Radiata pine (Pinus radiata) is the most important plantation tree in New Zealand forestry, and factors that reduce the quality of wood cause significant economic loss. Two of the most important of these issues are compression wood and spiral grain. Compression wood is a type of reaction wood, formed when a tree moves away from the vertical, and is characterised by biochemical and structural changes within the wood that reduce its quality and value. Spiral grain, however, is the alignment of the wood grain in a helix around the tree’s axis and away from the vertical. Again, this reduces the structural qualities of the wood and thus its value. Spiral grain and compression wood are notorious for their deleterious effect on the quality of wood produced and are very important for the forest industry due to the huge economic loss they cause. The demand for reliable tools to evaluate these wood quality issues in clonal planting material at an early stage, within 3 years of germination rather than at 8 to 15 years as in current practise, is of ever increasing importance from plant breeders and other industry stake holders. Therefore this research was undertaken with an overall aim to develop quick, easy and reproducible techniques to evaluate young radiata pine clones (up to 3 years old) based on compression wood content and presence of spiral grain. This is important because a shortened breeding cycle could provide significant economic benefits to the forest industry. The incidence of these commercially important wood quality parameters has been studied in this thesis in research conducted on young trees (1 to 3 years old). The research described in this thesis used a variety of different imaging approaches to investigate wood structure, including polarised light and confocal microscopy, and X-ray tomography and circular polarised light scanning. The images achieved have been analysed using a range of different software, including Photoshop, ImageJ and Matlab bringing a quantification approach to the imaging. Compression wood was quantified in young clonal material using images collected with a commercial document scanner, and processed using image analysis tools available in Photoshop. An easy, reliable and robust, automatic image analysis protocol was successfully developed and tested for the detection and quantification of compression wood in these young trees. This new technique to detect and quantify compression wood was based on the thresholding of the blue channel of the scanned RGB image as this was demonstrated to contain the greatest image contrast. Development of this new technique may reduce the waiting time for screening clonal planting materials based on compression wood content. To understand the organisation of the grain at a cellular level within these young trees, confocal microscopy techniques were utilised. The cell wall characteristics and fluorescence properties of compression wood in comparison with normal wood were investigated using a new cellulose specific dye, pontamine fast scarlet 4B. Staining protocols for this dye for confocal microscopy were optimised, and the potential of measuring the microfibril angle of the S1 and S3 layers of the pontamine treated opposite wood was demonstrated through either direct observations of these layers, or through the property of bifluorescence where the dye is excited only when aligned parallel to the polarisation of the incident light. Despite extensive work with confocal microscopy, this technique proved to be unsuitable for investigations of spiral grain because although it provided cellular detail, imaging was limited to the surface layers of sections, and the area over which observations were required was prohibitive. Instead of confocal microscopy, the incidence of spiral grain in young stems was investigated in two completely new ways. Resin canals, which are formed from the same cambial initials as the tracheids and which align with the grain, were used as a proxy to demonstrate the grain changes. A novel technique, using circular polarised light and a professional flatbed scanner, was developed to image whole serial transverse sections of the young stems to detect the resin canals. Using ImageJ, the number and location of resin canals was measured on vertical controls, and trees that had been rocked and leaned. The number and frequency of resin canals were less in tilted trees, especially in compression wood, compared to the higher number of canals formed in the rocked trees. More importantly, a combination of serial sectioning and this approach allowed a 3-dimensional view of the orientation of resin canals inside a stem to be generated with ImageJ, and the angles of these canals could be measured using Matlab. The resin canals were oriented with a left-handed spiralling near the stem surface whereas the canals near to the pith were nearly straight, consistent with previous observations of the development of spiral grain in radiata pine. However, it was observed that while vertical trees had a symmetric pattern of grain and grain changes around the stem, this was not the case in tilted trees. In these, the opposite wood often had severe spiral grain visible through formation of twist whereas the compression wood formed on the lower side had bending. Consistent with this, grain associated with compression wood was significantly straighter than in opposite wood. This hitherto unknown link between the incidence of compression wood and spiral grain was investigated and explained on the basis of the characteristics of resin canals in these types of wood. X-ray micro-tomography was also used to investigate resin canals in the stubs from which serial sections were collected. The 3D reconstructions of the resin canals showed exactly the same patterns as observed by polarised light scanning.