1. The role of arachidonic acid metabolites and oxygen radicals in carrageenin-induced rat paw oedema and dermal reverse passive Arthus reaction (RPA) have been investigated. 2. Indomethacin (10 mg kg-1, p.o.) inhibited carrageenin paw oedema when administered 30 min before, but not 2 h after carrageenin. BWB70C (10 mg kg-1, p.o.), a selective inhibitor of 5-lipoxygenase, had no effect whether administered before or after carrageenin. Administration of both indomethacin and BWB70C had no greater anti-inflammatory effect than indomethacin alone. 3. BW755C (20 mg kg-1, p.o.), which inhibits the cyclo-oxygenase and lipoxygenase pathways of arachidonic acid metabolism, or superoxide dismutase-polyethylene glycol conjugate (SOD-PEG, 3000 u, i.v.) inhibited carrageenin paw oedema whether administered either 30 min before, or 2 h after carrageenin. 4. Pretreatment with dexamethasone (0.1 mg kg-1) or colchicine (2 mg kg-1), likewise suppressed carrageenin paw oedema. 5. BW755C (25-100 mg kg-1, p.o.) dose-dependently reduced plasma leakage in the RPA, whereas indomethacin (5 mg kg-1, p.o.) or BWB70C either alone or in combination, did not. 6. SOD-PEG (300-3000 u, i.v.) dose-dependently inhibited plasma leakage in the RPA. In addition, the iron chelator and peroxyl radical scavenger, desferrioxamine (200 mg kg-1, s.c.) also inhibited plasma leakage. 7. Pretreatment with dexamethasone (0.1 mg kg-1) or colchicine (1 mg kg-1) reduced the plasma leakage in RPA, whereas MK-886 (10 mg kg-1) had no effect. 8. These results indicate an important role for oxygen radicals but not arachidonic acid metabolites in the maintenance of carrageenin paw oedema and the plasma leakage in RPA. Furthermore, the results suggest that the anti-inflammatory actions of BW755C can be dissociated from its effects on arachidonic acid metabolism and are attributed to its anti-oxidant activity.