A new species of free-living marine nematode, Fotolaimus cavus sp. nov. (Nematoda, Oncholaimida, Oncholaimidae), isolated from a submarine anchialine cave in the Ryukyu Islands, southwestern Japan

Fotolaimus cavus sp. nov. was described from a submarine anchialine cave called Akuma-no-yakata on the Shimoji Island,

In a recent classification system by Hodda (2022), the family Oncholaimidae belongs to the order Oncholaimida Siddiqi, 1983= suborder Oncholaimina De Coninck, 1965= superfamily Oncholaimoidea Filipjev, 1916with two other families, i.e., Enchelidiidae Filipjev, 1918and Thalassogeneridae Orton Williams & Jairajpuri, 1984.Thalassogeneridae is a terrestrial family that includes only the type genus Thalassogenus Andrássy, 1973, and several authors do not include Thalassogeneridae in Oncholaimoidea based on morphological data (Jensen 1976;Lorenzen 1981).Since molecular data do not allow any conclusion, only Oncholaimidae and Enchelidiidae are considered members of Oncholaimoidea sensu stricto.Several molecular phylogenetic analyses strongly support the monophyly of the clade composed of Oncholaimoidea sensu stricto (Meldal et al. 2007;van Megen et al. 2009;Bik et al. 2010a, b;Pereira et al. 2010;Smythe 2015;Smythe et al. 2019;Ahmed et al. 2022).However, analyses of ribosomal small subunit RNA sequences (Meldal et al. 2007;Bik et al. 2010a, b;Smythe 2015) and whole-genome/transcriptome (Smythe et al. 2019;Ahmed et al. 2022) do not support monophyly of Oncholaimidae.The monophyly of the seven subfamilies comprising Oncholaimidae (for three of which there are no molecular data) is also not supported by molecular analyses (Bik et al. 2010a, b;Smythe 2015), and phylogenetic relationships within Oncholaimoidea that have been supported by morphological analyses (e.g., Smol et al. 2014;Hodda 2022) are considered to require significant revision.

Sampling and morphological observation
Sediment samples were collected by scuba diving from a completely dark, anchialine zone at 7-20 m depth ("second slope zone," Osawa and Fujita 2019) in the submarine cave Akuma-no-yakata, Shimoji Island, Miyako Island Group, Ryukyu Islands, southwestern Japan (24°49'22.5"N,125°08'07.8"E)on 26 Oct. 2018.Thirteen nematode individuals were isolated from the sediments and preserved in DESS solution (Yoder et al. 2006).We considered that these individuals belonged to the same species in the family Oncholaimidae.Three males and four females were permanently mounted in anhydrous glycerin (Shimada et al. 2021), observed using a BX51 light microscope (Olympus, Japan) with differential interference contrast, and photographed with an PCM500 digital camera (AS ONE, Japan).One male and one female were dried using the hexamethyldisilazane method (Nation 1983), sputter-coated with gold to a thickness of 20 nm, and observed using an S-3000N scanning electron microscope (SEM; Hitachi, Japan).We edited digital photographs using GIMP ver.2.10 (https://www.gimp.org) and generated measurements and drawings from digital photographs using Inkscape ver.1.0 (https://inkscape.org).We deposited all specimens examined in the Invertebrate Collection of Hokkaido University Museum (ICHUM).The terminology used to describe the arrangement of morphological features such as setae follows that of Decraemer et al. (2014).The following de Man's ratios (Hooper 1986) were used: a, ratio of body length to maximum body diameter; b, ratio of body length to pharyngeal length; c, ratio of body length to tail length; c', ratio of tail length to body diameter at cloacal opening or anus; and V, position of vulva from anterior body end expressed as percentage of body length.

Taxonomy
Remarks.Cobb (1930) established the new genus Oncholaimium Cobb, 1930, which differs from Oncholaimus mainly by the presence of a distinct precloacal appendicule (papilla) in males.Subsequently, Kreis (1932) established Pseudoncholaimus Kreis, 1932 based on the lack of Demanian system.Kreis (1934) also distinguished Oncholaimium from Oncholaimus based on the Demanian system, which is without terminal pore in Oncholaimium and with terminal pores in Oncholaimus.However, Rachor (1969) synonymized Oncholaimium and Pseudoncholaimus to Oncholaimus, because the presence of a Demanian system is unknown for many species of Oncholaimus.Belogurov and Belogurova (1988), who proposed the currently accepted taxonomic system of Oncholaimidae, treated Oncholaimium and Pseudoncholaimus as valid genera.Pseudoncholaimus is still considered valid by several researchers (e.g., Smol et al. 2014;Tsalolikhin 2015;Milovankina and Fadeeva 2019).However, because it is unclear for numerous species whether they rather belong to Oncholaimus, Oncholaimium, or Pseudoncholaimus, we included the latter two in Oncholaimus sensu lato as described by Rachor (1969).Thus, the subfamily Oncholaiminae comprises six genera, i.e., Oncholaimus, Metoncholaimus, Prooncholaimus Micoletzky, 1924, Metaparoncholaimus De Coninck &  Remarks.The genus Fotolaimus can be distinguished from Metaparoncholaimus and Wiesoncholaimus by the left ventrosublateral tooth being larger than the right ventrosublateral tooth (left and right ventrosublateral teeth are equal in size in Metaparoncholaimus and Wiesoncholaimus).It also differs from Prooncholaimus by the absence of the large bubble-like cells in pseudocoelom (presence in Prooncholaimus).Fotolaimus, Oncholaimus, and Metoncholaimus are similar to each other and are distinguished by the morphological characters of the terminal ducts of the Demanian system.Belogurova and Belogurov (1974) characterized the Demanian system of Fotolaimus as follows: the posterior end of the main duct forms two symmetrical sacs, each of which is pierced by five terminal ducts ending in terminal pores.On the other hand, the main ducts of Oncholaimus and Metoncholaimus do not form a sac at the posterior end, but branch separately into two or more terminal ducts (Belogurov and Belogurova 1988).Belogurov and Belogurova (1977) distinguished Metoncholaimus from Oncholaimus in that the terminal ducts are covered with moniliform glands.However, some Metoncholaimus species do not mention the shape of the terminal duct in their description (e.g., Mawson 1958;Salma et al. 2017).For convenience, a species may be considered belonging to Oncholaimus if it has short (equal to cloacal body diameter) spicules and without gubernaculum, and belonging to Metoncholaimus if it has longer spicules (with or without gubernaculum) or short spicules with gubernaculum (cf.Platt and Warwick 1983).The distinction between Oncholaimus and Metoncholaimus will need to be reexamined in the future.
Key to genera in Oncholaiminae (cf.Belogurov and Belogurova 1988;Smol et al. 2014).Other material.Japan • one male (SEM specimen); same collection data as for paratypes; • one female (SEM specimen); same collection data as for preceding.
Etymology.The specific name cavus (cave) is a Latin noun in apposition derived from the type locality.
Description.Body (Fig. 1A, B) elongated, almost cylindrical but gradually tapering toward both extremities.Cuticle smooth throughout body besides oblique striations (Fig. 2A; cf.Leduc 2013) crossing at angle of ca.120° between amphids and anteriormost cervical setae visualized using SEM.Somatic sensilla arranged in eight longitudinal rows: setiform in anterior half of cervical (Figs 1C, 2B) and caudal (Fig. 2C) regions; papilliform or very short setiform in rest of body (Fig. 2D), difficult to observe without SEM.Cephalic region (Figs 1D-F, 2E-G, 3A-D) rounded at anterior end, with six lips, slightly     3H, I,  4A, C) plate-like, distally associated with spicules, proximally curved, shape similar to that of gubernaculum from Admirandus belogurovi Tchesunov, Mokievsky & Nguyen, 2010(Tchesunov et al. 2010), as long as 0.5-0.7 cloacal body diameters or 0.3-0.5 spicule lengths.No precloacal supplement.Circumcloacal setae (Fig. 2C, K) arranged in single circle, 9-12 setae in each body side.Three ventral papillae (Fig. 2L) present just anterior to cloacal opening, arranged in single longitudinal row, observed only with SEM.Male reproductive system (Fig. 4D) diorchic.Two opposed and outstretched telogonic testes, both situated on right side of intestine: germinal zone filled with small cells not arranged in rows; growth zone with larger cells arranged in single row.Blind end of anterior testis at 30%-40% of body length from anterior end; blind end of posterior testis at 70% of body length from anterior end.Seminal vesicle including two (dorsal and ventral) rows of sperms (Fig. 3J).Vas deference located on ventral side of intestine, anteriorly filled by single row of sperms.
Female reproductive system (Fig. 4E) monodelphic-prodelphic.Telogonic ovary antidromously reflexed, situated on right side of intestine, as long as 6%-10% of body length; anterior end connecting with oviduct at 50%-55% of body length from anterior end; blind end located at 60%-65% of body length from anterior end; germinal zone with small cells not arranged in rows; growth zone with gradually growing cells arranged in two (right and left) rows; ripening zone with grown oocytes arranged in single row.Oviduct situated on left side of ovary (between intestine and ovary).Uterus well developed in two specimens (ICHUM 8477 and 8478), ca.200 μm in length, each containing one egg (131-141 μm long and 48-62 μm wide).In other two individuals, uterus very short without egg and blind end of ovary located just anterior to vulva.However, the size and degree of maturity of the ovaries similar to those of egg-bearing specimens.Vulva (Figs 2M, 4E) transverse slit with thick cuticular walls, located at 2/3 from anterior body end.Vagina weakly sclerotized.Demanian system (Figs 3K, 4F, G) oncholaimoid type variant D (cf.Belogurov and Belogurova 1988).Ductus uterinus anteriorly inconspicuous.Uvette well-developed, spherical with thick sclerotized wall, situated on right side of intestine at 80%-85% of body lengths, connected to main duct with posterior tip.Main duct on dorsal side of intestine, anteriorly forming one short sac with thick sclerotized wall and filled with sperms, posteriorly forming two sacs on both sides.Ductus entericus very short, situated subterminally on the left side of anterior sac of main duct.Osmosium inconspicuous, may be simple pore.Terminal ducts branching off from posterior sacs of main duct, five or more in number on both sides.Terminal pores (Figs 2N, O, 3L, 4F, G) arranged in single circle surrounding body, 3-4 anal body diameters anterior to anus.Diagnosis.Fotolaimus cavus sp.nov. is characterized by small body size (2.1-3.0 mm), wide amphids (0.35-0.45 corresponding body diameters), a long buccal cavity (length/ width = 2.5-3.0), a long (c = 12-15, c' = 3.9-6.1)conicocylindrical tail, and a proximally curved gubernaculum.
The DNA sequences isolated from Fotolaimus species were deposited in the INSD for the first time.
The ML tree based on 18S sequences (Fig. 5) suggested, albeit with low support values, the presence of two major clades within the Oncholaimoidea.The first major clade was composed of two highly supported subclades: all sequences from Adoncholaimus Filipjev, 1918 (SH-aL-RT = 100% and UFBoot = 100%), and all sequences from Viscosia de Man, 1890 and several sequences from Oncholaimus, including Pseudoncholaimus (SH-aLRT = 97.3% and UFBoot = 99%).Oncholaimus from South Africa (SBA) and the Atlantic side of the USA (BUS, NAR, NUS, and OUS) did not form clades. Oncholaimus from the UK (AUK) and Japan placed in another major clade.Within the genus Viscosia, 11 sequences from the UK (AUK, HCL, LUK, and SBN) formed one clade, which did not include V. dossena Leduc & Zhao, 2023

Discussion
The 18S phylogenetic tree (Fig. 5) suggested that Fotolaimus cavus sp.nov.was assembled in one major clade with some members of Oncholaimus and all members of Wiesoncholaimus and Oncholaimellinae sp. with a high level of support (SH-aLRT = 96.3% and UFBoot = 97%).Fotolaimus and Wiesoncholaimus are included, with Oncholaimus, in the subfamily Oncholaiminae (Hodda 2022), and they all present an oncholaimid-type Demanian system (Belogurov and Belogurova 1988).Therefore, the position of Fotolaimus in the Oncholaiminae based on the morphology was supported by molecular phylogeny.
Oncholaimus sp.(KR265044) from Wakayama (Pacific side of central Japan) is considered to be O. secundicollis because DNA sequences obtained from the former was identical to those from the topotypes of O. secundicollis.Oncholaimus secundicollis is distributed on the Pacific and Sea of Japan sides of northeastern Japan (Shimada unpubl.)and on the Sea of Japan side of South Korea (Lee et al. 2015).Wieser (1955) also reported on O. dujardinii de Man, 1876 from Wakayama but provided limited morphological information and no illustration.A redescription of O. dujardinii by the same author (Wieser 1953) mentions a gubernaculum (Zhang and Platt 1983), which should be absent in Oncholaimus.Therefore, O. dujardinii sensu Wieser (1955) may not be a true Oncholaimus.
Meyersia branch has been assembled (SH-aLRT = 90.7% and UFBoot = 90%) with the Oncholaimus-Wiesoncholaimus-Fotolaimus clade.Meyersia did not form the clade with Adoncholaimus, indicating that monophyly of Adoncholaiminae is unlikely.Members of Adoncholaiminae possess two ovaries and a well-developed Demanian system, but other morphological features distinguish Meyersia from the other three genera.Adoncholaimus (including Metoncholaimoides Wieser, 1954), Admirandus Belogurov &Belogurova, 1979, andKreisoncholaimus Rachor, 1969 have larger teeth on the right side, and the terminal pores of the Demanian system are located near the anus (i.e., much posterior to both ovaries).In contrast, in Meyersia, right and left teeth are large, and the terminal pores of the Demanian system are located at the level of the vulva (i.e., between both ovaries).
Monophyly of Enchelidiidae was not supported by our phylogenetic tree, although previous studies by Bik et al. (2010a, b) evidenced monophyly with 95% of bootstrap Some members of Oncholaimus and all members of Pseudoncholaimus and Viscosia appear to be monophyletic (SH-aLRT = 97.3% and UFBoot = 99%).As aforementioned, Pseudoncholaimus is considered a junior synonym of Oncholaimus, but both taxa can be distinguished by the presence or absence of Demanian system.Our tree strongly suggested that unidentified Oncholaimus spp. was not clustered in a single clade.The species of Oncholaimus clustered with Pseudoncholaimus might not have a Demanian system.In fact, the species that doubtlessly had a Demanian system (O.secundicollis) belonged to a separate clade from Pseudoncholaimus.Fotolaimus and Wiesoncholaimus, suggested to be closely related to O. secundicollis, also have a Demanian system.Therefore, there was no evidence supporting the synonymization of Pseudoncholaimus with Oncholaimus.Because members of Oncholaimus (in which the left tooth is the largest) are included in two well-supported clades, which both contain Oncholaimellinae spp.(in which the right tooth is the largest), it is likely that the size of the left and right teeth does not reflect phylogenetic relationships.Additionally, Wiesoncholaimus, in which left and right teeth are large, was in the same clade as Oncholaimus and Fotolaimus, suggesting that the teeth size has evolved independently many times.
In Bik et al. (2010a, b) and Smythe (2015), the phylogenetic position of Adoncholaimus within Oncholaimoidea was not clear.Our phylogenetic analysis suggested that Adoncholaimus is a sister clade of Oncholaimus-Pseudoncholaimus-Viscosia clade with a certain degree of support (SH-aLRT = 78.9% and UFBoot = 91%).

Figure 5 .
Figure 5. ML tree of Oncholaimoidea based on 18S DNA sequences.Numbers at the nodes are SH-aLRT (left) greater than 70% and UFBoot (right) greater than 80%.values.Monophyly of the clade comprising Calyptronema and Symplocostoma was well supported (SH-aLRT = 94.8% and UFBoot = 97%).Both genera present a sexually dimorphic cephalic region, elongated spicules, and papilliform (not winged) precloacal supplements, and have been considered, together with Symplocostomella Micoletzky, 1930, which has the same three characteristics, closely related groups within Enchelidiidae.Pontonema appeared to be a sister group of the clade consisting of Eurystomina and Pareurystomina Micoletzky, 1930; however, the support values are quite low (SH-aLRT < 70% and UFBoot < 80%), consequently, the position of Pontonema remains uncertain.Some members of Oncholaimus and all members of Pseudoncholaimus and Viscosia appear to be monophyletic (SH-aLRT = 97.3% and UFBoot = 99%).As aforementioned, Pseudoncholaimus is considered a junior synonym of Oncholaimus, but both taxa can be distinguished by the presence or absence of Demanian system.Our tree strongly suggested that unidentified Oncholaimus spp. was not clustered in a single clade.The species of Oncholaimus

Table 1 .
List of nematodes sequenced from Japanese waters with INSD accession numbers.*Type locality.

Table 2 .
List of nematode sequences from the INSD including the phylogenetic analysis.Abbreviations: A = Atlantic Ocean side; FS = French Southern and Antarctic Lands; nd = no data; P = Pacific Ocean side.*Outgroup.Oncholaiminae.Left ventrosublateral tooth larger than the two other teeth.Spicules shorter than 2.0 cloacal body diameters.Gubernaculum present.Demanian system present, posterior end of main duct forming two symmetrical sacs each with five or more terminal ducts.

Table 3 .
Morphometrics of Fotolaimus cavus sp.nov.All measurements are in μm and in the form: mean ± s.d.(range).*Distance from the anterior body end.