386 W. M. DAVIS — OBSERVATIONS IN SOUTH AFRICA 



low country on either side of it are of earlier origin than the gap, seems 

 to be indigenous in South Africa as it is in Pennsylvania, thus adding 

 still another feature of similarity — a geographical habit of mind — be- 

 tween the two regions. The more rational explanation of all such gaDS 

 as of the same age as the open low countiy, when measured in years, 

 though of very different stage of development when measured by the 

 physiographic scale, is slow to find popular acceptance. 



The general drainage system to which the rivers of the deep-cut water 

 gaps belong is peculiar in that its members head on the inland side of the 

 Cape Colony ranges and maintain generally transverse courses southward 

 to the ocean. The river heads are in most instances on the Veld itself, 

 60 or 80 miles north of its south-facing escarpment, and flow from the 

 highland in valleys by which the escarpment is much dissected. All this 

 is noteworthy because it involves the drainage of a relatively undisturbed 

 area — the southern part of the Veld — by rivers that cross a broad belt of 

 strongly disturbed and upfolded strata in the Cape Colony ranges. It is 

 all the more noteworthy because it constitutes still another feature of 

 systematic resemblance with the Appalachians of Pennsylvania and Vir- 

 ginia, where the Delaware, Susquehanna, and Potomac exhibit essen- 

 tially the same peculiar relations to the areas of horizontal and folded 

 rocks. 



An origin for the South African drainage system has lately been sug- 

 gested by Schwarz (&). This writer points out the remarkable direct- 

 ness of the watershed between the Orange Eiver system and the Indian 

 Ocean drainage, from Cape Town to Deiagoa bay, and then adds : 



I can not conceive of any explanation for such an arrangement except that 

 which assumes that there was a vast plain stretching over the whole land when 

 the subcontinent first rose from the water's edge, and that the central ridge 

 [the watershed] was already then formed. If we examine the central parting 

 of the waters, we find that there is no structural cause for its existence ; there 

 is no backbone of igneous rock, nor is there a chain of folded mountains to 

 account for it ; neither, again, is there a wide anticlinal arch, for the beds dip 

 in toward it. I have adduced reasons elsewhere for supposing that the water- 

 shed owes its origin to the manner in which the Karroo sediments were laid 

 down; they were accumulated on [in ?] the thickest deposits about 150 miles 

 from the old Permian shoreline, which ran northeastward to the north of the 

 watershed. When elevation began, the curvature of the basin in which they 

 were lying was reduced, and consequently the thickest deposits formed an ele- 

 vation, which at once became a water-parting. 



This original plain was about on a level with the main watershed as it exists 

 today, for I have seen evidence elsewhere of the extremely little erosion that 

 takes place on a flat water-parting, and roughly we can say it is so now at an 

 elevation of 6,000 feet above sea level. 



