52 
THE TANGANYIKA PROBLEM. 
throughout a great part of equatorial Africa, and the extra- 
ordinary complexity of which we are only just beginning to 
understand. The two greater longitudinal series, which 
were first known, have been likened by Suess to the crack- 
like “ rills ” that traverse the surface of the moon ; but in 
their present complex aspect they appear to me to be far 
more closely comparable to the so-called canal systems 
which can be seen to form a dark lacework over the 
equatorial portion of the planet Mars. I do not intend 
to push this comparison any further, but it might be 
fruitful to bear in mind that these equatorial, terres- 
trial, and martian networks, are certainly in some respects 
analogous. 
In considering these remarkable folds still further, 
we find, as one of their most striking features, that 
their floors are more or less invariably strewn with 
active and extinct volcanic cones. We may say, in fact, 
that along any of these great valleys we have only 
to go far enough in order to encounter past or present 
evidence of intense volcanic activity and terrestrial imper- 
manence. Thus the short series to the east opposite 
Mombassa contains Kilima Njaro and its associated cones. 
The Beringo series contains many extinct volcanoes, such 
as Longonot, and further to the north the active cones 
which have been found in the neighbourhood of Lake 
Rudolf, while finally, all along the floor of the vast 
western series of valleys in relation to Tanganyika, we 
have equal evidence of past and present volcanic activities. 
Often as in the case of the Mfumbiro chain on a gigantic 
scale. 
This relation in Africa of the volcanoes of the continent 
to the floors of the eurycolpic folds is not, however, diffi- 
cult to understand ; we have seen that these valleys have 
