The reinforcing effect of the high affinity dopamine reuptake inhibitor GBR 12909 (1-[2-[bis(4-fluorophenyl)methoxy]ethyl)-4(3- phenylpropyl)piperazine) was compared with that of cocaine, and the effects of both GBR 12909 and cocaine pretreatments were evaluated on behaviour maintained by cocaine in rhesus monkeys. Increasing dose per injection of intravenously-delivered GBR 12909 or cocaine led to increased rates of lever-press responding. Maximum cocaine-maintained rates were higher and occurred at a smaller dose than maximum rates of GBR 12909-maintained responding. Presession intravenous administration of either GBR 12909 or cocaine (0.32, 1.0 or 3.2 mg/kg) resulted in dose-dependent decreases in rates of cocaine-maintained responding when high doses of cocaine, which engendered high response rates, were available early in the session. Under these conditions, the decrease in response rates was associated primarily with decreases in running rate rather than with a lengthening in post-reinforcement pause times. The decreases in running rate produced by both cocaine and GBR 12909 probably reflect an unconditioned rate-disruptive effect of these drugs on cocaine-reinforced responding rather than a reduction in the reinforcing efficacy of cocaine.