Genetic parameter estimates for interspecific Eucalyptus hybrids and implications for hybrid breeding strategy

Abstract

The Forestry Commission of Zimbabwe initiated a hybrid breeding program of Eucalyptus grandis with E. tereticornis (G × T) and E. camaldulensis (G × C) in order to provide suitable genotypes for planting in areas that are marginal for E. grandis in terms of drought and frost. A total of seven hybrid trials were established at four sites, representative of low to medium rainfall. The female parents were from E. grandis preselected for superior volume production. No pure species were included in the trials to act as controls and allow for estimation of hybrid vigour. Survival, growth and stem straightness were assessed at 18 and 43 months from planting. Survival was generally above 80% for all hybrids across sites. G × C hybrids performed better than G × T hybrids, showing a 20% difference in mean height at 43 months. Female and male variances were generally insignificant (P > 0.05) and in some cases, the estimates were 0. This was not unexpected, as the parents for the hybrids were preselected for superior height, diameter at breast height (DBH) and volume production as pure species. Ratio of dominance to phenotypic variance (\( \hat{d}^{2} \)) was significant for all traits for all hybrids (P < 0.05). Trait–trait dominance correlations (rD) at 43 months between height and DBH were generally large and positive, above 0.6. Type-B dominance correlation (rBD) suggested evidence of genotype-by-environment interaction (G × E), but one high elevation site contributed most to the observed G × E. If these results are confirmed in larger mating designs with progeny established on multiple sites, it seems that the most appropriate breeding strategy for E. grandis × E. tereticornis and E. grandis × E. camaldulensis hybrids may be one that exploits both additive and dominance genetic variance, such as the reciprocal recurrent selection (RRS) scheme or RRS with forward selection.