92 CONTRIBUTIONS TO NORTH AMERICAN ICHTHYOLOGY III. 



Hydrophlox ruhricrocens, Photogenis galactitriis, and Catostomns nigricans, 

 both in the Tennessee and Savannah, will illustrate this. Neither of the 

 two first-named species are as yet known from any other river-basins. 



VI. When the watershed between two streams is a swampy upland 

 instead of a mountain-range, the same si)ecies may be found in the head- 

 waters of both, although the species inhabiting the lower courses may be 

 different. In case the one stream flows northward and the other south- 

 ward, the common fauna will be nearest like that of the northern stream. 



In Northern Indiana, the same species are found in the waters of Saint 

 Joseph's, Maumee, Wabash, and Illinois Rivers, although these streams 

 discharge their waters in widely different directions. The swampy water- 

 shed between them is often overflowed in the spring, afi'ording to the 

 smaller fishes an easy means of migration. 



VII. In any river-basin, many of the species inhabiting small streams 

 are different from those occurring in the river-channels. Among the 

 brook si)ecies may be meutio.ued Eucalia inconsians, Pceciliclithys specta- 

 hilis, Xenotis lythrochloris, Xenisma steWferum, Salvelinus foniinalis, 

 Ericymha huccata, Semotiliis corporalis, Clirosomus erythrogaster, the 

 species of Rhinichthys, etc. Of channel specAes, Hojjloidonotys, Hyodon, 

 Dorosoina, Pomolohns, Eoccus chrysops, all the' " Buffalo-fishes", and the 

 larger Siluridce, Ichthcelurus pvnctatus, Pelodichthys oUvaris, Amiurus 

 nigricans, and the like, will serve as examples. 



Vlli. Many species inhabiting the upper course of a stream are differ- 

 ent from those of the lower. This subject has been ably discussed by 

 Professer Cope, but further investigations, especially of the rivers of the 

 Southern States, are very desirable. 



IX. This diflerence between the upper fauna and the lower is due to 

 differences in the character of the river itself, such as climate, condition 

 of water, character of river-bed, supply of food, etc. 



X. Hence, if in the same river basin there are two streams flowing 

 into a larger stream, the one near its source, the other near its mouth, 

 if the two streams are similar in all known physical respects, their 

 faunae will be similir, and if dissimilar, they will have different faunae. 

 The general identity of the faunae of Elk Eiver and Powell's River 

 may be noticed in this connection. 



XL Some species of fishes are confined strictly to a single river-basin, 

 while other species, with apparently no better means of diffusion or of 

 defense, are widely distributed, inhabiting many rivers. In illustration 

 of this, the limited range of each of the species of Codoma may be com- 



