Vol. 70.] MR. F. W. PENNY ON THE YREDEFORT GRANITE. 331 



against a further mass of diabase. Xorth of the Parijs road the 

 quartzite (Wl) is again absent, and the granite abuts against the 

 actinolite-magnetite slate as before, a tongue of it penetrating 

 into the sedimentary rocks and associated basic intrusions for a 

 considerable distance. 



So far, the granite has shown a remarkably sharp and regular 

 margin towards the various rocks with which it has come into 

 contact. From the map, one may presume that the diabase with- 

 stood the effect of the granite intrusion more successfully than 

 the sedimentary rocks ; that it has, in fact, formed some sort of 

 a protection for them. Even so, the adjoining sedimentary beds 

 are seen to have been removed to a considerable extent by the 

 granitic intrusion. The evidence seems to point to this having 

 been brought about by a process of stoping-away of the sedi- 

 mentary rocks and their sinking to lower levels, rather than to 

 direct absorption and assimilation of them by the granitic magma, 

 since the composition and structure of the granite thus far has 

 appeared to be uniform right up to its margin. 



However, in this tongue of granite which penetrates into the 

 sedimentary rocks and associated diabase for some distance, there 

 is ample evidence of action and reaction between the acid and the 

 basic magmas. At one spot the granite becomes a hornblende- 

 granophyre, at another practically an aplite with a considerable 

 amount of plagioclase, losing its ferromagnesian constituents and 

 its characteristic large felspar-phenociysts. In surface-specimens 

 the felspar assumes a brilliant red colour, suggesting that the pre- 

 dominant felspar of the rock is orthoclase ; but no determination 

 could be made in the microsections. owing; to the extent to which 

 kaolinization had proceeded. Thin streaks of the acid magma are 

 found veining the basic rock, which is here practically amphibolite. 

 That the granite partly re-fused this amphibolite is shown by the 

 presence of numerous dark patches of hybrid rocks embedded in 

 the former, varying in size from quite small inclusions to masses 

 several yards in diameter. Two such particularly large masses are 

 indicated on the map. They are completely recrystallized rocks, 

 consisting of small cloudy hornblende-crystals embedded in a very 

 fine-grained ground-mass, which appears to have a felspathic base 

 crowded with minute crystals of a decomposition -product like 

 zoisite. A considerable amount of acid material has been intro- 

 duced into this rock, as is proved by the presence of comparatively 

 large fragments of pink orthoclase intergrown with quartz, 

 evidently derived from the adjacent granite, corroded and pene- 

 trated by stringers of the hornblendic ground-mass (PI. XLVI, 

 fig. 3). 



Farther north, the margin of the granite is defined by part of a 

 really large intrusion of diabase. Usually, in the Witwatersrand 

 Beds of this area, the basic dykes have intruded themselves along 

 planes of weakness or junctions of slate and quartzite, and often 

 preserve a nearly constant thickness and horizon for many mile-. 

 This intrusion, however, has burst through the sedimentary rocks, 



