Stalked barnacles in the family Neolepadidae (Yamaguchi et al. 2004) are commonly found in diffuse flow and marginal areas of deep-sea hydrothermal vent fields in Pacific, Indian, and Southern oceans (Marsh et al. 2012; Chan et al. 2019). The family currently consists of four extant genera including Neolepas Newman, 1979, Leucolepas Southward & Jones, 2003, Vulcanolepas Southward & Jones, 2003, and Ashinkailepas Yamaguchi, Newman & Hashimoto, 2004, supported by molecular taxonomic analyses, and the extinct genus Stipilepas Gale, 2020 (Yamaguchi et al. 2004; Herrera et al. 2015; Buckeridge et al. 2013; Watanabe et al. 2018; Gale et al. 2020). Ashinkailepas is the only neolepadid genus occurring in both hydrothermal vents and hydrocarbon seeps, currently containing two recent and one fossil species (Gale et al. 2020). The genus shows highly specific morphologies with prominent apico-basal ridges of capitular plates and strongly projected scale plates; therefore, the other three neolepadid genera are currently placed in the tribe Neolepadini, and an arrangement supported by molecular phylogeny (Herrera et al. 2015). Neolepas was the first described vent barnacle genus reported from the East Pacific Rise (Newman 1979), and the second and third genera, Leucolepas and Vulcanolepas, were established by Southward and Jones (2003) based on morphological characteristics. Molecular phylogenetics have indicated that Vulcanolepas is likely not monophyletic and their gross external morphologies have been shown to be highly plastic within a single genetically homogeneous species, under different environmental conditions (Herrera et al. 2015; Buckeridge et al. 2013; Watanabe et al. 2018). Leucolepas is monotypic and only includes the type species Leucolepas longa Southward & Jones, 2003.

Eight species have been described in the Neolepas/Vulcanolepas group, all restricted to hydrothermal vents. These include Neolepas zevinae Newman, 1979 from the northeast Pacific, Neolepas rapanui Jones, 1993 from the southeast Pacific, Neolepas scotiaensis (Buckeridge and Linse, 2013 in Buckeridge et al. 2013) from the Southern Ocean, Neolepas marisindica Watanabe, Chen & Chan in Watanabe et al. 2018 from the Indian Ocean, Vulcanolepas osheai (Buckeridge, 2000) from Brothers Seamount on the Kermadec Arc, Vulcanolepas parensis Southward, 2005 from the Pacific-Antarctic Ridge, Vulcanolepas buckeridgei Chan & Chang, 2018 from the Lau Basin, and Vulcanolepas fijiensis Chan, Ju & Kim, 2019. Among them, a specialized feeding habit of cultivating epibiotic symbionts on their feeding appendages (i.e., cirri) has been reported in V. osheai and V. buckeridgei (Suzuki et al. 2009). Due to the high level of plasticity in the gross external morphology (i.e., the stalk and plates) on which the identification of barnacles often relies, the identification and taxonomy of these vent barnacles require a total evidence approach that puts weight also on other characteristics such as arthropodal morphology, ecological, and physiological characteristics and molecular data. Previous studies have shown that arthropodal (mouth part) morphology is able to reliably distinguish different species of Vulcanolepas (Watanabe et al. 2018; Chan et al. 2019).

Here, we report the fifth species of Vulcanolepas, collected from the Hafa Adai hydrothermal vent field in the Mariana Trough, northwestern Pacific (Fig. 1a; Baker et al. 2017). Mariana Trough is a back-arc basin located between Mariana and Western Mariana arcs (Fig. 1a). This is the second report of stalked barnacle from Mariana Trough vents after a single sighting in the Alice Springs field in the central part of Mariana Trough (Fig.1a), but unfortunately no specimen was collected from Alice Springs (Hessler and Lonsdale 1991). We characterize and describe the Vulcanolepas species from Hafa Adai using both morphology and genetics and show that its arthropodal morphology is distinct from other described species.

Fig. 1
figure 1

Map showing locations of Hafa Adai (black diamond) and Alice Springs (white diamond) hydrothermal vent fields (a) and distribution of currently known neolepadid stalked barnacles (b). Red star: Vulcanolepas verenae sp. nov., white star: V. osheai, gray star: V. fijiensis, yellow star: V. buckeridgei, white cirlce: Neolepas zevinae/rapanui complex, black circle: N. scotiaensis, gray circle: N. marisindica, white square: Leucolepas longa

Materials and methods


A single stalked barnacle was collected at the Hula Hoop site (16°57.701’N, 144°52.152′E, Depth = 3278 m) in the Hafa Adai hydrothermal vent field, Mariana Trough by ROV SuBastian on-board the Schmidt Ocean Institute R/V Falkor (Fig. 1b). The specimen is deposited in the National Museum of Nature and Science, Tsukuba (NSMT), Japan.

Morphological observation: The gross morphology of the stalked barnacle was observed in distilled and deionized water. The soft body was carefully extracted with detaching muscles and membranes from the inner side of the plates by sharpened tweezers. The cirrus on the left side of the body and oral cones were dissected, examined, and photographed using a digital camera (Panasonic DMC-GH3) under a light microscope (Olympus BX51). The terminology of descriptions of the capitular plates and arthropodal characters coined by Darwin (1854) are followed here.

Molecular phylogenetic reconstruction: Genomic DNA was extracted using DNeasy Blood and Tissue Kit (QIAGEN) from the adductor muscle. Partial sequence of the mitochondrial cytochrome c oxidase subunit I (COI) gene was amplified by PCR using the universal primer pair (LCO1490 and HCO2198) designed by Folmer et al. (1994) and the Hot Start version of Premix ExTaq (TaKaRa) in the following steps: initial denaturation at 94 °C for 3 min followed by 35 cycles of denaturation (94 °C for 30 s), annealing (50 °C for 30 s), and extension (72 °C for 90 s). The PCR products were purified using Exo-SAP-it (usb). BigDye Terminator reaction was carried out using BigDye Terminator ver 3.1, and the products were then sequenced using an ABI3130 automated sequencer (Applied Biosystems, Thermo Fisher). The electrophenograms obtained were checked by eye and assembled with Geneous ver. 9 (Biomatters Limited). The resulting sequence is deposited in DDBJ, with the accession number of: MT662001.

The obtained sequence was aligned with currently available neolepadid sequences in the International Nucleotide Sequence Database Collaboration database, using Clustal X mounted on MEGA ver 7.0.26 (Tamura et al. 2013). The model selection program in MEGA ver. 7.0.26 was applied to select the best model for maximum likelihood algorithm, which selected the Tamura 3-parameter + Gamma distribution model. Phylogenetic reconstruction was carried out using MEGA with the genus Ashinkailepas set as the outgroup, with 2000 bootstrap replicates for the 588 bp-length alignment.



Superorder Thoracica Darwin, 1854

Order Scalpelliformes Buckeridge & Newman, 2006

Family Neolepdidae Yamaguchi, Newman & Hashimoto, 2004

Tribe Neolepadini Yamaguchi, Newman & Hashimoto, 2004

Genus Vulcanolepas Southward & Jones, 2003

Vulcanolepas Southward and Jones 2003:81.-Southward 2005: 148.-Buckeridge et al. 2013: 556.

Type species. Neolepas osheai Buckeridge, 2000

Diagnosis (emended from Southward and Jones 2003). Neolepadid stalked barnacles whose adults have a ratio of peduncle length to capitulum length of between 1:1 and 7:1. Peduncle armed with numerous small scales, 20 or more scales per whorl in upper half of peduncle; scales less than 1 mm wide, almost flat, projecting up to 0.5 mm beyond cuticle. Capitulum broad, tergal apex blunter than in other neolepadid genera. Basal angle of scutum close to capitulo-peduncle margin. Basal angle of tergum slightly above capitulo-peduncular margin.

Distribution. Kermadec Arc, Southwest Pacific; North Fiji Basin, Southwest Pacific; Lau Basin, Southwest Pacific; Mariana Trough, Northwest Pacific; Pacific-Antarctic Ridge (depth range: 1200–3278 m).

Remarks. The latest diagnosis for this genus was provided by Buckeridge et al. (2013). The major diagnostic character of Vulcanolepas emended by Buckeridge et al. (2013) included the ratio of peduncle length to capitulum length reaching 20:1 and they also removed the degree of projection of peduncular scale in order to accommodate Vulcanolepas scotiaensis (Buckeridge et al. 2013), but that species is now considered to be a Neolepas and properly known as Neolepas scotiaensis from molecular evidence (Watanabe et al. 2018). Morphologically, N. scotiaensis has very large degree of peduncular projection (up to 1.5 mm beyond the cuticle), whilst in described Vulcanolepas species, this is only < 0.5 mm beyond the cuticle. Large peduncular scale projection is one of the key morphological characters of Neolepas. We therefore reverted back to the diagnosis of Southward and Jones (2003) and Southward (2005) and updated it to include the range of characters observed from the recently described V. buckeridgei and V. fijiensis as well as the new species Vulcanolepas verenae sp. nov. described herein, including the ratio of peduncle length to capitulum length of between 1:1 and 7:1 and peduncular scales less than 1 mm wide, almost flat, projecting up to 0.5 mm beyond cuticle. Among Neolepadini, the key character that distinguish Neolepas from Vulcanolepas is the larger degree of peduncular projection (up to 1.5 mm beyond the cuticle, vs 0.5 mm in Vulcanolepas) as mentioned above. Leucolepas consists of a single species, L. longa, which can be distinguished from Vulcanolepas by the separated capitular plates, narrower tergum, and the longer peduncle (more than five times longer than the capitulum).

The distribution of the genus Vulcanolepas was extended considerably with the collection of Vulcanolepas verenae sp. nov., as this is the first distribution record of Vulcanolepas in the northern hemisphere and extends the bathymetric range of the genus by 649 m (previously the deepest record was 2629 m held by V. buckeridgei; Chan and Chang 2018). This depth rivals that of the deepest known vent barnacle genus, the balanomorph barnacle Eochionelasmus Yamaguchi in Yamaguchi and Newman 1990, which has a depth range of 1764–2500 m for E. ohtai Yamaguchi in Yamaguchi and Newman 1990 in North Fiji, Lau and Manus Back-Arc Basins, Pacific Ocean (Yamaguchi and Newman 1997a). Eochionelasmus paquensis Yamaguchi and Newman, 1997b is known from sites up to 2578 m deep in the East Pacific Rise (Yamaguchi and Newman 1997b), Eastern Pacific, and E. coreana Chan et al. 2020 from 2625 m deep at Solitaire Vent field, Indian Ocean (Chan et al. 2020).

Vulcanolepas verenae sp. nov.

Type locality. Hula Hoop site (16°57.701’N, 144°52.152′E, Depth = 3278 m), Hafa Adai hydrothermal vent field, Mariana Trough (Fig. 1).

Type material. Holotype (NSMT-Cr 28234, Figs. 2, 3, 4, 5, 6, 7). Collected live from the type locality during dive #42 of ROV SuBastian on-board the Schmidt Ocean Institute R/V Falkor, cruise FK161129. Preserved in 99.5% ethanol.

Fig. 2
figure 2

In situ observations of Vulcanolepas verenae sp. nov. in Hafa Adai vent field, Mariana Trough. a The holotype (NSMT-Cr 28234) just before collection during ROV SuBastian dive #42; b Sighting of an uncollected cluster during ROV SuBastian dive #43

Fig. 3
figure 3

External morphology of Vulcanolepas verenae sp. nov., holotype (NSMT-Cr 28234). a left view; b right view; c carinal view; d rostral view. Abbreviations: paired tergum (t), scutum (s), unpaired carina (c), lateral (l), rostrum (r). Scale bars: 1 cm for all parts

Fig. 4
figure 4

Vulcanolepas verenae sp. nov., holotype (NSMT-Cr 28234). a maxilla; b setae on margin of maxilla; c maxillule; d setae on cutting margin of the maxillule; e mandible; f first tooth of the mandible; g inferior angle; h second and third teeth of the mandible. Scale bars: a, c, e: 500 μm; b, d, f, g, h: 100 μm

Fig. 5
figure 5

Vulcanolepas verenae sp. nov., holotype (NSMT-Cr 28234). a mandibular palps; b setae on palps in the box in a; c labrum with fine, sharp teeth on edge; d magnified view of teeth on labrum; e cirrus I; f simple type setae on cirrus I, which was magnified in the box in e; g cirrus II; h simple type setae with filamentous bacteria on cirrus II, which was magnified in the box in g. Scale bars: a, b, c, d, f, h: 100 μm; e, g: 1 mm

Fig. 6
figure 6

Vulcanolepas verenae sp. nov., holotype (NSMT-Cr 28234). a anterior ramus on cirrus III; b intermediate segment on cirrus III; c posterior ramus on cirrus IV; d spines on the surface of segments of cirrus IV; e posterior ramus on cirrus V; f simple setae on cirrus V coated by dense filamentous bacteria; g cirrus VI; h spines on the surface of segment of cirrus VI. Scale bars: a, c, e, g: 1 mm; b, f: 100 μm, d: 500 μm

Fig. 7
figure 7

Vulcanolepas verenae sp. nov., holotype (NSMT-Cr 28234). Overview of a cirrus I; b cirrus II; c cirrus III; d cirrus IV; e cirrus V; f cirrus VI; g penis, insert showing the tip under light microscopy; h drawing of caudal appendages (ca) and the penis. Pos: posterior, Ant: anterior. Scale bars: 1 mm for all parts

Diagnosis. Vulcanolepas with small peduncular scales, about 0.8 mm in length and protruding 0.3 mm from peduncle. First tooth of mandible small. Proximal segment of posterior and anterior rami of cirrus I protuberant. Segments of anterior and posterior segment of cirrus II protuberant. Setae of cirri III to VI extreme long, length of setae about 10 times the length of an individual segment. Surface of segment in cirri III to VI bear a line of spines.

Description. Hermaphrodite. Capitulum higher than wide; capitulum height to width ratio 1.4:1 (Fig. 3). Capitulum with eight approximate plates; plates with black mineral coating, inter-plate spaces occupied by thin cuticular membranes (Fig. 3). Carina umbo apical, slightly bowed, surface with 10 horizontal growth lines; carina height about 4/5 capitulum height (Fig. 3). Tergum quadrangular with apex angle 58o, surface with 10 V-shaped growth lines; basal angle of tergum (68o) slightly elevated from capitulum-peduncle edge, about 1/5 total height of capitulum (Fig. 3). Scutum quadrangular, umbo apical, apical angle 31o, located at capitulum-peduncle edge (Fig. 3), surface with 7 V-shaped growth ridges parallel to basal margin, basal angle 101o. Medial latus equilateral triangular, umbo apical, apical angle 53o, surface with eight straight horizontal growth lines. Rostrum triangular, with six horizontal growth lines (Fig. 4e).

Peduncle to capitulum ratio about 1:1 (Fig. 3). Peduncle with 52 rounded scales per whorl just below the capitulum region, 26 scales per whorl at mid region of capitulum, scales width 0.83 mm (averaged from 3 scales), height 0.66 mm (averaged from 3 scales) projected from peduncle.

Trophi. Maxilla subtriangular, with a convex exterior margin, simple type setae on exterior margin (Fig. 4a, b). Maxillule cutting edge straight, with dense simple type setae on cutting margin; interior margin straight (Fig. 4c, d). Mandibles tridentoid, first teeth separated from the remainder, small and sharp, can only be observed from inner side of mandible (Fig. 4e, f). Second and third teeth comb-shaped, with more than 30 sharp spines on cutting edge (Fig. 4g, h). Cutting edge of second and third teeth long, each occupy about half of mandible’s total length; inferior angle circular, composed of a series of sharp and large spines (Fig. 4e). Mandibular palp circular, with relatively sharp tip, with simple setae on tip and outer margin (Fig. 5a, b). Labrum not bullate, with single array of small sharp teeth on cutting edge (Fig. 5c, d).

Cirri. All six pairs of cirri are located close to each other. Cirrus I, both anterior and posterior rami are similar in length, proximal segments with width to length ratio about 2, rami become antenniform when approaching to the distal ends, with segment width to length ratio about 0.5 (Figs. 5e, g, 7a). The last 7 proximal segments of the posterior and last 8 proximal of anterior ramus protuberant, bearing dense bundles of simple setae (Fig. 7a). Cirrus II broken in holotype, segments protuberant at proximal region, with denser simple-type setae (Figs. 5g, h, 7b). Cirrus III incomplete in holotype, both anterior and posterior ramus bear very long setae (Fig. 6a, b). Length of setae about 10 times the length of an individual segment (Fig. 7c). Setae simple and surface coated by filamentous bacteria (Fig. 6f). Cirri IV–VI similar in morphology, anterior and posterior rami similar in length. Segments at distal 1/3 end of both rami become width twice greater than height (Figs. 6c–h, 7d–f). Setae are long, about 10 times the length of an individual segment, coated with filamentous bacteria (Fig. 6f). Intermediate segments of cirri IV–VI bear two pairs of short setae at distal end and bundles of long setae (Figs. 6c–h). Intermediate segment with scattered spines on lateral surfaces (Figs. 6c–h). Caudal appendage unsegmented, short, blunt, length about 1/8 length of pedicel on cirrus VI (Fig. 7h). Penis without basi-dorsal point, being half length of cirrus VI (Fig. 7g).

Etymology. Named after Verena Tunnicliffe, University of Victoria, Canada, for her contribution to hydrothermal vent research including the discovery of the present new species.

Distribution. At present, only collected from the Hafa Adai vent field. A stalked barnacle has previously been reported from the Alice Spring field, Mariana Trough (Hessler and Lonsdale 1991), but as no specimen was collected we cannot be sure if this record represents V. verenae sp. nov.

Remarks. This is the fifth species included in the genus Vulcanolepas. All Vulcanolepas have distinct morphological diagnostic characteristics (Watanabe et al. 2018; Chan et al. 2019). Vulcanolepas verenae sp. nov. is similar to V. osheai and V. buckeridgei in having extreme long setae, extending up to more than 3 mm on cirri III–VI in V. verenae sp. nov., and those setae are coated with filamentous bacteria. Mandibles of V. buckeridgei, V. osheai, and V. verenae sp. nov. all have very small first tooth, which differ from other members of Vulcanolepas. Vulcanolepas verenae sp. nov. differs from V. buckeridgei in having smaller peduncular scales. The number of scales just below the capitulum reached 52 in V. veranae. The number of scales just below capitulum in V. buckeridgei is about 23. In addition, peduncular scales in V. buckeridgei protrude more outwards from the peduncle in the basal region.

The first mandible tooth of V. parensis and V. fijiensis is much larger than the first tooth in V. buckeridgei and V. verenae sp. nov. In addition, the inferior angle of the mandible of V. scotiaensis contains dense setae, whilst the inferior angle of V. buckeridgei and V. verenae sp. nov. has several sharp spines, without any setae. Mandibles of V. oshaei have four teeth, which differs from the other four Vulcanolepas species. These characters together distinguish V. verenae sp. nov. from other described Vulcanolepas species.

Molecular phylogeny. Although only a single specimen was available for V. verenae sp. nov., the number of nucleotide substitution to the other species showed clear difference and barcoding gap (more than 15 bp in a 588 bp alignment) to the intraspecific nucleotide divergence in the genus (less than 5 bp; Table 1). The ratio of transition and transversion of the nucleotide substitution were larger than 1, and the region used in the present analyses does not appear to be saturated for the analyses.

Table 1 Mean number of nucleotide base substitution within species and among species, of 588 bp in COI sequences

The phylogenetic reconstruction showed that V. verenae sp. nov. is mostly closely related to V. osheai and V. buckeridgei. Vulcanolepas was recovered as paraphyletic, with Leucolepas longa being recovered sister to V. fijiensis, and a problematic position of Neolepas embedded within Vulcanolepas (Fig. 8). All species included, except for the N. zevinae/rapanui complex, were strongly supported (92% bootstrapping replicates in Maximum Likelihood and Bayesian posterior probability of 1.00). However, phylogenetic relationships among the genera were not resolved with strong support.

Fig. 8
figure 8

Phylogenetic reconstruction of neolepadid stalked barnacles based on 588 bp of the partial mitochondrial COI gene. Numbers on the branch show the Maximum Likelihood bootstrap value/Bayesian posterior probability supports. Only those higher than 70% / 0.70 are shown


The present results showed that the stalked barnacle from a hydrothermal vent field in the Mariana Trough constitutes a lineage independent from the previously described vent stalked barnacles in Neolepadidae. The mandibular and cirral morphologies, as well as molecular data (Figs. 4, 5, 6, 7, 8), support the Mariana Trough vent barnacle as an undescribed species, described as V. verenae sp. nov. herein. There appears to be two different feeding ecology among Vulcanolepas species, either harboring or feeding on epibiotic bacteria on their cirri, or simply suspension feeding. The former is applicable to V. osheai, V. buckeridgei, and V. verenae sp. nov. Dense bacteria on the cirri of Vulcanolepas was first observed in V. buckeridgei (referred to as “neolepadine Lau A” in Southward and Newman 1998) in Hine Hina site, Lau Basin. The relationship between bacteria on the cirri and the host barnacle was examined in V. oheai from the Brothers Caldera, Kermadec Arc (Suzuki et al. 2009). The combination of bacterial metabarcoding, in situ hybridization, as well as fatty-acid and carbon isotopic composition showed that the microbes on cirri of V. osheai mainly consisted of sulfur-oxidizing bacteria in Epsilonproteobacteria and Gammaproteobacteria and were likely consumed by the host barnacle (Suzuki et al. 2009). Similar dense bacteria were observed in V. verenae sp. nov., which is inferred to have a similar feeding ecology. These three species of Vulcanolepas possess elongated setae (2.59 mm in V. osheai, 4.6 mm in V. buckeridgei on average, over 3 mm in V. verenae sp. nov) compared with the other Vulcanolepas species without bacteria on their cirri; 1.62 mm on average in V. parensis; and just shy of 2 mm in V. fijiensis (Southward and Newman 1998, Southward 2005, Chan and Chang 2018). Neither Neolepas scotiaensis (originally described in Vulcanolepas and later transferred to Neolepas) nor Leucolepas longa harbor bacteria on their cirri and setae (Tunnicliffe and Southward 2004; Marsh et al. 2012; Buckeridge et al. 2013). Basic natural history observations and information can be valuable in separating closely related taxa (Sigwart and Chen 2018). Indeed, V. fijiensis apparently co-occurred with another stalked barnacle lacking ectosymbionts (“neolepadine Lau B” in Southward and Newman 1998) in the Lau Basin, and the two had different microhabitat preferences (Southward and Newman 1998). The feeding ecology of vent stalked barnacles and the behavior of cultivating epibionts may also be key characters to consider in the taxonomy of neolepadid stalked barnacles.

A key for presently known Vulcanolepas species is provided below:

  1. 1.

    First tooth of mandible large............................................2

    First tooth of mandible small...........................................3

  2. 2.

    Length ratio of antenniform to robust segments in both rami of cirrus I almost equal (~ 1).....................V. fijiensis

    Length ratio of antenniform to robust segments in both rami of cirrus I larger than 1..............................V. parensis

  3. 3.

    Setae on cirri IV–VI very long, length ratio of annulus to length of setae > 10...........................................................4

    Length ratio of annulus to length of setae in cirri IV–VI < 10 ........................................................................V. osheai

  4. 4.

    Peduncular scales small and non-protruded. V. verenae sp. nov.

    Peduncular scales large and protruded.......V. buckeridgei

The distribution of each Vulcanolepas species is limited to a specific area of the Pacific Ocean: V. osheai on the Kermadec Arc, V. buckeridgei in the Lau Basin, V. parensis on the Pacific-Antarctic Ridge, V. fijiensis in the North Fiji Basin, and V. verenae sp. nov. in the Mariana Trough. This differs from most species in the other Neolepadini genera (i.e., Leucolepas and Neolepas) which are more widely distributed (Fig. 1). Although the larval dispersal mechanisms of Vulcanolepas are not yet known, the limited existing knowledge on their reproductive traits suggest limited capability of larval dispersal. Vulcanolepas osheai brood the largest eggs among currently known neolepadids (1.05 mm in length; Buckeridge 2000), and brooding is not frequently observed, unlike in L. longa (Tunnicliffe and Southward 2004). The fragmented distribution of Vulcanolepas associated with specific geographic regions may be due to the larval dispersal and specialized in certain types of hydrothermal vents.

Our single-locus phylogenetic analysis could not fully resolve the phylogeny of the family Neolepadidae; however, the nodes with high statistical support were consistent with a multi-locus phylogenetic analyses of deep-sea vent barnacles by Herrera et al. (2015). The examination of V. verenae sp. nov. and comparisons with other Vulcanolepas species suggest that members of the genus Vulcanolepas may share a number of key characteristics, summarized in the genus diagnosis above. Nevertheless, results from the molecular phylogeny casts doubts on if all these characters are taxonomically informative, since Vulcanolepas was not recovered as monophyletic, a condition shared with previous studies (Herrera et al. 2015; Watanabe et al. 2018; Chan et al. 2019). Vulcanolepas fijiensis, which does not possess dense bacterial aggregation on the cirri, was recovered as the sister species of L. longa, both in the present analysis and in Chan et al. (2019). Vulcanolepas sp. from Hine Hina, Lau Basin, was also recovered in a clade with L. longa in a three-gene phylogeny (18S, 28S, and H3) (Pérez-Losada et al. 2008; Herrera et al. 2015). These infer that these two species may be more closely related to L. longa than other Vulcanolepas species. For the problematic V. cf. parensis sensu Plouviez et al. (2013) from the Manus Basin, the partial COI sequences (from Plouviez et al. (2013) were actually genetically indistinguishable from L. longa and should be regarded as misidentification of L. longa, as previously pointed out by Watanabe et al. (2018). Genetic data from the true V. parensis, from specimens confirmed to agree morphologically with the holotype, is therefore still unavailable. The placement of V. osheai, which has ectosymbionts, was not well-resolved in the present single-locus analysis (Fig. 8), but previous multi-locus analyses have resolved it in a basal position among a monophyletic clade with other ectosymbiont-hosting Vulcanolepas species and Neolepas species, instead of being closely related to L. longa (Pérez-Losada et al. 2008, Herrera et al. 2015). Together, these suggest that species currently included in Vulcanolepas may consist of two groups, one housing ectosymbionts that is more closely related to Neolepas (including V. osheai, V. buckeridgei, and V. verenae sp. nov.) and one without ectosymbionts that is more closely related to Leucolepas (including V. fijiensis and V. parensis). Given the topology of recent molecular phylogenies with Neolepas and Leucolepas species embedded within a paraphyletic Vulcanolepas, it is possible that hosting ectosymbionts is an ancestral character present in Neolepadini, subsequently lost in some species. Further phylogenetic analyses, as well as detailed examinations of morphological and ecological characters, will help resolve the issues with genera assignments in Neolepadidae and provide insights on the speciation and adaptation of these deep-sea barnacles.