EXCAVATIONS AT DEWLISH. 223 



junction of the gravel with the sand as an undulating 

 or sinuous line, the coarser material cutting laterally into 

 the finer sand. Near the base of the trench there were 

 some large angular boulders of chalk which did not rest upon 

 their longest side, but almost on end, as though they had 

 been torn out from the side of the trench or had fallen from 

 the top into the gravel and sand. The constituents of the 

 coarse gravel were mostly flints derived from the chalk and 

 only slightly abraded, but there were in addition many of 

 the brown coated flints which occur so abundantly scattered 

 upon the surface of the water-shed. Polished, white, flints 

 were also common, many of them being pebbles or well 

 water- worn. In this rubbly flint and sand the teeth and 

 bones of the elephants were found fairly near the surface of 

 the trench. The sequence of events was, therefore, first the 

 erosion and denudation of the valley system of the neighbour- 

 hood to a degree not widely different from its present state. 

 Next, sudden and powerful downrushes of water from the 

 watershed to the bottom of the valley, which gouged out 

 the trench, and in places formed waterfalls and cascades. Then 

 came the infilling of the trench so formed by the dust-like 

 sand derived from the chalk of the neighbourhood ; and 

 finally other torrential rushes of water which cut out the 

 sand, from part of the trench and replaced it with the brown 

 flints swept from the hill-top, masses of chalk, chalk flints and 

 sand, and also deposited the elephant remains with the other 

 materials. This last phase was brought about by much wetter 

 conditions than the previous one, when only fine sand was 

 carried in, for in the sand none of the brown flints are found, 

 nor any chalk boulders nor large chalk-flints. That the 

 trench was cut by torrents is probable from the evidence 

 cited, especially the remnants of the waterfall and cascades, 

 and also, by analogy with the known modern examples 

 described by Lamplugh. The conditions in which the sand 

 originated and accumulated are not certainly known, but 

 that they were arid has been suggested by Mr. Clement 

 Reid ; and this is further borne out by the oncoming of the 



