218 G. F. WEIGHT GLACI-\L DAM IN THE AliLEGHENY RIVER 



of necessity have been delivered lower down the steep channel at Tiona, Shef- 

 field, Barnes, and along the trough of the Tionesta ; but there are no foreign 

 components in the gravels east of Clarendon. 



"From Tiona eastward the gravels are the same slabby Lx^als with rounded 

 edges that we find in the unglaciated portions of the Kinzua Valley, and every 

 observer has reported no crystallines in the Tionesta Valley except at its 

 immediate debouchment into the Allegheny." 



CONCLUSIOXS 



From these facts it is clear that (a) the deepest rock erosion at Warren 

 was preglacial; (h) the earliest glacial drainage must have been through 

 the Tionesta outlet rather than down the Allegheny Yallev to Tionesta ; 

 (c) the absence of northern drift material east of Clarendon indicates 

 that the continuance of the glacial stream through the Tionesta was tem- 

 porary; (d) before the northern drift material had proceeded far in fill- 

 ing that valley the drainage was diverted down the Allegheny, which 

 would indicate that for a short period there was an ice-dam below Warren 

 which, on giving way, permitted the diversion to the present channel; 

 and (e) it is therefore improbable that there existed at that time a col at 

 Thompsons, as has been surmised. 



It is fair, however, to say that Prof. E. H. Williams supposes that all 

 the phenomena can be best accoimted for by the supposition that there 

 was at first a col somewhere below Tionesta which impounded the water 

 and raised the level to that of the highest glacial terraces at Warren and 

 Clarendon. 



This would seem to render the theory of an ice dam below Warren un- 

 necessary, since it would provide slack water at Warren. Stoneham, and 

 Clarendon long enough to account for the limitation of northern drift 

 to the upper part on the Tionesta Valley. On this supposition the de- 

 sertion of the Tionesta outlet would be due to the fact that the col at 

 Thompsons had already been worn lower than the level of the moraine 

 between Stoneham and Clarendon. But an examination of the map will 

 show that it is difficult to see why, if the channel at Thompsons had 

 always been open, the swollen glacial current coming down the main 

 channel of the Allegheny should not have pushed past the Tionesta open- 

 ing and carried its material along the present course of the channel below 

 Warren, instead of down Glade Eun toward the Tionesta, In view of 

 the evidence that the ice did actually cross the Allegheny below Warren, 

 the existence of the ice dam supposed seems to give the easiest and, in- 

 deed, the only explanation of all the facts connected with the singular 

 distribution of the gravel deposits described in the text. 



