Ototyphlonemertes is a cosmopolitan genus of meiofaunal nemerteans. Their morphological characters are insufficient to reliably identify and delimit species. Consequently, some of the species are considered cosmopolitan despite anticipated low dispersion capability of the adults and a short planktonic larval phase. Indeed, recent studies show that some species actually comprise cryptic species, and populations are connected by stochastic events of long-distance dispersion. Based solely on morphological traits, a Lactea and a Pallida morph of Ototyphlonemertes are recognized here from collections at eight and five locations respectively along the Chilean coast. To assess the phylogeographic patterns of their populations, two mitochondrial markers (COI and COX3) of 162 specimens of Lactea and 25 of Pallida were sequenced. Final sequences are 605bp and 362bp for COI and COX3, respectively. Results from phylogenetic and haplotype network analyses suggest that the Lactea morph comprises up to three independent evolutionary units (one with only COX3 sequences). A COI gene tree including other previously published Ototyphlonemertes sequences groups the Chilean Lactea with other Lactea, while the Chilean Pallida is grouped with other Pallida. Different structuring and gene flow patterns found for the four groups support the hypothesis that these are four independent evolutionary entities with different ecological, dispersal and demographical characteristics.
Citation: Mendes CB, Norenburg JL, Solferini VN, Andrade SCS (2018) Hidden diversity: Phylogeography of genus Ototyphlonemertes Diesing, 1863 (Ototyphlonemertidae: Hoplonemertea) reveals cryptic species and high diversity in Chilean populations. PLoS ONE 13(4): e0195833. https://doi.org/10.1371/journal.pone.0195833
Editor: Tzen-Yuh Chiang, National Cheng Kung University, TAIWAN
Received: January 4, 2018; Accepted: April 1, 2018; Published: April 26, 2018
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: All DNA sequence files are available from the GenBank database (accession numbers MG 926401 – MG926542 for COI sequences and MG987308 – MG987418 for COX3 sequences).
Funding: This work was funded by the following institutions: National Science Foundation Grant DEB-9712463 to JLN, funding the fieldwork and lab work; Fundação de Amparo à Pesquisa do Estado de São Paulo (http://www.fapesp.br), through the grant N° 2016/20005-5, funding CBM’s PhD scholarship; and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (http://www.capes.gov.br), through the grant N° 1629710, funding the first months of CBM’s PhD scholarship. 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.
The meiofauna comprises a huge variety of animals representing most of the invertebrate phyla. These animals can pass through a 500-μm mesh and, due to their taxonomic and ecological complexity, represent a very important component of marine ecosystems [1, 2, 3].
Unlike sessile organisms, which can actively disperse only during the larval stages, meiofaunal animals usually lack larval stages or larvae stay in the plankton for short periods , but adult dispersion can play a major role in connecting populations. For instance, harpacticoid copepods actually swim into the water column when the water velocity is slower [5, 6, 7]. Other animals, however, have body structures, such as the adhesive glands found in many marine meiofaunal worms, for attaching to sediment particles to avoid transport [8, 9, 10].
Most meiofaunal animals are inferred to have low dispersal ability; however, many of these species are widely distributed, and some are even considered cosmopolitan [11, 12, 13, 14]. Recent studies with these taxa show most of them to comprise morphologically cryptic species, probably isolated by the distance between populations (e.g., [15, 16, 17, 18, 14, 19, 20]), but some of those still have ranges of more than 5,000 km, likely due to adult transport with or without sediment grains in stochastic events [14, 20].
The genus Ototyphlonemertes
Ototyphlonemertes Diesing, 1863 is a cosmopolitan genus of interstitial nemerteans. They are unique among nemerteans (with the exception of two unconfirmed reports) in possessing a pair of cerebral statocysts. They are stenoecius and most commonly live in littoral or shallow sublittoral moderately sorted coarse sand and/or shell-hash. The adults avoid being swept away by waves by actively migrating deeper into the sediment [21, 22] and, in some species, using adhesive glands to attach to the sand grains .
Ototyphlonemertes, like many nemerteans, have a body with a simple morphology and no appendages, making the taxonomic delimitation of species difficult. Most of the species in the genus are identified based on some or all of the following morphology: characteristics of the statocysts; presence/absence of cerebral organs; sensory bristles and adhesive structures; as well as, especially, localization and size of the proboscis and characteristics of its components such as the stylet, accessory stylet pouches, papillae, and proboscis regions. Though there is the overall appearance of miniaturization and some simplification, as seen in other interstitial meiofauna, in the aforementioned features Ototyphlonemertes has significantly greater morphological disparity across the genus than other speciose nemertean genera (JLN pers obs). Unfortunately, some of those characteristics can be variable within even a genetically delimited species . In addition, many species, especially those named early in history, lack unambiguous morphological diagnoses. Several of those species are recorded worldwide. To facilitate identification, Envall and Norenburg , while recognizing the potential for non-monophyly, established six morphotypes with unique characteristics: Pallida, Fila, Lactea, Cirrula, Duplex and Macintoshi. For the most part these are unambiguous morphotypes but, unsurprisingly, two of these have been found to be paraphyletic or polyphyletic in recent genetic study .
Recent studies assessing the genetic structure of these species showed that many have a smaller distribution than was inferred by morphological studies [20, 23, 24]. Some species, notwithstanding, can have long-distance dispersal, as, for instance, Ototyphlonemertes parmula Corrêa, 1950 . Three studies covering five Ototyphlonemertes morphotypes found some lineages geographically restricted, while others showed evidence of long-distance gene flow [17, 18, 20]. A lineage of the Parmula morph present in both the Caribbean and Atlantic coasts of Florida, showed long-distance dispersal of as much as 2,000 km as well as sympatry of cryptic putative sister lineages, which could have been caused via dispersal of adults by storms and/or sediment transport . Some of the morphotypes along the Brazilian coast appear to be species complexes, of which some showed evidence of long-distance dispersal . The presence of sympatric cryptic species indicates that the major isolation force might be environmental change causing local extinction, rather than the geographic distance between populations .
Chilean coastal waters
The Chilean coast is more than 4,000 km long and has a complex system of ocean currents. The marine populations present along this coast have different patterns of connection and dispersion. Many coastal species seem to have gene flow restricted by a biogeographical transition zone around 30° S latitude, dividing the coast into North and South (e.g., [25, 26,27, 28]). This barrier is manifested as changes in wind patterns, upwelling regimes and associated variability in sea surface temperature [29, 30, 31].
Genetic analysis of the Chilean populations of the isopod Excirolana braziliensis Richardson, 1912 found three geographically separated groups related to its distribution, with strong evidence they are three different cryptic species . A similar pattern is found for populations of the barnacle Notochthamalus scabrosus (Darwin, 1854), which has a restricted gene flow between populations from North and South of the barrier. The confamilial species Jehlius cirratus (Darwin, 1854), however, has only a marginal signal for isolation by distance . This different pattern among similar barnacles might be a result of different larval behavior in the water column [32, 33, 25].
Our main goal is to understand how meiofaunal populations are connected along the Chilean Coast and if the biogeographical break plays a role in the genetic variability of meiofauna individuals. To address this, we analyze aspects of the phylogeography and population genetics of two morphotypes of Ototyphlonemertes found at eight localities along the Chilean coast: an Ototyphlonemertes Lactea morph and an Ototyphlonemertes Pallida morph. This is the first population genetics assessment for the Chilean meiofauna.
Materials and methods
Sampling and DNA sequence
One hundred and sixty-two individuals of Lactea and 25 individuals of Pallida were collected from sand samples in the intertidal zone following Corrêa’s  protocol. The animals were identified morphologically following Envall and Norenburg  by: (1) color of epidermis and brain region; (2) presence and position of the cephalic furrow; (3) presence and shape of cerebral organs; (4) proboscis papillae; (5) diaphragm size; (6) position and number of the accessory stylet pouches; (7) stylet structure; (8) shape of middle chamber; and (9) statolith type. The worms were collected in 2006, from eight sites along the Chilean coast: Chipana, Bellavista, Pozo Toyo, Huayquique (Tarapacá region), Totoralillo (Coquimbo region), Isla Negra, Las Cruces (Valparaíso region) and Dichato (Biobío region), with Lactea found at all eight and Pallida found at five of these sites (Fig 1 and Table 1). Living individuals were identified in the field utilizing a stereo and a compound microscope and preserved in ethanol 95% and, subsequent to field work, kept at -20°C until the DNA extraction. No permits were required for collecting meiofauna in Chile in 2006. All specimens were collected and transported in accordance with laws and permits.
BE, Bellavista; HQ, Huaiquique; PT, Pozo Toyo; CH, Chipana; TO, Totoralillo; IN, Isla Negra; LC, Las Cruces; DC, Dichato.
Genomic DNA was extracted using the CTAB method described by Thollesson . Partial regions of mitochondrial cytochrome c oxidase subunits 1 and 3 genes (hereafter COI and COX3) were amplified using the primer pairs LCO1490 ⁄ HCO2198  for COI and COX3F/COX3R for COX3 (50-TGCGWTGAGGWATAATTTTATTTATT-30 and 50-ACCAAGCAGC TGCTTCAAAACCAAA-30, respectively) . Polymerase chain reactions (PCR) were performed as follows: initial denaturation at 94°C for 1 min; 31 cycles of denaturation at 94°C for 30 s; annealing at 43°–48°C for 1 min, extension at 72°C for 1 min; and final extension at 72°C for 3 min. PCR products were purified with ExoSAP-IT (USB). Sequencing reactions for both strands of amplified markers were run using BigDye Terminator Cycle Sequencing Kit v3.1 (Applied Biosystems) and the amplification primers. Products were cleaned using Sephadex columns, dried, and sequenced using an Applied Biosystems automated sequencer. The products were sequenced using CEQ dye terminator chemistry and a CEQ 8000 Genetic Analysis System (Beckman Coulter, Brea, California). The sequence contigs were assembled using Sequencher, version 4.5 (Gene Codes, Ann Arbor, MI). Blast searches  were done on the NCBI website to check for possible contaminations. Sequences suspected of contamination or errors were removed. The analyses were performed with 128 and 100 Lactea individuals, and 19 and 9 Pallida individuals for COI and COX3, respectively (S1 Table). The final sequences are deposited in GenBank under the accession numbers MG 926401 –MG926542 for COI sequences and MG987308 –MG987418 for COX3 sequences.
Sequences were aligned using MAFFT L-INS-i v.7 . Phylogenetic trees were estimated with each morphotype partitioned for each marker in RAxML v8.2.9  using the general time reversible model with gamma distribution (GTR+GAMMA). In order to evaluate the species relationships, phylogenetic trees were estimated with both markers concatenated and, because some specimens did not have sequences available for both markers, with each marker separately. This was a way to not lose information about any population. Sequences from Nemertopsis tetraclitophila Gibson, 1990 (GenBank accession number KF572482.1) were included as outgroup in the analysis for COI and COX3. In addition, a tree with COI sequences from this study and from Leasi et al.  (GenBank accession number KM083821.1—KM083889.1, KT730596.1—KT730625.1, KT722707.1—KT722730.1, KT730626.1—KT730678.1, KT736303.1—KT736314.1, KU230031.1—KU230131.1, KU230132.1—KU230244.1, KU230245.1—KU230291.1) was also estimated to evaluate the relationship between the Chilean species and other species of Ototyphlonemertes. A sequence from Nipponemertes sp. (GenBank accession number KU230295.1) was included as outgroup for this tree. All trees were estimated with 5,000 bootstraps replicates.
The resulting trees from RAxML analyses of single and concatenated markers were used as input to estimate species delimitation using the Bayesian implementation of the Poisson Tree Processes (bPTP software, available at the web-service http://species.h-its.org/)  with 5,000 MCMC generations, 100 thinnings and a burn-in of 0.1. Since bPTP analysis from COX3 did not converge, we also estimate delimitation using the General Mixed Yule Coalescent model (GMYC) . Therefore, the sequences from COX3 were used to build an ultrametric tree in BEAST software v. 1.8.3  with 600,000 MCMC generations, 100 thinnings and a burn-in of 6,000. This tree was used as input to GMYC. The COI sequences also were used in the web server of Automatic Barcode Gap Discovery for primary species delimitation software (ABGD, http://wwwabi.snv.jussieu.fr/public/abgd/abgdweb.html)  with default conditions, to complement the species delimitation analyses. Haplotype networks of both genes singly and concatenated were assessed using statistical parsimony , as implemented in the program TCS v1.21 . The program connects haplotypes with the smallest number of differences as defined by 95% confidence criterion or the user can set the maximum number of accepted differences for connection between haplotypes. Here we set the maximum number of differences to 350, due to the highly polymorphic sequences. Results from the species delimitation analyses and TCS were compared to delimit possible cryptic species. All analyses performed after the species delimitation were in accord with these results. Therefore, the populations considered here are the ones delimited in such analyses.
The population analyses were performed with Arlequin v3.5 software . The parameters used to describe genetic diversity at each locality and each marker were the number of polymorphic sites (S), haplotype diversity (h) , the nucleotide diversity (π) , and the mean number of pairwise differences (k) . Nucleotide diversity (π) and haplotype diversity (h) were tested for significant differences among the groups by a non-parametric Kruskal-Walis in R software . To infer whether populations went through recent expansion, the mismatch distribution from each locality was calculated in Arlequin with 1,000 permutations. In addition, to test recent population expansion, Tajima’s D  and Fu’s Fs , were also calculated, with 1,000 permutations.
To infer the degree of genetic differentiation within and between the individuals (from same locality and group), as well as between the different groups (possible cryptic species found in the species delimitation analyses), AMOVA  was performed in Arlequin with 20,000 permutations. In addition, a permutation test of genetic differentiation based on a nearest-neighbor statistic (Snn)  was conducted in DNAsp v5  with 10,000 randomizations and an alpha value adjusted by Bonferroni correction for multiple tests.
The relationship between the genetic distance (Φst values)  and geographic distance (calculated as the minimum distance in kilometers between the two localities) was evaluated using Reduced Major Axis (RMA) regression. The correlation between the two distances was assessed by the Mantel test , implemented in the isolation-by-distance software (IBDWS v.3.23: http://ibdws.sdsu.edu/~ibdws/) , with 20,000 randomizations. Long-term gene flow inference was estimated by Migrate-n v3.6 software . Following a burn-in of 20,000, 2.0 x 106 genealogies were recorded at a sampling increment of 200 iterations. An adaptive heating scheme using four simultaneous Markov chains was implemented to increase the efficiency of searches. The same settings were used for all the groups. The long-term gene flow estimate M was converted to the average number of migrants per generation (xNm), using the formula xNm = θi x Mi→j. Two separated runs were performed and the xNm from each run was used to calculate a median. A non-parametric Kruskal-Wallis was performed in R  to assess the significance of the number of migrants per generation toward north and south. Also, a pairwise Student’s t was used to compare the number of migrants sent and received by each locality.
Haplotype networks and species delimitation
To avoid biasing or misdirecting the results we used only sequences with no ambiguities, which included trimming an average of 100 bp from each end of each gene fragment, resulting in final sequences of 605 bp and 362 bp for COI and COX3 respectively.
The species delimitation analyses for all specimens and concatenated genes (81 individuals) sorted the individuals into three major groups: two Lactea (Lactea 1, Lactea 2) and one Pallida, with support values above 50, plus seven singletons (S1 Fig). The same pattern is seen in the haplotype networks from these individuals (Fig 2). The two groups formed by Lactea individuals, Lactea 1 and Lactea 2, are 58 mutational steps apart from each other, and Pallida is 137 mutational steps from Lactea 2. It should be noted here that a group of ten Pallida individuals from DC evidently were mistakenly recorded in the field as Lactea and, on the strength of an abundance of COI sequence data for the genus in Genbank, are treated here as Pallida.
Lactea 1, Lactea 2 and Pallida (recovered from bPTP and TCS analysis). Pallida morphotype haplotypes are identified, the remaining are Lactea morphotype. Locality abbreviations as in Fig 1.
The species delimitation analyses with only COI sequences (147 individuals) has the same pattern as the concatenated analysis, with three major groups, but the number of singletons varies depending on the software used. With bPTP analysis, which gave the results with highest supports, there are 9 singletons (S2 Fig). In ABGD analysis there are 10 singletons and a fourth group consisting of two individuals (data not shown). GMYC allocates all individuals into three groups plus one singleton, but presents low values of support for the three groups (results not shown). The three groups are also recovered in the haplotypes networks from COI sequences (Fig 3). This network presents 68 haplotypes and 301 polymorphic sites with most of the haplotypes exclusive. Lactea 1 and Lactea 2 are separated by 25 mutational steps, while Lactea 2 and Pallida are 78 mutational steps apart. The singletons from bPTP are also excluded from the groups formed in the haplotype networks, some by large numbers of mutational steps. Therefore, those nine specimens were discarded from further analyses.
Locality abbreviations as in Fig 1.
GMYC results from COX3 sequences (109 individuals) subdivide Lactea 1 and Lactea 2 in two groups each, but Pallida remains as one group (S3 Fig). In this analysis a fourth group, Lactea 3, is also delimited with individuals of Lactea morphotype represented only by COX3 sequences. The haplotype network presents 91 haplotypes with 566 polymorphic sites. This network shows the four groups as separate entities where Lactea 1 is 82 mutational steps apart from Lactea 2, Lactea 2 is 72 steps apart from Pallida and 76 apart from Lactea 3 (Fig 4).
Locality abbreviations as in Fig 1.
The results from species delimitation analyses from concatenated and single markers were compared with each other and with the network haplotypes. From these comparisons, we are able to separate two morphological species as four different groups.
The ML analyses (lnL: -16353.45) for all COI sequenced specimens is presented in Fig 5. The phylogenetic tree with all Ototyphlonemertes placed Lactea 1 and Lactea 2 into two separate clades and with the Lactea morphotype individuals of Leasi et al.  also collected in Chile. Pallida individuals come out closely related to a Pallida group from Belize (Fig 5B) of Leasi et al. .
It should be noted here, inter alia, that this Chilean Pallida morphotype, with well-developed cerebral organs, a long, stout proboscis and its armature, is consistent with characteristics of the Pallida morphotype described by Envall and Norenburg  and could correspond to Ototyphlonemertes santacruzensis Mock and Schmidt, 1975, described from the Galapagos, for which there is no available genetic data for a morphologically confirmed specimen. The only field photomicrographs of one of our Chilean Pallida shows a pair of statocysts each with an irregular cluster of about six statolith granules (S3 Fig), rather than the four granules reported for O. santacruzensis and common for Pallida morphs found elsewhere in the world (JLN, pers obs). It is not unusual to see supernumerary granules in Pallida morphs but unusual to see almost identical irregularity in both statocysts (ibid.). Ototyphlonemertes cirrula Mock and Schmidt, 1975, also described from the Galapagos, resembles generally a Pallida morph but differs from all other Ototyphlonemertes in possessing epidermal cirri along the entire body length (these are easily overlooked by an inexperienced collector). More critically, the single O. cirrula described by Mock and Schmidt  is reported to have statocysts with up to 20 granules. However, one of us (JLN) found four individuals of Ototyphlonemertes on the Pacific coast of Panama with cirri along the entire body length but with statocysts having 6–8 granules, much as in S3 Fig, but we lack genetic data for these specimens. Hence, we need more morphological observations for the Chilean individuals, as well as morphological and molecular observations for those from the Galapagos.
Similarly, we have field photomicrographs for only a single Chilean Lactea. These reveal morphological affinity with Ototyphlonemertes lactea Corrêa, 1955 from Brazil, but also with Ototyphlonemertes americana Gerner, 1969 as reported from the Galapagos Islands by Mock and Schmidt . All lack cerebral organs and cephalic cirri, have a very short proboscis with a thin irregular stylet basis, accessory stylets next to the basis, and a short cylindrical proboscis mid-bulb. However, the mid-bulb in our specimen (S3 Fig) is slightly more elongate than depicted for the other two species but not as long as is characteristic for the Macintoshi morphotype .
Genetic diversity and gene flow estimates
To avoid causing an artifact by mixing probable different evolutionary units, the four evolutionary units were analyzed separately for intra- and interpopulational parameters. All localities from each group show moderate to high haplotype diversity (Table 2). There is no statistically significant difference in haplotype diversity between markers in Lactea 1 and Lactea 2, but COX3 values are higher and statistically different in Lactea 2 (Kruskal-Wallis x2 = 5.54, p = 0.0186). There also is no statistically significant difference among groups for COI or COX3. The nucleotide diversity is very variable, with no statistically significant difference between markers. Likewise, values of nucleotide diversity are not statistically different between groups for COI or COX3. The average pairwise differences (k) show high intra-population variation, with no statistically significant difference between markers or between groups in both COI and COX3 (Table 2).
Table 3 presents the AMOVA results for comparisons within and among localities for both markers. Most of the variability is within the localities in each group for both regions, except for Pallida. The pairwise Φst values show signs of differentiation between some populations in all groups (S4 Fig), but these signs don’t have a geographic pattern. In Lactea 1 for example, CH show the same degree of differentiation for both PT (a north population) and TO (a south population). Lactea 2 has low values of differentiation between all populations but HQ has high Φst values relative to BE, TO and LC. In Pallida and Lactea 3, due to the high intra-population variation, most of populations show high values of differentiation (S4 Fig).
Results from Tajima’s D analysis show most localities in each group are in equilibrium between mutation and genetic drift, with no evidence of recent geographic expansion. However, the results show evidence of geographic expansion and signs of recovery from a recent bottleneck in seven localities for COI region, and three for COX3 (Table 4). The Fu’s FS, on the other hand, shows only two localities in recent geographic expansion for COX3 and one, Pallida_DC, with conflicting results for each marker. In COI analyses, the localities show signs of population retraction and in COX3 show signs of population expansion (Table 4). Mismatch results shows signs of population expansion only in one locality, Lactea 1_DC. All other localities present no signs of expansion or retraction.
Results from Mantel tests show no statistically significant relationship between geographic and genetic distance in Pallida for COI and in all groups for COX3. Lactea 1 and Lactea 2 show a positive relationship between geographic and genetic distance for COI (r = 0.622, p = 0.0031, and r = 0.025, p = 0.0404, respectively), despite co-occuring in some localities.
Migrate-n results are shown in Tables 5 and 6. Pallida results are discordant between the two markers. However, this group is formed by individuals from different localities in COX3 and COI analyses. In general, all populations have high rates of migrants, with the exception of BE in Lactea 1 and all populations of Pallida when COI is analyzed. These results also show no statistical difference for migrant exchange between north and south populations for any group.
Presence of cryptic species
The presence of cryptic species is a common feature of meiofaunal organisms, since most of these tiny animals have few morphological characters to be used in identification. Therefore, many meiofaunal species have been recorded as cosmopolitan despite their low dispersion ability. This pattern is known as the Meiofauna Paradox  and has been discussed in many studies (e.g., [2, 20, 58, 59, 60, 61, 62]). Molecular data has increasingly revealed that some of these supposedly cosmopolitan species are actually complexes of cryptic species with low dispersal ranges (e.g., [14, 19, 20, 59, 63, 64, 65]).
The presence of cryptic species in Ototyphlonemertes is something already observed in other molecular studies from the western Atlantic coast [17, 18]. Delimitation between species and ecological morphotypes is difficult [23, 24, 66] because morphological characters frequently overlap. There has never, however, been a case of overlap for two fundamental character differences between Lactea and Pallida morphs: respectively, spiral vs. smooth stylet, and lack of cerebral organs vs. presence. It is, however, possible with the tight travel schedule and modest field microscopy employed during collecting in this study, to mis-allocate individuals.
The morphological group Lactea has different morphotypes distributed worldwide. However, the anatomical features of these morphotypes are not enough to reliably separate those groups into different species . The relationship between the cryptic species found here and previously characterized morphotypes is conditional on geographically more comprehensive study. Molecular data for all species described from the Galapagos not only could be very interesting for expanding this biogeographic study but they are needed to ascertain taxonomic affinities and status of the Chilean Ototyphlonemertes. The speciation might be caused by multiple colonization events causing genetic divergence and isolation in those groups , perhaps driven by ocean current, beach and weather dynamics.
The haplotype networks of both genes concatenated show a “parochial” pattern, where there is a high number of haplotypes, mostly restricted to one location. This pattern is characteristic of low-dispersal animals . Despite this, dispersion probably occurred over long distances, since the haplotypes have few mutation steps between each other, though most of the haplotypes are geographically restricted. Haplotype networks analyzed for each group separately, reveals a different pattern for Lactea 2. This haplotype network has a more star-like pattern, with one dominant haplotype and many rare haplotypes differing by few mutations. Star-like networks are common in species with a strong founder effect; e.g., when there is re-colonization in areas where local extinction occurred .
The genetic diversity indexes, like nucleotide and haplotype diversity, are very high for both markers. Genetic diversity values can be high in populations where environmental conditions are stable for long periods at evolutionary scales, allowing for accumulation of genetic diversity [68, 69]. This might not be the case for Ototyphlonemertes, many of which live in disturbed environments with seemingly ephemeral populations, present in a location in some periods and completely disappearing at others [18, 20]. High values of nucleotide and haplotype diversity are also related to large populations over evolutionary times . This most likely is the case for Ototyphlonemertes populations, since meiofaunal populations in general might occur in high numbers [70, 71, 72].
Neutrality tests show that most local groups are in equilibrium between mutation and genetic drift, with no expansion, but some have signs of recent expansion after a bottleneck. However, results from the three statistics used here (Tajima’s D, Fu’s Fs and Mismatch distribution) and for the two markers do not agree in some localities. The power of neutrality statistics decays with time since the expansion, but the pace is different for each test. For bottleneck events, Fu’s FS is the most powerful statistic; it, however, lacks power to detect bottlenecks that have just occurred. The same happens with population expansion and decrease, but Tajima’s D is also sensitive for population decrease . According to Ramos-Onsis and Rozas , tests based on mismatch distribution are very insensitive to detecting population growth, requiring large samples and stronger changes to indicate population growth. Therefore, for small sample sizes, as in this study, Tajima’s D and Fu’s FS seem to be more powerful to detect population fluctuations. In this sense, despite the seasonal environment, Ototyphlonemertes local groups are in equilibrium, except for Pallida_DC, which seems to be under expansion in both Tajima’s D and Fu’s FS statistics for both markers. Therefore, while these populations seem to be ephemeral they might have mechanisms to survive nearby or rapidly recolonize these environments.
Results from AMOVA and Migrate-n are congruent for most groups, with signs of gene flow in all groups, even between distant localities. Stochastic events, such as storms, were the mechanism suggested by Tulchinsky et al.  to explain long-distance dispersion found in Ototyphlonemertes along Florida and Caribbean coasts. Here the migration of these animals occurs most likely from larval  and occasional adult dispersal in ocean currents or by other dispersal mechanism (e.g., rafting–). The Chilean coast has a turbulent tectonic history (earthquakes) and a complex system of ocean currents with superficial, subsuperficial and deep currents flowing northwards and southwards. The currents that flow near the coast are the Chile Coastal Current, the Chile Coastal Countercurrent, the Peru Current (Humboldt Current) and the Peru-Chile Undercurrent (Gunther Current) [77, 78]. The north-flowing Humboldt Current is the one that most influences the coastal communities between 18°S and 42°S .
Some studies report a biogeographic break between 30°S and 33°S based on the occurrence and distribution of benthic species, separating this coast into North and South (e.g., [26, 27, 77, 79, 80, 81, 82, 83, 84]). This historical discontinuity, which originated in the last glaciation, disrupted gene flow in the past but low-dispersal species still retain this genetic signature . Species with high or medium dispersal do not present this genetic signature between populations from the North and South . Although the species in Ototyphlonemertes genus are presumed, like other mesopsammic meiofauna , to have low dispersal capability [20, 23, 24], our results show no differentiation between species from North and South for the four groups studied here, which may be evidence of a greater dispersal potential than expected. In addition, the results of pairwise Φst (S5 Fig) showing low values of differentiation between the localities support this idea. Another possibility for this lack of differentiation is the short time between extinction and recolonization of these populations. The environment where these populations develop is variable, subject to sudden changes that can easily extinguish very localized populations and erase some genetic signals.
Lactea 1 and Lactea 2 show patterns of isolation by distance only for COI, with significant results from Mantel’s test, AMOVA and highly significant Snn values. The rate of migrants in these two groups supports the idea of connected populations, but since they are isolated by distance, it is probable that the connection occurs by stepping-stones . For Lactea 2, comparing the results from both markers individually reveals that the local groups are connected, with a significant rate of migrants per generation between all localities. The Φst values are significant (S5 Fig) but low for mitochondrial markers, showing a level of structuring expected for populations with medium dispersion capability .
The different results found for the four groups further support the idea that these are four independent evolutionary entities with different ecological, dispersal and genetic characteristics. They each have most of their haplotypes geographically exclusive but the shape of the haplotype network is different for each group. In addition, Lactea 2 genetically is highly polymorphic, while Lactea 1, Lactea 3, and Pallida are less polymorphic, but still very diverse. The haplotype diversity is similar for all groups; the nucleotide diversity, however, is very low in Pallida, moderate in Lactea 1, and high in Lactea 2 and Lactea 3. Lactea 1 and Lactea 2 have significant isolation by distance, while Pallida and Lactea 3 do not. Also, the number of migrants per generation is very different between the groups. Therefore, these four groups do not present the same pattern in most of the characteristics analyzed, except for the lack of response to the historical biogeographic barrier present on the Chilean coast.
The results presented here suggest that these organisms exhibit patterns of gene flow similar to other organisms with high dispersal present in Chile, such as the isopods Limnoria chilensis and Limnoria quadripunctata, which have connected populations along the entire Chilean coast . They, somehow, are capable of long distance dispersal in sufficient frequency to connect the populations along the Chilean coast but have complicated evolutionary histories. In keeping with the Meiofauna Paradox, we see here more evidence that meiofauna species can successfully conquer new environments through dispersal.
S1 Table. List of all specimens, their geographic localization and GenBank access numbers.
Specimens from Leasi et al. 2016 are specified on the observations field.
S1 Fig. Resulting tree of species delimitation analysis in bPTP with concatenated sequences.
Numbers above the branches are the posterior probability of speciation events in each branch. Lactea 1 contains 24 specimens, Lactea 2 40 specimens and PallidaS 10 specimens.
S2 Fig. Resulting tree of species delimitation analysis in bPTP with COI sequences.
Numbers above the branches are the posterior probability of speciation events in each branch. Lactea 1 contains 24 specimens, Lactea 2 40 specimens and Pallida 15 specimens.
S3 Fig. Resulting tree of species delimitation analysis in GMYC with COX3 sequences.
The diagram in the upper left corner is the time at which the model infers that the threshold transition from the speciation-level events to the coalescent-level events takes place. Red branches were probably formed after the speciation events. Lactea 1 contains 29 specimens, Lactea 2 contains 53 specimens, Lactea 3 contains 11 specimens, and Pallida contains 15 specimens.
S4 Fig. Photomicrographs of squeezed live specimens of Pallida and Lactea morphotypes.
(A) Specimen of Pallida showing statocysts with six granules, marked with arrows. (B) Specimen of Lactea with everted proboscis showing the mid-bulb (MB).
The authors are grateful to the Universidad Católica del Norte (Chile) and, especially, Martin Thiel for providing lodging and microscope facilities. We thank Cynthia Santos and Nelson Valdivia for helping with the fieldwork. We are thankful to Diego Fontaneto, an anonymous reviewer and the Academic Editor Tzen-Yuh Chiang for their comments and insights on the previous version of this manuscript.
- 1. McIntyre AD, The meiofauna and macrofauna of some tropical beaches. J Zool 1968; 156: 377–392.
- 2. Giere O. Meiobenthology: The Microscopic Fauna in Aquatic Sediments. Berlin: Springer; 1993.
- 3. Leasi F, Norenburg JL. The necessity of DNA taxonomy to reveal cryptic diversity and spatial distribution of meiofauna, with a focus on Nemertea. PLoS One 2014; 9(8): e104385. pmid:25093815
- 4. Commito JA, Tita G. Differential dispersal rates in an intertidal meiofauna assemblage. J Exp Mar Bio Ecol 2002; 268(2): 237–256.
- 5. Armonies W. Drifting meio-and macrobenthic invertebrates on tidal flats in Königshafen: a review. Helgol. Meeresunters. 1994; 48(2): 299.
- 6. Sun B, & Fleeger JW. Field experiments on the colonization of meiofauna into sediment depressions. Mar Ecol Prog Ser 1994; 110: 167–175.
- 7. Fleeger JW, Yund PO, Sun B. Active and passive processes associated with initial settlement and post-settlement dispersal of suspended meiobenthic copepods. J Mar. Res. 1995; 53(4): 609–645.
- 8. Martin G. The duo-gland adhesive system of the Archiannelids Protodrilus and Saccocirrus and the turbellarian Monoceis. Zoomorphology 1978; 91(1):63–75.
- 9. Hermans CO. The duogland adhesive system. Oceanogr. Mar. Biol. Ann. Rev. 1983; 21: 283–339.
- 10. Tyler S, Rieger GE. Adhesive organs of the Gastrotricha. Zoomorphology 1980; 95(1): 1–15.
- 11. Knowlton N. Sibling species in the sea. Annu Rev Ecol Syst 1993; 24(1): 189–216.
- 12. Westheide W, Haß-Cordes E, Krabusch M, Müller M. Ctenodrilus serratus (Polychaeta: Ctenodrilidae) is a truly amphi-Atlantic meiofauna species-evidence from molecular data. Mar Bio 2003; 142(4): 637–642.
- 13. Bik HM, Thomas WK, Lunt DH, Lambshead PJD. Low endemism, continued deep-shallow interchanges, and evidence for cosmopolitan distributions in free-living marine nematodes (order Enoplida). BMC Evol. Biol 2010; 10(1): 389.
- 14. Jörger KM, Norenburg JL, Wilson NG, Schrödl M. Barcoding against a paradox? Combined molecular species delineations reveal multiple cryptic lineages in elusive meiofaunal sea slugs. BMC Evol. Biol. 2012; 12(1): 245.
- 15. Westheide W, Schmidt H. Cosmopolitan versus cryptic meiofaunal polychaete species: an approach to a molecular taxonomy. Helgol Mar Res 2003; 57(1): 1–6.
- 16. Casu M, Curini-Galletti M. Sibling species in interstitial flatworms: a case study using Monocelis lineata (Proseriata: Monocelididae). Mar. Biol 2004; 145(4): 669–679.
- 17. Andrade SC, Norenburg JL, Solferini VN. Worms without borders: genetic diversity patterns in four Brazilian Ototyphlonemertes species (Nemertea, hoplonemertea). Mar. Bio 2011; 158(9): 2109.
- 18. Tulchinsky AY, Norenburg JL, Turbeville JM. Phylogeography of the marine interstitial nemertean Ototyphlonemertes parmula (Nemertea, Hoplonemertea) reveals cryptic diversity and high dispersal potential. Mar Bio 2012; 159(3): 661–674.
- 19. Kieneke A, Martínez-Arbizu P M, Fontaneto D. Spatially structured populations with a low level of cryptic diversity in European marine Gastrotricha. Mol. Ecol. 2012; 21(5): 1239–1254. pmid:22257178
- 20. Leasi F, Andrade SCS, Norenburg JL. At least some meiofaunal species are not everywhere. Indication of geographic, ecological and geological barriers affecting the dispersion of species of Ototyphlonemertes (Nemertea, Hoplonemertea). Mol Ecol 2016; 25(6): 1381–1397. pmid:26840255
- 21. Corrêa DD. Ecological study of Brazilian Ototyphlonemertes. Comunicaciones Zoologicas del Museo de Historia Natural de Montevideo 1949; 3(55): 1–7.
- 22. Mock H, Schmidt P. Interstitielle Fauna von Galapagos XIII. Ototyphlonemertes Diesing:(Nemertini, Hoplonemertini). Akad. d. Wiss. ud Literatur, 1975.
- 23. Envall M, Norenburg JL. Morphology and systematics in mesopsammic nemerteans of the genus Ototyphlonemertes (Nemertea, Hoplonemertea, Ototyphlonemertidae). Hydrobiol. 2001; 456(1): 145–163.
- 24. Envall M, Sundberg P. Phylogenetic relationships and genetic distances between some monostiliferous interstitial nemerteans (Ototyphlonemertes, Hoplonemertea, Nemertea) indicated from the 16S rRNA gene. Zoo J. Linn. Soc 1998; 123(2): 105–115.
- 25. Zakas C, Binford J, Navarrete SA, Wares JP. Restricted gene flow in Chilean barnacles reflects an oceanographic and biogeographic transition zone. Mar Ecol Progr Ser 2009; 394: 165–177.
- 26. Tellier F, Meynard AP, Correa JA, Faugeron S, Valero M. Phylogeographic analyses of the 30°S south-east Pacific biogeographic transition zone establish the occurrence of a sharp genetic discontinuity in the kelp Lessonia nigrescens: vicariance or parapatry? Mol Phylogenet Evol 2009; 53: 679–693. pmid:19647087
- 27. Sánchez R, Sepúlveda RD, Brante A, Cárdenas L. Spatial pattern of genetic and morphological diversity in the direct developer Acanthina monodon (Gastropoda:Mollusca). Mar Ecol Prog Ser 2011; 434: 121–131.
- 28. Varela AI, Haye PA. The marine brooder Excirolana braziliensis (Crustacea: Isopoda) is also a complex of cryptic species on the coast of Chile. Rev Chil Hist Nat 2012; 85(4): 495–502.
- 29. Hormazabal S, Shaffer G, Leth O. Coastal transition zone off Chile. J Geophys Res: Oceans 2004; 109: C01021.
- 30. Navarrete SA, Wieters EA, Broitman BR, Castilla JC. Scales of benthic–pelagic coupling and the intensity of species interactions: from recruitment limitation to top-down control. Proc Nat Acad Sci U S A 2005; 102(50): 18046–18051.
- 31. Wieters EA, Broitman BR, Brancha GM. Benthic community structure and spatiotemporal thermal regimes in two upwelling ecosystems: Comparisons between South Africa and Chile. Limnol Oceanogr 2009; 54(4): 1060–1072.
- 32. Poulin E, Palma AT, Leiva G, Narvaez D, Pacheco R, Navarrete SA, Castilla JC. Avoiding offshore transport of competent larvae during upwelling events: the case of the gastropod Concholepas concholepas in Central Chile. Limnol Oceanogr 2002; 47(4): 1248–1255.
- 33. Guizien K, Brochier T, Duchêne JC, Koh BS, Marsaleix P. Dispersal of Owenia fusiformis larvae by wind-driven currents: turbulence, swimming behaviour and mortality in a three-dimensional stochastic model. Mar Ecol Prog Ser. 2006; 311: 47–66.
- 34. Thollesson M. Increasing fidelity in parsimony analysis of dorid nudibranchs by differential weighting, or a tale of two genes. Mol Phylogenet Evol 2000; 16(2): 161–172. pmid:10942604
- 35. Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol Mar Biol Biotechnol 1994; 3(5): 294–299. pmid:7881515
- 36. Turbeville JM, Smith DM. The partial mitochondrial genome of the Cephalothrix rufifrons (Nemertea, Palaeonemertea): Characterization and implications for the phylogenetic position of Nemertea. Mol Phylogenet Evol 2007; 43(3): 1056–1065. pmid:17210260
- 37. Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 1997; 25(17): 3389–3402. pmid:9254694
- 38. Katoh K, Standley DM. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol 2013; 30(4): 772–780. pmid:23329690
- 39. Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 2014; 30(9): 1312–1313. pmid:24451623
- 40. Zhang J, Kapli P, Pavlidis P, Stamatakis A. A general species delimitation method with applications to phylogenetic placements. Bioinformatics 2013; 29(22): 2869–2876. pmid:23990417
- 41. Fujisawa T, Barraclough TG. Delimiting Species Using Single-Locus Data and the Generalized Mixed Yule Coalescent Approach: A Revised Method and Evaluation on Simulated Data Sets, Sys Bio 2013; 62(5): 707–724. https://doi.org/10.1093/sysbio/syt033
- 42. Drummond AJ, Suchard MA, Xie D, Rambaut A. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Mol. Biol. Evol. 2012; 29(8): 1969–1973. pmid:22367748
- 43. Puillandre N, Lambert A, Brouillet S, Achaz G. ABGD, Automatic Barcode Gap Discovery for primary species delimitation. Mol Ecol 2012; 21(8): 1864–1877. pmid:21883587
- 44. Templeton AR, Crandall KA, Sing CF. A cladistic analysis of phenotypic associations with haplotypes inferred from restriction endonuclease mapping and DNA sequence data. III. Cladogram estimation. Genetics 1992; 132(2): 619–633. pmid:1385266
- 45. Clement M, Posada DCKA, Crandall KA. TCS: a computer program to estimate gene genealogies. Mol Ecol 2000; 9(10): 1657–1659. pmid:11050560
- 46. Excoffier L, Laval G, Schneider S. Arlequin (version 3.0): an integrated software package for population genetics data analysis. Evol Bioinfo Online 2005; 1: 47–50.
- 47. Nei M. Molecular evolutionary genetics. New York: Columbia University Press; 1987
- 48. Tajima F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 1989; 123(3): 585–595. pmid:2513255
- 49. Tajima F. Evolutionary relationship of DNA sequences in finite populations. Genetics 1983; 105(2): 437–460. pmid:6628982
- 50. R Core Team. R: A language and environment for statistical computing. 2013.
- 51. Fu YX. Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics 1997; 147(2): 915–925. pmid:9335623
- 52. Excoffier L, Smouse PE, Quattro JM. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 1992; 131(2): 479–491. pmid:1644282
- 53. Hudson RR. A new statistic for detecting genetic differentiation. Genetics 2000; 155(4): 2011–2014. pmid:10924493
- 54. Librado P, Rozas J. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 2009; 25(11): 1451–1452. pmid:19346325
- 55. Mantel N. The detection of disease clustering and a generalized regression approach. Cancer Res 1967; 27: 209–220. pmid:6018555
- 56. Jensen JL, Bohonak AJ, Kelley ST. Isolation by distance, web service. BMC genetics 2005; 6(1): 13.
- 57. Beerli P. How to use migrate or why are markov chain monte carlo programs difficult to use? In: Bertorelle G, Bruford MW, Hauffe HC, Rizzoli A, Vernesi C, editors. Population Genetics for Animal Conservation, volume 17 of Conservation Biology. Cambridge UK: Cambridge University Press; 2009. p. 42–79.
- 58. Todaro MA, Fleeger JW, Hu YP, Hrincevich AW, Foltz DW. Are meiofaunal species cosmopolitan? Morphological and molecular analysis of Xenotrichula intermedia (Gastrotricha: Chaetonotida). Mar Bio 1996; 125(4): 735–742.
- 59. Schmidt H, Westheide W. Are the meiofaunal polychaetes Hesionides arenaria and Stygocapitella subterranea true cosmopolitan species?—results of RAPD‐PCR investigations. Zoo Scr 1999; 29: 17–27.
- 60. Boeckner MJ, Sharma J, Proctor HC. Revisiting the meiofauna paradox: dispersal and colonization of nematodes and other meiofaunal organisms in low-and high-energy environments. Hydrobiol 2009; 624: 91–106.
- 61. Macfarlane C, Drolet D, Barbeau MA, Hamilton DJ, Ollerhead J. Dispersal of marine benthic invertebrates through ice rafting. Ecology 2013; 94: 250–256. pmid:23600259
- 62. Packmor J, Riedl T. Records of Normanellidae Lang, 1944 (Copepoda, Harpacticoida) from Madeira island support the hypothetical role of seamounts and oceanic islands as “stepping stones” in the dispersal of marine meiofauna. Mar Biodiv 2016; 46(4): 861–877.
- 63. Cvon Soosten, Schmidt H, Westheide W. Genetic variability and relationships among geographically widely separated populations of Petitia amphophthalma (Polychaeta: Syllidae). Results from RAPD-PCR investigations. Mar Bio 1998; 131(4): 659–669.
- 64. Westheide W, Haß-Cordes E. Molecular taxonomy: description of a cryptic Petitia species (Polychaeta: Syllidae) from the island of Mahé (Seychelles, Indian Ocean) using RAPD markers and ITS2 sequences. J Zoo Syst Evol Res 2001; 39(1/2): 103–111.
- 65. Todaro MA, Dal Zotto M, Bownes SJ, Perissinotto R. First records of Gastrotricha from South Africa, with description of a new species of Halichaetonotus (Chaetonotida, Chaetonotidae). ZooKeys 2011; (142): 1–13. pmid:22144860
- 66. Envall M. Ototyphlonemertes correae sp. nov. and a redescription of O. duplex (Nemertea: Monostilifera: Ototyphlonemertidae), with a phylogenetic consideration of the genus. J. Zoo 1996; 238(2): 253–277.
- 67. Allcock AL, Strugnell JM. Southern Ocean diversity: new paradigms from molecular ecology. Trends Ecol Evol 2012; 27(9): 520–528. pmid:22727016
- 68. Aleksić JM, Geburek T. Mitochondrial DNA reveals complex genetic structuring in a stenoendemic conifer Picea omorika [(Panč.) Purk.] caused by its long persistence within the refugial Balkan region. Plant Syst. Evol 2010; 285: 1–11.
- 69. Hellberg ME. Gene flow and isolation among populations of marine animals. Annu. Rev. Ecol. Evol. Syst. 2009; 40: 291–310.
- 70. McLachlan A. Composition, distribution, abundance and biomass of the macrofauna, and meiofauna of four sandy beaches. Afr Zoo 1977; 12(2): 279–306.
- 71. Murrell MC, Fleeger JW. Meiofauna abundance on the Gulf of Mexico continental shelf affected by hypoxia. Cont Shelf Res 1989; 9(12): 1049–1062.
- 72. Landers SC, Romano III FA, Stewart PM, Ramroop STEVE. A multi-year survey of meiofaunal abundance from the northern Gulf of Mexico continental shelf and slope. Gulf Mex Sci 2012; 1(2): 20–29.
- 73. Ramírez-Soriano A, Ramos-Onsins SE, Rozas J, Calafell F, Navarro A. Statistical power analysis of neutrality tests under demographic expansions, contractions and bottlenecks with recombination. Genetics 2008; 179(1): 555–567. pmid:18493071
- 74. Ramos-Onsins SE, Rozas J. Statistical properties of new neutrality tests against population growth. Mol Biol Evol 2002; 19(12): 2092–2100. pmid:12446801
- 75. Chernyshev AV. Nemertean larvae of the Ototyphlonemertidae family in the plankton of Peter the Great Bay, Sea of Japan. Rus. Jour. Mar. Biol 2000; 26(1): 48–50.
- 76. Thiel M, Gutow L. The ecology of rafting in the marine environment I. The floating substrata. Oceanogr. Mar. Biol. Annu. Rev. 2005; 42: 181–264.
- 77. Brattström H, Johanssen A. Ecological and regional zoogeography of the marine benthic fauna of Chile: Report no. 49 of the Lund University Chile Expedition 1948–49. Sarsia 1983; 68(4): 289–339.
- 78. Strub PT, Mesías JM, Montecino V, Rutlland J, Salinas S. Coastal Ocean Circulation off Western South America. In: Robbinson AL, Brink KH. The Sea. New Jersey: Jon Wiley and Sons; 1998
- 79. Thiel M, Macaya EC, Acuna E, Arntz WE, Bastias H, Brokordt K. et al. The Humboldt Current System of northern and central Chile: oceanographic processes, ecological interactions and socioeconomic feedback. Oceanogr Mar Bio 2007; 45: 195–344.
- 80. Marincovich L Jr. Intertidal mollusks of Iquique, Chile. Nat. Hist. Mus. Los Angeles County, Sci. Bull. 1973; 16:1–49.
- 81. Hoffmann AJ, Santelices B. Marine Flora of Central Chile. Santiago: Ediciones Universidad Católica de Chile; 1997.
- 82. Camus PA. Biogeografía marina de Chile continental. Rev. Chil. Hist. Nat. 2001; 74(3): 587–617.
- 83. Broitman B, Veliz F, Manzur T, Wieters EA, Randall Finke G, Fornes PA, et al. Geographic variation in diversity of wave exposed rocky intertidal communities along central Chile. Rev. Chil. Hist. Nat. 2011; 84: 143–154.
- 84. Montecinos A, Broitman BR, Faugeron S, Haye PA, Tellier F, Guillemin ML. Species replacement along a linear coastal habitat: phylogeography and speciation in the red alga Mazzaella laminarioides along the southeast Pacific. BMC Evol Biol 2012; 12: 97. pmid:22731925
- 85. Haye PA, Segovia NI, Muñoz-Herrera NC, Gálvez FE, Martínez A, Meynard A. et al. Phylogeographic structure in benthic marine invertebrates of the southeast Pacific coast of Chile with differing dispersal potential. PLoS One 2014; 9(2): e88613. pmid:24586356
- 86. Johansson ML, Alberto F, Reed DC, Raimondi PT, Coelho NC, Young MA, et al. Seascape drivers of Macrocystis pyrifera population genetic structure in the northeast Pacific. Mol. Ecol. 2015; 24(19): 4866–4885.