ORDER GANOIDEI. 959 



tebral column of certain members of this order which call for 

 passing notice. Thus in Eurycownus the ossifications in the dorsal 

 region consist of an upper and a lower hollow, wedge-shaped, semi- 

 cylinder, with their pointed extremities interlocking, and the former 

 bearing the neural arches and the latter the ribs ; while in the tail 

 we have two centra to each neural arch. From this it would appear 

 that the pieces bearing the ribs in the dorsal region, and the caudal 

 centra which have no arches, correspond to the intercentra of the 

 Amphibia, which are noticed below. The teeth of Ganoids vary 

 greatly in structure, and may be either conical and borne on the 

 margins of the jaws, when they are continuously replaced, or flat- 

 tened disks attached to the vomer, which have no successors. The 

 borders of the fins are frequently furnished with the modified scales 

 known as fulcra. 



Considerable diversity of view has prevailed as to the classification 

 of Ganoids ; but the system of Dr Traquair, who has paid especial 

 attention to the structure of the order, is adopted in this work. 



As regards their distribution in time, Ganoids first appear in the 

 Silurian nearly at the same time as the Elasmobranchs ; and from 

 the Devonian to the close of the Mesozoic they form a very large 

 proportion of the Fish-life of those periods. Their wane, however, 

 appears to have set in during the Upper Cretaceous, when the 

 Teleostei began to be numerically strong ; and from that date there 

 has been a rapid decrease to the present day, when we find only 

 one of the suborders (Amioidea) represented by several genera ; 

 while of the other three surviving suborders one is represented by 

 two genera, and each of the other two by a single genus — these four 

 genera each having a solitary species. 



A. Proganoid Series. — The first three suborders may be con- 

 veniently grouped in a single series, and are mainly characterised by 

 their exceedingly low development, their affinities being still doubt- 

 ful. Professor Cope is indeed disposed to regard one of these 

 suborders (Placodermata) as more nearly allied to the Tunicata 

 than to other Vertebrata ; but this view is scarcely likely to find 

 acceptation with the majority of palaeontologists. 



Suborder i. Cephalaspidea. — The members of this extraor- 

 dinary group have the head and the anterior part of the body 

 covered with a continuous shield, while the rest of the body is 

 coated with small angular plates or scales. No traces of an inner 

 skeleton, lower jaw, or teeth, have yet been discovered ; but at least 

 in the second family there was a strong pectoral fin at the hinder 

 extremity of the dorsal shield. The latter has been recently shown 

 to be furnished with a system of sensory canals. This group is 

 confined to the Silurian and Devonian ; and it is suggested by Pro- 

 fessor Huxley that it may be allied to the Acipenseroidea. 



