Abstract. The rate and extent to which biophysical resources are captured and utilized by the components of an agroforestry system are determined by the nature and intensity of interactions between the components. The net effect of these interactions is often determined by the influence of the tree component on the other component(s) and/or on the overall system, and is expressed in terms of such quantifiable responses as soil fertility changes, microclimate modification, resource (water, nutrients, and light) availability and utilization, pest and disease incidence, and allelopathy. The paper reviews such manifestations of biophysical interactions in major simultaneous (e.g., hedgerow intercropping and trees on croplands) and sequential (e.g., planted tree fallows) agroforestry systems.In hedgerow intercropping (HI), the hedge/crop interactions are dominated by soil fertility improvement and competition for growth resources. Higher crop yields in HI than in sole cropping are noted mostly in inherently fertile soils in humid and subhumid tropics, and are caused by large fertility improvement relative to the effects of competition. But, yield increases are rare in semiarid tropics and infertile acid soils because fertility improvement does not offset the large competitive effect of hedgerows with crops for water and/or nutrients. Whereas improved soil fertility and microclimate positively influence crop yields underneath the canopies of scattered trees in semiarid climates, intense shading caused by large, evergreen trees negatively affects the yields. Trees in boundary plantings compete with crops for above-and belowground resources, with belowground competition of trees often extending beyond their crown areas. The major biophysical interactions in improved planted fallows are improvement of soil nitrogen status and reduction of weeds in the fallow phase, and increased crop yields in the subsequent cropping phase. In such systems, the negative effects of competition and microclimate modification are avoided in the absence of direct tree-crop interactions.Future research on biophysical interactions should concentrate on (1) exploiting the diversity that exists within and between species of trees, (2) determining interactions between systems at different spatial (farm and landscape) and temporal scales, (3) improving understanding of belowground interactions, (4) assessing the environmental implications of agroforestry, particularly in the humid tropics, and (5) devising management schedules for agroforestry components in order to maximize benefits.