CONTRIBUTIONS TO THK GEOLOGY OF BENGUELLA. 517 



fine-grained, coarse-grained, and compact glassy varieties. Some of the more massive 

 varieties are similar to the Huambo Quartzite. On the edge of the plateau south of 

 the Chimboia River are large blocks of ironstone conglomerates containing quartzite, 

 and the soils indicate that the quartzites continue for 3 miles south of the river, 

 where the gneiss reappears with a coarsely speckled variety. Quartzite occurs again 

 on the high ground between this locality and the next stream, which is about 2^ 

 miles to the south, where half a mile to the east of the track is a boss of 

 gneiss. From this point the gneiss extends for a great distance to the south. It is 

 seen in all the stream beds as well as occasional large exposures on the intervening 

 ridges. On the Warwar River the gneiss is a dark-blue variety, rich in biotite ; it 

 strikes to 78°, and is penetrated by coarse veins of biotite-granite. At the large 

 village of Embala Andulo the foliation of the gneiss is very irregular, different 

 measurements in the same exposure giving the strike as 8°, 18°, 43°, and 48°. The 

 variations are due to disturbances occasioned by numerous faults and granitic veins. 

 Most of these veins and also some faulted quartz veins trend to the north-west, while 

 the faults trend to the north-east. The gneiss here is especially quartzose, and con- 

 tains pink felspars and many pegmatite veins. 



From Embala Andulo the track descends steeply to the Unyamba River, on the 

 south side of which is a coarse-grained gneissoid granite with a very irregular 

 foliation ranging from 48° to 110°. This gneissoid granite is invaded by an 

 amphibolite dyke, which Mr Tyrrell has determined as a hornblende-hyperite 

 belono;ino; to the Charnockite Series. 



About 3 miles south of the Unyamba and half way to the village of Etunda 

 the gneiss is of a dark-blue dioritic type. Mr Tyrrell has identified it as a rock 

 belonging to a granulitic intermediate division of the charnockites. The more basic 

 character of the rocks in this neighbourhood was indicated by the red soils upon 

 the plateau. 



About a mile and a half south of the charnockite is a red sandstone, and quartzite 

 fragments become conspicuous in the soil. Quartzite is seen in situ just north of 

 Etunda and for at least 3 miles to the south of that village. Then follows a long 

 interval with no exposures. To the north of Saccanjimba the track descends into 

 the valley of one of the uppermost tributaries of the Cutato, along which are extensive 

 exposures of gneiss of which the strike is at first to 52°. A mile further south, on 

 the floor of the valley to the north of the mission station at Saccanjimba, the rock is 

 a hornblendic or diorite gneiss with irregular, obscure foliation, its direction being in 

 places to 112°. The proximity of intrusive rocks is indicated by the existence of a 

 large pegmatite dyke beside the ford. This dioritic gneiss is succeeded higher up 

 the slope leading to the abandoned buildings of the Saccanjimba mission by white 

 saccharoidal quartzite and a pink quartzite which weathers into a white sandstone. 

 Its strike is to 58° and it dips 70° south, but this may be exaggerated by slip 

 down the hillside. 



