108 METAMOKPHOSES OF MAN 



on the development of oysters,* my own memoir on 

 the embryogeny of teredo, and the beautiful researches 

 of M. Lacaze du Thiers upon that of dentalium.f It 

 is in the latter that the most complex metamorphoses 

 are observed; these phenomena are of a far simpler 

 nature in the oyster, are still more so in the 

 anodon, and are entirely absent in other small 

 bivalve mollusks, which inhabit fresh-water ponds 

 and lakes. The larvae of all the marine members 

 of this group present at first the ciliated form, and 

 are naked ; but shortly afterwards they are furnished 

 with a shell, and with a rotatory apparatus, like that 

 of gastropods. This organ may be drawn within or 

 extended beyond the shell, at the desire of the animal, 

 which is in addition provided with a foot, which is 

 occasionally very long, and movable. By the assist- 

 ance of the latter, the creature can either crawl along 

 the sea-bottom or swim through the water, although 

 at a later date it may be soldered to a rock, like the 

 oyster, or fixed immovably to the chamber it has 

 constructed, like the teredo. 



* "Written by M. Davaine, and called " Kecherches sur la Kepro 

 duction des Huitres." These memoirs obtained the Academy's 

 prize for experimental physiology in 1855, and some of their results 

 have since been corrected by M. Lacaze du Thiers. 



t M. Lacaze du Thiers published a monograph on this peculiar 

 animal, and it is unquestionably the most complete work which has 

 ever been written upon a single mollusk (" Histoire de 1' Organisa- 

 tion, du Developpement, des Moeurs et des Kapports zoologiques du 

 Dentale "). I am sorry that I cannot give some of the details of 

 this creature's history ; but on account of its very exceptional 

 character, it does not come within the limits of my present scheme. 

 I may, however, state that, at a certain period of its life the larva 

 of dentalium is so like that of an annelid that it may be mistaken for 

 one, and that its shell, although symmetrical at first, is always univalve. 



