134 GEOLOGY- OF THE N l-IIG IIBOUK HOOD OF STECLENBoSC 1 1 . 



There is little doubt that the second and third lobes coalesce in 

 Jonker's Hoek ; and the trend of the margins of the first and 

 second lobes (see the map) makes it certain that these, too, unite 

 under Stellenbosch Mountain. The three lobes therefore repre- 

 sent one single intrusion. 



DoLERiTE Intrusions. 



Many dolerite dykes cut the granite in some districts, 

 notably the Cape Peninsula and Somerset West. In the more 

 immediate neighbourhood of Stellenbosch I have only located 

 one. on the farm Muratie, near the north end of Simonsberg. 

 This dyke is many yards wide, but its margins are covered, so 

 the exact width could not be ascertained ; it cuts coarse porphy- 

 ritic granite. It is more interesting than dolerite dykes usually 

 are, on account of the presence in it of a quantity of xenocrysts 

 of quartz and felspar derived from the granite. The rock i> 

 light grey, of decimillimetre grain, and contains white spots up 

 to a centimetre in diameter, which ap])ear to be felspars derived 

 from the granite. Besides {>lagioclase and augite the rock con- 

 tains brown biotite and some interstitial (juartz. The felspar^ 

 are rather unfresh, and their optical characters cannot l)e very 

 satisfactorily determined. The white spots are altered felspars, 

 filled with decomposition-prodticts ; their irregular shapes point 

 to their being xenocrysts, not phenocrysts. The larger frag-' 

 ments of quartz are, without any doubt, xenocrysts; each is 

 completely surrounded by a shell of aug'ite crystals, a phenome- 

 non which has been observed in other cases of reaction between 

 a magma and its xenocrysts. Taken all together, these features 

 suggest that there has been an appreciable amount of solution of 

 granite by the doleritic luaama. 



The Malmesburv Series 



is represented in the immediate neighbourhood of Stellenbosch 

 by a long tongue of slates and argillaceous sandstones, which has 

 been caught in between the western and central lobes of the 

 granite batholith.- These rocks decompose so easily that they 

 are generally hidden under a thick overburden, and it is hard 

 to form an idea of the normal composition of the series. On 

 the other side of the western granite mass, however, good sec- 

 tions of the Malmesbury beds are exposed in railway cuttings 

 between Lynedoch and Eerste River stations, and these sections 

 may be taken to be representative of the series. Just beyond the 

 granite margin, some two miles past Lynedoch station, there 

 appear in a cutting dark brown clayey rocks, dipping verticall} 

 and striking about magnetic north. They are very rotten, but 

 show spots of chiastolite, and are clearly decomjjosed shaly rocks. 

 They are cut by some interesting dykes of aplo-pegmatite, and 

 by quartz-tourmaline and ptire quartz veins. Further away the 

 rocks are fresher, and are best described as very cleavable 

 argillaceous sandstones, spotted by metamorphism. with occa- 



