1,218 smokers able to quit smoking for 48 hr were randomly assigned to one of 12 cells in a 4 x 3 fully crossed factorial experiment. A pharmacologic factor contained four levels: nicotine polacrilex (gum) delivered ad lib or on a fixed regimen, placebo gum, and no gum. A self-guided behavioral treatment factor contained three levels: self-selected relapse prevention modules, randomly administered modules, and no modules. Those receiving nicotine gum were more likely to be abstinent at the 2- and 6-month follow-ups. The fixed regimen accounted for most of the effect for gum. There was no effect for the relapse prevention module factor. Men and women showed a differential treatment response. Men who received nicotine gum were more likely to be abstinent at each follow-up (2, 6, and 12 months). No treatment was significantly better among women. We conclude that research on different gum chewing regimens is warranted and that further examination of possible gender differences in response to replacement therapy is needed.