We evaluated three maize ( Zea mays L.) crosses among BSCBI(R)C 5 , BSSS(R)C 5 , and BSSS(HT)C 6 populations in six environments. BSCBI(R)C 5 and BSSS(R)C 5 improved populations from reciprocal recurrent selection, and BSSS(HT)C 6 was obtained from testcross selection with a double‐cross tester. Also, we used five selected S 3 lines from each of BSCBI(R)C 5 and BSSS(R)C 5 and five selected S 3 or S 4 lines from BSSS(HT)C 5 in three sets of line ✕ line crosses among the sources. Average yields of the population crosses were essentially equal, which was unexpected for BSSS(R)C 5 ✕ BSSS(HT)C 6 Most of the variation among crosses in each set was accounted for by average line performance (general combining ability) in all traits. Average yield of line crosses in a set was significantly higher than the population cross in each set and in the BSCBI(R)C 5 ✕ BSSS(R)C 5 and BSCBI(R)C 5 ✕ BSSS (HT)C 6 sets, the best cross yielded 35% more than the population cross. Two BSCBI(R)C 5 ✕ BSSS(R)C 5 crosses yielded significantly higher than B37 ✕ Oh43, and one of these crosses was equal to B37 ✕ Oh43 for lodging resistance. None of the crosses in the other two sets was equal to B37 ✕ Oh45 for all traits, mainly because of lodging. A modified reciprocal recurrent selection procedure using an inbred tester is proposed as a more efficient method of simultaneously improving breeding populations and developing elite single crosses.
No takes yet. Share an insight, caveat, or question.
Russell et al. (1975) studied this question.