158 



ORDERS OF FISHES. 



have been referred to another genus under the name of 

 Thclodus (fig. 520, b). In the same bed are found jaw-like 

 bodies, with tooth -like processes of different sizes, which 

 have been named Plcdrodus, and have been supposed to be 

 the jaws of fishes. The true nature of these, however, is 

 doubtful, and they certainly do not belong to Plagiostomous 

 fishes. Possibly they are the prickly margins of the cephahc 

 bucklers of Cephalaspidean fishes. 



It should be mentioned, also, that M. Barrande enumerates 

 Ctenacanthtis amongst the fishes found in the Upper Silurian 

 rocks of Bohemia, this genus being otherwise only known in 

 the Devonian and Carboniferous formations. 



In the Devonian rocks the remains of Ccstrapliori are not 

 uncommon. The more important fossil spines of the deposits 

 of this period have been referred to the genera Onclius, Ctcna- 

 canthus, Homacanthus, and Cosmacanthus. The fossil teeth 

 belong chiefly to the genera Ctenoptycliius, Cladodus, and 

 Psammodus. 



We may also mention here the singular spines upon 

 which the genus Machmr acanthus (fig. 521) has been founded, 



Fig. 521.— Fin-spine of Machceracanthus inajor, reduced in size. Devonian of North 

 America. (After Newberrj'.) 



though it is quite an open question whether these spines are 

 referable to Cestracionts, Selachians, Chimceroids, or even 

 Siluroids. They present, in fact, the singular pecidiarity 

 that they are unsymmetrical, in the sense that they are 

 rights and lefts ; and it seems almost impossible to account 

 for this except upon the supposition that they were im- 

 planted in front of the pectoral and not of the dorsal fins. 

 They are flattened, curved, and hollow, with an enamelled, 

 smooth, or punctate surface ; and they must have belonged 

 to large and powerful fishes, as they are sometimes over 



