Smoking is a risk factor of coronary heart disease (CHD), while the role of testosterone in the development of CHD is controversial. The reported effects of cigarette smoking on testosterone levels in men are conflicting, and smoking may be an important confounding factor when evaluating the relationship between testosterone levels and CHD. Thus, the objective of the present study was to examine the associations of smoking status and number of cigarettes smoked per day with total and free testosterone levels in a cross-sectional population-based study of 3427 men participating in the fifth Tromsø study. Total testosterone, luteinizing hormone, follicle-stimulating hormone and sex hormone-binding globulin levels were measured with immunoassay while free testosterone levels were calculated. Waist circumference was also measured and two standardized questionnaires were completed, including smoking status and number of cigarettes smoked. The data were analysed with analysis of variance and covariance and multiple regression analysis. Smoking men had significantly higher levels of total and free testosterone compared with men who never smoked (p < 0.001 and <0.01 respectively). Both total and free testosterone levels increased significantly with increasing number of cigarettes smoked daily (p < 0.001). Smoking men had 15% higher total and 13% higher free testosterone levels compared with men who never smoked. Thus, smoking seems to be an important confounding factor when evaluating testosterone levels, and could possibly mask borderline hypogonadism.