10 THE ALLEGED DESICCATION OF EAST AFRICA 
monsoon, to the, at that time, comparatively dense population 
in South Arabia. 
In connection with the presumed desiccation of the country 
west of Lamu and Kismayu I may mention the drying up of 
Lake Stefanie. When Count Teleki and Hohnel visited it 
in 1888 it was a definite lake of large dimensions ; when 
Donaldson Smith visited it in 1884 it was rather larger than 
in Teleki’s day ; but of recent years it has become little more 
than a big puddle holding a certain amount of water during 
the rains. I do not, however, attach great importance to this 
evidence, as it extends over too short a period of time. Never 
having visited Lake Rudolf, I cannot adduce any direct evidence 
as to whether it exhibits traces of desiccation, but one fact 
strikes one, and that is that the Teleki volcano at the south 
end of the lake, wdiieh was very active in 1888, is now quite 
extinct, and this, I am inclined to think, may be due to the 
recession of waters of the lake, as the steam generated by the 
infiltration of lake water may have afforded motive power for 
its eruptions. 
To change the scene of our inquiry, let us proceed up 
country for a little. There are undoubted evidences of the 
greater extent of Lake Naivasha within geologically recent 
times ; Prof. Gregory, too, gives evidence as to the existence 
of a great lake he names after Prof. Suess, in the Rift valley, 
south of Naivasha ; Lake Magadi is, I believe, the attenuated 
relic of a sheet of water of much greater extent. Mr. J. Parkin- 
son, in his paper in the Geographical Journal (July 1914), 
has shown that in post- Pliocene times Lake Magadi was once 
a sheet of water of much greater extent than the area covered 
by the soda deposits at the present day, and he further states 
as follows : 4 Evidence is afforded by this district, i.e. the Rift 
Valley, of a general desiccation : the periodical floods of the 
Turoka river are not adequate for the formation of such a 
gorge as that now seen, and we have in addition the 
disappearance of Lake Suess. The old alluvial fans of Lorgo- 
salich, now overgrown by vegetation, point to torrential rains 
being more frequent at an earlier period.’ 
Lake Nakuru probably once also extended much further 
to the south than it does to-day. There is little doubt that 
