It has recently been proposed that in rat models of genetic hypertension, supplemental dietary potassium preserves release of endothelium-derived relaxing factor independently of its capacity to either attenuate hypertension or increase plasma potassium. To test this hypothesis in Dahl salt-sensitive rats given sodium chloride (4%) for 3 weeks, we supplemented dietary potassium (2.1%) with either KCl (n = 16) or KHCO3 (n = 16). Compared with unsupplemented rats (n = 16), rats supplemented with either potassium salt had a lower mean arterial pressure and a greater release of endothelium-derived relaxing factor, as assessed from acetylcholine-induced relaxation of precontracted aortic rings. However, the maximum relaxation response to acetylcholine correlated inversely with blood pressure (r = -.82, P<.001), not only in the KC] (r = -.68, P<.002) and KHCO3 (r = -.77, P<.001) groups but also in unsupplemented rats (r = -.86, P<.001). With potassium supplementation, plasma potassium concentrations measured between 4 and 6 PM did not increase, but those measured between 4 and 6 AM did increase (P<.05). In isolated ring segments, aortic compliance was greater in both the KCl and KHCO3 groups than in unsupplemented rats (0.015 and 0.017 vs 0.009 mm2/mm Hg) (P<.01). This greater compliance could not be related to differences in blood pressure, plasma potassium, or collagen or elastin content of the aortic wall. In salt-loaded Dahl salt-sensitive rats in which supplemental dietary potassium attenuates hypertension, (1) the extent to which release of endothelium-derived relaxing factor is preserved is directly related to the extent to which potassium restricts the salt-induced rise in blood pressure, (2) nocturnal but not diurnal plasma potassium is increased, and (3) aortic compliance is enhanced.