OBJECTIVES: To study changes in the family smoking profile and in the association between parental and child smoking from 1977 to 2005. METHODS: Data was based on biennial surveys using nationally-representative samples of 14-18-year-old Finns (n=58,279). Response rate ranged between 88% (1977) and 65% (2005). Parental smoking categories were: two smoking parents, smoking father, smoking mother, both currently non-smokers but one or both ex-smokers, and two never-smoking parents. Child smoking categories were: experimental, daily, and never. Associations between parental and child smoking were examined using multinomial logistic regression models. RESULTS: Over the study period, the proportion of never-smoking families (child and parents never-smokers) increased (9% vs. 18%). Age, sex, family structure, and survey decade adjusted odds ratios for child's daily smoking were 6.9 (95% CI: 6.4, 7.5) when both parents smoked, 4.7 (95% CI: 4.3, 5.2) when mother smoked, 3.8 (95% CI: 3.5, 4.1) when father smoked, and 2.8 (95% CI: 2.6, 2.9) when one/both were ex-smokers compared with children of never-smoking parents. Only a few non-systematic interactions between parental smoking and survey decade were found. CONCLUSIONS: The proportion of totally smoke-free families increased substantially. Association between parental and child smoking persisted strong and mainly similar over time.