379 



thread with which the blade has been secured to the haft 

 through two corresponding holes in its lower part. The upper 

 part of the blade has been driven home into a slit on the 

 lower side of the haft; but even if the blade could go right 

 to the bottom of this slit, its two holes can never have 

 come down far enough to tally with those in the haft, the slit 

 not being deep enough for this. The two holes in the blade 

 must accordingly have lain outside the slit, and been visible 

 under its lower edges, and the sinew cord must have run 

 from the holes in the blade up along both outer sides of the 

 bone haft to the holes in the latter in which it was secured. 

 Traces of the tautening of the lashing can still be seen in two 

 grooves in the upper surface of the bone under one of the 

 holes (cf. fig. 21a). 



The form and appearance of this blade correspond more 

 or less to the blade in one of the ulos which Ryder ^) brought 

 home with him, where it was found fixed in a wooden haft 

 and secured with a sinew lashing in a similar manner to the 

 knife found by Amdrup, namely with exteriorly fastened straps 

 passing from the holes in the blade to the holes in the haft. 



Most of the ulo blades hitherto found in North East Green- 

 land are of a somewhat different type, having the shape of a 

 section of a circle in which the edge describes the arc and the 

 two blunt side edges form the radii. BothRyder^) and Nathorst-'^) 

 found women's knives of this type, made entirely of stone, 

 without a haft of bone or wood. The curved edge, however, 

 is no doubt a feature which the two types have in common. For 

 the rest, several specimens of the type of a circular section 

 are imperfectly cut, tlie upper part (i. e. that nearest to the 

 centre) being broken off, and the quadrilateral blade thus formed 

 being perforated and inserted into a haft. As for other details 



'1 Ryder 331, ßg. 29c. 



*) Ryder 331, fig. b. 



') Stolpe PI. VI, ng. 19; PI. Ill, fig. 10; the same in Solbeig 56, fig. 48. 



