50 V. W. SARDESOA^ SAINT AKTHONY FALLS 



of the falls at the temporary head of the gorge. The stages indicated by 

 terraces and fallscarps show the height of Saint Anthony falls and rapids 

 together to have been as f oUotts : At Fort Snelling, no falls ; at 3 miles 

 the descent was 40 feet; at about 4 miles, 75 feet; at 6 miles, 55 feet;^° 

 at 8 miles, 50 feet. The time of the reduction of the river Warren may 

 therefore be correlated with the falls when at Lake Street bridge, or about 

 half way from the month to the present head of the gorge. 



Eegarding the rate of recession, it can not be said to have been uniform. 

 For the 5 miles below Fort Snelling I have no data from which to calcu- 

 late directl}^, but from above that point the first mile of the gorge of the 

 Mississippi was evidently taken up by the river very quickly, by reason of 

 a pre-Glacial valley there. Thence for a mile — that is, to Minnehaha 

 creek — the evidence of the waterfall is seen. It evidently reached in 

 depth only to the base of the limestones, where the stream, rebounding 

 from the limestone blocks which had fallen, could surge back and under- 

 mine the scarp by washing away the friable sandstone. The limestones 

 are jointed at intervals of 10 to 40 feet, and this facilitated the caving off 

 of blocks. That both limestones were thrown down together is evident, 

 since the upper is mingled with the lower in the debris now found in the 

 terraces. The lower limestone predominates, however, and this is evi- 

 dently because the upper is not so thick and, moreover, had been reduced 

 by erosion in the rapids, which extended 50 or 100 feet beyond the fall- 

 Bcarp. 



At the Lake Street, or 4-mile, stage the rate of recession appears to 

 have been slower than at the 2-mile stage. An upper rapids extended 

 back over 300 feet, indicating time for erosion of the upper limestone. 

 The caving ofE from the front of the fallscarp was evidently slow. The 

 blocks, which cover the terrace below the fall, are almost exclusively from 

 the lower limestone, testifying that the upper limestone was all eroded 

 preceding the falling of blocks. The recession of the fall involved evi- 

 dently first the complete removal of the close-jointed upper lixaestone; 

 second, the caving down of the lower limestone in blocks, because of 

 water entering its joints and undermining the sandstone; and next the 

 great blocks forming a lower rapids were slowly undermined, sinking to 

 the bed of the river. This manner of recession at the Lake Street stage 

 appears to have continued to the Tenth Avenue bridge, or 8-mile stage, 

 where the falls entered the jSTicollet Island rapids. 



In its last stage the falls entered the JSTicollet Island rapids and made 

 doubtless a greatly accelerated retreat, because the upper limestone was 



" sixty feet high at Riverside park Includes 4 feet above the top of the second Ume- 

 Btone. 



