The plant community is an assemblage of plant populations that live in certain area, and interact with and adapt to each other in the context of long-term environmental changes. The quantitative analysis of forest community is an important approach to research the community classification and sorting. The numerical classification and ordination of plant communities can objectively reveal the ecological relationship between plant communities, distribution and the environmental factors which provides the important theoretical basis for forest vegetation restoration and reconstruction, forest management, and biodiversity conservation. Although a fast-growing literature on the numerical classification and ordination of plant communities are published in recent years, yet little is known about the numerical classification and ordination of Cyclobalanopsis glauca communities in karst hills of Guilin, Southwest China, which prevents us from to determine the environmental factors influencing the distribution of Cyclobalanopsis glauca communities. Therefore, it is necessary to carry out a case study about the numerical classification and ordination of Cyclobalanopsis glauca communities in karst hills of Guilin. To explore this uncertainty, based on the environmental factors and community structure data by the systematic field investigation of 20 plots of Cyclobalanopsis glauca communities, we conducted the quantitative classification of Cyclobalanopsis glauca communities with Ward clustering, two-way indicator species analysis (TWINSPAN), and multivariate regression trees (MRT) methods. The relationship between the distribution of species and environmental factors was analyzed by the redundancy analysis (RDA) ordination method. The results showed that:(1) the results of the three classification methods were basically consistent. The 20 plots of Cyclobalanopsis glauca communities in karst hills of Guilin were divided into 3 associations types:Ass. Cyclobalanopsis glauca-Mallotus philippensis + Bauhinia championi + Alchornea trewioides-Carex tristachya, Ass. Cyclobalanopsis glauca-Bauhinia championi + Alchornea trewioides + Fordia cauliflora-Ophiopogon platyphyllus + Carex tristachya, Ass. Cyclobalanopsis glauca-Mallotus philippensis + Fordia cauliflora-Carex tristachya. (2) Both redundancy analysis and three numerical classification exhibited consistent results. The redundancy analysis clearly reflected the distribution range of three association types and the relationships between association types and the environmental factors. Among 10 environmental factors, the rock bare rate, soil water content, soil pH value, soil total nitrogen, soil total phosphorus, soil total potassium, and available potassium have significant effects on the distribution across the 3 associations types. The changes of soil water content and total potassium content were main ecological factors influenced the composition and distribution of Cyclobalanopsis glauca communities in karst hills of Guilin, Southwest China.