橐吾属（菊科-千里光族）的分类学研究

Approximately 140 species are now recognized in the genus Ligularia (Asteraceae), with 123 recorded to occur in China. The current taxonomy of this genus, however, has been revealed to be far from satisfactory through our previous studies. Errors in taxonomic treatment, particularly on species level, are rather prevalent in the genus. These mainly include uncritical citation of synonyms, misinterpretation of the identity of species, description of old species as new, and inaccuracy of geographical information of taxa. The current infrageneric classification system of Ligularia, proposed on the basis of few gross-morphological characters, is also of a rather low predictive value. The lack of critical survey of literature and type material by some earlier authors, the poor knowledge of both macro- and micro-morphological characters (such as floral micromorphology, palynology, and cytology) for the genus, the insufficient analysis of character variations, and the scantiness of specimens available for study in many species, are the major causes leading to these errors. Previous molecular systematic analyses sampled only a small portion of the species of the genus and utilized a few phylogenetically low-informative DNA fragments, and the results did not help much to improve the infrageneric division of the genus. This project aims to address these problems through multidisciplinary approaches. After a critical survey of all the relevant literature and herbarium specimens, extensive field observations, and an in-depth investigation of all the macro- and micro-morphological characters, we will carefully re-evaluate variation patterns of all these characters and determine their taxonomic applicability at different taxonomic levels. Our taxonomic decisions will be made based strictly on the results of re-evaluation of characters. In the light of our thorough revision of the genus, we then will try to re-construct its phylogeny using evidence from different disciplines, and apply the results of phylogenetic analyses to improve the infrageneric classification system and to clarify the evolutionary pattern of characters. Our final objective is to produce a taxonomic monograph of Ligularia in the true sense, which should synthesize all the information that is known about the genus. This project will help us gain a better understanding of species composition and differentiation in China, and will also provide invaluable basic scientific information for the utilization and protection of the resources of Ligularia, in which some species are medicinally and/or ecologically important.

菊科橐吾属目前约承认140种，其中我国123种。由于前人在文献考证、标本室和野外考察以及实验室等方面的工作有欠深入，许多种类标本过少，在性状（包括宏观和微观性状）的获取和性状的分析方面都存在明显不足，分子系统学研究涉及的种类和所用DNA片段偏少，因此该属中还有许多基本分类学问题（如异名错误归置、名实不符、目前承认的一些种类不能成立、地理分布记载混乱、属下分类系统仅基于少数性状建立而缺乏预见性等）有待解决。本项目拟将文献考证、标本室和野外观察、实验手段紧密结合起来，对橐吾属各主要宏观和微观性状进行深入研究，评价和确定性状的分类和系统学价值，完成种级水平的分类修订，并利用多学科证据重建该属的系统发育，改进其属下分类系统。最终目标为完成该属的分类学专著。橐吾属主要分布于我国，具有药用和生态价值，澄清其分类学问题可增加对我国物种组成和分化等问题的理解，同时为植物资源的利用和保护提供基础科学资料。

利用多学科手段对菊科橐吾属 (Ligularia)进行了分类学研究。全面收集了橐吾属及其近缘属的分类学文献，建立了该属非常完善的分类学文献数据库，研究了国内外主要标本馆的23000余份橐吾属及其他近缘属的标本，采集了85种橐吾属植物多种其他近缘属植物的标本、DNA材料和活材料。形态学研究表明，叶脉类型、花序类型、植株颜色、冠毛长短、头状花序的大小、叶形、小花数目等是在种级水平比较可靠的分类性状。对橐吾属60种110居群及其近缘属植物展开了染色体研究，确认了橐吾属及垂头菊属的染色体基数为x = 29、30，其中前者为比较常见的一个基数，后者虽不常见，但也确定存在。分子系统学研究表明橐吾属为多系群，共分为4支。这一分化式样否定了该属所有属级和属下系统分类性状的系统学价值，但和地理分布存在显著相关性。研究结果还表明，橐吾属的系统发育关系在叶绿体基因和核糖体基因树上存在显著矛盾，暗示该属存在复杂的网状进化历史。