6 Br. R. H. Traqiiair — On a Neiv Fossil SharJc — 



unfortunately the extremity of the upper lobe is cut off by the edge 

 of the stone : a thin layer of matrix also adheres very obstinately 

 in this part, and attempts which have been made to clear it off have 

 not been very successfuL Through this thin veil of matrix seven 

 fin cartilages may be traced projecting in the direction of the lower 

 lobe to a distance of 3^ inches. 



Immediately in front of the lower lobe of the caudal, there 

 projects a film or mass of minute shagreen granules, which may 

 either be the remains of an anal fin, or simply a mass of those 

 appendages displaced from the adjacent part of the body. But as 

 no radial cartilages can be detected in connection with it, I am rather 

 disposed to accept the latter alternative. 



Eemarhs. — The general form of the spines, showing a comparatively 

 short implanted portion, obliquely marked off from the sculptured 

 exposed surface ; the nature of that sculpture, consisting of longi- 

 tudinal ridges ornamented with tubercles; the evident presence of 

 a posterior area with recurved denticles along the margins between 

 it and each lateral surface — form an assemblage of characters afford- 

 ing. I think, ample justification for the reference of this remarkable 

 fossil shark to the genus Ctenacanthus of Agassiz. As regards the 

 species, there can be no doubt that it is new, as the system of tuber- 

 culation of the ridges differs sufficiently from that in any hitherto 

 described species. In general aspect the spine perhaps approaches 

 most nearly the SplienacantJms serrulatus of Agassiz than any other ; 

 but in the latter the tubercles of the ridges are much more distant 

 and less prominent, being in fact only observable on the ridges 

 towards the posterior margin and lower part of the side ; whereas, in 

 the Eskdale shark they are most prominent and closely set on the 

 ridges in front. Here also I may state, that taking the genus 

 Ctenacanthus as it at present stands, including as it does such species 

 as Ct. hybodoides and nodosus of Egerton, Ct. laivis of Davis, etc., I 

 cannot see any ground for the retention of Sphenacanthiis as a seimrate 

 genus. In describing it Prof. Agassiz seems to have been especially 

 struck by its resemblance to the spine of Hybodiis, but as a charac- 

 teristic difference he remarks that "au lieu de grosses dents a ses 

 bords posterieurs, on n'y remarque qu'une fine crenelure."' But the 

 type specimen of Sjjh. serrulatus, now in the Edinburgh Museum of 

 Science and Art, to which it was presented by the Eoyal Society of 

 that city, displays, on the only portion of one of the postero-lateral 

 margins which has been preserved, teeth as prominent as in most 

 species of Ctenacanthus, and which must therefore have escaped the 

 attention of the great pala3ichthyologist. For this new species I 

 accordingly propose the name of Ctenacanthus costellatus. 



The next important question is that regarding the dentition of 



Ctenacanthus. Except in a few instances, such as Pleur acanthus, 



Pleurodus, Tristychius,^ the correlation of the teeth and spines of 



Paleozoic Selachii has hitherto been a very hopeless subject ; how- 



^ Poissons Fossiles, vol. iii. p. 23. 



2 In connexion with the dentition of Tristychius, see a paper by Mr. T. Stock in 

 Ann. and Mag. Nat. Hist. Sept. 1883. 



