12 FRESH-WATER FISHES OF SOUTH AMERICA EIGENMANN. 



fouud in streams aud lakes at altitudes of 13,000* to 15,000 feet, while 

 some Alpine forms descend to the sea.t 



Some marine genera have, contrary to a priori conceptions, species 

 which are confined to some one river,| while species which are strictly 

 fresh-water have unexpectedly wide rauges.§ 



Many genera of wide distribution are conhned to the eastern slopes 

 while genera of narrower distribution || occur on both sides. Some 

 genera have few species which inhabit neighboring rivers,^ while the 

 species of some other small genera inhabit widely separated regions.** 



The distribution has been discussed by Agassiz, Wallace, Cope, and 

 Giiuther. Agassiztt speaks of the distribution of the lisbes found on 

 his journey from Para to Tabatinga. His discussions are, however, 

 more valuable as field notes and suggestions than as a contribution to 

 the subject, since he did not consult the works of previous writers. 

 He was especially impressed by the localization of species, which was in 

 great pit^rt due to mistaking the variations of a species as distinct 

 species, and to the fact noted above that many of the species supposed 

 by him to be restricted to a peculiar spot had beeu collected in other 

 localities by other explorers. On page 244 Agassiz says: "To this 

 day I have not yet recovered from my surprise at finding that shores 

 which, from a geographic point of view, must be considered simply as 

 opposite banks of the same stream, were, nevertheless, the abode of an 

 essentially diiierent ichthyological population." This is nothing more 

 than what is to be observed at a given locality of many rivers or 

 along most coasts. At Wood's Holl, Massachusetts, or at San Diego, 

 California, for instance, different species inhabit restricted areas within 

 a few square miles, one set of species rarely entering the locality of the 

 other. For this reason some species are always associated with cer- 

 tain other species. The same holds good of rivers and creeks. In a 

 small stream in Indiana the numerous species of darters are found at 

 one point ; half a mile further on are species of Noturus, beyond which 

 are species of Aniiurus, etc. To Professor Agassiz, however, belongs 

 the credit of first calling attentioii to this fact. 



Wallacefl devotes but little attention to fresh-water fishes, summar- 

 izing the accounts in Dr. Gunther's Catalogue of Fishes. 



* Orestias {Cijprinodoniido:), Gastropterus, ProHstius (Mugilidw). 



\ Fyr/idium pardum (Pygidida-) in Callao Bay. 



i TachiHurus ffravdoculis in tbe Rio Doce. 



§ Callichthys caUickthys; Eoplosiernum littorale, etc., La Plata to Trinidad ; Pbnelodus 

 cJarias, etc.. La Plata to Rio Magdalena. 



II Cetopsis. 



^Steindachneria with three species: (1) amhlyura in the Jequitinhonha; (2) doeeana 

 in the Rio Doce ; (3) parahyhw in the Rio Parahybiie. 



**Stegophiliis with six species : (1) maculaius in the La Plata ; (2) punctatus at Can- 

 elos, Ecuador ; (3) intermedins at Goyaz ; (4) macrops at Manacapuru ; (5) iyisidiosus in 

 the Rio das Volhas; (6) reinhai'dti in the Solimoens and its tributaries. 



tt A journey in Brazil. Boston : Tickuor & Fields, 1868. 



UThe Geographical Distribution of Animals. Harper »& Bros,: New York, 1876. 



