Cerebral blood flow (CBF) and its reactivity to alterations in arterial carbon dioxide tension were measured in 52 rats. Measurements were made in four groups of rats: 15 were controls, 14 had one carotid artery clamped, 11 had microvascular common carotid arteriotomies, and 12 had sham arteriotomies. Before carotid operation, a flow change of 1 ml/100 g of brain per minute per torr change in CO2 tension was demonstrated in the rats, a value for reactivity similar to that found in normal humans. After unilateral common carotid ligation, CBF was not significantly disturbed in either hemisphere, and reactivity was preserved. In rats subjected to arteriotomy, CBF fell by 30% from the control level in both hemispheres, and carbon dioxide reactivity was reduced by 80%. Very similar reductions in flow (38%) and reactivity (76%) were produced after sham arteriotomy, in which the common carotid artery was isolated between temporary clamps for 30 minutes, but the arteriotomy incision was not made. The observed flow reduction and carbon dioxide reactivity impairment must result from the reperfusion of the brain through the traumatized carotid artery, inasmuch as no impairment of either flow or reactivity followed extravascular dissection and permanent ligation of the vessel. These changes may be mediated by damage to the sympathetic plexus or the innervation of the carotid body chemoreceptors or by the release of histamine, prostaglandins, or other vasoactive substances from the incised vessel wall. If this finding is applicable to humans, it suggests that the entire cerebral circulation in patients with respiratory impairment may be precarious for some hours after a carotid endarterectomy and confirms the danger of hypercapnia during anesthesia for carotid surgery.