Positron emission tomographic (PET) measurements of regional cerebral blood flow (rCBF) with intravenously administered 15O-labeled water and an adaptation of the Kety autoradiographic model are well suited to the study of functional-anatomical correlations within the human brain. This model requires arterial blood sampling to determine rCBF from the regional tissue radiotracer concentration (Cr) recorded by the tomograph. Based upon the well-defined, nearly linear relation between Cr and rCBF inherent in the model, we have developed a method for estimating changes in rCBF from changes in Cr without calculating true rCBF and thus without arterial sampling. This study demonstrates that quantitative functional brain mapping does not require the determination of rCBF from Cr when regional neuronal activation is expressed as the change in rCBF from an initial, resting-state measurement. Patterned-flash visual stimulation was used to produce a wide range of increases in rCBF within the striate cortex. Changes in occipital rCBF were found to be accurately estimated directly from Cr over a series of 56 measurements on eight subjects. This adaptation of the PET/autoradiographic method serves to simplify its application and to make it more acceptable to the subject.