538 BR. W. A. CUNNINGTON ON THE 



relationship is sometimes recognisable even between species. The 

 genera found in the African lakes which best exemplify this 

 relationship, are Lcibeo, Discognathus, and Barbits (of the group 

 allied to Ba7'bus hynni). Mastacemhelus itself is the only genus 

 of the Mastacembelidfe which occurs in Africa, but it is also well 

 represented in India and the Malay Archipelago. This case is 

 specially interesting, since the genus was at one time considered 

 quite chai-acteristic of the Indian region, yet at the present day 

 an even larger number of forms is known from the continent of 

 Africa. 



Summary. — The lakes of Central Africa contain a large and 

 interesting assemblage of fishes. The bigger lakes as a rule possess 

 more forms than the smaller, but while Nyasa and Victoria 

 IS^yanza each possess an extensive iish fauna, Tanganyika can 

 show a far larger number of types. In Nyasa and Lake Victoria 

 a considerable percentage of these types are endemic, but this is 

 more strikingly the case in Tanganyika, where indeed not merely 

 endemic species, but a very exceptional number of endemic genera 

 are found. Of the 12 families which are represented, interest 

 centres in particular on the Oichlidge, which are very promi- 

 nent in all the lakes except Albert Nyanza. A large proportion 

 of the Oichlid species are endemic in the lakes where they occur, 

 the extreme case being that of Tanganyika, which, with 84 

 endemic species and 21 endemic genera, possesses the richest 

 Cichlid fauna in the world. There are reasons for believing that 

 these remarkable endemic Cichlidte show signs of specialisation. 

 An explanation of the unique fish fauna of Tanganyika is 

 probably to be found in a period of complete isolation of that 

 lake : it is possible that isolation for a lesser period has produced 

 an eflect in the cases of Victoria Nyanza and Nyasa. The fishes 

 of the families Cyprinidte and Mastacembelidas in particular 

 exhibit indications of Afro-Indian alfinities such as are displayed 

 by other groups of organisms. 



POLYZOA. 



The representatives of this group, as far as Lake Tanganyika 

 is concerned, are of particular interest. This is due in the first 

 place to the fact that while examples of the Phylactola^mata were 

 quite to be expected, the latter are accompanied in Tanganyika 

 by two species of Gymnola^mata, the members of which are, with 

 few exceptions marine. It is also true that the gymnolsematous 

 form which was first discovered by Moore (137, p. 295) has a re- 

 markable resemblance to the marine genus Arachnid'mm. With 

 the exception of this type, which Moore named ^.j-ac/ivioicZea *, 

 the species from Tanganyika were all obtained for the first time 

 by the author during the Third Tanganyika Expedition. 



* I follow AniiaiKlale (7, p. 198) and Harmer (99, p. 50) in adopting AracJi- 

 noidea as the correct spelling of this generic name, which Moore originally wrote 

 Arachnoidia. 



