The endogenous circadian oscillator (the body clock) is slow to adjust to altered rest-activity patterns. As a result, several negative consequences arise during night work and after time-zone transitions. The process of adjustment can be assessed by measurements of the sleep electroencephalogram (EEG), core temperature or melatonin secretion, for example, but these techniques are very difficult to apply in field studies, and make very great demands upon both experimenters and subjects. We have sought to establish if the activity record, measured conveniently and unobtrusively by a monitor attached to the wrist, can be treated in ways that enable estimates to be made of the disruption caused by changes to the rest-activity cycle, and the process of adjustment to them. In Part A, we describe the calculation and assessment of a series of "activity indices" that measure the overall activity pattern, activity when out of bed or in bed, or the activity in the hours adjacent to going to bed or getting up. The value of the indices was assessed by measuring changes to them in subjects undergoing night work or undergoing time-zone transitions. In both cases, there is a large body of literature describing the changes that would be expected. First, night workers (working 2 to 4 successive night shifts) were investigated during rest days and night shifts. The indices indicated that night work was associated with lower activity when the subjects were out of bed and higheut of bed and higher activity when in bed. Some indices also measured when subjects took an afternoon nap before starting a series of night shifts and gave information about the process of adjustment to night work and recovery from it. Second, in studies from travelers crossing six or more time zones to the east or west, the indices indicated that there were changes to the rest-activity cycle immediately after the flights, both in its overall profile and when activity of the subjects in bed or out of bed was considered, and that adjustment took place on subsequent days. By focusing on those indices describing the activity records during the last hour in bed (LHIB) and the first hour out of bed (FHOB), some evidence was found for incomplete adjustment of the body clock, and for differences between westward and eastward flights. In Part B, the battery of indices are applied to the activity records of long-haul pilots, whose activity patterns showed a mixture of effects due to night work and time-zone transitions. Actimetry was performed during the flights themselves and during the layover days (which were either rest or work days). The indices indicated that all pilots had disrupted rest-activity cycles caused by night flights, and that there were added problems for those who had also undergone time-zone transitions. Rest days were valuable for normalizing the activity profile. For those pilots who flew to the west, adjustment was by delay, though not all aspects of the rest-activity cycle adjusted immediately; for those who flew to the east, some attempted to advance their rest-activity cycle while others maintained home-based activity profiles. The indices indicated that the activity profile was disrupted more in those pilots who attempted to advance their rest-activity cycle. We conclude that objective estimates of the disruption caused to the rest-activity cycle and the circadian system can be obtained by suitable analysis of the activity record.