Previous experimental findings have led to the suggestion that guanosine 3',5'-cyclic monophosphate (cGMP) plays a permissive role in hypercapnic cerebral vasodilation. However, we recently reported that the technique used to reveal a permissive role for cGMP [cGMP repletion in the presence of nitric oxide synthase (NOS) inhibition] created a situation where CO₂ reactivity was normalized but where different mechanisms (i.e., K⁺ channels) participated in the response. In the present study, we examined whether that nascent K⁺-channel dependence is related in any way to an increase in the influence of the miconazole-inhibitable cytochrome P-450 epoxygenase pathway. Using intravital microscopy and a closed cranial window system in adult rats, we measured pial arteriolar diameters during normo- and hypercapnia, first in the absence and then in the presence of a neuronal NOS (nNOS) inhibitor [7-nitroindazole (7-NI)]. This was followed by suffusion of a cGMP analog and then cGMP plus miconazole. Separate groups of rats were used to evaluate whether miconazole either alone or in the presence of 8-bromoguanosine 3',5'-cyclic monophosphate (8-BrcGMP) or its vehicle (0.1% ethanol) had any effect on CO₂ reactivity and whether miconazole affected K⁺-channel opener-induced dilations. Hypercapnic (arterial PCO₂, 65 mmHg) pial arteriolar dilations, as expected, were reduced by 70-80% with 7-NI and restored with cGMP repletion. CO₂ reactivity was again attenuated after miconazole introduction. Miconazole, with and without 8-BrcGMP, and its vehicle had no influence on pial arteriolar CO₂ reactivity in the absence of nNOS inhibition combined with cGMP repletion. Miconazole alone also did not affect vasodilatory responses to K⁺-channel openers. Thus present results suggest that the nascent K⁺-channel dependence of the hypercapnic response found in our earlier study may be related to increased epoxygenase activity. The specific reasons why the pial arteriolar CO₂ reactivity gains a K⁺-channel and epoxygenase dependence only under conditions of nNOS inhibition and cGMP restoration remain to be identified. These findings again call into question the interpretations applied to data collected in studies evaluating potential permissive actions of cGMP or NO.