Abstract:The difficulty of ecological analysis in tropical forests due to high species diversity is a big problem. Therefore, to aggregate species with similar functional traits into functional groups for analysis is a new way to solve the problem. In this study, based on the certain functional traits (growth form, succession status, longevity, wood density, shade tolerance, elevation and forest type) and stand structural characteristics (relative biomass, breast area, height, density, frequency, canopy size, regeneration, mortality and germination), quantitative ordination analysis were used to identify the functional groups in this forest. Collected information is however summarized as follows: (1) By applying CCA (Canonical Correspondence Analysis) to analyze the factors of forest structure and species were classified into 6 functional groups, and changes were found in the related factors such as relative biomass, density, frequency, regeneration and mortality responding to the varieties of the diameters and heights of the stand; (2) By applying CCA to analyze plant functional traits, the species were classified into 5 functional groups and the related factors such as light demanding, succession status and wood density were found to be changing with the successional process; (3) based on integration of twice CCA analyses, the species of the secondary tropical montane forest could be classified into 11 functional groups, which could well represent the change of species with the forest structure and successional stages of the stand; (4) the functional group of shrubs in the early successional stage, which mainly distributed in the under-storey layer of the forest, died almost completely, coexisting with the functional groups of other successional stages; (5) the functional group of trees in the early successional stage also have high mortality, which distributed in the main canopy layer coexisting with the functional groups of the later successional stage. At present, the tropical montane secondary forest in Hainan island is in the intensive competitive stage of different functional groups.