AMMI analysis of grain yield in rice genotypes across environments

Abstract

The identification of the highest yielding cultivar for a specificenvironment on the basis of both genotype (G) and genotype xenvironment (GE) interaction would be useful to breeders andproducers since yield estimates based only on G and environment(E) effects are insufficient. The objective of this study wasto demonstrate the usefulness of additive main effects and multiplicativeinteractions (AMMI) model analysis. Replicated grain yielddata of rice (Oryza sativa L.) genotypes at three locations in Coimbatore (E1), Bhavanisagar (E2) and Aliyar Nagar (E3) were obtained andused for this purpose. Through AMMI model analysis, the magnitudeand significance of the effects of GE interaction and its interactionprincipal components relative to the effects of G and E wereestimated. Among the three, E3 environment was far away from the origin.  The environments E2 and E3 condition were exhibiting high interaction effect, whereas the environment E1 had moderate interaction effect with the other two environments. The parent TP 1021 R and the hybrids viz., IR 80559 A X RR 286 R, APMS 6 A X IR 62037 R, IR 75601 A X IR 72865 R, IR 75601 A X RR 347 R, IR 75596 A X MDU 5 R, IR 75596 A X ASD 06-8 R, CRMS 32 A X IR 62037 R, CRMS 32 A X TP 1021 R and IR 75608 A X ASD 06-8 R showed IPCA 1 score close to zero with high main effects indicating that these varieties were less influenced by environments. Hence, the above said hybrids were stable and had general adaptability for all the environments. Since, the environments E3 had positive IPCA 1 score and high mean, it had positive interaction with these genotypes and this environment was considered as the stable environment for these hybrids.