IDENTIFICATION OF STABLE FORAGE SORGHUM GENOTYPES USING UNIVARIATE AND MULTIVARIATE ANALYSES
ARUNA C., S. K. PAHUJA, P.K. SHROTRIA, S. K. JAIN, S. SIVA KUMAR, D. T. DESHMUKH
Indian Institute of Millets Research,
Hyderabad, India
*(e-mail : aruna@millets.res.in)
(Received : 21 November 2017; Accepted : 18 December 2017)
SUMMARY
Genotype selection and recommendation is influenced by genotype x environment interactions (GEI). Identifying stable and high yielding cultivars is the main objective of the multi-environment data analysis. The objective of this study was to analyze rank correlations among ANOVA, Eberhart and Russell and biplot analyses in ranking of genotypes for yield, stability and yield-stability. The study included 16 forage sorghum genotypes evaluated at 14 environments across two years. The results showed significant effects due to environments, genotypes and GEI, suggesting differential response of genotypes. Environment (E) main effects accounted for >80% of the variation, compared to <20% for genotype (G) and GEI effects together. For yield rankings, all the three methods are positively and significantly correlated, while for stability ranking, ER and biplot analysis had positive significant correlation, indicating that both methods
have identified the same genotypes for stability. GGE biplot has the advantage of identifying mega-environments and the genotypes for each mega-environment.
Key words: GGE biplot, Joint regression analysis, stability, yield ranking, correlations