Stevenson.] oltt \, 



The River Tekka< es 



The river terraces fall down stream and are covered by irregularly bed 

 ded sand, clay or gravel, containing transported fragments whicb have 

 been rounded by tbe action of running water. When followed up the 

 streams these terraces show differences among themselves in degree of 

 slope, so that each is merged successively into the next higher, until that, 

 which at the mouth of tire stream is the river "bottom," becomes the only 

 terrace and is lost at last in one of the lower horizontal benches. 



For the most part the terraces occur at the same elevation on both sides 

 of the stream, being divided by the channel-way just as the present " bot- 

 tom" is divided. Sometimes a single terrace and occasionally the whole 

 series of terraces is wanting on one side of the stream. In such a case it is 

 clear enough that erosion was confined to one side so as to remove all 

 traces of the terrace or to prevent the formation of the terraces, just as is 

 seen in the present channel- way, the " bottom" being present often on only 

 one side while stratified rocks reach to the water's edge on the opposite 

 side. 



As stated in my reporl on the Greene and "Washington district of Penn- 

 sylvania (1875), these terraces are merely shelves in the rock, on which a 

 thin coat of detritus rests. Mr. G. K. Gilbert, in his Memoir on the 

 Geology of the Henry Mountains, describes the occurrence of similar ter- 

 races in those mountains.* 



These terraces then do not fall in the same category with those described 

 by Dr. Hitchcock in his "Illustrations of Surface Geology," for those had 

 been eroded from valleys previously filled by gravels. No evidence has 

 been found suggesting that any valleys of that sort exist in the region 

 under consideration, while there is good reason for believing that the val- 

 leys lined by the river terraces, described in this paper, were not in exist- 

 ence to be filled with gravels. 



The terraces below Pittsburgh on the Ohio river are covered by a deposit 

 consisting largely of northern drift brought down by the Alleghany and 

 Beaver rivers. No such material is found along the Monongahela and other 

 rivers south from the Ohio, as they flow altogether beyond the southern 

 limits of the drift. But their age is as clearly shown as is that of the ter- 

 races on the Ohio. The fifth terrace, at New Geneva, on the Monongahe- 

 la, has a layer known as the "Swamp Clay." which contains much half- 

 rotted wood. With the wood are berries like those of the black haw and 

 in the same clay a well-preserved acorn cupule was found. In the same 

 vicinity, as well as near Morgantown in West Virginia, the third terrace 

 shows many Unio shells in an advanced stage of decay, while at Belver- 

 non, on the Monongahela, much wood is found on this terrace. That the 

 third terrace is older than the second and that this is older than the river 



- This work has not been published at the time of writing, but Mr. < Hlberl das 

 very kindly given me a sel of plate files for use during the preparation of this 

 paper. It contains an elaborate discussion of the whole subject of landsculp- 

 ture, embracing the results obtained by Mr. Gilbert during his long study of the 

 Colorado Plateau. 



