Agronomic performance of new common bean (
Phaseolus vulgaris
L.) varieties may vary considerably in diverse agro-ecological zones, due to genotype-environment (G x E) interactions. A multi-site evaluation is, therefore, crucial in elite lines selection as it considers simultaneously both yield and stability of performance, making selection of genotypes more precise and refined. This study aimed at assessing the agronomic performance and yield stability of 24 F
1.7 small red bean lines, selected for multiple disease resistance, using molecular markers from 16 inter-racial bean populations. The study lines were evaluated in low, medium and high altitude agro-ecological conditions in Kenya. G x E effects for grain yield were highly significant (P<0.001), implying that the agronomic potential of the small red bean lines varied with agro-ecological zones. Tigoni, the high altitude (2130 m above sea level) test site had the highest grain yield (3,809 kg ha
-1). There were no significant differences between medium altitude (1820 masl) Kabete (1,100 kg ha
-1) and low altitude (1150 masl) Mwea (1,025 kg ha
-1) sites. Number of pods per plant was strongly positively correlated with seed yield (r=0.85***), and could be adopted by breeders as an indirect selection method for grain yield. The G x E interaction on grain yield was high (14.4%), implying that most of genotypes should be selected and recommended to specific environments. The high yielding line, KMA13-25-09 (3,385 kg ha
-1), was the least stable across sites. Disease severity score for target diseases was low to intermediate regardless of the genotypes and sites.