BY H. LEIGHTON KESTEVEN. 333 



posing that we are presented with shells absolutely congeneric, 

 as far as they themselves show, but having, the one an almost 

 planorboid protoconch, the other an elevated conoidal protoconch, 

 would we be justified in separating them generically 1 I would 

 answer the question by instancing the conoidal Polinices conicus 

 Lam., and the almost planorboid P. incei Phil., and still more 

 striking Littorina tieritoides Brown, and L. scabra Linn., and ask 

 do we separate these generically for the same reason. Moreover, 

 our hypothetical case is almost paralleled in the genus Cymatium 

 by species which are undoubtedly congeneric in shell, radula, 

 operculum and every point of anatomy. Thus we see that the 

 plan of coiling is not of generic importance as opposed to adult 

 characters. Again, it is sometimes found that the protoconchs 

 of a single genus differ in that the later whorls of some are 

 sculptured whilst the whole protoconch of others is smooth. The 

 sculpture in such cavses will be found to be merely the first indi- 

 cation of the adult ornamentation, and to be attributable merely 

 to extreme tachygenetic tendencies of the species possessing it, 

 rather than to be regarded as an essential difference. Lastly, in 

 the one genus there sometimes occur symmetrical conoidal pro- 

 toconchs, and others having a large nucleus placed excentrically, 

 with the later whorls coiled regularly. All protoconchs at their 

 inception are thimble-shaped, therefore, if the trochosphere larva 

 is large, and therefore also the early veliger, then will the 

 thimble-shaped portion of the veloconch be large in proportion, 

 it follows that when the asymmetry begins to appear the resultant 

 coiling will be more or less excentric at first, unless the diameter 

 of the shell-mouth increases very rapidly, that is, unless the 

 mollusc itself enlarges very rapidly. From this we see that a 

 large nucleus, excentric or regular, results from a large trocho- 

 sphere larva; that is to say, the size of the nucleus is governed 

 by the size of the <d^^^^. It would therefore be just as wrong to 

 regard it as a character of classificatory value as it would be to 

 classify the Aves by the size of their eggs. 



Really critical differences between two protoconchs would be — 

 (I) two absolutely different types of sculpture, which were not 



