Abstract
Increasing production requires the development of high yielding cultivars adapted to various environments. Multi-environment trials (MET) remains the best approach to assess the performance of accessions across environments. The objective of this study was to select the best performing and stable accessions of sesame across different environments in Benin Republic. Nineteen sesame accessions were evaluated across eight environments using a randomized complete block design with four replicates. The accessions were evaluated for three traits: days to 50% flowering (D50F), thousand-seed weight (TSW), and seed yield (SY) during 2020 growing season. The stable and top-performing accessions across environments were determined using AMMI (Additive main effects and multiplicative interaction), GGE (Genotype main effect and genotype x environment interaction), and MTSI (Multi-trait stability index). AMMI analysis of variance showed a significant difference across environments for the three traits. The accessions were affected by environmental conditions for the three traits. The broad-sense heritability estimates were high (>0.60) for all the traits, indicating the improvement is achievable through selection. AMMI1 and AMMI2 biplots identified G10 and G13 as high seed yielding accessions adapted to environments E1, E2. The GGE biplot showed two mega-environments for TSW and three mega-environments for D50F and SY. For SY, G11 and G13 were the best accessions in the first mega-environment, G10 the best accession in the second mega-environment; G3 and G8 were the best accessions in the third mega-environment. AMMI and GGE analyses identified G10, G5, G12 as high seed yielding and stable accessions across environments. GGE biplot revealed that E1 and E2 were the most suitable environments for multi-location trials based on their discriminating ability and representativeness. MTSI indicated G10, G13, G19 as promising germplasm to be recommended for breeding program