English French

Bassirou Sani Boubacar Gaoh



Pearl millet is the most important staple crop for millions of people in Niger Republic. Germplasm biofortified with iron and zinc is now available but there is limited information on farmers’ varietal preferences and production constraints, gene effects controlling the inheritance of iron and zinc concentration and other agronomic traits, heterosis and hybrid stability. A participatory rural appraisal was conducted in three departments of Dosso region, the most afflicted region by anaemia in Niger. Data were collected using focus group discussions, and a semi-structured questionnaire was used to interview 150 farmers. The results showed that farmers’ preferred traits were long panicle, high seed set, earliness, bigger and white grains. Among these traits, long panicle and high seed set were the most preferred. The most important pearl millet production constraints were poor soil fertility followed by drought and the parasitic weed striga. All the interviewed farmers were willing to purchase and grow grain iron and zinc biofortified pearl millet varieties if they were made available. Women were found to be more aware of malnutrition and its consequences than men. A large majority of the respondents performed grain decortication that is reported to reduce grain micronutrient content. Fifty six hybrids from an 8 8 full diallel mating plus the eight parental lines, along with six checks were evaluated in an alpha lattice design with three replications in three locations (Gaya, Maradi, and Sadore) during two rainy seasons (2017 and 2018) to determine the combining ability and gene action conditioning the high level of iron and zinc concentration in grain and other selected agronomic traits; the heterosis and interrelationships among these traits; and the stability of hybrids rich in kernel iron and zinc concentrations and with high yield. The parental lines Jirani, LCIC9702 and MORO had positive significant general combining ability (GCA) effects for grain iron concentration while Jirani and MORO had the positive significant GCA effects for grain 4  zinc concentration. Positive significant GCA effects were recorded for plant height and panicle length (four parents), panicle girth and thousand seed weight (two parents) and for grain yield (ICMV 167006), whereas negative significant GCA effects were observed for days to flowering and panicle compactness (three parents) and downy mildew incidence (two parents). For the specific combining ability (SCA), among the 56 hybrids evaluated, only the hybrids LCIC9702 x Jirani and MORO x ZANGO had positive significant SCA effects for grain iron concentration across sites, and for grain zinc concentration the hybrids Gamoji x MORO, LCIC9702 x Jirani, and ICMV 167006 x Jirani had positive significant SCA effects. Positive significant SCA effects for grain yield were observed for seven hybrids, and among these hybrids ICMV 167006 was in the pedigree of five hybrids. Moreover no positive significant mid-parent or better parent heterosis was found for grain iron and zinc concentration. The reciprocal effects were significant for grain zinc concentration, grain yield, days to 50% flowering, plant height, thousand seed weight and downy mildew incidence which suggested that the choice of which is female or male parent is critical in the hybrid production when these traits are considered. The relatively high proportion of GCA effects over SCA effects with the predictability ratio close to unity revealed that grain iron and zinc concentration, flowering, plant height, panicle length, panicle girth, panicle compactness and downy mildew incidence were predominantly under additive gene actions, while grain yield and thousand grain weight were predominantly under non-additive gene action. A high positive correlation was found between grain iron concentration and grain zinc concentration, which may imply that these traits can be improved by selecting for only one of them. Plant height and panicle length had low to moderate significant negative correlation with grain iron and zinc concentration. Grain yield was positively correlated (low) to plant height, panicle length, panicle girth and thousand seed weight across all environments. The 5 stability analysis revealed that the hybrid ICMV 167006 x Jirani was the most stable and high yielding with high level of kernel iron and zinc micronutrients. Generation mean analysis model was used to study epistasis and estimate gene effects for grain iron and zinc content in pearl millet. The six parameters model revealed that for grain iron and zinc concentration, additive gene effects were predominant compared to non-additive gene effects in the inheritance of these traits. For grain iron content, the duplicate type of espistasis was determined and the additive x additive gene effects type of non-allelic gene effects were the most important. In contrast for grain zinc concentration the non-significance of the parameters h (dominance) and l (dominance x dominance) did not allow the determination of the type of epistasis.