This study addresses the question of whether the normal range for distribution of local cerebral blood flow (lCBF) in adults can be transferred to the 4- to 15-year-old age group. Twenty-three children (age: 4-15 years; mean 11+/-3 years, group I) and 10 adults (age: 27-56 years; mean 45+/-10 years, group II) without evidence of cerebrovascular disease or other brain diseases underwent technetium-99m ethyl cysteinate dimer single-photon emission tomography. Counts in cortical and subcortical regions of interest (ROIs) were related to those in cerebellar ROIs (= 100%). Relative cortical activity in group I exceeded that in group II, particularly in left parietal (107.6%+/-9.8% vs 84.1%+/-12.4%), left frontal (97. 7%+/-6.7% vs 79.4%+/-8.9%) and left temporal areas (99.7%+/-7.4% vs 84.9%+/-10.1%) and in the cingulate cortex (112.1%+/-9.1% vs 95. 9%+/-10.1%, P<0.05). Cerebral activity uptake per injected dose was inversely correlated with age in 19 children of group I (r = -0.77, P<0.001). In group I, there was also an inverse correlation between age and the relative local count density in the parietal (r = -0.42 to -0.57), frontal (r = -0.48), temporal (r = -0.42 to -0.58) and occipital cortex (r = -0.44). In these cortical regions relative counts differed when subgroups of children aged 4-10 and 11-15 years were analysed. It is concluded that there are systematic differences between 4- to 15-year-old children and adults with regard to normal lCBF. Diagnostic use of perfusion agents has to consider age-adjusted normal flow maps; normal ranges should be determined separately for the age groups 4-10 and 11-15 years.