CARTER : THE EVOLUTION OF THE DON RIVER-SYSTEM. 389 
strike for short stretches between. That a river should leave 
a straight course and form wide bends, not as it leaves but as 
it enters a hilly region, appears a remarkable fact, and one 
needing explanation. At Deepcar (440 feet), where it receives 
on its right bank the Little Don, the river resumes its straight 
course, and flows almost south, cutting through the great grits 
of the Grenoside, Greenmoor, and Whamcliffe Rocks of the 
Lower Coal Measures, with their imposing escarpments. After 
being reinforced by Ewden Beck on the right bank, it enters 
a V-shaped valley asrain at Worrall, receives the combined 
waters of the Loxley and the Rivelin on its right bank, and, 
flowing thence south-eastwards to Sheffield (150 feet), enters 
the Middle Coal Measures (Fig, 3). At Sheffield the Don, 
making a rectangular bend to the north-east, falls into what 
I regard as the valley of the Sheaf. The River Sheaf rises on 
Totley Moor (1,100 feet), flows X.X.E. over the Lower and 
Middle Coal Measures to Sheffield, where it joins the Don, and 
tlience passes in a broad alluvial valley to Rotherham (84 feet), 
where the Rother enters it on its right bank. At Denaby it 
receives the Dearne on its left bank (45 feet), and passes through 
a picturesque gorge in the Magnesian Limestone at Conisbrough 
(PL LII.) into the Triassic plain (40 feet) at Doncaster. Thence it 
continues its north-easterly direction to Thorne, where its channel 
has received much alteration from artificial diversion. The 
Old Don, the course of which can be traced north of Hatfield 
Chase and by Crowle to Adlingflcet on the Trent, was probably 
the pre-historic bed of the stream, which appears to have been 
artificially diverted at various times to the Trent at Keadby, 
to the Aire at Snaith. and finally by the Dutch River to the 
Ouse at Goole. 
II. — RivEK Diversion. 
The two principles of Physical Geology which are invoked 
to explain the peculiarities of the course of the Don and its 
tributary streams are :— 
(1) River capture ; (2) Glacial diversion. 
Both these appear to have influenced the course of the 
Don, but the former has been by far the more important agent 
in producing the erratic course of that river. As river capture 
