532 



DR. W. A. CUNNINGTON ON THE 



The family Cyprinidae is better represented than the others, 

 hitherto considered, but this is mainly due to the number of 

 species of the very widely distributed genuKS JJarbus, of which 

 about 250 forms have been ah^eady described from the fresh- 

 waters of Africa. Tiiere are types of this genus peculiar to each 

 of the lakes under review with the exception of Lake Albert, in 

 which, strange to say, no representative has yet been found. 

 The genus Varicorhinus, found in Asia and various parts of 

 Africa, occurs in the lakes only in Tangan3aka, where there 

 are two characteristic species. It is noteworthy of the genus 

 Engraulicypris that each of the three big lakes appeai-s to contain 

 a form peculiar to itself. 



The family Siluridfe cannot be expected to afford evidei:ce of 

 much value concerning geographical distribution, since many 

 of its members can remain for prolonged periods out of water 

 and even progress for some distance over land. There are there- 

 fore few points of significance to which attention need be drawn, 

 but it may be pointed out that even in this group, Tanganyika 

 exhibits two endemic genera. I'"urther, it is curious to note that 

 the genus Chryaichthys, while common in the Nile and Tropical 

 Africa, occurs m Tanganyika, alone, in wdiich six of the seven 

 species named are likewise endemic. 



Only a, few types of the family Cyprinodontid{Te are represented 

 in the list, but among them is Lamjyrichthys, a genus peculiar to 

 Tanganyika. — The genera Lates and Luciolates are the only 

 members of the 8erra.iiida3 to be found in the big lakes, and of 

 the species which occur, only one — tlie well-known Lates niloticus 

 — appears outside the confines of Tanganyika. Thus in this 

 group also thei'e is a genus (Luciolates) endemic in that lake. 



While the families hitherto considered show various noteworthy 

 features of distribution, it is the Cichlidge which far surpasses 

 them all, both in interest and in the number of forms represented 

 in the African lakes. Of all the species enumerated in the table 

 of distribution, more than 70 per cent, are Cichli<ls, or to put it 

 another way, there are considerably more species of the family 

 Cichlidpe than of all the other fainilies taken together. In the 

 case of the genera, exactly half are Cichlids, there being 36 out 

 of a total of 72. These are remarkable facts, but they are not 

 merely an expression of the common occurrence of this group of 

 fishes on the continent, as an examination of the list will show . 

 It is quite clear that there are groups of forms which are peculiar 

 to these big lakes, arul that the three biggest contain more 

 than the smaller ones, while a culminating point is reached in 

 Tanganyika with a succession of endemic genera. 



Lake Albert is alone in containing no endemic species, but the 

 numbers gi'ow from 2 and 3 endemic forms in Lakes Kivu 

 and Edw^ard respectively, to 31 in Nyasa, 40 in Victoria Nyanza, 

 and no less than 84 in Tanganyika. Nor is this all, for in the 

 number of endemic Cichlid genera the remarkable nature of 

 the fauna of Tanganyika is particularly exemplified. In the 



