FINDINGS WERE BASED UPON RECORDED CONTACTS WITH THE THE POLICE BY TWO COHORTS, ONE BORN IN 1942 AND THE OTHER IN 1949. THE REASONS FOR THESE CONTACTS, THEIR SERIOUSNESS IN THE EYES OF THE LAW, PLACE OF RESIDENCE, AND OTHER DATA WERE UTILIZED IN PREDICTING WHO IS MOST LIKELY TO INITIALLY ENGAGE IN DELINQENT BEHAVIOR, WHO WILL CEASE DELINQUENT BEHAVIOR, AND WHO WILL EXHIBIT ADULT CRIMINAL ACTIVITY. THE SPATIAL DISTRIBUTION OF MALES AND FEMALES, MEXICAN-AMERICAN, NEGRO, AND ANGLO WITH POLICE CONTACTS (ALTHOUGH SLIGHTLY SKEWED TOWARD THE INNER CITY) WAS NOT MARKEDLY DIFFERENT FROM THAT OF THE TOTAL COHORT. WHILE SOMEWHAT DISPROPORTIONATE NUMBERS OF NEGROES AND, TO A LESSER EXTENT, MEXICAN-AMERICANS HAD POLICE CONTACTS AT EACH AGE PERIOD ( 6-17; 18-21; OVER 21) IN THEIR CAREERS, NEITHER DELINQUENCY NOR ADULT CRIME COULD BE CONSIDERED A MINORITY GROUP PROBLEM. MOST CAREERS WERE NOT CONTINUOUS AND MOST DELINQUENCY DID NOT LEAD TO A CRIME. AMONG THOSE WHO HAD MORE THAN ONE POLICE CONTACT, CAREERS WERE HETEROGENEOUS; FACTOR ANALYSIS FAILED TO SHOW ANY RACE/ETHNIC OR SEX GROUP IN EITHER COHORT. COMPOSITE SCORES OR CAREERS BASED ON SERIOUSNESS AND TYPE OF CONTACT PROVIDED LITTLE INCREASE IN PREDICTIVE EFFICIENCY OVER SIMPLE NUMBER OF POLICE CONTACTS. DETAILED TABULAR DATA IS PROVIDED THROUGHOUT. (AUTHOR ABSTRACT MODIFIED)...ELW