The weather loaches of the genus Misgurnus (Cypriniformes, Cobitidae) are freshwater fishes widely distributed throughout Europe and East Asia (Kottelat and Freyhof 2007; Yashima et al. 2023). This genus is defined by the following diagnostic morphological characters: a slender and elongated body; three pairs of maxillary barbels; a suborbital spine hidden beneath the muscles; a rounded edge of the caudal fin; and no regular row of spots (Kim 1997; Kottelat and Freyhof 2007; Nakajima and Uchiyama 2017; Chae et al. 2019; Nakajima and Hashiguchi 2022). Currently, ten valid species are recognized within the genus Misgurnus globally (Vasil'eva 2001; Kottelat 2012; Nakajima and Hashiguchi 2022; Shedko and Vasil'eva 2022; Dyldin et al. 2023; Nakajima et al. 2025).

Recently, taxonomic revisions of the genus Misgurnus have been actively conducted in China, Russia, and Japan through molecular and morphological analyses. Since the 2000s, Russian scientists have described Misgurnus nikolskyi Vasil'eva, 2001 and Misgurnus chipisaniensis Shedko et Vasil'eva, 2022 as new species (Vasil'eva 2001; Shedko and Vasil'eva 2022). Meanwhile, in Japan, Misgurnus amamianus Nakajima et Hashiguchi, 2022 was identified as a new species based on features such as the shape of the lamina circularis in the male pectoral fin, the branching of the last ray of the dorsal fin, and the LPC/DPA ratio (the ratio of the projected length from the posterior edge of the anal fin base to the base of the caudal fin to the projected length from the pelvic fin insertion point to the anterior edge of the anal fin base) (Nakajima and Hashiguchi 2022). The genus Misgurnus is recognized as a taxonomically unstable group characterized by persistent controversies, largely due to frequent identification errors resulting from morphological variations associated with sex, life history, and environmental factors (Kottelat 2012; Han et al. 2013; Shedko and Vasil'eva 2022; Yashima et al. 2023). To address these challenges, recent studies describing new species utilize shape-based geometric morphometric techniques in addition to traditional meristic and morphometric comparisons (Sequeira et al. 2011; Sidlauskas et al. 2011; De La Cruz-Agüero et al. 2015). To this day, numerous studies remain underway globally to differentiate morphologically similar cryptic species.

In contrast, taxonomic studies of the genus Misgurnus in the Korean Peninsula have been limited. To date, only three species have been reported in this region: Misgurnus dabryanus (Guichenot, 1872) (of which Misgurnus mizolepis Günther, 1888 is a junior synonym according to Kottelat 2012) and Misgurnus anguillicaudatus (Cantor, 1842) in South Korea, and Misgurnus buphoensis Kim et Park, 1995 in the lower reaches of the Tumen River in North Korea (Yang et al. 1994; Kim and Park 1995; Kim and Park 2002; Chae et al. 2019). According to the studies in Russia (Vasil'eva et al. 2003; Perdices et al. 2012), there have been consistent suggestions that previously unrecorded Misgurnus species may inhabit the Korean Peninsula. This potential distribution is linked to the biogeographical connections between Korean waters and the Russian Amur River drainage (excluding Khanka Lake), northeastern Mongolia, and northeastern China, extending south to the upper reaches of the Liao River (Vasil'eva et al. 2003; Perdices et al. 2012).

In this study, while investigating the species of the genus Misgurnus in South Korea, a distinct group was identified in the rivers of Goseong-gun, Gangwon-do, which differed from the three species previously known to inhabit the Korean Peninsula. Morphological comparisons and mitochondrial DNA analysis confirmed these specimens as Misgurnus mohoity (Dybowski, 1869), a species previously known to be distributed in the Amur River basin across China, Mongolia, and Russia (Shedko and Vasil'eva 2022). This study provides the first specimen-based confirmation for the hypothesis of Perdices et al. (2012), who treated Misgurnus bipartitus (Sauvage et Dabry de Thiersant, 1874) as a junior synonym of M. mohoity and suggested a distribution extending from the Amur River basin and northeastern China to the Korean Peninsula. Consequently, by reporting the unrecorded species M. mohoity in Korea, this study provides essential baseline data for understanding the dispersal and differentiation history of the genus in Northeast Asia and presents a taxonomic re-evaluation of Korean Misgurnus species.

Sampling. A total of 61 specimens comprising two Misgurnus species were examined in this study. Specifically, 22 specimens of M. mohoity were collected from Goseong-gun (Fig. 1A), and 39 specimens of M. anguillicaudatus were collected from Eumseong-gun, Iksan-si, Sinan-gun, and Jeju-do in South Korea (Fig. 1B–E). All specimens were collected using a baited trap (commercial fish feed; 2 × 2 mm) and hand net (4 × 4 mm). Prior to fixation, all specimens were anesthetized with MS-222 to minimize distress and physical distortion. The specimens were then fixed in 10% neutral formalin and subsequently preserved in 70% ethanol. For genetic analysis, the right pelvic fin was sampled and fixed in 99.5% ethanol. To minimize growth-related morphological variation and account for the development of secondary sexual characteristics, only specimens of at least 70 mm of the standard length (SL) were collected (Yashima et al. 2023).

Morphological analysis. Meristic and morphometric methods followed Hubbs et al. (2004), Kottelat and Freyhof (2007), and Nakajima and Hashiguchi (2022) (Fig. 2). All measurements were taken using a digital caliper (Mitutoyo, CD-20CX, Japan) and recorded to the nearest 0.01 mm. Morphometric data are expressed as percentages of standard length (SL) and head length (HL) in Table 1, which presents the values for each population along with the number of specimens observed for each measurement. Fin ray and vertebral counts were obtained from soft X-ray radiographs (Hitex, HA-80, Japan), following the methodology of Sawada (1982). To identify the features of the lamina circularis, specimens were stained with Alizarin Red S following the protocol of Taylor and Van Dyke (1985) and subsequently preserved in 100% glycerin. Description of external morphology and the lamina circularis followed Nakajima and Uchiyama (2017) and Yashima et al. (2023). Statistical analyses were performed using SPSS version 29.0 (IBM, Armonk, New York, USA). Principal Component Analysis (PCA) was conducted to compare morphological variations among the five populations. Following verification of normality and homogeneity of variances, differences in factor scores were analyzed using one-way ANOVA and Duncan’s post-hoc test (P < 0.05).

DNA extraction and phylogenetic analysis. DNA was extracted from fin clips using a genomic DNA extraction kit (Bioneer, Korea). The mitochondrial cytb gene was amplified by polymerase chain reaction (PCR) using universal primers described by Palumbi (1991): F (5'- TGA CTT GAA GAA CCA CCG TTG-3') and R (5'- GGA TTA CAA GAC CGA TGC TTT-3'). The total reaction volume for PCR was 50 μL, containing 2 μL of template DNA, 1 μL of each primer (10 μM), 24 μL of PCR mastermix (Takara, Japan) and 22 μL of deionized water. PCR cycling conditions included an initial denaturation at 95°C for 5 min, followed by 30 cycles of denaturation at 94°C for 30 s, annealing at 56°C for 30 s, and extension at 72°C for 1 min, with a final extension at 72°C for 7 min. After checking the quality of the PCR products by electrophoresis, they were purified using a purification kit (Bioneer, Korea) according to the manufacturer’s protocol. The final PCR products were sequenced using an ABI-3730XL sequencer (Applied Biosystems, USA).

Phylogenetic reconstruction and genetic divergence. For phylogenetic reconstruction, the generated cytb sequences were subjected to manual inspection and multiple sequence alignment using Clustal Omega. The taxonomic positioning of the Korean M. mohoity was evaluated by reconstructing evolutionary relationships using the maximum likelihood (ML) method implemented in MEGA12. Prior to tree construction, the optimal nucleotide substitution model was identified by selecting the one with the lowest Bayesian Information Criterion (BIC) score. To assess the reliability of the resulting phylogeny, nodal support was estimated via bootstrap analysis with 1000 replicates. Additionally, genetic divergences between lineages were quantified using the Kimura 2-parameter (K2P) distance model. The final phylogenetic tree was rooted using Sabanejewia balcanica (Karaman, 1922) (family Cobitidae) and Pangio pangia (Hamilton, 1822) (both species: family Cobitidae) as designated outgroups to provide evolutionary orientation.

Family Cobitidae Swainson, 1838

Genus Misgurnus Lacépède, 1803

In this study, 22 specimens collected from the river in Goseong-gun, Gangwon-do, Korea, were identified as Misgurnus mohoity through morphological and genetic analyses. M. mohoity is morphologically similar to M. anguillicaudatus, which was previously known to inhabit Korea, but it can be clearly distinguished through detailed morphological analysis. The biggest diagnostic difference among meristic characters is that M. mohoity exhibits distinct differences from the M. anguillicaudatus group, with no overlap in total vertebrae (48–50 vs. 43–46 in M. anguillicaudatus), caudal vertebrae number (19–20 vs. 17–18 in M. anguillicaudatus), and lateral-line scales (164–180 vs. 144–161 in M. anguillicaudatus) (Table 1).

Furthermore, the lamina circularis in mature males is considered one of the most important diagnostic traits for species identification within the genus Misgurnus (see Vladykov 1935; Kubota 1961; Kim 2009; Nakajima and Hashiguchi 2022; Shedko and Vasil'eva 2022). Although M. mohoity and M. anguillicaudatus both possess morphologically similar poleaxe-shaped lamina circularis, the two species exhibit subtle differences in the neckline at the base; this feature is shallow in M. mohoity, whereas it is deep in M. anguillicaudatus. This poleaxe-shaped structure is clearly distinguishable from the narrow and elongated rod-shaped lamina circularis of M. mizolepis and the spatula-shaped one of M. buphoensis (Table 2).

These morphological differences were statistically validated through multivariate analyses (PCA and CDA). Misgurnus mohoity exhibits a relatively more slender and elongated caudal peduncle compared to M. anguillicaudatus, resulting in a complete separation in the morphospace (Figs 6, 7). Specifically, the CPL/CPD ratio (more than 1.7 vs. less than 1.7) and the LPC/DPA ratio (more than 1.1 vs. less than 1.1) serve as significant diagnostic indicators for differentiating the body shapes of the two species.

This finding provides the first specimen-based confirmation of the hypothesized presence of M. mohoity on the Korean Peninsula through actual specimens. This hypothesis was previously proposed by Russian scholars based on the biogeographical connectivity between the Korean Peninsula and the Amur River system (Vasil'eva et al. 2003; Perdices et al. 2012). Unlike the M. mohoity in Tokyo, Japan, which is classified as an alien species introduced as bycatch among Misgurnus loaches imported from China for bait or consumption (Nakajima et al. 2025), the Goseong-gun population is considered indigenous.

The indigenous status of this population reflects the historical connectivity between the Amur River and the Korean Peninsula within the ‘transition’ zoogeographical zone characterized by the absence of a major faunal break between the two regions (Bogutskaya et al. 2008). This is supported by the established biogeographical division of the Korean Peninsula, which categorizes the northern East Coast of Korea (including Goseong) into the “East north Korea Subdistrict”—a region defined by its high faunal affinity with the Amur River system (Kim 1997). Consequently, these findings provide crucial biogeographical evidence for understanding the dispersal and differentiation of the genus Misgurnus in Northeast Asia. However, due to the current lack of research on the distribution of M. mohoity within the Korean Peninsula, information regarding its precise biogeographical distribution and dispersal routes is limited. Therefore, subsequent studies are necessary to elucidate the evolutionary and dispersal history of M. mohoity in Northeast Asia.

Mitochondrial DNA (mtDNA) analysis results further support the morphological diagnosis and biogeographical hypotheses. Misgurnus mohoity from Korea was recovered in a monophyletic clade with populations from Russia, China, and North Korea (Fig. 8), showing significant genetic divergence (K2P distance: 10.04%) from M. anguillicaudatus populations. This indicates that, despite their superficial similarities, the two species have undergone independent evolutionary processes over a long period.

The M. mohoity specimens collected in this study clustered with Russian M. mohoity and Chinese M. bipartitus populations. These results genetically support the taxonomic view (Vasil'eva 2001; Kottelat 2012) that considers M. bipartitus to be a synonym of M. mohoity. Vasil'eva (2001) concluded through morphological comparisons that M. bipartitus in Chinese literature is actually M. mohoity, while the species previously misidentified as M. mohoity is the new species, M. nikolskyi. Although both names are still used interchangeably in some Chinese literature, these findings clearly demonstrate that M. bipartitus in Chinese records taxonomically corresponds to M. mohoity.

Based on clear differences in meristic characters, statistical separation, and high genetic distance, the population in Goseong-gun, Gangwon-do is confirmed as M. mohoity, representing the first record of this species in Korea. Accordingly, the Korean common name ‘ga-neun-mi-ggu-ri’ is proposed for this species, based on its most distinctive morphological feature-a slender and elongated body shape characterized by high CPL/CPD and LPC/DPA ratios.