THOMAS.] ORIGIN OF THE AMERICAN RACE. 723 



districts arc too apparent to ([ucstioii their existence. The chief draw- 

 back in attempting to use these as evidences of etlinic distinctions 

 arises from three causes: First, the lacii of sufficient data by wliich to 

 outline the diifereiit sections and districts; second, the overlapping and 

 intermingling of types in consequence of the shifting of positions by 

 tribes; and third, the fact that types of art and other archcologic char- 

 acteristics are not governed strictly by ethnic lines, but are often the 

 result of environment, materials, and contact with other tribes. Never- 

 theless race characteristics and tribal customs impress themselves to a 

 certain extent under all variations in locations and conditions so long- 

 as the identity of the race or tribe is maintained. There is no dithculty 

 in distinguishing the Mexican and Central American antiquities as a 

 whole from those of the mound area of the eastern half of the United 

 States, yet the geographical boundaries of these sections can, with our 

 present knowledge, only be determined approximately. If, however, 

 we move from the Mexican district southward or nmthward along the 

 western side of the continent we shall find the distinguishing features 

 less marked than when compared with the types of the mound area. 

 There is no difficulty in distinguishing the types of the Hnron-Iroquois 

 district from the works of the Dakotan .area (Wisconsin, Minnesota, 

 and the Dakotas), and we can point out some specimens of the former 

 types within the latter area, yet, where not fixed by natural conditions, 

 it is imxiossible with present data to draw a definite boundary to either 

 district. 



Although we meet with this difficulty in defining geographically the 

 boundaries of the districts and more comprehensive sections it does not 

 prevent us from drawing correct (conclusions from their general posi- 

 tions and peculiar types. That all the distinguishing types of a disti'ict 

 or section can not be attributed to the peculiar physical features of such 

 districts or sections must be adnn'tted. Will any one claim that the 

 vast difference ijetween the archcologic types of Mexico and Wisconsin 

 have resulted wholly from the physical differences of the two areas f 

 If not, it follows that so much as has not resulted from physical pecu- 

 liarities must be attributed to racial or tribal customs. 



It is necessary at this point, in order to present the thought in view,, 

 to repeat a few sentences given in the previous part of the volume 

 relating to "Archeological Areas and Distribution of Types." • 



A careful examination of wliat has been ascertained in regard to North 

 American archeology; with special reference to the question of arch- 

 cologic sections, leads in the first place to the conclusion that the 

 ancient remains belong in a broad and comprehensive sense to two 

 general classes. One of these classes is limited geographically to the 

 Atlantic sh)pe, the other chiefiy to the Pacific slope, the eastern or 

 liocky mountain range of the great continental mountain belt to the 

 Rio Grande, forming approximately the dividing line between the two 

 areas. According to this division the Atlantic section includes that 



