Submitted to: Journal of Agriculture of the University of Puerto Rico
Publication Type: Peer reviewed journal
Publication Acceptance Date: 8/15/2003
Publication Date: 1/20/2004
Citation: Smith, J.R. 2004. Selection protocols for increased yield and stress tolerance in common bean. Journal of Agriculture of the University of Puerto Rico. v. 88(1-2):27-43. Interpretive Summary: Soil-pathogenic fungi can cause severe losses to common bean production throughout the world and soil compaction can intensify those losses. New bean varieties are needed which can withstand these soil stresses, but better methods are needed to be able to develop them. Controlled field environments, with innovative statistical procedures, were used to identify ybean genotypes differing in their yield potential and yield stability unde soil stress. The new methodology and the identification of tolerant beans will allow scientists to develop, identify, and release to farmers bean varieties with improved yield performance under soil stress.
Technical Abstract: Superior common bean (Phaseolus vulgaris L.) cultivars are needed that perform well under both stressful and non-stressful soil conditions. Selection protocols are needed to assess yield potential across diverse environments, while utilizing only limited resources. The objective of this study was to compare the use of geometric means, deviations from regression, and arithmetic means to evaluate diverse bean lines for yield potential and resistance to yield reduction across multiple environments in short-row (1 m) plots. Seed yield was measured for forty-eight common bean lines grown in four field trials in Puerto Rico. Two trials were stressed with soil compaction, saturated soil, and a bean monoculture, whereas the other two trials (non-stress) were not. Based on a 10% selection intensity, geometric means and arithmetic means were equally effective in differentiating lines for yield potential across environments. Stress treatment affected which lines were selected. Hence, the best evaluation of lines was made utilizing both stressed and non-stressed data. Deviations from regression were effective in identifying lines with resistance to yield reduction, but not necessarily in identifying stable lines with yield potential. Lines G21212, MUS PM31, and T-39 had both the highest yield potential and the highest resistance to yield reduction based on combined means (arithmetic or geometric) and deviations from regression, respectively.