The aim of the present study was to determine whether the recording of chemosensory evoked potentials (CSEP) in healthy subjects (n = 11) can be helpful in differentiating the olfactory or trigeminal component possessed by odorants. By recording from several positions on the surface of the skull it was attempted to ascertain whether different generators are responsible for CSEP associated with the different sensory components of odorants. Birhinal stimulation was used in order to establish an interaction between the stimulated side and the stimulated sensory channel. The four substances carbon dioxide, menthol, hydrogen sulphide and vanillin were tested. EEG was recorded from eight positions.The CSEPs' topographical distribution revealed differences in the location of maximum amplitudes following stimulation with different types of stimulants. Largest amplitudes always appeared at the vet-tex when trigeminal stimulants (menthol, carbon dioxide) were presented, whereas olfactory substances (vanillin, hydrogen sulphide) elicited maximal amplitudes at parietal and central sites. This suggests that at least two neuronal populations are involved in the cortical generation of CSEP. Another interesting finding was that the evoked potentials differed in relation to the stimulated side. Generally, responses to carbon dioxide, menthol and hydrogen sulphide had shorter latencies and smaller amplitudes after stimulation of the left nostril. In contrast, after stimulation with vanillin latencies were shorter and amplitudes tended to be smaller after stimulation of the right side. Since vanillin was the only substance which always evoked pleasant and positive associations, it was assumed that the differences in CSEP after stimulation of the two nostrils are related to the different processing of emotional information within the two hemispheres.