714 ME. R. H. RASTALL ON THE GEOLOGY OF [Nov. I9H, 



black, and in thin sections a greenish-blue or bluish-grey. The 

 pleochroism is fairly strong, considering the paleness of the colour : 

 the usual tints being greenish-blue for rays vibrating nearly parallel 

 to cOOl, and very pale grey or colourless for rays vibrating normal 

 to this. The crystals are optically positive, and the absorption 

 scheme may be expressed as follows : a=fo >C. The extinction is 

 oblique, cj\z = about 9°. A peculiar form of twinning is con- 

 spicuous, so that the usual prismatic sections appear to be divided 

 into four portions, of which opposite diagonal pairs extinguish 

 simultaneously. The refractive index is fairly high and the bi- 

 refringence rather weak. All these characters agree with those of 

 ottrelite. 



This rock may be denned in general terms as an ottrelite-schist ; 

 and it is highly probable that, in this case at any rate, ottrelite 

 is a product of dynamic metamorphism, since there is no evidence 

 of any contact-action in this neighbourhood. The rock shows a 

 strong resemblance to the typical rocks of Ottre in the Ardennes. 



YTI. The Robertson District. 



In general structure this district closely resembles the neighbour- 

 hood of Worcester, but here a slight element of variety is introduced 

 by the occurrence of a large mass of granite in the Malmesbury 

 Series. This mass of granite forms a group of hills, of which the 

 highest rises to an elevation of just over 3000 feet, a height con- 

 siderably greater than that attained by the sedimentary rocks of 

 that series. As we shall see later, the presence of this mass of hard 

 rock has produced a considerable deviation in the general direction 

 of the Worcester-Swellendam Fault, and this gives rise to a question 

 of great interest : namely, what is the exact nature of the relations 

 between the dislocation and the granite mass ? 



In this district, just as near Worcester, we can divide the country 

 topographically into three regions : (1) the great escarpment of the 

 Table Mountain Sandstone, here rising to nearly 6000 feet, and 

 forming the range of the Langebergen ; (2) the narrow strip of 

 Malmesbury rocks forming foot-hills to the mountain-range ; (3) the 

 region south of the great fault, which here forms a district of some- 

 what more varied topography than farther west. It comprises a 

 hilly region about Roode Berg and Gorees Hoogte, composed of 

 highly folded beds of the Cape System ; another hilly region to the 

 west of this, extending from Lang Vley to Nuy, mostly composed of 

 Dwyka Conglomerate and Ecca Shales ; and a level or gently 

 undulating district east and south-east of Robertson, which seems 

 to consist for the most part of Enon Conglomerate. 



The highly-folded Cape rocks on this side of the fault present 

 some remarkable features : they are bent into a series of rather 

 steep anticlines and synclines, but the most striking feature is that 

 the axes of these folds are almost exactly at right angles to the 

 direction of the great fault, which is parallel to the strike of the Cape 

 rocks in the Langebergen. From the point of view of the physio- 



