Population Genetic Structure of the Tropical Two-Wing Flyingfish (Exocoetus volitans)

Eric A. Lewallen; Andrew J. Bohonak; Carolina A. Bonin; Andre J. van Wijnen; Robert L. Pitman; Nathan R. Lovejoy

doi:10.1371/journal.pone.0163198

Abstract

Delineating populations of pantropical marine fish is a difficult process, due to widespread geographic ranges and complex life history traits in most species. Exocoetus volitans, a species of two-winged flyingfish, is a good model for understanding large-scale patterns of epipelagic fish population structure because it has a circumtropical geographic range and completes its entire life cycle in the epipelagic zone. Buoyant pelagic eggs should dictate high local dispersal capacity in this species, although a brief larval phase, small body size, and short lifespan may limit the dispersal of individuals over large spatial scales. Based on these biological features, we hypothesized that E. volitans would exhibit statistically and biologically significant population structure defined by recognized oceanographic barriers. We tested this hypothesis by analyzing cytochrome b mtDNA sequence data (1106 bps) from specimens collected in the Pacific, Atlantic and Indian oceans (n = 266). AMOVA, Bayesian, and coalescent analytical approaches were used to assess and interpret population-level genetic variability. A parsimony-based haplotype network did not reveal population subdivision among ocean basins, but AMOVA revealed limited, statistically significant population structure between the Pacific and Atlantic Oceans (Φ_ST = 0.035, p<0.001). A spatially-unbiased Bayesian approach identified two circumtropical population clusters north and south of the Equator (Φ_ST = 0.026, p<0.001), a previously unknown dispersal barrier for an epipelagic fish. Bayesian demographic modeling suggested the effective population size of this species increased by at least an order of magnitude ~150,000 years ago, to more than 1 billion individuals currently. Thus, high levels of genetic similarity observed in E. volitans can be explained by high rates of gene flow, a dramatic and recent population expansion, as well as extensive and consistent dispersal throughout the geographic range of the species.

Citation: Lewallen EA, Bohonak AJ, Bonin CA, van Wijnen AJ, Pitman RL, Lovejoy NR (2016) Population Genetic Structure of the Tropical Two-Wing Flyingfish (Exocoetus volitans). PLoS ONE 11(10): e0163198. https://doi.org/10.1371/journal.pone.0163198

Editor: Giacomo Bernardi, University of California Santa Cruz, UNITED STATES

Received: March 29, 2016; Accepted: September 6, 2016; Published: October 13, 2016

This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.

Data Availability: DNA sequences used in this study are available from the Genbank database (Accession numbers HQ325634, HQ325635, KX912952, KX913215).

Funding: Funding for the study was provided by The Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant to NRL. We also acknowledge funding from the National Institute of Arthritis and Musculoskeletal and Skin Diseases (T32 AR56950 to EAL; R01 AR049069 to AJVW). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Understanding the genetic population structure of marine fishes is critical not only for documenting genetic diversity, speciation, and evolution, but also to inform global fisheries management efforts. Recent and rapid losses of global marine biodiversity make the genetic characterizations of marine organisms, and populations, a high priority [1]. In general, coastal and reef associated fish species are best studied [2, 3], probably because of their proximity to human populations and corresponding ease of sampling. Less is known about the population-level genetic diversity of organisms inhabiting the epipelagic zone (pelagic surface waters), one of the largest habitats on Earth (>130 million km²; [4]). This is in part due to the difficulty of sampling such vast geographic areas, and frequent challenges with species delimitation [5].

Because epipelagic habitats (surface waters of the open ocean) are among the most vulnerable to increasing sea surface temperatures and global climate change, it is imperative that we understand the population-level genetic diversity of open ocean inhabitants. Spatial patterns of genetic variation have been reviewed for populations of epipelagic and circumtropical fishes [5]. A number of traits are shared by many circumtropical species, including broadcast spawning, brevity of larval duration, and large adult range. Key questions that remain to be addressed are how circumtropical species achieve global geographic ranges and maintain global population connectivity. Population genetic investigation of the Tropical Two-Wing Flyingfish, Exocoetus volitans, (family Exocoetidae), may be particularly useful for resolving these questions.

Like other members of the Exocoetidae, E. volitans can leap from the water and glide through the air using enlarged pectoral fins, allowing evasion of epipelagic predators. Despite an exceptional predator-evasion strategy, flyingfishes are important prey for larger fish, sharks, squids, seabirds and marine mammals. As an epipelagic specialist, E. volitans completes all life stages in the upper stratum of the water column, and can be abundant in warm surface waters [6]. The patchy abundance of E. volitans at smaller spatial scales is likely due to extrinsic factors, such as primary productivity [7].

Exocoetus volitans spawns buoyant, pelagic eggs that should contribute to increased dispersal potential [8] and high gene flow. However, other life history traits such as short larval phase, coupled with a brief hatching time of approximately 1–2 weeks [9], and one-year life span [7] should limit population connectivity, especially on global scales. Further, E. volitans is small bodied (<210 mm; [6]), a relatively slow swimmer, and incapable of swimming great distances [10]. These characteristics suggest that E. volitans could be subdivided into discrete populations throughout its geographic range.

In this study, we tested for population subdivision using mtDNA sequences of E. volitans from throughout its entire circumtropical range, including far offshore epipelagic localities that are only accessible for research via dedicated long-range expeditions. Specifically, we addressed the following questions: (i) What is the worldwide genetic population structure of E. volitans? (ii) Are populations of E. volitans separated by previously described marine fish barriers? (iii) Is there evidence of a recent population expansion, or has the effective population size remained constant for a long period (i.e., drift-mutation equilibrium)? Using parsimony, AMOVA, Bayesian, and coalescent modelling methods, we measured and interpreted multiple population genetic parameters for this representative circumtropical epipelagic specialist

Materials and Methods

Specimen collection

Exocoetus volitans specimens were collected at night using flood lights and long-handled dip nets during 13 research cruises on the following research vessels from 1992 to 2010: Endeavor, Shoyo Maru, Gordon Gunter, McArthur II, David Starr Jordan, Kahana, and Oscar Elton Sette. The majority of specimens in the Pacific Ocean were collected during Stenella Abundance Research cruises, while Atlantic specimens were collected during the South Atlantic Black Carbon research cruise. Specimens were also donated by collaborators. In total, 266 E. volitans specimens were collected from 97 locations throughout the Atlantic, Pacific, and Indian Oceans (Table 1, Fig 1, S1 Table). All specimen collections were performed in accordance with all ethical care and animal welfare standards. No specific permissions or field permits were required for any of the localities where specimens were collected because these areas and species are not protected. In particular, these field studies did not involve endangered or protected species. All specimens were euthanized using a seawater ice bath to minimize undue pain and stress to the animals. Tissues for DNA analysis were removed post-mortem and preserved in 95% ethanol. Also post-mortem, whole-specimen vouchers were frozen in seawater, fixed in 10% formalin, and cataloged at the Royal Ontario Museum, Scripps Institution of Oceanography, and Los Angeles County Museum. This study was not conducted on private land, in a national park, or protected area of the land or sea.

Download:

Fig 1. Collection localities for Exocoetus volitans specimens used in this study.

In total, 266 individuals were collected from 97 localities (red circles). Grey polygons approximate the species' distribution, based on Parin and Shakhovskoy [6]. Previously-described marine barriers tested are indicated by blue bars (EPB = Eastern Pacific Barrier; IPB = Isthmus of Panama Barrier; BB = Benguela Barrier; SSB = Sunda Shelf Barrier). This figure was generated using ArcGIS, version 9.3.1 (ESRI, Redlands, CA, USA).

https://doi.org/10.1371/journal.pone.0163198.g001

Download:

Table 1. Summary statistics from population genetic analyses of Exocoetus volitans divided by ocean basin, and population clusters (northern and southern), and overall.

S = number of variable sites; h = haplotype diversity; π = nucleotide diversity.

https://doi.org/10.1371/journal.pone.0163198.t001

Genetic data collection

Genomic DNA was extracted using DNeasy kits (Qiagen, Hilden, Germany) and a portion of the mitochondrial cytochrome b (cytb) gene (1106 bps) was amplified using previously published protocols [11]. Cytb is a commonly used molecular marker of genetic diversity, facilitating comparison with other studies. Cytb has also been used previously to resolve the phylogenetic relationships among flyingfish species [11]. Sequences were edited and aligned using Sequencher v.4.6 (Gene Codes Corporation, Ann Arbor, MI), and have been deposited in Genbank (Accession numbers HQ325634, HQ325635, KX912952, KX913215).

Summary statistics and haplotype network

Population genetic summary statistics (number of haplotypes; haplotype diversity, h; nucleotide diversity, π; Θ_k (per sequence) and Θ_s (per site) were calculated using DNAsp v.5 [12]. To determine the relationships of mtDNA haplotypes, a cytb gene genealogy was generated using the statistical parsimony method implemented by TCS v1.21 [13]. We resolved ambiguous mutational relationships in the network using the criteria of Crandall, Templeton & Sing [14] and Templeton, Routman & Phillips [15].

Analyses of molecular variance

Because biogeographic barriers for epipelagic fishes are not well characterized, we postulated that barriers which have been demonstrated in other broadly distributed marine fishes could affect E. volitans. Globally, seven major biogeographic features limit gene flow among marine fishes, and have driven allopatric speciation events in numerous marine taxa [3]. Four of these barriers were relevant to our sampling scheme and the geographic range of E. volitans: the Isthmus of Panama Barrier (physically separates the Atlantic and Pacific Oceans, Central America), Sunda Shelf Barrier (numerous islands that create a division between the Indian and Pacific Oceans, southeast Asia), Benguela Barrier (a region of converging water currents between the Atlantic and Indian Oceans, South Africa), and the Eastern Pacific Barrier: a deep water division between the Western and Eastern Pacific (Fig 1). The Eastern Pacific and Benguela barriers are oceanic, can quickly change size and location depending on the year or season, and are therefore expected to be permeable for E. volitans. This is not the case for terrestrial barriers, such as the Isthmus of Panama, and the semi-terrestrial Sunda Shelf barrier. Despite the dynamic nature of oceanic barriers, we opted to include an assessment of all four potential barriers to flyingfish dispersal. To quantify genetic divergence among populations, defined a priori by collection locality, we conducted Analysis of Molecular Variance (AMOVA) within Arlequin v.3.5 [16] separately for each of the four barriers listed above (AMOVA analysis parameters: deletions, transitions, and transversions = 1; allowed missing data = 0.05; molecular distance = pairwise difference; distance matrix = minimum spanning network (inter-haplotypic); permutations = 10,000). All samples adjacent to these barriers were used in AMOVA comparisons.

Bayesian clustering analysis

MOVA requires that individuals are grouped a priori into testable gene pools. To delineate populations more objectively, we used the Bayesian clustering model implemented by Geneland v.1 [17, 18]. To reduce the potential impact caused by multiple individuals collected at the same locality, we set the spatial uncertainty of collection sites to 0.1 Universal Transverse Mercator. Ten million MCMC generations were run, with a thinning of 1,000, and the first 5,000 (50%) saved trees were discarded as burn-in. The conservative method for agglomerative "bottom up" clustering was specified by setting the maximum allowable gene pool number to match the total number of samples collected (n = 266). AMOVA was used to quantify genetic differentiation among the clusters identified by Geneland, (using the methods described above). Additionally, to gain a better understanding of the relative position of an equatorial barrier and balance sample sizes, we divided individuals into collection localities north and south of three latitudinal lines (0°N, 5°N, and 7°N) and conducted AMOVAs independently on each data partition.

Historic population growth

The cytb genealogy was evaluated against the null model of neutral evolution in a single gene pool using three methods described below. Individuals were grouped in putative populations according to distribution (worldwide, Pacific Ocean, Atlantic Ocean, Indian Ocean), as well as based on Bayesian cluster analysis ("northern cluster", and "southern cluster", see Results). First, DNAsp v.5 [12] was used to calculate three metrics commonly referred to as “neutrality tests”: Tajima's D [19] (which compares average divergence to the number of segregating sites), Fu and Li's D [20] (which compares the number of singleton mutations to the number of segregating sites), and Fu and Li's F [20] (which compares the number of unique haplotypes, the number of singleton mutations, and the number of segregating sites). Rejection of the null model in these tests may be due to natural selection, or a violation of the model assumptions (e.g., stable population size throughout the coalescent, population subdivision). Second, genealogy shape was analyzed using a mismatch distribution (MMD), which summarizes the number of mismatches between all possible pairs of sequences in a histogram [21]. We evaluated the demographic MMD using a test in Arlequin v.3.5 [16] for which the null hypothesis is range expansion during the time represented by the gene genealogy. Statistical significance was assessed using 1,000 bootstraps. We assumed the mutation rate (μ) was 2% per base pair (bp) per million years, which has been previously used for marine teleosts [22–24]. Exocoetus volitans lives for approximately one year [7], which was used as the estimate of generation time. The timing of population expansion was estimated as Tau (τ), which was converted to years based on τ = 2μkt (k = number of nucleotides assayed, 1106 bp; μ = mutation rate per nucleotide, 2 x 10⁻⁸; t = time since population expansion). Goodness-of-fit between the observed and null MMD was calculated using Harpending raggedness indices (HRI) and sum of square deviations (SSD). Effective population sizes before and after population expansion were estimated as Θ_k, which was converted to N_e based on Θ_k = 2N_ekμ. Finally, we used BEAST v.1.6.1 [25] to generate a Bayesian skyline plot (BSP), which estimates N_e through time from the shape of the gene genealogy [26] (10 million MCMC generations, GTR + I + Γ model of evolution, random starting tree, 1 tree saved every 1,000 generations, 10% discarded as burn-in). TRACER v.1.4 [27] was used to visualize the BSP.

Results

Sampling and Genetic Data

We sequenced cytb fragments (1106 bps) from 266 E. volitans specimens collected from 97 locations: 150 individuals from 22 locations in the Atlantic Ocean, 111 individuals from 70 locations in the Pacific Ocean, and 5 individuals from 5 locations in the Indian Ocean (Table 1; Fig 1; S1 Table). There were 258 polymorphic sites, and 281 total mutations. Overall haplotype diversity was high (h = 0.998 ± 0.001), with 242 singleton haplotypes and only 12 haplotypes shared among individuals. However, nucleotide diversity was low, with an average number of only 7.049 differences among pairs of sequences (π = 0.006 ± 0.0002). Global estimates of Θ_k, and Θ_s were 44.626 and 0.04125, respectively (Table 1). Independent analyses of specimens collected within the Pacific, Atlantic and Indian Oceans resulted in similar estimates of haplotype diversity (0.900–0.999). G/C content was 0.424 (and 0.424–0.426 within the three oceans). Genetic diversity metrics that are known to be biased by small sample size tended to be lower in the Indian Ocean, where only 5 individuals were collected (Table 1).

Gene genealogy

The gene genealogy revealed a well-connected set of 242 unique, 12 shared, and 324 unsampled haplotypes (Fig 2). The ancestral haplotype (inferred based on its high frequency and centrality [13]) was shared by 10 individuals that were sampled in all three oceans. All other shared haplotypes were restricted to single oceans (8 within Atlantic, 3 within Pacific; Fig 2). The gene genealogy shows that many lineages are composed of individuals from multiple oceans; supporting the idea that gene flow occurs between oceans. Three haplotypes (6255, 6299, and 8387) could not be unambiguously connected and are not shown in Fig 2. An analysis using DNAPARS within PHYLIP v3.69 [28] produced a very similar topology (not presented here).

Download:

Fig 2. Cytochrome b gene haplotype network.

This figure was obtained using statistical parsimony analysis within TCS, v.1.21 [13]. Circle sizes are proportional to the number of shared haplotypes. Lines connecting circles represent single mutations. Blue circles = Atlantic Ocean, Red circles = Pacific Ocean, and Green circles = Indian Ocean. The square in the center represents the ancestral haplotype shared by samples from all three oceans. Open circles represent un-sampled haplotypes.

https://doi.org/10.1371/journal.pone.0163198.g002

Analyses of molecular variance

We found low but statistically significant differentiation across the Isthmus of Panama (Φ_ST = 0.0352, p < 0.001), but not the other three putative barriers (Sunda Shelf, Benguela, and Eastern Pacific; Φ_ST < 0.0001 and p > 0.4 for each; Table 2). We also detected a significant effect of a worldwide equatorial barrier (see below). The limited number of samples from the Indian Ocean (n = 5) did not allow for robust analysis of the potential influence of the Sunda Shelf and Benguela barriers. Our analyses of the other barriers were more robust, because of higher numbers of samples (n > 45) for each putative population.

Download:

Table 2. Results of analysis of molecular variance (AMOVA) for cytochrome b sequence data between putative populations of Exocoetus volitans.

Collection localities, numbers of individuals (n), putative dispersal barriers, Φ_ST values, and p-values are listed. Significant results are highlighted in bold and any negative Φ_ST values were set to zero.

https://doi.org/10.1371/journal.pone.0163198.t002

Bayesian clustering analysis

The individual clustering algorithm implemented in Geneland (which does not use a priori groups) revealed two gene pools: a “northern cluster” (n = 83: average posterior probability of assignment = 0.855 ± 0.136), and a “southern cluster” (n = 183: average posterior probability of assignment = 0.819 ± 0.156). Low posterior probabilities of assignment (0.50–0.55) were detected for only 13 individuals. Fig 3 highlights the spatial segregation of these gene pools, as well as a region of overlap near the Equator that was most obvious in a densely sampled area of the Pacific Ocean. Differentiation across this global Equatorial barrier was low but statistically significant (Table 2; AMOVA: Φ_ST = 0.027, p < 0.001).

Download:

Fig 3. Map of Exocoetus volitans population assignments.

After cluster analysis using Geneland v.1 [17], posterior probabilities of assignment were loaded into ArcGIS, version 9.3.1 (ESRI, Redlands, CA, USA). Green circles = individuals assigned to the "Northern cluster", Yellow circles = individuals assigned to "Southern cluster", and Grey circles = individuals with low confidence in cluster assignment (posterior probability values < 0.525).

https://doi.org/10.1371/journal.pone.0163198.g003

Historic population growth

Cytb mutations within E. volitans departed from null expectations for a selectively neutral gene in a single gene pool with constant N_e. All three test statistics were statistically significant: Tajima's D (D = -2.63; p< 0.001), Fu and Li's D (D = -5.24; p<0.02) and Fu and Li's F (F = 4.64, p< 0.02; Table 3). These results are consistent with relatively recent population growth, recovery after a severe (but temporary) bottleneck, some amount of population subdivision, or natural selection. Analysis of the mismatch distribution (MMD) and Bayesian skyline plot (BSP) both supported population growth as the best interpretation. The MMD analysis resulted in a unimodal distribution that did not significantly differ from the null model of population expansion, estimated to occur 147,604 years before present (95% CI: 137,839–159,810; Fig 4A, Table 3). Similarly, the BSP analysis suggests a population expansion of approximately one order of magnitude, sometime between 125,000 and 175,000 years before present (Fig 4). Contemporary N_e is estimated to be on the order of 10⁹ individuals.

Download:

Fig 4. Assessments of Historical Population Growth for Exocoetus volitans.

(A) Mismatch distribution of cytochrome b sequence data for all individuals (n = 266) generated using Arlequin, v.3.5 [16] and DNAsp, v.5 [12]. Solid lines indicate observed frequencies of pairwise differences and dashed lines with circles indicate expected frequencies of pairwise differences. (B) Bayesian skyline plot of all cytochrome b sequences generated using BEAST, v.1.6.1 [25]. The solid line indicates the median change in population size over time, and the gray area represents the 95% confidence interval (CI). Note that on the y-axis # of individuals is an estimate of population size based on N_eτ.

https://doi.org/10.1371/journal.pone.0163198.g004

Download:

Table 3. Results of neutrality tests and mismatch distributions for each putative population of Exocoetus volitans.

https://doi.org/10.1371/journal.pone.0163198.t003

Discussion

Because E. volitans is short-lived and small-bodied, we hypothesized that populations separated by known biogeographic barriers would represent discrete gene pools. However, mtDNA population structure was very limited. We did detect statistically significant differentiation between the Pacific and Atlantic Oceans, as well as a global population barrier delineating gene pools north and south of the Equator.

Circumtropical patterns of population structure in an epipelagic specialist

Despite two statistically significant barriers to global dispersal, E. volitans individuals share high genetic similarity on a global scale. Population genetic structure has been studied in two other flyingfish species with smaller distributions. Based on mtDNA data, there is no population differentiation within the Bony Flyingfish, Hirundichthys oxycephalus, in the northwestern Pacific (Ф_CT = −0.0306, p = 0.1, [29]). However, the Fourwing Flyingfish, Hirundichthys affinis, is subdivided into three populations or "stocks" in the tropical Atlantic (Φ_CT = 0.49; Φ_ST = 0.42 to 0.80; [30]). Life history differences between E. volitans and Hirundichthys affinis may explain these contrasting patterns. E. volitans has a much larger geographic range, covering > 100 million kilometers of open ocean [6], while the distribution of H. affinis is ten-fold smaller [31]. Additionally, H. affinis lives in coastal waters (neritic) particularly during early life stages [32], while E. volitans primarily occurs in epipelagic waters throughout all stages of life (holo-epipelagic) [6]. Also, H. affinis spawns in aggregations, attaching eggs to floating vegetation [29, 30] while, E. volitans broadcasts buoyant eggs [9]. Whether the reproductive behavior of E. volitans contributes to high levels of population connectivity is not fully understood because the reproductive biology of this species remains to be fully documented. Major distinctions in life history and population sizes likely confer species differences in population-level genetic variation.

Our finding of minimal global-scale population structure in E. volitans can be compared with other genetic studies involving widespread marine fishes based on mitochondrial DNA data (Table 4). A number of widespread oceanic fish species have higher Φ_ST values than detected in E. volitans, including swordfish (Φ_ST = 0.09; [33], bigeye tuna (Φ_ST = 0.22; [34]) and blue marlin (Φ_ST = 0.22; [35]). Interestingly, [8], large body size and high mobility are imperfectly correlated with Φ_ST, since even lower Φ_ST values have been observed from intermediate-sized (e.g., pelagic wahoo; Φ_ST < 0.0001; [24]) and small-bodied species (e.g., tropical two-wing flyingfish; Φ_ST < 0.04; this study).

Download:

Table 4. Comparison of Φ_ST values from population genetic studies involving widely distributed marine teleosts including Exocoetus volitans (results of present study in bold), and two other flyingfish species (Hirundichthys affinis, and H. oxycephalus) for which a Φ_ST estimates are available.

https://doi.org/10.1371/journal.pone.0163198.t004

Extremely large effective population sizes, that are comparable to those we estimate for E. volitans, drive nonequilibrium conditions that can result in estimates of genetic divergence that are comparable to the effects of extensive contemporary gene flow [44]. Although near-zero values of Φ_ST are consistent with high gene flow, they are also consistent with the persistent signature of a recent population expansion under almost any level of current gene flow [21]. Our estimates of species-wide population expansion to > 1 billion fish only ~150,000 years ago (see below) are conditions that could lead to this nonequilibrium effect on Φ_ST estimates. Thus, based on our data, we cannot dismiss the possibility that current gene flow across the range of E. volitans is lower than indicated by Φ_ST estimates [45].

Evidence for barriers to E. volitans gene flow

Our analyses indicate at least two barriers to gene flow in E. volitans: the Isthmus of Panama barrier and a worldwide equatorial barrier (Fig 3). Small sample sizes limited our ability to adequately test whether other barriers (e.g., BB and SSB) are equally important. The effects of the Isthmus of Panama barrier and a worldwide equatorial barrier are detectable notwithstanding the potential effects of large population size and population expansion described above. The Isthmus of Panama currently represents an impermeable barrier for most marine fishes, and may have been a barrier for pelagic marine fishes for the past 15 million years [46].

The finding of comparable Φ_ST values (~0.03) for the Isthmian and equatorial barriers (Table 2) suggests that gene flow across the equatorial barrier is low. The worldwide equatorial barrier we propose for E. volitans has also been suggested for the pelagic copepod Haloptilus longicornis by Norton and Goetze [47]. These authors found a distinct genetic break at approximately 0–12°N in both the Atlantic and Pacific oceans, and hypothesized the presence of a physical or biophysical barrier restricting the latitudinal dispersal of pelagic copepods. A similar genetic subdivision has not been proposed for circumtropical fishes, although it has been discussed for a few species within ocean basins. For example, in the Atlantic, distinct swordfish populations have been detected on either side of the Equator [48]. Similarly, populations of striped marlin in the Pacific may partition in a north-to-south pattern [49]. However, hypothesizing a common cause for this divergence in multiple taxa is difficult due to variation in the biology of the species. Swordfish and marlin are large, fast-swimming, highly migratory, thermally-tolerant, and breed in specific areas, suggesting that behavior may play in role in their divergence. In contrast, for slow-swimming E. volitans and the planktonic copepod H. longicornis, oceanographic and/or biophysical barriers to dispersal are the most likely explanation for equatorial genetic divergence [47]. For E. volitans, it is possible that the species’ buoyant eggs are advected away from equatorial waters through natural processes of oceanographic upwelling and/or the Coriolis Effect.

E. volitans historical demography

Based on the star-like mitochondrial gene genealogy (i.e., high haplotype diversity and low nucleotide diversity), the species-wide effective population size of E. volitans appears to have been smaller in the recent past. The BSP and MMD analyses both suggested dramatic population growth beginning approximately 150,000 years ago. This time frame (the middle Pleistocene) was also suggested by Chou and colleagues [29] for demographic population expansion of H. oxycephalus. These authors suggested that range and/or demographic expansion may have been driven by the end of an era of maximum cooling sea surface temperatures (SST). Indeed, sea level changes (along with SST changes) during the Pleistocene have been suggested as correlated for population bottlenecks and/or by post-glacial range expansions in many species [50–54]. The preference of E. volitans for warmer tropical waters may have shaped its demographic history, and likely continues to influence the species' contemporary patterns of gene flow. However, given our study’s focus on a single genetic marker, we cannot discount other explanations for the mitochondrial gene genealogy. An alternative interpretation is that a selective sweep occurred within this species, leading to an abundance of distal branches, creating the "fireworks"-like appearance of this genealogy (see [55]). Multilocus assessments and genetic analyses of other epipelagic specialists will allow further validation of the post-glacial range expansion hypothesis, and provide key insight regarding current patterns of epipelagic marine biodiversity.

Supporting Information

S1 Table. Specimen sampling data for Exocoetus volitans specimens.

Museum voucher numbers (when available) and collection information are included for individuals used in genetic data analyses (n = 266). Assignments to southern or northern clusters resulting from analysis within Geneland v.1 [17] are also identified.

https://doi.org/10.1371/journal.pone.0163198.s001

(PDF)

Acknowledgments

We thank all who assisted with the collection of specimens aboard research vessels: Endeavor (R McMunn, R Chase, B Collins, P Roussell, J Montminy, P Quigley, K Walsh, G Maltby, T Varney, A Wright, M Brennan, B Wilson, B Fanning, A Tucker, K Pohl, V Pascal, R Cooper, L Koren, H Hamner, R Lohmann), Shoyo Maru, Gordon Gunter, McArthur II, David Starr Jordan, Kahana, and Oscar Sette. In particular, we thank J Cotton, JC Salinas, and L Ballance for their ongoing dedication to flyingfish research and assistance with dipnet sampling logistics. Additional tissues were donated by the Marine Fisheries Service (B Collette). Specimen curation was assisted by the Royal Ontario Museum (R Winterbottom, E Holm, M Burridge, D Stacy, and F Pardo), Scripps Institution of Oceanography (P Hastings, HJ Walker, P Ajtai) and University of Toronto (M Huang, D Xiao, J Gordon, JC Szamosi). Genetic data collection was assisted by A Shah and B Shah. The scientific results and conclusions, as well as any views or opinions expressed herein, are those of the author(s) and do not necessarily reflect those of NOAA or the Department of Commerce.

Author Contributions

Conceptualization: EAL AJB CAB AJVW RLP NRL.
Data curation: EAL AJB CAB AJVW RLP NRL.
Formal analysis: EAL AJB NRL.
Funding acquisition: EAL AJVW RLP NRL.
Investigation: EAL AJB CAB AJVW RLP NRL.
Methodology: EAL AJB RLP NRL.
Project administration: AJB AJVW RLP NRL.
Resources: EAL AJB AJVW RLP NRL.
Software: EAL AJB CAB NRL.
Supervision: AJB AJVW RLP NRL.
Validation: EAL AJB RLP NRL.
Visualization: EAL AJB CAB AJVW RLP NRL.
Writing – original draft: EAL AJB CAB RLP NRL.
Writing – review & editing: EAL AJB CAB AJVW RLP NRL.

References

1. Worm B, Barbier EB, Beaumont N, Duffy JE, Folke C, Halpern BS, et al. Impacts of biodiversity loss on ocean ecosystem services. Science. 2006;314(5800):787–90. pmid:17082450
- View Article
- PubMed/NCBI
- Google Scholar
2. Rocha L, Bowen B. Speciation in coral‐reef fishes. J Fish Biol. 2008;72(5):1101–21.
- View Article
- Google Scholar
3. Rocha L, Craig M, Bowen B. Phylogeography and the conservation of coral reef fishes. Coral Reefs. 2007;26(3):501–12.
- View Article
- Google Scholar
4. Longhurst AR, Pauly D. Ecology of tropical oceans: Academic Press San Diego; 1987.
5. Gaither MR, Bowen BW, Rocha LA, Briggs JC. Fishes that rule the world: circumtropical distributions revisited. Fish and Fisheries. 2015.
6. Parin N, Shakhovskoy I. A review of the flying fish genus Exocoetus (Exocoetidae) with descriptions of two new species from the southern Pacific Ocean. Journal of Ichthyology. 2000;40(1):S31.
- View Article
- Google Scholar
7. Grudtsev M, Salekhova L, Lushchina V, SSR—In Byum S. Distribution, ecology and intraspecific variability of flyingfishes of the genus Exocoetus of the Atlantic Ocean. Journal of Ichthyology. 1987;27:39–50.
- View Article
- Google Scholar
8. Bradbury IR, Laurel B, Snelgrove PV, Bentzen P, Campana SE. Global patterns in marine dispersal estimates: the influence of geography, taxonomic category and life history. Proc Biol Sci. 2008;275(1644):1803–9. Epub 2008/05/01. pmid:18445556; PubMed Central PMCID: PMC2587791.
- View Article
- PubMed/NCBI
- Google Scholar
9. Collette B, McGowen G, Parin N, Mito S. Beloniformes: development and relationships. Ontogeny and systematics of fishes. 1984:335–54.
10. Parin N. Ichthyofauna of oceanic epipelagic zone.“. Nauka Press. Moscow; 1968.
11. Lewallen EA, Pitman RL, Kjartanson SL, Lovejoy NR. Molecular systematics of flyingfishes (Teleostei: Exocoetidae): evolution in the epipelagic zone. Biological Journal of the Linnean Society. 2011;102(1):161–74.
- View Article
- Google Scholar
12. Librado P, Rozas J. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics. 2009;25(11):1451–2. pmid:19346325
- View Article
- PubMed/NCBI
- Google Scholar
13. Clement M, Posada D, Crandall KA. TCS: a computer program to estimate gene genealogies. Mol Ecol. 2000;9(10):1657–9. pmid:11050560
- View Article
- PubMed/NCBI
- Google Scholar
14. Crandall K, Templeton A, Sing C. Intraspecific phylogenetics: problems and solutions. Systematics Association Special Volume. 1994;52:273-.
- View Article
- Google Scholar
15. Templeton AR, Routman E, Phillips CA. Separating population structure from population history: a cladistic analysis of the geographical distribution of mitochondrial DNA haplotypes in the tiger salamander, Ambystoma tigrinum. Genetics. 1995;140(2):767–82. pmid:7498753
- View Article
- PubMed/NCBI
- Google Scholar
16. Excoffier L, Lischer HE. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol Resour. 2010;10(3):564–7. pmid:21565059
- View Article
- PubMed/NCBI
- Google Scholar
17. Guillot G, Mortier F, Estoup A. GENELAND: a computer package for landscape genetics. Molecular ecology notes. 2005;5(3):712–5.
- View Article
- Google Scholar
18. Guillot G, Leblois R, Coulon A, Frantz AC. Statistical methods in spatial genetics. Mol Ecol. 2009;18(23):4734–56. Epub 2009/11/03. pmid:19878454.
- View Article
- PubMed/NCBI
- Google Scholar
19. Tajima F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics. 1989;123(3):585–95. pmid:2513255
- View Article
- PubMed/NCBI
- Google Scholar
20. Fu Y-X, Li W-H. Statistical tests of neutrality of mutations. Genetics. 1993;133(3):693–709. pmid:8454210
- View Article
- PubMed/NCBI
- Google Scholar
21. Rogers AR, Harpending H. Population growth makes waves in the distribution of pairwise genetic differences. Mol Biol Evol. 1992;9(3):552–69. pmid:1316531
- View Article
- PubMed/NCBI
- Google Scholar
22. Avise JC. Phylogeography: the history and formation of species: Harvard university press; 2000.
23. Bowen B, Bass A, Rocha L, Grant W, Robertson DR. Phylogeography of the trumpetfishes (Aulostomus): ring species complex on a global scale. Evolution. 2001;55(5):1029–39. pmid:11430639
- View Article
- PubMed/NCBI
- Google Scholar
24. Theisen TC, Bowen BW, Lanier W, Baldwin JD. High connectivity on a global scale in the pelagic wahoo, Acanthocybium solandri (tuna family Scombridae). Mol Ecol. 2008;17(19):4233–47. pmid:19378403
- View Article
- PubMed/NCBI
- Google Scholar
25. Drummond AJ, Rambaut A. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol Biol. 2007;7(1):214.
- View Article
- Google Scholar
26. Drummond AJ, Rambaut A, Shapiro B, Pybus OG. Bayesian coalescent inference of past population dynamics from molecular sequences. Mol Biol Evol. 2005;22(5):1185–92. Epub 2005/02/11. pmid:15703244.
- View Article
- PubMed/NCBI
- Google Scholar
27. Rambaut A, Drummond AJ. Tracer v1. 4. 2007.
28. Felsenstein J. {PHYLIP}: phylogenetic inference package, version 3.5 c. 1993.
29. Chou C-E, Liao T-Y, Chang H-W, Chang S-K. Population structure of Hirundichthys oxycephalus in the northwestern Pacific inferred from mitochondrial cytochrome oxidase I gene. Zoological Studies. 2015;54(1).
- View Article
- Google Scholar
30. Gomes C, Dales RB, Oxenford HA. The application of RAPD markers in stock discrimination of the four‐wing flyingfish, Hirundichthys affinis in the central western Atlantic. Mol Ecol. 1998;7(8):1029–39.
- View Article
- Google Scholar
31. Shakhovskoy IB, Parin NV. A review of flying fishes of the subgenus Hirundichthys (Genus Hirundichthys, Exocoetidae). Part 2. Nerito-oceanic species: H. oxycephalus, H. affinis. Journal of Ichthyology. 2013;53(8):509–40.
- View Article
- Google Scholar
32. Parin N, Belyanina T. A review of flyingfishes of the subgenus Danichthys (genus Hirundichthys, Exocoetidae). Journal of Ichthyology. 2002;42(1):S23.
- View Article
- Google Scholar
33. Alvarado Bremer JR, Vinas J, Mejuto J, Ely B, Pla C. Comparative phylogeography of Atlantic bluefin tuna and swordfish: the combined effects of vicariance, secondary contact, introgression, and population expansion on the regional phylogenies of two highly migratory pelagic fishes. Mol Phylogenet Evol. 2005;36(1):169–87. Epub 2005/05/21. pmid:15904864.
- View Article
- PubMed/NCBI
- Google Scholar
34. Martinez P, Gonzalez EG, Castilho R, Zardoya R. Genetic diversity and historical demography of Atlantic bigeye tuna (Thunnus obesus). Mol Phylogenet Evol. 2006;39(2):404–16. Epub 2005/09/29. pmid:16188460.
- View Article
- PubMed/NCBI
- Google Scholar
35. Buonaccorsi V, McDowell J, Graves J. Reconciling patterns of inter‐ocean molecular variance from four classes of molecular markers in blue marlin (Makaira nigricans). Mol Ecol. 2001;10(5):1179–96. pmid:11380876
- View Article
- PubMed/NCBI
- Google Scholar
36. Gaither MR, Toonen RJ, Robertson DR, Planes S, Bowen BW. Genetic evaluation of marine biogeographical barriers: perspectives from two widespread Indo-Pacific snappers (Lutjanus kasmira and Lutjanus fulvus). Journal of Biogeography. 2009;37(1):133–47.
- View Article
- Google Scholar
37. Craig MT, Jeff AE, Brian WB, Robertson DR. High genetic connectivity across the Indian and Pacific Oceans in the reef fish Myripristis berndti (Holocentridae). Marine Ecology Progress Series. 2007;334:245–54.
- View Article
- Google Scholar
38. Klanten OS, Choat JH, van Herwerden L. Extreme genetic diversity and temporal rather than spatial partitioning in a widely distributed coral reef fish. Marine Biology. 2006;150(4):659–70.
- View Article
- Google Scholar
39. Ely B, Vinas J, Alvarado Bremer JR, Black D, Lucas L, Covello K, et al. Consequences of the historical demography on the global population structure of two highly migratory cosmopolitan marine fishes: the yellowfin tuna (Thunnus albacares) and the skipjack tuna (Katsuwonus pelamis). BMC Evol Biol. 2005;5:19. Epub 2005/02/24. pmid:15725349; PubMed Central PMCID: PMC554763.
- View Article
- PubMed/NCBI
- Google Scholar
40. Vinas J, Alvarado Bremer JR, Pla C. Inter-oceanic genetic differentiation among albacore (Thunnus alalunga) populations. Marine Biology. 2004;145(2).
- View Article
- Google Scholar
41. Graves JE, McDowell JR. Genetic analysis of white marlin (Tetrapturus albidus) stock structure. Bulletin of Marine Science. 2006;79(3):469–82.
- View Article
- Google Scholar
42. Horne JB, van Herwerden L, Choat JH, Robertson DR. High population connectivity across the Indo-Pacific: Congruent lack of phylogeographic structure in three reef fish congeners. Mol Phylogenet Evol. 2008;49(2):629–38. Epub 2008/09/23. pmid:18804542.
- View Article
- PubMed/NCBI
- Google Scholar
43. Carlsson J, McDowell JR, Carlsson JE, Graves JE. Genetic identity of YOY bluefin tuna from the eastern and western Atlantic spawning areas. J Hered. 2007;98(1):23–8. Epub 2006/12/13. pmid:17158466.
- View Article
- PubMed/NCBI
- Google Scholar
44. Bohonak AJ, Roderick GK. Dispersal of invertebrates among temporary ponds: Are genetic estimates accurate? Israel Journal of Zoology. 2001;47(4):367–86.
- View Article
- Google Scholar
45. Bohonak AJ. Dispersal, gene flow, and population structure. Quarterly review of biology. 1999:21–45. pmid:10081813
- View Article
- PubMed/NCBI
- Google Scholar
46. Lessios HA. The Great American Schism: Divergence of marine organisms after the rise of the Central American Isthmus. Annual Review of Ecology, Evolution, and Systematics. 2008;39(1):63–91.
- View Article
- Google Scholar
47. Norton EL, Goetze E. Equatorial dispersal barriers and limited population connectivity among oceans in a planktonic copepod. Limnology and Oceanography. 2013;58(5):1581–96.
- View Article
- Google Scholar
48. Chow S, Clarke S, Nakadate M, Okazaki M. Boundary between the north and south Atlantic populations of the swordfish (Xiphias gladius) inferred by a single nucleotide polymorphism at calmodulin gene intron. Marine Biology. 2007;152(1):87–93.
- View Article
- Google Scholar
49. McDowell JR, Graves JE. Population structure of striped marlin (Kajikia audax) in the Pacific Ocean based on analysis of microsatellite and mitochondrial DNA. Canadian Journal of Fisheries and Aquatic Sciences. 2008;65(7):1307–20.
- View Article
- Google Scholar
50. Grant W, Bowen B. Shallow population histories in deep evolutionary lineages of marine fishes: insights from sardines and anchovies and lessons for conservation. Journal of Heredity. 1998;89(5):415–26.
- View Article
- Google Scholar
51. Barber PH, Palumbi SR, Erdmann MV, Moosa MK. Biogeography: a marine Wallace's line? Nature. 2000;406(6797):692–3. pmid:10963585
- View Article
- PubMed/NCBI
- Google Scholar
52. Barber P, Palumbi S, Erdmann M, Moosa M. Sharp genetic breaks among populations of Haptosquilla pulchella (Stomatopoda) indicate limits to larval transport: patterns, causes, and consequences. Mol Ecol. 2002;11(4):659–74. pmid:11972755
- View Article
- PubMed/NCBI
- Google Scholar
53. Planes S, Lenfant P. Temporal change in the genetic structure between and within cohorts of a marine fish, Diplodus sargus, induced by a large variance in individual reproductive success. Mol Ecol. 2002;11(8):1515–24. pmid:12144670
- View Article
- PubMed/NCBI
- Google Scholar
54. Ludt WB, Rocha LA. Shifting seas: the impacts of Pleistocene sea-level fluctuations on the evolution of tropical marine taxa. Journal of Biogeography. 2015;42(1):25–38.
- View Article
- Google Scholar
55. Ho SY, Shapiro B. Skyline-plot methods for estimating demographic history from nucleotide sequences. Mol Ecol Resour. 2011;11(3):423–34. pmid:21481200
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Worm B, Barbier EB, Beaumont N, Duffy JE, Folke C, Halpern BS, et al. Impacts of biodiversity loss on ocean ecosystem services. Science. 2006;314(5800):787–90. pmid:17082450
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. Rocha L, Bowen B. Speciation in coral‐reef fishes. J Fish Biol. 2008;72(5):1101–21.
View Article
Google Scholar

[6] View Article

[7] Google Scholar

[ref3] 3. Rocha L, Craig M, Bowen B. Phylogeography and the conservation of coral reef fishes. Coral Reefs. 2007;26(3):501–12.
View Article
Google Scholar

[9] View Article

[10] Google Scholar

[ref4] 4. Longhurst AR, Pauly D. Ecology of tropical oceans: Academic Press San Diego; 1987.

[ref5] 5. Gaither MR, Bowen BW, Rocha LA, Briggs JC. Fishes that rule the world: circumtropical distributions revisited. Fish and Fisheries. 2015.

[ref6] 6. Parin N, Shakhovskoy I. A review of the flying fish genus Exocoetus (Exocoetidae) with descriptions of two new species from the southern Pacific Ocean. Journal of Ichthyology. 2000;40(1):S31.
View Article
Google Scholar

[14] View Article

[15] Google Scholar

[ref7] 7. Grudtsev M, Salekhova L, Lushchina V, SSR—In Byum S. Distribution, ecology and intraspecific variability of flyingfishes of the genus Exocoetus of the Atlantic Ocean. Journal of Ichthyology. 1987;27:39–50.
View Article
Google Scholar

[17] View Article

[18] Google Scholar

[ref8] 8. Bradbury IR, Laurel B, Snelgrove PV, Bentzen P, Campana SE. Global patterns in marine dispersal estimates: the influence of geography, taxonomic category and life history. Proc Biol Sci. 2008;275(1644):1803–9. Epub 2008/05/01. pmid:18445556; PubMed Central PMCID: PMC2587791.
View Article
PubMed/NCBI
Google Scholar

[20] View Article

[21] PubMed/NCBI

[22] Google Scholar

[ref9] 9. Collette B, McGowen G, Parin N, Mito S. Beloniformes: development and relationships. Ontogeny and systematics of fishes. 1984:335–54.

[ref10] 10. Parin N. Ichthyofauna of oceanic epipelagic zone.“. Nauka Press. Moscow; 1968.

[ref11] 11. Lewallen EA, Pitman RL, Kjartanson SL, Lovejoy NR. Molecular systematics of flyingfishes (Teleostei: Exocoetidae): evolution in the epipelagic zone. Biological Journal of the Linnean Society. 2011;102(1):161–74.
View Article
Google Scholar

[26] View Article

[27] Google Scholar

[ref12] 12. Librado P, Rozas J. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics. 2009;25(11):1451–2. pmid:19346325
View Article
PubMed/NCBI
Google Scholar

[29] View Article

[30] PubMed/NCBI

[31] Google Scholar

[ref13] 13. Clement M, Posada D, Crandall KA. TCS: a computer program to estimate gene genealogies. Mol Ecol. 2000;9(10):1657–9. pmid:11050560
View Article
PubMed/NCBI
Google Scholar

[33] View Article

[34] PubMed/NCBI

[35] Google Scholar

[ref14] 14. Crandall K, Templeton A, Sing C. Intraspecific phylogenetics: problems and solutions. Systematics Association Special Volume. 1994;52:273-.
View Article
Google Scholar

[37] View Article

[38] Google Scholar

[ref15] 15. Templeton AR, Routman E, Phillips CA. Separating population structure from population history: a cladistic analysis of the geographical distribution of mitochondrial DNA haplotypes in the tiger salamander, Ambystoma tigrinum. Genetics. 1995;140(2):767–82. pmid:7498753
View Article
PubMed/NCBI
Google Scholar

[40] View Article

[41] PubMed/NCBI

[42] Google Scholar

[ref16] 16. Excoffier L, Lischer HE. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol Resour. 2010;10(3):564–7. pmid:21565059
View Article
PubMed/NCBI
Google Scholar

[44] View Article

[45] PubMed/NCBI

[46] Google Scholar

[ref17] 17. Guillot G, Mortier F, Estoup A. GENELAND: a computer package for landscape genetics. Molecular ecology notes. 2005;5(3):712–5.
View Article
Google Scholar

[48] View Article

[49] Google Scholar

[ref18] 18. Guillot G, Leblois R, Coulon A, Frantz AC. Statistical methods in spatial genetics. Mol Ecol. 2009;18(23):4734–56. Epub 2009/11/03. pmid:19878454.
View Article
PubMed/NCBI
Google Scholar

[51] View Article

[52] PubMed/NCBI

[53] Google Scholar

[ref19] 19. Tajima F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics. 1989;123(3):585–95. pmid:2513255
View Article
PubMed/NCBI
Google Scholar

[55] View Article

[56] PubMed/NCBI

[57] Google Scholar

[ref20] 20. Fu Y-X, Li W-H. Statistical tests of neutrality of mutations. Genetics. 1993;133(3):693–709. pmid:8454210
View Article
PubMed/NCBI
Google Scholar

[59] View Article

[60] PubMed/NCBI

[61] Google Scholar

[ref21] 21. Rogers AR, Harpending H. Population growth makes waves in the distribution of pairwise genetic differences. Mol Biol Evol. 1992;9(3):552–69. pmid:1316531
View Article
PubMed/NCBI
Google Scholar

[63] View Article

[64] PubMed/NCBI

[65] Google Scholar

[ref22] 22. Avise JC. Phylogeography: the history and formation of species: Harvard university press; 2000.

[ref23] 23. Bowen B, Bass A, Rocha L, Grant W, Robertson DR. Phylogeography of the trumpetfishes (Aulostomus): ring species complex on a global scale. Evolution. 2001;55(5):1029–39. pmid:11430639
View Article
PubMed/NCBI
Google Scholar

[68] View Article

[69] PubMed/NCBI

[70] Google Scholar

[ref24] 24. Theisen TC, Bowen BW, Lanier W, Baldwin JD. High connectivity on a global scale in the pelagic wahoo, Acanthocybium solandri (tuna family Scombridae). Mol Ecol. 2008;17(19):4233–47. pmid:19378403
View Article
PubMed/NCBI
Google Scholar

[72] View Article

[73] PubMed/NCBI

[74] Google Scholar

[ref25] 25. Drummond AJ, Rambaut A. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol Biol. 2007;7(1):214.
View Article
Google Scholar

[76] View Article

[77] Google Scholar

[ref26] 26. Drummond AJ, Rambaut A, Shapiro B, Pybus OG. Bayesian coalescent inference of past population dynamics from molecular sequences. Mol Biol Evol. 2005;22(5):1185–92. Epub 2005/02/11. pmid:15703244.
View Article
PubMed/NCBI
Google Scholar

[79] View Article

[80] PubMed/NCBI

[81] Google Scholar

[ref27] 27. Rambaut A, Drummond AJ. Tracer v1. 4. 2007.

[ref28] 28. Felsenstein J. {PHYLIP}: phylogenetic inference package, version 3.5 c. 1993.

[ref29] 29. Chou C-E, Liao T-Y, Chang H-W, Chang S-K. Population structure of Hirundichthys oxycephalus in the northwestern Pacific inferred from mitochondrial cytochrome oxidase I gene. Zoological Studies. 2015;54(1).
View Article
Google Scholar

[85] View Article

[86] Google Scholar

[ref30] 30. Gomes C, Dales RB, Oxenford HA. The application of RAPD markers in stock discrimination of the four‐wing flyingfish, Hirundichthys affinis in the central western Atlantic. Mol Ecol. 1998;7(8):1029–39.
View Article
Google Scholar

[88] View Article

[89] Google Scholar

[ref31] 31. Shakhovskoy IB, Parin NV. A review of flying fishes of the subgenus Hirundichthys (Genus Hirundichthys, Exocoetidae). Part 2. Nerito-oceanic species: H. oxycephalus, H. affinis. Journal of Ichthyology. 2013;53(8):509–40.
View Article
Google Scholar

[91] View Article

[92] Google Scholar

[ref32] 32. Parin N, Belyanina T. A review of flyingfishes of the subgenus Danichthys (genus Hirundichthys, Exocoetidae). Journal of Ichthyology. 2002;42(1):S23.
View Article
Google Scholar

[94] View Article

[95] Google Scholar

[ref33] 33. Alvarado Bremer JR, Vinas J, Mejuto J, Ely B, Pla C. Comparative phylogeography of Atlantic bluefin tuna and swordfish: the combined effects of vicariance, secondary contact, introgression, and population expansion on the regional phylogenies of two highly migratory pelagic fishes. Mol Phylogenet Evol. 2005;36(1):169–87. Epub 2005/05/21. pmid:15904864.
View Article
PubMed/NCBI
Google Scholar

[97] View Article

[98] PubMed/NCBI

[99] Google Scholar

[ref34] 34. Martinez P, Gonzalez EG, Castilho R, Zardoya R. Genetic diversity and historical demography of Atlantic bigeye tuna (Thunnus obesus). Mol Phylogenet Evol. 2006;39(2):404–16. Epub 2005/09/29. pmid:16188460.
View Article
PubMed/NCBI
Google Scholar

[101] View Article

[102] PubMed/NCBI

[103] Google Scholar

[ref35] 35. Buonaccorsi V, McDowell J, Graves J. Reconciling patterns of inter‐ocean molecular variance from four classes of molecular markers in blue marlin (Makaira nigricans). Mol Ecol. 2001;10(5):1179–96. pmid:11380876
View Article
PubMed/NCBI
Google Scholar

[105] View Article

[106] PubMed/NCBI

[107] Google Scholar

[ref36] 36. Gaither MR, Toonen RJ, Robertson DR, Planes S, Bowen BW. Genetic evaluation of marine biogeographical barriers: perspectives from two widespread Indo-Pacific snappers (Lutjanus kasmira and Lutjanus fulvus). Journal of Biogeography. 2009;37(1):133–47.
View Article
Google Scholar

[109] View Article

[110] Google Scholar

[ref37] 37. Craig MT, Jeff AE, Brian WB, Robertson DR. High genetic connectivity across the Indian and Pacific Oceans in the reef fish Myripristis berndti (Holocentridae). Marine Ecology Progress Series. 2007;334:245–54.
View Article
Google Scholar

[112] View Article

[113] Google Scholar

[ref38] 38. Klanten OS, Choat JH, van Herwerden L. Extreme genetic diversity and temporal rather than spatial partitioning in a widely distributed coral reef fish. Marine Biology. 2006;150(4):659–70.
View Article
Google Scholar

[115] View Article

[116] Google Scholar

[ref39] 39. Ely B, Vinas J, Alvarado Bremer JR, Black D, Lucas L, Covello K, et al. Consequences of the historical demography on the global population structure of two highly migratory cosmopolitan marine fishes: the yellowfin tuna (Thunnus albacares) and the skipjack tuna (Katsuwonus pelamis). BMC Evol Biol. 2005;5:19. Epub 2005/02/24. pmid:15725349; PubMed Central PMCID: PMC554763.
View Article
PubMed/NCBI
Google Scholar

[118] View Article

[119] PubMed/NCBI

[120] Google Scholar

[ref40] 40. Vinas J, Alvarado Bremer JR, Pla C. Inter-oceanic genetic differentiation among albacore (Thunnus alalunga) populations. Marine Biology. 2004;145(2).
View Article
Google Scholar

[122] View Article

[123] Google Scholar

[ref41] 41. Graves JE, McDowell JR. Genetic analysis of white marlin (Tetrapturus albidus) stock structure. Bulletin of Marine Science. 2006;79(3):469–82.
View Article
Google Scholar

[125] View Article

[126] Google Scholar

[ref42] 42. Horne JB, van Herwerden L, Choat JH, Robertson DR. High population connectivity across the Indo-Pacific: Congruent lack of phylogeographic structure in three reef fish congeners. Mol Phylogenet Evol. 2008;49(2):629–38. Epub 2008/09/23. pmid:18804542.
View Article
PubMed/NCBI
Google Scholar

[128] View Article

[129] PubMed/NCBI

[130] Google Scholar

[ref43] 43. Carlsson J, McDowell JR, Carlsson JE, Graves JE. Genetic identity of YOY bluefin tuna from the eastern and western Atlantic spawning areas. J Hered. 2007;98(1):23–8. Epub 2006/12/13. pmid:17158466.
View Article
PubMed/NCBI
Google Scholar

[132] View Article

[133] PubMed/NCBI

[134] Google Scholar

[ref44] 44. Bohonak AJ, Roderick GK. Dispersal of invertebrates among temporary ponds: Are genetic estimates accurate? Israel Journal of Zoology. 2001;47(4):367–86.
View Article
Google Scholar

[136] View Article

[137] Google Scholar

[ref45] 45. Bohonak AJ. Dispersal, gene flow, and population structure. Quarterly review of biology. 1999:21–45. pmid:10081813
View Article
PubMed/NCBI
Google Scholar

[139] View Article

[140] PubMed/NCBI

[141] Google Scholar

[ref46] 46. Lessios HA. The Great American Schism: Divergence of marine organisms after the rise of the Central American Isthmus. Annual Review of Ecology, Evolution, and Systematics. 2008;39(1):63–91.
View Article
Google Scholar

[143] View Article

[144] Google Scholar

[ref47] 47. Norton EL, Goetze E. Equatorial dispersal barriers and limited population connectivity among oceans in a planktonic copepod. Limnology and Oceanography. 2013;58(5):1581–96.
View Article
Google Scholar

[146] View Article

[147] Google Scholar

[ref48] 48. Chow S, Clarke S, Nakadate M, Okazaki M. Boundary between the north and south Atlantic populations of the swordfish (Xiphias gladius) inferred by a single nucleotide polymorphism at calmodulin gene intron. Marine Biology. 2007;152(1):87–93.
View Article
Google Scholar

[149] View Article

[150] Google Scholar

[ref49] 49. McDowell JR, Graves JE. Population structure of striped marlin (Kajikia audax) in the Pacific Ocean based on analysis of microsatellite and mitochondrial DNA. Canadian Journal of Fisheries and Aquatic Sciences. 2008;65(7):1307–20.
View Article
Google Scholar

[152] View Article

[153] Google Scholar

[ref50] 50. Grant W, Bowen B. Shallow population histories in deep evolutionary lineages of marine fishes: insights from sardines and anchovies and lessons for conservation. Journal of Heredity. 1998;89(5):415–26.
View Article
Google Scholar

[155] View Article

[156] Google Scholar

[ref51] 51. Barber PH, Palumbi SR, Erdmann MV, Moosa MK. Biogeography: a marine Wallace's line? Nature. 2000;406(6797):692–3. pmid:10963585
View Article
PubMed/NCBI
Google Scholar

[158] View Article

[159] PubMed/NCBI

[160] Google Scholar

[ref52] 52. Barber P, Palumbi S, Erdmann M, Moosa M. Sharp genetic breaks among populations of Haptosquilla pulchella (Stomatopoda) indicate limits to larval transport: patterns, causes, and consequences. Mol Ecol. 2002;11(4):659–74. pmid:11972755
View Article
PubMed/NCBI
Google Scholar

[162] View Article

[163] PubMed/NCBI

[164] Google Scholar

[ref53] 53. Planes S, Lenfant P. Temporal change in the genetic structure between and within cohorts of a marine fish, Diplodus sargus, induced by a large variance in individual reproductive success. Mol Ecol. 2002;11(8):1515–24. pmid:12144670
View Article
PubMed/NCBI
Google Scholar

[166] View Article

[167] PubMed/NCBI

[168] Google Scholar

[ref54] 54. Ludt WB, Rocha LA. Shifting seas: the impacts of Pleistocene sea-level fluctuations on the evolution of tropical marine taxa. Journal of Biogeography. 2015;42(1):25–38.
View Article
Google Scholar

[170] View Article

[171] Google Scholar

[ref55] 55. Ho SY, Shapiro B. Skyline-plot methods for estimating demographic history from nucleotide sequences. Mol Ecol Resour. 2011;11(3):423–34. pmid:21481200
View Article
PubMed/NCBI
Google Scholar

[173] View Article

[174] PubMed/NCBI

[175] Google Scholar

Figures

Abstract

Introduction

Materials and Methods

Specimen collection

Genetic data collection

Summary statistics and haplotype network

Analyses of molecular variance

Bayesian clustering analysis

Historic population growth

Results

Sampling and Genetic Data

Gene genealogy

Analyses of molecular variance

Bayesian clustering analysis

Historic population growth

Discussion

Circumtropical patterns of population structure in an epipelagic specialist

Evidence for barriers to E. volitans gene flow

E. volitans historical demography

Supporting Information

S1 Table. Specimen sampling data for Exocoetus volitans specimens.

Acknowledgments

Author Contributions

References