Department of Botany, Swedish Museum of Natural History, Box 50007, SE-104 05 Stockholm, Sweden.
Department of Biological Sciences, Campbell University, 205 Day Dorm Road, Buies Creek, NC 27506, USA.
Syst Biol. 2019 Nov 1;68(6):1020-1033. doi: 10.1093/sysbio/syz025.
Oceanic islands originate from volcanism or tectonic activity without connections to continental landmasses, are colonized by organisms, and eventually vanish due to erosion and subsidence. Colonization of oceanic islands occurs through long-distance dispersals (LDDs) or metapopulation vicariance, the latter resulting in lineages being older than the islands they inhabit. If metapopulation vicariance is valid, island ages cannot be reliably used to provide maximum age constraints for molecular dating. We explore the relationships between the ages of members of a widespread plant genus (Planchonella, Sapotaceae) and their host islands across the Pacific to test various assumptions of dispersal and metapopulation vicariance. We sampled three nuclear DNA markers from 156 accessions representing some 100 Sapotaceae taxa, and analyzed these in BEAST with a relaxed clock to estimate divergence times and with a phylogeographic diffusion model to estimate range expansions over time. The phylogeny was calibrated with a secondary point (the root) and fossils from New Zealand. The dated phylogeny reveals that the ages of Planchonella species are, in most cases, consistent with the ages of the islands they inhabit. Planchonella is inferred to have originated in the Sahul Shelf region, to which it back-dispersed multiple times. Fiji has been an important source for range expansion in the Pacific for the past 23 myr. Our analyses reject metapopulation vicariance in all cases tested, including between oceanic islands, evolution of an endemic Fiji-Vanuatu flora, and westward rollback vicariance between Vanuatu and the Loyalty Islands. Repeated dispersal is the only mechanism able to explain the empirical data. The longest (8900 km) identified dispersal is between Palau in the Pacific and the Seychelles in the Indian Ocean, estimated at 2.2 Ma (0.4-4.8 Ma). The first split in a Hawaiian lineage (P. sandwicensis) matches the age of Necker Island (11.0 Ma), when its ancestor diverged into two species that are distinguished by purple and yellow fruits. Subsequent establishment across the Hawaiian archipelago supports, in part, progression rule colonization. In summary, we found no explanatory power in metapopulation vicariance and conclude that Planchonella has expanded its range across the Pacific by LDD. We contend that this will be seen in many other groups when analyzed in detail.
大洋岛屿起源于火山活动或构造活动,与大陆陆地没有联系,由生物体殖民,并最终由于侵蚀和沉降而消失。大洋岛屿的殖民化是通过远距离扩散(LDD)或物种形成发生分歧来实现的,后者导致谱系比它们栖息的岛屿更古老。如果物种形成发生分歧是有效的,那么岛屿的年龄就不能可靠地用于为分子定年提供最大年龄限制。我们探讨了广泛分布的植物属(Planchonella,山榄科)成员的年龄及其在太平洋的宿主岛屿之间的关系,以测试扩散和物种形成发生分歧的各种假设。我们从代表约 100 种山榄科的 156 个标本中采样了三个核 DNA 标记,并在 BEAST 中使用松弛时钟对这些标记进行分析,以估计分歧时间,并使用系统地理扩散模型来估计随时间的范围扩张。该系统发育树用一个二级点(根)和来自新西兰的化石进行校准。经时系统发育树表明,Planchonella 物种的年龄在大多数情况下与它们所栖息的岛屿的年龄一致。推测 Planchonella 起源于 Sahul 大陆架地区,并多次向该地区反向扩散。过去 2300 万年,斐济一直是太平洋地区范围扩张的重要来源。我们的分析在所有测试的情况下都拒绝了物种形成发生分歧,包括大洋岛屿之间、斐济-瓦努阿图特有植物区系的进化以及瓦努阿图和洛亚蒂群岛之间向西回溯的物种形成发生分歧。反复扩散是唯一能够解释经验数据的机制。识别出的最长(8900 公里)扩散是在太平洋的帕劳和印度洋的塞舌尔群岛之间,估计发生在 220 万年前(0.4-4.8 万年前)。夏威夷谱系(P. sandwicensis)的第一个分裂与内克尔岛(1100 万年)的年龄相匹配,当时它的祖先分化成两种以紫色和黄色果实区分的物种。随后在整个夏威夷群岛的建立支持了部分进展规则的殖民化。总之,我们在物种形成发生分歧中没有发现解释力,并得出结论认为 Planchonella 通过 LDD 扩展了其在太平洋的范围。我们认为,当详细分析时,许多其他群体也会看到这一点。