CLASSIFICATION OF DEVONIAN FISHES 459 
edge of the frontal region, is covered by two broad bony plates, which 
were, perhaps, divided in the middle line. Pander considers the an- 
terior of these to be frontals, the posterior to be parietals. At the 
sides of the parietals lie three narrow bones, which, perhaps, all belong 
to the skull, though the inner and uppermost may appertain to the 
shoulder girdle. The anterior edges of the other two bound the orbit 
posteriorly, and similarly elongated plates lie in front of the eyes, 
beside the frontals. The upper jaw is a large bone, rounded off pos- 
teriorly, and tolerably broad behind, while anteriorly its upper edge 
suddenly becomes excavated to form the lower boundary of the orbit 
and then tapers off; it is beset with small sharp conical teeth. The 
gape extends very far back, and the lower jaw is a long flat bone 
toothed like the upper. 
According to Agassiz, there were larger teeth interspersed among 
the smaller ones (“ Recherches,” p. 130; “ Vieux Grés Rouge,” p. 44), 
but all in a single row. Like Pander (1. ¢., p. 73), I have been unable 
to discover these larger teeth. The opercular apparatus and the 
branchiostegal rays, or their representatives, were not observed by 
Pander, nor have I seen indubitable evidence of their characters ; but 
Agassiz (“ Recherches,” p. 132) has described and figured the branchio- 
stegal rays of Ch. Uragus. “The branchiostegal rays are very well 
preserved on both sides of the head; the anterior are shorter and 
larger; they are well seen on the left side. The posterior ones, 
which are better preserved on the right side, are narrower and more 
elongated. I count at least ten of them.” 
According to Pander a large perforated plate surrounds the eye. 
Miller, Giebel, and Pander have agreed upon the propriety of 
separating Chezrolepis from the other Acanthodide of Agassiz, and 
Pander proposed to form for it a distinct family, that of the CHEIRO- 
LEPINI. Granting, as I think every one must do, the justice of this 
step, the question next arises in what suborder of the Ganoids does 
this family arrange itself. 
It certainly is not one of the Crossopterygida, for it has but a 
single, comparatively short, dorsal fin, neither pectorals nor ventrals 
are lobate, and there are no jugular plates ; still less can Chezrolepis be 
ranked among the Amiade or Chondrosteide. On the other hand, it 
presents certain points of resemblance with Paleontscus, and through 
those forms connects itself with that large body of fossil fishes which 
have more or less direct relations with Lepidosteus. Perhaps then, 
Cheirolepis ought to be regarded as the earliest known form of the 
great suborder of the Lepidosteide. 
In the absence of a full knowledge of the head, of the paired fins 
