C— GEOLOGY 



93 



independently by both Lapworth and Harmer, the main feature of which 

 was that a lake was held up by the ice sheet on the north-west side of the 

 pre-Glacial watershed at Iron Bridge ; and that this lake drained away 

 across the divide, and thus initiated a gorge that became so deep that it 

 has permanently retained the Upper Severn drainage which formerly went 

 out to sea either by the Dee or by the Mersey. This lake I named Lake 

 Buildwas (Fig. 6). 



At this stage then there were two lakes, Buildwas and Newport, on the 

 north-west side of the watershed, one draining to the Trent and one to 

 the Middle Severn. They were separated by the ice where it impinged 

 on the Wrekin. When the glacier melted back further and allowed the 

 lakes to join and form ' Lake Lapworth,' so nearly at the same level were 

 the outlets, that it was a mere matter of chance that the Upper Severn went 

 permanently to the Bristol Channel and not to the Humber. As it hap- 

 pened, the Iron Bridge outlet was, or at any rate soon became, the lower. 

 It took all the discharge and has retained it ever since. This implies that 

 when the ice left Cheshire, the drifts of the Cheshire plain formed across 

 the old pre-Glacial valley a barrier that was higher than the Iron Bridge 

 outlet at the time. It seems likely, therefore, that Lake Lapworth had by 

 then been lowered considerably by the partial destruction of the rock sill 

 at Iron Bridge. 



These glacial accidents have been the factors that have determined 

 much of the geography of the Midlands ; for they diverted into the rela- 

 tively small pre-Glacial catchment basin of the Lower and Middle Severn 

 great volumes of water which have rejuvenated the river, especially in its 

 middle reaches, on a stupendous scale. The rejuvenation is still operative 

 and can be seen to-day in the erosive activity of every tributary of the 

 Middle Severn. The relationship of the Main Terrace and its correlatives 

 in the Avon and Trent to the present valley-floor also displays in a striking 

 way the influence of the increased erosive activity on the shape of the 

 valley. The low terraces of the Tame and Trent system which has had 

 no rejuvenation of this type, rise a few feet above the alluvium and extend 

 downwards below the valley floor ; the surface of the Second Terrace in 

 the Avon keeps parallel to the present flood plain, but some 20-30 feet 

 above it. Here the rejuvenation has been slightly felt. On the other 

 hand, in the Severn valley the present floor lies 20-30 feet below the Main 

 Terrace at Tewkesbury, but 100 feet below it at Bridgnorth. In fact, 

 from Worcester upstream the Severn is increasingly incised, and its 

 tributaries, the Salwarpe, the Stour and the Worfe, join it in deep trenches. 



Climatic conditions during the Main Irish Sea glaciation were extremely 

 severe. Solifluxion and melt- water floods were on a correspondingly 

 grand scale in the periglacial region. There are vast spreads of local, 

 often angular, detritus at the foot of the Cotteswold and Malvern Hills, 

 and in the valleys draining the high ground of Enville and the Clent-Lickey 

 range, which resulted from these conditions. Most of these grade down 

 to the Main Terrace level in the adjacent valley, and may be correlated 

 with that terrace and thus with the third glaciation ; though some seem to be 

 still younger and to correlate with the Worcester Terrace and the Welsh 

 Re-advance. 



