Abstract:

This study used morphometric characteristics and fish condition to compare populations of
Barbus altianalis from the River Nile, Lake Edward and Kazinga Channel. Morphological
variation was determined using multivariate analysis of advanced truss and non-truss network
systems for 251 individuals collected from commercial catches. Fish condition was assessed
using relative condition index (Kn) computed using individual fish. Discriminant Function
Analysis (DFA) indicated significantly (P<0.00l) higher level of heterogeneity between
populations with the two factors (DFA1 and DFA2), collectively accounting for 100% of the
variance; (DFA1) 84% and (DFA2) 16%. Group means on the DFA1 showed the River Nile (-
0.8869) separated furthest from Lake Edward (0.796) and the Kazinga channel (0.6796). River
Nile had the highest number (80.53%) of individuals correctly classified, while Lake Edward
(67.2%) and Kazinga channel (60.5%) had more misclassifications between them, an indication
they could be much harder to separate. There was a significant difference in mean relative
condition among the three habitats (F2, 239 = 9.73, P <0.001, n = 240). The Lake Edward
population was in a better condition (1.05) compared to Kazinga Channel (1.04), while the River
Nile population had the lowest condition (0.96). Catch rates were higher in Kazinga Channel
6.85±L49 Kg boat’1 day"1) and Lake Edward (6.20±1.86 Kg boat"1 day"1) and lowest in the River
Nile (1.92±0.59 Kg boat"1 day"1). Dissolved oxygen was below the required minimum of 5 mgL"1
in all the water bodies. Conductivity was highest in Lake Edward (312 ±115.78 j.iS cm"1),
followed by Kazinga channel (298.06±l31.89 pS cm"1). The observed variations were mainly
associated with environmental condition faced by B. altianalis as indicated by water quality and
harvest pressure.