1894,] A Glacial Ice Dam. 491 
At K is a rock with 25 feet vertical front, and at T a rocky 
projection 45 feet high, through which the river has also cut, 
while there is an open channel 350 feet wide between K and 
HH, obstructed only by a gravel trail 15 feet high, extending 
from K to a low rocky exposure at KK. 
The rocks S and T are also separated by a channel about 
250 feet wide, presenting vertical walls for a short distance back 
from the river, and indicated farther south by a depression in 
the drift filling. 
At FF is a high hill with a rock cliff 25 feet vertical. At 
uuthere is a low rock wall which is extended to the dam 
at F. Wells in the drift terrace south of w show a buried 
channel there 50 to 75 feet deep. 
By tracing out the curves and sigmoids indicated by these 
rocky walls, evidence is found for two distinct rock gorges, be- 
sides the one occupied by the present river, as shown by the 
heavy, dotted and broken lines respectively. 
To make these data as clear as possible, five sections drawn 
to scales are presented in plate XV. These sections are taken 
along the lines bearing the same letters on the map and inthe 
- same position. The continuous, interrupted and crossed lines 
represent the courses of the rocky gorges, while the dotted por- 
tions represent the estimated depths of the drift-filled chan- 
nels. 
The interpretation of this peculiar locality is by no means 
certain. We venture an explanation which, in the light of 
Prof. I. C. Russel’s investigations on existing glaciers, seems to 
me quite possible. There is abundant evidence for believing 
that the preglacial drainage of Licking County was to the south, 
and that a great morainic dam backed the water up until it 
broke over a col into the Muskingum basin? 
2 This col was at the point represented by C at the extreme 
= left of plates XIV and XV. On each side of this low divide 
there was a ravine cut into the Waverly Conglomerate. 
__ As the water rose over the divide and began to cut 1t down, 
_ the gorge produced by this cutting to the west of the point C 
i “Full description of evidence for these conditions will be found in Bulletin of 
tific Laboratories of Denison University, Vol. VII, Pt. 2. 
