30 



1'. AV. SARDESOiS' SAINT ANTHOKY FALLS 



the glaciers from the north, as calculated by Winchell, seems fully evi- 

 dent from the j^henoniena and data as he represents them to us. His 

 presentation of the subject is such that little discoverable evidence would 



Figure 1. — Map slwii-iti'j the general liclation cf Glacial Lake Agassiz and River Warren 

 to the Minnesota and Mississippi Rivers 



appear to have escaped his attention, and therefore no great modification 

 or reversal of his conclusions might be expected. 



More than 10 years ago I came by accident iipon evidence which had 

 been previously overlooked, and which led me to think that Winchell's 

 interpretation might be open to important modification, This idea was 



