UNCONSOLIDATED FORMATIONS. 471 



the gravel is found to contain a large proportion of chert, in some cases 

 as much as 50 per cent. Next to the chert, schistose quartzite is the 

 most abundant constituent, and after that vein quartz, which makes 

 about 10 per cent. The finer portions of the gravel contain many small 

 fragments of shale and angular or partly rounded fragments of chert, 

 showiilg clearly that the material has been subjected to only a moderate 

 amount of stream or beach action. This gravel is overlain by from 5 to 

 15 feet of silt, colored deep red below and yellow^ or gray at the surface. 

 The line between the gravel and silt is sometimes well defined, but more 

 often the passage is gradual, and occasional pebbles occur several inches 

 or even feet above the base of the silt ] also lenses of coarse gravel are 

 sometimes seen in the silt. 



The high-level deposits form a cajj upon certain isolated hills, which ex- 

 tend from Center, Alabama, northeastward near the middle of the Coosa 

 valley. The composition and arrangement of their materials are almost 

 precisely the same as in the low-level deposits. The gravel is perhaps a 

 little coarser and the overlying silt more deej^ly colored, but otherwise 

 the deposits are indistinguishable. The intermediate slopes between the 

 upper and lower deposits are quite free from gravel, except the small 

 quantity which has washed down from above, and, so far as can be de- 

 termined, they are not 2)ortions of a single deposit originally spread 

 mantlewise over the entire surface and subsequently removed from the 

 slopes. On the contrary, the high-level deposit appears to be the older, 

 consisting of the remnants of a sheet once continuous over the whole 

 valley and almost entirely removed, together with 140 feet of the under- 

 lying shales before the deposition of the low-level deposits. 



No opportunity has yet been afforded of tracing these unconsolidated 

 formations to regions in which their age might be determined, and no 

 attempt is here made at their correlation, but it seems probable that 

 further study will show them to be the inland representatives of either 

 the Columbia or Lafayette formations or both. 



Overlaps. — In studying a region so extensively faulted as that between 

 Indian and Weisner mountains, there is a strong tendency to attribute 

 all unconformities to faulting, and especially is it difificult to discrimi- 

 nate l)etween the broad horizontal thrusts and deposition overlaps. 

 Sufficiently careful mapping, however, renders it possible to determine 

 in most cases to which class an unconformity belongs, even in presence 

 of a heavy residual mantle and deep decay of the rocks. 



At least two well-marked erosion overlaps occur in this region. The 

 first is at the top of the Knox dolomite. In the vicinity of Cedartown 

 and immediately southeast of the area covered by the map the contact 

 of the Knox dolomite and Chickamauga limestone is everywhere obscured 



