Rostrophontodes, Lee & Huys, 2019

Lee, Jimin & Huys, Rony, 2019, New Ancorabolidae (Copepoda: Harpacticoida) from Korea: two new genera, revisionary notes and updated identification keys, Zoological Journal of the Linnean Society 187, pp. 331-377 : 363-367

GENUS ROSTROPHONTODES GEN. NOV.

u r n: l s i d: z o o b a n k. o r g: a c t: A 7 B E F C 7 B - 3430-4470-AB6B-F17E7CA1F89A

Diagnosis: Laophontodinae . Body subcylindrical; body somites strongly chitinized, separated from each other by pronounced constrictions. Cephalothorax wider than remainder of prosome; with lobate extensions bilaterally. Cephalic shield and somites bearing P2–P6 (genital half of double-somite in ♀), with reticulated integument dorsally and laterally. Dorsal surface of P5-bearing somite, genital double-somite (anteriorly and posteriorly) in ♀ (or genital somite and first abdominal somite in ♂) and second abdominal somite with paired, backwardly directed, spinous projections around posterior margin. Anal somite with strongly protruding, round anal operculum. Caudal rami much longer than wide (about six times in the type species), slightly constricted in the middle; with seven setae; setae I and II originating from proximal third of outer margin, setae III – VII positioned around or near posterior margin .

Sexual dimorphism in antennule, P3 endopod, P5, P6 and segmentation of urosome.

Rostrum very large, about two-thirds of the length of cephalic shield, sharply pointed towards apex, and markedly recurved ventrally. Antennule five-segmented in ♀, with aesthetascs on segments 3 and 5; six-segmented and chirocer in ♂, with one segment distal to geniculation and aesthetascs on segments 5 and 6; posterior margin of segments 2 and 3 reportedly without spinules. Antenna with incompletely fused allobasis lacking abexopodal seta; exopod represented by minute segment with one seta. Mandibular palp one-segmented, with one outer, two apical and two inner setae. Maxilliped subchelate; syncoxa with one pinnate seta; endopod drawn out into curved claw with long pinnules in distal half, bearing one accessory seta at its base.

P1 with well-developed praecoxa and coxa; basis transversely prolonged in distal half, causing rami to be widely separated by its concave distal margin; outer and inner basal setae long and bipinnate. Exopod three-segmented; exp-1 with unipinnate spine bearing long pinnules; exp-2 with one outer, geniculate seta; exp-3 with four geniculate setae. Endopod two-segmented and prehensile; enp-1 very long, more than three times as long as exopod, with setules/spinules on both inner and outer margins; enp-2 short, with one small inner seta and an elongate slender claw apically, being almost as long as adjacent geniculate seta.

P2–P4 with well-developed praecoxae and coxae; bases transversely elongate, becoming progressively longer from P2 to P4. Exopods three-segmented; exp-1 without inner seta, exp-2 with inner seta, exp-3 with one inner seta, two apical setae and three outer spines; outer spines of all segments long and bipinnate. Endopods two-segmented in ♀; enp-1 minute, without armature; enp-2 elongate (becoming progressively shorter from P2 to P4), with one extremely long seta apically. P3 endopod three-segmented in ♂; enp-1 minute and unarmed; enp-2 with short apophysis; enp-3 very short, not extending beyond apophysis, with one very long apical seta.

P5 one-segmented in both sexes; outer basal seta arising from long, demarcated setophore; exopod represented by one outer and three apical elements; endopodal setae absent. Female genital field positioned anteriorly, with median copulatory pore; P6 represented by small cuticular plates, each with one minute seta. Male P6 unconfirmed.

Type and only species: Laophontodes gracilipes Lang, 1936b = Rostrophontodes gracilipes ( Lang, 1936b) , comb. nov.

Etymology: The generic name is derived from the Latin rostrum, meaning beak, and the suffix -phontodes, commonly used in the formation of generic names in subfamily Laophontodinae . It refers to the excessively large curved rostrum. Gender: feminine.

Remarks: According to Kornev & Chertoprud’s (2008) illustrations of the female and male antennules (their fig. 5.192B, И) segments 2 and 3 do not exhibit any spinules along their respective posterior margins. This observation requires confirmation, because the presence of spinules, usually on a rounded, bumplike expansion, on segment 2 is considered a potential synapomorphy for the Laophontodinae ( Gheerardyn & George, 2010; Gheerardyn & Lee, 2012; George, 2017).

Four character states lend support to a potential sister-group relationship between Rostrophontodes and Ancorabolina :

(1) Maxillipedal claw with ornamentation along distal half of inner margin. All known species of Ancorabolina display small spinules on the endopodal claw (e.g. Gheerardyn & George, 2010: figs 3F, 9D, 18C, 27D). In Rostrophontodes , the positional homologues of these ornamentation elements are secondarily enlarged, forming long pinnules along the distal inner margin. Spinular ornamentation has been reported in a few other members of the Laophontodinae (see above), but its presence does not appear to be strictly diagnostic for these genera (i.e. Calypsophontodes , Laophontodes and Tapholaophontodes ).

(2) P1 basis transversely elongate. In both Rostrophontodes and Ancorabolina , the distal half of the basis is bilaterally expanded, being at least twice as wide as the proximal margin and forming a distinct cylindrical pedestal for the endopod and a similar expansion for the exopod. The origins of both rami are typically widely separated by the concave distal margin of the basis. Some species show secondary modifications such as in Ancorabolina cavernicola George & Tiltack, 2009 , where the endopodal pedestal has undergone extreme elongation along the dorsoventral axis, reaching to about the distal end of the exopod (cf. George & Tiltack, 2009: fig. 6A). In Ancorabolina galeata Gheerardyn & George, 2010 , the outer distal portion of the basis shows a similar elongation along the transverse axis (cf. Gheerardyn & George, 2010: fig. 18A), causing a disproportionate displacement of the exopod. Transverse elongation of the basis has not been reported in other genera of the Laophontodinae .

(3) Claw of P1 enp-2 replaced by a slender element. In R. gracilipes , it is represented by a curved spine being almost as long as the adjacent geniculate seta. In Ancorabolina spp. , the distal endopod segment typically has one anterior claw-like seta and one posterior geniculate seta, with the former being slightly shorter than the latter (e.g. Gheerardyn & George, 2010: figs 4A, 9E). Although this condition is regarded here as the ancestral pattern in the genus, some secondary deviations have been observed in certain species, such as the presence of two non-geniculate setae in Ancorabolina chimaerae George, 2016 c, two geniculate setae in Ancorabolina divasecunda Gheerardyn & George, 2010 (cf. George, 2006c: fig. 7A; Gheerardyn & George, 2010: fig. 28A) and extreme reduction of the posterior geniculate seta in A. cavernicola (cf. George & Tiltack, 2009: fig. 6A’). Except for La. whitsoni , which shares the same condition with R. gracilipes (cf. George & Gheerardyn, 2015: fig. 21D), all other members of the Laophontodinae display a robust claw, being invariably and distinctly shorter than the geniculate seta.

(4) Body somites with dorsal spinous processes. In R. gracilipes , the dorsal surface of the P5-bearing somite, genital double-somite (both genital and first abdominal somite) and first free (second) abdominal somite exhibit paired, backwardly directed, spinous projections. Members of Ancorabolina have a propensity to develop similar tooth-like processes, but their distribution pattern differs between species (ranging from the cephalothorax to the penultimate somite). Hence, the significance of this character must be treated with caution when used in a phylogenetic context.

LAOPHONTODES MULTISPINATUS KORNEV & CHERTOPRUD, 2008

Conroy-Dalton (2004) proposed the genus Lobopleura to accommodate a new species Lobopleura ambiducti Conroy-Dalton, 2004 and Laophontodes expansus , and provided compelling evidence for its sister-group relationship with Probosciphontodes . Kornev & Chertoprud (2008) described Laophontodes multispinatus from specimens collected at 10 m depth in the Kandalaksha Gulf of the White Sea. Apparently unaware of Conroy-Dalton’s (2004) work, they considered their new species to be most closely related to La. expansus based on the shared similarity in body shape and ornamentation and in P1 morphology. They separated both species on the degree of dorsoventral flattening of the body, the relative length of the first antennulary segment in the female, the shape of the outer element of P1 exp-2, the morphology of the P2 endopod and the elongation of the bases of P2–P4.

Gheerardyn & Lee (2012: 263) had previously announced the imminent removal of La. multispinatus to Lobopleura . It is here formally transferred to this genus as Lobopleura multispinata ( Kornev & Chertoprud, 2008) comb. nov. based on the following combination of characters: dorsoventrally depressed body shape; somites with laterally produced lobate processes; wide, bell-shaped cephalothorax; rostral shape and morphology of antennule, antenna, mouthparts and P1–P5. It also shares three of the four autapomorphies of the genus (cf. Conroy-Dalton, 2004: 35): (1) paired genital system in both sexes (Kornev & Chertoprud do not discuss the condition of the male reproductive system, but their fig. 5.194a clearly shows the presence of paired spermatophores); (2) P4 endopod with only one apical seta; and (3) P5 baseoendopodal armature reduced to one vestigial seta. The fourth autapomorphy identified by Conroy-Dalton (2004) (i.e. male P3 endopod-2 not secondarily subdivided) cannot be confirmed for Lo. multispinata because according to Kornev & Chertoprud (2008) no sexual dimorphism is displayed on the swimming legs. This is likely to be based on an observational error, because all members of the Laophontodinae exhibit a sexually dimorphic P3 endopod (except for the two species of Probosciphontodes , which lack this ramus in both sexes).

Lobopleura multispinata is morphologically very close to the type species Lo. ambiducti . It is maintained as a distinct species based on the following characters: (1) the shape of the cephalothorax in both sexes, being broader and tapering more gradually towards the anterior end in Lo. ambiducti (particularly in the female); (2) shape of the rostrum (broader at base and with straight lateral margins in Lo. multispinata , with concave lateral margins in Lo. ambiducti ); (3) lateral spinulate processes on somites bearing P2–P5 and genital (double-)somite distinctly more produced in Lo. ambiducti ; (4) body ornamentation consisting of numerous spinules in Lo. multispinata vs. smooth in Lo. ambiducti ; (5) caudal rami distinctly longer in the male in Lo. ambiducti ; (6) second antennulary segment in female relatively shorter in Lo. multispinata ; (7) bases of P2–P4 more transversely elongate in Lo. ambiducti ; (8) P2 endopod represented by relatively larger unarmed protuberance in Lo. multispinata ; (9) P3 enp-2 relatively longer in Lo. multispinata and its inner distal seta being plumose and about as long as the outer distal one vs. naked and less than half the length of the outer distal seta; and (10) P4 enp-2 slightly longer in Lo. multispinata , but its apical seta distinctly shorter than in Lo. ambiducti . Other potential differences, such as the armature of the mandibular palp (six setae vs. five in Lo. multispinata ) and the number of accessory setae on P1 enp-2 (one long + two minute in Lo. ambiducti vs. one long + one minute in Lo. multispinata ), require confirmation. Differences based on observational errors in Kornev & Chertoprud’s (2008) description include the alleged absence of: (1) the vestigial antennary exopod; (2) the inner basal seta on P1; (3) the outer basal seta on P4; and (4) sexual dimorphism on the P3 endopod (no apophysis in male).

The genus Lobopleura assumes, so far, a northern Atlantic–Arctic distribution. The type species Lo. ambiducti is known from the Dalkey region (Co. Dublin) in Ireland ( Roe, 1958: as Laophontodes sp. ), the Gullmar Fjord in Sweden ( Lang, 1948: as Laophontodes expansus ) and the Isle of Iona in Scotland ( Conroy-Dalton, 2004). Lobopleura expansa was originally described from the Saltenfjord (Skjaerstadfjord) in Norway ( Sars, 1908). Steinarsdóttir & Ingólfsson (2004) reported it (as Lo. expansus ) from the eulittoral Ascophyllum nodosum (L.) Le Jol. zone in Hvassahraun in southwestern Iceland. Willems et al. (2009) collected the species in the Singlefjord, Sandviken along the Swedish west coast (Koster area), and Gheerardyn et al. (2009) (see also Buhl-Mortensen et al., 2010: fig. 10) recorded it from the coral degradation zone in the Porcupine Seabight (NE Atlantic). The species provisionally identified by Cuvelier et al. (2014) as ‘ Lobopleura cf. expansa ’ [and by Sarrazin et al. (2015) as Lobopleura and by Plum et al. (2017) as L. expansa ] from the Eiffel Tower edifice (Lucky Strike vent field) at 1700 m depth on the Mid-Atlantic Ridge south of the Azores probably refers to a different taxon. Lobopleura multispinata is, so far, known only from its type locality in the White Sea.

AUTHENTICITY OF LAOPHONTODES BICORNIS IN KOREA

The original description of La. bicornis was based on a single female dredged off Port Erin, Isle of Man, in the Irish Sea (A. Scott, 1896). The male was first reported, but not described, in dredgings from Kilbrennan Sound, Firth of Clyde in Scotland (T. Scott, 1896). It has subsequently been recorded from several localities in northwestern Europe, including Norway ( Sars, 1908), Sweden ( Lang, 1948), Ireland ( Farran, 1913; Roe, 1958, 1960; Sloane et al., 1961; Holmes, 1983), Northern Ireland ( Wells, 1963), England ( Norman & Scott, 1906; Wells, 1961; Ventham, 2011) and France ( Monard, 1935: as L. hamatus var. reflexus ; Bodin & Le Guellec, 1992). Mediterranean records include those from France ( Monard, 1928; Bodin, 1964; Soyer, 1971; Mascart et al., 2015), Italy ( Pesta, 1959), Croatia ( Brian, 1923), Greece ( Brian, 1928a, b), Turkey ( Noodt, 1955), Egypt ( Gurney, 1927) and Israel ( Por, 1964), whereas Griga (1964) reported it from the Crimea in the Black Sea. Additional records outside Europe were added from the Suez Canal by Por & Marcus (1973) and from the Red Sea by Nicholls (1944) and Por (1967).

George & Gheerardyn (2015) redescribed both sexes of La. bicornis in detail based on material from Kopervik (Western Norway, collected by G. O. Sars) and the Gullmar Fjord ( Sweden, collected by K. Lang) and concluded that its allegedly wide distribution documented in the literature is probably unfounded. The only records of the species they considered as authenticated are those from the type locality (A. Scott, 1896), Norway ( Sars, 1908) and Sweden ( Lang, 1948).

Within the genus Laophontodes , there is a welldelimited group of species characterized by the presence of: (1) paired posterolateral, backwardly directed, corniform processes on the cephalothorax; and (2) crenulate posterior margins on the body somites. This bicornis group includes La. bicornis , La. hamatus , Laophontodes brevis Nicholls, 1944 , Laophontodes ornatus Krishnaswamy, 1957 and La. horstgeorgei George & Gheerardyn, 2015 , but it is likely that other closely related species await discovery or have previously been confounded with described members of this group. The underlying reason for this unsatisfactory state of affairs is twofold.

Firstly, the available descriptions of La. hamatus , La. brevis , La. ornatus and, until recently, La. bicornis , are inadequate for unequivocal identification, and this has led to considerable confusion in the past. Thomson (1883) described the genus and species Merope hamata , on the basis of a single female from Dunedin Harbour in New Zealand. Gurney (1927) considered the species conspecific with La. bicornis , whose binomen loses in priority to M. hamata . However, since Merope Thomson, 1883 is preoccupied, being a junior homonym of Merope Newman, 1838 ( Mecoptera ), Merope Adams & Adams, 1856 (Mollusca) and Merope Albers, 1860 (Mollusca), he proposed Laophontodes hamata ( Thomson, 1883) as the new combination. Lang (1934) reinstated Thomson’s species with the corrected spelling, Laophontodes hamatus , and provided a partial redescription based on material from Campbell Island and Stewart Island ( New Zealand). Monard (1935) proposed a new variety, La. hamatus var. reflexus , based on differences in colour, rostrum size and shape of the somites observed in specimens from the Roscoff area, but its validity was subsequently refuted by Lang (1936b), who considered it a variety of and subsequently ( Lang, 1948: 1455) synonymous with La. bicornis . Laophontodes brevis was described by Nicholls (1944) based on six specimens from Ghardaqa, Egypt in the Red Sea. Nicholls (1944) did not disclose the sex of his individuals, but Lang (1965) recognized a similarity between La. brevis and the copepodid V stage of La. bicornis , believing that both species were probably conspecific. Contrary to certain authorities, who list La. brevis as a junior synonym of La. bicornis (e.g. Bodin, 1997) or omit it altogether ( Gee & Fleeger, 1986; Schizas & Shirley, 1994; Wells, 2007; George & Gheerardyn, 2015; George, 2018), Lang (1965) did not formally synonymize it with the latter, hence the species must remain as valid until forthcoming evidence proves otherwise. Krishnaswamy (1957) described La. ornatus from four females dredged off the coast of Chennai (Madras), Tamil Nadu ( India). Although the species is clearly a member of the bicornis group, its description is grossly inadequate, and various authors (e.g. Lang, 1965) have pointed out discrepancies between the text and accompanying illustrations. It should be noted that Gee & Fleeger (1986) erroneously listed the species under the name L. armatus in their table 2 (p. 158).

Secondly, Gheerardyn & George (2010) pointed out the superficial resemblance between La. bicornis and members of the genus Ancorabolina . Given that both share the paired corniform processes on the cephalothorax, George & Gheerardyn (2015) argued strongly that some earlier records of the former in reality were based on as yet undescribed species of the latter. This claim has been confirmed for Büntzow’s (2011) La. var. bicornis from the Seine seamount in the subtropical NE Atlantic (George, 2013) and extrapolated to the entire area where Ancorabolina spp. have been sighted so far, including the Mediterranean basin, where La. bicornis has allegedly been reported on numerous occasions (see above). Likewise, the authenticity of the records from the Black Sea and the Red Sea must remain unconfirmed, particularly those of Nicholls (1944), who reported two co-existing morphs of La. bicornis alongside La. brevis .

The records of Nicholls (1944) and Por (1967) from the Red Sea and the recent discovery of La. horstgeorgei from the Fiji Islands indicate that the bicornis group probably assumes a distribution throughout the Indo- Pacific. This is further corroborated by Kim’s (2013) report of La. bicornis from Seogwipo on Jeju Island, South Korea, where it was found in washings of wood infested by limnoriid isopods and teredinid bivalves and of coralline sand collected at moderate depths. Kim’s (2013) report consists of a text description of both sexes and three accompanying photographs illustrating the habitus, cephalothorax and urosome (all in dorsal aspect) of the female. However, the Korean material (six ♀♀, four ♂♂) differs from George & Gheerardyn’s (2015) redescription of La. bicornis in several significant aspects: (1) the lateral corniform processes on the cephalothorax are much shorter and blunt in appearance; (2) the pleural areas of the body somites are much more produced, so that the constrictions between individual somites appear more pronounced in dorsal aspect; (3) P3 and P4 exp-3 bears one inner seta instead of two; (4) P2–P4 endopod is one-segmented instead of two-segmented; (5) P3 endopod without outer seta instead of 0.021 pattern in La. bicornis ; (6) the caudal rami are relatively shorter; and (7) from Kim’s (2013) text description, it can be inferred that the male antennule shows two segments (instead of one) distal to the geniculation (cf. ‘Antennule 6-segmented …; fourth segment swollen, with aesthetasc arising from process near distal margin; sixth segment with an aesthetasc’). Within the bicornis group, Kim’s (2013) material is unique in the presence of only one inner seta on P3 and P4 exp- 3, indicating that it deserves distinct specific status. In accordance with ICZN (1999) Articles 16.4 and 72.5.6, we would have selected the female specimen illustrated by Kim (2013) in his fig. 19 as the holotype of a new species, but this would be premature. Huys & Lee (2018) pointed out that Kim (2013) copied the original text description of Folioquinpes mangalis Fiers & Rutledge, 1990 virtually verbatim, explaining the many discrepancies between the text and some of his illustrations of what appeared to be the new species Folioquinpes pseudomangalis Huys & Lee, 2018 . Upon closer inspection this practice of unorthodox copying and pasting text blocks turns out to be rampant in his two reports of the Korean marine harpacticoid fauna ( Kim, 2013, 2014), casting serious doubts on the scientific integrity of the work. For example, his 2014 text description (totally lacking in illustrations) of “ Laophontodes psammophilus ” – a species in reality belonging to another genus ( George, 2017) – is repeated in exactly the same words (and typographical errors) as were used originally in his 2013 description of “ L. bicornis ”. In addition, the same text (except for body length measurements) was reproduced verbatim for the description of Algensiella boitanii ( Kim, 2014: 88–89) which differs radically from members of Laophontodes . Clearly the taxonomic mess created by Kim (2013, 2014) in his two reports can only be cleared up by thorough re-examination of the original material and clearly demonstrates the pitfalls of publications that have not been subjected to peer-review.