This study estimated the variance components and heritability of yield and yield related traits of a population of 96 diverse bread wheat genotypes under drought-stressed and non-stressed conditions. The genotypes were evaluated across eight testing environments during the 2014/2015 and 2015/2016 growing seasons using an alpha lattice design with two replications. Combined analysis of variance and variance components were analysed following the General Linear Model (GLM)'s procedure. Results indicated the presence of significant effects of genotypes, seasons, sites, and water regimes and their interactions. From the eight testing conditions, high levels of genotypic variance (σ2g) were estimated for spike length (73%), number of spikelets per spike (44.19%), plant height (51.26%), number of kennels per spike (32.98%), number of days-to-heading (44.24%) and thousand seed weight (22.98%), resulting in high broad-sense heritability estimates of > 0.50. Conversely, genotypic variation was relatively moderate for the number of days to maturity, grain yield and number of productive tillers per plant, contributing to 15.03%, 8.46% and 6.13% of the total variation, respectively. The heritability estimates of the later traits were 20% ≤ H2 < 50% which may limit their selection gains under droughtstressed environments. Further, quantitative trait loci analysis and progeny testing are required to discern the number of genes and associated genetic effect and to pinpoint genomic regions in the tested wheat genetic resources for effective breeding for drought tolerance.