igo-j.] Records of tlic Iiuiiaii Museum. 389 



water of Indian ponds, for the " corpuscles " are found not only 

 in the closely allied S. reticulata but also in S. proliferens , a form 

 that I have frequently taken in the same pond as 5. alba. Some 

 peculiarity, structural or physiological, in the cells of the paren- 

 chyma is argued by their absence from 5. alba. Both 5. lacus- 

 tris and 5. alba vary greatly in external form ; but it is note- 

 worth}' that not only is 5. alba far more frequently devoid of 

 branches than 5. lacustris, but in the latter the branches appear 

 never to show any tendency to be laterally compressed — ^the 

 shape they always take in 5. alba, if they are present at all. 

 Very often the}^ occur in this species merely as ridges or irregular 

 projections on the surface, but frequently they are well developed. 

 Gemmules of 5. lacustris generally have a chitinous cup surrounding 

 the aperture ; such a cup is sometimes present in those of 5. alba 

 but often completely absent. 



For these reasons I think it advisable to regard S. alba 

 conventionally as a species distinct from S. lacustris, of which, 

 however, it is a close ally. 



My S. lacustris var. bengalcnsis is a synonym of 5. alba, 

 between the typical form of which and Bowerbank's S. cerebel- 

 lata I can draw no line, although Carter recognized 5. cerebellata 

 as a variety of his species. The arrangement, as well as the pro- 

 portions, of the gemmule spicules differs even in different gem- 

 mules of the same specimen, and I find that flesh spicules are 

 often present in one part of a sponge and absent from another. 



Specimens of S. alba were obtained during winter in salt water 

 in the Chilka Lake, Orissa, by Babu Oopal Chandra Chatterjee, 

 who has presented them to the Museum. They form a thin 

 layer, without a trace of branches, on and between the shells of 

 mussels {Mytilus striatulus) , are devoid of flesh spicules and have 

 larger and stouter skeleton spicules than any other form of the 

 species with which I am acquainted. Their finder tells me that 

 they were white in life. I name this form provisionally S. alba 

 var. marina, but it is possible that it is only a temporary phase. 

 In the Port Canning ponds 5. alba (bengalensis) was devoid of 

 branches in the winter of 1905-1906, but was profusely branched 

 in the succeeding cold weather, all the individuals of the first 

 phase having died down in the intervening seasons. It is worth}^ 

 of note that S. alba resembles S. lacustris not only in its struc- 

 ture and its variability, but also in being able to live in salt 

 water, a medium in which the latter species has frequently been 

 found in the Northern Hemisphere. 



Spongilla crassior, sp. nov. . 

 Subgenus Spongilla, Wierzejski. 



Sponge incrusting its support in a thin layer, very hard and firm, 

 of a yellowish colour, the external surface smooth, without 

 projecting spicules, the oscula situated on star-shaped areas, 

 the pores minute. Both vertical and transverse fibres of the 



