【作者】 周兰英；

【导师】 王永清；

【作者基本信息】 四川农业大学 ， 森林培育， 2008， 博士

【摘要】 本研究以分布于四川西南山地（海拔2020～3435m）的数十种杜鹃属植物（Rhododendron sp.）为材料进行杜鹃属植物亲缘关系和遗传多样性研究,主要方法和结果如下:表型性状数量分类研究。在全面衡量杜鹃属植物性状特征的基础上,选择44个性状用SPSS10.0进行分析,其中二元性状18个、多元性状13个、数量性状13个。O聚类采用平均欧氏距离平方系数（squared Euclidean distance coefficient）,R聚类采用相关系数距离（Pearson correlation）,聚类方法采用类间平均链锁法（within-groupslinkage）。Q聚类结果将46种杜鹃属植物分为特征明显的2大类群,即有鳞类杜鹃和无鳞的常绿类杜鹃,在2大类群下的聚类结果与传统的形态分类基本吻合,但美容杜鹃和汶川星毛杜鹃的亚组归属值得商榷。R聚类表明性状被分为明显不同的组,鳞片出现部位、花器构造、外部器官尺寸等性状群内有较强的相关关系。主成分分析中,到第5个主成分累计贡献率才达70%,而前3个主成分累积贡献率达60.39%,表明杜鹃分类中存在重要性状,主要是鳞片有无、腺体和毛被的有无和叶、花、果的大小等;对分类贡献不大的性状主要有雄蕊数、萼片是否具毛、萼裂片长短、花冠颜色、花丝基部毛、花冠毛等。主成分分析结果与Q聚类结果基本一致,揭示在杜鹃分类时应选择重要性状,避免过多次要性状的干扰。利用电子显微镜对26种野生杜鹃花粉进行观察和比较研究。26种杜鹃分属杜鹃亚属（Subgonus Rhododendron）杜鹃组（10种）,常绿杜鹃亚属（Subgenus Hymenanthes）常绿杜鹃组（14种）,糙叶杜鹃亚属（Subgenus Pseudorhodorastrum）（2种）。结果表明:1)本属花粉均为四合体花粉,呈正四面体排列,单粒花粉球形或近球形,具三孔沟,极少数为四孔沟,表面粘丝多少不等。花粉粒外壁呈现大小颗粒、裂纹等不同纹饰,沟界区和极区与其他部位的纹饰并无明显差异。2)单粒花粉形状、外壁颗粒大小与形状、裂纹形状与数量、四合体间缝深浅等在种内不同居群间表现一致,但四合体直径和萌发孔长宽有一些变异。3)外壁纹饰和萌发孔长宽在各亚属间未能呈现有规律的变化,不宜作为亚属划分的依据;四合体平均直径以常绿杜鹃亚属最大,杜鹃亚属次之,糙叶杜鹃亚属最小,可作为划分亚属的参考指标。4)在同一亚属内,花粉形态特征可作为划分组或亚组的辅助手段。花粉形态特征在种间既有差异也有相似之处,不宜单独用于种的鉴定,但结合组或亚组的划分,孢粉学特征对种的确定具有一定意义。探索杜鹃RAPD反应体系和反应程序,并利用RAPD标记分析43种杜鹃属野生植物49个样品的遗传多样性、亲缘关系和系统关系。总DNA提取采用核沉淀法,最佳扩增反应体系为:4μL模板DNA（20ng/μL）,2μL 10×PCR缓冲液,0.3μL随机引物（0.50D）、1.1μL dNTPs（2.5mmol/L）、2.5μL MgCl₂（25mmol/L）,1.2μL Taq-DNA聚合酶（2.5U/μL）,8.9μL双蒸水,总反应体系20μL。优化扩增程序为:94℃预变性3min,94℃变性1min,34℃退火39s,72℃延伸1min,共45个循环,最后72℃延伸10 min,4℃保存。按Sleumer分类系统,参试材料中常绿杜鹃亚属23种25个样品,杜鹃亚属18种18个样品,糙叶杜鹃亚属2种6个样品。用遗传相似系数按非加权成对算术平均法（UPGMA）进行遗传相似性聚类,结果显示:1)24条引物共扩增出407条谱带,其中多态谱带399条,多态率达98.03%,揭示出杜鹃属内高度的遗传多样性。2)各参试样品间的遗传相似系数（GS）变化范围为0.2623-0.9059,样本间平均GS值为0.4631。3)49个样品在GS值0.59附近聚为3大类群。第一类群为常绿杜鹃亚属的25个样品,第二类群为杜鹃亚属的18个样品,第三类群为属糙叶杜鹃亚属的6个样品,聚类结果与以形态区分的3个亚属完全吻合,且第一、二类群先聚类后再与第三类群聚类,说明杜鹃亚属与糙叶杜鹃亚属的亲缘关系远于与常绿杜鹃亚属的亲缘关系,因此支持将糙叶类杜鹃单独划分为一个亚属。更多还原

【Abstract】 In this study,the relationship and genetic diversity were investigated using Rhododendron materials from the southwest mountainous region of Sichuan where the altitude is 2020-3435m and the major methods and results are as follows.The mathematic classification of morphologic character was studied.44 morphologic characters including 18 dualistic characters,13 multi-characters and 13 quantitative characters,were selected and analyzed using SPSS10.0.Squared Euclidean distance coefficient was used in case clustering and Pearson correlation was used in variable clustering by within-groups linkage.The 46 species of Rhododendron were divided into two caboodles:one was lepidote and the other was evergreen by case clustering.The different groups were formed under two caboodles and the groups were similar to the traditional classification except R.calophytum and R.asterochnoum.The results of variable clustering showed that the characters were decided in different groups.In principal component analyses（PCA）,the accumulative contribution of the fifth principal component just up to 70%while the first three principal components was up to 60.9%, which showed that some important characters can be used in classification of Rhododendron,such as squama,gland,hair and the size of leaves,flowers and fruits. However,the number of stamen,hair on sepal,size of sepal,color of crown,hair on lower silk and hair on crown etc have a small contribution on classification.The results of PCA was in according with that in case clustering,which reflected that we should pay more attention to choose character in classifying.Pollen grains of 26 wild Rhododendron in southwest mountain region of Sichuan were observed by scanning electron microscope（SEM）,in which 21 species were reported for the first time.They belonged to 3 subgenera（Subgonus Rhododendron,Subgenus Hymenanthes,Subgenus Pseudorhodorastrum）,4 sections and 12 subsections.All of them are tetrahedron with tetrad construction.Pollen grains of Rhododendron were spherical or subspherical with tricolporate and few of them with four colporates.There was unequal visein thread on the surface.The size of tetrad and the pollen exine sculpture of them were different in SEM,but no conspicuous difference by the colpus and proximal exine.The diference of pollen exine sculptur and length and width of germinal apertures did not fit for sorting the subgenera,while the diameter of tetrahedron could provide proof for further taxonomy.In the same subgenera,the morphological data of pollen could assist to identify the section and subsect.It was not suitable to identify some certain species in Rhododendron only by morphological data of pollen,however,it might be used for identification of some species in certain section and subsection.Essential factors affecting the result of Rhododendron RAPD-PCR were studied with orthogonal design.And the optimal RAPD-PCR were built as follow:DNA4μL （20ng/μL）;10×PCR buffer 2μL;Taq E 1.2μL（2.5U/μL）;primer 0.3μL（0.5OD）、dNTPs1.1μL（2.5mmol/L）、MGCl₂2.5μL（25mmol/L）,in 20μL reaction solution. Amplification reactions was performed in a 96-well thermocycler（Eppendorf Authorized Thermal Cycler PCR） programmed as follows:an initial denaturizing at 94℃for 3min followed by 45 cycles of 1min at 94℃,39s at 34℃,1min at 72℃,and finally extended at 72℃for 10 min.49 samples of 43 wild Rhododendron species from the southwest mountainous region in Sichuan were studied by using RAPD,including 25 samples of 23 species in Subgenus Hymenanthes,18 samples of 18 species in Subgenus Rhododendron and 6 samples of 2 species in Subgenus Pseudorhodorastru,according to Sleumer’s classification system. Their genetic diversity,relationship and phylogeny were analyzed.They were clustered by UPGMA.The results showed that:1) 407 DNA fragments were amplified by 24 primers, among which 399 DNA fragments were polymorphic,up to 98.03%.Dendrogram obtained by UPGMA implied the genetic diversity of Rhododendron;2) Their coefficient of genetic similarity（GS） was ranging 0.25～0.82,the average GS was 0.4631;3) The division by RAPD was consistent with the 3 subgenus classification system by morphological characters.49 samples were divided into 3 groups by cluster analysis based on the GS:the first group contained 25 samples belonging to Subgenus Hymenanthes;the second group contained 18 samples belonging to Subgenus Rhododendron,and the third group consisted of 6 samples of Subgenus Pseudorhodorastrum.The first two groups were clustered at GS 0.40 then gathered with the third group at GS 0.37.Subgenus Hymenanthes had a closer relationship with Subgenus Rhododendron than with Subgenus Pseudorhodorastru,so Subgenus Pseudorhodorastru should not be mergered into Subgenus Rhododendron.更多还原