472 The Andes anb the Amazons. 



have the same grammatical structure, but different words. 

 This segregation of dialects, as Bates observes, is no doubt 

 due to the isolated life of each group. Along the banks 

 of the main Amazons, for a distance of 2500 miles, Tupi 

 is the common idiom for intercommunication. West of 

 Iquitos, we can almost say on the entire Maranou, this 

 llngoa geral is not heard. On the Ucayali, the Pano (the 

 language of an extinct tribe, and resembling the Quichua, 

 but the most difficult of all the Indian dialects) is the basis 

 for general intercourse on the Lower Ucayali ; an adulter- 

 ated Qnichua, however, is now supplanting it. According 

 to Hyde Clarke, the Guarani, Tupi, and Omagua tongues 

 are similar in roots and grammar to the Agaw of the Nile 

 region. All the Indians speak with very little modulation 

 of voice. 



To illustrate the utter distinctness of the Amazonian 

 dialects, I give the specimens shown on the following 

 page for comparison. It will be seen that tribes contigu- 

 ous are as incapable of social intercourse as those a thou- 

 sand miles apart. It will also be observed that some of 

 the words do not agree with other published vocabularies. 

 But no one will wonder at the discrepancy who has at- 

 tempted to take dowji a list of words from the lips of a 

 live savage : the rapid, indistinct utterance makes it ex- 

 tremely difficult to express tlie sound by English letters. 



It is a curious coincidence that while the two languages 

 of the Campas and Pirros have scarcely one word in com- 

 mon, they have this similar peculiarity, in that the names 

 of all parts of the body begin with the same letter. In 

 Campa,']N" is used ; in Pirro, W (Hu). In the list of words 

 on page 343, obtained by the author from a Campa boy, 

 there are many variations from the vocabularies of other 

 travelers, and the singular one that the parts of the body 

 besjin with A instead of N. 



