The objective of this work was to evaluate the contribution of the use of classes of cultivars with and without interaction with the environment, on the quality of joint analysis of corn trials for grain productivity. Corn grain productivity data from 99 trials, distributed in 12 groups, each with the same cultivars in different environments, were used. In each group, 9 to 40 cultivars were assessed in 5 to 12 environments during three farming years. For each group, joint analysis of variance was performed and each cultivar was tested for its contribution to the interaction, forming two classes of cultivars: CI, which contribute to the interaction; and SI, which do not contribute to the interaction. For each class, a new joint analysis of variance was performed and the contribution of the cultivar to the interaction was tested. The classification of cultivars based on their contribution to the genotype x environment interaction allows performing joint analysis for each cultivar class with better accuracy when comparing means of the SI cultivars and when analyzing the interaction of the CI cultivars.
Zea mays ; selective accuracy; adaptability analysis; cultivar recommendation
