Genetic selection for cold hardiness in coastal Douglas-fir seedlings and saplings

Abstract

Genetic control of cold hardiness in two-year-old seedlings was compared with that in 7-year-old saplings of 40 open-pollinated families in each of two breeding populations (Coast and Cascade) of coastal Douglas-fir (Pseudotsuga menziesii var. menziesii (Mirb.) Franco) from western Oregon. In addition, the efficacy of bud phenology traits as predictors of cold hardiness at the two stages was explored. Fall and spring cold hardiness were assessed using artificial freeze testing. Similar genetic control of cold hardiness in seedlings and saplings is suggested by strong type-B genetic correlations (rB) between the two ages for fall and spring cold injury traits (rB[Formula: see text] 0.78) and by similar trends in individual tree heritability estimates (hi2), e.g., hi2was greater in spring (h–i2= 0.73) than in fall (h–i2= 0.36) and greater in the Coast population (h–i2= 0.69) than in the Cascade population (h–i2= 0.40) at both ages. Strong responses to direct selection are expected for spring cold hardiness at both ages and for fall cold hardiness in seedlings, even under mild selection intensities. Similar heritabilities in seedlings and saplings, and strong genetic correlations between ages for cold-hardiness traits, ensure that selection at one age will produce similar gains at the other age. Type-A genetic correlations (rA) between fall and spring cold hardiness were near zero in the Cascade population (rA= 0.08 and -0.14 at ages 2 and 7, respectively) but were moderate and negative in the Coast population (rA= -0.54 and -0.36, respectively). Bud-burst timing appears to be a suitable surrogate to artificial freeze testing for assessing spring cold hardiness in both seedlings and saplings, as is bud set timing for assessing fall cold hardiness in seedlings, but bud set timing is a poor predictor of fall cold hardiness in saplings.