Botanical Museum, P.O. Box 7, FI-00014 University of Helsinki, Finland.
Mol Phylogenet Evol. 2010 May;55(2):381-98. doi: 10.1016/j.ympev.2010.02.004. Epub 2010 Feb 10.
Analysis of an extensive new molecular dataset for the moss class Polytrichopsida provides convincing support for many traditionally recognised genera and identifies higher level phylogenetic structure with a strong geographic component. A large apical clade that is most diverse in the northern hemisphere is subtended by a grade of southern temperate and tropical genera, while the earliest diverging lineages have widely separated relictual distributions. However, there is strongly supported topological incongruence between the nuclear 18S rRNA gene tree and the chloroplast and mitochondrial data for the positions of some taxa and notably for the status of Pogonatum. While Pogonatum is unambiguously paraphyletic in the 18S tree, it is well supported as monophyletic by the combined chloroplast and mitochondrial data, this being corroborated by several distinctive morphological synapomorphies and a 51-53 bp deletion in the rps4-trnS spacer. We explore various reticulate historical processes and methodological issues as possible explanations for incongruence, and suggest that either (1) the 18S topology is an artefact created by convergence of substitutions at specific sites due to functional and/or molecular-structural constraints not accounted for by the model, or (2) the incongruence is a product of ancient hybridization events. Under the latter scenario, incongruent topologies for Pogonatum are parsimoniously explained if Polytrichum (including Polytrichastrum sect. Aporotheca) is ultimately descended from a hybridization event involving an extinct maternal taxon derived from the branch ancestral to the combined Pogonatum/Polytrichum s.l. clade, and a paternal taxon belonging to (or ancestral to) the apical Pogonatum group to which the majority of extant species belong. Numerous novel relationships of taxonomic and evolutionary significance are supported. Notably, both Polytrichastrum and Oligotrichum are polyphyletic. While Polytrichastrum sect. Aporotheca is closely related to Polytrichum, other species, including the type, are not. The large majority of Oligotrichum species sampled occur in one of two distantly related clades with predominantly northern and southern hemisphere distributions, respectively, implying convergent evolution of this morphology in each of the two temperate zones.
对广泛的新分子数据集进行分析，为藓类植物的许多传统公认属提供了令人信服的支持，并确定了具有强烈地理成分的高级系统发育结构。一个在北半球最多样化的大型顶枝群由一个南温带和热带属的梯度支撑，而最早分化的谱系具有广泛分离的残余分布。然而，核 18S rRNA 基因树与叶绿体和线粒体数据之间在某些分类群的位置上存在强烈支持的拓扑不一致，特别是在 Pogonatum 的地位上。虽然 Pogonatum 在 18S 树中明显是并系的，但它在叶绿体和线粒体数据中被很好地支持为单系，这得到了几个独特的形态同源特征和 rps4-trnS 间隔区的 51-53bp 缺失的支持。我们探讨了各种网状历史过程和方法问题作为可能的不一致解释，并提出以下两种可能性：（1）18S 拓扑是由于特定位点的替代趋同造成的，这是由于模型未考虑的功能和/或分子结构约束造成的；（2）不一致是古老杂交事件的产物。在后一种情况下，如果 Polytrichum（包括 Polytrichastrumsect. Aporotheca）最终是由涉及一个已灭绝母系的杂交事件产生的，而这个母系是从与合并的 Pogonatum/Polytrichum s.l. 枝系祖先分支分离的，那么 Pogonatum 的不一致拓扑就可以很容易地解释，而父系分类群属于（或祖先于）现存物种绝大多数所属的顶端 Pogonatum 组。支持了许多具有分类学和进化意义的新关系。值得注意的是，Polytrichastrum 和 Oligotrichum 都是多系的。虽然 Polytrichastrumsect. Aporotheca 与 Polytrichum 密切相关，但其他物种，包括模式种，则不然。所采样的大多数 Oligotrichum 物种都属于两个相距很远的分支之一，分别分布在北半球和南半球，这表明这种形态在两个温带地区都经历了趋同进化。
