362 W. Thalbitzer 



(especially of the Greenlanders) probably have their archetype in 

 the four corner-supports of the Mackenzie house ; further, the illustra- 

 tion in Petitot seems to indicate, that roof-supports occur also along 

 the edge of the main platform'), just as in the Greenland house. 

 It is only in North-East Greenland, that the houses seem to have 

 no roof-supports; they have been so small, in fact, that these were 

 quite unnecessary. On the other hand, the supports (props) were 

 needed to bear the roof in the large, long house, which appears in 

 South Greenland, both on the east and west coast, especially in the 

 south. It remains uncertain, whether this is an ancient character, 

 preserved from the house-type of the Mackenzie Eskimo, or whether 

 it has arisen independently in Greenland. 



Hans Egede described the winter-houses of South-West Green- 

 land ^) as square, built of stones and turf, 4 to 6 feet high, with the roof 

 quite flat, the largest occupied by as many as 8 families, with 10 

 to 20 (?) lamps. This agrees with the largest houses at Ammassalik. 

 In northern West Greenland such large houses are unknown^). A 

 remarkable thing about the settlement in South Greenland is, that 

 both large and small houses are found in the same districts^). 

 Oblong houses alternate with those in which the depth is greater 

 than the length of the front, cf. the measurements from Ammassalik, 

 noted p. 356. In the latter I see an intermediate form between the 

 original small houses (for 1 or 2 families) and the long houses as 

 described by Egede which now only occur at Ammassalik. — 

 There is no doubt, that the internal arrangement of the Greenland 

 house agrees precisely with Eskimo custom and use elsewhere. The 

 platform along the back wall, where each family lives separate, 

 with the hanging skins between, each with a lamp in front of its 

 compartment, is a true Eskimo feature, which is found again in the 

 dome-shaped houses of the Siberian Eskimo'^). The long entrance 

 is also found everywhere, and the position of the windows as 

 well as other essential features appears again further to the 

 west. But the idea of the long and narrow, distinctly rectangular 

 houses in South Greenland, where practically a whole village of 



ij Petitot 1887) fig. V. 



•-) Hans Egede, Perlustration (1741) pp. 63—64; Relation (1738) pp. 153, 155, 157 



lone settlement consisting of houses with 30 families in all, another one of 7 



houses with 40 families . 

 "•J Cf. Rink (1857, p. 242. 

 ■•) Rink (ibid. pp. 242 — 245, 326—327) explained this fact by referring to the 



decline in the social conditions of the civilised West Greenlanders; but I do 



not tiiink he was right in this assumption. 

 •'-) Nelson (1899) p. 258. 



