Using data from the large, 30-year prospective Dunedin cohort study, we examined whether preexisting individual differences in childhood temperament predicted adulthood disordered gambling (a diagnosis covering the full continuum of gambling-related problems). A 90-min observational assessment at age 3 was used to categorize children into five temperament groups, including one primarily characterized by behavioral and emotional undercontrol. The children with undercontrolled temperament at 3 years of age were more than twice as likely to evidence disordered gambling at ages 21 and 32 than were children who were well-adjusted at age 3. These associations could not be explained by differences in childhood IQ or family socioeconomic status. Cleanly demonstrating the temporal relation between behavioral undercontrol and adult disordered gambling is an important step toward building more developmentally sensitive theories of disordered gambling and may put researchers in a better position to begin considering potential routes to disordered-gambling prevention through enhancing self-control and emotional regulation.