Specimens (NMMB-CD006299-006302, Table 1) were purchased at the Zhengbin fishing port in Keelung, Taiwan, on 28 January 2023. According to the voyage records, the Bathynomus were caught in the South China Sea about 150 km northeast of the Paracel Islands (Xisha Islands in Mandarin, 19.08333 N, 115.25 E). The specimens were taken from waters outside the protected area, preserved in ice aboard the ship and stored at -20°C in the laboratory.

Four Bathynomus specimens (Table 1) have been deposited and registered in the National Museum of Marine Biology, Checheng, Taiwan (registration code: National Museum of Marine Biology, NMMB, from NMMB-CD006299-006302).

Abbreviations. RS—robust seta/e; TMCD—National Taiwan Museum code, NMMB— National Museum of Marine Biology, Checheng, Taiwan; TL—total length; CL—cephalic length.

The specimen and dissected body parts (holotype, voucher singular: NMMB-CD006302, Fig. 1) were photographed using a high-resolution monocular camera (Canon EOS 90D, Tokyo, Japan). A digital camera (Sony MEX-5R, Tokyo) on a stereomicroscope (Olympus, SZ61, Tokyo) was used for illustration. The software CLIP STUDIO PAINT (https://www.clipstudio.net/ja/functions/) and Adobe Illustrator CS (Adobe Inc., San Jose, CA, USA) were used to make a line drawing illustration from an image file of a digital camera. The observed specimens were measured using a stereomicroscope's eyepiece micrometer. Dimensions from the measurements were labelled on each illustration. The terminology of body parts follows Huang et al. (2022). Total length (TL) and cephalic length (CL) were measured from the anterior tip of the head to the posterior end of the pleotelson plate with a central spine, respectively.

Figure 1.  

Holotype of Bathynomus paracelensis sp. nov. (female, voucher singular: NMMB-CD006300, 220 mm. The South China Sea, the water of Paracel Island (19.0833 N, 115.25 E), coll. Ming-Chih Huang, 28 January 2023). A dorsal view, the lower right corner is an ovum, slightly oval and about 12-15 mm in diameter; B ventral view. Scale bars: 1 cm.

Total genomic DNA of Bathynomus samples (NMMB-CD006299-006302) were extracted from ca. 25 mg of pereopod muscle resected from four specimens from off Paracel Islands waters, using a commercial genomic DNA extraction kit (QIAamp DNA Mini Kit, Hilden, Germany) according to the manufacturer’s protocol. PCR primers used for the amplification were designed, based on the sequences of the genes encoding COI (Folmer et al. 1994) and 16S ribosomal RNA (Palumbi et al. 1991). In addition, we also used our own designed primers as Kmae (F) and Kushi (R) for the test (Table 2). The four samples were all sequenced for COI and 16S rRNA (Table 1).

Amplification using the COI and 16S rRNA primers was based on a cycle of denaturation at 94ºC for 30 s, annealing at 48ºC for 40 s and extension at 72ºC for 30 s using a DNA thermal cycler model MyCyclerTM Thermal Cycler System (#1709703, Bio-Rad, Hercules, CA, USA). This procedure was carried out for 35 cycles and the final extension step was performed at 72ºC for 10 min. The 100 μL reaction medium contained 200 nM dNTPs, 10 mM each of forward and reverse primers, 2 units of Ex-Tag DNA polymerase (TaKaRa Ex Taq® DNA Polymerase, Takara Bio, Shiga, Japan), 10 μL of 2× Ex-Tag DNA polymerase buffer (Takara Bio) and 50 ng of genomic DNA. The PCR products were subjected to electrophoresis using 1% agar (VWR Funding Inc, West Chester, PA, USA) and visualized with SYBR Green (HealthView Nucleic Acid Stain, Thermo Fisher Scientific, Waltham, MA, USA). After confirming the success of PCR amplification, the products were sent to Biotech (Genomics, Xizhi District, New Taipei City, Taiwan) for sequencing. The obtained sequences were edited and aligned using editing software BioEdit 7.2 (https://bioedit.software.informer.com/7.2/) and Multiple Sequence Alignment (Clustal Omega – GenomeNet, Hinxton, Cambridgeshire, UK).

Comparisons of the edited and aligned COI and/or 16S rRNA sequences of the present specimens and seven reported sequenced species of Bathynomus were performed using Molecular Evolutionary Genetics Analysis 11 (MEGA 11) software (Tamura et al. 2021). COI sequence data were obtained from the National Center for Biotechnical Information (NCBI) for B. jamesi (KX417647) (from the sea off the southern part of Hainan Island, China, Kou et al. (2017)), B. kensleyi (OQ860751) (from Marion Plateau, Coral Sea, QLD, Australia, Huang et al. (2022)), B. yucatanensis (MZ354630) (from the Gulf of Mexico off the Yucatan Peninsula, Huang et al. (2022)), B. giganteus (MG229639) (from the northern Gulf of Mexico, except for De Soto Canyon, Timm et al. (2018)), B. maxeyorum (KT963292) (from Bahamas, West Atlantic, Shipley et al. (2016)), B. kapala (OQ970652) (from eastern Australian waters, unpublished) and B. doederleini (MZ723938) (from Sagami Bay, Japan, unpublished). B. decemspinosus from the Indian Ocean is registered in NCBI, but is considered a misidentification (Huang and Bruce 2024). 16S rRNA sequences for B. jamesi (KX417643) (from the sea off the southern part of Hainan Island, China, Kou et al. (2017)), B. kensleyi (OQ865221) (from Marion Plateau, Coral Sea, QLD, Australia, Huang et al. (2022)), B. yucatanensis (MZ042927) (from the Gulf of Mexico off the Yucatan Peninsula, Huang et al. (2022)), B. giganteus (MG229479) (from the northern Gulf of Mexico, except for De Soto Canyon, Timm et al. (2018)) and B. doederleini (OR239864) (from Suruga Bay, Japan, unpublished) were obtained. The nucleotide sequence for Cirolanidae (Atarbolana exoconta Bruce and Javed, 1987) COI (KX782999) and (Excirolana hirsuticauda Menzies, 1962), 16S rRNA (MK898194) were used as the outgroup control.

Using Drawtree (Phylip software package, http://bioweb.pasteur.fr/seqanal/interfaces/drawtree.html), phylogenetic trees were constructed by the Neighbour-joining (NJ) method under some different techniques (Nei and Kummer 2000). The percentage of replicate trees where the associated taxa clustered together in the bootstrap test (1000 replicates) is shown above the branches. Using Kimura 2 parameters (K2P) genetic distance (Kimura 1980) in MEGA 11, pairwise distance analysis was carried out (Tamura et al. 2007, Tamura et al. 2021).

Order Isopoda Latreille, 1989

Family Cirolanidae Dana, 1852

Genus Bathnomus A. Milne-Edwards, 1879

Comparative material

Bathynomus jamesi Kou, Chen and Li, 2017, male (TMCD003327), TL 355 mm, CL 197 mm, waters of the South China Sea about 300 km southwest of Pratas Island, 19.084N, 115.250E, bottom trawl, depth was about 420-550 m. 12 May 2020. Bathynomus giganteus A. Milne-Edwards, 1879, male (TMCD003336) exchanged with Japan's Shin Enoshima Aquarium (Fujisawa, Kanagawa, Japan), TL 316 mm, CL 172 mm, baited cage at a depth of 600-800 m on 19 April 2017 in the Gulf of Mexico off the Yucatan Peninsula. Bathynomus yucatanensis Huang, Kawai and Bruce 2022, holotype (TMCD003335) exchanged with Japan's Shin Enoshima Aquarium (Fujisawa, Kanagawa, Japan), male, TL 257 mm, CL 129 mm and wet weight 550 g, baited cage at a depth of 600-800 m on 19 April 2017 in the Gulf of Mexico off the Yucatan Peninsula. Bathynomus doederleini Ortmann, 1894, specimens (NMMB-CD003011), TL 128 mm, CL 47 mm, 28 Aug 2008, 122°2.751E. 24°53.324N, off Tai-chi, I-lan County, Taiwan, depth 600 m.

A phylogenetic tree of Bathynomus was drawn using molecular evolution theory and the COI sequence. Fig. 9 shows the relative relationship between different species of Bathynomus. The closest relative of B. paracelensis sp. nov. in this graph is a clade comprised of B. jamesi and B. kensleyi. B. doederleini is more distantly related to B. paracelensis sp. nov.

Figure 9.  

The phylogenetic tree is based on the cytochrome c oxidase I (COI) DNA sequences. The sequences were aligned using Clustal Omega and the Neighbour-joining method constructed the tree. Numbers at branches indicate bootstrap values. The sequences of Cirolanidae (Atarbolana exoconta, KX782999) COI were used as the outgroup. Evolutionary analyses were conducted in MEGA 11.

Fig. 10 is a phylogenetic tree drawn, based on 16S rRNA. The nearest relative of B. paracelensis sp. nov. in this graph is a clade comprised of B. doederleini and B. kapala.

Figure 10.  

The phylogenetic tree is based on the 16S rRNA sequences. The sequences of Cirolanidae (Excirolana hirsuticauda, MK898194) 16S rRNA were used as the outgroup. Evolutionary analyses were conducted in MEGA 11.

As shown in Table 4, the shortest pairwise distance between interspecies occurs in B. giganteus vs. B. yucatanensis (5.88%). The B. paracelensis sp. nov. distance to the closest related B. kensleyi is 14.14%, 2.4 times the shortest distance amongst known Bathynomus interspecies.

The average pairwise distance between Bathynomus interspecies calculated from Table 4 is 16.77%. The average pairwise distance of B. paracelensis sp. nov. for other Bathynomus is 17.16%, showing that the average pairwise distance of B. paracelensis sp. nov. is higher than the average.

Comparing B. paracelensis sp. nov. with other Bathynomus, B. kensleyi has the smallest pairwise distance (14.14%) and B. doederleini has the largest pairwise distance (23.22%), indicating that it has the closest relationship with B. kensleyi and the most distant relationship with B. doederleini. The average pairwise distance of B. paracelensis sp. nov. for other Bathynomus is 17.16%, which is greater than B. kensleyi (15.23%), B. giganteus (14.68%), B. jamesi (15.80%) and B. yucatanensis (14.70%), only smaller than B. doederleini (20.57%) and B. maxeyorum (17.42%).

B. jamesi, which has the closest geographical relationship, has a pairwise distance of 15.30% from B. paracelensis sp. nov., which is greater than B. kensleyi (14.14 %) and B. giganteus (14.31%).

B. paracelensis sp. nov. is the smallest supergiant Bathynomus in the South China Sea. The body shape is similar to B. doederleini. B. paracelensis sp. nov. may be an intermediate organism between giant and supergiant Bathynomus.

The morphological comparison of B. paracelensis sp. nov. with B. jamesi and B. vaderi is shown in Table 3. Characteristic features unique to B. paracelensis sp. nov., such as body shape sub-parallel, clypeus apex angle, number of maxilliped endite coupling five and number of maxilliped keratinised spine number 11, were compared. The comparison shows that B. paracelensis sp. nov. is a newly-described species.

Alignment of DNA sequences

The difference of DNA sequences in various Bathynomus is compared through DNA alignment. Fig. 7 and Fig. 8 show the comparison results of COI and 16S rRNA, respectively. From the DNA alignment in Fig. 7, it can be seen that the differences between B. paracelensis sp. nov. and other Bathynomus bases are: B. jamesi (70 bases different, base different ratios 13.5%), B. kensleyi (65, 12.5%), B. yucatanensis (74, 14.3%), B. giganteus (66, 12.7%), B. maxeyorum (75, 14.5%), B. kapala (89, 17.2%) and B. doederleini (101, 19.5%). The data show that the species with the greatest difference from B. paracelensis sp. nov. is B. doederleini (19.5%) and the least difference is B. kensleyi (12.5%). It is shown that, amongst the known Bathynomus, the closest relative to B. paracelensis sp. nov. is B. kensleyi.

Phylogenetic tree analysis

Phylogenetic trees, based on COI and 16S rRNA sequences (Figs 9, 10), showed different results. B. paracelensis sp. nov. was closer to supergiant B. kensleyi and B. jamesi in the analysis using COI as the marker. However, it was closer to giant B. doederleini and B. kapala in the analysis using 16S rRNA as the marker.

One of the reasons for this opposite result may be that there are fewer species and the selected DNA length is too short in the 16S rRNA sequence. Due to differences in length and location, there are limitations on the common lengths that can be compared. Compared with COI, 16S rRNA has fewer species (six species) and a comparable DNA length (345 bp). Another possibility is that B. paracelensis sp. nov. may be a Bathynomus species between supergiant and giant.

Barcoding gap analysis

Since the COI sequences of the four B. paracelensis sp. nov. are precisely the same and there is no base variation (Fig. 7), the DNA barcoding gap of intraspecies cannot be calculated, so only the pairwise distance of interspecies can be calculated.

The two data confirm that B. paracelensis sp. nov. and B. jamesi are different species. One is that the shortest pairwise distance amongst Bathynomus interspecies occurs between B. giganteus and B. yucatanensis (5.88%) (Table 4). The shortest pairwise distance between B. paracelensis sp. nov. is B. kensleyi (14.14%), 2.4 times the shortest distance amongst known Bathynomus interspecies. Another 16.77% of the data represents the average pairwise distance of Bathynomus interspecies. For B. paracelensis sp. nov., the average pairwise distance of Bathynomus is 17.16%, which is already higher than 16.77%, indicating that the possibility of its becoming an independent species is high. The above conclusion can confirm that B. paracelensis sp. nov. and B. jamesi are different species.

Bathynomus from the South China Sea and the Indian Ocean have been misidentified (Huang et al. 2022, Huang and Bruce 2024). Since genetic identification had not yet been developed, Lowry and Dempsey (2006) identified B. jamesi and B. kensleyi as the same species. However, without a morphological basis, misidentifications may also occur. Bathynomus species native to the Indian Ocean were identified as B. decemspinosus, B. doederleini and B. kensleyi (Sankar et al. 2011, PrasannaKumar et al. 2020), which are misidentifications due to a lack of morphological basis. Since the appearance of Bathynomus is quite similar, in the future species identification of Bathynomus, both morphological and genetic testing results need to be completed, becoming a necessary condition for the birth of new species.

Morphology is a very complex science. It compares all body structures and is the basis of taxonomy. However, many individuals have morphological differences unrelated to the species, making identification prone to errors.

Compared with morphology, genetic classification is sharper. Usually, when confirming a new species, double confirmation through morphological and genetic systems is more reliable and genetics also requires more than two markers (such as COI and 16S rRNA) to be more stringent. Bathynomus without gene information is worthy of re-examination and review to confirm whether it is a misidentification caused by intra-specific differences or an error caused by two different Bathynomus species being too similar in morphology. This study used morphological and Barcoding gap analysis to compare the structure and genetics of B. paracelensis sp. and B. jamesi and confirmed that the two are different species. Since B. vaderi only has a morphological description (Ng et al. 2025) and no COI and 16S rRNA data for comparison, this article compares B. paracelensis sp. with B. vaderi, which can only be restricted to morphology.