This study approaches wood quality in young trees by very early screening – and consequent selection for propagation – on the basis of physical and mechanical properties. In chapter 1 corewood properties are reviewed and the importance and problems associated with early screening are discussed. Due to randomly distributed reaction wood in young trees it is advantageous to lean trees to avoid intermixing of the two wood types and minimise any uncertainty in the results. In chapter 2 physical and mechanical properties are described for opposite and compression wood in a population of Pinus radiata comprising of 50 families, at a young (<3 years) age. The dynamic stiffness was determined using the resonance acoustic technique. Density was measured using water displacement method, and longitudinal and volumetric shrinkage were measured from green to ~5% moisture content. The compression wood and opposite wood differ significantly in all the measured properties. Compression wood was characterised by high density and high longitudinal shrinkage. The mean stiffness of opposite wood was 3.0 GPa with a mean standard deviation of 0.39, and the mean longitudinal shrinkage of opposite wood was 0.99% with mean standard deviation of 0.31 across the samples examined. This variation in stiffness and longitudinal shrinkage in opposite wood can be exploited to screen for wood quality. The variation in stiffness and longitudinal shrinkage within a family was comparable to variation among families. In spite of large within site variability it was possible to distinguish between the worst and the best families in opposite wood at young age. In chapter 3 ranking of selected families of Pinus radiata was done based on microfibril angle, which is considered as the main factor influencing both stiffness and longitudinal shrinkage. The ranking was compared with ranking done using acoustic velocity which is more practical and fast method of screening trees. The mean MFA in opposite wood was 39° with a mean standard deviation of 3.7 and in compression wood the mean MFA was 44° with a mean standard deviation of 2.9. The variation in MFA in opposite wood offers opportunities to breed for trees with low MFA. A strong negative correlation (R=-0.68) between acoustic velocity squared and MFA in opposite wood suggested that the resonance technique can be used effectively to screen very young wood rather than using MFA. At high MFA, the cell wall matrix also plays an important role in determining the mechanical and physical properties of the wood. At present the chemical composition of wood samples is determined by wet chemical analysis, which is time consuming and laborious. Therefore, it is impractical to characterise large numbers of samples. Mechanical properties, particularly tanδ (dissipation of energy), which changes with temperature and frequency as the structure of the material changes at the molecular level, was studied using dynamic mechanical analysis (DMA). The idea was to assess if it can be used as a quality trait for tree screening instead of wet chemical analysis. Compression wood and opposite wood were characterised for storage modulus and tanδ at constant moisture content. In practice the instrument used, TA instrument Q800, was unable to provide the desired range of temperature and humidity so no glass transition at 9% moisture content in the temperature range of 10°C to 85°C at 1 and 10 Hz frequency was observed that might be attributed to the hemicelluloses (or lignin). In spite of the huge difference in chemical composition of opposite and compression wood, the difference in their mean tanδ at 25°C and 1 Hz values was just 7%. The positive correlation between MFA and tanδ in opposite wood suggested that MFA also plays a role in the dissipation of energy. The strong relationship between storage modulus and dynamic modulus (R=0.74) again justifies the reliability of resonance technique to screen young wood for stiffness. Concurrently eighty seven, two-year-old leant Eucalyptus regnans were studied for growth strains along with other physical and mechanical properties, independently in tension and opposite wood. The leant trees in Eucalyptus regnans vary in their average growth strain. Strong correlation between measured and calculated strain (R=0.93) suggests that the quick split method can be used to screen large populations for growth stresses. Tension wood was characterised by high density and was three times stiffer than opposite wood and twice as high in volumetric shrinkage. The high longitudinal shrinkage in opposite wood could be due to comparatively high MFAs in opposite wood of the young trees. There was no correlation between growth strain values and other measured properties in opposite wood. It is possible to screen for growth strain at age two, without any adverse effect on stiffness and shrinkage properties.