126 PALAEONTOLOGY OF ILLINOIS. 



are much higher and more tooth-like, and the flattened and keeled convex bor-' 

 der of L. gcmmatus is here wholly wanting, still the style of ornamentation is so 

 similar that, with the striking character of a strong reversed curve in common, 

 it would be difficult to find good cause for their separation. 



Unfortunately the specimen before us wants the base, without which the 

 question of its curvature can only be conjecturally decided. The facts which 

 have led us to suspect that the curve of the tip was forward are: 1st. On the 

 concave side of the spine the bony wall of the central cavity is much thicker 

 than on the convex side, on which the pulp cavity runs out some distance above 

 the base. 2d. The tuberculation of the lateral surfaces become stronger toward 

 the concave border, fading out toward the convex. The natural inference from 

 these facts is, that the curve of the spine was reversed, as in most others; the 

 posterior wall is thinner than the anterior, and is broken through above the 

 base; the tuberculation of the lateral surfaces is, in most or all allied spines, 

 strongest on or near the anterior border. These indications are not wholly 

 "conclusive, however, and may be fallacious. If so, this would constitute the 

 type of a new genus, as, though allied to Oracanthus, Asteracanthus and Myria- . 

 canthus, perhaps in an equal degree, it could be united with neither of them. 

 From the first it is distinguished by its more slender form and the denticles of 

 the concave border. According pretty well with Asteracanthus in general form 

 (if not reversely curved) and in its stellate tubercles, it is separated from that 

 genus by the diversity in size, form and arrangement of its tubercles, and its 

 geological position. With Myriacanthus it agrees in the variety of its orna- 

 mentation, but not in its form nor geological place. Other specimens will be 

 required, before its relations can be fully determined. 



Formation and locality : Keokuk limestone, Warsaw, Illinois. 



Ctenocanthus gracillihus, N. and W. 



PI. XIII, Fig. 3. 



Spine very long and slender, (9-12 inches long, 4-5 lines 

 wide) gently arched, much compressed, scarcely more than a 

 line in thickness with a lenticular section; anterior margin 

 acute, posterior edge truncated below, above sulcated longitu- 

 dinally, and set with two rows of acute, compressed, strongly 

 depressed hooks, of which the bases are contiguous; each side 

 is ornamented with, about 10 below, 15 above, nearly equal 

 continuous, parallel, somewhat flattened, longitudinal carinae; 



