Quantitative genetics of spring and fall cold hardiness in seedlings from two Oregon populations of coastal Douglas-fir

Genetics of fall and spring cold hardiness were investigated in two western Oregon breeding populations (Coast and Cascade mountains) of Douglas-fir ( Pseudotsuga menziesii var. menziesii (Mirb.) France). Seedlings from 40 ope n-pollinated families from each population were grown in raised nursery bed s and subjected to two soil-moisture regimes (well-watered and mild drought ) to evaluate the influence of summer drought on ranking of families for co ld hardiness. Artificial freeze testing (AFT) of detached shoots, followed by visual scoring of injury, was used to evaluate needle, stem and bud cold hardiness on three dates in the fall (September, October and November) aft er the second growing season, and once in the following spring (March). The Cascade population suffered significantly less cold injury than the Coa st population in fall AFT. However, in spring AFT the Cascade population wa s less cold hardy, although population differences were seldom significant. Families within both breeding zones varied significantly in cold hardiness , with mean estimates of individual heritabilities greater in spring ((h) o ver bar (2)(i) = 0.57) than fall ((h) over bar (2)(i) = 0.37), greater in t he Coast ((h) over bar (2)(i) = 0.52) than in the Cascade ((h) over bar (2) (i) = 0.42) population, and greater in the wet ((h) over bar (-2)(i) = 0.54 ) than in the dry moisture regime ((h) over bar (-2)(i) = 0.40) (fall means based on October tests). A single Lest date seems adequate to assess fall cold hardiness, because estimated genetic correlations for cold injury betw een fall test dates were strong ((r) over bar (A) = 0.80). Genetic correlat ions between spring and fall cold injury, however, were moderately negative ((r) over bar (B) = -0.66 and -0.21, Coast and Cascade, respectively), ind icating that cold hardiness needs to be managed as two traits (i.e. fall an d spring cold hardiness). Selection for cold hardiness based on a single sh oot tissue is expected to increase cold hardiness in the other tissues as w ell, because genetic correlations between tissues in cold injury were moder ately-to-strongly positive in both fall ((r) over bar (B)= 0.67) and spring ((r) over bar (B) = 0.84). Seedlings grown under summer drought incurred s ignificantly less cold injury in the fall than those that were well-watered ; nevertheless, strong genetic correlations in fall cold injury between moi sture regimes ((r) over bar (B) = 0.91) indicate that summer moisture condi tions had little influence on family rankings for fall cold hardiness. Corr elations of injury resulting from a natural frost event in November of the first year with injury from AFT in the fall of the second year ((B) over ba r (A) = 0.72 and 0.78 for needle and bud injury, respectively) confirmed th at AFT reliably predicts cold hardiness to natural frost events. (C) 2001 E lsevier Science B.V. All rights reserved.