Evaluation of S2 Maize Lines Reproduced for Several Generations by Random Mating within Lines. II. Comparisons for Testcross Performance of Original and Advanced S2 and S8 Lines
An alternative to the use of highly inbred lines to produce maize hybrids (Zea mays L.) would be to use lines that have been only partly inbred, selected in early generation testing for their combining, ability, and maintained by randomating within lines (sib-mating). Our objective was to compare the testcross performance of original S2 lines with their maintained S2-1ine counterparts and S2-line derivatives to determine if changes have occurred in the S2 lines after several generations of sib-mating or S8 lines after further inbreeding.
We evaluated the performance of 14 original S2 lines; each S2 line maintained by using 10 and 20 plants generation −1, four lines also maintained by using mild selection for agronomically desirable traits, and their Ss line derivatives in testcrosses with two unrelated testers. Significant (P < 0.05) or highly significant (P < 0.01) differences among groups were observed for 8 of the 14 traits evaluated; grain yield differences were nonsignificant. Individual comparisons between each original S2 line and its maintained S2-1ine counterparts showed that significant differences occurred in both the positive and negative directions for many traits, thus indicatingenetic changes. Testcrosses of S2 lines maintained by using 10 plants generation −1 generally tended to be less stable than if they were maintained by using 20 plants. For grain yield, only one maintained S2 line differed significantly from its original S2-line counterpart~ whereas three Ss lines showed a significant change. Thus, maintaining S2 lines by sibmating by using 10 or 20 plants, or mild selection, had small effects on the combining ability of the individual S2 lines.
Copyright © 1989.