FORMATION OF SHELLS. 397 



is difficult to imagine, therefore, how any communication 

 could have existed between the animal contained within the 

 shell, and the cancellated structure which forms its thickness." 

 Now, as to the formation of these structures, Dr. Car- 

 penter is at antipodes with Mr. Bowerbank in opinion. He 

 views the shell as being not analogous to bone, but as the 

 representation in the Mollusca of the cutaneous membranes, 

 — the view which was taken of their relation to the animal 

 by Cuvier and his followers.* We may pass over, in our 

 statement of the theory, the formation of the basement or 

 primary membranes by the successive production and coa- 

 lescence of cytoblasts and cells, for this is merely a part of 

 the general theory of the formation of membranes, but com- 

 mence with the process after it has proceeded beyond this 

 initiative. 



I gather, then, from Dr. Carpenter's essays, that he con- 

 siders the periostracum or skin of the shell to be cast off 

 from the animal as an epidermis, and to be, consequently, 

 inorganic. Beneath it an organized membrane or epithe- 

 lium, derived from the mantle, is then detached, and, by 

 the secretion of calcareous matter in its cells, is hardened 

 into a layer of shell, which layer is thus nothing else but a 

 calcified epithelium analogous with the enamel of the teeth. 

 By the rapid and successive production of similar layers of 

 exuvial membrane and of lime within its cells, the shell is 

 ultimately completed in the following manner. 



The shell of the animal in the egg consists, perhaps, only 

 of periostracum, and at most merely of an additional layer of 

 epithelium in which no lime has as yet been deposited.")- By 

 the growth of the animal a new edge of periostracum is 

 added to the rim of that already formed, and underneath it a 

 new rim also of calciferous epithelium, enlarging as a matter 

 of necessity the shell in the direction of the animal's growth, 

 and with an equal step. The thickening of the shell is at 

 the same time carried on in the parts already traced by the 

 deposition of calcareous layers, — a deposition which is the re- 

 sult of the casts of the secreting epithelium furnished by the 

 underlying mucous skin of the mantle. The periostracum 

 and cellular layers are then made by an extension of the 

 margins of the aperture or valves by a secretion from the 

 collar or edges of the mantle ; but the nacreous layers pro- 

 ceed from secretions furnished by the mantle itself. 



* Mem. xi. 8. 



t " Cos deux valves primitives sont d'abord purement membrancuscs. 

 Le compresseur les aplatit sans les rompre," &c. — A. de Quatrefages, 

 Emlwyogcnie des TurelS in Ann. (Its Sc. Nat. (1849) xi. 211. 



