Abstract

The first goal of plant breeders in crop improvement is to develop genotypes with high yield and stable across environments. AMMI and GGE-bilpot are the most effective statistical tools for the analyses of stability, adaptability and helps for proper selection of sugarcane genotypes. The present investigation was carried out on eight introduced sugarcane genotypes excluding two standard checks for three cropping cycles at Finchaa and Metehara forming together six environments. The trial was laid out in completely randomized block design. This work was conducted with objective of evaluating G x E interactions on sugar yield performances of sugarcane genotypes using AMMI and GGE statistical tools. The results of combined ANOVA and AMMI analysis of variance for sugar yield showed that highly significant (p ≤ 0.001) difference among genotypes and environments and their interactions. The sugar yield of the genotype was influenced by environments which explained 73.77% of the total variation indicating the importance of environmental main effects over genotypic main effects. The first-IPCA1 (59.08%) showed highly significant level whereas the second-IPCA2 (18.66%) were not significant and totally explained 77.74% of the variations. The analysis of the AMMI resulted that genotypes G10 (1.928), G2 (1.744), G9 (1.683) and G4 (1.633) had high mean sugar yield in ton/ ha/month. The GGE-biplot analysis grouped the environments into two mega-environments and sub-divided the graph into five sectors. The first mega-environment was made up by three environments: E1, E2 and E4 while the second mega-environment was made up by E3, E5 and E6. G2 and G10 was located very nearest to the concentric circle; thus, considered to be ideal the genotype and therefore identified as the best genotype than the others. Consequently, these two genotypes could be selected for verification and recommended for commercialization