One possible explanation of the commonly reported associations between early onset cannabis use and elevated risks of other illicit drug use is that early onset cannabis use increases access and availability to other drugs. It was this argument that in part motivated policy changes in the Netherlands that led to the de facto legalization of cannabis there. This study examines, using a co-twin control design, whether previously observed associ- ations between early onset cannabis use and elevated lifetime rates of other illicit drug use would also be observed in a sample of 219 same sex Dutch twin pairs discordant for cannabis use before age 18. After adjustment for covariates, rates of lifetime party drug use (OR = 7.4, 95% CI = 2.3-23.4), hard drug use (OR = 16.5, 95% CI = 2.4-111.3), but not regular cannabis use (OR= 1.3, 95% CI = 0.3-5.1) were significantly elevated in individuals who reported early onset cannabis use, relative to their co-twin who had not used cannabis by age 18. The elevated odds of subsequent illicit drug use in early cannabis users relative to their non early using cotwins suggests that this association could not be explained by common familial risk factors, either genetic or environmental, for which our co-twin methodology provided rigorous control. © 2006 Springer Science+Business Media, Inc.