GREENSBORO, N.C., IS DIVIDED INTO FOUR PATROL DIVISIONS. DURING DECEMBER 1976, AN EXPERIMENT TESTING SATURATION PATROL WITH UNMARKED POLICE VEHICLES, ROVING PATROL, AND REDUCED USAGE OF RADAR EQUIPMENT WAS CARRIED OUT IN DIVISION 3, WITH THE OTHER DIVISIONS AS CONTROLS. THE EXPERIMENT WAS REPEATED IN JANUARY 1977 IN DIVISION 1. IN FEBRUARY 1977, A SECOND EXPERIMENT TESTING SATURATION PATROL WITH MARKED POLICE VEHICLES AND VARYING DEGREES OF RADAR USAGE WAS CARRIED OUT IN DIVISION 3. THIS EXPERIMENT WAS REPEATED IN DIVISION 1 IN MARCH. THE EXPERIMENTAL DESIGN WAS INTENDED TO PERMIT MAXIMUM ANALYSIS AND EVALUATION THROUGH PRE- AND POST-EXPERIMENT COMPARISONS WITHIN THE EXPERIMENTAL DIVISIONS, COMPARISONS BETWEEN THE EXPERIMENTAL DIVISIONS, AND COMPARISONS WITH CONTROL DIVISIONS. AMONG THE FINDINGS OF THE FIELD EXPERIMENTS ARE THE FOLLOWING: NO REDUCTION IN ACCIDENTS RESULTED FROM GIVING TRAFFIC OFFICERS MORE SPECIFIC ASSIGNMENTS (I.E., REDUCING THE AREA OF PATROL TO MORE SPECIFIC LOCATIONS); REDUCED USAGE OF RADAR EQUIPMENT DID NOT REDUCE THE NUMBER OF TRAFFIC ACCIDENTS; THERE WERE NO DIFFERENCES IN THE MEAN NUMBER OF ACCIDENTS WHEN OFFICERS OF THE TRAFFIC UNIT PATROLLED IN CLOSE PROXIMITY TO ONE ANOTHER OR WHEN THEY WERE SCATTERED THROUGHOUT A DIVISION; THERE WERE NO DIFFERENCES IN THE NUMBER OF ACCIDENTS AS A FUNCTION OF ROVING VERSUS STATIONARY PATROL; THE CONSIDERABLE VARIANCE IN NUMBER OF ACCIDENTS BY DIVISION BY MONTH IS NOT CONTROLLED BY SATURATION PATROL (VERSUS NORMAL PATROL), REGARDLESS OF THE TYPE OF POLICE VEHICLE (MARKED OR UNMARKED) USED; AND OFFICERS APPARENTLY CAN DETECT TRAFFIC LAW VIOLATIONS AS EASILY AND MAKE ARRESTS MORE FREQUENTLY IN MARKED POLICE VEHICLES. THE STUDY REPORT SERVES AS TECHNICAL DOCUMENTATION OF THE FIELD EXPERIMENTS AND CONTAINS INFORMATION NECESSARY FOR VERIFICATION AND/OR REPLICATION OF EXPERIMENTS BY AN EVALUATIVE AGENCY OR ANOTHER POLICE DEPARTMENT. GRAPHS, MAPS, AND TABLES PRESENT DATA ON TRAFFIC ARRESTS AND ACCIDENTS DURING THE TEST PERIOD. (LK