Population Genetics of the Deep-sea Acorn Barnacle Bathylasma hirsutum (Hoek, 1883) and the First Report of its Affiliation with a Hydrothermal Vent Field

Abstract

Confined by the Mid-Atlantic Ridge and the European continental shelf, the deep-sea acorn barnacle Bathylasma hirsutum (Hoek, 1883) lives in the northeast Atlantic deep sea, where it has been frequently reported in high current areas. Cemented to a solid substrate during its entire adult life, the species can only disperse by means of planktotrophic nauplius larvae. This study reports on the occurrence, ecology and genetic connectivity of B. hirsutum from four sites within the northeastern Iceland Basin and presents the first record of the species living affiliated with hydrothermal vent field on the Reykjanes Ridge axis. Vent-associated specimens were found to differ extrinsically from their naturally shaded conspecifics by a prominent brown-black shell precipitate. Energy Dispersive Spectroscopy revealed ferromanganese oxides to be the main component of these shell precipitates. Morphometric measurements of shell plates revealed specimens from the vent-associated habitat to be smaller compared to non-venting sites. Molecular species delimitation based on the mitochondrial COI and nuclear EF1 genetic markers aided species identification and revealed a low intraspecific genetic variability. Our findings suggest a pronounced genetic connectivity of B. hirsutum within the studied region and provide a first step towards a biogeographic study. As such, habitats of hydrothermal influence along the Mid-Atlantic Ridge are discussed as possible niches, as are deep-sea basins in the western Atlantic. In light of the reported affiliation with hydrothermal activity, we elaborate on the potential for the sister species Bathylasma corolliforme (Hoek, 1883) and Bathylasma chilense Araya & Newman, 2018 to utilise equivalent habitats in the Antarctic and Pacific Ocean, respectively. Our record of the unacquainted ecological niche occupation for B. hirsutum emphasises the need for further research on bathylasmatid acorn barnacles along the extensive Mid-Atlantic Ridge, where many biological communities remain to be discovered.

BACKGROUND

During the past 30 years, more than sixty hydrothermal vent systems have been discovered along the Mid-Atlantic Ridge (MAR), known as the longest obliquely spreading section of the global midocean ridge systems (Beaulieu and Szafrański 2020; Desbruyères et al. 2000 2006; German et al. 1994; Taylor et al. 2021; Wheeler et al. 2013). Elaborate studies on dominant faunal compositions of MAR vent communities, such as polychaetes (Kongsrud et al. 2013; Shields and Blanco-Perez 2013; Sigvaldadóttir and Desbruyères 2003), bivalves (Cravo et al. 2007; Duperron et al. 2006; Ó Foighil et al. 2001) and anemones (Fabri et al. 2011; Fautin and Barber 1999; López-González et al. 2003; Van Dover 1995), have contributed extensively to a better understanding of these fragile ecosystems and to map their particular biodiversity (Biscoito et al. 2006; Boschen-Rose and Colaço 2021; Desbruyères et al. 2000; Gebruk et al. 1997 2010). The Reykjanes Ride represents a ~950 km long submarine continuation of the MAR with an axial depth ranging from the shoreline southwest of the volcanic Iceland Peninsular to more than 2000 m at the Bight Fracture Zone (Keeton et al. 1997; Searle et al. 1998; Fig. 1). Due to erupted lava from volcanic seamounts, the ridge substrate is mainly composed of volcanic rock and pillow lava, providing hard-bottom habitats for a wide range of species. Until recently, the only recognised hydrothermal active field along this extensive ridge was the Steinahóll vent field (63°N) at ~300 m depth (German et al. 1994). Late bathymetric mapping and a detailed description of Steinahóll added another three vent sites to the list of hydrothermal activity in this fragile area (Taylor et al. 2021). Current research activities near 59°N lead to the discovery of yet another active vent field situated on the edge of a large, faulted seamount at ~650 m depth (Brix et al. 2020). Proximal to this novel hydrothermal vent field on the Reykjanes Ridge, the deep-sea acorn barnacle Bathylasma hirsutum (Hoek, 1883) was found attached to several volcanic boulders. Unlike the presence of characteristic vent barnacles in hydrothermal systems of the Pacific and Indian Ocean, Herrera et al. (2015) inferred cirripedes to be generally absent from MAR hydrothermal vent systems. Our findings present the first record of a barnacle species associated with hydrothermal activity on the northern section of the MAR. At current, the two groups of barnacles associated with deep-sea hydrothermal vents encompass all members of the superfamily Neolepadoidea as well as three species of the balanomorph genus Eochionealasmus Yamaguchi, 1990 (Chan et al. 2021 2020; Yamaguchi and Newman 1990 1997a b). We introduce a third group of balanomorph barnacle, the genus Bathylasma Newman & Ross, 1971, with at least one opportunist species occupying a hydrothermal habitat. Along with the species of study, the genus Bathylasma Newman & Ross, 1971 currently comprises three extant species, namely B. alearum (Foster, 1978), B. chilense Araya & Newman, 2018 and B. corolliforme (Hoek, 1883), distributed at depths of 37–2000 m from the coasts of New Zealand, Chile, and the Antarctic Peninsula, respectively (Araya and Newman 2018; Foster 1978; Meyer et al. 2021).

Fig. 1.

Distributional range of the deep-sea acorn barnacle Bathylasma hirsutum in the north-east Atlantic. Literature records are included from the: (1) Reykjanes Ridge, (2) Faroe-Bank Channel, (3) Wyville Thomson Ridge, (4) Rockall Trough, (5) Rosemary Bank, (6) George Bligh Bank, (7) Hatton Bank, (8) Anton Dohrn Seamount, (9) Hebrides terrace, (10) Celtic Shelf, (11) Gulf of Cádiz, (12) Great Meteor Seamount, and (13) Azores Archipelago. Schematic of the main circulation patterns adapted from Koman et al. (2020) and Puerta et al. (2020). Red arrows depict surface waters of North Atlantic Current (NAC) origin, with a branch of the Azores Current (AC). Yellow arrows depict Mediterranean Outflow Water (MOW). Blue arrows indicate deep-water pathways of the Iceland Scotland Overflow Water (ISOW) flowing towards the Faroe Bank Channel (FBC), as well as overflow waters that run in parallel to the East Reykjanes Ridge Current (ERRC). Orange arrows represent surface waters of the Irminger Current (IC) with mixtures of Arctic and Atlantic waters.

Distributional range of Bathylasma hirsutum

Bathylasma hirsutum was first described from 944 m depth south of the Wyville Thomson Ridge (59°40'N 07°21'W; Hoek 1883). The species is nested within the family Bathylasmatidae Newman & Ross, 1971, which includes sole deep-sea species of six genera (Chan et al. 2021; Jones 2000; Newman and Ross 1971). In the history of the controversial position of the Bathylasmatidae, Jones (2000), based on their morphological characters, classified the genera Hexelasma Hoek, 1913 and Bathylasma under the Pachylasmatoidea (now Coronuloidea Leach, 1817). In Chan et al. (2017), molecular evidence suggested the clade of Hexelasma and Bathylasma to be sister to the Tetraclitidae Gruvel, 1903, nested within the same clade. Finally, the comprehensive molecular approach by Chan et al. (2021) suggested that Bathylasmatidae is a discrete family, nested within the Coronuloidea.

Bathylasma hirsutum has a distributional range restricted to the North-East (NE) Atlantic, where it inhabits hard-bottom habitats in high current areas between 384–1829 m depth (Fig. 1). The species is known to occur along the Reykjanes Ridge (Copley et al. 1996; Murton et al. 1995; Poltarukha and Zevina 2006b), in addition to on banks and seamounts in the Rockall Trough, Rosemary Bank, George Bligh Bank, Hatton Bank, Anton Dohrn Seamount and Hebrides Terrace (Duineveld et al. 2007; Gage 1986; Narayanaswamy et al. 2006 2013; Young 2001). Further records extend to the south along the Celtic Shelf (Southward and Southward 1958; Young 1998) towards the Gulf of Cádiz (Foster and Buckeridge 1995) and near Tenerife (Gruvel 1903). Along the MAR, B. hirsutum has repeatedly been reported from the Azores Archipelago (Gruvel 1920; Poltarukha and Zevina 2006a; Southward 1998; Young 1998) as well as from the Great Meteor Seamount (Poltarukha and Zevina 2006a; Young 2001). The species populates a wide range of substrates, such as dropstones (Duineveld et al. 2007), telegraph cables (Gruvel 1903; Southward and Southward 1958), and echinoid spines (Hoek 1883; Southward and Southward 1958), as well as living and dead corals (Southward 1998). Furthermore, shell debris from dead barnacles that accumulates in the periphery of the substratum they are cemented to provides a solid layer, even on soft sediments, which can be re-populated by recruits (Bullivant and Dearborn 1967; Dayton et al. 1982; Foster and Buckeridge 1995; Gage and Tyler 1991).

Larval biology and shell growth

Cemented to a solid substrate during their entire adult life, barnacles can only disperse by means of larvae. In general, they develop through a maximum of six naupliar stages which are followed by metamorphosis into the terminal cypris stage (Dayton et al. 1982; Høeg and Møller 2006; Pineda et al. 2021; Walker et al. 1987). Adapted to specific habitats, cyprids show a broad array of sensory setae and attachment organs across species, facilitating them to sense the chemical environment (Bielecki et al. 2009; Maruzzo et al. 2012) and to bipedally move across the substratum (Lagersson and Høeg 2002; Walker et al. 1987; Yorisue et al. 2013 2016). Eventually, the cyprid induces the final cementation and develops into a juvenile sedentary barnacle (Aldred et al. 2018; Crisp 1955; DiBacco et al. 2011). In the deep sea, several species have developed a range of deviant larval development strategies (Yorisue et al. 2013). For example, some species hatch as cyprids which are immediately disposed to search for an attachment site (Buhl-Mortensen and Høeg 2006). Others, such as the endemic vent barnacle of the genus Neoverruca Newman 1989, develop lecithotrophic nauplii to ensure dispersal over far distances with a long larval development (Southward and Jones 2003). In addition, a special habitat adaptation in vent barnacles speeds up the metamorphosis into the sensory cyprid as their ontogeny responds to elevated water temperatures (Watanabe et al. 2004 2006; Yorisue et al. 2013). Currently, nothing is known about the ontogeny of the deep-sea barnacle B. hirsutum. Single larval stages of the Antarctic B. corolliforme have been examined in specimens from the McMurdo Sound region (Dayton et al. 1982; Foster 1989). Due to their relatedness, the naupliar and cyprid anatomy is suggested to be comparable in both species. Besides their conventional balanomorph anatomy which is explored in Dayton et al. (1982) and Foster (1989), the larvae of B. corolliforme can be distinguished from other balanomorph species by having a naupliar eye and an overall larger size. According to Foster (1989), larval development likely exceeds three weeks to complete, comparable to the boreal species Semibalanus balanoides (Linnaeus, 1767) which is suggested to have a planktonic existence of 3–6 weeks (Barnes and Barnes 1959 1958). The cypris develops into an undivided chitinous annulus with a single hirsute ring, including the aperture with terga and scuta adjacent to which a pair of photoreceptors is found (Dayton et al. 1982). Attachment, metamorphosis of settled cyprids and early post-larval mortality are the most critical steps and determine barnacle recruitment (Pineda et al. 2002; Thiyagarajan et al. 2002). As the barnacle grows, it produces distinct growth increments that undulate parallel to the basal margin and run continuously on the exterior of each parietal plate (Anderson 1994; Fig. 2). These increments are referred to as either growth lines (Checa et al. 2019; Varfolomeeva et al. 2008) or growth ridges (Anderson 1994; Blomsterberg et al. 2004; Bourget 1987; Bourget and Crisp 1975). With the terminology being used interchangeably in literature, the authors have agreed upon using the term ‘growth ridges’ throughout. In B. hirsutum, the growth ridges are well developed and hirsute, being covered with numerous chitinous setae projecting outwards from the shell plate, aggregated on small basal domes (Fig. 2H). According to Bourget and Crisp (1975), the rate of ridge formation corresponds roughly to the growth rate of barnacles. However, the ontogenic age of deep-sea species might be affected by the fluctuating food supply in the otherwise stable physical deep-sea environment (Anderson 1994; Foster 1983; Tyler and Young 1992). As generalist feeders, B. hirsutum captures suspended particulate matter using an erected cirral fan which it is able to adjust towards the prevailing water currents (Southward and Southward 1958; Dayton et al. 1982; Gallego et al. 2017). Adult specimens of B. hirsutum can grow more than 8 cm in height and have a white-yellow shell plate (Fig. 2A–C, G). The barnacle is of conical shape with the diameter of the orifice being less than half of the diameter of the membranous base (Newman and Ross 1971; Southward and Southward 1958). Main species-specific characteristics are narrow carinolateral plates, an elongate tergum, and compartments of the orifice which do not diverge from one another, as for example seen in B. corolliforme (Bullivant & Dearborn, 1967 plate 14; Dayton et al., 1982 fig. 3C; Meyer, 2020 fig. 3-3). The parietal plates of B. hirsutum show a wide plasticity, depending on the attachment site as well as the space given for the animal to grow. As many acorn barnacles, B. hirsutum is of gregarious nature and tends to settle crowded with epizoic growth, i.e., one established animal has one or several specimens laterally attached (Anderson 1994).

Fig. 2.

Plate organisation and general anatomy of B. hirsutum. External and internal view of A: carina, B: rostro-lateral, C: rostrum, D: tergum, E: scutum. F: Soft-bodied animal with serrated cirral appendages. G: Top view of an intact specimen with a carino-rostral diameter of 2.5 cm; c: carina, cl: carino-lateral, rl: rostro-lateral, r: rostrum, t: tergum, s: scutum. H SEM image of a shell plate showing growth ridges (gr) with protruding chitinous setae. Scale bars: A–C = 1.5 cm; D–F = 1 cm; H = 500 μm.

Fig. 3.

Collection sites of Bathylasma hirsutum targeted during the research cruise POS456 west of the Faroe Bank Channel (site A), as well as two expeditions along the Reykjanes Ridge MSM75 (sites B, C), and SO276 (site D). The white cross indicates the hydrothermal IceAGE vent field.

Scope of study

Processes involving successful recruitment, dispersal of planktotrophic larvae and settlement as benthic animals with future reproductive success determine the extent to which barnacle populations are genetically and biologically connected across their distributional range (Boschen-Rose and Colaço 2021; Cowen et al. 2007; Pineda et al. 2007). As these benthopelagic interactions are highly complex and challenging to assess, especially with regard to deep-sea habitats (Etter and Bower 2015), the genetic connectivity of B. hirsutum within the NE Atlantic is completely unknown. In the northeastern Iceland Basin, surface waters of the North Atlantic Current converge with deeper water masses from the Iceland-Faroe Ridge into the East Reykjanes Ridge Current (EERC), acting as a boundary between the Iceland Basin and the eastern flank of the Reykjanes Ridge (Koman et al. 2020). As the broad EERC flows southwest towards the Bight Fracture Zone, it crosses the ridge into the Irminger Basin (Fig. 1). This study reports on the opportunistic inhabitation of a hydrothermally influenced site for the species B. hirsutum and elaborates on habitat differences by comparing barnacle size distributions per site and depth, as well as chemical components of parietal shell plates from all sites. Furthermore, we attempt to investigate the genetic connectivity of B. hirsutum from four sites within the Iceland Basin, encompassing non-hydrothermal habitats in the Faroe Bank Channel and along the Reykjanes Ridge, as well as the ventassociated habitat. With the ERRC, an oceanographic connectivity pathway to western Atlantic basins, the potential forB. hirsutum nauplius larvae to be dispersed across the ridge is elaborated. Given that B. hirsutum is herein acknowledged to utilise a hydrothermally influenced habitat, vent fields within its range of depth and occurrence are discussed as potential habitats, as is the potential for B. corolliforme and B. chilense to live affiliated with hydrothermal activity.

MATERIALS AND METHODS

Taxon sampling and measurements

Specimens of Bathylasma hirsutum were collected at four sites (Fig. 3, Table 1) during three research cruises under the IceAGE (Icelandic marine Animals: Genetics and Ecology) project umbrella (Brix and Devey 2019; Brix et al. 2014; Meißner et al. 2018). At site A, situated west of the Faroe Bank Channel, barnacles were sampled by means of a triangular dredge onboard the RV Poseidon in 2013 (IceAGE_2 /POS456; Brix et al. 2013). In 2018, specimens were sampled along the Reykjanes Ridge axis at sites B and C using the Remotely Operated Vehicle (ROV) Phoca onboard the RV Maria S. Merian (IceAGE_RR /MSM75; Devey et al. 2018). In 2020, the Reykjanes Ridge was revisited with the RV Sonne and the ROV Kiel6000 (IceAGE_3 /SO276; Brix et al. 2020). Barnacles were collected at site D, affiliated with the hydrothermal ‘IceAGE vent field’ which was discovered during the SO276 expedition (Brix et al. 2020). Prior to and during animal sampling by means of the ROV, each site was documented by in situ video and photography records. Material collection was performed using the ROV’s operational titanium arms, with nets or the slurp sampler. On deck, specimens were labelled and fixed in sealed plastic bags using 96% undenatured ethanol and stored at -20°C to avoid DNA degeneration. Photographs of each specimen were taken with a Canon EOS 2000D prior to tissue sampling from the abdomen or one cirrus for DNA analysis. Height of carina parietal plates were measured using a digital vernier calliper, and growth ridges on the carina were counted with the aid of a Leica stereomicroscope. Morphometric measurements of each specimen are listed in the supplementary material (Table S1). Measurements were plotted using R 4.2.2 (R Core Team 2022) and RStudio 7.2.576 (RStudio Team 2022). Respective R packages are listed in the supplementary material (Table S2). Distribution records of B. hirsutum were reviewed from original literature and databases (www.jncc.gov.uk, www.gbif.org, www. obis.org), and geographic maps including bathymetric data were created using the software QGIS 3.10 (QGIS Development Team 2019).

Table 1.

Metadata of the four sampling sites A–D. Barnacles of the species Bathylasma hirsutum were collected by means of triangular dredge (TD) and remote operated vehicle (ROV)

DNA extraction, amplification and sequencing

DNA extractions were carried out either on board the research vessel or at the research institute. Therefore, three different extraction kits were used; E.Z.N.A^® Mollusc DNA Kit (Omega Bio-Tek Inc., Norcross, GA, USA), Qiagen DNeasy^® Blood and Tissue Kit (Qiagen, California, USA), and MachereyNagel NucleoSpin Tissue Kit (Macherey-Nagel, Düren, Germany); following the manufacturer’s instructions, leaving the tissue for digestion overnight in a shaking bath at 56°C/350 rpm. For all isolates, elution was carried out in two steps, using 50 µl elution buffer each turn. DNA concentration for each isolate (100 µl) was measured using a Qubit 4 Fluorometer (Thermo Fischer Scientific^TM Inc.), following the manufacturer’s protocol. DNA quality was comparable between all extraction methods. All DNA aliquots were stored at -20°C at the German Centre for Marine Biodiversity Research (DZMB), Hamburg. The mitochondrial marker cytochrome c oxidase subunit I (COI) and the nuclear marker elongation factor 1α subunit (EF1) were selected for species delimitation analyses (Table 2). Polymerase Chain Reaction (PCR) was carried out using three types of polymerases because of the workflow happening in different laboratories or due to troubleshooting. The polymerases included AccuStart II Taq PCR SuperMix (Quantabio, Beverly, MA, USA), Phire Green Hot Start II PCR Master Mix (Thermo Fischer Scientific^TM), and PCR-Beads, illustra^TM PuReTaq Ready-To-Go^TM (Avantor^®; VWR Int. GmbH, Darmstadt, Germany). Applied master mix compositions and thermal cycling conditions are listed in the supplementary material (Tables S3, S4). Quality and quantity of amplified product was assessed by gel electrophoresis using 1.5% TAE gels. Where band quantification yielded too high DNA concentration, samples were diluted 1:10 prior to purification. Samples with low DNA content were repeated using the same PCR settings but doubling the amount of amplified DNA. Successful PCR products were purified using 3 µl ExoSAP-IT PCR Product Cleanup Reagent (Thermo Fischer Scientific^TM) on 10 µl product and run on a thermal cycler (incubation: 37°C, 15 min; enzyme inactivation: 80°C, 15 min). Double stranded sequencing was carried out by Macrogen Europe Inc. (Amsterdam, Zuidoost, The Netherlands) and Eurofins Genomics Germany GmbH (Ebersberg, Germany) using ABI 3730xl sequencers.

Table 2.

List of primers with annealing temperature (AT) range

Sequence editing, alignment, and genetic analyses

Geneious Prime^® (Version 2022.1.1; Biomatters, Auckland, New Zealand; Kearse et al. 2012) was used to edit and assemble forward and reverse chromatograms as well as to check for potential contamination using the implemented BLAST search tool (Basic Local Alignment Search Tool; Altschul et al. 1990). Assembled sequences were aligned using MUSCLE (Edgar 2004), implemented in Geneious Prime^® with default settings, and ends trimmed with respective primer sequences. The software jModelTest 2 (Darriba et al. 2012; Guindon and Gascuel 2003) was used to estimate best-fit models of evolution by applying the Akaike Information Criterion (AIC; Sakamoto et al. 1986), resulting in GTR + I + G for COI and TIM1 + I + G for EF-1. Phylogenetic analyses of single-gene and concatenated alignments were performed using MrBayes 3.2.1 (Huelsenbeck and Ronquist 2001), with three parallel runs of 5 million generations, sampling every 1000 generations. Convergence of independent runs was examined in Tracer 1.7.2 (Rambaut et al. 2018) with a burn-in of 10%. Trees were reconstructed using Bayesian Inference (BI), assessing branch support by posterior probability (PP) with values ≥ 0.95 considered as highly supported (Felsenstein 1985; Huelsenbeck et al. 2001). Tree editing was performed in FigTree 1.4.4 (Rambaut 2009) and Affinity Photo 1.10.5 (Serif Ltd., Europe). Species were delimited using the two distance-based methods ASAP (Assemble Species by Automatic Partitioning; Puillandre et al. 2021) and ABGD (Automated Barcode Gap Discovery; Puillandre et al. 2012), using the three evolutionary models JukesCantor; Kimura 80 and Simple Distance with default settings. Haplotype networks were computed for both gene markers using the software PopArt (Leigh and Bryant 2015), applying the minimum spanning network algorithm to calculate nucleotide diversity, Tajima’s D test statistic and the number of segregating and parsimony-informative sites (Bandelt et al. 1999).

Energy Dispersive Spectroscopy

Energy Dispersive Spectroscopy (EDS) analysis for the elemental presence detection was performed to compare the mineral components of shell plates of B. hirsutum from sites with and without hydrothermal influence. EDS was done with a Hitachi TM3000 Scanning Electron Microscope (Hitachi HighTechnologies, Maidenhead, UK), equipped with an Oxford Instrument INCA system and Aztec software (Oxford Instruments, High Wycombe, UK) at the British Antarctic Survey. EDS maps for x150 magnified areas were acquired for five minutes, set to detect all elements, and quantified by the Aztec software. Elemental spectra of the bulk composition (% oxides) were displayed and presence of manganese (Mn), iron (Fe) and rare earth metals (e.g., Tungsten trioxide (WO₃), Ytterbium (III) oxide (Yb₂O₃), Vanadium pentoxide (V₂O₅)) were noted.

RESULTS

A total of 159 specimens of Bathylasma hirsutum were sampled at four sites (A–D) within the NE Atlantic. Site A extends the northern distributional range of the species from the Wyville Thomson Ridge to the Faroe Bank Channel. Sampling efforts at sites B–D present new records from the central section of Reykjanes Ridge and provide first-time in situ observations of the species in its natural habitat (Fig. 4). Video footage of the respective sites can be accessed via. The habitat at site B, situated at ~710 m depth, is dominated by large volcanic rocks and pillow lava with varying topography, featuring a range of niches with diverse megafaunal compositions including different species of cold-water corals (Fig. 4A; see also Devey et al. 2018). Barnacle coverage varies from individual specimens along the slope of a volcanic mount to localised patches of high abundance. Active predation by a sea star of unidentified species was observed, next to an empty barnacle shell which likely had been fed upon beforehand. At site C, barnacles occur with highest observed abundances along the northern and western rim of a large, cratered volcano structure between 858 and 966 m depth. At times, debris of dead barnacle shells is found below boulders on which living adult specimens sit attached (Fig. 4B). Active feeding behaviour by means of complete cirral extension was observed frequently (Fig. 4D). Proximal to the venting chimneys of the recently discovered IceAGE vent field (Brix et al. 2020), barnacles were found attached to volcanic boulders in localised groups, accompanied by several specimens of an orange deep-sea anemone, possibly belonging to the genus Phelliactis Simon, 1892 (Fig. 4E, F). During sampling, ambient water temperature was measured to ~5.6°C (see Brix et al. 2020 fig. 5.29). The influence of hydrothermal activity is apparent by the presence of white bacterial mats covering the sea bed. Prior to this study, the settlement of B. hirsutum in a hydrothermally influenced habitat had been undescribed. Of all species of acorn barnacles currently known, B. hirsutum represents the first to be found affiliated with a hydrothermal vent field in the Atlantic Ocean.

Fig. 4.

In situ images of B. hirsutum from the Reykjanes Ridge axis (A–D) and the IceAGE vent field (E, F). A, Dense cluster of specimens attached to pillow lava (site B). B, Barnacles aggregated with a cold-water coral. Accumulated dead shell plates visible to the lower left (site C). C, Epizoic growth on boulder with associated coral fauna (site C). D, Feeding behaviour by extended cirral fans in multiple specimens situated on volcanic rock (site C). E–F, Specimens with brown-black shell precipitate attached to volcanic rocks, surrounded by white bacterial mats (site D). Large orange anemones, possibly of the genus Phelliactis, share the habitat. Image courtesy: GEOMAR, Kiel.

Molecular species delimitation

DNA was successfully amplified for 146 specimens, yielding 287 novel sequences for the two genetic markers COI (146 sequences; 686 bp) and EF1 (141 sequences; 930 bp). Seventeen sequences were retrieved from GenBank and included in the final analysis. Chelonibia testudinaria (Linnaeus, 1758) was used to root trees, as the family Chelonibiidae Pilsbry, 1916 is sister to the Bathylasmatidae (Chan et al. 2021). The Bayesian analyses of the concatenated gene alignment (COI + EF1, 168 sequences; 1616 bp) show maximum support for a single clade of B. hirsutum (PP = 1), including specimens from all sampled sites (Fig. 5). The clade clearly separates from the Antarctic sister species B. corolliforme (PP = 0.95) as well as from the sister genus Hexelasma Hoek, 1913 (P = 1). ABGD and ASAP analyses were run on alignments for both genetic markers. Prior maximum divergence of intraspecific diversity (P) was set to 0.001–0.1 and relative gap width (X) to 1.0–1.5. Each of the evolutionary models yielded a partition consistent with the Bayesian analysis, suggesting one clade of B. hirsutum (P = 0.00167–0.100). Partitions yielding more than one clade of B. hirsutum likely resulted from splitting artefacts caused by the low P values (0.0010; Puillandre et al. 2012 2021). Sequence data is submitted to GenBank via the BOLD dataset doi:10.5883/DSIACIR. GenBank accession numbers are listed in the supplementary material (Table S5).

Fig. 5.

Bayesian phylogenetic tree with posterior probabilities based on the concatenated COI + EF1 dataset, including seven outgroup species. Names on the nodes refer to the Bathylasma hirsutum specimen code and the respective IceAGE expedition. Results of the ASAP and ABGD species delimitations are shown by yellow and green circles. The tree is rooted with Chelonibia testudinaria. Scale bar shows substitutions per site.

Haplotype network analyses

Each one minimum-spanning haplotype network was inferred from 146 COI and 134 EF1 sequences of B. hirsutum to visualise the genetic relationships between individual genotypes at population level (Fig. 6). Both networks depict a high degree of genetic connectivity amongst the sampled sites, reflected in the three (COI) and four (EF1) prominent clusters of genetically identical specimens. The genetic distance to smaller clusters or individual samples is low, with a maximum of four mutational steps visualised as hash marks. The low intraspecific genetic diversity is reflected in the nucleotide diversity and small number of parsimonyinformative sites.

Fig. 6.

Minimum spanning haplotype networks and analysis outputs of Bathylasma hirsutum from the four sampled sites A–D coded by colour. A, Network of the COI sequences. B, Network of the EF1 sequences.

EDS analyses of barnacle plates

Specimens collected at sites A, B, and C depict a natural shell colour, congruent with the morphological species description of B. hirsutum (Hoek 1883). Shell plates of specimens from site D, however, are of a darkened shade owing to a brown-black shell precipitate (Fig. 4). To examine these extrinsic differences and the composition of the precipitate, elemental spectra of the bulk composition (% oxides) of manganese (Mn), iron (Fe), Tungsten trioxide (WO₃), Ytterbium (III) oxide (Yb₂O₃), and Vanadium pentoxide (V₂O₅) were assessed of selected specimens from each site (Table 3). Shell plates from site D yielded the highest Fe/Mn ratios, with percentages of 2.54–34.82 (FeO) and 1.94–28.59 (MnO) measured from the youngest (base) to the oldest shell part (apex). This results in up to 60.24% ferromanganese (Fe-Mn), increasing towards the apex. Values of WO₃ double in top and mid sections, whereas Yb₂O₃ is absent. Vanadium pentoxide (V₂O₅) was solely found in the mid-section of one shell plate. Shell plates from sites B and C contained smaller Fe/Mn ratios, yielding traces of Fe-Mn oxide ranging from 0.26–1.35% and 0.44–1.95% in the mid-section, to 1.93% and 2.25% in the apex, respectively. No ferromanganese oxides were detected in the youngest shell part. Small fractions of WO₃ and Yb₂O₃ are present in the mid sections only. The shell plate from site A yielded 4.56% FeO in the mid-section, as well as 0.37% WO₃ in the apex and 0.64% Yb₂O₃ at the base. No Fe-Mn oxide precipitates were measured.

Table 3.

Energy Dispersive Spectroscopy (EDS) analyses for the elemental presence detection on barnacle plates from stations within the respective sampling areas. Elemental spectra of the bulk composition (% oxides) are shown for the top, mid and end sections of each shell plate with measured presences of iron oxide (FeO), manganese oxide (MnO), ferromanganese oxide (Fe-Mn) and the rare earth metals Tungsten trioxide (WO₃), Ytterbium (III) oxide (Yb₂O₃) and Vanadium pentoxide (V₂O₅). Measurements in the top section of specimen S039 failed and are not accessible (na)

Morphometric measurements of carinal shell plates

Intact carinal parietal plates were identified in 69 specimens (site A, B, C, D: 10, 24, 24, 11; see Table S1). Carinal height (CH) was measured from base to apex and plotted against the count of growth ridges (GR) for each parietal plate to map barnacle size distribution per site and visualise the relation between CH and GR (Fig. 7A). Overall, CH has a positive effect on the number of GR, illustrated by the positive slope. The largest CH of 65.48 mm was measured from a site C specimen, with the highest count of 116 GR. The sample set contained the fewest small specimens with a minimum height of 18.53 mm and a count of 28 GR. Barnacles from site A were second largest, with maximum and minimum heights of 29.00 mm and 78.30 mm, and respective counts of 31 and 63 GR. Specimens from site B depicted the largest size span from 5.15 mm to 56.30 mm, with 12 GR and 56 GR laid down on the carinal plates, respectively. Here, the highest GR count of 71 was from a specimen of 44.22 mm carinal plate height. The smallest barnacle from site D was 14.70 mm and had 12 GR. In the largest specimen of 50.50 mm height, 48 GR were counted. The highest count of 51 was retrieved from a carina measuring 43.20 mm. To test whether barnacle size distribution changes with ocean depth, average depth at each sampled site was plotted against CH (Fig. 7B). Our data denote a trend towards enhanced shell growth at the deeper sites A and C compared to the shallower sites B and D.

Fig. 7.

Morphometric measurements of carinal parietal plates.A, Relation between carinal height and count of growth ridges on the carinal parietal plate with a predicted linear model fits based on a 95% confidence interval. B, Barnacle size distribution represented by carinal height at average depth at the sampled sites. C, Bar chart showing the average number of growth ridges per site.

DISCUSSION

The herein presented results on the deep-sea acorn barnacle Bathylasma hirsutum from four sites within the northeastern Iceland Basin provide new distributional records with in situ footage of specimens occupying hard-bottom substrates in depths between 658 and 966 meters along the Reykjanes Ridge axis. For the first time, B. hirsutum was found affiliated with a hydrothermally active field. As it is unknown to what extent populations of this sessile crustacean with planktotrophic larval dispersal are connected, we have assessed the genetic connectivity between specimens from the Faroe Bank Channel and the Reykjanes Ridge. The molecular analyses revealed low levels of intraspecific genetic diversity, reflected in the low number of parsimony-informative sites. The latter indicates the observed intraspecific genetic variability to be lower than the expected genetic variation between specimens of B. hirsutum from different sites (Papathanassopoulou and Lorentzos 2014). Haplotype networks of both markers suggest a substantial gene flow between all sites (Fig. 6), indicating a pronounced genetic connectivity within the Iceland Basin. Site A, situated west of the Faroe Bank Channel, now marks the northern boundary of the species’ distributional range. Based on the basin-wide cluster analysis conducted by Schumacher et al. (2022), this site is categorised as nutrient-rich, with strong local currents and seasonal changes. It is highly hydrodynamic, influenced by the continuous flow of dense, well-oxygenated Iceland Scotland Overflow Water (Chafik et al. 2020; Hansen et al. 2016) with mean water temperatures at maximum depth between 4–9°C (Assis et al. 2018; Tyberghein et al. 2012). The hard-bottom habitat houses several dense aggregations of benthic suspension feeding organisms such as sponges and cold-water coral reefs, indicating a prominent food supply (Frederiksen et al. 1992; Kazanidis et al. 2019). Strong current velocities not only enhance particle suspension and thus food supply, making this a favourable habitat for B. hirsutum, but may also provide pathways for larval transport. As the deep-water currents interconnect with water flows along the eastern flank of the Reykjanes Ridge, we suggest a proportion of larvae dispersed from site A are transported towards the Reykjanes Ridge axis, facilitating the high localised abundances observed during the ROV dives (Fig. 4A). The ridge axis is mainly influenced by the ERRC, a weakly bottomintensified current resulting from a convergence of dense Iceland Scotland Overflow Water with the Iceland branch of the North Atlantic Current (Koman et al. 2020). With these nutrient-rich and well-oxygenated water masses flowing south, larvae are suggested to be transported along the ridge, facilitating the genetic connectivity between the herein assessed sites. We presume that recruits of B. hirsutum settled at site D originated from the dense barnacle fields north of this site. With ambient seawater temperatures measured to ~5.6°C where the vent-affiliated specimens were sampled (see Devey et al. 2018), the habitat suitability by means of temperature range is comparable to the non-hydrothermally influenced sites.

Analysis of morphometric measurements

The positive relation between carinal plate heights and number of growth ridges laid down on the respective shell plates reflects the formation of growth ridges to be a good indication for the growth rate of the deep-sea barnacle B. hirsutum, such as it is for shallowwater species (Bourget and Crisp 1975). In what way the number of growth ridges is related to the age of a specimen, however, seems to be depth-dependent. In an experimental study on growth ridge formation, Bourget and Crisp (1975) observed one growth ridge to be laid down for each exuviation in the intertidal species S. balanoides, reflecting an annual growth pattern. In the deep-sea species B. corolliforme, however, the authors counted many more ridges than the observed number of moults and were not able to distinguish between the ridges laid down on the shell during and in absence of a moulting event. Hence, in B. corolliforme, one growth ridge cannot be estimated to account for one year, but rather a growth period which in turn is affected by food availability. The principal food source in deep-sea habitats is owed to vertical flux of organic matter from both surface primary production and redistribution of suspended particles in the bottom mixed layers (Davies et al. 2009 2006; Ramirez-Llodra et al. 2010). With B. hirsutum being sister to the Antarctic B. corolliforme and occupying habitats with comparable physical environments, we suggest a similar pattern of growth ridge formation in this species where a fluctuating food supply in the else stable physical deep-sea environment might affect the growth rate (Anderson 1994; Foster 1983; Tyler and Young 1992).

Carinal parietal plates from the two deeper sites A and C were found to be larger compared to the shallower habitats at sites B and D, implying barnacle size to be greater with increased depth (Fig. 7). This trend is congruent with the general understanding of deep-sea species to grow larger with depth, adapted to invest already restricted energy resources in growth (Timofeev 2001). At site C, specimens were found in the highest abundance and are not only the largest but also show the highest amount of growth periods. We suggest the high abundances and enhanced growth at the volcanic crater to be a response to an enriched food environment accompanied by strong localised currents, providing a favourable habitat for B. hirsutum. The productive food environment is also reflected in a massive coral forest, dominated by large trees of Paragorgia arborea (Linnaeus, 1758), which was detected proximal to the barnacle fields (Devey et al. 2018). This is further supported by the frequently sighted active feeding behaviour (Fig. 4D). The accumulated shell debris observed below several boulders with living specimens still attached indicates a long-sustained population with several generations of settlement. The large size span of individuals collected at site B indicates a vital population with successful recruitment of juveniles, settled amongst their adult conspecifics (Fig. 7A). Since several specimens sizing less than 10 mm in carinal height were found, food availability in this habitat is assumed to be slightly lower compared to site C, yielding a lowered individual growth rate. At the hydrothermally-influenced site D, specimens were found to be smaller compared to the deeper sites (Fig. 7C). This is interesting, as B. hirsutum could potentially utilise matter from the widespread bacterial mats as an additional food source. This mixotroph mode has been confirmed in several stalked barnacle species from hydrothermal vent fields in the Pacific Ocean (Buckeridge 2000; Southward and Newman 1998) and was recently suggested for the acorn barnacle genus Eochionelasmus (Chan et al. 2020). If B. hirsutum were adapted to this feeding mode, we would expect to find barnacles of similar or even increased size which is not mirrored in our observations (Fig. 7). Accounting for the facts that the species rather opportunistically utilises the hard-rock substrate in vicinity of the vent field and does not live directly associated with the hydrothermally active chimneys, we assume that B. hirsutum is not adapted to integrate organic matter from bacterial mats. In addition to the smaller size range, less growth ridges were counted on carinal parietal plates that were of comparable height to specimens from non-hydrothermally influenced sites (Fig. 7A). These results imply that growth ridge formation might be impacted by the pronounced shell precipitate, possibly eroding the ridges in the process of the accumulation of ferromanganese oxides. Overall, our results imply barnacle shell size to increase with ocean depth (Fig. 7B). However, this snapshot of barnacle size distribution requires more detailed attention to verify whether this trend applies across the entire distributional range of the species.

Ferromanganese oxide shell precipitates

The formation of iron (Fe), manganese (Mn) and ferromanganese (Fe-Mn) mineral deposits happens by migration of Fe and Mn cations from less oxidising to more oxidising conditions (Glasby 2006; Park et al. 2023). Marine Fe-Mn deposits may form as crusts, nodules or are contained within layers of rock and are classified by three mineralisation types; hydrogenetic, diagenetic, and hydrothermal (Marino et al. 2019; Usui et al. 2020). Hydrogenetic Fe-Mn crust deposits form by precipitation of Fe oxyhydroxide and Mn oxide from ambient seawater onto the bedrock on i.e., continental shelves and seamounts (Lusty et al. 2018). Hydrothermal processes, on the other hand, lead to the accumulation of Fe-Mn oxides from plume discharge (Connelly et al. 2007; Muiños et al. 2013; Sujith and Gonsalves 2021). At the Reykjanes Ridge, plume discharge from the Steinahóll vent field contains a high concentration of dissolvable manganese (~60 nmol/l; Taylor et al. 2021). Equivalent discharge is expected to occur at the IceAGE vent field, situated ~400 km south on the same volcanic ridge axis. Inactive ridges such as the Iceland-Faroe Ridge do not display such enrichments (Horowitz 1974). Since there are no records of hydrothermal activity in the vicinity of site A, it is plausible that the chemical component analysis yielded no traces of ferromanganese (Table 3). None of the specimens from site B, despite being situated just ~450 m north of the IceAGE hydrothermal vent field, were observed with shell precipitate. In accordance with the circulation patterns and bottom currents prevailing along the eastern flank of the Reykjanes Ridge, the plume discharge by the IceAGE vent field is likely to be carried in a south-western direction relative to the active chimneys (Horowitz 1974; Koman et al. 2020; Ruddiman 1972). Hence, Fe-Mn traces from hydrothermal origin seem not to reach barnacles from site B. Situated south of the vent field, specimens collected at site C could be in range to be affected by plume discharge. However, throughout the distance of ~130 km, trace elements such as Fe and Mn are likely to have been accreted to sediments or precipitated as crust deposits in the proximate environment before reaching the barnacle field (Connelly et al. 2007; Horowitz 1974). The minor percentage of Fe-Mn oxide measured in shells from both sites (Table 3) is rather thought to reflect ambient water Mn and Fe concentrations as barnacles show a strong affinity for incorporating both Fe and Mn into their shells compared to other calcifying animals (Pilkey and Harriss 1966; Ullmann et al. 2018). Only specimens at site D had pronounced Fe-Mn precipitates on their parietal plates, visible as dark brown-black deposits from the apex to the base. The high Fe/Mn ratios reflect an enrichment from hydrothermal plume discharge emanated from the IceAGE vent field. Detailed geochemical analyses on the enrichment of trace elements in the hydrothermal plume are desirable to assess whether the Fe-Mn enrichment could additionally be a product of diagenetic processes. In this case, high concentrations of typical diagenetic elements such as Ni, Cu and Co would need to be present (Schiellerup et al. 2021). In general, shell plates from site D contained a higher percentage of rare earth metal components compared to sites A–C, especially with regard to Tungsten trioxide. This is predictable, as this metal has been found to accumulate in ferromanganese deposits (Kunzendorf and Glasby 1992).

Previous to our study, the affiliation of B. hirsutum with hydrothermal vent systems was unexplored. Still, the presence of shell precipitates has not gone fully unnoticed. During a geological survey, intact barnacles were collected together with coral fragments at 800 m depth on the Reykjanes Ridge, at a station situated central between the herein sampled sites B and C (Murton et al. 1995). Without further commenting on a possible source of origin, the authors reported a manganese staining on the barnacles, which were later identified as B. hirsutum by Copley et al. (1996), who however neglected to mention the precipitate. The reported manganese staining is likely to be FeMn oxide and could have originated from plume discharge emanated from site D, or even from another yet undetected hydrothermal source in this area. On the Azores Archipelago, Southward (1998) examined several hundred dead shell plates of B. hirsutum collected together with dead pieces of Desmophyllum pertusum (Linnaeus, 1758) between 1250 and 1630 m depth. Both barnacles and corals were described as having a black manganese coating. As the collected material had died off, we assume that the precipitate formed through either hydrogenous or diagenetic sources, as for example observed in shark teeth or coral skeletons (Edinger and Sherwood 2012; Iyer 1999). Based on the high amount of barnacle shells, Southward (1998) estimated that the sampled area must have been densely populated by B. hirsutum in earlier times and suggested a possible enrichment from hydrothermal sources which would have enabled the high abundance of specimens. The Lucky Strike vent field, situated ~12 km west of this area and described just shortly before (Langmuir et al. 1997), could be considered as a potential hydrothermal source of enrichment. However, the once high abundance of B. hirsutum could also have sustained due to strong prevailing currents that transport nutrient-rich, well-oxygenated water masses along the MAR (Puerta et al. 2020; Fig. 1). With regard to our observations of a rather low abundance of B. hirsutum at the IceAGE vent field (Fig. 4E, F), the transport of nutrients by ocean currents is regarded to have a much higher impact on barnacle abundance compared to the influence of hydrothermal activity. If latter were to be the main source, the abundance of B. hirsutum ought to have been much greater than what we observed at site D.

On the affiliation of Bathylasma with hydrothermally influenced habitats

This study acknowledges the habitat adjacent to the IceAGE vent field as a suitable living space for B. hirsutum. As the recognised maximum depth for B. hirsutum is 1829 m, NE Atlantic hydrothermal vent fields deeper than 2000 m are anticipated to be out of the species’ ecological range for successful settlement in these habitats. Within its known depth range, however, several records from the MAR spreading centre on the Azores Archipelago overlap with recognised hydrothermal active sites (Beaulieu and Szafrański 2020). Given from west to east, B. hirsutum has been recorded ~800 m off Ribeira Quente (Young 1998), and Don Joao de Castro Bank (Gruvel 1920), ~784 m south of Espalamaca (Young 1998), ~860 m south of LUSO (Poltarukha and Zevina 2006a), and ~12 km west of the Lucky Strike vent field (Southward 1998). Repeated sampling efforts by means of a ROV are highly encouraged to investigate whether B. hirsutum is in fact affiliated with the recognised hydrothermal active fields and whether the barnacles utilise these habitats in a comparable fashion to what has been observed at the IceAGE vent field, including observations on possible shell precipitates.

Detailed biological surveys of the myriad hydrothermal systems in the world oceans are still a major task (Tunnicliffe et al. 2003; Van Dover et al. 2006). We hypothesise that future research activities on hydrothermal vent systems within the distributional range of the bathylasmatid sister species Bathylasma corolliforme and Bathylasma chilense will reveal their potential to utilise hydrothermally influenced habitats, in a comparable manner to what has been found for B. hirsutum in this study. As for the sister species B. corolliforme, Dayton et al. (1982) reported specimens from 400 m depth attached to volcanic rocks. The sampling locality is situated south of the PacificAntarctic Ridge. In this region, hydrothermal activity has been inferred from several localities (Beaulieu and Szafrański 2020; Bell et al. 2016; Linse et al. 2022; Rogers et al. 2012; Tao et al. 2012) and distal hydrothermal influences have been suggested (Sieber et al. 2021) within the given depth range of the species. Furthermore, a high abundance of unidentified deepsea acorn barnacles was detected on an active volcanic mount structure at 2221 m depth during the Marion Rise Expedition programme (Koepke et al. 2020). Though not mentioned in the respective cruise report, publicly available video footage documents this observation from the Marion Rise (marumTV 2020; 43°53.6822'S, 39°13.3234'E). We believe these barnacles to belong to the species B. corolliforme as species-specific characters such as the strong diverging orifical compartments are clearly visible (see Bullivant and Dearborn 1967, plate 14; Dayton et al. 1982, fig. 3C; Meyer 2020, fig. 3-3) and the observations were captured from a site west of the species’ type locality (Hoek 1883). Unfortunately, no material was sampled and examined in further detail during the Marion Rise expedition.

Just recently, the sister species B. chilense was described based on three specimens collected from 1800–2000 m depth off Caldera, northern Chile (Araya and Newman 2018). The type locality is situated in vicinity of the Peru-Chile Trench, an area of subduction and earthquake activity. As the vast majority of seafloor venting in the southern hemisphere remains unexplored, there may be many hydrothermal vent sites on this actively spreading ridge to discover (Baker et al. 2016; Beaulieu and Szafrański 2020). In line with further deep-sea explorations off the Chilean coast, we hypothesise B. chilense to be found affiliated with hydrothermal activity, utilising these habitats in a comparative manner to what has been presented here for B. hirsutum from the NE Atlantic.

A connectivity pathway to the northwest Atlantic?

To date, no records of B. hirsutum are known from any of the north-western (NW) Atlantic basins. There is, however, one fossil record of the extinct sister species B. corrugatum from the continental shelf break off North Carolina, indicating that these bathylasmatid acorn barnacles managed to sustain at least one population in the NW Atlantic during the Neogene and Paleogene periods (Zullo and Baum 1979). In fact, the westward flow of the East Reykjanes Ride Current merging with the Irminger Current could serve as a connectivity pathway into the Irminger Basin as well as the David Strait, facilitating larval distribution of B. hirsutum to suitable hard-bottom habitats along the shelf breaks. Such habitats were recently surveyed by Kenchington et al. (2017) who discovered a reef of Desmophyllum pertusum at 886–932 m depth on a steep slope in the Davis Strait. The herein reported temperature range of 4.13–5.03°C as well as current velocities of 10–15 cm s-1 could indeed facilitate an eligible habitat for barnacle recruits and already established, actively feeding individuals (see Southward and Southward 1958). Increased sampling efforts in NW Atlantic basins and along the continental shelf breaks could possibly reveal yet undiscovered occurrences of B. hirsutum. Special focus ought to be given to the coastline off America as the type of Hexelasma americanum Pilsbry, 1916, a species of sister genus to Bathylasma, was sampled here (Pilsbry 1916). Showing high morphological similarity and occupying overlapping habitats on the Azores Archipelago (Poltarukha and Zevina 2006a; Young 1998 2001), we suggest that both species require similar environmental conditions to sustain vital populations. Hence, it might be possible that a co-existence of Bathylasma and Hexelasma along the western Atlantic shelf is present, but has been left unrecognised due to their morphological similarities (Newman and Ross 1971; Southward and Southward 1958).

CONCLUSIONS

The Reykjanes Ridge provides a variety of hard-bottom habitats which are dominated by saline, nutrient-rich and well oxygenated water masses as well as strong current regimes (Puerta et al. 2020; Schumacher et al. 2022). Much effort is still needed to understand the extent and functions of biological communities found in these habitats, with special regard to hydrothermally influenced sites. Our investigations on the deep-sea acorn barnacle B. hirsutum contribute to a better understanding of the benthic sessile fauna with planktotrophic larval dispersal and related connectivity patterns in the North Atlantic. Our molecular sampling approach has not only extended the genetic dataset of this species, but aided to investigate the genetic connectivity between four sampled sites in the northeastern Iceland Basin. The comprehensive literature review on the species revealed a high number of sampling records as well as a series of detailed studies on favourable habitats in this species. However, previous to this study, in situ footage of these large barnacles remained absent, as did the acknowledgement of hydrothermally influenced habitats as a suitable living space. Our findings of small local abundances on volcanic rocks in vicinity of the IceAGE vent field do not only extend the knowledge on a favourable habitat, but also emphasise the need for further research that ought to be focused on the Reykjanes Ridge axis, as well as on hydrothermal habitats shallower than 1800 meters along the extensive Mid-Atlantic Ridge.

Supplementary materials

Table S1.

Morphometric measurements of carinal parietal plates from 69 specimens of Bathylasma hirsutum.

Table S2.

Applied Rpackages to plot the morphometric measurements.

Table S3.

Components of master mix cocktails used to perform PCR reactions.

Table S4.

Thermal cycling conditions of the polymerase chain reactions. Reaction steps are: 1 Initial denaturation, 2 Denaturation, 3 Annealing, 4 Elongation, 5 Final extension, 6 Cooling. Temperatures marked with an asterisk were used for the EF1 gene locus.

Table S5.

List of GenBank accession numbers for 146 novel COI and 141 novel EF1 sequences of Bathylasma hirsutum.

List of abbreviations

ABGD

Automated Barcode Gap Discovery.

ASAP

Assemble Species by Automatic Partitioning.

BI

Bayesian Inference.

BOLD

Barcode Of Life Database.

CH

Carinal height.

EDS

Energy Dispersive Spectroscopy.

ERRC

East Reykjanes Ridge Current.

GR

Growth ridge.

MAR

Mid-Atlantic Ridge.

PP

Posterior Probability.