236 The American Geologist. October, looi. 
course of a great river, an interglacial Niagara, which was al- 
lowed the time requisite for the erosion of a gorge about three 
and a half miles long, from the escarpment near St. David's 
to the south side of the Whirlpool, but was then interrupted by 
the accumulation of ice again deeply enveloping all this region. 
This explanation, however, seems to me inadmissible, because 
the St. David's channel expands northward in that distance 
to the width of more than a mile before it intersects the es- 
carpment. If it had been cut by a great interglacial cataract, 
its width would be nearly uniform, like the present river gorge. 
A comparatively small preglacial stream, on the contrary, 
w^orking slowly through many million years, would have the 
older part of its valley thus widened by the very long sub- 
aerial decay and retreat of its rock cliffs on each side. So great 
lateral erosion cannot be ascribed to glaciation, which was light 
upon this area of confluent ice currents from the northeast and 
northwest, with consequent deep drift deposition. 
Immediately after the melting of this southern part of the 
ice-sheet and the withdrawal of the ice-dammed lake Warren, 
the Niagara river began to erode its gorge, and it has continued 
in this work, under varying conditions, to the present time. It 
found a lower passage along the course of the gorge to Lewis- 
ton than in the course of the preglacial channel, deeply drift- 
covered, between the Whirlpool and St. David's. From my re- 
newed examination of these areas this year, with the aid of the 
contoured map distributed by the U. S. Geo'logical Survey at 
the Pan-American Exposition, of the latest studies by Gilbert,* 
published on the reverse side of the map, and the valuable 
"Guide to the Geology and Paleontology of Niagara Falls and 
Vicinity," prepared by Prof. A. W. Grabau, summarizing the 
conclusions of all preceding geologists, for the many visitors 
who come to this Exposition and to Niagara during the present 
year, I have to add to these and to my own former studies an- 
other factor in the Niagara history, namely, that the erosion of 
the gorge below the Whirlpool had been partly accomplished by 
a small preglacial stream which flowed along nearly the entire 
length of that earliest part of the gorge, after draining at its 
head, farther east, probably nearly the same area as the present 
Fish creek. Joining the St. David's channel at the Whirlpool. 
♦Reprinted in the Am. Gbologist, vol. xxvii, pp. 375-377, June, 1901. 
