Dorothy Nyingi
Ichthyology Department, National Museums of Kenya.

The species which belong to the genus Barbus constitute a very diverse group in many respects. They do not seem to be monophyletic and they occupy a wide range of different habitats. The African barbs all belong to the genus Barbus usually divided into large and small Barbus. B. oxyrhynchus is grouped with the large Barbus (Berrebi, 1998). This group has great variations in plastic characters among populations; so much so that the species B. oxyrhynchus itself has been described in 15 different species. Banister (19??) grouped these together. This study aims to verify the classification of these morphotypes. The African Labeo is a separate monophyletic unit from the Asian Labeo. This genus is defined on its uniquely specialised anatomy and the species level taxonomy has been revised severally (Reid, 1985). However, few of these revisions are considered valid. The taxonomy of the Nile perch in Lake Victoria has always been assumed to be resolved. This is far from the truth since it is believed that both Lates niloticus and Lates macrophtalmus could have been introduced at the same time. The Lates sp. In the lake could also consist of a hybrid of both or even from introductions from the Lake Turkana (Harrison, 1990). The species of Lates occurring in L. Victoria are therefore unknown.

The specimens for this study have been collected from various sites along the Tana and Athi river and preserved at the Ichthyology department of the National Museums of Kenya. The study was carried out in the Ichthyology department in collaboration with the Molecular Genetics department.

The method used involves application of both morphologic and genetic techniques simultaneously. Morphological studies involves conducting certain measurements and meristic counts to establish the phenotypic features of the species. The data are applied to Principals Component Analysis. Genetic studies involve various techniques such as Allozyme/Isozyme and mtDNA. MtDNA is preferred to nDNA because it has a faster mutation rate and thus is more sensitive to tests for variations within natural populations.