The present study was designed to determine whether depressive symptoms are independently associated with smoking and nicotine dependence among cigarette smokers, using 1990-1991 data from the Coronary Artery Risk Development in Young Adults (CARDIA) Study. A total of 3,933 participants (788 black men, 1,090 black women, 974 white men, and 1,081 white women) aged 23-35 years were included. Analyses were stratified by race and sex. Depressive symptoms were measured by means of the Center for Epidemiologic Studies Depression (CES-D) Scale. Nicotine dependence was defined as smoking one's first cigarette of the day within 30 minutes of awakening. Analysis of covariance was used to control for potential covariates (age, body mass index, alcohol consumption, and education). In unadjusted comparisons, smokers had more depressive symptoms than never smokers in all groups except white men; this relation showed little change after adjustment for age, body mass index, and alcohol consumption. However, after adjustment for education in addition to the above variables, these differences became attenuated and were significant only among white women (adjusted CES-D score difference = 1.9, p < 0.02). When analyses were further stratified by nicotine dependence, dependent smokers had higher CES-D scores than never smokers in all groups. The differences again became attenuated when education was added to the model, and were significant only among black women (adjusted CES-D score difference = 2.3, p < 0.01). These results indicate that although smoking in general and nicotine-dependent smoking in particular are related to symptoms of depression, controlling for educational level attenuates these relations.