<jats:p> The importance of nitric oxide (NO) for CBF variations associated with arterial carbon dioxide changes was investigated in halothane-anesthetized rats by using an inhibitor of nitric oxide synthase, N<jats:sup>G</jats:sup>-nitro-l-arginine (NOLAG). CBF was measured by intracarotid injection of <jats:sup>133</jats:sup>Xe. In normocapnia, intracarotid infusion of 1.5, or 7.5, or 30 mg/kg NOLAG induced a dose-dependent increase of arterial blood pressure and a decrease of normocapnic CBF from 85 ± 10 to 78 ± 6, 64 ± 5, and 52 ± 5 ml 100g<jats:sup>−1</jats:sup> min<jats:sup>−1</jats:sup>, respectively. This effect lasted for at least 2 h. Raising P<jats:sub>a</jats:sub>co<jats:sub>2</jats:sub> from a control level of 40 to 68 mm Hg increased CBF to 230 ± 27 ml 100g<jats:sup>−1</jats:sup> min<jats:sup>−1</jats:sup>, corresponding to a percentage CBF response (CO<jats:sub>2</jats:sub> reactivity) of 3.7 ± 0.6%/mm Hg P<jats:sub>a</jats:sub>co<jats:sub>2</jats:sub> in saline-treated rats. NOLAG attenuated this reactivity by 32, 49, and 51% at the three-dose levels. Hypercapnia combined with angiotensin to raise blood pressure to the same level as the highest dose of NOLAG did not affect the CBF response to hypercapnia. l-Arginine significantly prevented the effect of NOLAG on normocapnic CBF as well as blood pressure and also abolished its inhibitory effect on hypercapnic CBF. d-Arginine had no such effect. Decreasing P<jats:sub>a</jats:sub>co<jats:sub>2</jats:sub> to 20 mm Hg reduced control CBF to 46 ± 3 ml 100g<jats:sup>−1</jats:sup> min<jats:sup>−1</jats:sup> with no further reduction after NOLAG. Furthermore, NOLAG did not change the percentage CBF response to an extracellular acidosis induced by acetazolamide (50 mg/kg). The results suggest that NO or a closely related compound is involved in the regulation of CBF in normocapnia and even more so in hypercapnia. </jats:p>