Pleistocene Rock Gorges. — Hershey. :?17 
that a stream one-tenth as targe as the main creek flows in a 
valley ten times as large. 
Comparison with postglacial valleys. — From the published 
accounts of valleys in the newer drift, and from personal ex- 
amination of such gorges as those of the Wisconsin river at 
the Dells, and the Niagara river below the falls, it is found to 
be the rule that the great majority of postglacial rock gorges 
are but little, or often not at all. wider than the streams which 
occupy them. Frequently their sides are perpendicular walls 
of bare rock, with the stream touching the base on either 
side of the canon. The streams also are contracted to less 
width in the gorge than where the valley is wider and the 
drainage gradient less. Postglacial time has not been of suf- 
ficient length to allow the excavation of the valleys to pro- 
ceed in many cases to such an extent as to drain the lakes 
above them. 
But in the rock gorges of northwestern Illinois we find quite 
a different state of affairs. Not a single lake or pond remains, 
and the swamp areas that do exist are due to the postglacial 
deposition of river alluvium. And not only have the gorges 
been cut to the normal drainage gradient, but they have been 
widened to many times the width of the present streams 
flowing in them. 
The determination of the exact length of time that would 
be required in the erosion of the Pleistocene valleys is impos- 
sible at the present time, as many of the governing conditions 
are not thoroughly understood. One of the chief factors is 
the size of the streams which carried on the excavation. Al- 
though there is no direct evidence bearing on the width of 
the early streams, it is probable that they were larger and 
more efficient than those of the present day. Hut even though 
they may have carried away twice as much water as at the 
present time, they are not likely to have been much wider 
within the gorges than those which now occupy them. for. as 
the eroding power depends largely on the strength of the cur- 
rent, a concentration of the waters would occur within the 
narrow parts of the valley. That they were not Dearly so 
wide as the present gorges is evident, for it would be prepos- 
terous to suppose that a stream 50 feet wide could flow from 
