Genetic Potential for Pearl Millet Hybrid Development in West Africa

Pearl millet represents the first staple food grain and constitutes the bedrock of food security in Niger where it is alternatively used as processed foods and feed for animals and poultry. With a population growth faster than food progress and limitation of favorable lands, increased grain yield is a high priority in breeding programs to provide food security. Cultivars with high and stable yield are thus an integral part of meeting this goal and variety improvement remains a goal of both national and international agencies. Phenotypic selection of parents for hybridization based on their performance per se alone may not always be a viable procedure but parents shall be selected on the basis of their combining ability. For the reason, heterosis and combining ability of sixteen hybrids developed from eight varieties belonging to two putative heterotic groups were studied. Entries were tested in two locations during two years in Randomized Complete Block Design with three replicates regardless the location and the year. Data on grain yield and productivity related traits were recorded. Eight hybrids outyielded the check Zatib which in turn performed better than the best parent Souna-3. All the eight hybrids were more stable than all of the parents. Parents HKB, Ankoutess, and Gamoji had the best GCA for grain yield. The cross Gamoji x HKB showed the highest SCA and the best mid-parent heterosis for grain yield resulted from the two parents with high and positive GCA. Ex-Borno, H80-10Gr, Moro, and HKP-GMS had negative GCA for grain yield, but the hybrids Ex-Borno x H80-10Gr and Moro x HKP-GMS showed high and positive SCA for this trait. Increased grain yield was found to be positively and highly correlated with peduncle girth while hybrids showed more phenotypic uniformity than their parents. Mean squares for GCA were significant for most of the traits underlining the importance of additive gene effects. The ratio σ2 gca / σ2 sca for grain yield was less than one indicating the prevalence of non-additive gene effects in the yield control. There were G x E interactions suggesting the necessity of further testing of the germplasm.