Genetic characterization of pearl millet(Pennisetum glaucum (L.) R.Br) genotypes in Zambia

The most important objective in any crop improvement programme is to increase yield. Current grain yields of pearl millet among small-scale farmers in Zambia are very low (500 – 650kg/ha) due to the fact that farmers have continued to use landraces and varieties that are low yielding. There are no hybrid varieties that have been developed simply because the basis for developing hybrid varieties has not been established in Zambia. This study, therefore, aimed at characterising pearl millet genotypes in order to establish a basis for exploiting the genetic potential of pearl millet for development of high yielding varieties (hybrid varieties) in Zambia. To contribute to this realisation, the study involved the determination of combining ability for grain yield in cytoplasmic male sterile lines and restorer lines as well as determining the nature of gene action controlling grain yield and other important traits. A total of 104 crosses developed as lines x testers (i.e. 13 lines and 8 testers) combinations were evaluated during the 2012/2013 growing season at ZCA-Monze (52 crosses i.e. 13 x 4) and Longe (52 crosses i.e. 13 x 4) in a triple lattice design. Results showed that the performance of the crosses in terms of grain yield, days to 50% flowering, plant height, productive tillers per plant, panicle length, panicle girth, and panicle weight was variable. General combining ability (GCA) effects revealed that male parents; ZPMV 28001, ZPMV 28010 and ZPMV 28011 and female parents; NCD2A4 and ICMA4 02999 that showed significant (P≤0.05) positive GCA effects for grain yield, can be used for generating hybrid pearl millet varieties. On the other hand, specific combining ability (SCA) effects revealed that the five crosses; ZPMV 28010 x ICMA1 00444, ZPMV 28011 x ICMA1 92888, ZPMV 28011 x ICMA4 04777 and ZPMV 28013 x ICMA1 97111 and ZPMV 28015 x ICMA1 97111 that showed significant (P≤0.05) positive SCA effects for grain yield, can be utilized as high yielding hybrid varieties. The determination of the nature of gene action conditioning grain yield, productive tillers per plant and panicle weight was largely controlled by non-additive gene action while panicle girth and panicle weight were largely controlled by additive gene action. Since grain yield is largely controlled by non-additive gene action which can only be exploited by developing hybrid varieties, it can be concluded that these inbred lines and cross combinations represent a good choice to make future strategy for the development of pearl millet hybrid varieties in Zambia.