504 STEVENSON— THE FORMATION OF COAL BEDS, [^'ov. i, 



immediately adjacent parts of Ohio and West Virginia, in all less 

 than one fifth of the present area of the formation. The deposits 

 are especially important in Pennsylvania between the Monongahela 

 and Ohio Rivers, perhaps 2,000 square miles, but they disappear 

 rapidly in all directions. The extreme development is near Wheel- 

 ing on the Ohio, where one finds 154 feet of limestone in 190 feet 

 of measures ; but nearly the same thickness is shown near the 

 Monongahela, where, however, the section is longer, as it has sand- 

 stones and shales which are wanting at Wheeling. These deposits 

 vary from almost pure limestone to calcareous shale, and in much 

 of the area they are in layers one to 4 feet thick, separated by much 

 thinner layers of calcareous shale. Their appearance is so char- 

 acteristic that one familiar with them in any locality could hardly 

 mistake IMonongahela limestones elsewhere for those of any earlier 

 formation. Animal remains are confined to a few types. Minute 

 forms, resembling the ostracoids found in the upper limestones of 

 the Conemaugh, are abundant at many places. Teeth of Helodus 

 and a spine of Ctenacanthus inarshii have been collected from mid- 

 way in the formation within Washington county of Pennsylvania ; 

 the fourth limestone is rich in Naiadites near Uniontown, Penn- 

 sylvania, and a blue shale near Morgantown, West Virginia, con- 

 tains abundance of SoJenomya. The fish remains are the same with 

 those which abound in the marine limestones of Illinois, but these 

 marine forms are evidence of sea invasions so brief as to have no 

 significance, for they are unaccompanied by marine molluscs. All 

 the features indicate that the Monongahela limestones are, in 

 greatest part, of freshwater origin. 



The Washington formation has six limestones, confined prac- 

 tically to Greene and Washington counties of Pennsylvania, all, 

 except one, disappearing quickly in each direction. They resemble 

 those of the INIonongahela, some being covered with similar forms, 

 thought to be of freshwater types. Larger fossils are ver}- rare. 

 I. C. White obtained from a black shale in the upper portion scales 

 of Rhi:;odus and other fishes, which are most probably freshwater 

 in their relations, as the same genera occur in a cannel layer at 

 Linton, Ohio. The Greene formation has many limestones within 



