Field experiments were carried out under unstressed conditions of soil water during two summer crop growing seasons (1998-99 and 1999-2000 seasons) in a South African semi-arid region (Bloemfontein, Free State, South Africa). The aim of this study was to investigate shade effects on beans intercropped with maize in terms of plant mass and radiation use. The experimental treatments were two cropping systems (no shading/sole cropping and shading/intercropping) and two row orientations (northsouth and east-west). At the top of bean canopies shaded by maize, incident radiation was reduced by up to 90 %. Shading reduced total dry matter of beans by 67 % at the end of the growing season, resulting in yield losses. The dry matter partitioning into leaf and stem (the ratios of leaf and stem to total biomass) was about 50 % higher in intercropping than sole cropping. In contrast, intercropped beans had 40 % lower dry matter partitioning into pod (the ratio of pod to total biomass). Fraction of radiation intercepted by sole-cropped beans steeply increased until canopy closure (0.9) and then slowly decreased, while fraction of radiation intercepted by intercropped beans remained constant between 0.0 and 0.2 throughout the growing seasons. However, intercropped beans had 77 % higher radiation use efficiency (RUE) than sole-cropped beans. In contrast, for maize, no effect of intercropping (shading) was found on growth, partitioning, yield, radiation interception or RUE. Consequently, lower bean yield losses can be attained in association with late shading rather than early shading. This can be controlled by growing crops with different temporal and spatial treatments. As regards row treatment, no effect of row direction was found on growth, partitioning, yield, radiation interception or RUE.