J86 RIYERS OF SOUTH AMERICA. 



tion non interrompue de 1400 lieus sans sortir du 

 domain des etablismens monastiques," writes Hum- 

 boldt; and what a field is there not here, without 

 crossing the boundary from comparative civilization, 

 for every species of investigation. Again, the same 

 author states, " Le Cassiquare, par example, forme 

 une ligne navigable entre deux bassins de rivieres * 

 (I'Amazone et 1'Oronoque) dans 1'area est a 190,000 

 lieues-carrees. La temperature, la profondure, et la 

 vitesse aes eaux, leur limpidite, leur proprietes chi- 

 miauea le lit des fleuves tantot bossux, tantot rem- 

 pli d ecueils influent puissant sur 1'organization 

 animale." If arrangements can be made to exceed 

 these bounds, and to trace these rivers to their 

 scources, there is another as extensive field entered 

 upon, and ground altogether unbroken to be sur- 

 veyed. The dangers and difficulties are undoubt- 

 edly great, but the rewards just as undoubtedly 

 certain. After making these remarks, and before 

 closing the volume, we have thought it right to bring 

 together a few of the species observed by Humboldt 

 and more lately by D'Orbigny, and which do not 

 seem to have occured to Mr. Schomburgk ; so that 

 as little of our present knowledge as possible may 

 be withheld from future investigators ; and we shall 

 take them in the order which has been already 

 followed. 



Among the Loricaringe and Siluridse, with which 

 we commenced our first volume, we introduced a 

 short notice of most of the allied fishes which should 



