We set out to determine the minimum intensity of light able to suppress nocturnal melatonin levels as measured in normal human saliva. Five healthy male volunteers were exposed to light at different intensities (<10, 500, 1000, 2500, and 5000 lux) in a repeated measure design. Suppression of melatonin was dependent on both light intensity and duration of light exposure. Minimum intensities of light suppressing nocturnal melatonin levels were calculated as 393, 366, 339, and 285 lux for exposure durations of 30, 60, 90, and 120 min, respectively. Minimum effective intensity and duration of light exposure showed a linear inverse relationship. These results suggest that less intensity of light than previously reported suffices to suppress melatonin in humans, and that caution is required in interpreting studies using long exposure to dim light as a background condition.