|dc.description.abstract||Drought stress, maize lethal necrosis (MLN) and storage pests, mainly maize weevil (Sitophilus zeamais Motschulsky) and larger grain borer (Prostephanus truncatus Horn), are among the most important maize production constraints and storage problems in tropical and subtropical environments. Drought stress and MLN can cause grain yield losses of up to 90% depending on the severity and stage of growth when they affect the crop, while post-harvest pests can cause 10-60% grain loss. There are no practical agronomic practices that can control these stresses under small-scale farming conditions since investments in irrigation and pesticides are unaffordable for the majority of small-scale farmers in developing African countries. The study was, therefore undertaken to: a) estimate the heritability and gene effects controlling maize weevil and larger grain borer resistance in tropical maize germplasm; b) determine whether resistance to maize lethal necrosis and tolerance to drought can be combined in F1 hybrids developed from tropical maize inbred lines; c) determine the combining ability of tropical maize inbred lines for drought tolerance and resistance to maize weevil and assess the possibility of combining the two traits in one genotype and d) determine gene action controlling the morpho-physiological and agronomic traits of tropical maize under maize lethal necrosis virus infected conditions and maize weevil infestation.
Populations involving six generations; two parents (P1 and P2), F1, F2 and backcrosses (BCP1 and BCP2) were developed from cross one, CKDHL120731 (resistant) × CKDHL120918 (susceptible) and cross two, CKDHL120517 (resistant) × CKDHL120918 (susceptible). The generations were evaluated under artificial infestation of Sitophilus zeamais Motschulsky and Prostephanus truncatus Horn in separate experiments in a post-harvest laboratory at Kiboko, Kenya. Data was recorded for percentage kernel weight loss, kernel damage and the final number of living insects and data were analysed using generation mean analysis. Results revealed that resistance traits for both crosses did not fit a simple additive-dominance model for S. zeamais, suggesting the existence of epistasis effects. However, for P. truncatus resistance, cross one fitted a simple additive-dominance model, but cross two did not, suggesting both simple additive-dominance model and digenic interaction model were important in the inheritance of P. truncatus resistance. Additive, dominance and epistasis gene effects played a role in the inheritance of resistance to both insects in the selected maize genotypes. This was further confirmed by the moderate narrow-sense heritability estimates which suggested the involvement of additive and non-additive gene effects in the expression of resistance to both insect pests.
Three separate half-diallel analyses involving eight inbred lines each were conducted involving (1) lines with varying reactions to drought and maize lethal necrosis (MLN), (2) lines with varying drought tolerance and post-harvest pest resistance backgrounds, and (3) inbred lines with varying reactions to maize lethal necrosis (MLN) resistance and maize weevil resistance. The F1 hybrids from these diallel crosses were evaluated in different locations under optimum conditions and managed drought stress, artificial MLN infestation or artificial infestation with maize weevil depending on the objective. For artificial weevil infestation, grain samples were obtained from sites with optimum conditions.
Hybrids differed significantly (p<0.001) for MLN resistance and drought tolerance traits, including MLN scores, senescence, days to anthesis and anthesis-silking interval. The yield reduction due to MLND was 93% of the optimum mean grain yield of 6.04 t/ha, while reduction due to drought stress was 67% of the same. Genetic analysis detected highly significant mean squares (p< 0.001) due to both general combining ability (GCA) and specific combining ability (SCA) for most of the recorded traits, including grain yield under all environments, suggesting the importance of both additive and non-additive gene effects. However, additive gene action was generally predominant across all evaluation conditions. The results suggest that it is possible to improve tropical maize for combined drought and MLN tolerance and it can be faster when the evaluation is conducted under combined drought and MLN conditions.
Highly significant genotype and genotype × environment interaction mean squares (p<0.001) for grain yield and days to anthesis were observed under drought and optimum conditions for the drought tolerance and weevil resistance. Highly significant genotypic effects (p<0.001) were also observed on the key parameters for maize weevil resistance; Dobie’s Susceptibility Index (SI), living insects, weight loss (WL) and seed damage (SD) revealing different reactions of the tested hybrids. In addition, highly significant mean squares (p<0.001) due to both GCA and SCA for grain yield under drought and significant (p<0.001) under optimum conditions were detected, suggesting the importance of both additive and non-additive effects. Under maize weevil infestation, highly significant mean squares (p<0.001) due to both GCA and SCA for the key parameters were observed except for GCA mean squares for weight loss which was significant (p<0.05). Additive gene action was predominant over non-additive for grain yield under drought, SI, SD and living insects. The cross CKDHL120731 × CKDHL120517 showed tolerance to drought and resistance to maize weevil, while 24 hybrids showed tolerance to drought only.
The maize lethal necrosis (MLN) resistance and maize weevil resistance F1 hybrids showed highly significant (p<0.001) genotype differences for field weight and grain yield under MLN infestation. Highly significant (p<0.001) genotype and genotype × environment interaction effects were also observed for MLN scores at the early and late-stages under artificial MLN infestation, grain yield under optimum conditions, SI, WL and SD under maize weevil infestation. Significant mean squares (p<0.01) due to only GCA for grain yield under MLN and weight loss under maize weevil infestation were detected, while highly significant mean squares (p<0.001) due to both GCA and SCA for MLN scores under MLN infestation and grain yield under optimum growing conditions, SI and SD under weevil infestation were observed suggesting the importance of both additive and non-additive effects.
However, for most of the traits under the three evaluation conditions, additive gene action was predominant. Three hybrids CKDHL120918 × CKSBL10060, CKSBL10060 × CKDHL120731 and CML494 × CKDHL120731 showed good performance under the three evaluation conditions. The observed importance of both additive and non-additive gene action, with predominance of additive gene action, especially under the stressed environments, is an indicator of the feasibility of breeding for resistance to combined stresses, and suggests that recurrent selection can be applied for rapid breeding progress. Furthermore, the improvement of tropical maize for combined stress resistance can be faster when the inbred lines and hybrids are developed and evaluated under the combined stress environments, than under a single stress. The identified superior genotypes across environments in this study can be used immediately in breeding programs, especially in sub-Saharan Africa.||en_US