Field performance, C effects, and their relationship to initial rooting ability for western hemlock clones

Rooted cuttings from 30 clones of western hemlock (Tsugaheterophylla (Raf.) Sarg.) were established in a field trial in northwest Oregon. A secondary cloning approach was used to partition C effects variance from genetic variance. Clonal variance in total height was significant at 1 (HT1) and 5 (HT5) years and represented 9.2% and 10.6% of the total variance at the two ages, respectively. C effects variance (among primary ramets within clones) was nonsignificant and represented 2.4% and 0% of the total variance at the two ages, respectively. Broad-sense heritabilities of HT1 and HT5 were relatively low, being 0.10 and 0.13, respectively. When calculated on a clone-mean basis, the broad-sense heritabilities increased substantially, reaching 0.53 and 0.61 for HT1 and HT5, respectively. Phenotypic correlations paired between the rooting traits and either HT1 or HT5 were low and not significant. Genetic correlations were 0.42 and 0.29 between an indirect measure of initial root system volume (VOL) and HT1 or HT5, respectively. The genetic correlations between rooting ability and HT1 or HT5 were 0.61 and 0.69, respectively. No genetic correlation existed between length of the longest initial root and field height. Environmental correlations between initial rooting traits and field height were essentially zero. Of the three initial rooting traits tested via regression analysis, VOL served as the only significant predictor or either HT1 or HT5, although the R2 values were less than 0.03, hence, the regression equations were of no practical value. Based on these results, clonal selection of western hemlock for increased early height growth should be aided by indirect selection for clonal rooting ability via a multiple trait selection.