VOL. XVI. (3) THE LOWER SEVERN 251 
process of “ working backwards” is not to be reconciled 
with the words of Sir A. Geikie -_— A river can only act 
upon the rocks over which it flows.” Every stream flows 
away from, not over, the interval between its head water 
and the divide which separates it from the head water of 
another stream. 
On full consideration of all the arguments that I have 
heard or read, I cannot see any reason for believing that 
while the river-system of the Severn, as a whole, origin- 
ated in the Welsh Mountains, at the top of an incline, a 
part of the system originated in the Bristol Channel, at 
the bottom of the incline. I regard such expressions as 
“river capture,” “working” or “ cutting,” headwards or 
backwards, and the terms “consequent,” “subsequent” 
and “ obsequent ” as incorrect and misleading. 
Passing now to the Lower Severn Valley, as seen at 
the present*time, fig. I, we find that it narrows between 
Bredon Hill, an outlier of the Cotteswolds, and the 
Malvern Hills, at the point where the valley of the Avon 
joins that of the Severn. Below this is the Vale. of 
Gloucester, which extends downwards to the point where 
the river comes close to the western hills at Newnham: 
here the Vale of Berkeley succeeds. Above Gloucester 
the valley is divided into three minor valleys, fig. 7A. 
COTTES WOLD HILL 
m v9 
Fo So 8 
oe ew 
CORSE HILL 
WESTBURY TIBBERTON DOBSHILL 
QUE ag 
ASEVERN 
R.GEVERN 
Bowe oem 
Fic. 7.—Diagram-section of the Severn Valley above Gloucester. A, transverse. B, longitudinal, show- 
ing, by a broken line, an arm of the river flowing along the valley at a higher level, and the 
present conditions below. : 
