Abstract: To establish HPLC fingerprints of methanolic extract from Bauhinia aurea of different batches. HPLC method was used to establish fingerprints. The chromatographic column was Inertsil ODS-SP C₁₈ column （4.6 mm×250 mm, 5 μm）, the mobile phase was gradient acetonitrile-0.1% phosphoric acid solution, the flow rate was 1.0 mL/min, the column temperature was 30 ℃, the injection volumn was 10 μL, and the detection wavelength was 212 nm. Similarity evaluation, cluster analysis and principal component analysis were used for data analysis. The HPLC fingerprints of Bauhinia aurea of 10 batches were established. 11 common peaks were found in the fingerprints, three of them were identified as components epicatechin, hyperoside and quercitin. The similarities of 10 samples were all over 0.9. The results of cluster analysis and principal component analysis were divided different batches of Bauhinia aurea into four categories. The established fingerprint method and analysis models are stable and reliable, which can provide the basis for quality evaluation of this Zhuang medicinal herb.

Key words: Zhuang medicinal herb, Bauhinia aurea, HPLC, fingerprint, cluster analysis, principal component analysis