The results of this study demonstrate that a high quality of the smoking-specific communication was related to a lower likelihood of adolescent smoking, whereas the frequency was positively associated with adolescent smoking. Both the quality and frequency of parental smoking-specific communication were related to adolescents’ selective affiliation with non-smoking friends. This study investigated whether parental smoking-specific communication was related to adolescents’ friendship-selection processes. Additionally, the work investigated whether adolescents and their best friends influenced each other over time, and what role parents played in this process. The findings also suggest that parental smoking-specific communication is associated with adolescent smoking directly, but also indirectly by influencing the friends the adolescents will associate with. In the present study, data were derived from the Family and Health Project in which, at baseline, 428 full families participated. Analyses were conducted by means of Structural Equation Modeling. In this 2-year, three-wave longitudinal study, data were available from fathers, mothers, early adolescents (mean age 13.4 years), and middle adolescents (mean age 15.2 years). The majority of the participating adolescents were of Dutch origin, over 95 percent, with an almost equal distribution of male and female; adolescents with lower, middle, and higher educational levels were equally represented. Tables, figure, and references