OBJECTIVE To establish the fingerprints of Ligusticum sinense from different habitats ，screen differential components and determine their contents. METHODS Using Z-ligustilide as reference ，HPLC fingerprints of 12 batches of L. sinense were established by using Similarity Evaluation System of Chromatographic Fingerprints of TCM （2012 edition）；common peaks were identified and their similarities were evaluated. Cluster analysis （CA），principal component analysis （PCA）and orthogonal partial least squares-discriminant analysis （OPLS-DA）were performed to screen differential components with variable importance in the projection （VIP）＞1 as standard ；meanwhile，the contents of above differential components were determined by the same HPLC method. RESULTS There were 17 common peaks in the fingerprints of 12 batches of L. sinense ，and their similarities ranged 0.989-1.000. A total of 9 common peaks were identified ，i.e. chlorogenic acid （peak 1），ferulic acid （peak 2）， senkyunolide Ⅰ（peak 7），coniferyl ferulate （peak 9），E-ligustilide（peak 13），senkyunolide A （peak 14），Z-ligustilide（peak 17）. CA results showed that 12 batches of L. sinense were divided into 3 categories，S1-S5（Wuning）were clustered into one category，S6-S8（Ruichang）were clustered into one category ，S9-S12（De’an）were clustered into one category ；the VIP values of peaks 2，13，14 and 17（corresponding to ferulic acid ，E-ligustilide，senkyunolide A ，and Z-ligustilide respectively ）were all greater than 1，respectively. In S 1-S5，S6-S8 and S 9-S12 samples，the contents of ferulic acid were 0.488-0.533，0.603-0.658 and 0.415-0.433 mg/g，respectively；senkyunolide A were 1.184-1.295，1.450-1.588 and 1.307-1.377 mg/g，respectively；E-ligustilide were 0.118-0.125，0.130-0.135 and 0.223-0.229 mg/g，respectively；Z-ligustilide were 7.200-7.681，8.076-8.643 and 4.508-4.996 mg/g， respectively；the differences between two groups were statisti-cally significant （P＜0.05）. CONCLUSIONS Established ARS-11）；fingerprint is simple and accurate ，and can be used for overall quality evaluation of L. sinense from different habitats by combining with multivariate statistical analysis. Ferulic acid ， senkyunolide A ，Z-ligustilide and E-ligustilide may be the differential components that affect the quality of L. sinense from different habitats ，the contents of the first 3 components in L. sinense from Ruichang are the highest ，and the content of E-ligustilide in samples from De’an is the highest.

OBJECTIVE To screen the primary processing methods of Ligusticum sinense slice based on differential metabolites， and provide theoretical basis for the scientific processing of L. sinense. METHODS Using 13 groups of L. sinense slice processed by fresh-cutting or traditional methods as samples， UHPLC-QE-MS was employed for metabolite identification. Multivariate statistical analysis was applied to screen differential metabolites among the 13 sample groups， analyzing the effects of washing， soaking， drying methods， and drying cycles on both the relative expressions of differential metabolites and the contents of carboxylic acids and their derivatives in the samples （to reflect the total amino acid content）. RESULTS Principal component analysis and partial least squares-discriminant analysis both showed significant intergroup differences among the 13 sample groups. A total of 688 differential metabolites were screened from the 13 sample groups， with carboxylic acids and their derivatives showing the highest proportion. The relative expression levels of phosphatidylcholine significantly increased after washing treatment， while tryptophan expression significantly decreased after soaking treatment. Samples dried at 50-60 ℃ showed significantly increased expression of psoralen， whereas those dried at 40 ℃ showed significantly decreased expression of methyl -p- methoxycinnamate. Both washing and soaking treatments significantly reduced the total amino acid content in samples， while secondary drying significantly increased it. The three controlled-temperature drying methods maintained relatively stable total content of amino acids in samples. CONCLUSIONS The optimal processing protocol for L. sinense slice is as follows: fresh L. sinense slice should be freshly cut at the production site， undergo quick washing after soil removal， and be dried twice at 40 ℃ （before and after slicing）.