The focal point of this paper is the transition from drug use to drug dependence. We present new evidence on risk for starting to use marijuana, cocaine, and alcohol, as well as risks for progression from first drug use to the onset of drug dependence, separately for each of these drugs. Data from the National Comorbidity Survey (NCS) were analyzed. The NCS had a representative sample of the United States population ages 15-54 years (n = 8,098). Survival analysis techniques were used to provide age- and time-specific risk estimates of initiating use of marijuana, cocaine, and alcohol, as well as of becoming dependent on each drug. With respect to risk of initiating use, estimated peak values for alcohol and marijuana were found at age 18, about two years earlier than the later peak in risk of initiating cocaine use. With respect to risk of meeting criteria for the clinical dependence syndrome, estimated peak values for alcohol and marijuana were found at age 17-18. Peak values for cocaine dependence were found at age 23-25. Once use began, cocaine dependence emerged early and more explosively, with an estimated 5-6% of cocaine users becoming cocaine dependent in the first year of use. Most of the observed cases of cocaine dependence met criteria for dependence within three years after initial cocaine use. Whereas some 15-16% of cocaine users had developed cocaine dependence within 10 years of first cocaine use, the corresponding values were about 8% for marijuana users, and 12-13% for alcohol users. The most novel findings of this study document a noteworthy risk for quickly developing cocaine dependence after initial cocaine use, with about one in 16 to 20 cocaine users becoming dependent within the first year of cocaine use. For marijuana and alcohol, there is a more insidious onset of the drug dependence syndrome.