486 MR. A. W. WATKRS ON 



avicularium with the lai-ge pore at the base and the minute 

 dorsal avicularia suggest its being separated as a variety. 



The structure of the vittse has not received much attention, 

 though Harmer * has aUuded to it ; but in this species it has 

 been possible to obtain some explanation. The vittse are sunken 

 perforated grooves in the calcareous wall, and along each groove 

 there is a cylindrical tube, and within this, from the pore-tubes 

 (the perforations just mentioned), organic cords spread out and 

 reach the upper free surface at definite spots or pores (fig. 10). 

 It thus seems that the vittae should be compared with pore- 

 chambers of many Cheilostomata in so far as there is indirect 

 communication from the interior to the water-surface, through 

 the vittfe. 



Loc. Prison Island, Zanzibar Channel, 8 fath. (505); Wasin, 

 Brit. East Africa, 10 fath. (500), collected by Crossland. Algoa 

 Bay and Natal (Brit. Mus.). 



Membranipora savartii Audouin, (PI. LXXI. figs. 1-4.) 



In my Repoi-t on the Bryozoa from the Red Sea (Journ. Linn. 

 Soc, Zool. vol. xxxi. p. 138), I refer to the astonishing amount 

 of anastomosing protoplasmic threads in a specimen from Zanzi- 

 bar, and as some from the Sudan are also very full, this seems to 

 be a specific character. It certainly seems strange to find such 

 an extraordinary quantity, for though in my collection there are 

 preparations of a large number of species showing the threads 

 exceedingly well, I have never seen anything approaching these, 

 and further study of the funicular cords is desirable. 



These threads are very abundant in zooecia with active poly- 

 pides having digestion in full activity. In these threads are 

 included small granular patches, either round or filiform, and 

 where the polypides have degenerated or are degenerating there 

 are large masses of this granular substance also surrounded by 

 and connected with the protoplasma (fig. 2). In earlier stages 

 the protoplasmic threads are in some cases surrounding the 

 granular cord (fig. 4) ; in others there are only one or two plasmic 

 thi'eads by the sides of the granular cord or mass. 



Tiie collecting together of these masses naturally suggests that 

 waste products are thus brought together and afterwards got 

 rid of. 



Loc. Zanzibar Channel from the under side of buoy (528) ; 

 Has Osowamembe, Zanzibar Channel, 10 fath. (504) ; Prison 

 Island, Zanzibar Channel, 8 fath. (505), collected by Crossland. 



?Membranipora armata Haswell («o?i-Koschinsky). (PI. LXYII. 

 fig. 10, & PI. LXXI. figs. 5-10.) 



Biflustra armata Haswell, " On some Polyzoa from the Queens- 

 land Coast/' Proc. Linn. Soc. N. S. Wales, vol. v. p. 38, pi. i. 

 fig. 7 (1880). 



* " Morph. Cheil.," Quart. Journ. Micr. Sc. vol. xlvi. p. 306 (1902). 



