Içi6.] Fauna of the Chilka Lake : Mollusca Gastropoda, etc. 363 



length of shell and concavity of margin. The former are those which repose in com- 

 paratively short cavities with a smooth lining and straight or nearly so, while the 

 latter are esconced in deeper holes of irregular shape and are pressed either against 

 one another or against the walls. In each case the shell takes the shape of the space 

 it occupies; in some instances it forms practically a cast of that space and the degree 

 of concavity of the lower margin is most strictly correlated with the degree of curva- 

 ture of the surface against which it is pressed. 



In the Chilka Lake we noticed exactly the same phenomenon. Shells with a 

 straight margin, like the types of M. jenkinsi and M. cochinensis, were those which 

 were attached to flat objects such as the inner surfaces of oyster-shells, while extreme 

 forms such as M. annandalei were living in crevices in rocks or on uneven stones. 

 The byssus is always very short and the shell is pressed closely against the object of 

 attachment. In the case of forms resembling Mr. Preston's M. celator, we believe 

 that we are dealing merely with abnormalities produced by growth in unusally con- 

 fined spaces. The shell is always greatly thickened and eroded on the surface, either 

 all over or in parts. 



Colouration is to some extent correlated with environment, shells from rocks or 

 logs overgrown with algae being paler and greener than those on bare stone of a dark 

 shade, while those on the inside of oyster-shells are often quite pale ; the correlation, 

 however, is not of a precise nature. Specimens from some localities, e.g. the Cochin 

 and Madras backwaters, are browner than those from the Chilka Lake. They have 

 been named by Mr. Preston, M. cochinensis and M. taprobanensis respectively. 



The shell seems to be thicker in specimens from marine localities than in those 

 from estuaries and backwaters. 



All our specimens from the Chilka Lake are small, exceptionally large shells not 

 exceeding 20 mm. in length, while in many series none reach 15 mm. The largest 

 specimen from the log of wood to which we have referred above is 31*5 mm. in 

 length, others from the Andamans are scarcely smaller, while Reeve figures an indivi- 

 dual 39 mm. long and von Martens l notes that the largest he examined was 36 mm. 



It is clear, therefore, that all the individuals we found in the Chilka Lake are 

 dwarfed and we are convinced that our investigations were sufficiently exhaustive in 

 this respect to include the whole range of variation. There is, however, a small 

 series of specimens in Blanford's collection, labelled as coming from the Chilka Lake, 

 some of which are more than 35 mm. in length. It is unfortunate that no precise 

 information is available as to their provenance, but in general appearance they bear 

 a remarkably close resemblance to those on the log of wood referred to above; we 

 have good reason to suspect that they may have been introduced on driftwood. 



Specimens from the backwaters of Cochin and Madras are even smaller than 

 those from the Chilka Lake, rarely, if ever, exceeding 10 mm. in length, but in these 

 places they live in confined spaces between the valves of dead oyster-shells. Those 

 from the Chilka Lake oyster-beds are almost as small. Among those we have our- 



1 Weber's Zool. Ergebn. Nied. Ost-Ind., IV, p. 227 (1897). 



