CANADIAN FOSSILS. 47 



their normal position by the introduction of the area of small plates. 

 The other three plates of the second series form a regular row across 

 the body above the pelvic plates. The central one is hexagonal, 

 it has one-half of the posterior pectinated rhombs on the right lower 

 side, and is flanked on each side by a large heptagonal plate. 



The third series consists of four large plates, elongated vertically ; 

 one of these, situated on the right hand of the centre of the back, is 

 pentagonal, and next to it, on the left, is another of nearly the same 

 form, but made hexagonal by the truncation of one of the upper 

 angles. These two plates are narrowed above to correspond with 

 the decreasing dimensions of the body, which here begins to contract. 

 The other two plates of this series are either heptagonal or slightly 

 octagonal, and at their upper extremities they fold around the body 

 and unite on the ventral side by narrow projections, which arch over 

 the area of small plates. Above the third series are ten smaller 

 plates, which close the summit and form a solid support for the arms. 



The ventral side, as before mentioned, is mostly occupied by an 

 area of small plates (see plate 1. fig. Ic, and plate ii. fig. 16) which 

 are altogether of a different character from those of the dorsal side. 

 They vary greatly in size in the different species ; for instance, in 

 P. s(piamosus there are several hundreds of them, and in P.Jilitextus 

 only about forty or fifty. The margins of the dorsal plates being- 

 folded over, form a solid smooth border around the whole space, and 

 to this border the small plates do not seem to be connected. 

 According to my view, the whole of the oval space on the ventral 

 side surrounded by the border, was covered by a flexible integument, 

 strengthened by the small plates, and these cannot therefore be 

 regarded as normal plates in the sense in which the term is used in 

 describing genera of Crinoids or Cystidea, but rather as the remains 

 of a partially calcified dermal covering for a part not protected by 

 the true skeleton. It follows that in this genus the dorsal side is 

 composed of a definite number of plates, arranged according to a 

 permanent plan, and in this respect is related to that group to which 

 Echino-encrinites belongs ; while the ventral side, by the indefiniteness 

 of the number and arrangement of the plates, is more like the body 

 of a Sphasronite. 



The mouth is situated on the left hand side of the base, and opens 

 out through a notch excavated in the border. In most specimens 

 the sutures between the plates cannot be detected in the border, 

 owing to the close manner in which they have been anchylosed. It 

 can be seen however, that if the two anterior basal plates were 



