38o The American Geologist. December, 1902.- 
partly filled, and are in turn overlain by lowan drift. The 
gracefully curvang surfaces of valleys and uplands are sprink- 
led with lowan bowlders. Nearly all the streams of this south- 
western area have their origin within the limits of the county, 
and they are practically branchless so far as development of 
tributary channels is concerned. Broad "sloughs," in place 
of eroded creek beds, serve to collect the waters from the ad- 
jacent slopes. While the main drainage courses seem to have 
been largely determined by the position of preglacial valleys, 
the post-Iowan streams have accomplished very little in the 
way of erosion. They have neither valleys nor flood plains 
in the ordinary sense. They run in simple shallow trenches 
cut only ,a few feet below the level of the surface on which 
they began to flow after the withdrawal of the lowan ice. 
The facts bearing on the relative age of the Kansan and 
lowan stages of glaciation, presented by the topography of the 
county we are considering, are consistent with all the facts 
which may be gathered from other portions of the state. The 
Kansan drift is deeply eroded; it was deeply eroded before the 
loess was deposited upon it. Taking the state as a whole, the 
entire Kansan surface, with the exception of a small undissect- 
ed plateau here and there on the divides, was! similarly 
trenched and carved by erosive agents. Mature erosional top- 
ography, with reliefs ranging from fifty to 200 feet, was devel- 
oped by water carving of the drift surface, before the loess 
period began. Since the time of the loess, since the withdraw- 
al of the lowan ice, the erosion of the surface of the lowan 
drift is too small to be measured, too small for relative numer- 
ical expression. The erosion of the old drift surface in the 
inter-Kansan-Iowan interval w^as certainly many hundreds of 
times as great as the erosion of the lowan surface in all post- 
Iowan time. If any other measure of the relative age of the 
two drift sheets of Howard county be applied, the same aston- 
ishing results are reached. In the valley of the Turkey river 
at New Oregon, for example, there is a terrace of old, altered, 
rusty Buchanan gravel deposited when the Kansan ice was 
waning. About a mile above New Oregon there are young 
terraces of sand and fine gravel of late lowan age, as fresh as 
if the material had been deposited by last season's floods-. The 
same thing occurs at scores of places. At Iowa City, in the 
