13 



This subelass embraces, according to the views of Agassiz and Miiller, 

 the recent sturgeons and the gar-pikes, and the Amias of the fresh water 

 streams and lakes of America. 



The subclass of Ganoids, as here restricted, is one of the most interest- 

 ing divisions of the class of Fishes. Some of its representatives are so 

 nearly similar in external form and appearance to true fishes or Teleostei, 

 that a naturalist, unacquainted with the anatomical characters of the 

 species, might well be excused for considering them as members of the 

 same family. Such are the Amise of North America, and the Erythrinoids 

 of South America. The former have the optic nerves connected by a 

 simple commissure and not decussating, and the bulbus arteriosus fur- 

 nished with many valves ; it is, therefore, a true Ganoid. The Erythri- 

 noids exhibit decussating optic nerves, and a single pair of opposite valves 

 in the bulbus arteriosus ; they ax - e, consequently, true fishes or Teleostei. 

 Yet in external characters, there is a very strong resemblance between 

 them, and they were indeed placed next to each other and in the same 

 family by Cuvier. Amia is provided with a sublingual hone, but this is 

 not a character peculiar to the Ganoids ; for the family Elopoidas, com- 

 posed of the genera Elops of Linnaeus and Megalops of Lacepede, is 

 distinguished by the presence of a similar bone. Professor Agassiz has 

 indeed expressed an opinion that, on account of the structure of the scales, 

 and on other grounds, the genus Megalops may be a member of his order 

 of Ganoids. The reasons for arriving at such a conclusion have not been 

 given by him. One reason might well be the presence of such a sublingual 

 bone, especially if, as appears to be the case, such an appendage is peculiar 

 to the Elopoids among the subclass of Teleostei. And there is, indeed, 

 no very inconsiderable resemblance between the Elopoids and the species 

 of some of the families of Holostean Ganoids which have no living represen- 

 tatives. Such are the Leptolepoids, which are generally regarded as true 

 Ganoids. But unless the Elopoids have the structure of the brain and 

 the simple chiasma of the optic nerves as well as the two or more rows of 

 valves in the bulbus arteriosus, they cannot, without a new conception of 

 the characters of Ganoids, be referred to that subclass. 



In other forms, characters are seen which indicate their affinity with 

 the Amphibians and Reptiles ; traits which were formerly supposed to be 

 peculiar to those classes have now been found in representatives of this 

 subclass of fishes. The most singular and remarkable of those types are 

 the paradoxical Lepidosirenes of South America, and Protopteri of 

 Africa. The former were first described by Fitzinger and Natterer as 

 amphibian reptiles, most nearly allied to the Sirenoids of North America; 

 the latter, first named by Owen Protopteri, and afterwards, on the publi- 

 cation of the memoir of Fitzinger, referred to the genus Lepidosiren, were 

 placed among the fishes. The rank and affinities of those animals have since 

 attracted much attention from naturalists, and besides the eminent ones 

 above-mentioned, the accomplished anatomists and zoologists, Bischoff, 



