The effect of glucagon on renal haemodynamics in sheep is controversial. In this study we have examined the effects of perfused glucagon on renal blood flow (RBF) in six conscious sheep bilaterally implanted with transit-time ultrasonic flow metering probes around the renal arteries. Glucagon was perfused intravenously over 90 min at doses of 3.12, 6.25, 12.5, 25, 50 and 100 ng kg-1 min-1. Mean RBF was calculated over 10 min periods. Blood samples were taken to monitor the time course of the changes in glycaemia and glucagonaemia. The perfusions of glucagon induced rapid and progressive dose-dependent increases in RBF (9-19.2 %, P < 0.05) and glycaemia (29-155 %, P < 0.05) for doses of 25-100 ng kg-1 min-1. High positive correlations were found between the increases in RBF and glucagonaemia (R2 = 0.95) and between the increases in RBF and glycaemia (R2 = 0.96). At the lowest doses of glucagon (3.12-12.5 ng kg-1 min-1), the increase in RBF was highly significant; however, the rise in glucose level was not. At the highest doses (25-100 ng kg-1 min-1) the time course of the changes in RBF was parallel to that of glucagonaemia throughout the perfusion time. However, between minutes 45 and 90 of the glucagon perfusion, the increase in RBF was the inverse of the change in glycaemia, which decreased. One hour after the end of the 50 and 100 ng kg-1 min-1 perfusions, the glucose levels were still significantly higher than the baseline, while the RBF values were not. These results are consistent with the idea that the enhanced RBF cannot be attributed to a rise in blood glucose level. They also show that the haemodynamic response to glucagon perfusion was more sensitive than the metabolic response. It is concluded that the intravenous perfusion of physiological doses of glucagon induced a highly sensitive dose-dependent increase in RBF in sheep. Experimental Physiology (2003) 88.5, 575-580.