No abstract

The identification of stable genotypes with high yield in diverse multiple‐stress environments is important to increase maize (Zea mays L.) grain yield under tropical environments. Our objective was to assess the yield performance and stability of experimental hybrids and broad‐based populations of tropical maize across diverse environments in Southeastern Brazil. We evaluated two sets of maize genotypes for grain yield: 190 experimental hybrids along with 6 commercial hybrids and 45 population hybrids along with their 10 parental populations across 8 environments in Southeastern Brazil. Multiple statistical methods were used and compared in the analyses. Combined analysis of variance indicated that genotypic main effect (G), environmental main effect (E), and genotype by environment interaction were highly significant (p < 0.0001) for grain yield. The E accounted for 42% of the total variation for both hybrids and populations, and they were more similar within the growing season than between seasons, mainly for populations. Low nitrogen (N) stress was a key factor in hybrid evaluation and recommendation, particularly under drought stress conditions. Among the environment classification methods, genotype main effect plus genotype × environment interaction (GGE) biplot provided more accurate information about environments grouping and selection of the genotypes than the Eberhart and Russell method. Harmonic mean of the relative performance of the predicted genetic values (HMRPGV) based on mixed models ranked the hybrids and populations according to mean grain yield and stability, penalizing hybrids, and populations with lower stability. Therefore, we recommend the GGE biplot and HMRPGV for genotype evaluation based on multi‐environment trials data. These methods identified 92V2144 and 92V2033 as the most promising hybrids for favorable and 92V2141, 92V2153, and 92V2137 as the most promising for unfavorable environments. 92VX033 and 92VX043 were identified as broadly adapted and stable populations across multiple environments in Southeastern Brazil.