398 W. M. DAVIS 



seashore, and that its new-made rivers should not dissect it, and 

 that there should be no drifted sands and loess sheets on adjoining 

 areas, and no signs of submergence on neighboring coasts. An 

 untrenched plain of erosion in such an attitude would be properly 

 interpreted as the result of normal processes, long and successfully 

 acting with respect to normal baselevel. There would therefore 

 appear to be no serious danger of confusing an actual peneplain of 

 normal origin, still standing close to baselevel, with a depressed 

 plain of desert origin. For the reasons above given I am not 

 disposed to follow Passarge in the suggestion that the old land mass 

 of Guiana may be an Inselberglandschajt in the process of destruction. 

 He cites it as a flat, gently undulating surface of gneiss, above which 

 rise knobs and mountains of granite; the divides are so low that 

 one may pass in canoe between the headwaters of the Orinoco and 

 Amazon systems (&, p. 194). Until further details are given, it 

 would seem appropriate to regard this region, like the interior of 

 Brazil to which Lapparent refers (a, p. 148), as an example of a 

 normal peneplain, not raised so as to be attacked by its rivers. 



In the same way a high-standing plain of erosion in a desert 

 region might be possibly explained as an evenly uplifted peneplain 

 whose climate had in some way been changed from humid to arid, 

 whose deep weathered soils had been removed and replaced by thin 

 sheets of stony, sandy, or saline waste, and whose residual reliefs 

 had been modified to the point of producing shallow basins. But 

 in this case there should be some indications of recent uplift around 

 the margin of the area, either in the form of uplifted marine forma- 

 tions whose deposition was contemporaneous with the erosion of 

 the peneplain, or in the form of fault-escarpments separating the 

 uplifted from the non-uplifted areas. Moreover, it is extremely 

 unlikely that the uplift of an extensive peneplain could place it in 

 so level a position that it should not suffer dissection even by desert 

 agencies; hence a high-standing desert plain is best accounted for 

 by supposing that it has been leveled in the position that it now 

 occupies. According to Passarge, there are no sufficient indications 

 of elevation associated with the South African desert plains, and 

 their explanation as the product result of long- continued desert 

 erosion in a still-standing region would therefore seem to be assured. 



