19J2.] STEVENSON— THE FORMATION OF COAL BEDS. 437 



it deepened rapidly toward the south. It was dug in soft rocks of 

 the Mississippian, shale, limestone or comparatively fine grained 

 sandstone ; but it headed eventually in the Archaean rocks of Canada, 

 while an important branch rose in the Mississippian area of Michi- 

 gan, covered with a cherty limestone. A gradual though probably 

 not continuous subsidence is shown by the distribution of New- 

 River beds, which terminate one after another in progressive over- 

 lap northward until beyond the Ohio River one finds only the 

 newest members of the formation. In like manner, overlap is 

 distinct on the eastern or Alleghania side of the valley; at the north, 

 the Beaver rests on Mississippian beyond termination of the New 

 River ; Hayes and others saw that, in White, Bledsoe and Cumber- 

 land 'counties of Tennessee, the lower members of the New River 

 thin away in succession on the west slope of Alleghania until at last 

 the Bonair sandstone, midway in the formation, crosses the ridge 

 where it rests on Mississippian. A similar condition exists on the 

 eastern side of the ridge, facing the old trough of sedimentation. 



The valley was filled eventually by New River deposits. In 

 Ohio, only the latest members appear; coarse sandstone, of which 

 the lower portion is dense, hard and white, containing quartz pebbles 

 from the Canadian Archaean and pebbles of fossiliferous chert from 

 Michigan. At many places in northern Ohio, this is a mass of 

 pebbles with hardly enough sand to bind them, while mingled with 

 them at times are irregular pieces of shale — the whole giving unmis- 

 takable proof of river action. The chert pebbles continue almost to 

 the Ohio River at the south, more than 400 miles from the source, 

 the measurement being made in direct line. The upper portion con- 

 tains no chert but abundance of quartz pebbles, and in Jackson and 

 Pike of Ohio, more than 300 miles from the only possible source, 

 those pebbles, occurring throughout the deposit, are often as large 

 as a hen's egg. In most of Kentucky, the equivalents of these later 

 deposits are merely coarse sandstones with layers of small pebbles, — 

 it must be remembered that there one sees only the border, not the 

 deeper part of the valley. In the southern part of the state the 

 beds consist largely of " hailstone grit," for there the exposures 

 reach beyond those at the north. Soon after passing into Tennessee, 



