QUANTITATIVE GENETICS OF GROWTH AND DEVELOPMENT IN POPULUS .2. THE PARTITIONING OF GENOTYPE X ENVIRONMENT INTERACTION IN STEM GROWTH

Two interspecific hybrid pedigrees of Populus trichocarpa and P. delto ides, each containing parental, F-1 and F-2 generations, and planted i n two contrasting environments in northern Oregon, served to estimate the interactions of genotypes x environments for stem growth during th e first two years of growth. All three generations showed greater mean growth in the warm, high-radiation, and well-watered regime of the in terior (Boardman) than in the cooler coastal conditions of Clatskanie. In the two F-2 families, all stem traits displayed significant genoty pe x environment interactions which accounted for approximate to 10 pe r cent of the total phenotypic variance. Genetic correlations across e nvironments were larger for basal area and volume growth than for stem height and proportion. Genotype x environment interaction variance wa s further partitioned into two components attributable to the heteroge neity of genetic variance between environments and the lack of genetic correlation across environments. For radial and volume growth, a larg e amount of genotype x environment interaction was associated with the heterogeneity of genetic variance across environments. For stem heigh t and proportion, lack of genetic correlation across environments expl ained more of the genotype x environment interaction. The Spearman ran k correlations of phenotypic means between the two environments were m oderately high but significantly less than unity in most cases. Much o f the growth difference in the two F-2 families had a significant gene tic component, with broad-sense heritabilities at each location rangin g from 0.70-0.90. The implications for breeding and selection of super ior poplars for these environments are discussed.