OP THE HALOLIMNIC FAUNA OF LAKE TANGANYIKA. 367 



denudation. A new era had arrived, and some faint shadow of 

 modern conditions was inaugurated. Unfortunately, there is 

 no evidence, as far as I can make out at present, of what took 

 place between the close of- the White Friable Sandstone period 

 and that of sub-recent and recent deposits. This interval 

 doubtless was, during part of the time, a period of great inland 

 waters, where the basins of the Congo, Shari and White Mle 

 inosculated and where the fauna now exisiting in Equatorial 

 Africa was to a considerable extent evolved, and the rivers 

 themselves partly marked out. But enough perhaps has been 

 said on this subject, and I must now conclude this geological 

 disquisition with a brief description of the more immediate 

 surroundings of Lake Tanganyika itself, inasmuch as a proper 

 understanding of the peculiar physical features of this lake 

 may help us to consider, if not to explain, the origin of its 

 still more remarkable molluscan fauna. 



Structure of a Graten. — Before proceeding to consider the 

 geological features uf Lake Tanganyika, I would draw the 

 attention of members to the structure of a Graben as depicted 

 by Mr. Moore, in the case of Lake Nyassa. This traverse which 

 is taken through Mount Waller towards the north end of the 

 lake, shows the relation of the Red Felspathic Grits to the 

 underlying granitoid rocks (Archeean) ; and it also exhibits the 

 system of trough-faulting which may be taken as one form of 

 the structural arrangement of a Graben. 



Geology of Lake Tanganyika. — As regards Tanganyika itself 

 the lake occupies the principal depression in the western 

 arm of the Graben-system of Equatorial Africa, running due 

 north and south for 400 miles, and the present elevation 

 of the surface of the water is stated to be 2,700 feet. 

 There are several affluents, the principal one being the Euzizi 

 at the head of the lake, whilst there is only one effluent, 

 viz., the Lukuga, which escapes through a chasm in the 

 vv-estern walls (? vide Gregory, Tlie Great Rift Valley, p. 3), 

 and ultimately joins the Congo drainage system, to which 

 at present it belongs. The discharge of the Lukuga seems to 

 be a precarious one, and it is clear that there have been times 

 when the water does not escape, in which case one would expect 

 an increase in its salinity. Great depths are reached in this 

 lake, and Mr. Moore considers that it is not all of one age, the 

 central portion between Karema and Ujiji being regarded as 

 the oldest. This circumstance is also true of Lake Nyassa, 

 where in some places the bottom is so bare of recent deposit as 

 to suggest that such portions may have been added to that lake 



