The application of optical densitometry in the study of wood structure and Properties

SUMMARY The importance of density as an indicator of wood quality, and the need to measure density rapidly, accurately and on a large scale, have led to the development of the techniques described in this paper. The principles of X‐ray radiography on which the method is based are described briefly. Problems that have arisen in the development of the techniques are discussed, together with the solutions proposed for them. The routine procedure now used at the Department of Forestry, Oxford, for X‐ray radiography of wood specimens, and for examination of radiographs by optical densitometry, is explained. A system for the sorting and presentation of data for the application is given in outline, together with a note on the statistical techniques that have been found useful for analysis of data and interpretation of results. Of the structural features and physical and chemical properties that determine the behaviour of wood in manufacturing processes and its characteristics in use, density is the most important. Density is closely related to physical properties, such as hardness and mechanical strength, it gives a good estimation of the yields likely to be obtained in pulping, and it may be a useful indicator of behaviour in drying and resistance to cutting and machining. The wood technologist and the wood user have to take account of the influence of density in several ways. There is a very great range of variation between different species of trees in their mean values for density. This is a primary consideration in the selection of species for a particular purpose, and in general is an advantageous situation, in that tree species can be found for a wide range of applications. When a species has been selected as suitable for a particular purpose, there may still be problems concerning density, if there is considerable variation in density within trees, particularly from the pith outwards, or if there is considerable variation between trees in mean values for density. To make this point more precisely, it has been demonstrated, with some species, that differences in density for mean values of trees from the same site are greater than the differences for mean site values, over a very wide range of geographical distribution and of climatic and soil types. The variation in density within the stem of individual trees, although this is not generally recognized, may be an even more serious matter for concern with some species, particularly in fast grown plantation crops, where the wood near the pith may be less than half of the density of the wood near the bark. Because of the occurrence of considerable variability in wood density, some species are regarded as unsuitable for purposes where the specifications are exacting, for example for some constructional purposes. With many species, however, it should be possible to minimize the adverse effects of within‐tree and between‐tree variability, by appropriate silvicultural treatment of the crops and by selection of trees that have an inherently more uniform pattern of development of density. It is this possibility that has provided the incentive to find and to improve the methods for the measurement of wood density that are described in this paper. If silvicultural treatments for the improvement of wood quality are to be applied on a worthwhile scale, then methods are required by which large numbers of specimens can be accurately and rapidly measured, and that also facilitate the estimation and comparison both of whole tree values for density and within‐tree patterns of the development of density.