THE TANGANYIKA PROBLEM. 71 



it would appear probable that the lake must have continued 

 to extend northwards as the fold increased, and this view of 

 the matter seems to be supported by the very singular fact 

 that, as we found, in several places off the Livingstone 

 range, the bottom of the lake although the water was 270 

 fathoms deep, is composed of practically bare rock, and 

 quartz fragments.- whereas further south the whole floor of 

 Nyassa is covered with fine deep mud. or, in other words, 

 it is suggested that the origin of this portion of the lake's 

 floor is so modern that up to the present time very little 

 fine sediment has had time to collect upon it. Similar 

 evidence is afforded by the upraised lake deposits in the 

 neighbourhood of Mount Waller. These deposits are not 

 more than 70 to 100 feet thick, they do not appear to be 

 much denuded, and consequently they cannot have been 

 deposited for any great length of time. In the south, on 

 the other hand, we have in the region of Fort Johnston and 

 Pamalombi. obviously vast depressions which are filled with 

 lake deposits, and these latter, on the very lowest estimates, 

 must be many hundreds of feet thick. It would appear 

 therefore that the old fold which orioqnallv contained the 

 southern portion of Nyassa has progressively extended 

 northward, and, consequently, we encounter here for the first 

 time, a fact which we shall encounter ao-ain. namelv. that 

 different portions of the same Lake valley may be ol 

 very different ages. 



We saw that towards the north end of Nyassa the old 

 Atrican sandstones and Drummond's beds were both much 

 contorted and often sloped at a high angle out of the floor of 

 the present lake. There is no conformity between these 

 series, and there must consequently have been in this area, 

 first a period when the old African sandstones were laid 

 down in deep water, afterwards there was a rising ol the 



