Wood preservatives are essential to the continued use of wood in situations where fungal decay and damage from wood boring insects such as termites, would otherwise occur. Significant risks to health and the environment arising from their correct use are unproven but new legislation and consumer trends continue to cause a decline in the use of traditional and highly effective preservatives such as arsenical copper chromates (CCA). It is not certain that wood treated with alternative biocides will perform well in soil, an environment containing diverse decay flora and fauna and an aggressive preservative depletion hazard. The objective of this work was to determine the effects of diverse test site conditions on in-ground performance of Pinus radiata D. Don. and Fagus sylvatica L. treated with new and traditional wood preservatives, determine wood decay type diversity, and test its significance on wood preservative performance and the ability of current decay classification nomenclature to describe it. P. radiata test stakes, 20 x 20 x 500 millimeters, were treated with ground contact retentions of CCA (0.72% mass/mass (m/m) active element), copper-azole(CuAz) (0.59% (m/m) active ingredient (a.i.): 0.57% m/m copper plus 0.02% m/m tebuconazole), an ammoniacal copper quaternary (ACQ) (1.56% m/m (a.i.): 0.87% m/m copper oxide plus 0.69% m/m didecyl dimethyl ammonium chloride (DDAC)), chlorothalonil plus chlorpyriphos in oil (CC) (1.07% (m/m) chlorothalonil) and a 60:40 mixture of creosote plus oil (37% m/m creosote). F. sylvatica stakes were treated with CCA (0.42% m/m a.i.), CuAz (0.54% m/m a.i. Cu) and CC (0.98% m/m chlorothalonil). For pine, 3 - 4 lower retentions of a 1.5 geometric series were also exposed to ensure advanced decay occurred within the 6.5 year exposure period. For beech, 1 - 2 lower retentions of a 2.25 geometric series were exposed for 5.5 years. Eleven New Zealand test sites of highly diverse soil type, vegetation and climate, plus a tropical and sub-tropical site in Australia were used. A Zeiss Axioplan II light microscope (LM) was used to determine decay micromorphology for each treatment (preservative + wood species) at each site and the effects of site, preservative and wood species on preservative performance, decay type and preservative depletion was determined. Suitability of epifluorescent confocal laser scanning microscopy (CLSM) for imaging decay micromorphology on the limits of LM resolution capability was determined. Preservative performance 1. Across 13 sites, pine treated with CuAz (0.59% m/m a.i), ACQ (0.87% m/m CuO plus 0.69% m/m DDAC) and CCA (0.72% m/m a.i.) had mean soundness reduction (MSR) values of 15, 19 and 19% respectively. CuAz performed significantly better (5% level of probability (P)) than ACQ and CCA. 2. 6.5 years exposure is too short for prediction of long-term performance, however over widely varying conditions, including sites with extreme preservative depletion hazards, CuAz treated pine performed consistently. Performance of 0.39% m/m a.i., a retention 14% below the NZS3640 minimum requirement of 0.46% m/m a.i., gave significantly (5% P) poorer performance than the NZS3640 minimum requirement for CCA (0.72% m/m a.i.). 3. Pine treated with CuO plus DDAC (ACQ) at retentions 29 and 64% above theNZS3640 minimum requirements respectively, was susceptible to rapid failure from brown rot at some pastoral sites likely to be encountered in service. In contrast, CuAz treated pine gave superior performance against brown rot at the same sites compared to CCA treated pine. 4. Site had a significant (5% P) effect on preservative performance and a significant site-preservative interaction effect meant that overall site hazard index was not always a reliable indicator of performance, particularly at sites containing aggressive brown rot fungi (7 out of 13 sites). 5. Knowledge of the distribution of different decay types across sites, coupled with known decay hazards and preservative depletion hazard, suggested that 4 test sites of clearly defined features would enable comprehensive field testing of preservative treated wood. Selection of sites is not straightforward and requires detailed knowledge of soil type, geology and climate, or comprehensive knowledge of the decay types present. Preservative depletion 1. Site and wood species had a major effect on copper (Cu), chromium and arsenic depletion from CCA treated pine and beech and depletion of Cu and tebuconazole from CuAz treated pine and beech. Mean Cu depletion for radiata pine treated with 0.72% m/m a.i. CCA after 5.5 years, across 13 sites was less than 1% for above ground portions of stakes compared to 30% for below ground. However, below ground depletions at acidic sites located at a peat bog and a Nothofagus beech forest were 43 and 73% respectively. Mean below ground chromium and arsenic depletions were 9 and 21 % respectively but were 22 and 41 % at the most severe depletion site (Nothofagus forest). 2. Extremes of soil water availability and pH were the most important factors affecting depletion, where waterlogged sites with a high organic matter content (by inference high organic acid concentration) had the most severe depletion hazard. 3. Across all sites above ground depletion of copper and tebuconazole from radiata pine treated with 0.59% m/m a.i. CuAz was 19 and 42% compared to 47 and 55% for below ground. Substantially greater loss of copper from Cu-Az treated wood compared to CCA treated wood, especially for above ground exposure, across all sites, may become significant for wood in service situations where aquatic toxicity of copper is important. 4. Beech was more susceptible than pine to loss of copper for both CCA and CuAz. This may have been attributable to less efficient fixation reactions and preservative distribution (macro- and micro-) in beech. Decay type diversity and classification nomenclature 1. Preservative treated wood was decayed by a complex mixture of decay types except at those sites where brown rot alone caused rapid failure. On the basis of unique mixtures of decay micromorphology, 18 decay types were identified of which 5 fitted within current decay classification nomenclature. 2. Based on mean soundness reduction caused by macroscopic decay types and their associated microscopic decay types, fifty percent of decay across all sites was caused by decay types not described (or inadequately described) by the currently accepted decay types: brown rot, soft rot (SR) (Corbett’s type 1, 2 and diffuse cavitation), white rot (WR) (simultaneous and preferential) and bacterial decay (tunnelling, erosion and cavitation). This was in part attributed to presence of fungi which possessed diverse decay capabilities that produced mixtures of decay micromorphology preventing a neat fit within current classification, a finding that is concurrent with increasing evidence within the literature that basidiomycetes possess multiple decay capabilities. Current decay classification boundaries that place emphasis on taxonomic affinity and decay type expression under laboratory conditions, failed to accommodate the diversity of wood decay types found under field conditions for preservative treated pine and beech. 3. Close association of hyphae possessing clamp connections, with several unusual types of tunnelling and cavitation micromorphology in wood with a colour and texture typical of WR or intermediate between WR and SR, suggested that WR basidiomycetes were responsible for a significant proportion of the decay types observed. 4. Unidentified decay fungi that bypass the decay resistant S3 layer, causing complete concentric removal of the S2 layer by multi-branched hyphae under an overlying lumen wall, were very important in preservative treated beech fibres and pine tracheids. Whilst some of the micromorphology associated with this type of decay has previously been reported as diffuse type 1 SR, its significance has been largely overlooked. 5. New insights into cavitation decay that forms near or at the S3—S2 interface, and where the overlying lumen wall disintegrates early in the decay sequence, has in the past been misdiagnosed as Corbett’s type 2 (erosion) SR. 6. Simultaneous WR and type 2 SR were of little significance in preservative treated pine and beech, suggesting that decay fungi that produce it in other situations (e.g. untreated wood) adopt a tunnelling or cavitation mode in preservative treated woods. This was thought to be a stress response similar to the previously observed responses of basidiomycete decay fungi present in pathological reaction wood that contains allopathic chemicals and physical blockages of wood cell lumenae such as gums and resins. Therefore, latent genes encoding decay capabilities not expressed under laboratory conditions may be highly significant in preservative treated wood in ground contact. 7. Cellulose microfibril orientation in both the S2 and S1 layers determined orientation and positioning of the great majority of decay types, including those previously thought to be essentially random, such as tunnelling bacteria. Contrary to the literature, orientation parallel to, or affected by, the S1 cellulose microfibrils (CM) was of similar significance as alignment with the S2 CM. Position, orientation and limits of expansion of novel types of tunnelling and cavitation in both pine and beech suggested that concentric radial wood cell wall structure (cell lamellae and/or planes of resistance caused by differential lignin content) were important in determination of decay micromorphology. 8. A putative model was derived to account for the high frequency and morphology of concentrically aligned SR cavities, and cavities and tunnels of other decay types. The S1–middle lamella interface triggered formation of tangential hyphal branches followed by non-oscillatory or suppressed oscillatory growth and preferential expansion in the tangential plane. Whilst the S1 triggered tangential cavity orientation, it is too narrow to trigger normal SR oscillatory growth. 9. Protracted staining procedures, the need to embed samples and the time taken to manipulate images using computer assisted 3D modelling software, made epifluorescent CLSM unsuitable for examination of large numbers of samples, as required in ecological studies. Furthermore, the resolving capabilities and depth of field of the LM optics used in this study provided greater insight into the 3D nature of decay micromorphology.