This thesis is based on a field study of human-elephant conflict (HfiC) in the Masai Mara dispersal areas of Transmara District, Kenya, during 1999 and 2000. The district experiences high HEC because it supports a small resident elephant population and receives elephants that disperse up the escarpment seasonally from the Masai Mara. The study examines attitudes and perceptions of people towards elephants, land use changes, changes in elephant distributions and densities, types and patterns of HEC, and the success of different mitigation methods. Attitudes and perceptions towards elephants were determined using participatory workshops and questionnaire surveys. Land use patterns were examined using aerial photo surveys, secondary data and field surveys. Elephant abundance and distribution were determined from secondary data and ground surveys. Vegetation plots were established to measure seasonal changes in natural forage, and changes in elephant diet were examined through dung analysis. Conflict data were obtained from KWS records and daily monitoring, including interruption of learning in schools. Data were analysed at fine scale using a Geographical Information System. and appropriate statistical tests were used to determine which of many variables most determined prevailing levels of HEC, and the success of mitigating HEC. Forestland was increasingly converted to cultivation, which in tum reduced elephant range and confined resident elephants to the remaining forest and to group ranches. Corridor usage by elephants increased with migration of wildebeests into the Masai Mara. Seasonal and spatial patterns in the occurrence of crop raiding incidents were determined by the maturity of maize, the area under farming, distance from the road and distance from the market centres. The success of crop protection measures depended on using a combination of traditional methods. Men who are drunk most risk being attacked by elephants. Elephants may also directly and/or indirectly interfere with learning in schools, but pupil performance was mainly determined by distance from school, absenteeism and tribe. The local community fosters negative attitudes towards elephants from which they currently receive no benefits. The future for elephants in Transmara District is bleak unless a benefit sharing, compensation and active problem elephant control programmes are implemented. Effective land use planning and participation of the community in conservation could help achieve these goals. The findings of this study have important implications for the future of elephant conservation in the face of competing human needs, both in Transmara District and elsewhere in Africa.