Introduction
Marine fouling communities are important reservoirs of non-indigenous and cryptogenic species, particularly in coastal regions with extensive shipping activity, aquaculture, fishing infrastructure, and artificial hard substrata (Tyrrell and Byers, Reference Tyrrell and Byers2007; Sorte et al., Reference Sorte, Williams and Carlton2010; Pederson et al., Reference Pederson, Carlton, Bastidas, David, Grady, Green-Gavrielidis, Hobbs, Kennedy, Knack, McCuller, O’Brien, Osborne, Pankey and Trott2021; Tamburini et al., Reference Tamburini, Keppel, Marchini, Repetto, Ruiz, Ferrario and Occhipinti-Ambrogi2021). Although rapid assessment surveys and fouling panel deployments have substantially improved knowledge of introduced marine invertebrates and algae, many such surveys emphasize accessible floating docks, marina structures, and other easily sampled or accessed substrates (Pederson et al., Reference Pederson, Carlton, Bastidas, David, Grady, Green-Gavrielidis, Hobbs, Kennedy, Knack, McCuller, O’Brien, Osborne, Pankey and Trott2021; Davinack et al., Reference Davinack, Carlton and Pederson2024). Because these approaches are necessarily biased towards near-surface fouling habitats, benthic or semi-permanent artificial structures deployed below the surface, including fishing gear, remain comparatively under-sampled and may harbour taxa that are overlooked by conventional survey approaches.
Spionid polychaetes are common members of marine fouling, boring, and soft-sediment communities, and several species have achieved broad distributions through human-mediated transport (Rouse et al., Reference Rouse, Pleijel and Tilic2022; Davinack and Williams, Reference Davinack, Williams, Byers, Blakeslee and Wares2026). Among these, Boccardia proboscidea Hartman, 1940 is a morphologically distinctive spionid that inhabits tubes and can also bore into firm substrata, including stone and mollusc shells. The species was originally described from northern California (Hartman, Reference Hartman1940) and subsequently redescribed and compared with related Boccardia species by Woodwick (Reference Woodwick1963), with later accounts providing additional information on its morphology, reproduction, and larval development (Woodwick, Reference Woodwick, Reish and Fauchald1977; Blake and Kudenov, Reference Blake and Kudenov1978, Reference Blake and Kudenov1981; Gibson, Reference Gibson1997; Oyarzun, Reference Oyarzun2010; Oyarzun et al., Reference Oyarzun, Swalla, Mahon and Halanych2011; Radashevsky et al., Reference Radashevsky, Pankova, Malyar, Neretina, Wilson, Worsfold, Diez, Harris, Hourdez, Labrune, Houbin, Kind, Kuhlenkamp, Nygren, Bonifacio and Bachelet2019). B. proboscidea has since been reported from temperate regions throughout the world, including the Pacific coast of North America, the Asian Pacific (Imajima and Hartman, Reference Imajima and Hartman1964; Sato-Okoshi, Reference Sato-Okoshi2000; Abe et al., Reference Abe, Takeuchi, Taru, Sato-Okoshi and Okoshi2019), Australia and New Zealand (Blake and Kudenov, Reference Blake and Kudenov1978; Pollard and Hutchings, Reference Pollard and Hutchings1990; Jones, Reference Jones1991; Hewitt et al., Reference Hewitt, Campbell, Thresher, Martin, Boyd, Cohen, Currie, Gomon, Keough, Lewis, Lockett, Mays, McArthur, O’Hara, Poore, Ross, Storey, Watson and Wilson2004; Read, Reference Read2004), South Africa (Simon and Booth, Reference Simon and Booth2007; Simon et al., Reference Simon, Thornhill, Oyarzun and Halanych2009; Mead et al., Reference Mead, Carlton, Griffiths and Rius2011; David, Reference David2015), Argentina (Diez et al., Reference Diez, Radashevsky and Orensanz2011; Jaubet et al., Reference Jaubet, de Los Angeles Sanchez, Rivero, Garaffo, Vallarino and Elias2011), and Europe (Martinez et al., Reference Martinez, Adarraga and Lopez2006; Hatton and Pearce, Reference Hatton and Pearce2013; Kerckhof and Faasse, Reference Kerckhof and Faasse2014; Lopez and Richter, Reference Lopez and Richter2017). Its broad distribution, association with anthropogenic habitats, and repeated detection outside its originally described range have led to its treatment as an introduced or cryptogenic species in many regions.
Molecular work has shown that geographically distant populations assigned to B. proboscidea are genetically similar across mitochondrial and nuclear markers, supporting their conspecificity despite broad geographic separation (Radashevsky et al., Reference Radashevsky, Pankova, Malyar, Neretina, Wilson, Worsfold, Diez, Harris, Hourdez, Labrune, Houbin, Kind, Kuhlenkamp, Nygren, Bonifacio and Bachelet2019). Radashevsky et al. (Reference Radashevsky, Pankova, Malyar, Neretina, Wilson, Worsfold, Diez, Harris, Hourdez, Labrune, Houbin, Kind, Kuhlenkamp, Nygren, Bonifacio and Bachelet2019) recovered low sequence divergence among specimens from distant regions and found that 16S haplotypes from introduced or putatively introduced populations were often shared with north-eastern Pacific populations. They also reported the highest 16S haplotype diversity in the north-eastern Pacific and interpreted this pattern as consistent with a northern Pacific native range followed by human-mediated introductions elsewhere.
However, elevated genetic diversity should be interpreted cautiously when inferring native ranges, particularly for aquatic species with histories of shipping, aquaculture, fisheries-related transport, and repeated introductions (Davinack, Reference Davinack2025). High diversity in a region may reflect long-term existence, but it may also result from admixture of multiple introductions, uneven sampling in the true native range, or secondary contact among introduced lineages (Roman and Darling, Reference Roman and Darling2007). Conversely, true native populations may exhibit reduced genetic diversity because of historical bottlenecks, local extirpations, or incomplete sampling (Davinack, Reference Davinack2025). Thus, while the north-eastern Pacific remains a plausible source region for B. proboscidea, genetic diversity alone should not be treated as conclusive evidence of nativity without complementary historical, ecological, morphological, and biogeographical evidence.
Despite the extensive literature on B. proboscidea, confirmed records from the western North Atlantic remain absent or poorly documented. Curiously, Bailey-Brock (Reference Bailey-Brock2000) documented B. proboscidea in March 1990 from Atlantic oysters cultured in Hawai’i that were sourced from Maine. Carlton and Eldredge (Reference Carlton and Eldredge2009) reported that the fate of the adult worms and larvae in the Hawaiian site remains unknown and that there were no further published reports of the species in Hawai’i. Despite this report, there has been no direct record of B. proboscidea in Maine nor anywhere else in the north-east and larger western Atlantic region. This is notable given that extensive sampling and monitoring of nearshore benthos in Maine waters, in Massachusetts (New Bedford Harbor, Buzzards Bay, Boston Harbor, and Massachusetts Bay), and Long Island Sound dredge disposal sites have not recovered any specimens of Boccardia spp. (J. A. Blake, personal communication). This absence has been reinforced by recent rapid assessment work, aquaculture monitoring, and fouling-community studies, especially on the east and mid-Atlantic coast of the United States (Wells et al., Reference Wells, Pappal, Cao and Carlton2013; Kennedy et al., Reference Kennedy, Pappal, Bastidas, Carlton, David, Dijkstra, Duffey, Gibson, Grady, Green-Gavrielidis, Harris, Hobbs, Mauk, McCuller, Neefus, O’Brien, Osborne, Pederson, Robidoux, Tyler and Van Volkom2020; Pederson et al., Reference Pederson, Carlton, Bastidas, David, Grady, Green-Gavrielidis, Hobbs, Kennedy, Knack, McCuller, O’Brien, Osborne, Pankey and Trott2021; Fowler et al., Reference Fowler, Blakeslee, Davinack, Aguilar, Andersen, Benadon, Choong, Green-Gavrielidis, Greenberg, Hartshorn, Hobbs, Labbe, Larson, McCuller, Moloney, Parsons, Stancil, Thornber, Pederson and Carlton2025).
In this study, we report the occurrence of B. proboscidea from submerged abandoned lobster traps recovered from Long Island Sound, USA. We confirm the identification using adult morphology, scanning electron microscopy (SEM), and mitochondrial 16S rRNA sequence data. We further compare these specimens with published global B. proboscidea occurrences and sequences to assess their morphological and genetic similarity. This record represents the first confirmed occurrence of B. proboscidea from the Atlantic coast of North America and highlights the value of sampling subtidal substrates as a complementary habitat for detecting non-indigenous and cryptogenic marine species.
Materials and methods
Sampling and study location
Northeast Atlantic specimens of B. proboscidea were collected from Fishers Island Sound, an eastern extension of Long Island Sound, off Avery Point, Groton, Connecticut (USA), at a depth of 6–9 m near marker buoy 3 (Black Ledge) (41o18′27.21″ N, 72 o04′10.97″ W) (Figure 1). The worms were collected from fouling communities on abandoned lobster traps during trap retrieval by Project Oceanology (collectors James T. Carlton and Rebha Raviraj) (Figure 2). Temperature and salinity were not recorded at the time of trap retrieval. Fouling samples were collected from the traps soon after landing, returned to the laboratory, sorted alive, with selected material then preserved in 95% ethanol. These specimens, collected on 22 September 2023, were found in the crevices of colonies of the native cheilostome bryozoan Schizoporella variabilis (Leidy, 1855).
Map showing sampling locations for Boccardia proboscidea in the United States.
Photographs of recovered lobster traps in September 2023 from which samples were obtained.
Comparative specimens of B. proboscidea from Oregon in the North-east Pacific were collected by hand from a large intertidal aggregation at North Cove, Cape Arago (USA) (43°18′28.55″ N, 124°24′0.43″ W) by hand on 15 July 2005 (coll. Jason Williams) (Figure 1). This locality is ∼465 km north of the type locality for B. proboscidea in Caspar, Mendocino County, California (see Radashevsky et al., Reference Radashevsky, Pankova, Malyar, Neretina, Wilson, Worsfold, Diez, Harris, Hourdez, Labrune, Houbin, Kind, Kuhlenkamp, Nygren, Bonifacio and Bachelet2019). Temperature and salinity were not recorded at the time of collection.
Voucher specimens of both populations are deposited in the United States National Museum of Natural History, Smithsonian Institution (USNM), Washington, D.C., USA (Oregon [OR] samples: USNM 1776284–1776288; Connecticut [CT] samples: USNM 1776276–1776283); see also Supplementary Table 1.
Morphological analysis
For SEM, specimens were dehydrated through a graded ethanol series, and then critical point dried using a Samdri-795 critical point dryer (Tousimis Research Corporation, USA). Dried specimens were mounted on aluminium stubs using double-sided conductive adhesive tape and coated with gold (∼21 nm) using an EMS-550 sputter coater (Electron Microscopy Sciences, USA). Imaging was conducted using an FEI Quanta 250 scanning electron microscope (FEI, USA).
Molecular analysis
A subsample of the 2023 specimens morphologically identified as B. proboscidea, including two specimens that were used for SEM examination, was prepared for molecular identification. Four whole worms and the midsection of two others were digested in a Proteinase K and lysis buffer solution (QIAGEN, Hilden, Germany). Genomic DNA was then extracted using the DNeasy Blood and Tissue Extraction Kit (QIAGEN, Hilden, Germany) to produce aliquots of 12–38 ng/μL. A ∼400-bp fragment of the mitochondrial 16S rRNA gene was amplified using the primers of Palumbi et al. (Reference Palumbi, Martin, Romano, McMillan, Stice and Grabowski1991) and cycling conditions of Simon et al. (Reference Simon, Thornhill, Oyarzun and Halanych2009), with the only exception being an annealing temperature of 51°C for 30 s. Polymerase chain reaction (PCR) products were visualized in a 1% agarose gel stained with ethidium bromide and purified using a gel extraction kit (QIAGEN, Hilden, Germany). Amplicons were sequenced by Azenta LLC (South Plainfield, NJ, USA) using the forward primer. All sequences were deposited into the GenBank database (accession codes: PZ534488–PZ534493) (Supplementary Table 1).
Sequences were first compared against the GenBank database using BLASTn to confirm initial identifications. Raw sequences were then combined with all available 16S sequences identified as B. proboscidea in Radashevsky et al. (Reference Radashevsky, Pankova, Malyar, Neretina, Wilson, Worsfold, Diez, Harris, Hourdez, Labrune, Houbin, Kind, Kuhlenkamp, Nygren, Bonifacio and Bachelet2019) and associated GenBank records, along with select 16S rRNA sequences from other Boccardia species available on GenBank, specifically B. pseudonatrix, B. perata, and Boccardia sp.1 and Boccardia sp. 2. Sequences were aligned and edited using the MUSCLE alignment algorithm in AliView (Larsson, Reference Larsson2014). Both intra- and interspecific uncorrected genetic distances were then calculated in MEGA11 (Tamura et al., Reference Tamura, Stecher and Kumar2021) to determine the range of genetic differences among individuals. Finally, to determine genetic interrelationships of populations, sequences were collapsed into haplotypes and a minimum spanning network was constructed in Python using the HapNet package (Davinack, Reference Davinack2026).
Results
Morphological description
Specimens from OR (Figures 3a, c, e; 4a, c, e, g) and CT (Figures 3b, d, f; 4b, d, f, h) matched B. proboscidea in their suite of diagnostic characters. Prostomium was rounded to weakly incised anteriorly; caruncle extending posteriorly to near end of chaetiger 2 (Figure 3c, d). Occipital tentacle absent. Three pairs of eyes present; fixed specimens with faint bands of pigmentation along sides of prostomium.
Scanning electron micrographs of Boccardia proboscidea specimens from Oregon (a, c, e; a = USNM 1776286, c = USNM 1776284, e = USNM 1776285) and Long Island Sound, Connecticut (b, d, f; b, f = USNM 1776281, d = USNM 1776278). (a, b) Anterior end in lateral view. (c, d) Anterior end in dorsal view; arrows indicate the posterior extent of the caruncle. (e, f) Middle portion of palp, oblique ventral view; arrow labeled “f” indicates the food groove, the unlabeled arrow indicates remnants of latero-frontal cilia rows.
Scanning electron micrographs of Boccardia proboscidea specimens from Oregon (a, c, e, g; a, c, e = USNM 1776285, g = USNM 1776287) and Long Island Sound, Connecticut (b, d, f, h; b, d, f = USNM 1776281, g = USNM 1776282). (a, b) Chaetigers 4–6, showing major spines of chaetiger 5, including heavy falcate and brush-topped chaetae and ventral fascicle labeled “vf.” (c, d) Close-up of chaetiger 5 major spines. (e, f) Hooded hooks, lateral view. (g, f) Posterior segments and pygidium (g = ventral view; h = dorsal view).
Chaetiger 1 with short capillaries in both noto- and neuropodial rami. Posterior notopodia with capillaries (no modified spines present) (Figure 3a–d).
Palps ∼1/3 to 1/2 the length of body, with a distinct longitudinal food groove extending along the frontal surface (Figure 4c, d), frontal ciliary bands present; and evidence of a row of latero-frontal cilia and non-motile cirri on papillae scattered on abfrontal surface (latter two groups in poor condition due to the fact that specimens were not fixed in glutaraldehyde) (Figure 4c–f).
Chaetiger 5 larger than anterior and posterior chaetiger, with modified spines of two distinct types: heavy falcate spines and bristle-topped spines (Figure 4a–d). OR specimens typically with three falcate and four bristle-topped spines, whereas CT specimens with two falcate and three brush-topped spines (based on SEM observation; therefore, additional spines below surface). Ventral fascicle of 6–7 capillaries present in all specimens (Figure 4a, b); dorsal capillaries and postchaetal lamellae absent.
Bidentate hooded hooks from chaetiger 7, shaft slightly curved, without constriction (Figure 4e, f); up to 6 and 12 in series (CT and OR, respectively), accompanied by inferior capillaries.
Branchiae on chaetigers 2–4 and from chaetiger 6 to near the posterior end, becoming minute or absent in posterior most chaetigers.
The pygidium forming a large disc-disk shaped structure with pair of dorsal lobes and ventral lobes; lobes more prominent in OR specimens (Figure 4g), reduced in CT specimens (Figures 3b, 4h).
Remarks: Although the OR and CT specimens were consistent in diagnostic characters, the specimens were distinct in size, with CT specimens being smaller, up to 6.0 mm in length and width of 0.6 mm for 36 chaetigers, whereas OR specimens were larger, up to 23.7 mm and width of 1.7 mm for 104 chaetigers. The differences in number of fifth chaetiger spines, number of hooded hooks, and pygidium morphology between the OR and CT specimens likely reflect this size difference.
Molecular analyses
After alignment and trimming to a common length with available Boccardia GenBank sequences, 264 bp were retained for molecular analyses. BLASTn searches recovered top matches to B. proboscidea (99.23–100% identity match), supporting the morphological identification of the CT specimens.
Uncorrected p-distances further supported the assignment of the CT specimens to B. proboscidea. Genetic distances among the six CT specimens ranged from 0.000 to 0.015. Distances between the CT specimens and published B. proboscidea sequences from GenBank were similarly low, ranging from 0.000 to 0.015. In contrast, distances between the CT specimens and other Boccardia taxa were substantially higher, ranging from 0.100 to 0.224. CT specimens differed from B. perata by 0.100–0.108, from Boccardia sp. 1 by 0.140–0.147, from Boccardia sp. 2 by 0.143–0.151, and from B. pseudonatrix by 0.216–0.224. Thus, the CT specimens were genetically indistinguishable or nearly indistinguishable from published B. proboscidea sequences, but clearly divergent from other Boccardia taxa included in the analysis.
Haplotype analysis of the 53 aligned B. proboscidea 16S rRNA sequences recovered 20 haplotypes, 3 of which were shared among two or more geographic regions and 17 of which were restricted to a single region in the present dataset. The haplotype network recovered the CT specimens within the broader global 16S haplotype diversity of B. proboscidea (Figure 5). The six CT specimens were distributed among five haplotypes. One CT specimen belonged to a broadly distributed haplotype shared with specimens from Argentina, Australia, California, England, France, South Africa, and Vancouver, Canada. Two CT specimens belonged to a haplotype shared only with California. The remaining three CT specimens each represented unique haplotypes not observed among the published sequences included in the analysis.
Haplotype network for Boccardia proboscidea based on mitochondrial 16S ribosomal RNA sequence data. Size of circles is representative of the number of individuals with that haplotype. Smallest circles represent a haplotype frequency of one. Each connecting line between haplotypes represent one mutation step.
Discussion
The present record represents the first confirmed occurrence of B. proboscidea from the Atlantic coast of North America. Although the species has been reported from multiple temperate regions worldwide, including Europe, Argentina, South Africa, Australia, New Zealand, the Asian Pacific, and the Pacific coast of North America, confirmed records from the western North Atlantic have remained absent. The CT specimens agree morphologically with B. proboscidea, and molecular data further support the identification, with CT specimens showing 99.23–100% BLAST identity to published B. proboscidea sequences and low 16S divergence from global B. proboscidea material.
B. proboscidea now joins the more than 100 non-indigenous marine and estuarine invertebrates and algae in New England (Carlton, Reference Carlton and Latimer2014; JTC, personal observations). Many of these species are thought to have arrived in the Northwest Atlantic through maritime vectors, particularly vessel hull and niche-area fouling and ballast water. Given the global invasion history of B. proboscidea and its association with anthropogenic habitats, a similar introduction pathway is plausible for the Long Island Sound population. The CT specimens shared one 16S haplotype with several globally distributed populations and another haplotype with California specimens, while three additional CT haplotypes were not observed among the published sequences included in our analysis. These data place the CT material within the known global mitochondrial diversity of B. proboscidea, but they do not identify a source population or introduction pathway. Because the present analysis is based on a short mitochondrial 16S fragment and does not include nuclear markers, it cannot distinguish among potential introductions from the north-eastern Pacific, Europe, another introduced region, or an incompletely sampled source population.
The recovery of five 16S haplotypes among only six CT specimens is notable given the limited sample size. This pattern contrasts with the expectation of a single, genetically depauperate founding event and may instead reflect a long-established but previously overlooked population, repeated introductions, or introduction from a genetically diverse source. However, this inference should be treated cautiously. The 16S rRNA fragment used here is relatively conserved, and the available global dataset is geographically uneven. Radashevsky et al. (Reference Radashevsky, Pankova, Malyar, Neretina, Wilson, Worsfold, Diez, Harris, Hourdez, Labrune, Houbin, Kind, Kuhlenkamp, Nygren, Bonifacio and Bachelet2019) found the highest 16S haplotype diversity in north-eastern Pacific populations and interpreted this as consistent with a northern Pacific native range, but as noted above, elevated genetic diversity alone does not definitively establish nativity. Multiple introductions, admixture, incomplete sampling, and historical bottlenecks can all alter regional patterns of genetic diversity (Roman and Darling, Reference Roman and Darling2007; Davinack, Reference Davinack2025). Thus, while the north-eastern Pacific remains a plausible source region for B. proboscidea, additional sampling and more variable markers will be needed to reconstruct the invasion history of the Long Island Sound population.
The detection of B. proboscidea on abandoned lobster traps also highlights the importance of sampling submerged artificial substrata. Rapid assessment surveys and fouling-community surveys with deployed panels have contributed substantially to knowledge of marine invertebrates in New England and adjacent regions, but these efforts often emphasize floating docks, marine structures, and other accessible near-surface habitats. In contrast, lobster traps and other benthic or semi-permanent fishing gear provide hard substratum at depth and may support fouling assemblages that are not well represented in conventional survey datasets (Enrichetti et al., Reference Enrichetti, Bavestrello, Betti, Rindi, Tregrosso and Bo2021; Legare et al., Reference Legare, Nichols and Ludwig2024). The absence of earlier records of B. proboscidea from New England may therefore reflect limited sampling of appropriate habitats rather than true prior absence from the region.
Preliminary examination of additional lobster trap-associated fouling material collected from a nearby site in the lower Thames River, Connecticut, in December 2025 at 5.8–7.6 m depth revealed spionids consistent with the B. proboscidea described here, both in terms of morphology and size, and also associated with the bryozoan S. variabilis (AAD unpublished data). While these 2025 specimens have not yet been subjected to the same morphological, molecular, and vouchering workflow as the 2023 material, these additional collections suggest that the species could be established in Long Island Sound given their occurrence over multiple years and sites.
It should be noted that the CT specimens were recovered from tubes located within crevices in the matrix of colonies of the native cheilostome bryozoan S. variabilis on abandoned lobster traps; they were not observed boring into the bryozoan, calcareous substrata, or mollusc shells. This is important because B. proboscidea is known to occupy several habitat types, including sandy or silty tubes, firm substrata, and in some cases, mollusc shells. Radashevsky et al. (Reference Radashevsky, Pankova, Malyar, Neretina, Wilson, Worsfold, Diez, Harris, Hourdez, Labrune, Houbin, Kind, Kuhlenkamp, Nygren, Bonifacio and Bachelet2019) confirmed the conspecificity of tube-dwelling, stone-boring, and shell-associated individuals, suggesting substantial habitat flexibility within the species. The present record adds trap-associated bryozoan colonies to the range of habitats in which the species may be encountered. Whether B. proboscidea in Long Island Sound is primarily associated with bryozoan colonies, trap fouling, soft-sediments, or shell-bearing habitats remains unknown relative to their role in hosting this species, and is an area ripe for further investigation. Although, to our knowledge, B. proboscidea has not been reported from bryozoan colonies previously, this is not surprising given that the species opportunistically settles in crevices around and in a wide range of benthic invertebrates and algae (see supplement of Radashevsky et al. Reference Radashevsky, Pankova, Malyar, Neretina, Wilson, Worsfold, Diez, Harris, Hourdez, Labrune, Houbin, Kind, Kuhlenkamp, Nygren, Bonifacio and Bachelet2019).
The specimens of B. proboscidea from OR matched the length and number of chaetigers of those reported from the region of the type locality where they obtain robust sizes (typically 100–150 chaetigers) (Radashevsky et al., Reference Radashevsky, Pankova, Malyar, Neretina, Wilson, Worsfold, Diez, Harris, Hourdez, Labrune, Houbin, Kind, Kuhlenkamp, Nygren, Bonifacio and Bachelet2019). However, the CT specimens were notably smaller (<40 chaetigers). Given that the CT specimens were collected in late September and December, these individuals likely would have had sufficient time to grow to larger sizes if resulting from larvae having been released in the spring/early summer. However, spionid populations can exhibit distinct size distributions, reflecting timing of recruitment and potential variation in growth conditions (Zajac, Reference Zajac1991). In addition, the CT specimens were also much smaller in size than those of B. proboscidea reported from California, which attained sizes ranging from 30 to 35 mm (Woodwick, Reference Woodwick1963). The observed size disparities are in need of investigation. Differences in adult size may reflect recruitment timing, local growth conditions, temperature, productivity, food availability, sediment or microhabitat characteristics, or life-history variation. If CT and OR populations differ in reproductive mode, fecundity, or the relative production of planktotrophic versus adelphophagic larvae, these differences could also influence allocation between growth and reproduction and contribute to geographic variation in body size. Additional seasonal sampling and reproductive observations will be needed to evaluate these alternatives.
Reproductive biology is also relevant to interpreting its potential establishment and spread. The species is poecilogonous, exhibiting multiple developmental modes that differ in planktonic duration and dispersal potential (Woodwick, Reference Woodwick, Reish and Fauchald1977; Blake and Kudenov, Reference Blake and Kudenov1981; Gibson, Reference Gibson1997; Oyarzun, Reference Oyarzun2010; Oyarzun et al., Reference Oyarzun, Swalla, Mahon and Halanych2011; Blake et al., Reference Blake, Maciolek, Meißner, Purschke, Westheide and Boggemann2020). Woodwick (Reference Woodwick, Reish and Fauchald1977) described lecithotrophic development in the species, while Blake and Kudenov (Reference Blake and Kudenov1981) reported Australian populations producing capsules with nurse eggs and two larval forms, including larger adelphophagic larvae capable of short or non-pelagic development and smaller larvae capable of longer planktonic development. Gibson (Reference Gibson1997) further documented variable brood types in California, including broods producing planktotrophic larvae, broods with shorter planktonic durations, and broods containing adelphophagic offspring that hatched as advanced larvae or juveniles. Oyarzun (Reference Oyarzun2010) and Oyarzun et al. (Reference Oyarzun, Swalla, Mahon and Halanych2011) later showed that reproductive variation occurs along the west coast of North America and is associated with population structure. A later study by David and Simon (Reference David and Simon2014) emphasized that this developmental flexibility can influence invasion dynamics: planktotrophic larvae may facilitate dispersal through the water column, whereas adelphophagic larvae or juveniles may settle rapidly after hatching and enhance local retention. Such flexibility could allow B. proboscidea both to disperse among suitable habitats and to maintain local populations once established.
The potential ecological and economic significance of this record depends on how abundant B. proboscidea is or will become, how widely it occurs, and which habitats it occupies in the Northwest Atlantic. At present, the CT specimens were not observed boring into oysters, mussels, or other mollusc shells, but potential impacts on local shellfish should not be inferred from these limited data alone. B. proboscidea has been documented from mollusc shells elsewhere and belongs to a broader group of spionid polychaetes that includes economically important shell-boring species (see Davinack and Williams, Reference Davinack, Williams, Byers, Blakeslee and Wares2026). In addition to possible shellfish impacts, dense populations of B. proboscidea can also function as auto-ecosystem engineers. In Argentina, the species formed extensive reef-like aggregations of tubes on intertidal abrasion platforms, particularly in organic-enriched habitats (sewage outflow), where dense settlements altered benthic community structure and displaced the native mussel Brachidontes rodriguezii as a dominant structuring species (Jaubet et al., Reference Jaubet, Garaffo, Sanchez and Elias2013). Although the CT specimens were collected from submerged trap-associated bryozoan colonies rather than intertidal reef-forming aggregations, these records indicate that established populations of B. proboscidea can have ecological effects that are mediated by the new environment.
Targeted surveys focusing on biofouling communities are now needed to determine the distribution, abundance, habitat, and associations of B. proboscidea in the Northwest Atlantic. Priority targets should include abandoned and actively deployed lobster traps, other submerged fishing gear, marina and aquaculture infrastructure, soft sediment habitats, and colonies of S. variabilis. Because S. variabilis occurs broadly along the western Atlantic coast, from at least the Gulf of Maine to Cape Hatteras (Winston and Hayward, Reference Winston and Hayward2012), bryozoan colonies may provide a useful search target for detecting additional populations. Historical wet-preserved collections of Schizoporella, especially material collected from Long Island and adjacent New England waters, may also help determine whether B. proboscidea was present prior to its recent detection. More extensive sampling, combined with additional mitochondrial and nuclear markers, will be necessary to determine whether the CT specimens’ occurrence represents a recent introduction, a long-established overlooked population, or part of a broader but previously undocumented western North Atlantic invasion.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0025315426101763
Acknowledgements
We thank Rebha Raviraj of the Maritime Aquarium at Norwalk, Connecticut, Lili Kane of the Mystic Seaport Museum, Mystic, Connecticut and Callie Scheetz, Haylie Schwarzenbach, as well as Captain Walter Roderick, Captain Ian Morrison, Amy Stockburger, and Amanda Fall, of Project Oceanology, Avery Point, Groton, Connecticut, for collection assistance. We thank Emily Adamo and Leah John (Hofstra University) for help in the SEM work. The helpful comments of two reviewers are greatly appreciated.
Author contributions
J.T.C. performed fieldwork and conceived of the study; A.A.D., J.D.W., and A.D. conducted laboratory work and data analysis. All authors contributed to writing the manuscript.
Declaration of interest
The authors declare no conflict of interest.
Data availability statement
All 16S sequence data in this study are publicly available on GenBank (accession numbers: PZ534488–PZ534493). Specimen vouchers have been deposited in the National Museum of Natural History, Washington, DC.
Ethics and approach
Samples were collected under the authorization of the U.S. state of Connecticut (Connecticut Department of Energy and Environmental Protection, Permit # SC-23029) and Oregon (Oregon Department of Fish and Wildlife scientific taking permit 29612).
Funding statement
This study received no funding.
