488 STEVENSON— THE FORMATION OF COAL BEDS. l^ov. i, 



position that they owe their red color to any widely-acting cause. 

 The especial localization of the deposits in the central part of the 

 area, the lateral passage into fine shales and sandstones of wholly 

 dififerent color and the many isolated occurrences of small deposits 

 seem to exclude explanations based on supposed aridity or any other 

 general condition of climate. The shale, at times, contains small 

 areas of coal in distinct beds, and occasionally one finds coal at the 

 horizon of beds which are persistent around the borders. Where 

 the mass is interrupted by other deposits, which continue to within 

 the limits of the reds, a coal bed also at times continues from the 

 border, though at the east only a few miles away it is wanting as 

 the red is continuous vertically. It is certain that occasionally the 

 conditions, favoring accumulation of coal, existed for considerable 

 periods within the Central area of reds. No matter which 

 hypothesis respecting the formation of coal beds be accepted, the 

 condition of general aridity becomes inadmissible, because the exist- 

 ence of coal beds, great or small, is proof of humid atmosphere 

 and dense vegetation not far away. One finds a great mass of 

 reds at the Pittsburgh coal horizon at less than a score of miles 

 from localities where that bed is of workable thickness ; and the 

 same statement is true respecting the Harlem and Anderson coal 

 beds. There is every reason to suppose that in a general way the 

 climate, in respect of rainfall, was very much as now ; the direction 

 of the winds was the same and there is no reason to suppose that, 

 at any time during the Pennsylvanian. a mountain chain existed on 

 the west side of the Appalachian basin. Yet alternation of wet 

 and dry conditions, as suggested by Barrell, may have been prevalent, 

 though due only indirectly to atmospheric influence. 



Topographic changes would seem to be the preferable explana- 

 tion for conditions in the Central area. The subsidence, convert- 

 ing that area into vast tidal flats, continued until, just prior to the 

 Ames limestone, the region subject to river and tidal overflow may 

 have embraced more than 20,000 square miles. It must be remem- 

 bered that in the vertical space occupied by the great reds of the 

 Conemaugh one finds the Cambridge and Ames limestones, both 

 marine. The rivers during long periods of little change had ac- 



