Polychaetes distributed across oceans—examples of widely recorded species from abyssal depths of the Atlantic and Pacific Oceans

Abstract

Distributional ranges of selected deep-sea annelids are examined in an integrative approach using genetic markers (COI, 18S) and morphology. The source material comes from various deep-sea expeditions to the Pacific and Atlantic Oceans realized between 1998 and 2015. Selection criteria for the eventual target species are a reliably documented widespread distribution in the deep-sea, and the presence in sufficient numbers of specimens in our source material. Specimens from museum collections are also incorporated. Species studied are Sigambra magnuncus, Bathyglycinde profunda and B. sibogana, Progoniada regularis, P. cf. regularis, and Spiophanes cf. longisetus, plus three newly described species: Octomagelona borowskii sp. nov., Spiophanes australissp. nov., and Spiophanes pacificus sp. nov. Illustrated descriptions are provided and the morphological distinction to congeners discussed. Genetic diversity is highest in most frequently found species, also reflected by the large numbers of genetically divergent haplotypes. The majority of haplotypes are singletons. Pan-oceanic distribution is observed for Progoniada regularis, Bathyglycinde profunda and Sigambra magnuncus, but even species restricted to a single ocean have distributions spanning hundreds or even thousands of kilometres. Our data suggest multiple and possibly ongoing dispersal and genetic exchange between oceans, most cogent for Sigambra magnuncus.

pan-oceanic distribution, trans-oceanic distribution, deep-sea, genetic diversity, genetic marker, morphology, taxonomy, abyssal

Introduction

Biogeographic patterns of deep-sea macroinvertebrate species remain poorly known, and studying their distribution seems challenging. A review of publications on population genetics of marine invertebrates emphasized that the number of studies available for deep-sea invertebrates was very low (Taylor and Roterman 2017). The authors explained that it has been almost impossible in the past to gather the necessary number of specimens for statistically robust analyses. However, despite the paucity of information, a general pattern emerged from the review by Taylor and Roterman (2017): horizontal connectivity is extensive at the regional and oceanic scales, whereas vertical connectivity (between various depths) is limited. This hypothesis is positively tested by France and Kocher (1996) for an amphipod species, and for different bivalve species by Zardus et al. (2006) and Jennings et al. (2013), to name a few examples. In general, it appears that our knowledge about population connectivity, and consequently distribution of macroinvertebrates inhabiting the deep-sea, is restricted to a limited number of species of various groups. Employing morphospecies concepts, many species appear to be widely distributed across the deep-sea floor (McClain and Hardy 2010, Rex and Etter 2010). This also holds true for abyssal annelids (e.g. Schüller and Ebbe 2007, 2014, Fiege et al. 2010, Thiel et al. 2012, Böggemann and Dietz 2016, Maciolek 2020). However, such observations triggered the discussion about cosmopolitan species whose existence has been considered doubtful and, in contrast, a certain morphological plasticity of species was explained by the presence of cryptic species (Brasier et al. 2017, Hutchings and Kupriyanova 2018, Drennan et al. 2019, Wiklund et al. 2019, Arias and Paxton 2020). Meanwhile there are examples of molecular studies confirming the wide distribution ranges reported in morphological studies. Here the spionid Aurospio dibranchiataMaciolek, 1981a is a good representative. It was originally described from the Atlantic Ocean based on morphological characters only, with the type locality off the South American continent in about 2050 m water depth. Moreover, it was recorded by the same author over most of the Atlantic Ocean from 300 to 3600 m water depth (Maciolek 1981a). After its description, the species was also found in the tropical Pacific not far from geothermal vents and manganese nodule fields of the Pacific Ocean (Maciolek 1981b, Borowski 1996, Mincks et al. 2009), and supposedly also in Antarctic waters [Scotia and Weddell Seas, according to Mincks et al. (2009) referring to Hilbig (2004)]. Based on morphological studies of specimens from the eastern Australian abyss, Aurospio cf. dibranchiata was reported by Meißner in Gunton et al. (2021). First sequence information (16S and 18S rRNA) was provided by Mincks et al. (2009) from the manganese nodule sites in the central East Pacific and then again analysed by Neal et al. (2018), including additional molecular data from deep-water sedimentary habitats off Nova Scotia (NW Atlantic). Eventually the authors suggested that A. dibranchiata may include two morphologically similar species (Neal et al. 2018). Guggolz et al. (2020) analysed all publicly available sequences (here 16S rRNA and 18S) for Aurospio Maciolek,1981 and Prionospio Malmgren, 1867, complemented by new sequences for these genera from the Clarion Clipperton Fracture Zone (CCZ) in the Pacific and the tropical Atlantic. The authors identified different lineages of Aurospio cf. dibranchiata of which two evidently had a pan-oceanic distribution. In the same publication, trans-oceanic distribution of several Prionospio species was shown as well, in some instances even with 16S rRNA haplotypes shared between Atlantic and Pacific populations (Guggolz et al. 2020). Sequence information also provided evidence for a pan-oceanic distribution of the bone-eating worm Osedax rubiplumusRouse et al., 2004, which occurs on both sides of the North Pacific, in the Southern Ocean, and in the Indian Ocean (Zhou et al. 2020). Other publications, employing either exclusively molecular tools or in combination with morphological studies, report the wide distribution of selected polychaete species in one ocean (e.g. Hurtado et al. 2004, Bors et al. 2012, Schüller and Hutchings 2012, Neal et al. 2014, Georgieva et al. 2015, Eilertsen et al. 2018, Kobayashi et al. 2018, Drennan et al. 2019, Guggolz et al. 2019).

In conclusion, our knowledge about distributional ranges of deep-sea polychaetes has grown over the last decades mainly due to increased sampling effort and the integration of molecular tools into the analyses. But this accounts for few species only and is still rare on the trans- and pan-oceanic scale. Another downside is that molecular studies are often not accompanied by detailed studies of morphology, which reduces the potential benefit for other fields of research.

During several major expeditions between 1998 and 2015 to the North and South Atlantic, as well as the central and NW Pacific, benthic organisms have been collected from abyssal and bathyal depths, and became available for further studies. Based on our preliminary morphological identifications, several annelid species documented to be widely distributed in the deep-sea were present in this material. Thus, the opportunity for studying them in detail and verifying their potential occurrences in different oceans was given. The key question of the present study was whether these species are indeed geographically widespread, i.e. with wide occurrences within the Atlantic or Pacific (trans-oceanic) or even across both oceans (pan-oceanic), or whether they constitute an assemblage of closely-related species with geographically restricted distributions. The observed distribution patterns were discussed and compared to those of other deep-sea taxa.

Material and methods

Source material

Specimens were collected during various deep-sea expeditions to the Pacific and Atlantic Oceans conducted under the leadership of different institutions (Supporting Information, Table S1). The German Centre of Marine Biodiversity Research (DZMB), Senckenberg Research Institute and Museum, was involved in all these cruises, either by logistic support, participation of staff members, data and sample management, subsequent sample processing in the lab, or by contributions to the scientific analysis. Some of the studied specimens had already been registered in scientific collections, and were provided on loan for re-examination. In the present study, the focus is on selected deep-sea annelid species (‘polychaetes’) that were morphologically identified from the source material to species or at least generic level. Primary selection criterion for the target species was a documented wide distribution in benthic deep-sea habitats, preferably in different oceans. The respective species were also expected to be found regularly in the deep-sea and to be present in higher numbers (>10 specimens) in our samples in order to have sufficient specimens for both molecular analysis and morphological examination. Eventually, species of the following genera were included: SigambraMüller, 1858 (Pilargidae), OctomagelonaAguirrezabalaga et al., 2001 (Magelonidae), ProgoniadaHartman, 1965 and BathyglycindeFauchald, 1972 (both Goniadidae), and SpiophanesGrube, 1860 (Spionidae).

Although the terms ‘Polychaeta’ and ‘polychaetes’ refer to a paraphyletic taxon and should be replaced by ‘Annelida’ and ‘annelids’, respectively, we decided to use them here as colloquial terms since they are still widely used for marine bristle worms in applied sciences and by environmental agencies and consultants.

Sampling and sample processing

The majority of examined specimens were collected using a Brenke-type epibenthic sledge (EBS) following standard deployment procedures (Brenke 2005). Few specimens came from box core samples. As a standard procedure all samples were sieved with cold seawater through 500 μm and 300 μm mesh. Samples designated for molecular studies were transferred to pre-cooled 96% ethanol after sieving, and subsequently stored in a cold storage room. The ethanol was exchanged repeatedly until stable concentrations were reached. Sorting of material for molecular studies was partially done on board of research vessels and completed in the home institutions. Samples preserved in borax-buffered formalin were kept in their originial fixation and thoroughly washed with water before sorting in the lab.

Polychaetes from projects DISCOL 1–3 (Peru Basin, SE Pacific), not yet deposited in any collection, were also available for study. Christian Borowski (Max Planck Institute for Marine Microbiology, Bremen, Germany) studied these specimens in the scope of his Ph.D. thesis (Borowski 1996). However, this publication does not constitute a published work in the sense of the International Commission of Zoological Nomenclature (ICZN) (1999) with regard to the nomenclatorial acts contained because it does not meet the necessary requirements, especially with regard to the number of physical copies, wide dissemination and archiving (ICZN 2012: Article 8.1., Recommendation 8A, B, E). In the present study, Borowski’s work is now reviewed and put into the current scientific context. The description of Octomagelona borowskii sp. nov. presented herein is based on specimens collected and tentatively identified by Borowski (1996) as Magelona sp. A. Additional specimens of this species became available from expeditions BioNod, MANGAN and ‘JPI Oceans—DISCOL revisited’ (Supporting Information, Table S1).

Detailed information about the origin of samples/specimens is listed in the Supporting Information, Table S2. Voucher specimens are deposited at the Senckenberg Museum Frankfurt, Germany (SMF) and the Zoological Museum Hamburg (ZMH), Germany.

Molecular methods

Total genomic DNA was extracted from ethanol-preserved animals. Depending on the size of the specimens, either the complete specimen (individuals <3 mm) or a small body part (individuals >3 mm) was used for DNA extraction using the QIAmp Tissue Kit (Qiagen GmbH, Hilden, Germany) or the NucleoSpin Tissue Kit (Macherey-Nagel, Düren, Germany), following the manufacturers’ protocol.

The COI barcoding fragment was amplified using the primer pairs (LCO1490 GGTCAACAAATCATAAAGATATTG/HCO2198 TAAACTTCAGGGTGACCAAAAAATCA; Folmer et al. 1994) either without or with M13-FP and M13-RpUC tails for sequencing (Messing 1983). Amplification and purification steps were done according to protocols described earlier (Meißner and Götting 2015). Purified polymerase chain reaction (PCR) products were sequenced in both directions at the sequencing facilities of Macrogen (Amsterdam, The Netherlands) or GATC (Konstanz, Germany). Forward and reverse sequences were assembled and quality checked using GENEIOUS v.7.0.4 (Kearse et al. 2012). Sequences were deposited in GenBank and accession numbers are listed in the Supporting Information, Table S2 (GenBank accession numbers OP965564–OP965637). Additional sequences used in the analyses were mined from GenBank (https://www.ncbi.nlm.nih.gov/genbank/). For each taxon studied, COI sequences were aligned with ClustalW with default settings in BIOEDIT (Hall 1999). The COI alignments were checked for indels and potential stop-codons in MEGA X (Kumar et al. 2018) and cropped to reduce the amount of missing data.

We also generated 18S rRNA sequences using the primers 18Sfw CCTAYCTGGTTGATCCTGCCAGT/18L GAATTA­CCGCGGCTGCTGGCACC (Halanych et al. 1995, Englisch and Koenemann 2001). However, the obtained sequences were too conservative to be informative for our purposes. Neither do they contribute to the discussion of gene flow between different localities, nor do shared 18S genotypes between species indicate on-going gene flow or similar here but probably represent the ancestral genotype that remained unchanged in and among these species. We, therefore, decided to not present or discuss this data in any more detail but the sequences were deposited at GenBank for future studies (see Supporting Information, Table S2; GenBank accession numbers OP980999–OP981046).

Molecular genetic species delineation and population genetics

To assess the geographic distribution of species and to identify putative cryptic species, we used a combination of genetic distance analyses, species delimitation analyses [assemble species by automatic partioning (ASAP); Puillandre et al. 2021] and haplotype networks (Figs 1–5).

Figure 1.

Geographic distribution and median-joining COI haplotype network for Sigambra magnuncusPaterson & Glover, 2000. The colour-code in the network corresponds to the geographic localities, each vertical line represents one mutational step and the size of each haplotype its frequency. Haplotypes are numered 1–28, details are listed in the Supporting Information, Table S2. Available data were from both the Atlantic and Pacific Oceans; the type locality of the species located in the Porcupine Abyssal Plain, NE Atlantic Ocean, is also indicated by an asterisk in the map.

Figure 2.

Geographic distribution and median-joining COI haplotype networks for Bathyglycinde profunda (Hartman and Fauchald, 1971) and B. sibogana (Augener and Pettibone in Pettibone, 1970). The colour-code in the network corresponds to the geographic localities, each vertical line represents one mutational step and the size of each haplotype its frequency. Haplotypes are numered 1–28, details are listed in the Supporting Information, Table S2. Available data came from both the Atlantic and Pacific Oceans; type locality of B. profunda in the Sargasso Sea, NE Atlantic Ocean, and of B. sibogana off Sulawesi, Gulf of Boni, is indicated by asterisks in the map.

Figure 3.

Geographic distribution and median-joining COI haplotype networks for Progoniada regularisHartman, 1965 and P. cf. regularisHartman, 1965. The colour-code in the network corresponds to the geographic localities, each vertical line represents one mutational step and the size of each haplotype its frequency. Haplotypes are numered 1–11, details are listed in the Supporting Information, Table S2. Available data came from both the Atlantic and Pacific Oceans; the type locality of P. regularis located off New England, NE Atlantic Ocean, is indicated by an asterisk in the map.

Figure 4.

Geographic distribution and median-joining COI haplotype network for Octomagelona borowskii sp. nov.. The colour-code in the network corresponds to the geographic localities, each vertical line represents one mutational step and the size of each haplotype its frequency. Haplotypes are numered 1–6, details are listed in the Supporting Information, Table S2. Data available came from the Pacific Ocean exclusively; the type locality of the new species is in the DISCOL area, indicated by the colour-coded asterisk.

Figure 5.

Geographic distribution and median-joining COI haplotype networks for Spiophanes spp.. The colour-code in the network corresponds to the geographic localities, each vertical line represents one mutational step and the size of each haplotype its frequency. Spiophanes adriaticusD’Alessandro et al., 2019 from shallow waters of the Adriatic Sea is also displayed because sequence information deposited for this species suggest a close relationship to S. pacificus (which seems dubious). Haplotypes are numered 1–30, details are listed in the Supporting Information, Table S2. Data available came from both the Atlantic and Pacific Oceans; type localities of S. longisetus and S. abyssalis are located in the North Atlantic Ocean, those of S. australis sp. nov. is in the Argentine Basin and of S. pacificus sp. nov. in the NW Pacific, each indicated by colour-coded asterisks.

Pairwise genetic distances were calculated separately as uncorrected p-distances for each family in MEGA X (Kumar et al. 2018). ASAP was run using the web-based version of the software (https://bioinfo.mnhn.fr/abi/public/asap/asapweb.html#); all available sequences were combined into a single alignment and jointly analysed.

Median-joining haplotype networks were calculated with NETWORK 10.1.0.0 (Bandelt et al. 1999) and redrawn with Adobe Illustrator 2019. Haplotype networks were calculated for COI.

Population genetic inferences were performed with ARLEQUIN 3.5.2.2 (Excoffier and Lischer 2010). Haplotype diversities (h) and nucleotide diversities (π) were calculated, as well as the neutrality tests Tajima’s D and Fu’s Fs, to detect deviations from neutrality (e.g. to assess whether populations are stable or expanding). These parameters were calculated for each species (as identified by ASAP, see above) and within species for each ocean and/or population comprising at least five individuals. The neutrality tests assess deviations from the mutation-drift-equilibrium. Positive values are indicative of mutation-drift-equilibrium, which is typical of stable populations or populations with a recent bottleneck if the values are particularly large. Negative values indicate an excess of rare haplotypes, typically the result of recent expansions, often preceded by a bottleneck. Significantly negative values (significant at the 0.05 level) reveal in both tests historic demographic expansion events. Significance was tested by 1000 permutations.

Population differentiation within species was assessed via ϕ_ST and AMOVAs. In an AMOVA (analysis of molecular variance), genetic subdivision among and within different predefined hierarchical levels is assessed. In each AMOVA three hierarchical levels are defined: (i) between groups (=oceans; in case of Spiophanes pacificus sp. nov. East and West Pacific); (ii) among populations within groups (here single populations localities within oceans; each population is represented by a different colour in the figures); and (iii) within populations. AMOVAs were performed if at least five individuals were present in each ocean (=group). Pairwise ϕ_ST was also only assessed between populations with at least five individuals each, which was the case in two instances only (Sigambra magnuncusMüller, 1858 and Spiophanes pacificus sp. nov.).

Morphology

Morphology was studied using light and scanning electron microscopy (LM, SEM) and confocal laser scanning microscopy (cLSM). For LM studies of Spiophanes, methyl green staining was applied as described in Meißner et al. (2011), allowing the study of the external morphology in the best possible way, as well as the detection of glandular features in the integument. Drawings were made using a camera lucida. Light micrographs were taken with various digital cameras attached to the light microscope (Olympus SC50 and UC90, Canon EOS 5D). For SEM studies, specimens were dehydrated in a graded ethanol series, critical-point dried, sputter coated with gold/ gold-paladium, and examined with Leo 1525 and CamScan CS24 scanning electron microscopes. For cLSM studies, specimens were stained with Shirlastain A as described in Meißner et al. (2019). High-resolution datasets were collected using a confocal laser scanning microscope Leica TCS SPE equipped with an ACS APO 10.0 Å~ 0.3 dry lens using a red-light laser (excitation wavelength: 635 nm; absorption spectrum 650–762 nm). ImageJ 1.51j/FIJI (Schindelin et al. 2012) was used for post-processing digitally recorded image datasets. All plates were compiled using Adobe TM Adobe Creative Suite 6 (including Photoshop and Illustrator). The body width of all target taxa was measured as the maximum width of the anterior part of the body, in Spiophanes parapodia included but chaetae omitted (see further details in the ‘Results’ section for each taxon). The body length refers to the distance between the apical and terminal tips, cirri disregarded. Abbreviation with regard to completeness of specimens are: af (=anterior fragment), cs (=complete specimen), mf (=middle fragment), pf (=posterior fragment). Synonymy lists presented in this manuscript include only those references for which confusion with other species can by the greatest possible extent be ruled out either because the studied material has been re-examined by the authors of the present paper or was listed in publications by taxonomic specialists of the group. Following abbreviations are used for the various museums and institutions involved in deposition and loan of examined material: DZMB, Deutsches Zentrum für Marine Biodiversitätsforschung, Hamburg and Wilhelmshaven, Germany; ZMH, Zoological Museum Hamburg, Germany; SMF, Senckenberg Museum Frankfurt, Germany.

Results

Alignments

We successfully amplified and sequenced the mitochondrial COI of 74 individuals and 18S rRNA of 49 individuals (Supporting Information, Table S2, new GenBank accession numbers in bold). The COI alignments, including the sequences downloaded from GenBank, were 619 bp (Progoniada, 13 individuals), 619 bp (Bathyglycinde, 61 individuals), 621 bp (Octomagelona, nine individuals), 621 bp (Sigambra, 31 individuals), and 587 bp (Spiophanes, 38 individuals) long. No stop codons or indels were detected. The 18S rRNA alignments were 482 bp (Progoniada and Bathyglycinde, 12 individuals), 427 bp (Octomagelona, seven individuals), 476 bp (Sigambra, 15 individuals), and 459 bp (Spiophanes, 22 individuals) long.

Species delimitation

The best-scoring partition in ASAP suggested nine species in our dataset: Bathyglycinde sibogana (Augener & Pettibone in Pettibone, 1970), Bathyglycinde profunda (Hartman & Fauchald, 1971), Octomagelona borowskii sp. nov., Sigambra magnuncusPaterson & Glover, 2000, Spiophanes australis sp. nov., Spiophanes pacificus sp. nov., Spiophanes cf. longisetus and two putative species within Progoniada regularisHartman, 1965 (in the following referred to as Progoniada regularis and Progoniada cf. regularis) (Table 1; Figs 1–5). For Bathyglycinde sibogana, which was collected from different localities in the Atlantic Ocean, the intraspecific p-distances were 0.0–0.6 %, and for B. profunda, with occurrences in the Pacific and Atlantic Oceans, values were similarly low with 0.0–1.6 %. Sigambra magnuncus featured distances up to 4.2 %, but without any obvious gap in the distance distribution (Table 1; Fig. 1). Octomagelona borowskii sp. nov. had up to 1.7 % intraspecific p-distances due to a single divergent haplotype from the central Pacific (French license area) (Figure 4). However, other individuals from the same locality were identical to the remaining individuals, suggesting the observed variation represents intraspecific diversity. Specimens initially identified as Progoniada regularis include two putative species (here referred to as P. regularis and P. cf. regularis) that were separated by COI p-distances of 12.0–13.1 % and exhibited up to 1.5 % intraspecific distances each (Table 2; Fig. 3). Although ASAP suggested two putative Spiophanes species, COI p-distances, as well as morphological data (see below), suggest the presence of three species. Spiophanes australis sp. nov. differs by 12.4–14.4 % uncorrected p-distance from the other two species and is unambiguously delimited by ASAP (Table 3). The other two species, S. cf. longisetus and S. pacificus sp. nov., are differentiated by 4.0–5.8% uncorrected p-distance, with up to 2.2 % intraspecific p-distances (Tables 1, 3; Fig. 5). These two probably clustered together in ASAP due to the extensive intraspecific diversity of Sigambra magnuncus (up to 4.2%), which exceeds the interspecific distances between them. When the Spiophanes species were analysed separately, all three species were differentiated. It is furthermore noteworthy that COI sequences of Spiophanes pacificus sp. nov. are highly similar to those published for the shallow water species Spiophanes adriaticusD’Alessandro et al., 2019, separated by only six mutational steps (Fig. 5).

Table 1.

Intraspecific COI p-distances: Intraspecific uncorrected p-distances within all putative species. For interspecific distances between closely related species pairs see Tables 2 and 3

Species	Intraspecific distances
Progoniada regularis	0.0–1.5 %
Progoniada cf. regularis	0.0–0.7 %
Bathyglycinde sibogana	0.0–0.6 %
Bathyglycinde profunda	0.0–1.6 %
Octomagelona borowskii sp. nov.	0.0–1.7 %
Sigambra magnuncus	0.0–4.2 %
Spiophanes australis sp. nov.	0.0–0.7 %
Spiophanes pacificus sp. nov.	0.0–2.2 %
Spiophanes cf. longisetus	0.0–1.0 %

Table 2.

COI Intra- and interspecific uncorrected p-distances in percent for all studied species of Goniadidae

	Progoniada regularis	Progoniada cf. regularis	Bathyglycinde sibogana	Bathyglycinde profunda
P. regularis	0.0–1.5
P. cf. regularis	12.0–13.1	0.0–0.7
B. sibogana	23.3–24.1	21.6–22.0	0.0–0.6
B. profunda	26.7–28.1	24.1–25.7	22.0–23.1	0.0–1.6

Table 3.

COI Intra- and interspecific uncorrected p-distances in percent for all studied species of Spiophanes.

	Spiophanes australis sp. nov.	Spiophanes pacificus sp. nov.	Spiophanes cf. longisetus
S. australis sp. nov.	0.0–0.7
S. pacificus sp. nov.	12.4–13.6	0.0–2.2
S. cf. longisetus	13.1–14.4	4.0–5.8	0.0–1.0

Species distributions and population genetics

Wide geographic distributions, including trans-oceanic distributions, were confirmed for several of the species studied herein (Figs 1–5). Pan-oceanic distribution was observed for three species: Progoniada regularis, Bathyglycinde profunda, and Sigambra magnuncus. Other species were recorded either in a single ocean (Progoniada cf. regularis, Bathyglycinde sibogana, Octomagelona borowskii sp. nov., Spiophanes cf. longisetus and S. pacificus sp. nov.) or only from a single locality (Spiophanes australis sp. nov.). However, the species that were restricted to a single ocean had distributions spanning hundreds or even thousands of kilometers (Figs 1–5). For P. regularis, the presence of a cryptic species had no impact on the species large-scale geographic distributions. For example, P. regularis still occurs in both oceans, whereas P. cf. regularis is currently recorded only from the Atlantic.

All species with large sampling sizes (>20 COI sequences per species; e.g. Bathyglycinde profunda, Sigambra magnuncus, and Spiophanes pacificus sp. nov.) revealed high levels of genetic diversity, with large numbers of genetically divergent haplotypes as seen in the species nucleotide and haplotype diversities (Figs 1–5; Table 4). Not only did the species as a whole yield high genetic diversity, also each ocean and each single population yielded high genetic diversities. The majority of haplotypes was recorded only once or a few times; only B. profunda featured several closely related haplotypes that were collected multiple times each.

Table 4.

Overview of population genetic parameters. Key parameters are provided for each species across its studied distribution as well as for intraspecific groups (e.g. grouped by oceans or single populations), the latter only if at least five individuals were available from the respective ocean or population. Provided parameters include the number of studied individuals, the number of retrieved haplotypes, nucleotide (π) and haplotype (H) diversities as well as the results from the neutrality tests Fu’s F and Tajima’s D (including the corresponding P-values)

Species	Area	# ind.	# haplotypes	π	H	Fu’s F (P-value)	Tajima’s D (P -value)
Sigambra magnuncus	Atlantic + Pacific	31	28	0.0220 +/– 0.0113	0.991 +/– 0.012	–12.55 (0.001)	–1.23 (0.103)
	Atlantic	16	15	0.0239 +/– 0.0126	0.992 +/– 0.025	–3.97 (0.044)	–0.77 (0.218)
	MAR East	7	7	0.0241 +/– 0.0141	1.000 +/– 0.076	–0.893 (0.179)	–0.21 (0.428)
	Pacific (Peru Basin)	15	13	0.0182 +/– 0.0098	0.971 +/– 0.039	–2.836 (0.114)	–0.87 (0.202)
Bathyglycinde profunda	Atlantic + Pacific	58	25	0.0061 +/– 0.0035	0.947 +/– 0.014	–13.38 (0.0)	–1.54 (0.041)
	Atlantic	7	6	0.0066 +/– 0.0042	0.952 +/– 0.096	–1.57(0.107)	–0.95 (0.195)
	Pacific	51	20	0.0053 +/– 0.0031	0.935 +/– 0.017	–9.26 (0.0)	–1.56 (0.034)
	Pacific (CCZ German license area)	50	20	0.0052 +/– 0.0030	0.935 +/– 0.017	–9.46 (0.0)	–1.58 (0.038)
Bathyglycinde sibogana	Atlantic	3	3	0.0054 +/– 0.0049	1.000 +/– 0.272	–	–
Progoniada regularis	Atlantic + Pacific	10	8	0.0095 +/– 0.0056	0.933 +/– 0.077	–1.56 (0.182)	–0.09 (0.498)
	Atlantic	5	5	0.0087 +/– 0.0059	1.000 +/– 0.127	–1.23 (0.101)	–0.45 (0.396)
	Pacific	5	3	0.0082 +/– 0.0056	0.700 +/– 0.218	2.35 (0.859)	1.28 (0.889)
Progoniada cf. regularis	Atlantic	3	3	0.0043 +/– 0.0039	1.000 +/– 0.272	–	–
Octomagelona borowskii sp. nov.	Pacific	9	6	0.0047 +/– 0.0031	0.833 +/– 0.127	–1.25 (0.186)	–1.89 (0.005)
	Pacific (CCZ German license area)	5	3	0.0013 +/– 0.0013	0.700 +/– 0.218	–0.83 (0.089)	–0.97 (0.103)
Spiophanes pacificus sp. nov.	Pacific + Mediterranean	22	20	0.0115 +/– 0.0063	0.991 +/– 0.017	–12.52 (0.0)	–1.36 (0.087)
	West Pacific	9	9	0.0068 +/– 0.0042	1.000 +/– 0.052	–5.75 (0.002)	–0.81 (0.205)
	East Pacific	12	10	0.0099 +/– 0.0057	0.970 +/– 0.044	–3.11 (0.061)	–0.74 (0.232)
Spiophanes cf. longisetus	Atlantic	11	7	0.0053 +/– 0.0033	0.891 +/– 0.074	–1.59 (0.126)	–1.07 (0.170)
	MAR West	7	5	0.0046 +/– 0.0032	0.857 +/– 0.137	–0.91 (0.209)	–0.89 (0.220)
Spiophanes australis sp. nov.	Atlantic	5	3	0.0027 +/– 0.0022	0.700 +/– 0.218	0.28 (0.494)	–1.10 (0.740)

In all species occurring in the Atlantic, the observed genetic diversity was evenly distributed across all populations, without any obvious association between geography and haplotypes. Though haplotypes were usually not shared among populations (not surprising, given the fact that most were singletons), the haplotypes recovered at each population were not more closely related to each other than to those from other populations (Figs 1–3, 5). This was also supported by AMOVA, which assigned only a low percentage of the observed genetic diversity to the ‘among populations within oceans’ level and the majority to the ‘within populations’ level (59.6–92.4 %) (Table 5). For the East Pacific, the same overall pattern was observed as for the Atlantic. However, between eastern and western Pacific populations of Spiophanes pacificus sp. nov., the genetic differentiation was more pronounced and a clear geographic separation was evident, except for a single Eastern Pacific haplotype (Fig. 5; Table 5). This was supported by a high and significant ϕ_ST value of 0.46 between eastern and western Pacific populations.

Table 5.

Overview of AMOVA results. Populations were grouped by oceans (for S. pacificus sp. nov. by East and West Pacific), at least five individuals had to be present in each group. Variances and the corresponding variation percentage are reported as well as the respective fixation indices F_ST, F_CT and F_SC

Species	Comparisons/ groups	Among groups variance (% variation)	Among populations within groups variance (% variation)	Within populations variance (% variation)	FST (P value)	FCT (P value)	FSC (P value)
Sigambra magnuncus	Atlantic vs. Pacific	0.51 (7.2)	0.02 (0.3)	6.55 (92.4)	0.08 (0.099)	0.072 (0.486	0.00 (0.736)
Bathyglycinde profunda	Atlantic vs. Pacific	0.76 (28.7)	0.31 (11.8)	1.59 (59.6)	0.40 (0.0)	0.29 (0.309)	0.17 (0.263)
Progoniada regularis	Atlantic vs. Pacific	0.31 (9.6)	0.77 (24.0)	2.13 (66.4)	0.33 (0.060)	0.10 (0.197)	0.27 (0.203)
Spiophanes pacificus sp. nov.	East vs. West Pacific	1.06 (27.6)	0.42 (10.9)	2.36 (61.5)	0.38 (0.0)	0.28 (0.263)	0.15 (0.155)

Notably, in two of the three pan-oceanically distributed species, Sigambra magnuncus and Progoniada regularis, the genetic diversity was more or less evenly distributed across both oceans with no apparent geographic association of haplotypes (Figs 1, 3). This was particularly striking in S. magnuncus with its highly divergent haplotypes. In both species, AMOVA assigned <10 % of the observed genetic variability to occur among oceans and the corresponding F_ST and ϕ_ST (ϕ_ST = 0.06) values were low and non-significant (Table 5). Bathyglycinde profunda showed stronger geographic structure among oceans (Fig. 2; Table 5); however, only one or a few mutations separated Atlantic and Pacific haplotypes and the Atlantic haplotypes were partly nested within those from the Pacific.

Both neutrality tests were consistently negative, sometimes even strongly negative, across nearly all species and oceans (Table 4). Exemptions were the Pacific populations of Progoniada regularis and Spiophanes australis sp. nov. in the Atlantic with low positive values. Particularly strong negative and largely significant (P < 0.05) values were observed for species-wide analyses of Bathyglycinde profunda, Sigambra magnuncus, and Spiophanes pacificus sp. nov., as well as the Pacific populations of B. profunda, suggesting recent expansions in these species and populations. All other tests were closer to zero and (mostly) non-significant, suggesting long-term stability or slight expansions.

Morphology of delimited species

The morphology of suggested putative species is studied in detail. We here provide illustrated descriptions. Information about examined material is provided in the Supporting Information, Table S2.

Systematic account
Pilargidae Saint-Joseph, 1899,
Genus SigambraMüller, 1858

Type species:

Sigambra grubiiMüller, 1858. Gender: male (original spelling).

Diagnosis:

(Salazar-Vallejo et al. 2019, emended.) Body depressed. Prostomium with three antennae, longer than palps; palps biarticulate. Tentacular cirri as long as half width of tentacular segment. Parapodia biramous. Dorsal and ventral cirri foliose to tapered, dorsal ones usually longer than ventral ones. Notopodia with dorsal hooks along numerous segments, sometimes with accessory capillaries. Neurochaetae flattened capillaries of varying length often twisted distally, one edge finely denticulate; additional shorter pectinate chaetae present in some species.

Remarks:

The diagnosis by Salazar-Vallejo et al. (2019) is herein emended in order to include species without shorter pectinate chaetae (pectinates), which have not been observed in specimens of Sigambra magnuncus studied here. A most recent key to species of Sigambra, together with a synoptic table of characters for all 26 species currently accepted (Read and Fauchald 2022), is presented in Bhowmik et al. (2021).

Sigambra magnuncus Paterson & Glover, 2000
(Figs 6A–J, 7A, B, 8A–H)

‘Sigambra armata’ Borowski, 1996: 116–121, figs 24A–D, 25A–D (manuscript name).

Figure 6.

Sigambra magnuncusPaterson & Glover, 2000, SEM studies: A, anterior end, dorsal view (arrows pointing to papillae). B, anterior end, dorsal view. C, anterior end, frontal view. D, anterior end, lateral view, right side. E, posterior end, lateral view, right side. F, neurochaetae, chaetiger 6, left side. G, posterior region, dorsal view. H, notopodia, chaetiger 10/11 with hooks and accompanying capillaries. I, dorsal hook and accompanying capillaries, chaetiger 7. J, detail of papillae row, chaetiger 1, dorsal.—A, E, J, SMF 30569 (SEM stubs 1332, 1333), Peru Basin; B, SMF 30538 (SEM stub 1322), Peru Basin; C, D, H, I, SMF 30562 (SEM stub 1321), central Atlantic; F, G, SMF 26945 (SEM stub 617), Angola Basin. Scale Bars: A–E, 300 µm; F, J, 10, µm; G, 100 µm; H, I, 30 µm.

Figure 7.

Sigambra magnuncusPaterson & Glover, 2000, SEM studies: A, posterior chaetigers, lateral view (arrows pointing to secretions at openings of interramal parapodial glands). B, chaetigers 22–24, dorsal view (arrows pointing right to segmental rows of papillae; arrows pointing left to secretion of interramal parapodial gland).—A, SMF 30569 (SEM stub 1333), Peru Basin; B, SMF 26945 (SEM stub 617), Angola Basin. Scale Bars: A, B, 30 µm.

Figure 8.

Sigambra magnuncusPaterson & Glover, 2000: A, anterior end, dorsal view, with interramal secretions as dark patches. B, anterior end, dorsolateral view. C, posterior end, dorsal view. D, cross-section of posterior chaetiger, anterior view. E, first chaetiger, right side, anterior view. F, 12th chaetiger, right side, anterior view with interramal secretions protruding. G, notopodial capillary chaeta. H, neurochaeta.—A, C, SMF 30537; E, F, SMF 30534; B, D, G, H, SMF 30536, all Peru Basin. Scale Bars: A, B, C, 500 µm; D, 200 µm; E, F, 250 µm; G, H, 10 µm. Originals C. Borowski (1996), reprinted with permission of original publisher.

Sigambra magnuncusPaterson and Glover, 2000: 167–169, figs 1–5. – Nishi et al. 2007: 65 (in table 1). – Böggemann 2009: 376–378, figs 114A–F, 115A–K. – Fiege et al. 2010: appendix, tables 3 and 5 (name only). – Salazar-Vallejo et al. 2019: 46 (key only). – Bhowmik et al. 2021: 56 (in table 3), 62 (key).

Material examined: Central Atlantic Ocean,

Meteor Seamount deep sea, M 79-1 (DIVA 3), station 636-1, 4339 m (SMF 30563). Mid-Atlantic Ridge, Central, SO 237 (VEMA-Transit), station 8-4, 5176 m (SMF 30564, ZMH-P 27949, ZMH-P 27950, ZMH-P 29710, ZMH-P 29711, ZMH-P 29712). Mid-Atlantic Ridge, East, SO 237 (VEMA-Transit), station 6-7, 5085 m (ZMH-P 29707); station 6-8, 5119 m (SMF 30562, +SEM 1321, ZMH-P 27947, ZMH-P 27948, ZMH-P 29708); station 4-8, 5735 m (ZMH-P 27946, ZMH-P 29705,ZMH-P 29706); station 2-6, 5520 m (ZMH-P 27943). SW Atlantic Ocean, Brazil Basin N, M 79-1 (DIVA 3), station 605-1, 518 m, (SMF 30585). SE Pacific Ocean, Peru Basin (DISCOL area), SO 61 (DISCOL 1), station 1264-1, 4133 m (SMF 30534); station 1270-1, 4154 m (SMF 30535); SO 64 (DISCOL 2), station 1417-1, 4150 m (SMF 30536, SMF 30538, +SEM 1322, SMF 30540); station 1421-1, 4175 m (SMF 30541, ZMH-P 30445); station 1425-1, 4162 m (SMF 30537); SO 77 (DISCOL 3), station 1456-1, 4165 m (SMF 30543); station 1460-1, 4162 m (ZMH-P 30444, SMF 30545); station 1464-1, 4154 m (SMF 30546, SMF 30547); SO 242 (JPIO-DISCOL 1), station 37-1, 4161 m (SMF 30577, SMF 30578, SMF 30579, SMF 30580, SMF 30581, SMF 30582, SMF 30588); station 81-3, 4157 m (SMF 30465, SMF 30466); station 85-4, 4147 m (SMF 30468); station 93-5, 4185 m (SMF 30571, SMF 30574, SMF 30575, SMF 30576); station 117-7, 4154 m (SMF 30572, SMF 30573); station 122-8, 4078 m (SMF 30569, SEM 1332 + 1333; SMF 30570); station 126-9, 4257 m (SMF 30583, SMF 30584). – for details and additional specimens see the Supporting Information, Table S2.

Measurements of largest specimen studied: central Atlantic (SMF 30562): length: 8.7 mm, width: 0.5 mm (excl. parapodia), 57 chaetigers.

Description:

Body slightly dorsoventrally flattened except anteriormost chaetigers, anterior chaetigers widest (Figs 6A, B, 8A, B). Prostomium wider than long, rounded anteriorly, slightly indented laterally at insertion of lateral antennae and peristomium (Fig. 6A, C, D). One pair of palps and three antennae. Palps biarticulate, palpophores fused basally; palpostyles digitiform, pointing antero-ventrally (Figs 6B, 8B). Median antenna inserted near posterior margin of prostomium, ca. 1.5 to more than 2× longer than laterals, inserted medio-laterally. Nuchal organs not observed. Eyes absent. Pharynx with eight papillae of equal size (Fig. 8B). Peristomium 2 to 3× longer than first chaetiger, two pairs of tentacular cirri inserted at anterior end, dorsal cirri slightly longer than ventral; longer than lateral antennae but shorter than median (Figs 6A, B, 8A, B). Transverse row of dense small papillae across dorsum (best visible in SEM). Some specimens with additional row along posterior margin of following segments but fewer papillae and less dense; papillae inverted pyriform, apically indented [Fig. 6A (arrow), B, J].

Parapodia biramous; parapodia in chaetigers 1–3 small, oriented laterally; middle and posterior parapodia larger (largest ca. above chaetigers 9–10) with notopodia shifted dorsally resulting in wide gap between neuro- and notopodia; notopodia of both sides coming close together dorsally (Figs 6G, H, 8A–D). Numerous specimens with fibrous secretions from pores between parapodial rami in chaetigers from midbody to posterior (from chaetigers 4–10 in specimens from DISCOL 1–3) (Figs 7A, B, 8A, B, F). Notopodia of chaetigers 1–2 small, bluntly triangular; increasing in length from chaetiger 3. Dorsal cirrus of first chaetiger cirriform, longer than median antenna and 2× length of tentacular cirri; cirri of following chaetigers shorter than tentacular cirri (Figs 6A, B, D, 8A, B, E). Neuropodia triangular with wide base and conical tip with fascicle of chaetae. Ventral cirri from chaetiger 1 but absent in chaetiger 2, similar in shape to dorsal ones and subequal or slightly longer (Fig. 6D). Notochaetae absent from chaetigers 1–2; single prominent dorsal hook with tip curved strongly inwards from chaetiger 3 to near posteriormost chaetigers; hooks of both sides crossing in midline resembling zipper in dorsal view. Hooks accompanied by short, thin capillaries, one in anterior chaetigers, two in posterior ones (Figs 6D, G–I, 8D, F, G). Neurochaetae flattened capillaries of varying length tapering into fine tip, finely denticulate along one edge (Fig. 6F, 8H).

Pygidium rounded with two long anal cirri inserted ventrally, extending back to chaetigers 4–5 before last (Figs 6E, 8C).

Remarks:

Specimens match well the descriptions of Sigambra magnuncus published by Paterson and Glover (2000) and Böggemann (2009), and main distinguishing characters listed for known species of Sigambra by Nishi et al. (2007: table 1) and Bhowmik et al. (2021: table 3). Revision of the specimens described by Borowski (1996) as ‘Sigambra armata’ showed that they belong to Sigambra magnuncus.

Among the 26 described Sigambra species currently recognized (Read and Fauchald 2022), Sigambra magnuncus is closest to S. qingdaoensisLicher & Westheide, 1997, S. bidentataBritayev and Saphronova, 1981, S. healyaeGagaev, 2008, and S. papagayu Bamber in Muir and Bamber 2008, which all share eight pharyngeal papillae, the presence of notopodial capillaries, and dorsal cirri that are longer than ventral ones. However, S. magnuncus is the only species with large notopodial hooks always starting from chaetiger 3, and with ventral cirri subequal or slightly longer than dorsal ones. The median antenna in specimens of S. magnuncus is up to 2× longer than laterals but variable among specimens, making it a less reliable character for distinction among species. In S. qingdaoensis, notopodial hooks start from chaetigers 3–8, and are accompanied by a single capillary chaeta, which is also present in anteriormost chaetigers without notopodial hooks; the median antenna is twice the length of laterals, and anal cirri are relatively short, i.e. as long as lateral antennae. Sigambra bidentata shares with S. magnuncus dorsally-shifted notopodia along most of the body and hooks crossing in the dorsal midline. It differs from the latter by the presence of neurochaetae with bidentate tips (neurochaetae tapering to fine tip in S. magnuncus) and median antenna being barely longer than lateral ones. In S. healyae, notopodial hooks start from chaetiger 4 and are accompanied by a single capillary starting from chaetiger 20; neurochaetae comprise two types, both serrated and with a bidentate tip; the median antenna is 1.5× longer than lateral ones; anal cirri extend back to chaetiger 7 or 8 before last (to chaetigers 4–5 before last in S. magnuncus). Sigambra healyae is the only species among those listed above with a ventral cirrus present on chaetiger 2 and it shows reddish-brown eyespots which are absent in S. magnuncus. In S. papagayu notopodial hooks start from chaetigers 3–5, the median antenna is 1.75× length of laterals, neurochaetae are simple capillaries with a number of slender, curved, serrate capillaries located supraacicular with serrations low and rounded, and a pair of distinct pectinate chaetae located subacicular with 14–17 teeth each and a naked slender distal extension; two anal cirri extend back to chaetigers >5 before last (Bamber in Muir and Bamber 2008: fig. 1C). Sigambra papagayu is the only species among those listed above without notopodial capillaries. Sigambra magnuncus has been recorded only from abyssal depths between 3900 and 5700 m in the Atlantic and Pacific Ocean. Sigambra healyae was found at ca. 1800 m depth in the Arctic Ocean, S. bidentata between 510 and 2220 m in the Sea of Japan, S. quingdaoensis and S. papagayu in coastal waters in the Yellow Sea (near Qingdao, China) and South China Sea (Hong Kong), respectively.

Rows of characteristically shaped papillae across the dorsum, as described for Sigambra magnuncus (Figs 6A, B, 7B), have also been noted for S. bassi (Hartman, 1947), S. hanaokai (Kitamori, 1960), S. phuketensisLicher and Westheide, 1997, S. tentaculata (Treadwell, 1941), S. parva (Day, 1963), S. grubii, and S. hernandeziSalazar-Vallejo et al., 2019 (Licher and Westheide 1997, Nishi et al. 2007, Moreira and Parapar 2002, Salazar-Vallejo et al. 2019). A sensory function for these papillae has been suggested by Paterson and Glover (2000).

In specimens of S. magnuncus collected in the Peru Basin (SE Pacific), Borowski (1996) described tufts of fibrous matrix extruding from interramal pores as one to two amorphous strands fraying distally, sometimes dotted with singular cells. Pores start from chaetigers 4–10 and extend into the posterior region in most of the larger specimens, but are missing in small juveniles. Similar structures can also be seen in Paterson and Glover (2000: figs 1A and 2: whitish tufts in mid-body chaetigers and chaetiger 4, left side). They were also observed in our more recently collected specimens from the Peru Basin (SE Pacific), as well as in specimens originating from the Angola Basin (SE Atlantic) (Fig. 7A, B). These structures have been described as ‘hypertrophied gonopores’ by Salazar-Vallejo et al. (2019) for S. diaziSalazar-Vallejo et al., 2019, S. hernandezi, S. olivaiSalazar-Vallejo et al., 2019, and S. grubii. Bhowmik et al. (2021) described and figured them as parapodial glands for S. sundarbanensisBhowmik et al., 2021 between chaetigers 5 and 60, rudimentary in small and fully developed in larger specimens. Methyl green staining showed similarities to the chromophile glands in neuropodial pinnae of Tomopteridae but further histochemical and morphological studies are needed to clarify the function of these glands. Picture of a live specimen of S. hanaokai in Nishi et al. (2007: figs 2A, B) shows tufts of greenish material possibly originating from such parapodial glands, which appear to be a quite common feature in species of Sigambra.

Sigambra magnuncus was originally described from the deep North Atlantic (Porcupine Basin to Cape Verde Islands) by Paterson and Glover (2000). The extended distribution of S. magnuncus indicated for the Atlantic by our samples from the Angola and Brazil Basins, and from the Guinea Basin studied by Böggemann (2009), as well as our samples from the Peru Basin initially studied by Borowski (1996), suggest that this species might not only have a continuous longitudinal distribution along the whole deep Atlantic east and west of the Mid-Atlantic Ridge but is also present in the SE Pacific. Böggemann (2009) did not find clear population patterns in his molecular data, suggesting no restricted gene flow within the Guinea Basin. The presence of planktonic larvae among Pilargidae (Bhaud 1974, Achari 1975, Bhaud and Cazaux 1987) might provide an explanation for the wide and obviously even pan-oceanic distribution of this species.

Distribution:

NE Atlantic: 4000–5085 m (Paterson and Glover 2000); S Atlantic, Angola and Brazil Basins, 5179–5495 m (Böggemann 2009, this study); Central Atlantic, Guinea Basin and Vema Fracture Zone, 3945–5735 m (Böggemann 2009, this study); SE Pacific, Peru Basin, 4078–4257 m (this study). Molecular data confirm the distribution of Sigambra magnuncus for the Atlantic (Meteor Seamount deep sea, central Atlantic and Brazil and Guinea Basin) and the Pacific (Peru Basin) (Fig. 1).

Goniadidae Kinberg, 1865,
Genus BathyglycindeFauchald, 1972

Type species:

Bathyglycinde mexicanaFauchald, 1972. Gender: female.

Diagnosis:

(See: Böggemann 2022.) Segments uni- or biannulate. Prostomium smooth or indistinctly annulated; tip with biarticulate appendages. Proboscis with several different types of papillae, arranged in distinct longitudinal rows; macrognaths and dorsal micrognaths present; chevrons absent. First segment sometimes with only a pair of small lateral cirri and without parapodia or chaetae. Anterior part of body with uniramous parapodia, following region with biramous parapodia. Notochaetae capillaries; neurochaetae compound spinigers.

Bathyglycinde profunda (Hartman & Fauchald, 1971)
(Figs 9A–G, 10G)

Glycinde profundaHartman and Fauchald, 1971: 74, pl. 4, figs c–e.

Figure 9.

Bathyglycinde profunda (Hartman & Fauchald, 1971) SEM studies: A, prostomium, dorsal view. B, prostomium, lateral view. C, chaetigers 21/22, right side, posterior view. D, chaetigers 37/38, left side, anterior view. E, notochaetae chaetiger 49, anterior view. F, neurochaetae chaetiger 32, left side, anterior view. G, posterior end, dorsal view.—A–G, SMF 30589 (SEM stub 1318), central Atlantic. Scale Bars: A–D, 100 µm; E, F, 30 µm; G, 300 µm.

Figure 10.

LM and cLSM studies of Bathyglycinde sibogana (Augener and Pettibone in Pettibone, 1970) (A, B), SEM studies of Progoniada regularisHartman, 1965 (C–F), and Bathyglycinde profunda (Hartman & Fauchald, 1971) (G): A, LM study of proboscidial papillae of area I and area II, median part of proboscis. B, cLSM study of similar part of proboscis. C, prostomium with extended proboscis. D, chaetigers 20–25, right side, posterior view. E, chevrons. F, proboscidial papilla. G, proboscidial papillae of areas I and II (as labelled) on posterior part of proboscis.—A, B, SMF 30491, central Atlantic. C–F, ZMH-P 26697 (SEM stub 1341), central Atlantic. G, SMF 30589 (SEM stub 1319), central Atlantic. Abbreviations: rtm = retractor muscle of proboscis, long arrows pointing to papillae of area I, arrowheads pointing to papilla of area II, asterisk marking protruding bulge on papilla of area II. Scales A, B, 50 µm; C, D, G, 100 µm; E, F, 10 µm.

Bathyglycinde profundaRizzo and Amaral, 2004b: 938–942, figs 1–15, 16–21, table 1. –Böggemann, 2005: 189–193, figs 111, 112. – Böggemann 2009: 308–311, figs 41–43, 47. – Fiege et al. 2010: appendix, tables 3 and 5 (name only). – Moreira and Parapar 2015: 58–60, figs 24A–R. – Böggemann 2016: 236–237, figs 8, 11. – Böggemann & Sobczyk in Gunton et al. 2021: 47 fig.11, B, D, E.

Figure 11.

Progoniada cf. regularisHartman, 1965, SEM and LM studies: A, macro- and micrognaths of proboscis. B, macrognaths. C, micrognaths. D, chevrons. E, proboscidial papillae stained with methyl green. F, prostomium, dorsolateral view. G, chaetigers 51–53, posterior midbody, left side, anterior view.—A–G, SMF 23791, Brazil Basin N. Scale Bars: A, 200 µm, B, D, F, G, 100 µm, C, E, 50 µm.

Material examined:

Central Atlantic Ocean, Mid-Atlantic Ridge, West, SO 237 (VEMA-Transit), station 9-8, 5004 m (ZMH-P 26696); station 11-4, 5112 m (ZMH-P 26698); Mid-Atlantic Ridge, Central, SO 237 (VEMA-Transit), station 8-4, 5176 m (SMF 30493, SMF 30494, ZMH-P 26695, (SMF 30589, SEM 1318 + 1319: 33 uniramous parapodia). SE Pacific Ocean, Peru Basin (DISCOL area), SO 242 (JPIO-DISCOL 1), station 93-5, 4185 m (SMF 30496: 29 uniramous parapodia); station 126-9, 4257 m (SMF 30495: 31 uniramous parapodia). – for details and additional specimens see the Supporting Information, Table S2.

Measurements for largest specimens studied: central Atlantic (SMF 30589): length 14.2 mm, width 0.5 mm (excl. parapodia), 55 chaetigers; Peru Basin, Pacific (SMF 30496): length 7.5 mm, width 0.3 mm (excl. parapodia), 31 chaetigers.

Description:

Prostomium much longer than wide, smooth, indistinctly annulated (ca. 10 rings) (Fig. 9A, B), two longitudinal lateral grooves on each side, terminal annulus with bi-articulate appendages. Eyes absent. Proboscis with papillae of different types arranged in longitudinal rows of different areas (I–VI). Area I: short oval to rounded, pointed, II: II-1 short, bifid to tridentate, II-2–6 elongate, decreasing in length and tip less bent, uni and bidentate (Fig. 10G).

Parapodia increasing in length from anterior to posterior. Anterior uniramous chaetigers with one digitiform prechaetal and one shorter, triangular postchaetal lobe. Dorsal cirri inserted at parapodial base, foliaceous, slightly shorter than prechaetal lobe, ventral cirri about same length, digitiform (Fig. 9C); 29–33 uniramous chaetigers, notopodia reduced; following parapodia biramous, notopodia with digitiform prechaetal lobe, smaller than slender triangular neuropodial prechaetal lobe (Fig. 9D). Notopodia with simple capillaries with pointed tips (Fig. 9E). Neuropodia with heterogomph compound spinigers with long, finely serrated blades, increasing in length towards center of fascicle (Fig. 9F). Pygidium short, rounded (Fig. 9G).

Remarks:

The specimens studied from the central Atlantic and the Peru Basin agree well with the description published by Böggemann (2005). The number of anterior uniramous parapodia varies among specimens and lower numbers (29/31) were found in specimens from the Peru Basin studied here than in those reported earlier by Rizzo and Amaral (2004b) and )

Böggemann (2005, 2009) for specimens from the South Atlantic. Rizzo and Amaral (2004b) reported slightly higher numbers of uniramous parapodia for specimens originating from the equatorial region of the Atlantic and lower numbers for Brazilian specimens. Morphological details of proboscidial papillae differ slightly between Rizzo and Amaral (2004b) and Böggemann (2005) and our limited observations, which is probably due to different points of inspection along the proboscis.

Bathyglycinde profunda was originally described from abyssal depths of 2862–5023 m (type locality 4800 m) in the Atlantic north of Bermuda (Hartman and Fauchald 1971) but our morphological and molecular data, as well as those of Böggemann (2009) and morphological data of Gunton et al. (2021), suggest it is a widespread deep-water species occuring not only in the Atlantic, but also in the Central, Southern, and Eastern Pacific Ocean. Gunton et al. (2021) reported it as one of only two species occurring all along the east coast of Australia from the Coral Sea Marine Park and off Fraser Island in the north to Flinders and Freycinet Marine Parks in the south, i.e. covering a distance of more than 1900 km. However, the species has to date not been recorded in faunal studies from the NW Pacific (Böggemann 2005, Alalykina 2020) and the Atlantic Sector of the Southern Ocean (Böggemann and Dietz 2016).

Molecular data confirm and extend the distribution of Bathyglycinde profunda for the Atlantic (Guinea and Cape Basin (s. Böggemann 2005) and the Central Atlantic) and the Central and Eastern Pacific (Clarion Clipperton Fracture Zone (s. Janssen et al. 2019) and Peru Basin) (Fig. 2).

Distribution:

Central and Eastern Pacific [1400–4900 m, Böggemann 2005; 4127–5023, Janssen et al. (2019); 4127–4257 m, this study], SE Pacific off Eastern Australia [2093–4280 m, Gunton et al. (2021)], Southern Ocean off Southern Australia [2063 m, MacIntosh et al. (2018)], NW Atlantic [1000–5023 m, Böggemann 2005], NE Atlantic [2400–5500 m, Böggemann 2005], W Atlantic [630–3783 m, Böggemann 2005], E Atlantic [5389–5450 m, Böggemann 2005], Central Atlantic (Vema Fracture Zone: 5004–5176 m, this study), SE Atlantic [Cape, Angola, and Guinea Basin: 3961–5494 m, Böggemann (2009)], SW Atlantic [350–5189 m, Böggemann 2005, 2016)].

Bathyglycinde sibogana (Augener & Pettibone in Pettibone, 1970)
(Fig. 10A, B)

Glycinde sibogana Augener and Pettibone in Pettibone, 1970: 244, figs 40, 41.

Bathyglycinde siboganaBöggemann, 2005: 196–198, figs 115, 116. – Böggemann 2009: 311, figs 43, 44, 47. – Fiege et al. 2010: appendix, tables 3 and 5 (name only). – Moreira and Parapar 2015: 61–62, fig. 25A–M. – Böggemann and Dietz 2016: 1508–1509, fig. 1.

Material examined:

Central Atlantic Ocean, Mid-Atlantic Ridge, East, SO 237 (VEMA-Transit), station 4-8, 5735 m (SMF 30491, SEM 1340, SMF 30492, ZMH-P 29694). – for details and additional specimens see the Supporting Information, Table S2.

Measurements for largest specimen studied (ZMH-P 29694): length 14.5 mm, width 0.5 mm (excl. parapodia), 67 chaetigers, 35 uniramous parapodia.

Description:

Prostomium in all three specimens with four irregular indistinct annulations and longitudinal lateral grooves along each side. Posterior chaetigers with two neuropodial prechaetal lobes from chaetiger 36 in specimen ZMH-P 29694 but absent in specimen SMF 30491 with only 37 chaetigers. Proboscidial papillae in area I small conical papillae (Fig. 10A, B; long arrows) and area II with five rows of long, slender partly sclerotized papillae similar in shape to those of B. profunda (Fig. 10A, B; short arrows). Some with protruding bulge in lower part (Fig. 10A, B; asterisk).

Remarks:

Bathyglycinde sibogana was originally described from bathyal depths of 1886 m in the Gulf of Boni (Indonesia) (Augener and Pettibone in Pettibone 1970) and has been reported to occur in the Central and SE Pacific (Böggemann 2005) and the Central and SE Atlantic (Böggemann 2009; this study) as well as the Atlantic Sector of the Southern Ocean (Böggemann and Dietz 2016). The three specimens studied agree well with the description published by Böggemann 2005. Molecular data confirm and extend the distribution of Bathyglycinde sibogana to the Atlantic (Cape Basin (s. Böggemann 2009) and the central Atlantic) (Fig. 2).

Distribution:

Indo-Pacific, Gulf of Boni (1886 m, Augener and Pettibone in Pettibone 1970), Central Pacific [4978 m, Böggemann 2005], SE Pacific [3739–3824 m, Böggemann 2005], Central Atlantic (Vema Fracture Zone: 5735 m, this study), South Atlantic [Cape and Angola Basin: 5058–5648 m, Böggemann (2009)], and Scotia Sea [3789–4720 m, Böggemann and Dietz (2016)].

Genus ProgoniadaHartman, 1965

Type species:

Progoniada regularisHartman, 1965. Gender: female.

Diagnosis:

(See: Böggemann 2022.) Segments bi- or triannulate. Prostomium annulated, consisting of eight rings; tip with biarticulate appendages. Proboscis with a few different types of papillae; with macrognaths and dorsal and ventral micrognaths; chevrons present. First segment with only a pair of small lateral cirri and without parapodia or chaetae. All parapodia uniramous. Notochaetae absent; neurochaetae compound falcigers and/or spinigers in all parapodia.

Progoniada regularisHartman, 1965
(Figs 10C–F)

Progoniada regularisHartman, 1965: 100, pl. 16a–f. – Rizzo and Amaral 2004a: 48–52, figs 1A–E, 2A–I, 3A–J. – Böggemann, 2005: 155–160, figs 86–88. – Böggemann 2009: 312, figs 43, 45–47. – Fiege et al. 2010: appendix, tables 3 and 5 (name only). – Moreira and Parapar 2015: 85–87, fig. 36A–J. – Böggemann and Dietz 2016: 1509–1510. – Böggemann 2016: 239–240, figs 10, 11. – Böggemann & Sobczyk in Gunton et al. 2021: 48 (name only).

Material examined:

Central Atlantic Ocean, Mid-Atlantic Ridge, West, SO 237 (VEMA-Transit), station 9-8, 5004 m (ZMH-P 26697, SEM 1341). SW Atlantic Ocean, Brazil Basin S, M 79-1 (DIVA 3), station 561-1, 4484 m (SMF 23790). SE Pacific Ocean, Peru Basin (DISCOL area), SO 242-1 (JPIO-DISCOL 1), station 37-1, 4161 m (SMF 30498, SMF 30499, SMF 30500, SMF 30501, SMF 30502); station 85-1, 4147 m (SMF 30497). – for details and additional specimens see the Supporting Information, Table S2.

Measurements largest specimens studied:

Central Atlantic (ZMH-P 26697): length 4.6 mm, width 0.2 mm (excl. parapodia), 35 chaetigers for largest specimen studied; SE Pacific, Peru Basin (SMF 30502): length 7.5 mm, width 0.35 mm (excl. parapodia), 30 chaetigers.

Description:

Prostomium with eight annulations (Fig. 10C). Tip with biarticulate appendages. Eyes absent. Proboscis with paired lateral rows of chevrons (five in specimen collected from the central Atlantic, ZMH-P 26697) (Fig. 10C, E) and dotted with several rows of small cordiform papillae (Fig. 10E, F). Parapodia all uniramous (Fig. 10D). Prechaetal lobe long, digitiform; postchaetal lobe short, rounded. First segment apodous and with lateral cirri only. Dorsal and ventral cirri digitiform of similar length, slightly shorter than prechaetal lobe (Fig. 10D). Chaetae compound spinigers and falcigers.

Remarks:

Specimens studied from the central Atlantic and Peru Basin agree morphologically with the descriptions published by Rizzo and Amaral (2004a) and Böggemann 2005. The number of chevrons found for specimen ZMH-P 26697 collected from the central Atlantic is lower than given with 14–22 by Rizzo and Amaral (2004a) but falls into the range from 4 to 30 as given by Böggemann 2005. Some specimens from central Atlantic (ZMH-P 26697) and from Peru Basin (SMF 30497) with a median, brownish pigment dot ventrally on segments (from ca. chaetiger 13 onwards in ZMH-P 26697), as described and figured by Rizzo and Amaral (2004a: fig. 3D). Böggemann (2022) considers P. oahuensisBarret & Bailey-Brock, 2005 a synonym of P. regularis with the difference in length of falcigerous blades as the main distinguishing character, i.e. shorter in P. oahuensis and longer in P. regularis, to be related to specimen size. This leaves P. regularis the only species in the genus currently valid as P. simplexHartman 1971 was already synonymized by Böggemann 2005. According to Böggemann (2009), P. regularis is a complex of cryptic or sibling species based on his molecular results, while Böggemann and Dietz (2016) consider P. regularis a cosmopolitan species occuring at depths from 10 to 5655 m. Specimens of P. regularis considered here collected from the Pacific (CCZ and Peru Basin), as well as from the central Atlantic (VEMA Fracture Zone), could not be separated based on our molecular data, thus extending the already wide distribution of P. regularis to abyssal depths in the central Atlantic and the central and SE Pacific (Fig. 3).

Distribution:

Central Pacific (CCZ: 4280 m, this study), SE Pacific (Peru Basin: 4147–4161 m, this study); Central Atlantic (VEMA Fracture Zone: 5004 m, this study), South Atlantic, Guinea (3945–5470 m), Angola (5424–5495 m), and Cape Basin (5058–5655 m) (Böggemann 2009), Brazil Basin and Argentine Basin (4481–5188 m) (Böggemann 2016), and Scotia Sea (4551–4720 m) (Böggemann and Dietz 2016).

Progoniada cf. regularisHartman, 1965
(Fig. 11A–G)

Progoniada regularisHartman, 1965. – Böggemann 2009: 312–315, figs 46D (neuropodial chaeta), table 7, specimen ZMH-P 25157. – Böggemann 2016: 239–240, fig. 10B–I (anterior to posterior parapodia), specimen SMF 23791.

? Progoniada sp. Böggemann and Sobczyk in Gunton et al. 2021: 47, fig.11, B, D, E.

Material examined:

Central Atlantic Ocean, Mid-Atlantic Ridge, East, SO 237 (VEMA Transit), station 4-8, 5737 m (ZMH-P 26694, bad condition). SW Atlantic Ocean, Brazil Basin N, M 79-1 (DIVA 3), station 605-1, 5189 m (SMF 23791, SEM 1339). SE Atlantic Ocean, Angola Basin, M 63-2 (DIVA 2), station 45-5, 5647 m (ZMH-P 25157a(SEM) + 25157b). – for details and additional specimens see the Supporting Information, Table S2.

Measurements largest specimen studied:

Central Atlantic (SMF 23791): length 25 mm, width 0.6 mm, 84 chaetigers.

Description:

(Based on specimen SMF 23791.) Prostomium with eight indistinct annulations (Fig. 11F). Tip with biarticulate appendages. Eyes absent. Proboscis with paired lateral rows of chevrons (15) (Fig. 11D) and dotted with several rows of small cordiform papillae (Fig. 11E). Macrognaths with three to four tips. Nine micrognaths arranged in dorsal arc as x-shaped or jointed H + w-shaped pieces and one to two larger jointed H + w-shaped pieces located ventrally (Fig. 11A–C).

Parapodia all uniramous (Fig. 11G). Prechaetal lobe long, digitiform; postchaetal lobe short, rounded. First segment apodous and with lateral cirri only. Dorsal and ventral cirri digitiform with pointed tip. Ventral cirri slightly longer than dorsal ones. Chaetae compound falcigers and spinigers with long blade (Fig. 11G).

Remarks:

The specimen collected from the eastern central Atlantic (ZMH-P 26694) is in very bad condition, unsuitable for morphological description. However, the specimens listed here can be separated from specimens of P. regularis based on our genetic results. The specimens collected from Angola Basin (ZMH-P 25157) and Brazil Basin (SMF 23791, SEM stub 1339) were originally included in descriptions of Progoniada regularisHartman, 1965 by Böggemann (2009, 2016)), respectively. Our molecular data strengthen the view of Böggemann (2009: 318, fig. 48) that P. regularis probably is a complex of cryptic or sibling species. Because our morphological studies of the specimens listed did not reveal significant differences to P. regularis, we refrain from describing a new species solely based on genetic differences.

Distribution:

Eastern Central Atlantic (VEMA Fracture Zone) (5735 m, this study), SW (Brazil Basin) and SE Atlantic (Angola Basin) (5189 and 5647 m, respectively) (Böggemann 2009, 2016) (Fig. 3).

Magelonidae Cunningham & Ramage, 1888,
Genus OctomagelonaAguirrezabalaga, Ceberio and Fiege, 2001

Type species:

Octomagelona bizkaiensisAguirrezabalaga et al., 2001. Gender: female.

Diagnosis:

[Aguirrezabalaga et al. (2001) and Mortimer (2019), emended.] Body divided into thoracic region comprising achaetous first segment and eight chaetigers, and abdominal region with unknown number of segments. Prostomium large, flattened, wider than long with small frontal horns. Parapodia biramous, with noto- and neuropodial lateral lamellae filiform if present, superior dorsal lobes [DML of Jones (1963)] absent. Branchiae absent. Thoracic chaetae long, limbate capillaries, abdominal chaetae tridentate hooded hooks. Posterior part and pygidium unknown. Lateral pouches not observed.

Remarks:

Members of the genus Octomagelona are characterized by the presence of eight instead of nine thoracic chaetigers in contrast to all species of MagelonaMüller, 1858, the only other genus in Magelonidae. Instead of being a kind of intermediate segment sometimes equipped with special chaetae, chaetiger 9 represents the first abdominal segment in Octomagelona. The generic diagnoses provided by Aguirrezabalaga et al. (2001) and Mortimer (2019) has been slightly emended due to abdominal hooded hooks being tridentate in both species belonging to the genus Octomagelona, i.e. O. bizkaiensis and O. borowskii sp. nov.. Mortimer et al. (2021), considered Octomagelona a junior synonym of Magelona following their phylogenetic analysis based on morphological characters. We agree with these authors that the ‘magelonid-like body regionation’ is an important synapomorphy for Magelonidae (Mortimer et al. 2021: 67 and abstract) but we regard the two different character states for the thoracic body region, i.e. the presence of eight versus nine thoracic chaetigers, as an easy to observe and taxonomically sufficient character to distinguish between the two genera. Therefore, we prefer to maintain Octomagelona as a valid genus for the time being and hope that studies of genetic markers will add significant information to characterize the two genera. Additionally, SEM and LM studies of thoracic chaetae in O. borowskii sp. nov., as well as the paratype of O. bizkaiensis, revealed that thoracic capillary chaetae are, in fact, unilimbate in O. bizkaiensis(Fig. 12L) as stated in the original description (Aguirrezabalaga et al. 2001) and not bilimbate as coded by Mortimer et al. (2021: table S1, coding matrix character 31). In specimens of Octomagelona borowskii sp. nov. collected from the Peru Basin, thoracic capillaries were partly found to be bilimbate (Figure 12J) as reported for the majority of Magelona species by Brasil (2003: table 5, coding matrix character 19) and Mortimer et al. (2021). But irregularly bilimbate chaetae are also present, i.e. with limbus irregularly indented (Figure 12K) as defined by Brasil (2003: fig. 28B, C). Limbation of chaetae needs to be judged at high magifications and it remains to be evaluated whether irregularly indented limbation represents a natural character state or an artefact due to the effect of fixative or mechanical treatment during sampling.

Figure 12.

SEM and LM studies of Octomagelona borowskii sp. nov. (A–F, I–K), Octomagelona cf. borowskii (G, H), and O. bizkaiensisAguirrezabalaga, Ceberio and Fiege, 2001 (L): A, anterior end, dorsolateral view. B, thoracic chaetiger, left side, posterior view. C, thoracic chaetigers 1–4, right side, lateral view. D, thoracic chaetigers 3–5, right side, lateral view. E, thoracic chaetigers 6–8, right side, lateral view. F, chaetigers 8–11, right side, lateral view. G, anterior fragment, dorsolateral view. H, anterior fragment, ventral view. I, abdominal hooks, lateral view. I’, tip of abdominal hook, frontal view. J, bilimbate thoracic capillary chaeta. K, irregularly bilimbate thoracic capillary chaeta. L, limbate thoracic capillary chaetae, chaetiger 1, right side.—A–F, SMF 30528 (SEM stub 1320), paratype, Peru Basin; G, H, SMF 30503 (SEM stub 1323), Angola Basin; I–K, SMF 30512, Peru Basin; L, O. bizkaiensis, SMF 10026 (SEM stub 604), paratype, Bay of Biscay. Scale Bars: A–F, 100 µm; G, H, 1 mm; I, 20 µm; I’–L, 10 µm.

Figure 13.

Distribution of lateral ciliated patches in neuropodia along chaetigers of the middle body region in Spiophanes pacificus sp. nov. and S. australis sp. nov.; trendlines of calculated mean and median values for counted patches also shown; character only reliably observable in formalin fixed specimens examined with SEM.

Octomagelona borowskii sp. nov. Fiege, Knebelsberger and Meißner
(Figs 12A–F, I–K)

Magelona sp. A Borowski, 1996: 54–55.

Type material:

Holotype. SE Pacific Ocean, Peru Basin (DISCOL area), SO 242-1 (JPIO-DISCOL 1) stn 117-7 EBS, 4154 m depth, 19 Aug 2015, one af (11 chaetigers, length 3.0 mm, width at posterior end of prostomium 1.1 mm) (SMF 30508).

Paratypes:

SE Pacific Ocean, Peru Basin (DISCOL area), SO 64 (DISCOL 2): one af, stn 1272 BC 1, 4159 m, 13 Feb 1989 (SMF 30518); one af, stn 1302 BC 1, 4136 m, 22 Mar 1989 (ZMH-P 30452); one af, stn 1379 BC 1, 4152 m, 9 Nov 1989 (SMF 30520); one af + tentacle, stn 1388 BC 1, 4167 m, 13 Sep 1989 (SMF 30521); one af, stn 1419 BC 1, 4156 m, 19 Sep 1989 (SMF 30522). SO 77 (DISCOL 3): one af, stn 1471 BC 1, 4152 m, 13 Feb 1992 (ZMH-P 30451); one af, stn 1476 BC 1, 4174 m, 14 Feb 1992 (SMF 30528, SEM stub 1320); one af, stn 1492 BC 1, 4163 m, 21 Feb 1992 (SMF 30530); all originally identified by C. Borowski as Magelona sp. A.

Measurements for largest paratype (SMF 30530): length 2.4 mm, width 0.8 mm (excl. parapodia) at posterior end of prostomium, 11 chaetigers.

Additional material examined:

Central Eastern Pacific, CCZ French licence area, BioNod12, station 101-1, 5055 m (SMF 30531); Central Eastern Pacific, CCZ German license area, BioNod12, station 6-1, 4259 m (SMF 30531, SMF 30532, SMF 30533, SMF 32223, SMF 32224). SE Pacific, Peru Basin (DISCOL area), SO 242-1 (JPIO-DISCOL 1), station 37-1, 4161 m (SMF 30510); station 126-9, 4257 m (SMF 30509). SO 61 (DISCOL 1), station 1276-1, 4153 m (SMF 30511). SO 64 (DISCOL 2), station 1377-1, 4124 m (SMF 30512); station 1403-1, 4166 m (SMF 30514); station 1414-1, 4149 m (SMF 30515); station 1416-1, 4142 m (SMF 30516); station 1419-1, 4156 m (SMF 30517); station 1423-1, 4145 m (SMF 30513). SO 77 (DISCOL 3), station 1449-1, 4148 m (SMF 30524); station 1465-1, 4161 m (SMF 30525); station 1471-1, 4152 m (SMF 30526); station 1476-1, 4174 m (SMF 30523); station 1485-1, 4167 m (SMF 30527); station 1485-1, 4167 m (SMF 32261). – for details and additional specimens see the Supporting Information, Table S2.

Description:

All specimens incomplete, i.e. anterior fragments with only few abdominal segments. Tentacles lost in all specimens, free tentacle fragments only found with two specimens from Peru Basin (SMF 30515, SMF 30521) with four rows of pinnules along tentacle, pinnule length twice diameter of tentacle resulting in frilly appearance. Distal part of tentacle with densely arranged pinnules and free tip.

Prostomium wider than long, length/width/ratio: 0.75 mm/1.25 mm/0.6 (paratype SMF 30530: 0.5 mm/0.8 mm/0.625, paratype SMF 30528: 0.25 mm/0.55 mm/0.45), truncate, with small frontal horns with free tips, anterior margin straight, smooth. Posterior ends of prostomium bent down forming postero-lateral flaps. Fixed specimens with prostomium pushed up and backward by partly everted bulbous burrowing organ. Eyes absent. Palps lost. First segment (peristomium) achaetous, best visible ventrally (Fig. 12A, B).

Thoracic region with eight chaetigers (Fig. 12A–F). Length 2.19 mm, width decreasing from 0.94 mm at chaetiger 1 to 0.57 mm at chaetiger 8 (paratype SMF 30530: length 1.2 mm, width 0.6 to 0.42 mm; paratype SMF 30528: length not measured, width 0.45 mm to 0.20 mm). All thoracic parapodia without superior notopodial process [DML of Jones (1963)]. Notopodial lateral lamellae postchaetal, present in chaetigers 1–4 (1–4 in paratypes SMF 30528 and 30530) filiform with slightly broadened base; decreasing in length from chaetigers 1–4 to less than half length, and absent posteriorly (Fig. 12A–C). In chaetigers 5–8 shallow ridge encircling chaetal fascicles like cuffs (Fig. 12D, E). Neuropodial lateral lamellae postchaetal, present in chaetigers 1–3(4) similar in size and form to notopodial lateral lamellae (slightly smaller in paratype SMF 30528). Gradually decreasing in length from chaetigers 1–4, lacking in posterior segments. Noto- and neuropodia of chaetigers 5–8 forming shallow ridges encircling chaetal fascicles like cuffs. Neuropodia similar to notopodia but slightly larger (Fig. 12D, E). Ventral neuropodial process present in chaetiger 1 only, very small (absent in paratype SMF 30528, Fig. 12A–F). Thoracic chaetae all long uni- or irregularly bilimbate capillaries, similar in length; most numerous in chaetigers 1–3 (paratype SMF 30528 with 10–14 in each fascicle), reduced in number posteriorly (?broken) (paratype SMF 30528 ca. 3–5 in chaetigers 7 and 8). Constriction between chaetiger 8 and 9, i.e. thorax and abdomen (Fig. 12A, F).

Abdomen starting with chaetiger 9. Width 0.63 mm (paratype SMF 30530: 0.6 mm; paratype SMF 30528: 0.3 mm). Abdominal parapodia without superior notopodial (DML) and ventral neuropodial process (VML) at upper- and lowermost ends of chaetal rows. Noto- and neuropodial lateral lamellae short filiform with slightly broadened base, present in chaetigers 9–11 positioned symmetrically, i.e. sub- and suprachaetal, respectively, at lateral ends of rows of abdominal hooks; increasing in length from chaetigers 9–11 (abdomen broken) (Fig. 12F). Abdominal chaetae tridentate hooded hooks with main fang surmounted by two small teeth; hooks arranged in single line on low parapodial ridge, 6–10 hooks per parapodial ramus, all same size, with three to five each in vis-à-vis arrangement, i. e. noto- and neuropodia with two groups each, teeth facing (Fig. 12I, I’). Lateral pouches not observed. Pygidium unknown.

Thorax dorsolaterally with fields of whitish spots, best developed in chaetigers 2–3, less in chaetigers 4–6, situated more laterally; only laterally in chaetiger 7 and 8, mid-dorsum always without spots. Fields of white spots almost touching mid-ventrally in chaetigers 1–3; only few spots in chaetiger 4 and following. Pigment spots in abdominal chaetigers not observed.

Remarks:

Until now, Octomagelona bizkaiensis has been known as the single species in the genus originating from about 1000 m depth in the Cap Breton Canyon (NE Atlantic, Gulf of Biscay) (Aguirrezabalaga et al. 2001). The specimens reported here represent the second species for the genus likewise collected from abyssal depths. Additional yet undescribed species have been reported by Mortimer (2019).

Specimens collected in the Pacific from the Peru Basin and the Clarion–Clipperton Fracture Zone do not differ significantly with regard to COI sequences (Fig. 4). Unfortunately, sequencing of specimens collected from the South Atlantic was unsuccessful and we also do not have any genetic information for O. bizkaiensis from the Bay of Biscay.

Octomagelona borowskii sp. nov. differs from O. bizkaiensis by slightly more distinct prostomial horns, a thorax decreasing significantly in width posteriorly, the presence of filiform postchaetal noto- and neuropodial lateral lamellae only in thoracic chaetigers 1–4 replaced by parapodial cuffs encircling chaetal fascicles in more posterior thoracic chaetigers, while parapodia in O. bizkaiensis appear completely reduced. Pigmentation is less dense and less pronounced in O. borowskii sp. nov. and spots are whitish instead of dark pigment granules as in O. bizkaiensis.

One specimen collected from the CCZ (SMF 30533) showed pigmentation spots in dense patches dorsolaterally on chaetigers 2–7, and on chaetigers 2–5 also laterally; ventrally pigmentation spots are scattered and more dense in anterior thorax.

Etymology:

The specific name refers to Dr. Christian Borowski, who studied the polychaetes collected during the initial phase of project DISCOL for his Ph.D. project. He was the very first to observe magelonids with only eight thoracic chaetigers and described them tentatively as ‘Magelona sp. A’ (Borowski 1996).

Distribution:

Central Pacific (CCZ, 4128–5055 m), SE Pacific (Peru Basin, 4124–4257 m).

Octomagelona cf. borowskii Fiege, Knebelsberger and Meißner
(Fig. 12G, H)

Octomagelona bizkaiensisAguirrezabalaga, Ceberio and Fiege, 2001– Fiege et al. 2010: 1338, tables 3 and 5, appendix (name only).

Material examined:

SE Atlantic, Angola Basin, M 48-1 (DIVA 1), station 330-3, 5469 m (SMF 30503, SEM 1323; SMF 30506; SMF 30587); station 338-8, 5439 m (SMF 30504); station 340-1, 5464 m (SMF 30507); station 341-8, 5464 m (SMF 30505). – for details see the Supporting Information, Table S2.

Description:

(Based on specimen SMF 30503.) All specimens incomplete, i.e. anterior fragments with only few abdominal segments. Tentacle fragment (free) found only with one posterior fragment (SMF 30587) showing same characters as described for Octomagelona borowskii sp. nov..

Prostomium wider than long (length/width n.d./1.15 mm), truncate, with small, rudimentary frontal horns with free tips, anterior margin smooth. Posterolateral ends of prostomium bent down forming posterolateral flaps. Prostomium pushed up and backwards by bulbous burrowing organ partly everted, longitudinally ridged ventrally. Eyes lacking. Palps broken. First segment achaetous (Fig. 12G, H).

Thoracic region with eight chaetigers (Fig. 12G, H). Thoracic width decreasing from chaetiger 1 (0.63 mm) to chaetiger 8 (0.50 mm). Constriction between thorax and abdomen. All thoracic parapodia without superior notopodial or ventral neuropodial process (Fig. 12G). Notopodial lateral lamellae postchaetal; filiform in chaetigers 1–4; length decreasing gradually posteriorly, absent in chaetigers 5–8. Neuropodial lateral lamellae postchaetal, similar in shape, as long as or longer than notopodial lateral lamellae, likewise gradually decreasing in length from chaetigers 1–4, the latter as rudimentary knob only, lacking posteriorly. Noto- and neuropodia of thoracic chaetigers 5–8 represented as low ridges encircling chaetal fascicles like cuffs (Fig. 12G). All thoracic chaetae long, uni- and bilimbate capillaries, similar in length, decreasing in number from chaetiger 1 (12–14 noto-, 15 neurochaetae) to chaetiger 8 (one noto-, two neurochaetae, some broken).

Abdomen starting with chaetiger 9. Width 0.6 mm. Abdominal parapodia without superior notopodial (DML) and ventral neuropodial process (VML) at upper- and lowermost ends of chaetal rows. Notopodial lateral lamellae from chaetiger 9, subchaetal, filiform, gradually increasing in length to chaetiger 12 (abdomen broken). Neuropodial lateral lamellae from chaetiger 11, suprachaetal, filiform, slightly longer in chaetiger 12. Abdominal chaetae tridentate hooded hooks with main fang surmounted by two small teeth, arranged in single line on parapodial ridge. About 10 per parapodial ramus, all about same size, with five each in vis-à-vis arrangement, i. e. noto- and neuropodia with two groups each, teeth facing. Lateral pouches not observed.

Pygidium unknown.

Remarks:

Specimens of Octomagelona cf. borowskii from the South Atlantic, are morphologically very similar to specimens of O. borowskii sp. nov. collected in the Pacific. Unfortunately, the limited number of specimens and their poor state of preservation make a detailed morphological distinction difficult. For example, a thoracic ventral neuropodial process is absent in thoracic chaetigers of O. cf. borowskii, whereas it is present in chaetiger 1 in some specimens of O. borowskii sp. nov. but absent in paratype SMF 30528 (Fig. 12A–F). Thoracic chaetae are uni- or bilimbate capillaries in O. cf. borowskii and uni- or irregularly bilimbate capillaries in O. borowskii sp. nov.. No conspicuous pigmentation pattern is discernable in specimens of O. cf. borowskii, whereas it is present in some specimens of O. borowskii sp. nov.. Molecular data are highly desirable for distinction among these two taxa. Unfortunately, sequencing of specimens of O. cf. borowskii was not successful.

Distribution:

SE Atlantic, Angola Basin, 5439–5473 m.

Spionidae Grube, 1850,
Genus SpiophanesGrube, 1860

Type species:

Spiophanes kroyeriGrube, 1860.

SpiophanesGrube, 1860: 88–89, pl. 5, fig. 1. Type species: Spiophanes kroyeriGrube, 1860, by monotypy. – Meißner and Hutchings 2003: 118–120, figs 1, 2. – Meißner 2005: 6. – Meißner and Blank 2009: 6–7.

MorantsChamberlin, 1919: 17. Type species: Morants duplexChamberlin, 1919, by monotypy. Junior synonym.

Diagnosis:

(See also Meißner and Blank 2009 for further details.) Prostomium rarely rounded, usually subtriangular or bell-shaped; anterolateral horns of different length present or absent, anterior margin not incised; occipital antenna present or absent. Branchiae absent. Nuchal organs as dorsal ciliated organs of different types present. Body divided into three regions: anterior region extending to chaetiger 4, parapodia with well-developed postchaetal lamellae in both rami; middle region from chaetiger 5 to last chaetiger with capillaries rather than neuropodial hooks in neuropodium, neuropodial postchaetal lamellae reduced; middle chaetigers with parapodial glandular organs, their presence most conspicuous in chaetigers 5–7(8) where they open via chaetal spreaders of different types (see: Meißner and Hutchings 2003) rather than simple vertical slits in the neuropodia; posterior region starts between chaetigers 13–16 with first appearance of neuropodial hooks. Chaetae include one to two conspicuous crook-like spines in neuropodia of chaetiger 1; stout sabre chaetae in inferiormost position usually present from chaetiger 4; bacillary chaetae maybe exposed in chaetigers 5–9; capillaries present in both parapodial rami along the body arranged in one to three distinct or indistinct rows; hooks present in posterior region, usually quadridentate with main fang surmounted by single tooth and two smaller apical teeth in parallel position; with or without secondary hoods. Transverse dorsal ciliated crests usually present. Clusters of cilia herein referred to as ‘lateral ciliated patches’ present or absent in parapodia of the middle body region. Ventrolateral pouches present or absent in posterior region between neuropodia. Pygidium with two or more anal cirri.

Remarks:

The generic diagnoses by Meißner and Blank (2009), and more recently by Blake et al. 2019, have not undergone considerable change over the last 20 years. The presence of up to three Spiophanes species in our source material from abyssal depths was suggested based on sequence analysis. Two species are newly described here; the identity of the third species and its potential assignment to a known species is discussed. The three putative species are all morphologically very similar and characters useful for morphological species delimitation are difficult to find. Prostomial characters, nuchal organs, both distribution and details of parapodial glandular organs along the body, as well as many chaetal characters, are in good agreement among the three species (see details in the species descriptions). In all three species sabre chaetae are present from chaetiger 4 throughout the body, but are often strikingly long in anterior chaetigers compared to those in more posterior and hook-bearing segments. Neuropodial quadridentate hooks with reduced hoods start on chaetiger 15. All three species possess very long capillaries in notopodia of the posterior part of the median region, and even longer capillaries in posterior notopodia. Morphological differences concern the ciliated patches, a character that has not been formerly described for Spiophanes spp., and has been discovered in formalin-fixed specimens during our SEM studies. Ciliated patches are clusters of cilia. They are present in parapodia of the middle body region where they are arranged in more or less regular rows along the distal edge of neuropodia. Observed variation in the presence of ciliated patches was intraspecific but also interspecific (Table 6; Fig. 13). Intraspecific variation concerned the number of patches found on a neuropodium in a certain chaetiger. Differences between putative species became apparent when comparing median values of the number of patches for each chaetiger for all studied specimens (Table 6; Fig. 13). In Spiophanes pacificus sp. nov., the highest number of patches was observed in chaetigers 8–10 with up to seven patches arranged in a regular row, or four to five patches considering the median values for counts in all specimens of this species (Table 6; Figs 13, 14F, I). In preceding parapodia the number of patches was reduced to only one and maximally three. In Spiophanes australis sp. nov. the highest number of patches was observed more anteriorly along the body in chaetigers 5–7, with the maximum of seven patches in chaetiger 6 in one specimen, and maximally four to six patches in chaetigers 4–7 for other specimens (Table 6; Figs 13, 16D, F, G). In S. australis sp. nov., the arrangement of patches in rows was less strict and rather seemed slightly irregular. In the third species, Spiophanes cf. longisetusMeißner, 2005, patterns of ciliated patches could not be securely detected, only single patches randomly on some parapodia of the middle body region were detected. Unfortunately, ciliated patches are not reliably observable in ethanol-fixed specimens and their recognition, in general, also depends on the quality of preservation, and moreover relies on SEM studies. Thus, the number of collected data regarding this character was restricted and, admittedly, not fully reliable in every detail. But since observed differences between S. pacificus sp. nov. and S. australis sp. nov. were distinct, this newly found character might turn out to be useful, and is here newly added to the diagnosis of Spiophanes. Also, there was slight variation in the number of neuropodial quadridentate hooks in posterior parapodia. In S. pacificus sp. nov., usually three hooks in a row are present (rarely four); in S. australis sp. nov., usually four (rarely three); and in S. cf. longisetus, the number of neuropodial hooks is usually five.

Table 6.

Counts of lateral ciliated patches observed in neuropodia along chaetigers (Ch) of the middle body region in different specimens of Spiophanes pacificus sp. nov. (P1–P7) and S. australis sp. nov. (A1–A4), together with calculated mean and median values

Chaetiger	S. pacificus sp. nov									S. australis sp. nov.
	P1	P2	P3	P4	P5	P6	P7	Mean	Median	A1	A2	A3	A4	Mean	Median
Ch 1	0	0	0	0	0	0	0	0.0	0	0	0	0	0	0.0	0
Ch 2	0	0	0	0	0	0	0	0.0	0	0	0	0	0	0.0	0
Ch 3	0	0	0	0	0	0	0	0.0	0	0	0	0	0	00.0	0
Ch 4	0	0	0	0	0	0	0	0.0	0	1	0	0	0	0.3	0
Ch 5	3	3	0	0	0	0	0	0.9	0	5	4	4	5	4.5	4.5
CH 6	3	3	0	0	0	0	2	1.1	0	7	4	4	5	5.0	4.5
Ch 7	3	3	1	0	0	0	3	1.4	1	3	4	3	6	4.0	3.5
Ch 8	5	5	5	0	1	1	4	3.0	4	0	4	3	0	1.8	1.5
Ch 9	4	5	5	0	6	5	4	4.1	5	0	2	0	0	0.5	0
Ch 10	0	0	5	4	7	0	5	3.0	4	0	0	0	0	0.0	0
Ch 11	0	0	4	4	0	0	3	1.6	0	0	0	0	0	0.0	0
Ch 12	0	0	0	0	0	0	0	0.0	0	0	0	0	0	0.0	0

Spiophanes pacificus sp. nov. Meißner, Schwentner and Fiege
(Figs 13–15; Table 6)

Spiophanes sp. NHM_1897 Neal et al., 2022: 42–43, fig. 40a—d, table 1.

Type material:

Holotype. NW Pacific Ocean, Kuril-Kamchatka trench, SO 223 (KuramBio), stn 4-5 BC, 5766 m, 7 Aug 2012, complete specimen (68 chaetigers, nearly 16 mm long, width 0.72 mm), now preserved in 70% EtOH (ZMH-P 30460).

Paratypes:

NW Pacific Ocean, Kuril-Kamchatka trench, SO 223 (KuramBio), stn 2-10 EBS Epi, 4859 m, 3 Aug 2012, one af (ZMH-P 30453); stn 5-5 BC, 5379 m, 10 Aug 2012, four af (SMF 32272); stn 5-10 EBS Epi, 5375 m, 11 Aug 2012, 1af (SMF 30593). SE Pacific Ocean, Peru Basin (DISCOL Experimental Area, DEA), SO 242 (JPIO-DISCOL 1), stn 85-4 EBS, 4147 m, 14 Aug 2015, one af (SMF 30610).

Measurements for largest paratype from the Kuril-Kamchatka trench (SMF 32272): af of about 21 chaetigers, length ~5.5 mm, width ~0.6 mm. Largest paratype from the Peru Basin (SMF 30610): af of about 21 chaetigers, length 4.1 mm, width 0.9 mm (chaetae omitted).

Non–type material:

NW Pacific Ocean, Kuril-Kamchatka trench, SO 223 (KuramBio), stn 2-2, 5247.1 m (ZMH-P 28455); stn 3-4, 4987.7 m (ZMH-P28445, ZMH-P28446, ZMH-P28447); stn 4-4, 5773.6 m (ZMH-P28451); stn 4-5, 5766 m (SMF 32273). SE Pacific Ocean, Peru Basin (DISCOL area), SO 242 (JPIO-DISCOL 1), Reference area Hill NW, stn 117-7 EBS, 4154 m, 19 Aug 2015, one af (SMF 30612, SEM 1336); reference area S, stn 45-2 EBS, 4184 m, 4 Aug 2015, two af (SMF 30616, SEM 1324). – for details and additional specimens see the Supporting Information, Table S2.

Description:

Holotype complete specimen with 68 chaetigers, 0.72 mm wide and nearly 16 mm long (Fig. 14A, B). Examined specimens between 0.3 and 0.9 mm in width, except for the holotype all anterior fragments. Body slender, subcylindrical.

Prostomium bell-shaped with straight anterior margin extending into short anterolateral projections (Fig. 14B, C), short stout cirriform occipital antenna with rounded tip present (Fig. 14C, E). Dorsal ciliated organs (suggested to represent nuchal organs) as dorsal ciliated grooves posterior to the prostomium (ciliation detectable in SEM, Fig. 14C, E), if viewed with LM appearing as thick straight double lines of ochre or greenish colour reaching the end of the 2nd chaetiger (Fig. 14A, B), sometimes outer margins of ciliated grooves demarcated resulting in a U-shaped appearance of the nuchal organ. Eyes absent. Peristomium moderately developed. First parapodium oriented dorsally, second dorsolaterally, thereafter orientation of parapodia lateral (Fig. 14E). Notopodial postchaetal lamellae of chaetigers 1–4 narrow subulate, longest in first three chaetigers, neuropodial postchaetal lamellae in first chaetiger also subulate with broad base and slender tip, only slightly shorter than in notopodium, in following chaetigers neuropodial lamellae much shorter than notopodial, and gradually changing into a subtriangular shape (Fig. 14B, C, E, G, 15A, B); chaetigers 5–8 with short, rounded notopodial and reduced neuropodial postchaetal lamella (Figs 14H, 15C); from chaetiger 9 notopodial lamella low with short acute tip, gradually base becoming more voluminous and the tip more slender (Fig. 14F), neuropodial lamella reduced; from about chaetiger 14 throughout the end of the body notopodial lamella with broad base and cirriform tip, neuropodial lamella reduced (Figs 14A, 15D). Chaetal spreader of the ‘0 + 1 type’ with semicircular glandular opening developed in chaetigers 5–7, in chaetiger 8 chaetal spreader present but with small hole-like opening allowing the passage of very few or even only single bacillary chaeta (Fig. 14F, H), glandular organ of chaetigers 9–14 opens as a lateral vertical slit, without chaetal spreader (Fig. 14I). Ciliated patches might be observed laterally on neuropodia of chaetigers 5–11, with the highest number of patches in chaetigers 8–10 and up to seven patches arranged in a regular row (Fig. 14F, I; see discussion above regarding this character for Spiophanes). Ventrolateral intersegmental genital pouches absent. Dorsal membranous transverse crests very poorly developed from about chaetiger 18.

Chaetiger 1 bearing one or two stout, crook-like chaetae in neuropodium. Other chaetae in chaetigers 1–4 neuropodial capillaries arranged in two rows, chaetae in anterior row shorter and appearing granulated viewed with light microscopy, chaetae in second row smooth; notochaetae smooth capillaries, arranged in a tuft (Figs 14C, G, 15B). Chaetigers 5–14 with granulated neurochaetae with narrow sheaths, arranged in one row (Fig. 14F, H, I, 15C); notochaetae both granulated and smooth capillaries arranged in two to three irregular rows (Fig. 14F, H), longest chaetae in superior position. Posterior region starting at chaetiger 15 with first presence of neuropodial quadridentate hooks with main fang surmounted by single tooth and two smaller teeth in parallel position in uppermost position, hooks with half-hood from the tip of the main fang to the shaft, usually three rarely four hooks arranged in one row; thin accompanying capillary in inferior position next to sabre chaeta present (Fig. 14J); notopodia with slightly granulated capillaries arranged in a tuft, among those few strikingly long capillaries, longest chaetae observed in posteriormost notopodia of the few complete specimens and in fragmented specimens (af) with many chaetigers.

Bacillary chaetae as thin hirsute bristles can be exposed on chaetigers 5–8 though never protruding much in chaetiger 8 compared to chaetigers 5, 6, and 7 (Figs 14F, H, 15C). Ventral sabre chaetae from chaetiger 4, very long in anterior and middle chaetigers (Figs 14F–H, 15B, C); appearing granulated near the tip under light microscope.

Pygidium with one pair of anal cirri in terminolateral position, both cirri about equal in length, thin cirriform (Fig. 14A).

Pigmentation:

Neuropodia of chaetigers 10–13(14) with faint yellowish to light brown pigment (Fig. 14D); pigmentation only observed in few specimens originally preserved in formalin.

Methyl green staining pattern:

Chaetigers of the anterior middle body region, and particularly chaetiger 8, most intensely and moreover most persistently stained compared to other parts of the body.

Biology:

None of the studied specimens was observed bearing gametes.

Remarks:

The species is morphologically very similar to other congeners from the deep-sea also discussed in this paper: S. australis sp. nov. (found in the SW Atlantic and adjacent Antarctic waters) and S. cf. longisetus from the Atlantic Ocean. These species are best distinguished by the distribution of lateral neuropodial ciliated patches along the middle body region and the number of neuropodial hooks (Table 6; Figs 13, 14F, I). See introductory paragraph on Spiophanes above for details of morphological distinction. The species can also be distinguished based on information from molecular markers (COI).

Searching GenBank for nucleotide sequences close to the ones gained in our study we came across a COI sequence deposited for Spiophanes adriaticus. The close genetic similarity between Spiophanes pacificus sp. nov. and S. adriaticus is surprising and the genetic distance is low enough to imply that these could be conspecific. However, ecologically it seems highly unlikely that these are the same species, one occurs in the deep sea of the Pacific, the other in shallow Mediterranean waters, Adriatic Sea (<30 m depth). This would be a truly remarkable vertical and horizontal distribution for a single species and even for sister taxa such different depth preferences would be surprising. The S. adriaticus sequences came from D’Alessandro et al. (2019) who reported to have sequenced three specimens, which are also depicted in their phylogenetic tree. However, only one sequence has been deposited in GenBank. Unfortunately, no information (e.g. collection details) was provided for the numerous other Spiophanes species that they studied and also none of these sequences appears to have been deposited in GenBank. This makes it impossible to verify their species identifications and raises the question whether the correct sequence has been deposited for S. adriaticus. Unfortunately, the authors have only very recently responded to our repeated attempts to contact them but eventually new information could not be provided and thus the problem is not resolved. As already argued by Jourde et al. (2020), who studied the morphology of Spiophanes spp. from French coasts of the NE Atlantic and the Mediterranean Sea, we here too question the validity of S. adriaticus.

Etymology:

The specific name pacificus refers to the known distribution of the species in soft bottom habitats of the deep Pacific Ocean.

Distribution:

The species has a wide distribution in the Pacific Ocean. Records came from the central (Clarion Clipperton Fracture Zone CCZ) and SE Pacific (Peru Basin), and from the NW Pacific (Kuril–Kamchatka trench) (Fig. 5; Supporting Information, Table S2). Water depths were between 4900 and 5800 m in the NW Pacific, 4200–5100 m in the CCZ, and 4100–4200 m in the Peru Basin.

Spiophanes australis sp. nov. Meißner, Götting and Fiege
(Figs 13, 16, 17; Table 6)

Type material:

Holotype. South Atlantic Ocean, Argentine Basin, M 79-1 (DIVA 3), stn 533-1 EBS, 4602 m, 15 Jul 2009, one af fragmented into two pieces (21 chaetigers, 3.1 mm long, width 0.6 mm), original fixation formalin, now in 70% EtOH (SMF 30647).

Paratypes:

SW Atlantic Ocean, Argentine Basin, M 79-1 (DIVA 3), stn 533-1 EBS, 4602 m, 15 Jul 2009, four af original fixation formalin, now in 70% EtOH (SMF 30634 incl. one af SEM 1334); stn 534-1 EBS, 4608 m, 16 Jul 2009, six af, original fixation formalin, now in 70% EtOH (SMF 30632 incl. one af SEM 1335); stn 534-1 EBS, 4608 m depth, 16 Jul 2009, one af, fixed in 96% EtOH, tissue sample (SMF 32277).

Measurements for largest paratype (SMF 30632) af of about 19 chaetigers, length ~3 mm, width ~0.6 mm (chaetae omitted); other paratypes all af max. 21 chaetigers, between 0.9–4.7 mm long, 0.3–0.6 mm wide.

Description:

Holotype anterior fragment with 21 chaetigers, 0.6 mm wide and 3.1 mm long. Specimens between 0.3–0.9 mm in width, and up to 4.7 mm long, all anterior fragments of max. 21 chaetigers. Body slender, subcylindrical.

Prostomium bell-shaped with straight anterior margin extending into short rounded anterolateral projections (Fig. 16A–C), short papilliform occipital antenna present (Fig. 16A). Dorsal ciliated organs (suggested to represent nuchal organs) as dorsal ciliated grooves posterior to the prostomium (ciliation detectable with SEM), if viewed with LM appearing as thick straight double lines of ochre or yellow colour reaching the end of 2nd chaetiger (Fig. 16B, C), sometimes outer margins of ciliated grooves demarcated and then appearing U-shaped. Eyes absent. Peristomium moderately developed. First parapodium oriented dorsolaterally, second laterally to almost completely laterally, thereafter orientation of parapodia lateral (Fig. 16A–D). Postchaetal lamellae of chaetigers 1–4 narrow subulate in notopodia (Figs 16A, B, 17A), longest in first two to three chaetigers, thereafter notopodial lamellae decreasing in length; neuropodial postchaetal lamellae in first chaetiger also subulate with broad base and slender tip, slightly shorter than in notopodium and less narrow (Figs 16A, 17A), from 3rd chaetiger neuropodial lamellae distinctly shorter and more stout than notopodial lamellae, gradually changing into a subtriangular shape (Figs 16D, 17B); chaetigers 5–8 with short broad subulate notopodial lamella, sometimes distally with minute sharp tip, reduced neuropodial postchaetal lamella (Figs 16D, G, 17C); from chaetiger 9 notopodial lamella low with short distal tapering, gradually base becoming more voluminous and the tip more slender and longer; from about chaetiger 14 throughout the end of the body with broad base and short cirriform tip, neuropodial lamella reduced (Fig. 17D). Chaetal spreader of the ‘0 + 1 type’ with semicircular glandular opening developed in chaetigers 5–7 (Fig. 16D, F, G), in chaetiger 8 chaetal spreader present but with hole-like opening allowing the passage of only single bacillary chaeta (Fig. 17C); glandular organ of chaetigers 9–14 opens as a lateral vertical slit, without chaetal spreader. Ciliated patches might be observed laterally on neuropodia of chaetigers 4–9, with the highest number of patches in chaetigers 5–8 and up to seven patches arranged in less strict rows in a single neuropodium (Table 6; Figs 13, 16D, F, G; see discussion regarding this character above for Spiophanes). Ventrolateral intersegmental genital pouches absent. Dorsal crests not observed but none of the specimens very well preserved.

Chaetiger 1 bearing one or two stout, crook-like chaetae in neuropodium. Other chaetae in chaetigers 1–4 capillaries, in neuropodia arranged in two rows with chaetae in anterior row shorter and appearing slightly granulated when viewed with light microscopy, in second row longer and smooth, in notopodia arranged in a tuft or forming irregular rows; from chaetiger 4 inferior sabre chaeta present, stout and granulated near the tip, sabre chaeta of considerable length, becoming shorter in middle and posterior chaetigers (Figs 16A, D, 17A, B). Chaetigers 5–14 with stout granulated, narrow bilimbate neurochaetae, arranged in one (to two) row; notochaetae both granulated and smooth capillaries arranged in two to three irregular rows, longest chaetae in superior position (Figs 16D, F, G, 17C). Posterior region starting at chaetiger 15 with first presence of neuropodial quadridentate hooks with main fang surmounted by single tooth and two smaller teeth in parallel position in uppermost position, hooks with half-hood from the tip of the main fang to the shaft, usually four (rarely three) hooks arranged in one row (Figs 16E, 17D); single accompanying capillary next to sabre chaetae present; notopodia with slightly granulated capillaries arranged in a tuft, often broken but also distinctly longer if intact (Fig. 17D). Bacillary chaetae as thin, hirsute bristles can be exposed on chaetigers 5–8 though more numerous in chaetigers 5–7 compared to chaetiger 8 and here never protruding. Ventral sabre chaetae from chaetiger 4, of conspicuous length in anterior and middle chaetigers, shorter in hook-bearing chaetigers (Fig. 16D–G); appearing granulated near the tip under light microscope. Pygidium unknown since all specimens incomplete.

Pigmentation:

Only yellowish pigment associated with nuchal organ discernable (Fig. 16B, C).

Methyl green staining pattern:

Chaetigers of the anterior middle body region, and particularly chaetiger 8, most intensely stained.

Biology:

None of the studied specimens was observed bearing gametes.

Remarks:

The species is morphologically very similar to other congeners from the deep-sea also discussed in this paper: S. pacificus sp. nov. (found at different localities in the Pacific Ocean) and S. cf. longisetus (found at more northerly locations of the Atlantic Ocean and perhaps synonymous with S. abyssalisMaciolek, 2000 from the NE Atlantic, see Remarks below for S. cf. longisetusMeißner, 2005). The species are best distinguished by the distribution of lateral neuropodial ciliated patches along the middle body region (Table 6; Fig. 13) and the number of neuropodial hooks (see above introductory paragraph under Spiophanes for details of morphological distinction). The species can also be distinguished based on information from molecular markers (COI).

Etymology:

The specific name australis Latin (=southern); refers to the known distribution of the species in soft bottom habitats of the South Atlantic Ocean and probably also in adjacent Antarctic waters (Weddell Sea).

Distribution:

The species has been collected from soft bottoms of the Argentine Basin in the SW Atlantic Ocean (Fig. 5; Supporting Information, Table S2). Water depth was about 4600 m. The species probably also occurs in more southerly waters but this only refers to preliminary identification of specimens collected in the Weddell Sea at 2046 m water depth. Since the acquisition of COI molecular markers failed the specimen is listed as S. cf. australis sp. nov., and gained 18S sequences are deposited in GenBank under this name. Tissue all used in molecular analysis.

Spiophanes cf. longisetusMeißner, 2005
(Fig. 18)

Spiophanes longisetusMeißner, 2005: 45–48, figs 26–28, table 1.

Figure 14.

Spiophanes pacificus sp. nov., LM and SEM studies: A, holotype, complete specimen original preservation in 96% ethanol; note nuchal organ with associated pigment, and anal cirri. B, close-up of anterior region, latero-oblique view. C, anterior region, dorsal view; note short cirriform occipital antenna with rounded tip, also note nuchal organ reaching end of chaetiger 2. Inset C, close-up of antenna of same specimen. D, middle body region with pigment in parapodia of chaetigers 11–14, dorsal view. E, anterior end, lateral view; note stout cirriform occipital antenna and extension of dorsal ciliated nuchal organs to the end of chaetiger 2. F, chaetigers 8–10 in lateral view; note vertical row of ciliated patches at chaetigers 9 and 10, and single patch at chaetiger 8 next to hole-like opening of glandular organ, arrows indicate upper patches. G, left parapodium of chaetiger 4; note long sabre chaeta in inferiormost position; anterolateral view. H, parapodia of chaetigers 5–6, lateral view; note semi-circular openings of glandular organs and exposed bacillary chaetae. I, right neuropodium of chaetiger 10, lateral view; note ciliated patches and opening of parapodial glandular organ as vertical slit with fibre-wool protruding from the gland, arrow heads indicate upper- and inferior-most patches. J, neuropodial quadridentate hooks with half-hood in chaetiger 18; note three hooks are arranged in a row together with thin accompanying capillary in inferior position, sabre chaeta not shown.—A, B, ZMH-P 30460, holotype, NW Pacific; C, E, F, G, SMF 30612 (SEM stub 1336); D, SMF 30610, paratype, Peru Basin; H, SMF 30616 (SEM stub 1324), Peru Basin; I, SMF 30612 (SEM stub 1336), Peru Basin; J, ZMH-P 28455, NW Pacific. Scale Bars: A, B, 1 mm; C, E, 100 µm; D, 200µm; F, G, 50 µm; H, I, 30 µm; J, 10µm.

Figure 15.

Spiophanes pacificus sp. nov.: A, parapodium from 2nd chaetiger; notochaetae partially lost. B, same from 4th chaetiger. C, same from 7th chaetiger; neurochaetae partially lost. D, same from 15th chaetiger with three neuropodial hooks and broken sabre chaeta. All anterior view.—A–C, SMF 30593, paratype, NW Pacific; D, SMF 30610, paratype, Peru Basin. Scale Bars: 100 µm.

Figure 16.

Spiophanes australis sp. nov., LM and SEM studies: A, prostomium and anteriormost chaetigers, antero-oblique view; note short papilliform occipital antenna; crook-like chaeta lost on right side in first neuropodium. B, anterior region, dorsal view; note nuchal organ with associated yellow pigment. C, same but different specimen; note well-preserved nuchal organs. D, parapodia of chaetigers 4–7, lateral view; note sabre chaeta at 4th chaetiger, and vertical row of ciliated patches at chaetigers 5–6/7, arrowhead pointing at 2nd patch in line (counted from above) at 5th chaetiger. E, neuropodial quadridentate hooks with half-hood in chaetiger 15; note four hooks arranged in a row. F, chaetiger 7, lateral view; note ciliated patches and opening of glandular organ with bacillary chaetae protruding from inside the gland. G, chaetigers 7–8, lateral view; note reduced number of ciliated patches in chaetiger 8.—A, B, E, paratypes SMF 30634 (SEM stub 1334); C, holotype SMF 30647; D, F, G, SMF 30632; all Argentine Basin. Scale Bars: A–C, 100 µm; D, G, 20 µm; E, F, 10 µm.

Figure 17.

Spiophanes australis sp. nov.: A, parapodium from 2nd chaetiger. B, same from 4th chaetiger. C, same from 8th chaetiger; glandular organ detectable inside the neuropodium, two bacillary chaetae emerging laterally from hole-like opening. D, parapodium from 15th chaetiger with four neuropodial hooks (mostly seen subepidermally). All anterior view. A–D, SMF 30634, paratype. Scale Bars: 100 µm.

Figure 18.

Spiophanes cf. longisetusMeißner, 2005, LM and SEM studies: A, specimen stained with methyl green, dorsal view; note nuchal organ with associated yellow pigment, also note intensively stained glandular organs in middle chaetigers and distally stained neuropodium of chaetiger 8. B, anterior end until chaetiger 4, dorsal view; note dorsal nuchal organ extending to end of 2nd chaetiger. C, same in lateral view; note short stout cirriform occipital antenna with rounded tip. D, chaetigers 6–8, lateral view; note bacillary chaetae emerging from hole-like opening distally on chaetiger 8, also see enlarged detail below. E, left parapodia of chaetigers 7–9, stained with methyl green, lateral view; note intensively stained spot in 8th chaetiger. F, micrograph of quadridentate neuropodial hook from posterior chaetiger; note half-hood from the tip of the main fang to the shaft. G, right parapodium of chaetiger 4, lateral view; note long sabre chaeta in inferior position. H, row of five neuropodial hooks of chaetiger 15 with sabre chaeta in inferior position (to the right in image). I, neuropodial quadridentate hook, lateral view; note half-hood from tip of main fang to shaft.—A, ZMH-P 28154, MAR East; B, G, SMF 30641, MAR East; C, D, H, I, SMF 30640 (SEM stub 1327), MAR central; E, ZMH-P 28157, MAR East; F, ZMH-P 28155, MAR East. Scale Bars: A, 200 µm; B–D, F, 100 µm; E, 50µm; G, 20 µm; H, 10 µm; I, 5 µm.

Spiophanes kroyeriGrube, 1860. – Hartman and Fauchald, 1971: 105–106.

Material examined:

Central Atlantic Ocean, Mid-Atlantic Ridge, East, SO 237 (VEMA-Transit), stn 2-6, EBS, 12 Jan 2015, 5520 m, one af, tissue sample (ZMH-P 28149), one af, tissue sample (ZMH P-28150); stn 2-7, EBS, 20 Dec 2014, 5507 m, one af (ZMH P-28154), one af (ZMH-P28155), one af (SMF 30642, SEM 1330), one af (SMF 30643, SEM 1331), one mf (SMF 30644); stn 4-8, EBS, 26 Dec 2014, 5725 m, one af, tissue (ZMH P-28156); stn 4-9, EBS, 27 Dec 2014, 5733 m, two af (ZMH P-28157); stn 6-7, EBS, 2 Jan 2015, 5079 m, four af (SMF 30641, SEM 1342); stn 6-8, EBS, 2 Jan 2015, 5079 m, one af, tissue (ZMH-P 28160). Mid-Atlantic Ridge, Central, SO 237 (VEMA-Transit), stn 8-4, EBS, 6 Jan 2015, 5176 m, one af, tissue (SMF 30640, SEM 1327). Mid-Atlantic Ridge, West, SO 237 (VEMA-Transit), stn 9-8, EBS, 12 Jan 2015, 5004 m, one af, tissue (ZMH-P 28162, SEM), one af, tissue (SMF 30639), one af, tissue (ZMH P-28164); stn 11-4, EBS, 14 Jan 2015, 5108 m, one af (SMF 30638). SW Atlantic Ocean, Brazil Basin N, M 79-1 (DIVA 3), stn 604-1, EBS, 31 Aug 2012, 5180 m, one af, tissue (SMF 30645), one af, tissue (SMF 30646). – for details and additional specimens see the Supporting Information, Table S2.

Description:

(Focusing on most important characters for specimens examined in the course of the present study.) Specimens all anterior fragments with up to 23 chaetigers, between 0.2 and 0.9 mm in width, and up to 6.4 mm long. Prostomium bell-shaped with straight anterior margin extending into short anterolateral projections (Fig. 18A, B), short, stout cirriform occipital antenna with rounded tip (Fig. 18B, C). Dorsal ciliated organs as dorsal ciliated grooves posterior to the prostomium (ciliation detectable with SEM), if viewed with LM appearing as thick, straight, double lines of ochre or yellowish colour reaching the end of the 2nd chaetiger (Fig. 18A, B), sometimes outer margins of ciliated grooves demarcated and nuchal organ then appearing as pair of short U-shaped double lines. Ventral sabre chaetae from chaetiger 4, often of imposing length (Fig. 18C, D, G). Chaetal spreader of the ‘0 + 1 type’ with semicircular glandular opening developed in chaetigers 5–7, in chaetiger 8 chaetal spreader present but with only small hole-like opening (Fig. 18D); glandular organ of chaetigers 9–14 opens as a lateral vertical slit, without chaetal spreader. Bacillary chaetae as thin hirsute bristles can be exposed on chaetigers 5–8 though small opening of glandular organ on chaetiger 8 allowing only the protrusion of distal tips of very few bacillary chaetae (Fig. 18D). Few single ciliated patches randomly present in parapodia of the middle body region or completely absent. Ventrolateral intersegmental genital pouches absent. Posterior region starting at chaetiger 15 with first presence of neuropodial quadridentate hooks with main fang surmounted by single tooth and two uppermost smaller teeth in parallel position, hooks with half-hood from the tip of the main fang to the shaft, usually four to five hooks, in some juveniles only three hooks arranged in one row (Fig. 18F, H, I); single, thin accompanying capillaries often present, observed in position next to sabre chaeta; notopodia with slightly granulated capillaries arranged in a tuft, among those few strikingly long capillaries. Pygidium not observed in examined specimens (all anterior fragments).

Pigmentation:

All examined specimens pale without pigmentation, only nuchal organ with yellowish or ochre pigment as in related species (Fig. 18A). According to original description for S. longisetusMeißner, 2005 yellow to orange pigment in neuropodia of chaetigers 11–14 present. Remnants of faint brownish pigment in neuropodia 11–14 in few specimens discernible (e.g. SMF 30645), but usually absent.

Methyl green staining pattern:

Chaetigers of the anterior middle body region, and especially their subepidermal glandular organs, most intensely stained and also most persistently stained compared to other parts of the body. In chaetiger 8, lateral part of the neuropodium most intensively stained, in lateral view observed as dark circular area (Fig. 18A, E).

Biology:

Information about reproduction and development not available, because none of the studied specimens was bearing gametes.

Remarks:

The species is morphological very similar to other congeners from the deep-sea discussed in this paper: S. australis sp. nov. (found in the SW Atlantic and adjacent Antarctic waters) and S. pacificus sp. nov (collected from the Pacific Ocean). All three species are morphologically best distinguished by the distribution of lateral neuropodial ciliated patches along the middle body region, which are arranged in distinct patterns in S. australis sp. nov. and S. pacificus sp. nov. but are only randomly found as single patches in S. cf. longisetus, or also completely absent in the latter. The number of neuropodial hooks in posterior chaetigers is with (3)4–5 highest in S. cf. longisetus whereas in S. australis sp. nov. (3–)4 hooks are present and in S. pacificus sp. nov. usually not more than three hooks are found. The species can also be distinguished based on information from molecular markers (COI). However, the identity of specimens here referred to as S. cf. longisetus is not entirely resolved. The problem is mainly caused by the lack of information on molecular markers for S. longisetusMeißner, 2005 from type material or other specimens from the type locality. Moreover, our search in public sources (GenBank, BOLD) for sequences in good agreement with our putative species here referred to as S. cf. longisetus was unsuccessful. Based on what we know today molecular information is indispensable for solving the problem of the species identity since another Spiophanes species from the abyssal NE Atlantic is known: S. abyssalisMaciolek, 2000. Records for this species come from two different localities in the Bay of Biscay and the type locality close to the Canary Islands. Spiophanes abyssalis is morphologically extremely close to S. longisetus. Morphological differences concern the hood of the neuropodial hooks which are described as rudimentary and hard to observe in S. longisetus whereas they are clearly visible in S. abyssalis (Meißner 2005). In the latter species, four to five hooks were observed, in S. longisetus three to five. Spiophanes longisetus can also be identified by the presence of long granulated notopodial chaetae from chaetiger 14 whereas for S. abyssalis posterior notopodial chaetae are described as simple narrowly sheathed capillaries [updated species description in Meißner (2005)]. Also, in the original description Maciolek (2000) describes the notochaetae as comparatively short and moreover states the presence of two eyes in the holotype. Since the here examined specimens from the central Atlantic present very long notopodial capillaries, up to five neuropodial hooks and no eyes, we here refer to them as S. cf. longisetus. However, a more reliable conclusion will be possible if additional molecular information becomes available and a subsequent review of the morphology of all involved species, especially in regard to the newly discovered ciliated patches and the number of neuropodial hooks in relation to specimen size, can be undertaken.

Distribution:

Spiophanes longisetusMeißner, 2005 has been described from localities in the NW Atlantic Ocean (Meißner 2005). Type material and additional non-type material studied while describing the species all came from slope and abyssal depths off New England and Bermuda in the NW Atlantic Ocean (Meißner 2005). Specimens studied in the course of the present study and tentatively suggested to belong to S. longisetus came from the VEMA fracture zone of the Mid-Atlantic Ridge and abyssal plains east and west of it, as well as from the Brazil Basin in the northern South Atlantic Ocean (Fig. 5). Water depths were 3753–4663 m in the Western North Atlantic, 5004–5733 m at locations near the Mid-Atlantic Ridge, and 5200 m in the Brazil Basin. As soon as new information on genetic markers for specimens from the type locality becomes available and the uncertainty concerning the identity of the different specimens can be dispelled, the distribution of S. longisetus has to be revalidated.

Discussion

Species delimitation by integrative taxonomy

The main focus of the present study was to analyse the distributional ranges of selected deep-sea polychaetes and to assess the presence of morphologically cryptic species. Delineating species and unambiguously assigning individuals to known species is the prerequisite for any assessment of biodiversity, species distributions, or population genetic inferences. Although approaches like DNA barcoding (Hebert et al. 2003, Hebert and Gregory 2005), which rely on sequencing mitochondrial COI only, are invaluable to detect potential cryptic species, the lack of a universal barcoding gap requires additional data to solve problematic cases. Such data can be, for example, additional nuclear markers or morphological and ecological data. The strength of combining morphological and molecular genetic techniques has been well proven for many marine invertebrates, including polychaetes (e.g. Meißner and Blank 2009, Meißner et al. 2011, 2017, 2019, Bonifácio and Menot 2019, Drennan et al. 2019, Wiklund et al. 2019, Elgetany et al. 2020, Parapar et al. 2020, Surugiu et al. 2022). Following this approach in the present study, we suggest that, based on all data available to us, putative species delineated herein by COI are indeed separate entities, with our molecular delimitation of species supported by morphological data. As a result, the following species are identified from our source material: Sigambra magnuncusPaterson and Glover, 2000 (Pilargidae), Bathyglycinde profunda (Hartman and Fauchald, 1971), Bathyglycinde sibogana (Augener and Pettibone in Pettibone, 1970), Progoniada regularis Haartman, 1965, Progoniada cf. regularis (Goniadidae), Octomagelona borowskii sp. nov., Octomagelona cf. borowskii (Magelonidae), and also Spiophanes australis sp. nov., S. pacificus sp. nov., and Spiophanes cf. longisetus (Spionidae). For all newly described species, detailed illustrated descriptions are provided and the morphological distinction to congeners is discussed. Identification based on morphological information alone is difficult for some species studied here and will rely on well-preserved material and additional molecular information. However, after the now published descriptions of the new species, more information could be gathered in the future and eventually species diagnoses refined and further improved.

Patterns of geographic distribution

Among our target species of abyssal polychaetes, three species, Sigambra magnuncus, Progoniada regularis, and Bathyglycinde profunda, were confirmed to have a pan-oceanic distribution, i.e. in the Atlantic as well as in the Pacific. Sigambra magnuncus was one of the most frequent species in our source material. It presented comparatively high levels of genetic diversity and a large number of genetically divergent COI haplotypes, most of them singletons being equally distributed across the species distributional range. Our molecular and morphological data suggest that S. magnuncus not only has a continuous longitudinal distribution along the whole deep Atlantic Ocean east and west of the Mid-Atlantic Ridge, but also occurs in the SE Pacific Ocean, with very low variability of morphological characters. The other two pan-oceanic species, i.e. Progoniada regularis and Bathyglycinde profunda, also displayed comparatively high haplotype diversity. For P. regularis, eight different haplotypes from 10 individuals were determined, including five haplotypes alone from five individuals from the Atlantic and the remaining three haplotypes from five Pacific individuals. Intraspecific diversity of Bathyglycinde profunda was on the same level as in P. regularis but here several of the closely related haplotypes from the Pacific were collected multiple times each, and an inter-oceanic differentiation could be detected. One haplotype of B. profunda was found to be shared between oceans, which was not the case in any other species. Interestingly, it came from individuals collected from locations with the greatest possible geographic distance in our dataset, being the Cape Basin in the SE Atlantic and the CCZ German licence area in the central Eastern Pacific. Haplotype diversity was not reflected in variability of morphological characters studied, which means consistent morphological differences related to the origin of different haplotypes could not be documented.

Pan- and trans-oceanic distributions have been documented before (e.g. Georgieva et al. 2015, Eilertsen et al. 2018, Guggolz et al. 2020, Zhou et al. 2020, Jażdżewska et al. 2021) and might not be rare among invertebrates from the deep-sea, though detecting such distribution patterns is usually impeded by the lack of available samples. Even some of the geographically restricted species studied herein may have wider geographic and possibly even trans-oceanic distributions, but this could be masked by the limited number of samples currently available. Guggolz et al. (2020), who studied Atlantic and Eastern Pacific polychaete species, have suggested that dispersal and gene flow occurred either recently (and possibly continuously) or historically, based on very similar and, in some instances, even shared 16S rRNA haplotypes. For the three species with pan-oceanic distribution we suggest a similar scenario, though only B. profunda includes a shared pan-oceanic haplotype. The latter species also had the strongest inter-oceanic differentiation. Here it seems likely that the pan-oceanic distribution is the result of a single historic dispersal or range expansion event. The higher genetic diversity observed in the Pacific probably implies that the origin of the species lies here. However, it can not be ruled out that the difference in genetic diversity between oceans is due to the lower sampling size in the Atlantic. The distribution of the genetic diversity of Sigambra magnuncus and Progoniada regularis was more complex and suggested multiple and possible ongoing dispersal and genetic exchange events between oceans. This is most evident in S. magnuncus with the many highly divergent haplotypes occurring in both oceans.

Among the remaining species, all restricted to one ocean, the newly described Spiophanes pacificus sp. nov. shows a huge distribution area spanning over the whole Pacific. It is noteworthy that the genetic differentiation observed between West and East Pacific populations within S. pacificus sp. nov. is more pronounced than between Atlantic and Pacific populations of S. magnuncus, P. regularis, and B. profunda. Given the low number of involved species, we are hesitant to put too much emphasis on this finding, which might just reflect biological differences between the species. On the other hand, it might indicate that dispersal and gene flow rates between the Atlantic and Pacific are higher than within the Pacific. According to Hollister et al. (1984) the global pattern of strong near-bottom currents creates more potential for redistribution of sediment and organic matter in the Atlantic than in the Pacific. Although all major oceans are directly connected to the southern polar region where they receive inflow from deep, cold currents, the connection to the Arctic is quite different. The flow of deep, cold water from the Arctic into the Pacific is regarded as negligible, whereas large volumes of deep water (estimates of 10 million m³/s) forming the Western Boundary Undercurrent move south into the Atlantic (Hollister et al. 1984).

Isolation versus connectivity

The only alternative to inter-oceanic dispersal and gene flow would be vicariance related to the closure of the Isthmus of Panama >3 million years ago (O’Dea et al. 2016). According to Chevaldonné et al. (2002) passage of deep water between the Pacific and Atlantic was interrupted long before the final closure of the Isthmus (ca. 34 Mya). COI substitution rates are largely unknown for deep-sea organisms; however, they would need to be much lower than the 3.5–4.7% per million years as recently suggested for Arctic polychaetes (Loeza-Quintana et al. 2019). Chevaldonné et al. (2002) have suggested a rate of ca. 0.2% per Myr, and regarded this estimation in good agreement with the scenario proposed by Jacobs and Lindberg (1998) postulating the eradication of deep-sea fauna by anoxic events in the Cenomanian/Turonian and latest Paleocene, caused by warming of bottom waters (Rohling et al. 1997). This extinction was potentially global but almost exclusively occurred in the deep sea (Röhl and Ogg 1996). Accordingly, these authors concluded that the deep-sea and vent faunas were possibly eliminated from bathyal and abyssal habitats and most likely (re-)populated the deep-sea after these events in the Late Cretaceous and Early Tertiary (Jacobs and Lindberg 1998). We think that the estimated rate of 0.2 % per Myr by Chevaldonné et al. (2002) may be too conservative. It was based on three pairs of sister-species (Amphisamytha galapagensis/A. n.sp. (JdF = A. caldarei); Paralvinella grasslei/P. palmiformis, and Oasisia alvinae/Ridgeia piscesae) that were deemed to have been separated by the split-up of the Farallon–Pacific Ridge 28.5 Mya. Thus, dispersal between the diverging rift systems may have persisted for a long period after the initial separation, resulting in shorter divergence times and higher substitution rates. Even if we assume that the substitution rate lies between the estimates of Chevaldonné et al. (2002) and Loeza-Quintana et al. (2019), it is highly unlikely that the current distribution of the species studied here can be explained by vicariance associated with the closure of the Isthmus of Panama. More recent dispersal between oceans appears to be the more plausible explanation for the trans-oceanic distribution of the species studied herein.

Connectivity by larval dispersal and sediment transport in abyssal depths

Free-swimming larvae have been described for many Polychaeta (eg. Bhaud and Cazaux 1987) and likewise for groups treated herein [Pilargidae: Bhaud (1974), Achari (1975); Goniadidae: Böggemann (2005); Magelonidae: Wilson (1982); Spionidae: Blake (2006), Blake et al. (2019)] and it might consequently be suggested that genetic connectivity is maintained by larval dispersal (e.g. Brasier et al., 2017).

Examples of invertebrate larvae with morphological and physiological adaptations are known to be capable of travelling great distances in different depths using favourable ocean currents (Thorson 1961, Scheltema 1968, 1986a, b). Among polychaetes, so-called long-distance or ‘teleplanic larvae’ (Scheltema 1971) have been reported for Spionidae and Chaetopteridae (Scheltema 1986a, 1988), and larvae of these taxa and other polychaetes have been found to be commonly represented in surface waters of the vast central Pacific (Scheltema 1986a). For deep-sea species, demersal development is regarded as most common (Levin and Bridges 1995) and near-bottom currents could be most important for larval dispersal. However, by ontogenetic vertical migration the larvae of benthic deep-sea species could also benefit from major ocean currents in different layers of the water column. Diversity of reproduction and development among polychaetes shows a plethora of adaptations (Giangrande 1997, Rouse and Pleijel 2006), but given the limited number of studies we are unable to assess the importance of larval dispersal for abyssal polychaete species.

Another more plausible option how connectivity might be maintained in the deep-sea is the possible dislodgement of benthic invertebrates by sediment transport. Hollister and McCave (1984) assessed sediment dispersal and deposition along the Nova Scotian Rise and concluded that ‘… mean deposit thickness of 10³ yr net accumulation can be removed in a few weeks and redeposited in a few months’. They also examined implications for deep-sea areas on a global scale and eventually identified several regions displaying evidence of abyssal storm activity. Such areas are seen as major sources of suspended material, which is raised from the bottom and then advected across ocean basins. These areas are to be found in all oceans (e.g. Hollister et al. 1984), including locations where specimens studied in this paper originate, e.g. the North Atlantic basins beneath the Gulf Stream system and the Argentine Basin beneath the confluence of the Falkland and Brazil currents (Hollister and McCave 1984: fig. 5; Harris 2014: fig. 6). Measurements from the Argentine basin documented storms lasting one to a few weeks with maintained particle concentration of several hundred µg/L (Richardson et al. 1993). During abyssal storms, currents are expected to transport up to a ton of sediment per minute (Hollister et al. 1984). Seibold and Berger (1982) state that the fastest normal ocean currents run at ca. 2 m/s (3.9 kn), and sediment starts moving with current speed of 0.5–2 m/s, depending on grain size, with finer sediments being more resilient. According to these authors, only short events are needed to invoke considerable transport and once sediment is brought into suspension even relatively weak currents can keep it moving (Seibold and Berger 1982). It appears that sediment transport, in particular as it is known to occur also in abyssal depths (Hollister et al. 1984, Thistle et al. 1991), should be given more attention. We here want to stress the potential of sediment transport, involving the dislodgment of benthic fauna, to operate as an effective means of maintaining connectivity between populations in the deep-sea. Sediment transport might explain why deep-water species are much wider distributed than expected.

Conclusion

Our results, based on morphological and genetic data, show that populations of the studied polychaete species are connected over great distances and even between oceans. As possible means of dispersal over such wide distances we discuss larval transport by ocean currents, as well as dislodgment of benthic specimens, together with the sediment, by abyssal storms and subsequent transport by near-bottom currents. To further elucidate the distribution and dispersal of deep-sea species the collection of additional samples from all oceans would be helpful, as well as studies on reproduction and larval retention in the plankton. Further assessment of large-scale hydrological events in the deep-sea affecting the sediment and the respective infauna would certainly deliver interesting results.

urn:lsid:zoobank.org:pub:65B60DD3-64C9-4262-B7B2-74DA4D3D889Furn:lsid:zoobank.org:act:B98F17A7-AFA2-4071-8AD2-97433E249A4Aurn:lsid:zoobank.org:act:D4D1BFC8-8B45-49FD-96D3-1D257E412682urn:lsid:zoobank.org:act:9394DE22-D1A5-4DF7-A4AA-3248BC993038

Acknowledgements

We would like to thank Pedro Martinez Arbizu, Lenaick Menot, Angelika Brandt, and Jens Greinert for making available polychaetes collected during expeditions led by them for our study. Special thanks goes to Viola Fischer who took great care of the polychaetes on board of RV Sonne during the KuramBio expedition and while sorting specimens later in the lab. Theresa Guggolz did the same during the VEMA-Transit expedition and organized the material for deposition in the collections of the ZMH. Working with Theresa has always been a pleasure and resulted in different publications. Christian Borowski (MPI for Marine Microbiology, Bremen, Germany) made available polychaetes from expeditions DISCOL 1–3 for this study. Especially the description of S. magnuncus draws on specimens studied by him and on his early observations. The drawings of S. magnuncus (Fig. 8) were originally prepared by Christian Borowski and are here reproduced with permission. Karen Jeskulke and Nicole Gatzemeier (DZMB Hamburg, Senckenberg) helped with DNA barcoding and kept track of specimens used for lab work. Antje Fischer (DZMB Hamburg, Senckenberg) retrieved literature from the libraries and gave support with sample management in our database. Lenka Neal (Natural History Museum, London) was happy to discuss and share her knowledge on available information for deep-sea species from current projects. Janna Peters (DZMB Hamburg, Senckenberg) discussed with us statistical analyses with regard to evaluation of morphological characters in Spiophanes. Frank Friedrich (Hamburg University) and Torben Riehl (Senckenberg Frankfurt) gave advice for the use of CLSM. Renate Walter (University of Hamburg) assisted with SEM studies. Marie-Louise Tritz (Senckenberg Frankfurt), Katrin Philipps-Bussau, and Petra Wagner (Zoological Museum, Hamburg, Germany) provided specimens from, as well as their final integration to, the respective museum collections. Thoughtful comments by Lenka Neal and three additional reviewers helped to improve the manuscript.

Conflict of interest

The authors declare that they are not aware of any conflict of interest.

Data availability

All links and identifiers for our data are presented in the manuscript and in supplementary data Tables S1 and S2. Additional data are available from the authors upon request.