Search

Breeders ought to consider the confounding effects of the environment and genotype x environment (G x E) interaction on response to early generation selection. To meet this requirement, honeycomb breeding was performed at a low density within two dry bean populations (Phaseolus vulgaris L.) under typical open-field conditions and in an adjacent greenhouse. Nineteen progeny lines were formed through selection of nine and ten of the highest-yielding plants in the greenhouse and the field, respectively. Honeycomb progeny testing at the low density in the two distinct environments showed up to 75% improvement in plant-yield potential. Under farming-density conditions in five environments, six of the lines outyielded the respective original population by 12 to 38% and exhibited the greatest stability according to the genotype and genotype x environment (GGE) biplot model. At low density, the greenhouse evaluation demonstrated less acquired variance than the field evaluation and was especially useful for selection and progeny evaluation. Three of the six outstanding lines originated from the greenhouse. Honeycomb progeny estimation on a single-plant yield basis in the greenhouse, rather than in the field, gave a better prediction of yield potential on an area basis. The results showed that honeycomb breeding performed in two environments to address the G x E interaction may be successful for developing varieties that exhibit both high and stable productivity. doi: 10.2135/cropsci2009.08.0459