Using the National Longitudinal Study of Adolescent Health, this study examined the impact of school-le I vel smoking and drinking on adolescent-peer similarity for smoking and drinking. Hierarchical linear modeling revealed that adolescent-peer similarity was significantly moderated by school-level substance use for both tobacco and alcohol use. For tobacco, similarity between adolescent and peer use increased from .18 in the lower quartile of tobacco-using schools to .44 in the upper quartile of tobacco-using schools. Corresponding similarities for alcohol use ranged from .25 to .34. These results suggest that schools with relatively few substance-using peers provide less opportunities for adolescents to pick niches that expose them to risk factors that correspond to their own substance-use behaviors.