1897.] NEW YORK ACADEMY OF SCIENCES. 309 



Allowing for certain minor differences this description holds 

 good for what I have heretofore called the working form of 

 Fholadidea and there seems to be no doubt that Parapholas 

 passes through a stage of growth marked by like characters. 

 In Fholadidea, however, the antero-dorsal mantle-lobe is some- 

 what larger than in Zirphsea. 



The shell of the resting forms of both Fholadidea and Fara- 

 pholas correspond in a very high degree and are characterized 

 by the presence of the antero-dorsal and antero-ventral plates, 

 Fholadidea differs only in the greater posterior extent of shell 

 and the leathery flaps. 



The mantle of the working forms of Fholadidea, and Fara- 

 pholas and of Zirphsea is fused throughout its length excepting 

 at the antero-ventral region where there is an opening sufficiently 

 large for the protrusion of a muscular cylindrical foot. The size 

 of the mantle opening at this point is regulated by a sphincter 

 muscle, the fibres of which are arranged tangentially. While 

 this condition obtains in Zir2oh8ea throughout life a marked 

 change comes over the other two. With the cessation of their 

 task of providing a home for themselves in the rocky matrix, 

 the muscular foot degenerates and the mantle opening becomes 

 smaller, leaving but a small slit-shaped pore at the extreme an- 

 terior end. This opening may be called the anterior mantle 

 pore. This pore is provided internally with a two lipped valve 

 by means of which egress of water may be prevented, and serves 

 for the indraught of silt and other debris which may chance to 

 collect at the bottom of the animal's burrow. Subsequently this 

 material is expelled from the mantle cavity through the siphon. 

 The anterior mantle pad is naked during the working state, but 

 is protected by the antero-dorsal plates. In Zirphsea it is naked 

 throughout life. 



The anterior adductor is made up of two masses of tissues, a 

 posterior portion encircling the urabones dorsally, and extend- 

 ing backwards and forwards of the same, and an anterior portion 

 (accessory anterior adductor) which appears to be made up of 

 pallial muscles running transversely from valve to valve. This 

 portion antagonizes the posterior adductor while the posterior 

 portion is the opposite of a ventral adductor which is made up 

 of transverse fibres from the pallial muscles at the end of the 

 lower limb of the pallial sinus. There are thus, it will be seen, 

 four separate adductor muscles instead of three as I stated in a 

 previous paper. The action of these muscles will be discussed 

 a little further on. 



The foot of the working forms of Fholadidea and Paranholas 

 and of Zirphsea is cylindrical and strongly muscled. The points 



