This investigation had the objective to evaluate the grain yield adaptability and stability of 18 wheat dihaploid inbred lines and two wheat cultivars, at four locations of the State of São Paulo, Brazil, from 2001 to 2004, utilizing three methods (Eberhart and Russell; Cruz and colleagues and Lin and Binns, modified by Carneiro). During each year the experiments were carried out at: Monte Alegre do Sul and Tatuí (sprinkler irrigation and limed soil); Capão Bonito (upland and acid soil) and Tietê (upland and limed soil), totalizing 16 experimental areas. Randomized block design with four replications was utilized. The joint analysis of variance for the 16 experiments and the analysis of variance for each location, showed that all sources of variations presented significant effects, except genotypes x years interactions in Monte Alegre do Sul. In these location cultivars IAC-24 and IAC-370 were superior for grain yield considering the average of the four years. Adaptability and stability methods were calculated for the other locations. Considering the best genotypes for grain yield taking in account the 16 experiments, the three methods indicated that the lines 5, 6 and 7 were adapted to unfavorable environments. IAC-24 and IAC-370 cultivars were adapted to suitable environments. Lin and Binns method indicated by the analysis of variance of each location that cultivar IAC-24 and inbred line 7 as the more stable to Capão Bonito, Tietê and Tatuí.
Triticum aestivum L.; grain yield; genotype x environment interaction; acid and limed soil; upland and irrigation conditions
