316 D. WHITE — POTTSVILLE SERIES ALONG NEW RIVER, W. VA. 



second column refers to stages or groups in the Pottsville sections in 

 Pennsylvania, Ohio, Tennessee, Virginia, Arkansas and Alabama. Thus 

 " Ohio " indicates the Sharon coal in Ohio ; "Ark." is used for the '" coal- 

 bearing shale " of Washington county, Arkansas, the flora of which is 

 found to be largelj^ identical with and clearly belonging to the Sewanee 

 group of Tennessee, indicated by " Tenn." " Horsepen " is used for 

 present convenience, without any intention to add to geologic nomen- 

 clature, to indicate a group of coals above the Pocahontas coal in the 

 lower half of the Pottsville sections of Tug river and Great Flat Top 

 mountain. They are more or less exposed near the school-house at 

 Horsepen, near the fault-line at Smiths Store, near the mouth of War 

 creek on Dry fork, and in Clarks gap on Great Flat Top. 



A brief inspection of the second column in the accompanying tabula- 

 tion shows that nearly one-half of the varieties or forms collected from 

 this stage are characteristic or predominant in the lower middle portion 

 of Tug River section, while a third of the entire number are either of 

 too great a vertical range to possess any precise correlative value or 

 they have not been noted by me from any of the other sections yet ex- 

 amined. Considering, then, only those species, forms or phases of 

 species which have thus far been observed to be limited in their ascent 

 and to be characteristic of certain horizons or groups, it becomes at 

 once obvious that the great preponderance of the forms of this particular 

 stage is also found in and mostly characteristic of the group represented 

 at Horsepen and near Peery ville, in the Tug River basin, at Smiths Store, 

 and at Clarks gap, on Great Flat Top mountaiii."^ Several of the forms 

 are more characteristic of higher stages, such as the Sewanee group in 

 Tennessee and Arkansas and the Sharon coal of Ohio, while a few come 

 from horizons whose comparative paleontologic stage is not yet known 

 to me. 



FLORA OF THE SEWELL COAL AND ITS RANGE. 



Treating in the same way the plants obtained from the Sewell coal at 

 various localities, it appears that nearly three-fourths of the forms are 

 characteristic of the Sewanee group in Tennessee, which includes the 

 " coal-bearing shale " of Washington county, Arkansas, and to which 

 probably belongs the flora of the Sharon coal in Ohio. At all the locali- 

 ties along the river the plants f gathered from the Sewell coal are at once 



* It requires no extensive study of the floras of the coals at Horsepen and Smiths Store— both 

 localities concerning which there has been much doubt and difference of opinion on account of 

 their being more or less isolated and against the fault-line — to show their intimate relation and 

 identity with those between Jacobs creek and Peeryville, on Dry fork, or in Clarks gap. 



t It should be borne in mind that, as remarked at the beginning of this paper, I treat as forms, 

 often less than varietal in rank, those variations of species which appear, so far as the study of the 

 Pottsville floras has extended, to have been much restricted in time, and seem therefore tolerably 

 characteristic, at least locally, of fairly definite stages or groups. 



