formed of large tessellated 

 divisions, about 20 (fig. 7); and curiously, another species, which I 

 ca.ll C. wa^inensis, sp. n., has also a tessellated area, but the 

 avicnlarium of that species is triangular with an acute mandible, 

 and belongs to the C. tenuirostins group. These large tessellated 

 ai-ens are unusual, and have not been found in various species of 

 Cellaria in my collection ; nor in the British Museum ' Challenger ' 

 collections ai-e they found in C bicornis B., G. duhia B., C. malvi- 

 nensis B., C. variabilis B., G. dAvaricata B., G. australis Hincks, 

 G. rigida MacG. The significance both of the trabeculse and of 

 the divisions is at present obscure. 



The opercula are a trifie larger than the type from Holborn 

 Island, but the mandibles are the same. 



Loc. Type : Cumberland Island ; Cape Capricorn ; Victoria, 

 8 fath. ; Torres Straits; Holborn Island (Queensland); Katow, 

 New Guinea, 7 fath. Variety : Has Osowamembe, Zanzibar 

 Channel, 10 fath. (504), collected by Grossland. 



Cellaria wasinensis, sp. n. (PI. LXVII. figs. 1-6.) 



Zoa.rium about 25 mm. high, with small branches dichotomously 

 jointed and connected by two or three stout straight chitinovis 

 tubes and a " knot." The zoarium swells out where the ovicells 

 occur, as is frequently the case in Cellaria, there being sometimes 

 two such swellings in an internode ; and it will be seen on fig. 6 

 how these are equidistant from the joint in the two branches. 



The zooecia are rather elongate, hexagonal, or rhombic, with the 

 lateral walls of the zooecium much raised, surface finely granular ; 

 oral orifice wide, with two teeth on the distal edge, which is 

 slightly turned up, forming a lip. 



The ovicellular aperture is large, round, with a plate from the 

 proximal border spreading out and partially closing the apei'ture. 



used in saying the first is not the type, but surelj^ it never meant that any individual 

 could subsequently pick and choose which was the t.ype among those agreeing 

 equally well. However, in my work, to which Norman refers, I and other specialists 

 were under an obligation to follow the rules of the Zoological Congress, and the rule 

 in question is "other things being equal the name is to be preferred which stands 

 first in the book or article." In anything I write now the same obligation does not 

 exist, though the rule entirely commends itself, and it does not seem to clash with 

 the rules previously mentioned, drawn up by Strickland. Both practically say that 

 when there are adequate reasons to indicate which was the type intended, that maj' 

 be considered the type although not first mentioned, but without good cause to the 

 contrary the first is the type, and independently this is what workers are constautb* 

 doing. 



