GLACIAL BOUNDARY IN OHIO. 763 
the north), and rises from twenty to thirty feet above the present valley. 
It resembles in nearly every respect the post-glacial conglomerate, 
known as “Split Rock”, at the mouth of Woolper Creek, about twenty- 
five miles below Cincinnati, where the glacial boundary recrosses the 
Ohio, and enters Indiana near Aurora. Whether there are granite 
pebbles in the conglomerate at Flag’s Spring I am unable to say, 
owing to the haste with wnich I was compelled to examine it. But at 
Woolper Creek, granitic pebbles in small quantity form a constituent 
element of the conglomerate. One was observed which was two feet in 
diameter. The limestone pebbles in this conglomerate are frequently 
three or four feet in diameter. As pointed out forty years ago, by Prof. 
Locke, and noticed later by Dr. Sutton (see Indiana Geological Report 
for 1878, pp. 108-113), this conglomerate at Woolper Creek is not 
confined to the immediate vicinity of the Ohio; though there it rises 
more than one hundred feet above low-water mark. The conglomerate 
is conspicuously developed on the summit of the Kentucky hills for 
three or four miles southeast, and four hundred or five hundred feet 
above the river, and here, as at Flag’s Spring, on the other side of Cin- 
cinnati, the formation marks the true glacial boundary. It would seem, 
however, that the ice nowhere extended into Kentucky more than four or 
five miles from the river. Near Burlington, in Boone county, on one of the 
tributaries to Gunpowder, which flows to the south, and whose source is 
between five hundred and six hundred feet above the river, there is a 
noticeable collection of granitic boulders marking the southern extent 
of the ice. Fifteen or twenty, from one to three feet in diameter, were 
counted in a small space. Three or four of these were composed of a 
metamorphic conglomerate containing jasper pebbles peculiar to the 
eastern shore of Lake Superior. 
Prof. Lewis supposes (see Journal of Franklin Institute, April, 
1883) that near Olean, in New York, where the ice extended for a short 
distance across the Allegheny river, a sub-glacial channel was kept 
open. It would scarcely seem possible that this was the case at Cincin- 
nati; for the trough of the Ohio is considerably wider than that of the 
Upper Allegheny, and not far from fifty miles of the Ohio Valley 
bordering Campbell, Kenton, and Boone counties, Ky., must have been 
covered by glacial ice. Probably, for a short time, the ice at Cincin- 
nati formed an obstruction to the channel; but what was the course of 
its overflow [ am not prepared to say. The obstruction must have been 
