54 ORKilX OF RAXD P.ANKETS. 



continent, and secondly after their being compacted into the 

 Httoral deposits of the Bokkeveld ; hence the grains would be on 

 the whole finer than in the Table Mountain sandstone. The 

 higher iron content of the Witteberg rocks as contrasted with 

 the Table Mountain sandstone is, I think, largely a secondary 

 efifect. The iron is frequently arranged in nodules and concre- 

 tions and along special layers in the rocks ; occasionally these 

 nodules are so rich that the earthy varieties are worked for red 

 and yellow ochre : these facts show clearly that the iron in 

 solution was actively circulating in the rock, and it is obvious 

 that since a finer-grained sandstone will have a larger pore-space 

 than a coarser-gramed one, so more iron would circulate in the 

 finer than in the coarser sandstone — in other words, the Witte- 

 berg would be more ferruginous than the Table Mountain sand- 

 stone. However the conditions of deposition may have rendered 

 the Witteberg beds originally more ferruginous, owing to the 

 fixing of iron in ironstone gravels on the surface of plains, such 

 as we see going on all over South Africa at the present day. 

 The loose littoral deposits of the Bokkeveld would rise from the 

 sea as a more or less featureless plain on which drainage was 

 imperfect. The roots of the Lepidodendra and other olants 

 which we find as fossils in the Witteberg rocks would yield 

 organic acids which would dissolve the iron from the rocks, 

 and the organisms in the soil would precipitate this iron as 

 moor-bed-stone beneath ; some such concentration of the iron 

 on the land-surface is quite probable, and on the final cutting 

 away of the plateau these beds of hard ironstone gravel would 

 help to enrich the sediments forming ofif-shore. 



In the Transvaal System there is a similar three-fold system 

 consisting of the Black Reef below, made up of quartzites and 

 conglomerates ; then a dolomite series ; and the Pretoria series 

 on top, consisting of quartzites and shales, often highly fer- 

 ruginous. The same sequence in the nature of the deposits 

 signifies similar methods of deposition, the dolomite taking the 

 place of the deep-water Bokkeveld shales ; but the parallelism 

 otherwise is very striking: the quartzite bands in the Bokkeveld 

 series nearer the original shore are frequently copied in the 

 dolomite by bands of ferruginous quartzite appearing towards 

 the top of the series. The Transvaal System is represented on 

 the east in the escarpment facing the low country, and then 

 right through the breadth of South Africa to German Namaqua- 

 land ; much of this country is untraversed, so that it is impossible 

 yet to use the teaching of the Neo-xA.fric beds for the recon- 

 struction of the continent from which these Pal-Afric beds were 

 derived. It is significant that they form a broad band, longer, 

 it is true, than the Cape formation is known to extend, but of 

 a similar nature, and that further north, although there are 

 Waterberg and Karroo beds, as at Lake Tanganyika, the area 

 shows continental conditions during the deposition of the Trans- 

 vaal Formation. 



