84 SAXICAVID^. 



siderably larger than these shell-fish require in order 

 that they may freely open their valves. This gives 

 room for the foot to expand and work. The side 

 of the shell in all such cases is often more or less 

 rubbed or worn^ in the same Avay as the spinous 

 fringes of Pholas dactylus, in which the last-formed 

 rows are uninjured; and the epidermis is seldom pre- 

 served on that part. In specimens of the typical form 

 of >S^. rugosay excavating limestone in Shetland, the 

 ventral and exposed border of the mantle has sometimes 

 delicate sessile Foraminifera [Truncatulina lobatula and 

 Disco7'bina globularis) living on it, which proves that the 

 mantle is not the organ of attrition. If an acid were em- 

 ployed by the Saxicava in dissolving calcareous rocks, it 

 would assuredly destroy that portion of the shell from 

 which the epidermis had been removed, as well as the 

 shells of the Foraminifera. The edges of the excavation 

 are sharply defined, and present an appearance very un- 

 like that which would be produced by a solvent action. 

 Therefore, either the shell or the foot must be the opera- 

 tive agent. Were it the former, the epidermis in front 

 would be entirely abraded; and such is never the case. The 

 Saxicava do not work, if they can meet with ready-made 

 holes. The late Dr. Lukis, in one of his letters to me, 

 said, " Successive generations will occupy the same hole. 

 The last inhabits the space between the valves of its 

 predecessor. In this way four or five pairs of shells may 

 be frequently seen nested one within the other, and not 

 unusually a Sphenia Binghami in the centre of all."^ 

 Cailliaud observed a Saxicava within a specimen of Vene- 

 rupis Irus, which it had perforated. Malm found a cylin- 

 drical variety in the burrows of Limnoria lignorum. The 

 form of the shell is so variable and dependent on habitat, 

 that (as the late M. Bouchard-Chantereaux remarked) 



