Forty elderly patients, aged 68 to 89 years, with congestive cardiac failure, who were attending a hospital out-patients department, entered an open, parallel group, comparative study of two diuretic combinations, 40 mg frusemide plus 5 mg amiloride per tablet and 0.5 mg bumetanide plus 573 mg slow-release potassium chloride per tablet. Patients were assigned at random to receive one or other combination for 8 weeks, dosage being determined by the severity of the individual patient's condition (range 1 to 3 tablets frusemide/amiloride; 2 to 6 tablets bumetanide/potassium chloride). Clinical assessments, including visual analogue scores for dyspnoea at rest and on effort, and laboratory measurements of serum potassium and magnesium levels were carried out on entry and after 2, 4 and 8 weeks of treatment. Other variables were monitored before, during and/or after treatment. Although significant decreases were reported in dyspnoea severity scores at rest and on effort only in the bumetanide/potassium chloride group, global assessment of the patients' condition by patient and clinician at the end of the study indicated that both treatments produced improvement, and a greater proportion of patients considered treatment as satisfactory in the frusemide/amiloride group. Both drug combinations were well-tolerated and only a few minor side-effects were reported. Serum potassium levels were maintained in both treatment groups but there was a significant decrease in mean serum magnesium levels in patients on bumetanide/potassium chloride. Hyponatraemia was also detected in 2 patients on this combination. An increase in body weight was recorded in both groups, the increase being significant in patients receiving bumetanide/potassium chloride.