Relative performance of testers to identify elite lines of corn (Zea mays L.)
Seven experiments were conducted at seven contrasting environments in Guatemala to determine the relative performance of seven testers in ranking a set of 21 lines of corn (Zea mays L.) of different origin and different levels of inbreeding. A randomized complete block design with a split-plot arrangement and two replications was used to evaluate the 147 testcrosses (21 lines x seven testers), the seven testers per se, and a set of seven hybrid checks. The 21 lines were assigned to the whole plots while the seven testers were assigned to the sub plots. The seven testers per se (five single-crosses, one synthetic, and one S[subscript]3 inbred) and the seven checks were assigned to two whole plots, but they were analyzed separately from the 147 testcrosses;The combined analysis of the 147 testcrosses for yield (t/ha) showed highly significant differences (P ≤ 0.01) for the sources of variation of environments, lines, testers, and the interactions of lines and testers with environments. Estimates of coefficient of concordance (W) and Pearson correlation coefficients (r) were highly significant for the ranking of the 21 lines by the seven testers. Significant differences among testers occurred for days to silk, plant and ear height, and husk score. Three testers (No. 4, 5, and 6) had positive estimates of general combining ability (GCA) across the 21 lines and eight lines had positive estimates of general combining ability across the seven testers;The best 10 testcrosses across the seven environments had significantly greater yield than the best check ICTA HB-85 (0.829 t/ha or greater). Line No. 13 and tester No. 5 were involved in five of the best 10 testcrosses. Tester No. 4 seems to include acceptable features to be considered as the convenient tester for the hybrid program under consideration in Guatemala. Three-way testcrosses significantly better than the best check ICTA HB-85 were identified and should be extensively evaluated as potential new hybrids for release in Guatemala. Selection pressure should be emphasized to screen lines for southern corn rust (Puccinia polysora Underw) and northern corn leaf blight (Exserohilium turcicum Pass. = Helminthosporium turcicum Pass.) because of the highly significant correlation found in this study between these two diseases and yield.