This study evaluates the effects of respiratory motion on breast radiotherapy delivered using segmented multileaf collimator (SMLC) intensity modulation. An anthropomorphic breast phantom was constructed of polystyrene plates between which radiographic films were inserted. The phantom was mounted on a moving platform to simulate one-dimensional sinusoidal oscillation with variable amplitude and frequency. The motion effect on two breast IMRT techniques, a beamlet-based plan created using the Corvus treatment planning system and an aperture-based plan, was evaluated via film comparison. Motion-induced differences in the treatment region are generally within , with the exception of the posterior field edge and the apex of the breast in the Corvus IMRT plan. Considering the experimental uncertainty arising from the setup and film dosimetry, this result indicates that respiratory motion-induced dose variations are generally relatively insignificant. It appears that the anterior hot spots observed in the Corvus IMRT plan result from the high intensity fluence delivered to the “virtual bolus” area which must be created during the planning process in order to provide “flash” for the respiratory motion. The potential magnitude of such effects resulting from the interplay between fluence modulation and patient motion are unique to the individual planning system and planning technique, as well as the delivery equipment and technique. Such effects should be carefully investigated prior to the implementation of IMRT for breast radiotherapy.