Night shift work is associated with a myriad of health and safety risks. Phase-shifting the circadian clock such that it is more aligned with night work and day sleep is one way to attenuate these risks. However, workers will not be satisfied with complete adaptation to night work if it leaves them misaligned during days off. Therefore, the goal of this set of studies is to produce a compromise phase position in which individuals working night shifts delay their circadian clocks to a position that is more compatible with nighttime work and daytime sleep yet is not incompatible with late nighttime sleep on days off. This is the first in the set of studies describing the magnitude of circadian phase delays that occurs on progressively later days within a series of night shifts interspersed with days off. The series will be ended on various days in order to take a "snapshot" of circadian phase. In this set of studies, subjects sleep from 23:00 to 7:00 h for three weeks. Following this baseline period, there is a series of night shifts (23:00 to 07:00 h) and days off. Experimental subjects receive five 15 min intermittent bright light pulses (approximately 3500 lux; approximately 1100 microW/cm2) once per hour during the night shifts, wear sunglasses that attenuate all visible wavelengths--especially short wavelengths ("blue-blockers")--while traveling home after the shifts, and sleep in the dark (08:30-15:30 h) after each night shift. Control subjects remain in typical dim room light (<50 lux) throughout the night shift, wear sunglasses that do not attenuate as much light, and sleep whenever they want after the night shifts. Circadian phase is determined from the circadian rhythm of melatonin collected during a dim light phase assessment at the beginning and end of each study. The sleepiest time of day, approximated by the body temperature minimum (Tmin), is estimated by adding 7 h to the dim light melatonin onset. In this first study, circadian phase was measured after two night shifts and day sleep periods. The Tmin of the experimental subjects (n=11) was 04:24+/-0.8 h (mean+/-SD) at baseline and 7:36+/-1.4 h after the night shifts. Thus, after two night shifts, the Tmin had not yet delayed into the daytime sleep period, which began at 08:30 h. The Tmin of the control subjects (n=12) was 04:00+/-1.2 h at baseline and drifted to 4:36+/-1.4 h after the night shifts. Thus, two night shifts with a practical pattern of intermittent bright light, the wearing of sunglasses on the way home from night shifts, and a regular sleep period early in the daytime, phase delayed the circadian clock toward the desired compromise phase position for permanent night shift workers. Additional night shifts with bright light pulses and daytime sleep in the dark are expected to displace the sleepiest time of day into the daytime sleep period, improving both nighttime alertness and daytime sleep but not precluding adequate sleep on days off.