392 W. M. DAVIS 



general presence of graded wadies and of drainage slopes in desert 

 regions, confirm this statement. 



As soon as a shallow wind-blown hollow is formed, that part of 

 the integrated drainage system which leads to the hollow will supply 

 waste to it whenever rain falls there; the finer waste will be blown 

 away, the coarser waste will accumulate, and thus the tendency of 

 the winds to overdeepen local hollows will be spontaneously and 

 effectively counteracted. As incipient hollows are formed in advan- 

 cing old age, and the maturely integrated drainage system disinte- 

 grates into many small and variable systems, each system will check 

 the deepening of a hollow by wind-action; hence no deep hollow can 

 be formed anywhere, so long as occasional rain falls. 



It is conceivable that, in some special cases, there might be a 

 peculiar balance of the various factors involved which would result 

 in the development of wind-carved hills and hollows, even if the 

 region were not absolutely rainless. The occurrence of permeable 

 sandstones might favor such a result, because the rain falling on them 

 would sink into the ground instead of running off of it, while fine 

 grains weathered from the sandstone would be disposed of by the 

 winds. But for the present no desert sandstone region with hills 

 and hollows is known while such regions with hills and valleys 

 are common. Hence it must be inferred that even in sandstone 

 deserts the occasional rains suffice to wash the surface and 

 to prevent the formation of anything more than very shallow 

 depressions. 



As the drainage becomes more and more disintegrated, and the 

 surface of the plain is slowly lowered, rock masses that most effectually 

 resist dry weathering will remain as monadnocks— /w^/forge, as 

 Bornhardt and Passarge call them in South Africa. At the same 

 time, the waste will be washed away from the gathering grounds 

 of maturity and scattered in the shallow hollows that are formed 

 here and there by the winds as old age approaches. The removal 

 of the basin deposits by the winds may be delayed where the hygro- 

 scopic action of saline clays keeps the surface firm; but wherever 

 the integrated centripetal slopes are locally reversed by the hollowing 

 action of the wind, some of the central deposits will be washed back 

 again and exposed to renewed search for fine material by the wind, 



