546 Dr. F. H. Hatch — Witwatersrand Beds, Trammal. 



a sufficiently high power, and tlien show a redrlish-yellow colour. 

 It will be interesting, and of considerable importance as a dis- 

 tinguishing character, should this rutile prove to be confined to the 

 slates that occur above the Main Eeef. There is a remarkable 

 petrological resemblance between these slates and the phyllite of 

 the Ardennes.^ According to Renard, who has made a close study 

 of the latter rocks, they are characterised by the presence of quartz, 

 sericite, chlorite, rutile, and micaceous hematite, and these are just 

 the minerals that characterise the Witwatersrand slates. 



Although, as I have shown above, the slates are not of igneous 

 origin, 1 would like to point out that rocks of true igneous origin also 

 occur in sheets or 'flows,' interbedded with the Witwatersrand Beds. 

 For instance, boreholes that have been put down in the extreme 

 Eastern Rand on the farms Geduld, Welgedacht, and Grootvlei, 

 have all intersected a sheet of amygdaloidal diabase varying in 

 thickness from 100 to 300 feet, and situated some little distance 

 above the Modderfontein Series. This, however, is quite different in 

 character from the slates. Although very fine-grained or aphanitic 

 in the hand-specimen, it can at once be distinguished under the 

 microscope by the presence of lath-shaped felspar crystals. These 

 are turbid tlirough kaolinisation, and what was probably augite is 

 now only represented by patches of chlorite. The rock therefore 

 shows signs of considerable decomposition, which is remarkable when 

 we consider that the specimens examined come from a great depth 

 (2,600 feet in the Grootvlei borehole). It is, however, clearly 

 recognisable as a diabase. Its chief characteristic in the hand- 

 specimen is the presence of almond-shaped infilled vesicles of calcite, 

 chlorite, and other secondary minerals. That this rock should be so 

 decomposed at such a considerable depth below the surface bears 

 witness to the power of circulating underground waters in efi'ecting 

 mineralogical changes. With regard to the main contention of 

 Mr. Curtis's paper, viz., that the auriferous contents of the banket 

 beds were introduced by the agency of underground waters sub- 

 sequently to the deposition of the conglomerate beds, this has always 

 been my view; ^ but it is scarcely correct to designate this as 'lateral 

 secretion,' as Mi'. Curtis does. Sandberger introduced this terra in 

 1877 to denote the process by which the metallic contents of mineral 

 veins had been derived from their enclosing ' country' or wall rocks. ^ 

 It is true that in America the term has been extended to include 

 derivation from subjacent rocks, and in this sense no doubt 

 Mr. Curtis meant to use it. 



An important fact to be borne in mind when discussing the origin 

 of the gold in the banket is the close association of the gold with 

 sulphide of iron in the form of pyrites. It seems to me that if it 

 can be shown that the pyrites in the conglomerates is of secondary 

 origin, it must also be admitted that at least the gold which is 



1 Geol. Mag., 1883, pp. 322-324 (abstract). 

 - Quart. Journ. Geol. Soc, vol. liv (1898), p. 81. 



3 Sandberger, " Zur Tbeorie cler Bildiuig der Erzgiiuge Berg. u. Hiitteum." : 

 Zeitung, xxxvi, Nos. 44-45. 



