Coenraad Brand Wessels

Eucalyptus is the most widely planted hardwood genus globally and an important source of industrial roundwood. In many Eucalyptus species, a substantial proportion of shrinkage arises from cell collapse occurring above the fibre saturation point (FSP). Shrinkage and collapse are commonly evaluated using increment cores or small specimens prepared as wedges or strips, yet the influence of specimen geometry on measured responses remains poorly understood. This study investigated the effect of specimen geometry on shrinkage and collapse behaviour in Eucalyptus grandis . Specimens from five trees were dried from the green state to an operational fibre saturation point and subsequently to oven-dry condition. Collapse-induced shrinkage was defined as shrinkage occurring above FSP, while normal cell-wall shrinkage occurred below FSP. The effects of longitudinal length, tangential length and specimen shape (wedge versus strip) were evaluated. Mean radial shrinkage and radial collapse did not differ significantly between wedge- and strip-shaped specimens. Tangential collapse was greater in wedges based on specimen means, although this difference was not significant when analysed using tree means. Longitudinal length had a significant but non-linear influence on collapse and total shrinkage, with 70 mm specimens generally showing the highest collapse, 45 mm specimens the lowest, and 20 mm specimens intermediate values. Tangential length had no significant effect over the tested range. Wedge and strip specimens also exhibited contrasting pith-to-bark trends in tangential collapse, indicating that specimen shape influenced pith-to-bark collapse profiles. These findings demonstrate that specimen geometry is not a neutral sampling consideration and can influence measured collapse behaviour in Eucalyptus grandis . Longitudinal length and specimen shape should therefore be standardised and reported explicitly when comparing results among studies or using laboratory specimens in breeding and screening programmes.

Eucalyptus is one of the world’s most widely planted hardwood genera and is increasingly supplying higher-value solid-wood and engineered products. However, wood cell collapse during drying—manifesting as surface waviness, dimensional instability, and internal checks—remains a concern for product and value recovery in many fast-grown taxa. Collapse in Eucalyptus has been studied for over a century, with numerous articles and postgraduate studies describing the collapse mechanisms and industrial pretreatment methods. Yet practically applied and standardised evaluation protocols for cell collapse within tree improvement programs are scarce. This review identifies the mechanisms, measurement approaches, and genetic selection potential for collapse, and outlines operational phenotyping options for breeding programs. Collapse typically initiates above the fibre saturation point (FSP) when capillary tension coincides with peak internal compressive stresses, with evidence indicating that collapse is also governed by the interaction of moisture transport, nanoscale weakening of the cell wall, and anatomical constraints. Across species and sites, density tends to correlate negatively with collapse but is not a reliable sole predictor, while cellulose content and stiffness (MOE) are informative indicators, especially where tension wood is implicated. Increment-core evaluation remains the most practical phenotyping method, although the effects of sample geometry on collapse are not yet fully understood. Technologies such as near-infrared (NIR) spectroscopy enable scalable screening in tree improvement programs. Collapse shows moderate heritability with limited genotype-by-environment interaction in several taxa, indicating that breeding against collapse is feasible, provided phenotyping is robust and comparable across sites. New insights into sap ion effects on permeability/collapse and strong geometry effects highlight the need for standardised test conditions. Process options such as intermittent schedules, reconditioning, and non-capillary dewatering can prevent or facilitate recovery from collapse but are species- and geometry-dependent. Integrating cost-effective, standardised phenotyping with multi-trait selection and genomic tools can materially reduce collapse risk in Eucalyptus breeding while informing drying strategies for value recovery.

Using nail plates to connect timber structural elements is a way to reduce costs and time during assembly, but moisture-induced backout can compromise these connections in the medium and long term. Most nail plate backout models proposed are based on a single tooth backout, and therefore, do not consider the influence of the metal plates at their bases. This study aimed to verify the effect of the plate connection on the backout. The dimensional changes, including a holistic approach to nail plate behaviour were evaluated in two treatments: S-D cycle, where wood started saturated and then dried to 0% moisture content; and D-S cycle, where wood started at 0% moisture content and was then saturated with moisture. The hole made on saturated wood got tighter when wood dried, and enlarged when dry wood gained moisture, so that teeth squeeze may happen when wood shrinks. These results were confirmed in holes made with nails hammered into the wood using the same moisture regimens. The nail plates were bent in opposite directions with the wood dimensional changes: with the S-D cycle the concave side faced the wood while with the D-S cycle the convex side faced the wood. Although the moisture gradients assessed here were high, realistic moisture changes have the potential to generate high stresses (450–650 MPa) on the nail plates and the dry-wet cycles might promote nail plate bending in opposite directions resulting in nail plate backout. To minimize this effect, wide nail plates can be substituted by multiple narrower ones. The amount of stress transferred from the wood to the nail plate should be investigated prior to adding it to the backout models.

Commercially grown Eucalyptus plantation trees are mostly used for low value applications such as pulp, board, and energy products. The broad aim of the research described in this paper was to assess and develop the use of fast-growing Eucalyptus species in South Africa for use in high value engineered wood products. As a first step the variation in the most important wood and processing properties was established for two species (E. grandis and E. nitens) and a hybrid (E. saligna x urophylla) commonly used in commercial forestry in South Africa. Standing tree dynamic modulus of elasticity, log splitting and log dynamic modulus of elasticity (MOE _dyn ) were determined as a function of species, site, and log height position. Log and board end-splitting is arguably the defect having the largest value impact for Eucalyptus species processed into sawn timber. Log end-splitting was assessed using a visual splitting score which included length, opening width and position of splits in a log. The stiffness of timber is an important mechanical property affecting the grade and value of engineered wood products and was assessed firstly on standing trees using the Fakopp Treesonic device, and then on the logs using Fibre-gen’s Director HM200 device through evaluation of the acoustic wave speed (AWS) and subsequently MOE _dyn . Linear mixed effects models were developed to examine the effects of species, site, and log height position on the MOE _dyn of logs. All three factors played a role, to some extent, in the measured properties of logs. The research showed that an understanding of the variation of end-splitting and MOE _dyn of Eucalyptus logs is important to select the best resource for engineered wood products and also to help improve processing decisions for each log resource.

Key message: The study investigates how wind exposure influences wood density and microfibril angle in two Pinus taxa, revealing greater sensitivity in Pinus elliottii × caribaea and highlighting MFA as a better wind-response indicator. Wind is one of the most persistent natural forces affecting plants, driving them to adapt their properties for survival. With the projected increase in wind intensity and coverage, commercial forests face growing vulnerability. This study investigated how environmental factors, particularly wind, influence wood density and microfibril angle (MFA) in 14.5-year-old South African-grown Pinus taxa. A 2 × 2 factorial experiment was conducted with two taxa— P. elliottii × caribaea and P. radiata —under high and low wind exposure. Over nine months, solar-powered anemometers recorded wind speed in the plots. Sixteen randomly selected trees were sampled and bark-to-bark wood cores extracted from the north–south and east–west directions at 1.3 m height. Density and MFA were analyzed using SilviScan, and mixed-effects models were developed using weather variables as predictors. Pinus radiata demonstrated better growth in diameter, height, and crown length but showed no significant wind-related effects on wood properties. In contrast, P. elliottii × caribaea was more responsive to wind, with MFA significantly lower in trees from less windy plots. Pinus elliottii × caribaea exhibited distinct annual rings with wide earlywood and narrow latewood, whereas P. radiata had less defined rings with more latewood. Growth rings near the pith were indistinct in both taxa. Cardinal direction significantly influenced MFA, with the highest values in the northern direction. Weather variables notably impacted density and MFA on the northern side across treatments. This study contributes valuable insights into the effects of environmental factors on wood properties. However, growth and development during the juvenile stage remain a complex process, requiring further research to clarify the factors driving property variation in commercial forestry species.

In the last two decades, wood has regained relevance in the construction industry of Europe, mainly driven by the adoption of cross laminated timber (CLT). This trend has reached other regions, but specially in developing countries this demand is facing a bottleneck: the high costs of the equipment for CLT manufacturing and the skilled labour to operate it. This study has the objective to test CLT samples manufactured with a novel and simple method using shank containing screws to generate pressure for the adhesion of layers. Three screw densities were used (30, 50 and 100 screws m⁻²) and compared to control samples (manufactured in a press under 0.5 MPa). Screws generated an average load of 0.85 kN and the interaction of wood and thread seemed more important than the interaction of wood with the head. All treatments met the shear strength on the glue line requirements from standards, with equivalent average values, approximately 3 MPa. Only the control treatment met the technical standard requirements for total and maximum delamination (10 and 40%) or minimum and average wood failure (50 and 70%). The 50 screws m⁻² treatment was close to meeting one of the delamination requirements, 40% maximum delamination. Although not fulfilling all the requirements, the CLT standard considers the use for high-rise buildings and the proposed method could be adopted for less demanding applications, such as single and double storey residences. Changes can be made on the manufacturing parameters to decrease the delamination, starting from the adhesive used and the screw geometry. With more research, the new method described can unlock the adoption of CLT in less developed regions that will be beneficial for the local economy since the forest and timber industry can generate many jobs and promote the adoption of wood buildings.

Key message: Understanding bark allocation in juvenile stages of commercially viable Pinus trees can be useful in dating wood formation, thus improving the accuracy of correlating wood quality to environmental factors. Bark is an important and multifunctional part of plant anatomy that has been researched mostly in the context of fire history, timber resource assessments and more recently as a bioresource. Few studies have comprehensively examined bark thickness in some commercially valuable Pinus species. More importantly, the role of bark in accurately dating wood formation has seldom been researched. This study was conducted to model and compare bark thickness variation between different species of young South African-grown Pinus trees including Pinus radiata var. radiata D. Don., Pinus elliottii Engelm., Pinus elliottii × Pinus caribaea var. hondurensis, Pinus patula × Pinus tecunumanii (Low Elevation), and Pinus patula × Pinus tecunumanii (High Elevation). Measurements were taken at 1.6 and 2.3 years including absolute and relative bark thickness and distribution along the stem. Results showed species-specific variation in absolute and relative bark thickness with the highest means recorded in Pinus elliottii × Pinus caribaea and lowest in Pinus radiata. A positive linear relationship was observed between bark thickness and diameter, consistent with all species and ages. Absolute bark thickness decreased along the stem from bottom to top while more nuanced patterns of variation were observed for relative bark thickness. These findings underscore the importance of understanding bark thickness in young trees for various applications, including dating wood formation, anticipatory breeding strategies for quality wood and predicting stand quality among others.