GENOTYPE X ENVIRONMENT INTERACTION AND PHENOTYPIC STABILITY ANALYSIS OF DUAL- PURPOSE TRITICALE (X TRITICOSECALE WITTMACK)

Abstract

Ten dual-purpose triticale genotypes were evaluated in four different locations of Bangladesh to determine the genotype x environment interaction and phenotypic stability of six traits including yield and forage biomass using the Eberhart and Russell's model and the AMMI model. Significant differences were observed for genotypes, environment and genotype x environment interactions for all the six characters studied. The G x E interaction was further partitioned into linear and non-linear components. The G x E interaction (linear) was highly significant for all the traits revealed that there were considerable genetic differences among the genotypes for their regression coefficients on the environmental index. Significant pooled deviation (non-linear component) for G x E interaction for all the six traits suggested that the triticale genotypes also differed considerably in respect of their stability. Considering stability parameters of the Eberhart and Russell's model; phenotypic index greater than zero (Pi>0), regression coefficient around unity (bi = 1) and least deviation from regression (non-significant S2di), the triticale genotype E-5 with low mean yield was the most stable for grain yield, whereas the genotypes E-7, E-4 and E-5 were regarded as the most stable for green forage weight. Similarly, the genotypes E-1, E-7 and E-8 for days to maturity, the genotypes E-4 and E-8 for plant height, the genotypes E-6 and E-2 for grains per spike and the genotypes E-4 and E-10 for thousand grain weight were regarded as the most stable triticale genotypes. The genotypes E-5, E-4 and E-7 were regarded as the most stable considering their overall stability performances. Similarly, the genotypes E-6 and E-10 were highly responsive to environmental changes and were considered suitable for favorable environments. Whereas the genotypes E-2, E-3 and E-8 were least responsive to environmental changes and were considered suitable for poor environments. According to the AMMI model, E-5 with low mean yield and E-4 with higher mean yield than grand mean yield were the most stable triticale genotype for grain yield, whereas E-5 and E-7 were regarded as the most stable genotypes for green forage weight. Similarly, the genotypes E-8 and E-1 for days to maturity, the genotypes E-4 and E-8 for plant height, the genotypes E-6, E-1 and E-8 for grains per spike and the genotypes E-4 and E-7 for thousand grain weight were regarded as the most stable genotype. With overall ranking, the genotypes E-5, E-4 and E-7 were regarded as the most stable triticale genotypes considering their overall stability performances. Similarly, the genotypes E-9 and E-3 were the most unstable genotype with higher grain yield and green forage weight. However, these genotypes showed high performance in favorable environments. A comparison of the stability analysis for the triticale genotypes with the above mentioned two models gave the similar stability ranking of the genotypes. With both models, the triticale genotype E-5 with low mean yield and the triticale genotype E-4 with higher mean yield than the grand mean with average green forage biomass were regarded as the most stable dual-purpose triticale genotypes considering overall stability performances. Comparatively, the AMMI analysis model was found more suitable for accurate estimate and meaningful interpretations in analyzing the G x E interaction vis-a-vis in estimating the phenotypic stability of some dual-purpose triticale genotypes across the environments.