72 H. P. STEENSBY. 
12,500 belonging to Danish Greenland and Labrador and 14,500 to 
South of Bering Strait. In other words the central Arctic regions 
which are several times greater in extent and coast lengths than the 
Subaretie regions have only !/, of the total number of Eskimo, here 
grouped in a few districts, while on the Subartic coasts the form of 
settlement is more regular. 
For the rest it must be remembered that the Eskimo are not 
essentially resident, but that they change their residence in accordance 
with the change of the seasons and the altered conditions of hunting. 
It would therefore be more correct to talk about constantly and 
regularly visited regions than about fixed inhabited ones. The deserted 
settlements which are found on numerous coasts, and may sometimes 
be judged to be of considerable age, are evidently also connected with 
this roaming. Boas, who has subjected this question to investigation, 
rejects the conclusion which has at times been drawn that the Eskimo 
in past days inhabited several Arctic coasts which they left later in 
consequence of deterioration in the climate and the freezing of the 
sea. He finds sufficient explanation in the periodical shiftings which 
take place even today on the boundaries of the areas which are in- 
habited permanently. He even draws a parallel between the boundary 
line for the distribution of the Eskimo and the northern limit of woodland, 
which latter is not, however, a definite line, either, but a belt within 
which the woodland region advances towards the north in favourable 
periods only to recede during subsequent and unfavourable ones, 
leaving dead and dried up stumps. As reasons why the Eskimo feel 
inclined to look for new hunting grounds Boas mentions the periodical 
changes in the conditions of the ice, the evacuation of old hunting 
grounds by animals worth hunting, and the pressure from neighbouring 
tribes, and he thinks that the question of the periodical migrations 
and the origin of the deserted settlements may be solved by a study 
of the wanderings which the Eskimo undertake at the present time. 
That Boas is right in his view that the deserted dwellings are not 
due to an earlier fixed settlement is indubitable. In my paper on 
the Polar Eskimo! I have tried to give a more precise explanation 
for the occurrence of the deserted abodes in the northern and western 
parts of the Archipelago, in that I have associated the migrations 
with the occurrence of the musk ox and the comparatively rapid decima- 
tion of this animal by reckless hunting, and also with the poorness 
of these somewhat northern regions in seals and fish, which might 
serve as a support in times of stress. The views then expressed by 
me have later been confirmed by the observations of V. STEFANSSON, 
published in 1913. This author, who speaks Eskimo, and has lived 
in near contact with the Eskimo from Point Barrow to Coronation 
1 M. о. G., Vol. 34, pp. 395 sqq. 
en RE 
