Résumé : Five normal male volunteers were submitted to blood sampling at 15-min intervals during seven 24-h periods staggered over 10 weeks while they underwent successively a westward and an eastward 7-h time shift by jet. Total proteins (TPP), ACTH, and cortisol concentrations were measured in each plasma sample. Possible psychological stresses were minimized by careful selection and adjustment of the volunteers before the beginning of the investigation. Physical facilities and time of the year of the execution of the study were chosen in order to avoid the influence of differences in air temperature and duration of daylight. The westward and eastward travels involved, respectively, periods of 23 and 33 h of sleep deprivation without recumbency. TPP levels were measured on each sample in order to monitor plasma dilution. Rapid fluctuations of TPP of an average magnitude of 12% were observed in all studies independently of posture or travel. They did not result from artefactual dilution due to fluid infusion. Seventy percent of the individual 24-h TPP patterns showed significantly higher TPP concentrations during ambulation than during recumbency, and a circadian rhythm with an acrophase during daytime could be detected in 21 of 34 profiles. These data thus generally support the concept of posture dependence of TPP daily variations. Transient disruptions (abolished or reversed differences between ambulation and recumbency TPP levels and/or absence or reversal of circadian rhythm) of the 24-h TPP pattern was observed in 7 of the 9 profiles obtained 1 day after the flights and was more severe after the eastward shift. It is suggested that the prolonged ambulation periods as well as the change in the type of ambulatory activity involved in commercial airline transportation caused the delay in the resumption of the usual circadian pattern of TPP levels. The time of maximal secretion (acrophase) and the quiescent period of the 24-h pattern of ACTH and cortisol adapted differently to the time shifts, suggesting that the various components of the pituitary-adrenal periodicity may be under different controls. Partial shifts of the acrophase toward the new clock time occurred as early as 1 day after travel in both directions, and the synchronization of the acrophase was complete 10 days after both westward and eastward flights. In contrast, the quiescent period needed at least 3 weeks to adapt to Chicago time (alterations consisted of desynchronization and fragmentation) but had returned to normal on the 11th day after arrival in Brussels. 'Jet lag' failed to produce quantitative secretory alterations, since no significant changes were observed in the 24-h mean levels, the amplitude of the circadian rhythms, or the frequency or global magnitude of episodic fluctuations of both hormones. Disruption in the sleep patterns, with an increase in rapid eye movement sleep, and subjective psychological discomfort, rated by Hamilton's anxiety and depression scales, were highly significant after the eastward flight only. In contrast to the persistent disruption in the pituitary-adrenal periodicity, sleep and psychological indexes had returned to normal when recorded 11 days after flight. Although no consistent correlation was found between the disturbances in the pituitary-adrenal periodicity and the level of psychological discomfort, subjects with the highest and lowest scores on the Hamilton scales showed, respectively, the slowest and fastest adaptation of the cortisol acarophase to the new clock time.