142 On the PalmoUthic Flint Implements from Knowle. 



the Knowle beds are situated. The rains, too, would have run off 

 the clay-with-flints, instead of soaking into the chalk. These 

 causes would have been amply sufficient to supply the waters 

 needed for the excavation of the now dry vallies. The most im- 

 portant of these had its rise a short distance from the eastern 

 slope of Martinsell Hill, and after traversing Savernake Forest 

 from west to east came out slightly to the north of Knowle. It 

 was here joined by other vallies that had also taken their rise in 

 the same direction, and the waters from all of them went south- 

 eastward into the Valley of Kennett. The period when these vallies 

 were formed must be assigned to the Glacial Epoch, which so 

 largely affected the surface of the country. England north of the 

 Thames is believed to have been covered with an ice-cap, and 

 although this did not extend further south the climate must have 

 been the same over the whole country. It was one of intense cold, 

 varied at intervals by inter-glacial periods accompanied with 

 an elevation of temperature. Considerable meltings of the snow 

 and ice would take place in these periods, and the water thus 

 produced would come down' from the high lands in violent floods, 

 bringing with it the stones, trees, and other debris that it had 

 previously collected, or that obstructed it in its flow. The under- 

 lying ground would still be deeply frozen and quite prevent the 

 water from soaking away, and the mechanical effect of such a mass 

 of water, stone, and other matter would each time tend to deepen 

 the vallies. Although many of the implements at Knowle were 

 made at the place where they are now found, many others would 

 have been brought down from other settlements in this rush of 

 water, and Knowle, being a piece of rising ground at the confluence 

 of the vallies, would arrest this material, the water receding but 

 leaving the heavier matter behind. Many implements are found 

 rolled and worn and still retaining, the marks of glacier scratching. 

 It is reasonable to suppose that during some one or more of these 

 inter-glacial periods the settlement may have been suddenly 

 abandoned by the inhabitants (or overwhelmed by the floods), and 

 all their belongings left , there. Nothing but the imperishable 

 stone implements have survived and there is no trace of any other 



