AND ON THE CLASSIFICATION OF FISH. 501 
copper slate, in the structure of the tail-fin, and in the insertion 
of all the rays into its inferior surface; that among such as are 
now in existence, Lepisosteus alone, sia in a less degree the 
Sturgeon, possess this character ; that he should rather arrange 
the fossils with the Lepisosteus, and that they belong with the 
latter to the abdominal Malacopterygians*. 
Valenciennes and Pentland likewise state that Dipterus and 
Osteolepis should form new genera in the order of the Abdominal 
Malacopterygians. 
Agassiz deserves the greatest praise for having discovered the 
identity in the structure of the scales of Lepisosteus and Poly- 
pterus with those of all osseous fishes of the old formation down to 
the Chalk, for having separated the Ganoidei as a distinct order, 
and for having discovered and distinctly separated their numerous 
genera, and ascertained their species. He said truly}, “ L’éta- 
blissement de lordre des Ganoides est 4 mes yeux le progrés 
le plus important que j’ai fait faire a Vichthyologie.” Equally 
important is the deduction from these investigations, that the 
types, according to which the vast majority of fish at present in 
existence are formed, first commence with the Chalk. 
The Ganoid scales, as remarked by Agassiz, are in general 
imbedded in depressions in the skin exactly like the ordinary 
scales. The outer portion of the membrane of the depression is 
extremely delicate and adherent, and may even appear to be 
wanting, asin Polypterus; but in Lepisosteus the enamel of the 
scale is seen to be covered by an extremely fine membrane, in 
which a slight silvery lustre and even pigment is discoverable, 
and which is readily removable by friction. 
In the structure of the skeleton, the Ganoidei again differ 
very materially from one another, for many have a skeleton which 
is completely bony, such as we find in the living Lepisosteus 
and Polypterus ; while in other fossils the spinal column has re- 
mained partly in the foetal condition, of a soft chorda dorsalis, 
with bony apophyses in rows, just as among the Sturgeons in 
living fish. They also exhibit the greatest variations in the form 
of the body, just as the two living genera are completely different 
from one another. 
In many genera the vertebral column extends to the extremity 
of the upper tail-piece, as we find among living fish, in the 
* Geol. Trans. 2nd Series, vol. iii. p. 125. 
+ Poiss. Foss. vol. ii. 
: 202 
