172 THE MECHANICAL FUNCTIONS. 



strength than is possessed by the porcellaneous shells, which, 

 in consequence of the tenuity and uniform intermixture of 

 the animal cement with the calcareous particles, present a 

 harder and more transparent, but at the same time more 

 brittle compound. It is these qualities, together with their 

 smooth enamelled surface, often beautifully variegated with 

 brilliant colours, and presenting altogether a close resem- 

 blance to porcelain, that have procured them the name they 

 * bear. 



When the transparency and brittleness of these shells are 

 very great, they have been considered as forming another 

 class, and they have been termed Vitreous shells, from their 

 making a nearer approach to glass. Some shells present in- 

 termediate textures between the membranous and the por- 

 cellaneous. 



All those surfaces of the shell on its outer side which are 

 not in contact with any part of the animal, are originally co- 

 vered with an epidermis:* which, however, is frequently 

 rubbed off by friction. 



The process employed by nature for the formation and en- 

 largement of the shells of the mollusca was very imperfect- 

 ly understood prior to the investigations of Reaumur, who 

 may be considered as having laid the first solid foundations 

 of the theory of this branch of comparative physiology. t 

 His experimental inquiries have fully established the two 

 following general facts: first, that the grow^th of a shell is 

 simply the result of successive additions made to its surface, 

 and secondly, that the materials constituting each layer, so 

 added, are furnished by the organized fleshy substance, 

 which he termed the skin of the animal, but which is now 

 known by the name of the mantle, and not by any vessels 

 or other kind of organization belonging to the shell itself. 



If a portion of the shell of a living snail, for instance, be 

 removed, which can be done without injury to the animal, 

 since it adheres to the flesh only in one point, there is 



* This membrane has been termed the Perlostracum. 



f Memoires dc I'Academie des Sciences, 1709, p. 367, and 1716, p. 303. 



