OBJECTIVE To optimize extraction technology of couplet medicinals of Astragalus membranaceus-Puerariae lobatae. METHODS With contents of puerarin，daidzin，calycosin-7-O-β-D-glucopyranoside，daidzein，calycosin and formononetin and the yield of dry extract as index，the analytic hierarchy method was used to determine the weight coefficient of each index and calculate the comprehensive score. The effects of solid-liquid ratio， extraction times and extraction time on the comprehensive score were investigated by single factor test. The level of each factor was determined. By multi-index comprehensive scoring method， using comprehensive scores of above 7 indexes as indexes，the extraction technology of couplet medicinals of A. membranaceus-P. lobata was optimized by orthogonal experiment，and the validation tests were conducted. RESULTS The weight coefficient for the contents of puerarin，daidzin，calycosin-7-O-β-D-glucopyranoside，daidzein，calycosin and formononetin and the yield of dry extract were respectively 0.304 7，0.065 2，0.185 8，0.185 8，0.107 8，0.107 8 and 0.042 7. The optimal extraction technology was determined as follows： solid-liquid of 1∶8（g/mL），extracting 3 times and for 1 h each time. RSD of each evaluation index in the validation test results was lower than 3.00% （n＝3）. CONCLUSIONS The optimized extraction technology for A. membranaceus-P. lobata is stable and feasible.

OBJECTIVE To establish the fingerprint of Guiqi tongmai mixture and evaluate its quality in combination with chemical pattern recognition . METHODS Using narirutin as reference peak ，HPLC fingerprint of Guiqi tongmai mixture was established with Similarity Evaluation System of Chromatographic Fingerprint of TCM （2012 edition）. Common peaks identification and similarity evaluation were conducted . Hierarchical cluster analysis （HCA），principle component analysis （PCA）and orthogonal partial least square -discriminate analysis （OPLS-DA）were performed ，and the value of variable importance in the projection （VIP） greater than 1 was taken as the standard to screen the differential markers affecting the quality of Guiqi tongmai mixture . RESULTS Nineteen common peaks in the HPLC fingerprints of thirteen batches of Guiqi tongmai mixture were calibrated ，and the similarities of them were 0.928-0.990；eight common peaks were identified ，including the peak 7（hydroxysafflor yellow A ），peak 8（albiflorin），peak 9（paeoniflorin），peak 10（calycosin-7-O-glucoside），peak 12（liquiritin），peak 14（ammonium glycyrrhizinate ）， peak 16（narirutin）and peak 19（hesperidin）. According to the results of HCA ，thirteen batches of samples could be clustered into three categories ：S5，S9 and S 11-S13 were clustered into one category ；S4 and S 8 were clustered into one category ；S1-S3，S6- S7 and S 10 were clustered into one category . The results of PCA indicated that the accumulative variance contribution rate of the principle components 1-7 was 92.115%. The results of OPLS -DA showed that VIP values of narirutin ，paeoniflorin，liquiritin， ammonium glycyrrhizinate ，the components corresponding to peak 11 and peak 1 were greater than 1. CONCLUSIONS HPLC fingerprint of Guiqi tongmai mixture is established ，and it could be used for quality evaluation of Guiqi tongmai mixture by combining with chemical pattern recognition techniques . Six components such as narirutin are differential markers affecti ng the quality of Guiqi tongmai mixture .