In this study, the appearance and texture of E. rutaecarpa were linked with the chemical constituents to explore methods of classification of E. rutaecarpa. The Chemometrics such as Hierarchical cluster analysis (HCA), principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) models were used for analysis. According to the models, samples of E. rutaecarpa were divided into three categories based on their source: Evodia, Stone Tiger and Sparse Evodia. The Evodia category could be subdivided into two categories, one representing large fruits with a greater degree of cracking and the other representing large fruits with little or no cracking. The method provided by this study combines chemometrics with HPLC fingerprints, which can provide a basis and reference for the identification of E. rutaecarpa and enables establishment of a grade standard.