M. Agassiz on Fossil Fishes. 333 



but they differ in this, that the one, the Ctenoids, have the 

 posterior edge of the scales denticulated, while in the Cycloids 

 the same edge is entire. The author endeavours to prove 

 that this distinction, apparently insignificant, is, notwithstand- 

 ing, founded in nature ; that it is the expression of a funda- 

 mental feature, which, in like manner, appears in the other 

 parts of the body. It is thus that the fishes possessing den- 

 ticulated or pectinated scales, are in general stuck over with 

 spines on the head, the operculum, and various parts of the 

 body, while the others, the Cycloids, are smooth fishes, without 

 any armature. M. Agassiz considers the family of perches 

 and its allies as the type of his order of Ctenoids ; and the 

 family of carps, salmon, and pikes, as the type of the order 

 Cycloid. This division, therefore, corresponds to a certain 

 point with Cuvier's division, into Acanthopterygians and Mala- 

 copterygians ; and this coincidence will be sufiicient of itself 

 to prove the necessity that exists for separating these two 

 types, since we arrive at the same result by ways entirely 

 opposite. Is M. Agassiz's method more successful in this than 

 that of Cuvier ? We scarcely believe that it is ; and already 

 the researches of other naturalists have pointed out fishes be- 

 longing to the same family, some of which have the scales of 

 the cycloids, and others the scales of the ctenoids. Yet we 

 willingly admit, that, for the study of fossil fishes, the distinc- 

 tion derived from the scales is of greater practical value than 

 that which is founded on the structure of the dorsal fins. 



M. Agassiz's second order. Ganoid, appears to us much better 

 founded. Two fishes exist in the Nile and in the rivers of 

 South America, which have at all times embarrassed ichthy- 

 ologists. One of them, that of the Nile, is known under the 

 name of Bichir {Folypterus Bichir) ; the other, that of Ame- 

 rica, under the name of the Osseous Pike {Lepidosteus), be- 

 cause in its exterior it resembles our pike. Both these fishes 

 are covered with scales, of a form and structure quite peculiar. 

 Instead of being placed one over another like tiles on a roof, 

 as among ordinary fishes, they are simply placed beside each 

 other, and their surface is covered with a coat of enamel, 

 which forms a very solid cuirass. M. Agassiz has examined 

 these fishes in an anatomical point of view, and diff^ercnces 



