Hubbard — High Level Terraces in Southeastern Ohio. 109 



wave-cut forms, and suggests lake conditions due " to obstruc- 

 tions to the drainage beyond the limits of the basin." He 

 further states that they bear no relation to the ice dam of 

 "Wright. The author does not believe an ice dam was ever 

 effective near Cincinnati nor that any of the phenomena up the 

 valley require such a dam for explanation. 



To sum up the previous literature on the subject. There 

 have been described, slender, high-level horizontal terraces from 

 the upper Ohio river region and from southeastern Ohio. They 

 have been generally ascribed to wave work, but a few are due 

 to structure and degradation or bear perfect relation to the 

 strata. Sea invasion to an altitude of 2600 feet for those in the 

 upper Ohio region, and lake conditions for the others, have been 

 invoked to explain them. The ice dam theory has evidently 

 been considered but has been discarded as insufficient. 



During the past summer a number of examples of these neat, 

 slender, hillside terraces have been seen and carefully examined. 

 The following descriptions will suffice to show the character of 

 the forms seen. 



In Monroe county, southeast of Woodsfield about 12-13 

 miles, and northwest of Hannibal about five miles along a 

 little branch of Opossum creek near Winkler's Mill, are 

 three horizontal terraces on a rather steep slope near the sum- 

 mit of a hillside. Two little brooks run down over them and 

 these have cut notches down into the terrace tops and terrace 

 fronts. Careful examination of these cuts showed the whole 

 terrace in each case to be of bed-rock with a thin cover of 

 mantle rock. There is no possibility that these terraces are 

 wave-cut unless the waste removed to make them has been 

 entirely removed. They are due to the weathering of alternate 

 hard and soft layers of the prevalent shaly sandstones of the 

 Pennsylvanian formations. 



In Washington county back of the village of Lowell two 

 localities possessed terraces. In the first place the general 

 survey of the conditions led to the conclusion that there was 

 a single, horizontal rock-fronted terrace and a waste slope 

 below, but a close examination disclosed the fact that a landslide 

 was the cause of the terrace. A long horizontal strip of rock 

 and waste had broken loose from the firmer bed-rock and had 

 fallen a few feet. At the second place, the semblance of ter- 

 races, at a distance, was striking, but on going up among them 

 there was absolutely no question but that they were caused by 

 landslides. An acre or more slid down^ crowding into the road 

 and pushing a rail fence down. The vertical slip was probably 

 three or four feet. Earth came down in blocks or sections, 

 three larger ones and several minor ones, extending north and 

 south horizontally along the valley wall. Cracks opened be- 

 Am. Jour. Sci. — Fourth Series, Vol. XXV, No. 146. — February, 1908. 



