82 
THE TANGANYIKA PROBLEM. 
the lake, the more desiccated, inland steppes became covered 
with euphorbia trees, clumps of bush and groups of park- 
trees, the valley assuming more and more the characters of 
a park, until typical African-park scenery was reached some 
miles before the village of Buta-gata. All this country pre- 
sented, in fact, in a striking manner, the peculiar zoned- 
character in its vegetation, which, as* I have shown else- 
where, marks the recession of water from land under a 
tropical sun. We were, indeed, passing over older and 
older ground as we moved away from the lake, and as 
we gradually rose from ioo to 200 feet above its 
surface, the plains composed of modern alluvium were 
found in places to be intersected by the deep cuttings 
formed by the storm torrents, which periodically llow from 
the eastern scarps, and in some of these cuttings, which 
were at times from 80 to 90 feet deep, older stratified 
materials were exposed. These strata were composed of 
brown and yellow sandstones, with a slight dip to the south, 
and embedded in them there were apparent numerous 
fossilised remains. Examination of these showed that they 
consisted chiefly of the fragments of shells, but I obtained 
a number of fairly preserved specimens, and these un- 
questionably belonged to several of the molluscs which now 
inhabit Tanganyika ( Neothaunia , Nassopsis and Para- 
mellania). 
From these observations, it becomes clear that, at some 
time, Tanganyika extended many miles north of its present 
limit as a deep lake, and that besides its general water-level 
having fallen, the floor of the valley, north of Tanganyika, 
has also actually been raised, in the same manner that 
portions of the east coast have been raised, between Ujiji 
and Usambora. Beyond this point, i.e ., north of Buta-gata, 
* See Chapter VI. 
