Matthew G. Hamilton

Genetic variation in Eucalyptus nitens pulpwood and wood shrinkage traits

Eucalyptus nitens plantations are generally established for pulpwood production but an increasing area is being managed for solid wood. Genetic variation in, and correlations among, three Kraft pulpwood traits (diameter at breast height, basic density and near-infrared-predicted cellulose content) and three 12-mm wood-core shrinkage traits (recoverable collapse, net shrinkage and gross shrinkage) were examined, utilising data from two 9-year-old first-generation progeny trials in Tasmania. These trials contained approximately 400 open-pollinated families (over 100 of which were sampled for wood properties) representing three central-Victorian E. nitens races. Significant genetic variation at the race and/or within-race level was identified in all traits. Within races, relative levels of additive genetic variation were higher for shrinkage traits, although narrow-sense heritabilities were lower and the expression of genetic variation less stable across sites than for other wood property traits. Heterogeneous intertrait genetic correlations were identified across sites between growth and some wood property traits. However, where significant, genetic correlations indicated that within-race selection for growth would adversely affect core basic density and all core shrinkage traits. Furthermore, results based on cores suggested that within-race selection for higher basic density would favourably impact on cellulose content and collapse but selection for either higher basic density or cellulose content would adversely affect net shrinkage. Most within-race genetic variation in gross shrinkage appeared to be due to genetic variation in collapse. The implications of these results for sawn timber breeding will depend on the strength of genetic correlations between core traits and rotation-age objective traits and objective trait economic weights.

A latitudinal cline in disease resistance of a host tree

The possible drivers and implications of an observed latitudinal cline in disease resistance of a host tree were examined. Mycosphaerella leaf disease (MLD) damage, caused by Teratosphaeria species, was assessed in five Eucalyptus globulus (Tasmanian blue gum) common garden trials containing open-pollinated progeny from 13 native-forest populations. Significant population and family within population variation in MLD resistance was detected, which was relatively stable across different combinations of trial sites, ages, seasons and epidemics. A distinct genetic-based latitudinal cline in MLD damage among host populations was evident. Two lines of evidence argue that the observed genetic-based latitudinal trend was the result of direct pathogen-imposed selection for MLD resistance. First, MLD damage was positively associated with temperature and negatively associated with a prediction of disease risk in the native environment of these populations; and, second, the quantitative inbreeding coefficient (QST) significantly exceeded neutral marker FST at the trial that exhibited the greatest MLD damage, suggesting that diversifying selection contributed to differentiation in MLD resistance among populations. This study highlights the potential for spatial variation in pathogen risk to drive adaptive differentiation across the geographic range of a foundation host tree species.

Stiffness and checking of Eucalyptus nitens sawn boards: genetic variation and potential for genetic improvement

A trial was undertaken to assess the extent to which variation in sawn-board quality traits of plantation-grown Eucalyptus nitens is under genetic control and amenable to genetic improvement. Five hundred and sixty trees from 129 families and three central Victorian races were sampled from an open-pollinated progeny trial in Tasmania, Australia. Acoustic wave velocity (AWV) was assessed on standing trees and sawlogs. Wedges from disks extracted from sawlogs were assessed for basic density and checking. Processed boards from 496 of the trees were assessed for board stiffness (static modulus of elasticity, MOE), and internal and surface checking. Genetic differences among races were significant for AWV and MOE traits. The Southern race had the highest mean values for these traits. Significant additive genetic variation within races was observed in all traits, demonstrating that the quality of plantation-grown E. nitens boards could be improved through breeding. Estimated narrow-sense heritabilities were 0.85 for standing-tree AWV, 0.71 for log AWV, 0.37 for board MOE, and ranged from 0.20 to 0.52 for checking traits. A strongly positive genetic correlation (rg = 1.05) was observed between standing-tree AWV and board MOE, indicating that AWV could be used as a selection trait to improve E. nitens board stiffness. The genetic correlation between basic density and board MOE was also positive (rg = 0.62). However, a significant and adverse genetic correlation (rg = 0.61) was identified between basic density and surface check length. Wood stiffness and checking traits were more-or-less genetically independent, and genetic correlations between surface and internal checking were positive but only moderate (rg = 0.48–0.52).

Achievements in forest tree improvement in Australia and New Zealand 9. Genetic improvement of Eucalyptus nitens in Australia

Summary Eucalyptus nitens is the second most widely planted eucalypt species in Australia. The species is principally grown for the production of pulpwood but substantial areas are also managed for solid-wood production. The first large-scale E. nitens progeny trials were established in the 1970s and up to two cycles of breeding have since been completed. Estimates of genetic gains achieved through breeding are not routinely published, but numerous genetic gain trials have been established. Advances in the understanding of E. nitens genetic architecture and reproductive biology have been integrated into operational breeding and deployment programs. Despite extensive research into alternative deployment strategies, improved E. nitens genotypes are almost universally deployed as seedlings derived from open-pollinated seed-orchards.

Chemical Variation in a Dominant Tree Species: Population Divergence, Selection and Genetic Stability across Environments

Understanding among and within population genetic variation of ecologically important plant traits provides insight into the potential evolutionary processes affecting those traits. The strength and consistency of selection driving variability in traits would be affected by plasticity in differences among genotypes across environments (G×E). We investigated population divergence, selection and environmental plasticity of foliar plant secondary metabolites (PSMs) in a dominant tree species, Eucalyptus globulus. Using two common garden trials we examined variation in PSMs at multiple genetic scales; among 12 populations covering the full geographic range of the species and among up to 60 families within populations. Significant genetic variation in the expression of many PSMs resides both among and within populations of E. globulus with moderate (e.g., sideroxylonal A h2op = 0.24) to high (e.g., macrocarpal G h2op = 0.48) narrow sense heritabilities and high coefficients of additive genetic variation estimated for some compounds. A comparison of Qst and Fst estimates suggest that variability in some of these traits may be due to selection. Importantly, there was no genetic by environment interaction in the expression of any of the quantitative chemical traits despite often significant site effects. These results provide evidence that natural selection has contributed to population divergence in PSMs in E. globulus, and identifies the formylated phloroglucinol compounds (particularly sideroxylonal) and a dominant oil, 1,8-cineole, as candidates for traits whose genetic architecture has been shaped by divergent selection. Additionally, as the genetic differences in these PSMs that influence community phenotypes is stable across environments, the role of plant genotype in structuring communities is strengthened and these genotypic differences may be relatively stable under global environmental changes.

The genetic variation in the timing of heteroblastic transition in Eucalyptus globulus is stable across environments

Eucalyptus globulus is one of the best known examples of a heteroblastic plant. It exhibits a dramatic phase change from distinctive juvenile to adult leaves, but the timing of this transition varies markedly. We examined the genetic variation in the timing of heteroblastic transition using five large open-pollinated progeny trials established in north-western Tasmania. We used univariate and multi-variate mixed models to analyse data on the presence/absence of adult or intermediate foliage at age 2 years from a total of 14 860 trees across five trials, as well as height to heteroblastic phase change from one trial. Up to 566 families and 15 geographic subraces of E. globulus were represented in the trials. The timing of the heteroblastic transition was genetically variable and under strong genetic control at the subrace and within-subrace level, with single-trial narrow-sense heritability estimates for the binary trait averaging 0.50 (range 0.44–0.65). The degree of quantitative trait differentiation in the timing of heteroblastic transition among subraces, as measured by QST, exceeded the published level of neutral molecular marker (FST) differentiation in all cases, arguing that diversifying selection has contributed to shaping broad-scale patterns of genetic differentiation. Most inter-trial genetic correlations were close to one at the subrace and additive genetic levels, indicating that the genetic variation in this important developmental change is expressed in a stable manner and that genotype-by-environment interaction is minimal across the environments studied.

Patterns of longitudinal within-tree variation in pulpwood and solidwood traits differ among Eucalyptus globulus genotypes

Wood discs were sampled from 6 heights up the stem of 248 trees representing 10 subraces and 116 families grown in an E. globulus base-population progeny trial. The lower stem had the least favourable wood properties for kraft pulpwood and most solidwood applications: bark was thickest, basic density was lowest and kino, decay and shrinkage traits were greatest at or below 12% of tree height. Significant genetic differences at the subrace level were revealed in diameter, bark thickness, basic density, decay and gross shrinkage and at the family within subrace level in diameter, basic density and decay. However, subrace-by-height-category interactions in bark thickness, basic density, decay and gross shrinkage indicated that differences among subraces were dependent on height in these traits. Examination of longitudinal trends revealed some evidence that the zone of thick basal bark extended further up the stem in thicker-barked subraces and that the Southern Tasmania subrace might be less effective than other subraces in restricting the longitudinal spread of decay after infection.RésuméDes disques de bois ont été prélevés à 6 hauteurs différentes dans 248 arbres représentant 10 provenances et 116 familles d’un dispositif de provenance-descendance d’E. globulus. La partie inférieure des troncs présente les propriétés du bois les moins bonnes pour la pâte Kraft et la plupart des utilisations du bois massif : l’écorce est plus épaisse, l’infradensité plus faible tandis que le lino, la dégradation biologique et les retraits sont plus importants jusqu’ à 12 % de la hauteur des tiges. Des différences génétiques significatives ont été établies au niveau provenance pour le diamètre, l’épaisseur d’écorce, l’infradensité et la dégradation biologique et au niveau famille dans une provenance pour l’infradensité et la dégradation biologique. Cependant, pour ces propriétés, les interactions provenance par catégorie de hauteur, pour l’épaisseur d’écorce, l’infradensité, la dégradation biologique et le retrait total, indiquent que les différences entre provenances dépendent de la hauteur. L’analyse des variations longitudinales fait apparaître que la longueur de la bille de pied ayant une écorce plus épaisse est plus importante pour les provenances présentant des écorces épaisses et que les provenances de sud de la Tasmanie pourraient être moins efficaces que les autres pour limiter la diffusion des pourritures après infection.

Genetic and environmental variation in heartwood colour of Australian blackwood (Acacia melanoxylon R.Br.)

Abstract Australian blackwood (Acacia melanoxylon R.Br.) is a high-quality appearance-grade timber species native to eastern Australia. It is characterised by dark-coloured heartwood, with colour varying from pale straw to red-brown and walnut brown, and pale cream-coloured sapwood. This wide range in heartwood colour is expressed as between- and within-tree variation, and is regarded as a problem in markets where colour consistency is important. To understand the genetic and environmental control of heartwood colour, multiple colour measurements were taken from stem cores from 16 open-pollinated family collections of blackwood planted across three 19-year-old progeny trials. Data was analysed with a mixed model with within-tree variation modelled using a three-node cubic spline. Significant genetic, environmental and genetic-by-environment interactions were detected, not only in the variation in between-tree mean heartwood colour but also in the pattern of within-tree heartwood colour variation. In general, heartwood colour became darker, less yellow and more red towards the sapwood/heartwood boundary. To control the between- and within-tree heartwood colour variation in blackwood plantations, care is needed to not only select genotypes with the desired wood colour and colour variation but also site characteristics that allow expression of that colour.

Genetic correlations between pulpwood and solid-wood selection and objective traits in Eucalyptus globulus

Abstract• If selective breeding is to be successful, significant genetic variation must be present in the traits targeted for improvement (i.e. “objective traits”).• This study aimed to quantify genetic variation in Eucalyptus globulus pulpwood and sawn-timber objective traits (rotation-age whole-tree volume, survival, whole-tree basic density, sawn-board Janka hardness and sawn-board internal checking) and estimate additive genetic correlations between these and inexpensively-assessed “selection traits”.• Significant genetic variation was identified in all objective traits at the subrace and/or family within subrace level.• Selection-age diameter at breast height (1.3 m, DBH) was strongly genetically correlated with rotation-age volume (0.78) and survival (0.82). Subrace and additive genetic correlations of selectionage Pilodyn penetration with rotation-age 12-×-12-mm-sample basic density (−0.70 and −0.75 respectively) and whole-tree basic density (−0.83 and −0.91 respectively) were also strong.• No significant subrace or additive genetic correlation between wood-sample gross shrinkage and sawn-board internal checking was detected. However, subrace and additive genetic correlations of sawn-board Janka hardness with Pilodyn penetration (−0.75 and −0.58 respectively) and sample gross shrinkage (−0.77 and −0.73 respectively) were significantly different from zero.• These findings suggest that genetic improvement of the examined objective traits is possible through selective breeding, although none of the assessed selection traits were strongly correlated with internal checking.Résumé• Pour qu’un programme d’amélioration soit couronné de succès, une variabilité génétique significative doit être présente dans les traits cibles de la sélection.• Cette étude visait à quantifier les variations génétiques des traits cibles du bois à pâte et du bois de sciage d’Eucalyptus globulus (âge de révolution, volume de l’arbre entier, survie, infradensité de l’arbre entier, indice de dureté de Janka des sciages et fissures internes des sciages) et à estimer les corrélations génétiques additives entre ces traits et une sélection de caractéristiques évaluées à peu de frais.• Des variations génétiques significatives ont été détectées pour tous les traits cibles au niveau de la provenance et/ou de la famille dans la provenance.• Le diamètre à hauteur de poitrine (1,3 m, DBH) présentait une forte corrélation génétique avec le volume en fin de révolution (0,78) et la survie (0,82). Les effets de provenance et les corrélations génétiques additives ont été forts entre le coefficient de pénétration de Pilodyn et l’infradensité d’un échantillon de 12-×-12-mm (−0,70 et −0,75 respectivement) et l’infradensité des arbres entiers (−0,83 et −0,91 respectivement).• Aucun effet provenance et aucune corrélation génétique additive n’ont été détectés entre le coefficient de retrait du bois de l’échantillon brut et les fissures internes des sciages. Toutefois, des effets provenance et des corrélations génétiques additives dont été détectés entre l’indice de dureté Janka des sciages et l’indice de pénétration Pilodyn (−0,75 et −0,58 respectivement) ou le coefficient de retrait de l’échantillon brut (−0,77 et −0,73 respectivement).• Ces résultats suggèrent que l’amélioration génétique des traits-cibles examinés est possible par le biais d’une sélection bien qu’aucun des traits cibles évalués n’était fortement corrélé avec un contrôle interne.

Short note: the genetic correlation between air-dried density and basic density in Eucalyptus nitens wood cores

Abstract Cores extracted from trees to assess wood chemistry are generally not used to assess basic density in eucalypt pulpwood breeding programmes, as the measurement of basic density requires high temperature drying. However, both wood chemistry and air-dried density can be assessed on the same core. This study found that the inter-trait genetic correlation between core air-dried and basic density to be effectively equal to one in two Tasmanian Eucalyptus nitens progeny trials. This implies that selection for basic density could be undertaken using air-dried density with little or no reduction in genetic gain, thus negating the need to extract a separate core to assess basic density and wood chemistry. The adoption of this practice could considerably reduce the cost of assessing these traits in eucalypt breeding programmes.