Terebratulina septentrionalis ( Couthouy, 1838 ) Eucalathis ergastica Fischer & OEhlert, 1890

Terebratulina septentrionalis ( Couthouy, 1838) View in CoL

Figures 21 View Figure 21 , 22 View Figure 22

Terebratula septentrionalis Couthouy, 1838, p. 65 –66.

Terebratulina septentrionalis View in CoL – ( Davidson 1852b), p. 366.

Terebratulina caput-serpensis var. septentrionalis View in CoL – Davidson (1880), p. 33 –36, pl. 1, figs. 3–9.

Description: This species is very similar to Terebratulina retusa and has overlapping ranges of almost all diagnostic characters. As noted by Sars (1878) and Wesenberg-Lund (1941), T. septentrionalis tends to have the following characters: more egg-shaped outline; rectimarginate or weakly uniplicate anterior commissure; thinner shell-wall; weaker growth lines; denser ribs with 14–22 usually low subtriangular or rounded radiating ribs per 5 mm counted 10 mm anterior of ventral umbo; tubercles on umbonal crest missing or less strongly developed; stronger tubercles usually scattered laterally on umbone; ribs lower and tend to be nearly absent distally on larger specimens; often slightly wider pedicle opening (11–15% as wide as valve). In the present study, the most stable difference from T. retusa was found to be the absence of or denser, lower and more transverse tubercles never overhanging deep interspaces between the ribs on the ventral umbonal crest. The largest measured specimen was 27.5 mm long.

Depth range: 0–3614 m depth ( Dall 1920; Brunton & Curry 1979), but mostly occurring at 10–550 m depth, with the seconddeepest recorded occurrence at approximately 1325 m ( Thomsen 2012).

Temperature range: -1.7–10.6˚C ( Allmen et al. 2010; this study).

Salinity range: 28–35.4 ( Brand et al. 2003; this study).

Oxygen range: 55–104% saturation (this study).

Current velocity: Mean current velocity at four locations measured at 2.7–6.9 cm /s, with a normal daily maximum of 2–18 cm /s and 10.8–24.8 cm /s as the absolute maximum over 1 month of measurement (this study).

Substrate: Commonly attached to serpulid tubes, but also to shells, shell sand, branched bryozoans, the coral Desmophyllum , Porifera, other polychaete tubes, or gravel (this study).

Geography: S, W and E Greenland, Svalbard (Bjørnøya, Spitsbergen, and verified from off northern Nordaustlandet in 2009, and from the Hinlopen Strait), Murmansk, the White Sea, Norway (Hordaland and North), Baffin Bay, East Canada, Jan Mayen, Iceland, Scotland, Ireland and E USA (Grieg 1933; Wesenberg-Lund 1941; Brunton & Curry 1979; Curry & Endo 1991; Thomsen 2001; Allmen et al. 2010; Thomsen 2012; this study).

Remarks: Terebratulina septentrionalis has been synonymised with the species Terebratulina retusa several times (e.g., Jeffreys 1878; Davidson 1880; Posselt 1898; Knipowitsch 1901; Hägg 1905; Zezina 1997a,c) since the species are very similar and intermediate morphotypes are found to occur in places where both species are present (e.g., Wesenberg-Lund 1941; Lüter et al. 2017). Cohen et al. (1991) addressed this question through a molecular study on specimens from British and Icelandic waters. They concluded that there were two distinct species and found no evidence of hybridisation. Recently, in their follow-up study, Lüter et al. (2017) found hybrids in places where both species co-occur in Iceland. Hybridisation could explain some observed morphological trends along Norway, where T. retusa tends to show finer costellae, thinner shells and weaker tubercles in northern regions dominated by T. septentrionalis .

Sars (1878) found that the crural processes in T. septentrionalis were not connected, in contrast to those of T. retusa . This difference was not supported by the examined material in the present study, which included both specimens with disconnected and connected processes for both species. Logan (1979) found there to be an ontogenetic trend with the connection generally developing latter in T. septentrionalis than in T. retusa . Blochmann (1908) used the spicules in the tissue to distinguish the two species. However, Wesenberg-Lund (1941) questioned the usefulness of this method. In the present study, I also failed to separate the two species based on spicules.

The variability and usefulness of all the expected diagnostic features of the two species were rigorously examined based on a large collection containing a wide range of shell sizes, population samples and geographically separated material of both species. As expected, the shell shape and ornamentation of both species show strong variability between specimens within the same population, but also along different growth stages of single specimens, with most features varying independently of each other. It can be concluded that the range of variation within each single feature shows a considerable overlap between the two species. The most stable differences surprisingly turned out to be the ornamentation on the umbonal part, especially in specimens 2–10 mm in length.

Eucalathis ergastica Fischer & OEhlert, 1890 View in CoL

Figures 23 View Figure 23 , 24 View Figure 24

Eucalathis ergastica Fischer & OEhlert, 1890, p. 73 View in CoL .

Description: Shell low to moderately ventribiconvex and slightly longer than wide. Outline subtriangular to subpentagonal and widest close to front. Ventral umbo pointed and with moderately high interarea bisected by elongate subtriangular pedicle opening. Pedicle divided into many rootlets. Anterior commissure rectimarginate. Dorsal valve often with slight sulcus. Ornamentation of 14–27 costellate ribs, as well as abundant and strong lamellar growth breaks. Ribs triangular in cross section. Middle dorsal rib not larger than the others. Shell matrix endopunctate with very tiny punctae. Colour porcelain white to slightly yellowish. Broad hinge teeth not supported by dental plates. Short and anteriorly directed loop. Crural processes not meeting. Dorsal median septum lacking. Maximum shell size 7 mm.

Depth range: 280–2736 m depth (Logan 2007).

Temperature range: 9.5–12˚C (this study).

Substrate: Mud, sand, stones and rock, and often associated with corals ( Fischer & OEhlert 1891; Logan 1983; Anadón 1994; Anadón et al. 2022). Can also be attached to Porifera as in the illustrated examples.

Due to the past synonymisation and confusion of the two species, the published geographic distribution—especially in the Arctic region—may be based on a mix of the species in many cases. Therefore, in the present study, the more uncertain observations were excluded. Some of the earlier identifications made in the present study are also erroneous since a greater understanding of the value of various diagnostic features evolved along the way. This latter problem has been addressed to some extent by re-examining the specimens from many of the border occurrences.

Geography: Celtic Sea, Bay of Biscay, W Spain, Azores, Morocco, the Canary Islands and Western Sahara ( Fischer & OEhlert 1890; Cooper 1981a; Logan et al. 2007).

Material collected from southern Norway during the present century was not available for the present study. Therefore, no conclusions can be drawn from the absence of data from this area.

Like the previous species, this species is often overgrown by sponges, most likely of the genus Hymedesmia (belonging to Demospongia) ( Tendal & Thomsen 2005; Thomsen 2012).