We studied the effects of light quality and defoliation on the rate of phytomer appearance and axillary bud outgrowth in white clover. The treatments were applied to one phytomer, a phytomer being defined as the structural unit comprising a node, internode, axillary bud, subtending leaf and two nodal root primordia. Light of a low red:far-red (R:FR) ratio (0.27) was applied to a ''target'' phytomer either (i) within the apical bud and then to the axillary bud after emergence of the phytomer from the apical bud, or (ii) to the axillary bud only after emergence. The Light conditions were directed to these specific parts of the plant by collimating light from small FR light-emitting diodes; with this technique we were able to change the light quality without any change in the level of photosynthetically active radiation. The subtending leaf of the target phytomer was retained or defoliated when it had emerged from the apical bud. FR light applied from the time the phytomer was within the apical bud caused a delay in branch appearance at the target phytomer. In contrast, direct treatment of the axillary bud with FR light after it had emerged from the apical bud did not result in any delay in branch appearance. As the light treatment of the apical bud may have changed the light environment of any of the organs contained in the bud we were unable to ascribe the delay in branch appearance to light perception by any particular organ. However, indirect evidence leads to the conclusion that the likely site of light perception was the developing leaf subtending the axillary bud while it was the outermost phytomer within the apical bud. These results do not support the hypothesis that the R:FR ratio of light incident at an axillary bud site is the environmental factor that controls bud development. Defoliation of the unfolding leaf reduced the rate of phytomer appearance on the main stolen but had no immediate effect on branch appearance. As a consequence there was a reduction in the number of phytomers between the stolen apical meristem and the first phytomer with a branch. This is frequently taken to indicate a relaxation of apical dominance, but in this case was found not to involve a direct effect on bud activity. A current model of white clover growth suggests that there is integration of activity between apical meristems but independence of activity and response to the local micro-environment by axillary buds. In contrast, we found that (i) defoliation reduced phytomer appearance only at the main stolen apical meristem and not at all the meristems in the plant and (ii) that a change in the local light environment of an axillary bud had no discernible effect on bud activity once the bud had emerged from the apical bud but could delay branching if applied before emergence. These results are at variance with the predictions of the model.