ALTHOUGH THERE IS A PAUCITY OF STUDIES ON THE INCIDENCE, VICTIMS, AND MANAGEMENT OF FAMILY VIOLENCE, A NUMBER OF GENERALIZATIONS MAY BE MADE FROM EXISTING DATA. FAMILY VIOLENCE OCCURS BETWEEN SPOUSES WITH BOTH PARTNERS HAVING RECOURSE TO VIOLENCE. WOMEN TEND TO BE MORE FREQUENTLY AND SEVERELY INJURED THAN MEN. POLICE INTERVENTION APPEARS TO BE SOUGHT TO PREVENT FURTHER ESCALATION OF HOSTILITIES, NOT TO PRESS CRIMINAL CHARGES. POLICE INTERVENTION IN FAMILY DISPUTES IS USUALLY HIGHLY DISCRETIONARY; THE GENERAL PRACTICE IS TO SEPARATE THE DISPUTANTS, ARRANGE FOR MEDICAL AID IF NEEDED, AND DETERMINE IF THERE IS PROBABLE CAUSE FOR ARREST. POLICE APPROACHES MAY INCLUDE THE USE OF AUTHORITY TO IMPOSE AN END TO THE CONFLICT AND INFORM THE PARTIES OF THEIR LEGAL STANDING, NEGOTIATION OR MEDIATION OF THE CONFLICT, OR COUNSELING THE DISPUTANTS. THE GENERALLY UNSATISFACTORY OUTCOMES OF ARREST AND PROSECUTION, THE PRACTICAL DIFFICULTIES INVOLVED IN ARREST AND PROSECUTION, AND THE IMMEDIATE AVAILABILITY OF THE POLICE OFFICER, MAKE FAMILY CRISIS INTERVENTION BY THE POLICE AN APPROACH OF CHOICE FOR MENTAL HEALTH SERVICE DELIVERY. A 1967 DEMONSTRATION PROJECT IN NEW YORK CITY UTILIZING A POLICE FAMILY CRISIS INTERVENTION APPROACH HAS SET THE TREND FOR IMPLEMENTATION OF SIMILAR PROGRAMS. THE NEW YORK PROJECT EMPHASIZED CLASSROOM AND FIELD TRAINING IN FAMILY CRISIS INTERVENTION AND MANAGEMENT, AND UTILIZATION OF THE COMMUNITY REFERRAL NETWORK. IN A GENERAL-SPECIALIST MODEL OF IMPLEMENTATION, A SELECT GROUP OF OFFICERS IS TRAINED AND HANDLES ALL FAMILY DISPUTES IN A SPECIFIED AREA. THE GENERALIST MODEL IN WHICH ALL MEMBERS OF THE FORCE ARE GIVEN TRAINING IN FAMILY CRISIS THEORY AND PRACTICE IS MORE SUITABLE FOR A SMALL FORCE. A THIRD MODEL, ALTHOUGH APPARENTLY THE LEAST DESIRABLE, IS THE SPECIALIST MODEL IN WHICH HIGHLY TRAINED OFFICERS HANDLE FAMILY DISPUTES AND HAVE NO OTHER POLICE RESPONSIBILITIES. ALTHOUGH MANY SIMILAR PROGRAMS HAVE BEEN IMPLEMENTED, FEW HAVE BEEN STRINGENTLY EVALUATED. FINDINGS OF TWO EVALUATIVE STUDIES OF SUCH PROGRAMS WERE GENERALLY FAVORABLE BUT NOT CONCLUSIVE. FACTORS SIGNIFICANTLY ASSOCIATED WITH PROGRAM SUCCESS WERE FOUND TO BE TRAINING CONTENT, ADMINISTRATIVE SUPPORT, AND LINKAGES AND RELATIONSHIPS WITH THE COMMUNITY AND ITS REFERRAL RESOURCES. ISSUES IN POLICY FORMULATION RELATING TO WOMEN'S RIGHTS AND FEMALE EQUALITY AND POLICE DISCRETIONARY DECISIONS ARE DICUSSED. THE NEED FOR FURTHER RESEARCH INTO FAMILY CRISIS INTERVENTION AND THE POLICE ROLE IS ALSO EXAMINED. MANY REFERENCES ARE INCLUDED. (JAP)