Plasma prolactin response to postural change and variation in dietary sodium was evaluated in five normal volunteers and 15 patients with essential hypertension. Values at 0800 hours (11.9 +/- 3.5 ng/ml, mean +/- 1 S.D.) were uninfluenced by the duration of recumbency (10 or 34 hr) and were significantly higher than those obtained at noon (6.9 +/- 3.5 ng/ml, mean +/- 1 S.D., p less than 0.001). The latter were uninfluenced by postural change. There was no correlation between sodium intake and plasma prolactin, nor was there any apparent correlation between prolactin and plasma renin activity. There was no significant difference in prolactin concentrations between normotensive and hypertensive subjects. In 10 additional patients with unilateral renal disease, renal vein prolactin concentrations did not differ significantly from simultaneously obtained peripheral concentrations. Renal vein prolactin was uninfluenced by the presence of renal disease and did not correlate with renal blood flow. It is concluded that there is no evidence of feedback between sodium intake and prolactin in man. Human kidneys do not seem to clear significant amounts of prolactin. It appears unlikely that alterations in prolactin concentration, at least as assessed by daytime values, participate in the maintenance of either essential or renovascular hypertension. Since values at 0800 hours are frequently elevated as a reflection of preceding sleep-related peaks, sampling at 1200 hours may be preferable when search is undertaken for hypothalamic-pituitary disease.