Children's behavior problems, both internalizing and externalizing, are a function of both genetic and environmental influences. One potentially important environmental influence is the classroom environment. This study of 1941 monozygotic twin pairs examined whether children whose parents rated them as similarly high or low on a number of problem behaviors were rated in the same way by teachers when they were together versus separated in their classrooms at school. Results showed that twins in the same classrooms were rated by their teachers more similarly on each dimension of problem behavior than were twins who were separated into different classrooms, suggesting that the classroom environment is important in predicting differences between twins in terms of problem behaviors at school. In addition, parents' ratings of problem behaviors were lower for twins placed in the same classroom versus twins in different classrooms. Thus, there is reason to consider whether, at least during the early school years, parents should be allowed to make the decision to not separate twins at school.