Department of Botany, University of Wisconsin-Madison, 430 Lincoln Dr., Madison, WI 53706, USA.
Department of Plant Biology, Carnegie Institution for Science, 260 Panama St., Stanford, CA 94305, USA.
Mol Phylogenet Evol. 2018 May;122:59-79. doi: 10.1016/j.ympev.2018.01.014. Epub 2018 Feb 2.
Inferring interfamilial relationships within the eudicot order Ericales has remained one of the more recalcitrant problems in angiosperm phylogenetics, likely due to a rapid, ancient radiation. As a result, no comprehensive time-calibrated tree or biogeographical analysis of the order has been published. Here, we elucidate phylogenetic relationships within the order and then conduct time-dependent biogeographical and diversification analyses by using a taxon and locus-rich supermatrix approach on one-third of the extant species diversity calibrated with 23 macrofossils and two secondary calibration points. Our results corroborate previous studies and also suggest several new but poorly supported relationships. Newly suggested relationships are: (1) holoparasitic Mitrastemonaceae is sister to Lecythidaceae, (2) the clade formed by Mitrastemonaceae + Lecythidaceae is sister to Ericales excluding balsaminoids, (3) Theaceae is sister to the styracoids + sarracenioids + ericoids, and (4) subfamilial relationships with Ericaceae suggest that Arbutoideae is sister to Monotropoideae and Pyroloideae is sister to all subfamilies excluding Arbutoideae, Enkianthoideae, and Monotropoideae. Our results indicate Ericales began to diversify 110 Mya, within Indo-Malaysia and the Neotropics, with exchange between the two areas and expansion out of Indo-Malaysia becoming an important area in shaping the extant diversity of many families. Rapid cladogenesis occurred along the backbone of the order between 104 and 106 Mya. Jump dispersal is important within the order in the last 30 My, but vicariance is the most important cladogenetic driver of disjunctions at deeper levels of the phylogeny. We detect between 69 and 81 shifts in speciation rate throughout the order, the vast majority of which occurred within the last 30 My. We propose that range shifting may be responsible for older shifts in speciation rate, but more recent shifts may be better explained by morphological innovation.
在真双子叶植物目石南目内推断家族间的关系一直是被子植物系统发育中比较棘手的问题之一,这可能是由于其快速而古老的辐射所致。因此,尚未发表关于该目的全面的时间校准树或生物地理分析。在这里,我们阐明了该目的系统发育关系,然后使用一种分类群和基因座丰富的超级矩阵方法,对三分之一的现存物种多样性进行时间相关的生物地理和多样化分析,该方法经过了 23 个宏观化石和两个二级校准点的校准。我们的结果证实了以前的研究,也提出了一些新的但支持度较低的关系。新提出的关系包括:(1)全寄生的肉实科与千屈菜科为姐妹群,(2)肉实科+千屈菜科组成的分支与不包括紫金牛科的石南目为姐妹群,(3)山茶科与安息香科+肉实科+石南科为姐妹群,(4)石楠科的亚科关系表明,鹿蹄草亚科与水晶兰亚科为姐妹群,而Pyroloideae 与除鹿蹄草亚科、安石榴亚科和水晶兰亚科以外的所有亚科为姐妹群。我们的结果表明,石南目于 1.1 亿年前在印度-马来西亚和新热带地区开始多样化,这两个地区之间的交流以及从印度-马来西亚向外扩展成为许多科现存多样性形成的重要区域。快速的分支发生在 1.04 亿至 1.06 亿年前的目主干上。跳跃扩散在目内很重要,在过去的 30 万年中,扩散是谱系中较深层次的分裂的最重要的分支发生驱动因素。我们在整个目中检测到 69 到 81 次物种形成率的转变,其中绝大多数发生在过去的 30 万年中。我们提出,范围转移可能是导致较早的物种形成率转变的原因,但最近的转变可能更好地解释为形态创新。