THE TANGANYIKA PROBLEM. 57 



supposed, and in the neighbourhood of the central range 

 it in general possesses none of those characters which have 

 so often been spoken of as showing that Central Africa is, 

 geologically speaking, without a history. Granting that 

 portions of the continent to the north and east may be the 

 remains of a possibly paleozoic land-mass, even these ancient 

 terrestrial areas are by no means without interest in a 

 Geological sense. Consider for a moment the characters of 

 the Shiri highland district. It is in reality a tongue of 

 the mountainous interior which stretches from the east of 

 the Shiri river into the vast Zambesi swamps. Once 

 over the steep forest-clad hills which bound the region to 

 the south, we are in a hilly upland country, which in the 

 configuration of its masses closely resembles the moun- 

 tainous districts encountered in the interior of Sicily. The 

 hills are of granitoid material, and wherever they enclose 

 extensive valleys, these valleys are true rock valleys ; 

 and are generally filled up to a certain level with the 

 silt of old lakes, and ancient alluvium ; the strange flats 

 produced in this way between the hills presenting a very 

 curious and typical appearance (see illustration on page 59). 

 From the absolutely horizontal character of the layers of 

 mud and gravel and ancient lake deposit of which these 

 plains are composed, it is evident that they have never been 

 disturbed by local elevation or distortion ; and the mass of 

 horizontal material encountered in such a valley is, in a 

 sense, the measure of the disintegration which has gone 

 on since the hills surrounding it assumed their present 

 characters. If we could make a cutting, or sink an artesian 

 well in the centre of such a valley, we should find it to 

 be composed of older and older strata, and the age of these 

 beds might, and probably would, be proclaimed by the nature 

 of the animal remains we should encounter in the course 



