450 SIDNEY PAIGE 
to the origin of this oversteepening of topographic gradient. 
Agegrading streams flow out upon fans and fans have a cross- 
section concave upward. Such a cross-section implies that from 
time to time the stream will flow against the mountain wall and 
initiate lateral cutting. It may be argued in opposition to this 
view that it will build up here as elsewhere and prevent any per- 
sistent lateral cutting at a definite level. This objection may be 
answered by pointing out that a desert level of great expanse is a 
feature of great stability, far more permanent than the local filling 
at the side of a fan. This filling may be swept out and renewed 
many times before the general level of the desert is raised an appre- 
ciable amount by accumulation. This process then, viz., lateral 
cutting, combined with that outlined in the preceding paragraph, 
seems to account for all the conditions that need explanation. 
Both must have worked from the beginning of the cycle. 
CONCLUSIONS 
The considerations which have been outlined above suggest the 
following tentative conclusions: (a) Processes of erosion within an 
inclosed basin system in an arid climate tend ultimately to produce 
surfaces of very low relief about the borders of the gravel sheet 
which accumulate within the basin. (6) The gradual rising of the 
gravel fillimg implies an equally gradual rising of the local base 
level. (c) The surface resulting from such a shifting system tends 
ultimately to take the form of a sloping planated surface, most per- 
fect at its mountainward side and progressively more irregular 
valleyward beneath the gravel cover. (d) Interstream erosion, 
lateral cutting at edges of accumulating fans, and progressive 
burial of low-lying areas are the factors which govern the formation. 
of the rock-cut surface. (e) The abnormally steep mountain flank 
against which the rock-cut plain abuts is considered the normal 
product of the three processes mentioned above. (f) Sheet-flood 
erosion is considered a result of the rock-cut plains and not a cause 
of the plains as hypothecated by W. J. McGee. (g) The old pla- 
nated surfaces near Silver City, though now dissected because of 
readjustments of drainage due to faulting, are regarded as examples 
of the type described in the Sonoran district. 
