OBJECTIVE： To establish HPLC fingerprint of Paeonia suffruticosa， and to conduct cluster analysis. METHODS： HPLC-MS was adopted. The determination was performed on Ecosil C18 column with mobile phase consisted of 0.2％ formic acid solution-acetonitrile （gradient elution） at the flow rate of 1.0 mL/min. The column temperature was 25 ℃. The sample size was 10 μL. With the reference of paeonol， HPLC chromatograms of 69 batches of samples （S1-S69） were determined. The similarity evaluation was conducted by Similarity Evaluation System for Chromatographic Fingerprint of TCM （2004 A edition）. The common peaks were further identified. Corresponding compenents of commen peaks were identified by HPLC-MS [same HPLC condition as above； MS condition included ESI， positive and negative ion model， mass scanning range of m/z 100-1 200， dry gas of N2 （99.999 9％）， dry gas velocity of 15 L/ min， dry gas temperature of 350 ℃， collision energy of 4 V， and the collision energy of 10-40 V]. Cluster analysis was carried out by using SPSS 20.0 statistical software. RESULTS： There were 29 common peaks in HPLC chromatograms of 69 batches of samples， and the similarity was more than 0.900 cexcept for 6 batches of samples； the corresponding components of 24 peaks were identified totally by using HPLC method. 69 batches of samples were divided into 5 categories. S19， S22， S26， S29， S30， S33， S34， S38， S43， S46， S49-S51， S53， S56， S57， S64， S65 could be grouped into category Ⅰ； S4-S6， S41 could be grouped into category Ⅱ； S1-S3， S7-S10， S12， S24， S25， S54， S59 could be grouped into category Ⅲ； S15， S23， S35-S37， S39， S40， S47， S48， S52， S58， S66， S67， S69 could be grouped into category Ⅳ； S11， S13，S14， S16-S18， S20， S21， S27， S28， S31， S32， S42， S44， S45， S55， S60-S63， S68 could be grouped into category Ⅴ. CONCLUSIONS： The established HPLC fingerprint， component identification and cluster analysis can provide reference for identification and quality evaluation of P. suffruticosa.