THE OFFICAL CRIMINAL CAREERS OF MEMBERS OF TWO MIXED-SEX BIRTH COHORTS WERE ESTABLISHED BY COMPARING THEIR NAMES AGAINST THE MASTER RECORD FILE OF THE POLICE DEPARTMENT IN A MEDIUM-SIZED INDUSTRIAL CITY IN THE MIDWEST. ALL RECORDED CONTACTS WITH POLICE INVOLVING ALLEGATIONS OF CRIMINAL ACTIVITY WERE ENTERED AS PART OF A COHORT MEMBER'S OFFICIAL CAREER. ABOUT 68 PERCENT OF THE COHORT BORN IN 1942 AND 69 PERCENT OF THOSE BORN IN 1949 WITH CONTINUOUS RESIDENCE IN THE CITY HAD AT LEAST ONE POLICE CONTACT AT SOME TIME DURING THEIR LIVES. THESE DATA CONSTITUED THE BASIS FOR CREATING A TYPOLOGY BASED ON THE FREQUENCY AND SERIOUSNESS OF POLICE CONTACTS. LIFE HISTORIES WERE ALSO OBTAINED THROUGH EXTENSIVE INTERVIEWS. DATA WERE ANALYZED BY MEANS OF THE DISCRIMINANT ANALYSIS PROGRAM. FINDINGS INDICATED THAT CREATING EMPIRICAL TYPOLOGIES OF CRIME IS A DIFFICULT TASK, AND THERE IS RELATIVELY LITTLE CROSS-COHORT COMPARABILITY ON EITHER FREQUENCY OR SERIOUSNESS DIMENSIONS. ANOTHER FINDING WAS THAT CATEGORIES ESTABLISHED FOR A TYPOLOGY WILL TEND TO BE SOMEWHAT ARBITRARY AND MUST BE LINKED TO THE NEEDS OF THE RESEARCHER. ANY CONCLUSIONS REACHED ON THE BASIS OF A GIVEN TYPOLOGICAL SYSTEM WILL, IN ALL LIKELIHOOD, BE DIFFERENT FROM THOSE REACHED WHEN ANOTHER SYSTEM IS USED. EXTENSIVE TABULAR DATA, FOOTNOTES, AND A BIBLIOGRAPHY ARE INCLUDED. (MJW)