Genetic Improvement of Timber Yield and Wood Quality in Eucalyptus Grandis (Hill) Maiden

SYNOPSIS Results of a partial diallel progeny test of Eucalyptus grandis assessed at ages 5, 36, 60 and 105 months are presented. Large genetic variations among full-sib families were found for all traits studied, e.g. the mean tree volume per family at 105 months varied from 0,226 m3 to 1,325 m3. Estimates of additive genetic variance (σ2 a) for volume production was slightly less than dominance variance (σ2 d) but for stem and crown form, and for taper σ2 a clearly exceeded σ2 d. Estimates of individual tree heritabilities were moderately high, varying from 0,28 to 0,54. Reciprocal effects were not significant. Genetic values, breeding values and specific combining ability values indicate that the relatively high level of additive genetic variation can be utilised for predicting which parent will produce the best offspring. Such offspring could be particularly useful for vegetative propagation. Genetic correlations are favourable among most growth traits, indicating that simultaneous improvements are possible even if selection is practised on one trait only. Strong age-age and trait-trait correlations imply that assessment procedures on E. grandis could be simplified substantially and that a strategy with rapid generation turnover could be adopted.