1 We have recently developed a new model of transient focal ischaemia in the rat utilising topical application of endothelin-1 to the left middle cerebral artery (MCA). In order to validate this approach the present study assessed the neuroprotective efficacy of the NMDA receptor antagonist dizocilpine (MK-801) in the endothelin-1 model. The anti-ischaemic efficacy of the nitric oxide (NO) synthase inhibitor N -nitro-L-arginine methyl ester (L-NAME) was subsequently evaluated, and contrasted with its efficacy against permanent focal ischaemia, to determine the utility of the endothelin-l model for identification of novel pharmacoprotective agents. 2 MK-801 (0.12 mg kg-' bolus, 108 Ag kg-'h-' infusion i.v., either 1 or 2.5 h pre-transient MCA occlusion (MCAO)) induced hypotension that persisted for approximately 1.5 h so that mean arterial blood pressure (MABP) at the time of MCAO was significantly lower in the 1 h group compared with control (MABP: 86 ± 11, 68 ± 6 and 84± 4 mmHg (mean ± s.d.) for saline, 1 h MK-801 and 2.5 h MK-801 groups respectively). The 2.5 h pretreatment schedule resulted in significant reduction (71%) in the volume of hemispheric damage (assessed 4 h post onset of ischaemia) while the 1 h pretreatment schedule did not (volumes of hemispheric damage: 59 ± 38, 51 ± 51 and 17 ± 28 mm3 for saline, 1 h and 2.5 h MK-801 groups). 3 Thus the considerable neuroprotective effect of MK-801 in the endothelin-1 model of transient focal cerebral ischaemia was highly sensitive to drug-induced hypotension. This result is in contrast to previous studies of permanent MCAO where MK-801-induced hypotension did not compromise its neuroprotective action. 4 L-NAME (3 mg kg-', i.v. 30 min pre-MCAO) moderately, but significantly, reduced (16%) the volume of ischaemic damage 4 h post-permanent MCA occlusion, whereas the 29% reduction in volume of damage achieved in the model of transient focal ischaemia did not attain significance due to the greater variability associated with this model. L-NAME did not significantly alter MABP in either model. 5 The modest neuroprotection achieved with NO synthase inhibition suggests NO is of relatively minor importance as a mediator of neurotoxicity following permanent focal cerebral ischaemia. In addition the comparable efficacy of L-NAME against transient focal ischaemia suggests the presence of reperfusion does not enhance the contribution of NO to neuronal injury in the acute (4 h) phase following a focal ischaemic insult.