Soybean stability and adaptability in Southern and Central Brazil
Marco Antonio Rott de Oliveira, Valéria Carpentieri Pípolo, Ivan Schuster, Dorival Vicente, Marisa Dellagostin, and Edson Feliciano de Oliveira
The use of effective methodologies for the identification of more stable lines with wide adaptation is an indispensable tool in genetic improvement. Diverse methods of estimation are presently available, each one with specific features. This study evaluated the adaptability and stability of 21 soybean lines/cultivars grown in 78 environments in the southern and central regions of Brazil, between 2001 and 2003, using four methodologies. In the stability analyses, the four methods led to the same results. In the analyses of adaptability, the methods Eberhart and Russel (1966) and of Cruz et al. (1989) presented basically the same results, though the latter presented additional information regarding the adaptability of lines to favorable and unfavorable environments. The analysis by the AMMI method led to the establishment of the models AMMI3 (Central Region) and AMMI4 (southern region), making the analysis of adaptability by biplots very complex. Six stable lines were identified in the central region and two in the South.