PLANATION SURFACES IN THE KARROO 399 



arc in the 100-mile Buff els river from Laingsburg to the sea, and it can 

 hardly be doubted that so small a change of slope would be easily pro- 

 ducible by climatic control without any change whatever in continental 

 attitude. On the other hand, it is entirely possible, as far as purely theo- 

 retical considerations are considered, that the dissection of the terrace 

 here described may be due wholly to revival of river erosion by crustal 

 movement, independent of climatic change. When fuller details are at 

 hand regarding the distribution of similar gravel-covered terraces in the 

 valleys of the Cape Colony ranges it may be possible to choose between 

 these alternative hypotheses. It will then be important to examine the 

 distribution of terraces with respect to the direction of the rivers that 

 they accompany, for under the hypothesis of climatic change all the rivers- 

 of the region will terrace their valleys, irrespective of their direction, 

 the depth of terracing being dependent on the amount of climatic change, 

 on the distance of the terrace from the river mouth (measured along the 

 river), and on the time since the change of climate occurred. On the 

 hypothesis of regional warping or tilting, terraces will be much better 

 developed along rivers whose courses are in the direction of tilting than 

 along those whose courses are in the opposite direction. 



It may be mentioned that some members of our excursion party seemed 

 to regard the Buffels Eiver terrace as possibly a result of the difficulty 

 encountered by the river in deepening its notches through the mountains 

 farther down valley without any aid from climatic change or from re- 

 vival by tilting — that is, the terrace was regarded as a possible product 

 of a continuous cycle of erosion, uninterrupted by land movement and 

 unaffected by climatic change. Such an explanation is untenable in the 

 case of a river which is cutting a notch in an anticlinal ridge; for here 

 the difficulty of cutting increases with increased depth of cutting below 

 the anticlinal crest, while the valley eroded beneath the terrace indicates 

 a relatively sudden increase in the efficiency of erosion. Such an in- 

 creased efficiency might be associated with a river which had been cutting 

 a notch through a synclinal mountain, for the broadened valley floor 

 eroded in weaker strata upstream from the notch might be trenched when 

 the river had cut down through the harder synclinal strata into weaker 

 underlying strata. The Buffels Kiver terrace is evidently associated with 

 notches cut in anticlinal ridges, and hence can not be explained merely 

 by the varying relations of the river to the hard rocks in the notches. 



THE DISSECTION OF GRADED MOUNTAIN SLOPES 



On the eastern side of the open anticlinal valley between the Draken- 

 stein range (the southern part of the Olifants Eiver mountains) on the 



