108 The Atnerican Geologist. Marcii.isss 
had withdrawn a short distance into the watershed of Paint creek, 
which flows into the Scioto at Chillicothe, and which is strikingly 
marked by glacial terraces. The most likely explanation of this 
complicated series of facts is furnished b}" the Cincinnati dam. 
The additional evidence which I now have to present is similar 
to that afforded at Beech flats, and was discovered last December 
while exploring the lower portion of the valley of Big Beaver creek, 
Pa., under the guidance of Mr. Richard E, Hice. It will be I'e- 
membered that two or three years ago Dr. P. Max Fosha}" and 
Mr. Hice made an important discovery of glacial furrows in the 
Beaver valley three or four miles south of the limit which Prof. 
Lewis and 1 had set to glacial action in that vicinity. These fur- 
rows were on the west side of the Beaver, near the southern line 
of Lawrence county, a little above the mouth of the Conno([uenes- 
sing creek, which comes in from the east. 
On reviewing this ground with Mr. Hice, we found a pretty 
well-marked terminal moraine on the promontory just above the 
junction of the Connoquenessing with the Beaver. Below this 
point, as well as above, the rock shelf of the old l)ase- level 
spoken of was covered for some miles with overwash gravel, and, 
upon the western side especially, lay occasional bouldei's of large 
size, which had probably been distributed by water more or less 
choked with floating ice. The elevation of this rock terrace here 
above the river which occupies the narrow eroded gorge is about 
1(J0 feet, or 2!^5 feet above low water mark in the Ohio at the 
mouth of the Beaver. The significant discovery made by us was 
that, as we followed this rock shelf down stream a few miles into 
Beaver county, suddenly the overwash gravel nearly disappeared 
at Stackman's run, in Big Beaver township, the ridge which 
we had been following ending abruptly at the face of the pre- 
cipice, on the west side of the narrow trough of the river. From 
this point a mile or more down to Clark's run, which enters the 
Big Beaver at Homewood, there was but little superficial material 
on the rock}' bench. 
Clark's run comes in at right angles to Big Beaver, and has 
worn a deep gorge through the Homewood sandstone directly 
across the bench we were following. Though the stream is small, 
the gorge is ;^>00 or 400 feet in width, and is worn dowm nearly 
to the depth of the Big Beaver. The significant facts meeting 
us here were that, while the south side of this gorge of Clark's 
