AbstractParoxetine, a selective inhibitor of serotonin uptake and an antidepressant, was used in conjunction with quantitative ex vivo autoradiography to study the feasibility of imaging serotonin terminals in the living brain. Tritiated paroxetine was injected in the rat tail vein, and the brain was processed for quantitative autoradiography 3 hours later. Animals received either [3H]paroxetine alone (100 μCi/animal) or a mixture of labeled paroxetine (100 μCi) and an excess of unlabeled drug (0.5 or 2 mg/kg intravenously [i.v.]). Computerized image analysis of the resulting autoradiograms revealed high densities of radioactivity in brain regions known to contain high densities of serotonergic terminals and high specific binding of [3H]paroxetine in vitro, such as the raphe nuclei, interpeduncular nucleus, basolateral amygdala, substantia nigra, and some hypothalamic nuclei. Radioactivity uptake in these brain regions was effectively blocked (50‐72%) by coadministration of excess unlabeled paroxetine. However, cortical and hippocampal binding of paroxetine in vivo was moderately high, in contrast to the relatively sparse serotonergic innervation in these regions. Only a relatively small proportion of cortical and hippocampal binding (20‐40%) could be blocked by excess unlabeled paroxetine, indicating that most of the radioactivity in these regions is not associated with serotonin terminals or uptake sites. The usefulness of [3H]paroxetine as an in vivo ligand for imaging serotonin terminals in the human brain is limited by these nonserotonergic binding sites. © 1993 Wiley‐L