1916.] Fauna of the Chilka Lake : Mollusca Gastropoda, etc. 359 



words to objects that sway freely in the water. The algae may be growing on stones 

 and of no great length and a few living shells were found apparently lying free on a 

 muddy bottom, but they may have been shaken from weeds by the net. Large 

 numbers of dead shells were noted in December on the shore at Rambha after a 

 strong breeze. % 



The shell in specimens of M.undulata from the Chilka Lake is thin, as a rule 

 semi-transparent and lightly tinged with yellowish-green ; markings when present, as 

 they usually are, are of a bright reddish-purple. There is considerable variation in 

 outline ; but the upper margin is always strongly elevated at or near the middle and 

 is sometimes subangulate : the exact position of this point is not always the same. 

 On account of the elevation the proportional depth of the shell is always considerable, 

 but in this character also there is much variation. The lower margin of the shell is 

 often quite straight, but perhaps more frequently very slightly concave : it is never 

 emarginate. In some specimens one valve is a little more inflated than the other, 

 but this peculiarity is sometimes so slight as to be almost imperceptible and may exist 

 in either valve. The type of Mr. Preston's M. chilkaënsis (pi. xv, fig. 5) is an indivi- 

 dual in which it is particularly well marked, but his fig are exaggerates the asym- 

 metry. 



The surface of the shell is usually devoid of radiating ridges, except immediately 

 in front of and below the umbo, where there are distinct transversely-striated costae. 

 Faint traces of similar costae are, however, often to be observed on the posterior 

 edge of the shell and occasionally extend along its whole length in a well-developed 

 condition. It is to this form that Mr. Preston has given the name var. crassicostata 

 (pi. xv, fig. 6). 



■ 



In the commonest type of colouration the shell is marked with zig-zag purple 

 lines, which run transversely and are frequently interrupted, and also with finer 

 straight radiating lines of the same colour. Lines of both kinds frequently disappear 

 almost completely on the lower half of the shell and the longitudinal ones are almost 

 always most strongly developed on the posterior half. Both kinds of lines may be 

 obsolete or even entirely absent and the whole surface of a uniform pale yellowish- 

 green ; shells of this type are not uncommon. On the other hand the purple lines 

 often develop into irregular blotches, and occasionally the whole surface, except the 

 extreme margin, becomes deeply suffused with purple pigment, definite markings 

 being indistinguishable. This type of colouration is, however, very rare. Photo- 

 graphs illustrating variation in colour-pattern are reproduced on Plate xv, figs. 1-4. 



The shells described by Dunker evidently belong, so far as colouration is con- 

 cerned, to the form commonest in the Chilka Lake, but Reeve has figured and 

 described a unicolourous specimen which appears to have been browner than any 

 in our collection. Our descriptions of colour have, however, been drawn up from 

 specimens preserved in spirit, in which the differences are much better seen than 

 in dried shells. 



Neither in colouration, shape, degree of asymmetry or presence or absence of 

 ribs on the surface are we able to find correlation of any kind, and specimens from 



