Abstract

Maize plays an important role in the national and global economy, continuously increasing its total production due to advances in technology and access to new land areas. Thus, new sources of germplasm are fundamental to generate cultivars more adapted to the diversity of environments and planting times. The objective of this study was to evaluate 36 populations of maize in three environments, aiming to identify the existence of genotype-by-environment interaction, classify populations based on adaptability and stability using the methods of regression and mixed models, indicate the best populations, and compare the two methodologies. The environments evaluated were: E1-second crop (safrinha) season of 2016 in an experimental area of latosol, with incidence of water stress; E2-crop season 2016/2017 in sandy soil, in family farm area; and E3-crop season 2016/2017 in an experimental area of latosol, no incidence of water stress. Grain yield was evaluated, adaptability and stability analysis was performed. Population 36 achieved high productivity, adaptability and general stability in three tested environments. Both methodologies showed similar results regarding adaptability and stability of some populations in three environments, but mixed models were more suitable for providing better selective accuracy.

This work is licensed under a Creative Commons Attribution 4.0 License.