A contralateral signal presented at the same time and at the same frequency and intensity as an ipsilateral signal leads to an apparent sound source in the medial plane. The level of the contralateral signal required to produce this effect (the lateralization level) has been used as an estimator of the effective level of the ipsilateral signal in studies of suppression. The apparent contradictions between the properties of suppression measured with lateralization and other techniques are probably attributable to an oversimplified interpretation of the lateralization data. We examined lateralization level as a function of the temporal position of the binaural signal with respect to an ipsilateral masker for several levels of the ipsilateral component of the signal. We found a reduction in lateralization level associated with the introduction of the masker when the masker onset preceded signal onset by about 25 ms or more. We could not obtain an image in the medial plane by presenting a brief ipsilateral signal during an ongoing contralateral signal. The simple interpretation of lateralization as a measure of the effective level of the ipsilateral signal component fails to account for these data, and we argue that the conclusions reached in some earlier lateralization papers should be reconsidered.