Following a brief discussion of both the benefits and the risks of police patrols in contributing to crime prevention, the author discusses eight major hypotheses about policing and crime by addressing several varieties of police crime prevention. After discussing the theories concerning the numbers of police, rapid response to emergency calls, random patrols, directed patrols, reactive arrests, proactive arrests, community policing, and problem oriented policing, the author details the challenges these theories pose when subjected to scientific testing. Turning to a discussion of the scientific evaluation of these eight hypotheses concerning policing and crime, the author reviews and interprets a wide variety of tests assessing the success and failure of policy attempts to combat crime. The author concludes his evaluation of policing approaches to crime reduction arguing that evidence shows that the effects of policing on crime is a complex, surprising relationship. The author maintains that the more focused the police strategy, the more likely it is to prevent crime. Furthermore, the author found that increased directed patrols in areas of high crime; proactive arrests of repeat offenders, drunk drivers, and domestic assault suspects; and problem oriented policing work well in preventing crime. Contrary to findings in earlier research studies, neighborhood block watches, arrests of juveniles for minor offenses, unemployed suspects of domestic abuse, and drug market individuals, and community policing in areas with no crime-risk factors were not found to be particularly effective in reducing community crime. Tables, references