1887.] il* [Stevenson. 



Byt there have been great changes in the water- waj'S. The fragment- 

 ary deposit on the line between Russell and Scott counties, carrying 

 quartz pebbles at 640 feet above Clinch river, tells the story of one great 

 change, for nothing along the upper Clinch could furnish the material for 

 this deposit, of which so little remains. The deep erosion near the North 

 fork of llolston river, from Saltville in Smyth county eastward for sixteen 

 miles, cannot be referred to any present drainage system ; its bottom is 

 more than 600 feet below the present bed of the river and the excavation 

 has been filled with a deposit of gypsum and rock salt. This was digged 

 out after tlie faulting, for the excavation crosses and re-crosses the Salt- 

 ville fault.* 



It is sufficiently clear that the courses of many of the present streams 

 are due in no small degree to the geological structure. The fans formed by 

 the Clinch and llolston with their tributaries show a co-incidence with the 

 general course of the rocks and faults which cannot be merely fortuitous. 

 Tributaries to New river in Bland and Giles counties flow irregularly with 

 the strike of the beds ; between outcrops of sandstone they follow the 

 more readily yielding rocks, so that they are often brought near to the 

 fault lines ; and many streams belonging to the other systems do the same. 

 But in the broad limestone areas, some other cause has determined 

 the direction, for not a few streams exist there whose courses appear to 

 bear no relation to the geological structure. No especial weakness now 

 exists in the immediate vicinity of the faults, for the streams flow with 

 utter indifference to them. New river crosses all of the faults. Big 

 Walker creek and the North fork of Holston flow back and forth over the 

 Saltville fault, and the latter at times wanders to a distance of two or 

 three miles, apparently without reference to the cliaracter of the rocks. 

 Clinch river coquettes in the same way with the Clinch faults and event- 

 ually deserts them to cross the limestone area in Scott county and to cut 

 the Copper creek fault at ten miles toward the south-east. The North 

 fork of Clinch crosses two faults and many of the smaller streams flow 

 directly across one or more. 



It is altogether probable that the present lines of the faults are verj'^ far, 

 in some cases at least, from the original lines. The lateral thrust in more 

 than one case must have been enough to push the upthrowu rocks to a 

 considerable distance over upon the downthrown series ; so that the areas 

 of weak or crushed or much distorted roclis lay north or northwest from 

 the lines as now observed. The crushed portions have been removed by 

 erosion and the streams have changed their channel-ways as the erosion 

 advanced. It is diflicult, therefore, to determine much respecting the 

 former drainage ways. 



The thickness of rock removed by erosion in this region, though not 

 equal to that removed from Western Colorado and adjacent parts of Utah, 

 is still sufficient to challenge respect. There is no room for doubting that 

 the Coal Measures reached at one time beyond the "Valley " to the Blue 



♦See Proc. Amer. Phil. Soc, Vol. XXII, p. 15-t et seq. 



PROC. AMER. PHILOS. SOC. XXIV. 125. W. PRINTED JUNE 7, 1.887. 



