Conditioned stimuli (CS) can be devalued by exposure to those stimuli in the absence of primary reward. We tested the hypothesis that dopamine (DA) mediates the control of behavior by conditioned appetitive stimuli. Long-Evans rats were trained to respond for sucrose under a heterogeneous chain schedule in which seeking responses (lever press) turned on a houselight [variable interval (VI)-120 s]; taking responses (wheel turn or chain pull) in the presence of the houselight were reinforced [fixed ratio (FR)-1] by a sucrose pellet. When responding on this schedule was stable, the levers were retracted and subjects had access to the sucrose-taking manipulandum only. Sucrose-taking responses were either extinguished or reinforced under the influence of the DA antagonist, pimozide. Control groups were also reinforced for sucrose-taking responses but received no injection or a vehicle injection prior to each session. Responses of extinction and pimozide-treated groups declined over sessions. Sucrose-seeking responses were measured in a later test when subjects had no access to the sucrose-taking manipulandum or to the reinforcer. Both extinction and pimozide manipulations reduced seeking responses, relative to the respective control groups. Pimozide injections in the home cage had no effect. These data support the idea that DA mediates the conditioned reinforcing properties provided by access to the taking link of the chain.