It has been suggested that administration of a cannabinoid CB(1) (SR141716A ¿N-(piperidin-1-yl)-5-(4-chlorophenyl)-1-(2, 4-dichlorophenyl)-4-methyl-1-H-pyrazole-3-carboxamide) and CB(2) (SR144528 ¿N-[(1S)-endo-1, 3, 3-trimethyl bicyclo ¿2.2.1 heptan-2-yl]-5-(4-chloro-3-methylphenyl)-1-(4-methylbenzyl)-pyr azo le- 3-carboxamide¿) receptor antagonists to mice potentiates inflammatory hyperalgesia by removing an endogenous cannabinoid tone. We examined whether the behavioural response to s.c. formalin injection in rats is similarly enhanced. A total of 30 animals received SR141716A (0.5 or 5 mg/kg) or SR144528 (0.3 or 3 mg/kg) 30 min before 1% formalin. Pain behaviour was quantified using the composite weighted pain score technique (CPS-WST(0,1,2)). An overall CPS-WST(0,1,2) was calculated for each phase and groups were compared (analysis of variance). The results obtained in the control group confirmed the characteristic biphasic behavioural response to formalin injection. None of antagonist groups had a significant increase in overall CPS-WST(0,1,2) compared to the control. Indeed, a significant decrease in CPS-WST(0,1,2) scores for both phases was detected in most of all of the groups, except SR141716A at 5 mg/kg. Levels of endogenous cannabinoids (anandamide, palmitoylethanolamide, 2-arachidonylglycerol) were measured from rats hind-paw skin 1 h after s.c. injection of 0.9% saline (100 microl), 1% (50 microl), 2. 5% (50 microl) and 5% (100 microl) formalin. The concentration of endocannabinoids did not differ between control and formalin-induced inflammation groups. The activity of anandamide amidohydrolase in hind-paw skin also did not change after treatment with formalin. In conclusion, cannabinoid antagonists do not enhance formalin-evoked pain behaviour. These results suggest that, in this model, endogenous cannabinoids do not tonically attenuate inflammatory hyperalgesia.