In alley cropping systems, herbaceous strips planted with trees are added into cropped fields. These systems could provide a favorable habitat for many organisms, producing shade in summer and litter in autumn, but their impacts on soil macrofauna are poorly studied, with most samplings limited to one date and therefore one season only. This study investigated the seasonal variations in the spatial distribution of soil macrofauna in a Mediterranean alley cropping plot. We sampled three different shading treatments: a strong shading treatment with lightly-pruned trees, a light shading treatment with pollarded trees, and a no-shade treatment without trees in the herbaceous strips. We measured the density of macrofauna individuals classified into different taxonomic and functional groups by the extraction of 25*25*25 cm soil cores in the herbaceous strip (with or without trees), and in the crop alley at 1 m and 2.5 m from the herbaceous strip. We found a similar response of macrofauna density to the distance from the herbaceous strip for all shading treatments. However, this response was strongly dependent on the season. In spring, just after soil tillage, most soil macrofauna presented higher densities in the herbaceous strip than in the crop alley, except for plant feeders. In early summer, this trend was mostly visible for earthworms, Diplopoda and adult Coleoptera. In autumn, higher densities in the herbaceous strip than in the crop alley were only significant for anecic and endogeic earthworms, while other macrofauna categories presented similar densities at all distances from the herbaceous strip, due to an increase in macrofauna density in the crop alley. These large seasonal variations in spatial distribution of soil macrofauna highlight the importance of temporal repetitions when studying community responses to agroforestry system design and management. The mechanisms by which population densities increased in the crop alley a few months after soil perturbation require further investigation.