Recurrent selection provides an established methodology for host improvement which should be applicable to increasing N₂ fixation. This study was conducted to evaluate response to three cycles of phenotypic selection for fresh plant weight, nodule number and dry weight, and acetylene reduction when individual peanut (Arachis hypogaea L.) plants were evaluated in the greenhouse using a mixture of four Bradyrhizobium strains as inoculum. The F₂ generation derived from a single cross, Florigiant x CES 101, constituted the base population. In each of three cycles the 20 superior individuals were selected and randomly intermated to generate the subsequent cycle for selection. The 20 selections from each cycle were bulked and evaluated with the original parents in two greenhouse trials to evaluate response to selection. No variability for fresh plant weight, nodule number or dry weight, or acetylene reduction was detected among cycles. Realized heritability estimates were 0.01 ± 0.02, 0.05 ± 0.03, −0.06 ± 0.06, and 0.31 ± 0.32 for fresh plant weight, nodule number, nodule dry weight, and acetylene reduction, respectively. Agronomic traits of the same entries were evaluated in one field study. No variability for these traits was detected among cycles. Insufficient control of environmental variation was suggested as a possible cause of lack of response to selection.