and Systematic Position of Cheirolepis. 247 
palato-quadrate apparatus, and of the hyoid and branchial 
arches. 
Regarding the dentition of Chetrolepis there has also pre- 
vailed some little obscurity. Agassiz describes the teeth as 
being indeed of two sizes, but all arranged in one line, and 
in that res spect differing from the unequal dentition of his 
““ Sauroids”’ and “‘Coelacanths,”’ in which the smaller teeth 
form a continuous external range. Pander and Huxley describe 
the jaws as being set with small conical teeth, but they were 
unable to find any of the larger ones referred to by Agassiz ; 
while Powrie, on the other hand, returns to the statement of 
Agassiz regarding the larger and smaller teeth being in one 
row. According to the specimens which have come under 
my own observe ation, the jaws of Chetrolepis were set along 
the inner aspect of ‘their’ dental margins with one row of 
tolerably equal and rather closely set, sharp, and acutely 
conical teeth, each having a marked inward curve, and, when 
broken, displaying a large simple internal pulp- cavity. "These 
are undoubtedly the teeth referred to and figured by Pander, 
who, however, seemed to expect that, according to Agassiz’s 
description, lar ‘ger ones would be found among “them. Now, 
other teeth of a different size do exist—not larger, however, 
but smaller; and these form a row eaternal to those first 
described. ‘The outer row of smaller teeth, the discovery of 
which at once breaks down Agassiz’s demarcation between the 
dentition of Checrolepis and that of his so-called “ Sauroids”’ 
and “Ccelacanths,’”’ is not often seen, from the fact that the 
edge of the jaw on which they are placed i is almost invariably 
found split off and adherent to the matrix of the ‘“ counter- 
part,” and thus the little teeth in question are hidden. But 
by careful working out with the point of aneedle, I have been 
able to display some of them in two cases where a portion of 
the edge of the jaw remained, as shown in Plate XVII. figs. 4 
and 5. They are indeed very minute, being only about one 
third or one fourth the length of the larger ones, which them- 
selves only measure 1; inch in specimens of the ordinary size. 
The dentition of Cheirolepis i is thus reduced to a type very 
frequent in Ganoid fishes, and which notably occurs in many, 
if not im most, of the genera comprised in the family of 
Paleoniscide. 
The facts adduced in the preceding pages seem most satis- 
factorily to prove not only that Chedrolepis, as Prof. Huxley 
has already indicated, must take its place among those Ganoids 
which he has brought together under his suborder of Lepi- 
dosteidee, but also that among those Lepidosteids it must 
