»9i2.] STEVENSON— THE FORMATION OF COAL BEDS. 435 



The area of Dunkard deposits seems to have been much less 

 than that of the Wheeling, but erosion has stripped the borders 

 and one can make few positive statements respecting the condi- 

 tions. The full section, as it now exists, is found only at rare 

 places in the long narrow strip within western Pennsylvania and 

 West Virginia. There appears to be the same tendency to north- 

 ward shortening of the section ; there is evidence in Pennsylvania 

 that there were local foldings, that, as in the earlier periods, there 

 were not merely general movements leading to decrease of the area 

 of deposition but also others due to strains varying in direction 

 at different times. 



Evidence of elevation and subsidence is found in buried valleys, 

 marking the courses of subaerial streams. These are so numerous 

 that one need select only a few instances. 



That Alleghania and the Ohio basin were dry land at the close 

 of the Mississippian, is clear from the evidence of erosion as well 

 as of corrosion. Hyde^* has summarized the observations of his 

 predecessors and has added the results of his own studies. The 

 base of the Coal Measures along the outcrop belt in southern Ohio 

 rests directly on the Logan formations, which for the most part are 

 Lower Mississippian. These were subjected to great erosion, which 

 left a relief of 200 to 300 feet. Hyde followed this surface of 

 unconformity across Ohio from the central line of the state south- 

 ward to the Ohio River, finding frequent variations of 100 feet in 

 the elevation. The Coal Measures sandstones are often let down 

 into the Logan, but ordinarily the non-conformity is gentle. Here 

 and there one sees old valleys which can be traced for considerable 

 distances. The erosion was mostly post-Maxville. 



Soon after the elevation of Alleghania and the Ohio basin, a 

 stream began to cut for itself a valley, which, before the end of 

 the New River stage, extended from the Canadian highlands south- 

 wardly across Ohio, Kentucky and Tennessee to an outlet at the 



"J. E. Hyde, "Notes on the Absence of a Soil Bed at the Base of the 

 Pennsylvanian of Southern Ohio," Amer. Journ. Sci., Vol. XXXI., 191 1, pp. 

 557-560. 



