110 Hubbard — High Level Terraces in Southeastern Ohio. 



tween the blocks, which gaped to a depth of from 3 to 6 feet 

 and wide enough, in places, to receive a man. From below, 

 the series looked extremely like wave-cut terraces, but from 

 above they looked just as much like landslide forms. A num- 

 ber of other slides in the vicinity were old enough to have 

 become wholly healed. 



In Gallia county, near Gallipolis on the walls of a little brook 

 leading into Raccoon creek, two terraces were seen. They 

 rose slightly, southward as did the strata. Careful examination 

 revealed the coincidence of one with a hard calcareous layer 

 among softer shaly layers and of the other with a soft layer 

 between two harder series of layers. Apparently there was in 

 the latter case an accumulation of detritus as if the material 

 taken out of the hillside to make the notch had been piled be- 

 low to build up a terrace, but examination showed that the 

 terrace, from below the notch, was of rock with a thin covering 

 of residual rock waste. It became apparent that a soft layer 

 among hard layers could so weather out as to leave a terrace 

 below. 



The author has seen a number of such terraces as these two 

 types in central New York, notably in and around Newark 

 Valley 15-20 miles northwest of Binghamton. In Gallia county, 

 6-8 miles from Gallipolis along the pike to Cora, one fine ter- 

 race was seen on two sides of a hill. It was horizontal on the 

 south side and synclined on the east side and folloAved the 

 same soft statum on both sides. This was throughout a slender 

 rock terrace, with no mantle accumulation at all commensurate 

 with the notch. About a mile down the Raccoon* from 

 JSTorthup on the left side of the valley, at altitudes of 570-620 

 feet U. S. G. S., were found four terraces. The upper one 

 was perfect, the second poor, and the two lower ones good. 

 The upper one now shows rock along its crest instead of waste, 

 and the steep slope back of it is also of rock. A little rill over 

 the top and crest of the second shows rock all the way beneath 

 a foot or less of waste, and similarly two rills across the front 

 of the third reveal bed-rock, while one rill over the front of 

 the lower one likewise discloses the rock in place. There has 

 been some sliding in connection with the second terrace from 

 the top. These beautiful slender forms at this point then are 

 rock terraces in shales of varying hardness. They all extend 

 along the hill side a hundred yards, and the third is more per- 

 sistent — perhaps to 250 yards. 



The valley of Raccoon creek was examined for seven miles, 

 partly above and partly below Northup, and those described 



* This valley was visited in this connection because it is thought by Tight 

 to be one of the best localities to study the terrace. U. S. G. S. Prof. Paper 

 13, p. 89. 



