247 
SOME EVIDENCE INDICATING THE IMPORTANCE OF FROST 
ACTION IN WIDENING VALLEYS. 
GLENN CULBERTSON. 
At several points along the lower reaches of Clifty gorge in Jefferson 
county, Clifty Creek in its first attempts at valley widening has eroded the 
sides of the valley to such an extent as to cause more or less extensive 
landslide areas. In the case of several of these, the naked soil and rock are 
exposed for a horizontal distance of several hundred feet, and vertically 
several scores of feet. About one-half mile from the Madison and Hanover 
road across the mouth of the valley, there is such an erosion area on the 
slope facing west. Near the south end of this exposure a large mass of the 
original rock extends out into the valley, as may be seen by referring to 
the figure. 
The formation into which the stream is cutting at this point is that of the 
Lorraine shales of the Ordovician. At this horizon the Lorraine is com- 
posed .of alternating layers of limestone from one to three inches in thickness 
and comparatively soft blue shale comprising at least 75 per cent of the 
whole. At the point ‘‘A” in the figure, the stream at every flood stage under- 
cuts the steep slope and removes all debris consisting of broken lime stone 
and shale resulting from the weathering of the exposure. The projection 
at “B” is composed of the same material as that at “‘A’’, and is from its 
position especially exposed to the action of the stream at flood stage and 
thruout its course in the two mile gorge the stream gradient is high. 
Why has this projection “B” been able to endure while a similar material 
both upstream and down has been carried away? In the judgment of the 
writer, it is due to the fact that a spring emerges from the hillside above, 
the waters of which flow over the projecting rocks at “B’’, and not only pre- 
vents freezing, but keeps them uniformly at the same temperature and 
moisture. The part “B” which stands out some twelve or fourteen feet be- 
yond the part “A” and at least six feet beyond the part “C”’ which is com- 
pletely protected from the force of the flood waters by the projection, is 
kept well covered by the spring waters at all times. The difference in the 
rate of valley widening here which is quite marked, is probably due to 
two, if not three processes. First and foremost is frost action. The 
alternate freezing and thawing which occurs many times during the winter, 
loosens annually large quantities of material from the steep slope at ‘“A” 
which is entirely unprotected by vegetation. The Spring floods periodically 
remove this. Some material of this sort, however, is loosened and accumu- 
lated during the summer and autumn. Hence another cause of weathering 
must be sought, that does not take place at “B”. Alternate expansion and 
contraction due to change in temperature is responsible for part of this 
work, but in the writers judgment, alternate wetting and drying, which 
takes place many times during the summer, may be equally as efficacious. 
At two other points in Jefferson County, a similar protective influence of 
the waters of springs has been noticed. In these cases the formations 
protected were resistant Devonian limestones, and the mass of the pro- 
jecting area was not nearly so great as in the case of the less resistant 
