Objective To determine whether cigarette smoking is associated with the incidence and progression of diabetic retinopathy. Research Design and Methods Younger-onset diabetic subjects who had been diagnosed at >30 yr of age and taking insulin (n = 1210) and a random sample of older-onset diabetic subjects diagnosed at ≥30 yr of age (n = 1780) were selected. Baseline examinations were conducted on 996 youngerand 1370 older-onset subjects. Incidence of retinopathy was based on 138 younger-onset and 154 older-onset insulin-taking subjects and 321 older-onset non-insulintaking subjects who were free of retinopathy at baseline. Progression of retinopathy was based on 530 youngeronset and 418 older-onset insulin-taking subjects and 486 older-onset non-insulin-taking subjects with less than proliferative diabetic retinopathy at baseline. Results Baseline smoking history was categorized by status (nonsmoker, ex-smoker, current smoker) and pack-years smoked while diabetic. Retinopathy was documented by stereoscopic fundus photography. In univariate analyses, the only significant association was between pack-years and progression to proliferative diabetic retinopathy in older-onset insulin-taking subjects (P < 0.01). After controlling for known risk factors for the incidence and progression of retinopathy, pack-years smoked was borderline significant (P = 0.052) in predicting incidence of retinopathy in youngeronset subjects. Smoking was not associated with incidence in older-onset subjects or with progression or progression to proliferative diabetic retinopathy in any of the groups. Conclusions Smoking is not likely to be an important risk factor for diabetic retinopathy.