Case reports and laboratory research indicate the existence of a cannabis withdrawal syndrome. However, the data tell us little about the prevalence and clinical characteristics of a marijuana withdrawal syndrome in people who have used the drug but who did not enter treatment for cannabis dependence. Face-to-face semi-structured interviews applying standard diagnostic criteria were used in the present study to gather data from 5611 men and women, recruited between 1991 and 1995 through the Collaborative Study of the Genetics of Alcoholism (COGA). Almost 41% of the sample had no history of marijuana use (Group 1), 28% had consumed this drug less than 21 times in any single year (Group 2), and 31% used it at least that frequently (Groups 3 and 4). Almost 16% of the more frequent marijuana users related a history of a marijuana withdrawal syndrome, and these Group 4 subjects had used the drug almost daily for an average of almost 70 months. The typical withdrawal symptoms included "nervous, tense, restlessness", "sleep disturbance" and "appetite change". While Group 4 subjects were more likely to have developed dependence on most types of drugs, even when alcohol and drug use patterns were statistically taken into account, marijuana use was still significantly related to a self-report of a history of marijuana withdrawal.