4 " THE CANADIAN NATURALIST. [Vol. X. 



a series of swellings or ridges, and to these are attached the 

 remains of what seems to have been membranous partitions, 

 which must have formed successive interior cones as the siphuncle 

 grew in size, terminating in the final one, which became com- 

 pletely calcified. 



It would thus appear that the shell had a wide conical siphuncle 

 which, as it grew in height, it partitioned ofl", with flattened cones 

 within. The effect would be to give protection to whatever part 

 of the body extended back into the siphuncle, to give great 

 strength to the shell, and to form a double series of air 

 chambers, that within the outer shell and that in the apex of the 

 conical siphuncle, by which great buoyancy would be secured. 



As Salter has already remarked, this shell has affinities with 

 such shells as Cameroceras and Endoceras, though in magnitude 

 of siphuncle it exceeds these types, as well as possibly in the 

 property of possessing a double series of air cells. It is, however, 

 not improbable, from the manner of the filling of the partitioned 

 parts of the siphuncle of Endoceras, that this also was hollow in 

 the living state. 



I would propose for the present species the name Piloceras 

 amjjlum, with reference to the great width of the siphuncle. Its 

 description will be as follows : 



Length of siphuncle, 12 centimetres ; longest diameter at the 

 top, 6 centimetres: shorter diameter, 3.5 centimetres; greatest 

 angle of divergence of siphuncle, about 27°. Siphuncle annula- 

 ted with raised lines, marking the attachment of the septa of the 

 exterior shell. These lines are inclined to the axis of the shell 

 at an angle of about 20°. They are, however, slightly curved 

 and on the dorsal (?) side of the shell bend slightly downward. 

 The internal cone of the siphuncle is 5 centimetres in depth. It 

 is flatter than the siphuncle, ending at the apex in an edge, which 

 is attached to a central shelly plate crossing the lower part of the 

 siphuncle. This plate shows, at intervals, slight projections 

 giving rise to delicate internal cones apparently membranous. 

 The space between the inner cone and the wall of the siphuncle 

 must have been empty and closed, as it has been tilled not with 

 the surrounding coarse dolomite, but with calcite, introduced by 

 infiltration. Whether the siphuncle was central or lateral does 

 not appear. There are, however, distinct marks of the partitions 

 of the chambers all around it. 



