Assessed perceptions of stress prior to quitting and at 1, 3, and 6 months following quit date. Self-reported smoking status was augmented by a bogus pipeline procedure at each interview, and abstinence at 6 months was verified by carbon monoxide and saliva cotinine. The analyses provide strong evidence for a relation between changes in stress levels and changes in smoking status. Those who failed to quit smoking for more than 24 hr maintained a relatively high and consistent level of stress over the entire 6-month period. For those remaining continuously abstinent over the course of the study, stress decreased as duration of abstinence increased. Increases in stress with relapse were found across all three panel lags (prequit to 1 month, 1 to 3 months, and 3 to 6 months), and decreases in stress with quitting were found across the two lags where that comparison was possible. The possibility that the relation between smoking and stress is bidirectional is discussed as a possible interpretation of the results.