484 THOMAS THOMSEN 



he found at Scoresby Sound, 1 and a like feature is observable in 

 the illustrations given by KOLDEWEY and NATHORST of the houses 

 they encountered in the interjacent regions. 2 Peculiar to this type 

 is the deeper-cut passage-way, and niches in the lateral walls are 

 also frequent. The size varied, in those seen by the Danmark Ex- 

 pedition, from 1-80x1-80 to 3'25x3'00 or 3'50x2'80; in the case of 

 RYDER'S from I'60x2'50 to 2'70x3'80 metres. 



The same type of house was found by AMDRUP in the settle- 

 ments he investigated at Scoresby Sound and south of there as far 

 as the Kangerdlugsuak glacier: at Cape Tobin and the Sksergaards 

 Halv0. 3 South of this glacier, we enter the territory known to 

 have been traversed by the Angmagsalik Eskimo, and where we 

 might in consequence expect to encounter the larger houses with 

 passage-way running out from one of the longer sides. Such were 

 also found, from Nordre Aputitek southward, but in addition to 

 these, the smaller houses, with greatest extent in direction of the 

 passage-way, were likewise encountered as far down as Depot Island, 

 south of Kangerdlugsuatsiak Fjord. AMDRUP notes, with regard to 

 this point, that it to some extent confirms RYDER'S hypothesis, viz; 

 that the Eskimo of North-east Greenland gradually moved southward 

 via Angmagsalik to the southern East Coast: "It is necessary, how- 

 ever" he adds, "to be .very cautious in drawing conclusions from the 

 size of the houses as to the peregrinations of the Eskimos; for the 

 size of the houses is no doubt determined by local circumstances, 

 and not by any peculiarity of the different tribes". 4 



That the question is a difficult one to determine is certain, and 

 there are good grounds for recommending caution, the more so, as 

 these houses have not been subjected to excavation. Nevertheless, 

 for my own part I consider the occurrence of the small, deep houses 

 here of some importance as indicating that families from North-east 

 Greenland must have penetrated down to somewhere near Angmag- 

 salik; I cannot but regard it as significant, that both types of house 

 are met with at one and the same spot, i. e. under identical local 

 conditions, as for instance between Cape Warming and Lang0, at 

 Nordfjord, between Kangerdlugsuatsiak and Nordfjord, and on Stor0. 5 

 At the last-named place we encounter the peculiar circumstance of 

 two large houses having been reduced in size by building new side 

 walls. If my theory holds good, then what must have happened 

 here would be this : Eskimos coming down from the North had taken 

 up their quarters in a house previously built by a party from Ang- 

 magsalik, in the same manner as the man from there just mentioned 



1 RYDER I, p. 296. 2 KOLDEWEY, p. 589; STOLPE, PI. III. 3 AMDRUP II, p. 31i 

 and 312. 4 AMDRUP II, p. 320. 5 AMDRUP II, p. 303, 301, 300 and 299. 



