284 Report S.A.A. Advancement of Science. 



not sufficient data to construct a similar diagram, since the specimens 

 are few in number, and the limits of each rock-type are not stated. 



At Bultfontein and Dutoitspan the general succession in the 

 upper part, above the Dwyka conglomerate, is very similar to that 

 already described, but below this level the thicknesses seem to be 

 very different. The pre-Karroo sedimentary series appears to be 

 reduced to about 150 feet, and volcanic rocks come in at from 550 

 to 600 feet from the surface. In the diagram given by Gardner 

 Williams these rocks are labelled " melaphyre, " but they have 

 nothing in common with the rocks so-called above the sedimentary 

 series. As we should naturally expect from their position below the 

 quartzite, they clearly belong to the acid volcanic type, and they 

 show in great perfection the micropoecilitic structure which is so 

 characteristic of these rocks in the other shafts, and with these they 

 are clearly identical. 



The specimens of " basalt " (201, 100, loob, 304, 402a, 405) 

 from all the mines are very similar, in composition and general 

 structure, though some are slightly coarser in grain than others. 

 The rock is apparently holocrystalline, and consists essentially of 

 feldspar, augite, and olivine. The latter mineral is abundant, so 

 that the rock is of a distinctly basic character. 



The feldspar occurs in large idiomorphic prismatic crystals, 

 which invariably show twinning on the albite law, and generally also 

 on the Carlsbad law. The extinctions are somewhat variable, but 

 the majority show angles of 35° to 45°, so that the feldspar clearly 

 includes basic labradorite and bytownite. Zonary banding is not 

 conspicuous. 



The olivine forms rounrled, colourless grains, showing the usual 

 irregular cracks, and in some cases serpentinisation has ju.st begun, 

 but most of the grains are almost unaltered. 



Both the foregoing constituents are enclosed in typical ophitic 

 fashion by large plates of colourless augite, which show good 

 cleavages and extinction angles up to 45°. This augite is perfectlv 

 normal, and calls for no special remark, except that the ophitic 

 structure is more than usually perfect. 



The only accessories commonly visible are small flakes of a rather 

 pale brown pleochroic biotite and small crystals of iron ore, which 

 are probably ilmenite, since some of them show slightly translucent 

 spots. 



From the evidence at our disposal it is impossible to determine 

 whether the rock in question is of an intrusive or extrusive character. 

 Considering its petrographical characters alone, it appears to be more 

 closely allied to the intrusive tvpes than to the lavas. Some of the 

 coarser varieties present a striking resemblance to the intrusive olivine 

 rocks which occur in the Cai'boniferous Limestone of Derbyshire, 

 and, apart from any knowledge of their field relations, would be 

 unhesitatingly described as typical olivine dolerites, using the latter 

 term in its modern sense, as a substitute for the older term, diabase, of 

 British petro^raphers. Included in the collection are one or two 

 rather finer-grained specimens, which are of more strictly basaltic 



