512 DR. W. A. CUNNINGTON ON THE 



1913, and brought back valuable information concerning the 

 depth of the lake and the salinity of its water, as well as repre- 

 sentatives of its fauna. A large collection of fishes was found to 

 contain a number of new species. l:>rief diagnoses of which have 

 been published (32, 33, 37), but beyond this, so far as I am aware, 

 no further particulars have been forthcoming, owing to the out- 

 break of the European war and the subsequent lamented death 

 of Stappers. 



2. GENERAL INTEODUCTION. 



This paper aims, in particular, at giving an account of the 

 animal forms now known to occur in Lake Tanganyika, but while 

 special attention has been directed to enumerating these, types 

 from other lakes have been noticed and discussed. The account 

 is therefore a comparative survey, and shows clearly the re- 

 markable nature of the fauna of Tanganyika to which reference 

 has frequently been made. 



It must, of course, be understood that the systematic i^eview 

 and the deductions drawn therefrom, are of necessity based upon 

 present knowledge, and ma}^ need modification in the light of 

 future discoveries. 



In order not to exceed reasonable limits, the scope of the 

 paper has been restricted, and it does not deal with ail the big 

 lakes of Africa. At the same time, the largest and most impor- 

 tant lakes have been considered, and only those of lesser interest 

 are outside the field of enquiry. Whenever a tabular form has 

 been adopted, columns for the lakes are uniformly arranged in 

 the following order: — Tanganyika, Victoria Nyauza, jSTyasa, 

 Albert ISTyanza, Edward Nyanza, and Kivu. 



It is not possible to consider hei-e at any length the geography 

 and physical features of the lakes concerned, but certain out- 

 standing particulars may be mentioned. 



With the exception of Victoria Nyanza, all the lakes under 

 review occupy portions of the Great Kift Valley, as it has been 

 called, and are, therefore, of a very characteristic type. Without 

 entering into details, it may be briefly explained that Suess, in 

 discussing the geology of Central Afiica, put forward the view 

 that the long deep valleys which run through the country in an 

 approximatel}^ north and south direction have been caused by 

 trough-faulting on an extended scale. Each " Graben "^-as he 

 terms these long narrow valleys, with their steep, almost preci-. 

 pitous sides — he believes to have been produced by the faulting 

 and sinking of a long narrow tract of country, and it is but 

 natural that a number of these characteristic depressions are 

 filled with water. Thus Tanganyika and the other typical 

 *' Graben " lakes ai'C situated in long trough-like vallej'^s, bounded 

 for the most part by steeply rising sides, which may reach in 

 places a height of two or three thousand feet above the level of 

 the water. 



