DESCRIPTION AND ECOLOGY OF THE NEW FORAMINIFERAL SPECIES ELPHIDIUM BALTICUM – WITH A GENETIC AND MORPHOLOGICAL COMPARISON TO ELPHIDIUM INCERTUM (WILLIAMSON, 1858) AND ELPHIDIUM ASKLUNDI BROTZEN, 1943

Elphidiidae are benthic foraminifera found throughout the world’s oceans from low to high latitudes. They are primarily found in inner to outer shelf areas, from brackish to hypersaline habitats; most species are epifaunal or shallow infaunal, while others may live clinging to macroalgae (e.g., Murray, 1991, 2006). It is a highly diverse group with many species whose relatively small differences in morphology may lead to difficulties in taxonomic determinations and cause confusion for ecological and palaeoecological studies.

Until now, more than 25 phylotypes of elphidiids have been identified with molecular approaches based on the SSU rDNA gene (Pillet et al., 2013; Darling et al., 2016; Brinkmann et al., 2023). By combining morphological and molecular recognition of elphidiids, these studies helped to disentangle taxonomic problems such as synonymy or cryptic species. For example, the species that Haake (1962) described as Elphidium voorthuyseni has turned out to correspond to Elphidium incertum [originally named Polystomella umbilicatula (Walker) var. incerta Williamson, 1858] (Darling et al., 2016). Therefore, the name E. voorthuyseni, widely used in Scandinavian and North European literature, is now regarded as a junior synonym to E. incertum (phylotype S6 of Darling et al., 2016) and should not be used anymore, as also proposed by Haynes (1973). In addition, in Scandinavian, as well as in much European and Russian literature, the name E. incertum has traditionally been used for two different morphotypes, one of which thus now remains unnamed. This issue was also mentioned by Darling et al. (2016) and highlights the limitations of morphology-based taxonomy, when cryptic (indistinguishable morphologically but genetically distinct) or pseudo-cryptic (first indistinguishable morphologically but can be discriminated after morphological re-examination) species are studied. A new phylotype, S23, identified recently by Brinkmann et al. (2023), corresponds to this hitherto unnamed morphotype, which can be both morphologically and genetically distinguished from E. incertum (Williamson).

The aim of the present paper is to describe a new species representing the commonly occurring morphotype found in boreal and Arctic regions, which has previously erroneously been referred to Elphidium incertum (Williamson). Since Williamson’s type specimens (syntypes) of E. incertum hold the morphology of the phylotype S6 described by Darling et al. (2016), corresponding to the morphotype that Haake (1962) described for Elphidium voorthuyseni, we are now left with a very common morphotype without a species name but already defined as a phylotype (S23). Based on the morphological description of sequenced, recent non-sequenced and fossil specimens, as well as new molecular studies (Brinkmann et al., 2023), we therefore define this new species as Elphidium balticum n. sp.

The present work includes foraminiferal material from two marine sediment cores (M1 and M5) collected in 2009 in Aarhus Bay off the east coast of Jutland, Denmark (Fig. 1), using a gravity corer onboard the Aarhus University research vessel Aurora (Rasmussen et al., 2019). Core M1 was retrieved from the center of the bay (56.117°N, 10.347°E; water depth 16 m), while core M5 is from a more easterly site (56.100°N, 10.450°E; water depth 28 m). The 1101-cm-long M1 core covers the period ca. 10,000–3,700 cal. years BP, whilst the shorter M5 sequence covers the last ∼4,400 years (645 cm long). Additional material for comparison was collected from last glacial (Weichselian) deposits of the Apholm 7.131b core (57.463°N, 10.524°E; Knudsen, 1984) and from the Late Holocene of the Skagen 5 borehole (57.722°N, 10.577°E; Thorsen & Ibsen, 2006), both drilled on land in northern Jutland, Denmark. All these foraminiferal samples were prepared using standard techniques described in Feyling-Hanssen et al. (1971), with undried sediment samples being wet sieved on 100-μm sieves and subsequently concentrated with heavy liquid (C₂Cl₄). Recent elphidiid specimens from Holywood, Ireland (54.639°N, 5.825°W; intertidal mudflat; Schweizer et al., unpublished data), and from Vie Estuary, NW France (46.70°N, 1.93°W; intertidal mudflat; Fouet et al., 2022; Jorissen, 2023; Jorissen et al., 2023), were also included for comparison. The recent sediment samples were collected at low tide on mudflats, stained with Rose Bengal and wet sieved at 125 μm. Stained foraminifera were picked under a Leica MZ16 stereomicroscope.

Living specimens of Elphidium were collected in June 2019 at a water depth of 30 m in the Havstens Fjord (west coast of Sweden, 8.313°N, 11.773°E) for molecular studies using a GEMAX Gemini-type twin-barrel corer during the TO₂PICal project (Brinkmann et al., 2023). An additional sample for molecular studies was collected at low tide on the Vie Estuary mudflat (Jorissen, 2023).

As described in Brinkmann et al. (2023), samples from the top 2 cm of the sediment in the GEMAX cores were stored with ambient seawater in bottles in a cooler, then transported to the Laboratory of Planetology and Geosciences (LPG) at the University of Angers (France). The same was done with sediment collected in the Vie Estuary. In Angers, all samples were sieved (>100 μm) with artificial seawater and kept at 4°C. Within a week after collection, the sieved sediment was examined under a stereomicroscope (Leica S9i). Foraminifera with a colored cytoplasm and an empty last chamber were selected and placed in Petri dishes containing artificial seawater with fine sediment. Vitality was confirmed when observing individual activity overnight. The live foraminifera were then picked, cleaned with a brush and air dried for imaging with an environmental Scanning Electron Microscope (SEM) Zeiss EVOLS10 (Havstens Fjord specimens) or a Hitashi TM4000 (Vie specimen). Specimens were then crushed individually in a DNA extracting buffer (DOC, see Pawlowski, 2000). Foraminiferal-specific primers s14F3-J2 and s14F1-N6 (Pawlowski, 2000; Darling et al., 2016) were used with two rows of PCR (Polymerase Chain Reaction) following the protocol described in Darling et al. (2016) to amplify the DNA extractions. The amplified region (∼500 base pairs) is situated at the 3′ end of the SSU rDNA, in the 18S V9 region, and is used for foraminiferal barcoding (Pawlowski & Holzmann, 2014). Positive amplifications were sequenced with the Sanger method (Macrogen Europe, Amsterdam) and edited with Seaview v.4 (Gouy et al., 2010). Three specimens identified as Elphidium sp. S23 (GF885, GF886, and GF922) were deposited in the GenBank database under the accession numbers ON818447, ON818448, and ON818460 (Brinkmann et al., 2023). One specimen identified as Elphidium incertum (Vi031) was deposited in the GenBank database under the accession numbers OQ682703 (Jorissen, 2023).

DNA sequences from clades B, C, D, E, and G of elphidiids (Pillet et al., 2013; Darling et al., 2016) were retrieved from GenBank (http://www.ncbi.nlm.nih.gov) (Table 1) and aligned together in Seaview v.4 (Gouy et al., 2010). Sequences published by Pillet et al. (2013) represented the complete SSU rRNA gene, but were cut to adapt to the ∼1,000bp 3′ end of the SSU rRNA gene fragment of the other sequences, which is the barcode for foraminifera. Newly published 500bp long sequences belonging to S6 (Jorissen, 2023), S23, and S24 phylotypes (Brinkmann et al., 2023) were then added to the alignment. A maximum likelihood (ML) molecular phylogenetic tree (Fig. 2, Table 2, Appendix Fig. 1) was built with the PhyML program (Guindon & Gascuel, 2003) implemented in Seaview, choosing the GTR (General Time Reversible) evolutionary model (Tavaré, 1986) and the aLRT (approximate Likelihood Ratio Test) branch support (Anisimova & Gascuel, 2006). To correct for across site rate variation, gamma distribution (Γ) was optimized by PhyML (GTR + Γ). Among the 1149 sites used in the phylogenetic analysis, 465 (40.5%) had no polymorphism. A BioNJ tree (Gascuel, 1997), implemented in Seaview, was also built under the Kimura’s two parameters (K2P) model (Kimura, 1980) with 1,128 sites and 100 bootstrap replicates (Appendix Fig. 2). According to the topology of previous elphidiid trees (Pillet et al., 2013; Darling et al., 2016), clade B was chosen as the outgroup to root the tree for both analyses.

The topology of the ML tree (Fig. 2, Table 2) based on 41 sequences of elphidiids (Appendix Fig. 1) is slightly different from the ones shown in Darling et al. (2016, fig. 2) and Pillet et al. (2013, fig. 1). In our study, as in the study by Darling et al. (2016), the more basal nodes are statistically not well supported, and the topologies are not the same in the different analyses (ML and BioNJ here; Appendices 1, 2), with the addition of Bayesian analysis (BA) in Darling et al., 2016). This can be explained by a low phylogenetic signal present in the studied fragment of SSU rDNA (about 1,000 nucleotidic sites), as the phylogenetic analyses of the complete SSU rDNA gene (about 3,000 sites) gives much better statistical supports (Pillet et al., 2013, ML and BA trees). Clades B, D, and E were recognized in the present study with high to very high statistical supports (1.00/99 for clade B, 1.00/98 for clade D, and 1.00/83 for clade E), but clade C (S16 and Canada) was polyphyletic in both ML and BioNJ (Fig. 2). The new species Elphidium balticum (S23) forms a well-supported clade (Fig. 2, 1.00/100%), and groups with Elphidium asklundi (S19) and Haynesina nivea (S20), its genetic closest relatives, in clade D. In contrast, Elphidium incertum (S6) is genetically less related to E. balticum, as it groups in clade B with Elphidium sp. (S14) and Elphidiella tumida (S22), the latter referred by Pillet et al. (2013) to Elphidiella groenlandica.

Images of Elphidium balticum n. sp., E. incertum, and E. asklundi are illustrated in Figures 3–9, including the sequenced specimens GF885, GF886, GF922, and Vi031. Light microscope multifocal images were prepared using a DeltaPix M12Z, while scanning electron images were obtained using a Tescan VEGA Scanning electron microscope (SEM), both at the Department of Geoscience, Aarhus University, Denmark, and a TM4000Plus HITACHI SEM at the Laboratory of Planetology and Geosciences, University of Angers, France. Computed Tomography (CT) scans were performed using a Zeiss Xradia 620 Versa X-ray microscope at the Interdisciplinary Nanoscience Center (iNano), Aarhus University. Images and videos of the CT scans of three specimens were made using Dragonfly Software, Version 2022.2 for Windows.

The classification below follows Adl et al. (2019), Pawlowski et al. (2013), and Holzmann & Pawlowski (2017) with recognition of phylum status for the foraminifera.

• RHIZARIA Cavalier-Smith, 2002

• RETARIA Cavalier-Smith, 2002

• Phylum FORAMINIFERA d’Orbigny, 1826

• Class GLOBOTHALAMEA Pawlowski et al., 2013 

• Order ROTALIIDA Lankester, 1885

• Superfamily ROTALIOIDEA Ehrenberg, 1839

• Family ELPHIDIIDAE Galloway, 1933

• Genus Elphidium de Montfort, 1808

• Elphidium balticum Knudsen, Schweizer & Seidenkrantz, n. sp.

• Figs. 3–7 

• Elphidium incertum (Williamson), Rottgardt, 1952, p. 182, pl. 2, fig. 27; Feyling-Hanssen et al., 1971, p. 277, pl. 12, figs. 11, 12; pl. 21, figs. 8, 9; Hansen & Lykke-Andersen, 1976, p. 35, pl. 12, figs. 5–9; Grobe & Fütterer, 1981, pl. 1, figs. 1–6; pl. 2, figs. 3–8; Knudsen, 1982, p. 170, fig. 14:12, no. 16; Knudsen, 1993, p. 86, pl. 2, figs. 4–6; Polyak et al., 2002, p. 257, pl. 1, figs. 1–4 (not figs. 5–7); Polodova et al., 2009, p. 135, pl. 1, figs. 12–15; Schönfeld, 2018, p. 389, pl. 1, figs. 1–2 and figs. 8–11(not fig. 3); figs. 6, 7, 12, 13, 14, and 15: presumably also synonyms, but this cannot be determined with certainty from the images; Knudsen et al., 2021, p. 493, pl. 8.4, fig. 16.

• Cribrononion incertum (Williamson), Brodniewicz, 1965, p. 207, pl. 10, figs. 9–11; text figs. 30, 31; Lutze, 1965, p. 103, pl. 21, figs. 43, 44; text fig. 17; Lutze, 1974, text fig. 9.

• Cribroelphidium incertum (Williamson), Wollenburg, 1995, p. 41, pl. 6, fig. 14.

• Elphidium sp. (S23), Brinkmann et al., 2023, p. 6, fig. 2, image (P); Supporting information, p. 33, GF885-Elphidium_sp-S23 and GF886-Elphidium_sp-S23; p. 34, GF922-Elphidium_sp-S23.

Refers to the common occurrences of the species in the Baltic Sea around Denmark, Sweden, northern Poland, and Germany.

Planispiral, involute, moderately depressed elphidiid with rounded periphery and 7–11 chambers in the last whorl. It has only a few retral processes in the sutures and a multiple aperture at the base of the apertural face. The sutures and the foramina are covered by dense turberculation.

Specimen from Aarhus Bay, Core M5 (56°100 N, 10°450 E, water depth 28 m), core depth 94–95 cm. Natural History Museum of Denmark catalog number NHMD 001768457 is housed at The Natural History Museum of Denmark, Copenhagen (Figs. 3.1a–d).

Sixteen specimens from Cores M1 and M5 Aarhus Bay. Twelve specimens are stored at The Natural History Museum of Denmark, Copenhagen: Paratype 1 (NHMD 001768458) from Aarhus Bay M5, 94–95 cm (Figs. 3.2a–c); Paratype 2 (NHMD 001768459) from Aarhus Bay M5, 60–61 cm (Figs. 3.3a–d); Paratype 4 (NHMD 001768460) from Aarhus Bay M1, 958–959 cm (Fig. 4.6); Paratype 6 (NHMD 001768461) from Aarhus Bay M1, 958–959 cm; Paratype 7 (NHMD 001768462) from Aarhus Bay M1, 812-813 cm; Paratype 9 (NHMD 001768463) from Aarhus Bay M5, 94–95 cm; Paratype 10 (NHMD 001768464) from Aarhus Bay M5, 94–95 cm; Paratype 11 (NHMD 001768465) from Aarhus Bay M5, 94–95 cm; Paratype 12 (NHMD 001768466) from Aarhus Bay M5, 60–61 cm; Paratype 14 (NHMD 001768467) from Aarhus Bay M1, 987–988 cm; Paratype 15 (NHMD 001768468) from Aarhus Bay M1, 978–979 cm; Paratype 16 (NHMD 001768469) from Aarhus Bay M1, 958–959 cm.

Two specimens are housed at the Natural History Museum, U.K.: Paratype 3 (NHMUK PM PF 75275) from Aarhus Bay M5, 60–61 cm (Figs. 4.8a, b); Paratype 13 (NHMUK PM PF 75276) from Aarhus Bay M1, 987–988 cm. Two specimens are housed at the Smithsonian Institution, Washington, D.C., U.S.A.: Paratype 5 (USNM PAL 795605); Paratype 8 (USNM PAL 79606); both from Aarhus Bay M5 94–95 cm.

Holocene; Aarhus Bay, East Jutland, Denmark.

Aarhus Bay off the east coast of Jutland, Denmark.

Holocene.

Test planispiral, involute with 7–11 (mostly 8–9) embracing chambers in the final whorl. Outline smooth to slightly lobulated, particularly in the final part. Sutures curved, only slightly depressed with only a few retral processes, and rounded periphery. The thickness of the wall differs from thick, white with poorly visible pores to thin and transparent, with a distinct perforation of the wall. The apertural face is rounded, smooth, finely perforated and with an interiomarginal, multiple aperture at the base of the apertural face. Dense fine turberculations (papillae) occur around the multiple aperture and on the earlier part of the final whorl (Fig. 6.1), as well as in the depressed umbilicus and in the depressed part of the sutures (fossettes/slits). On juvenile forms, papillae also occur on the lower part of the apertural face (Figs. 7.3a, 7.3b). The last chamber is often missing, and the foramina are similar to the aperture, but with a less dense tuberculation (Fig. 6.2).

The hyaline calcite wall is perforate, laminated (bilamellar) and optically granular. In modern (living) specimens, the sutural fossettes may be hidden or obscured by the additional, secondary lamination, but they become visible in fossil tests, which are often slightly etched on the surface. The test wall is often translucent in well-preserved specimens, but opaque tests are also common. Sometimes small dark spots are seen on the surfaces, probably from remnants of a protective cocoon of agglutinated material that often covers living specimens (cf., Polodova et al., 2009). For description of the canal system (based on SEM of plastic moulds), see Hansen & Lykke-Andersen (1976).

Holotype: largest diameter: 0.53 mm; smallest diameter: 0.41 mm; thickness: 0.27 mm.

The morphological variation in Elphidium balticum n. sp. is only minor. There is some variation in size, and the outline may be lobulate throughout or the lobulation may be restricted to the later part of the last whorl. The number of retral processes differs quite significantly, and in some living specimens the fossettes are not visible at all. Other variations are mainly linked to preservation.

Difficulties in morphological separation of E. asklundi and E. balticum n. sp. have been mentioned by several authors, and Feyling-Hanssen et al. (1971) and Hansen & Lykke-Andersen (1976) even suggested that E. asklundi and E. balticum n. sp. (as E. incertum) may be considered as ecophenotypes. The problem with a lack of morphological hiatus between these two species was also mentioned by Voltski et al. (2015). The present genetic studies, however, show that E. asklundi and E. balticum n. sp. represent two related, but distinct phylotypes and can be regarded as different species (see above, Fig. 2).

The specimens attributed here to Elphidium balticum n. sp. were previously merged within the morphospecies, Elphidium incertum Williamson. It has therefore been important to separate the specimens truly belonging to E. incertum from the ones with a different morphology. Elphidium incertum is a relatively small [Haake, 1962 (as E. voorthuyseni): largest diameter 0.24–0.35 mm, thickness 0.08–0.11 mm], thin-walled and very compressed species. The two species Elphidium balticum n. sp. and E. asklundi are both larger and less compressed than E. incertum. However, E. balticum n. sp. is generally smaller than E. asklundi, and its periphery is most commonly less lobulate. The relatively large umbilical region of E. asklundi is often partly covered by large calcareous knobs and pustules, a character which is not seen either in E. balticum n. sp. or in E. incertum.

This morphological profile is based on sequenced specimens in Brinkman et al. (2023), which are also illustrated in the present study (Figs. 5.8–5.10). Test relatively large with 8–9 chambers in the last whorl, rounded smooth to slightly inflated periphery and distinct perforation of the wall. Sutures backwards curving, only slightly depressed with few, often indistinct sutural bridges. Dense papillae occur around the multiple apertures, in the relatively narrow umbilicus, and along the innermost part of the sutures. This phylotype is linked to the morphospecies Elphidium balticum n. sp.

Boreal, subtidal species, which is common in brackish, inner shelf areas with salinity higher than 20 (Lutze, 1965, 1974: as Cribrononion incertum). In stratified waters, it is particularly frequent just below the discontinuity layer (the halocline). The halocline was observed at about 20–25 m water depth in the western Baltic by Lutze (1965, 1974), but the position of the halocline was the determining factor for the distribution rather than the water depth (e.g., Lutze, 1974; Schönfeld & Numberger, 2007). Polodova et al. (2009) found E. balticum (as E. incertum) to be an infaunal, opportunistic species which occurred frequently in the uppermost muddy sediment layer of the inner part of Flensburg Fjord in the westernmost Baltic Sea, where it might reflect the seasonal anoxic conditions in the sediments of this area. It is often found to be cocooned by an agglutinated cyst around its tests (e.g., Exon, 1972; Wefer, 1976; Linke & Lutze, 1993; Gustafsson & Nordberg, 1999), presumably to provide a shelter against mechanical and chemical disturbances as an adaption to rapid changes in the environment (Linke & Lutze, 1993; Polodova et al., 2009). The specimens for DNA barcoding were sampled in the Havstens Fjord in June 2019 at a depth of 30 m with a temperature of 7°C, 4 ml/l of oxygen, and a salinity around 33 (Brinkmann et al., 2023, fig. S1). Nevertheless, the bottom of this fjord (30 m) is periodically hypoxic (Brinkmann et al., 2023, fig. S2).

In the Arctic, E. balticum has been identified in inner shelf areas of the Laptev Sea (Wollenburg, 1995: as Cribroelphidium incertum) and from the Kara Sea (Polyak et al., 2002: as E. incertum). However, Polyak et al. (2002) apparently included other species into this taxon (e.g., Elphidium albiumbilicatum and Elphidium asklundi), and, therefore, it is not possible to work out the detailed distribution of the present species in the Kara Sea.

Pliocene to Recent.

The following morphological diagnoses of the phylotypes S6 and S19 are based on the available SEM images of sequenced specimens and descriptions from Darling et al. (2016) and Pillet et al. (2013).

This morphological profile is based on four sequenced specimens from Darling et al. (2016), supplemented by one specimen (Vi031) from Vie Estuary included in the present study (Jorissen, 2023; Fig. 8.6): Test relatively small with rounded and relatively smooth periphery, 9–10 chambers in the final whorl. Sutures slightly depressed, backwards curving with few (1–4), short and sometimes poorly developed sutural bridges, which leaves longitudinal depressed slits along the sutures. The sutures typically merge towards a very small umbilical region. This phylotype is linked to the morphospecies Elphidium incertum (Williamson, 1858).

This morphological profile is based on six sequenced specimens described by Pillet et al. (2013): Test relatively large, slightly compressed with broadly rounded lobate periphery, 8–9 inflated chambers (Pillet et al., 2013: plate 2, Q–R). The intraseptal space is broad, covered with papillae and with irregular row of septal bridges and pores. The relatively large umbilical area is only slightly depressed. This phylotype is linked to the morphospecies Elphidium asklundi Brotzen, 1943.

Because of the morphological affinities between the species E. balticum n. sp., E. incertum (Williamson), and E. asklundi (Brotzen), we have studied well-illustrated images in the literature and established the following synonymy lists for E. incertum and E. asklundi.

• Elphidium incertum (Williamson, 1858)

• Fig. 8 

• Polystomella umbilicatula, var. incerta Williamson, 1858, p. 44, pl. 3, fig. 82a.

• Elphidium incertum (Williamson), part Cushman, 1930, p. 18, pl. 7, fig. 4a only; part Cushman, 1939, p. 57, pl. 15, fig. 21a only (after Williamson, 1858); Haynes, 1973, p. 199, pl. 22, fig. 6; pl. 24, figs. 14–16; pl. 28, figs. 8, 9; Horton & Edwards, 2006, pl. 4, figs. 18a, b (Williamson’s syntype); Darling et al., 2016, p. 9, 17, figs. 3–4 (S6); Jorissen et al., 2023, p. 58–59, pl. 22, figs. 1, 2.

• Elphidium voorthuyseni Haake, 1962, p. 51, pl. 5, figs. 6, 7; Hansen & Lykke-Andersen, 1976, p. 9, pl. 4, figs. 8–12; Knudsen, 1980, p. 210, pl. 7, figs. 4–6.

• Elphidium asklundi Brotzen, 1943 

• Fig. 9 

• Elphidium? asklundi Brotzen, in Hessland, 1943, p. 267, fig. 109-1.

• Cribrononion obscurus Gudina, 1966, p. 36, pl. 2, figs. 4, 5.

• Elphidium asklundi Brotzen; Feyling-Hanssen et al., 1971, p. 270, pl. 10, figs. 20, 21; pl. 11, figs. 1–5; Hansen & Lykke Andersen, 1976, p. 35, pl. 12, figs. 10–12; Knudsen, 1978, pl. 2, figs. 8, 9; pl. 3, fig. 1; pl. 4, figs. 6–8; Knudsen, 1982, fig. 14:12, no. 1, 2; Feyling-Hanssen, 1990, p. 28, pl. 5, figs. 16, 17; Pillet et al., 2013, p. 7, pl. 2, images Q–R.

• Cribroelphidium asklundi (Brotzen), Wollenburg, 1995, p. 41, pl. 6, fig. 15.

Through many years, Elphidium balticum n. sp. has incorrectly been referred to E. incertum (Williamson) in European literature, and, as a result, two different species have been included into the taxon E. incertum. However, several authors have previously been aware of this problem. For instance, in connection with her work in the Baltic Sea, Brodniewicz wrote in 1965, “the species E. incertum (Williamson), described and illustrated by different investigators, is not identical with Polystomella umbilicatula var. incerta Williamson (1858).”

In 1973, Haynes pointed out that E. voorthuyseni of Haake (1962) is probably a synonym to E. incertum (Williamson), an assumption that was based on the study of a topotype sent to him by F. W. Haake. Based on a comparison with the specimens figured by Brodniewicz (1965), Haynes (1973) also mentioned that the Baltic population of E. incertum consisted of larger specimens (greatest diameter 0.33–0.54 mm, thickness 0.20–0.29 mm). This agrees with the fact that the Baltic population figured by Brodniewicz (1965) now is assumed to represent the new species E. balticum n. sp. and not E. incertum (Williamson).

Later, in connection with their genetic studies of elphidiids, Darling et al. (2016) wrote that “in the literature, the name E. incertum has been used to describe a much wider morphology, which remains an issue to be resolved in future studies.” The aim of the present separation of E. balticum n. sp. from Williamson’s E. incertum is to clear up that problem.

As shown in the molecular phylogeny (Fig. 2), the phylotype S23 (Elphidium balticum n. sp.) groups in clade D with its genetic closest relatives, S19 (Elphidium asklundi) and S20 (Haynesina nivea). In contrast, S6 (Elphidium incertum), which resembles E. balticum n. sp. morphologically, is genetically less related to S23, as it groups with S14 (Elphidium sp.) and S22 (Elphidiella tumida) in clade B.

At the species level, S23 (Elphidium balticum n. sp.) sequences form a well-supported clade with 1.00/100% support. The sequence GF922-ON818460 diverges slightly from the two others (longer branch, see in Fig. 2), therefore a deeper investigation with clones from the same individuals would be needed to investigate the intra-specific variability of S23.

To link molecular and morphological data, the morphological profiles of phylotypes (i.e., the morphological description based on shell SEM images belonging to sequenced specimens) was provided for E. balticum n. sp. (S23), E. incertum (S6), and E. asklundi (S19). These profiles fit with the morphological descriptions of the species and allow for taxonomic names to be given to the phylotypes, which have the rank of species (Darling et al., 2016).

The geographical distribution of the species E. balticum n. sp., E. incertum (Williamson), and E. asklundi Brotzen are different, albeit overlapping. Elphidium balticum n. sp. is mainly distributed in brackish inner shelf areas of the boreal region (as E. incertum or Cribrononion incertum by e.g., Rottgardt, 1952; Lutze, 1965, 1974; Exon, 1972; Wefer, 1976; Linke & Lutze, 1993; Schönfeld & Numberger, 2007; Nikulina et al., 2008; Polodova et al., 2009). In stratified waters of the Baltic area, it is particularly frequent just below the discontinuity layer (the halocline; e.g., Lutze, 1965, 1974). Elphidium balticum n. sp. has also been identified in a few studies from the Arctic Ocean (Wollenburg, 1995: as Cribroelphidium incertum; Polyak et al., 2002: as E. incertum).

In contrast, Elphidium incertum appears to be solely restricted to boreal and lusitanian environments, where it occurs in relatively open marine areas (Haake, 1962: as E. voorthuyseni; Haynes, 1973; Horton & Edwards, 2006; Darling et al., 2016; Jorissen et al., 2023). On the other hand, Elphidium asklundi is a purely Arctic species, restricted to shallow inner shelf areas (Wollenburg, 1995; Pillet et al., 2013). It occurs frequently in Pleistocene glacial deposits from the Scandinavian and circum-Arctic regions (e.g., Gudina, 1966: as Cribrononion obscurus n. sp.; Feyling-Hanssen et al., 1971; Knudsen, 1978, 1982; Feyling-Hanssen, 1980, 1990). Consequently, despite some potential difficulties in distinguishing in particular Elphidium balticum and Elphidium asklundi, their different ecological niches make them valuable indicators in both modern and palaeoenvironmental studies.

A new boreal and Arctic foraminiferal species Elphidium balticum n. sp., which was previously often included in the boreal-lusitanian species Elphidium incertum (Williamson, 1858), is described from the Baltic Sea area. The differences in the morphology of these two groups have previously been pointed out in the literature, but the groups have not been separated in different species. The morphology of the Arctic species Elphidium asklundi Brotzen, 1943, is also relatively close to that of E. balticum n. sp., and these two species are sometimes difficult to separate, particularly the juvenile individuals. Molecular phylogenetic analyses reveal that these three morphospecies are genetically well separated (i.e., Elphidium balticum n. sp. = phylotype S23, E. incertum = phylotype S6, and E. asklundi = phylotype S19. Morphological profiles based on sequenced specimens of the three species are provided to link molecular and morphological data and help in morphological identification. Synonym lists for each of these three different species are based on literature studies of well-illustrated specimens, and the ecological preferences of each species are provided.

We are grateful to Trine Ravn-Jonsen, Department of Geoscience, Aarhus University, for help with the SEM imaging, with advice from Pia Bomholt Jensen, Interdisciplinary Nanoscience Center (iNano), Aarhus University. Our thanks also to Nina Kølln Wittig, Interdisciplinary Nanoscience Center (iNano), Aarhus University, for carrying out the Computed Tomography scanning. MS thanks Inda Brinkmann, Helena Filippson, Marie Fouet, and Frans Jorissen for help in sampling foraminifera, Paul Baraton-Métois for picking Irish foraminifera, the University of Angers, Pays de la Loire, the French Office of Biodiversity and the Swedish Research Council VR (grant number 2017-04190) for funding projects FORESTAT (PI: Frans Jorissen), PhylForBen (PI: Magali Schweizer), and TO₂PICal (PI: Helena Filipsson). MSS acknowledges funding from the Independent Research Fund Denmark (grant no. 0135-00165B GreenShelf) and the European Union’s Horizon 2020 research and innovation program under Grant Agreements No. 869383 (ECOTIP) and 101136480 (SEA-Quester). Finally, we greatly appreciate helpful comments and suggestions from the editor and from two anonymous reviewers.