536 I>R. W. A. CUNNINGTO^^ ON THE 



14 are known also from the Congo system, a.nd that of these 

 about one-third are of very wide distribution, by no means 

 <3haracteristic of the Congo. 



A few lines may now be devoted to a general survey of the 

 fishes of the other big lakes, for the pnrpose of pointing out any 

 characteristic features and affording a gi'ouud of comparison with 

 Tanganyika. It has already been seen that Victoria Nyanza, 

 while containing representatives of nearly as many families as 

 Tanganyika, contains a much smaller number of species. Very 

 few endemic genera are found in the lake, but a considerable 

 number of endemic species occur, especially among the Cichlidse, 

 It may also be noted that Lake Victoria shows a larger number 

 ■of Mormyrids (including two endemic types) than does Tan- 

 ganj'ika. Again, as regards the Cyprinidae, this lake is licher in 

 forms, witli 11 endemic species mostly belongiug to the genus 

 Barbus, while the Anabantidte is represented here alone. 

 Speaking generally it may be said, that although now one of 

 the sources of the Nile, only a small proportion of its fish are 

 found also in that river, so that there are good I'easons for sup- 

 posing that Victoria Nyanza was also isolated for a certain 

 period, and only included in the Nile basin in times, geologically 

 speaking, recent. Meanwhile it is clear that the modifications 

 of form which exist are not comparable in importance to those 

 occurring in Tanganyika, but go to prove that the lake has a 

 peculiar fish fauna, not closely related to that of either Nile or 

 ■Congo. 



Nyasa,, having a lesser area, shows something of that decrea,se 

 in number of types which is related to the smaller size of the lake. 

 Nevertheless it still contains more C^yprinidte and Mormyridse 

 than Tanganyika, and it is relatively lich in Cichlid fishes. Of 

 the latter there are indeed five endemic genera, yet out of a totsal 

 of 38 species, 27 belong to the two genera Tilapia and Para- 

 tilajna, which are of very wide distribution. Thus, altliough 

 Nyasa possesses a fish fauna which is in part peculiar and 

 characteristic, it has a far less striking assemblage of forms 

 than Tanganyika, or even than Victoria Nyanza. While it is 

 probable that a period of isolation favoured the production of 

 species peculiar to Nyasa, it is nevertheless clear that its fishes 

 belong essentially to the Zambezi river system, and do not differ 

 from those of that river to the same extent as do the fish of 

 Tanganyika, and to a less degree of Victoria Nj'anza, from those 

 respectively of the Congo and the Nile. 



Speaking of the fishes, it may therefore be said that each of the 

 three largest lakes contains a group of forms peculiar to and 

 characteristic of itself. These forms differ from the fish of the 

 rivers which drain the lakes, but not to the same degree in each 

 instance. Nyasa exhibits a number of types which are not fou.nd 

 in the Zambezi, and Victoria Nyanza a larger and more diver- 

 gent series of forms which are not found in the Nile ; but in 

 these cases the difierences are for the most part only specific. 



