370 The American Geologist. June, issy 
carried upward by glacial currents from lowlands. During 
the twelve years since my examination of Bird's hill I have 
often thought over its evidence, which seems to me incontro- 
vertible, that much drift existed there within the ice-sheet at 
greater altitudes than 500 feet above the very level country 
on which it rested, and from which the drift had been eroded 
and boine upward into the ice.* 
No other locality known to me is so clear and certain in 
ts testimony of abundant englacial drift transportation upon 
plain legions; although the high and solitary kame named 
Devil's Heart hill, in North Dakota,! would apparently sur- 
pass Bird's hill in this evidence, except for the very significant 
relation of the latter to the highest stage of the glacial lake 
Agassiz. It may be added also that to my mind almost equally 
ccnvaicing proof of much englacial drift in the same region 
is supplied by its conspicuous marginal moraines, which attain 
altitudes of 200 to 359 feet in the Leaf hills of western central 
Minnesota, and half as great altitudes on their best developed 
tracts in North Dakota. 
Proceeding southward to the upper part of the Mississippi 
basin, we have near New Ulm and Rush City, respectively in 
southern and eastern Minnesota, sheets of till at the surface, 
15 to 23 feet thick, v^-hich I think to have been englacial and 
deposited from the ice when it melted, lying on thick and ex- 
tensive beds of modified drift.t 
The Rush City plain, occupying the northeast part of Chi- 
sago county, Minnesota, drained by the St. Croix river to the 
Mississippi, is especially instructive. The surface of this tract 
is yellow till, 10 to 20 feet deep, brought by an ice current 
from the northwest. Under this till, on an area measuring 
five miles or more in both length and width, is a greater but 
undetermined thickness of red sand and gravel, supplied from 
the melting of ice which had earlier flowed thus far from the 
region of lake Superior on the northeast. Unlike derivation 
of these drift deposits is known by their material. The early 
*Geol. and Nat. Hist. Survev of Canada, Annual Report, new series, 
vol. iv, for 1888-89 Part E, pp. 38-42, with a section. U. S. Geol. Sur- 
vey, Monograph xxv, 1896, pp. 183-188, with map and section. 
^Bulletin, G. S. A., vol. v, 1894, pp. 76-79. U. S. Geol. Survey, Mon. 
xxv, p. 157. 
JGeol. and Nat. Hist. Survev of Minnesota, Final Report, \oi. i, 1884, 
pp. 581, 582; vol. ii, 1888, pp. 413-417- 
