Age trends of heritabilities and genotype-by-environment interactions for growth traits and wood density from clonal trials of Eucalyptus grandis Hill ex Maiden

Obtaining accurate and precise genetic parameter estimates is fundamental t o determining breeding strategies, and for choosing genotypes for commercia l propagation. Results for survival and growth from seven clonal genetic te sts of Eucalyptus grandis in Colombia supported the a priori contention of sub-dividing them into three different environments for deployment and poss ibly breeding purposes. The genotype-by-environment interactions (GxE) for growth trails were moderate at six years of age in the target environment ( 5 sites representative of 95% of the E. grandis planting area for the clona l program). Therefore, it is recommended to breed and select for clones tha t perform well across the range of sites within the target environment. The clonal rankings for growth traits at the two extreme sites differed marked ly between these two distinct environments, and between each extreme enviro nment and the five sites in the target environment. Thus, the extreme envir onments require separate clonal test locations and deployment populations. Broad sense heritabilities for survival, individual tree volume and mean an nual increment (MAI) tended to increase over time for the three environment s, but the trends for height were quite different among environments. The b road sense heritabilities for mean wood density declined with age, but GxE interaction for wood density was low indicating that clonal rankings were s table among the five sites within the target environment. The estimation of genetic gains by two methods, predicted clonal values and the classical fo rmula, gave similar results and showed great potential for increasing produ ctivity in the target environment through selection of the top clones.