Macroevolutionary Patterns of Flowering Plant Speciation and Extinction
Total Page:16
File Type:pdf, Size:1020Kb
PP69CH25_Vamosi ARI 4 April 2018 8:6 Annual Review of Plant Biology Macroevolutionary Patterns of Flowering Plant Speciation and Extinction Jana C. Vamosi,1 Susana Magallon,´ 2 Itay Mayrose,3 Sarah P. Otto,4 and HerveSauquet´ 5,6 1Department of Biological Sciences, University of Calgary, Calgary, Alberta T2N 1N4, Canada; email: [email protected] 2Instituto de Biologıa,´ Universidad Nacional Autonoma´ de Mexico,´ Ciudad de Mexico´ 04510, Mexico´ 3Department of Molecular Biology and Ecology of Plants, George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv 69978, Israel 4Department of Zoology and the Biodiversity Research Centre, University of British Columbia, Vancouver, British Columbia V6T 1Z4, Canada 5Laboratoire Ecologie,´ Systematique,´ Evolution,´ Universite´ Paris-Sud, CNRS UMR 8079, 91405 Orsay, France 6National Herbarium of New South Wales (NSW), Royal Botanic Gardens and Domain Trust, Sydney, NSW 2000, Australia Annu. Rev. Plant Biol. 2018. 69:685–706 Keywords First published as a Review in Advance on biome, conservation, dispersal, diversification, pollination, polyploidy, February 28, 2018 sexual system The Annual Review of Plant Biology is online at plant.annualreviews.org Abstract https://doi.org/10.1146/annurev-arplant-042817- Species diversity is remarkably unevenly distributed among flowering plant 040348 Annu. Rev. Plant Biol. 2018.69:685-706. Downloaded from www.annualreviews.org lineages. Despite a growing toolbox of research methods, the reasons under- Access provided by University of British Columbia on 05/17/18. For personal use only. Copyright c 2018 by Annual Reviews. lying this patchy pattern have continued to perplex plant biologists for the All rights reserved past two decades. In this review, we examine the present understanding of transitions in flowering plant evolution that have been proposed to influence speciation and extinction. In particular, ploidy changes, transitions between ANNUAL REVIEWS Further tropical and nontropical biomes, and shifts in floral form have received at- Click here to view this article's online features: tention and have offered some surprises in terms of which factors influence • Download figures as PPT slides speciation and extinction rates. Mating systems and dispersal characteristics • Navigate linked references • Download citations once predominated as determining factors, yet recent evidence suggests that • Explore related articles • Search keywords these changes are not as influential as previously thought or are important only when paired with range shifts. Although range extent is an important 685 PP69CH25_Vamosi ARI 4 April 2018 8:6 correlate of speciation, it also influences extinction and brings an applied focus to diversification research. Recent studies that find that past diversification can predict present-day extinction risk open an exciting avenue for future research to help guide conservation prioritization. Contents 1. INTRODUCTION . 686 2. KEY INNOVATIONS AND DIVERSIFICATION PATTERNS OVERLONGTIMESCALES................................................. 690 2.1.DiversificationRateShiftsinAngiosperms................................... 690 2.2.KeyFloralInnovationsinAngiosperms...................................... 691 3. GENOMIC AND GEOGRAPHICAL INFLUENCES ON ANGIOSPERM DIVERSIFICATION.......................................................... 693 3.1. Whole-Genome Duplications in the Evolutionary History of Angiosperms . 693 3.2.GeographicalPatternsinAngiospermDiversification......................... 695 4. MECHANISMS UNDERLYING PATTERNS OF ANGIOSPERM DIVERSIFICATION.......................................................... 695 4.1.TheContextDependenceofTraitTransitions............................... 696 4.2. Constraints of Trait Evolution and Their Influence on Angiosperm Diversification............................................................... 696 5. EXTINCTION PATTERNS AND ANGIOSPERM DIVERSIFICATION . 697 5.1. Macroevolutionary Patterns of Extinction. 697 5.2. The Fate of Flowering Plants in the Anthropocene . 698 1. INTRODUCTION Angiosperms, or flowering plants, are a diverse group consisting of 304,000 named species—and potentially as many as 156,000 unnamed (102)—that originated more recently than any other clade of vascular plants, between approximately 140 and 250 Mya (37, 77, 131). Because of this comparatively recent origin, flowering plants have been subject to fewer major mass extinction events that obscure their patterns of speciation and background extinction (131). The remarkable variation in form and habit among extant vascular plant clades has generated a proliferation of Annu. Rev. Plant Biol. 2018.69:685-706. Downloaded from www.annualreviews.org investigations into macroevolutionary patterns. If we examine the time line of land plant evolution Access provided by University of British Columbia on 05/17/18. For personal use only. over the entire Phanerozoic (18), paleontological approaches find that the high species diversity of angiosperms is a result of patterns of speciation outpacing extinction early in their evolutionary history. Although this is similar to patterns exhibited by ferns (18, 127), some characteristics of flowering plants are unique (18). The uneven distribution of species within the angiosperm clade was noticed early in paleobi- ological studies yet given relatively little attention (76), to some degree because species richness values could be biased by differences in the quality of the fossil record between major plant clades (158). Nevertheless, the fossil record consistently produces patterns that show younger clades (i.e., that originated in the Tertiary) contribute disproportionately to angiosperm species diversity due to higher speciation rates and lower extinction rates (76). Phylogenetic perspectives of speciation and extinction surged after the development of sister-group tests by Slowinski & Guyer (133). Since then, our ability to incorporate phylogenetic evidence into diversification analyses has expanded 686 Vamosi et al. PP69CH25_Vamosi ARI 4 April 2018 8:6 rapidly (e.g., see References 107, 116, 120) (Figure 1), yet the methods differ in their assumptions, the range of models they explore, and the way they deal with incomplete and biased sampling (sum- marized in Figure 2; see also the sidebar titled Summary of Trait-Independent Methods to Identify Shifts in Diversification Rates). Understanding the intricacies of these alternatives is critical if we are to understand when and why plants have diversified over evolutionary time (summarized in Future Issues). a d Slope ≈ l 500 200 100 50 20 10 Slope ≈ λ – μ Number of lineages 5 b 2 50 100 150 200 Time (millions of years) e c Annu. Rev. Plant Biol. 2018.69:685-706. Downloaded from www.annualreviews.org Access provided by University of British Columbia on 05/17/18. For personal use only. (Caption appears on following page) www.annualreviews.org • Macroevolutionary Patterns of Flowering Plants 687 PP69CH25_Vamosi ARI 4 April 2018 8:6 Figure 1 (Figure appears on preceding page) Disentangling speciation and extinction rates. The shape of a phylogenetic tree provides information about the relative rates of speciation (λ) and extinction (μ), not just the net diversification rate (r = λ – μ). Example trees were generated using diversitree in R (35), with (a) λ= 0.1 and μ= 0.075 (r = 0.025) in green; (b) λ= 0.1 and μ= 0.08 (r = 0.02) in blue; and (c) λ= 0.02 and μ= 0(r = 0.02) in red. (d ) These differences can be detected in a plot showing lineages over time (88), here shown as the expected number of species (on a log scale) over time for those lineages that survive to the present. Early on, the slope reflects the diversification rate (r), which is the same for blue and red clades [e.g., the number of lineages at time 100 (dashed lines)]. Near the present, recently formed taxa have not yet had time to go extinct and the slope approaches the speciation rate (λ) (116), which is the same for green and blue lineages (note that these clades have many more younger taxa). (e) A simulated tree in which a character affecting the speciation and extinction rates was allowed to change between green, blue, and red states at r = 0.003 (other parameters as above). Most of the youngest species are in the states with a high speciation rate ( green or blue), whereas the overall high number of red lineages may be attributed to this state being ancestral. Hence, this upturn near the present primarily provides information about the effects of a trait on speciation (λ), and the overall growth in size of a lineage provides information about the diversification rate (r). Extinction μ1 Continuation Transition q01 Speciation λ0 Figure 2 Schematic of the processes considered in the Binary State Speciation and Extinction (BiSSE) model for a binary trait that can exist in two states, 0 and 1, denoted in the figure by blue and red, respectively (74). The core logic of BiSSE involves calculating the probability, DNi, of observing all the data (N) that descend from a particular point on the tree, given that the trait is in state i at that point. These data include both the shape of the tree (its topology and the distribution of branch lengths) and the distribution of the character states among the extant species. As we move down a branch toward the root by an amount of time t, each possible type of transition illustrated in the figure could occur and change DN0 with probability: μ + + λ + − μ + λ + 0 t 0 q01 tDN 1 0 t2DN 0 E0 (1 ( 0 0 q01) t)DN 0 . In words, if extinction occurs extinction