In this paper, we analyze cow endemic fluorosis spatial distribution characteristics in the target district based on combination of spatial statistical analysis and GIS. This disease geographical distribution relates closely to the environmental factors of habitats and the correct analysis of the spatial distribution of the disease for monitoring and prevention plays an important role. However, the amount of monitoring spots is smaller and the monitored data are very limited. How to use the limited data to show the general spatial distribution of the disease is a key issue in building the efficient spatial monitoring method. We use GIS application software ArcGIS9.1 to overcome the lack of data of sampling sites and establish a prediction model in estimation of the disease distribution. Comparing with the true disease distribution, the prediction model has a well coincidence and is feasible and provides a fundamental application for other study on space-related cow diseases.