Reproductive morphology and redescriptions of some Neanthes Kinberg, 1865 (Annelida: Nereididae) species from the southeastern Asian seas, with comparative synoptic tables of accepted species

Abstract Neanthes Kinberg, 1865 is a speciose nereidid genus with some species exhibiting epitoky during the timing of reproduction. The epitokal morphology of males and females has long been agreed to be diagnostic to distinguish species. In the present study, we provide redescriptions of the type material and reassess the current taxonomic status of four Neanthes species originally described based on the reproductive forms of specimens collected from Southeast Asia: Neanthes augeneri (Gravier & Dantan, 1934), N. gisserana (Horst, 1924), N. trifasciata (Grube, 1878) and N. trifasciata vandersandei (Horst, 1924). The non-metamorphosed and epitokous diagnostic features, including new characters found in this study, were used to redescribe and distinguish these species from other congeners. The four species are here recognised in Neanthes, whereby N. vandersandei is elevated to species level. Neanthes gisserana is considered as distinct from its previous senior synonym N. rubicunda (Ehlers, 1868). Morphology of both N. trifasciata and N. vandersandei, originally described based on mixed species, is restricted. Additionally, some remarks on the status of N. thysanota (Ehlers, 1908) are given. Distributional and reproductive data of the reviewed species are provided. An updated checklist of 80 valid species in Neanthes and synoptic tables of diagnostic features including the atokous and epitokous morphology are also presented.


Introduction
Nereididae de Blainville, 1818 includes species with reproductive modes that are conspicuously variable. However, most exhibit swarming behaviour to release gametes into the water, accompanied by a series of physiological, anatomical and morphological modifications when they become sexually mature (Smith 1958;Clark 1961;Schroeder & Hermans 1975;Sato 2017). The metamorphosis into reproductive forms is named epitoky (Ehlers 1868;Gravier & Dantan 1927;Defretin 1949;Clark 1961).
Generally, the epitokes show sexual dimorphism, which is more noticeable during the marked epitokous form, also known as heteronereis. The males undergo a more significant structural metamorphosis than the females by usually showing papillae in the pygidium, crenulate papillae in some parapodial cirri, and more marked swellings in the first parapodial cirri. Some other sexual differences can be found in the number of body regions and pre-natatory chaetigers, the number of basally swollen first parapodial cirri, and the colour and size of individuals (Fauvel 1923;Fage & Legendre 1927;Gravier & Dantan 1928;Reish 1954;Clark 1961;Boilly-Marer 1972). Nonetheless, the epitokal morphology has rarely been recorded in sufficient detail in the taxonomic literature to describe and distinguish species, until recently (Read 2007 Neanthes Kinberg, 1865 is a speciose and heterogeneous nereidid genus with the number of valid taxa currently fluctuating between 74 and 88 Bakken et al. 2018;Read & Fauchald 2018a;Hsueh 2019a). It comprises several species whose morphology is identified solely from distinctive epitokal modifications; among them are three little-known taxa originally described from the southeastern Asian seas: Neanthes augeneri , N. gisserana (Horst, 1924) and N. trifasciata vandersandei (Horst, 1924).
Neanthes augeneri was described using males and females from Vietnam  and has been recorded from Djibouti (Fauvel 1951). Although the species has prevailed within Neanthes (e.g. Hartman 1959Hartman , 1974Fauchald 1972;Wilson 1984), it has been allocated interchangeably in Nereis (Fauvel 1939;Read & Fauchald 2018b). Moreover, N. gisserana, described originally from the Moluccas (Indonesia), was synonymised with the Adriatic Sea species N. rubicunda (Ehlers, 1868) (Núñez 1995), but the former is still recognised as valid without justification (see Glasby 2015;Pamungkas & Glasby 2015). Finally, N. trifasciata vandersandei was described in Nereis using male specimens from Timor (Indonesia) and distinguished from the stem species Nereis trifasciata Grube, 1878 (type locality: the Philippines) based solely on the pigmentation patterns (Horst 1924). Although the subspecies was treated within Nereis (Hartman 1959;Salazar-Vallejo et al. 2014;Pamungkas & Glasby 2019), it was recently allocated in Neanthes without reasons being provided (Hsueh 2019a(Hsueh , 2019b. The current generic assignment of both N. trifasciata (Glasby 2015) and N. trifasciata vandersandei is still uncertain, and the subspecies position has not been addressed since originally described. All the Southeast Asia species mentioned above have not been redescribed based on type materials, even though most of them were described and illustrated only briefly if at all.
In the present study, we redescribe N. augeneri, N. gisserana and N. trifasciata vandersandei using type material in the epitokous stage to reassess their current taxonomic status and to describe their morphology based on the non-metamorphosed (i.e. atokous) and epitokous characters. We also redescribe N. trifasciata to re-evaluate its generic position and the subspecies status of N. trifasciata vandersandei. Lastly, we provide lists of all the accepted Neanthes species, with atokous and epitokous features, for convenience in comparison and identification.

Literature review
The valid species of Neanthes listed in the World Polychaeta Database (Read & Fauchald 2018a) were examined, compared and updated with earlier lists (Hartman 1959;Fauchald 1972;Wilson 1984) and more recent literature. The original descriptions, redescriptions and/or characterisations of topotypes of the currently accepted species were gathered and analysed. These were only considered to disregard possible misidentifications within the literature based upon non-type materials.

Morphological observation
The type materials examined in this study are deposited in the following museums: Muséum National d'Histoire Naturelle, Paris, France (MNHN); Museum of Natural History, Wroclaw, Poland (MPW); and Zoological Museum, University of Amsterdam (collections now in Naturalis Biodiversity Center, National Museum of Natural History), Leiden, The Netherlands (ZMA).
Total length (TL), length from the distal end of the prostomium to chaetiger 15 (L15), and body width at chaetiger 15 excluding parapodia (W15) were measured; and the total number of chaetigers was counted for complete specimens. Denticles on the jaws and paragnaths on unpaired and paired sides (denoted "a" for left and "b" for right) of pharyngeal areas were counted (range: min-max). Features of the non-everted pharynx were observed by conducting a longitudinal dissection in the midventral oral region. Parapodia were dissected and mounted on glass slides to examine parapodial features. Decimal numbers are used for practical purposes when measurements between two structures exceeded one unit (e.g. 1.2 times, 2.5 times, twice); whereas written fractions are used when those measurements are less than one unit (e.g. half, twothirds, four-fifths).
Light microscopy observations were carried out using both stereo and compound microscopes. Specimens were photographed using a digital camera (Nikon D5100) mounted on each microscope with a portable microscope adaptor; around 15-20 photos were stacked to improve the visual field's depth using Helicon Focus® 6 (Method C). The figure background was cleaned and darkened or lightened as necessary, and the final figures were assembled in plates using Adobe Photoshop® CS6. Parapodia are shown in anterior view unless otherwise stated.
Descriptions of the species are based upon the atokous and epitokous characters of one primary type (holotype, lectotype or syntype) unless variations are indicated in parentheses for other material. The relative extension of parapodial structures and the relative width of ligules and lobes were described following recent studies (Villalobos-Guerrero & Carrera-Parra 2015; Conde-Vela 2018). However, the dorsal cirri were measured considering the length of the proximal lobe of dorsal ligules (from now on: proximal dorsal ligule), whereas the distal lobe of dorsal ligules (from now on: distal dorsal ligule) was measured regarding the length of proximal dorsal ligules (Villalobos-Guerrero 2019).
The atokous and epitokous nereidid parapodial terminology and standardised definitions of the articulations of chaetae by Villalobos-Guerrero & Bakken (2018) were followed. The size of falcigers' blade (b/ a ratio) and the length of its serrated edge concerning the total blade length were described following Bakken & Wilson (2005) and Glasby & Hsieh (2006), respectively. The first chaetiger with an additional parapodial lobe determined the start of the natatory region, particularly the lower lobe of the cirrophore of ventral cirri. The paragnath terminology of Bakken et al. (2009), partially readapted by Conde-Vela (2018), and the characterisation of the jaws by Jansonius & Craig (1971) were used. Finally, the description of the pharynx arrangement at the dorsal areas of the oral ring, namely the "area VI-V-VI ridge pattern", is based on Villalobos-Guerrero (2019).
The definition that separates the two genera prevails nowadays (e.g. Fauchald 1972;Wu et al. 1981;Khlebovich 1996;de León-González et al. 1999;Núñez 2004;Bakken & Wilson 2005), and it is also followed in the present study. Nonetheless, the difference is sometimes problematic, such as in heteronereis specimens having conical paragnaths on all (or most) pharyngeal areas. The replacement of homogomph falcigers by specialised chaetae can occur in the posterior natatory parapodia of some Nereis species (McIntosh 1910;Pettibone 1963;Chambers & Garwood 1992;TFVG, pers. obs.). This chaetal feature is insufficient to distinguish between Neanthes and Nereis specimens in the heteronereis stage and in those missing the posterior end. A comprehensive revision of Neanthes that defines the genus using other characters remains to be performed.
Palpophores sub-conical, thick, 1.2 times wider than long (Figure 1(b,c)), as long as two-fifths of entire prostomium; with distinct sub-distal transverse groove. Palpostyles bluntly conical, with diameter as wide as one-third of palpophore.
Left antenna missing but with marked scar; right antenna tapered, slender, medium size (Figure 1(c)), extending forwards beyond palpophore and posteriorly to distal two-fifths of length of prostomium; antennae close together with gap as wide as onethird of antennal diameter.
Paired eyes blackish, markedly enlarged, arranged in a rectangular form, with both pairs overlapping ( Figure 1(b,c)). Anterior pair of eyes ovoid, 5 times as wide as antennal diameter, gap between both eyes as wide as diameter of eyes (Figure 1(c)), with lens

566
T. F. Villalobos-Guerrero and I. Idris inconspicuous, purplish, covering 30% of eye. Posterior pair of eyes ovoid, 5.5 times as wide as antennal diameter, with lens inconspicuous, purplish, placed posterolaterally in eyes and covering 35% of it. Apodous anterior segment 3 times wider than long, as long as chaetiger 1, with flattening anterior margin, dorsum without marked transverse wrinkle.
Pharynx semi everted, yellow-amber jaws not dissected, with at least 10 (9) well-developed and blunt denticles; pulp cavity with two canals. Pharynx consisting of paragnaths without abnormal development due to epitokal metamorphosis ( Figure 1(c,d)), brownish to yellow-amber on maxillary and oral rings, consisting of conical and merged paragnaths; plate-like basements absent. Area I: 4 (5), cross-shaped patch of even cones; areas IIa: 17 (24) and IIb: 18 (24), three slightly regular rows of even cones in eyebrow-shaped patch (Figure 1(c)); area III: 28 (28), four irregular rows of even cones in oval patch, without distinct laterally isolated cones (Figure 1(d)); areas IVa: 18 (22) and IVb: 15 (23), crescent-shaped patch with three slightly regular transverse rows of even cones located proximally and two (three) medium size and thickened merged paragnaths (1.5-2 times longer than wide) located distally; area V: 1 (1), cone placed slightly behind level of paragnaths on area VI; areas VIa: 4 (6) and VIb: 6 (6), one rounded patch of even cones (Figure 1(c)), none centrally placed; areas VII-VIII: 105 (90), two bands of cones ventrally well separated, with anterior band consisting of two transversely aligned rows (furrow row and ridge row with one cone on each region), and posterior band with three transverse rows slightly displaced from each other having cones on furrows and ridges (Figure 1(d)). Areas VI-V-VI ridge pattern, λ-shaped. Gap between area VI and areas VII-VIII narrow, as wide as palpostyle.

Ecology. Unknown.
Reproduction. Males undergo marked epitokal metamorphosis and possibly swarm in the water column to release the sperm through the pygidial papillae.
Remarks. Neanthes augeneri from Vietnam is similar to N. crucifera (Grube, 1878) from the Philippines, N. kermadeca (Kirkegaard, 1995) from the Kermadec Trench, and N. roosevelti Hartman, 1939 from the Galapagos Islands. These species share proximal dorsal ligule of similar size throughout the body (or slightly enlarged in posterior parapodia), absence of notopodial prechaetal, neuropodial superior and neuropodial postchaetal lobes, and area VII-VIII with 80 or more paragnaths (Table II).
Neanthes augeneri differs from them all by the following features: (1) 4-5 paragnaths on area I and one on area V, in contrast to more than 50 on both areas in N. roosevelti and one on area I in N. kermadeca; (2) the rounded patch of paragnaths on area VI, in comparison to the transverse patch in N. roosevelti; (3) the presence of eyes, in contrast to its absence in N. kermadeca; (4) the bluntly conical distal dorsal ligules in anterior chaetigers, in comparison to those conical in N. kermadeca; (5) 17-24 paragnaths on area II, whilst there are seven in N. kermadeca; (6) and the shallow marine habitat, in contrast to the abyssal habitat (4410 m depth) of N. kermadeca. Notwithstanding, N. augeneri seems more similar to N. crucifera in both morphological and geographical terms. The latter species' type material is lost, the original description is brief with few illustrations, and no records from the type locality exist. Several features remain uncertain; however, a couple of recognisable characters are useful to separate the two species. In N. augeneri, there is one paragnath on area V, and the paragnaths are divided into two bands (anterior and posterior) on areas VII-VIII; whilst N. crucifera has six paragnaths on area V and one broad band of paragnaths on areas VII-VIII.
Regarding the epitokal morphology, N. augeneri is similar to N. biseriata Hutchings & Turvey, 1982 from Australia, N. nanhaiensis Wu, Sun & Yang, 1981 from China, and N. vandersandei from Indonesia. These species share males with body divided into two regions, up to 15 pre-natatory chaetigers, first seven pre-natatory parapodia with dorsal cirri modified, and dorsal cirri of natatory chaetigers with several papillae (Table III). However, the males of N. augeneri can be distinguished from them all by the following features: (1) the presence of a distinctly enlarged basal end on the median ligule of natatory chaetigers, in contrast to its absence in N. nanhaiensis and N. vandersandei; (2) the dorsal pigmentation on the first 8-10 anterior chaetigers, in comparison to that present throughout the body in N. nanhaiensis and N. vandersandei and absent in N. biseriata; and (3) the dorsal cirri of chaetigers 6-7 with a blunt sub-distal end, in contrast to that llama head-like in N. vandersandei. Moreover, N. augeneri and N. biseriata can be separated only by (4) the smooth dorsal cirri present in the last 10-12 chaetigers, in contrast to those in the last 20 chaetigers in N. biseriata.
Other pharyngeal and parapodial features can separate N. augeneri from those three species: (5) the higher number of paragnaths on areas I (4-5), III (28), V (1) and VII-VIII (90-105), in contrast to the lower number (area I: 0-1, III: up to 15, V: 0, VII-VIII: up to 3 paragnaths) in N. nanhaiensis and N. vandersandei (Table II); (6) the λ-shaped ridge pattern of areas VI-V-VI, in contrast to that πshaped in N. nanhaiensis and N. vandersandei; (7) the heterogomph spinigers in the subacicular neurochaetae of pre-natatory chaetigers, in comparison to those homogomph in N. nanhaiensis; (8) the conical and merged paragnaths on area IV, in contrast to only conical in N. vandersandei; (9) one paragnath on area V, compared with 9-19 in N. biseriata (Table II); and (10) the absence of the neuropodial superior lobe, in contrast to its presence in N. biseriata.
According to , one male and two females constitute the original material. Solís-Weiss et al. (2004) indicated that only a single syntype remains. Upon examining the type material here, three epitokous male syntypes were found.  assigned N. augeneri to the subgenus Nereis (Neanthes) sensu de Saint-Joseph (1898) by having conical paragnaths on all areas. Hartman (1959Hartman ( , 1974 listed N. augeneri in Neanthes possibly founded on her genus definition (see Neanthes remarks), and it has been recognised within this genus until the present (Fauchald 1972;Wilson 1984;Kim Hong 2015). Atokous specimens of N. augeneri need to be examined to identify middle and posterior parapodia features, which can confirm the generic assignment of the species.
Neanthes augeneri has not been recorded from the type locality since it was described, although it has been incorporated into the species lists that contain Nha Trang (Hartman 1974;Salazar-Vallejo et al. 2014;Kim Hong 2015). The species has been reported from the Arabian Peninsula (Fauvel 1951;Wehe & Fiege 2002), but these records are regarded as doubtful. Fauvel (1951) described the male from Djibouti briefly, without illustrations, and the general epitokous morphology provided is also similar to N. biseriata, whilst Wehe & Fiege (2002) did not give a description.
Palpophores ovoid, thick, as long as wide ( Figure  3(c)), as long as two-fifths of entire prostomium; without sub-distal transverse groove. Palpostyles bluntly conical, with diameter as wide as half of palpophore.
Antennae tapered, thick, medium in size; extending forwards beyond palpophore and posteriorly to distal two-fifths of length of prostomium; antennae close together with gap as wide as one-third of antennal diameter (Figure 3(c)).
Paired eyes blackish, markedly enlarged, arranged in a rectangular form, with both pairs overlapping (Figure 3(b)); anterior pair of eyes ovoid, 3 times as wide as antennal diameter, gap between eyes as wide as diameter of eyes, with lens conspicuous, whitish, covering 35% of eye; posterior pair of eyes ovoid, 2.7 times as wide as antennal diameter, with lens conspicuous, whitish, placed posterolaterally in eyes and covering 35% of it (Figure 3(b)).

Distribution. Banda Sea.
Ecology. Coral reefs and thalloid red algae Lithothamnion sp. in 18 m depth.
Reproduction. Males undergo marked epitokal metamorphosis, and possibly swarm in the water column and release sperm through pygidial papillae. Females unknown.
Remarks. Neanthes gisserana from the Maluku Province (Indonesia) resembles N. trifasciata from the Philippines and N. vitiazi Khlebovich, 1996 from the Japan Trench. These species share proximal dorsal ligule of similar size throughout the body (or slightly enlarged in posterior parapodia), presence of neuropodial postchaetal lobes at least in some few anterior chaetigers, absence of notopodial prechaetal and neuropodial superior lobes, and areas VII-VIII with one single transverse row of fewer than 10 paragnaths (Table II). Likewise, N. gisserana and N. trifasciata are similar in sharing area IV with five long merged paragnaths, area VI with a rounded patch of paragnaths, and π-shaped ridge pattern in areas VI-V-VI. Also, N. gisserana resembles N. vitiazi by presenting spiniger-like falcigers.
Neanthes gisserana differs from these two species by the following features: (1) the multi-articulated tentacular cirri, in contrast to those smooth in N. trifasciata and N. vitiazi; (2) the band of pigmentation on the apodous anterior segment, in comparison to two prominent and round spots in the apodous segment of N. vitiazi and one band in several anterior segments in N. trifasciata; (3) the presence of neuropodial falcigers with both spiniger-like and convex blades, in contrast to those only convex in N. trifasciata; (4) the spinigers with long blade (b/a ratio 10-16.5), in comparison to that shorter (b/a ratio 5.5-7.2) in N. trifasciata; (5) the higher number of paragnaths on areas I (6) and III (20), in comparison to the lower number (areas I: 0 and III: 8) in N. trifasciata; (6) the presence of heterogomph falcigers only in subacicular neurochaetae, in contrast to also homogomph in N. vitiazi; (7) the thick and short (b/a ratio 0.9-1.1) blade of falcigers, in comparison to that slenderer and more elongate (b/a ratio 3-3.5) in N. vitiazi; and (8) the shallow-water habitat, in contrast to that abyssal (3342-4160 m depth) in N. vitiazi.
Concerning the epitokal morphology, N. gisserana is similar to N. kerguelensis (McIntosh, 1885) from Kerguelen Islands (Southern Ocean), N. pachychaeta (Fauvel, 1918) from Madagascar, and N. rubicunda from the Adriatic Sea. In these species, the males have the body divided into two regions, up to 16 prenatatory chaetigers, median ligule with an enlarged basal end in natatory chaetigers, and dorsal cirri of natatory chaetigers with one papilla or without papillae (Table III). Likewise, N. gisserana and N. rubicunda are similar in having dorsal cirri of natatory chaetigers with a single papilla (Table III).
The males of N. gisserana can be distinguished from all of these species by the following characters:

574
T. F. Villalobos-Guerrero and I. Idris (1) the presence of one papilla on dorsal cirri of natatory chaetigers, in contrast to its absence in N. kerguelensis and N. pachychaeta (Table III); (2) the 15 pre-natatory chaetigers, whilst there are 13-14 in N. kerguelensis and 16 in N. rubicunda; (3) the two upper lobes in the cirrophore of ventral cirri of natatory chaetigers, whilst there is a single one in N. kerguelensis and N. rubicunda; (4) the nonprojected base of dorsal cirri in natatory chaetigers, in contrast to that markedly projected and cordiform in N. kerguelensis; (5) the broadly enlarged and reniform lower lobe of the cirrophore of ventral cirri in natatory chaetigers, in comparison to that slender and digitiform in N. kerguelensis; (6) the presence of an upper basal lobe in the ventral ligule of natatory chaetigers, in contrast to its absence in N. kerguelensis; (7) the non-branched lobes and lamellae in parapodia of natatory chaetigers, in comparison to those distinctly ramous in N. pachychaeta; (8) the paragnaths located on soft pharyngeal tissue, in contrast to those attached on a plate-like basement in N. pachychaeta; (9) the first five ventral cirri of pre-natatory chaetigers modified, in comparison to the first 6-7 in N. pachychaeta; and (10) the reddish-brown band on the apodous segment, in contrast to the reddish-brown patches on the dorsum of the anterior segments in N. rubicunda. Some other pharyngeal, parapodial and chaetal features can separate N. gisserana from those species: (11) the absence of the notopodial prechaetal lobe, in contrast to that well-developed in N. kerguelensis, N. pachychaeta and N. rubicunda (Table II); (12) the π-shaped ridge pattern of areas VI-V-VI, in contrast to that λ-shaped in N. kerguelensis and N. pachychaeta; (13) the conical and merged paragnaths on area IV, in contrast to only conical in N. kerguelensis and N. pachychaeta; (14) the higher numer of paragnaths on area III (16-27), in comparison to the lower number (area III up to 14 paragnaths; Table II) in N. kerguelensis and N. pachychaeta; (15) the heterogomph spinigers in the subacicular neurochaetae of pre-natatory chaetigers, in contrast to those homogomph in N. kerguelensis; (16) the non-fused neuropodial falcigers, in comparison to those partially and fully fused in N. pachychaeta; and (17) the higher number of paragnaths on area I (2-6), in contrast to the lower number (area I: 0-1) in N. rubicunda. Horst (1924) described N. gisserana without comparisons based upon six epitokous males and two atokous specimens, although six syntypes remain (Bleeker & van der Spoel 1992). Two male syntypes were available in the present study and fit with the original description. Monro (1939) identified some incomplete atokous specimens from the Maldives and Seychelles islands as N. gisserana, particularly by the presence of bars (i.e. merged paragnaths) on area IV. However, some differences were found in the present study between Horst's (syntype) specimens and Monro's (non-type) description. The non-types have areas I (0), II (9), IV (3-4 merged paragnaths) and VI (3) with a lower number of paragnaths than in the syntypes (areas I: 2-6, II: 18-20, IV: 5 merged paragnaths, VI: 4-12). The non-types have paragnaths located in a transverse row on area VI, in contrast to a round patch in the syntypes. Lastly, the nontypes have posterodorsal tentacular cirri nearly twice as long as those in the syntypes. Hence, we judge that Monro's non-types may represent a new species. Núñez et al. (1984) identified specimens from the Canary Islands with some doubts as N. gisserana based upon Horst's (1924) and Monro's (1939) descriptions. Later, based on specimens from the Canary, Maldives and Seychelles islands, Núñez (1995) synonymised N. gisserana with N. rubicunda from the Adriatic Sea. This nomenclatural act was followed in subsequent studies, rendering N. rubicunda a widely distributed species in tropical and subtropical waters (Núñez & Brito 2002;Núñez 2004). Nonetheless, we consider both species valid because they can be separated by atokous and epitokous features (see above). Glasby (2015) and Pamungkas & Glasby (2015) treated N. gisserana as valid in discussing a similar species referred to as 'Neanthes sp. cf. N. gisserana' from Eastern Indonesia and Northeastern Australia, but they provided no comments on the validity. These authors considered that 'Neanthes sp. cf. N. gisserana' may represent a new species by having 21-22 prenatatory chaetigers, a few paragnaths in a single row on area III, a brown band on chaetiger 4, and a much larger body.
Neanthes gisserana has not been recorded from the type locality since described but has been included in a few species lists covering Indonesian waters (Hartman 1974;Salazar-Vallejo et al. 2014). The records from the Maldives and Almirante islands (Monro 1939;Fauvel 1953) may represent a new species.
Description. Lectotype epitokous female, incomplete, slightly damaged, 17 mm TL, 4.5 mm L15, 1.6 mm W15, and with 53 chaetigers. Body colour brown ( Figure 5(a)), with tiny black spots forming ovoid patches in distal end of inner ( Figure 5(b,c)) and lower parts of palpophore, and in distal end of dorsum of prostomium, same spots forming a midanterior row in dorsal and ventral sides of apodous anterior segment and first 29 chaetigers (more obvious in first anterior chaetigers), without further pigmentation patterns in long-term preserved specimens; reddish-brown band on dorsum of apodous anterior segment (segment 1), chaetigers 2-3
Palpophores ovoid, slender, as long as wide ( Figure 5(c,d)), as long as one-half of entire prostomium; with sub-distal transverse groove. Palpostyles bluntly conical, with diameter half as wide as palpophore.
Antennae tapered, slender, medium in size; extending forwards slightly beyond palpophore and posteriorly to distal two-fifths of length of prostomium; antennae close together with gap as wide as two-thirds of antennal diameter ( Figure 5(c)).
Paired eyes dark purple, markedly enlarged, arranged in a rectangular form, with both pairs barely overlapping (Figure 5(c)); anterior pair of eyes ovoid, 3 times as wide as antennal diameter, gap between eyes 1.2 times as wide as diameter of eyes, with lens conspicuous, purplish, covering 85% of eye; posterior pair of eyes rounded, 3 times as wide as antennal diameter, with lens conspicuous, purplish, placed posterolaterally in eyes and covering 85% of it ( Figure 5(c)).
Pharynx not everted, previously dissected. Jaws orange-amber in distal third, remaining yellowamber, with 12 well-developed and blunt denticles; pulp cavity as long as three-quarters of jaw, with two canals. Pharynx without abnormal development of paragnaths due to epitokal metamorphosis ( Figure 5 (e)), brownish on maxillary and oral rings, consisting of conical and merged paragnaths; plate-like basements absent. Area I: 0; areas IIa: 14 and IIb: 11, two irregular rows of uneven cones in eyebrowshaped patch, distal cones longer ( Figure 5(e)); area III: with two paragnaths but large fragment of area missing, small patch of eight cones according to original description; areas IVa: 17 and IVb: 19, lemniscate patch with three irregular transverse rows of uneven cones located proximally and five long and thick merged paragnaths (2-4 times longer than wide; Figure 5(e)) located distally; area V: 0 (0); areas VIa: 5 and VIb: 6, one small rounded patch of uneven, close-together cones; areas VII-VIII: 5, single transverse row of uneven cones, gradually enlarging dorsoventrally. Areas VI-V-VI ridge pattern, π-shaped. Gap between area VI and areas VII-VIII broad, 1.3 times as wide as palpophore.
Remarks. Neanthes trifasciata from the Philippines resembles N. gisserana from the Maluku islands (Indonesia) and N. vitiazi from the Japan Trench. These species share proximal dorsal ligule of similar size throughout the body (or slightly enlarged in posterior parapodia), presence of neuropodial postchaetal lobes at least in some few anterior chaetigers, absence of notopodial prechaetal and neuropodial superior lobes, and areas VII-VIII with a single transverse row of fewer than 10 paragnaths (Table II). Neanthes trifasciata and N. gisserana are also similar in having five long merged paragnaths on area IV, a rounded patch of paragnaths on area VI, and a π-shaped ridge pattern in areas VI-V-VI. However, they can be distinguished by several non-metamorphosed features (see remarks on N. gisserana for detailed comparison). Moreover, N. trifasciata differs from N. vitiazi by the following diagnostic features: (1) the presence of heterogomph falcigers only in subacicular neurochaetae, in contrast to also homogomph in N. vitiazi; (2) the thick and short/medium (b/a ratio 0.8-1.8) blade of falcigers, in comparison to that slenderer and longer (b/a ratio 3-3.5) in N. vitiazi; (3) the higher number of paragnaths on areas II (11-14), IV (17-19) and VI (5-6), in comparison to the lower number (areas II: 7, IV: 6-7, VI: 1-2) in N. vitiazi; and (4) the shallow-water habitat, in contrast to that abyssal (3342-4160 m depth) in N. vitiazi.
Concerning the epitokal morphology, N. trifasciata resembles N. donghaiensis from China and N. pachychaeta from Madagascar. Females of these species have 24 pre-natatory chaetigers (Table III). However, the females of N. trifasciata can be distinguished from these two species by the following features: (1) the less-developed epitokal morphology, in contrast to that well developed in N. donghaiensis and N. pachychaeta; (2) the non-branched lobes and lamellae in parapodia of natatory chaetigers, in comparison to those distinctly ramous in N. pachychaeta; (3) the paragnaths located on soft pharyngeal tissue, in contrast to those attached on a plate-like basement in N. pachychaeta; (4) the presence of a small and blunt basal end on median ligule of natatory chaetigers, in comparison to that distinctly large and flabellate in N. donghaiensis (Table III); (5) the absence of an upper secondary lobe in ventral ligule, in contrast to its presence in N. donghaiensis (Table III); and (6) the first five dorsal and four ventral cirri modified, in comparison to those six and five in N. donghaiensis, respectively (Table III).
Neanthes unifasciata (Willey, 1905) from the Gulf of Mannar has been considered similar to N. trifasciata except in the presence of notopodial homogomph falcigers (Fauvel 1930(Fauvel , 1932(Fauvel , 1935(Fauvel , 1936(Fauvel , 1951Fishelson & Rullier 1969;Rullier 1972). Indeed, both N. trifasciata (this study) and N. unifasciata (Fauvel 1932) lack that type of chaetae. The morphology of N. unifasciata has not been delimited, although based upon the original description (Willey 1905) and characterisation of the topotypes (Fauvel 1932), N. trifasciata differs from N. unifasciata in the few following atokous features: (1) no paragnaths on area I, whilst there are 3-6 in N. unifasciata (Table II); (2) dorsal cirri 1.5 times as long as distal dorsal ligule, in contrast to those twice as long as in N. unifasciata; and, (3) several segments with a broad reddish-brown band, in comparison to only one (chaetiger 2) pigmented in N. unifasciata. A more comprehensive study based on topotypes is required to detect other differences. The following relevant characters need consideration: the ridge pattern of areas VI-V-VI, the presence of neuropodial postchaetal lobe and merged paragnaths on area IV, and the number of pre-natatory chaetigers. Grube (1878) mixed the fragments of two species to describe N. trifasciata. The holotype's lot was here examined, and it consists of two components: one anterior end fragment and one posterior end. The anterior end fragment is an ovate epitokous female with 54 segments (including the apodous segment), dorsal ligules of similar size throughout, dorsal cirri cirriform and medium in size, and pale brown body colour. In contrast, the posterior end fragment is an atokous specimen with 34 chaetigers, somewhat expanded dorsal ligules, long and digitiform dorsal cirri, body of brown colour, and without gametes ( Figure 5(f)). The marked differences imply that the fragments derived from two species. The anterior end is treated as holotype, whereas the extraneous component, possibly belonging to a Perinereis species, is excluded (ICZN 1999, Art. 73.1.5). Neanthes trifasciata is redescribed here based on the anterior end to fix the species definition.
Two of the more relevant features to distinguish N. trifasciata from morphologically similar Neanthes species have been treated differently in the literature. The supposed presence of notopodial homogomph falcigers in N. trifasciata was relevant to separate it from N. unifasciata. Grube (1878) did not address this feature in the original description, although no traces were found in this study, even though the atokous chaetae are not replaced in the incomplete epitokous holotype. Possessing a notopodial homogomph falciger would be an improbable condition. Nereis species with subequal or reducing dorsal ligules towards the posterior end and few paragnaths in a transverse row on areas VII-VIII have notopodial falcigers commonly starting from chaetigers 13-22 but rarely from chaetiger 8 or even from chaetigers 39-40 (see Hutchings & Turvey 1982;Wilson 1985;Dean 2001;Santos & Lana 2003;Núñez 2004;Santos 2007;Ramírez-Hernández et al. 2015). However, it is not yet present in the 53 chaetigers of the holotype, and, therefore, N. trifasciata is retained within Neanthes.
Regarding the second feature, the body pigmentation of N. trifasciata displays a unique pattern among all Neanthes species but this has already faded in the holotype. Grube (1878) described the primary pigmentation of the preserved specimen as follows: "segments … 1st [apodous segment], 3rd with 4th, and 8th entirely reddish-brown" (Grube 1878: 74), where "the 2nd [band] belongs to the 3rd [chaetiger 2] and 4th [chaetiger 3] segment, and it is twice 580 T. F. Villalobos-Guerrero and I. Idris as wide as the others" (Grube 1878: 75). This pattern has not been mentioned elsewhere but was misinterpreted as transverse bands on segments 2-4 (Glasby 2015) or segments 2-5 (Horst 1924), or described differently for specimens from other regions; for instance, all anterior segments with dorsum reddish-brown (Ehlers 1901;Augener 1922), chaetigers 3, 6 and 7 with dorsal brown bands (Fauvel 1951), or even colour absent throughout body (Fauvel 1935). The pigmentation is a relevant feature to distinguish among closely related species when comparing fresh or recently preserved material. Glasby et al. (2013) demonstrated that body pigmentation patterns are diagnostic to distinguish cryptic species based on molecular and morphological evidence. Neanthes trifasciata has been recorded, characterised or both on several occasions based on specimens from distant localities: the southeastern Pacific (Ehlers 1901;Augener 1922;Fauvel 1936;Hartman 1964), the western and central Indian Ocean (Fauvel 1932;Silva 1965;Day 1967;Fishelson & Rullier 1969;Hartmann-Schröder 1974;Amoureux et al. 1978;Cantone 1982), the eastern Indo-Pacific (Fauvel 1930;Augener 1933;Rullier 1972) and the northwest Pacific (Khlebovich & Wu 1962;Wu et al. 1981). Only a few studies addressed the species from regions near the type locality, such as Ambon Island, Indonesia (Augener 1933), or Vietnam and Spratly Islands (Fauvel 1935(Fauvel , 1939; however, none of the specimens in these studies fits entirely with the morphology of the species. The author citation of N. trifasciata has been misinterpreted in the literature. Hartman (1964) doubtfully referred to some Chilean specimens similar to Nereis trifasciata as ?Neanthes trifasciata sensu Ehlers (1901) in lacking notopodial homogomph falcigers but the species was later erroneously adopted as Neanthes trifasciata (Ehlers, 1901) (e.g. Rozbaczylo 1974Wilson 1984;Read & Fauchald 2018a;Hsueh 2019aHsueh , 2019bPamungkas & Glasby 2019). (Horst, 1924)  Diagnosis. Neanthes species with a dark band on dorsum of all segments (absent in long-term preservation); tentacular cirri smooth; area III without laterally isolated paragnaths, area IV without merged paragnaths, areas VI-V-VI ridge pattern πshaped, areas VII-VIII with one band of a few cones; oesophageal caeca absent. Males with prenatatory region consisting of 14 chaetigers, lacking neuropodial postchaetal lobe; natatory region with dorsal cirri multi-papillated and cirrophore of ventral cirri bilobate; falcigers present only in prenatatory region, with convex blades only.
Palpophores ovoid, thick, as long as wide ( Figure  7(c,d)), as long as two-fifths of entire prostomium; without sub-distal transverse groove. Palpostyles bluntly conical, with diameter half as wide as palpophore.
Antennae tapered, slender, long (Figure 7(c)), extending forwards beyond palpophore and posteriorly to distal two-fifths of length of prostomium; antennae close together with gap as wide as onethird of antennal diameter.
Paired eyes blackish, markedly enlarged, arranged in a rectangular form, with both pairs overlapping (Figure 7(b)); anterior pair of eyes ovoid, 3 times as wide as antennal diameter, gap between eyes 1.5 times as wide as diameter of eyes (Figure 7(b)), with lens conspicuous, whitish, covering 40% (60%) of eye; posterior pair of eyes ovoid, 2.7 times as wide as antennal diameter, with lens conspicuous, whitish, placed posterolaterally in eyes and covering 40% (60%) of it.
Pharynx not everted, previously dissected. Jaws not dissected, with at least five well-developed and blunt denticles; pulp cavity with two canals. Pharynx without abnormal development of paragnaths due to epitokal metamorphosis, yellow-amber to translucent on maxillary and oral rings, consisting of conical paragnaths only; plate-like basements and merged paragnaths absent. Area I: 0; areas IIa: 16 and IIb: 17, two slightly regular rows of uneven cones in eyebrow-shaped patch, middle cones larger; area III: 15, three irregular rows of even, barely pigmented cones in oval patch, without distinct laterally isolated cones; areas IVa: 16 and IVb: 16, lemniscate patch with three slightly regular rows of even cones only; area V: 0; areas VIa: 2 and VIb: 2, one transverse row of small, unpigmented cones; areas VII-VIII: 3, one transverse band of cones ventrally well separated. Areas VI-V-VI ridge pattern, π-shaped. Gap between area VI and areas VII-VIII broad, twice as wide as palpophore.
Distribution. The species is known only from the type locality (Nusa Tenggara Timur, Flores, Indonesia).

Ecology. Unknown.
Reproduction. Males undergo marked epitokal metamorphosis, and possibly swarm in the water column and release sperm through pygidial papillae. Females unknown.
Remarks. Horst (1924) described N. trifasciata vandersandei as a variety of N. trifasciata from the Philippines by differences in the body pigmentation. Additional differences were found between these and other species in the present study, and the subspecies is elevated to species level.
Neanthes vandersandei was described briefly, without illustrations, based on an uncertain number of epitokous males collected in the east of Flores Island (Indonesia). Nine syntypes exist at present (Bleeker & van der Spoel 1992), and five of them were available for the present study. Although Horst did not indicate distinct differences among the epitokous material, two species can be recognised in the type series. A couple of male syntypes are regarded as N. vandersandei, and the best-preserved specimen is here selected as lectotype to fix the species definition (ICZN 1999, Art. 74.1, 74.7.3), whereas the other is regarded as a paralectotype (ICZN 1999, Art. 74F). The other three specimens may correspond to a new species, referred hereto as Neanthes cf. vandersandei.
True N. vandersandei is distinguished from N. cf. vandersandei according to the following differences: (1) the 14 pre-natatory chaetigers, in contrast to 13 in N. cf. vandersandei; (2) the llama head-like dorsal cirri of parapodia 6-7, in comparison to those subdistally blunt in N. cf. vandersandei; (3) the crenulated papillae of dorsal cirri starting from chaetiger 16, in contrast to 14-15 in N. cf. vandersandei; (4) the dorsal ligule slightly shorter than the median ligule in pre-natatory chaetigers, in contrast to that longer in N. cf. vandersandei; and (5) the digitiform dorsal and median ligules in the first seven chaetigers, in contrast to conical in N. cf. vandersandei. The description of N. cf. vandersandei is being addressed in a subsequent study.
Neanthes vandersandei resembles N. dawydovi (Fauvel, 1937) from Vietnam and N. egregicirrata (Treadwell, 1924) from Antigua. These species share proximal dorsal ligule of similar size throughout the body (or slightly enlarged in posterior parapodia), absence of notopodial prechaetal, neuropodial superior and postchaetal lobes, spinigers all heterogomph in subacicular neurochaetae, areas VII-VIII with one single transverse row of fewer than 10 paragnaths, area III with paragnaths, and area I bare (Table II).
Neanthes vandersandei can be distinguished from these two species by the following features: (1) the higher number of paragnaths on area III (15), compared with the lower number (up to seven paragnaths; Table II) in N. dawydovi and N. egregicirrata; (2) the absence of paired oesophageal caeca, in contrast to their presence in N. egregicirrata; (3) the conical paragnaths only on area IV, in contrast to both conical and merged paragnaths in N. egregicirrata; (4) the higher numbers of paragnaths on areas II (16-17) and IV (16), in contrast to the lower numbers (areas II: 7-8, IV: 8-9) in N. dawydovi; (5) the long blade (b/a ratio 2.1-4.3) of heterogomph falcigers in anterior chaetigers, in contrast to the short blade (b/a ratio 1.1-1.4) in N. dawydovi; and, (6) the dorsal dark band present on all segments, in contrast to its absence on chaetiger 2 in N. dawydovi.
Concerning the epitokal morphology, N. vandersandei resembles N. augeneri from Vietnam, N. biseriata from Australia and N. nanhaiensis from China. In these species, males have the body divided into two regions, up to 15 pre-natatory chaetigers, first seven pre-natatory parapodia with dorsal cirri modified, and dorsal cirri of natatory chaetigers with several papillae (Table III). However, the males of N. vandersandei can be distinguished from N. biseriata and N. nanhaiensis by the following features: (1) the llama head-like dorsal cirri of parapodia 6-7, in contrast to those sub-distally blunt in N. biseriata and N. nanhaiensis; (2) the presence of a basal end distinctly enlarged on the median ligule of natatory chaetigers, in contrast to its absence in N. biseriata (Table III); and (3) the first five ventral cirri modified, in comparison to the first six modified in N. nanhaiensis.
Some other pharyngeal and parapodial features can separate N. vandersandei from N. biseriata and N. nanhaiensis: (1) the lower number of paragnaths on areas I (0), III (15), V (0) and VII-VIII (3), in contrast to the higher number (area I: 2-8; III: 20-34; V: 9-19; VII-VIII: 58-115; Table II) in N. biseriata; (2) the absence of oesophageal caeca, in contrast to their presence in N. biseriata (Table II); (3) the lack of neuropodial superior lobes, in contrast to their presence in N. biseriata (Table II); (4) the dorsal dark band on all segments, in contrast to pigmentation absent throughout in N. biseriata; (5) the absence of neuropodial inferior lobes, in contrast to their presence in N. nanhaiensis; and (6) the heterogomph spinigers in the subacicular neurochaetae of prenatatory chaetigers, in comparison to those homogomph in N. nanhaiensis (Table II). See the remarks on N. augeneri for a detailed atokous and epitokous comparison with N. vandersandei.
Moreover, N. vandersandei can be distinguished from its similar and previous stem species N. trifasciata based on the following features: (1) the conical paragnaths only on area IV, in contrast to those conical and merged in N. trifasciata; (2) the absence of neuropodial postchaetal lobes, in contrast to their presence in N. trifasciata (Table II); (3) the two paragnaths in a transverse patch on area VI, in contrast to 5-6 in a rounded patch in N. trifasciata; (4) the falcigers with long blade (b/a ratio 4.1-4.3) in the anteriormost chaetigers, in contrast to the shorter blade (b/a ratio 1.5-1.8) in N. trifasciata; (5) the spinigers with long blade (b/a ratio 7.5-14), in contrast to the shorter blade (b/a ratio 5.5-7.2) in N. trifasciata; (6) the anterior margin protruded in the apodous anterior segment of N. vandersandei, in contrast to that rounded in N. trifasciata; (7) the posterodorsal tentacular cirri extending to chaetiger 10, in comparison to those reaching chaetiger 5 in N. trifasciata; and (8) the posterodorsal tentacular cirri twice as long as the antero-dorsal cirri, in contrast to those being 1.3 times as long as the antero-dorsal cirri in N. trifasciata.
Neanthes vandersandei has not been recorded from the type locality or any other region since originally described but is included within Nereis in a few species lists covering Indonesian waters (Hartman 1974;Salazar-Vallejo et al. 2014). Recently, the species was assigned to Neanthes without reasons being given (Hsueh 2019a(Hsueh , 2019b. Nonetheless, owing to the close resemblance of the species to N. trifasciata, and because the presence of two canals in the jaws is a feature also shared with the type species N. vaalii Kinberg, 1865 (three canals in the type species Nereis pelagica Linnaeus, 1758; TFVG pers. obs.), N. vandersandei is a better fit for Neanthes than for Nereis.
Neanthes thysanota (Ehlers, 1920) Nereis thysanota Ehlers, 1920: 51-54  Remarks. Neanthes thysanota was originally described by Ehlers (1868) based on a single epitokous female collected by J. Brock at Batavia (Jakarta Bay, Indonesia) and acquired by the Zoological Institute of the University of Gottingen in the summer of solved, many annelids were transferred to the Zoologischen Museum Berlin in 1929. However, some types were lost during World War II (Hartwich 1993). The type material of the species may be no longer available.
Neanthes thysanota resembles N. pachychaeta, which was described originally from atokous material collected in Madagascar (type locality) and Djibouti, according to two remarkable features: the presence of paragnath base plates on the pharynx, and ramous lobes and lamellae on epitokal parapodia (Ehlers 1920;Horst 1924;Glasby et al. 2011). Glasby et al. (2011) redescribed N. pachychaeta using type and non-type specimens from the Eastern Indian and Western Pacific oceans. Also, they regarded two similar species as junior synonyms, including N. (Ceratonereis) ramosa Horst, 1919, described originally with epitokous specimens from Indonesia and the Philippines. Nonetheless, these authors did not study epitokal topotypes of N. pachychaeta, nor did they describe in detail the types of N. (Ceratonereis) ramosa, particularly the topology of the natatory parapodia, which was instead based upon the brief original description and one epitokous male from Australia. Considering that the thorough study of epitokous morphology has aided us to distinguish nereidid species groups, plus the distance between the type localities and the presence of two similar species in Indonesia with remarkable and unique characteristics, the possibility remains that N. pachychaeta consists of more than one species as currently defined.
Neanthes thysanota differs from N. pachychaeta sensu Glasby et al. (2011) by having a higher number of paragnaths on the areas III (about 13), V (about 4-5) and VII-VIII (two bands of paragnaths) than in the latter species (areas III: 0-8, V: 0, VII-VIII: single band of paragnaths; Table II). Also, the epitokous female of N. thysanota has 27 pre-natatory chaetigers, in contrast to 24 in N. pachychaeta (Table III). Glasby et al. (2011) transferred N. thysanota to Neanthes. Later, Salazar-Vallejo et al. (2014 treated the species within Perinereis in a species list of annelid polychaetes from the Philippine and China seas. Neanthes and Perinereis are still pending redefinition Villalobos-Guerrero 2019;Villalobos-Guerrero et al. 2021;this study). Nonetheless, owing to the close resemblance of the species to N. pachychaeta, and because the absence of a smooth bar on area VI fits the concept of Neanthes much better than Perinereis Kinberg, 1865, we re-allocate N. thysanota to Neanthes.

Discussion
In the present study, four little-known Neanthes species (N. augeneri, N. gisserana, N. trifasciata and N. vandersandei) were redescribed using type specimens in the epitokous stage collected from the Southeast Asian seas. These species are distinguishable from all other similar congeners by several atokous and epitokous diagnostic features. Table II presents a summary comparison of atokous diagnostic features of the 80 valid Neanthes species, and Table III shows the same but for epitokous characters of the 24 species of Neanthes hitherto known to undergo epitokal modifications.

Epitokal morphology in the taxonomy of Neanthes
Nowadays, the distinct patterns of epitokal characters are considered a relevant source of information to distinguish among closely related species (Read 2007;Glasby et al. 2013; Conde-Vela & Salazar-Vallejo 2015; Villalobos-Guerrero & Carrera-Parra 2015; Villalobos-Guerrero & Bakken 2018). Epitokal characters are useful to distinguish nereidid species. Among those related to sexual dimorphism are (1) the presence and (2) the form of papillae in (3) the parapodial cirri and (4) the pygidium, (5) the total number of body regions, (6) pre-natatory chaetigers and (7) first swollen parapodial cirri; some features not directly related to sexual dimorphism are (8) the enlargement and (9) the form of (10) prostomium, (11) the neuropodial postchaetal lobe, (12) the first dorsal cirri and (13) the cirrophore of ventral cirri, as well as (14) the appearance and (15) the size of additional projections in (16) the parapodial cirri and (17) lobes, and (18) the type of epitokous chaetae and (19) their appearance in neuropodial fascicles (Fauvel 1923;Fage & Legendre 1927;Gravier & Dantan 1928;Reish 1954;Clark 1961;Boilly-Marer 1972 Although there are still broad gaps in the knowledge of epitokal modifications in both males and females of several Neanthes species (Table III), we found that the features mentioned above regarding sexual dimorphism (1-7) and some others (9, 12-17) are relevant to distinguish species within the genus. Four additional epitokous characters in natatory chaetigers are recognised in the present study: (20)  Our results show that epitokous characters alone are not informative in all cases, and atokous characters of reproductive forms are also required to describe and distinguish species (Tables II, III). Diagnostic characters in non-metamorphosed Neanthes species include (1) the form and (2) enlargement of dorsal ligules throughout the body; (3) the presence/absence of oesophageal caeca, (4) notopodial prechaetal lobe, (5) neuropodial postchaetal lobe, (6) superior lobe, (7) inferior lobe, (8) articulations in the tentacular cirri and (9) merged paragnaths on area IV; (10) the number of paragnaths and (11) their arrangement mainly on areas I, III, V and VII-VIII; (12) the ridge pattern of areas VI-V-VI; (13) the presence of homogomph spinigers in the subacicular neurochaetae; (14) the form of heterogomph falcigers; (15) the length of the blade of spinigers and (16) heterogomph falcigers; and (17) the pigmentation patterns, when present, on the dorsum. Pamungkas & Glasby (2015) showed that epitokal patterns could be used to recognise nereidid genera according to the modification of body regions and first parapodial cirri, and the presence of dorsal cirri papillae in males. Based on this latter feature, the authors stressed that Neanthes is non-monophyletic. We also conclude, from the variability of the epitokal characters across the genus as shown in Table III, that Neanthes is non-monophyletic. It could be hiding further genera and containing species from currently accepted genera.
Counting the species of Neanthes Hartman (1959) listed 36 valid species in Neanthes. The subsequent list by Fauchald (1972) increased the number to 45 species, and this was later updated by Wilson (1984) to 61 species. In the succeeding years, several species were described, transferred or retained within Neanthes, rendering a total of 88 species (Read & Fauchald 2018a). As a result of taxonomic changes in this paper, Neanthes now comprises 80 valid species (Table I). The genus is well represented in the Southeast Asian seas, with approximately a quarter of the valid Neanthes species originally described from this region (18 species; Table I).

Current status of Neanthes
Neanthes is a polyphyletic genus requiring revision (Bakken & Wilson 2005; this study). Based on current taxonomic concepts, Neanthes differs from Nereis by having homogomph spinigers only in notopodia (falcigers present in posterior chaetigers of Nereis sensu Hartman 1954), and from Perinereis by lacking bar-shaped paragnaths on area VI (present in Perinereis sensu Bakken & Wilson 2005;Villalobos-Guerrero et al. 2021).
An alternative for grouping Neanthes species could be using the enlargement of the proximal dorsal ligule throughout the body, as suggested previously, superseded by two or three relatively constant features -for instance, the presence of neuropodial postchaetal and superior lobes, the presence of homogomph spinigers in the subacicular neurochaetae, or the arrangement of bands of paragnaths on areas VII-VIII (Table II). The following characters, which are rarely included or illustrated in the descriptions, may also be useful for practical purposes to distinguish species: (1) the presence of oesophageal caeca, (2) merged paragnaths on area IV (3) and laterally isolated paragnaths on area III, and (4) the ridge pattern of areas VI-V-VI.
Both the epitokous and atokous characters, where comprehensively available, are valuable sources for testing generic groupings. Further studies on Neanthes species are required to fill the information gaps on both characters' systems and to re-assess those with an unclear or doubtful condition in a particular species. A revision of the genus that involves a phylogenetic analysis using reproductive and non-metamorphosed characters could both delimit the morphology of Neanthes and recognise hidden generic patterns.
provided much-needed facilities to examine their collections or other gathered materials. Robin S. Wilson (Museums Victoria) kindly shared information on and pictures of Neanthes biseriata that were valuable for comparison purposes. Masaatsu Tanaka (Kagoshima University), Rolf Beutel (Friedrich-Schiller-Universität Jena), Sergio I. Salazar-Vallejo and Víctor M. Conde-Vela (ECOSUR), Taeseo Park (National Institute of Biological Resources) and Xuwen Wu (Chinese Academy of Sciences) were very helpful in providing digital copies of several publications. The reading and useful recommendations by Adriana Giangrande and two anonymous reviewers resulted in a much-improved final contribution.

Funding
The first author was supported by the DAAD under grant no. 91673478, and by CONACYT under grants no. 291062, 291212.

Disclosure statement
No potential conflict of interest was reported by the authors.