Sorghum is the second most important cereal crop in Kenya after maize, grown on an area of 117000 ha -1 with about 144000 tonnes being produced annually. Striga hermonthica is among the major causes of sorghum yield loss especially in Western and Nyanza regions of the country. Farmers have traditionally managed Striga using cultural methods but the most `effective and practical solution to poor smallholder farmers is to develop Striga resistant varieties. A field trial consisting of Sixty-four sorghum genotypes comprising of wild relatives, landraces, improved varieties and F4 progenies were evaluated in a sickplot (field with Striga inoculum capable of causing up to 100% incidence in susceptible sorghum genotypes) and in a potted trial at KALRO-Kenya Agricultural and Livestock Research Organization Alupe during the 2019 rainy season. The experiment was laid out in a square lattice design with three replications. These accessions were also genotyped using Diversity Array Technology markers to assess their diversity. In another experiment, Marker Assisted Selection (MAS) was used to transfer Striga resistance quantitative trait loci into adapted farmer preferred varieties Gadam and Kari-Mtama-1. Crosses were also made between known Striga resistance namely N13, Framida, SRN39 and Hakika as donor sources and Gadam and Kari- Mtama-1 as the female parents to obtain F1 and BC1F1 generations. Backcross generation crosses were genotyped using DArT markers to trace heterozygous alleles and to confirm successful backcrossing. The (ASNPC) selection criteria was used to identify resistant genotypes in the trials. Wild genotypes GBK045827, GBK044336, GBK047293 and GBK048921, improved varieties F6YQ212, ICSV III_IN and F4 population F6YQ212 × B35, B35 × Lodoka and B35 × ICSVIII_IN had lower ASNPC values than N13, the resistant check under sickplot conditions. Four wild genotypes GBK016109, GBK016085, GBK045827, GBK048152, one improved variety F6YQ212 and three F4 population crosses F6YQ212 × B35, LODOKA × Landiwhite, ICSVIII_IN × E36-1 had lowest ASNPC values in the potted trial. Genotypes SRN39, F6YQ212, GBK045827 and F6YQ212 × B35 were the most resistant to Striga in both field and potted trials. MACIA, B35, E36-1, OKABIR × AKUOR-ACHOT and LODOKA × ICSVIII_IN were the most tolerant to Striga recording superior yield performance in both trials. Negative correlation was observed between yield traits (100 grain weight, dry panicle weight, yield (t/ha) and Striga related traits across both trials while Striga response related traits (ASNPC, NSmax, NSFC) significantly (<0.001) correlated positively with each other in both trials. Days to flowering and plant height were also negatively correlated to yield and Striga resistance. xiv The overall best performing genotypes in terms of Striga resistance and yield in both trials were Macia, SRN39,GBK 045827 and GBK 016085. SNPs generated from DArT-sequencing grouped the genotypes into three major clusters, with all resistant checks grouping in the same cluster except N13. The results from this analysis revealed successful backcrosses for the crosses Gadam × N13 × Gadam, Gadam × Framida × Gadam and Gadam × SRN39 × Gadam with heterozygous allele percentages ranging from 63% to 77%. High heritability values for yield and ASNPC suggest additive gene action and selection for improvement of these trait will be beneficial. Demonstrated genetic gain for Striga tolerance points the possibility of development of Striga tolerant varieties that give substantial yield under Striga pressure. The study showed that Striga resistance and Striga tolerance alleles are available within the local wild relatives, in local landraces and in improved sorghum genotypes and there is need tap into this potential to improve sorghum production in the crop.

show more