Two portable light sources, comprising light emitting diodes (LEDs) of two different wavelengths, were compared to a standard light box in suppressing and phase shifting nocturnal salivary melatonin. All light sources were equated for illuminance of 2000 lux. Sixty-six volunteers participated in the 2-day study and were randomly allocated to one of four conditions; light box, white LED, blue/green LED, or no light control group. Light was administered to the experimental groups from midnight to 02.00 hr on the first night. Half-hourly saliva samples were collected from 19.00 to 02.00 hr on night 1 and until 01.00 hr on night 2. Percent melatonin suppression on night 1 and dim light melatonin onset (DLMO) for each night were calculated. The experimental groups showed significant melatonin suppression during light stimulation, with the blue/green LEDs producing the greatest (70%) suppression. There was no significant difference between the light box at 63% and white LED at 50% suppression. Similarly, the blue/green LED had a significantly greater DLMO delay of 42 min and no difference between the light box of 23 min and the white LED of 22 min. These data suggest the portable LED light source is an effective way of delivering light to phase shift the melatonin rhythm, with the blue/green LED being the more effective of the two LEDs.