EIGHTY STANDARD, METROPOLITAN STATISTICAL AREAS WERE SELECTED IN A STRATIFIED, RANDOM SAMPLE. ALL AGENCIES CONDUCTING GENERAL AREA PATROL, TRAFFIC PATROL, TRAFFIC ACCIDENT INVESTIGATION, RESIDENTIAL BURGLARY INVESTIGATION, OR HOMICIDE INVESTIGATION FOR RESIDENTS OF EACH STANDARD METROPOLITAN STATISTICAL AREA WERE INCLUDED. DATA FOR THE STUDY WERE GATHERED FROM THESE POLICE AGENCIES THROUGH PERSONAL INTERVIEWS AND INSPECTION OF DOCUMENTS SUPPLEMENTED BY TELEPHONE AND MAIL COMMUNICATION. USING THIS INFORMATION, THE RESEARCHERS INVESTIGATED THREE BROAD AREAS: 1) WHAT AGENCIES PRODUCE POLICE SERVICES AND WHAT LEVELS OF SERVICE ARE PROVIDED? 2) HOW MUCH DUPLICATION AND COOPERATION IS THERE IN THE DELIVERY OF SERVICES? AND 3) HOW IS THE SIZE OF AGENCIES RELATED TO SERVICE DELIVERY? ON THE BASIS OF DATA GATHERED IN THIS STUDY, THE RESEARCHERS CONCLUDE THAT THE DIVERSITY IN SIZE OF AGENCIES AND THE DIFFERENT COMBINATIONS OF SERVICES THEY PRODUCE ARE NOT NECESSARILY WASTEFUL OR CONFUSING. THERE APPEARED TO BE LITTLE DUPLICATION BY SMALL- TO MEDIUM-SIZED POLICE AGENCIES. MOST POLICE DEPARTMENTS SEEMED TO KNOW WHAT SERVICES THEY ARE RESPONSIBLE FOR AND WHERE TO DELIVER THEM. JURISDICTIONAL BOUNDARIES WERE GENERALLY OPEN FOR FRESH PURSUIT AND FOR MUTUAL ASSISTANCE IN EMERGENCIES. INSTANCES OF DUPLICATION, OF CONFUSION, AND OF NONCOOPERATION WERE FOUND, BUT THEY WERE NOT THE COMMON PATTERN. THE AUTHORS CONCLUDE THAT REORGANIZATION OR METROPOLITAN POLICING NEEDS TO BE BASED ON A REALISTIC ASSESSMENT OF THE RESOURCES AND NEEDS OF EACH PARTICULAR METROPOLITAN AREA. (AUTHOR ABSTRACT MODIFIED)