122 
Micmac of Nova Scotia, and perhaps the Yellowknives in the vicinity 
of Great Bear lake. But, in general, whether we examine the 
Algonkian-speaking peoples, the Athapaskan, or the Eskimo, the 
only clearly defined political unit was the band. It is true that 
neighbouring bands differed little except in the possession of different 
hunting-grounds, and that they frequently joined together for mutual 
support, but the compositions of these united groups varied con- 
tinually as a band coalesced with its kindred, now in one direction, 
43343 
Summer migration of an Eskimo community, Coronation gulf. 
(Photo hy K. G. Chipman.) 
now in another. Customs, too, like the colours in the spectrum, 
changed almost imperceptibly from one district to another as the 
bands encountered different influences; and the farther one receded 
from any given point, the greater, as a rule, became the variations 
in dialect. Hence the usual divisions into tribes, Naskapi, Montag- 
nais, etc., for the eastern woodlands, Chipewyan, Yellowknife, and 
the rest for the Mackenzie basin, are to a considerable extent 
arbitrary; less so now than formerly, perliaps, because the bands are 
less numerous and more restricted in their wanderings, and the 
establishment of Indian reserves and trading-posts subjects them to 
local influences and conditions that were formerly lacking. 
There is, of course, a justifiable basis for the tribal divisions; they 
provide a convenient method of classification that depends for its 
validity, like the differentiation of species in the biological world, on 
the relative amount of agreement or disagreement in certain more 
