Br. F. II. Hatch— Witwatersrand Beds, Transvaal. 545 



this is due to the abundant presence of chlorite. The ground-mass is 

 too minutely granular to allow of specific minerals being recognised, 

 but it reacts under polarised light, and probably consists mainly 

 of very finely comminuted quartz, together with a micaceous mineral, 

 either muscovite or sericite, in minute blades. It is possible that 

 there is also some admixture of clayey matter, kaolin or some similar 

 mineral. No evidence of lath-shaped crystals or microlites of 

 felspar, or augite granules, or indeed of any of the usual characteristics 

 of igneous rocks is foi'thcoming. The differences observable under 

 the microscope between quartzite and slate consist, firstly in the 

 size of the quartz grains, and secondly in the relative proportion 

 of quartz grains and ground-mass in the two rocks. In the quartzite, 

 the quartz grains are fairly large and the ground-mass subordinate ; 

 in the slate, on the other hand, the quartz grains are small to minute, 

 and the ground-mass predominant. The difference might be com- 

 pared to that existing between sands and slimes, when speaking 

 of tailings. Where a banded structure is observable in a hand- 

 specimen of slate, under the microscope this is seen to be due solely 

 to minute variations in texture and relative proportion of quartz 

 granules, such as might be expected to occur under conditions 

 of sedimentation from currents. It certainly is not ' flow-structure,' 

 as this term is applied to igneous rocks. In short, the rocks I have 

 examined are not igneous, but of detrital origin; or, in other words, 

 they are true sediments of a quartzitic nature, and are correctly 

 to be designated as slates on account of their minutely comminuted 

 character, their banded or zonal variation due to sedimentation, and 

 their slaty cleavage, which is of course a superimposed structure 

 due to pressure, and may or may not be coincident with the 

 bedding planes. 



The presence of abundant iron pyrites in the slates is noteworthy, 

 and may be due to the original sediments having contained car- 

 bonaceous matter, which subsequently acted as a reducing agent on 

 sulphates dissolved in the circulating underground waters, sulphates 

 being probably a common constituent of these waters.' The red 

 oxides of iron, which characterise the slates when outcropping at the 

 surface, must be products of decomposition of the pyrites. Can the 

 decomposition of the pyrites also have given rise to specular iron 

 and to magnetite, as these minerals also frequently occur in the 

 surface rocks ? 



Another interesting feature, which I have only observed in the 

 slates which occur in the neighbourhood of the Bird Eeef (Cloverfield 

 and Grootvlei), is the presence of rutile in innumerable minute 

 hair-like needles or spicules, and only clearly definable by a high 

 magnification. It is characterised by its high refractive index, 

 straight extinction under crossed nicols, and the occasional presence 

 of knee- and heart-shaped twins. The smallest needles appear as 

 opaque lines on account of their needle character coupled with their 

 high refractive index ; but the larger prisms are transparent under 



' Cf. Van Hise, " Some Priuciples coutrolliug the Deposition of Ores": Traus. 

 Amer. Inst. Miu. Eug., vol. xxx (1901), p. 94. 



DECADE IV. — VOL. X. — NO. XII. 35 



