Comparison of Three Selection Procedures for Yield in Soybeans 1

Four segregating soybean [ Glycine max (L.) Merr.] populations were used to evaluate the effectiveness and efficiency of three selection procedures in isolating highyielding lines; early‐generation yield testing (EGT), pedigree selection (PS), and single‐seed‐descent selection (SSD). Lines were obtained from the PS procedure in the F 4 generation and from the EGT and SSD procedures in the F 5 generation. These lines were yield‐tested and selected in either the F 6 or F 7 generation in each procedure. In the F 8 generation of each cross the selected lines in each procedure were compared. The means of all selected lines, the means of the five highest‐yielding lines, and the highest‐yielding line from each population showed no consistent differences in procedures. The lines from the EGT procedure were consistently later in maturity than the lines from the other two procedures. This was attributed to the yield‐testing in unbordered plots in the early generations by the EGT procedure. The SSD procedure emerged as the most efficient procedure, because it required less selection effort than the EGT and PS procedures, allows a rapid advance of the early‐generation segregating populations, and did not use expensive yield‐testing until later generations, when yield‐testing is more efficient.