SPONGES. 127 



recorded the same species from the vicinity of the Basse Rocks, off the south-east 

 coast of Ceylon. It appears, therefore, that the two species, melobcsioulex and 

 cannata, hoth occur in Ceylon waters. 



R.N. 118 (Stat. V., off Chilaw, 10 fathoms). 



Negombo, n. gen. 



Spirastrellidre consisting (? always) of tubular processes (? arising from a common 

 base). Megascleres smooth styli ; microscleres sanidasters. 



This genus may, perhaps, have arisen independently from some sanidastrose form 

 of Tetractinellid, but as regards its existing characters it is so closely related to 

 Spirastrella that it may be included in the same family. It is also quite possible 

 that its sanidaster may be merely a modified spiraster. 



Negombo tenuistellata, n. sp. Plate V., fig. 8. 



Sponge consisting of a group of short, rather thin-walled tubes of very variable 

 diameter, growing up close together, side by side, and more or less fused with one 

 another laterally. Each tube ends above in a single widely-open vent, ranging in 

 diameter from about 3 millims. to about 8 millims. All the tubes are broken off' and 

 widely open below, so that it is impossible to say whether or not there was a basal 

 mass from which they sprung, but probably there was. The walls of the tubes contain 

 a great number of large sand-grains embedded rather sparsely in them. The colour 

 (in spirit) is pale yellowish-grey, translucent ; the texture rather soft and flexible, but 

 fairly tough. The tubes do not vary greatly in height, the height of the longest 

 being about 31 millims., while its width in the middle is about G millims. ; the tube 

 next to it is of about the same height, but as much as 12 millims. wide in the middle. 

 The walls of the tubes are scarcely 2 millims. thick in the middle, thinning out 

 somewhat towards the margin of the vent and thickening slightly towards the base. 

 The outer surface of the tube-wall is rough, with more or less embedded sand-grains, 

 and also, between the grains, irregularly reticulate with slightly-projecting ridges ; I 

 have not been able to find dermal pores in it. The inner surface of the tube-wall, on 

 the other hand, is covered by a kind of dermal membrane, strengthened by a reticulation 

 of megascleres, and bearing numerous small pores, resembling dermal pores but 

 presumably exhalant, in the interstices of this reticulation. 



The main skeleton consists of long styli, not forming definite fibres but sometimes 

 collected into loose wisps. They mostly lie lengthwise in the thickness of the tube- 

 wall and are more abundant in the middle of its thickness than elsewhere. On the 

 inside of the tube-wall there is, as already indicated, a well-developed "dermal" 

 reticulation of styli, crossing one another singly, or in twos or threes, in all directions 

 parallel with the surface. On the outer surface of the tube-wall the dermal 

 membrane contains very numerous microscleres and the megascleres lie at a slightly 

 lower level. On both surfaces the dermal membrane is supported to some extent on 



