A descriptive cross-sectional study was carried out to determine the prevalence and to identify some socio-demographic factors associated with wife beating in the Medical Officer of Health (MOH) area of Kantale in the Trincomalee district of eastern Sri Lanka. A random sample of 417 women in the age category 18-49 years constituted the sample population. Data were obtained by focus group discussions followed by the administration of a structured questionnaire by trained interviewers. The prevalence of reported wife beating among ever-married women was 30% and the prevalence of wife beating in the year preceding the study was 22%. There was no significant association between wife beating and ethnicity of the study population or a particular age group of either the batterer or the victim. Moreover, wife beating was associated with an early age at marriage for women, low-income, a low standard of living index, large families and alcohol consumption by the batterer. A significant inverse relationship between domestic violence and the level of education of both the batterer and the victim was also identified. Contusions, typically distributed in the region of the head, face and neck were found to be the commonest type of injury suffered by battered women. A majority of women, irrespective of their level of education and employment status placed the welfare of their children as the prime reason for continuing to stay in an abusive relationship. The study concludes that wife beating is a serious health and social problem for the women population of Kantale. Intervention is recommended in relation to key issues identified by the study, including alcohol abuse by men, relative lack of education among the population, lack of family planning, societal influences promoting teenage marriages of the girl-child and absence of programs aimed at creating awareness on wife beating.