An optimised schedule has been developed using a model predictive control technique for drying 43-mm-thick (green) Eucalyptus pilularis Sm. boards, based on an original schedule generally recommended for greenhouse solar kilns equipped with good control of temperature and relative humidity. The predicted drying time for this schedule was 14% shorter than the original schedule using the drying model. Experimental tests confirmed this, with drying time in laboratory kiln being 10% shorter for this schedule (die initial moisture content was 10% higher) than the original schedule. The original schedule produced boards with a large number of end splits, a few surface checks, and some distortion, when drying from an initial moisture content of 60% to a final moisture content of 12%. In comparison, the optimised schedule produced boards with less degrade, from an initial moisture content of 70% to a final moisture content of 12%. Overall, the quality was slightly better and the drying time shorter for the optimised schedule than for the original schedule

Roots with under-bark diameters of between 1 and 4 mm from 11 New Zealand indigenous riparian plant species — lacebark (Hoheria sexstylosa Col,), kowhai (Sophora tetraptera J.S.Mill.), manuka (Leptospermum scoparium J.R. et G.Forst.), fivefinger (Pseudopanax arboreus (Murr.) Philipson), kohuhu (Pittosporum tenuifolium Sol. ex Gaertn.), rewarewa (Knightia excelsa R.Br.), cabbage tree (Cordyline australis (Forst.f.) Endl.), ribbonwood (Plagianthus regius (Poit.) Hochr.), lemonwood (Pittosporum eugenioides A.Cunn), tutu (Coriaria arborea Lindsay), and karamu (Coprosma robusta Raoul) — were tested to determine their live rootwood tensile strength using a Floor Model 1195 Instron Universal Testing Machine. These results were coupled with those from earlier tests on the native tree species southern rata (Metrosideros umbellata Cav.), red beech (Nothofagus fusca (Hook.f.) Oerst.), hard beech (N. truncata (Colenso) Cockayne), mountain beech (N. solandri var. cliffortioides (Hook.f.) Poole), manuka, kanuka (Kunzea ericoides var. ericoides (A.Rich.) Joy Thomps.), and kamahi (Weinmannia racemosa L.f.), and the exotic plantation species Douglas fir (Pseudotsuga menziesii (Mirb.) Franco) and radiata pine (Pinus radiata D.Don) to allow a wider-ranging comparison of live root-wood tensile strengths of those trees and shrubs that can commonly be found growing in potentially unstable slope and/or riparian environments throughout New Zealand. The mean live root-wood tensile strengths of these plant species ranged from 8 to 52 MPa.

Estimates of the economic benefit of biosecurity and forest health research to the New Zealand forest-growing industry and urban forest estate were made using cost-benefit analysis. The cost associated with the arrival of exotic forestpests was the sum of costs of eradication and control programmes, reduced harvest value, household expenditures to control the exotic pest, and replacement of affected trees in the urban forest. The expected cost associated with each of these was dependent on the likelihood of pest arrival, detection, eradication, and successful control, and the effect of research on these. Depending on the assumed efficacy of the research, the net present benefit of the $3.5 million annual research cost ranged from $3,519,million to $5,888 million, so there is considerable benefit to the New Zealand forest-growing industry and urban forest estate from biosecurity and forest health research. Sensitivity analysis showed benefits of research to be most sensitive to the estimate of the value of the urban forest estate

The four fungi described in this paper are either new or have not been fully described from New Zealand. The fungi are: Foliicolous Ascomycota: Calonectria acicola sp. nov. and its anamorph Cylindrocladium acicola sp. nov. on Pinus radiata D.Don. Xylophilous Anamorphic fungi: Hyphomycetes: Sporothrix nothofagi sp. nov. on Nothofagus fusca (Hooker f.) Oersted. Corticolous Ascomycota: Valsaria rubricosa on Pinus radiata. Radicicolous Anamorphic fungi: Hyphomycetes: Leptographium alethinum on Pinus radiata and P. strobus L.

A 27-year-old Pinus radiata D. Don clonal trial (with clones propagated from 7 year-old seedlings) was screened to identify clones with a wide range of branch size, internode length, and mature wood density (outer five rings at breast height). Four trees (ramets) from each of 10 such clones were subsequently chosen for use in wood processing/product performance studies, and wood properties were measured on discs taken from each end of 5-m logs up to the merchantable limit (200-mm small-end diameter). Properties assessed included wood density, microfibril angle, spiral grain, and incidence of compression wood. Wood density values di ffered markedly between clones, but within each the overall patterns of density variation were consistent from pith to bark and between stem levels, indicating high heritability. Broad-sense heritability estimates were 0.6 and higher for wood properties, except compression wood (0.43) and knot area ratio (0.38). Over all wood samples, there was strong evidence for heritability of density and spiral grain, good evidence for heritability of microfibril angle, and weak evidence for heritability of compression wood and knot area ratio. In juvenile wood samples there was strong evidence for heritability of density and spiral grain, good evidence for heritability of microfibril angle and knot area ratio, and weak evidence for heritability of compression wood. Values of Silviscan-2 variables were imputed for each piece of lumber on the basis of the actual sawing patterns for each log. Subsequent analyses examined the relationships between fundamental wood properties, predicted performance, and observed performance. Two stems from each cl one were processed into structural lumber and assessed for drying distortion and stiffness. The major effects on board stiffness were, in order of importance: density, board orientation, and knot area ratio. There was no evidence for an effect of microfibril angle. The major factors contributing to twist were spiral grain and density

Preliminary stand growth and yield models were constructed for even-aged Cupressus lusitanica Mill. and C macrocarpa Hartw. plantations in New Zealand. Models that predict mean top height, basal area, initial basal area, post-thinning basal area, mortality, and total standing volume per hectare were fitted to permanent sample plot data biased towards younger ages. A wide range of height growth rates were observed, with site index estimates ranging from 15 to 35 m mean top height at age 30. Basal area models predicted greater basal area growth for C macrocarpa. Exponential models predicted different rates of C macrocarpa mortality in the North and the South Islands of New Zealand. All suitable data were used to fit models, preventing separation of independent validation data. The models were tested as a system of equations by comparing total standing volume predictions with data used to fit the models. Volume predictions were relatively imprecise, but unbiased overall across the range of available data

A field trial was initiated in 1986 to evaluate nursery methods for producing aged cuttings of varying physiological age (PA) from stool-beds. Seedlings and cuttings taken from 5-year-old Pinus radiata D.Don seedlings grown in the field were included as controls. Three seedlots of different genetic improvement levels from unimproved bulk seed (GF3) to control-pollinated seed from a seed orchard (GF21) were used to study differences in growth rate, tree form, and wood properties (density and acoustic velocity) between cuttings of five different physiological ages up to 5 years and seedlings, and any interaction between physiological age and levels of genetic improvement. Results showed no persistent long-term height growth differences, and by 4 years of age, any height differences were no longer significant. GF16 and GF21 plants had significantly larger diameters than GF3 plants at both 4 and 11 years of age. Also, seedlings (PAO), and cuttings from 1-year-old seedlings (PA1), had significantly larger diameters than PA5 cuttings from field-grown trees; the cuttings of other physiological ages were intermediate at both 4 and 11 years of age. Physiologically older cuttings had better butt log straightness and freedom from malformation than seedlings and PA1 cuttings. There was no effect of seedlot or physiological age on wood density, up to a physiological age of 5 years. However, acoustic velocity at breast height (as an indicator of stiffness) increased significantly and consistently with increasing physiological age. Previous research has shown that cuttings with a phy siological age of 1 to 3 years will perform as well as or better than seedlings on both farm and forestry sites, with an optimum physiological age of about 3 years, when there will be improved stem form without any early loss of growth rate. This trial on a fertile ex-farm site has confirmed these trends and also shown improvement in breast-height acoustic velocity of around 6.4% from PA3 cuttings and more than 11% from PA5 field cuttings, compared with seedlings

Timber harvesting can result in adverse physical, chemical, and biological alterations to soil. The objective of this study was to examine the effects of site disturbance to determine the extent and duration of possible harvesting impacts on soil chemical and biological properties, including fine roots and mycorrhizas of Pinus radiata D.Don. The disturbance study was located in the North Island of New Zealand and was examined 9 years after organic matter and compaction treatments were installed. Treatments included undisturbed control plots, O horizon removed with no compaction, and O and A horizons removed with heavy compaction. Soil was examined for soil solution nitrogen, extractable nitrogen, fine root biomass, mycorrhizal root tips, and specific mycorrhizal root tip ammonium uptake rates. Results showed that total fine root biomass was reduced with loss of both organic-rich soil horizons and compaction to approximately one-third of that in the control treatment, but that mycorrhizal infection rates were higher (averaging over 60%). With removal of only the O horizon, the largest effect was simply the loss of rooting volume and the roots that would normally occur in this horizon, with little reduction in root biomass or change in mycorrhizal infection rates in the mineral horizons. Specific mycorrhizal root uptake rates of ammonium did not appear to have been changed by the most severe disturbance treatment. However, the unaltered uptake rate may be due to the predominant mycorrhizal morphotype found in the most-severe treatment which was different from the dominant morphotype found in the two less-severe treatments. Reduced fine root biomass of the severe disturbance treatment correlated with reduced tree growth and foliar nitrogen

Plant growth in unstable coastal sand is limited by nitrogen supply, and the promotion of in situ biological nitrogen fixation can reduce the need for fertiliser application where rapid growth of continuous vegetation cover is required after preliminary stabilisation with sand-binding species. Lupinus arboreus Sims (yellow tree lupin) was once used for this purpose, but susceptibility to a blight fungus has necessitated a search for alternative plant species. Eighteen leguminous species raised in a nursery were transplanted into spaced plant trials at three, widely separated, North Island west coast, sand dune sites. Trial areas were just behind the current foredune where planted Ammophila arenaria L. (marram grass) or Spinifex sericeus R.Br (spinifex) was being used to commence the sand stabilisation process. Survival rates, growth characteristics, and nitrogen-fixing potential were compared over a period of 4.5 years (1993–97). Assessments in the third year after planting gave mean values across all sites of 0–85% for survival, 7–121 cm for plant height, and 6–256 cm for plant spread (canopy diameter). Highest estimates of nitrogen accumulation in plant material in the fourth year were 38 kg/ha at Ninety Mile Beach (Lathyrus latifolius L.), 733 kg/ha at Muriwai Beach (Dorycnium rectum (L.) Ser.), and 869 kg/ha at Santoft Beach (Medicago arborea L.). Plants of 11 species survived for at least 5 years at all three sites. Six of the species tested (Dorycnium hirsutum (L.) Ser., D. pentaphyllum Scop., D. rectum, Lotus corniculatus L., L. pedunculatus Cav., and L. tenuis Waldst. & Kit. ex Willd.) are recommended for use in the planting succession because they fix nitrogen and also grow at a moderate rate under a wide range of open dune conditions. Other species may be useful although more site-specific. Reliance on a single leguminous species for a continuous nitrogen supply should be avoided.

Static bending of small clear specimens is one of the most commonly used methods for assessing the stiffness (modulus of elasticity, MoE) of sawn timber and trees. Small clear specimens have traditionally been cut at breast height from the same growth rings on opposing radii, thus seeking to minimise the radial and longitudinal variation. The remaining (residual) variation between small clear specimens determines the precision of the estimate of the tree mean MoE, but has rarely been analysed in detail because the method originally was not intended for tree-level analyses. To investigate this, axial stiffness measurements previously collected from small clear specimens taken from opposing radii at breast height on New Zealand-grown Pinus radiata D.Don and Pseudotsuga menziesii (Mirb.) Franco (Douglas fir) were reanalysed to ascertain the magnitude of the residual variation at breast height. Expressed as coefficient of variance between small clear specimens from the same radial position (growth ring), the variation ranged from 8% to 32% for P. radiata and from 7% to 13% for Douglas fir. Using two small clear specimens, the associated margin of error for estimates of mean stiffness of individual trees ranged from 40% to 140% for P. radiata and 40% to 60% for Douglas fir. It is recommended that at least four small clear specimens are used (margins of error of 10-40%) when estimating the mean MoE of individual trees from small clear specimens extracted at the same height from the same growth ring

A sawing simulator, AUTOSAW, was used to examine the effect of increasing sweep on lumber recovery. Sample material consisted of 51 logs from 22 western hemlock (Tsuga heterophylla (Raf.) Sarg.) trees in western Oregon, United States. All knots on the 4.9-m logs were measured, mapped, and converted into 3-dimensional digital formats. The digital logs were then increasingly bent, in 25.4-mm (1-inch) increments, with the bend occurring at the mid-point for one set of logs and at quarterway from the log small-end for another, and sawn into primarily structural grade dimension lumber. On average, conversion decreased at a rate of 10% for each 100-mm. increase in sweep, equivalent to a decrease of 5-7% for each 0.1 unit increase in sweep-to-diameter (s/d) ratio. Conversion losses were represented by an exponential decay function. There was no significant difference in conversions between the two sets of logs; however, the rate of loss of Select Structural and 1 Common lumber was greater for logs bent at the mid-point when sweep was within 51 to 152 mm. (2 to 6 inches). Whereas the rate of loss of lower grades was represented by linear functions, those for higher grades and lumber value per cubic metre of log volume were described by exponential decay functions. A 5% loss in value was recorded when s/d equaled 0.09 for logs with diameter less than 200 mm, and for larger logs the ratio was 0.07

The enormous variation in wood stiffness both within and amongst trees, results in the production of low-grade solid wood products from some trees. Accordingly, it would be highly desirable to segregate logs to ensure that only those logs with predominantly high stiffness wood are processed into structural timber products. This study examined whether sound flight velocity (km/s) could be used as a measure of wood stiffness to allow such segregation. Butt logs were cross-cut from 316 Pinus radiata D. Don trees and measured with three non-destructive acoustic devices, before and after harvest, to establish whether there was a relationship between stress wave velocity along the wood grain and the machine stress-grades of boards sawn from those logs. The wave velocity along the grain of logs was closely correlated with wood stiffness, whilst tree size and basic density estimated from depth of pin penetration of a Pilodyn were only moderately related. The outcomes of this study indicate that non-destructive acoustics tools offer a means of sorting logs according to wood stiffness prior to milling. A highly significant and positive relationship was found for acoustic resonance measurements made on logs; a weaker, but still significant relationship was obtained for time of flight measurements from standing trees. The challenge now is to develop a non-destructive tool that is able to more accurately measure the wood quality of standing trees to assist with wood quality inventories, yield mapping, and tree selections within breeding programmes

Use of mechanical harvesting/processing systems in timber harvesting is increasing worldwide, with advantages in terms of increasing productivity and safety. However, despite these systems giving operators access to advanced computer and measuring systems, their ability to extract the maximum value from a tree is, on average, less than motor manual log bucking systems. The productivity, cost, and value recovery of several simulated procedures for scanning and bucking Pseudotsuga menziesii (Mirb.) Franco (Douglas fir) and Pinus ponderosa Lawson & C.Lawson (ponderosa pine) trees were evaluated from a log seller's perspective. The procedures evaluated were (a) conventional operating where quality changes and bucking decisions were made by the machine operator, (b) an automatic full scan of the stem prior to optimisation and bucking, and (c) partial scanning where a portion of the stem was scanned and then qualities and dimensions were forecast before the optimal bucking took place. After subtracting costs, the net value improvement for the automated scanning procedures over the conventional procedure ranged from -7% to 8%. The best net value improvement for both species was obtained using the procedure that fully scans the stem prior to bucking. Breakeven capital investment costs for new scanning, forecasting, and optimisation equipment ranged between zero and US$2,120,000 depending on tree species, markets, scanning speed, volume scaling rules, and scanning procedure

Crown characteristics of 4-year-old Pinus radiata D.Don originating from seedlings and clonal trees propagated by tissue culture, and growing in the presence and absence of an understorey of luceme (Medicago sativa L.), in an agroforestry experiment were investigated to explain differences in foliage efficiency between the treatments. There was no difference in the tree height between selected sample trees for any of the treatments, but the diameter at breast height (1.4 m above ground-level) and the total tree biomass were larger for the clonal trees than for the trees originating from seedlings in the no-understorey treatment. The distribution of biomass within the crown showed marked differences between treatments. Branch basal area and the number of medium-sized branches were greater for the clonal trees than for the seedling trees, and foliage area per unit branch basal area was lower for the clonal trees than the seedling trees. Internode length was longer and the crown shape ratio higher for the clonal trees than the seedling trees. These differences in crown architecture resulted in differences in foliage area distribution within the crown, possibly leading to differences in the fraction of solar radiation intercepted. The allocation of above-ground biomass was changed, resulting in an increase in the stem wood fraction and a decrease in the branch fraction for trees growing with the lucerne understorey, compared with trees with no understorey present. Foliage area per unit branch basal area was lower for trees grown with the lucerne understorey than in the no-understorey treatment and these changes were more pronounced for the seedling than for the clonal trees. These results confirm that the competitive effects of understorey vegetation result in changes in the growth patterns of trees, including the allocation of biomass to above-ground components

The effectiveness of early selection for growth and form traits to maximise gain per annum in Pinus radiata D.Don was examined in a "farm-field" experiment based at two farm sites and one field site in the North Island of New Zealand. "farm-field" experiments are tests that are established on highly fertile "farm" sites, with intensive site preparation, intensive weed control, and close spacing, paired with tests of the same material on "field" sites. Family mean correlations among growth traits at early ages (1-3 years) on the farm sites were relatively high (0.54-1.0) but correlations between early ages at the farm sites and measurements at age 8 years at the field site were low to moderate (0.29-0.65). Correlations between form traits measured at age 3 and then age 8 years showed similar trends. Early assessments of growth and form did not provide good prediction of later-age performance, although the best prediction was obtained from the age-3 measurement. Selecting the highest-ranked 20 families for diameter at age 3 years gave higher relative efficiency of selection than selection at earlier ages. The use of bare-root seedlings rather than container stock, and persistent nursery effects may have contributed to the low selection efficiencies. Maternal effects were also likely to have affected early-age measurements. However, although including 100-seed weight in the model did lower heritability estimates and some genetic correlations for height growth, genetic correlations were not greatly changed by including this factor and early-age selection efficiency did not improve. The potential for early selection for growth and for form in farm-field experiments with P. radiata in New Zealand appears to be very limited"

The neutral sulphite-anthraquinone delignification of Pinus radiata D. Don samples is readily followed by determining the variation in absorbance at 280 nm of (diluted) pulping liquor with cooking time. The 280 nm wavelength, despite its low absorption maximum, is preferable to that of the 205 nm wavelength used in a previous study since it is less prone to interference by changes in pH, or liquor oxidation. Excellent correlations between liquor absorbance at 280 nm and the amount of lignin dissolved in liquor were obtained for a number of cooks (and for three digester systems). The results showed that the extent of lignin removal and sodium sulphite concentration (or amount consumed) were linearly correlated, and indicated that absorbance at 280 nm, when combined with other procedures, could be used to examine the kinetics of neutral sulphite-anthraquinone pulping

Identification and selection of superior trees in forest management and breeding programmes provide a means to improve the properties and value of future wood products. Non-destructive stiffness assessment of standing trees enables selection of individuals for their stiffness, and so the accuracy and cost of four methods for assessing stiffness were evaluated: (1) IML hammer, (2) 5-mm outerwood density cores, (3) Pilodyn penetrometer, and (4) SilviScan-2®. Sixty 18-year-old Pseudotsuga menziesii (Mirb.) Franco (Douglas fir) trees were assessed for stiffness and the results compared with static modulus of elasticity (MoE) measurements of small clears centred on the tenth annual ring at breast height. Data were analysed using linear models and descriptive statistics, and the effects and costs of selection were modelled. The IML Hammer and outerwood density cores both gave corrected selection differentials of 11–16% with respect to stiffness at a cost of NZ$20–30 per tree selected. The Pilodyn was also quite cheap, but failed to give an informative measure of stiffness. SilviScan-2® provided a more accurate assessment and subsequent higher estimated selection differential of 22% at a cost of around NZ$500 per selected tree. Technology developments currently being implemented may reduce this cost over time. Selection for stem volume growth alone decreased average stiffness by around 10%.

Some coastal Californian provenances of Pseudotsuga menziesii (Mirb.) Franco (Douglas fir) have bark that is visibly thicker and more deeply furrowed than more northern and inland provenances. From a literature study it was evident that these variations in bark thickness most likely constitute adaptation to spatial and temporal patterns of wildfires within the natural range of Douglas fir. Six provenances from the latitudinal range of Douglas fir in the Pacific Northwest of the United States (37°–48°N) were sampled for bark thickness and compared with a New Zealand landrace (Kaingaroa) seedlot at two New Zealand trial sites (38° and 46°S). The analyses showed that Californian provenances had significantly thicker bark than both the Kaingaroa (ex Washington) control seedlot and the Oregon and Washington provenances. The most southern provenance (Santa Cruz, California) had the thickest bark. Thus there was a steady reduction in bark thickness with increasing latitude of the seed sources. The bark thickness of the Kaingaroa seedlot was not significantly different from the Washington and Oregon provenances. The provenance variations in bark thickness caused a bias in under-bark volume estimates from volume function “T136”. Errors in volume estimation were greatest for Santa Cruz (+7.1%), Jackson State Forest (+2.8%), and Mad River (+2.0%). It is recommended that volume equation “T136” should be revised to account for differences in bark thickness with provenance.

The genotypic variation of two commonly planted Pseudotsuga menziesii (Mirb.) Franco (Douglas fir) provenances (Ashley strain (Seedlot 93-273) referred to as 93 and Tramway strain (i.e., Beaumont strain) (Seedlot 98-514) referred to as 98) was investigated in growth and physiological responses of the seedlings to nitrogen (N) supply levels (40 and 100 mg N/litre) in sand culture in a glasshouse. The growth of Douglas fir seedlings was greater at 100 mg applied N/litre (high nitrogen) than 40 mg applied N/litre (low nitrogen). The high nitrogen supply treatment increased seedling growth rate and caused more biomass to be allocated to the shoots. Increased uptake rate of nitrogen and other nutrients (per unit mass of roots), increased percentage allocation of the nutrients to shoots, and enhanced net photosynthesis in needles were observed with high nitrogen supply. There were very significant provenance differences in most measured parameters and significant nitrogen × provenance interactions in some parameters. Provenance 93 showed better growth than Provenance 98 at both nitrogen levels, with a larger difference in growth between the two provenances at the low nitrogen supply. Differences in response of the two provenances to nitrogen treatment levels were related to the ratio of needles to whole-plant dry weight, nitrogen productivity, uptake rate and allocation of nitrogen and other nutrients, but unrelated to the photosynthetic rate per unit leaf area. Provenance 93 had a greater ratio of needle to whole-plant dry weight, nitrogen productivity, uptake rate of nitrogen and other nutrients, and allocated greater proportions of absorbed nutrients to shoots for photosynthesis and new growth, which sustained a greater growth rate in Provenance 93, especially in the shoot. Among the growth parameters measured in this study, the relative growth rate of seedling height (RGR-Ht), the root /shoot ratio, and the ratio of needle to whole-plant dry weight seem to be the reliable and simple indicators to discriminate the provenance difference in response to nitrogen

Foliar application of chitosan on Pinus radiata D.Don seedlings resulted in the induction of resistance to wound-inoculation with Fusarium circinatum Nirenberg & O’Donnell, the causal agent of pitch canker. Induced resistance was expressed as the suppression of symptom development on treated seedlings compared to the controls. Chitosan treatment protected seedlings against wound inoculation with ca100 spores per plant, and reduced disease incidence by 60% and deadtop development by 50%, compared to water-treated controls. However, chitosan efficacy was inoculum-dose dependent and no significant disease control was observed when treated seedlings were challenged with 500 or 8500 spores. At lower inoculum levels (10 spores/seedling) chitosan-treated seedlings remained more resistant to inoculation for at least 6 weeks after application. Furthermore, chitosan-induced resistance was shown to be systemic on 4-year-old P. radiata trees when chitosan application and inoculation were spatially separated. Due to the quarantine status of F. circinatum in New Zealand, inoculations were performed using Sphaeropsis sapinea (Fr.) Dyko & B.Sutton for this aspect of the study. Localised chitosan application induced an 86% reduction in average lesion length on adjacent untreated branches compared to their water-treated counterparts. These results demonstrate that chitosan can induce systemic disease resistance in P. radiata and there is potential for induced resistance in forest nursery disease management.

Bare-root seedlings and bare-root cuttings of Pinus radiata D.Don at three ages (11, 27, and 36 months) were excavated by hand from a nursery and two trial sites in the North Island, New Zealand. A total of 48 trees were sampled, 24 from each plant type. Root system characteristics were compared by investigating differences in above- and below-ground biomass, root length, root size, and root biomass distribution. It was found that by 36 months a greater portion of wind-induced biomass had been allocated to the near-stem lateral roots of bare-root cuttings than bare-root seedlings. The presence of this indicated growth over and above what could be expected solely from age, and was likely to be related to a wind-induced adaptive growth response. The development of juvenile-tree stability can be attributed primarily to the increasing rigidity of those near-stem areas of the root system subjected to higher concentrations of wind-induced stress. As only a portion of the root system, i.e., the near-stem roots, contributes to tree stability, the use of root:shoot ratio as a primary indicator of tree wind-stability ranking could be suspect

Soil humic substances are important components of soil organic matter and contain a significant portion of total soil organic carbon (C) and nitrogen (N). Solid-state ¹³C nuclear magnetic resonance (NMR) with cross-polarisation and magic angle spinning (CPMAS) was applied to humic acids extracted from 0-10 cm soils collected from areas under windrows of harvest residues and those areas between the windrows, 3 years after implementation of residue management in a second-rotation plantation of Araucaria cunninghamii Ait. ex D.Don (hoop pine). In addition, nitrogen availability of under-windrow and between-windrow soils was also assayed by anaerobic incubation with either water or ¹⁵N-labelled ammonium sulphate solution in the laboratory. The NMR spectra of the humic acids showed that the carbon composition of the under-windrow humic acids was different to that of the between-windrow humic acids. Potentially mineralisable nitrogen of the under-windrow soils was greater than that of the between-windrow soils, as was gross nitrogen mineralisation (ammonification, m _g). Soil potentially mineralisable nitrogen was also positively correlated with humic acid-alkyl and humic acid-O-alkyl carbon (p<0.05), while gross nitrogen mineralisation was positively correlated with humc acid-aromatic carbon (p<0.01). The gross nitrogen mineralisation was 33 - 45 mg N/kg dry soil as determined by isotope dilution with 15N-labelled ammonium sulphate (100 mg N/kg and 99 atom% ¹⁵N excess) and was greater in under-windrow than between-windrow soil after the 7-day anaerobic incubation. In addition, gross ¹⁵N immobilised (NH ₄+ consumption, m) was positively correlated with humic acid-aromatic carbon (p<0.05). Humic acid-iron content was positively correlated with humic acid-alkyl and O-alkyl carbon (p<0.05)

Genetic gains were simulated for backwards and forwards selection (1.5- and second-generation respectively), using offspring of 300 plus-trees from a base population of 30 000, and making 15 backwards selections throughout. Options were: (1) backwards selection from half-sib progeny test; (2) forwards selection within same; (3) forwards selection within select-pollen polycrosses; (4) forwards selection within single-pair crosses. Parameters varied were: heritability (h²) (0.1– 0.6 in progeny, 0.05–0.3 in base population); dominance : additive genetic variance (d) (0–1); total offspring (N) (9000 or 30 000); and forwards selections made (n) (15–120). Each set of conditions involved 200 replicate runs. For mean gain across runs (Mean) Option 4 was consistently best, except with high d; Option 3 excelled Option 2 at higher heritabilities and lower n