Abstract: We analyzed the spatial heterogeneity and patch shape complexity of a forest landscape within 3000 m on both sides of the Ming Great Wall in Beijing. A complexity evaluation index was developed based on the most commonly used landscape indexes, along with the use of fractal theory. The buffer analysis method was used to examine the landscape heterogeneity and patch complexity under different buffer amplitudes. The results showed that: 1) Forest land and shrub land were the most widely distributed and dominant landscape types, accounting for about 83% of the total study area. Forest land represented the landscape matrix with the strongest aggregation, best connectivity and lowest degree of fragmentation; open forest and non-stumpage forest land had a high degree of discrete patches, poor landscape connectivity and high fragmentation. 2) With the increase of buffer distance, the proximity of the same kind of landscape patches, landscape connectivity, average patch size and landscape heterogeneity increased. The amount of discrete patches and fragmentation decreased, and patch types approached a uniform distribution. 3) The expansion of buffer distance also gradually increased patch shape complexity and the influence of edge effects, but diminished landscape structure stability. These results show that a historical monument, like the Ming Great Wall of China, may have a significant effect on the surrounding landscape.