270 N. ANNANDALE. 



pi. xvi, Vol. v., Mem. Ind. Mus., with figs. 14-16, pi. x, Mem. As. Soc. Bengal, 

 Vol. VI. 



Unfortunately we have very few precise data as to most of the specimens of this 

 species examined, but I think that we may safely take type A as the true marine 

 phase and assume that all the others are either from water of lower salinity than the 

 open sea or from water abnormal in other respects as regards chemical composition- 

 Further, the smallest shells examined, those found fresh or living in the Chilka Lake, 

 are certainly from water of very low and variable salinity and are very much smaller 

 than those found subfossil in the same neighbourhood. They are also distinctly 

 broader in proportion to their height, more nearly equilateral and in some instances 

 distinctly thinner. 



It is natural to compare what little we know of the plasticity of Area granosa 

 with Bateson's' well-known account of the cockle-shells of the Sea of Aral and the 

 Nilotic delta, for, though Cardium edulc and Area granosa are very distinct in ana- 

 tomical structure, there is a remarkably close superficial resemblance between their 

 shells. It must be noted, however, that the physical changes correlated with changes 

 in shell-form, etc., in the two species have taken place in opposite directions, for 

 while in the cockle these changes occurred as an apparent result of concentration of 

 the salts dissolved in the water in which it lived, in the Chilka Lake we are dealing with 

 a decrease of salinity.* It may be accepted as a fact that the phase of the Area that 

 survived in the lake in 191 4 was dwarfed and had a thinner, more transverse, more 

 swollen and le.ss inequilateral shell than that of those known in a subfossil condition 

 on the shores and islands ; but whether the production of this phase was directly 

 connected with decreased or more variable salinity, or due to some other factor such 

 as the deposition of alluvial silt on the bottom, we have as yet no information to 

 show. 



Shells of type B of Area granosa are not uncommon on the sheltered northern 

 shore of Barkuda but are usually broken and always more or less waterworn. A few 

 were also found in a fragmentary condition with those of Meretrix in the soil of the 

 island. I can detect no difference between these shells and specimens from the beds 

 at the head of Rambha Bay and from the shores of the outer channel of the lake. 

 Recent shells from the Ennur backwater near Madras are also similar, but are not so 

 readily separable into two subtypes. 



Fam. Veneridae. 

 Meretrix casta (Chemnitz). 



1917. Meyctrix casta (with vars.) Hornell, Rec. Ind. Mus. XJII, pp. 166-73, pis. iv, v, vi. 



The phase of this species found .subfossil on Barkuda may be included in the 

 race satparaensis as defined by Hornell in the paper cited. As to the status of this 

 race I cannot do better than to quote what he says : — 



1 Bateson, Phil. Trans. Roy. Soc. B, LXXX, p. 297 (1889). 

 '^ Annandale & Kemp, Mem. Ind. Mui., V, p. 5 (1915). 



