164 Dr. F. H. Match — Geological History of South Africa. 



Transvaal, and probably included a portion of Southern Rhodesia ; 

 it may have been landlocked, for the facts that the Waterberg 

 formation has been found resting on an uneven floor, and that near 

 the base very coarse irregular conglomerates ^ are met with, seem 

 to militate against this formation being a true marine deposit. 



The Cape Colony representative of the Waterberg Sandstone,, 

 namely, the Table Mountain Sandstone, also consists of sandstone, 

 but with less conglomerate. Conglomerates, although scarce, however, 

 do occur, and Rogers has described in the Pakhuis Pass, near 

 Clanwilliam, the occurrence of ice-scratched pebbles which appear 

 to have come from a glaciated region, and to have been carried to 

 their present site by the agency of floating ice.- From an increase 

 in the degree of coarseness of the sediments towards the west, and 

 the greater frequency of conglomerates in the Piquetberg division 

 and the Olifants River Mountains, Rogers argues that their source 

 must have lain in that direction, and that the sediments were 

 accumulated at no great distance and in shallow water. In his view 

 the Table Mountain Sandstone is probably of fluviatile oi'igin."' 



The Table Mountain Sandstone is followed conformably at the 

 Cape by the Bokkeveld Beds, the character of which points to 

 a continuance of the subsiding phase. The fossil remains indicate 

 marine conditions, although the frequent occurrence of false bedding 

 in the sandstones precludes the possibility of very deep waters. 

 The fossil fauna of the Bokkeveld Beds shows that they are 

 homotaxial with the Devonian System, but whether Upper, Middle, 

 or Lower has not yet been decided. The upper beds of the Bokkeveld 

 and the succeeding Witteberg Beds indicate a change to lacustrine or 

 fluviatile conditions, the beds consisting of mudstones and sandstones 

 which, with the exception of a few plant remains, are barren of 

 organisms. It is possible, as suggested by Mr. Rogers,* that at 

 about the middle of the Bokkeveld period open communication with 

 the sea may have been again cut off, deposition subsequently taking 

 place by fluviatile agencies in inland seas. This would account for 

 the absence of marine organisms in these beds. At the Cape the 

 Witteberg Beds are followed without a break by the shales forming 

 the lowest beds of the Dwyka Series. Northwards, however, the 

 Witteberg and Bokkeveld Beds gradually thin out, and the Dwyka 

 Series finally rests unconformably on the Table Mountain Sandstone. 

 From this the deduction can be made that owing to the elevation of 

 the beds above sea-level, they became subject to denudation before 

 the commencement of the Dwyka epoch, and it is even possible that 

 still further north denudation may have commenced before the 

 Witteberg epoch, so that the latter series was never deposited in 



1 E. T. Mellor, "The "Waterberg Sandstone Formation ": Trans. Geol. See. 

 S. Air., vol. vii (1904), p. 40. Mr. Mellor instances the occurrence of boulders 

 ranging up to 8 feet in diameter. These must have been transported by streams of 

 a torrential character. 



- A. "W. Rogers, " The Glacial Conglomerate in the Table Mountain Series near 

 Clanwilliam": Trans. S. Afr. Phil. Soc, vol. xvi (1905), p. 1. 



3 A. W. Rogers : "The Geology of South Africa," p. 395. 



* " Geologv'of the Cape Colonv,"" p. 396. 



