The Synonymy of the Supratidal Crab Species Parasesarma cognatum Rahayu & Li, 2013 with P. liho Koller, Liu & Schubart, 2010 (Decapoda: Brachyura: Sesarmidae) Based on Morphological and Molecular Evidence, with a Note on P. paucitorum Rahayu & Ng, 2009

© 2019 Academia Sinica, Taiwan Open Access The Synonymy of the Supratidal Crab Species Parasesarma cognatum Rahayu & Li, 2013 with P. liho Koller, Liu & Schubart, 2010 (Decapoda: Brachyura: Sesarmidae) Based on Morphological and Molecular Evidence, with a Note on P. paucitorum Rahayu & Ng, 2009 Hsi-Te Shih1,2,§,*, Pei-Yi Hsu1,§, Adnan Shahdadi3,§, Christoph D. Schubart4, and Jheng-Jhang Li5,6 1Department of Life Science, National Chung Hsing University, Taichung 402, Taiwan. *Correspondence: E-mail: [email protected] 2Research Center for Global Change Biology, National Chung Hsing University, Taichung 402, Taiwan 3Department of Marine Biology, Faculty of Marine Sciences and Technology, University of Hormozgan, Bandar Abbas, Iran 4Zoologie & Evolutionsbiologie, Universität Regensburg, 93040 Regensburg, Germany 5Department of Marine Biotechnology and Resources, National Sun Yat-sen University, Kaohsiung 804, Taiwan 6East Peak Ecological Consultants, Inc., Linyuan, Kaohsiung 832, Taiwan §HTS, PYH, and AS contributed equally to this paper. Received 9 May 2019 / Accepted 15 July 2019 / Published 10 September 2019 Communicated by Benny K.K. Chan Parasesarma liho Koller, Liu & Schubart, 2010 and P. cognatum Rahayu & Li, 2013 from East and Southeast Asia are similar species that have been confused in several studies. Here, we re-examined the type specimens of both species and found identical main characters, which is supported by the molecular identity of the mitochondrial cytochrome oxidase subunit I gene. As a result, we treat P. cognatum as a junior subjective synonym of P. liho. We also show that the male paratype of P. paucitorum Rahayu & Ng, 2009 is conspecific with P. liho, although P. paucitorum s. str. remains a distinct but allied species. The distribution of P. liho is updated to include Japan (Ryukyus), Taiwan, Philippines (Cebu) and Indonesia (Sulawesi). Key words: Parasesarma, P. liho, P. cognatum, P. paucitorum, Morphology, Mitochondrial COI. Citation: Shih HT, Hsu PY, Shahdadi A, Schubart CD, Li JJ. 2019. The synonymy of the supratidal crab species Parasesarma cognatum Rahayu & Li, 2013 with P. liho Koller, Liu & Schubart, 2010 (Decapoda: Brachyura: Sesarmidae) based on morphological and molecular evidence, with a note on P. paucitorum Rahayu & Ng, 2009. Zool Stud 58:21. doi:10.6620/ZS.2019.58-21. BACKGROUND Crabs of the family Sesarmidae Dana, 1851 form part of the dominant macrofauna of the Indo-West Pacific mangrove ecosystems, playing an especially important ecological role as “ecosystem engineers” (Lee 1998; Kristensen 2008). They inhabit landward regions of mangrove fringes and can tolerate high temperature and salinity fluctuations (Theurkauff et al. 2018). With 69 recognised species, Parasesarma De Man, 1895 is the most speciose genus of this family, especially after most species of Perisesarma De Man, 1895 have been moved to Parasesarma and the recent descriptions of more new species (see Ng et al. 2008; Shahdadi et al. 2017 2018a 2019; Li et al. 2018; Shahdadi and Schubart 2017; Fratini et al. 2019). In Taiwan, 12 species of this genus have been reported, of which eight, viz. P. cognatum Rahayu & Li, 2013, P. corallicum Ng, Davie & Li, 2016, P. kuekenthali (De Man, 1902), P. kui Li, Rahayu & Ng, 2018, P. lepidum (Tweedie, 1950), P. liho Zoological Studies 58: 21 (2019) doi:10.6620/ZS.2019.58-21 1 © 2019 Academia Sinica, Taiwan Koller, Liu & Schubart, 2010, P. macaco Li, Rahayu & Ng, 2018, and P. ungulatum (H. Milne Edwards, 1853), were added after the latest revision of Taiwanese crabs in 2001 (Ng et al. 2001 2016 2017; Koller et al. 2010; Rahayu and Li 2013; Li 2015; Hsu and Shih 2018; Li et al. 2018 2019). Recently, several taxonomic studies of crabs have used molecular evidence to support the descriptions or recognition of new or reinstated species (cf. Chu et al. 2015); this is also the case within the family Sesarmidae (Schubart et al. 1998 2009; Gillikin and Schubart 2004; Koller et al. 2010; Naderloo and Schubart 2010; Ragionieri et al. 2012; Thiercelin and Schubart 2014; Cannicci et al. 2017; Shahdadi et al. 2017 2018a b). To help identify species of Taiwanese sesarmids, a DNA barcoding approach using the cytochrome oxidase subunit I (COI) marker (Hebert et al. 2003a b) was undertaken. A preliminary result of the COI analyses showed that only one clade was obtained for specimens labelled as P. liho and P. cognatum from Taiwan; and this required further investigation. Parasesarma liho Koller, Liu & Schubart, 2010 (type locality: Hualien, Taiwan) is distributed in Hualien and Taitung, eastern Taiwan (Koller et al. 2010), as well as the Ryukyus, Japan (Okinawa, Miyako and Ishigaki; Maenosono and Naruse 2015). Parasesarma cognatum Rahayu & Li, 2013 (type locality: Pingtung, Taiwan) is distributed in Pingtung (southern Taiwan) and Hualien (eastern Taiwan), as well as in Cebu (the Philippines) according to Rahayu and Li (2013). Rahayu and Li (2013: 639) mentioned that P. cognatum is different from P. liho in the proportions of ambulatory propodi, structure of the male first gonopod, number and shape of dactylar tubercles of male chela, and coloration. However, the similarity between the two species has been remarked by Maenosono and Naruse (2015: 22), who questioned their identities. In the present study, the types of these species, as well as more specimens of different sizes, were examined and their COI sequences compared, including specimens from various localities. The types of the allied P. paucitorum Rahayu & Ng, 2009 from Indonesia were also studied. MATERIALS AND METHODS New specimens with the appearance of Parasesarma liho and P. cognatum were collected from their type localities and other areas in southern and eastern Taiwan, as well as from Cebu, Philippines. The holotypes and paratypes of P. liho, P. cognatum and P. paucitorum were also included. Individuals of other related species (see below) were also studied for comparison (Table 1). Those specimens were deposited in the Biodiversity Research Museum, Academia Sinica, Taipei, Taiwan (ASIZ); the Muséum national d’Histoire naturelle, Paris (MNHN); the Museum Zoologi Bogor, Indonesian Institute of Sciences, Indonesia (MZB); the Zoological Collections of the Department of Life Science, National Chung Hsing University, Taichung, Taiwan (NCHUZOOL); the National Museum of Marine Biology and Aquarium, Pingtung, Taiwan (NMMBA); National Museum of Natural Science, Taichung, Taiwan (NMNS); the Senckenberg Museum, Frankfurt am Main, Germany (SMF); Zoological Reference Collection of the Lee Kong Chian Natural History Museum, National University of Singapore (ZRC); and the Zoologische Staatssammlung, München (Munich), Germany (ZSM). The abbreviations P4 is used for the fourth pereiopods (third ambulatory legs), and G1 for male first gonopods. Measurements, all in millimeters (mm), are of the maximum carapace width (CW) and carapace length (CL). The G1s of different sized specimens were compared to determine whether there is a size effect on the morphology of G1. The ratios of length/width of the P4 propodus for different-sized specimens were calculated, as it was used to distinguish P. liho and P. cognatum by Rahayu and Li (2013). Because different authors could measure different positions from the photograph (Shih and Do 2014), both the proximal and distal widths of the propodi were measured to obtain the range of width in our study (Fig. 1). The morphology of the upper margin of the cheliped merus of the holotype was also examined to confirm the presence of a large subdistal spine, as mentioned in Rahayu and Li (2013: 637) and Maenosono and Naruse (2015: 22, fig. 7B). Genomic DNA was isolated from the muscle tissue of the pleon or walking leg with different kits (Shih et al. 2016; Shahdadi and Schubart 2017). A portion of the COI gene was amplified with PCR using the primers LCO1490, HCO2198 (Folmer et al. 1994) or COL6 and COH6 (Schubart 2009). The PCR conditions for the above primers were: denaturation for 50 s at 94°C, annealing for 70 s at 45–47°C (45 s at 48–50°C for COL6-COH6), and extension for 60 s at 72°C (40 cycles), followed by extension for 10 min at 72°C. Sequences were obtained by automated sequencing (Applied Biosystems 3730) and verified with the complementary strand. Sequences of the different haplotypes were deposited in the DNA Data Bank of Japan (DDBJ) (accession numbers in Table 1). For comparative purposes, we included the species or species complexes used in the study of P. liho, P. cognatum and P. paucitorum in Koller et al. (2010), Rahayu and Li (2013) and Rahayu and Ng (2009), respectively, using the following as outgroups: P. affine page 2 of 13Zoological Studies 58: 21 (2019) © 2019 Academia Sinica, Taiwan Table 1. Found haplotypes of the COI gene of Parasesarma liho, P. cognatum and P. paucitorum, as well as the outgroups. For abbreviations of museums and universities see MATERIALS AND METHODS Locality sample size Catalogue no. Haplotype of COI Access. no. of COI P. liho or P. cognatum Taiwan Hualien: Meilun R. estuary 1 SMF 36266 (holotype of P. liho) —aLC490879 1 ZSM A20100040 (paratype of P. liho) —aLC490880 1NCHUZOOL 15027 PRL-C1 LC490881 Taitung: Dulanwan 2 NCHUZOOL 15025 PRL-C1 LC490881 Pingtung: Niou R. estuary 1 NCHUZOOL 15031 PRL-C1 LC490881 Pingtung: Gangkou R. estuary 1 NMMBCD 3975 (holotype of P. cognatum) PRL-C1 LC490881 2 NMMBCD 3976 (paratypes of P. cognatum) PRL-C1 LC490881 1 NMMBCD 3976 (paratype of P. cognatum) PRL-C2 LC490882 1NCHUZOOL 15028 PRL-C1 LC490881 1 ZRC 2013.1757 —aLC490883 1 ZRC 2013.1757 —aLC490883 Pingtung: Houwan 3 NCHUZOOL 15022 PRL-C1 LC490881 1NCHUZOOL 15425 PRL-C1 LC490881 Pingtung: Baoli R. 1 NCHUZOOL 15024 PRL-C1 LC490881 Philippines Cebu: Kawasan 1 ASIZCR PRL-C1 LC490881 1 ASIZCR PRL-C1 LC490881 1NCHUZOOL 15034 PRL-C2 LC490882 Indonesia Sulawesi: Manado 1 ZRC 2019.0578 (male paratype of P. paucitorum) —aLC490884 Others P. paucitorum: Sulawesi, Indonesia 1 MZB Cru 2243 (holotype) LC490885 1 ZRC 2008.0869 (female paratype) LC490886 P. kui: Gangkou R. estuary, Pingtung, Taiwan 1 NMNS 7779-015 (holotype) LC490887 P. macaco: Baoli R. estuary, Pingtung, Taiwan 1 NMNS-7779-005 (holotype) LC490888 P. tripectinis: Dajia R. estuary, Taichung, Taiwan 1 NCHUZOOL 15428 LC490889 P. pictum: Nangan, Matsu, Taiwan 1 NCHUZOOL 15427 LC490890 P. affine: Danshuei, New Taipei, Taiwan 1NCHUZOOL 15426 LC490891 P. dumacense: Cebu, Philippines ZRC 2008.0833 KX400929 asequences are shorter and not included for further analyses (see RESULTS). Fig. 1. Schematic drawing showing the measurement of the length, as well as the maximum and minimum widths of the propodi of the fourth pereiopods (third ambulatory leg, P4) used in this study. page 3 of 13Zoological Studies 58: 21 (2019) © 2019 Academia Sinica, Taiwan (De Haan, 1837), P. dumacense (Rathbun, 1914), P. leptosoma (Hilgendorf, 1869) (now corresponding to P. kui Li, Rahayu & Ng, 2018, P. macaco Li, Rahayu & Ng, 2018 of this species complex; see Li et al. 2018), P. pictum (De Haan, 1835), and P. tripectinis (Shen, 1940) (Table 1). The COI sequences of P. dumacense was downloaded from GenBank (accession number: KX400929). The best-fitting model for sequence evolution was determined by Partition Finder (vers. 2.1.1; Lanfear et al. 2017), selected by the Bayesian information criterion (BIC). The obtained best model (GTR + I) was subsequently used for a Bayesian inference (BI) analysis. The BI analysis was performed with MrBayes (vers. 3.2.3, Ronquist et al. 2012). The phylogenetic analyses were run with four chains for 10 million generations and four independent runs, with trees sampled every 1000 generations. The convergence of chains was determined by the average standard deviation of split frequency values below the recommended 0.01 (Ronquist et al. 2019) and the first 3000 trees were discarded as burnin. The maximum likelihood (ML) analysis was conducted in RAxML (vers. 7.2.6, Stamatakis 2006). Because RAxML does not accept the GTR + I model, the second best model, GTR + G (i.e., GTRGAMMA), was used with 100 runs, and the best ML tree was found by comparing the likelihood scores. The robustness of the ML tree was evaluated by 1000 bootstrap pseudoreplicates under the model GTRGAMMA. The relationships of the COI haplotypes among P. liho and other related species were examined by using the program PopART (vers. 1.7, Leigh and Bryant 2015). Basepair (bp) differences and the pairwise estimates of Kimura 2-parameter (K2P) distances (Kimura 1980) for genetic diversities between haplotypes were calculated with MEGA (vers. 10.0.5, Kumar et al. 2018). Material examined (see Table 1): Holotype of P. liho: male (13.0 × 12.3 mm) (SMF 36266), Meilun R. (= River) estuary, Hualien, Taiwan, coll. H.-C. Liu, 31 October 2009; paratypes: 1 male (12.7 × 11.5 mm) (ZSM A20100040), same data as holotype; 1 male (14.2 × 13.1 mm) (SMF 36269), 1 male (14.5 × 13.1 mm) (MNHN B32312), same locality as holotype, coll. H.-C. Liu, 6 November 2000. Holotype of P. cognatum: male (14.3 × 13.1 mm) (NMMBCD 3975), Gangkou R. estuary, Manjhou, Pingtung, Taiwan, coll. J.-J. Li, 1 September 2012; paratypes: 2 females (11.5 × 10.3 mm; 14.4 × 12.6 mm) (NMMBCD 3506), same locality as holotype, coll. J.- J. Li, 8 June 2012 (the data is different from that in Rahayu and Li 2013). Others: Taiwan: 1 female (10.8 mm) (NCHUZOOL 15024), Baoli R. estuary, Pingtung, coll. P.-Y. Hsu et al., 11 July 2017; 1 male (14.2 mm) (NCHUZOOL 15030), Houwan, Pingtung, coll. P.- Y. Hsu, 26 June 2012; 2 males (11.1–16.2 mm), 5 females (12.6–16.0 mm) (NCHUZOOL 15022), Houwan, Pingtung, coll. R.-H. Lee, 19 September 2013; 3 males (9.7–14.9 mm), 1 female (9.3 mm) (NCHUZOOL 15425), Houwan, Pingtung, coll. P.- Y. Hsu and C.-Y. Chi, 3 December 2016; 1 female (16.2 mm) (NCHUZOOL 15029), Leidashih, Kenting, Pingtung, coll. J.-H. Lee, 18 August 2012; 1 male (16.7 mm) (NCHUZOOL 15031), Niou R. estuary, Pingtung, coll. P.-Y. Hsu, 19 January 2016; 2 males (13.3–13.0 mm), 1 female (11.8 mm) (ZRC 2013.1757), Gangkou R. estuary, Pingtung, coll. J.- J. Li, 19 February 2013; 2 females (11.7–12.6 mm) (NCHUZOOL 15023), Fushuei Bridge, Gangkou R. estuary, Pingtung, coll. P.-Y. Hsu et al., 12 July 2017; 2 females (8.6–13.9 mm) (NCHUZOOL 15028), Gangkou R. estuary, Pingtung, coll. P.-Y. Hsu et al., 4 September 2017; 8 males (4.8–12.4 mm), 2 females (5.6–6.0 mm) (NCHUZOOL 15025), Dulanwan, Taitung, coll. P.-Y. Hsu et al., 9 August 2017; 1 male (6.1 mm), 2 females (10.9–10.9 mm) (NCHUZOOL 15032), Jihuei Fishing Port, Taitung, coll. P.-Y. Hsu et al., 10 August 2017; 1 male (11.6 mm) (NCHUZOOL 15027), Meilun R. estuary, Hualien, coll. J.-H. Lee, 18 May 2012; 1 male (11.6 mm), 3 females (10.2–11.3 mm) (NCHUZOOL 15026), Meilun R. estuary, Hualien, coll. P.-Y. Hsu et al., 10 August 2017. Philippines: 1 male (11.2 mm), 1 female (16.4 mm) (ASIZCR), Kawasan, Cebu, coll. H.-C. Liu, 4 December 2001; 2 males (13.0–13.3 mm) (NCHUZOOL 15034), Kawasan, Cebu, coll. J.-J. Li, 6 September 2018. Indonesia: 1 male (15.5 mm) (ZRC 2019.0578, ex ZRC 2008.0869 partim) (paratype of P. paucitorum Rahayu and Ng, 2009), Manado, northern Sulawesi, Indonesia, coll. P. K. L. Ng, 17 July 2003. Comparative material: Parasesarma paucitorum: 1 male (19.7 mm) (MZB Cru 2243, holotype), 1 female (19.2 mm) (ZRC 2008.0869, paratype), Manado, northern Sulawesi, Indonesia, coll. P. K. L. Ng, 17 July 2003. P. affine: 1 male (29.6 mm) (NCHUZOOL 15426), Danshuei, New Taipei City, Taiwan, 30 June 2006. P. dumacense: 1 male (20.4 mm) (ZRC 2008.0833), Kawasan Waterfall, Cebu, Philippines, coll. H.-C. Liu, 25 November 2001. P. pictum: 1 male (15.0 mm) (NCHUZOOL 15427), Nangan, Matsu, Taiwan, coll. P.-Y. Hsu et al., 26 August, 2011. P. tripectinis: 1 male (9.7 mm) (NCHUZOOL 15428), Dajia R. estuary, Taichung, Taiwan, 31 May 2014. page 4 of 13Zoological Studies 58: 21 (2019) © 2019 Academia Sinica, Taiwan RESULTS Morphology The CWs of the holotypes of Parasesarma liho (SMF 36266) and P. cognatum (NMMBCD 3975), as well as the paratypes of P. liho (SMF 36269; MNHN B32312) used for description and/or drawing are 13.0 mm, 14.3 mm, 14.2 mm and 14.5 mm, respectively. Different sized specimens with CW from 4.8 to 16.7 mm of males (n = 19) and 5.6 to 16.4 mm of females (n = 22) were included in this study. The distal part of the G1s of the specimens of P. liho and P. cognatum, with CW ranging from 11.1 to 16.7 mm, are quite similar in the form of the apical processes (Fig. 2A–E, G, H). The morphology of G1 of the holotype (SMF 36266) was examined and also agrees with other G1s shown in figure 2. The length/width ratios of the P4 propodi are shown in table 2; the differences in the maximum and minimum ratios are large, ranging from 2.67 to 4.76, for different sizes of CW (7.0–16.7 mm), including the paratype of P. liho (SMF 36269 and MNHN B32312; ratios: 2.79 and 2.67, respectively; based on Koller et al. 2010: figs. 2f, 3c) and the holotype of P. cognatum (NMMBCD 3975; ratio: 2.79). The chelar morphology of the types of P. liho and P. cognatum are similar, including the 11–12 elongate tubercles on the dorsal margins (Fig. 3A–D). The holotype of P. cognatum (NMMBCD 3975) was re-examined, confirming only a subdistal angle, not a spine, on the outer margin of the cheliped merus (Fig. 4). For the coloration of P. liho in the field, the carapace is light brown with dark brown blotches, whereas the chelipeds and legs are uniform yellow to brownish yellow (Fig. 5). The male paratype of P. paucitorum (ZRC 2019.0578) was also found to resemble other specimens of P. liho (and P. cognatum), but somewhat different from the holotype of P. paucitorum (MZB Cru 2243), including the G1s (Fig. 2F, I), chelae (Fig. 3E–H), as well as in the fresh coloration of carapace and legs (Rahayu and Ng 2009: fig. 1). DNA analysis The molecular results comprised 22 P. liho-like and P. cognatum-like specimens, including the holotype and paratypes of P. liho and P. cognatum (Table 1). Available COI sequences for the holotype of P. liho, the paratype of P. cognatum and the male paratype of P. paucitorum were a bit shorter (614–635 bp see below), but otherwise identical to most other sequences with 658 bp. Therefore, they were omitted from further analyses. The male paratype of P. paucitorum (ZRC 2019.0578) has a similar genetic sequence as P. liho (see above) and is also referred to as being in the P. liho clade (see below). In total, three haplotypes of P. liho (including P. cognatum) are found from among the studied specimens (Table 1). The reconstructed COI phylogenetic tree (Fig. 6) shows that the three haplotypes of P. liho (including the male paratype of P. paucitorum) form a distinct clade, sister to another clade including real P. paucitorum s. str. The mean pairwise nucleotide divergence with the K2P distances and bp differences of haplotypes of the two clades are shown in table 3. The intraspecific K2P nucleotide divergence within P. liho (including P. cognatum) is 0.05% (0–0.32%), and the interspecific K2P distance between P. liho and the P. paucitorum clade is 1.49% (1.46%–1.64%). Molecular data shows that the male paratype of P. paucitorum belongs to the P. liho clade and is different from the other types of P. paucitorum (Figs. 6, 7, Table 3). The haplotype network based on COI haplotypes (Fig. 7) further shows P. liho and P. paucitorum to be separated by 11 steps, and both species to be different from P. kui and P. macaco by 34–38 steps. Table 2. The ratios of length/maximum width and length/minimum width of P4 propodi of specimens of Parasesarma liho with different size. The ratios of paratypes of P. liho was measured by Koller et al. (2010: figs. 2f, 3c). For abbreviations of museums and universities see MATERIALS AND METHODS CW (mm) cat. no. Length / maximum width Length / minimum width 7.0 NCHUZOOL 15025 3.19 4.76 11.1 NCHUZOOL 15022 2.89 4.17 11.6 NCHUZOOL 15027 2.81 4.15 11.6 NCHUZOOL 15026 2.67 3.79 14.2 SMF 36269 (paratype of P. liho) 2.79 4.01 14.5 MNHN B32312 (paratype of P. liho) 2.67 3.88 14.3 NMMBCD 3975 (holotype of P. cognatum) 2.79 4.17 16.7 NCHUZOOL 15031 3.36 4.53 page 5 of 13Zoological Studies 58: 21 (2019) © 2019 Academia Sinica, Taiwan Fig. 2. The morphological variation of distal part the right G1s of Parasesarma liho (A–H), and P. paucitorum (I). A, CW 11.1 mm (NCHUZOOL 15022), Pingtung, Taiwan; B, CW 11.6 mm (NCHUZOOL 15027), Hualien, Taiwan; C, CW 12.7 mm (ZSM A20100040, paratype of P. liho), Hualien, Taiwan; D, CW 13.29 mm (NCHUZOOL 15034), Cebu, Philippines; E, CW 14.3 mm (NMMBCD 3975, holotype of P. cognatum), Pingtung, Taiwan; F, CW 15.5 mm (ZRC 2019.0578, paratype of P. paucitorum), Sulawesi, Indonesia; G, CW 16.2 mm (NCHUZOOL 15022), Pingtung, Taiwan; H, CW 16.7 mm (NCHUZOOL 15031), Pingtung, Taiwan; I, CW 19.7 mm (MZB Cru 2243, holotype of P. paucitorum), Sulawesi, Indonesia. Scales bars = 0.5 mm. (A) (D) (G) (B) (E) (H) (C) (F) (I) page 6 of 13Zoological Studies 58: 21 (2019) © 2019 Academia Sinica, Taiwan Fig. 3. The morphology of left chela of types of Parasesarma liho (A–B), P. cognatum (C–D) and P. paucitorum (E–H). A, C, E, G, outer view; B, D, F, H, upper view. A, B, holotype of P. liho (CW 13.0 mm, SMF 36266); C, D, holotype of P. cognatum (CW 14.3 mm, NMMBCD 3975); E, F, paratype of P. paucitorum (CW 15.5 mm, ZRC 2019.0578); G, H, holotype of P. paucitorum (CW 19.7 mm, MZB Cru 2243). Scales bars = 2 mm. (A) (D) (G) (B) (E) (H) (F) (C) page 7 of 13Zoological Studies 58: 21 (2019) © 2019 Academia Sinica, Taiwan Fig. 4. The outer surface of chelipedal meri of Parasesarma cognatum (holotype, NMMBCD 3975). A, left cheliped; B, right cheliped. Arrow indicates a subdistal angle on the upper margin of chelipedal merus. Fig. 5. The coloration of Parasesarma liho in the field in Taiwan. A, specimen (not captured) from Gangkou R. estuary, Pingtung; B, specimen (not captured) from Meilun R. estuary, Hualien. (A) (A) (B) (B) page 8 of 13Zoological Studies 58: 21 (2019) © 2019 Academia Sinica, Taiwan DISCUSSION In this study, we confirm that Parasesarma cognatum Rahayu & Li, 2013 is a synonym of Parasesarma liho Koller, Liu & Schubart, 2010 based on the evidence from morphology (Figs. 2–4, Table 2) and mitochondrial DNA (Figs. 6–7, Table 3). The main differences between the two species was supposedly the morphology of G1 and the ratio of length/width of P4 propodus (Rahayu and Li 2013). The apical processes of the presently examined G1s are quite different from that of the paratype of P. liho (SMF 36269), which has a very tapering tip (Koller et al. 2010: fig. 2e). After comparing the line drawings of Fig. 6. A Bayesian inference (BI) tree of Parasesarma liho, as well as the outgroups, based on the cytochrome oxidase subunit I genes (COI). Probability values at the nodes represent support values for BI and maximum likelihood (ML). For haplotype names, see table 1. Table 3. Matrix of percentage pairwise K2P nucleotide divergences (lower left) and mean number of differences (upper right) based on COI within the clade of Parasesarma liho (including types of P. cognatum and the male paratype of P. paucitorum) and other related species (see Table 1). Values of range are shown in parentheses Intraspecific Interspecific Nucleotide divergence Mean nucleotide difference P. liho P. paucitorum P. kui P. macaco P. liho 0.05 (0-0.32) 0.32 (0-2) 9.17 (9-10) 33.11 (33-34) 35.11 (35-36) P. paucitorum 0 0 1.49 (1.46-1.64) 33 (33-33) 33 (33-33) P. kui 5.26 (5.22-5.58) 5.5 (5.5-5.5) 4 (4-4) P. macaco 5.6 (5.56-5.94) 5.5 (5.5-5.5) 0.61 (0.61-0.61) page 9 of 13Zoological Studies 58: 21 (2019)