144 BUREAU OF AMEETCAN ETHNOLOGY [bull. 00 



gold are brought to light in the Amazon Valley and along the 

 northern border. It is regarded as probable that objects of these 

 classes had their origin in the Andean highland. 



Pottery is found in many sections of the vast area from I\itagonia 

 to the Caribbean coast, and the use of large vessels for burial pur- 

 poses, observed in the east and north, indicates a stage of advance- 

 ment considei'ably above the average of the area. In the middle 

 regions and far to the south the ware is of primitive ty[)e; the 

 forms are of the simple round-bodied pots and bowls, and the deco- 

 rations consist of conventional incised and impressed figures. In 

 the north many of the vessels are modeled to imitate or to suggest 

 life forms and the incised and i)ainted decorations are tasteful and 

 elaborate. 



Xaturally the art of the noilhern margin of the continent has 



somewhat marked analogies with that of the Westln- 

 Picto^rapbs dlcs, but the contrast between the art of these regions 



and that of northern America is very decided. Picto- 

 graplis. engraved and jiainted, are connnon. especially in the north, 

 and all appear to be of the simple ty})e connnon to primitive America 

 generally. jSIauy represent very rudely the figures of men and ani- 

 mals, but the nuijority are formal unintelligible devices descending to 



mere scrawls. Kitchen middens, the refuse of food 

 Kitchen Middens consumption, are found along the coast from the 



mouth of the Amazon to the Rio de la Plata, and also 

 on the Fuegian and Chilean coasts. In cases they are of remarkable 

 dimensions and the reports of explorers indicate that the accumula- 

 tions of these extended over very considerable periods of time. They 

 seem to owe their existence to very primitive peo])les, however, 

 since they contain but limited traces of works of art of any kind. 

 The Tupi tribes, or, according to Church, the Tapuya, who largely 

 occupied eastern Brazil on the arrival of the Spanish, were preceded, 

 it is said, by the Cles stock, a ^ery pi'iuiiti\e people, and the middens 

 may be attributed, at least in part, to them. 



In recent years uuich attention has l)een given by anthropologists 



to supposed traces of early man in Brazil and Argen- 

 Anti(3uity tiua, but it is doubtful whether there exists suflicient 



reason for assumiug the geological anti(iuity of either 

 the osseous remains of man or the works of his hands, and there is 

 certainly no substantial ground for assigning any of the artifacts 

 thus far discovered to a stage of culture or to a period corresponding 

 to the European ))ale()lithic. No human renuiins have been brought 

 forth that afford suflicient ground for assuming an early people of 

 distinct type from the Indian, and no form of artifact has been 



