BRYOZOA OF THE CAPE VERDE ISLANDS. 25 



points, this genus again received further examination, confirming most of 

 the conclusions previously arrived at. 



The ovicells are generally of three forms * : — 



1st. Those that are free, with the oceciostome on the dorsal surface, as 

 Crista Edward. nana, d'Orb., C. biciliata, MacG-., C. howensis, MacGr., C. ker- 

 guelensis, Busk. C. infiata, Waters, C. cuneata. Maplestone. 



2nd. Those with a pomiform ovicell, as C. aculeata, Hass., C. ebumea, Hincks 

 (pars), C. elongata, Harmer (M. Ed. ?), all of which have an opening with 

 more or less of a slit and no pronounced tube ; then C. con/erta, Busk, and 

 C. sertularoides, d'Orb., have a tubular oceciostome. A funnel occurs in 

 C. conferta, Busk, which in many other respects corresponds with C. elongata, 

 Harmer. 



3rd. There are a number with piriform ovicells, most of which seem to 

 have a tubular oceciostome, as C. tubidosa, Busk, and t'.Jistulosa, Heller, with 

 large wide zooecia about 0\1 mm. or more in the aperture. C. operculataf, 

 Rob., C. pacifica, Rob., C. maxima, Rob., C. pugeti, Rob., have also elongate 

 pyriform adnate ovicells, but the zooecia are smaller, beiug about O07 mm. ; 

 then C. ramosa, Harmer, C. Holdsworthii, Busk, C. eburneo-denticidata, Smitt, 

 C. tenuis, MacG., C. eribraria, Stimp., C. genicidata, M. Ed. 



We can only give typical shapes of the ovicells as there is a certain amount 

 of variation — for instance, I have C. denticxdata, from Roscoff, sent to me by 

 Joliet, in which there are several long ovicells (Zeppelin form) almost 

 tapering at the end (see Busk, Cat. Mar. Poly. pt. iii. pi. ±. figs. 2, 3), whereas 

 from the same colonies there is one with a shorter ovicell and flattened distal 

 end. A colony of Harmer's elongata from the ' Siboga ' Expedition shows 

 considerable variation of the ovicells, so that they might be called pyriform, 

 pomiform, or wider. ( 'anu J gives the photograph of the fossil Crisia Corbini, 

 Canu, with a very large and wide ovicell, about double the width of the 

 branch, with the end flat. 



Jullien and Calvet § say that, they consider the "formules algebroides" 

 introduced by Smitt, and followed by Harmer, with one slight simplification, 

 to be valueless, and these have never appealed much to me, for in description 

 we must try and give the usual or typical characters, rather than take an 

 individual in such a genus. However, there may be cases where such a 

 formula may be useful, as, for instance, using it instead of an extra figure 

 or in sending particulars to a correspondent. The formula has, however, been 



* " Bry. from Zanzibar," Proc. Zool. Soc. 1914, p. 840. 



t Some of Robertson's species are only judged from the tigures, and this is the case with 

 C. tenuis, MacG. 



% "Bry. des Terrains Tert. des Environs de Paris," Ann. de Pal^ont. vol. ii. (1907) p. 104, 

 pi. xii. fig. 6. 



§ Jullien et Calvet, Bry. prov. des Camp, de lTlirondelle, p. 109 (1903). 



