SILUEIAN ROCKS. 469 



through the shales and in others as very massive beds, frequently several 

 hundred feet in thickness. 



The lowest division consists wholly of fine clay shales, which appear 

 yellow at the surface and dark bluish gray below drainage. 



Throughout Coosa valle}^ there is no single stratum sufficiently char- 

 acteristic as to be recognizable in different exposures and thus used as a 

 datum for correlating other beds. For this reason and for others already 

 m<3ntioned the above classification is simply tentative. It cannot be 

 asserted with any confidence that the divisions represented on the map 

 are made at the same stratigraphic point for any considerable distance, 

 either along the strike or across the outcrops. 



When the rocks of the Coosa valley are compared with the Cambrian 

 rocks south of the Coosa fault but little similarity is found. Nothing 

 resembling the upper silicious division occurs in the Rome and Cedar 

 creek sections, and, on the other hand, the characteristic oolitic lime- 

 stone of these sections is wholly w^anting north of the fault. On litho- 

 logic grounds alone these rooks Avould never be correlated. The presence 

 of identical faunas, however, in the two regions aff'ords unmistakable 

 evidence of their contemporaneous deposition. These facts have an im- 

 portant bearing upon the structure of the region, and will be again 

 referred to in connection with the Coosa thrust-fault. 



Silurian Rocks. — The Knox dolomite probably represents a deposition 

 l^eriod covering the upper Cambrian and extending well into the Silu- 

 rian. The almost complete absence of fossils makes it imj)ossible to fix 

 tlie limit between the Cambrian and Silurian, and aj^parenth'' there is 

 no break in the continuity of deposition from bottom to top of the forma- 

 tion. The Knox dolomite has been described by various writers and 

 needs only brief mention in this place. It is probably between three 

 and four thousand feet in thickness, although the nature of its exposures 

 and the fact that it is usually covered with a heavy mantle of residual 

 material render measures of its thickness uncertain. 



North of the Coosa valley the Knox dolomite is followed by the 

 Chickamauga, a series of l)lue, dove-colored and purple limestones ; also 

 at a few i:)oints south of the Coosa fault the dolomite is overlain by blue 

 limestone, although in the greater part of this region a stratigraphic 

 break, which will be described more fully later, occurs at this point; 

 also south of the fault tlie purple earthy limestones are re2:)laced by a 

 great thickness of black slates. They have been placed in a sejDarate 

 formation, the Rockmart slate, although the}^ probably represent the 

 same period of deposition as the earthy limestones further north, but it 

 is impracticable to separate the latter from the purer limestones beneath. 



North of the valley the Chickamauga limestone is followed without 



