A new estuarine moray eel, Uropterygius hades sp. nov., is described based on 14 specimens from Japan, Taiwan, the Philippines, southern Indonesia, and Fiji. It is a small-bodied, slender, uniformly dark-brown moray separated from congeners within the U. concolor species complex. The new species can be distinguished from congeners by the anteriorly positioned small eyes (5.0–7.2% of head length), absence of branchial pores, and extended inner rows of teeth which reach the posterior end of the jaws. Uropterygius hades sp. nov. represents a rare species of moray eel that inhabits turbid estuarine environments, preferring soft, muddy substrates, and burrowing and hiding among rocks or in fallen mangrove leaves. Additionally, Uropterygius mactanensis Huang, Balisco, Evacitas & Liao, another species recently separated from the U. concolor species complex, is reported for the first time from Iriomote Island in the Ryukyu Archipelago based on two specimens; this new record expands the geographic range of U. mactanensis from the central Philippines to southern Japan.

DNA barcoding, mangroves, unicolor snake moray, Uropterygiinae, Uropterygius mactanensis

Uropterygius Rüppell, 1838 is the most speciose genus of the subfamily Uropterygiinae (family Muraenidae), comprising 23 of the 38 valid species in the subfamily (Smith 2012; Huang et al. 2023a, 2023b; Fricke et al. 2024). The type species of the genus is Uropterygius concolor Rüppell, 1838, a small-bodied, uniformly brown moray eel described from Eritrea in the Red Sea. Three nominal species described outside the Red Sea have been synonymized with U. concolor, including Anarchias insuetus Whitley, 1932, Anarchias vermiformis Smith, 1962, and Gymnomuraena fusca Peters, 1866 (Böhlke and Smith 2002) (Fig. 1). Consequently, U. concolor was considered a widespread species in the Indo-Pacific region, distributed from South Africa to the Marquesas Islands, north to southern Japan, and south to New Caledonia (McCosker et al. 1984; Fricke et al. 2011, 2018; Delrieu-Trottin et al. 2015).

Figure 1. 

The distribution of nominal species and genetic lineages of the Uropterygius concolor species complex in the Indo-Pacific region. Each color represents a species or genetic lineage. Solid stars indicate the type localities of valid species; empty green stars indicate type localities of the three junior synonyms of U. concolor. Figure modified from Huang et al. (2023a).

However, recent molecular studies based on cytochrome c oxidase subunit I (COI) sequences have revealed that, in addition to the true U. concolor, there are at least five deeply divergent genetic lineages identified as “U. concolor” (Smith et al. 2019; Huang et al. 2023a). These studies suggested that the true U. concolor is currently known only from the Red Sea, while other genetic lineages outside the Red Sea likely represent a species complex. Among these lineages, one was recently described as Uropterygius mactanensis Huang, Balisco, Evacitas & Liao, 2023 from Cebu, the Philippines, while the remaining lineages were only identified as U. cf. concolor 1 (New Caledonia and the Society Islands), U. cf. concolor 2 (South Africa), U. cf. concolor 3 (Western Australia), and U. cf. concolor 4 (Okinawa) (Smith et al. 2019; Huang et al. 2023a) (Fig. 1). Additionally, the synonymization of Anarchias insuetus, A. vermiformis, and Gymnomuraena fusca with U. concolor remain uncertain (Huang et al. 2023a).

Sakai and Sato (1982) provided the first brief description of U. cf. concolor 4 based on two specimens collected from estuaries of the Amami and Okinawa islands in southern Japan. However, they identified it as U. concolor, and this misidentification was perpetuated in all subsequent Japanese reports. Although McCosker et al. (1984) questioned the identification of Japanese specimens by noting the absence of a branchial pore compared to those from the Red Sea, they attributed this variation to different environmental conditions, as they did not find any other morphological differences. According to literature, photos, and specimen records worldwide, U. cf. concolor 4 is mainly found in estuarine mangrove swamps in the Ryukyu Archipelago, such as Amami, Okinawa, Ishigaki, and Iriomote islands (Hatooka and Yoshino 1982; Sakai and Sato 1982; McCosker et al. 1984; Maeda and Tachihara 2006; Kanda et al. 2009; Hibino 2018; Miyake et al. 2019). It has been listed as a Critically Endangered (CR) species in the Red Data Book by the Okinawa Prefectural Government due to its rarity and habitat destruction (Tachihara 2017). However, the molecular characteristics of U. cf. concolor 4 were not available until Miyake et al. (2019) sequenced the mitochondrial genome of two specimens from Okinawa Island. Subsequently, Huang et al. (2023a) used these sequences for comparison and identified diagnostic morphological characteristics for U. cf. concolor 4, suggesting that it is possibly an undescribed species. Furthermore, they examined two specimens from Fiji and southern Java that seem similar to U. cf. concolor 4 and proposed that it may be widely distributed in the central and western Pacific Ocean, in addition to its occurrence in the Ryukyu Archipelago.

In the present study, we conducted detailed examinations of the specimens initially identified as U. cf. concolor 4 and describe it as a new species based on 14 specimens from Japan, Taiwan, the Philippines, southern Indonesia, and Fiji. This new species represents a rare case of a widespread moray eel specifically inhabiting estuarine environments. Additionally, during the examination, we found two specimens from Iriomote Island that can be recognized as U. mactanensis. This finding represents the first record of U. mactanensis in this area and signifies a northward range expansion from the central Philippines to the Ryukyu Archipelago.

Sixteen specimens (14 Uropterygius cf. concolor 4 and two U. mactanensis) were examined, mostly from museum collections, with a few newly obtained samples. Fresh specimens were photographed, and a piece of muscle tissue was obtained from a small incision in the abdomen near the anus. Tissue samples were preserved in 95% ethanol in a −20 °C freezer prior to DNA extraction, while the voucher specimens were fixed in 10% formalin before gradually transferred to 70% ethanol for long-term preservation. Morphological data were collected from the specimens deposited in different institutions, including National Museum of Marine Biology and Aquarium, Pingtung (NMMB-P), Kagoshima University Museum, Kagoshima (KAUM–I), Kitakyushu Museum of Natural History and Human History, Kitakyushu, Fukuoka (KMNH VR), Kyushu University Museum, Fukuoka (KYUM-PI), Okinawa Churashima Foundation, Motobu, Okinawa (OCF; the two URM-P specimens have been donated to OCF), National Museum of the Philippines, Manila (PNM), Australian Museum, Sydney (AMS I), and Zoological Reference Collection, Lee Kong Chian Natural History Museum, Singapore (ZRC). Specimens from the Ryukyu Archipelago identified as “U. concolor” were deposited in various Japanese institutions. In addition to the previously mentioned museums, specimens from Graduate School & Faculty of Bioresources, Mie University, Tsu, Mie (FRLM), National Museum of Nature and Science, Tsukuba, Ibaraki (NSMT-P), and the University Museum, the University of Tokyo, Tokyo (ZUMT), were also examined and confirmed by the second author. However, due to their condition, only 11 specimens (nine U. cf. concolor 4 and two U. mactanensis) were selected for detailed examination.

Morphometrics were measured following Böhlke et al. (1989), and these characters are presented as percentages of total length (TL) or head length (HL). Meristic counts include vertebrae, teeth, and cephalic sensory pores. Vertebral numbers were counted from radiographs, and the vertebral formula is presented as pre-anus, pre-dorsal fin, pre-anal fin, and total vertebrae, following the definitions of Böhlke (1982) with slight modifications. Dentition and head pores were examined under a stereomicroscope, with terminologies following Böhlke and Smith (2002) and Smith et al. (2019), respectively.

DNA was extracted from muscle tissues of the holotype (NMMB-P039570) and a paratype (PNM 15806) of the new species. A fragment of partial COI gene (680 bp) was amplified by polymerase chain reaction (PCR) using the primers FishF2 (5′-TCG ACT AAT CAT AAA GAT ATC GGC AC-3′) and FishR2 (5′-ACT TCA GGG TGA CCG AAG AAT CAG AA-3′) (Ward et al. 2005). Details for PCR thermal cycling conditions, PCR product purification, and DNA sequencing can be found in Huang et al. (2021a). Obtained sequences were manually edited and assembled in MEGA version 11 (Tamura et al. 2021), and the newly generated COI sequences were submitted to GenBank (refer to Fig. 2 for their accession numbers).

Figure 2. 

Maximum-likelihood tree of the Uropterygius concolor species complex based on COI sequences (652 bp). Gymnothorax kidako and G. mucifer (subfamily Muraeninae) are outgroups. Numerals beside the internal branches are bootstrap values, and values below 60 are not shown. (B) = sequence from BOLD Systems; (G) = sequence from GenBank.

All available COI sequences of U. concolor and U. cf. concolor, as well as three sequences of U. mactanensis, from the online databases (GenBank and BOLD Systems) were downloaded for molecular comparisons. After aligning and trimming, a length of 652 bp was retained for analyses. A genetic tree of COI sequences was reconstructed based on the maximum-likelihood (ML) method conducted in MEGA v. 11. The HKY + Γ + I substitution model (Hasegawa et al. 1985) was applied, and a bootstrap analysis with 1,000 replicates was conducted (Felsenstein 1985) for tree building. Sequences of Gymnothorax kidako (Temminck & Schlegel, 1846) and Gymnothorax mucifer Snyder, 1904, from the subfamily Muraeninae, were used as outgroups. Genetic distances of COI sequences between the new species and U. concolor, U. cf. concolor, and U. mactanensis were also calculated using the Kimura 2-Parameter (K2P) model (Kimura 1980).

The absence of branchial pores is an important characteristic of Uropterygius hades sp. nov. Although one examined specimen has a branchial pore on the left side of the head, no other morphological differences were found. Intraspecific variations in head pore number are common in moray eels. For example, in the genus Strophidon McClelland, 1844, a fourth infraorbital pore is a diagnostic characteristic of Strophidon tetraporus Huang & Liao in Huang et al. 2020, distinguishing it from congeners with only three pores. Despite this, a few specimens of Strophidon dorsalis (Seale, 1917), Strophidon sathete (Hamilton, 1822), and Strophidon ui Tanaka, 1918 were observed with a fourth infraorbital pore on only one side of the head (Huang et al. 2020). Similarly, the additional branchial pore in U. hades sp. nov. should be considered a rare intraspecific variation and does not affect the general diagnostics of the taxon.

The salinity at the type locality in the Zhuan River was promptly measured using a refractometer, a day after its collection during a spring tide. Measurements showed a salinity of 5‰ during low tide and 19‰ during high tide. Similarly, surface water salinity at the sampling site in Puerto Princesa Subterranean River (PPSR) was measured at 9‰ during the high tide. These findings support that U. hades sp. nov. is one of the rare cases of moray eels inhabiting brackish-water environments. Another estuarine moray, Echidna rhodochilus Bleeker, 1863, was observed co-occurring with U. hades sp. nov. in Japan, Taiwan, and the Philippines (Masuda and Kobayashi 1994; Huang et al. 2021b; Wada and Hibino 2024; W.C. Huang pers. obs.).

Uropterygius hades sp. nov. seems to be closely associated with mangroves, as most of its known habitats in the Ryukyu Archipelago are mangrove swamps. The habitat on Okinawa Island is an estuary with many fallen leaves from mangrove trees (Fig. 9A). However, U. hades sp. nov. has never been discovered inside the mangrove forest. Instead, it rests in gaps among mangrove aerial roots and leaves (Fig. 9B). When there are fewer fallen leaves, it uses scattered stones as an alternative for concealment. On Amami Island, one collection site is far outside the river mouth and lacks mangrove trees. It consists of a muddy bottom with many gravels and stones, with a tiny freshwater seepage from the land side, similar to the environment of Kakeroma Island. For Philippine specimens, although U. hades sp. nov. was collected inside the PPSR, the entrance of the cave also harbors a few mangrove trees (Fig. 9C), analogous to the estuarine environment of the Matutinao River in Cebu. The Zhuan River, where the holotype was discovered, emerges as another known habitat of U. hades sp. nov. without mangrove trees (Fig. 9D). These observations imply that while mangroves may be important for the survival of U. hades sp. nov., brackish water is a more essential factor.

Figure 9. 

Different habitats of Uropterygius hades sp. nov. A Oura River, Okinawa Island, photographed by Koki Takatsuki B Live photo of a U. hades sp. nov. hiding in gaps among mangrove aerial roots, taken at night at the edge of a mangrove forest zone along the Oura River, Okinawa Island, photographed by Hirozumi Kobayashi C Puerto Princesa Subterranean River, Palawan, photographed by Wei-Cheng Jhuang D Zhuan River, northeastern Taiwan, photographed by Shan-Yu Yang.

One possible interpretation for the preference of U. hades sp. nov. to mangroves as habitat is that mangroves can facilitate the accumulation of fine sediment, thereby creating a soft-mud substrate that may be suitable for U. hades sp. nov. (Perry and Berkeley 2009; Tachihara 2017). We observed that U. hades sp. nov. exhibits tail-first burrowing behavior when kept in an aquatic tank, similar to snake eels (family Ophichthidae) which typically inhabit muddy or sandy substrates. Additionally, the reduction in the number of head pores is hypothesized to help avoid clogging by the substrate, as this phenomenon is observed in certain eel species that inhabit sand and mud burrows (McCosker 1977; McCosker and Randall 2007). Although there are no mangroves directly adjacent to the sampling sites at the Zhuan River and the PPSR, both locations feature muddy, silty bottoms with rocks, similar to the substrates found in mangrove swamps.

Furthermore, we observed that U. hades sp. nov. is highly sensitive to light and consistently attempts to hide when exposed to it. This suggests that it may typically inhabit turbid waters such as estuaries, resulting in its lack of acclimation to light exposure. The small eye proportion of U. hades sp. nov. may also indicate its adaptation to low-light conditions, wherein they primarily use their chemoreception rather than vision to detect prey or avoid predators. A reduction in eye size is also observed in some congeners, such as U. cyamommatus (eye diameter 3.0–4.6% of HL) and Uropterygius oligospondylus Chen, Randall & Loh in Loh et al. 2008 (eye diameter 3.9–7.1% of HL). The former is found in anchialine caves, whereas the latter inhabits intertidal zones consisting of boulders frequently hit by strong waves. Both environments pose challenges to visual perception, making reliance on other senses more important (Loh et al. 2008; Hibino et al. 2020; Koreeda et al. 2020; Huang et al. 2023b).

Combining information from habitat type, body structure, and behavior, we propose that U. hades sp. nov. is an estuarine moray eel that inhabits turbid waters with muddy and soft substrates, using its tail to burrow and hide in sediments, among rocks, or in fallen mangrove leaves. While this study addresses a portion of the U. concolor species complex conundrum, the diversity of these small, uniformly brown moray eels may still be underestimated. For instance, morphological and genetic data reveal the presence of at least three sympatric species found in the Philippines (i.e., U. hades sp. nov., U. mactanensis, and U. cf. concolor 1), while both U. hades sp. nov. and U. mactanensis can be found at Iriomote Island. This high diversity could be triggered by niche segregation and adaptation to different environments among species; for example, U. concolor inhabits shallow fringing reefs, U. mactanensis prefers reef-seagrass interfaces, and U. hades sp. nov. thrives in muddy estuaries (Smith et al. 2019; Huang et al. 2023a). Additionally, a phylogenetic analysis reveals the non-monophyly of the U. concolor species complex, indicating that these morphologically similar moray eels have multiple evolutionary origins (Smith et al. 2019). Further studies are needed to clarify their taxonomic status, as three synonyms of U. concolor and at least three unresolved genetic lineages persist within the species complex.

We thank Hong-Ming Chen, Shing-Lai Ng, Jian-Fu Huang (National Taiwan Ocean University, NTOU), Yung-Chieh Chiu (Fisheries Research Institute, FRI), Heok Hui Tan, Kelvin Lim, Danwei Huang (Lee Kong Chian Natural History Museum, LKCNHM), Seishi Kimura, Taiga Yodo (FRLM), Hiroyuki Motomura, Ryusei Furuhashi, Shintaro Hashimoto, Reo Koreeda (KAUM), Hidetoshi Wada (Kanagawa Prefectural Museum of Natural History, KPM), Munetoshi Maruyama, Noritaka Mochioka (KYUM), Masanori Nakae, Gento Shinohara, Mao Sato (NSMT), Kei Miyamoto (OCF), Ken Maeda (Okinawa Institute of Science and Technology, OIST), and Atsushi Tawa (Japan Fisheries Research and Education Agency, FRA) for their assistance in examining the specimens. We extend our gratitude to Wei-Cheng Jhuang, Shan-Yu Yang (National Sun Yat-sen University, NSYSU), Ryusei Furuhashi (KAUM), Koki Takatsuki (University of the Ryukyus, URM), Hirozumi Kobayashi (Natural History Museum and Institute, Chiba, CBM), Atsushi Tawa (FRA), and Hidetoshi Wada (KPM) for providing information and photo-documenting of the habitats. We also appreciate the logistical support provided by Wei-Cheng Jhuang (NSYSU), Bergenius Shalah, and Elizabeth Maclang (PPSR Management Office) during the sampling in PPSR. Finally, we are indebted to the subject editor Hsuan-Ching Ho, the three reviewers, copy editor Robert Forsyth, and layout editor Polina Petrakieva, all of whom helped improve the manuscript.