Part of the hardening hypothesis to explain the persistence of smoking-despite powerful antismoking forces-links smoking with psychopathology, especially depression. It has been proposed that the association between depression and smoking has emerged in more recent cohorts as smoking rates declined, disproportionately leaving among current smokers those who found it more difficult to quit because of their psychopathology. We examined the association of regular smoking and depression in a cohort who began smoking prior to the decline in smoking rates in the United States and assessed a corollary hypothesis that smokers with depression were more likely to persist in smoking than were those without depression. Data were from the Wisconsin Longitudinal Study of a random sample of high school graduates from the class of 1957. In the 1992 follow-up, a subset of these 53-54-year-olds were assessed for lifetime and current depression and smoking (n = 4,858). A modest association between regular smoking and depression was found (OR = 1.4, 95% CI = 1.2-1.6); persistence of smoking (current smoking among ever regular smokers) was unrelated to single-episode or recurrent depression (OR = 1.1, 95% CI = 0.8-1.5). The results do not support the proposition that the association between smoking and depression emerged when smoking rates declined, or that self-medication of depression through smoking is a likely mechanism for the persistence of smoking.