Abstract: We studied the spatial autocorrelation and its pattern of datum land prices of cities in a region,and took Hubei Province as a case study.We employ Moran's I,local Moran's I and Moran's I correlogram to analyze the spatial autocorrelation degree and its change along with contiguity order.We use semi-variance statistic of land price based on route distance to find the spatial autocorrelation scale.We also adopt spatial clustering based on a kind of composite distance to probe into the clustering characteristic of land prices.In the case study,we find datum prices of the cities in Hubei province has faint spatial autocorrelation degree at the first and the second order contiguity.The scale of the autocorrelation is about 200 km in route distance.The spatial distribution of city land prices is a kind of hierarchcal structure.We find that land prices of cities are spatially related with the development,land profit and urban populations by principal component analysis and stepwise linear regression analysis,and exploit the mechanism of the spatial autocorrelation of urban land prices.