188 The Geology of the Berks 8f Hantf Extension, 



The section shews how slight the partitions between the valley, 

 and the coombs near St. Ann's Hill now are, and we cannot doubt 

 that a marine erosion greater either in duration or intensity, acting 

 between a and b or near b only, would have reduced the "col" 

 by some 250 feet in height. 



There are many cases in which such connections between the 

 valley and the coomb have been made, and which are now passes at 

 a high level between the higher hills of the escarpment. Such 

 is the hollow leading from Bratton to the downs, and that from 

 West Lavington by Gore Cross to Tilshead, and such a connection 

 may well have been made at the upper end of the Avon valley. 

 Without supposing a case so extreme as that which has nearly 

 occurred on the opposite side of the valley, we can readily see how 

 a passage could have been opened at as low a level as that of the 

 present river. But marine denudation which acts only at or near 

 the sea level, could not open a connection between the valley and 

 the coomb at a lower level than the bottom of the valley as it 

 existed when it rose above the sea, and when marine denudation 

 ceased. At what height then in the present valley are we to draw 

 a line which shall roughly represent the limit of marine denudation, 

 and below which everything has been removed by subserial causes ? 

 It will clearly be higher than the post-pliocene drift at Pewsey, 

 and as high or higher than the oldest and highest gravel in that 

 part of the valley. The flint gravel of Pewsey Common is the 

 highest with which I am acquainted. It is 38 feet higher than the 

 post- pliocene drift at Pewsey Station, and 90 feet above the road 

 near Lydeway. At the level of this gravel the line in the section is 

 drawn, and lower than the line, if it truly represent the limit of 

 marine denudation, the connecting gap between the valley and the 

 coomb could not have been opened by that agency. The difficulty 

 then does not lie in the depth of the coomb, and the low level of 

 the connecting gap, but the difficulty is that if the waters took 

 their course down the coomb when the connecting gap was at the 

 level of the line in the section, the watershed between the valley 

 and Urchfont bottom must have been tilted up higher than the 

 line in the section, and subsequently lowered to its present level, 

 that is, 90 feet below the line. 



