CONTRASTING VALLEYS 35 



channel below the other. The higher lies above the limestone ledge and 

 is bounded by clay-shale and glacial morainic deposits, into which it has 

 the appearance of having been cut by a large stream. The other is half 

 as wide as the higher one and lies below the limestone ledge and in the 

 Saint Peter sandstone, which appears in its bounding walls. This condi- 

 tion of double valley extends from above Fort Snelling to the center of 

 Saint Paul city. Eelated to the lower channel are terraces of limestone 

 shingle within the valley at west Saint Paul, and potholes and loosened 

 limestone blocks along the left wall of the channel, such as a gradually 

 receding cataract would produce in making the channel ; also an island of 

 Saint Peter sandstone stands' midway in the valley of the Minnesota 

 river, southwest quarter of section 33, Mendota, marking the place where 

 such a receding cataract should have died out upon encountering the pre- 

 Glacial channel of the Minnesota river. These phenomena are readily 

 explained by the hypothesis that a cataract once receded from Saint Paul 

 to and above Fort Snelling, even though they may not constitute indis- 

 putable proof of the same. 



Further evidence appears from comparison of this double valley with 

 that of other channels which are known to have been cut by recession of 

 falls or rapids. Dr U. S. Grant, in studying the abandoned gorge at 

 Minnehaha, emphasizes "the fact that the river did not cut its gorge of 

 the same width as the channel in which it was flowing before wearing 

 through the limestone."* 



Whether the view which I present of the cutting of this double valley 

 is accepted or not, it will be agreed by all who may study this part of the 

 valley that it is swept clear of glacial drift and owes its present form and 

 size to the Glacial river, a stream vastly greater than that which now 

 meanders within its rocky walls. The deeper, narrower channel now 

 serves as a valley for the Mississippi and Minnesota, with accompanying 

 floodplain, swamps, and lakes. It will be conceded, too, I hope, that the 

 river did not necessarily recede in volume as it was first drawn from the 

 wide u])per channel into the deeper, and therefore narrower, one. 



According to the view which I have taken from the evidence already 

 presented, the contrast in the valleys at the confluence of the Minnesota 

 and Mississippi rivers is due directly to the size of tlie streams which 

 made them. The valley from Saint Paul to and beyond Fort Snelling, 

 which may be termed the River AVarren valley, was made in a way very 



'Upham : Op. clt., p. 81. 

 ' Grant : Op. clt., p. 4. 



