46 SCIENCE PROGRESS 



will remove a flake almost devoid of ripple-marks, and it is 

 supposed that such a blow has less jarring effect upon the flint, 

 and consequently produces less rippling of the fractured surface, 

 than one delivered in an oblique direction. The explanation 

 of the frequent occurrence of prominent ripple-marks upon 

 fortuitous flakes, and their comparative scarcity upon those 

 produced by human blows, may be as follows. It may be 

 assumed that there are 180 angles at which a flake may be 

 removed from any given length of edge, but it is only more or 

 less vertical blows, delivered at a few of the higher angles, 

 that will remove flakes exhibiting insignificant ripple-marks. 

 Thus in fortuitous haphazard blows the chances are that the 

 greater number of blows will be delivered at the lower angles, 

 which are oblique to the edge of the flint, and which, as has 

 been shown, remove flakes exhibiting prominent ripple-marks 

 upon their surfaces. It seems, then, that in the shape of the 

 flakes, their character at the point of final separation from the 

 parent block, and the presence or absence of prominent ripple- 

 marks upon their surfaces, afford further criteria for determining 

 whether any particular series of flints has been flaked by man 

 or by nature. 1 



It was noticed that the edges of all the wedge-shaped stones 

 fractured in the sack experiment showed a tendency to assume 

 a sinuous outline due to blows having fallen upon either side 

 of these edges. 



The majority of the stones exhibited this sinuosity in an 

 incomplete manner, but one specimen showed a markedly 

 sinuous edge 4! inches in length. A critical examination 

 (based upon the tests already described in this paper) of the 

 flaking to be seen upon this flint demonstrated clearly its 

 " natural " origin, but another characteristic of the flaking was 

 detected which seems to still further differentiate it from the 

 work of man. It was found that the edge of this flint, though 

 only 4\ inches in length, exhibited seventeen truncated flakes 

 (fig. 3). 



On the other hand an examination of various neolithic and 

 palaeolithic implements with sinuous edges demonstrated that 

 the average number of truncated flakes on an edge 7 inches long 

 is only six. 



1 In arriving at a decision it is necessary to examine not less than twenty or 

 thirty specimens ; a lesser quantity examined might lead to erroneous conclusions. 



