CARTER : THE EVOLUTION OF THE DON RIVER-SYSTEM. 397 
its channel and tapped the stream higher up at Wortley. That 
this was a much later effect is shown by the fact that A had 
been able to cut down its bed at least 115 feet in the interval. 
IV. — The Geological History of the Capture of the Don. 
From the close of the Carboniferous Period, with the ex- 
ception of the Trias uplift, one may say that there was a general 
tendency for the Pennine Chain to slowly subside, with the 
gradual encroachment of the Permian, Jurassic, and Cretaceous 
rocks, overlapping each other against this old land area. The 
probability is that the Pennine range was not submerged beneath 
the Cretaceous sea until Middle or Upper Cretaceous times,* 
and that there was never any great thickness of Chalk on the 
higher parts of the range. Thus when the Tertiary uplift com- 
menced the Pennine arch would reappear surrounded by a wide- 
stretching plain of oozy chalk, which would easily be disintegrated 
by sub -aerial denuding agencies, and the underlying Carboniferous 
land surface would be revealed before the major lines of drainage 
had had time to establish themselves. The early drainage 
would consist of endless runlets wandering about on the soft 
surface of the Chalk, and quickly denuding it. Where, as the 
uprise continued, an old Carboniferous river channel became 
revealed, it would at once determine the main line of drainage 
for that area, and even if it were filled with chalk the unequal 
denudation of the hard old and the soft new rocks would soon 
reveal the old features, and the new valley would be compelled 
to follow the line of the old channel. Where this commenced 
at the earliest period of Eocene denudation, there would be 
nothing to prevent the old channel being followed at all subse- 
quent phases of the uplift. This is my explanation of the pre- 
dominance of the Sheaf as the capturing river of this system. The 
valley of the Sheaf has many signs of being an ancient geological 
feature. It follows the line of a marked Carboniferous synclinal, 
which is traversed by two parallel valley faults (Fig. 3). As there 
appears to be evidence for the subsequent movement of these lines 
of fault in Permian times, and it is common for lines of weakness 
* Jukes-Browne, Building of the British Isles, p. 191. 
