Historical review of Dutch Pectinariidae with Pectinaria belgica as a new taxon for the Netherlands (Annelida: Polychaeta: Pectinariidae)

Pectinariidae, “trumpet worms” or “ice-cream-cone worms” (in Dutch: Goudkammetjes), are easily recognizable by their distinctive anterior golden paleae (Fig. 1). They live in a fragile and thin cone-shaped tube consisting of sand, small shells, or other small particles. The diagnosis of the family is presented in Rouse & Pleijel (2002), Zhang & Hutchings (2019) and Hutchings et al. (2020), and some relevant characters of the European species are reproduced below. The anterior-dorsal part of the head (operculum) is marked by two bundles of large golden paleae that extend forward. The dorsal edge of the operculum, the opercular rim, can be either serrated (Amphictene) or smooth (other genera). The thoracic region bears two pairs of branchiae and three segments with notopodial hairs only. The abdominal region has 12 or more segments with both notopodial hairs and transverse rows of neuropodial uncini (Fig. 2). The total number of segments with notochaetae and neurochaetae has been used by Malmgren (1866) as a generic character (Table 1), but with the description of more recent species this number proved to vary within a genus and to be stable within a species. At least in Dutch specimens of A. auricoma and L. koreni, the number of segments with notochaetae and neurochaetae remains the same as individuals of the species grow (T. van Haaren, unpublished observations). The scaphe, the posterior part of the body, is usually well demarcated from the abdomen, except in Petta. The scaphe segments are achaetous, though there may be up to 20 or more pairs of stout spine-like chaetae, called scaphal hooks, around the area where the scaphe joins the abdomen. In most 19th-century taxonomic papers, only the number of paleae, notochaetae and neurochaetae, the shape of the opercular rim (smooth or serrated) or cephalic veil (with or without cirri) are reported (e.g. Savigny 1822; Grube 1851; Johnston 1865; Levinsen 1884). Sometimes the shape of the sand tube is also reported (curved or straight) (Savigny 1822; Johnston 1865; Malmgren 1866 and others), but in the author's opinion it has no diagnostic value. The shape of the tube depends on the size, and often the thinner, fragile end section breaks off, “changing” the shape from a curved to a straight tube. Other research indicates that one has to pay attention to several other characteristics, such as the number of vertical rows of major teeth on the uncinus (e.g. Annenkova 1929; Fauvel 1936; McIntosh 1922; Hutchings & Peart 2002), which was mostly neglected in the 18th and 19th century papers. Nowadays, it is used in the generic diagnosis and regarded as a diagnostic feature to distinguish Cistenides from Pectinaria (Fauchald 1977; Hartmann-Schroder 1996; Hutchings & Peart 2002; Hutchings et al. 2020).

Figure 1

Figure 2

Peter Simon Pallas (1741–1811) was born in Berlin and studied at the universities of Berlin, Halle, Göttingen, Leiden and London. From 1762 to 1766 he stayed in ‘s-Gravenhage (The Hague), after which he completed Miscellanea zoologica (1766) where he described several vertebrate and invertebrate species new to science, including Nereis cylindraria var. belgica. Several of his works, including Miscellanea zoologica, have been translated by P. Boddaert from Latin into Dutch (Boddaert 1770). In 1767, Pallas became a professor at the St. Petersburg Academy of Science (Russian Kunstkamera at the Academy of Sciences), where he led several expeditions in Russia. He finished his work in Crimea and moved to Berlin, where he died in 1811. His material on Nereis cylindraria var. belgica is untraceable, or at least not to be found in the St. Petersburg Museum (N. Slepkova, personal communications) or the Naturalis Biodiversity Center (Leiden).

In Miscellanea zoologica, Pallas describes two varieties of Nereis cylindraria, N. cylindraria var. capensis (from the Cape of Good Hope, South Africa) and N. cylindraria var. belgica, based on specimens washed up on Dutch beaches. As Pallas was staying in The Hague at that time, his material of var. belgica was most likely from the beaches of Scheveningen. In that period, there was only one main road from Den Haag to Scheveningen (Scheveningse weg), leading to the Oude Kerk (English: old church), so it can be assumed that his material originates from the beach near the Oude Kerk, Scheveningen (52°6′28″N; 4°16′12″E).

Both varieties can be attributed to Pectinariidae. N. cylindraria var. capensis (as “Kaapsche Verscheidenheid”, the corresponding Dutch vernacular name of this species in Boddaert 1770) differs from N. cylindraria var. belgica (“Hollandsche verscheidenheid” in Boddaert 1770) in a crenulated dorsal edge behind the head, while in belgica this edge is smooth (Pallas 1766; Boddaert 1770). Furthermore, it appears from the original description of N. cylindraria var. capensis that N. cylindraria var. capensis has 13 neurochaetae. In the following description of N. cylindraria var. belgica, Pallas (1766) writes that this species is the same as the previous one, except for the size (three times smaller; the tube is up to 2.5 duim in N. cylindraria var. belgica and up to 5 duim in N. cylindraria var. capensis; 1 duim = 1 inch), the dorsal edge (crenulated in N. cylindraria var. capensis and smooth in N. cylindraria var. belgica), and the scaphe (long in N. cylindraria var. capensis and short in N. cylindraria var. belgica). Several other details can be derived from the plates. It is clear from Figure 5, Plate IX in Pallas (or Plate IV in Boddaert 1770) that var. belgica has 14 or 15 notochaetal bundles posterior to the gills (Fig. 3). The number of neurochaetae is impossible to determine and one can only guess whether Pallas ever saw a difference in the number of neurochaetae between N. cylindraria var. capensis and N. cylindraria var. belgica. It cannot be inferred from that paper whether var. belgica also has 13 neurochaetae, as does var. capensis.

Figure 3

There is also the question of the etymology of the species epithet “belgica”. The Netherlands was formerly divided into two regions, the northern region and the southern region. The northern region, known as the “united provinces of the Netherlands” or in Latin “Belgium Foederatum”, was a federal republic from 1588 until the Batavian revolution in 1795. It largely covered the territory of the present-day Netherlands. The southern region covered the territory of today's Belgium and was occupied by Spain (1585–1713) and Austria (1714–1795). The latter period is also known as the “Austrian Netherlands” or in Latin “Belgium austriacum”. Thus, when Pallas stayed in The Hague (1762–1766), he actually was in the northern region or Belgium Foederatum, which explains the name belgica”. Papers from before 1795 in which Belgium or Belgii is reported refer either to the present-day Netherlands or to Belgium, e.g. Linné (1788, p. 3749) says “habitat ad Belgii littora”. Later reports of P. belgica from Belgium (e.g. Southern 1910, 1914; McIntosh 1922) are either due to this geographical misunderstanding of the name Belgium/Belgii, or to the assumption that Malmgren described the same species as that described by Pallas.

In 1767, Linné describes Sabella granulata, but his description of the species [“Lives in a curved dark colored sandy tube (…) in deeper parts of the North Atlantic Ocean”] is very scanty.

Houttuyn (1771) provides a Dutch translation of Linnaeus’ S. granulata (species 809 on pages 611–612) saying that the tube is composed of brown sandy grains and has dark color. Houttuyn (1771) also writes that such tubes of sand and shell debris can also be found on Dutch beaches. According to the current state of knowledge, however, it appears that these tubes, containing shell remains, may represent tubes of Lanice conchilega (Pallas, 1766), another common species on the Dutch coast. Furthermore, species with features of the current concept of C. granulata have never been found in the (Dutch) North Sea. It is therefore most likely a common mistake in 18th and 19th century papers to name each species with a sandy tube S. granulata, P. belgica, or perhaps L. conchilega. Malmgren (1866) wrote a more detailed description, also adding that the paleae of P. granulata lack a thin tip (“apice haud tenui attenuatto”), a character now considered diagnostic.

The description of Amphitrite auricoma in Müller (1776) is very scanty. He describes the species as having golden fans of stiff bristles on both sides of the head and living in fragile sand tubes. However, in 1779 and 1788, Müller gave a more detailed description and a very nice color plate (Plate XXVI, Fig. 5) clearly showing the serrated edge of the opercular rim, a characteristic feature of this species at that time. Nowadays, this serrated edge is regarded as a generic feature of the genus Amphictene.

It is also interesting that Müller creates a new name for Sabella granulata L., as he lists it under the synonymy of A. auricoma. Since Fabricius was Müller's student and they were in contact, Müller published some data from Fabricius before the famous book of this author on the Greenland fauna was published (Fabricius 1780). It seems that Fabricius did indeed find the species in Greenland, and it is possible that Müller is referring to the findings of Fabricius in Greenland. However, Müller (1766) does not make any direct reference to the work of Fabricius, except maybe the use of the Greenlandic (vernacular) name (Imab Polia). As Fabricius was a linguist, he used to name each species with a Greenlandic vernacular name as well, and most likely Müller copied that in his 1776 book. The extensive description by Fabricius (1780) in his book “Fauna groenlandica” on A. auricoma is therefore interesting. Fabricius writes “Segmenta 3 sequentia utrinque papillo conico setis paucis aureis armato, sursum retroque tendente; 12 sequentia similia, fed fetis pluribus et longioribus (…)”; translated by the author as: “3 segments papillose with golden hairs, 12 others more or less similar, but with more and longer hairs”. The presence of 15 pairs of notochaetae suggests that he actually referred to L. koreni, the only NW European species with this character. Grube (1871) suggests that Fabricius’ A. auricoma is synonymous with P. granulata.

In 1866, a revision on Scandinavian polychaetes was published by Anders Johan Malmgren. In his chapter on the family Amphictenea, he gives a redescription of P. belgica (Pallas) and A. auricoma (Müller), describes the genus Petta and the type species Petta pusilla, the genus Lagis with the type species L. koreni and the genus Cistenides with the type species Sabella granulata L. and the new species C. hyperborea. The diagnostic features of these species are presented in Table 1.

These features of the European (read: Scandinavian) species were generally followed by subsequent authors, but the number of paleae is more variable. Malmgren's material of P. belgica from Bohuslän (Swedish west coast along the Kattegat) is still available for study and has been assigned as the neotype of P. belgica (Pallas) by Nielsen et al. (1977). It is present in the Swedish Museum of Natural History, Invertebrate Zoology, type no. 3138. The nomenclatural problem of these European Pectinariidae, and especially L. koreni vs P. belgica, is extremely complex and results from the accumulation of successive errors and misinterpretations. The main problem was that Malmgren assigned the name belgica to an actually new species, and created a new species (L. koreni) for what actually was belgica Pallas. Morphology, ecology and distribution data clearly show that Nereis cylindraria var. belgica Pallas must be identical to L. koreni Malmgren. It is most likely that Pallas was collecting Pectinariidae specimens with 15 segments with notochaetae and 12 segments with neurochaetae, as this is the only species on the west coast of the Netherlands (see the next paragraph). On the other hand, faunistic changes over the last 250 years cannot be entirely ruled out. Consequently, Malmgren's material of P. belgica (17 notochaetae, 13 neurochaetae) cannot be identical with Pallas’ material of Nereis cylindraria var. belgica. This matter was addressed by Lucas & Holthuis (1975). However, returning to the correct meaning of belgica would mean that many contemporary papers would have to change the meaning and create a third name, while suppressing koreni (as a junior synonym of belgica). To avoid this inconvenience, the committee decided, at the request of Nielsen et al. (1977), to continue using the names sensu Malmgren. This meant announcing the neotype belgica. There were objections to codify the misidentification (Holthuis 1978), but the request was supported by a large number of marine biologists (Nielsen & Kirkegaard 1978). The final decision was issued by the ICZN in 1982 (Opinion 1225), putting an end to the discussion. The committee ruled that Malmgren's species of Pectinaria belgica Pallas from Bohuslän was to be regarded as a valid type, with P. belgica retaining the original author – Pallas. However, it was overlooked at that time that a candidate name for the new species – misidentified by Malmgren as P. belgica – was actually available. French Professor Giard (1899, 1913) attempted to solve the problem earlier by introducing Pectinaria septentrionalis as an alternative to Malmgren's P. belgica and preserving the name P. belgica as originally intended by Pallas. Since then the name has been overlooked (or neglected), even by Lucas & Holthuis (1975), and consequently also by the committee. Giard (1899: p. 489) and Giard (1913: p. 122) writes (translated by the author from French): “I have preserved the name Pectinaria belgica assigned by Bouchard-Chantereaux for an Amphictenien so common on our sandy beaches. It is indeed the species that Pallas had before his eyes when describing Nereis cylindrica var. belgica. It is the same species that Malmgren has since called Lagis koreni, reserving the name of Pectinaria belgica Pallas for a species of the Arctic seas whose presence on the coasts of France and Belgium seems to us very doubtful. To avoid any confusion, it would be advisable to restore the name of Pectinaria belgica Pallas to the species from the coasts of Holland, Belgium and France and to give the name Pectinaria septentrionalis to the species from Norwège, which does not penetrate the Baltic”. This available name P. septentrionalis Giard, 1899 does not change the validity of both L. koreni and P. belgica. The committee ruled to maintain the prevailing use of L. koreni Malmgren, 1866 (type locality Finnmark, Norway) and P. belgica (Pallas, 1766) (type locality Bohuslän, Sweden) as valid species.

There are numerous reports on Pectinariidae washed up on Dutch shores, especially after SE storms and in spring. They mostly report empty tubes, which cannot be linked to any Pectinariid species, but sometimes they provide some information on the worm itself.

17th century: Dutch biologist Jan Swammerdam (1637–1680) worked until his death on his major work, published posthumously in 1737–1738 by Leiden University Professor Herman Boerhaave as Bijbel der natuure. In Volume 2 (1738), Jan Swammerdam describes a trumpet worm (as ‘pijpken’) from material he found on the beaches of the North Sea. This means that Pectinariidae (in this case most likely L. koreni) were already washed up on Dutch beaches in the second half of the 17th century. The material he found, although circumstantial, probably originated from a beach in Zuid-Holland, as he probably made his observations while studying at Leiden University (1661–1667).

1762: In Natuurkundige Uitspanningen, Baster writes about Amphitrite (translated from Dutch by the author): “When walking along the beaches on the isle of Schouwen in February and March, one may notice a number of small ‘blisters’ filled with pink or reddish-brown eggs. From these eggs ‘zand-aaltjes’ hatch, Amphitrite, which hide in the sand immediately after hatching” (Baster 1762, p. 44). It is impossible to tell from his figure (Table V, Figure 1) what species this might be.

1766: Pallas describes Nereis cylindraria var. belgica, which he collected on the beach at Scheveningen (see above).

1826: Bennet & van Olivier (1826) report three species of tube-dwelling worms from the Dutch beaches, Amphitrite auricoma (p. 81), Sabella chrysodon (p. 159) and Sabella belgica (p. 160). The current status of the A. auricoma specimen they reported is not certain, but it is definitely a Pectinariid [translated from Dutch by the author: “slanting head, with 4 palps and with 2 transversely arranged, comb-shaped and gold-colored ‘leafs’ that glisten very strongly. Body tapering and a row of brush-shaped nipples on both sides (…) second and third segment with four comb-shaped gills. Length 2–2.5 inches. Tube of fine sand.”]. Since their A. auricoma is reported from Dutch beaches, this may be the current concept of L. koreni. Their S. chrysodon belongs to the current concept of Lanice conchilega (Pallas). Nereis conchilega is on their list of synonyms and they describe the typical habitus of L. conchilega (e.g. tube of sand and shell debris, flexible). Their Sabella belgica is definitely the current concept of L. koreni, as they describe it as having a sandy tube and 12 feet. They also write (translated from Dutch by the author): “They live along the Dutch western beaches along the North Sea, especially at Scheveningen, were they can be found among shrimps”. The last sentence, at least, is noteworthy, as this is exactly what Pallas wrote – “(…) among shrimps”. It is worth noting that Bennet & van Olivier (1826: p. 160) gave Sabella belgica the vernacular name “Belgische zandkoker”, while the species does not even occur in Belgium. Again a misinterpretation of the species epithet belgica.

1851: Maitland (1851: p. 204) reports Amphitrite auricoma from the Netherlands with a long list of synonyms, including Sabella granulata L. and Nereis cylindraria Pallas, and he also refers to the record by Bennet & van Olivier (1826). The species is reported from Schouwen (Baster) and Scheveningen (Bennet & Olivier). Baster's record from Schouwen is probably not a Pectinariid (see above) and the latter record may still refer to the original Pallas record. Maitland also writes (translated from Dutch by the author): “sometimes in spring, the beaches of our coasts are covered with tubes of these animals”. From his very short Latin description it is impossible to tell which Pectinariid he actually described.

1866: In a note on Dutch caddisflies (Fam. Phryganeidae), van Hasselt (1866) compares the beautiful cases with sandy tubes of saltwater annelids he often found on Dutch beaches. He liked the paleae (“gulden lokken”) of our “Amphitrite Belgica (Amphictena, ook Pectinaria, auricoma)” (Hasselt, 1866, p. 211). This quotation shows that Hasselt (1866) considered Amphitrite belgica and Pectinaria auricoma to be the same species.

1898: In Horst's checklist of Dutch Polychaetes, two species are reported on pages 26–27, Pectinaria belgica Pallas and Pectinaria auricoma Müll. He writes about P. belgica (translated from Dutch by the author): “The ‘belgische zandkoker’ is very common along our coast and sometimes can be found in large numbers on our beaches”. However, in his chapter on Pectinaria auricoma he writes that “Hornell (Report upon the fauna of Liverpool bay) claims that Malmgren's description of the species having 17 bundles of hairs and 13 comb-brushes is false, and that there are only 15 pairs of hair bundles and 12 pairs of the latter are present, like in P. belgica. This is, however, not the case. P. auricoma and belgica have 17 feet, but in the latter species, the two last segments are without bristles, unlike auricoma”. This is interesting as we may conclude from this that Horst's and Hornell's belgica are in fact koreni. Horst's P. belgica can be assigned to L. koreni for he describes it as having 15 pairs of notochaetae and 12 neurochaetae. His P. auricoma has an extra pair of neurochaetae, so this one probably belongs to A. auricoma (see also Horst 1922). P. belgica is reported to be common along the Dutch coast and can sometimes be found in large numbers on the beach. P. auricoma is rare and was only found in dredge samples from the Zuiderzee at Wieringen.

1913: In a popular book on beach species by Heinsius & Jaspers (1913, p. 7 & Figure 1F), Pectinaria belgica is reported as very common, but from the drawings made by these authors it is impossible to tell which species of Pectinariidae they actually described. They even made an erroneous observation by calling the paleae “comb-shaped gills”.

1922: In the chapter on polychaetes in the Zuiderzee (Horst 1922, p. 274), Pectinaria auricoma is reported from Wieringen again. In this paragraph, it is very clear that the specimens described are indeed A. auricoma (translated from Dutch by the author): “This Pectinaria can be distinguished at first sight from Pectinaria belgica, common on our coast, by the slightly bent tube that it inhabits, which in the latter species is entirely straight. Furthermore, in Pectinaria auricoma, the opercular rim behind the paleae is equipped with fringe-like appendages, while it is smooth in Pectinaria belgica (Fig. 17a). There are 17 hair bundles on each side of the body, while the number of rows of combs is 13. The combs (Fig. 17b) are pedunculate, furnished with five to six larger teeth along the free edge and a group of finer teeth and an incision below that. At the base of the tail lobe (scaphe), which ends in a thin tip, there is a row of six anal hooks”. He clearly describes the current concept of A. auricoma and confirms the current knowledge about its distribution in the (northern) Dutch North Sea. It was already clear from Horst's earlier publication (Horst 1898) that his Dutch material of belgica is the same as the current concept of L. koreni, with 15 notochaetae and 12 neurochaetae.

1939: Hana (1949, pp. 81–84) describes all that is washed up on the beach after a storm (from the northwest) and spring tide. Based on his story (numerous straight tubes with worms having golden fans of stiff corneous bristles) and a photograph, there is no clear species description to determine which of the Pectinariidae species he may be referring to. An exact copy of his photograph from Hana (1949) is available in the “Nederlands Fotomuseum”. According to the available information, this is the original photograph by Kees Hana taken in April 1939 on the beach at Scheveningen.

1951: Korringa (1951) reports one juvenile specimen of Pectinaria koreni from his culture of experimental oysters in the Eastern Scheldt on 25 June 1941 with an additional comment that “it apparently does not like that habitat”.

1942–1968: In the Dutch magazine “Het Zeepaard”, there are paragraphs entitled “C.S. feuilleton” where enthusiastic biologists could report on interesting marine species they found in the Netherlands. Pectinariidae (or in Dutch “Goudkammetjes”) are reported in large numbers from Dutch beaches. Mass strandings of “Goudkammetjes” were reported on 15 February 1942 (Scheveningen, Katwijk: Bremer 1942), 11 April 1947 (Scheveningen: Leenhouts 1947a), September 1947 (Katwijk-terheijde: Leenhouts 1947b), 9 October 1955 (Toorne, De Beer, Hoek van Holland, Scheveningen, Wassenaarse slag, Katwijk, Bloemendaal, IJmuiden: Swennen 1955), 14 January 1958 (Hoek van Holland : Wolff 1966), 6 February 1960 (Katwijk-Wassenaar: Spaink 1960), 17 November 1963 (Tholen: Compaan 1965), February–March 1965 (Scheveningen, Terheide: Compaan 1965) and 20 January 1968 (between Noordwijk and Noordwijkerhout: Heerebout 1968). These mass strandings are either reported as Pectinaria sp., Pectinaria belgica, Pectinaria koreni or Pectinaria auricoma. This seems strange as Bloklander (1947) had already written that specimens from Dutch beaches were in fact Pectinaria koreni, not Pectinaria belgica. For some reason, however, most reporters neglected this information until Wolff (1970) clarified the issue (again). For example, Wolff (1970) writes (translated from Dutch by the author): “If one now tries to name the ‘Goudkammetjes’ from the Dutch beach using the most common identification keys for polychaetes (Fauvel, 1927; Friedrich, 1940), one always arrives at one and the same species: Pectinaria koreni Malmgren 1865. Also all material from Zeeland and from the North Sea off the coast at the Delta area belongs, as far as I have seen, to this species”. As Fauvel (1927) and Friedrich (1938, Wolff incorrectly indicated 1940) are correct according to the current knowledge about European Pectinariidae, one may conclude that all coastal species from the SW region are indeed L. koreni. This is confirmed by more recent findings (this paper).

1957: In his popular book on what can be found on Dutch beaches, Prud’homme van Reine (1957, p. 74) writes about two species of “Goudkammetjes” Pectinaria auricoma (Müller) and Pectinaria koreni (Malmgr.). He gives a very brief description of both species and depicts them on Plate XXII, Figure 7. It is not possible to tell from the descriptions which species is involved (about 15 tentacles on both sides of the mouth in P. auricoma and about 20 in P. koreni; the latter also being slightly larger). It is clear from aforementioned Figure 7 that his P. auricoma has a crenulated dorsal edge, while P. koreni has a smooth one. Apparently Prud’homme van Reine (1957) caught or expected both species on the Dutch beaches, but did not provide any data.

1973: Entrop (1973) writes about Pectinaria koreni washed up on the Dutch beaches. His short description (15 segments, 12 with hook shaped chaetae) leaves no doubt as to their identity: L. koreni.

1999: Prins (1999) reports on Pectinariidae (with original drawings) living in shallow waters along the Dutch coast, washed up on beaches after storms. According to him, the specimens had 15 fleshly protuberances, each ending in a brush of hairs. The original drawings show a smooth opercular rim, 15 notochaetae and a straight tube. It is impossible to determine as to how many neurochaetae they have, but they are also clearly L. koreni.

2002/2019: In two guides on marine species, Leewis (2002) and de Bruyne & Gmelig Meyling (2019) report L. koreni from the beaches along the Dutch coast. Since they both mention 15 segments, they clearly refer to L. koreni.

Based on the material from the Dutch North Sea, especially the descriptions by Malmgren (1866; see also Table 1), which is widely accepted as the correct diagnosis of the European species, the following list of Dutch species of Pectinariidae can be presented, together with a list of synonyms. The list is strictly based on morphological descriptions from the historical papers. C. granulata L. is not a Dutch species. It is an amphiboreal-Arctic species (Holthe 1977) and should not to be expected in the Dutch North Sea. However, due to misinterpretations of the original and subsequent synonymization with other species of Pectinariidae, this species is also included.

At least Malmgren (1866), Levinsen (1884), Meyer (1912), Wollebaek (1912), Hessle (1917), McIntosh (1922), Fauvel (1927), Friedrich (1938), Holthe (1986), Hartmann-Schröder (1996), Kirkegaard (1996), Böggemann (1997) and Jirkov (2001) gave the correct species diagnosis for the NW European species. However, the generic characters need to be revised. Hutchings & Peart (2002, p. 109 + Table 6) are incorrect in claiming that 16 notochaetae is a generic character of the genus Lagis. This would be impossible as the type species L. koreni Malmgren, 1866 (by monotypy) has only 15 notochaetae. It is therefore worth noting that Icelandic specimens of L. koreni (Parapar et al. 2020) have 16 notochaetal bundles, which may suggest another Lagis species. Furthermore, Zhang & Hutchings (2019) indicated in their generic diagnosis of Lagis that these species should have 13 pairs of neurochaetae, but again Lagis koreni (the type species) have only 12 of them. On the other hand, they also write that the neuropodia are present on segments 8–19, i.e. on 12 segments. Furthermore, Fauchald (1977) and Hutchings & Peart (2002) are incorrect about Cistenides. At least C. hyperborea and C. granulata are species that have 12 neurochaetae, while the aforementioned papers report that they should have 13 neurochaetae. The number of neurochaetae in C. granulata is not reported in the original paper by Linné (1767), but Malmgren (1866) clearly writes that both species of Cistenides have 12 neurochaetae. Thus, it is unclear why Fauchald (1977) considered 13 neurochaetae as a generic character of Cistenides. Also the generic characters of Pectinaria (including belgica) and Petta in Hutchings & Peart (2002) are incorrect, as P. belgica does not have 14 but 13 neurochaetae, and P. pusilla does not have 18 but 17 notochaetae (see Malmgren, 1866). Zhang et al. (2019) corrected the number of notochaetae for Petta to 17. The number of vertical rows of the major teeth on the neuropodial uncini is reportedly of diagnostic value for distinguishing between Cistenides (1 row) and Pectinaria (at least 2 rows; Hutchings & Peart 2002, Hutchings et al. 2020). According to Annenkova (1929), however, C. hyperborea has 2–3 vertical rows, which places this species within Pectinaria. Thus, the number of vertical rows is not used in the key.

The six (possible) North Sea species can be identified using the information from Table 1 and the following simplified key:

Cephalic veil smooth, Scaphe indistinctly separated from posterior segments, 14 neurochaetae............................................................................................ Petta pusilla

- Cephalic veil cirrated, scaphe distinctly separated from posterior segments, 12–13 neurochaetae........2

Distribution in the Netherlands (Fig. 6)

Although Pallas’ Nereis cylindraria var. belgica is in fact morphologically the same as Malmgren's Lagis koreni, the neotype of P. belgica (Pallas) from Bohuslän, Sweden is designated as the correct type. Consequently, and ironically, P. belgica (Pallas) is a new species for the Netherlands.

Only recently a single, fully grown specimen of P. belgica (Pallas) sensu Malmgren (non Pallas) was found in a sample collected at the Dutch Oyster Grounds (9 April 2018; Figs 4–5). Another single, somewhat smaller, specimen was collected in the last week of October 2019 at the Oyster Grounds during a trawl survey by NIOZ, Texel. The exact location and date is not known, but it was collected at three possible locations (see the material section; L. Kleine Schaars, NIOZ, pers. comm.).

Figure 4

Figure 5

Figure 6

Figure 7

There is no evidence that P. belgica (Pallas) sensu Malmgren was ever found in the Netherlands prior to this study. It seems likely that the northern part of the Dutch North Sea (north of the Frisian Front) may be the southernmost limit of this species. It is therefore unlikely that P. belgica will ever be found washed up on the Dutch coast, also due to the southern current.

Distribution in the Netherlands (Fig. 8)

Lagis koreni is widely distributed in the Dutch North Sea. It is very common in the coastal zone and has scattered records north of the Frisian Front. It is rare in the Wadden Sea, the delta area and offshore locations, and has only one record in the North Sea Canal.

Figure 8