1899] THE DEVELOPMENT OF RIVERS 281 



which must be necessary to wear a given area down nearly to base 

 level. 



Drainage equilibrium would be established when the distance from 

 the divide to the sea is the same in either direction. That would be 

 brought about by the growth of obsequent streams. 



It seems necessary to suppose that drainage equilibrium would be 

 established before a given area could be worn down to base level. 



I confess that it is easiest to start the Thames drainage afresh on 

 an outcrop plain representing a base-level of erosion — a line drawn 

 from about 1300 feet at the Malverns to about 700 on the Chilterns. 

 Starting afresh, however, means marine denudation. 



In this plain the Chalk would extend to about Witney, the Severn 

 valley would be filled with Jurassic strata, the Inferior Oolite extending 

 about 3 miles to the west, and the Lias to the Malverns. 



An outcrop plain, produced by subaerial denudation, might have 

 been formed in Eocene or Miocene times ; but by marine erosion it 

 only seems possible in early Pliocene. That does not seem to give 

 sufficient time. All but the latest details of the Severn valleys are 

 Pre-Pleistocene. However, suppose an outcrop plain, and consider 

 the drainage. 



The plain has a general south-east dip, so far as the main area of 

 England and part of Wales is concerned. But this plain seems to 

 have been bounded on the south by a more or less continuous and 

 irregular line of hills — the anticlines of the Mendips, of the vales of 

 Pewsey and Kingsclere, and of the Weald. The consequence was that 

 the drainage of what may be called the great central plain was towards 

 the south-east. But at the southern end of the plain it was checked 

 by the anticlines ; from their north side came streams draining north- 

 wards, and hence the drainage was given an easterly turn at the south 

 of the plain. Fig. 7 illustrates the position of the supposed lines of 

 original drainage and the positions of the anticlines. 



The initiated streams are all consequents; but as they cut down 

 their valleys subsequent streams would be started. 



As Professor W. M. Davis points out, the Ock and the Thame 

 are both subsequent streams. So is the Thames from beyond its 

 junction with the Churn to its junction with the Evenlode. This 

 Berkshire Thames was evidently developed before the Ock, by a 

 subsequent growth working west from the Evenlode ; it successfully 

 cut into and diverted all the Cotswold streams. Evidently the Wind- 

 rush once joined the Thames direct somewhere near Abingdon ; but the 

 Evenlode, perhaps helped by the Cher well, had so successfully lowered 

 the country around Oxford that curiously enough it was able to divert 

 the Windrush to a more immediate lower level though giving it a 

 slightly longer course. This was done by the initiating Berkshire 

 Thames subsequent. 



The Evenlode was a very large river, draining by one branch the 



