CONTRIBUTIONS TO THE GEOLOGY OF BENGUELLA. 531 



that the report may be based on Cretaceous shells. I saw no raised beaches or other 

 indications of marine action except upon the low coast plain between Lobito and 

 Benguella. The evidence of the wells near Bimbas Road (p. 504) is not in favour of 

 the marine origin of the recent deposits in the valley of the Cavoca. 



The age of the coast faults is post-Cretaceous, and it is probably Pleistocene. The 

 cliff on the south-east shore of Lobito Bay appears modern, for some of the valleys 

 which notch it are still in the condition of hanging valleys ; and as the rocks on 

 their beds are soft, these valleys must be quite young. Earthquakes frequently 

 happen along this coast, and they may be due to continued movements on these 

 recent coast faults. 



A more difficult problem is whether the great steps by which the country rises 

 over 6000 feet above sea-level to the Bihe plateau are also fault scarps. The section 

 across the Western Congo by Cornet represents the country as built of a succession 

 of plateaus with steep faces toward the west, and according to his interpretation 

 (Cornet 1897, No. 1, p. 22) the faces are escarpments and the plateaus are long dip- 

 slopes to the east. In one part of his section, however, near the " crete de Bonza 

 Manteka," the plateau cuts across the bedding and its surface is therefore due to 

 denudation. In Benguella the general succession of the plateaus to the west of the 

 coastal belt appears also to be due more to denudation than to faulting. The 

 arrangement of the north-western and south-western faces of the Oendolongo Hills 

 is, however, suggestive of their having been determined by faults. 



The edges of the main plateaus are so intensely dissected that even if the steps 

 were due to faulting their most conspicuous present features are due to denudation. 

 The successive plateaus are not sharply separated in the section showing the relief 

 of the railway line, for the railway naturally follows the gulleys and sometimes 

 adopts a sinuous course in order to increase the distance and thus lessen the grade. 

 If the railway had adopted a shorter course, rack sections would have been necessary, 

 and they were proposed at two localities where they have been avoided by lengthen- 

 ing the line. The plateau fronts are therefore not well seen from the railway, and 

 the determination of their nature would require long traverses north and south from 

 the railway. If plateaus were originally bounded by faults the original scarps have 

 been destroyed except near the coast, and in the interior the fault lines are buried to 

 the west of the steps. 



The explanation of the steps by denudation alone would be difficult. The terraces 

 might be regarded as successive pene-planes due to halts in the uplift of the land ; 

 but though that explanation would explain the surface of the plateaus, it does not 

 appear consistent with the nature of the steps. The step-like structure of this 

 country appears to require either a series of monoclinal folds or of step faults ; as 

 these two structures are tectonically equivalent, the choice between them is com- 

 paratively unimportant. I saw, however, nc evidence of monoclinals, while faults 

 are numerous. 



