A major objective of the apple (Malus domestica Borkh.) breeding program in Stanthorpe, Australia, is to develop early ripening, high-quality cultivars. The heritability and inheritance of ripening date was investigated. Regression of offspring on midparent harvest dates and estimation of best linear unbiased predictions for parents were used to demonstrate that apple harvest date is highly heritable. Predominantly, additive genetic components of variance are responsible for the variation. Despite the existence of some specific combining ability variance and some non-normal family distributions, the best strategy for a breeder to predict the harvest date of progeny is to calculate the mean harvest date of parents.

In order to investigate the effect of long term recurrent selection on the pattern of gene diversity, thirty randomly-selected individuals from the progenitors (p) and four selection cycles (C0, C3, C6 and C11) were sampled for DNA analysis from the tropical maize (Zea mays L.) breeding populations, Atherton 1 (AT1) and Atherton 2 (AT2). Fifteen polymorphic Simple Sequence Repeat markers amplified a total of 284 and 257 alleles in AT1 and AT2 populations, respectively. Reductions in the number of alleles were observed at advanced selection cycles. About 11 and 12% of the alleles in AT1 and AT2 populations respectively, were near to fixation. However, a higher number of alleles (37% in AT1 and 33% in AT2) were close to extinction. Fisher's exact test and analysis of molecular variance (AMOVA) showed significant population differentiations. Gene diversity estimates and AMOVA revealed increased genetic differentiations at the expense of loss of heterozygosity. Population differentiations were mainly due to fixation of complementary alleles at a locus in the two breeding populations. The estimates of effective population at an advanced selection cycle were close to the population size predicted by the breeding method.

The objectives of this study were to quantify the components of genetic variance and the genetic effects, and to examine the genetic relationship of inbred lines extracted from various shrunken2 (sh2) breeding populations. Ten diverse inbred lines developed from sh2 genetic background, were crossed in half diallel. Parents and their F1 hybrids were evaluated at three environments. The parents were genotyped using 20 polymorphic simple sequence repeats (SSR). Agronomic and quality traits were analysed by a mixed linear model according to additive-dominance genetic model. Genetic effects were estimated using an adjusted unbiased prediction method. Additive variance was more important than dominance variance in the expression of traits related to ear aspects (husk ratio and percentage of ear filled) and eating quality (flavour and total soluble solids). For agronomic traits, however, dominance variance was more important than additive variance. The additive genetic correlation between flavour and tenderness was strong (r = 0.84, P \ 0.01). Flavour, tenderness and kernel colour additive genetic effects were not correlated with yield related traits. Genetic distance (GD), estimated from SSR profiles on the basis of Jaccard's similarity coefficient varied from 0.10 to 0.77 with an average of 0.56. Cluster analysis classified parents according to their pedigree relationships. In most studied traits, F1 performance was not associated with GD.

Understanding the nature of complex genotype‐by‐environment‐by‐management interactions is crucial to identify risks and opportunities for increasing maize yield and profitability in rainfed production systems. The objectives of this study were to (i) define the conditions where hybrids of different maturity and plant densities are viable options in terms of improving productivity, and (ii) quantify the risk levels associated with different genotype‐by‐management combinations in relation to target environments. Responses to plant density were analysed on field experimentation with different genotypes representing early, medium and late maturity types and 2, 4 and 6 plant/m2 plant densities at three major or potential dryland maize production environments in Queensland, Australia. Agricultural Production Systems sIMulator (APSIM)‐Maize module was employed to simulate yield responses and compute the cumulative probability distribution. APSIM simulations suggested that the risk of expecting a yield level less than 2 t/ha increased up to about 17 and 27% for quick and late maturing types, respectively, when density increased to 10 plants/m2 in marginal environments such as Emerald. In relatively better environments, however the risk increased only up to 10% for late hybrids, and 7% for a quick hybrid at 10 plants/m2. In both high and low potential environments, choice of hybrids and plant densities should be based on seasonal weather forecasts to minimize risks and maximize opportunities for higher yields.

With 6 tables Abstract The objectives of this study were to evaluate the importance of heterosis for agronomic and quality traits in shrunken (sh2) sweet corn, assess the usefulness of combining ability to predict the value of parents and their crosses for further genetic improvement and examine whether genetic divergence can predict heterosis or F1 performance. Ten genetically diverse shrunken (sh2) sweet corn inbred lines were used to generate 45 F1s. F1s and parents were evaluated for agronomic and quality traits across environments. Heterosis was more important for yield‐related traits than it was for ear aspects and eating quality. Heterosis for most traits was mostly dependent on dominance genetic effects of parental lines. Parents and F1per se performance were highly correlated with general combining ability effects and mid‐parent values, respectively, for most traits. Hybrid performance for flavour and plant height was significantly but weakly related to simple sequence repeat (SSR)‐based genetic distance (GD). Phenotypic distance (PD), estimated from phenotypic traits was correlated with heterosis for total soluble solids, ear length and flavour.