Abstract This paper reports on prospective studies of the effects of psychological and social variables measured in first grade on the use of various drugs, alcohol, and cigarettes by teenagers 10 years later. The population consisted of all of the 1966–1967 first graders of Woodlawn, a poor, black, Chicago neighborhood. In a 10-year follow-up, the population, which had been studied three times in first grade and once in third grade, was reassessed for family, psychological, and social data, in addition to drug, alcohol and cigarette use. The former Woodlawn first graders, now aged 16–17, used beer or wine, hard liquor, marijuana or hashish, and cigarettes with considerable frequency. Psychedelics, amphetamines, barbiturates, tranquilizers, opiates, cocaine, inhalants, and cough syrup or codeine were used with much lower frequency. Three separate kinds of characteristics observable in first grade were associated with drug use by adolescents ten years later. (1) Higher first-grade IQ test scores or school readiness test scores predicted more frequent drug use for both sexes. (2) Males used drugs and alcohol (not cigarettes) more often than females, and, antecedents of later drug use by males were more clear than those for females. (3) Those children whose first-grade teachers rated them as shy used drugs least often 10 years later; first graders rated as aggressive used drugs most often 10 years later; adapting first graders and those with learning problems only were found to be moderate drug, alcohol and cigarette users. These results were much more clear for males. Among females, higher levels of psychiatric symptoms in first grade predicted, to some extent, lower teenage drug use. Teenage antisocial behavior was an important mediator of teenage drug use for first-grade shy-aggressive males and somewhat less important for first-grade aggressive males.