Maize landraces have the highest genetic variation and adaptation to less favourable farming environments. In Eswatini, maize landraces occupy about 20 percent of the maize crop planted annually. The current study was undertaken to assess the morpho-agronomic diversity, heritability and genotype by environment interaction among 70 genetically diverse maize landraces collected from different farmers in the country. The maize landraces (genotypes) were grown in replicated trials at Malkerns in Eswatini for three consecutive years (environments) from 2016 to 2019. Data were recorded on 15 morpho-agronomic traits, where significant differences were observed for genotypes on individual traits in all three environments. The combined ANOVA also indicated significant differences for genotypes, environments and genotype by environment interaction in most of the traits. Close resemblance between genotypic coefficient of variation (GCV) and phenotypic coefficient of variation (PCV) was observed in number of ears per plant, ear diameter and ear length indicating minimum influence of the environments in the expression of these traits. Broad-sense heritability ranged from 18-99% with anthesis silking interval with the least and ear length with the highest. Analysis of genotype by environment interaction (GEI) indicated that genotype, environment and genotype by environment interaction explained variable amounts of total variance across traits, respectively. Number of kernels per row, plant height and one thousand seeds weight were more affected by environmental changes, while anthesis silking interval and kernel size traits were less affected thus indicating greater resistance to environmental changes. Based on the deviation from regression stability parameter, accessions H328, M640 and L161 showed stability of grain yield and adaptability in the Malkerns area. Further testing of these landraces in different locations across the country’s five agro-ecological zones is recommended.

Keywords: Maize Landraces, Diversity, Morpho-agronomic Traits, Heritability, GxE Interaction, Cluster Analysis, PCA