Somniosus (Rhinoscymnus) cheni sp. nov., A New Species of Sleeper Shark (Squaliformes: Somniosidae) from Eastern Taiwan, with Aspects of Embryo Biology

© 2020 Academia Sinica, Taiwan Open Access Citation: Hsu HH, Lin CY, Joung SJ. 2020. Somniosus (Rhinoscymnus) cheni sp. nov., a new species of sleeper shark (Squaliformes: Somniosidae) from eastern Taiwan, with aspects of embryo biology. Zool Stud 59:48. doi:10.6620/ZS.2020.59-48. BACKGROUND The family Somniosidae contains five genera: Centroscymnus, Scymnodalatias, Scymnodon, Somniosus, and Zameus. Most Somniosid sharks inhabit near the seabed on continental and insular slopes, but few species are oceanic and semi-oceanic (Ebert et al. 2013a). Somniosus differs from the other genera in having its first dorsal fin situated at the middle of the dorsum, well behind the pectoral fin, but well before the pelvic fin, both dorsal fins without spines, second dorsal fin slightly smaller than first, and paddle-shaped caudal fin with long lower lobe (Compagno 1984; Yano et al. 2004; Ebert et al. 2013a). Yano et al. (2004) divided the genus Somniosus into two subgenera and recognized five nominal species. The total length (TL) of the subgenus Somniosus is over four meters; it is a large species group and comprises the Greenland shark S. (S.) microcephalus (Bloch and Schneider 1801) from the Arctic and North Atlantic; the Antarctic sleeper shark S. (S.) antarcticus Whitley (1939), from the Southern Hemisphere; and the Pacific sleeper shark S. (S.) pacificus Bigelow and Schroeder (1944), from the Arctic and North Pacific. The subgenus Rhinoscymnus is a small species group (< 150 cm TL) comprising the frog shark S. (R.) longus (Tanaka 1912) from the western Pacific, and the little sleeper shark S. (R.) rostratus (Risso 1827) from the eastern North Atlantic and Mediterranean Sea (Compagno 1984; Francis et al. 1988; Yano et al. 2004; Yano et al. 2007; Ebert et al. 2013a). Cigala Fulgosi and Gandolfi (1983) redescribed S. (R.) rostratus based on the examination of external morphology of eight free-swimmers and 15 embryos; Francis et al. (1988) recorded S. (R.) rostratus that occurred in New Zealand waters of the Southern Hemisphere for the first time, however Roberts et al. Somniosus (Rhinoscymnus) cheni sp. nov., A New Species of Sleeper Shark (Squaliformes: Somniosidae) from Eastern Taiwan, with Aspects of Embryo Biology Hua-Hsun Hsu1, Chia-Yen Lin2,3, and Shoou-Jeng Joung2,3,* 1Marine Studies Section, Center for Environment and Water, Research Institute, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia. E-mail: [email protected] (Hsu) 2Department of Environmental Biology and Fisheries Science, National Taiwan Ocean University, Keelung 20224, Taiwan. E-mail: [email protected] (Lin) 3George Chen Shark Research Center, National Taiwan Ocean University, Keelung 20224, Taiwan. *Correspondence: E-mail: [email protected] (Joung). Tel: +886-2-24622192 ext. 5039 Received 3 June 2019 / Accepted 28 July 2020 / Published 22 September 2020 Communicated by Hin-Kiu Mok A new species of sleeper shark is described based on a 134.0 cm total length pregnant female collected from off Hualien, eastern Taiwan. The species belongs to a small species group (subgenus Rhinoscymnus) and can be distinguished from its congeners in having a smaller second dorsal fin, smaller eyes, more rows of teeth in the upper jaw and fewer in the lower jaw, and a different body proportion. Moreover, the specimen had 33 embryos, which is substantially more than the two other small species in the subgenus. Information on embryo biology are also provided. Key words: Elasmobranch, Litter size, Morphology, Sex-bias, Taxonomy. Zoological Studies 59:48 (2020) doi:10.6620/ZS.2020.59-48 1 © 2020 Academia Sinica, Taiwan (2015) considered as S. (R.) longus pregnant females of S. (R.) rostratus females from Mediterranean Sea were reported by Barrull and Mate (2001) and Vella et al. (2013), and length-weight relationships of S. (R.) rostratus were reported by Loyola Fernández et al. (2017). In contrast, records or information on S. (R.) longus are very rare. The holotype of S. (R.) longus, a 136 cm TL female, was collected from the Tokyo Market, Japan, and only five additional individuals have been reported from Japan and New Zealand waters, western Pacific (Yano et al. 2004). In Taiwan, three somniosids, Zameus squamulosus, Scymnodon ichiharai and Somniosus pacificus, have been reported with vouchers (Ebert et al. 2013b; White et al. 2014). A pregnant female sleeper shark was caught recently by a deep sea longliner, collected by Mr. WenJong Chen (landings consignee) from a fish market on 17 March 2017, and sent to Shoou Jeng Joung. A closer examination of the specimen revealed that it belongs to the small species group (subgenus Rhinoscymnus) with a short lateral keel present on the caudal fin base (Compagno 1984). However, it is clearly different from the two currently-recognized congeners and represents a new species. Herein we describe this new sleeper shark and report aspects of its embryos. MATERIALS AND METHODS The type specimen was collected from the ChengKung fishing harbor and fish market, eastern Taiwan (Figs. 1, 2). This individual was caught by a deep sea longliner operating in Hualien waters on 21 March 2017. Total length (TL) and head length (HL) are used throughout. Methods for taking morphometric measurements followed Yano et al. (2004). Data used for comparisons were from Francis et al. (1988), Barrull and Mate (2001), and Yano et al. (2004). All measurements are reported in mm and presented as percentage of TL. Meristic characters including upper and lower tooth row counts were measured from the holotype, and its vertebrae were counted from soft X-radiograph. A 147.0 mm TL male embryo and a 134.9 mm TL female embryo were dissected to investigate their vertebral counts and numbers of turns in the spiral valve, and another 145.5 mm TL male and 134.9 mm TL female embryos were also stained to count their vertebrae. Dermal denticles from the lateral trunk below the first dorsal fin were photographed. A t-test was used to distinguish between the measurement of the present individual and the mean of measurements of specimens from literature when measurements of the present specimen were out range of other specimen measurements. Differences in the embryonic sex ratio (male: female) were tested using a chi-square test, and average TL and body weight (non-yolk total weight, BW) between the sexes were tested using t-tests. The holotype and 29 embryos are preserved at the Department of Environmental Biology and Fisheries Science (EBFS), National Taiwan Ocean University, Keelung, Taiwan. The other four embryos are preserved at the National Museum of Marine Biology and Aquarium, Pingtung, Taiwan (NMMB-P33387). RESULTS Somniosus (Rhinoscymnus) cheni sp. nov. Taiwan Sleeper Shark (Fig. 2, Tables 1, 2) urn:lsid:zoobank.org:act:F3F4B53B-3970-49CB-B42A0CE7BA2F8020 Holotype: EBFSFSX001, 1340 mm TL, pregnant female, off Hualien, ca. 23°50'N, 121°50'E, eastern Taiwan, northwestern Pacific Ocean, landed at ChengKung fish market, Taitung, long line, 17 March 2017. Paratypes: NMMB-P33387, 2 males 145.5–147.0 mm TL and 2 females 134.9–135.8 mm TL, embryos taken from EBFSFSX001. Etymology: The species is named in honor of Mr. Wen-Jong Chen, Taitung Xin Gang District Fisherman’s Association, for his contributions of chondrichthyan fish samples and research assistance with Taiwan and international research for over 30 years. Diagnosis: A species of Somniosus (Rhinoscymnus) can be distinguished from its congeners by a combination of the following characters: no anal fin; two spineless dorsal fins, first dorsal fin higher and larger than second dorsal fin; a short lateral keels on the caudal peduncle, no precaudal pits; caudal fin asymmetrical and paddle-shaped, with a relatively short upper lobe and long lower lobe; rhomboid-shaped dermal denticles; teeth dissimilar in upper and lower jaws, upper-jaw teeth small with lanceolate, in 72 rows, lower-jaw teeth semioblique with low roots, in 28 rows; precaudal vertebrae 60–61, caudal vertebrae 15–18, and 75–78 in total; number of turns in spiral valve 25–26. In mid-term embryo stage (12.8–15.0 cm TL), body color light brown to yellow-grayish; teeth and dermal denticles absent; sexes identifiable; caudal keel present. Description: Proportional dimensions in percentage of TL are given in table 1. Body slender, almost cylindrical, no anal fin. Fork length 93.1% TL, precaudal length (PCL) 82.5% TL (Fig. 2). Snout page 2 of 11Zoological Studies 59:48 (2020) © 2020 Academia Sinica, Taiwan Fig. 1. The landing position (Cheng-Kung) and fishing boat’s base (Hualien). Fig. 2. Holotype of Somniosus (Rhinoscymnus) cheni sp. nov. (EBFSFSX001), pregnant female, 1340 mm total length. Photo by C. Y. Lin. page 3 of 11Zoological Studies 59:48 (2020) © 2020 Academia Sinica, Taiwan rounded, short and conical (Fig. 3); preoral length 26.2% HL. Head moderately long, HL 25.5% PCL. Gill opening moderately wide, last one about as long as first four, length of fifth gill opening 8.5% HL (Fig. 3). Mouth slightly arched, its width 35.1% HL (Fig. 3). Eye oval, horizontal diameter of eye 21.3% of length from snout tip to eye. Spiracle small, almost circular, maximum diameter of spiracle 6.1% prebranchial length. Two spineless dorsal fins, first dorsal fin higher than second dorsal fin, height of first dorsal fin 4.9% PCL and height of second dorsal fin 2.9% PCL; overall length of second dorsal fin 88.7% of that of first dorsal fin (Fig. 4). Interdorsal space longer than HL, its length 30.7% PCL. Pectoral fins short, broadly rounded free rear tips, length of anterior margin 15.5% PCL (Fig. 4). Pelvic fins small, overall length 85.3% of distance between base of pelvic fin and caudal fin (Fig. 4). Distance between base of pectoral fin and pelvic fin 42.3% PCL. Caudal peduncle short, distance between base of second dorsal fin and caudal fin 9.6% PCL; short lateral keels on the caudal peduncle, length of keel 5.1% PCL. No precaudal pits. Caudal fin asymmetrical and paddle-shaped, with a relatively short upper lobe and long lower lobe, and strong subterminal notch; length of lower caudal lobe and 90.2% of length of upper lobe (Fig. 4). Color uniformly brownish, with narrow but not conspicuous dark brown edges on dorsal, pectoral and pelvic fins, and caudal fin with wider but also not conspicuous dark brown edge (Fig. 2). Teeth dissimilar in upper and lower jaws, upper small, with lanceolate, arranged in 72 rows; lower with semioblique with low roots, arrange in 28 rows (Fig. 5). Precaudal vertebrae 60 (60–61), caudal vertebrae 17 (15–18), and 77 (75–78) in total; number of turns in spiral valve 25–26 (Table 2). Dermal denticles of lateral trunk flat basically Table 1. Proportional dimensions expressed as the percentage of the total length for Somniosus (Rhinoscymnus) cheni (Sc) collected from Taiwan (TW); S. (R.) rostratus (Sr) including the holotype of S. bauchotae (Sb) collected from the Mediterranean Sea (MD), New Zealand (NZ) and North Atlantic (NA); and S. (R.) longus (Sl) collected from Japan (JP) and NZ. Data sources: 1. this study; 2. Francis et al. 1988; 3. Barrull and Mate 2001; 4. Yano et al. 2004 Locality TW NZ MD MD MD NA NA JP JP JP NZ NZ Species Sc Sr Sr Sr Sr Sr Sb Sl Sl Sl Sl Sl Sources 1 2 3 444444444 Sex female female female female female female female female male male female female Total length (cm) 134.0 143.3 100.0 27.9 82.5 103 131.4 130.1 98.3 109.8 106.7 129.5 Snout tip to: 1 outer nostrils 2.54#$ 2.1 2.2 1.5 2.1 1.1 1.2 1.6 1.8 1.3 1.4 2 eye 5.60 4.2 6.0 6.1 3.9 5.3 4 4.6 4.9 5.6 4.9 4.1 3 spiracle 9.18 8.2 11.5 11.8 8.8 10.3 8.1 8.1 9.6 10.3 9.1 8.5 4 mouth 5.52 7.2 5.7 6.6 4.5 6.5 4.8 6.5 6.1 5.5 5 1st gill opening 14.63 13.9 17.9 15.8 16.6 13.2 14.9 15.6 16.6 15.7 14.1 6 2nd gill opening 16.12$19.7 17.1 14.2 16.8 16.7 17.9 7 3rd gill opening 17.54 21.5 18.2 18.2 15.9 17.8 17.3 18.9 8 4th gill opening 18.88 22.9 19.8 16.5 19.3 18.5 19.9 9 5th gill opening 19.85$24.7 20.6 20.9 17.6 20.4 20.2 21.3 10 pectoral origin 21.04 18.4 25.1 21.5 21.7 18.3 21.1 20.3 21.9 21.9 19.9 11 pelvic origin 62.39 60.2 60.9 63.0 62.9 64.5 61.0 61.0 61.9 62.6 62.1 12 cloaca 64.48 64.9 66.5 67.0 67.5 64.6 63.6 65.6 13 1st dorsal origin (AO) 27.84 37.6 34.5 38.8 35.8 34.9 38.8 36.2 35.7 32.9 14 1st dorsal origin (PO) 36.34 35.9 37.0 15 2nd dorsal origin 68.58 66.2 71.0 65.2 69.0 67.6 69.4 67.4 65.7 65.6 68.7 68.8 16 upper caudal origin 82.46$80.7 80.3 82.7 81.7 82.5 81.6 80.6 81.5 82.3 81.3 17 lower caudal origin 81.04 78.7 74.0 77.8 81.5 79.8 80.4 80.1 79.9 79.2 81.2 79.0 Distance between bases: 18 1st and 2nd dorsal 25.30 23.9 26.5 21.1 26.3 24.5 24.4 23.8 18.5 22.6 24.4 27.4 19 2nd dorsal and caudal 7.91 8.6 6.5 9.0 7.9 7.9 8.8 8.2 7.9 7.3 7.7 7.7 20 pectoral and pelvic 34.85 41.8 30.1 35.2 35.1 39.9 35.1 33.8 35.1 21 pelvic and caudal 11.64$11.5 11.8 11.2 11.0 11.5 11.5 11.4 Distance between origin of: 22 pectoral and pelvic 41.34 36.9 39.0 41.0 45.3 39.6 40.0 40.1 41.2 43.0 page 4 of 11Zoological Studies 59:48 (2020) © 2020 Academia Sinica, Taiwan Locality TW NZ MD MD MD NA NA JP JP JP NZ NZ Species Sc Sr Sr Sr Sr Sr Sb Sl Sl Sl Sl Sl Sources 1 2 3 444444444 Sex female female female female female female female female male male female female Nostrils: distance between 23 inner corners 2.99 3.3 4.3 2.1 3.3 3.3 2.8 3.4 3.4 3.6 3.2 24 Mouth: width 7.39$6.0 9.7 6.9 6.0 5.7 5.8 6.0 6.8 6.3 6.2 Gill opening lengths: 25 1st 1.94 1.7 1.8 1.5 1.6 2.0 1.5 1.5 2.0 2.1 1.9 26 2nd 2.01#1.8 1.7 1.6 1.5 1.5 2.0 27 3rd 1.64 1.8 1.7 1.8 1.5 1.5 1.5 1.9 28 4th 1.72 1.8 1.9 1.6 1.5 1.5 1.9 29 5th 1.79 2.4 2.2 2.2 2.6 2.2 2.3 1.8 2.4 2.3 2.7 30 spiracle 0.90#0.7 0.5 0.8 0.6 0.8 0.4 0.7 Eye: 31 horizontal diameter 1.19 1.5 2.9 2.2 1.9 1.8 1.7 1.7 1.6 1.5 1.3 32 vertical diameter 1.12 1.4 0.8 1.6 1.2 1.4 0.7 1.5 1st dorsal fin: 33 overall length (from AO) 19.10#$ 14.7 13.3 14.6 10.4 12.5 13.7 13.7 34 overall length (from PO) 10.60 13.5 35 length base (from AO) 7.99#$ 9.0 8.1 8.4 8.4 8.9 7.4 8.6 8.1 7.8 36 length base (from PO) 6.94 6.4 7.5 37 length posterior margin 5.60 5.1 5.7 6.7 6.6 3.0 4.2 7.6 5.9 38 height 4.03 4.0 5.5 3.6 4.2 3.7 3.6 3.9 4.4 3.8 4.0 4.0 39 Interorbital width 8.66 9.7 9.0 8.3 7.8 8.8 7.6 8.3 2nd dorsal fin: 40 overall length 9.40 10.0 11.8 11.8 11.2 10.4 11.3 13.6 13.7 41 length base 5.97 5.9 5.0 6.1 6.8 5.6 5.9 6.3 7.5 8.8 5.7 4.7 42 length posterior margin 6.27 4.5 7.2 6.7 6.0 5.5 7.3 7.5 6.8 43 height 2.39 2.2 7.0 2.5 2.1 2.2 2.3 2.4 2.5 2.1 2.5 2.4 Pectoral fin: 44 length base 6.49 5.3 5.5 6.8 4.2 5.6 5.5 5.6 5.4 5.5 6.4 6.9 45 length anterior margin 12.76 12.7 12.0 13.3 11.8 12.7 11.6 11.5 13.5 11.4 12.2 12.4 46 length distal margin 2.84 2.5 6.8 4.8 5.3 4.7 3.8 5.0 4.6 47 length posterior margin 7.99#$ 6.1 7.3 6.6 5.5 7.3 7.8 6.4 6.6 6.6 Pelvic fin: 48 overall length 9.93 10.0 9.3 9.5 8.8 9.5 11.5 9.6 9.4 9.7 49 length base 7.01#5.6 5.0 6.8 5.7 6.8 6.4 7.1 6.5 5.7 50 length anterior margin 6.42 5.5 6.5 5.1 6.0 5.2 6.5 6.5 6.1 5.6 6.8 51 length distal margin 5.30#$ 3.5 3.6 3.9 2.9 2.7 2.5 0.3 0.7 52 length claspers 9.0 7.1 53 (from pelvic axil) 1.3 1.6 Caudal fin: 54 length dorsal lobe 15.97 19.90 17.5 19.7 17.6 18.9 16.7 18.4 19.6 18.2 17.8 18.9 55 length ventral lobe 14.40 15.5 14.0 14.7 13.3 14.4 14.4 13.1 15.0 14.8 13.6 14.0 56 dorsal tip to notch 7.91 5.7 8.4 7.2 8.1 8.9 9.0 8.3 57 depth notch 1.64#7.5 2.9 2.1 2.4 1.7 1.5 1.3 1.5 Trunk at pectoral origin: 58 width 11.34 10.8 11.5 13.0 11.8 11.5 10.1 10.7 59 height 17.01#$ 9.7 10.3 10.6 11.6 12.5 7.6 11.2 60 Length from snout tip to pectoral end 28.73 36.6 33.7 34.0 30.1 32.1 32.9 32.6 61 Distance between preoral clefts 8.51 62 Length eye to 1st gill opening 7.84#8.2 9.5 8.0 9.0 9.0 8.8 63 Keel length 4.18$4.3 3.6 4.4 4.6 2.2 3.2 3.6 2.8 3.9 Each number of measurements is from figure 1 in Yano et al. 2004. #: significantly different to the mean of Sr and Sb (P < 0.05). $: significantly different to the mean of Sl (P < 0.05). Table 1. (Continued) page 5 of 11Zoological Studies 59:48 (2020) © 2020 Academia Sinica, Taiwan rhomboidal in shape with wide crowns and without conspicuous horizontal cusps, giving the skin a smooth texture (Fig. 6). Distribution: Currently known from the holotype collected in canyons surrounded by deep-sea longliner off eastern Taiwan at a depth greater than 500 m. Comparison: According to comparison of length proportional measurements among the present individual and two other Rhinoscymnus species, a total of 18 item values are outside of the measurement range of S. (R.) rostratus. Of these 18 items, 11 were significantly different to the average length (P < 0.05) (Table 1). In addition, a total of 25 item values are outside of the measurement range of S. (R.) longus, and 12 of these were significantly different to the average length (P < 0.05) (Table 1). In total, 16 item values are outside of the measurement range of both S. (R.) rostratus and S. (R.) longus, six items of these were significantly different to the average length of both species (P < 0.05) (Table 1). In meristic counts, the number of tooth rows in the upper jaw 72 and number of tooth rows in the lower jaw 28 are clearly different from 53–57 tooth rows in upper jaw and 31–36 in lower jaw in two other Rhinoscymnus species (Table 2, Fig. 5). The number of turns in the spiral valve (25–26) is similar to that of S. (R.) longus (26–27) but higher than S. (R.) rostratus (23) (Table 2). The total vertebral number (77) is similar to those of two other species (77–79), with similar combinations of precaudal and caudal vertebrae. The arrangement of dermal denticles (Fig. 6) is loose, and there are spaces between denticles, similar to in S. (R.) longus (Yano et al. 2004: fig. 8C), but unlike S. (R.) rostratus (Cigala Fulgosi and Gandolfi 1983: fig. 8G) whose denticles are arranged closer together with more overlaps. The shape of the dermal denticles without conspicuous horizontal cusps is different from those of two other Somniosus (Rhinoscymnus) species (Fig. 6; vs. Cigala Fulgosi and Gandolfi 1983: fig. 8G; Yano et al. 2004: fig. 8C). Moreover, Yano et al. (2004) reported that the overall length of the second dorsal fin and the first dorsal fin and horizontal diameter of eye of length from Fig. 3. Closeup of head portion of Somniosus (Rhinoscymnus) cheni sp. nov. from the holotype. A, dorsal view. B, ventral view. C, lateral view of right side (reversed laterally). Not to scale. Table 2. Meristic characters of Somniosus (Rhinoscymnus) cheni sp. nov., S. (R.) rostratus and S. (R.) longus Species Somniosus (R.) cheni sp. nov. S. (R.) rostratus S. (R.) longus Sources This study Yano et al. 2004 Yano et al. 2004 Number of tooth rows Upper jaw 72 53 56–57 Lower jaw 28 31–36 31–32 Number of spiral valve turns 25–26 23 26–27 Vertebral count Precaudal 60–61 58 58–59 Caudal 15–18 20 18–21 Total 75–78 78 77–79 page 6 of 11Zoological Studies 59:48 (2020) © 2020 Academia Sinica, Taiwan Fig. 4. Closeup of fins of Somniosus (Rhinoscymnus) cheni sp. nov., from the holotype. A, right pectoral fin, dorsal view. B, right pelvic fin, dorsal view. C, first dorsal fin, left side, lateral view. D, second dorsal fin, left side, lateral view. E, caudal fin, left side, lateral view. Not to scales, anterior to left. Fig. 5. Closeup of teeth of Somniosus (Rhinoscymnus) cheni sp. nov., from the holotype. Fig. 6. Dermal denticles on the trunk below first dorsal fin of holotype, Somniosus (Rhinoscymnus) cheni sp. nov., anterior to up. page 7 of 11 Zoological Studies 59:48 (2020) © 2020 Academia Sinica, Taiwan snout tip to eye are diagnostic for distinguishing S. (R.) rostratus and S. (R.) longus. The overall length of the second dorsal fin 49.2% of first dorsal fin is lower than those of S. (R.) rostratus (76–88%) and S. (R.) longus (91–100%); horizontal diameter of the eye of the length from snout tip to eye 21.3% is lower than 33% of S. (R.) rostratus and 46% of S. (R.) longus. Key to species of the genus Somniosus 1a. Small sharks with adults not exceeding 1.5 m in total length; rhomboid-shaped dermal denticles; semioblique cusps teeth on lower jaw, number of tooth rows in the lower jaw 28–36; number of turns in spiral valve less than 28; vertebral counts more than 74 ................................................................................................ 2 1b. Large sharks (adults about 4 m or more); hook-shaped dermal denticles; strongly oblique cusps lower teeth, number of tooth rows in the lower jaw 45–63; number of turns in spiral valve more than 29; vertebral counts less than 45 ............................... 4 2a. Number of tooth rows in the upper jaw 72, number of tooth rows in the lower jaw 28 ....................... S. (R.) cheni (Eastern Taiwan) 2b. Number of tooth rows in the upper jaw 53–57, number of tooth rows in the lower jaw 31–36 ...................................................... 3 3a. Overall length of second dorsal fin 76-88% of that of first dorsal fin; horizontal diameter of eye 27–37% (mean 33%) of length from snout tip to eye; number of tooth rows in the upper jaw 53; number of turns in spiral valve 23 ................................................. ..................... S. (R.) rostratus (Mediterranean and Atlantic Ocean) 3b. Overall length of second dorsal fin almost equal to first dorsal fin (91–100%); horizontal diameter of eye more than 30–56% (mean 43%) of length from snout tip to eye; number of tooth rows in the upper jaw 56–57; number of turns in spiral valve 26–27 .............. ....................................... S. (R.) longus (Japan and New Zealand) 4a. Interdorsal space almost equal to length from snout tip to first gill opening (prebranchial length); length from snout tip to first dorsal origin less than 45% of total length; number of turns in spiral valve 29-34 (mode 31); precaudal vertebral number 31–36 ......................... S. (S.) microcephalus (Arctic and North Atlantic) 4b. Interdorsal space less than length from snout tip to first gill opening; length from snout tip to first dorsal origin more than about 45% of total length; number of turns in spiral valve more than 32; precaudal vertebral number less than 31 ...................... 5 5a. Interdorsal space about 70% of prebranchial length; height of first dorsal fin about 3.7% of precaudal length (PCL), height of second dorsal fin about 3.4% of PCL; number of turns in spiral valve 32–37 (mode 33); precaudal vertebral number 28–30 (mode 29) ............................................... S. (S.) pacificus (North Pacific) 5b. Interdorsal space about 80% of prebranchial length; height of first dorsal fin about 3.0% of PCL, height of second dorsal fin about 2.9% of PCL; number of turns in spiral valve 36–41 (mode 39); precaudal vertebral number 30-31 (mode 30) ........................ ................ S. (S.) antarcticus (South Indo-Pacific, South Atlantic) Embryos biology: A total of 33 mid-term embryos—eight males and 25 females—were found in the specimen and preserved in EBFS and NMMBA (Fig. 7). The sex ratio of 8:25 indicated that there were significantly more female than male embryos (X2 = 8.76, P = 0.003). All embryos contained a huge yolk sac, with body pigments deposited as light brown to yellow-grayish, teeth and dermal denticles were totally absent, sexes could be identified, and a caudal keel had developed (Fig. 7). However, only two male and two female embryos with complete yolk sac were preserved, one female remained partial yolk, and others’ yolk sacs were damaged during transportation and dissection (Figs. 7 and 8). Embryonic TLs were 13.7–14.7 cm for males (average 14.3 cm, n = 8) and 12.8–15.0 cm for females (average 14.0, n = 25), and BWs were 15.70–19.62 g for males (average 17.4 g, n = 6) and 13.47–19.54 g for females (average 16.3 g, n = 22). The average TL was not significantly different between sexes (t = 1.54, P > 0.05), and neither was the average BW (t = 1.58, P > 0.05), thus the average TL and BW of embryos were 14.0 cm and 16.55 g, respectively. Total weights of the other four embryos with complete yolks were 90.92–100.00 g, with an average of 95.47 g. DISCUSSION Before the present specimen, only S. (S.) pacificus of genus Somniosus had been recorded in Taiwan, and most individuals were caught in Hualien by deep-sea longline fishery, a small and localized fishery where very few coastal fishing boats operate. This kind of deep-sea fishery deployed 500–1000 m deep bottom-set longlines in canyons surrounding seamounts in a small region, and the present specimen was also caught by the same fishery in a similar operation area (Wang and Yang 2004). Small sleeper sharks were found and named in 1827 and 1912; however, there are very few recorded encounters, particular for S. (R.) longus (Risso 1827; Tanaka 1912). The S. (R.) rostratus specimens were also only found from the Mediterranean, the North Atlantic, and New Zealand waters in the Southern Hemisphere (Yano et al. 2004; Ebert et al. 2013a). In the report from Yano et al. (2004), the ratio of the overall length of two dorsal fins and the proportion of the horizontal diameter of eye of the length from snout tip to eye are two of keys to distinguish S. (R.) longus and S. (R.) rostratus. In S. (R.) cheni, these two values are clearly smaller than in the two other species, although these measurements might be different because they were preserved specimens (Sotola et al. 2019). In addition, at least six morphometric measurements in S. (R.) cheni are significantly different to those of S. (R.) longus and S. (R.) rostratus (Table 1). The other keys using tooth count and the number of turns in the spiral valve to distinguish S. (R.) longus and S. (R.) rostratus were also used to distinguish S. (R.) cheni (Table 2; Yano et al. 2004). The tooth count in the upper jaw is clearly higher in S. (R.) cheni than in S. (R.) rostratus or S. longus, and the tooth count in page 8 of 11Zoological Studies 59:48 (2020) © 2020 Academia Sinica, Taiwan the lower jaw is clearly lower in S. (R.) cheni than in S. (R.) rostratus or S. (R.) longus (Yano et al. 2004). The number of turns in the spiral valve of S. (R.) cheni is similar to that of S. (R.) longus but clearly higher than that of S. (R.) rostratus (Table 2; Yano et al. 2004). There were some observations on the reproduction of S. (R.) rostratus in previous studies. Two pregnant females from the Mediterranean Sea contained nine and eight embryos, one litter of which contained six males and two females (Cigala Fulgosi and Gandolfi 1983; Barrull and Mate 2001). Litter sizes of two S. (R.) rostratus individuals were similar (8–9), but relatively Fig. 7. Embryos of holotype, Somniosus (Rhinoscymnus) cheni sp. nov., individuals of the first column from the left are males, the others are females. page 9 of 11Zoological Studies 59:48 (2020)