STRUCTURE OF SHELLS. 391 



remains of exceedingly minute vessels or elongated cavities, 

 which measured 1 -20,000th of an inch in diameter, and which 

 are either the remains of minute inter-cellular vessels, or of 

 cavities analogous to the so-called corpuscular bodies of true 

 bone ; and here the discrepancy in the size of the organs is 

 not so great, as one of the smallest of the bone-corpuscles 

 measured 1 -9.346th of an inch, while one of the calcigerous 

 tubes radiating from it was 1 -45,456th of an inch in diameter." 

 Bivalved shells exhibit too great a variety in structure 

 to be described summarily in a paragraph, but in the greater 

 number the shell is lamellated, the lamellae being formed of 

 fleshy membranes whose secreting cells are filled and hard- 

 ened to stone by the deposition within them of carbonate of 

 lime. The cells may be distinctly columnar and placed at 

 right angles to the surface, or they may be less definite in 

 form and laid parallel with it ; and either structure may pre- 

 vail throughout the thickness of the shell, as in Pinna, or 

 may be combined to form it, as in Modiola. In the genus 

 Ostrea, the structure consists of alternating layers of fleshy 

 membrane and cellular structure, " which are produced in 

 succession from the inner surface of the shell." But what- 

 ever may be the modification 'of structure it is, in all 

 bivalves, permeated with Haversian canals, and with an 

 interstitial vascular system more or less developed. There 

 are, according to Mr. Bowerbank, other vessels in the appa- 

 rently stony shell. " The vessels which I have hitherto 

 described," he says, " are not the only ones which exist in 

 the shells of molluscous and conchiferous animals, for there 

 is scarcely a membranous film that can be separated from 

 the animal remains obtained by the maceration of these 

 bodies in a weak solution of hydrochloric acid, in which ex- 

 ceedingly fine ramifying vessels may not be observed, pro- 

 vided a sufficiently high power be used in their examination ; 

 but this is absolutely necessary to a successful investigation 

 of their structures, which requires powers varying from five 

 hundred to one thousand linear to ensure distinct and satis- 

 factory results. After the description of the tissues already 

 named, it is scarcely necessary to say that, as a matter of 

 necessity, there must be a free vascular communication be- 

 tween the animals inhabiting both univalve and bivalve 

 shells and their habitations, through the medium of their 

 points of attachment : but this fact is exceedingly difficult 

 of demonstration ; and, although I have used my best en- 

 deavours to trace the vessels both from the animal to the 

 shell, and from the shell to the animal, 1 have not yet suc- 

 ceeded in detecting them in their passage from the one to the 



