AMMI, GGE-BIPLOT, AND JOINT REGRESSION TECHNIQUE AS A TOOL IN MEASURING G × E INTERACTION IN 3-WAY CROSS MAIZE (Zea mays L.) HYBRIDS

Abstract

In order to meet the food requirements of ever increasing population, effectiveness of yield selection is important. Genotype by Environment Interaction (GEI) has a large impact in selecting adapted and predictable genotypes. Therefore, there is the need to evaluate maize genotypes across different environments, seasons or locations for a successful selection. Twelve 3-way cross maize hybrids obtained from International Institute of Tropical Agriculture (IITA) were evaluated on the field of Federal University of Agriculture, Abeokuta, Nigeria (Latitude 7⁰ 15⁰ N and Longitude 3⁰ 25⁰ E) across 3 growing seasons of 2021 and 2022. The experiment was laid out in a Randomized Complete Block Design (RCBD) with three replicates. Data were collected on 50 % days to tasseling, 50% days to silking, plant height, ear height, field weight, 1000 seed weight and grain yield. Data were subjected to analysis of variance and means were separated using Least Significance Difference at 5% probability level.  Additive Main Effect and Multiplicative Interaction (AMMI), Genotype plus Genotype × Environment (GGE biplot) and Joint Regression Techniques were used to identify stable and high yielding genotypes. The AMMI analysis showed that the total variance in the yield of the three-way maize hybrids accounted for by Genotype (G), Environment (E) and genotype × environment interaction (G × E) were 30.6 %, 44.19% and 25.31% respectively. Based on AMMI biplot, genotypes LW1701-10 and OBA SUPER-9 which combined high yield with stability were the most desirable. The GGE biplot showed that hybrids LW1701-10, OBA SUPER-9 and LW1701-6 were the most stable and desirable genotypes. The Joint Regression Technique showed that the performance of the genotypes could not be revealed in a linear manner as the deviation component variance accounted for 81.05 % and identified LW1701-6, LW1701-16, LW1701-12, LW1701-21, LW1701-4 as stable and desirable genotypes. The study concluded that GGE and AMMI models were effective in the study of yield stability of maize hybrids than the Joint regression technique.