NUCULA. 147 



structive and elaborate examination of this problem will 

 be found in the essay of Professor Lacaze-Duthiers, 

 entitled "Natural History of the Purple of the Ancients," 

 to which I have referred in the Introduction to the first 

 volume (p. Ixvii), and which I shall again have occasion 

 to notice. There seems to be no doubt that all colours 

 are of mineral origin, and that they are secreted, by the 

 Mollusca and other animals, from inorganic matter by 

 special glands. But there is no necessary connexion 

 between the colour of the mollusk and that of its shell. 

 The plain variety of our native Cowry (Cypr&a Europ&a) 

 is a familiar illustration of this fact. While the shell 

 is of a uniform porcelain-white colour, its inhabitant 

 and fabricator exhibits most varied and brilliant tints of 

 vermilion, yellow, brown, green, and red. The coloured 

 markings of shells cannot be relied upon as a specific 

 test, especially when the same hue predominates; and 

 N. radiata is in this respect undistinguishable from 

 N. nucleus. Taking into consideration the question of 

 locality, with reference to the remarks which I made in 

 the Introduction to Vol. I. (pp. xix and xx), I believe it 

 will be found that the two forms do not live together. 

 I have never taken them in the same spot ; and Forbes 

 and M f Andrew have observed that N. radiata occurs 

 at Milford Haven " always in separate parts of the bay 

 from nucleus " This circumstance would of itself in- 

 cline me to doubt the propriety of separating these forms 

 unless as varieties. Sufficient weight does not seem to 

 have been given to the remarkable occurrence in the 

 same locality of different species of marine animals 

 which are bisexual but require mutual impregnation *. 



* Oppian speaks of this mode of generation with abhorrence : 

 Eire irpbs a\\r]\(Dv, Tepas dypiov, eicQvovTai, 



iroQtiJv, Kal voaipi ya/iwv, Kai v6ff<pi TOKOIO. 



H2 



