中文摘要：

采用高效液相色谱法，建立了滇重楼的 ＨＰＬＣ指纹图谱，使用 ＳＰＳＳ１７０软件对 １９批不同产地滇重楼根茎样品的 ＨＰＬＣ指纹图谱进行了主成分分析，在主成分得分系数矩阵的基础上，对其进行了系统聚类分析；并使用相似度评价软件进行了相似度评价验证。结果表明，１９批滇重楼根茎样品共提取出 ３个主成分，被分为 ３大类；与共有峰直接聚类相比，主成分—聚类分析更符合相似度评价结果。本研究所建立的模式识别方法，操作简便，统计结果具有可靠性，可对滇重楼化学计量分类及其质量评价提供参考。

英文摘要：

HPLC fingerprint for rhizome of Paris polyphylla var. yunnanensis was established by HPLC method; the HPLC fingerprint was analyzed with principal component analysis and systematic cluster method by means of SPSS 17.0 software, and assessed the similarity by the software of similarity evalu- ation. The results showed that three principal components were extracted from 19 samples of P. polyphylla var. yunnanensis and the samples were divided into three major categories. Compared with the direct cluster analysis according to common peaks, the results of principal component-cluster analy- sis were closer to the truth of similarity evaluation. The method of chemical pattern recognition estab- lished in this study is convenient and reliable, which can provide a reference for the stoichiometry clas- sification and quality evaluation of P. polyphylla var. yunnanensis.