A study is presented of the application of GIS to landslide hazard mapping. Two methods of statistical analysis are employed: the failure rate (FR) and quantification scaling type II (Q-S II). The study area is the Kulekhani watershed , located in central Nepal. A map showing the distribution of landslides was produced from air photograph interpretation and field checking. This was used to analyze the important terrain factors influencing landslide initiation. The factors employed were slope gradient, slope aspect, elevation, geology, land-use/cover, proximity to fault lines, and drainage density. Each factor was divided into several classes. The influencing factors were first determined from the FR analysis and were then considered for the Q-S II analysis. Also, two of the influencing factors were overlaid in turn to produce combined-factor layers (consisting of combined-classes). This was used as the basis for another FR analysis. Geology was found to be the most important factor influencing landslide activity. Other important factors were elevation and land use/cover. Based on the results of the three analyses, three hazard maps were produced each with four levels of hazard classes: high; moderate; less; and least. The area occupied by the different hazard classes do not show substantial variation from map to map. In general, about four percent of the study area fell into the "high hazard class." The "moderate," "less," and "least" hazard classes occupied about a quarter, a half, and a fifth of the total study area, respectively.