THE SACRAMENTO STEP INVOLVED SEVEN DIFFERENT EXPERIMENTS. THE MOST SIGNIFICANT WAS ENFORCEMENT FOR DRIVING UNDER THE INFLUENCE (DUI) WHICH WAS A CONTINUING EXPERIMENT OVER THE LIFE OF THE PROJECT. THE EVALUATION REVEALED AN EXACT CORRELATION BETWEEN FATAL ACCIDENTS AND THE TIMES OF DUI ENFORCEMENT. OTHER COUNTERMEASURES TESTED UNDER THE SACRAMENTO STEP WERE PATROL, RADAR, AND FIXED-POINT EFFORTS, THE FIXED-POINT COUNTERMEASURES DID NOT REDUCE ACCIDENTS AND, IN ADDITION, HAD AN ADVERSE EFFECT ON MORALE OF STEP OFFICERS. THE RADAR AND PATROL COMBINATION WAS EFFECTIVE IN SOME EXPERIMENTS BUT NOT IN OTHERS. AS PART OF THE EVALUATION EFFORT, A MATHEMATICAL MODEL WAS DEVELOPED TO TEST WHETHER ENFORCEMENT PRODUCED ANY 'HALO' EFFECTS EITHER SPATIALLY OR TEMPORARILY. THE MODEL INDICATED THAT NO SUCH HALO EFFECTS OCCURRED. PART OF THE SACRAMENTO STEP WAS THE DEVELOPMENT OF JUDGES' CONFERENCES. THESE CONFERENCES WERE MEETINGS WITH REPRESENTATIVES OF THE POLICE, PROSECUTION, AND JUDICIAL ELEMENTS OF THE TRAFFIC SYSTEM. THESE MEETINGS WERE VERY EFFECTIVE IN IMPROVING THE SYSTEM AND ALLEVIATING PROBLEMS THAT PREVIOUSLY EXISTED. (AUTHOR ABSTRACT MODIFIED).