Cerebral blood flow (CBF) was measured by means of Celabeled microspheres in infant (20-day-old) and adult (3-month-old) rats, anesthetised with Na-5-ethyl-5-(1-methylpropyl)2-thiobarbituric acid. Cerebral arteriovenous differences of acetoacetate, D-beta-hydroxybutyrate, glucose, lactate, and oxygen and brain DNA content were determined in other groups of similarly treated infant and adult animals fed or starved for 48 or 72 hr. The mean CBF values of 0.48+/-0.04 and 0.62+/-0.07 ml/(g X min), +/- SEM, in infant and adult animals, respectively, were not significantly different. CBF was unaffected by starvation. At any given arterial concentration the cerebral arteriovenous difference of acetoacetate was significantly higher in infant than adult rats. The same was true for D-beta-hydroxybutyrate at arterial concentrations above 1 mmol/liter. There was an approximately linear relationship between arterial concentration of acetoacetate and its cerebral arteriovenous difference in both infant and adult rats. A similar relationship was found for D-beta-hydroxybutyrate only in infant animals. In the fed state, the cerebral uptake of glucose and ketone bodies (micromoles per (mg DNA X min)) was not different in infant and adult rats. During starvation, cerebral uptake of ketone bodies expressed as micromoles per (mg DNA X min) was higher in infant than adult rats, indicating a higher rate of utilization of ketone bodies per cell in these animals. For glucose, no such difference was found in either fed or starved groups (Table 3). The average percentage of the total cerebral uptake of substrates (micromoles per min) accounted for by ketone bodies increased in both infant and adult rats during starvation. This percentage value was clearly higher in infant than adult rats during starvation. After 72 hr of starvation the values were 38.8% and 15.2% in infant and adult rats, respectively (Fig. 3). Calculated cerebral metabolic rate for oxygen (CMRO2), assuming complete oxidation of glucose and ketone bodies and expressed as micromoles per (mg DNA X min), was similar in fed and starved rats of both age groups (Table 3), indicating that ketone bodies serve as an alternative substrate for glucose during starvation. Calculated CMRO2 for glucose plus ketone bodies was similar to the measured CMRO2 in adult rats both in the fed and the starved groups. For infant rats, calculated CMRO2 for glucose plus ketone bodies was higher than measured CMRO2, indicating that in this age group a portion of substrate was used for synthesis or storage rather than for complete oxidation.