The subjects were a stratified random sample from a population of 6,149 known narcotic users arrested or identified by the police between 1952 and 1976. The sample was unselected for criminality but stratified by race and year of police contact. To be eligible for the study, subjects must have used narcotics on at least 4 days a week for at least 1 month while in the community. Subjects described their narcotic addiction, abstinence, and incarceration periods, with criteria for successive addiction periods being the same as that for inclusion in the study. All crimes reported by the subjects were placed in one of the following categories: theft, violence, drug dealing, confidence games, and other. Subjects were also questioned about their use of nonnarcotic substances during each period of addiction and nonaddiction. Frequency and type of nonnarcotic drugs used were a joint function of race (black and white) and current narcotic addiction status (addicted or nonaddicted). The relationship of such use to the commission of types of crime also depended on race and narcotic addiction status. Both races used more nonnarcotic drugs during addiction periods compared to nonaddiction periods. Bivariate and multiple correlational analyses indicate that higher use rates for certain nonnarcotic drugs were associated with higher rates for certain crimes. 2 tables and 12 references. (Author abstract modified)