C— GEOLOGY 75 



able that in pre-Glacial times the upper Severn went to the Irish Sea, that 

 the watershed of England separated it at Iron Bridge from the middle and 

 lower Severn, which then had its source where now the Worfe rises. 

 From here, as indicated diagrammatically on Fig. i, it may have followed 

 the line of the great depression now occupied by the valleys of the Worfe 

 and Claverley Brook, the Lower Stour, the Elmley Brook and Salwarpe, 

 and the Bow and Piddle Brooks. 



In throwing our minds back to this distant period, however, it is essen- 

 tial to remember that a study of the sub-drift surfaces and of the river 

 terraces makes it abundantly clear that the whole river system of the 

 Severn and Avon was then at a considerably higher level than it is to-day, 

 though naturally the differences in level between the present and the pre- 

 Glacial valley-floors decreases almost to the vanishing-point as the water- 

 sheds are approached. These relationships do not obtain, at any rate on 

 the same scale, in neighbouring drainage basins, particularly in those 

 draining to the Humber and Wash. The difference in behaviour is in 

 my opinion due primarily to the fact that the Severn and Avon were, all 

 through the Glacial period, principal lines of drainage from the ice front, 

 and in the later stages also received from the Upper Severn water that 

 should have gone to the Dee and Mersey. Thus they carried far more 

 water then than they did in pre-Glacial times and than they do to-day. 

 The overdeepening of their valleys is more the outcome of increased 

 volume than of increased velocity due to elevation. This hypothesis 

 explains the differences between the Severn and the other rivers without 

 having to invoke differential uplift. 



A thorough appreciation of the vast extent to which erosion has gone 

 on, and of the enormous length of time involved at once helps us to under- 

 stand the apparently anomalous distribution of Glacial drifts in this region, 

 in which, as a rule, the vales and lower ground are free from Glacial 

 deposits, whereas the higher country and the watershed areas are exten- 

 sively and often heavily drift-covered — a disposition that is the exact 

 converse of the usual arrangement in a glaciated region. I shall attempt 

 to show in the sequel that the drifts formerly extended beyond the regions 

 where they now form large outcrops and that their absence from an area 

 need not be taken as an indication that it was never under ice. 



It is advisable that at this point a brief statement be made of the geo- 

 graphical distribution of the areas that may be termed drift-covered. In 

 some cases the Glacial deposits are thick and continuous, in others they 

 may be more scattered and are often thinly developed. The term ' drift- 

 covered ' is nevertheless appropriate when comparison is made with the 

 rest of the region which can fitly be called ' drift-free,' though here again 

 the description is not literally true. The drift-covered areas may be 

 grouped in relation to the major watersheds, as follows : 



I. Watershed between the Tern and Penk on the north and the Severn, 

 Worfe and Smestow on the south, and between the Penk and the 

 Shenstone Brook. This area extends over high ground from the 

 Wrekin towards Wolverhampton, and thence north-east towards 

 Cannock Chase. It also covers lower ground in the Worfe basin 



