132 AXES. KNIVES. 



varied from six inches to one, 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 was the principal material. Most of the larger 

 settlements were evidently manufacturing places, and many 

 spoilt pieces and half-finished specimens have been found. 

 The process of manufacture is thus described by M. Troyon : 

 After having chosen a stone, the first step was to reduce it 

 by blows with a hammer to a suitable size. Then grooves 

 were made artificially, which must have been a very tedious 

 and difficult operation, when flint knives, sand, and water 

 were the only available instruments. Having carried the 

 grooves to the required depths, the projecting portions were 

 removed by a skilful blow with a hammer, and the implement 

 was then sharpened and polished on blocks of sandstone. 



Sometimes the hatchet thus obtained was simply fixed in a 

 handle of horn or wood. ^Generally, however, the whole 

 instrument consisted of three parts. A piece of horn, two or 

 three inches in length, received the stone at one end, and was 

 squared at the other, so as to fit into a longer handle either 

 of wood or horn. These intermediate pieces present several 

 variations ; some are simply squared, while others have a pro- 

 jecting wing which rested against the handle ; some few are 

 forked as if to receive a wedge, and one has a small transverse 

 hole, apparently for the insertion of a peg. It is remarkable 

 that while in some places these horn axe-handles are numerous, 

 this being especially the case at Concise, whence several hun- 

 dred have been obtained, in other Lake- villages they are very 

 scarce: at Wangen, for instance, though more than 1,100 stone 

 axes have been found, M. Lohle has as yet met with only a 

 few handles, all of which were of wood. The axes appear to 

 have been fastened into the handles by means of bitumen. 



The stone knives may be considered as of two sorts. Some 

 differ from the axes principally in having their width greater 



