ON THE NOIITII -WESTERN TRIBES OF CANADA. 799 



where two or more young cbildren of diflFerent sexes, left by the death of 

 their parents to giow up secluded from all other society, were thus com- 

 pelled to frame a language of their own, which would become the mother- 

 tongue of a new linguistic stock. This result, it is clear, would only 

 follow in those regions where, from the mildness of the climate and the 

 spontaneous fruitfulness of the soil, young children would be able to find 

 subsistence for themselves through all seasons of the year. 



It is evident that, along with their new language, these children and 

 their descendants would have to frame a new religion, a new social policy, 

 and, in general, new customs and arts, except so far as reminiscences of 

 the parental example and teachings might direct or modify the latter. 

 All these conclusions accord precisely with the results of ethnological 

 investigations in America. 



It sliould, however, be borne in mind that, whether the theory which 

 I thus proposed is accepted or not, the fact will still remain that the 

 existence of a linguistic stock involves the absolute certainty that the 

 tribe speaking such a form of language, differing entirely from all other 

 tongues, must have lived for a very long period wholly isolated from all 

 other communities ; otherwise this idiom would not have had time to be 

 formed and to become the speech of a tribe sufliciently numerous and 

 strong to maintain its independence. In this long isolation (however 

 it might arise) the tribe would necessarily acquire by continual inter- 

 marriage a peculiar mental character, common to the whole tribe, and 

 with it the modes of thought and the social institutions which are the 

 necessary outcome of such a character. Thus the linguistic stock, Avhat- 

 ever its origin, must naturally and necessarily be, as has been said, the 

 proper ethnological unit of classification. 



The experience of the able philologists of the American Bureau of 

 Ethnology entirely confirms these views. Special attention, of course, 

 has been given by them to the investigation of the stocks in North 

 America. Mr. J. C. Pilling, of the Bureau, the author of the valuable 

 series of bibliographies of American lingiaistic stocks now in course of 

 publication, informs me that the number of these stocks in North America 

 (north of Mexico), so far as at present determined, is fifty-eight — a greater 

 number, perhaps, than can be found in the whole eastern hemisphere, 

 apart from Central Africa. Of this number no less than thirty-nine are 

 comprised in the narrow strip of territory west of the Rocky Mountains, 

 which extends from Alaska to Lower California. Why a great number 

 of stocks might naturally be looked for along this coast, with its mild and 

 equable climate, and its shores and valleys abounding in shell-fish, berries, 

 and edible roots, is fully explained in my essays already referred to. 



From what has been said it follows that in our studies of commnnities 

 in the earliest slage, we must look, not for sameness, but for almost end- 

 less diversity, alike in languages and in social organisations. Instead of 

 one ' primitive human horde ' we must think of some two or three hun- 

 dred primitive societies, each beginning in a single household, and ex- 

 panding gradually to a people distinct from every other, alike in speech, 

 in character, in mythology, in form of government, and in social usages. 

 The language may be monosyllabic, like the Khasi and the Othomi ; or 

 agglutinative in various methods, like the Mantshu, the Nahuatl, the 

 Eskimo, and the Iroquoian; or inflected, like the Semitic and the Sahaptin. 

 Its forms may be simple, as in the Maya and the Haida, or complex, as 

 in the Aryan, the Basque, the Algonkin, and the Tinneh. The old tbeo- 



