AXES. KNIVES. 19.") 



water, rising and sinking with it ; * or whether they were 

 fixed, and rested on a sort of artificial island, formed by the 

 clay, branches, etc., which now occupy the interspaces between 

 the different platforms. Subsequent observations, however, 

 confirmed as they have been by discoveries elsewhere, as, for 

 instance, at Inkwyl and Niederwyl, have decided the question 

 in favour of the latter hypothesis. 



During my visit at Wauwyl we obtained four small stone 

 axes, one arrow-head, four flint flakes, fifteen rude stone ham- 

 mers, eight whetstones, thirty-three slingstones, eight instru- 

 ments of bone, and two of wood, besides numerous bones, and 

 a great quantity of broken pottery. Col. Suter regarded this 

 as a fair average day's work. Altogether about 500 instru- 

 ments of stone and bone had been discovered at Wauwyl; 

 at Moosseedorf more than 3300 ; at Wangen no less than 

 5800, while M. Troyon estimated that those at Concise must 

 have amounted to 25,000, and these numbers have since been 

 largely increased. 



The axe was pre-eminently the implement of antiquity. It 

 was used in war and in the chase, as well as for domestic 

 purposes, and great numbers have been found, especially at 

 Wangen (Lake of Constance) and Concise (Lake of Neuf- 

 chatel). With a few exceptions, they are small, especially 

 when compared with the magnificent specimens from Den- 

 mark ; in length they varied from one to six inches, while the 

 cutting edge had generally a width of from fifteen to twenty 

 lines. Flint was sometimes used, and nephrite or jade in a 

 few cases, but serpentine and diorite were the principal mate- 

 rials. Most of the larger settlements were evidently manu- 

 facturing places, and many spoilt pieces and half -finished 

 specimens have been found. After having chosen a stone, 



* Dwellings of this character sion to Sumatra, p. 395. Squier, 

 occur in the East and in S. America. American Naturalist, vol. iv. p. 18. 

 See, for instance, Anderson's Mis- 



2 



