The bioavailability of the thiazide diuretic bemetizide from a tablet containing 25 mg of this drug and 50 mg of the chemically unrelated diuretic triamterene was lower than, and significantly different (p less than 0.01) from that from a tablet containing 25 mg bemetizide alone. The mean peak plasma level of bemetizide after administration of the combination tablet (68.3 ng ml-1) was lower than that after administration of bemetizide alone (87.9 ng ml-1), although the times of occurrence of the peak levels were similar. The bioavailability of triamterene from the combination tablet was greater than, but not significantly different from that after administration of a capsule containing 50 mg triamterene alone. The mean peak plasma level of triamterene after administration of the combination tablet (44.6 ng ml-1) was higher than and significantly different (p less than 0.001) from that after administration of triamterene alone (15.7 ng ml-1). Although bemetizide is unstable in urine, measurement of the apparent excretion of unchanged drug in the 24 h post-dose urine (less than 4 per cent of the dose) agreed with the estimate of drug bioavailability from the plasma level data. Less than 2 per cent of the dose of triamterene was excreted unchanged in the 24 h post-dose urine, but the urinary excretion data also agreed with the bioavailability estimates from the plasma level data. The results of this study and those reported in the literature suggest that because of their physicochemical properties, the bioavailability of some thiazides and triamterene needs to be evaluated when new formulations of these drugs are produced. However, with respect to the combination formulation reported in this paper, the difference in bioavailability of the thiazide component did not detectably effect the diuretic activity of the formulation.