226 



are quite incapable of resisting its action, and a cliff formed of 

 these beds would, under its influence, in a few years become a 

 mere slope, covered with Oolitic detritus ; thus the frost, in a 

 certain sense, commences the process, which is subsequently 

 completed by rain. No doubt when a colder climate prevailed 

 in England, the action of frost, ice, and snow was much more 

 effective in denuding the country than in recent times. The 

 accumxdation of snow in winter, and its melting in summer, 

 would tend to give increased volimae and an excavating force to 

 the rivers, of which no idea can be formed from their present 

 condition. So the accumulation of ice on the slopes, and its 

 periodical removal, would tell with effect upon the soft materials 

 of which the hills are composed. 



If we take a general survey of one of the long valleys of the 

 Cotteswolds, with its numerous branches and combes, it is 

 difficult to resist the impression that atmospheric action, 

 running water, and landslips, have together accomplished its 

 formation. As we ascend the valley, we find the stream which 

 runs through it becomes reduced in volume, the denudation is 

 less, imtil at length the valley becomes a mere tmdulation of 

 the surface, and finally it ceases to be discernible. But what 

 do we then observe ? Why vdthin a few yards, perhaps wit hin 

 a few feet, the surface begins to slope in the opposite direction, 

 and it gradually sinks into a valley, very similar to that we 

 have just ascended; and this condition of things is applicable 

 not only to the main valleys, but, in districts where the country 

 is much intersected, to their branches and combes. I do not 

 give instances, because they are so common as to form the rule 

 . rather than the exceptions. Thus the whole of the Cotteswold 

 district is intersected by valleys, running either in the direction 

 of the Severn on the one side, or the Thames and Avon on the 

 other. 



I have alluded to some of the combes of the Cotteswolds as 

 dry valleys. They usually commence in undulations on the tops 

 of the hiUs, form deep hollows in the slopes, and have springs 

 issuing from them at the level of the Upper Lias. The 

 circumstance of these combes being dry might at first appear to 



