The relative roles of genetic and environmental factors in bruxism are not known. In 1990 a questionnaire sent to the Finnish Twin Cohort yielded responses from 1298 monozygotic and 2419 dizygotic twin pairs aged 33-60 years. We used structural equation modelling to estimate genetic and environmental components of variance in the liability to bruxism. There was a significant gender difference both in childhood (P = 0.001) and adult (P = 0.007) bruxism. Females compared to males reported childhood bruxism 'often' 5.2% vs 4.1% and 'sometimes' 17.4% vs 17.3%, and as adults 'weekly' 3.7% vs 3.8% and 'monthly' 3.9% vs 4.6%, respectively. Bruxism in childhood and adulthood is highly correlated (0.86 in males and 0.87 in females). The proportion of total phenotypic variance in liability to bruxism attributed to genetic influences in childhood bruxism was 49% (95% CI 37-60%) in males and 64% (55-71%) in females, and for adults 39% (27-50%) among males and 53% (44-62%) among females. The correlation between the genetic effects on childhood bruxism and the genetic effects on adult bruxism was estimated in a bivariate model to be 0.95 (95% CI 0.94-0.96) in males and 0.89 (0.88-0.90) in females. Bruxism appears to be quite a persistent trait. There are substantial genetic effects on bruxism both in childhood and as adults, which appear to be highly correlated.