DATA WERE COLLECTED IN A COMPREHENSIVE STUDY OF BLACK POLICEMEN IN A METROPOLITAN AREA IN 1974. OF THE 576 MALE OFFICERS STUDIED, 83 PERCENT WERE IN THE LOWEST RANK; 68 PERCENT WERE MARRIED; AND 90 PERCENT WERE LESS THAN 35 YEARS OLD. THE AVERAGE LENGTH OF SERVICE WAS 5 YEARS, AND THE AVERAGE PERIOD OF EDUCATION WAS 13 YEARS. THE INDEPENDENT VARIABLES WERE RACE, POLICE RANK, EDUCATIONAL LEVEL, LENGTH OF SERVICE, MARGINALITY 1 (RESULTING FROM CONFLICTING REFERENCE GROUPS), AND MARGINALITY 2 (RESULTING FROM LOCATION ON THE FRINGES OF TWO GROUPS). THE DEPENDENT VARIABLES WERE PERCEPTION OF POLICE AUTHORITY AND PERCEPTION OF PUBLIC INVOLVEMENT IN POLICE AFFAIRS. MULTIPLE REGRESSION WAS USED TO IDENTIFY THE RELATIVE EFFECTS OF THE INDEPENDENT VARIABLES ON THE DEPENDENT VARIABLES. THE COEFFICIENT OF MULTIPLE DETERMINATION WAS ALSO USED TO IDENTIFY THE PROPORTION OF VARIATION IN THE DEPENDENT VARIABLES ACCOUNTED FOR BY THE INDEPENDENT VARIABLES. RESULTS INDICATE THAT THE CONCEPT OF 'DOUBLE MARGINALITY' HAS LITTLE EXPLANATORY POWER IN ACCOUNTING FOR VARIATION IN EITHER BLACK OFFICERS' PERCEPTIONS OF PUBLIC INVOLVEMENT IN POLICE AFFAIRS OR THEIR PERCEPTION OF POLICE AUTHORITY. THIS GROUP OF POLICE OFFICERS MAY REPRESENT A NEW BREED OF POLICE WHOSE CONCEPTION OF BLACKNESS IS MUCH LESS NEGATIVE THAN THAT OF THEIR PREDECESSORS. FUTURE RESEARCH SHOULD CONSIDER FACTORS OTHER THAN ETHNIC GROUP MEMBERSHIP; FOR EXAMPLE, INADEQUATE REWARD SYSTEMS, ADVERSE WORK SCHEDULES, AND DISTORTED NEWS MEDIA ACCOUNTS) WHICH MAY CREATE UNCERTAINTIES IN REFERENCE GROUP TIES FOR BLACK OFFICERS AND BLACKS IN GENERAL. NOTES AND REFERENCES ARE PROVIDED. (RCB)