ATOMIC CONSTITUTION, ETC., IN MINERALOGY. 435 



the case that he would be a bold man who should venture in speaking 

 of the southern tribes of Russian America, to say : here the Eskimo 

 area ends, and here a different area begins?'* The difference 

 thus pointed out may be accounted for, to a considerable extent, by 

 the diverse geographical conformation of the continent, on its eastern 

 and western sides, which admit in the latter of such frequent and 

 intimate intercourse as is not unlikely to lead to an intermixture of 

 blood, and a blending of the races, however primarily distinct and 

 diverse. The evidence presented here, however, refers to tribes 

 having no such intercourse with the Esquimaux, and distinguished 

 from them by many important characteristics, in manners, social 

 habits, and external physiognomy. Nevertheless if these conclusions, 

 deduced from an examination of Canadian crania, are borne out by 

 the premises and confirmed by further investigation, this much at 

 least may be affirmed : that a marked difference distinguishes the 

 Northern tribes, now or formerly occupying the Canadian area, in 

 their cranial conformation, from that which pertains to the abori- 

 gines of Central America and the southern valley of the Mississippi ; 

 and that in so far as the Northern differ from the Southern tribes, 

 they approximate more or less, in the points of divergence, to the 

 characteristics of the Esquimaux : — that intermediate ethnic link be- 

 tween the Old and the New "World, acknowledged by nearly all 

 recent ethnologists to be physically a Mongol and Asiatic, if philo- 

 logically an American. 



ON ATOMIC CONSTITUTION AND CRYSTALLINE FORM 

 AS CLASSIFICATION CHARACTERS IN MINERALOGY.f 



BY PROFESSOR CHAPMAN, 



PROFESSOR OF MINERALOGY AND GEOLOGY, UNIVERSITY COLLEGE, TORONTO. 



The difficulties that beset the framing of a satisfactory classification 

 in Mineralogy are altogether unknown in other departments of Natural 

 Science. These difficulties originate essentially in the compound 

 nature of minerals, or, in other words, in the apparent absence of 

 corelation between the chemical and physical characters of these 

 bodies. So long as we are forced to admit the existence of dimorphous 

 substances — so long as chemistry remains unable to distinguish or 



* Varieties of Man, p. 291. 



t Abridged from a paper read before the Montreal Meeting of the American Association 

 for the Advancement of Science. 



