MR. T. W. BRIDGE ON THE OSTEOLOGY OF POLYODON FOLIUM. 
721 
Cephalaspidse, the remaining recent and fossil Ganoids being distributed among seven 
other suborders. 
Owen* included Cephalaspis, Pterichthys, Pteraspis, Coccosteus, and the Stu- 
rionidse (gen. Acipenser, Chondrosteus, Polyodon ) in the suborder Placoganoidei ; 
while the remaining Ganoids, both recent and extinct, are included in his second 
suborder of Lepidoganoidei. 
Dr. Gunther, t in his catalogue of living Fishes, retains H. Muller’s subdivisions 
of Holostei and Chondrostei, and uses them in the sense originally suggested by their 
author, the latter group including the families of the Acipenseridse ( Acipenser and 
Scaphirhynchus), and Polyodon. Dr. Gunther further removes Polyodon from the 
Acipenseridse and elevates it to the rank of a distinct family, under the name of 
Polyodon tidse. 
It will be seen from the foregoing brief account of the various schemes for the 
classification of the extinct and living Ganoids which have been proposed, that in 
the majority of cases the systematic importance of the Chondrosteous Ganoids is 
regarded as being equivalent only to that of each of the other families of Ganoids. 
But I submit that the facts elucidated in the earlier part of this paper, and briefly 
summarised above, afford abundant evidence in favour of adhering to Id. Muller’s 
two primary divisions of Ganoidei (Chondrostei and Holostei), or to Lutken’s two 
groups of ‘ Ganoidei proper ’ and c Sturiones.’ 
Though there can be but little doubt as to the primary group of Fishes to which 
the Chondrostei should be relegated, yet the remarkable combination of characters 
generally regarded as more or less distinctive of the Teleostei, the Elasmobrancbs, and 
the Amphibia possessed by Polyodon and, though to a less extent, by Acipenser also, 
even if they do not justify the elevation of these genera to a primary subdivision of 
the class, at all events render it necessary that these facts, which so clearly demon- 
strate their distinctness from all other Ganoids, should receive adequate expression in 
any natural system of classification. 
Omitting for a moment any reference to the Amphibian tendencies of Polyodon, 
it seems probable that the true position of Polyodon is that of an annectent group, 
between the Elasmobranchs on the one hand, and the divergent stems of the Teleostei 
and Ganoidei on the other. The existence of such transitional families as the Cbon- 
drosteidse and the Palseoniscidte, by which the Acipenseroid and Polyodontoicl types 
are related to the more typical Ganoids, renders doubtful the necessity of establishing 
a primary group for the reception of those types. While, therefore, the Chondrostei 
may be retained with the remaining Ganoids in the same order, yet it seems necessary 
that the distinction between the two groups should be definitely expressed. We may, 
I think, retain the two subdivisions first instituted by IT. Muller, though it seems 
advisable to remodel the definitions on which they were based. As before stated, the 
* Palaeontology. 1861. 
f ‘ British Museum Catalogue of Fishes,’ vol. viii. 
4 Z 2 
