Ethnographical collections from East Greenland. 411 



division of this weapon into toggle and barbed harpoons ^) can 

 thus in so far not be used here. But, as he himself admits, the 

 two types grade insensibly into one another, and in many Green- 

 land harpoon-heads the line is drawn through so close to the butt 

 ■of the head, that the direction of the pull differs very little from 

 the longitudinal axis of the head; in such heads it is probable, that 

 я toggling of the head does not take place in the wound, and it is 

 only the barbs of the head which prevent its falling out. — A more 

 detailed description of the Ammassalik types of harpoon-heads will 

 Ъе found pp. 424—430. 



Regarding the form and function of the loose shaft — the char- 

 acteristic bone-joint of the harpoon between the head and the shaft 

 proper — I may refer to G. Holm's description here p. 46 and to 

 'O. Mason's detailed description^), which like his other studies on 

 the Aboriginal American Harpoons, is very instructive. 



On only one point in Mason's account is a correction necessary. In his 

 description of the East Greenland harpoon he states: "The foreshaft is in 

 this specimen a cap of ivory, squared off on top, and the middle left pro- 

 jecting for the socket on the base of the loose shaft" (1. с p. 238). According 

 to this the loose shaft would have a socket on the basal surface, covering a 

 corresponding projection on the top of the foreshaft. The same explanation 

 is repeated in describing a second example from South Greenland and his 

 illustration of this part of the harpoon (fig. 49 in Mason) shows the same 

 peculiar feature''). It is probably based on some mistake. The condition in 

 all the Greenland harpoons, which I have seen, has always been that the 

 tenon (or projection) was on the base of the loose shaft and the socket on the 

 flat top of the foreshaft*). There is some doubt, as to whether Mason has 

 , described his own specimens correctl}'^ on this point. In the first place it is 

 unusual, that the two Greenland harpoons, he describes, should differ from 

 the Greenland type known elsewhere; in the second place, there is a con- 

 tradiction in Mason's description. On the same page, namely, where he 

 describes the foreshaft erroneously (p. 238), he explains, in full agreement 

 with the usual type of Greenland harpoon, that the loose shaft has a "flat 

 surface at the base, with a projection in the middle, fitting into a cavity on 

 the front of the foreshaft", and his drawing of this harpoon (PI. 4) is accurate 

 and correct. It thus appears, that through forgetfulness he has given an 

 erroneous description of the ball-and-socket joint which he had described 

 correctly on referring to the loose shaff^). 



The foreshaft of the harpoon, as Mason states, is "the working 

 end of the shaft and is usually a block of bone or ivory neatly 

 ütted on". The remark agrees with the harpoons and lances of the 



1) O. Mason (1900) p. 199, cf. 237. 



2) Id. (ibid.) pp. 2Ü3— 204; cf. 242. 



^) 0. Mason, 1. с pp. 255 — 256: "The loose shaft is — — square at the base and 

 socketed to fit over a small projection on the foreshaft," cf p. 242 (below). 



*) Otto Fabricius (1810) p. 135. 



■•■') In F. Nansen's "Eskimoliv" (1891) p. 31 there is a drawing of the front end of 

 a harpoon, which, though indistinctly, shows the same error in confusing the 

 loose shaft and foreshaft. Nansen is cited by Mason (1. с p. 240). 



