To study the relationship between extracellular potassium concentration and renal excretion of potassium, seven chronically adrenalectomized dogs were maintained on a constant intravenous infusion of aldosterone (50 micrograms/day), and constant sodium intake (30 meq/day ) while they received four levels of potassium intake--10, 30, 100, and 200 meq/day--for 7-10 days each. At the conclusion of each level of intake, plasma potassium and renal excretion as well as other variables known to influence potassium excretion were measured. There were minimal changes in arterial pH, mean arterial pressure, extracellular fluid volume, or glomerular filtration rate at any level of potassium intake. The values for plasma potassium and renal potassium excretion attained at each level of intake were: 3.13 +/- 0.24 and 10 +/- 2; 4.18 +/- 0.18 and 21 +/- 6; 4.31 +/- 0.11 and 66 +/- 10; and 4.75 +/- 0.10 meq/liter and 170 +/- 16 meq/day, respectively. Under these experimental conditions in which the levels of aldosterone, sodium intake, arterial pH, arterial pressure, extracellular fluid volume, and glomerular filtration rate remain constant, plasma potassium concentration appears to have a week effect on renal potassium excretion below the normal level of plasma potassium (approx. 11 meq/day change in excretion for each milliequivalent per liter change in concentration). Above the normal level, however, plasma potassium concentration has a powerful effect, 260 meq/day per milliequivalent per liter. The characteristics of the relationship between plasma potassium and renal potassium excretion make it ideally suited for controlling potassium excretion in response to greater than normal potassium intake.