Mechanisms of peripheral phylogeographic divergence in the indo-Pacific: lessons from the spiny lobster Panulirus homarus
© The Author(s). 2017
Received: 7 June 2017
Accepted: 11 August 2017
Published: 18 August 2017
There is increasing recognition of the concordance between marine biogeographic and phylogeographic boundaries. However, it is still unclear how population-level divergence translates into species-level divergence, and what are the principal factors that first initiate that divergence, and then maintain reproductive isolation. This study examines the likely forces driving population and lineage divergences in the broadly-distributed Indo-Pacific spiny lobster Panulirus homarus, which has peripheral divergent lineages in the west and east. The study focuses particularly on the West Indian Ocean, which is emerging as a region of unexpected diversity. Mitochondrial control region (mtCR) and COI sequences as well as genotypes of 9 microsatellite loci were examined in 410 individuals from 17 locations grouped into 7 regions from South Africa in the west, and eastward across to Taiwan and the Marquesas Islands. Phylogenetic and population-level analyses were used to test the significance and timing of divergences and describe the genetic relationships among populations.
Analyses of the mtCR revealed high levels of divergence among the seven regions (ФST = 0.594, P < 0.001). Microsatellite analyses also revealed significant divergence among regions, but at a much lower level (FST = 0.066, P < 0.001). The results reveal different patterns of mtCR v. nDNA divergence between the two distinct peripheral lineages: a subspecies in South Africa and Madagascar, and a phylogeographically diverged population in the Marquesas. The results also expose a number of other more fine-scale population divergences, particularly in the Indian Ocean.
The divergence of peripheral lineages in the west and east of the species’ range appear to have been initiated and maintained by very different processes. The pattern of mitochondrial and nuclear divergence of the western lineage, implicates processes of parapatric isolation, secondary contact and introgression, and suggests possible maintenance through adaptation and behavioural reproductive isolation. In contrast, the eastern lineage appears to have diverged through a rare colonisation event, maintained through long-term isolation, and matches expectations of the core-periphery hypothesis. The process of active peripheral speciation may be a common force in the Indo-Pacific that helps drive some of the regions’ recognized biogeographic boundaries.
KeywordsIndo-West Pacific Larval dispersal Marine biogeography mtDNA Microsatellites Panulirus Peripheral speciation Phylogeography West Indian Ocean
Our understanding of the origins of marine biodiversity has been greatly enhanced through investigation of the evolutionary history of biogeographic provinces. This is particularly so in the vast Indo-Pacific region, which harbours the bulk of marine diversity, and has long been of great biogeographic interest across a wide taxonomic range of species [1–5]. Marine biogeographic provinces are largely defined by distributions of species and/or endemism [6, 7]. However, there is increasing recognition of the concordance between marine biogeographic and phylogeographic boundaries [8, 9], and there have been a number of within-species genetic studies that have greatly expanded our understanding of some of the common evolutionary patterns in the Indo-Pacific [10, 11]. However, it is still unclear how population-level divergence translates into species-level divergence , and what the differing factors are that first initiate that divergence, and then maintain reproductive isolation [12, 13]. This is particularly puzzling in the marine environment, where many species have the ability to disperse widely through planktonic larvae .
Although there are common biogeographic and phylogeographic boundaries among Indo-Pacific species that hint at common causes of divergence, there are also many discrepancies in the patterns observed among species [15, 16]. These discrepancies suggest that a diversity of forces act on marine species in this region to initially create genetic and biogeographic boundaries. These may include not only the obvious restrictions to dispersal of drifting larvae by the prevailing ocean currents [14, 17], but also the impact on both currents and habitat availability by recent sea-level changes, or more distant geological events .
It is also unclear what the forces are that maintain reproductive isolation as diverged populations become species, and what is the relative importance of physical versus ecological boundaries in maintaining that divergence . That is, how is it that the dynamic processes of gene flow reduction and population subdivision within species may ultimately lead to biogeographic divergence among species. Although it may at first appear to be mediated in a simple way through allopatric speciation, the complex overlapping distributions of most marine taxa through the region show that the process is far from simple, and have led some to invoke the potential impact of adaptation to differing environments .
Another major gap in our understanding of Indo-Pacific biodiversity is the fine-scale pattern of population substructure across the region. The broad biogeographic patterns and provinces are now somewhat clear , but there is still much doubt about the locations of less obvious barriers to gene flow within each province, and if these too may be common among species . Such fine-scale questions are of particular importance for management of commercial or endangered species, and require much more intensive sampling than common in past studies across this vast region, especially within the less-studied Indian Ocean, which appears to harbour unrecognised levels of biodiversity [15, 22].
We investigate here the process of phylogeographic divergence in this species across biogeographic boundaries, to shed light on the evolutionary forces driving this. Using a greatly expanded data set that includes substantially increased sample sizes from a much wider diversity of locations, examined using two mitochondrial DNA genes and nine nuclear microsatellite loci, we address two groups of related questions: (1) What can multiple loci tell us about the processes driving divergence at the western and eastern peripheries? When did the divergences occur, have they both been driven and maintained by similar allopatric processes, do they match the predictions of the core-periphery hypothesis, and are they concordant with those of other species? (2) Do more fine-scale divergences exist between populations separated by other recognised biogeographic boundaries, the Indo-Pacific barrier (IPB), and the proposed mid-Indian Ocean barrier (MIOB ). Can we help clarify the nature and location of the MIOB .
Sampling, PCR and genotyping
Sample locations and sizes for each genetic marker type
India west (Kannur)
India south (Colachel)
India east (Chennai)
A total of 14 loci were amplified individually or multiplexed in 10 μl PCR reactions. Details of loci, and PCR and sequencing conditions are provided in Additional file 1 in Supporting Information. Each PCR and Genescan run included positive and negative controls to prevent cross contamination and to estimate the genotyping error rates. Null allele frequencies, stuttering, and large allele dropout were checked using Micro-Checker 2.2.3 .
Aligned and edited 618 bp sequence fragments of mtCR (using Geneious vR9 ), were used for further analysis. A median-joining haplotype network  was constructed using mtCR and COI sequences in Network with star contraction . After preliminary analysis, samples from adjacent locations with small sample size that were not significantly different genetically, and did not cross proposed biogeographic barriers, were pooled together for further regional analysis, but also compared with analyses from unpooled samples (Table 1). Gene (haplotype) and nucleotide (л) diversities, Tajima’s D  as well as Fu’s F  tests and analysis of molecular variance (AMOVA) were calculated using Arlequin version 3.5 . The most likely clusters of genetically similar individuals were estimated from mtCR data using a Bayesian approach in the program Baps v.6 with varying numbers of populations (k) .
Microsatellite alleles were sized in Geneious using the microsatellite plugin version 1.4.0 . Of the 14 loci initially genotyped, 9 were retained for further population analysis following quality control analysis, taking regard of reliability, interpretability, and fit of genotype distributions to null expectations. Measures of variability for each locus analysed are reported in Additional file 1: Table A1.2. Analysis of Molecular Variance (AMOVA), and pairwise FST comparisons between populations and regions were carried out with Arlequin 3.5 , Genepop  and GenAlEx . Principal coordinate analyses (PCoA) of pairwise genetic distances were undertaken in GenAlEx to display the genetic relationships among sampled populations. Genalex was used to estimate allele frequency, effective number of alleles, observed and expected heterozygosity and fixation index. Allelic richness and gene diversity was analysed in Fstat 2.9.3 . The number of private alleles in each region were adjusted for sample size with a rarefaction approach implemented in Adze . Spatial genetic discontinuities were determined using the Bayesian clustering algorithm Structure  either using or excluding a priori locations and groupings (see Additional file 1 for detailed methods). The Benjamini and Hochberg  method was used to adjust the family-wide false discovery rate in microsatellite and mtCR sequence data. Additional analysis parameters are provided in Additional file 1.
The ratio of male to female gene flow (mm/mf) between the two lineages was estimated from the differences between mtDNA ФST and nDNA FST, assuming neutral divergence ( eq. 7c).
The time to most recent common ancestor (TMRCA), was calculated from the COI data using the Bayesian MCMC approach implemented in Beast 2.4.2 . To enable as direct comparisons as possible with divergence dates calculated in Iacchei et al., , the same COI calculation parameters were used, including the 1.39% per lineage divergence rate in a strict clock model. Unfortunately, there are no confirmed divergence times of closely related species on which to base a confident prior. Thus, although there may remain some imprecision in the dates calculated, we can be confident that they are directly comparable with the estimated dates presented in the related studies we discuss. Specific parameters and further details are again provided in Additional file 1.
Gene flow between the P. h. rubellus and P. h. homarus lineages were estimated in the isolation-with-migration model using IMa2  on the mitochondrial and microsatellite data. Estimates were made independently from the mtDNA and nDNA data sets, as they differed substantially, using search parameters detailed in Additional file 1.
The mitochondrial control region was highly polymorphic with high haplotype diversity (mean 0.99) (Additional file 1: Table A1.1). Nucleotide diversity was moderate (mean 6.9%), and was significantly lower in the Marquesas (2.1%) (Additional file 1:Table A1.1). Tajima’s D values were negative for most of the locations (Additional file 1: Table A1.1), except for the sample from Marquesas Islands, although not significantly so. The Fu’s F values were also negative for most samples, including the Marquesas.
The mtCOI analysis of a representative subset of individuals enabled direct comparisons with other studies of P. homarus and other species [27, 53, 55]. The haplotype network (Additional file 1: Fig. A1.2) revealed very similar patterns of genetic variation across the distribution of P. homarus as seen in the mtCR, but with lower haplotype diversity. Further tests of population subdivision on the mtCOI data (not shown) revealed very similar patterns to those shown in the mtCR analyses presented below, but with less discriminatory power (overall ФST = 0.596, P < 0.001).
Overall, the AMOVA test of mtCR sequence variation showed a very high level of differentiation among all samples (ФST = 0.71, P < 0.00001). Hierarchical analysis of all 17 locations grouped into seven regions (areas over which no significant pairwise site differences occurred, and that did not cross proposed biogeographic barriers, Table 1) revealed high levels of differentiation among regions (ФRT = 0.59, P < 0.001), but little divergence among locations within each region (ФSR = 0.01, n.s.). Excluding the divergent P. h. rubellus and Marquesas clades also resulted in a significant AMOVA test among the remaining five “central” regions (ФST = 0.09, P < 0.001).
Pairwise regional differentiation. Above diagonal, mtCR differentiation (ФST). Below diagonal, microsatellite differentiation (FST)
A Beast analysis of divergence times between the mtCOI lineages, showed that the P. h. rubellus lineage diverged approximately 2.9 million years ago (Ma) (CI: 2.2–3.7), while the Marquesas lineage diverged around 1.25 Ma (CI: 0.9–1.6) (see Additional file 1: Fig. A1.5). In this analysis, the next most closely related species, P. ornatus, was estimated to have diverged around 6.8 Ma (CI: 5.4–8.2), which closely matches the previous estimate of approximately 7 Ma .
Of the nine most reliable microsatellite loci included in the final analyses, all loci exhibited relatively high diversity, with high numbers of alleles per locus (mean Na = 15.6) and high heterozygosity (mean He = 0.83) (Additional file 1: Table A1.2). Deviation from neutral expectations was observed only in the Marquesas population, where a positive inbreeding coefficient indicated fewer heterozygotes than expected (F = 0.115, P < 0.05 after correction for false discovery rate).
Overall, the AMOVA test of microsatellite variation showed a significant, but much lower level of differentiation among all samples compared to the mtCR (FST = 0.035, P < 0.001), even after correction for high heterozygosity (F’ST = 0.115, P < 0.001) . Hierarchical analysis of all 17 locations grouped into seven regions (see Table 1) revealed significant differentiation among regions (F’RT = 0.189, P < 0.001), but also significant differentiation among locations within each region (F’SR = 0.168, P < 0.001). Excluding the divergent P. h. rubellus and Marquesas clades also resulted in a significant AMOVA test among the remaining five “central” regions (F’RT = 0.097, P < 0.01).
Analyses of pairwise differences between collection regions revealed where the main microsatellite divergences lie within the species (Table 2). Again, the Marquesas region is highly significantly different from all other regions, but the S. African (P. h. rubellus) region is less significantly diverged. Somewhat surprisingly for a recognised distinct subspecies, P. h. rubellus does not exhibit any fixed allelic differences. In contrast to the mtCR results, of the remaining five “central” regions, only the E. Africa and NWIO regions are significantly different from the other regions (Table 2). Within regions, there were several location pairs that were significantly different from each other. In particular, Oman was significantly different from the Iranian samples (FST = 0.026 P < 0.001) (Additional file 1: Table A1.5). Other significant fine-scale differences among locations were not consistent among loci, and these are likely to be artefacts due to small sample size. Again, there was no significant difference between P. h. rubellus samples from S. Africa and Madagascar (FST = 0.004, P = 0.33).
The microsatellite pattern of relationships among regions is again well represented in the PCoA of pairwise FST values (Fig. 3b). Unlike the mtDNA analysis, this shows that the Marquesas is by far the most divergent region, with P. h. rubellus (from S. Africa and Madagascar) much less so in comparison. Of the five “central” regions, Indonesia and India are the most similar, with the NWIO the most divergent of these. The significantly different microsatellite geographic clusters are also highlighted by the Structure analysis, which shows the P. h. rubellus and Marquesas samples to be distinct (Additional file 1: Fig. A1.4b, with optimal k of 5). An analysis excluding the use of a priori locations and regions gave a very similar result.
A Mantel test on the microsatellite data showed an isolation by distance pattern (R2 = 0.826, P < 0.01), but this pattern was not significant for mtCR data (R2 = 0.24, P = 0.21), or if the samples from the Marquesas and P. h. rubellus populations were excluded (mtCR: R2 = 0.0002, P = 0.46; microsatellites: R2 = 0.0362, P = 0.31).
The ratio of male to female gene flow (mm/mf) between the lineages was estimated from the differences between mtDNA ФST and nDNA FST . Using FST gave a ratio of 54.2, while using the adjusted F’ST gave a ratio of 3.6. The same ratio calculated between the Marquesas and adjacent populations was 1.1 (0.4 using F’ST). The considerable difference in mtDNA and nDNA gene flow between the lineages P. h. rubellus and P. h. homarus was highlighted by Bayesian estimates from the isolation-with-migration model in IMa2. Maximum likelihood estimates of gene flow from mtDNA peak at zero (mo>1 (migration from P. h. homarus to P. h. rubellus) = 0.0003, 95%HPD 0.0–0.53; m1>0 (migration from P. h. rubellus to P. h. homarus) = 0.0003, 95%HPD 0.0–0.47), while estimates from nDNA were significantly greater than zero (mo>1 = 0.51, 0.11–0.59; m1>0 = 0.26, 0.16–0.58). (Additional file 1: Table A1.6).
A more extensive sampling of individuals, locations and genetic loci confirms our previous findings of high levels of diversity within and among regions of P. homarus [25, 36]. The analyses of mtCR and microsatellite variation provide further evidence that P. h. rubellus and the Marquesas population are strongly diverged from the remainder of the species, but low differentiation among centrally-located samples. The differences between the mtDNA and nDNA patterns of divergence of these peripheral populations enable a closer examination of the evolutionary processes driving and maintaining their divergence.
P. h. rubellus (south African) lineage
There is still strong evidence to support a distinct sub-species status for P. h. rubellus . This sub-species has a distinct morphological form, is substantially geographically isolated on the extreme of the P. homarus distribution, is characterized by belonging to a unique mtDNA lineage that is considerably diverged from the remainder of the species, and is distinct at the ITS-1 nuclear locus . However, the microsatellite results indicate that, surprisingly, there is considerably less nuclear than mtDNA divergence between these sub-species than under neutral expectations . They have very divergent mtDNA lineages (16% K2P for mtCR, compared with 4% divergence of the Marquesas lineage), and a substantial number of private microsatellite alleles, yet only small mean microsatellite allele frequency differences, resulting in significantly larger than expected differences between mtDNA ФST and nDNA FST. The P. h. rubellus lineage is less divergent than the Marquesas population in its microsatellite DNA, even though it is far more divergent in its mtDNA.
It is known that there is currently some overlap in the distribution of the distinct P. h. rubellus and P. h. homarus forms along the north-eastern coast of South Africa and Mozambique, with the homarus form decreasing in frequency southward . It is also known from previous work that some degree of hybridization occurs between the two forms, as a very small number of individuals with the rubellus morphological form have been found to carry the homarus mtDNA [25, 61]. Given the geographically highly demarcated boundary to the divergent mtDNA lineages, and the very limited sharing of mtDNA among morphological forms, it is clear that there are very low levels of mtDNA gene flow between lineages. In contrast, the microsatellite results suggest relatively high nDNA gene sharing between these lineages. There are several potential explanations for this phenomenon, and all of them may be relevant to the mechanisms that initiate and maintain marine species divergence and biogeographic boundaries in the Indo-Pacific.
Potentially, the discrepancy in divergence of mtDNA and nuclear microsatellite alleles could be due simply to stochastic lineage sorting with random fixation of mtDNA alleles and/or insufficient time for nuclear alleles to diverge in frequency [62–64]. However, this is unlikely here, given the ratio of mtDNA/nDNA divergence and estimated gene flow, and the extreme contrast in the divergence ratio between P. h. “brown” (in the Marquesas) and P. h. rubellus. This pattern is close to expectations for neutral divergence in P. h. “brown”, which has a mtDNA lineage 4% divergent, no private microsatellite alleles, a mtDNA ФST/nDNA FST ratio of 0.59/0.09 and an estimated ratio of male to female gene flow (mm/mf)  of 1.1. In contrast, this pattern is significantly divergent from neutral expectations in P. h. rubellus, which has a mtDNA lineage 16% divergent, and a high proportion of private microsatellite alleles, but the mtDNA ФST/nDNA FST ratio is 0.75/0.02 and the estimated ratio of male to female gene flow (mm/mf) is 54.2. This departure from neutral expectations is highlighted by the difference between mtDNA and nDNA estimates of gene flow from the isolation-with-migration model. It is unlikely that either neutral divergence with no gene flow, or high levels of current gene flow in both males and females would leave behind the pattern seen in P. h. rubellus.
Given the deep mtDNA divergence, it is more likely that the P. h. rubellus lineage diverged allopatrically (or parapatrically) at around 2.9 Ma, and it is only recently that the distributions of the two lineages have begun to overlap, leading to secondary contact and introgression [65, 66]. On initial allopatric divergence, it is not uncommon to see rapid divergence of a new mtDNA lineage, and slower divergence of nDNA alleles, depending on effective population size [59, 62]. Upon secondary contact, there are several possible factors that may have eroded the nDNA divergence more rapidly than the mtDNA divergence, but still maintained lineage divergence through a degree of reproductive isolation . Firstly, marked differences between the sexes in migration behaviour, especially related to breeding, is commonplace among spiny lobsters [68–70], although this seems unlikely to have sufficient impact on a broad scale. Alternatively, asymmetric developmental fitness of hybrids may exist, whereby crosses of one type (i.e., homarus males with rubellus females) are substantially less fit, resulting in asymmetric male gene flow but restricted female gene flow. The fact that only rubellus father/homarus mother hybrids have so far been reported may lend some support to this hypothesis . Another possibility is that of a frequency-dependent male preference in both subspecies for conspecific mating. This would result in fewer cross-species matings with the rare female subspecies. Each of these scenarios would result in greater male than female gene flow between lineages. A final possibility (discussed further below) is that of strong environmental selection on differentially adapted phenotypes. The distribution of P. h. rubellus extends the furthest south of any P. homarus population, across the recognised biogeographic boundary separating warm tropical from cooler subtropical waters along the south-westerly progress of the Agulhas current . If P. h. rubellus did diverge allopatrically in the past in this region, there is every likelihood over a substantial period of time that it would adapt to the cooler temperatures of its range. If so, the resulting differential selection along this coast on the two lineages could maintain their largely non-overlapping parapatric distributions, and the degree of reproductive isolation evident today. Further exploration of this potential adaptation is worth pursuing.
P. h. rubellus divergence likely originated with the advent of modern glacial cycles at the beginning of the Pleistocene ~2.5 Ma, when a relatively warm, stable interval of global climate transitioned to cooler, high-amplitude cycles, curtailing a warm-water corridor around southern Africa, and intermittently reducing connectivity along this coastline . There is evidence of local adaptation to cooler waters in some other species in this region (Nassarius snails , Hymenosoma crabs [74, 75]), and extensive evidence of ongoing biogeographic barriers that separate tropical, subtropical, warm-temperate and cool-temperate waters along this coast . The breakdown of physical separation between the two lineages likely arose through population expansion during one or more inter-glacial cycles.
Regardless of the exact mechanisms, the data are clear that P. h. rubellus likely diverged about 2.9 Ma (CI: 2.2–3.7), that there is a surprising discrepancy in mtDNA and nDNA divergence between these lineages, and that, despite apparently high recent nDNA gene flow, distinct lineages are able to persist in parapatry. This highlights that marine lineages may still be able to maintain divergence despite substantial gene flow, a matter of considerable interest in marine speciation and biogeography [19, 76].
Like P. h. rubellus, the Marquesas population (P. h. “brown”) is also a peripherally isolated unique mtDNA lineage, but the data suggest that very different processes have led to its origin and maintenance. Unlike P. h. rubellus, its mtDNA lineage is only slightly divergent, and appears to have diverged only around 1.25 Ma (CI: 0.9–1.6). Its mtDNA diversity represents just a small subset of the total diversity present in the neighbouring P. h. homarus populations. The nDNA pattern in this population is one of reduction in allelic diversity, but with no unique alleles (unlike P. h. rubellus). As such, it may best be regarded as an isolated and distinct population rather than a distinct subspecies, although it does appear to have a distinct morphology .
Since its colonisation around 1.25 Ma (CI: 0.9–1.6), likely through a rare period of favourable current flow from the West Pacific (see below), the Marquesas population appears to have diverged genetically through random drift in allele frequencies, with no unique nDNA genotypes. The population appears to have been completely marooned through much of this time, and is now isolated by a very large distance from the main population range of the species (i.e., >7000 km). (The geographically closest recorded specimen is from New Caledonia, >6000 km distant). The lowered levels of mtDNA and nDNA diversity, and positive value of Tajima’s D for the mtCR, all suggest that this population has a reduced effective size, although not sufficient to leave a significant signal of a bottleneck in its microsatellite allele frequencies. This genetic pattern is as predicted by the core-periphery hypothesis: lowered diversity in the periphery compared to central populations, as measured by mtDNA and nDNA allelic diversity and private alleles, and increased genetic divergence of the peripheral population .
It has been recognized previously that the Marquesas Islands display a high level of endemism and genetic distinction of its marine fauna [3, 16, 20, 77]. This has been linked to the formation of these islands and local retentive currents  and/or prevailing westward currents that now prevent gene flow toward the East Pacific .
Thus very different processes appear to have led to the divergence and maintenance of the western and eastern peripheral lineages of P. homarus. Further, the eastern periphery matches the expectations of the core-periphery hypothesis, while the western periphery does not.
Fine-scale population divergences
Examination of mtDNA in P. h. rubellus has previously indicated the Madagascan population of this lineage is genetically divergent from the South African population . However, we found no evidence of this distinct gene lineage or any significant divergence in microsatellite alleles in Madagascar. Our sample size from Madagascar may have been insufficient to detect an allele frequency difference, but it should have been possible to detect a distinct mtDNA lineage. Given the distinct lineage observed by Reddy et al.  was a COI-like nuclear copy (numt), unlike our COI sequences, it is possible that the divergence identified by this marker is not widespread throughout the mitochondrial or nDNA genomes.
We found strong evidence for reduced gene flow across the distribution of P. h. homarus, despite the long larval dispersal period (around 6 months). The NWIO region appears to be genetically isolated (at both mitochondrial and nuclear loci) from regions to the south (Kenya and Tanzania) and to the east (India and beyond). This unexpected degree of isolation observed in both marker types may be explained by the Arab Sea Gyre and east-west monsoon currents [28, 36] which likely act to limit larval transport and gene flow along the north-east African coast and Arabian Peninsula. Larval connectivity with populations further to the east around the Indian subcontinent may also be limited by the significant outflows of turbid fresh water in the vicinity of Karachi (such as the Indus). Similarly, the huge outflows from the Meghna River delta (Brahmaputra outflow) into the Bay of Bengal may help to explain the limited coastal larval connectivity to the east of the Indian subcontinent. Both these regions to the west and east of India appear to be breaks in the natural distribution of P. homarus, and may also reflect breaks in larval connectivity. These genetic patterns do not clearly locate the existence of a single mid-Indian Ocean barrier (MIOB), as has been seen in other species . The genetic divergence to the west of India is greater than that to the east, but there is clearly an isolation-by-distance effect along this coastline.
There appear to be few breaks in connectivity between the Indonesian (Aceh and Lombok samples) and East Asian (Vietnam and Taiwan samples) regions, at either mtDNA or nDNA markers. Although our sampling is less intense across this region, the continuous habitat, strong currents, and many stepping-stone islands suggest that connectivity is likely to be strong in this species across this region. Indeed, more intense sampling of this region in the closely related P. ornatus (with similar larval duration) has not detected any population differentiation in mtDNA or nDNA [79, 80], and broad sampling of P. penicillatus across this region also failed to find any marked genetic differentiation [53, 81].
Peripheral divergence and biogeographic implications
This study provides further evidence of the ability of marine Indo-Pacific populations to diverge genetically, despite a capacity for extensive planktonic larval dispersal. It is apparent that the levels of gene flow within the species are sufficient to maintain cohesion of the species, except at the peripheries of the distribution range. In the east, the Marquesas population appears to be an isolated remnant of past dispersal, which is now gradually diverging genetically and morphologically. In the west, the great mtDNA divergence and reduced nDNA divergence in the P. h. rubellus lineage strongly suggests past allopatric divergence of this lineage, with more recent secondary contact and introgression. The evidence that nDNA gene flow between the lineages is now much higher than mtDNA gene flow, suggests that a partial reproductive barrier has developed during their allopatric isolation, and that this barrier may act through a combination of male mating preferences, and ecological selection on differentially adapted phenotypes. Such a scenario may also be a common feature behind some of the marine biogeographic boundaries commonly observed across the Indo-Pacific.
A similar situation is seen in the closely related P. ornatus, where a very distinct mtDNA lineage has been found in the WIO . However, in that species the mtDNA barrier appears to have started breaking down, and this lineage has now spread much more widely eastwards at low frequency. It is possible that this species experienced a similar allopatric divergence history in the WIO as P. h. rubellus, but that a behavioural reproductive barrier did not develop in P. ornatus before secondary contact and introgression occurred. P. ornatus exhibits its greatest mtDNA lineage boundary further north than P. homarus (north of Kenya). Another related species, P. penicillatus, has its most divergent Indian Ocean mtDNA lineage in its NWIO population, specifically in the Red Sea [53, 82], which itself has now been recognised as being a common biogeographic province . The one factor in common among all these genetically divergent Panulirus populations is that they all lie at the extreme western periphery of their species’ distribution. Thus it seems that peripheral populations in the WIO may have a high probability of forming divergent lineages and perhaps species, even if there is potential for long-distance larval dispersal.
Recent studies in other diverse species have also highlighted the importance of the WIO as a location of unique marine lineages and biodiversity. A number of finfish species have been identified as having divergent populations in the WIO [15, 22, 83], and a similar pattern is now also being described in several invertebrate species in the same region [84–86]. Therefore, there appears to be considerable scope for phylogeographic and biogeographic boundaries to be present in the WIO, which has only recently been recognised . As further studies are undertaken in the WIO, we propose that additional common biogeographic barriers will be found, and it appears highly likely that the WIO in general will exhibit higher levels of population divergence than the Indo-Polynesian province to the east for widespread species. We also suggest that the potential mechanisms of population divergence described for P. homarus in this study may also apply more widely in this region.
Analysis of mtDNA divergence times in P. penicillatus revealed that the East Pacific lineage of that species diverged approximately 1.5 Ma (0.8–2.31), while the NWIO Red Sea lineage diverged only about 0.7 Ma (0.4–1.1) . Similar calculations on the mtDNA divergences in P. homarus show that the P. h. rubellus lineage likely diverged around 2.9 Ma (2.2–3.7), while the Marquesas sub-population diverged around 1.25 Ma (0.9–1.6). Identical calculations on the P. ornatus mtDNA data estimate that the East African lineage diverged around 0.66 Ma , and it does not appear to have a peripherally isolated eastern lineage. It is clear from these estimates that there is some, but not complete, overlap in time and place where genetic lineages diverged in these species with similar biology. Around 0.7 Ma, both P. ornatus and P. penicillatus had peripheral WIO populations that genetically diverged. Around 1.5 Ma, both P. penicillatus and P. homarus had peripheral Eastern Pacific populations that genetically diverged. The P. h. rubellus lineage is by far the oldest divergence among this group, and appears to pre-date these divergences by several million years, but coincides with the estimated date of most common divergence across the MIOB in a range of fish species .
These estimated dates of divergence point toward the possibility of common periods of phylogeographic divergence across the Indo-Pacific, which may help to identify the potential initial drivers of divergence. The dramatic impact of sea-level changes on both sea shelf habitat availability and fluctuations in prevailing currents have previously been highlighted, especially in the Indian Ocean, where coastal shelves are generally narrow . However, the divergences reported here fall well before the most recent sea-level changes (around 0.02 Ma), and thus it is very difficult to accurately align any of these divergences with any specific sea-level events. Intersecting with these long-term temporal changes are the complex seasonal changes in currents and upwelling in the WIO that have been identified [36, 86]. Cowman and Bellwood  have alerted us to the changes over time in biogeographic processes that have likely taken place in the Indo-Pacific; “accumulation”, followed by “survival”, “origin”, and then most recently a period of “survival and export” of species. All the Panulirus lineage divergences reported here fall into the most recent Pliocene period that they characterise as a period of “survival and export” of species. The present study confirms that, even within the tropical spiny lobsters with enormous larval dispersal potential, this has certainly been a period of “export” of unique genetic lineages at the peripheries of the Indo-Pacific.
The data presented here show that very different evolutionary processes have led to the initial divergence and long-term maintenance of the western and eastern peripheral lineages of P. homarus. The pattern of mitochondrial and nuclear divergence of the western lineage, implicates processes of parapatric isolation, secondary contact and introgression, and suggests possible maintenance through adaptation and behavioural reproductive isolation. In contrast, the eastern lineage appears to have diverged through a rare colonisation event, maintained through long-term isolation, and matches expectations of the core-periphery hypothesis. This may provide an explanation of some of the common evolutionary processes that translate phylogeographic divergence into biogeographic divergence, and continue to generate marine diversity throughout the Indo-Pacific.
The laboratory work for this study was carried out in the Molecular Ecology and Evolution Lab in the School of Biological Sciences at the University of Auckland. Sincere thanks are extended to the Iranian Fisheries Research Organization and the University of Tehran, Iran, Matthew Slater, John Kochey, Peter Fielding, Renee Albertson, Nguyen Thi Bich Ngoc and Tin Yam Chan for kindly providing us with samples. The authors would like to thank Vibhavari Thakur and Bhargavi Yellapu for laboratory assistance and sample collection. We also thank the University of Auckland Centre for Genomics and Proteomics for sequencing and genotyping services.
The study was conducted with the financial support from Nexus Aquasciences Sdn Bhd and the Australian Research Council’s Industrial Transformation Research Hub (Project number IH 120100032).
Availability of data and materials
Genotype data will be accessible through the Dryad repository, sequence data are accessible through NCBI Genbank and accession numbers are provided in manuscript.
AF, AJ, HF and SL conceived the ideas. AJ, AF, TSR collected samples, AJ, SL, GS, provided materials and laboratory equipment, AF produced the data, AF, SL analysed the data. AF, SL, AJ wrote the manuscript. All authors read and approved the final manuscript.
Ahmad Farhadi is a postdoctoral researcher interested in population genetics and phylogeography of marine taxa. The work was undertaken under the supervision of Shane Lavery, who studies the molecular ecology and evolution of a broad range of marine taxa, and Andrew Jeffs, who works on the ecology and aquaculture of spiny lobsters.
All animal sampling in this study complied with the Council of the European Communities Directive 86/609/EEC regarding the protection of animals used for experimental and other scientific purposes, and fully complied with local fisheries management and marine protected area controls. Dead reptant lobsters were purchased or provided by local commercial fishers and sampled and therefore no specific permits were required for the described field sampling as the fishers were required to comply with local laws regarding capture. The species sampled are not endangered or protected.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Briggs JC. Marine zoogeography. New York: McGraw-Hill; 1974.Google Scholar
- Briggs JC. The marine east indies: diversity and speciation. J Biogeogr. 2005;32(9):1517–22.Google Scholar
- Briggs JC, Bowen BW. A realignment of marine biogeographic provinces with particular reference to fish distributions. J Biogeogr. 2012;39(1):12–30.Google Scholar
- Ekman S. Zoogeography of the sea. London: Sidgwick and Jackson; 1953.Google Scholar
- Kulbicki M, Parravicini V, Bellwood DR, Arias-Gonzalez E, Chabanet P, Floeter SR, Friedlander A, McPherson J, Myers RE, Vigliola L, et al. Global biogeography of reef fishes: a hierarchical quantitative delineation of regions. PLoS One. 2013;8(12):e81847.PubMedPubMed CentralGoogle Scholar
- Reaka ML, Rodgers PJ, Kudla AU. Patterns of biodiversity and endemism on indo-West Pacific coral reefs. PNAS. 2008;105:11474–81.PubMedPubMed CentralGoogle Scholar
- Springer VG. Pacific plate biogeography, with special reference to shorefishes. Smithson Contribut Zool. 1982;367:1–182.Google Scholar
- Bowen BW, Gaither MR, DiBattista JD, Iacchei M, Andrews KR, Grant WS, Toonen RJ, Briggs JC. Comparative phylogeography of the ocean planet. PNAS. 2016;113(29):7962–9.PubMedPubMed CentralGoogle Scholar
- Cowman PF, Parravicini V, Kulbicki M, Floeter SR. The biogeography of tropical reef fishes: endemism and provinciality through time. Biol Rev. 2017:n/a-n/a.Google Scholar
- Gaither MR, Rocha LA. Origins of species richness in the indo-Malay-Philippine biodiversity hotspot: evidence for the centre of overlap hypothesis. J Biogeogr. 2013;40(9):1638–48.Google Scholar
- Keyse J, Crandall ED, Toonen RJ, Meyer CP, Treml EA, Riginos C. The scope of published population genetic data for indo-Pacific marine fauna and future research opportunities in the region. B Mar Sci. 2014;90(1):47–78.Google Scholar
- Palumbi SR. Genetic divergence, reproductive isolation, and marine speciation. Annu Rev Ecol Syst. 1994;25:547–72.Google Scholar
- Rocha LA, Bowen BW. Speciation in coral-reef fishes. J Fish Biol. 2008;72(5):1101–21.Google Scholar
- Cowen RK, Sponaugle S. Larval dispersal and marine population connectivity. Annu Rev Mar Sci. 2009;1:443–66.Google Scholar
- Borsa P, Durand JD, Chen WJ, Hubert N, Muths D, Mou-Tham G, Kulbicki M. Comparative phylogeography of the western Indian Ocean reef fauna. Acta Oecol-Int J Ecol. 2016;72:72–86.Google Scholar
- Bowen BW, Shanker K, Yasuda N, Malay MCMD. Heyden Svd, Paulay G, Rocha LA, Selkoe KA, barber PH, Williams ST et al: Phylogeography unplugged: comparative surveys in the genomic era. B Mar Sci. 2014;90(1):13–46.Google Scholar
- Hellberg ME, Burton RS, Neigel JE, Palumbi SR. Genetic assessment of connectivity among marine populations. B Mar Sci. 2002;70(1):273–90.Google Scholar
- Ludt WB, Rocha LA. Shifting seas: the impacts of Pleistocene sea-level fluctuations on the evolution of tropical marine taxa. J Biogeogr. 2015;42(1):25–38.Google Scholar
- Bowen BW, Rocha LA, Toonen RJ, Karl SA, ToBo L. The origins of tropical marine biodiversity. Trends Ecol Evolut. 2013;28(6):359–66.Google Scholar
- Gaither MR, Bernal MA, Coleman RR, Bowen BW, Jones SA, Simison WB, Rocha LA. Genomic signatures of geographic isolation and natural selection in coral reef fishes. Mol Ecol. 2015;24(7):1543–57.PubMedGoogle Scholar
- 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). J Biogeogr. 2010;37(1):133–47.Google Scholar
- Hodge JR, Bellwood DR. The geography of speciation in coral reef fishes: the relative importance of biogeographical barriers in separating sister-species. J Biogeogr. 2016;43(7):1324–35.Google Scholar
- Holthuis L. FAO species catalogue. Marine lobsters of the world: an annotated and illustrated catalogue of species of interest to fisheries known to date, vol. vol. 13. Rome: FAO; 1991.Google Scholar
- Phillips BF, Booth JD, Cobb JS, Jeffs AG, McWilliam P. Chapter 7. Larval and postlarval ecology. In: Phillips BF, editor. Lobsters: biology, management, aquaculture and fisheries. Oxford: Blackwell; 2007. p. 231–62.Google Scholar
- Lavery SD, Farhadi A, Farahmand H, Chan T-Y, Azhdehakoshpour A, Thakur V, Jeffs AG. Evolutionary divergence of geographic subspecies within the scalloped spiny lobster Panulirus homarus (Linnaeus 1758). PLoS One. 2014;9(6):e97247.PubMedPubMed CentralGoogle Scholar
- Berry PF. A revision of the Panulirus homarus group of spiny lobsters (Decapoda, Palinuridae). Crustaceana. 1974;27(1):31–44.Google Scholar
- Ptacek MB, Sarver SK, Childress MJ, Herrnkind WF. Molecular phylogeny of the spiny lobster genus Panulirus (Decapoda : Palinuridae). Mar Freshw Res. 2001;52(8):1037–47.Google Scholar
- George RW. Tethys Sea fragmentation and speciation of Panulirus spiny lobsters. Crustaceana. 2005;78:1281–309.Google Scholar
- Hodge JR, Read CI, van Herwerden L, Bellwood DR. The role of peripheral endemism in species diversification: evidence from the coral reef fish genus Anampses (family: Labridae). Mol Phylogenet Evol. 2012;62(2):653–63.PubMedGoogle Scholar
- Puckridge M, Last PR, Andreakis N. The role of peripheral endemism and habitat associations in the evolution of the indo-West Pacific tuskfishes (Labridae: Choerodon). Mol Phylogenet Evol. 2015;84:64–72.PubMedGoogle Scholar
- Eckert CG, Samis KE, Lougheed SC. Genetic variation across species' geographical ranges: the central-marginal hypothesis and beyond. Mol Ecol. 2008;17(5):1170–88.PubMedGoogle Scholar
- Cahill AE, Levinton JS. Genetic differentiation and reduced genetic diversity at the northern range edge of two species with different dispersal modes. Mol Ecol. 2016;25(2):515–26.PubMedGoogle Scholar
- Lal MM, Southgate PC, Jerry DR, Bosserelle C, Zenger KR. Swept away: ocean currents and seascape features influence genetic structure across the 18,000 km indo-Pacific distribution of a marine invertebrate, the black-lip pearl oyster Pinctada margaritifera. BMC Genomics. 2017;18(1):66.PubMedPubMed CentralGoogle Scholar
- Liggins L, Gleeson L, Riginos C. Evaluating edge-of-range genetic patterns for tropical echinoderms, Acanthaster planci and Tripneustes gratilla, of the Kermadec Islands, southwest Pacific. B Mar Sci. 2014;90(1):379–97.Google Scholar
- Cowman PF, Bellwood DR. Vicariance across major marine biogeographic barriers: temporal concordance and the relative intensity of hard versus soft barriers. Proc R Soc B. 2013;280(1768):20131541.PubMedPubMed CentralGoogle Scholar
- Farhadi A, Farhamand H, Nematollahi MA, Jeffs A, Lavery SD. Mitochondrial DNA population structure of the scalloped lobster Panulirus homarus (Linnaeus 1758) from the West Indian Ocean. ICES J Mar Sci. 2013;70(7):1491–8.Google Scholar
- Van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P. MICRO-CHECKER: software for identifying and correcting genotyping errors in microsatellite data. Mol Ecol Resour. 2004;4(3):535–8.Google Scholar
- Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, et al. Geneious basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics. 2012;28(12):1647–9.PubMedPubMed CentralGoogle Scholar
- Bandelt HJ, Forster P, Rohl A. Median-joining networks for inferring intraspecific phylogenies. Mol Biol Evol. 1999;16(1):37–48.PubMedGoogle Scholar
- Forster P, Torroni A, Renfrew C, Rohl A. Phylogenetic star contraction applied to Asian and Papuan mtDNA evolution. Mol Biol Evol. 2001;18(10):1864–81.PubMedGoogle Scholar
- Tajima F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics. 1989;123:585–95.PubMedPubMed CentralGoogle Scholar
- Fu YX. Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics. 1997;147(2):915–25.PubMedPubMed CentralGoogle Scholar
- Excoffier L, Lischer HEL. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and windows. Mol Ecol Resor. 2010;10(3):564–7.Google Scholar
- Cheng L, Connor T, Siren J, Aanensen D, Corander J. Hierarchical and spatially explicit clustering of DNA sequences with BAPS software. Mol Biol Evol. 2013;30:1224–8.PubMedPubMed CentralGoogle Scholar
- Raymond M, Rousset F. Genepop (version 1.2) population genetics software for exact tests and ecumenicism. J Hered. 1995;86(3):248–9.Google Scholar
- Peakall R, Smouse PE. GenAlEx 6.5: genetic analysis in excel. Population genetic software for teaching and research-an update. Bioinformatics. 2012;28(19):2537–9.PubMedPubMed CentralGoogle Scholar
- Goudet J. FSTAT (version 2.9.3): a computer program to calculate F- statistics. J Hered. 2013;104:586–90.Google Scholar
- Szpiech ZA, Jakobsson M, Rosenberg NA. ADZE: a rarefaction approach for counting alleles private to combinations of populations. Bioinformatics. 2008;24(21):2498–504.PubMedPubMed CentralGoogle Scholar
- Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155(2):945–59.PubMedPubMed CentralGoogle Scholar
- Benjamini Y, Hochberg Y. Controlling the false discovery rate - a practical and powerful approach to multiple testing. J R Stat Soc Series B Methodol. 1995;57(1):289–300.Google Scholar
- Hedrick PW, Allendorf FW, Baker CS. Estimation of fmale gene low from measures of nuclear and female genetic differentiation. J Hered. 2013;104(5):713–7.PubMedGoogle Scholar
- Bouckaert R, Heled J, Kühnert D, Vaughan T, Wu C-H, Xie D, Suchard MA, Rambaut A, Drummond AJ. BEAST 2: a software platform for bayesian evolutionary analysis. PLoS Comput Biol. 2014;10(4):e1003537.PubMedPubMed CentralGoogle Scholar
- Iacchei M, Gaither MR, Bowen BW, Toonen RJ. Testing dispersal limits in the sea: range-wide phylogeography of the pronghorn spiny lobster Panulirus penicillatus. J Biogeogr. 2016;43(5):1032–44.Google Scholar
- Hey J. Isolation with migration models for more than two populations. Mol Biol Evol. 2010;27(4):905–20.PubMedGoogle Scholar
- Senevirathna JDM, Munasinghe DHN, Mather PB. Assessment of genetic structure in wild populations of Panulirus homarus (Linnaeus, 1758) across the south coast of Sri Lanka snferred from mitochondrial DNA sequences. Int J Mar Sci. 2016;6(6):1–9.Google Scholar
- Tourinho J, Solé-Cava A, Lazoski C. Cryptic species within the commercially most important lobster in the tropical Atlantic, the spiny lobster Panulirus argus. Mar Biol. 2012;159:1897–906.Google Scholar
- Hedrick PW. A standardized genetic differentiation measure. Evolution. 2005;59(8):1633–8.PubMedGoogle Scholar
- Cornuet JM, Luikart G. Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genetics. 1996;144:2001–14.PubMedPubMed CentralGoogle Scholar
- Larsson LC, Charlier J, Laikre L, Ryman N. Statistical power for detecting genetic divergence - organelle versus nuclear markers. Cons Genet. 2009;10(5):1255–64.Google Scholar
- Berry P. The biology of the spiny lobster Panulirus homarus (Linnaeus) off the east coast of southern Africa, report 28. Durban: Oceanographic Research Institute; 1971.Google Scholar
- Reddy Maggie M, Macdonald Angus HH, Groeneveld Johan C, Schleyer Michael H. Phylogeography of the scalloped spiny-lobster Panulirus homarus rubellus in the southwest Indian Ocean. J Crust Biol. 2014;34(6):773–81.Google Scholar
- Avise JC. Phylogeography: The history and formation of species. Cambridge: Harvard University Press; 2000.Google Scholar
- Hare MP, Cipriano F, Palumbi SR. Genetic evidence on the demography of speciation in allopatric dolphin species. Evolution. 2002;56(4):804–16.PubMedGoogle Scholar
- Marko PB, Hart MW. Retrospective coalescent methods and the reconstruction of metapopulation histories in the sea. Evol Ecol. 2012;26(2):291–315.Google Scholar
- Tillett BJ, Meekan MG, Broderick D, Field IC, Cliff G, Ovenden JR. Pleistocene isolation, secondary introgression and restricted contemporary gene flow in the pig-eye shark, Carcharhinus amboinensis across northern Australia. Conserv Genet. 2012;13(1):99–115.Google Scholar
- Reid K, Hoareau TB, Graves JE, Potts WM, dos Santos SMR, Klopper AW, Bloomer P. Secondary contact and asymmetrical gene flow in a cosmopolitan marine fish across the Benguela upwelling zone. Heredity. 2016;117(5):307–15.PubMedGoogle Scholar
- Servedio MR. Geography, assortative mating, and the effects of sexual selection on speciation with gene flow. Evol Appl. 2016;9(1):91–102.PubMedGoogle Scholar
- Gardner C, Frusher S, Haddon M, Buxton C. Movements of the southern rock lobster Jasus edwardshi in Tasmania, Australia. B Mar Sci. 2003;73(3):653–71.Google Scholar
- Kelly S. Temporal variation in the movement of the spiny lobster Jasus edwardsii. Mar Freshw Res. 2001;52(3):323–31.Google Scholar
- Santos J, Rouillard D, Groeneveld JC. Advection-diffusion models of spiny lobster Palinurus gilchristi migrations for use in spatial fisheries management. Mar Ecol Prog Ser. 2014;498:227–41.Google Scholar
- Teske PR, Winker H, McQuaid CD, Barker NP. A tropical/subtropical biogeographic disjunction in southeastern Africa separates two evolutionarily significant units of an estuarine prawn. Mar Biol. 2009;156(6):1265–75.Google Scholar
- Dwyer GS, Cronin TM, Baker PA, Raymo ME, Buzas JS, Correge T. North-Atlantic deep-water temperature-change during late pliocene and late quaternary climatic cycles. Science. 1995;270(5240):1347–51.Google Scholar
- Teske PR, Papadopoulos I, McQuaid CD, Newman BK, Barker NP. Climate change, genetics or human choice: why were the shells of mankind's earliest ornament larger in the pleistocene than in the holocene? PLoS One. 2007;2(7):e614.PubMedPubMed CentralGoogle Scholar
- Teske PR, McLay CL, Sandoval-Castillo J, Papadopoulos I, Newman BK, Griffiths CL, McQuaid CD, Barker NP, Borgonie G, Beheregaray LB. Tri-locus sequence data reject a "Gondwanan origin hypothesis" for the African/South Pacific crab genus Hymenosoma. Mol Phylogenet Evol. 2009;53(1):23–33.PubMedGoogle Scholar
- Teske PR, von der Heyden S, McQuaid CD, Barker NP. A review of marine phylogeography in southern Africa. S Afr J Sci. 2011;107(5–6):43–53.Google Scholar
- Briggs JC, Bowen BW. Marine shelf habitat: biogeography and evolution. J Biogeogr. 2013;40(6):1023–35.Google Scholar
- Delrieu-Trottin E. Al. E: shore fishes of the Marquesas Islands: an updated checklist with new records and new percentage of endemic species. Check List. 2015;11(5):1–13.Google Scholar
- Bonjean F, Lagerloef G. Diagnostic model and analysis of the surface currents in the tropical Pacific Ocean. J Phys Oceanogr. 2002;32:2938–54.Google Scholar
- Dao HT, Smith-Keune C, Wolanski E, Jones CM, Jerry DR. Oceanographic currents and local ecological knowledge indicate, and genetics does not refute, a contemporary pattern of larval dispersal for the ornate spiny lobster, Panulirus ornatus in the south-east Asian archipelago. PLoS One. 2015;10(5):e0124568.PubMedPubMed CentralGoogle Scholar
- Yellapu B, Jeffs A, Battaglene S, Lavery SD. Population subdivision in the tropical spiny lobster Panulirus ornatus throughout its indo-West Pacific distribution. ICES J Mar Sci. 2016;74(3):759–68.Google Scholar
- Abdullah MF, Chow S, Sakai M, Cheng JH, Imai H. Genetic giversity and population structure of pronghorn spiny lobster Panulirus penicillatus in the pacific region. Pac Sci. 2014a;68(2):197–211.Google Scholar
- Abdullah MF. Alimuddin, Muththalib M, Salama AJ, Imai H: genetic isolation among the northwestern, southwestern and central-eastern Indian Ocean populations of the pronghorn spiny lobster Panulirus penicillatus. Int J Mol Sci. 2014b;15(6):9242–54.PubMedPubMed CentralGoogle Scholar
- Muths D, Tessier E, Bourjea J. Genetic structure of the reef grouper Epinephelus merra in the West Indian Ocean appears congruent with biogeographic and oceanographic boundaries. Mar Ecol. 2015;36(3):447–61.Google Scholar
- Hoareau TB, Boissin E, Paulay G, Bruggemann JH. The southwestern Indian Ocean as a potential marine evolutionary hotspot: perspectives from comparative phylogeography of reef brittle-stars. J Biogeogr. 2013;40(11):2167–79.Google Scholar
- van der Ven RM, Triest L, De Ryck DJR, Mwaura JM, Mohammed MS, Kochzius M. Population genetic structure of the stony coral Acropora tenuis shows high but variable connectivity in East Africa. J Biogeogr. 2016;43(3):510–9.Google Scholar
- Vogler C, Benzie J, Barber PH, Erdmann MV. Ambariyanto, Sheppard C, Tenggardjaja K, Gerard K, Worheide G: Phylogeography of the crown-of-thorns starfish in the Indian Ocean. PLoS One. 2012;7(8):e43499.PubMedPubMed CentralGoogle Scholar
- Cowman PF, Bellwood DR. The historical biogeography of coral reef fishes: global patterns of origination and dispersal. J Biogeogr. 2013;40(2):209–24.Google Scholar