Ligustrazine, the synthesized principle of a Chinese herbal remedy shown previously to be a pulmonary vasodilator, was tested in chronically hypoxic and normal rats. Pressure/flow, (P/Q), relations were measured in isolated perfused lungs during normoxia, hypoxia and after reversal of hypoxic vasoconstriction by increasing doses of ligustrazine. P/Q lines were linear over a wide range and extrapolation to the pressure axis gave an intercept which was the effective downstream pressure for flow. In chronically hypoxic rats the slope of the line was steeper and the intercept greater than in control rats, which we attributed to newly muscularized arterioles with tone. Hypoxia caused an increase in slope and intercept in both groups but the intercept increase was greater in chronically hypoxic rats. In both groups of rats increasing doses of ligustrazine given during continued hypoxia caused a fan of lines which moved progressively towards the control normoxic line. In chronically hypoxic rats it required only 2 mg of ligustrazine to bring the line back to the normoxic position, whereas in controls it required 4 mg. In chronically hypoxic rats the change in intercept with every dose was greater than in control rats; this suggests that ligustrazine mainly relaxes the muscle of small collapsible vessels. The action of ligustrazine remained in both control and chronically hypoxic rats after administration of an arginine analogue which blocks synthesis of the endothelial relaxant factor nitric oxide. This and previous evidence suggest that ligustrazine is a non-endothelial-dependent pulmonary vasodilator.