Received: 13 June 2018 | Revised: 4 January 2019 | Accepted: 13 January 2019 DOI: 10.1111/jbi.13531 RESEARCH PAPER Understanding the origin of the most isolated endemic reef fish fauna of the Indo‐Pacific: Coral reef fishes of Rapa Nui Erwan Delrieu-Trottin1 | Laura Brosseau-Acquaviva1 | Stefano Mona2,3 | Valentina Neglia1 | Emily C. Giles1 | Cristian Rapu-Edmunds4 | Pablo Saenz-Agudelo1 1 Instituto de Ciencias Ambientales y Evolutivas, Universidad Austral de Chile, Valdivia, Chile Abstract Aim: To understand the origin of the most isolated endemic fish fauna of the Indo‐ 2 Institut de Systématique, Evolution, Biodiversité (ISYEB), UMR 7205 - CNRS, MNHN, UPMC, EPHE, Ecole Pratique des Hautes Etudes, Paris Sorbonne Universités, Paris, France 3 EPHE, PSL Research University, Paris, France 4 Mike Rapu Diving Center, Caleta Hanga Roa O'tai, Isla de Pascua, Chile Pacific, Easter Island (Rapa Nui), and to infer divergence times and colonization routes of the endemic coral reef fish fauna from their closest relatives. Location: Easter Island, Pacific Ocean. Methods: Samples of ten species were used: six small‐range species endemic to Rapa Nui and Motu Motiro Hiva (Salas y Gómez) (i.e. small‐range endemic species) and four large‐range species endemic to the southern subtropical Pacific (i.e. large‐ range endemic species). We present phylogenetic reconstruction results based on Correspondence Erwan Delrieu-Trottin, Instituto de Ciencias Ambientales y Evolutivas, Universidad Austral de Chile, Valdivia, Chile. Email: erwan.delrieu.trottin@gmail.com mitochondrial (1 to 5) and nuclear (1 to 6) loci to place these endemic species in their respective family phylogenies (8). Using these newly calibrated phylogenetic trees, information of species distributions and geological data, we inferred the divergence times from the closest relatives of these ten endemic fishes, compared bio- Funding information Fondo Nacional de Desarrollo Científico y Tecnológico, Grant/Award Number: 11140121, 3160692 geographical history estimation models to reconstruct their ancestral geographic Editor: Cynthia Riginos of the small‐range endemics studied were more recent than the age of Rapa Nui and ranges, colonization routes and underlying mechanisms of speciation. Results: The divergence times (i.e. divergence times from the closest relatives) of all Motu Motiro Hiva; thus, these species can be considered as neoendemics. Biogeographical history estimation models indicated that speciation following a founder‐ event is the most likely scenario. In contrast, the divergence estimates of the large‐ range endemic species were highly variable. This being said, the divergence times of all species were more recent than the age of the oldest islands within their distributions. Main conclusions: Taken together, these results demonstrate that Rapa Nui acts as a cradle of coral reef biodiversity, being responsible for the emergence of small‐ range endemic fish species, but also a route of dispersion for several large‐range endemics and as a stepping stone in the diversification of the Myripristis and Pseudolabrus genera. While no common divergence time was recovered for all of the ten endemic species studied here, the common mechanism of speciation following a founder event was recovered for most of the small‐range endemic species. KEYWORDS coral reef fishes, Easter Island, endemism, neoendemism, phylogeny Journal of Biogeography. 2019;1–11. wileyonlinelibrary.com/journal/jbi © 2019 John Wiley & Sons Ltd | 1 2 | 1 | INTRODUCTION DELRIEU‐TROTTIN ET AL. as ecological conditions evolve through space and time (Anderson, 1994; Kruckeberg & Rabinovitz, 1985; Pigot, Owens, & Orme, One of the most striking aspects of coral reef fishes is their particu- 2012). In practical terms, palaeoendemic species should be distin- larly uneven geographic distributions where species diversity is high- guishable from neo or ecological endemics as these should display est in the Indo‐Australian Archipelago (IAA) and tends to decline deep divergence times with signals of contraction. In contrast, with latitude and longitude (Bellwood & Meyer, 2009; Bellwood & neoendemics and ecological endemics are more challenging to tell Wainwright, 2002; Mora et al., 2003; Reaka, Rodgers, & Kudla, apart (Cowman, Parravicini, Kulbicki, & Floeter, 2017), and are not 2008). Of particular interest are endemism hotspots, which are geo- necessarily mutually exclusive. A species could be young and graphic regions that harbor an unusually high number of species that restricted to a specific geographic area because it arose as the pro- have restricted geographic distributions (Anderson, 1994) and that duct of local adaptation to particular environmental conditions. are found in the periphery of the IAA (Allen, 2008; Bellwood & Overall and despite the complexity of how endemic species origi- Wainwright, 2002; Hughes, Bellwood, & Connolly, 2002). For nate, understanding whether endemic species are the product of instance, despite their relatively low species diversity, the Hawai'ian recent speciation or ancestral extinctions is of great importance to archipelago, Rapa Nui (Easter Island), the Marquesas Islands and the assess their likelihood of extinction (Myers, Mittermeier, Mittermeier, Red Sea have an unusually high proportion of endemic species with da Fonseca, & Kent, 2000). For instance, small‐range size is one of 14%–25% of their fish communities being endemics (Delrieu‐Trottin the characteristics that defines endemics and is also used to deter- et al., 2015; Randall, 2005, 2007; Randall & Cea, 2011). These par- mine if a species is likely to be threatened or to become extinct (Pin- ticular regions constitute centres of endemism in the Indo‐Pacific heiro et al., 2017). In addition, regions that have unusual proportions region and have been an important part in a long‐standing debate of endemic species (centres of endemism) have been proposed as regarding the origin and current patterns of the biodiversity of Indo‐ conservation priorities (Roberts et al., 2002) because they can harbor Pacific coral reefs. From this debate, two major hypotheses have both genetic novelty (neoendemics) and genetic history (paeloen- emerged to explain the current distribution of coral reef biodiversity. demics). Thus, despite the fact that elucidating how a species origi- One proposes the IAA as a Centre‐of‐Origin, a cradle of tropical mar- nates remains challenging, elucidating their time of origin and their ine biodiversity where lineages originate, and the other suggests that demographic history is a first step towards this end. It can also help the IAA is rather a Centre‐of‐Accumulation where distinct faunas the to evaluate their risk of extinction and help to design and implement Pacific and Indian Oceans overlap (Connolly, Bellwood, & Hughes, conservation strategies accordingly (Foote et al., 2007). 2003; Hughes et al., 2002; Mora et al., 2003; Reaka et al., 2008). In general, little information is available regarding the evolution- Understanding the evolutionary history of endemic species from ary processes that have shaped the histories of endemic fish species peripheral centres of endemism can thus contribute to this ongoing and thus the distribution of fish fauna in and among centres of debate. Endemism by definition should be described within a specific endemism. From an evolutionary point of view, these centres have spatial and temporal framework (Anderson, 1994). Spatially, this is graveyards (palaeoendemism) of tropical marine biodiversity (Bowen, done by defining the geographic boundaries or ecological character- Rocha, Toonen, & Karl, 2013). However, the mechanisms by which istics of the area that encompass a specific species distribution. Tem- these reservoirs of endemic fish species have been generated are porally, endemic species have long been classified as either old or not well characterized. Only a handful of studies have analysed new endemics (Engler, 1882) and alternative hypothesis have been divergence times and distribution of coral reef fishes to infer proposed to explain their current endemic status. One of these whether areas were hosting neoendemics or palaeoendemics (e.g. hypotheses suggests that endemic species occupy a restricted geo- Cowman et al., 2017; Hodge, van Herwerden, & Bellwood, 2014; graphic area because they are young and newly established species Pinheiro et al., 2017). These studies have found that endemic spe- (neoendemic species) (Willis, 1922). The alternative hypothesis sug- cies display a wide range of divergence times and so far support gests that endemic species are the remnants of an ancestral wide- both the palaeo and neoendemism hypothesis. However, whether spread species (palaeoendemic species) (Stebbins & Major, 1965). these emerging patterns can be generalized remains an open ques- Both of these hypotheses imply changes in the demographic history tion as divergence times for most endemic species still remain of the species that in turn should leave detectable genetic footprints unknown. been proposed to be either evolutionary cradles (neoendemism) or (signals of expansion or contraction). These hypotheses are thus Rapa Nui (Easter Island, Chile) is among the most remote islands compatible with classical neutral models of speciation (allopatric, on Earth and is the most isolated tropical island of the Pacific Ocean. parapatric and peripatric speciation). However, it has also been sug- Located at 27°07′10 S and 109°21′17 W in the Pacific, Rapa Nui gested that endemic species can arise by ecological speciation via lies 3,700 km from South America and 2,100 km east of the Pitcairn local adaptation (“Gene pool—Ecological niche interaction hypothe- Islands. This small triangular island (166 km2) and the islet Motu sis” sensu Stebbins (1980)). Under this hypothesis, the distribution of Motiro Hiva (Salas y Gómez, 0.15 km2), located 400 km further east, endemic species is restrained by the availability of specific ecological are the easternmost islands of Polynesia. Rapa Nui and Motu Motiro conditions. Endemics arising from ecological speciation could main- Hiva constitute the only two emerged islands of the Easter Chain tain or expand their range size throughout their evolutionary history that also includes dozens of seamounts extending 2,232 km east DELRIEU‐TROTTIN | ET AL. 3 until the Nazca seamount (23°360 S et 83°300 W) (Clouard & Bon- at the origins of new genetically isolated lineages. Alternatively, neville, 2005; Ray et al., 2012). Due to a subtropical gyre, the mean finding divergence times for Rapa Nui endemic species older than current field in the Rapa Nui region is to the east, favouring the trans- the age of this island and no historical biogeographical support for portation of waters from southern Pacific islands to Rapa Nui (Marti- founder‐event speciation would indicate that the island represents nez, Maamaatuaiahutapu, & Taillandier, 2009) and more locally from a refuge of biodiversity that originated elsewhere. In the following Rapa Nui to Motu Motiro Hiva (Andrade, Hormazabal, & Correa sections, we test these predictions to resolve the evolutionary his- Ramirez, 2014). The two islands are relatively young. Vezzoli and Aco- tories of ten endemic species from eight different families. We cella (2009) date Rapa Nui as being 0.8 My while Clouard and Bon- place these species with their presumed sister species within cur- neville (2005) date Rapa Nui as being 2.5 My and Motu Motiro Hiva rently available phylogenies constructed using mitochondrial and as being 1.7 My. The fish fauna of these two islands, which mainly nuclear loci and reconstruct their biogeographical history. We dis- consists of coral reef fishes, is notable for its low diversity and high cuss the potential underlying evolutionary processes that led to proportion of endemic species. Randall and Cea (2011) have identified present‐day patterns. a total of 169 fish species (139 shore fishes), this being an order of magnitude lower than that found in the Hawai'ian Islands (1,250 species) or in Indonesia (3,000) (Allen & Erdmann, 2012). Of the fish species, 21.7% found in Rapa Nui are endemic to this island (Cea, 2016; Randall & Cea, 2011). Overall, little is known regarding the evolution- 2 | MATERIALS AND METHODS 2.1 | Selection of taxa and specimen collection ary history of the island's fish fauna though three of the 30 endemic Randall and Cea (2011) have reported 30 fish species as endemic to species of Rapa Nui are estimated to have diverged from their closest Rapa Nui and 17 fish species as southern subtropical species. The relatives 2.6 > Mya (Cowman et al., 2017). defined Rapa Nui endemics are only present around Rapa Nui and This study aims to shed light on the evolutionary history of the Motu Motiro Hiva and have small ranges (<500 km in linear dis- endemic reef fish fauna of Rapa Nui. In this context, we refer to tance, see Delrieu‐Trottin, Maynard, & Planes, 2014). The southern neoendemic species as those species with divergence times post‐ subtropical species or regional endemics (see Friedlander et al., dating the origin of the island of Rapa Nui (this means that most 2013) have large ranges (1,000–8,000 km in linear distance, see Del- likely they originated on the island during or after its appearance; rieu‐Trottin et al., 2014) and are usually distributed from Southern 2.5 My). We refer to palaeoendemics as those species with diver- Polynesia to Rapa Nui. We referenced all of the presumed sister gence times that pre‐date the formation of Rapa Nui (meaning that species of the 47 species of these two groups using species descrip- most likely they originated elsewhere, colonized the island and tions of endemic species and books of reference (Randall, 2005, went extinct everywhere else). We anticipated that if Rapa Nui 2007; Randall & Cea, 2011). To avoid overestimating divergence constitutes a cradle of coral reef biodiversity, a) divergence times dates, it was essential that the phylogenies contained all the pre- for endemic species should post‐date the formation of Rapa Nui; b) sumed sister species of each endemic species. Of the 47 species we should find historical biogeographical evidence that supports investigated, complete information was only available for nine spe- founder‐event speciation models, that is, rare jump dispersal events cies; four large‐range endemic species: Centropyge hotumatua T A B L E 1 Distribution of the endemic species (Rapa Nui name) under study and of their presumed sister species based on morphology Species & Distribution Sister species Cantherhines rapanui (koreva) RN + MMH Species complex with C. fronticictus (East coast of Africa to the Marshall Is.), C. verecundus (Hawai'ian Is.), C. longicaudus (Cook Is. and Society Is.) (Randall, 2005; Siu et al., 2017) Chromis randalli (mamata) RN + MMH Chromis pamae (Austral Is., and RI, Gambier Is., Pitcairn) (Randall, 2005) Chaetodon litus (tipi tipi ‘uri) RN + MMH C. smithi (RI and Pitcairn) (Randall & Caldwell, 1973) Coris debueni (teteme) RN + MMH Species complex with C. caudimaculata (Indian Ocean), C. dorsomaculata (western Pacific), C. roseoviridis (Cook Is. to Pitcairn), C. venusta (Hawai'ian Is.) (Randall & Cea, 2011) Kuhlia nutabunda (mahore) RN + MMH K. sandvicensis (Pacific Ocean) (Randall & Cea, 2011) Chrysiptera rapanui (mamata) RN + MMH Chrystiptera galba (Austral Is., Gambier Is. and Pitcairn) (Randall, 2005) Centropyge hotumatua (kototi para) RN+ MMH + Pitcairn + RI+ Australs Species complex with C. multicolor (Central Pacific), C. interrupta (Japan and Hawai'ian Is.), C. joculator (Coco & Christmas Is.), C. nahackyi (Johnston atoll), C. debelius (Mauritius) (Pyle, 2003) Cheilodactylus plessisi (ra'ea) RN + MMH + RI C. zebra (Japon), C. francisi (Lord Howe Is. to New Caledonia) & C. vittatus (Hawai'ian Is.) Burridge, 2004) Myripristis tiki (marau) RN + MMI + Pitcairn + RI M. amaena (Japan to Pitcairn) (Greenfield, 1974) Pseudolabrus fuentesis (kotea) RN + MMH+ Pitcairn + RI+ Australs Pseudolabrus gayi (Juan Fernandez Is. and Desventudaras Is.) (Russell & Randall, 1980) RN: Rapa Nui; MMH: Motu Motiro Hiva; RI: Rapa Iti. 4 | DELRIEU‐TROTTIN ET AL. (Pomacanthidae), Cheilodactylus plessisi (Cheilodactylidae) Myripristis Randall and Cea (2011). Upon collection, a small piece of fin tissue tiki (Holocentridae), Pseudolabrus fuentesi (Labridae) and five small‐ was preserved in 96% EtOH. range endemic species: Cantherhines rapanui (Monacanthidae), Chromis randalli, Chrysiptera rapanui (Pomacentridae), Coris debueni (Labridae) and Kuhlia nutabunda (Kuhliidae). We added a tenth species, the 2.2 | Laboratory procedures small‐range endemic Chaetodon litus (Chaetodontidae), albeit no Whole genomic DNA was extracted from fin tissue preserved in genetic data are available for one of its presumed sister species to 96% EtOH at ambient temperature. DNA extraction was performed obtain preliminary results regarding its evolutionary history (Table 1, using the GeneJet Genomic DNA purification kit according to the Figure 1). Sampling was performed using pole‐spears or an anaes- manufacturer's protocols (Thermo Fisher Scientific). For each species, thetic (clove oil) while SCUBA diving around Rapa Nui in October we amplified the same gene fragments using the corresponding pri- 2016. Identification of collected specimens was done according to mers as used in previously generated phylogenies. It is worth noting (a) (b) (c) F I G U R E 1 Geographic distribution of the species under study and divergence times from their closest relatives. (a) Geographic distributions of large‐range endemic species and of (b) small‐range endemic species in this study. The large‐range endemics have larger distributions while the small‐range endemics are only present around Rapa Nui and Motu Motiro Hiva (MMH) (see Table 1 for detailed distributions). The map was generated using the R package ggplot2 (Wickham, 2009) and drawn with rectangular projection (c) Divergence times of species endemic to Rapa Nui from their closest relatives. Median (dot) and 95 % HPD (lines) are represented. Colors denote the distribution. Vertical dotted lines represent the age of the oldest emerged island of the archipelago (Austral Islands: 12 Ma, Gambier Islands: 8 Ma and Pitcairn: 1 Ma Clouard and Bonneville (2005); Vezzoli and Acocella (2009)). An asterix next to the species name indicates that there was strong statistical support for jump dispersal according to the Biogeographical history estimation models DELRIEU‐TROTTIN | ET AL. that we also generated molecular data for three presumed sister spe- 5 & Gibson, 1994) or MAFFT (Katoh & Standley, 2013) and edited cies, Chrysiptera galba (presumed sister species of C. rapanui), Chro- using mis pamae (presumed sister species of C. randalli) and Coris roseovidis (2012)). GENEIOUS 9.0.5 (http://www.geneious.com, Kearse et al. (presumed sister species of C. debueni) to complete the dataset (see Maximum Likelihood (ML) analyses were performed for all of the Table 2 for details). Fragments were amplified using PCR protocols families under study using the online version of IQ‐TREE (Minh, and sequencing as described by Williams, Delrieu‐Trottin, and Planes Nguyen, & von Haeseler, 2013; Nguyen, Schmidt, Von Haeseler, & (2012). Minh, 2015) available at http://iqtree.cibiv.univie.ac.at (Trifinopoulos, Nguyen, von Haeseler, & Minh, 2016). The best model of evolution 2.3 | Phylogeny of each gene was retrieved with ModelFinder (Kalyaanamoorthy, All generated sequences were desposited in GenBank (Accession TREE using the Bayesian Information Criterion (BIC) prior to the numbers: MK100716‐MK100761). Species were placed in the phylo- construction of the ML trees (see Appendix S1 in Supporting Infor- Minh, Wong, Von Haeseler, & Jermiin, 2017) implemented in IQ‐ genies of cheilodactylids (Burridge & Smolenski, 2004), holocentrids mation Table S1). To assess branch support, all IQ‐TREE analyses (Dornburg et al., 2012), kuhliids (Feutry et al., 2013), monacanthids included the ultrafast bootstrap approximation (UFboot) with 1000 (Santini, Sorenson, & Alfaro, 2013), pomacanthids (Gaither et al., replicates (Minh et al., 2013) and the SH‐like approximate likelihood 2014) and pomancentrids (Frédérich, Sorenson, Santini, Slater, & ratio test (SH‐aLRT) also with 1,000 bootstrap replicates (Guindon et Alfaro, 2013) while several phylogenies were used to compose the al., 2010). The time‐calibrated phylogenies (BI) were constructed dataset of chaetodontids (Cowman and Bellwood (2011) for eight of with the software BEAST2 2.4.6 (Bouckaert et al., 2014) using the the nine genes and Gaboriau, Leprieur, Mouillot, and Hubert (2018) fossil calibrations of each phylogeny. We used secondary calibration for COI). For labrids, we used the 12S, 16S, Tmo‐4C4, RAG2, S7 and points from the Actinopterigian super‐phylogeny (Near et al., 2013) COI genes as described in Aiello, Westneat, and Hale (2017) and fol- when studies did not include time‐trees (Table S2). In the same way, lowed the methodology of Wainwright et al. (2018): we included we used tree priors (Yule model for all the analyses except for only a few representatives of all genuses other than that of the tar- pomacentrids (Birth‐Death model)) and partitioning (by individual get species (Table 2). This strategy allowed us to optimize computa- molecular markers for all analysis except for pomacentrids (all genes tion time without compromising the accuracy of divergence date concatenated)) according to the reference phylogenies while the sub- estimates. stitution models were selected according to the ModelFinder results We accessed sequence data stored in GenBank (https://www. using the SSM 1.1.0 package in BEAUti (Bouckaert & Xie, 2017) ncbi.nlm.nih.gov/genbank/) for each coral reef fish family using the R Table S1). We used a relaxed log normal clock model as the molecu- package seqinr (Charif & Lobry, 2007). The endemic species Kuhlia lar clock. The number of generations was set to reach analysis con- nutabunda and Cheilodactylus plessisi were already present in the vergence (Table S2). We assessed the convergence and appropriate phylogenies of their genera, so we did not add sequences for these burn‐in of each analysis using TRACER 1.5 (Drummond & Rambaut, species. Sequences were aligned with Clustal W (Thompson, Higgins, 2007). Three independent analyses were run to ensure convergence. T A B L E 2 Loci used and total number of species included in each phylogeny to place endemic species in their respective phylogenies. In bold are loci that amplified successfully (Accession numbers MK100716‐MK100761). In italics are loci that were already published. Phylogenies of Cowman and Bellwood (2011)1; Gaboriau et al. (2018)2; Burridge and Smolenski (2004)3; Dornburg et al. (2012)4; Feutry et al. (2013)5; Westneat and Alfaro (2005)6; Aiello et al. (2017)7; Santini et al. (2013)8; Gaither et al. (2014)9; Frédérich et al. (2013)10 Family Chaetodon litus 1,2 Cheilodactylus plessisi3 Myripristis tiki 4 Kuhlia nutabunda5 Coris debueni 6,7 Coris roseoviridis6,7 Pseudolabrus fuentesi 6,7 Cantherhines rapanui8 Genes Nb of sp included 12S, 16S, COI, Cytb, ETS2, ND3, Rag2, S7, TMO4C4 122 COI, Cytb 30 COI, ENC1, Glyt, Myh6, Ptr, Rag1, Sreb 42 COI, Rh, 16S, Tmo4c4, IRBP 37 12S, 16S, Rag2, COI, Tmo4c4 105 12S, 16S, Rag2, COI, Tmo4c4 105 12S, 16S, Rag2, COI, Tmo4c4 105 COI, Cytb, Myh6, Rag1, Rh 9 94 Centropyge hotumatua COI, Cytb, Rag2, S7, Tmo4C4 121 Chromis randalli10 COI, Cytb, 12S, 16S, Bmp4, Nd3, Rag1, Rag2 163 Chrysiptera rapanui10 COI, Cytb, 12S, 16S, Bmp4, Nd3, Rag1, Rag2 163 Chrysiptera galba COI, Cytb, 12S, 16S, Bmp4, Nd3, Rag1, Rag2 163 Chromis pamae10 COI, Cytb, 12S, 16S, Bmp4, Nd3, Rag1, Rag2 163 10 6 | DELRIEU‐TROTTIN ET AL. We constructed a maximum clade credibility tree using TREEANNOTA- colonization of the Juan Fernandez Islands by Pseudolabrus fuentesis 1.8.1 (Drummond & Rambaut, 2007) to get median ages and (2 Ma, Clouard & Bonneville, 2005) and of the Marquesas Islands by 95% highest posterior density (HPD) intervals for each node. The Cantherhines, Kuhlia and Myripristis (5.5 Ma, Clouard & Bonneville, 95% HPD represents the smallest interval that contains 95% of the 2005). For each analysis, we set the maximum number of ancestral posterior probability and can be loosely thought of as a Bayesian areas to the maximum number of areas inhabited by a single species. TOR analogue to a confidence interval (Gelman, Carlin, Stern, & Rubin, 2004). 3 | RESULTS 2.4 | Phylogeography 3.1 | Phylogenetic analysis We used the R package BioGeoBEARS (Matzke, 2013) in R (R Core We worked on aligned sequence matrices including 31 to 232 spe- Team, 2017) to estimate the ancestral geographic ranges and to cies per phylogeny and consisting of two to nine genes (932 to investigate the biogeographical history of the endemic species. This 7157 base pairs per matrix) (Table 2).The ML analyses consistently package includes three of the most common biogeographical history returned the same tree topology across all runs with strong boot- estimation models in a likelihood framework, that is, the Dispersal‐ strap support at most nodes (Figures S1–S8). Similarly, the Bayesian Extinction‐Cladogenesis model (DEC; (Ree, Moore, Webb, & Dono- analyses produced the same tree topology across all three runs with ghue, 2005), DIspersal Vicariance Analysis (DIVA; (Ronquist, 1997) high posterior probabilities (PP, Figures S9–S16). For all analyses and the BayArea model (Landis, Matzke, Moore, & Huelsenbeck, except those of Labridae and Pomacanthidae, parameters reached 2013). All of these models allow for anagenetic processes (range effective sample sizes higher than 200 in the Bayesian inferences. expansion and/or range contraction), but differ in the range of clado- For Labridae and Pomacanthidae, most parameters reached effective genetic processes that are permitted (vicariance/sympatry) among/ sample sizes higher than 200 and only few parameters had ESS within‐areas or single areas. As an example, all models allow for a within 100 and 200. sympatry event within one area, but only the DEC model allows for For all eight families under study, we found topologies congruent sympatry event within a subset of the range of a species. In the with those of the published reference phylogenies. For eight of the same manner, only the DEC and the DIVA models allow for a vicari- ten endemic species of this study, the Maximum Likelihood analysis ance event within a subset of the range of a species, and only the and the Bayesian analysis retrieved the same topology. Differences DIVA model allows for a vicariance event resulting in the existence among the ML and BI topologies were observed between Cantherhi- of species with multiple areas; but see the supermodel figure of nes rapanui and Coris debueni and their closest relatives. Matzke for more details (2015). Here, we assessed the fit of each of Large‐range endemic species were found among species‐rich the different models using the corrected Akaike information criterion clades (Cheilodactylus plessisi and Pseudolabrus fuentesi) but also (AICc). The three models were tested including or not a “jump dis- among species poor clades (Centropyge hotumatua and Myripristis tiki) persal” parameter J that permits founder‐event speciation in the spe- while all small‐range endemic species were part of species‐rich cies's history (Paulay & Meyer, 2002; Templeton, 2008). Founder‐ clades. Several of the species presumed to be sister species based event speciation is the result of a long‐distance colonization event of on morphology were confirmed by the molecular analysis (Chromis a small number of individuals founding a population that is then randalli / C. pamae; Chrysiptera rapanui / C. galba; Kuhlia nutabunda / genetically isolated from the ancestral population; these events have K. sandvichensis; Pseudolabrus fuentesi / P. gayi). been shown to be common in island systems (Matzke, 2014). Bio- Our molecular analysis helped to resolve five species complexes GeoBEARS requires a clade including the endemic species and all of that to date are recognized as polytomies. First, our results clearly its closely related species as an input. We used the maximum clade show that Coris debueni is closely related to Coris roseoviridis, a credibility tree from BEAST2 that we obtained for each family and Southern endemic species distributed from the Cook Islands to Pit- extracted the clade of interest using the R package ape 5.0 (Paradis, cairn (Table 1). Second, Cantherhines rapanui is part of a four‐species Claude, & Strimmer, 2004). The geographic ranges of each species complex that was poorly resolved by the ML analysis, yet the BI were coded with one or more areas. showed that C. rapanui is closely related to C. longicaudus (Cook Is. Finally, given that the colonization of an island before its exis- and Society Is.). Third, for Cheilodactylus plessisi, both the BI and ML tence is technically impossible, and before its emergence is unlikely, analyses placed it as the closest relative to C. vittatus from Hawai'i. we constrained the timing of dispersal as 1.7 Ma and 2.5 Ma for Fourth, by adding Centropyge hotumatua to the phylogeny of Motu Motiro Hiva and Rapa Nui respectively (“area allowed” param- Pomacanthidae, we show that the position of Centropyge hotumatua eter in BioGeoBEARS). Thus, this resulted in the removal of those is not within the “multicolour” complex as previously thought. Both areas in the ancestral range estimation of nodes that were older than the ML and BI placed C. hotumatua at the base of the “multicolour” the ages of these islands. We chose to implement the oldest age and “bispinosa” complexes. The BI indicated that C. hotumatua is the estimate for Rapa Nui to try to balance the fact that BioGeoBEARS closest relative of the Hawai'ian endemic C. potteri. Fifth and last, cannot take into account the 95% highest posterior density (HPD) Myripristis tiki was not recovered as the closest relative of M. intervals of each node. We also constrained the timings of amaena. Instead both our BI and ML analyses provided strong DELRIEU‐TROTTIN | ET AL. 7 support for M. tiki as the closest relative of M. leiognathus, a species Australian species and a Juan‐Fernandez endemic species. Cheilo- distributed in the Eastern Tropical Pacific (from Gulf of California to dactylus plessisi, Chaetodon litus and Centropyge hotumatua were Ecuador). recovered as part of an antitropical clade with a close relative from Finally, acknowledging that the current available Chaetodontidae phylogeny does not include Chaetodon smithi, our Hawai'i. Finally, Kuhlia nutabunda was closely related to a species lar- results indicate that Chaetodon litus is the closest relative to the gely distributed in the Pacific while Myripristis tiki likely emerged Hawai'ian endemic C. miliaris in a clade that also includes the West from a founder‐event and was found to be related to an East Tropi- Pacific species C. guentheri. cal Pacific species. 3.2 | Dating divergence time between endemic species and their closest relatives 4 | DISCUSSION We generated time‐trees for the eight families studied here, and these This study represents the first attempt to better understand the ori- include the first‐ever generated time‐trees for Cheilodactylidae, Kuhli- gin of Rapa Nui endemic reef fishes including all their presumed sis- idae and Holocentridae. For chaetodontids, labrids, monacanthids, ter species using newly computed time calibrated phylogenic data pomacanthids and pomancentrids, the divergence times that we and employing biogeographical analysis. Our results show that the obtained between species for the internal nodes of the maximum coral reef fishes endemic to Rapa Nui have diverse evolutionary his- clade credibility trees were similar to those of the available published tories and two trends seem to emerge strongly related to the type phylogenies. Overall, we obtained divergence times between endemic of range distribution: (a) Species strictly endemic to Rapa Nui (small‐ species and their closest relatives that ranged from 0.54 Ma (0.07– range endemic species) have divergence dates with their closest rela- 1.3, 95% HPD, Cantherhines rapanui) to 9.52 Ma (6.77–12.4, 95% tives that are less than the geological age of Motu Motiro Hiva and HPD, Myripristis tiki) (Figure 1, Figures S9–S16). Rapa Nui and thus can be considered as neoendemic species that We observed that all of the divergence estimates for the small‐ likely emerged from a founder‐event. Additionally, these species range endemic species (Cantherhines rapanui, Chaetodon litus, Chromis have sister species with restricted distributions only in the central randalli, Chrysiptera rapanui, Coris debueni, Kuhlia nutabunda) were less south Pacific Islands (Austral Islands, Gambier Islands, Rapa Iti, Pit- than the geological ages of Motu Motiro Hiva (1.7 Ma) and Rapa Nui cairn). (b) Large‐range endemic species have divergence times with (2.5–0.8 Ma). We observed a wider range of divergence times for the their closest relatives that differ drastically (up to an order of magni- large‐range endemic species and species with divergence times older tude), and their sister species are widely distributed among the than Rapa Nui and Motu Motiro Hiva include Cheilodactylus plessisi Hawai'ian Islands, East Pacific and Central Pacific. with 2.80 Ma (0.88–4.00% HPD), Centropyge hotumatua with 6.29 Ma The posterior distribution of divergence times (HPD) between (4.46–8.25 Ma 95% HPD) and Myripristis tiki with 9.52 Ma (6.77– endemic species and their sister species should be compared to the 12.4 Ma 95% HPD). Despite these results, we also found a large‐ geological ages of the islands where endemic species are found. An range endemic species with divergence times that were younger than HPD smaller or containing the values of the geological age of the the geological ages of Motu Motiro‐Hiva and Rapa Nui (Pseudolabrus island where the endemic species is found should support neoen- fuentesi 2.21 Ma (0.88–4.21% HPD)). None of the species under study demism while HPDs encompassing older ages should support had divergence times older than the oldest island that it inhabits (Aus- palaeoendemism. From our phylogenetic reconstruction, we found tral Islands: 12 Ma, Gambier Islands: 8 Ma, and Pitcairn: 1 Ma). that most of the endemic species emerged since the Pliocene (5 Ma), which was a period of high diversification of coral reef fishes in the 3.3 | Phylogeography Indo‐Australian Archipelago (Alfaro, Santini, & Brock, 2007; Cowman & Bellwood, 2011; Hodge, Read, van Herwerden, & Bellwood, 2012; The BioGeoBEARS analyses of the maximum clade credibility trees Hodge et al., 2014; Klanten, van Herwerden, Choat, & Blair, 2004). allowed us to estimate the ancestral range and the evolutionary ori- Geological data suggest that Rapa Nui emerged between 2.5 and gin of the different endemic species (Figure S17–S25). Interestingly, 0.8 M ago while Motu Motiro Hiva is dated at 1.7 Ma (Clouard & the majority (7/9) of the best‐fitting models included a founder‐event Bonneville, 2005; Vezzoli & Acocella, 2009). Based on these dates, (+J) (BayAreaLike+J for Myripristis tiki and Chrysiptera rapanui; DIVA- more than half of the species (6 out of 10) are strictly younger than Like+J for Cantherhines rapanui, Coris debueni and Cheilodactylus ples- Rapa Nui and most (8 out of 10) have HPDs that encompass the age sisi; DEC+J for Chromis randalli and Centropyge hotumatua, Table S3). of Rapa Nui. Additionally, we have found that only Myripristis tiki and Overall, BioGeoBEARS analyses showed that the best‐fitting Centropyge hotumatua, two large‐range endemics, are older (both the model was very often the one including a founder‐event speciation median and HPD bounds) than Rapa Nui. Overall, the estimated diver- process. Four of the small‐range endemic species (Chromis randalli, gence times between the small‐range endemic species and their sister Chrysiptera rapanui, Coris debueni and Cantherhines rapanui) were clo- species taken together with the geological data indicate that most of sely related to species distributed in the South Pacific Islands (Aus- these species are likely neoendemics. trals Islands, Gambier Islands, and Pitcairn) and likely emerged from Founder‐event speciation was expected for neoendemic species a founder‐event. Pseudolabrus fuentesi was closely related to an East while range contraction was expected for palaeoendemic species. 8 | DELRIEU‐TROTTIN ET AL. Using phylogenies to disentangle the biogeographical histories of neoendemism from palaeoendemism in Rapa Nui as contrasting species is a complex task (Heads, 2009), and estimating the internal levels of abundance are found even among neoendemics. nodes of the phylogenies generated here proved difficult. Despite Taken together, our results indicate that Rapa Nui acts as a cra- this, the analyses indicated that most of the models selected (7/9 dle of coral reef biodiversity, being responsible of the emergence of analysis) included a founder‐event, which have been shown to be of small‐range endemic fish species. We have found a common specia- prime importance in other island systems (Matzke, 2014; see Litsios, tion mechanism for the small‐range endemic species, highlighting the Pearman, Lanterbecq, Tolou, & Salamin, 2014; Pinheiro et al., 2017 importance of founder events. Such results are of high importance and Wainwright et al., 2018 for examples). Overall, our biogeograph- from a conservation point of view: Rapa Nui and Motu Motiro‐Hiva ical analyses show that dispersal and founder‐event speciation constitute centres of endemism in the Indo‐Pacific region and are played a major role in the colonization of Rapa Nui. For the small‐ among the most unique marine environments in the Pacific. The cre- range endemics, two speciation schemes can be recognized. First, a ation of a Marine Protected Area in 2018 excluding extractive indus- southern Pacific species diverged from a largely distributed species tries and industrial fishing is an important first step towards the and then colonized and diverged in Rapa Nui via a founder event. conservation of this centre of speciation. This scenario is likely for Coris debueni, Chrysiptera rapanui and Chromis randalli. All three species have a sister species distributed in several islands of the South Pacific (from Australs to Pitcairn), and the ACKNOWLEDGEMENTS emergence of the Rapa Nui archipelago was very likely followed by We thank Rebeca Tepano, Nina, Taveke Olivares Rapu, Liza Garrido colonization and divergence. This scenario is also very likely for Toleado (SERNAPESCA), Ludovic Burns Tuki (Mesa del Mar), and the Chaetodon litus whose presumed sister species, the Southern ende- people of the Rapa Nui Island for their kind and generous support. mic species C. smithi (Randall & Caldwell, 1973), is absent from our This study was funded by FONDECYT Postdoctorado fellowship dataset. Second, a largely distributed species colonized and diverged N°3160692 to E. Delrieu‐Trottin and FONDECYT Iniciación fellow- in Rapa Nui through a founder event, a scenario that is likely for ship N°11140121 to P. Saenz‐Agudelo. The authors declare no con- Kuhlia nutabunda and Cantherhines rapanui. This scenario differs from flict of interest. All applicable institutional guidelines for the care and the first in that the largely distributed species (Cantherines longi- use of animals were followed. Specimens were collected under per- caudus and Kuhlia sandvicensis) colonized the southern islands of the mit No. 724, 8 March 2016 obtained from the Chilean Subsecretary Pacific but not Rapa Nui. of Fishing. The Universidad Austral de Chile Ethical Care Committee Finally, given the young age of Rapa Nui compared to the Austral and Biosecurity Protocol approved our use and handling of animals. Islands (20 Ma; Clouard & Bonneville, 2005), Rapa Nui has likely rep- Finally, we thank C. Riginos, H. Pinheiro and three anonymous resented a new area for the colonization of the large‐range endemic reviewers for providing constructive reviews of earlier versions of species included in this study (Myripristis tiki, Pseudolabrus fuentesi, the manuscript. Centropyge hotumatua, Cheilodactylus plessisi). Rapa Nui could have acted as a stepping stone in the diversification of the Myripristis and Pseudolabrus genera; Myripristis tiki and Pseudolabrus fuentesi are pos- DATA ACCESSIBILITY sible links between the fauna of the Tropical Eastern Pacific, of Juan These sequence data have been submitted to the GenBank data- Fernandez and of the Desventuradas islands. bases under accession numbers MK100716‐MK100761. All the .tree Local abundance and geographical range size are not correlated files from ML and Bayesian analyses are available on FigShare in marine fishes (Hobbs, Jones, & Munday, 2011; Hughes, Bellwood, (https://figshare.com/s/2c45e9f680ddf96632b7; Connolly, Cornell, & Karlson, 2014). Indeed, many marine endemic m9.figshare.7643132). https://10.6084/ reef fish are abundant (Delrieu‐Trottin et al., 2014; Hobbs, Jones, Munday, Connolly, & Srinivasan, 2012) and the local population size of small‐range endemics can be an order of magnitude greater than the population size of widespread species at the same location ORCID Erwan Delrieu-Trottin http://orcid.org/0000-0002-4120-9316 (Hobbs, 2010). Similar to what has been found in Hawai'i (DeMartini & Friedlander, 2004; Kosaki et al., 2017) and the Marquesas islands (Delrieu‐Trottin et al., 2015), the endemic fishes of Rapa Nui constitute a major component of the total assemblage. According to Friedlander et al. (2013), the small‐range endemic species Chromis randalli, Chrysiptera rapanui and Chaetodon litus are among the most abundant species in Rapa Nui, and here we show that all three have HPDs that are younger or encompassing the geological age of Rapa Nui and Motu Motiro‐Hiva. In contrast, we found Kuhlia nutabunda to have a similar HPD, yet it is rather rare on the island. Overall, abundance does not seem to be a good proxy to differentiate REFERENCES Aiello, B. R., Westneat, M. W., & Hale, M. E. (2017). Mechanosensation is evolutionarily tuned to locomotor mechanics. Proceedings of the National Academy of Sciences, 114(17), 4459–4464. https://doi.org/ 10.1073/pnas.1616839114 Alfaro, M. E., Santini, F., & Brock, C. D. (2007). Do reefs drive diversification in marine teleosts? Evidence from the pufferfish and their allies (Order Tetraodontiformes). Evolution, 61(9), 2104–2126. https://doi. org/10.1111/j.1558-5646.2007.00182.x DELRIEU‐TROTTIN ET AL. Allen, G. R. (2008). Conservation hotspots of biodiversity and endemism for Indo‐Pacific coral reef fishes. Aquatic Conservation: Marine and Freshwater Ecosystems, 18(5), 541–556. https://doi.org/10.1002/aqc. 880 Allen, G. R., & Erdmann, M. V. (2012). Reef fishes of the East Indies. Honolulu: University of Hawaii Press. Anderson, S. (1994). Area and endemism. The Quarterly Review of Biology, 69(4), 451–471. https://doi.org/10.1086/418743 Andrade, I., Hormazabal, S., & Correa Ramirez, M. (2014). Time space variability of satellite chlorophylla in the Easter Island Province, south‐eastern Pacific Ocean. Latin American Journal of Aquatic Research, 42(4), 871–887. https://doi.org/10.3856/vol42-issue4-fullte xt -13 Bellwood, D. R., & Meyer, C. P. (2009). Searching for heat in a marine biodiversity hotspot. Journal of Biogeography, 36(4), 569–576. https://doi.org/10.1111/j.1365-2699.2008.02029.x Bellwood, D. R., & Wainwright, P. C. (2002). The history and biogeography of fishes on coral reefs. In Peter F. Sale (ed.), Coral reef fishes (pp. 5–32). San Diego, CA: Elsevier. https://doi.org/10.1016/b978012615185-5/50003-7 Bouckaert, R., Heled, J., Kühnert, D., Vaughan, T., Wu, C. H., Xie, D., & Drummond, A. J. (2014). BEAST 2: A software platform for Bayesian evolutionary analysis. PLoS Computational Biology, 10(4), e1003537. https://doi.org/10.1371/journal.pcbi.1003537 Bouckaert, R., & Xie, D. (2017). Standard Nucleotide Substitution Models v1.0.1. https://doi.org/10.5281/zenodo.995740 Bowen, B. W., Rocha, L. A., Toonen, R. J., & Karl, S. A., the ToBo Laboratory (2013). The origins of tropical marine biodiversity. Trends in Ecology and Evolution, 28(6), 359–366. https://doi.org/10.1016/j.tree. 2013.01.018 Burridge, C. P. (2004). Cheilodactylus (Goniistius) francisi, a new species of morwong (Perciformes: Cirrhitoidea) from the Southwest Pacific. Records of the Australian Museum, 56(2), 231–234. https://doi.org/10. 3853/j.0067-1975.56.2004.1425 Burridge, C. P., & Smolenski, A. J. (2004). Molecular phylogeny of the Cheilodactylidae and Latridae (Perciformes: Cirrhitoidea) with notes on taxonomy cand biogeography. Molecular Phylogenetics and Evolution, 30(1), 118–127. https://doi.org/10.1016/S1055 -7903(03) 00157-X Cea, A. (2016). Ika Rapa Nui. Rapa Nui Press. Charif, D., & Lobry, J. R. (2007). SeqinR 1.0-2: A contributed package to the R project for statistical computing devoted to biological sequences retrieval and analysis. In U. Bastolla, M. Porto, H. E. Roman & M. Vendruscolo (Eds.), Structural approaches to sequence evolution (pp. 207–232). New York: Springer Verlag. https://doi.org/ 10.1007/978-3-540-35306-5 Clouard, V., & Bonneville, A. (2005). Ages of seamounts, islands, and plateaus on the Pacific plate. In Gillian R. Foulge (Ed.), Plates, plumes and paradigms (pp. 71–90). Boulder, CO: Geological Society of America. doi: 10 .1130/0-8137-2388-4.71 Connolly, S. R., Bellwood, D. R., & Hughes, T. P. (2003). Indo‐Pacific biodiversity of coral reefs: Deviations from a mid‐domain model. Ecology, 84(8), 2178–2190. https://doi.org/10.1890/02-0254 Cowman, P. F., & Bellwood, D. R. (2011). Coral reefs as drivers of clado‐ genesis: Expanding coral reefs, cryptic extinction events, and the development of biodiversity hotspots. Journal of Evolutionary Biology, 24(12), 2543–2562. https://doi.org/10.1111/j.1420-9101.2011. 02391.x Cowman, P. F., Parravicini, V., Kulbicki, M., & Floeter, S. R. (2017). The biogeography of tropical reef fishes: Endemism and provinciality through time. Biological Reviews, 92(4), 2112–2130. https://doi.org/ 10.1111/brv.12323 Delrieu-Trottin, E., Maynard, J., & Planes, S. (2014). Endemic and widespread coral reef fishes have similar mitochondrial genetic diversity. | 9 Proceedings of the Royal Society London B, 281, 20141068. https://d oi.org/10.1098/rspb.2014.1068 Delrieu-Trottin, E., Williams, J. T., Bacchet, P., Kulbicki, M., Mourier, J., Galzin, R., … Planes, S. (2015). Shore fishes of the Marquesas Islands, an updated checklist with new records and new percentage of endemic species. Check List, 11(5). https://doi.org/10.15560/11.5.1758 DeMartini, E., & Friedlander, A. (2004). Spatial patterns of endemism in shallow‐water reef fish populations of the Northwestern Hawaiian Islands. Marine Ecology Progress Series, 271, 281–296. https://doi.org/ 10.3354/meps271281 Dornburg, A., Moore, J. A., Webster, R., Warren, D. L., Brandley, M. C., Iglesias, T. L., … Near, T. J. (2012). Molecular phylogenetics of squirrelfishes and soldierfishes (Teleostei: Beryciformes: Holocentridae): Reconciling more than 100 years of taxonomic confusion. Molecular Phylogenetics and Evolution, 65(2), 727–738. https://doi.org/10.1016/ j.ympev.2012.07.020 Drummond, A. J., & Rambaut, A. (2007). BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology, 7(1), 214. https://doi.org/10.1186/1471-2148-7 -214 Engler, A. 1882. VersucheinerEntwick lungsgeschichte der Pflanzenwelt. Leipzig: Engelmann. Feutry, P., Castelin, M., Ovenden, J. R., Dettaï, A., Robinet, T., Cruaud, C., & Keith, P. (2013). Evolution of Diadromy in Fish: Insights from a Tropical Genus (Kuhlia Species). The American Naturalist, 181(1), 52– 63. https://doi.org/10.1086/668593 Foote, M., Crampton, J. S., Beu, A. G., Marshall, B. A., Roger, A., Maxwell, P. A., & Matcham, I. (2007). Rise and fall of species occupancy in Cenozoic fossil mollusks. Science, 318, 1131–1134. https://doi.org/ 10.1126/science.1146303 Frédérich, B., Sorenson, L., Santini, F., Slater, G. J., & Alfaro, M. E. (2013). Iterative Ecological radiation and convergence during the evolutionary history of Damselfishes (Pomacentridae). The American Naturalist, 181(1), 94–113. https://doi.org/10.1086/668599 Friedlander, A. M., Ballesteros, E., Beets, J., Berkenpas, E., Gaymer, C. F., Gorny, M., & Sala, E. (2013). Effects of isolation and fishing on the marine ecosystems of Easter Island and Salas y Gómez, Chile. Aquatic Conservation: Marine and Freshwater Ecosystems, 23(4), 515–531. https://doi.org/10.1002/aqc.2333 Gaboriau, T., Leprieur, F., Mouillot, D., & Hubert, N. (2018). Influence of the geography of speciation on current patterns of coral reef fish biodiversity across the Indo‐Pacific. Ecography, 41, 1295–1306. https://d oi.org/10.1111/ecog.02589 Gaither, M. R., Schultz, J. K., Bellwood, D. R., Pyle, R. L., DiBattista, J. D., Rocha, L. A., & Bowen, B. W. (2014). Evolution of pygmy angelfishes: Recent divergences, introgression, and the usefulness of color in taxonomy. Molecular Phylogenetics and Evolution, 74(1), 38–47. https://d oi.org/10.1016/j.ympev.2014.01.017 Gelman, A., Stern, H. S., Carlin, J. B., Dunson, D. B., Vehtari, A., & Rubin, D. B. (2013). Bayesian data analysis. London, UK: Chapman and Hall/ CRC. Greenfield, D. W. (1974). Revision of the squirrelfish genus Myripristis Cuvier (Pisces: Holocentridae). Science Bulletin Natural History Museum Los Angeles County, 19, 1–54. Guindon, S., Dufayard, J.-F., Lefort, V., Anisimova, M., Hordijk, W., & Gascuel, O. (2010). New algorithms and methods to estimate maximum‐likelihood phylogenies: Assessing the performance of phyml 3.0. Systematic Biology, 59(3), 307–321. https://doi.org/10.1093/sysbio/ syq010 Heads, M. (2009). Inferring biogeographic history from molecular phylogenies. Biological Journal of the Linnean Society, 98(4), 757–774. https://doi.org/10.1111/j.1095-8312.2009.01320.x Hobbs, J.-P. A. (2010). Poleward range expansion of a tropical coral reef fish (Centropyge flavissima) to Lord Howe Island, Australia. Marine Biodiversity Records, 3, 1–2. https://doi.org/10.1017/S1755267210000990 10 | Hobbs, J. P. A., Jones, G. P., & Munday, P. L. (2011). Extinction risk in endemic marine fishes. Conservation Biology, 25(5), 1053–1055. https://doi.org/10.1111/j.1523-1739.2011.01698.x Hobbs, J. P. A., Jones, G. P., Munday, P. L., Connolly, S. R., & Srinivasan, M. (2012). Biogeography and the structure of coral reef fish communities on isolated islands. Journal of Biogeography, 39(1), 130–139. https://doi.org/10.1111/j.1365-2699.2011.02576.x Hodge, J. R., Read, C. I., van Herwerden, L., & Bellwood, D. R. (2012). The role of peripheral endemism in species diversification: Evidence from the coral reef fish genus Anampses (Family: Labridae). Molecular Phylogenetics and Evolution, 62(2), 653–663. https://doi.org/10.1016/ j.ympev.2011.11.007 Hodge, J. R., van Herwerden, L., & Bellwood, D. R. (2014). Temporal evolution of coral reef fishes: Global patterns and disparity in isolated locations. Journal of Biogeography, 41(11), 2115–2127. https://doi. org/10.1111/jbi.12356 Hughes, T. P., Bellwood, D. R., & Connolly, S. R. (2002). Biodiversity hotspots, centres of endemicity, and the conservation of coral reefs. Ecology Letters, 5(6), 775–784. https://doi.org/10.1046/j.1461-0248. 2002.00383.x Hughes, T. P., Bellwood, D. R., Connolly, S. R., Cornell, H. V., & Karlson, R. H. (2014). Double jeopardy and global extinction risk in corals and reef fishes. Current Biology, 24, 2946–2951. https://doi.org/10.1016/ j.cub.2014.10.037 Kalyaanamoorthy, S., Minh, B. Q., Wong, T. K., Von Haeseler, A., & Jermiin, L. S. (2017). ModelFinder: Fast model selection for accurate phylogenetic estimates. Nature Methods, 14(6), 587–589. https://doi. org/10.1038/nmeth.4285 Katoh, K., & Standley, D. M. (2013). MAFFT multiple sequence alignment software version 7: Improvements in performance and usability. Molecular Biology and Evolution, 30(4), 772–780. https://doi.org/10. 1093/molbev/mst010 Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., & Drummond, A. (2012). Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics, 28(12), 1647–1649. https://doi. org/10.1093/bioinformatics/ bts199 Klanten, S. O., van Herwerden, L., Choat, J. H., & Blair, D. (2004). Patterns of lineage diversification in the genus Naso (Acanthuridae). Molecular Phylogenetics and Evolution, 32(1), 221–235. https://doi. org/10.1016/j.ympev.2003.11.008 Kosaki, R. K., Pyle, R. L., Leonard, J. C., Hauk, B. B., Whitton, R. K., & Wagner, D. (2017). 100% endemism in mesophotic reef fish assemblages at Kure Atoll, Hawaiian Islands. Marine Biodiversity, 47(3), 783– 784. https://doi.org/10.1007/s12526-016-0510-5 Kruckeberg, A. R., & Rabinovitz, D. (1985). Biological aspects of endemism in higher plants. Annual Review of Ecology and Systematics, 16, 447–479. https://doi.org/10.1146/annurev.es.16.110185. 002311 Landis, M. J., Matzke, N. J., Moore, B. R., & Huelsenbeck, J. P. (2013). Bayesian analysis of biogeography when the number of areas is large. Systematic Biology, 62(6), 789–804. https://doi.org/10.1093/sysbio/ syt040 Litsios, G., Pearman, P. B., Lanterbecq, D., Tolou, N., & Salamin, N. (2014). The radiation of the clownfishes has two geographical replicates. Journal of Biogeography, 41(11), 2140–2149. https://doi.org/10. 1111/jbi.12370 Martinez, E., Maamaatuaiahutapu, K., & Taillandier, V. (2009). Floating marine debris surface drift: Convergence and accumulation toward the South Pacific subtropical gyre. Marine Pollution Bulletin, 58(9), 1347–1355. https://doi.org/10.1016/j.marpolbul.2009.04.022 Matzke, N. J. (2013). Probabilistic historical biogeography: New models for founder‐event speciation, imperfect detection, and fossils allow improved accuracy and model‐testing. Frontiers of Biogeography, 5(4), 242–248. DELRIEU‐TROTTIN ET AL. Matzke, N. J. (2014). Model selection in historical biogeography reveals that founder‐event speciation is a crucial process in island clades. Systematic Biology, 63(6), 951–970. https://doi.org/10.1093/sysbio/ syu056 Matzke, N.J. (2015). Stochastic mapping under biogeographical models. Available at: http://phylo.wikidot.com/biogeobears#stochastic_map ping (accessed 11 September 2018). Minh, B. Q., Nguyen, M. A. T., & von Haeseler, A. (2013). Ultrafast approximation for phylogenetic bootstrap. Molecular Biology and Evolution, 30(5), 1188–1195. https://doi.org/10.1093/molbev/mst024 Mora, C., Chittaro, P. M., Sale, P. F., Kritzer, J. P., Ludsin, S. A., & Africa, S. (2003). Patterns and processes in reef fish diversity. Nature, 421, 933–936. https://doi.org/10.1038/nature01421.1 Myers, N., Mittermeier, R. A., Mittermeier, C. G., da Fonseca, G. A. B., & Kent, J. (2000). Biodiversity hotspots for conservation priorities. Nature, 403, 853–858. https://doi.org/10.1038/35002501 Near, T. J., Dornburg, A., Eytan, R. I., Keck, B. P., Smith, W. L., Kuhn, K. L., … Wainwright, P. C. (2013). Phylogeny and tempo of diversification in the superradiation of spiny‐rayed fishes. Proceedings of the National Academy of Sciences, 110(31), 12738–12743. https://doi. org/10.1073/pnas.1304661110 Nguyen, L. T., Schmidt, H. A., Von Haeseler, A., & Minh, B. Q. (2015). IQ‐ TREE: A fast and effective stochastic algorithm for estimating maximum‐likelihood phylogenies. Molecular Biology and Evolution, 32(1), 268–274. https://doi.org/10.1093/molbev/msu300 Paradis, E., Claude, J., & Strimmer, K. (2004). APE: Analyses of phylogenetics and evolution in R language. Bioinformatics, 20(2), 289–290. https://doi.org/10.1093/bioinformatics/btg412 Paulay, G., & Meyer, C. (2002). Diversification in the tropical pacific: Comparisons between marine and terrestrial systems and the importance of founder speciation. Integrative and Comparative Biology, 42 (5), 922–934. https://doi.org/10.1093/icb/42.5.922 Pigot, A. L., Owens, I. P. F., & Orme, C. D. L. (2012). Speciation and extinction drive the appearance of directional range size evolution in phylogenies and the fossil record. PLoS Biology, 10, e1001260. https://doi.org/10.1371/journal.pbio.1001260 Pinheiro, H. T., Bernardi, G., Simon, T., Joyeux, J.-C., Macieira, R. M., Gasparini, J. L., … Rocha, L. A. (2017). Island biogeography of marine organisms. Nature, 549, 82. https://doi.org/10.1038/nature23680 Pyle, R. L. (2003). A systematic treatment of the reef-fish family Pomacanthidae (Pisces: Perciformes). University of Hawaii at Manoa. R Core Team. (2017). R: A language and environment for statistical computing. doi: http:// www.R-project.org/ Randall, J. E. (2005). Reef and shore fishes of the South Pacific: New Caledonia to Tahiti and the Pitcairn (Vol. 1). Honolulu: University of Hawaii Press. Randall, J. E. (2007). Reef fishes of Hawaii. Honolulu: University of Hawaii Press. Randall, J. E., & Caldwell, D. K. (1973). A new butterflyfish of the genus Chaetodon and a new angelfish of the genus Centropyge from Easter Island. Los Angeles County: Natural History Museum. Randall, J. E., & Cea, A. (2011). Shore fishes of Easter Island. Honolulu, HI: University of Hawai'i Press. Ray, J. S., Mahoney, J. J., Duncan, R. A., Ray, J., Wessel, P., & Naar, D. F. (2012). Chronology and geochemistry of lavas from the Nazca Ridge and Easter Seamount Chain: An ˜30 myr hotspot record. Journal of Petrology, 53(7), 1417–1448. https://doi.org/10.1093/petrology/egs021 Reaka, M. L., Rodgers, P. J., & Kudla, A. U. (2008). Patterns of biodiversity and endemism on Indo‐West Pacific coral reefs. Proceedings of the National Academy of Sciences of the United States of America, 105 (Suppl), 11474–81. https://doi.org/10.1073/pnas.0802594105 Ree, R. H., Moore, B. R., Webb, C. O., & Donoghue, M. J. (2005). A likelihood framework for inferring the evolution of geographic range on phylogenetic trees. Evolution, 59(11), 2299. https://doi.org/10.1554/ 05-172.1 DELRIEU‐TROTTIN | ET AL. Roberts, C. M., McClean, C. J., Veron, J. E. N., Hawkins, J. P., Allen, G. R., McAllister, D. E., … Werner, T. B. (2002). Marine biodiversity hotspots and conservation priorities for tropical reefs. Science, 295, 1280–1284. https://doi.org/10.1126/science.1067728 Ronquist, F. (1997). Dispersal‐vicariance analysis: A new approach to the quantification of historical biogeography. Systematic Biology, 46(1), 195–203. https://doi.org/10.1093/sysbio/46.1.195 Russell, B. C., & Randall, J. E. (1980). The Labrid Fish Genus Pseudolabrus from Islands of the Southeastern Pacific, with description of a new species from Rapa. Pacific Science, 34(4), 433–440. Santini, F., Sorenson, L., & Alfaro, M. E. (2013). A new multi‐locus timescale reveals the evolutionary basis of diversity patterns in triggerfishes and filefishes (Balistidae, Monacanthidae; Tetraodontiformes). Molecular Phylogenetics and Evolution, 69(1), 165–176. https://doi. org/10.1016/j.ympev.2013.05.015 Siu, G., Bacchet, P., Bernardi, G., Brooks, A., Carlot, J., Claudet, J., … Galzin, R. (2017). Shore fishes of French polynesia. Cybium, 41(3), 1–34. Stebbins, G. L. (1980). Rarity of plant species: A synthetic viewpoint. Rhodora, 82(829), 77–86. Stebbins, G. L., & Major, J. (1965). Endemism and speciation in the California flora. Ecological Monographs, 35(1), 1–35. https://doi.org/10. 2307/1942216 Templeton, A. R. (2008). The reality and importance of founder speciation in evolution. BioEssays, 30(5), 470–479. https://doi.org/10.1002/ (ISSN)1521-1878 Thompson, J. D., Higgins, D. G., & Gibson, T. J. (1994). CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position‐specific gap penalties and weight matrix choice. Nucleic Acids Research, 22(22), 4673–4680. https://doi.org/10.1093/nar/22.22.4673 Trifinopoulos, J., Nguyen, L.-T., von Haeseler, A., & Minh, B. Q. (2016). W‐IQ‐TREE: A fast online phylogenetic tool for maximum likelihood analysis. Nucleic Acids Research, 44, W232–W235. https://doi.org/10. 1093/nar/gkw256 Vezzoli, L., & Acocella, V. (2009). Easter Island, SE Pacific: An end‐member type of hotspot volcanism at Easter Island. GSA Bulletin, 121(5– 6), 869–886. https://doi.org/10.1130/B26470.1 Wainwright, P. C., Santini, F., Bellwood, D. R., Ross Robertson, D., Rocha, L. A., & Alfaro, M. E. (2018). Phylogenetics and geography of speciation in New World Halichoeres wrasses. Molecular Phylogenetics and Evolution, 121, 35–45. https://doi.org/10.1016/j.ympev.2017.12.028 Westneat, M. W., & Alfaro, M. E. (2005). Phylogenetic relationships and evolutionary history of the reef fish family Labridae. Molecular Phylogenetics and Evolution, 36(2), 370–390. https://doi.org/10.1016/j.ym pev.2005.02.001 Wickham, H. (2009). ggplot2: Elegant graphics for data analysis. New York: Springer-Verlag. https://doi.org/10.1007/978-0-387-98141-3 11 Williams, J. T., Delrieu-Trottin, E., & Planes, S. (2012). A new species of Indo‐Pacific fish, Canthigaster criobe, with comments on other Canthigaster (Tetraodontiformes: Tetraodontidae) at the Gambier Archipelago. Zootaxa, 3523(3523), 80–88. Willis, J. C. (1922). Age and area—A study in geographical distribution and origin of species. Cambridge, UK: Cambridge University Press. https://doi.org/10.5962/bhl.title.70451 BIOSKETCHES Erwan Delrieu‐Trottin is broadly interested in molecular ecology and the evolution of fishes. Stefano Mona and Pablo Saenz‐Agudelo are interested broadly in population genetics. Laura Brosseau‐Acquaviva and Cristian Rapu Edmunds are interested in conservation. Valentina Neglia and Emily Giles are interested in the ecology and molecular ecology of marine organisms. Author contributions: E.D.T. and P.S.A. conceived the project; E.D.T., V.N., C.R.E., E.C.G. and P.S.A. designed and conducted sampling; E.D.T. and L.B.A. performed molecular experiments; E.D.T. and L.B.A. analysed and interpreted the data; E.D.T. wrote the first draft and all co‐authors contributed. SUPPORTING INFORMATION Additional supporting information may be found online in the Supporting Information section at the end of the article. How to cite this article: Delrieu-Trottin E, BrosseauAcquaviva L, Mona S, et al. Understanding the origin of the most isolated endemic reef fish fauna of the Indo-Pacific: Coral reef fishes of Rapa Nui. J Biogeogr. 2019;00:1–11. https://doi.org/10.1111/jbi.13531