90 SECTIONAL ADDRESSES 



indicates a lake in the upper Rea valley, south-west of Birmingham. This 

 expresses the hypothesis that certain clays, such as the ' india-rubber 

 clay ' of California and the similar deposits at Parson's Hill, King's Norton, 

 may have originated as lake clays when the Rea valley was obstructed by 

 ice that impounded water up to about 550 O.D. 



The line further coincides to the south-west of Birmingham with the 

 gravel deposits of Rowney Green near Alvechurch which may be regarded 

 as marginal in origin. The lobe stretching southwards complies with 

 the necessity for an extension into the lowlands of an ice-sheet that was 

 mighty enough to overspread the high ground of the Black Country and 

 the Lickey Hills. Drainage from this was carried away along the Sal- 

 warpe into the Severn, and was responsible for part of the erosion of 

 these valleys before the Kidderminster Terrace came to be formed. It 

 will be noted that the overflow from the Blythe lake coupled with drainage 

 from the retreating Eastern glacier produced similar erosion-eft'ects in the 

 Avon valley before Avon No. 4 Terrace was laid down. 



Returning to the melting glacier, the next event seems to have been 

 the splitting of the ice on the high ground of the Black Country. The 

 lobe on the east may have been concerned with initiation of the peculiar 

 drainage of the Shenstone valley which is being investigated both by 

 Mr. Wilfred Bullows and by Mr. S. J. Martin (Fig. 5, line with four 

 offsets). The lobe on the west I picture as occupying the low ground 

 west of the coalfield and of the Clent-Lickey range as far south as the 

 Salwarpe valley. It was this ice that held up the lake or lakes near Wild- 

 moor and Barnt Green which have left their record in the horizontally 

 bedded high-level sands and gravels of that district. 



As the ice shrank back the thick mounds of sand and gravel in the 

 Stour vale and near the Churchill brook were deposited. These clearly 

 antedate the Kidderminster Terrace and so fall into their correct position 

 in the scheme. 



The final stage, indicated on Fig. 5 by a line with five offsets, was sug- 

 gested to me by Mr. T. H. Whitehead. There is much evidence to justify 

 the assumption that in pre-Glacial times the Stour flowed northwards as 

 far as Hinksford, where it rounded the end of the then-unbroken Bunter 

 Pebble Bed escarpment. Ice in the position shown on the map would, 

 as already suggested (p. 88), have impounded a lake in the upper Stour 

 valley, the overflow from which might have initiated the present gorge of 

 the Stour through the Bells Mill Gap. The sands of the so-called Kings- 

 winford Esker can be regarded as having originated in this lake. 



I am very grateful to Mr. Whitehead for allowing me to use this idea 

 which he is setting out in the forthcoming memoir on the Dudley 

 district ; though I am far from confident that he will agree with the date 

 to which I assign the event. 



All the records of the further retreat of the Welsh ice sheet have been 

 obliterated by the invasion of the later Main Irish Sea glacier. 



The evidence relating to the Older Drifts that we have been considering 

 is scattered, difficult to interpret and usually ambiguous ; but nevertheless 

 I feel some confidence in the correctness of the main features of its inter- 



