Correlation between parametric and non parametric stability measures in soybean

Abstract

The adaptability and the stability of soybean (Glycine max L.)genotypes were evaluated through the classical methodology of Eberhart and Russell and the stability was also evaluated through the non-parametric methodology of Huhn. The experiments were conducted in the completely randomized blocks with three replications and with 30 treatments (soybean genotypes) during three consecutive years. The experimental plots were constituted by four rows 0.50 m apart and with 25 plants per meter. Data were collected in the two central rows being eliminated 0.50 m in each edge. To compare the two methodologies the character grain yield was considered. Positive and significative rank correlation was observed between the regression deviations and the two non-parametric stability parameters (S_i⁽¹⁾ and S_i⁽²⁾) for the three studied characteristcs. The parameters S_i⁽¹⁾ e S_i⁽²⁾ were nearly perfectly correlated to each other.