The absorption of three combination formulations of hydrochlorothiazide and either triamterene or amiloride was studied over a 5-year period in seven separate investigations under varying conditions of food and fasting. The most widely prescribed combination, containing 25 mg of hydrochlorothiazide and 50 mg of triamterene, demonstrated impaired absorption in the fasting state that was partially corrected by the addition of a breakfast high in fat. The increase in the fat content of the food appeared to correlate directly with the amount of both drugs absorbed from this formulation. The second formulation studied, a new combination formulation of 50 mg of hydrochlorothiazide and 75 mg of triamterene, demonstrated acceptable absorption in the fasting state that was not altered by the concurrent administration of a high-fat breakfast. The absorption of the third formulation, a combination of 50 mg hydrochlorothiazide and 5 mg amiloride, was acceptable in the fasting state and demonstrated a slight reduction in the absorption of the amiloride component when administered concurrently with a high-fat meal. The clinical and biopharmaceutic implications of these observations are discussed.