THE ILLINOIAN DRIFT SHEET. 89 



EFFECT OF THE ILMNOIAN ICE INVASION ON THE OUTER-BORDER 



DRAINAGE. 



A most important question connected with the Illinoian invasion is its 

 effect upon the drainage lines of the border districts. The slope of the 

 region west of its western border in Iowa is directly toward that border, 

 and the large streams had apparently opened channels to about the present 

 line of the Mississippi prior to the invasion of this ice sheet. The ice 

 nowhere extended more than 20 miles beyond the present valley of the 

 Mississippi, but it covered that valley for a distance of over 100 miles. It 

 apparently furnished an obstruction of such consequence that the question 

 of a more or less complete displacement of the main artery of drainage 

 naturally suggests itself. 



On the southwest, below Keokuk, Iowa, the ice sheet failed to cover 

 the Mississippi and the obstruction to the drainage disappears. At the south 

 the terminus was near the crest of the elevated ridge that crosses southern 

 Illinois, and no streams discharged toward the ice margin. On the south- 

 east a part of the border touches an elevated tract, whose natural course of 

 drainage appears to have been toward the ice margin. The amount of drain- 

 age obstruction in that region becomes, therefore, an important question. 



TEMPORARY DISPLACEMENT OF THE MISSISSIPPI. 



That the Illinoian ice invasion did not permanently displace the portion 

 of the Mississippi which it covered is certain, for a considerable section of 

 the present course of the stream crosses the territory covered by the Illi- 

 noian drift. Furthermorej this section of the river follows, in the main, the 

 line of a preglacial valley which appears to have been the main artery of 

 drainage for this region down to the time of the Illinoian invasion, except 

 perhaps during the Kansan invasion. The rapids between Montrose and 

 Keokuk, 12 miles in length, and the rapids and narrow portion of the 

 Mississippi Valley between the mouth of the Wapsipinnicon and Muscatine, 

 40 miles in length, are the only departures made by the river from the line 

 of broad preglacial valleys. The latter displacement is not such as to 

 coincide with the border of the Illinoian drift, and was apparently not 

 determined by this invasion. There remains, then, but the stretch of 12 

 miles at the lower rapids, in which the present stream has opened a new 

 course, that even approximately coincides with the Illinoian drift border. 



