34 F. W. SARDESON SAINT ANTHOKY FALLS 



the Minnesota, with that of the Mississippi below the confluence, as said, 

 has been heretofore considered as a pre-Glacial valley, reexcavated by the 

 Glacial river, the river Warren, and its successor, the Minnesota. 



It is my view that the larger valley, the Minnesota-Mississippi course, 

 from 2 miles above Fort Snelling to 6 miles below it (see figure 2), is 

 also not pre-Glacial, but is a new course cut by the glacial floods and the 

 river Warren. The pre-Glacial course may have run by a now buried 

 valley south of the present course — that is, from a few miles above Fort 

 Snelling, eastward, in the direction of Hastings, or Gray Cloud island, 

 there joining the present Mississippi's valley. The pre-Glacial river cor- 

 responding to the Mississippi may have entered the Minnesota valley 

 above the present confluence, coming by way of the present lake Minne- 

 tonka and Purgatory creek. At Saint Paul the new course of the river 

 joins another pre-Glacial valley, which it follows thence southeastward. 



The histor}' of the two contrasting valleys at the confluence of the 

 Minnesota and Mississippi rivers is therefore the same excepting in the 

 matter of magnitude and changes of the rivers which made them. The 

 rock formations are alike and the same lime rock ledge and fresh talus 

 slopes bound their sides. Only in size the valleys are contrasted, indi- 

 cating that the Minnesota valley originally carried the main stream, but 

 now the tributary. 



The glacial river which followed the present Minnesota valley and 

 thence the Mississippi was the main stream at the time of the ice-retreat. 

 As shown by terraces in the Minnesota valley"* and by the scarp above the 

 lime rock bench which lies between Fort Snelling and Saint Paul, this 

 river flowed for a time at a high" level, to or above the present 820-foot 

 contour line. At that stage it was one mile wide in its steep-sided chan- 

 nel at Saint Paul. It spread out over a large part of the present Fort 

 Snelling reservation. Its bottom was above the limestone ledge. Corre- 

 spondingly the Glacial Mississippi was about one-half mile wide, as indi- 

 cated by scarps and terraces. 



As noted by Winchell,'^ there could have been no Saint Anthony falls 

 at that high stage of the river. The flood in the Minnesota Biver 

 valley must first "recede," or rather the channel from Saint Paul to 

 Fort Snelling must be cut down below the limestone ledge, before the 

 falls could have formed at Fort Snelling. In what exact manner the 

 channel was cut down is somewhat uncertain. It may be noted, however 

 (see figure 2) that this valley is double — that there is one distinct 



° Warren Upham : Geological Survey of Minnesota. Final Report, vol. ii. 1888, p. 91. 

 " Geological Survey of Minnesota, Final Report, vol. ii, 1888, p. 338. 



