BACKGROUND Acamprosate (calcium acetyl homotaurinate) has been used clinically to treat relapse in alcoholics. In rats, it has been shown to decrease ethanol, but not water, self-administration after ethanol deprivation. METHODS To further investigate the effect of acamprosate on reinforced behaviors in rats, the present experiment used: (1) both ethanol and sucrose reinforcer solutions to better assess the distinct effects of acamprosate on ethanol-directed behaviors, and (2) an operant model that procedurally separates the "cost" to begin drinking from consuming the reinforcer solutions to dissociate the effects of acamprosate on appetitive versus consummatory processes. In daily sessions (5 days/week), rats (n = 6/group) were trained to make 30 lever-press responses to gain access for 20 min to a sipper tube containing either ethanol (10%) or sucrose (3%). After stable responding, acamprosate treatment was given. Three doses were tested (50, 100, and 200 mg/kg/injection, intraperitoneally), one dose per week. Each week, a total of four injections were given (21 and 2 hr before the operant sessions over 2 consecutive days). RESULTS At these doses, acamprosate had no effect on the measures of appetitive responding for either solution. However, all doses reliably decreased ethanol consumption on the 2nd day of treatment (from an average of 0.83 to 0.63 g/kg). Analysis of the pattern of ethanol consumption showed that the effects of acamprosate occurred after the onset of a normal pattern of intake, as measured by lick rate and size of the initial bout of drinking, which suggested that acamprosate is most effective when combined with the pharmacological effects of ethanol. Sucrose intake was unaffected by all acamprosate treatments, which indicated that the treatment effects were specific to ethanol and not due to a general decrease in consummatory behavior. CONCLUSIONS Overall, these results suggest that acamprosate is effective at reducing total ethanol intake, but may not reliably alter subjects propensity to begin a drinking bout as measured by this model. However, whether this applies to the clinical use of acamprosate, where other types of reinforcement may also precipitate relapse drinking, is not certain.