130 MR. A. MONTGOMERIE BELL ON" [May I9Q4, 



river-bed was prolonged ; that there is a great interval, perhaps as 

 great as that which separates the river-bed from ourselves. The 

 evidence consists in the patination of the two groups of flints. We 

 must remember that these two groups of flints lie under similar 

 influences ; they are both in gravei and sand ; they are both 

 permeated by water depositing carbonate of lime and oxide of 

 iron. Why should they be so completely different — one class 

 deeply patinated, the other little altered? Why were they as 

 deeply patinated when the second stage began as they are now ? 

 They seem to have been so, from the evidence of secondary work 

 on the edges of ochreous implements. I, at least, cannot explain 

 it, except on the supposition that the ochreous class has been 

 exposed to weathering influences for a far greater length of time 

 than the other: in other words, that there is a great gap between 

 the two beds. If the Drift-bed has been rightlv attributed to 

 a return of cold conditions, and if the return of cold conditions 

 destroyed all the ordinary flora of the country, then the return 

 of the flora, so marked a character of the river-bed, would be a 

 kind of measure : it would mean that between the first bed and the 

 second practically the whole of our recent flora had returned to our 

 shores. 



I endeavoured for many years to work on Sir John Evans's 

 theory of river-action alone. If I may be allowed to give my 

 present hypothesis, it is that implement-bearing deposits are of 

 different kinds, and fall into three classes : — (1) River -gravels, 

 which Sir John has eloquently described. (2) Rainwash-drifts, 

 which occur at high or low levels under circumstances well 

 described by Mr. Clement Reid in the following words ' : 



' The South of England, during the second period of glaciation, seems to have 

 suffered from dry, cold winters, which froze the ground unprotected by snow, 

 and allowed the summer rains to fall on soils rendered impervious by deep 

 freezing. This led to enormous and rapid denudation, over areas where the 

 rain now sinks in and is slowly given out as springs. Masses of loose flint and 

 chalk-debris were swept off the South Downs, and spread out in a wide sheet 

 extending several miles over the lowlands.' 



If any geologist, with these words in his mind, will examine the 

 section at Knowle Farm, in Savernake Forest, he will find in them 

 a fitting explanation. The gravel is loosely thrown together, and 

 has no horizontal layers ; the newest implements are fresh, and lie 

 at the base, where they were covered by the descending debris, 

 which frequently contains weathered and worn implements of earlier 

 age. It is a gravel caused by excessive rainwash on a sloping 

 hill. (3) These are ice-drifts, such as that which has been described 

 in the foregoing pages. I would set in this class the gravel-bed 

 of Limpsfield Common in Surrey, a deposit which, more than twenty 

 years ago, was the origin of my interest in these studies, and of my 

 resolution to grapple with their perplexities. 



1 « The Origin of the British Flora ' 1899, pp. 44-45. 



