PLATE XXXt. 



RICE COUNTY, 1884. N. H. WINCHELL. 



A little more than the western one-third portion of this county has the same 

 morainic characters as the eastern part of Le Sueur county, and was similarly wooded, 

 especially through the western tier of towns. In the northern part of this rolling 

 belt the forest is wanting. It is not altogether certain that this rough surface is due 

 to morainic accumulations alone. It is quite possible that the line of superposition 

 of the Trenton limestone on the St. Peter sandstone was marked, as in Winona and 

 Fillmore counties, by a rock-terrace, and that that terrace was broken and was 

 accompanied by knobs and isolated mounds. The effect of this on the thinner drift 

 sheet is seen in the northeastern part of Eice county. This effect apparently extends 

 across the Cannon valley in Bridgewater and Cannon City townships. It is reason- 

 able to suppose that some part of the rolling surface in Webster, and northward in 

 Scott county, is likewise due to an originally broken rock surface. The same cause 

 operates to increase the roughness of the drift along the Straight river south from 

 Faribault. The southeastern one-fourth of the county is characterized by broad 

 undulating prairies based on the till, and has an older topography than the rest of 

 the county, which is dependent on the later action of the main glacier. 



The history of the latest glacier and its drainage in Rice county is very inter- 

 esting, equalling in that respect the glacial phenomena of Le Sueur county. In the 

 same manner a lake was formed in the Straight valley, which had a discharge to the 

 north branch of the Zumbro river. This is marked by high terrace gravels. When 

 the ice uncovered the Cannon valley the lake of the Minnesota valley was discharged 

 through the Cannon valley, forming a lower series of gravelly flats. Between these 

 two events the Straight river for a time ran over the eastern margin of the ice, 

 forming the remarkable kame^ which is traceable through portions of Bridgewater 

 and Cannon City townships for a distance of about five and a half miles. The gravel 

 in each instance, whether of the river valleys or of the kame, was derived from the 

 cotemporary drift-laden ice adjacent. The existing terraces are due to the later 

 action of the streams in cutting their present channels into the once continuous 

 gravel plains. Still there is an element of confusion introduced, in any effort to 

 assign the terraces to their cause, by the earlier rock-terraces which have also been 

 covered and obscured by the drift. One is due to the superposition of the Trenton 

 over the St. Peter, and another to the Shakopee over the Richmond sandstone. This 

 pre-glacial topography was not destroyed in Rice county, and probably not entirely 

 in any county. It manifests itself where the drift deposits were laid down thin, as 



