J. J. Stevenson—Surface Geology of Pennsylvania. 249 
at which the drift was spread over northern Pennsylvania. 
Along the Monongahela and other rivers south from the Ohio, no 
such material occurs, but the deposits afford sufficient evidence 
of another kind to enable us to fix their origin within compara- 
tively recent times. At New Geneva on the Monongahela near 
the West Virginia line, the highest terrace has a thick coating, 
in which a layer known as the “swamp clay” holds much half 
rotted wood, such as is frequently seen in peat . In the 
same neighborhood the third terrace shows many Unio shells in 
an advanced stage of decay. A similar condition exists on the 
same terrace at Morgantown, farther up the river in West Vir- 
ginia. At Belvernon, on the same river, near the northern line 
of Fayette county, this terrace yields many fragments of wood. 
In this way it can be shown that the deposits on the first, third, 
fourth and fifth benches are of recent origin. 
Since these deposits are of recent origin, there would seem to 
be good reason for supposing that the valleys through which 
their streams flow are also of recent origin, at least so much of 
them as lies below the level of the highest terrace. But it has 
been suggested that these terraces are only the result of re- 
working the sides of thé valleys, which had been eroded pre- 
vious] 
the sides are gently sloping, whereas below it they become steep 
at once. Above the line of that terrace, the smaller valleys a 
very few miles of its mouth. ; 
ese river terraces are relics of river beds, which at one 
time stretched across the valleys, just as the river “ bottoms 
now do; and the valleys below the line of the highest terrace 
have been eroded since the drainage system was reésta ‘ished 
by withdrawal of the ocean below the lines of the former 
stream beds. 
Conclusions. 
The general conclusions to which I have come are :—1. That 
he erosion, to which is due the general configuration of th 
surface above the highest river terrace, began even before the 
elevation of the anticlinal axes and con Papi 
region was submerged in post-glacial time. 2. That t it orl- 
zontal benches are due to re-working of preéxisting valleys, 
