356 
BRIDGING THE AGES OF ICE 
may be styled a suhmedial moraine; (2) the attenuation of the 
ice. brought the fine debris, undoubtedly disseminated throughout 
the lower portion of the ice, to the surface in unusual quantities, 
thereby faciliting superficial melting along the ridges, and thus 
determined the course of a supra-glacial stream; (3) the rapid 
melting of the ice along such lines so reduced its thickness and 
diminished its pressure upon the subjacent surface as to divert 
thither all sub-glacial water, which accordingly formed sub¬ 
glacial streams coincident and finally confluent with the supra- 
glacial rivers; These streams formed canyons in the ice, and 
when eroded through, either formed basins, or extended their 
corrasion into the subjacent deposits, according to the slope of the 
surface. In either case, when the canyons were long, the streams 
so deeply corraded their beds before the bounding walls of ice 
disappeared as to permanently retain the water-ways in the ridges 
over which they were first defined; while, when the canyons were 
short, the streams left the ridges as soon as they reached the 
margin of the ice. 
“Passing now to the region shown in the accompanying map of 
Des Moines we find to the eastward a typical as in which the 
ice-canyon was too short to define a water-way, and about the 
confluence of the rivers a plateau which formerly existed as an 
ice-bound basin, and through which the rivers corraded their 
valleys before the final disappearance of the ice. In such a basin 
the loess was deposited, just as was all of that of eastern Iowa; 
the coldness of the waters and the low temperature of the air be¬ 
ing attested by the depauperate shells found imbedded in it. 
Here, however, a re-advance of the glacier occurred before the 
ice was melted from the plains east of Capitol Hill and west of 
Walnut Creek, which disturbed, contorted and broke up the 
superficial portions of the loess, and mingled its materials with 
the clays and bowlders of a super-imposed sheet of drift; the re¬ 
advance being too slight to completely remove the loess even 
from exposed localities. 
''Apropos to the re-advance of the ice-sheet here suggested is 
Upham’s discovery that Des Moines lies approximately in the 
course of the southernmost lobe of the great terminal moraine.^® 
10 Ninth Ann. Kept. Minnesota Geol. and Nat. Hist. Surv., p. 304, plate VI, 1880. 
