The purpose of this research was to use five selection procedures to develop inbred lines of maize (Zea mays L.):

(1) to compare the selected lines for performance per se and in testcrosses at various plant densities, thereby evaluating the selection procedures; and

(2) to determine the effects of plant densities on the relationships among plant, ear, and grain traits of the inbred lines and between these traits and hybrid yield performance.

The source population was an Iowa Synthetic, BS1. Five groups of lines were obtained: Group 1, selected visually at a density of 59,000 plants/ha in rows spaced 102 cm; Group 2, selected visually and for one ear per plant at a density of 29,500 plants/ha in rows spaced 102 cm; Group 3, selected visually and for two ears per plant at a density of 29,500 plants/ha in rows spaced 102 cm; Group 4, selected visually at a density of 59,000 plants/ha in rows spaced 51 cm; Group 5, selected for testcross performance (double-cross tester) at 59,000 plants/ha in rows spaced 51 cm. Lines in groups 1 to 4 were selected in the S0 to S4 generations and lines in Group 5 in the S0 to S2 generations with some visual selection of the lines and plants when the testcrosses were produced. Nineteen lines in each of groups 1 to 4 and a control (BS1) were evaluated in testcrosses (double-cross tester) in eight environments, three plant densities per environment (Experiment I). Also, the lines in groups 1 to 4, 10 lines in Group 5, and the control were evaluated in testcrosses (double-cross tester) in two environments, 51-cm row width and 59,000 plants/ha (Experiment II).