i899 S. BUCKMAN. — WYE VALLEY 27 



Kennet ; and to this river system belonged a stream 

 occpi)ying somewhat the course of the i)resent Lower 

 Wye. It was a river which had developed in an area of 

 Mesozoic strata. 



Later this river system was broken into by the growth 

 of the Severn from the (present) Bristol Channel north- 

 eastwards. Working its way back, it cut into these 

 Thames-flowing rivers, and, tapping them, diverted their 

 head waters to itself. It could give them so much 

 quicker fall to the sea than if they travelled all the wav to 

 East England. 



This will account well enough for the steep, gorge-like 

 valley of the Lower Wye. For when the Wye was 

 flowing at a level high enough to carry it over the Cottes- 

 wold escarpment at Bath, and the growing Severn was 

 working back in a channel 200 or 300 feet lower, as soon 

 as the Wye w^as tapped by the Severn it would be enabled 

 to lower its bed by the difference between them. And in 

 thus rapidly lowering its bed it would cut a gorge-like 

 steep-sided channel. 



The Cottesw^old Hills, relatively to the Severn Vale, 

 furnish many parallels to this presumed state of affairs, 

 before the Wye had been tapped by the Severn. Thus, 

 near Andoversford, the Coin is flowing in a bed 500 feet 

 above sea-level. The bed of the Chelt, at Cheltenham, 

 some five miles distant, is 300 feet lower, and that level 

 the Coin does not reach till it has travelled some 40 miles, 

 not counting windings. So when in time the Chelt taps 

 the Upper Coin, the latter will be able to deepen its bed 

 very rapidly, and will make a gorge-like valley. Such has 

 happened with the upper waters of the Stroud stream, 

 the Frome, and its wonderful valley at Sapperton.* 



* 'l"he views upon this and other river developments have since been put forward in 

 a paper in "Natural Science," vol. xiv., p. 273, 1899, to which the reader is referred. 



