VERTEBRATES. 431 



spine, with a prominent sharp crested median keel, moderately chan- 

 neled either side, with obscure secondary longitudinal ridges, and faintly 

 striated, bearing along the inner side of the postero -lateral angles, 

 which latter are sharply defined, a row of irregularly developed, strag- 

 ling, obtuse tubercles, whose apices culminate in a rounded rim which 

 is in some instances obliquely directed downward and inward, perhaps 

 more frequently in the opposite direction, and extending from the apex 

 to a point at least below the middle. Base unknown. Pulp cavity, in 

 the middle of the spine, less than half the greater diameter of the spine, 

 in transverse section subcircular, or lateral diameter exceeding the 

 antero-posterior diameter, posterior walls about half the thickness of 

 that in front. Length of spine at least three inches, lateral diameter 

 between the postero-lateral angles near the middle of the spine .13 inch, 

 or but slightly less than the antero-posterior diameter. 



The present form is described from a unique example of a nearly 

 entire and well-preserved spine, but which is so imbedded in its matrix 

 as to exhibit only the posterior face and the posterior portion of the 

 lateral surfaces, except a section from the middle of the spine which 

 shows the entire lateral surface and anterior margin. Towards the 

 apex, the costse are more or less smoothly worn, and the tubercles in 

 the posterior face are relatively larger than in the middle region below. 

 The great lateral breadth of the spine, as compared with its antero- 

 posterior diameter, the absence of a distinct angulation in the broadly 

 rounded anterior margin, and the irregular development of posterior 

 denticles, constitute the chief distinguishing features by which the pre- 

 sent spine may be recognized from other forms of the genus to which it 

 is here referred. 



Position and locality: Upper beds of the St. Louis limestone; St. 

 Louis, Missouri. 



Ctenacantatjs siiiilis, St. J. and W. 



PI. XV, Tig. 11 a, b, c, d. 



We here refer to a form apparently not uncommon in the Chester 

 formation, but which is represented by imperfect specimens in the mate- 

 rial accessible to us. In general outline it is exceedingly like C. angu- 

 latus, X. and W., a fin-spine described from the same horizon, being 

 rigid or very slightly curved in outline. It is, however, apparently 

 distinguishable from the above, with the original of which we have 

 compared the present examples, by the more uniform and relatively 

 stronger costs;, of which there are also fewer, and the peculiar decussa- 

 tion of the same, which latter recalls C. gradocostus, or perhaps even 

 more strikingly C. ezcavatus of the Keokuk formation. It is possible 



