142 Mr. A. S. Woodward on the 



XVI. — On the Devonian Ichthyodorulite^ Byssacanthus. 

 By A. Smith Woodward, F.L.S. 



Among the remarkable Paleozoic fish-spines still awaiting 

 determination, the Devonian fossil Byssacanthus is one of the 

 most interesting-. It was originally described by Agassiz * 

 as related to the contemporaneous Elasmobranch tin-spine 

 Onchvs, differing from the latter in the great expansion of its 

 hollow base. Tlie typical specimens, ascribed to two species 

 both by Agassiz t and E, von Eichwald |, were obtained from 

 the Devonian of the Government of St. Peteisburg, Russia. 

 Another form was described by Barrois in 1875 § from the 

 Upper Devonian of Couvin, in the Ardennes ; while some 

 difficultly determinable spines (perhaps Cephalaspidian cornua) 

 from the Old Bed Sandstone of Bromyard, Herefordshire, 

 •were also placed under the generic name Byssacanthus in 

 Agassiz's work already cited 1|. 



The specimens from. Herefordshire may be neglected as 

 worthless ; and the present writer has not yet had the privi- 

 lege of examining the fossil described by Dr. Barrois. A 

 study of all the examples of the typical Byssacanth^is, how- 

 ever, now to be seen in the Museums of Jurjeff (Dorpat) and 

 St. Petersburg has convinced the writer that a slight advance 

 can already be made in determining its affinities. The 

 following notes are thus published in the hope of directing 

 attention to the subject. 



The most satisfactory specimens, all nevertheless frag- 

 mentary, are preserved in the Geological Museum of the 

 University of Jurjeff (Dorpat); and, by the kindness of 

 Professor Loewinson-Lessing, one duplicate example has now 

 been exchanged with the British Museum. All these were 

 obtained from the Devonian of Livonia (in association with 

 Psamrnosteus, IJeterosieus, Bojnosteus, and Asterolejns) ^ and 

 they appear to belong to the type-species of the genus, jS^ssa- 

 canthus crenulatus. 



The spine exhibits a small central cavity even in its more 

 constricted part, and its expanded basal portion is correctly 

 described as hollow ; but, as shown by the specimen in the 



* L. Agassiz, ' Poiss. Foss. Vieux Gres Rouge ' (184o), pp. Ill, 116. 



t Loc. cit. 



X E. von Eichwald, ' Lethsea Rossica,' vol. i. (1860) p. 1598. 



§ C. Barrois, "Sur le Byssacanthus Gosseleti/' Comptes Rendus Assoc. 

 Fran?. 1874 (1875), pp. 3«1, 382. 



II Op. cit. p. Ill ( = Onchus arcuatus, L. Agassiz, ' Poiss. Foss.' vol. iii. 

 p. 7, pi. i. figs. .3-5). 



