The structure of an aggregated soil is characterized by the distribution of the distance from an arbitrary point in the soil to the nearest macropore or crack. From this distribution an equivalent model system is derived to which a diffusion model can be more easily applied. The model system consists of spherical, or cylindrical or plane aggregates, which do not represent the individual aggregates of the soil, however. The radii of the spheres, cylinders or plane sheets represent different length scales occurring in the soil structure and the relative abundance of each radius is expressed as a weight factor. These weight factors are derived by conserving the distance distribution of the soil: the real soil and the model system have the same distance distribution. This implies that the length scales occurring in the two systems are the same and that diffusion processes take place in approximately the same way. A model of diffusion in soil aggregates is evaluated by solving a differential equation for a spherical, a cylindrical or a plane geometry. The overall result for the soil is then found as a weighted sum of the results for the various length scales or radii.In case of an isotropic soil structure, the weight factors for a cylinder system can also be obtained from distances measured in a cross section of the soil. The measurements may be carried out by means of image analysis.The "scale method" is also applied to root systems with a non-regular root distribution. Theoretically calculated nutrient uptakes show good agreement with exact results in the literature obtained with an electrical analogue.For a cracked clayey soil, the crack pattern observed in the field is compared with the distribution of anoxic soil. In case of a uniform soil activity, anoxic soil is expected to occur in the centres of the crumbs only. Part of the anoxic soil, however, was found within 1 mm from a crack. From the measured local diffusivity of oxygen it is concluded that the anaerobiosis near cracks is caused by a local soil activity exceeding the overall soil respiration by at least a factor 100. The presence of highly active and anoxic "hot spots" near macropores or cracks implies that denitrification may take place by the flow of water through the hot spots and without nitrate diffusion.