A transfontanellar range-gated ultrasound Doppler technique for recording blood flow velocity in an artery on the base of the skull was validated in eight anesthetized newborn lambs during hypo-, normo-, and hypercarbia. Blood flow velocity was linearly related to PaCO2 from 20 to 80 mm Hg; mean blood flow velocity (Vmean) (r = 0.86, p less than 0.001), peak systolic blood flow velocity (r = 0.83, p less than 0.001), and end-diastolic blood flow velocity (r = 0.87, p less than 0.001). Vmean changed 2.0% per mm Hg of PaCO2. A linear relationship was demonstrated between brain blood flow (BBF), as determined by the microsphere method, and PaCO2 (r = 0.91, p less than 0.001), with BBF changing 3.6%/mm Hg of PaCO2. Blood flow velocity was linearly related to BBF in the PaCO2 range studied; Vmean (r = 0.89, p less than 0.001), peak systolic blood flow velocity (r = 0.87, p less than 0.001), and end-diastolic blood flow velocity (r = 0.87, p less than 0.001). However, Vmean predicted only approximately 55% of the change in BBF, which suggests a concomitant change in the cross-sectional area of the artery being studied. Despite this limitation, these data suggest that blood flow velocity, recorded by a transfontanellar range-gated Doppler technique from one of the two main arteries perfusing the brain, provides qualitative information on changes in BBF.