Allozyme marker loci associated with favorable alleles for grain yield in maize.

The evaluation of germplasm to identify its potential as a source of new favorable alleles is a time-consuming phase of maize hybrid breeding programs. The objective of this paper was to study the relationship between allozyme diversity and quantitative estimators of the relative number of new favorable alleles for grain yield, present in donor lines but not present in the elite hybrid. Twenty-two maize inbred lines representing heterotic groups from the United States (US) and Yugoslavia (YU) were used as donors to estimate the presence of new favorable alleles for grain yield improvement for the hybrid B73 x Mo17. In a second experiment, a 15-line diallel was grown, and 13 single crosses differing in allozyme relatedness measure (ARM) and heterotic grouping were considered as targets to be improved by the remaining 13 lines. Minimally biased estimates of new favorable alleles for grain yield (μG) and ARM values were made for all donor lines within each target hybrid. Donor lines were grouped in four allozyme-pedigree classes for each target hybrid to compare the effect of allozyme diversity with pedigree diversity. Pedigree dissimilarities had significant effects on μG estimates. Dissimilar pedigree classes had higher μG estimates than similar pedigree classes. Allozyme differences between donor inbred lines and target hybrids had inconsistent effects on μG estimates. Significant differences in μG estimates among allozyme classes were found for 31% of the target hybrids. Classes with similar allozymes had higher μG estimates more frequently than classes with disimilar allozymes. Correlation coefficients between μG estimates and ARM values were low and not significant for 12 of the 14 target hybrids.