Variability of selected wood characteristics in 40 half-sib families of black spruce (Picea mariana)

Summary Based on 15-year-old black spruce (Picea mariana) trees from 40 half-sib families sampled from 9 blocks of a family test in New Brunswick, this study examined intertree and intratree variation in various wood density and ring width characteristics. Of various variance components of the intertree variation, a remarkable variance component due to family was found in wood density characteristics (viz. average wood density, average earlywood density and latewood density of the tree), and these characteristics are thus under strong genetic control (h_i² ranging from 0.60 to 0.86, and h_f² from 0.56 to 0.68). It, to a lesser extent, applies to ring width characteristics at the tree level (viz. average ring width, and average earlywood width, latewood width and latewood percent of the tree) that show a lower heritability (h? from 0.18 to 0.28, and h_f² from 0.22 to 0.36). Both block and block × family interaction contribute little to the total intertree variation encountered in 40 families from 9 blocks, while tree-to-tree variation within the family accounts for most (over 3/4) of the total intertree variation.Compared to the intertree variation (tree-to-tree variation within the family), the intratree variation in various wood characteristics studied is considerably larger in this species. It appears that most intraring wood density characteristics show a relatively smaller intertree variation but a relatively larger intratree variation as compared to ring width characteristics (except latewood width and latewood percent). Latewood width and latewood percent show the smallest intertree variation and the largest intratree variation. Between the two sources of the radial intratree variation, cambial age explains much more variation in most intraring wood density characteristics, while ring width accounts for more variation in earlywood width, latewood width and intraring density variation. This indicates that wood density of growth rings in this species is dependent more on cambial age than ring width (growth rate). Among various wood density and ring width characteristics studies, maximum (latewood) density shows the strongest response to calendar year. This characteristics is thus a useful dendroclimatic parameter in this species.I would like to thank Dr. E.K. Morgenstern and Mr. D. Simpson for their involvement in the planning of this study. Thanks are also due to G. Chauret, T. Keenam, R. Ploure, V. Steel and C. Reitlingshoefer for their technical assistance