84 CEYLON PEARL OYSTER REPORT. 



It appears, then, that the basin-like character of the oscular area, which was 

 figured by BOWKRBANK (22) in the type specimen, and which appears to form a charac- 

 teristic feature of the adult sponge, is not developed until comparatively late in life, 

 and it appears also that its development is due simply to more rapid growth of the 

 surrounding parts. According to the classification adopted by SOLLAS in his work 

 on the " Challenger " Tetractinellida, this mode of development would necessitate our 

 placing the species in the genus Cydonium. SOLLAS himself, however, places the 

 species in the genus Geodia, having been insufficiently acquainted with the characters 

 in question. LENDENFELD, in his work on the Tetractinellida of the Adriatic, merges 

 the genus Cydonium in Geodia, and in this I must agree with him, for the present 

 instance demonstrates very clearly the impracticable character of a classification 

 which involves a study of the development, at any rate until our knowledge is much 

 more complete than at present. 



The general surface of the sponge is smooth and, perhaps, even glabrous, where the 

 dermal membrane is intact, but from a few points on the lower surface, in the 

 immediate neighbourhood of the attachment, a few long, hair-like spicules (mostly 

 anatrisenes) project. The colour of the general surface (in spirit) is purplish brown, 

 mottled with dull yellow, the pigment being lodged in an immense number of 

 granular brown pigment cells which lie between the layer of sterrasters and the 

 dermal membrane. Internally the sponge is nearly white. 



The cortical layer of sterrasters is about 1'2 millim. thick in the largest specimen, 

 and between this layer and the dermal membrane lies a zone about 0'27 millim. thick, 

 in which the cladi of most of the triseues are found. The manner in which the shafts 

 of these trisenes pierce the layer of sterrasters, so that their cladi come to lie 

 externally to it, is, as already pointed out by BOWERBANK, very characteristic of the 

 species, and forms a conspicuous feature even in the smallest specimen which I have 

 seen. 



The spiculation, as observed in the largest specimen, is as follows : 



(1.) Dichotriaenes ; shaft stout, about 3'2 millims. by 0'07 millim.; chord, 0'48 

 millim. to 0'64 millim. 



(2.) Protrisenes ; scarce (seen only in another specimen) ; cladi about 0'075 millim. 

 long. 



(3.) Anatrieenes ; shaft slender, about 4'2 millims. long; cladi about 0'07 millim. 

 long. 



(4.) Somal oxea ; about 2'67 millims. by 0'042 millim. 



(5.) Cortical oxea; about 0'3 millim. by 0'0083 millim. 



(6.) Sterrasters; about 0'13 millim. in diameter. 



(7.) Somal chiasters or spherasters ; about O'OO!) millim. in diameter. 



(8.) Choanosomal spherasters with spined rays; about 0'04 millim. in diameter. 



The spherasters with spined rays are found just beneath the layer of sterrasters 

 and appear to be characteristic of the species. The fact that the spination of the 



