Evaluation of three test statistics used to identify maize inbred lines with new favorable alleles not present in elite single cross

Abstract

The identification of inbred lines useful for improvement of an elite single cross hybrid is an essential part of a pedigree maize (Zea mays L.) breeding program. The objectives of this study were to identify lines that could be useful for improvement of hybrid B73 × Mo17 and to relate the values of estimators of new favorable alleles with test cross yields. Crosses of parents of hybrid B73 × Mo17 with 10 public lines from the United States (US), and 14 Maize Research Institute "Zemun Polje" proprietary lines (lines per se, and test crosses from 3 F2 populations) were evaluated at 4 locations in Yugoslavia in 1986. Significant differences in grain yield were found among lines in minimally biased estimates of favorable alleles (μG) present in a donor inbred but not present in a B73 × Mo17, in minimum upper bound (UBND) estimates and in predicted three-way performance (PTC). Twenty-one lines had a significant number of dominant favorable alleles for grain yield not present in B73 × Mo17. The highest values for all estimators of new favorable alleles were found for donor lines which belonged to different heterotic groups than the B73 and Mo17. For most of the inbreds, the (C + F) - (D + E) statistics agreed with predigree information. Simultaneous increases in grain yield and decreases in grain moisture content for B73 × Mo17 are possible with several donor inbred lines. All of the lines with a high number of new favorable alleles for grain yield not present in B73 × Mo17 had negative μD (F)-μG values for low plant height. Line N152 had the most new favorable alleles for grain yield not present in single cross B73 × Mo17. Population (N152 × Mo17) F2 had the highest difference of observed test cross means from check mean, the most test crosses with significantly higher yields than the check, and the largest estimate of number of segregating loci.