122 CEYLON PEARL OYSTER REPORT. 



grouping of the actines, and that we have here the first stage in the evolution of the 

 typical Spirastrella microsclere. From this point of view it is of great interest to 

 compare RIDLEY'S Spirastrella transitoria (16), from the Amirante Group. This is a 

 thinly encrusting species, with " spinispirular, extremely concentrated, composed of 

 only one entire bend," in fact, very closely resembling the aster of Hymedesmia 

 curvistellifera. As RIDLEY further observes, "in S. transitoria we have the spiiii- 



spirular almost in the form of the stellate, with which SCHMIDT and 



CARTER consider it to be homologous." 



S. transitoria is evidently closely related to S. curvistellifera, but differs in the 

 proportions of the spicules. Both species, together with H. tristellata, and possibly 

 some others, undoubtedly occupy a position intermediate between the typical species 

 of Hymedesmia on the one hand and of Spirastrella on the other, so that they might, 

 with almost equal propriety, be referred to either genus. 



R.N. 320 (Ceylon seas). 



Spirastrella, SCHMIDT. 



Massive Spirastrellidse with styli or tylostyli for megascleres and spirasters for 

 microscleres ; the spirasters usually forming a dermal crust. 



Spirastrella vagubunda, RIDLEY. 



1884, Spirastrella vagabunda, RIDLEY (16). 



There are in the collection a number of specimens which, while differing greatly in 

 external form, agree so closely in spiculation that I am obliged to regard them merely 

 as varieties of one and the same species, which appears to be identical with RIDLEY'S 

 Spirastrella vagabunda. This species was originally described by RIDLEY from 

 Torres Straits, but at the same time he assigned to it certain specimens in the British 

 Museum Collection which came from Trincomalee and the Galle Coast, Ceylon, a 

 fact which, of course, strongly supports my identification of Professor HERDMAN'S 

 specimens. The Trincomalee specimen referred to was very briefly described by 

 CARTER (33) under the name " Suberites ? sp." RIDLEY, having examined the same 

 specimen, suggested (loc. cit.) that it should be distinguished as " S. vagabunda 

 var. trincomaliensis" on account of certain slight differences in spiculation. In 

 Mr. CARTER'S cabinet, now in my possession, there is a preparation, labelled in his 

 handwriting " Suberitcs trincomaliensis," which is evidently from the specimen 

 examined and described by him. Subsequently, in 1886, Mr. CARTER (20) described 

 a sponge from the Mergui Archipelago under the name Saberites trincomaliensis, 

 identifying it with the Ceylonese form. 



Spirastrella vagabunda thus appears to be a widely- distributed and variable 

 species. RIDLEY'S Spirastrella congenera, from Torres Straits, is probably a mere 

 variety of the same, and perhaps, also, his S. punctulata, from Mozambique and 

 Mauritius. The principal characters of the species appear to be the dense, confused 



