Cutaneous vibration is able to reduce both clinical and experimental pain, an effect called vibratory analgesia. The traditional explanation for this phenomenon is that it is mediated by lateral inhibition at the segmental (spinal cord) level, in pain-coding cells with center-surround receptive fields. We evaluated this hypothesis by testing for two signs of lateral inhibition-namely (1) an effect of the distance between the noxious and vibratory stimuli and (2) an inhibition-induced shift in the perceived location of the noxious stimulus. The experiment involved continuous ratings of the pain from pressure applied to the back of a finger, alone and in the presence of vibration delivered to sites on the palm of the hand both near to and far from the site of painful stimulation. Neither prediction of the segmental hypothesis was supported. There was also little evidence to support the view (widely held by subjects) that distraction is the primary mechanism of vibratory analgesia. The results are more consistent with a recently proposed theory of interactions between two cortical areas that are primarily involved in coding pain and touch, respectively.