The Azorean barnacle, Megabalanus azoricus (Pilsbry, 1916), is a Macaronesian endemic whose obscure taxonomy and the unknown relationships among forms inhabiting isolated Northern Atlantic oceanic islands is investigated by means of molecular analysis herein. Mitochondrial data from the 16S rRNA and COX1 genes support its current species status, tropical ancestry, and the taxonomic homogeneity throughout its distribution range. In contrast, at the intraspecific level and based on control region sequences, we detected an overall low level of genetic diversity and three divergent lineages. The haplogroups α and γ were sampled in the Azores, Madeira, Canary, and Cabo Verde archipelagos; whereas haplogroup β was absent from Cabo Verde. Consequently, population analysis suggested a differentiation of the Cabo Verde population with respect to the genetically homogenous northern archipelagos generated by current oceanographic barriers. Furthermore, haplogroup α, β, and γ demographic expansions occurred during the interglacial periods MIS5 (130 Kya - thousands years ago -), MIS3 (60 Kya), and MIS7 (240 Kya), respectively. The evolutionary origin of these lineages is related to its survival in the stable southern refugia and its demographic expansion dynamics are associated with the glacial-interglacial cycles. This phylogeographic pattern suggests the occurrence of genetic discontinuity informative to the delimitation of an informally defined biogeographic entity, Macaronesia, and its generation by processes that delineate genetic diversity of marine taxa in this area.
Citation: Quinteiro J, Manent P, Pérez-Diéguez L, González JA, Almeida C, Lopes E, et al. (2015) Phylogeography of a Marine Insular Endemic in the Atlantic Macaronesia: The Azorean Barnacle, Megabalanus azoricus (Pilsbry, 1916). PLoS ONE 10(4): e0124707. https://doi.org/10.1371/journal.pone.0124707
Academic Editor: Andrew R. Mahon, Central Michigan University, UNITED STATES
Received: November 6, 2014; Accepted: March 3, 2015; Published: April 28, 2015
Copyright: © 2015 Quinteiro et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All nucleotide sequences are avaliable under GenBank/EMBL/DDBJ accession numbers: KM575947-KM576099.
Funding: This work was supported by the EU project BANGEN (MAC/1/C070) 701 BANGEMAC Network: Genetic Bank for Macaronesia Transnational Cooperation 702 Program Azores-Madeira-Canaries (PCT-MAC 2007-2013). 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 Azorean barnacle belongs to the taxonomically problematic genus Megabalanus . In the latest major revision of the genus it was classified as Megabalanus azoricus (Pilsbry, 1916) . It was formerly included in Darwin’s  subdivision A of the genus Balanus, which contained the larger forms of Balani, as a variety of Balanus tintinnabulum (Linnaeus). Hoek  proposed the term Megabalanus for these forms and Pilsbry  used it as a subgenus. Newman and Ross  elevated it to the generic rank, Megabalanus, which currently includes 23 species . Recently, it has been referred to as Megabalanus tintinnabulum . In some instances, it is cited as M. azoricus when located in the Azores and as M. tintinnabulum when found in the other Macaronesian archipelagos . Although a recent morphological re-examination clearly described M. azoricus from the Azores , their taxonomic status and phylogenetic relationships with the other Macaronesian forms remains unclear . It is apparently endemic to the Macaronesian archipelagos, although some records suggest the presence of similar forms as far south as the St. Helena Islands . Additionally, their fouling ability confers a high dispersion capacity, which may have altered their original distribution . This occurs in several Megabalanus species, including M. coccopoma, M. rosa, and M. volcano, whose current cosmopolitan distributions are associated with fouling in ocean-crossing vessels .
Megabalanus azoricus, common, native endemic form in the Azores islands, has been reported from different collections . Historically, the species is a highly appreciated seafood and is consumed locally . Topotypes are well characterized  and their reproductive biology described . Darwin  assigned the Madeira islands Megabalanus specimens to the M. tintinnabulum taxa. A re-examination of these specimens assigned them to the M. azoricus taxa . González et al.  reported the occurrence of M. azoricus in the Canaries, tracing archeological remains of the species back to 800 years ago. Although present throughout the archipelago, it is more abundant in the eastern islands, closest to the African coast . Although M. azoricus has been cited for the Cabo Verde Islands, were it is also consumed as a seafood, there is no detailed study of this population [2, 14].
The absence of genetic data for M. azoricus means that morphological characters have been exclusively used for species diagnosis to date. The wall plates, terga, and scuta have been well described  and facilitate differentiation from other co-generic species. Recently, a re-examination of Azorean specimens confirmed the validity of these external characters, the thoracic and mouth appendages were also used . However, genetic data is necessary to confirm the expected congruence of genetics with morphology, providing a standard barcode for species identification, to identify taxonomic distinctiveness and relationships with co-generic species, and to define the homogeneity or partition of genetic diversity along the Macaronesian geographic range under discussion. The resulting species delimitation is a fundamental prerequisite for biogeographical hypothesis testing . Available mitochondrial DNA data for co-generic species available include partial COX1 and 16SrDNA sequences and the complete mitochondrial genome for M. volcano (NC_006293). Homologous data has been used in similar biological and biogeographic studies to explore taxonomic and phylogeographic issues. Consequently, the highly complex taxonomic ranks of barnacles have been phylogenetically evaluated [16–19]. However, there are also cases where genetic analysis has assisted low-level taxonomic revisions. For example, the naked barnacle, Heteralepas japonica, located in Philippine waters, shows a distinctive clade divergent from the Japanese and Taiwanese populations and is now a new species, Heteralepas cantelli . Similar work contributed to new species descriptions in the Tretaclita squamosa complex in Asia [21, 22]. Analysis of genetic diversity through the East Atlantic distribution range of Pollicipes pollicipes detected a distinctive, monophyletic population in Cabo Verde , which was subsequently described as a new species . Phylogeographical inferences include the relationship between local oceanography and clade structure in the morphologically homogeneous acorn barnacle, Chthamalus malayensis, along the Indo-West Pacific . The effect of historic glacial events on the genetic diversity of European and North American populations of Semibalanus balanoides was investigated by analyzing control region variation . The mechanisms delineating the range distribution in Tetraclita rubescens, focusing on migration load and current climate change effects, have also been investigated .
A prerequisite for Macaronesian endemism is the existence of a well-defined biographic entity supporting it. Throughout the last 60 My (million years) the Macaronesian region has been evolving geologically, and at present is comprised of diverse seamounts and the Azores, Madeira, Canaries, and Cabo Verde archipelagos . However, their biogeographic delimitation has also been evolving and is mainly inferred from vegetation  but also from ichthyological distribution data , from a nuclear biogeographic unit including the volcanic provinces of the Canaries and Madeira , the Azores archipelago , the Cabo Verde Islands , and frequently incorporating the Iberian and African continental coastal areas . As a result, Macaronesia spans an extensive latitudinal range from 40–15° N, a climate gradient from temperate, Mediterranean, and tropical oceanic climate regimes, a gradient of distances to the mainland, and diverse floristic affinity patterns. This all raises the question of whether or not Macaronesia as a biogeographic entity is a valid zoogeographical concept .
Some crucial features of the M. azoricus species are digonic hermaphroditism in sessile adults, development of planktotrophic larvae, dependence on exposed rocky shores in the intertidal habitat, and insular distribution restricted to the northeastern Atlantic oceanic islands . These Macaronesian islands are a source of evolutionary paradigms . They emerged in the open-ocean where contrasting hydrographic mechanisms that favors either dispersion or isolation of marine organisms occurs [23, 36, 37]. Sessile intertidal organisms, such as barnacles, show a hydrographic-dependent mobility throughout the planktotrophic larval phase. The genetic signature expected from this dispersal pattern can be recovered after discriminating it from signals produced by a number of species-specific processes and factors delineating genetic diversity within a marine species, including life history traits, behavior, selection, and historical demography . Therefore, the delineation of intraspecific genetic diversity in this barnacle species must match the eco-geographical marine continuum expected for the insular Macaronesia biogeographical region.
The high level of direct exploitation as a food source has been pinpointed as the cause of declining populations in the Azores archipelago [6, 11], a situation that can likely be extrapolated to the rest of Macaronesia. Accordingly, based on the “global/regional importance, rarity, sensibility, keystone status, decline, and threat”, M. azoricus has been placed on the OSPAR 2010 list of threatened and/or declining species and habitats . In this critical context, taxonomic, genetic, and phylogeographic data will make a valuable contribution to resource management and conservation strategies of the Azorean barnacle .
The present work includes the characterization of diverse conserved codifying and non-codifying hypervariable mitochondrial sequences to allow both taxonomic and population structure analysis. Thus, the phylogenetic and descriptive analysis of the intraspecific genetic diversity in M. azoricus allow us to evaluate hypotheses on their taxonomic and endemic status throughout their Macaronesian geographic range. Subsequently, inferences on gene flow and historical demography facilitate the definition of a specific phylogeographic pattern consistent with the hypothesis of a putative Macaronesian biogeographic marine entity, and with the local hydrography and paleoclimatology processes.
Materials and Methods
Sampling and DNA isolation
Four Megabalanus azoricus (Pilsbry, 1916) samples (N ≥ 50 individuals) were collected from each of the Macaronesian archipelagos, including the Azores, Madeira (including Savage), Canaries, and Cabo Verde Islands (Fig 1, Table 1), comprising a total of 207 specimens. No regulations were adopted in response to the suspected declining status ; thus, no specific permissions were required at the sampling locations during the collection period. Therefore, the work did not involve endangered or protected species. However, at present, capture is formally prohibited at the Canary Islands to stock recovery. Immediately after collection, the complete individuals were fixed in 70% ethanol and preserved at −20°C until analysis. A specimen of the sympatric species Megabalanus tulipiformis (J. González, ICCM Collection: ICCM21A-SP370) was also collected in the Canaries. A muscle tissue section was excised to obtain a 30 mg tissue sample for total DNA isolation with the E.Z.N.A.® Tissue mollusk kit (Omega Bio-Tek, Norcross, GA). DNA was spectrophotometrically quantified with SmartSpec Plus (Bio Rad, Hercules, CA) and its integrity was then evaluated by 1.5% agarose gel electrophoresis. The BANGEMAC Network (http://bangen-pct.org/) maintains a collection of the analyzed tissues and DNA samples.
Image (NASA) of the north-eastern Atlantic area indicating the sampling locations for each of the Macaronesian archipelagos. Arrows indicate the main surface currents and their directions. Haplogroup (α, β, and γ) frequencies estimated for each sampling site are presented in pie charts.
The complete individuals and the external and internal views of the scuta and terga plates were observed to verify their congruence with the morphological descriptions available for M. azoricus [2, 8].
Sequence amplification and sequencing
The intraspecific analysis was carried out with the complete mitochondrial control region sequences flanked by the 12S rRNA and tRNA-Ile genes. Moreover, the partial mitochondrial 16S rRNA and COX1 gene sequences were obtained in a subset of the sample for the interspecific comparative analysis of sequences in phylogenetic reconstruction and DNA barcoding. The near complete mitochondrial 16S rRNA was amplified with Cirr-16S-5F (5′-CAACCTGGCTYACGCCGGTCT-3′) and Cirr-Val-3R (5′-GCTGTTTAAGCATTTCATTTACACTGAAAAGG-3′) primers. The fragment in the COX1 mitochondrial gene was amplified using the LCO1490 and HCO2198 primers . Finally, amplification of a fragment containing the complete mitochondrial control region and flanking sequences was carried out with the primers Mazo-12S-1F (5′-CTATCAAAGTAATCCTTTTGTCAGGCA-3′) and Mazo-Ile-2R (5′-′GATAACAACACGGACCTCAACGAT-3′).
Amplifications were performed in a 15 μL reaction volume using GeneAmp 9700 thermocyclers (Applied Biosystems, Carlsbad, CA). PCR reactions comprised 1X AmpliTaq 360 buffer (Applied Biosystems), 3.5mM MgCl2, 200 μM each dNTP, 0.13 μM each primer, 0.15 units of AmpliTaq 360 (Applied Biosystems), and 10–50 ng of total DNA. However, amplification within the COX1 gene was performed with 2.5mM MgCl2. The consensus PCR profile was 94°C for 3 min, followed by 36 cycles of 94°C for 30 s, 50°C for 30 s, and 72°C for 60s, then 72°C for 7 min.
The PCR products were digested with Exo-SAP-It (Affimetrix, Santa Clara, CA), to remove primers and to deactivate unused dNTPs, and sequenced in both directions using the BigDye 3.1 sequencing kit (Applied Biosystems). The extension products were purified with DyeEx columns (Qiagen, Hilden, Germany) and separated and detected in a 3730xl Genetic Analyzer (Applied Biosystems). Following chromatogram revision and trimming with Sequence Scanner software (Applied Biosystems), the sequences were aligned with ClustalX  implemented in BioEdit .
Taxonomic and phylogenetic inference
The external morphology of the specimens was verified for compliance with previous descriptions of the Azorean barnacle [1, 2, 8]. A meta-analysis of the phylogenetic relationships among Megabalanus taxa based on the available homologous 16S rRNA partial gene data was performed to corroborate the phylogenetic singularity of the sampled M. azoricus and define their phylogenetic relationships in a limited but informative data set. Additionally, our dataset (N = 54) was compared with the available Megabalanus COX1 sequences through BLAST (http://blast.ncbi.nlm.nih.gov/) and BOLD  to find homologous sequences with maximum similarity. The ribosomal phylogenetic inference was based in the Bayesian criterion in MrBayes v.3.2 . The HKY model  was selected for 16S rRNA alignment based on AIC estimations in jMoldeltest v.2.1.1  and implemented in MrBayes with an α parameter of the γ distribution = 0.32 and the proportion of invariable sites = 0.57 sets as priors. Markov chain Monte Carlo (MCMC) analysis was run twice for 10,000,000 generations with default parameters being obtained the 50% majority rule tree.
Population structure analysis
Genetic diversity estimations for the control region and COX1 sequences were obtained in Arlequin v. 3.5  and DnaSPv.5.10 , these included nucleotide diversity (π), haplotype diversity (h), and Tajima’s D in populations and observed haplogroups.
The demographic expansion hypothesis within this species was tested with alternative tests based on different criteria. First, using the distribution of the observed number of differences between sequence pairs obtained with Arlequin v3.5 , the so called mismatch distributions, which are usually unimodal in populations under recent demographic expansion [48, 49]. The mismatch distributions were obtained for both populations and haplogroups. The adequacy of the expansion model was assessed using a parametric bootstrap approach , by means of i) the sum of squared deviations (SSD) between the observed and expected distributions, and ii) computing the raggedness index (H’s Ri) from the observed distribution as suggested by Harpending . The corresponding estimators of time of expansion (τ) and mutational parameters θ0 and θ1, before and after the expansion event, were also obtained in Arlequin by parametric bootstrapping .
We also performed the Tajima’s  and the Fu’s  tests for neutrality. Moreover, the significant values for the Tajima’s D statistic can be related to demographic dynamics  and the Fu’s Fs statistic shows large negative values when estimated from a population undergoing demographic expansion . For small sample sizes the R2 test is superior in detecting population growth .
The pairwise FST among the four sampled populations and their significance were computed in Arlequin v.3.5 . The significance of the partition of genetic diversity within a jerarquized structure including populations and their groups was tested with AMOVA, also in Arlequin v.3.5 . The population size dynamics over time were reconstructed by Bayesian coalescent inference . A Bayesian skyline plot of effective breeding population size (Ne) through time was obtained with BEAST (version 1.7.5)  under the HKY model  with site rate heterogeneity, specifying an estimated fixed mean substitution rate value, using a linear model, and 50,000,000 generations for the MCMC algorithm.
Relationships among the sampled haplotypes, their relative frequencies, and geographic locations were graphically described by means of a median-joining network  in Network v.4.6 (Fluxus Technology Ltd., Suffolk, UK). Additionally, the phylogenetic relationships and assemblage of haplotypes were corroborated in a neighbor-joining optimal mid-point rooted tree inferred with MEGA 5 , with the branch support estimated by bootstrapping (2,000 replicates). Considering the available models of sequence evolution for Arlequin and MEGA calculations, the model that best fit the control region sequence data based on AIC estimations from jMoldeltest v2.1.1  was the Tamura-Nei , with an α value = 0.2220 for the gamma distribution.
Major revisions of the complex Megabalanus taxonomy include Megabalanus azoricus at the species rank and define their external diagnostic characters [1, 2]. The individuals sampled throughout the Macaronesian archipelagos, including the Azores type location (Terceira, Azores), exhibit morphological characters that agree with the available descriptive data for M. azoricus . Thus, the parietal plates are predominantly ribbed. The straight scutum has prominent growth ridges in the external view marked by longitudinal striae, whereas internally there is a prominent adductor ridge. The scutum—tergal segment is narrow. The terga spur furrow is almost closed, and internally, the blue-violet to purple and white parities present a weakly developed crest for the depressor muscle (Fig 2).
Exterior (A) and interior (B) views of scuta and exterior (D) and interior (C) view of terga from a Megabalanus azoricus individual sampled in the Azorean archipelago (Terceira Island).
The partial 16S rRNA sequences obtained from the M. azoricus (Mazo CAN03) and M. tulipiformis (Mtul 01) specimens, both sampled from the Canary Islands, were 1143 bp and 1192 bp length. The ribosomal alignment consisted of 11 sequences, 484 characters long and comprising 8 sequences from the available partial mitochondrial 16S rRNA gene from diverse Megabalanus species (GenBank/EMBL/DDBJ accession numbers: KM575947 and KM575948). This limited phylogenetic inference allowed us to verify the genetic distinctiveness and taxonomic congruence of the sample analyzed, belonging to a species without any available phylogenetically relevant molecular information at present. The Megabalanus species (97%) formed a monophyletic group that included Austromegabalanus psittacus and M. tulipiformis in a divergent and basal position. M. azoricus belonged to a clade that includes its closest relatives M. spinosus, and basally, M. occator. The sister clade, which comprises the rest of the species for which 16S rRNA data is available, assembled M. tintinnabulum in a basal position, but with low support (46%), with respect to the derived clade containing M. stultus and the closely related M. volcano and M. californicus (Fig 3).
Phylogenetic tree based on the Bayesian criterion and elaborated with MrBayes from 16SrRNA sequences with distance calculated by the GTR model. Megabalanus sequences available from Genbank were included in the analysis, which incorporated 16S rRNA sequences obtained from M. tulipiformis and M. azoricus specimens (in blue color). Branch support is indicated by the numbers on branches, reflecting the Bayesian analysis posteriori probabilities in percentages. The Elminius modestus sequence was used as an outgroup.
Moreover, the publicly available COX1 barcode data from alternative Megabalanus taxa, not previously considered in the ribosomal dataset, also suggest its divergence with respect to M. azoricus. Thus, the M. azoricus specimen shows a low value of similarity (<86%) with respect to M. zebra in the BOLD repository and to M. rosa (JX503004), M. coccopoma, and M. ajax records in the Genbank database.
The intraspecific variability, calculated from the COX1 partition data set (N = 54, length = 658 bp), was 0.0036 (S.D. = 0.0013), which is in the range of other intraspecific barnacle values: 0.006–0.011 in Pollicipes species  and 0.022 in Balanus glandula .
The population genetic structure of M. azoricus in Macaronesia was inferred from the complete mitochondrial control region sequences acquired from samples (N ≥ 50 individuals) collected along the Macaronesian range, at sites located in the Azores, Madeira, Savage, Canary, and Cabo Verde archipelagos (Table 1, Fig 1). The control region exhibits the expected AT rich (79.35%) and hypervariable profile (24.9% of polymorphic sites). This data set elucidated for population analysis consists of an alignment of 207 sequences, 430 bp length, including 107 polymorphic sites, and 15 gap sites. The control region sequences were deposited in GenBank/EMBL/DDBJ under the accession numbers KM575967-KM576099. All haplotypes from the Cabo Verde sample were private. The nucleotide and haplotype diversity values obtained from the CR in M. azoricus were moderate. The overall species haplotype diversity (h) was 0.9649, being 133 haplotypes with a nucleotide diversity (π) = 0.01813. The population values suggest lower values in the northern Azores population (h = 0.8967; π = 0.01242) compared to those of the southern Cabo Verde population (h = 0.9984; π = 0.02115) (Table 1). Moreover, the h (R2 = 0.735, P = 0.0022) and π (R2 = 0.9035, P = 0.0073) slope values were significant according to the latitudinal regression analysis (Fig 4). As is expected for a codifying sequence, the COX1 h and π values were lower (N = 54, h = 0.809, π = 0.00356). These COX1 sequences were deposited in GenBank/EMBL/DDBJ under the accession numbers KM575949—KM575966. The mean intraspecific p-distances among COX1 and control region (CR) haplotypes were 0.004 and 0.016, respectively.
Regression analysis of haplotype diversity (h) (blue line) and nucleotide diversity (π) (green line) from mitochondrial control region sequences of the Azorean barnacle (Megabalanus azoricus), with respect to sampling site latitudes (Azorean, Madeira, Canary, and Cabo Verde archipelagos).
In all samples, the θ estimates (4Neμ) based on segregating sites exceeded those for π, as is expected for expanding populations. Thus, Tajima’s D estimate was generally negative although without significant values (P > 0.05). Therefore, we could not reject the null model of a neutral genealogy with a constant effective population size, with the exception of the Cabo Verde sample (P = 0.012). For the complete set of populations, D was negative, being −1.78818 (P = 0.005). The Fu’s Fs was significant for all samples whereas the R2 estimators were only significant for the Canary and Cabo Verde samples, with similarly low Fs and R2 values at lower latitudes (Table 1).
The haplotype network revealed three haplogroups (α, β, and γ) disjointed by five mutations. They are equally distributed among the Azores, Madeira, and Canary archipelago samples, whereas in the Cabo Verde sample haplogroup β is absent (Fig 1). The α haplogroup (N = 77, h = 0.985, π = 0.01073) includes two highly frequent haplotypes. One of them is located mainly in the Azores and Canary Islands; the other occurs, in identical proportion, in the Madeira and Canary Islands. Additionally, the majority of the other divergent and rare haplotypes have been detected in the Cabo Verde sample. The β haplogroup (N = 107, h = 0.9607, π = 0.004718) is compact, absent from the Cabo Verde, with a most common haplotype (64%), and forms a star conformation with closely related haplotypes. The third haplogroup, γ (N = 23, h = 0.9921, π = 0.01574), consists of a small number of low frequency haplotypes, mainly from the Cabo Verde sample (Fig 5A). This haplotype relationships and assemblage were also recovered in the neighbor-joining tree where haplogroups were monophyletically clustered and located the α and γ haplogroups in a basal position (Fig 5B). The mean sequence divergence among the haplogroups ranged from 0.019 ± 0.004 to 0.030 ± 0.004, being equidistant haplogroup α, with respect to β and γ. In contrast, mean sequence divergence within haplogroups was only 0.009 ± 0.002. Both Tajima’s D and Fu’s Fs revealed low, significant (P < 0.001) values for the α and β haplogroups as is expected from an expansion event; whereas a non-significant value was observed for haplogroup γ. The R2 values, being lower values for the α and β haplogroups under the demographic expansion scenario, are congruent with the Tajima’s D and Fu’s F values (Table 1).
A. Haplotype network using a median joining algorithm from the Megabalanus azoricus mitochondrial control region DNA sequence alignment. Each code-colored circle represents a single haplotype, whose area is proportional to the haplotype frequency. The length of the connecting lines is proportional to the number of mutational steps. B. Neighbor-joining optimal mid-point rooted tree with the distance measured by the selected Tamura-Nei model, including a rate variation among sites with shape α = 0.21 for the gamma distribution. Branch support was estimated by bootstrapping (2,000 replicates).
Population genetic structure
The computed pairwise FST were only significant (P < 0.001) for the population pairwise comparisons involving the Cabo Verde population with a mean value of 0.366 (S.D. = 0.009). The rest of the populations comparisons were close to 0 and non-significant (Table 2).
In agreement with the FST estimations, significant AMOVA values were only observed when the Cabo Verde sample was included in a separate group from the one containing the Azores, Madeira, and Canary Islands. The majority of percentage variation (65.1%) was significantly allocated within populations within the defined groups, resulting in a ΦST of 0.35. The percentage variation associated with the separation of the Cabo Verde sample from the other populations had (35.33%) an identical and also significant ΦCT of 0.35 (Table 3).
The mismatch distributions estimated from the whole species or sampling sites at the four Macaronesian archipelagos are either bi- or multi-modal, reflecting the population data arrangement containing three divergent haplogroups or currently sympatric lineages. The almost superimposed distributions of the Azores, Madeira, and Canary samples clearly differentiated from the right-displaced Cabo Verde bimodal mismatch distribution (Fig 6A). In contrast, the distributions calculated separately for each M. azoricus haplogroup revealed its historical demography. They were unimodal and consistent with distributions expected for populations undergoing a sudden population expansion (Fig 6B).
Mismatch distributions estimated for (A) each of the sample sites in the four Macaronesian archipelagos (Azores, Madeira, Canary, and Cabo Verde Islands) and (B) for the three identified haplogroups distributed along the Macaronesian sampling sites.
The non-significant SSD and Ri estimates were generally lower in the case of haplogroup distributions, suggesting a better fit with the ones expected for demographic expansion events. The β haplogroup distribution shows a maximum for a small number of differences, reflecting the majority of pairwise comparisons among haplotypes in a star-like conformation. The α and γ haplogroup specific waves were right-displaced, with the highest τ value estimated for the γ distribution (Table 1).
The Bayesian coalescent analysis suggested that the oldest haplogroup γ dates back to 180 Kya (thousands years ago) but with a signal of significant growth after 100 Kya. However, this haplogroup has the smallest population size, almost an order of magnitude less than that of haplogroups α and β. The emergence and exponential growth of haplogroup α also occurred around this date (110 Kya). Haplogroup β was dated back to 40 Kya, showing the highest exponential growth to achieve a present population size similar to that of haplogroup α (Fig 7).
Bayesian skyline plot based on mitochondrial control region coalescent analysis for the Megabalanus azoricus haplogroups.
In geminate barnacle species (Euraphia genus) around the Panamanian Isthmus, a divergence rate of 3.1% My-1 (million years) was estimated for the COX1 gene being used for time divergence estimations in Chthamalus and Pollicipes genera [23, 63, 64]. The present work produced COX1 and control region (CR) comparable data sets for a subsample (N = 54). From these data, we estimated the relationship between COX1 and CR gene mutation rates to apply an indirect estimation of the mutation rate to the CR sequences. This approach was based on the estimation of the ratio between each gene-specific θ (θ = 2Nμ, N = effective population size, and μ = mutation rate for each loci). Because the effective population size (N) is identical among linked loci within the same sample, the ratio between θCOX1 and θCR equals the ratio of mutation rates between loci. The θ estimates were 3.731 and 8.558 for the COX1 and CR data sets, respectively. Consequently, the control region mutation rate (μCR) was 2.294 times the value observed for COX1 gene and, thus, it is estimated as 3.56 x 10-8 per site per year. This estimated μCR value was useful for population time estimation using the t = τ/2u formula, t being the number of generations since demographic expansion, τ the mismatch-distribution parameter value, and u the substitution rate per sequence per generation. Considering a generation time of 0.583 years , the time since demographic expansion has been estimated as 108,000, 61,000, and 224,000 years for the α, β, and γ haplogroups, respectively. The highest lineage divergence was 4.5% (mean = 2.25%) and considering the 3.56% per million years mutation rate, the oldest haplotype divergence dates back to approximately 632 Kya.
The morphological and molecular species identification carried out was necessary to confirm the taxonomic congruence within the analyzed sample set, considering i) the geographically isolated sample sites, ii) the complex taxonomy, with incomplete descriptions, major revisions, and a number of synonymies [2, 9]; iii) the invasive ecology of these taxa [10, 66], and iv) the presence of morphologically related Macaronesian sympatric taxa and forms, with unknown relationships among them . As a result, our Azorean barnacle, Megabalanus azoricus, sample sets have been defined as a well-supported species based on diagnostic morphological characters and molecular data, confirming its taxonomic consistency throughout the Macaronesian region.
The taxonomic assignation of the samples analyzed was based on morphological characterization [2, 8], the sampling from the M. azoricus type locality (Terceira, Azores Islands), and their monophyly with relatively low level of intraspecific genetic variability. Comparison of our mitochondrial sequences with the homologous data available for other Megabalanus species, M. tulipiformis, M. occator, M. spinosus, M. stultus, M. californicus, M. tintinnabulum, M. volcano, M. zebra, M. rosa, M. coccopoma, and M. ajax, suggest the distinctiveness and taxonomic singularity of the present taxa. Moreover, among the species not considered above and without available genetic information only M. vesiculosus, with distinctive characters (e.g. spinose ribs), has an Atlantic distribution, although it is apparently extinct from the Brazilian coast. Among the species analyzed in the 16S rRNA data set, M. azoricus was closely related to M. spinosus, which shows a southern distribution across the Guinean Gulf, the Principe, Sao Tomé, and Annobon islands , suggesting a tropical ancestry for M. azoricus. The phylogenetically divergent M. tulipiformis, with shared distribution in the East Atlantic although absent in the Azores , also shows clearly differentiated characters (e.g. equal orifice and basal diameter) and has a deeper bathymetric range than the intertidal M. azoricus [2, 13].
Despite the limited ribosomal sequence data, both in taxa and sequence length, the present Megabalanus data supplements the extensive barnacle phylogeny inferred in other studies [16, 19], which do not include M. azoricus, M tulipiformis, and other Megabalanus species considered here. These phylogenies included independent molecular and fossil calibrations that support the validity of our COX1 calibration. Thus, Austromegabalanus victoriensis  has been used to calibrate speciation in the Megabalanus genus, which has been dated to 11.6–23 Mya. The highest Megabalanus interspecific distance, among the available COX1 sequences, was 0.215 when comparing the pairwise M. coccopoma and M. ajax. Using the 1.55%/My mutation rate a minimum speciation time for the origin of the Megabalanus genus can be roughly dated to ~14 Mya, which agrees with barnacle divergence time estimations based on fossil calibration .
Genetic diversity and population demography
The haplogroup pattern of ancestral lineages retained after a lineage sorting process falls within the category II in Avise’s classification of phylogeographic patterns. This category is characterized by the presence of highly divergent and broadly sympatric lineages, theoretically occurring in species with a large evolutionary Ne and abundant gene flow . Nevertheless, there is a disparity within haplogroup-specific π values, as also revealed by the mismatch distributions, with the lowest value estimated for haplogroup β. Likewise, the overall species haplotype diversity (h) (0.9649) differs from the higher h value estimated for haplogroup γ (0.9921). The maximum values from haplogroup γ agree with the expected estimates for a population characterized by a stable, large and long-term Ne . In the case of haplogroup β, the low haplotype and nucleotide diversity are consistent with a recent demographic expansion, spatially circumscribed to northern archipelagos, with a single majoritarian haplotype and closely related haplotypes in a star-like structure.
Both π and h values at the species level (π = 0.0181, h = 0.9649) were lower than those estimated for other Cirripedia species. Considering the homologous control region sequence, the stalked barnacle, Pollicipes pollicipes, had a π = 0.0330 and h = 0.9936 . Also in Semibalanus balanoides, π = 0.0298 and h = 0.9967 were estimated from a European sample (UK, N = 43) . Similarly, the π from COX1 (0.0036) is below the mean calculated for Cirripedia species (0.0093) . Higher π and h COX1 values are obtained in other species such as Tetraclita rubescens (π = 0.0094, h = 0.9998), which also included the control region  and Chthamalus malayensis (π = 0.005–0.017 and h = 0.95–1.00) . Furthermore, estimated values are slightly higher for the Japanese populations of congeneric M. volcano (π = 0.0099, h = 0.847) and M. coccopoma (π = 0.00701, h = 0.906) species, but were similar to those for M. rosa (π = 0.00386, h = 0.819) . Repeated demographic bottleneck events and a relatively small population size are likely responsible for the low π and h observed in this insular endemic species  when compared to species that have a larger distribution range.
The negative relationship between latitude and genetic diversity (h and π) observed is a recurrently described pattern, also observed in diverse crustacean species [23, 70] and other marine invertebrates along this latitudinal range . A parallelism with respect to the species richness diversity patterns at tropical latitudes , promoted by diverse debated mechanisms, such as larval dispersal capacity , life history, demography, population structure, rate of molecular evolution, and rate of speciation has been suggested . However, at the intraspecific level, the above cited variables are fixed and the process responsible for this pattern is frequently associated with post-glacial range expansion from southern refugia, and likely relies on low lineage extinction rates at low latitudes .
All of the Tajima’s D population values were negative but also non-significant, with the exception of the Cabo Verde sample. Significant D values were only estimated for the α and β haplogroups, suggesting a recent expansion. The non-significant Megabalanus azoricus D value (−1.15875, P>0.1) was similar to the mean D value (−1.087) estimated from Cirripedia COX1 data . However, this mean value is above the significant D estimation for the species (−1.7882; P<0.01) for haplogroups α and β (−2.46126; P<0.01 and −2.64192; P<0.001; respectively) when using the non-codifying control region sequence. Although a recurrent negative skewness in the empirical distribution of D statistic values for Cirripedia  has been described, the present estimations seem to recover a true expansion dynamics signal. Accordingly, in M. azoricus, negative, significant D values are not universal, thus haplogroup γ shows a non-significant D value as expected for a long standing stable lineage in agreement with inferences from sequence diversity, mismatch distributions, and dating estimates. Moreover, we could not reject the null hypothesis based on the Fu’s Fs statistic of constant size for the haplogroup γ, whereas that of the R2, was a higher, although significant value (P = 0.026) for this haplogroup (Table 1).
The genetic signature of putative demographic bottlenecks associated with extinction and re-colonization dynamics in northern archipelagos contrast with long-term stable population signals in the low latitude Cabo Verde population. Thus, the low number and closely related haplotypes, particularly in the Azores, reflect the genetic diversity erosion in northern latitudes by means of low-frequency haplotype losses, re-colonization associated founder events, and the persistence of common haplotypes. In contrast, the high diversity values in Cabo Verde reflect the long-term persistence of lineages in an evolutionarily stable environment.
The FST and AMOVA analysis of the geographic partition of genetic diversity suggested two opposing patterns. First, a homogeneous gene pool comprises the northern archipelagos of the Azores, Madeira, Savage, and Canary Islands, which exhibit identical distributions of the three haplogroups detected and with the most recent haplogroup β (52%) comprising the majority. Xavier et al.  observed a similarly homogeneous pool at these same locations for the sponge, Phorbas fictius, except for the Madeira differentiated population. Second, an isolated southern population in the Cabo Verde Islands, where genetic diversity is high but haplogroup β is absent. Gene flow between these two population sets is low enough to promote the significant genetic isolation observed between northern and southern populations. The northern populations are subject to common oceanographic and paleoclimatic regimes. These archipelagos are under the influence of the Azores and Canary Currents, they have a similar geological origin including historically emerged intermediate seamounts available as stepping-stones during colonization  and a lot of human-mediated maritime connections . Consequently, shared historical and actual homogenizing processes are likely responsible for the common pattern observed. In contrast, the Cabo Verde archipelago is under a different climate and oceanographic regime with a composite biogeographic character, having both Caribbean and Atlantic-Mediterranean affinities regarding the marine biota . As a result, this biogeographic divergence is to varying degrees reflected in genetic discontinuities observed in related and other marine invertebrates [23, 24, 76]. The observed isolation of Cabo Verde is supported by favorable contemporary hydrographic dynamics influencing planktotrophic larval dispersion . At the northern branch of the subtropical gyre, the Azores current transports North Atlantic Central Water (NACW) eastward. The southeasternmost branch originates the Canary Current (CC), which flows from 33° N to 20–25° N and southward along the African coast. However, it is prevented from reaching Cabo Verde at the Cabo Verde Frontal Zone (CVFZ), which is characterized at the northern limit by the predominance of NACW water, whereas waters south of the CVFZ resemble equatorial conditions . The southwestern flow drives CC waters to the open ocean where it feeds the North Equatorial Current. This hydrographic discontinuity generated by the CVFZ agrees with the genetic divergence observed in populations on both sides of the CVFZ . Therefore, the front presents a crucial physical component of a biologically effective barrier for larval dispersion. Additionally, recruitment is hindered in insular habitats by the intense upwelling systems induced by the southwestward trade winds in the region between Cape Blanc and the Canary Islands .
In contrast, mesoscale hydrographic structures play a key role in the settlement and recruitment of other crustacean species [23, 79], particularly in the critical littoral zone. For example, the intense mesoscale events and the island mass effect in the Azores , Madeira , Canary [81, 82], and Cabo Verde Islands .
Dating and process for haplogroup pattern
Pleistocene climate changes with the associated sea surface temperature (SST) and sea-level fluctuations are a paradigm in marine biogeography and sustain the current phylogeographic hypothesis. Both variables affect the highly sensitive rocky intertidal habitat, particularly for the sessile M. azoricus. During maximum glacial episodes the glacial front reached as far south as 40° N  and is associated with negative consequences for the warm fauna of the Azores archipelago . However, the effect of SST variation on the southernmost archipelagos, particularly Cabo Verde, is probably negligible. Additionally, fluctuations in sea-level alter habitat availability with respect to seamounts .
The small populations inhabiting isolated insular habitats and/or glacial refugia fuel population differentiation . The intensification of the CCA, NEC, and coastal upwelling at the southernmost Macaronesian refuges are conditions favoring insular isolation . Less astringent current patterns and intermediary and dispersed seamounts operating as stepping-stones [28, 76] likely supported a limited northward colonization process.
Within the Macaronesian marine region, it is noteworthy that the paleoclimatic records from ODP Site 658, off Mauritania, agree with the abundant paleo-oceanographic data. Both ln (Si/Al) and records of percentage sedimentary opal reflect the deglacial events, allowing an accurate dating of glacial termination II and the start of interglacial period MIS 5 to 130 Kya, whereas termination III and MIS 7 interglacial occurred 240 Kya . The MIS3 epoch took place between 60 and 24 Kya . The widely distributed haplogroups with the highest polymorphism level (α and γ) are expected to be older than haplogroup β. Moreover, the mismatch distributions, Bayesian coalescent analysis, and the basal position of the α and γ haplogroups within the mid-point rooted phylogenetic tree support this relatively broad temporal dating. Thus, the estimated expansion dates of the ubiquitous α and γ haplogroups place these events within the MIS 5 and MIS 7 interglacial periods, respectively. Similarly, the β haplogroup dating preceded the Last Glacial Maximum, putting it in the MIS 3 interglacial. Coincidentally, demographic expansion of European and African stalked barnacle, Pollicipes pollicipes, populations date back to the same paleoclimatic period 87–118 Kya, with the most recent common ancestor dated between 200 and 170 Kya [23, 64].
Cabo Verde has well-established refugia and ancestral area features , as suggested by the close phylogenetic relationships between M. azoricus and the eastern Atlantic tropical M. spinosus. The higher diversity values along a latitudinal gradient for both the α and γ haplogroups suggest that these lineages originated in the stable refuge habitat of the Cabo Verde islands. Additionally, haplogroup α possesses a limited number (2) of highly frequent haplotypes in contrast to a number of private haplotypes from Cabo Verde, as is expected from few founder events or genetic diversity erosion. In contrast, the most recent haplogroup, β (40–61 Kya), is distributed exclusively throughout northern locations (Canaries, Savage, Madeira, and Azores) suggesting the persistence of an effective oceanographic barrier to gene flow, as detailed above, from MIS3, selection against Cabo Verde subtropical habitat, and/or a strong contemporary human-mediated northern dispersion. This dispersion is sustained by a high level of marine connectivity through fouling on vessels circulating among the northern archipelagos, which is significantly less in Cabo Verde. This human-mediated vagility is responsible for the cosmopolitan distribution of the congeneric species M. coccopoma, M. rosa, and M. volcano  and other Balanidae colonization events in Macaronesia . However, the absence of any shared haplotypes between the Cabo Verde population and the other northern archipelagos agrees with the long-term barrier hypothesis, as suggested by the persistence of the Azores front from interglacial and LGM times .
A similar colonization pattern was suggested for another Macaronesian endemic, Tectarius striatus [76, 92]. Here, Cabo Verde is also considered the ancestral area, with dispersion mediated by counterclockwise currents and available seamounts, and, in contrast, a recently established gene flow barrier between Cabo Verde and the northern archipelagos.
Macaronesian endemism in two biogeographic provinces
The level of local endemism is highly variable throughout Macaronesia. Cabo Verde and the Azores have the lowest species richness and endemism ranking for oceanic islands, in terms of native higher plant species . The biogeographic data shows a low level of Azorean marine endemism [85, 93–95]. Particularly concerning intertidal space occupiers, marine animal diversity and endemism are lower in the Azores compared to the Canaries and Madeira . The stalked barnacle P. pollicipes and C. montagui are absent from the Azorean intertidal, whereas they are recorded from other Macaronesian islands. Similarly, other intertidal taxa such as mussels, patellid limpets, and large trochids are absent or barely represented, likely because of limited food availability for filter feeders .
Until now endemic character and evolutionary origin hypotheses concerning M. azoricus were obscured by the lack of information regarding the relationships between the Azorean type and specimens from the rest of Macaronesia . Thus, the present data support the hypothesis of a widely distributed endemism throughout Macaronesia, with an evolutionary origin in the Cabo Verde Islands, and a historical colonization of the Canaries and Madeira (both are considered the biodiversity nuclei of this region based on the observed endemism levels), and the Azores.
The Macaronesia sensu lato, including Cabo Verde, spans two classic biogeographic provinces, Lusitanian and West African, with their common boundary located in the Cabo Verde Islands . Around central Macaronesia, with accepted zoogeographical affinities, differences in dispersal potential confers varying affinity degrees among close regions. In contrast, distant islands with low affinities, the Azores and particularly Cabo Verde, lack a dispersal support . Congruently, the geographic affinity of the well characterized ichthyofauna shows a primary clustering of the Azores, Madeira, and the Canaries, whereas the highest divergence is observed in Cabo Verde . However, Ophioblennius atlanticus from the Azores and Cabo Verde cluster in a divergent Lusitanian clade differentiated from the West-African province populations .
The concept of a Macaronesian biogeographic entity has been challenged because of its flora, for example, the divergent phytogeographic affinities of the Cabo Verde pteriodophytes . Present data and other phylogeographic analyses of marine invertebrates suggest a similar divergent pattern [23, 24, 76, 92] modified by life cycle, vagility, behavior, and ecology. Further study is necessary to evaluate the phylogeographic hypothesis on a generalized marine pattern defined by the Cabo Verde divergence within a Macaronesian sensu lato biogeographic unit or with respect to the Macaronesian sensu stricto, originating from and sustained by paleoclimatology and oceanography, respectively. Any effort in pattern definition is crucial in guaranteeing the conservation of genetic diversity and the sustainability of marine resources, particularly for those species that are heavily exploited, as is the case for M. azoricus.
We are grateful to Pedro López (NATURALIA-CV), Rui Freitas (Uni-CV) and Prof. Dr. Alberto Brito (ULL) for assistance with sample collection; and Msc Raül Triay Portella for photographic assistance.
Conceived and designed the experiments: JQ PM MR-M NG-H. Performed the experiments: JQ PM LPD. Analyzed the data: JQ PM. Contributed reagents/materials/analysis tools: JAG CA EL RA GPC. Wrote the paper: JQ PM.
- 1. Newman WA, Ross A. Revision of the Balanomorph barnacles; including a catalog of the species San Diego Soc Nat Hist Mem. 1976;9:108.
- 2. Henry DP, McLaughlin PA. The recent species of Megabalanus (Cirripedia: Balanomorpha) with special emphasis on Balanus tintinnabulum (Linnaeus) senso lato. Zoologische Verhanelingen. 1986;235:3–60.
- 3. Darwin C. A monograph on the sub-class Cirripedia,with figures of all species. The Balanidae (or sessile cirripedes); the Verrucidae, etc. London: Ray Society; 1854. 1–684 p.
- 4. Hoek PPC. The Cirripedia of the Siboga-Expedition, B: Cirripedia sessilia. Siboga-Expedition Monographies. 1913;31b:129–275.
- 5. Pilsbry HA. The sessile barnacles (Cirripedia) contained in the collections of the U.S. National Museum; including a monograph of the American species. Bulletin US National Museum. 1916;93:1–76.
- 6. Santos RS, Hawkins S, Monteiro LR, Alves M, Isidro EJ. Marine research, resources and conservation in the Azores. Aquat Conserv: Mar Freshwat Ecosyst. 1995;5(4):311–354.
- 7. López DA, López BA, Pham CK, Isidro EJ, De Girolamo M. Barnacle culture: background, potential and challenges. Aquacult Res. 2010;41(10):e367–e375.
- 8. Dionísio M, Costa A, Rodrigues A. Re-examination of Megabalanus azoricus (Cirripedia: Balanomorpha): a natural and edible resource to preserve. Mar Biodivers Rec. 2012;5:6.
- 9. Southward AJ. New observations on barnacles (Crustacea: Cirripedia) of the Azores region. Arquipélago Life and Marine Sciences. 1998;16(A):11–27. pmid:25606133
- 10. Yamaguchi T, Prabowo RE, Ohshiro Y, Shimono T, Jones D, Kawai H, et al. The introduction to Japan of the Titan barnacle, Megabalanus coccopoma (Darwin, 1854) (Cirripedia: Balanomorpha) and the role of shipping in its translocation. Biofouling. 2009;25(4):325–333. pmid:19219673
- 11. Dionisio M, Rodrigues A, Costa ANA. Reproductive biology of Megabalanus azoricus (Pilsbry), the Azorean Barnacle. Invertebr Reprod Dev. 2007;50(3):155–162.
- 12. Wirtz P, Araújo R, Southward AJ. Cirripedia of Madeira. Helgol Mar Res. 2006;60(3):207–212.
- 13. González JA, Martín L, Herrera R, González-Lorenzo G, Espino F, Barquín-Diez J, et al. Cirripedia of the Canary Islands: distribution and ecological notes. J Mar Biol Assoc UK. 2011;92(01):1–13.
- 14. Morri C, Cattaeno-Vietti R, Sartoni G, Banchi CN. Shallow epibenthic communities of Ilha do Sal (Cape Verde Archipelago, eastern Atlantic). Arquipel Life Mar Sci Suppl. 2000;2(A):157–165.
- 15. Palumbi SR. What can molecular genetics contribute to marine biogeography? An urchin's tale. J Exp Mar Biol Ecol. 1996;203(1):75–92.
- 16. Perez-Losada M, Hoeg JT, Crandall KA. Unraveling the evolutionary radiation of the thoracican barnacles using molecular and morphological evidence: a comparison of several divergence time estimation approaches. Syst Biol. 2004;53(2):244–264. pmid:15205051
- 17. Hayashi R, Chan BK, Simon-Blecher N, Watanabe H, Guy-Haim T, Yonezawa T, et al. Phylogenetic position and evolutionary history of the turtle and whale barnacles (Cirripedia: Balanomorpha: Coronuloidea). Mol Phylogenet Evol. 2013;67(1):9–14. pmid:23306306
- 18. Perez-Losada M, Harp M, Hoeg JT, Achituv Y, Jones D, Watanabe H, et al. The tempo and mode of barnacle evolution. Mol Phylogenet Evol. 2008;46(1):328–346. pmid:18032070
- 19. Simon-Blecher N, Huchon D, Achituv Y. Phylogeny of coral-inhabiting barnacles (Cirripedia; Thoracica; Pyrgomatidae) based on 12S, 16S and 18S rDNA analysis. Mol Phylogen Evol. 2007;44(3):1333–1341. pmid:17560131
- 20. Chan BKK, Tsang LM, Shih FL. Morphological and genetic differentiations of the stalked barnacle Heteralepas japonica Aurivillius, 1892, with description of a new species of Heteralepas. Raffles Bull Zool. 2009;20:83–95.
- 21. Chan BKK, Tsang LM, Chu KH. Morphological and genetic differentiation of the acorn barnacle Tetraclita squamosa (Crustacea, Cirripedia) in East Asia and description of a new species of Tetraclita. Zool Scr. 2007;36(1):79–91.
- 22. Chan BKK, Tsang LM, Chu KH. Cryptic diversity of the Tetraclita squamosa complex (Crustacea: Cirripedia) in Asia: Description of a new species from Singapore. Zool Stud. 2007;46(1):46–56.
- 23. Quinteiro J, Rodríguez-Castro J, Rey-Méndez M. Population genetic structure of the stalked barnacle Pollicipes pollicipes (Gmelin, 1789) in the northeastern Atlantic: influence of coastal currents and mesoscale hydrographic structures. Mar Biol. 2007;153(1):47–60.
- 24. Quinteiro J, Rodríguez-Castro J, López P, López-Jurado LF, González-Henríquez N, Rey-Méndez M. Morphological and molecular characterization of a new species of Atlantic stalked barnacle (Scalpelliformes: Pollicipedidae) from the Cape Verde Islands. J Mar Biol Assoc UK. 2011;91(07):1537–1545.
- 25. Tsang LM, Wu TH, Shih HT, Williams GA, Chu KH, Chan BK. Genetic and morphological differentiation of the Indo-West Pacific intertidal barnacle Chthamalus malayensis. Integr Comp Biol. 2012;52(3):388–409. Epub 2012/04/24. pmid:22523127
- 26. Flight PA, O’Brien MA, Schmidt PS, Rand DM. Genetic Structure and the North American Postglacial Expansion of the Barnacle, Semibalanus balanoides. J Hered. 2012;103(2):153–165. pmid:21885571
- 27. Dawson MN, Grosberg RK, Stuart YE, Sanford E. Population genetic analysis of a recent range expansion: mechanisms regulating the poleward range limit in the volcano barnacle Tetraclita rubescens. Mol Ecol. 2010;19(8):1585–1605. pmid:20345681
- 28. Fernández-Palacios JM, de Nascimento L, Otto R, Delgado JD, García-del-Rey E, Arévalo JR, et al. A reconstruction of Palaeo-Macaronesia, with particular reference to the long-term biogeography of the Atlantic island laurel forests. J Biogeogr. 2011;38(2):226–246.
- 29. Vanderpoorten A, Rumsey FJ, Carine MA. Does Macaronesia exist? Conflicting signal in the bryophyte and pteridophyte floras. Am J Bot. 2007;94(4):625–639.: pmid:21636431
- 30. Lloris D, Rubacado J, Figueroa H. Biogeography of the Macaronesian Ichthyofauna. Bol Mus Munic Funchal. 1991;43(234):191–241.
- 31. Stearn WT. Philip Barker and Canarian Botany. Monogr Biol Canar. 1973;4:15–29.
- 32. Engler A. Versuch einer Entwicklungsgeschich, insbe- sondere der Florengebiete seit der Tertiärperiode. Leipzig, Germany.: Engelmann; 1879.
- 33. Danserau P. Etudes macaronesiennes I: geographie des Cryptogames vasculaires. Agronomic Lusit. 1961;23:151–181.
- 34. Sunding P. Origins of the Macaronesian flora. In: Bramwell D, editor. Plants and islands. London, UK.: Academic Press; 1979. p. 13–40.
- 35. Whittaker RJ, Fernández-Palacios JMa. Island biogeography: ecology, evolution, and conservation. 2nd ed. Oxford; New York: Oxford University Press; 2007. xii, 401 p. p.
- 36. Leung PTY, Morton B, Ng WC. Meso-scale genetic structure of the intertidal, crevice-dwelling, stalked barnacle Ibla cumingi (Crustacea: Cirripedia): an interplay of life history and local hydrographic conditions. J Mar Biol Assoc UK. 2011;91(01):173–180.
- 37. Palumbi SR. Genetic divergence, reproductive isolation, and marine speciation. Annu Rev Ecol Syst. 1994;25:547–572.
- 38. OSPAR. Quality Status Report 2010. London: OSPAR Commission, 2010.
- 39. 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):249–299.
- 40. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 1997;24:4876–4882. pmid:9396791
- 41. Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp Ser. 1999;41:95–98.
- 42. Ratnasingham S, Herbert PDN. BOLD: The Barcode of Life Data System. Mol Ecol Notes. 2007;(7):355–364. pmid:18784790
- 43. Ronquist F, Teslenko M, van der Mark P, Ayres DL, Darling A, Höhna S, et al. MrBayes 3.2: Efficient bayesian phylogenetic inference and model choice across a large model space. Syst Biol. 2012.
- 44. Hasegawa M, Kishino H, Yano T. Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. J Mol Evol. 1985;22(2):160–174. pmid:3934395
- 45. Darriba D, Taboada GL, Doallo R, Posada D. jModelTest 2: more models, new heuristics and parallel computing. Nat Methods. 2012;9(8):772. pmid:22847109
- 46. Excoffier L, Laval G, Schneider S. Arlequin ver. 3.0: An integrated software package for population genetics data analysis. Evol Bioinform Online. 2005;1:47–50.
- 47. Librado P, Rozas J. DnaSP v5: A software for comprehensive analysis of DNA polymorphism data. Bioinformatics. 2009;25(11):1451–1452. pmid:19346325
- 48. Rogers AR, Harpending H. Population growth makes waves in the distribution of pairwise genetic differences. Mol Biol Evol. 1992;9(3):552–569. pmid:1316531
- 49. Slatkin M, Hudson RR. Pairwise comparisons of mitochondrial DNA sequences in stable and exponentially growing populations. Genetics. 1991;129(2):555–562. pmid:1743491
- 50. Schneider S, Excoffier L. Estimation of past demographic parameters from the distribution of pairwise differences when the mutation rates vary among sites: Application to human mitochondrial DNA. Genetics. 1999;152:1079–1089. pmid:10388826
- 51. Harpending HC. Signature of ancient population growth in a low-resolution mitochondrial DNA mismatch distribution. Hum Biol. 1994;66(4):591–600. pmid:8088750
- 52. Tajima F. Statistical method for testing the neutral mutation hypothesis by DNA Polymorphism. Genetics. 1989;123(3):585–595. pmid:2513255
- 53. Fu YX. Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics. 1997;147(2):915–925. pmid:9335623
- 54. Tajima F. The effect of change in population size on DNA polymorphism. Genetics. 1989;123:597–601. pmid:2599369
- 55. Ramos-Onsins SE, Rozas J. Statistical properties of new neutrality tests against population growth. Mol Biol Evol. 2002;19(12):2092–2100. pmid:12446801
- 56. 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;131:479–491. pmid:1644282
- 57. Drummond AJ, Rambaut A, Shapiro B, Pybus OG. Bayesian coalescent inference of past population dynamics from molecular sequences. Mol Biol Evol. 2005;22(5):1185–1192. pmid:15703244
- 58. Drummond AJ, Suchard MA, Xie D, Rambaut A. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Mol Biol Evol. 2012;29(8):5.
- 59. Bandelt HJ, Forster P, Rohl A. Median-joining networks for inferring intraspecific phylogenies. Mol Biol Evol. 1999;16(1):37–48. pmid:10331250
- 60. Kumar S, Nei M, Dudley J, Tamura K. MEGA: a biologist-centric software for evolutionary analysis of DNA and protein sequences. Brief Bioinform. 2008;9(4):299–306. pmid:18417537
- 61. Tamura K, Nei M. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol Biol Evol. 1993;10(3):512–526. pmid:8336541
- 62. Sotka EE, Wares JP, Barth JA, Grosberg RK, Palumbi SR. Strong genetic clines and geographical variation in gene flow in the rocky intertidal barnacle Balanus glandula Mol Ecol. 2004;13(8):2143–2156. pmid:15245390
- 63. Wares JP. Patterns of speciation inferred form mitochondrial DNA in North American Chthmalus (Cirripedia: Balanomorpha: Chthmaloidea). Mol Phylogen Evol. 2001;18:104–116.
- 64. Campo D, Molares J, Garcia L, Fernandez-Rueda P, Garcia-Gonzalez C, Garcia-Vazquez E. Phylogeography of the European stalked barnacle (Pollicipes pollicipes): identification of glacial refugia. Mar Biol. 2009;157(1):147–156.
- 65. Pham CK, De Girolano M, Isidro EJ. Recruitment and growth of Megabalanus azoricus (Pilsbry, 1916) on artificial substrates: first steps towards commercial culture in the Azores. Arquipel Life Mar Sci Suppl. 2011;28:47–56.
- 66. Carlton JT, Newman WA, Pitombo F. Barnacle invasions: Introduced, cryptogenic, and tanga expanding Cirripedia of North and South America. In: Galil BS, Clark PF, Carlton J, editors. In the Wrong Place—Alien Marine Crustaceans: Distribution, Biology and Impacts. Invading Nature, Springer Series in Invasion Ecology: Springer; 2011. p. 159–214.
- 67. Buckeridge JS. Fossil barnacles (Cirreipedia: Thoracica) of New Zealand and Australia. NZ Geol Surv Paleontol. 1983;50:1–151.
- 68. Avise JC. Phylogeography: the history and formation of species. Cambridge, MA: Harvard University Press; 2000. 447 p.
- 69. Ewers C, Wares JP. Examining an outlier: molecular diversity in the Cirripedia. Integr Comp Biol. 2012.
- 70. Wares JP. Insights from Population Genetics: Are all Crustaceans Created Equal? J Crust Biol. 2011;31(2):339–344.
- 71. Xavier RJ, van Soest RWM, Breeuwer JAJ, Martins AMF, Menken SBJ. Phylogeography, genetic diversity and structure of the poecilosclerid sponge Phorbas fictitius at oceanic islands. Contrib Zool. 2010;79(3):10.
- 72. Foltz DW. Invertebrate species with nonpelagic larvae have elevated levels of nonsynonymous substitutions and reduced nucleotide diversities. J Mol Evol. 2003;57(6):607–612. pmid:14745529
- 73. Dowle EJ, Morgan-Richards M, Trewick SA. Molecular evolution and the latitudinal biodiversity gradient. Heredity (Edinb). 2013;110(6):501–510. pmid:23486082
- 74. Martin PR, McKay JK. Latitudinal variation in genetic divergence of populations and the potential for future speciation. Evolution. 2004;58(5):938–945. pmid:15212375
- 75. Morton B, Britton JC. Origins of the Azorean intertidal biota: The significance of introduced species, survivors of chance events. Arquipel Life Mar Sci Suppl. 2000;2(A):22–51.
- 76. Van den Broeck H, Breugelmans K, De Wolf H, Backeljau T. Completely disjunct mitochondrial DNA haplotype distribution without a phylogeographic break in a planktonic developing gastropod. Mar Biol. 2007;153(3):421–429.
- 77. Pastor MV, Peña-Izquierdo J, Marrero-Díaz A. Meridional changes in water mass distributions off NW Africa during November 2007/2008. Cienc Mar. 2012;38:223–244.
- 78. Mittelstaedt E. The ocean boundary along the northwest African coast: circulation and oceanographic properties at the sea surface. Prog Oceanogr. 1991;26:307–355.
- 79. O'Riordan RM, Arenas F, Arrontes J, Castro JJ, Cruz T, Delany J, et al. Spatial variation in the recruitment of the intertidal barnacles Chthamalus montagui Southward and Chthamalus stellatus (Poli) (Crustacea: Cirripedia) over an European scale. J Exp Mar Biol Ecol. 2004;304(2):243–264.
- 80. Caldeira RMA, Groom S, Miller P, Pilgrim D, Nezlin NP. Sea-surface signatures of the island mass effect phenomena around Madeira Island, Northeast Atlantic. Remote Sens Environ. 2002;80(2):336–360.
- 81. Peña-Izquierdo J, Pelegrí JL, Pastor MV, Castellanos P, Emelianov M, Gasser M, et al. The continental slope current system between Cape Verde and the Canary Islands. Sci Mar. 2012;76(S1):65–78.
- 82. Ruiz S, Pelegrí JL, Emelianov M, Pascual A, Mason E. Geostrophic and ageostrophic circulation of a shallow anticyclonic eddy off Cape Bojador. J Geophys Res (C Oceans). 2014;119(2):1257–1270.
- 83. Meunier T, Barton ED, Barreiro B, Torres R. Upwelling filaments off Cap Blanc: Interaction of the NW African upwelling current and the Cape Verde frontal zone eddy field? J Geophys Res (C Oceans). 2012;117(C8):C08031.
- 84. Ruddiman WF, McIntyre A. Northeast Atlantic Paleoclimatic Changes over the Past 600,000 Years. Geol Soc Am Mem. 1976;145:111–146.
- 85. Briggs JC. Global biogeography. Amsterdam; New York: Elsevier; 1995. xvii, 452 p. p.
- 86. Flores J-A, Bárcena MA, Sierro FJ. Ocean-surface and wind dynamics in the Atlantic Ocean off Northwest Africa during the last 140 000 years. Palaeogeogr, Palaeoclimatol, Palaeoecol. 2000;161:459–478.
- 87. Meckler AN, Sigman DM, Gibson KA, Francois R, Martinez-Garcia A, Jaccard SL, et al. Deglacial pulses of deep-ocean silicate into the subtropical North Atlantic Ocean. Nature. 2013;495(7442):495–498. pmid:23538831
- 88. Huber C, Leuenberger M, Spahni R, Flückiger J, Schwander J, Stocker TF, et al. Isotope calibrated Greenland temperature record over Marine Isotope Stage 3 and its relation to CH4. Earth Planet Sci Lett. 2006;243:504–519.
- 89. Hewitt G. The genetic legacy of the Quaternary ice ages. Nature. 2000;405(6789):907–913. pmid:10879524
- 90. Torres P, Costa A, Dionísio M. New alien barnacles in the Azores and some remarks on the invasive potential of Balanidae. Helgol Mar Res. 2011:1–10.
- 91. Rogerson M, Rohling EJ, Weaver PPE, Murray JW. The Azores Front since the Last Glacial Maximum. Earth Planet Sci Lett. 2004;222(3–4):779–789.
- 92. De Wolf H, Verhagen R, Backeljau T. Large scale population structure and gene flow in the planktonic developing periwinkle, Littorina striata, in Macaronesia (Mollusca: Gastropoda). J Exp Mar Biol Ecol. 2000;246(1):69–83. pmid:10699219
- 93. Briggs JC. Oceanic islands, endemism and marine paleotemperatures. Syst Zool. 1966;15:153–163.
- 94. Briggs JC. Marine zoogeography. Ehrlich PR, Holm RW, editors. New-York: McGraw-Hill; 1974.
- 95. Hawkins SJ, Corte-Real HBSM, Pannacciulli FG, Weber LC, Bishop JDD. Thoughts on the ecology and evolution of the intertidal biota of the Azores and other Atlantic islands. Hydrobiologia. 2000;440(1):3–17.
- 96. Muss A, Robertson DR, Stepien CA, Wirtz P, Bowen BW. Phylogeography of Ophioblennius: the role of ocean currents and geography in reef fish evolution. Evolution. 2001;55(3):561–572. pmid:11327163