A total of 39 subjects were studied after a flight from the UK to either Sydney or Brisbane (10 time-zones to the east). Subjects varied widely in their age, their athletic ability, whether or not they were taking melatonin, and in their objectives when in Australia. For the first 6 days after arrival, subjects scored their jet-lag five times per day and other subjective variables up to five times per day, using visual analogue scales. For jet-lag, the scale was labelled 0 = no jet-lag to 10 = very bad jet-lag; the extremes of the other scales were labelled - 5 and + 5, indicating marked changes compared with normal, and the centrepoint was labelled 0 indicating 'normal'. Mean daily values for jet-lag and fatigue were initially high (+ 3.65 +/- 0.35 and + 1.55 +/- 0.22 on day 1, respectively) and fell progressively on subsequent days, but were still raised significantly (p < 0.05) on day 5 (fatigue) or day 6 (jet-lag). In addition, times of waking were earlier on all days. By contrast, falls in concentration and motivation, and rises in irritability and nocturnal wakings, had recovered by day 4 or earlier, and bowel activity was less frequent, with harder stools, on days 1 and 2 only. Also, on day 1, there was a decrease in the ease of getting to sleep (- 1.33 +/- 0.55), but this changed to an increase from day 2 onwards (for example, + 0.75 +/- 0.25 on day 6). Stepwise regression analysis was used to investigate predictors of jet-lag. The severity of jet-lag at all the times that were measured was strongly predicted by fatigue ratings made at the same time. Its severity at 08:00 h was predicted by an earlier time of waking, by feeling less alert 30 min after waking and, marginally, by the number of waking episodes. Jet-lag at 12:00 and 16:00 h was strongly predicted by a fall of concentration at these times; jet-lag at mealtimes (12:00, 16:00 and 20:00 h) was predicted by the amount of feeling bloated. Such results complicate an exact interpretation that can be placed on an assessment of a global term such as jet-lag, particularly if the assessment is made only once per day.