Abstract: The paper is adopting rank-size rule and nearest neighbor analysis to analyze the evolution of size distribution and spatial distribution of the top 300 cities in China. Then the correlation between size distribution and spatial distribution was studied by using Granger causality test. The results show that: The urban district non-agricultural population’s size distribution of the top 300 cities is in line with rank-size rule. From the point of view trends over the past 27 years, the size distribution index Q showed a three stage change characteristics of decline-rise-fluctuation. It’s indicating that the driving forces of population aggregation and diffuseness are intertwined. R statistical values of the spatial distribution although had large fluctuations but less than 1, and it showed a significant concentration distribution patterns. Natural and socio-economic conditions of the region and the national development strategy had a profound impact on the development and spatial distribution of urban system. It had a great consistency between the size distribution index Q and spatial distribution index R of China's top 300 cities in the past 27 years. There existed a causal relationship between size distributions and spatial distribution in Chinese urban systems can be seen from the Granger causality test results. The existence of widespread urban system spatial correlation means that cities in planning development process need to not only focus on their own initiative as subjective, but also consider the impact of other neighboring cities. Make full use of space link to strengthen their development and let the urban systems gradually form a complementary industrial structure and ordered functions aggregates, to enhance their overall competitiveness.