226 The American Geologist. April, i89i 
amount of erosion by the lake waves during postglacial time which 
very far exceeds the total erosion that was effected upon the shores 
of lake Agassiz during all its stages, the proportion between them 
being surely not less than ten to one ; and lake Michigan has a 
similarly greater amount of beach deposits, which upon a large 
area about its south end are raised by the wind in conspicuous 
dunes. This contrast indeed suggests that the duration of lake 
Agassiz, and the recession of the ice-sheet from lake Traverse to 
the lower part of the Nelson river, may have been included within 
less than one thousand years. 
Before lake Agassiz began to exist, the receding Minnesota and 
Dakota ice-lobes had each given place to a large lake on the central 
part of the area from which they withdrew. By the barrier of the 
Minnesota ice-lobe a lake having an elevation of about 1,150 feet 
above the sea was formed in southern Minnesota in the basin of 
the Blue Earth and Minnesota rivers, outflowing southward b} T way 
of Union Slough to the East Fork of the Des Moines. In its 
maximum extent this lake probably had a length of 160 miles, 
from Waseca to Big Stone lake, with a width of forty miles in 
Blue Earth and Faribault counties, attaining an area of more than 
3,000 square miles. The continued glacial recession afterward 
opened lower outlets eastward to the Cannon river, and at the 
time of the Waconia moraine had uncovered the lower part of the 
Minnesota valley, permitting the lake to be wholly drained north- 
eastward to the Mississippi.* The modified drift from the retreat- 
ing ice on the upper Minnesota basin was deposited along the 
lower half of this valley, filling it with stratified gravel, sand and 
clay, to a depth 75 to 150 feet above the present river from New 
Ulm to its mouth, which shows that at least this portion of the 
valley was excavated in the sheet of till during the interglacial 
epoch, and remained with nearly its present form through the 
later glaciation. It seems also probable that the upper part of 
the channel above New Ulm, occupied by the river Warren at the 
time of the Herman beaches, remained from such interglacial ero- 
sion, so that the first outflow from lake Agassiz was at a level 
some twenty-five feet below the general surface adjoining lakes 
Traverse and Big Stone and Brown's valley, being thus approxi- 
mately marked by the Milnor beach, t As long as streams poured 
* Geology of Minnesota, vol. i, pp, 460, 622, (142. 
f Compare with Geology of Minnesota, vol. i, pp. 479-485, describing 
