NO. 1704. A COLLECTION OF SHELLH FROM PERU—DALL. 163 



nodules on the inside of the ventral points of the valves. They 

 spring from a thick reinforcement of the hinge line, simulating a 

 hinge plate, and they have nothing to do with the muscular system, 

 but, as in the Pholads, are buried within the mass of the body and 

 are probably of use in supporting the internal organs against the 

 violent shocks resulting from its boring operations. From the pos- 

 terior end of the ''hinge plate" to a point on the margin of the valves 

 corresponding to the angle between the postmedian and posterior 

 areas, extends in the adult a broad septum in each valve, continuous 

 on its inner edge with the margin of the valves and on the opposite 

 edge free, with a recess behind it equal in depth to about half the whole 

 width of the septum. The surfaces of these two plates form the 

 third myophoric area above referred to and carry a relatively im- 

 mense mass of muscular fibers uniting and holding closed the two 

 valves and counteracting the action of the muscles massed on the 

 exterior myophores. In other words, these muscles correspond to 

 adductors of ordinary bivalves as regards their function, while the 

 external muscles operate like a ligament. The nodules on the inside 

 I of the distal or ventral ends of the valves are of a rather unusual 

 shape, subcylindrical and blunt at the opposed ends, rapidly cuneate 

 at the proximal ends. The type, an adult shell, (Cat. No. 207695 

 U.S.N.M.), measures dorsoventrally 20, in length 19, and transverse 

 diameter 19 mm. The soft parts, in alcohol, of this specimen were 

 about 8 inches (20 cm.) in length. The pallets, set rather far back 

 from the siphonal ends, measured about 45 mm. in length, of which 

 25 mm. is smooth cylindrical stalk, the remainder being vane, of 

 which the mass is set on the stalk inequilaterally, the segments being 

 closely crowded with a serrate profile, and pretty well covered by a 

 thin brownish periostracum which passes over the segments on the 

 back without interruption for the interspaces. The width of the 

 vane is about 5 mm. near the base, gradually narrowing to a point at 

 the tip. 



It is somewhat odd that, in comparing the shell of this species 



with that of other Teredinidse, the most similar shell found was not 



that of any Xylotrya, but a shell belonging to another genus, the 



Teredo norvegica of vSpengler. From this the valves of X. dryas 



I differ in having the posterior area axially longer, the postmedian 



! shorter, the premedian wider, and the anterior about the same pro- 



! portion. The styloid processes are longer; between the root of the 



i process and the anterior end of the thickening I have for convenience 



I called a ''hinge plate," there is a small but prominent denticle which 



I have not found in any other species. 



It is not improbable that this species may be confined to the man- 

 groves and not attack dry wood; if so this would account for the 

 form not being reported before. 



