The ability to detect a mistuned component in a harmonic complex was measured as a function of harmonic number using three different stimulus procedures. The harmonic complex and target component were presented either simultaneously to the same ear (simultaneous monotic), successively to the same ear (successive monotic), or simultaneously to different ears (simultaneous dichotic). For harmonic numbers above the third, the thresholds obtained in the simultaneous-monotic procedure were much lower than those obtained in the other two procedures. It is argued that detection in the simultaneous-monotic procedure is primarily based on a cue associated with the lower-frequency components in the power spectrum of the stimulus envelope. The envelope cue seemed to be weakened by either decreasing the stimulus duration from 410 ms or increasing the fundamental frequency from 200 Hz. Thresholds did not differ for successive-monotic and simultaneous-dichotic procedures, suggesting that the detection of the mistuned component was mediated above the level of the cochlea. [Work supported by NIH.]