Genetic analysis of inbred and topcross progeny of an elite, single-cross maize population
The utility of restriction fragment length polymorphisms (RFLPs) in detecting quantitative trait loci (QTL) in maize (Zea mays L.) populations has been established for many traits. More complex topics in quantitative trait inheritance are being explored such as heterosis, QTL-by-environment interaction, and QTL detection in more complex progeny types such as testcross progeny in maize. The main objectives of this study were to investigate the consistency of QTL detection in diverse environments in a population of F[subscript]2:3 lines of an elite, maize single-cross, detect QTL for general and specific combining effects in three testcross populations, and compare QTL among testcross progeny with QTL for the same traits in lines per se;For the lines per se, differences existed in the number of QTL that were commonly detected between two diverse environments. Over all traits, only 50% of the QTL detected in one environment were also detected in the other environment. The QTL that were detected in both environments were consistent with respect to the parent contributing to increased trait values and the size of the effect. An effective strategy to obtain the most information from both environments seems to be mapping QTL by using the mean of the trait values in the environments tested. The mean environment allowed the detection of 69 to 74% of all QTL and 71 to 81% of the QTL detected in each of the individual environments;Mapping of SCA and GCA QTL in the testcross populations was possible for all traits. No tester-by-parental allele interactions occurred for any of the QTL of these traits. For grain yield and moisture, most SCA and GCA QTL seemed to be detected randomly among testers and environments. The morphological traits had a higher proportion than grain yield and moisture, albeit not a large proportion, of regions having QTL for SCA and GCA effects in which QTL were also detected for these same traits in the lines per se. In each of these instances, the testers ranked the parental alleles the same as the alleles were ranked in the per se evaluations.