Stability Analysis of Rice (Oryza sativa L.) Hybrids Utilizing Regression and AMMI models

Four CMS lines were crossed with 13 restorers in a line x tester fashion during kharif, 2010. Fifty two hybrids along with three checks were evaluated for yield and yield contributing characters over four different agro-climatic zones in Andhra Pradesh, India during rabi 2010–11. The analysis of variance of Eberhart and Russell indicated that G x E interaction was significant for all the characters under study and that genotypes differed significantly. AMMI is a useful tool for interpreting genotype ´ environment interaction in multi environment trials. Among the AMMI components first IPCA axis explained most of the portion of G x E interaction than other IPCA axis. According to Eberhart and Russell model the hybrids like APMS 10A x MTU 1071 and APMS 10A x MTU II-190-1-1-1-1-1 showed stable performance whereas in AMMI analysis the hybrids viz., APMS 10A x MTU 1071, APMS 10A x MTU II -187-6-1-1, APMS 9A x MTU II -187-6-1-1, APMS 10A x MTU II-190-1-1-1-1-1 and IR 58025A x MTU II-110-9-1-1-1-1 were more stable as they recorded IPCA score near zero and less interaction with environments. AMMI model ANOVA brings out G x E interaction through IPCA1 which gets combined with error in the other ANOVA and points out the utility of AMMI models in studying the G x E interaction and identifying stable genotypes for characters which so undetected in the earlier analysis.