In studying the metabolic effects of diet potassium (K+) variation in normal humans, we noted that varying diet K+ within its normal range influenced inorganic phosphorus (Pi) homeostasis and serum calcitriol (1,25-dihydroxyvitamin D) levels. In six men who ingested a constant whole-foods diet containing (per 70 kg body wt) 27 mmol/day Pi and 52 mEq/day K+, we increased diet K+ to 156 mmol/day with supplements first of potassium bicarbonate (KHCO3) alone and then of potassium chloride (KCL) alone, each for eight days interrupted by an eight-day recovery period of no K+ supplement. Urine Pi decreased promptly with either K(+)-salt, each inducing a persisting retention of 7 to 10 mmoles Pi, which was dumped during recovery. Fasting serum [Pi] increased with either K+ supplement (P = 0.022, repeated measures analysis of variance); the composite mean serum [Pi] for the two K(+)-supplement periods exceeded that for the two periods without supplements (P less than 0.01, paired t-test). Conversely, the concentrations of serum calcitriol decreased with either K+ supplement (P = 0.020). Among subjects, the diet K(+)-induced increases in serum [Pi] correlated with those in plasma [K+] (r = 0.64, P = 0.027); the decreases in serum calcitriol concentration correlated with the increases in serum [Pi] (r = -0.69, P = 0.014). There were no significant differences among periods in serum parathyroid hormone, ionized calcium, urine cyclic AMP excretion, plasma renin activity, body weight, serum albumin, or creatinine clearance; plasma volume decreased slightly during KCL but not during KHCO3 periods.(ABSTRACT TRUNCATED AT 250 WORDS)