LAUSR

An inter-specific hybrid breeding program involving Eucalyptus urophylla (U) and Eucalyptus camaldulensis (C) was implemented in order to provide genotypes better adapted to southern China with improved growth rate, stem-straightness and wind-resistance. A trial involving 36 reciprocal crosses from six parents each of C and U that had been preselected for superior growth and stem-straightness was established at a site in Luokeng in Guangdong province. Ten, pure-species families using the hybrid parents as open-pollinated female parents were included as controls. Survival and growth traits at ages 2 and 8.3 years and stem-straightness at age 2 years were assessed. Inter-specific hybrids generally performed better than the pure species in terms of survival, growth traits and stem-straightness. Female U by male C crosses generally outperformed those involving male U and female C. Further indication of the significant reciprocal effect was supported by negative and low correlations between paired groups of full-sibs that differed only in the direction of the cross. Components of female additive genetic variance and narrow-sense heritability calculated from this estimate ($$\sigma_{\rm Af}^{2}$$σAf2 and $$h_{\rm f}^{2}$$hf2), respectively) were generally higher than those of the male $$\left( {\sigma_{\rm Am,}^{2} h_{\rm m}^{2} } \right)$$σAm,2hm2, providing evidence for maternal effects. The narrow-sense heritability (h2) estimates based on general hybridizing ability for growth traits and stem-straightness were generally low at both ages, and of low precision at 8.3 years. The ratio of hybrid additive-to-dominance variance $$\left( {{{\sigma_{\rm A}^{2} } \mathord{\left/ {\vphantom {{\sigma_{\rm A}^{2} } {\sigma_{\rm D}^{2} }}} \right. \kern-0pt} {\sigma_{\rm D}^{2} }}} \right)$$σA2σD2 was of little practical consequence at age 2 years and had further decreased by age 8.3 years. Trait-trait genetic correlations amongst hybrids were generally positive and moderate to high. Hybrid vigour, gauged by comparison with the performance of the pure species progeny was significant, though correlations between pure species and hybrid progeny-based estimates of parental performance were weak, indicating that making parental selections in pure species trials may not be a successful strategy.