﻿342 THE EEV. E. C. SPICEE OS! [Aug. I908, 



of shallow water stretching from bank to bank when the rocks on 

 each side are saturated with water that oozes quietly out into the 

 main yallej' and deposits more material upon the level floor. 



Valleys of a similar character are found (as at Stonesfield, and 

 above Box near Bath) in all the Great Oolite region, and a grada- 

 tion is suggested to the Mendip, Yorkshire, and Derbyshire valleys 

 with their underground streams and huge unroofed valleys similar 

 in origin to that of Cheddar. 



It may be asked why, if these Great Oolite valleys are solution- 

 valleys, the same theory does not apply to the valleys of the Chalk ? 

 The answer is that it does, and it is hoped to show later that the 

 solution-theory may be appKed in a still more striking way to 

 the Chalk-valleys. 



The adoption of any violent mechanical hypothesis, such as a 

 Welsh Thames, or an Oxford-Lake gorge, or glacial scour, or the 

 rush of marine currents, or the march of an ice-cap, will become 

 unnecessary, if the long-continued action of an efficient agent 

 proceeding quietly is found sufficient to account for the facts. The 

 meandering Thames above Oxford fed by water from the limestone- 

 plateau may lose the interest of a legendary past, but will be 

 studied with increased interest in the present as the rational result 

 of the causes that I have ventured to indicate, and therefore as a 

 river still in vigorous solvent activity, and not in a moribund 

 condition of senile mechanical decay. 



EXPLANATION OF PLATES XXXVIII & XXXIX. 



Plate XXXYIII. 



Fig. 1. Head of dry valley near Stonesfield : Evenlode Valley in the distance. 

 The plateau (here covered with drift) sinks suddenly into several 

 winding valleys similar to that which is represented. All these 

 vallej-s merge into the Evenlode Vallej', thus widening its banks 

 locally, and largely producing its winding curves. The Evenlode- 

 flows under the tree-covered ridge on the right of the picture, on the 

 side nearest the observer. 



2. Brook-valley (with trees) entering the Glyme Valley. The plateau, 



quite free from drift, sinks in a manner similar to that seen in fig. 1, 

 into dry vallej-s entering the brook-valley (with trees), which in its 

 tuz*n similarly enters the stream-valley of the Grlyme, surrounded by 

 marshy ground, at the right of the picture. The Glyme Valley ruu& 

 lengthwise across the picture. 



3. Mouth of developed dry valley, with characteristic cross-contours, 



entering the left bank of a brook-valley near Grimm's Dyke. The 

 observer is looking up the valley. 



Plate XXXIX. 



Fig. 1. The Dorn Valley is seen emerging on the left just beyond the soakage- 

 bowl. The Dorn stream enters the Glyme near the arrow, but 

 makes no impression upon the straight spur on the right. The 

 Grlyme-Dorn then turns round this spur, and flows down the valley to 

 the right towards Woodstock, along a valley similai-ly flattened by 

 solution-debris. 

 2. The same area flooded with quiet water covering the stream-meanders. 

 This flooded condition sometimes lasts for several months when the 

 rocks are saturated. 



