We investigated the role of the exercise-induced elevation of plasma prolactin (PRL) concentration on sweat rate and composition during prolonged exercise in men. Two groups of healthy young males (20-26 yr old) showing a high (high responders; n = 8) or a low (low responders; n = 7) response of plasma PRL concentration to exercise were studied during a 60-min period of exercise on a cycle ergometer (65% maximum O2 consumption) in warm conditions (26.2 +/- 0.1-degrees-C; 57 +/- 1% relative humidity), 1 h after receiving 1.25 mg bromocriptine (BRC) per os or a placebo. In high responders, administration of BRC totally abolished the threefold increase in plasma PRL observed in response to exercise with placebo [placebo, 10 +/- 2 (rest) and 30 +/- 2 mug/l (exercise); BRC, 9 +/- 1 (rest) and 8 +/- 1 mug/l (exercise)]. The latter was associated with a significant decrease in sweat rate (2.7 +/- 0.5 to 1.9 +/- 0.3 mul.cm-2.min-1) and a significant increase in sweat Na+ concentration (57 +/- 7 to 68 +/- 5 mmol/1). BRC also reduced the small response in plasma PRL concentration observed in low responders [placebo, 10 +/- 1 (rest) and 15 +/- 1 mug/l (exercise); BRC, 9 +/- 1 (rest) and 7 +/- 1 mug/l (exercise)], but this was not associated with any change in sweat rate (2.2 +/- 0.2 to 1.9 +/- 0.3 mul.cm-2.min-1) or in sweat Na+ concentration (63 +/- 10 to 64 +/- 9 mmol/l). A significant increase was observed in plasma aldosterone, vasopressin, and atriopeptin concentrations and in plasma renin activity in response to exercise in both groups with placebo as well as with BRC. The inhibition of plasma PRL response by BRC in high responders was associated with slightly larger increase in plasma aldosterone and vasopressin concentrations at the end of the exercise period. Although PRL seems to promote a slight increase in sweat loss and a slight decrease in sweat Na+ concentration, the absence of a clear compensatory response of fluid-electrolyte regulating hormones to the inhibition of plasma PRL response to exercise by BRC suggests that PRL is not a key factor in the control of sweat production and composition, under the experimental conditions used in the present study.