Many GABAergic anaesthetics reduce gap junction coupling but it is currently unknown whether this effect contributes to anaesthetic anticonvulsant action. In this study we examined the possible role of connexin36 gap junctions in the anticonvulsant action of isoflurane and compared this to etomidate, an anaesthetic known for having proconvulsant effects. We compared the effect of anaesthetic concentrations of isoflurane (1 MAC) and etomidate (16 μM) on low-magnesium-induced interictal-like activity in isolated neocortical slices. The effect of connexin36 gap junction blockade was explored by comparing effects in slices from wild-type mice and from a transgenic mouse strain lacking the gene for connexin36. In slices from wild-type mice, both isoflurane (1 MAC) and etomidate (16 μM) reduced interictal-like event frequency; mean(S.D.) reduction of 44(13)% (P < 0.0001) and 25(24)% (P < 0.0001), respectively. The reduction in event frequency was greater for isoflurane (P < 0.005). Isoflurane had no effect on the amplitude of interictal-like events, but event amplitude was enhanced by etomidate (18(28)% increase, P < 0.005). The capacity for isoflurane to reduce event frequency was significantly reduced, but not eliminated in slices from connexin36 knock-out mice (33(15)% reduction, P < 0.05 for the difference with wild-type), while that of etomidate remained unchanged (23(39)% reduction). The etomidate-mediated increase in event amplitude was eliminated in connexin36 knock-out slices. The results from this study support the hypothesis that the anticonvulsant effect of isoflurane is in part mediated by gap junction blockade. The role of gap junction modulation by etomidate is more complicated and may be important in the mechanism of action of etomidate's proconvulsant effects.