The kraft pulping properties of New Zealand-grown Picea abies (L.) Karst and Picea sitchensis (Bong.) Carr were compared with Pinus radiata D. Don. At kappa number 30, kraft pulp yield was approximately 48% from P. sitchensis and 46% from P. abies. Handsheet properties of the P. sitchensis pulps were similar to those from Pinus radiata of equivalent wood density but the P. abies pulps were somewhat inferior. Pulps from the 72-year-old P. sitchensis trees examined were approximately equivalent in yield and strength to those from 25-year-old Pinus radiata.

Pinus radiata D. Don plywood panels of two constructions and four different widths were tested in third-point bending over a span of 1050 mm. No significant difference was found between moduli of rupture for the different widths. An ASTM centre-point bending test was carried out on 50-mm-wide specimens cut from the panels. It was found that the moduli of rupture of these specimens tested in third-point bending were significantly lower than the ASTM test values. Moduli of elasticity were similar in all tests.

Pinus radiata D. Don wood residues can be divided into two categories - top logs and whole trees <20 years old (predominantly of corewood quality), and the slabwood from sawn sawlogs (predominantly of outerwood quality). These two residue types are further defined by their wood basic density with top logs generally <400 kg/m³ and slabwood generally >400 kg/m³. The over-all basic density range for either residue category normally lies within 100 kg/m³. The actual density values of a given wood supply are dependent on tree age, and the growing site and silvicultural history of a forest stand. Moisture contents are higher and basic densities are lower for whole trees (<20 years old) than for top logs, which contain the same number of growth layers.

It is estimated that up to 60-70% of New Zealand's P. radiata resource could be utilised as residues rather than as solid wood. Residue utilisation options include the manufacture of chemicals, panel products, and a range of pulp and paper products. Pulp and paper products are considered to be the most sensitive to wood quality differences.

A series of test roofs clad with radiata pine (Pinus radiata D. Don) shingles that had been pressure-treated with various water-borne preservative formulations were installed at a Forest Research Institute test site in Rotorua in 1977. A further series of test roofs installed in 1978 included radiata pine shingles pressure-treated with a commercial light organic solvent preservative.

Evaluation of selected radiata pine shingle test roofs in early 1984 showed that, although preservative leaching had occurred from shingles treated with an alkyl ammonium compound, a copper-chrome-arsenic preservative, and a light organic solvent preservative, the shingles appeared sound when examined microscopically. Slight lichen growth was evident on some shingles. In contrast, untreated radiata pine shingles displayed marked fungal infection, early stages of decay, and plentiful mould and lichen growth. Untreated western red cedar (Thuja plicata D. Don) shingles, which were used as a reference point from which to judge radiata pine shingles, showed lichen and mould growth after 7 years' exposure. Checking and cupping of western red cedar shingles were attributable in part to the inclusion of some flat-sawn material.

The strength and stiffness properties of Pinus radiata D. Don plywood at angles to the grain of the face veneer have been modelled by theoretical equations and compared with test results. Stiffness properties were calculated assuming linear elastic behaviour of each wood veneer in the plywood, and strength properties were determined using a two-dimensional plane stress failure criterion and assumed values for shear strength and axial strength of wood. Experimental values were close to the theoretical relationships derived for in-plane bending strength and stiffness, tension strength, and compression strength, and for stiffness bending perpendicular to the plane of the plywood. The derived relationships were used to calculate strength ratios which may be used for the design of plywood gusset plates and other components where stress is applied at angles other than parallel or perpendicular to the grain of the face veneer. Experimental results verify that the current code basic working stresses parallel to the grain derived from out-of-plane bending tests can be used for in-plane bending.

Australian-grown Pinus radiata D. Don (radiata pine) timber of 35 x 70 mm, 35 x 90 mm, and 45 x 90 mm cross-section was mechanically graded and then tested in bending to determine modulus of elasticity (MOE) and modulus of rupture (MOR). Knots contributing to failure were measured and classified and their knot area ratio calculated. The air-dry density of the timber was also determined.

Accuracy of prediction of MOR using MOE was shown to be independent of timber density but was improved by including knot area ratio, and sometimes knot position, in the regression.

Reductions in timber density and increases in knot size reduce theoretical mechanical stress-grade yields. Therefore a reduction in the market value of sawn structural timber is predicted if forest management aims to increase the rate of growth of plantation trees without compensatory attention to density and branch size.

When the radiata pine timber used in the study was graded to the normal Australian visual grading rules, it generally failed to meet normal bending strength requirements. A re-assessment of the grading rules is recommended.

A procedure has been developed for the measurement of undried (uncollapsed), and dried and reconstituted (collapsed) fibre cross-sections in unbleached and bleached kraft pulps made from Pinus radiata D. Don wood of low and medium basic density. The data obtained show that pulp drying causes fibres to be collapsed and fibre walls to be somewhat reduced in thickness. The widths of unbleached kraft fibres are generally unchanged by pulp drying although those of corresponding bleached fibres are significantly decreased by the drying process.

The wood anatomy of five hardwood species grown as exotics in Western Samoa has been examined. The species are Anthocephalus chinensis (Lamk) Rich, ex Walp., Cedrela odorata L., Eucalyptus deglupta Blume., Swietenia macrophylla King, and Tectona grandis L.

There should be no difficulty in distinguishing between the timbers described, and between these timbers and the indigenous timbers of Western Samoa. However, when the origin of the specimen is unknown, identification of C. odorata may be difficult because of its similarity to other species of Cedrela and, because of limited information, it is not certain whether A. chinensis can be separated from the other two species of Anthocephalus.

A clonal seed orchard of Pinus radiata D. Don growing in Victoria, Australia, was irrigated and fertilised with urea and superphosphate over a 4-year period.

In two of the years, irrigation significantly increased female "flowering". Water input (rainfall and irrigation) in the year prior to flowering accounted for 71% of the variation in flower crops. A positive flowering response to nitrogen was observed in the 2 years of higher rainfall, and a positive interaction between irrigation and nitrogen fertiliser in the drier years. Response to nitrogen was thus dependent upon an adequate water supply.

Stem basal area increment was greatest in irrigated plots treated with urea, and our results were consistent with the hypothesis that female flowering increased as a direct consequence of increased vegetative growth.

It was estimated that mean annual seed yield over the 4-year period from the most productive treatment was 29.1 kg/ha compared with 17.4 kg/ha from untreated trees.

Changes in total, soluble, and insoluble nitrogen and chlorophyll concentrations in Pinus radiata D. Don seedlings were monitored for 56 days after a single undercutting (root pruning). After undercutting there was a marked reduction in total soluble and insoluble shoot nitrogen concentration. Some of this mobilised nitrogen contributed towards lateral root growth while 15% was lost, possibly from the damaged root as exudate. Twenty-one days after undercutting, total seedling nitrogen content began to improve, coinciding with lateral root dry matter gains. By 56 days, undercut seedling nitrogen content approached that at Day 0, while control seedlings had acquired an additional 20.6 mg nitrogen.

Lateral root dry weights and nitrogen content of undercut seedlings were higher than those of control plants after 56 days. However, soluble nitrogen concentration in lateral roots of undercut seedlings was not significantly different from the controls, although total and insoluble nitrogen concentrations were significantly below control plant values.

There was no significant difference in chlorophyll concentration of needles between control and undercut seedlings.

Soil sampling during early forest growth on coastal sand showed little indication of nitrogen accumulation in the mineral soil. In spite of 7 years’ intermittent lupin (Lupinus arboreus Sims) growth and regular 6-monthly applications of fertiliser (a total of 900 kg N/ha, plus other nutrient elements) over a 10-year period, total-nitrogen levels remained below 0.03%.

A bioassay of soil samples using Pinus radiata D. Don seedlings showed that potentially-available-nitrogen was influenced by experimental treatment (fertiliser addition, lupin exclusion) to a greater extent than total-nitrogen. Potentially-available-nitrogen responses to the presence of lupins and to fertiliser were observed in the second year of the trial and from the fifth year onwards. The lupin effect was very similar to that of fertiliser treatment. Highest values resulted from the combined influence of lupins and fertiliser, but the effect was additive rather than interactive.

Foliar-nitrogen levels declined during the first 5 years of tree growth and showed no effect of lupins or fertiliser treatment. In the sixth year the onset of nitrogen stress in trees growing without fertiliser or lupins (foliar-nitrogen range 0.9-1.2%) was associated with significant and recurring treatment differences. The effect of fertiliser (foliar-nitrogen range 1.1-1.5%) was sometimes, but not always greater than that of lupins (foliar-nitrogen range 1.0-1.4%) and highest values (1.2-1.6%) were recorded where trees were influenced by both lupins and fertiliser. The lupin effect lasted for at least 3 years after lupins had died out of the understorey at tree age 4 years and the fertiliser effect for at least 4 years after the final application.

Evidence from foliar analysis indicated that tree nitrogen demand was not met by fertiliser treatment alone, and the efficiency of fertiliser-nitrogen utilisation appeared to be low. All improvements in tree nitrogen status attributable to lupins were achieved without additional management costs.

The effect of temperature, particle size, sodium carbonate applied, and bark consistency on total extractive yield and on the extract tannin content from Abies concolor (Gord. & Glend.) Lindi, (white fir) whole bark was investigated. A Graeco-Latin square design was used to evaluate the statistical significance of the variables on extractive yield and a completely random, uneven design replication was used to indicate the magnitude and trends that any of these variables might have. The most feasible and economic, but nonexhaustive, extraction process possible for the production of polyphenols from white fir bark for use in tannin-formaldehyde wood binders is extraction of particles smaller than 0.23 mm at 60 °C and at 20% or less consistency, with sodium sulphite/bisulphite added to stabilise the extracts.

Seedlings of Eucalyptus delegatensis R.T. Bak., E. maculata Hook, and E. brockwayi CA. Gardn. were grown for up to 10 weeks in a solid rooting medium and flushed daily with one of five different nutrient solutions. Stable relative growth rates and internal nitrogen concentrations were established after an initial nutrient adjustment (lag) phase. Relative growth rate and relative nutrient uptake rate were directly related after the lag phase. Partitioning of assimilate to roots decreased from 0.35 (root weight/total weight) at low concentrations of nutrients, to 0.10 at high concentrations.

The relationship between maturation state of Pinus radiata D. Don and its early resistance to Dothistroma pini Hulbary var. linearis infection (measured as number of needle blight bands) was examined. Two-year-old seedlings, grown in the field, were found to have much higher levels of infection than related stecklings of a more advanced maturation state. By the end of the second season of growth, seedlings were almost completely defoliated as a result of the infection. There was no evidence that hedging P. radiata trees at different heights up to 4m provided cuttings with any differences in susceptibility to infection. Seedlings from the Guadalupe population showed a lower level of infection than seedlings of the other populations tested.

Fungicides were screened for ability to control needle-cast of Pinus radiata D. Don caused by Cydaneusma minus (Butin) DiCosmo et al. Undetached shoots were dipped in water-based suspensions at fortnightly intervals for 20 months from the time of flush. Significant improvements in levels of needle retention were achieved with dodine (86% retention), anilazine (76%), benomyl (74%), and dichlone (66%), in comparison with untreated controls (39%). Injections of acidified aqueous solutions of carbendazim into stems of 8- to 10-year-old trees resulted in reductions in foliage yellowing, needle loss, and numbers of C. minus colonies isolated from the foliage. Needle loss was reduced two- to three-fold on 0- to 1-year-old shoots after one season of injections; after 2 consecutive years, of injections needle loss was reduced seven-fold on 1- to 2-year-old shoots. Yellowing and defoliation were also reduced by injections of the non-fungicidal compounds ortho-phenylenediamine and L-arginine monohydrochloride, implying that not all the effects of carbendazim are a consequence of its fungicidal properties. Aerial applications of benomyl (0.25 kg/ha) in an emulsion of water (6 litres/ha) and BP crop oil (4 litres/ha) in June and July failed to check the disease in a 10-year-old P. radiata plantation.

An aerial survey of 80 000 ha of Pinus radiata D. Don plantations in New Zealand in 1983 showed that the needle-cast caused by Cyclaneusma minus (Butin) DiCosmo et al. was present in all forests surveyed. Stands aged between 11 and 20 years showed a considerable range of disease severity (trace to 60% crown infection) and trees with more than 20% crown infection occupied 11% of the survey area in this age-class. Stands younger than 6 years and older than 26 years of age showed very low disease severity (less than 20% of the crown infected) but in the >25-year age-class such lightly diseased trees occupied 77% of the area surveyed. The survey values were extrapolated to the entire estate of P. radiata and it was estimated that the annual increment loss incurred through infection by C. minus was of the order of 0.2% in the >25-year age-class and 3.8% in the 11-20 age-class. Over the forest estate as a whole, it equalled 2.3%.

Twenty pairs of 9-year-old final-crop trees were selected in a stand of Pinus radiata D. Don heavily infected with Cyclaneusma minus (Butin) DiCosmo et al. at Kaingaroa State Forest in 1977. Each pair consisted of one heavily diseased tree and one healthy tree. Disease severity, in terms of percentage of crown infected, was recorded annually. Relationships between disease severity and growth were studied by complete stem analysis at tree age 15 years.

Annual volume increments of the diseased trees were significantly reduced from age 7 onwards. Reduced diameter growth in the lower parts of the stem of the diseased trees caused small changes in the relative diameter distribution. The relationship between disease severity and volume loss showed a reduction of the average volume increment of approximately 60% at an average disease severity of 80%. Stand growth was projected to age 30 for various proportions of diseased trees. For each 10% increase in the proportion of diseased trees a reduction in total volume of 10-14 m3/ha and a subsequent reduction in revenue at clearfelling of $600-$700/ha may be expected under the given stand conditions.

Nine pairs of trees were injected with a solution of carbendazim fungicide in 1980 and 1981 in an attempt to control the disease. Injection with the fungicide gave a significant growth response associated with a trend to reduced disease severity. Some of the growth response may have been attributable to a nutritional effect as it occurred in healthy and diseased trees.

Rooted cuttings of three clones and seedlings of Pinus radiata D. Don were inoculated with ascospores or mycelium of Cyclaneusma minus (Butin) DiCosmo et al. and kept for 3 months in growth rooms at different temperatures. All cuttings belonging to two of the clones developed typical symptoms of Cyclaneusma needle-cast and C. minus was isolated from needles taken from these cuttings. Needles from cuttings of the third clone yielded very few isolates of C. minus and the cuttings did not exhibit any symptoms of ill health. Seedlings were not infected by the fungus. An estimate was made of the number of spore-bearing apothecia of C. minus present on the litter layer of a stand of 10-year-old P. radiata, subject to the needle-cast, by collecting needles shed over fortnightly periods and counting the number of spore-bearing apothecia present on a sub-sample. The collected needles were returned to the stand, placed in a mesh-lined frame, and examined every fortnight until no more apothecia were seen. The two peak periods of needle-cast were in spring and autumn, and the greatest number of apothecia was found in autumn-winter (May to August). Records from a Hirst spore trap set up in the same stand showed that the C. minus ascospores occurred most frequently in autumn-winter and that ascospore release was dependent on rainfall. Monthly isolations from needles showed that current season's needles were first colonised by C. minus in autumn-winter (May-June) when they were about 8-9 months old and by Lophodermium spp. about 2 months later. Most of the infected needles were shed when they were about a year old, but some needles were retained until the following winter.

A largely subjective provisional classification of virgin indigenous forests has been produced for the South Island of New Zealand. Three geographical zones were recognised, and within each zone forest types were identified from a combination of two approaches. Canopy-tree stocking data collected during the systematic line-plot sampling of the National Forest Survey of New Zealand (1946-55) were examined with the assistance of a computerised cluster analysis technique. In areas not covered by the National Forest Survey plot sampling, forest descriptions from more recent work were used. The resultant 94 forest types were distributed between 10 forest classes to produce a three-level classification.

Australian-grown Pinus radiata D. Don (radiata pine) timber, 35 x 150 mm and 35 x 200 mm in cross-section, was mechanically and visually graded, then subjected to mechanical testing to determine modulus of elasticity and modulus of rupture. Results were compared to previous ones for radiata pine 35 x 120 mm and 35 x 90 mm in cross-section, and to normally accepted in-grade strength and stiffness requirements. Timber stiffness met the accepted requirements for all the visual and mechanical grades. Under the 1973 Australian visual grading rules the material had comparable strength ratings to that of mechanically graded timber. If, however, the current visual rules had been used the timber strength of the F8 and F7 grades would have been inadequate. A return to the 1973 rules appears warranted. For machine grading, the lower 2.5 percentile of modulus of rupture was, for most grades, slightly below the normally accepted limits and a minor change to the current grading programmes appears to be warranted for the larger cross-sections. Mean modulus of rupture, for all machine grades, was lower than found previously for radiata pine of smaller cross-sectional width.

The mathematical solution for a two-dimensional cutting problem of an irregular space presented is capable of being extended to a third dimension. The computer program based on this algorithm uses information on marketable timber assortments (dimensions, prices for the various grades, and dressing) to optimise the sawing pattern for a specified log with a certain defect core. In the two-dimensional examples three grades are recognised: clear boards, one face clear, and knotty grade boards. The run time on a single processor is extensive and real-time on-line applications will be realistic only using a multi-processor, for which the dynamic programming algorithm described is highly suited.

Sixty 5.5-m pruned butt logs, from two managed stands of Pinus radiata D. Don in Mangatu State Forest, were selected on the basis of three size classes and three sweep categories. All logs were sawn to 25-mm boards using a standard conversion procedure in a single-band sawmill at Rotorua. The produce was graded according to commercial grades and provisional Clear and Clearcutting grade classifications.

Conversion factors increased with increasing log size and decreased with increasing sweep. A change in the sweep : diameter ratio (deviation : small-end diameter (s.e.d.)) of 0.1 was associated with an average drop in timber recovery of about 5%. Straight logs (average s.e.d. = 390 mm) yielded 45% of the outturn in Clear grades compared to 34% and 28% for moderately swept (average sweep 16 mm/m) and severely swept (30 mm/m) logs respectively. Increasing sweep also caused a marked reduction in the recovery of Clearcuttings longer than 3 m.

Regression equations linking log characteristics and grade outturn indicated that only the higher grades can be predicted on an individual log basis with any degree of confidence. The most significant log variables were small-end diameter, defect core diameter, sweep, and taper.

The dry matter content of the above-ground components of a 7-year-old plantation of Pinus radiata D. Don was recorded for 4 years after treatment in a thinning x nitrogen fertiliser factorial experiment. Fertiliser application significantly accelerated maximum canopy development. Stem production was closely related to the amount of stem material present after initial treatment. Nitrogen fertiliser increased the fraction of dry matter allocated to crowns and especially to needles. Thinning increased the fraction of dry matter allocated to crowns in the year after treatment. As canopies closed there was a tendency to allocate an increasing fraction of dry matter to stems at the expense of branches. The increase in dry matter production after fertiliser application was related to an increase in foliage amount rather than to changes in foliage efficiency.

Published biomass equations, based on a large and widely dispersed sample, for the above-ground components of individual Pinus radiata D. Don trees were tested with a regional sample of up to 62 trees from Gippsland, Victoria. The masses of some components were considerably under-estimated, indicating that the widely based equations are inappropriate to the region. Biomass equations for unpruned trees growing in closed-canopy stands were therefore developed from our data for the Gippsland region. The mass of needles, of live branches, of total branches, of stem wood, of stem bark, and of total stem are given as functions of diameter at breast height, diameter at the base of the live crown, and total height. These equations are considered suitable for use in regional surveys.

A methodology for modelling the growth of managed even-aged stands has been developed. The state of a stand is represented by a number of variables - typically basal area, stocking, and top height. Changes of state through growth and mortality are given by a system of differential equations which are a multivariate generalisation of the Bertalanffy-Richards model. For estimation purposes random perturbations are included, and the parameters are estimated by maximum likelihood based on the resulting stochastic differential equations. The growth equations are complemented by models for thinning, early growth, volume per hectare, and diameter distributions. The methods have been applied to the development of a model for Pinus radiata D. Don in Golden Downs Forest, Nelson, with satisfactory results.

Two hundred and ninety-nine plots of Pinus radiata D. Don in forests throughout the North Island of New Zealand were sampled. At each plot a site index (tree height at 20 years) estimate was obtained, and soils were sampled to determine soil depth, strength, and nutrient concentration. Climatic data were extrapolated from the nearest available station. Principal component analysis was used to reduce the large number of items of data available for each plot to a smaller, mutually independent set. A model was constructed by multiple regression analysis which related site index to this set of variables. The model has a standard error of 2 m around a mean site index of 29.2 m, and predicts increasing site index with increasing rainfall, nutrients, topsoil depth, and soil penetrability, the optimum average annual temperature for growth being 12°C and the optimum soil pH 6.

Validation of the model on a New Zealand-wide set of plots, subjectively chosen to be in extreme conditions, showed that it generally behaved reliably. It did, however, overpredict growth for an unusual forest on shallow slopes receiving 4000 mm of rainfall per year.

On three pasture sites at Putaruru, Rotorua, and Taradale, and on one forest site at Kaingaroa, the establishment requirements of Pinus radiata D. Don cuttings were found to be similar to those of seedlings. The removal of grass competition for a minimum of 1 year improved height growth of both cuttings and seedlings at age 5 by an average of 1.7 m (30% improvement) and diameter growth by an average of 3.2 cm (32% improvement) (equivalent to 1 year's growth). Grass release for 2 years gave no significant increase in growth and would not justify the additional cost, even on drier sites. Only at Kaingaroa did the application of fertiliser at establishment result in some improvement in growth of both stock types.

Although by age 5 the cuttings and seedlings had shown similar height growth, the diameter growth of the seedlings was found to be 1.0 to 2.4 cm better than the cuttings, depending on site but not establishment treatment. The rapid diameter growth recorded on the farm sites in this study indicates the advantage of growing trees on sites with a history of pastoral farming.

Disease severity of Dothistroma pini Hulbary and growth data on Pinus radiata D. Don after aerial spraying with copper oxychloride for 4 consecutive years were analysed at Kaingaroa State Forest, Tiritea Plantings, and Mawhera State Forest, and in a 14-year-old spray trial in Cpt 360 at Kaingaroa State Forest. Reduction of disease levels after the spray treatments was variable between sites and between seasons. The disease levels decreased to trace levels after pruning and thinning, whereafter the rates of reinfection were variable.

Differences in height associated with spraying for disease control could not be demonstrated. Basal area increased in sprayed plots after 1 year or more of substantial disease reduction. Thinning, however, eliminated basal area differences between treatments. Volume differences between treatments in the 14-year-old spray trial were not significant.

A marginal cost analysis and a Silvicultural Stand Model were used to estimate the net financial gain from spraying. It was estimated that volume savings of 3-3.5 m³/ha/spray would be needed to break even at an average cost of $22.50/ha/spray. The data suggest that in the experimental plots spraying was not justified on purely financial criteria as a response of this magnitude was not obtained.<

Over a period of 5 years a stand of Pinus radiata D. Don annually sprayed for protection against Dothistroma pini Hulbary showed 6.0 m²/ha more basal area than unsprayed neighbouring stands. This difference arose from losses in diameter increment, as well as mortality, in the unsprayed areas. The growth of dominant trees in the unsprayed areas was not significantly affected. The standard operational treatment, spraying at the 25% level of green crown infection, was only marginally effective in preserving total basal area increment.

Disease assessment of Dothistroma needle blight in stands of Pinus radiata D. Don was evaluated for accuracy. Disease levels were scored visually by estimating the percentage of the normal crown depth infected. Seven independent observers made a tree-by-tree scoring of eight 200-tree transects on the ground and six of the observers assessed the transects from the air; subsequently 11 ground observers assessed two transects of 100 trees and four of the observers re-assessed them 1 week later.

Estimates of variance components for ground assessments showed that appreciable variation was attributable to interactions between observer and tree and the residual error effects. This can be largely eliminated by increasing sample size. The variation caused by change in observer bias in assessments repeated after 1 week was of only minor importance. The effect of observer bias was appreciable but can be reduced by increasing the number of observers. Although the size of the different variance components varied among transects, their relative contribution remained fairly constant. Greater accuracy was achieved by more-experienced observers. Corresponding variance component estimates for aerial assessments were slightly smaller than for ground assessments. Disease ratings of the transects showed good agreement in ranking between ground and aerial assessments.

Accuracy of disease assessments depends on site conditions, observer skill, and sample size and structure. The mean disease level of a 100-tree transect obtained from ground assessments by three observers will have a coefficient of variation of about 10%. An aerial assessment by two observers of a single transect will have a coefficient of variation of about 25% since it involves assigning a single global figure per observer per transect.