225 



which are exposed to the south or south-west are more denuded, 

 and much less precipitous than those on the opposite sides of 

 the valleys. I attribute this difference to the washing of rain, 

 which, on the one side, falls with full force upon the slope, while 

 the opposite side is comparatively protected.* 



Again, it may be observed, that on the edge of a wood on 

 sloping ground, there is usually a considerable drop from the 

 surface of the woodland to the cleared ground. This is caused 

 by rainfall, and by the water dropping from the overhanging 

 branches. If the wood were to be cleared and converted into 

 tillage land, its ancient boundary would be marked by a terrace 

 which might weU be mistaken for an ancient water line.f 



Another cause of denudation by rainfall is the carrying away 

 of earthy matter from springs. An analysis of the spring 

 waters from the Fullers Earth gives 28 grains per gallon, 

 and that of the Upper Lias 20 grains per gallon, being an 

 average of 24 grains; 290 gallons would give a pound, and 

 650,000 gallons a ton; taking the amount of rainfall annually 

 thrown out in springs at 9 inches, would give upwards of 200 

 tons per square mile, which will roughly represent the same 

 number of cubic yards. Multiply this quantity by the area of 

 the Cotteswolds, and we may form some idea of the prodigious 

 quantity of solid matter abstracted every year in an invisible 

 form in apparently clear water. Then we must allow at least 

 as large a quantity of earthy matter to be carried off by storm 

 waters, making a total of 400 yards per mile. At this rate of 

 denudation, in about half the period assigned by Sir Charles 

 LvEiiL for the subsidence and subsequent upheaval during the 

 glacial epoch, soil would be removed equal to the area of a 

 valley 600 feet wide, and of equal depth in the centre throughout 

 the whole mile of ground from which the above quantities are 

 taken. 



It is scarcely necessary to refer to the effects of frost upon 

 the soft Oohte. It is well known what an important agent it 

 is in widening valleys. The Freestone and OoHte Marl beds 



* See observations on Denudation by Eainfall in "Earn and Eivers," by Colonel 

 Greenwood, which are peculiarly applicable to the Cotteswold District, 

 t Vide "'Rain and Rivers." 



