J. W. DAVIS ON THE GENTJS PLEFEACANTHUS, AGASS. 331 



general similarity, they differ considerably in detail. P. rohustiis is 

 a larger and stronger spine than either of the others ; it has quite 

 double the number of denticles on its posterior surface ; and the 

 space between the two rows of denticles is produced so as to form 

 a very distinct median keel ; it is deeper from back to front ; and the 

 internal orifice is proportionally small. P. cdatus is a broader spine ; 

 its diameter is greatest from side to side : whilst P. alternidentatus is 

 as nearly round as possible — neither of the latter two possessing apos- 

 terior median keel. The posterior teeth differ in each of the species. 

 In P. rohustus the teeth are long, arched, and terminate in a fine 

 point ; they present the appearance of being implanted in alveolar 

 cavities. The denticles of P. cdatus are short, broad at the base, and 

 obtusely pointed ; they seem rather to be produced from the body 

 of the spine than implanted ; they are opposite to each other, and 

 in this respect differ from those of P. alternidentatus^ in which the 

 denticles alternate ; in the latter also the denticles are much wider 

 apart, the spaces between them being quite double the breadth of 

 the base of the tooth. 



An ichthyodorulite from the Linton coal-beds of America is de- 

 scribed by Prof. J. S. Newberry in the Palseontological part of the 

 ' Geological Survey of Ohio,' p. 56, pi. lix. fig. 7. It is somewhat 

 similar, judging from the figure, to P. alatus. It is more slender ; 

 the teeth are more numerous, acute, and recurved ; it is also straight 

 and round. P. alatus is slightly curved, and the teeth are blunt. 

 The American specimen is named by Dr. l^ewberry Orthacanthus 

 gracilis. 



9. PlEFE ACANTHUS CTLINDEICFS. Pig. 8. 



Orthacanthus ci/lindricus, Agass. Poiss. Foss. vol. iii. pi. 45. figs. 

 7, 8, 9. 



Prof. Agassiz gives a figure of this species along with the name, 

 but does not describe it. A reference is made to the genus in the 

 third volume of the descriptive text, p. 330, as a straight spine, of 

 circular form, with two rows of sharp teeth, the specimen being 

 from the Coal-formation near Manchester. 



The length of the spine in a perfect state and full-grown is from 

 16 to 18 inches. The one I have before me is 16 inches in length. 

 Its greatest diameter, 2 inches from the basal extremity, is *7 of an 

 inch. Prom this maximum thickness it tapers gradually and per- 

 sistently towards the apex, which ends in a sharp point. The spine 

 is circular in form throughout its entire length. There is a round 

 cavity extending from the base to about two inches from the point. 

 Where the thickness of the spine is greatest the cavity occupies one 

 third the diameter, and slightly approaches towards the anterior 

 surface ; thence the cavity gradually contracts to a point towards 

 the apex. In the opposite direction the cavity is terminal, the 

 orifice becomes rapidly extended, its walls thin out, and end in a 

 sharp edge at its basal extremity. One inch and a half appears to 

 have been imbedded, the remainder exposed. The external surface 

 is covered with well-marked fine longitudinal striations, which be- 



