AimsTo evaluate gender differences in the role of positive and negative affect on smoking uptake. DesignProspective longitudinal cohort study of adolescent health behaviors. SettingFour suburban secondary schools outside Philadelphia, Pennsylvania, USA. ParticipantsAdolescents (n=1357) were surveyed every 6months for 4years (age 14-18years). MeasurementsSmoking and affect were measured via survey at each of the eight time-points. FindingsA two-group associative process latent growth curve model revealed that baseline positive affect was related negatively to smoking progression for females (b=-0.031, Z=-4.00, P<0.0001) but not for males (P=0.33). This gender difference was significant, (2)((df=1))=8.24, P=0.0041, indicating that for every standard deviation (SD) decrease in positive affect (SD=2.90), there was a 10% increase [odds ratio (OR)=1.10, 95% confidence interval (CI)=1.04, 1.14] in the odds of smoking progression for females. Baseline negative affect was related significantly and positively to smoking progression for males (b=0.038, Z =2.874, P=0.004) and females (b=0.025, Z =3.609, P<0.0001), but the gender difference was not significant, (2)((df=1))=0.82, P=0.37. Thus, on average, for every standard deviation (SD=4.40) increase in baseline negative affect there was a 15% (OR=1.15, 95% CI=1.06, 1.26) increase in the odds of smoking progression for males and for females. ConclusionsThe impact of affect on adolescent smoking uptake varies by gender. Low positive affect (low experience of positive feelings or emotions) for females and high negative affect (high experience of negative feelings or emotions) for both males and females increases the risk for adolescent smoking.