Cerebral blood flow increases at high altitude, but the mechanism of the increase and its role in adaptation to high altitude are unclear. We hypothesized that the hypoxemia at high altitude would increase cerebral blood flow, which would in turn defend O2 delivery to the brain. Noninvasive Doppler ultrasound was used to measure the flow velocities in the internal carotid and the vertebral arteries in six healthy male subjects. Within 2–4 h of arrival on Pikes Peak (4,300 m), velocities in both arteries were slightly and not significantly increased above sea-level values. By 18–44 h a peak increase of 20% was observed (combined P less than 0.025). Subsequently (days 4–12) velocities declined to values similar to those at sea level. At altitude the lowest arterial O2 saturation (SaO2) and the highest end-tidal PCO2 was observed on arrival. By day 4 and thereafter, when the flow velocities had returned toward sea-level values, hemoglobin concentration and SaO2 were increased over initial high-altitude values such that calculated O2 transport values were even higher than those at sea level. Although the cause of the failure for cerebral flow velocity to increase on arrival is not understood, the subsequent increase may act to defend brain O2 transport. With further increase in hemoglobin and SaO2 over time at high altitude, flow velocity returned to sea-level values.