are brought together. The borings of the sponge frequently 

 weaken the shells decidedly and it is not uncommon to find bor- 

 ings that come so near the inner surface that light may readily 

 be seen through them, but I have never found actual perforation 

 of the shells. It seems probable that the secretion from the 

 sponge that acts in dissolving the shell soaks through the shell, 

 sufficiently to stimulate the mantle to renewed secretion of shell 

 substance before the perforations are actually complete. This is 

 indicated by the fact that shells that are badly attacked are quite 

 universally thicker in these places than those that are not so badly 

 attacked, and that their inner surfaces are likely to be rough- 

 ened (fig. 7). The meaning of the roughened appearance is 

 especially striking if shells that are bored only in patches are 

 examined. In such cases the end of each of the deep burrows 

 is marked on the inner surface of the shell by a rounded eleva- 

 tion, each of which reminds one at first sight of a minute pearl 

 that has become fastened to the shell, and partially covered up 

 by later secretions of nacre. 



The markings on the inner surface of the shell are much more 

 distinct in some individuals than in others and they differ some- 

 what on the two valves. 



The adductor muscle scar is quite large and fairly distinct, 

 and is placed somewhere near the center of the shell, a little 

 nearer the posterior than the anterior border. The scar is much 

 larger on the left (fig. 5), than in the right valve (fig. 6), 

 . where it is distinctly double, consisting of a large rounded ante- 

 rior portion, (pas.) and a small somewhat crescent shaped 

 posterior portion (pas'.). The scars usually form depressions, 

 but sometimes elevatons, on the inner surface of the shell, and 

 are more or less distinctly marked by lines that indicate the 

 larger bundles into which the muscle is divided. 



The foot retractor muscle is attached to the left valve of the 

 shell along the dorsal border of the adductor muscle, just ante- 

 rior to the upper end of the posterior crescent shaped portion, 

 and the scar is not distinguishable from the scar of the adductor 

 muscle. 



The mantle is very firmly attached to each valve along a rather 

 broad and indistinct pallial line, (pi.) that is very far removed 

 from the margin of the shell. • The muscles attached along the 



