As far as we know, this is the first national study that reports compelling evidence for the main effects of genetic variants on serious and violent delinquency among adolescents and young adults. This study investigated the association between the self-reported serious and violent delinquency and the TaqI polymorphism in the DRD2 gene and the 40-bp VNTR in the DAT1 gene. The study was based on a cohort of more than 2,500 adolescents and young adults in the National Longitudinal Study of Adolescent Health in the United States. The trajectories of serious delinquency for the DAT1*10R/10R and DAT1*10R/9R genotypes are about twice as high as that for the DAT1*9R/9R genotype (LR test, P = 0.018, 2 df). For DRD2, the trajectory of serious delinquency for the heterozygotes (A1/A2) is about 20% higher than the A2/A2 genotype and about twice as high as the A1/A1 genotype, a phenomenon sometimes described as heterosis (LR test, P = 0.005, 2 df). The findings on violent delinquency closely resemble those on serious delinquency. The trajectories of violent delinquency for the DAT1*10R/9R and DAT1*10R/10R genotype are again about twice as high as that for DAT1*9R/9R (LR test, P = 0.021, 2 df). The two homozygotes of DRD2*A1/A1 and DRD2*A2/A2 scored lower (LR test, P = 0.0016, 2 df) than the heterozygotes. The findings in the models that consider DAT1 and DRD2 jointly (serious delinquency P = 0.0016, 4 df; violent delinquency P = 0.0006, 4 df) are essentially the same as those in the single-gene models, suggesting the absence of a significant correlation between the two genetic variants. These results only apply to males. Neither variant is associated with delinquency among females.