Two T. chinensis communities, from Niushiping and Lishuzhou in Taoyuandong National Nature Reserve, are studied by using the methods of plots for the composition and structure of the communities, using ShannonWeiner index, Simpson index and Pielou index for species diversity of every layers and using diameter class frequency distribution of T. chinensis to analyze the age class frequency distribution and the regeneration of the target population. The results are as follows: (1) since the evergreen plants are dominant, the physiognomy of the two communities show some seasonal changes of dark green in summer and light yellowgreen in autumn and winter. (2) Twentysix species in 21 genera of 13 families compose the T. chinensis community of Niushiping and the T. chinensis community of Lishuzhou contains 22 families, 30 genera with 36 species. Their proportions of temperate genera of seed plants are higher than that of tropical genera, which shows the subtropical montane property of geographical elements. (3) The vertical structure of above two communities can be divided into three layers, i.e. tree layer, shrub layer and herb layer and the tree layer includes three sublayers, i.e. upper layer, middle layer and lower layer. (4) ShannonWiener index and Simpson index arrange as lower layer of tree > middle layer of tree > shrub layer > upper layer of tree in both communities, and Pielou index of middle layer of tree and lower layer of tree is higher than that of upper layer of tree and shrub layer in both communities. (5) The dominant populations of the T. chinensis community in Niushiping are Rhododendron ovatum, T. chinensis, Camellia cuspidata and Rhododendron latoucheae while the dominant populations of the T. chinensis community in Lishuzhou are Schima argentea, T. chinensis, Rhododendron latoucheae and Rhododendron hypoblematosum. (6) According to the age frequency distribution and the survival analysis, the T. chinensis population from Niushiping is a declining population and unable to regenerate itself because of lacking of enough younger age individuals; the T. chinensis population from Lishuzhou is a increasing population with a certain number of younger age individuals and relatively reasonable age structure while the high canopy density could be a limiting factor that challenge its natural regeneration. In conclusion, we suggest that in situ conservation and proper human intervention are necessary for the T. chinensis community of Niushiping and further monitoring is essential for the T. chinensis community of Lishuzhou.