High-level Gravels of the Cape. 45 



extends out to the Agulhas Bank was almost certainly at one time 

 dry land, which has now been covered by the sea, and formed a 

 base-levelled plain at some remote period. The level of a plain need 

 not be all reduced to sea-level in order to produce base-levelling, as 

 a stoppage low down in a river- course, whether from its confluence 

 with the ocean or from a bar in its course, reacts very strongly on 

 the denudation of its upper course ; it is as if the river when rapidly 

 flowing was to a certain extent rigid. Good instances of this are 

 afforded by dams across the dry rivers of the Karroo : the water is 

 stopped by the dam-wall, and this reacts all along its course up ; 

 floods occur far beyond what one would at first sight think possible 

 for the effect of the dam-wall to reach. 



The rocks cut by this bevelling process include all the known rock 

 formations in the west of the Colony, except the Eecent ; Uitenhage 

 Beds are bevelled equally with the Malmesbury clay-slates, so that 

 the plains were subsequent to the mountain-building and to the 

 Jurassic or Lower Cretaceous rock-formations. 



These plains are only plains in the centres of their areas ; on one 

 side, or on both, the plain rises to the mountains from which the 

 streams flow (Plate X. 1). Ealph Tarr* has applied to the action of 

 rivers when they cut inclined plains like this the term of "bevelling," 

 and by an extension of the term I shall allude to the intra-mountain 

 plains as double bevels. Of such double bevels there are a number 

 of series between Ladismith and Uitenhage, but the effect of them 

 all was to produce wide-open valleys all longitudinally on about the 

 same level. On the north of this area, in the Karroo, there was a 

 single bevel with a wide peneplain extending some way towards the 

 Nieuweveld and Camdeboo, and on the seaward side there was the 

 single bevel forming the Euggens of Caledon and Swellendam, and 

 -the table-topped hills all along the coast. 



As the evidence is best in the double-bevelled region, I will begin 

 my description with that. There are two rivers running westwards 

 towards Oudtshoorn, the Kammanassie and the Oliphant's Eivers. 

 The former has had to excavate its bed through the loose sands and 

 •conglomerates of the Uitenhage Series (Fig. 1), while the latter has 

 had to work its way through solid Bokkeveld slates ; hence the 

 Oliphant's Eiver has stolen a march on the Kammanassie, and has 

 not only been able to clear its bed and lay down wide alluvial flats, 

 but has captured the head- waters of its rival, so that the Kamma- 

 nassie Eiver no longer reaches the water-parting between the Gouritz 

 and Gamtoos river-systems. The smaller river, however, is much 

 more interesting, as the stream and its side tributaries has simply 

 * " The Peneplain," American Geologist, xxi., 1898. 



