The study obtained data from the Mater-University Study of Pregnancy, an Australian longitudinal study of women's and children's health and development. Because sample selection occurred over 3 years, some women returned to the hospital to give birth to a second child, which resulted in a subsample of 685 sibling pairs. These 1,370 children, ages 1 to 3 years apart, were the focus of the current study. At the 14-year follow-up assessment of mothers and children, 376 sibling pairs (55 percent) remained in the study, including 93 sister pairs, 97 brother pairs, and 186 mixed-sex sibling pairs. Adolescent tobacco and alcohol use was based on self-reported information from each sibling at the 14-year follow-up assessment, conducted from 1995 to 1997. Information on parental tobacco use and maternal alcohol use was obtained from the mothers at the 14-year follow-up. The link with sibling tobacco and alcohol use was consistent and moderately strong, and it remained significant when controlling for a number of family-related factors, indicating that the shared family environment could not fully explain the extent of similarity in siblings' behaviors. Also, sibling substance use had a greater effect on adolescent substance use than did parents' drinking or smoking. These findings emphasize the need to include siblings in substance abuse prevention and intervention programs. 6 tables and 26 references