Response to selection for seed yield and nitrogen (N-2) fixation in commonbean (Phaseolus vulgaris L.)

This study examined the feasibility of recurrent selection for enhanced N-2 fixation in common bean (Phaseolus vulgaris), and sought to generate breed ing lines with the high N-2-fixing ability. Following their selection as su perior in N-2 fixing ability, the landrace Puebla 152, and the CIAT breedin g lines BAT 271 and RIZ were intercrossed, and 17 F-2.3 lines derived from the progeny used to initiate two cycles of recurrent selection. Each cycle included crossing, selfing, multiplication and field testing phases, and oc cupied one year. Field testing was undertaken on an irrigated, leached, loa my sand with 9 to 14 mg total N kg(-1). Plants were inoculated with Rhizobi um etli UMR 1632, with 17 of 490 F-2.3 derived lines per cycle selected for seed yield and shoot dry weight under conditions of N limitation. Selectio n progress was evaluated by comparing the C-0, C-1, and C-2 parents, and se lections from the C-2 population, for biological and seed yield, and N accu mulation under the same N-limited field conditions. The C-2 parents yielded an average of 11% more than the C-0 parents, and were 15% and 13% better i n shoot dry weight and shoot N per plant, respectively. Several lines perfo rmed significantly better than any parent or check cultivar under the low N field conditions. Among cross-differences for seed yield and shoot dry wei ght occurred in both the C-1 and C-2 populations, with GCA accounting for m ore of the variation than SCA in both populations. Not all the parents were well adapted to Minnesota, and several Cycle 1 F-2.3 lines were late to fl ower and mature, resulting in a negative correlation between these traits a nd seed yield and seed N. Selection emphasis on lines maturing in 98 days o r less reversed this relationship in the C-2 parental group. We conclude th at selection based on both plant and seed parameters is essential. (C) 1999 Elsevier Science B.V. All rights reserved.