BRYOZOA FROM ZANZIBAR. 469 



and S. annulata were both obtained from the ' Challenger ' Station, 

 oif Bahia. In the British Museum ' Challenger ' specimen there 

 are six spines to the primary of the tuft, not a number as figured ; 

 there are no avicularia or ovicells, but there is a central spine in 

 the same position as the aviculaiium in S. glabra H. 



Loc. Port Phillip Heads [MacG.). Wasin, Brit. East Africa, 

 20 fath. (522), collected by Crossland. 



Stirparia zanzibariensis, sp. n. (PI. LXVIII. figs. 1,2; 

 PI. LXIX. fig. 14.) 



The stem throws ont branches which may bear a tuft 6-7 mm, 

 long, and the internodes of the stem are approximately equal. 



The zooecia are alternate and turn partly away from each otlier, 

 that is, the central line of the branch is raised so that the zooecia 

 slope laterally downwards. The zooecium is much wider at the 

 distal end than below, and the area is rather more than half 

 the length of a zooecium. The first zooecium of a tuft has 9-11 

 long spines, usually six on one side and four on the other 

 (or dorsal side) with the central spine long; then the second, 

 third, and sometimes fourth zooecium have several spines, whereas 

 the normal zooecia have usually only one stout spine at the upper 

 inner angle, though occasionally there is also one at the outer 

 angle. 



There is sometimes an avicularium to the second zooecium, and 

 this and the subsequent avicularia, which are long and narrow, 

 are situated close to the base of the zooecium. 



There are about 14-15 tentacles. 



On the lower part of the stem or stolon the radicles are some- 

 times replaced by capsules, similar to those described in my paper, 

 " Bryozoa from Rapallo," * and they may be filled with bright 

 yellow homogeneous contents. Levinsen f refers to similar 

 capsules as occurring in Bugida caliculata Lev. 



The ovicells are pedunculate, and there is a calcareous cover 

 over a part only, not exceeding the half of a globe, so that the 

 embryo is thereby but slightly protected, sometimes not at all. 

 The calcareous wall of the ovicell is made of plates deposited 

 from centres and looking like the shell of a turtle. 



The ovaria are central immediately proximal to the caecum, 

 and the testes fill up the proximal part of the zooecium. Ko 

 ovarium in my sections has more than one ovarian cell. 



The zoarial growth is similar to that of S. dendrograpta, sp. n., 

 with the long nodes as depicted in PI. LXVI. fig. 4 ; but the 

 tufts are longer, and both the stout spinous processes and the 

 long avicularia are distinguishing characters. There is no line of 

 chitinous thickening as in S. dendrogra'pta. 



Loc. West Australia, some imperfect specimens in my col- 

 lection. Chuaka, Zanzibar, 3 fath. (506), collected by Crossland. 



* Joiirn. Linn. Soc, Zool. vol. xxvi. p. 19, woodcut fig. 6 (1896). 

 t Morph. & Syst. Studies on Cheil. Bry. p. 102. 



32* 



