To investigate the possibility that specific symptom patterns contribute differentially to excess depression among smokers, we analyzed Center for Epidemiological Studies-Depression (CES-D) and subscale scores in 931 women current smokers (CS), ex-smokers (ES), and never-smokers (NS). After adjusting for differences in age, education, and marital status, significant group differences were found for the CES-D overall and for Depressed Affect, Anhedonia, and Somatic Features, but not Interpersonal Distress. Both CS and ES scored significantly higher than NS on most measures, whereas differences between CS and ES failed to reach significance. CS were significantly more likely than NS to be taking antidepressants. Alcohol intake also showed significant group differences, with CS>ES>NS, suggesting that the link between current smoking and alcohol intake is mediated by factors other than depression, since CS and ES were comparably depressed. We conclude that studying specific symptom patterns may elucidate the smoking-depression link. Our observation that ES and CS exhibit similar levels of depressive symptomatology further suggests that interventions for depressed smokers need to take into account the likely persistence of depression following cessation.