1914] Northwestern Denes and Northeastern Asiatics 147 



This is from the writings of an erudite author who was merely 

 echoing one of Humboldt's declarations.^ 



North of Mexico alone, we have no less than fifty-eight families, whose 

 languages have not anything more in common than vague and exceed- 

 ingly broad characteristics, which cannot be adduced as valid criteria 

 of ethnic assimilation. ^ Not one word of a particular stock will ever be 

 found repeated in another, unless it be a loan word due to commercial 

 intercourse; the grammar and morphology of each are irreconcilably 

 different, with scarcely any trait of resemblance. 



Does not that wonderful diversity point to extraneous origins, to 

 some accidental importation from unrelated quarters? 



The answer is ready at hand, and we implicitly find it in a third 

 ethnological fact which is no less established: "A marked feature of 

 the distribution of Indian linguistic families north of Mexico is the 

 presence, or former existence, in what are now the States of California 

 and Oregon, of more than one-third of the total number, while some 

 other stocks . . . have a very wide distribution. The Pacific coast 

 contrasts with the Atlantic by reason of the multiplicity of its linguistic 

 families as compared with the few on the eastern littoral". 



These remarks are not mine. I take them from the sketch of the 

 American linguistic families in the "Hand-book of American Indians" 

 by our common lamented friend, the late Dr. Alexander Chamberlain.* 



That anthropologist had chiefly in mind the aboriginal population 

 of what is now the United States. If we turn to Canada, the apposite- 

 ness of his last remark will be greatly enhanced. As everyone knows, 

 fully five-sixths of the territory of that country lies east of the Rocky 

 Mountains. Yet, within that immense stretch of land, we find only 

 three aboriginal stocks, namely the Iroquois, the Algonquin and the 

 Dene. I do not mention the Sioux, who are a recent intrusion, or the 

 Eskimos, who can hardly be considered an American race, since they 

 are found in Asia and the inter-continental islands. Moreover, they 

 belong as much to the west as to the east slope of the Rockies. 



Now there are no less than six unrelated families within the remain- 

 ing sixth part of Canada's territory, that is, on the Pacific coast and the 

 land intervening between it and the Rocky Mountains. These are the 

 Kootenays, the Salish, the Kwakwiutl, the Haidas, the Tsimpsians 



^Cardinal Wiseman, "Twelve Lectures on the Connexion between Science and 

 revealed Religion", pp. 78-79; London, 1842. 



^ I am well aware of the fact that an effort is now being made to find analogies 

 between some of these stocks with a view to reducing their number (See "American 

 Anthropologist", Vol. XV, p. 647 et seq.). But these pretended verbal assimilations 

 will seem far-fetched and little convincing to more than -yne American philologist. 



' Vol. I, p. 767. 



7- 



