The spatial organization of industries is the main force in shaping the urban spatial structure and is the result of balance between forces of agglomeration and dispersion. With the knowledge of spatial organization of urban industries, the underlying mechanism of the evolution of the urban spatial structure can be better understood, and proper urban planning or zoning can be achieved. However, the characteristics of the spatial organization of industries have not been discussed in depth. This paper aimed at the classification and characteristics of the industrial spatial organization in cities with two considerations: industrial concentration and geographical agglomeration. Industrial concentration reveals the extent to which the market of industries is competitive, which in a way contributes to the geographical concentration of industries. Geographical agglomeration is attributable to the external economies of scale which is the main force for the formation of industrial clusters. In methodology, this paper measures industrial concentration by Herfindal-Herschman index (HHI), which provides a complete picture of industrial concentration by calculating the sum of squared employment shares. Geographical agglomeration is measured by MS index based on the probabilistic model of location choice developed by Maurel and Sedillot. The study case is Beijing which is divided into 1049 geographical units (4 km ×4 km) for spatial research. With the economic census data of Beijing in 2001, this paper examines the patterns of spatial organization of industries by analyzing two aforementioned dimensions. This paper classifies 29 2-digit manufacturing industries and 40 2-digit service industries into 6 patterns of oli-cluster, oli-disperse, leading-cluster, leading-disperse, competing-cluster and competing-disperse. The oli-cluster pattern features oligopolistic market and geographical agglomeration. The oli-disperse pattern features oligopolistic market and geographical dispersion. The leading-cluster pattern features less competitive market and geographical agglomeration. The leading-disperse pattern features less competitive market and geographical dispersion. The competing-cluster pattern features competitive market and geographical agglomeration. The competing-disperse pattern features competitive market and geographical dispersion. Sample industries from every pattern are analyzed, and factors responsible for each pattern are discussed in short. Finally, a comparison between manufacturing industries and service industries was made regarding patterns of spatial organization, and the findings include that manufacturing industries and service industries have similar structure of industrial concentration while a larger proportion of manufacturing industries is geographically dispersed compared to service industries. Moreover, the mean values of HHI from manufacturing industries across three classifications types are close to that from service industries while the mean values of MS index from manufacturing industries across two classification types are significantly higher than that from service industries.