Evaluation of regression models and variance measures as stability parameters of some soybean genotypes [electronic resource].

Includes references.

The presence of genotype x environment (G x E) interaction is a major concern to plant breeders, since large interaction can reduce gains from selection and complicate identification of superior genotypes. Fifteen soybean genotypes were grown in a randomized complete block design, with three replications, in five locations (Etay Elbarood, Gemmeiza, Sakha, Sids and Mallawy) through 2010 and 2011 summer seasons. The objectives were to assess the yield performance, determine the magnitude of(G x E) interaction and investigate the stability of the assessed genotypes, using twelve stability statistics, derived from two types of statistical approaches (regression and variance analyses). Also, Spearman rank correlation coefficient, principal components analysis and biplot graph were applied to obtain a good understanding of the interrelationship and overlapping among the used stability statistics. Results showed highly significant mean squares for genotypes, environments and (G x E) interaction, indicating that the tested genotypes exhibited different responses to environmental conditions. Also, the terms of predictable (linear) and unpredictable (non- linear) interaction components were highly significant, which confirmed that the tested soybean genotypes considerably differed in their relative stability. The greatest seed yield was produced by Giza 111, followed by H2L12, H30, DR1O1, H117, Giza21, H32 and H15L5 genotypes that surpassed the grand mean over environments. It is evident that the genotype, Giza 111, in addition to its high seed yield, was the most stable because it met the assumptions of a stable genotype, as described by the stability models of Eberhart & Russell (1966), Tai (1971), Francis and Kannenberg (1978), Kang and Magari ( 1995) and Sharaan and Ghallab (2001).

Summary in Arabic.

1