THE STUDY SAMPLE CONSISTED OF 2,750 ADOLESCENTS, 18 YEARS OR YOUNGER SURVEYED IN 1976-1977. THE SUBJECTS WERE DRAWN FROM A NATIONAL SAMPLE OF 97 YOUTH DRUG-ABUSE TREATMENT PROGRAMS FROM 37 STATES. SOME 1,035 OF THE SAMPLE (38 PERCENT) REPORTED AT LEAST ONE INSTANCE OF A PRIVATE PROPERTY OFFENSE, 627 (23 PERCENT) HAD BEEN ARRESTED AT LEAST ONCE FOR THIS OFFENSE, AND 377 (14 PERCENT) HAD BEEN CONVICTED FOR A PRIVATE PROPERTY OFFENSE. THE DATA INDICATE THAT THOSE WHO HAVE USED THE EIGHT DRUG CLASSES CONSIDERED (HEROIN, BARBITURATES, AMPHETAMINES, COCAINE, MARIHUANA, HALLUCINOGENS, INHALANTS, AND PHENCYCLIDINE) ARE MUCH MORE LIKELY TO REPORT INVOLVEMENT IN PRIVATE PROPERTY OFFENSES THAN THOSE WHO HAVE NOT USED THESE SUBSTANCES. THE STRENGTH OF THE ASSOCIATION BETWEEN DRUG USE AND CRIME IS SEEN IN THE PERCENTAGE DIFFERENCES--FROM A LOW OF 19 PERCENTAGE POINTS DIFFERENCE FOR THE HEROIN-USE VARIABLE TO A 26-POINT DIFFERENCE FOR AMPHETAMINES AND HALLUCINOGENS. DATA CONCERNING OTHER OFFENSES TYPICALLY ASSOCIATED WITH DRUG ABUSE (DRUG USE AND POSSESSION, DRUG SALES, PUBLIC PROPERTY OFFENSES) SHOWED VIRTUALLY THE SAME RESULTS. THE MEAN NUMBER OF SELF-REPORTED ARRESTS BY THE CATEGORIES OF THE MARIHUANA/HEROIN INDEX, WITH SEX AND RACE CONTROLLED, WAS ALSO DETERMINED. FOR BOTH SEXES, THE DATA SHOW THAT ANY CONTACT WITH HEROIN IS MORE LIKELY TO LEAD TO ARRESTS FOR CRIMINAL BEHAVIOR THAN IS TRUE WITH MARIHUANA USE. THERE ARE NO CLEAR OR CONSISTENT RESULTS WHEN WHITES ARE COMPARED WITH BLACKS BY SEX. A COMPARISON OF DRUG USE BY SOURCE OF REFERRAL--CRIMINAL JUSTICE VERSUS VOLUNTARY--INDICATES THAT A GREATER PERCENTAGE OF YOUTHS COMING FOR TREATMENT HAVE BEEN REFERED BY CRIMINAL JUSTICE OFFICIALS. SOME IMPLICATIONS OF THESE FINDINGS ARE DISCUSSED. SUPPORTING TABULAR DATA AND REFERENCES ARE PROVIDED. (RCB)