2 BULLETIN OF THE LABORATORIES 



southward, he finds Paint creek again at C, and follows it thence 

 through a broad valley to the Scioto bottoms. Coming back to C 

 and ascending Paint creek he observes that the hills on either side 

 contract in a V shape toward the mouth of Ralston's run, which runs 

 through a level bottom about 500 feet wide with steep hills on either 

 side. From here up to a ravine putting in from the west, there is a 

 strip of bottom land on one side of Paint creek, nowhere wider than 

 400 feet ; and from this ravine up to the point E the hills ascend 

 from the edge of the water which flows on solid rock all the way. 

 At E he finds the creek in its proper channel at the foot of the hill, 

 under the elms and sycamores, just as he had thought when looking 

 from the pike. 



The order of events that gave rise to these conditions is appar- 

 ently about as follows : 



It is plain that the glacier reached, as a mass, to the old valley 

 of Paint creek and that it did not ascend the hills on the southern 

 slope or even reach to them anywhere below the point E, except 

 at the points B and D. About two miles below Bainbridge, the 

 drift is piled half-way to the tops of the hills to the south, and the 

 valley along here must, for a time, have been entirely closed by ice. 

 There is no doubt that it thus followed the valley nearly or quite to 

 its head, leaving the deposits above Bainbridge, probably forming 

 the Beech Flats, and filling up all the valleys when it passed out 

 upon the limestone table-land beyond the rugged hills of shale and 

 sandstone ; thus deflecting toward the Ohio all the waters which in 

 this region had flowed into Paint creek. But, as above stated, this 

 is still to be worked out. At the point B where the creek formerly 

 discharged into the Scioto the drift is fully 100 feet higher than the 

 highest river terrace ; the distance between the hills, measured on 

 the drift, is a little less than one mile. The flat-topped hills at this 

 place are about 100 feet higher than the drift. This denotes a sufti- 

 cient thickness of ice to dam Paint creek and form in its basin a 

 lake which, fed by the natural drainage and the floods from the 

 melting ice would rapidly rise until it found a new outlet. 



The Scioto having a deep pre-glacial channel, it is very probable 

 that a lobe closed up the mouth of the creek at B some time before 

 the main body of ice surmounted the hill and filled its bed. Between 

 Slate Mills and the point C there was evidently a low depression 

 formed by two ravines, one opening north into Paint valley, the 



