1912.] STEVENSON— THE FORMATION OF COAL BEDS. 465 



Whether or not the sandstone existed in the northern area, where 

 no trace remains, cannot be determined. Erosion has removed the 

 whole section in the space where the sandstone should be present ; 

 but it seems probable that the sandstone was not wholly continuous, 

 as an outlying exposure in JeiTerson county of Ohio shows only 

 shale. Aside from this gap at the northwest, the sandstone is prac- 

 tically continuous around the whole Monongahela area ; it is often 

 coarse, sometimes shaly, at others massive, is seldom pebbly, except 

 at the southwest, where it has conglomerate layers with pebbles up 

 to an inch in diameter ; occasionally it is replaced with sandy, even 

 argillaceous shale. It thins away rapidly toward the center of the 

 area ; well records in the deeper part of West Virginia frequently 

 note sandy shale or sandstone, just above the Pittsburgh coal bed, 

 but, in view of the disappearance of the deposit at the border of the 

 area, these occurrences cannot be taken as its equivalent. 



The material came from all sides of the basin, as is clear from 

 the distribution, but on the western and southwestern outcrop the 

 rock is coarse, at times even conglomerate, whereas the earlier sand- 

 stones are fine grained. If the conglomeratic sandstones at top of 

 the New River were exposed to erosion at this time, one would have 

 no difficulty in determining the source of the material. That those 

 sandstones were exposed appears the more probable when one con- 

 siders that the area of deposition .had become greatly restricted. 

 The coarse rocks of the Logan in southern Ohio, long exposed, may 

 have yielded some of the material, but Hyde's description of those 

 deposits makes clear that they were not the source of the larger 

 pebbles. The character of the material along the eastern outcrop 

 suggests that it had come from a distance and had been rehandled 

 many times. 



The higher sandstones of the Monongahela in Pennsylvania are 

 mostly of small areal extent and are present for the most part on 

 the borders of the Monongahela area; but in West \"irginia they are 

 thicker and more extended. Yet there as in Pennsylvania they 

 decrease toward the center of the area and are replaced with finer 

 deposits. 



The Waynesburg sandstone, separated by a thin deposit of shale 



