Men (n = 161) and women (n = 51) seeking treatment for marijuana use were randomly assigned to either a relapse prevention (RP; G.A. Marlatt & J.R. Gordon, 1985) or a social support (SSP) group discussion intervention. Data collected for 12 months posttreatment revealed substantial reductions in frequency of marijuana use and associated problems. There were no significant differences between the cognitive-behavioral RP intervention and the SSP group discussion conditions on measures of days of marijuana use, related problems, or abstinence rates. Men in the RP condition were more likely than men in the SSP condition to report reduced use without problems at 3-month follow-up. Posttreatment increases in problems associated with alcohol did not appear to relate to reduced marijuana use. Results are discussed in terms of the need for further research with marijuana-dependent adults and the efficacy of RP.