Genetic variation of growth and log end-splitting in second-generation Eucalyptus grandis in Sri Lanka

Abstract

ABSTRACTEucalyptus grandis is one of the main commercial forest plantation species in the highlands of Sri Lanka. An E. grandis genetic improvement program was initiated in 1995 with the establishment of a first-generation provenance-family trial cum seedling seed orchard (SSO), and several seed production areas (SPAs) using seeds from natural stands in Australia. In 2007 a second generation was initiated, comprising a seed-source/family trial with 132 families from plus-trees selected in Sri Lanka: 72 families from the first-generation SSO, 28 families from two of the first-generation SPAs, and 32 families from plus-trees selected in local plantations. Diameter at breast height (DBH) and tree height measurements were taken at ages 2, 5 and 7 years. Trees from a first thinning conducted just after the 5-year assessment were used for assessment of log end-splitting. Significant differences between seed sources and families within seed sources were observed for individual tree volume but not for log end-splitting. The seed source with the highest 7-year tree volume was the SSO, and the second and third best seed sources were the SPAs. Heritabilities for individual tree volume ranged from 0.04±0.05 at age 2 years (all trial families) up to 0.27±0.12 at age 7 years (subset of the 72 SSO families). Heritabilities for end-splitting were close to zero across all trial families and 0.16±0.15 for the subset of 72 SSO families. Phenotypic and genetic correlations between tree volumes across different ages were positive (favourable) with genetic correlations ranging from 0.39±0.38 to 1.00±0.40. The phenotypic and genetic correlations between individual tree volumes at ages 2 and 5 years with end-splitting at age 5 years were generally positive (unfavourable), with the former ranging up to 0.70 and the latter ranging up to 0.90±0.95. This correlation between tree volume and end-splitting could make simultaneous improvement of these traits difficult. Options for further tree improvement work with E. grandis in Sri Lanka are discussed.