Rice bacterial blight (
Xanthomonas oryzae pv.
oryzae
) is a major constraint to rice (
Oryza sativa
L.) production in Uganda and as part of strategies to develop resistant cultivars, it is important to evaluate resistance of commonly used cultivars. A full-diallel mating design involving three resistant and three susceptible rice cultivars was used to produce F
1 and F
2 progenies in a screen-house at the National Crop Resources Research Institute (NaCRRI), Namulonge in Uganda. The parents and F
2 populations were challenged with the ≤
Xanthomonas oryzae pv.
oryzae isolate (UX00) and lesion lengths were scored 21 days after inoculation (DAI). Griffing’s combining ability analysis showed significant specific combining ability (SCA) and non-significant general combining ability (GCA) effects, indicating the preponderance of non-additive gene effects in controlling the resistance to bacterial leaf blight (BLB) in rice. Rice genotypes, NERICA14, NERICA10 and NERICA4 had desirable GCA estimates, and were, therefore, the best general combiners. Crosses CO39 x NERICA10 and NERICA14 x IRAT104 had favorable SCA values. These hybrids are thus, promising in developing the BLB resistant progenies. Significant reciprocal effects indicate the importance of maternal contribution in controlling the BLB virulence. For this, resistant lines can be used as female parents for fear of affecting transfer of resistance to the progenies, and the hybrids and their reciprocals would be handled separately. Low estimates of narrow sense coefficient of genetic determination (NSCGD) (0.9%) and medium broad sense coefficient of genetic determination (BSCGD) estimates (16.4%) highlight the influence of non-additive gene action in controlling the resistance to BLB, confirming an effective selection of superior genotypes at advanced generations when the maximum homozygosity is fixed.