The effects of arterial pressure on cerebral reactive hyperaemia were studied in anaesthetized goats measuring electromagnetically middle cerebral artery flow and performing arterial occlusions of 5-30 s. Under normotension (mean arterial pressure, MAP = 11 +/- 0.3 kPa), reactive hyperaemia (peak hyperaemic flow to control flow and repayment to debt ratios) increased, and cerebrovascular resistance during peak hyperaemic flow decreased, as ischaemia duration lengthened; the virtual maximal changes were obtained after 20 s ischaemia. During hypertension by aorta constriction (MAP = 18 +/- 0.7 kPa) or by i.v. infusion of noradrenaline (MAP = 19 +/- 0.8 kPa) middle cerebral artery flow did not change significantly and cerebrovascular resistance increased 25 and 46%, respectively (P < 0.05). During both types of hypertension reactive hyperaemia was over 50% higher, and the decrement in cerebrovascular resistance during peak hyperaemic flow was also higher, than under normotension. During hypotension by constriction of the inferior vena cava (MAP = 5 +/- 0.5 kPa) or by i.v. infusion of isoproterenol (MAP = 6 +/- 0.5 kPa), middle cerebral artery flow decreased 35% or did not change, and cerebrovascular resistance decreased 41 and 45%, respectively (P < 0.05). In these conditions, reactive hyperaemia and the decrement in cerebrovascular resistance during peak hyperaemic flow were reduced 80%, and it was similar in both types of hypotension. The absolute levels of cerebrovascular resistance obtained during peak hyperaemia were similar during normotension, hypertension and hypotension. Thus, arterial pressure is a main determinant of postocclusive cerebral reactive hyperaemia, and myogenic mechanisms may be of significance in determining the early stage of cerebral reactive hyperaemia after brief ischaemias. Adrenergic mechanisms might be of minor significance in this type of cerebral reactive hyperaemia.