ABSTRAa. Cerebral blood flow velocity was recorded for an average of 23 4-min epochs during natural sleep in 11 normal full-term newborn babies. Intracranial pressure, core temperature, and respiration were simultaneously and non-invasively monitored. Sleep state was classified using information from EEG, pattern of respiration, and eye and body movements by a trained observer. From a total of 238 epochs, 66 were considered to occur in quiet sleep, 101 in active sleep, and in 77 the baby was awake, in a transitional state or moving excessively. Slow cyclical variations in cerebral blood flow velocity were observed with a frequency of between 2 and 6 cycles/min, and these were of significantly greater amplitude during quiet sleep (24%) com- CBF is linked to metabolic demand, and it is therefore not surprising that total CBF is about 30% lower during quiet sleep when compared with active sleep in adult humans (1). CBF measured with jugular occlusion plethysmography was also lower in quiet sleep in term newborns (2,3). No difference in CBF between sleep states was shown in a group of 15 preterm infants, who have lower CBF than adults and whose cerebral metabolic needs may be less variable between sleep states (4). Using Doppler ultrasound to monitor the velocity of blood flow in the anterior cerebral artery Jorch et al. (5) demonstrated a lower velocity during quiet sleep than active sleep in a group of preterm and term neonates. Doppler ultrasound is unable to quantitate CBF