Squalus shiraii sp. nov. (Squaliformes, Squalidae), a new species of dogfish shark from Japan with regional nominal species revisited

A new species of deep-water dogfish shark, Squalus shiraii sp. nov., is described herein as endemic to the tropical waters off Southern Japan. This species has been largely misidentified with S. mitsukurii. However, morphological, meristic and morphometric evidence support it to be a separate and undescribed species. Squalus shiraii sp. nov. differs from this species by having body brown in colour dorsally, caudal fin with ventral and dorsal tips markedly tapered and broadly white, dermal denticles uniscuspidate and lanceolate and larger number of precaudal (91–94) and total vertebrae (120–123) (vs. body dark grey to black; caudal fin with ventral and dorsal tips rounded and not white in colour; denticles tricuspidate and rhomboid; 86–90 precaudal and 116–117 total vertebrae). Squalus shiraii sp. nov. is also clearly separated from other Japanese congeners which are herein revisited to include six species, based on the examination of over 150 specimens caught from Japanese waters that were available in ichthyological collections: S. mitsukurii, S. japonicus, S. acutirostris, S. brevirostris and S. suckleyi. Squalus mitsukurii, S. japonicus and S. brevirostris are re-described in detail and the neotype of S. japonicus is herein designated. Squalus acutirostris is treated as a valid species with occurrences in Japan, China and Taiwan and, thus, a provisional diagnosis is given, as well as an updated diagnosis of S. suckleyi. A key to Squalus species from the North-western Pacific Ocean is given and main morphological differences between S. shiraii sp. nov. and the closest related species are discussed.


Introduction
Dogfish sharks of the genus Squalus Linnaeus, 1758 represent commercially-important taxa within the world fish trade, in which individuals are caught through direct or indirect fisheries and traded for consumption of meat, fins and liver oil as primary products. Although highly exploited, the landing reports, observer and logbooks data and/or fisheries surveys have constantly pointed out that accurate identification of species is scarce (e.g. da Silva et al. 2015;Hacohen-Domené et al. 2020). Fisheries management and conservation of dogfish sharks are thus still outstanding as species-specific catch and landing statistics, population threats and trends are not yet acknowledged.
Japan represents one of the world's leading shark fisheries country whose estimated annual shark landing data for between 1992 and 2000 was 19,600-28,700 t (Sonu 1998;Matsunaga et al. 2003). Over 78 elasmobranch species currently evaluated as threatened (VU, vulnerable; CR, critically endangered; EN: endangered) in the Red List of Threatened Species of the International Union for Conservation of Nature (IUCN) are traded in the Japanese shark fin markets, highlighting it as a priority country for elasmobranch conservation (Dulvy et al. 2017), even though shark catches have decreased drastically (Sonu 1998). Squalus and other squalid sharks have been exploited in Japan as by-catch fisheries of Pacific salmon gillnets, tuna longliners and squid driftnets (Wilson and Seki 1994;Nakano and Nagasawa 1996;McKinnell and Seki 1998), but fishery statistics related to Squalus stocks are unreported or deficient and usually inserted into the category of "other species".
Five valid species (of 36 valid Squalus species) are often recognised in the country (Chen et al. 1979;Nakaya 1985;Shirai 1989;Nakaya and Shirai 1992;Zhu and Meng 2001;Nakabo 2013;Shinohara et al. 2014;Dyldin 2015): S. blainvillei (Risso, 1827), S. suckleyi (Girard, 1854), S. mitsukurii Jordan & Snyder, 1903, S. japonicus Ishikawa, 1908and S. brevirostris Tanaka, 1917. Two additional nominal species, S. acutirostris Li, 1984 andS. wakiyae Tanaka, 1917, originally described from Japan, are considered junior synonyms of S. mitsukurii and S. acanthias, respectively (White et al. 2007c;White and Last 2013). A sixth species, S. formosus White & Iglésias, 2011, a Taiwanese endemic, is likely to inhabit Japanese waters as well, but requires confirmation. These species are found on lower continental shelves, upper continental slopes and, more rarely, on insular slopes and seamounts of the North Pacific and Eastern Indian Oceans between Russia, Japan, China, South Korea, Taiwan and the Philippines. These species are all now classified as IUCN Data Deficient (DD), except for S. suckleyi (Least Concern, LC) (IUCN 2020). A key criterion to be inserted into the first category is unreliability of species identification and, thus, species-specific data are inaccurate or doubtful and may not be incorporated into the assessments.
Systematic approaches of the last decade support that S. mitsukurii comprises a species complex that also includes S. blainvillei, a species originally described from the Mediterranean Sea (Ward et al. 2007;Naylor et al. 2012;Veríssimo et al. 2017). Distribution records of S. mitsukurii from outside the North Pacific Ocean were successively refuted: Duffy and Last (2007b) and Last et al. (2007d) for the South Pacific Ocean; Viana et al. ( , 2017bViana et al. ( , 2018 for the South Atlantic and Western Indian Oceans;and Verissimo et al. (2017) for the North-eastern Atlantic and Pfleger et al. (2018) for the North-western Atlantic Oceans. Additional species, previously misidentified with S. mitsukurii, were also more recently recognised in the North Pacific Ocean, S. hawaiiensis Daly-Engel, Koch, Anderson, S. boretzi Dolganov, 2019. Records of S. japonicus and S. brevirostris are exclusive to the Indo-Pacific region (Chen et al. 1979;Compagno 2000;Last et al. 2007b;Manjaji-Matsumoto 2017) with uncertainties regarding their occurrences in Northern Australia and Indonesia due to morphological similarities with S. megalops (MacLeay, 1881) and S. nasutus Last, Marshall & Stevens, 2007. Taxonomic confusions within this Linnaean group are ordinary because of fair original descriptions, in-distinct morphological diagnostic characters and exaggerated availability of synonyms. Species delimitation using DNA barcoding of mitochondrial genes alone has shown to be ineffective on the genus (e.g. Bineesh et al. 2016;Vella et al. 2017) when non-designation of species identifications in the DNA reference libraries persists as collection-based data are not incorporated within the taxonomic investigations. Comparative examination of over 150 specimens of Squalus from the North Pacific Ocean, particularly from Japan, support the recognition of a novel species that has not been included in the latest molecular genetic and morphological taxonomic analyses of the genus. The present study aimed to describe the new species, based on morphological characters and to revise the nominal species of Squalus occurring in the region.

Morphological analysis and species description
Comparative analyses of external morphology were made, based on specimens preserved in 70% ethanol. Maturity stage was determined according to Compagno (1988). Description of external morphology and colouration is in accordance with Last et al. (2007f). Colouration is described from preserved specimens unless otherwise noted. Other terminology follows Cappetta (1987) for dentition and Deynat and Séret (1996) for dermal denticles. Synonyms for species indicate authorship, date and page when possible. Tooth samples were taken from upper and lower jaws (three lateral teeth towards the first series) and later examined using a stereoscopic microscope. Squamation is based on skin samples measuring 1 cm 2 obtained from below the first dorsal fin (right side) and was analysed under a Scanning Electron Microscope (SEM) at the Instituto de Biociências, Universidade de São Paulo (IBUSP) and Rhodes University (RU).

Morphometric analysis
External measurements were obtained from preserved specimens using a digital caliper with 0.1 mm precision and/or a metric tape for measurements larger than 150 mm. Morphometrics follow Last et al. (2007f) and  and are expressed as percentages of total length (% TL). Morphometrics of the species are provided in Tables 1 to 5. Single values in the diagnosis and description correspond to holotype and range values to all other material from which data was obtained, except when first mentioned. Values for mean and standard deviation in tables include all specimens for which data were taken. Range values for diagnosis, discussion and identification key include adult specimens only, unless otherwise stated in text.

Meristic data
Tooth counts were taken according to Herman et al. (1989) from preserved specimens. Vertebral counts were obtained through film and digital radiographs and follows Springer and Garrick (1964). Meristic data of the species are provided in Table 6. Single values in the diagnosis and description correspond to holotype and range values to additional material from which data was obtained except otherwise noted.

Description.
Single values correspond to the holotype and range to paratypes, respectively.
Pelvic-caudal space 1.4 (1.1-1.6) times pectoral-pelvic space. Caudal keel evident laterally in the caudal peduncle from second dorsal fin free rear tip to caudal-fin origin. Caudal fin with conspicuously thin upper and lower caudal lobes (Fig. 1I, M); caudal fork concave with its width corresponding to 6.4% (5.7%-6.4%) TL; dorsal caudal margin straight and markedly elongate, its length 0.9 (0.9-1.0) times head length and 2.1 (1.9-2.1) times length of preventral caudal margin; upper postventral margin straight on its upper half and slightly convex on its lower half; lower postventral margin convex; preventral caudal margin convex and short, its length 1.8 (2.1-2.3) times length of pelvic-fin inner margin; dorsal caudal tip rounded; ventral caudal tip markedly pointed. Squamation (Fig. 7A, B). Dermal denticles lanceolate and imbricate, broad at the crown base (conspicuously broad in females); length of dermal denticles larger than its width (although it is about equal to its width in females); denticles with median cusp slightly pointed lateral cusps inconspicuous; median and lateral ridges thick; median and lateral ridge bifurcated anteriorly; anterior furrow conspicuously profound and oval.
Colouration (Fig. 1). Body brown dorsally, light brown laterally, although whitish latero-ventrally from the pelvic fin to the caudal fin and pale ventrally. Dorsal fins brownish, somewhat light brown at dorsal fin base; first dorsal-fin inner margin white; second dorsal-fin posterior margin fairly white at its lower half; second dorsal-fin apex somewhat dark brown. Dorsal-fin spines brownish anterior-laterally and white at the tip. Pectoral fins dark brown, whitish ventrally at pectoral-fin base; pectoral-fin posterior margin fairly white (not uniform); pectoral-fin apex broadly white. Pelvic fins brown and whitish at the pelvic-fin base; pelvic-fin anterior and posterior margins slightly white. Caudal fin dark brown at upper and lower caudal lobes, whitish over vertebral column; postventral caudal margins white, except at the caudal fork where it is dark brown; dorsal and ventral caudal tips broadly white; preventral caudal margin white; dark caudal stripe and bar inconspicuous; black upper caudal blotch somewhat evident in adults. Juveniles with body light brown dorsally, pale lateral and ventrally; first and second dorsal-fin posterior margins slightly white at its lower half; dorsal-fin apex blackish to dark brown at the tip; pectoral fins light brown to reddish-brown with pectoral-fin posterior margins and apex broadly white; pelvic fins light brown to reddish-brown; caudal fin dark brown with postventral caudal margins broadly white, except at the caudal fork; dorsal and ventral caudal tips broadly white; conspicuous upper black caudal blotch in the midline of dorsal caudal margin; lower caudal lobe with oblique black marking at the origin of lower caudal lobe; small black caudal bar at the caudal fork. Geographical distribution. This species is apparently a Japanese endemic, occurring in the shallow waters of the upper continental slope off Southern Japan in the North-western Pacific Ocean at 310-390 m depth (Fig. 8A).

Vertebral counts (
Remarks. Squalus shiraii sp.nov. is often misidentified with S. mitsukurii due to similarities concerning the general shape of body, pectoral and caudal fins and snout length. However, it clearly differs from this species in addition to the characters provided in the diagnosis above by having an obtuse snout, dorsal and ventral caudal tips conspicuously pointed, upper and lower caudal lobes markedly tapered, pectoral fins conspicuously falcate, dermal denticles lanceolate and unicuspidate (vs. snout rounded, dorsal and ventral caudal tips rounded, upper and lower caudal lobes broad, pectoral fins not falcate, dermal denticles rhomboid and tricuspidate in S. mitsukurii). Squalus shiraii sp. nov. has body brown in colour, postventral and preventral caudal margins whitish, dorsal and ventral caudal tips broadly white and black upper caudal blotch evident in adults. S. mitsukurii has body conspicuously black to dark grey and caudal fins black throughout with post-ventral caudal margin fairly whitish and black upper caudal blotch not evident in adults.
Preoral length 1.2 (1.0-1.5) times greater than mouth width. Mouth somewhat straight and conspicuously broad (mouth width 1.8, 1.5-2.1 times internarial width and 1.5, 1.2-1.8 times prenarial length); upper labial furrow elongate, its length 2.5% (2.1%-2.7%) TL, bearing a thin fold; lower labial furrow markedly elongate, lacking fold. Teeth unicuspid, flattened labial-lingually, similar in both jaws; teeth somewhat rectangular, very broad at the crown; cusp short, thick, pointed and oblique; mesial cutting edge convex and oblique; distal heel conspicuously rounded; mesial heel sharp; apron thick and small, placed more laterally in the upper teeth and in the midline of crown base in the lower teeth, slightly more elongate in the lower teeth than in the upper teeth; median teeth present on upper jaw in adult paratype only and distinct from the subsequent teeth by: teeth hexagonal with cusp and apron placed more medially; both mesial and distal heels pointed (Fig. 3C, D). One to two series of functional teeth on upper jaw and two series in lower jaw; 13-0-15 for holotype (13-1-13) teeth rows on upper jaw; 12-0-12 (11-0-11) teeth rows on lower jaw.
Caudal keel prominent and situated laterally in the precaudal peduncle since insertion of second dorsal fin to behind origin of caudal fin. Caudal fin conspicuously slender in the upper caudal lobe; dorsal caudal margin somewhat convex proximally and conspicuously straight distally (Fig. 2I, M); caudal fin elongate with dorsal caudal margin length 1.0 (0.9-1.1) times head length and 2.0 (1.7-2.0) times larger than length of preventral caudal margin; upper postventral caudal margin straight; lower postventral caudal margin somewhat convex; preventral caudal margin convex, its length 1.9 (1.5-3.8) times length of pelvic-fin inner margin; both posterior and ven-tral caudal tips rounded and thin; caudal fork discontinuous and conspicuously concave, its width 7.0% (6.1%-7.7%) TL. Squamation (Fig. 7E, F). Dermal denticles tricuspid, rhomboid to heart-shaped and imbricate, conspicuously broad at the crown, its length equal to its width; cusps pointed and posterior; lateral cusps often inconspicuous and much shorter than median cusp; median cusp conspicuous and wide; median ridge prominent, tall and thick with anterior furrow oval and profound; anterior margin of the crown broad, convex and arrow-shaped with one small ridge on each side; one to two lateral ridges, thin and smaller than median ridge. Dermal denticles slightly larger and more imbricate in adults than juveniles.
Colouration (Fig. 2). Body black dorsally, dark grey ventrally and laterally since first dorsal fin insertion to origin of caudal fin. Dorsal fins blackish, dark grey at fin base near the dorsal-fin spines; lower half of anterior margin of dorsal fins whitish. Dorsal-fin spines brownish anteriorly and white posteriorly and at the tips. Pectoral fins black dorsally and ventrally, although dark grey near pectoral-fin inner margin; pectoral-fin posterior margin fairly white. Pelvic fins dark grey dorsally and ventrally, somewhat dark brown at the fin base; pelvic-fin posterior margins slightly white. Caudal fin black, dark grey near caudal fin origin and proximally over vertebral column; dorsal caudal margin dark grey proximally; postventral caudal margins slightly white, except near the caudal fork; preventral caudal margin blackish. Embryos, neonates and young juveniles differ from adults by having body light grey, dorsal fins light grey with dorsal-fin apex blackish and dorsal-fin free rear tips white, caudal fin light grey with postventral caudal margins broadly white, dark caudal bar conspicuous, black upper caudal blotch prominent and black caudal blotch at the base of lower caudal lobe, located horizontally.
Vertebral counts (   This species possibly reaches depths of up to 980 m in the Western Pacific Ocean, but it is often caught between 180-300 m depth (Chen et al. 1979).

Description.
Single values correspond to the neotype and ranges to all other examined material from which data were obtained.
Caudal keel prominent and lateral since insertion of second dorsal fin to behind origin of caudal fin. Caudal fin small with dorsal caudal margin length 0.9 (0.8-0.9) times head length and 1.9 (1.8-3.1) times length of preventral caudal margin; dorsal and ventral caudal lobes slender (dorsal lobe rectangular in juveniles); dorsal caudal tip pointed; dorsal caudal margin slightly convex; upper postventral caudal margin straight; lower postventral caudal margin convex (Fig. 4I); preventral caudal margin convex and short, its length 2.0 (1.4-2.4) times greater than length of pelvic-fin inner margin; ventral caudal tip rounded; caudal fork discontinuous, markedly concave and narrow, its width 6.6% (6.0%-7.4%) TL. Squamation (Fig. 7G, H). Dermal denticles tricuspid and markedly imbricated, its length slightly greater than its width; denticles conspicuously broad at the crown base; cusps posterior, prominent and pointed; lateral cusps shorter than median cusp, forming well-marked concavity with median cusp; median ridge thick and elon-gate; one to two lateral ridges, very thin, almost equal in length to median ridge; anterior furrow shallow with one small ridge on each side; anterior margin of the crown arrow-shaped. In juveniles, denticles are slender at the crown base, with cusps conspicuously pointed and more elongate and anterior margin of the crown tapered.
Colouration (Fig. 4). Body grey to light grey dorsally and white laterally from insertion of pectoral fins to origin of caudal fin. Dorsal fins greyish, darker at its apex to the upper half of the posterior margin, whitish at fin base and free rear tips; dorsal-fin spine brown anteriorly, white posteriorly and at the tips. Pectoral fins grey, slightly darker near pectoral-fin anterior margin; pectoral-fin posterior margin strongly white and uniform. Pelvic fins grey dorsally and completely pale ventrally; pelvic-fin posterior margin white. Caudal fin mostly greyish, light grey over vertebral column; dorsal caudal margin whitish proximally; upper postventral caudal margin white, lower postventral caudal margin broadly white; lower caudal lobe mostly white; preventral caudal margin whitish; black caudal stripe conspicuous; black upper caudal blotch conspicuous in the upper caudal lobe, nearest to dorsal caudal tip; black caudal bar oblique, placed in the caudal fork to midline of upper postventral caudal margin; dorsal and ventral caudal tips broadly white. Embryos, neonates and young juveniles exhibit body light grey with caudal bar and upper caudal blotch more conspicuous than in adults.
Remarks. Ishikawa (1908) mentioned three syntypes of S. japonicus that were deposited at the Zoological University Museum of Tokyo (formerly Imperial Museum of Tokyo). A visit to the fish collection from ZUMT was undertaken in 2013 to verify the existence of the type material. However, it was noticed that there was a single specimen of S. japonicus held in the museum, assuming that syntypes are lost and no longer exist. Dr. M. Nakae (NSMT, person. comm. 2014) stated that the NSMT fish collection has never been in possession of this material and ratified that the syntypes are lost. Squalus japonicus has been misidentified in waters outside Japan (e.g. Vietnam by Quang et al. 2013) and many morphological variations were reported earlier in Compagno et al. (2005b) and Last et al. (2007b) for material from Indonesia, Philippines and Western Australia, compromising the correct application of this nominal species. To clarify the taxonomic status of S. japonicus across the Western Pacific Ocean, the designation of a neotype along with a re-description of this species is provided herein. The neotype was collected from Southern Japan (Senoumi Bank, Suruga Bay), which is within the known range distribution of S. japonicus, based on the examined material, the original description of S. japonicus and information in literature. The designation of the neotype was further based on the sex, maturity and total length that were approximate to those provided for a syntype of S. japonicus in Ishikawa (1908). The neotype is securely deposited in the scientific fish collection from the NSMT in Tokyo at the Department of Zoology for preservation and accessibility purposes. This species is easily separated from its congeners through many characteristics of the external morphology, morphometrics and meristic data, including shape of head, snout and pectoral fins, length of snout, width of mouth and preoral length and several other external measurements as shown in the diagnosis above. The current diagnosis and re-description of S. japonicus are congruent with observations of Ishikawa (1908) and Chen et al. (1979) with the exception of the presence of two hooks in the clasper. These cartilages of the clasper probably concern the accessory terminal 3 cartilage (spur) and dorsal terminal cartilage (claw) that are commonly noticed in species of Squalus. Differences in morphometrics are observed between the data provided in Ishikawa (1908) and those from the present analysis, but this is due to divergences of the applied methodology.
This species together with S. melanurus and the Australian species S. nasutus, comprise the S. japonicus subgroup of species whose members are characterised by having a conspicuously elongate and obtuse snout, thin body, mouth very narrow and pectoral fins tapered (Viana and Carvalho 2018a). The Japanese species is more closely related to S. nasutus according to morphological and genetic data ). Shape of dermal denticles, dentition and shape of dorsal, pectoral, pelvic and caudal fins are intrinsically similar between these two species. Morphometrics are also mostly congruent, although differences in measurements are apparent when adults are compared: second dorsal fin height 5.4%-6.5% TL for S. japonicus vs. 4.9%-5.3% TL for S. nasutus; length of second dorsal-fin anterior margin 10.8%-12.7% TL vs. 8.6%-10.5% TL; length of second dorsal-fin spine 4.7%-5.6% TL vs. 3.9%-4.4% TL; length of dorsal caudal margin 19.6%-20.4% TL vs. 18.1%-19.4% TL. Preoral length, eye length, pectoral-fin inner margin length, trunk height and width overlaps between these two species which contradicts Last et al. (2007b) in supporting these measurements as diagnostic characters for these species. Vertebral counts differ between S. japonicus and S. nasutus as well: monospondylous (41-43), precaudal (84-88) vertebrae (vs. 37-40; 80-83 for those from Australia). These results are congruent with recent molecular analyses of the cytochrome c oxydase 1 (COI) and NADH dehydrogenase 2 (NAD2) mitochondrial genes in Ward et al. (2007), Naylor et al. (2012), Straube et al. (2013), Veríssimo et al. (2017) and Daly-Engel et al. (2018) which support genetic divergence between S. japonicus and S. nasutus. The latter study also noticed high intraspecific genetic distances in S. japonicus when including data taken from populations from Japan and Taiwan, but variations in the species are not yet supported through morphological evaluation, except for the black caudal marking near the base of the lower lobe in Taiwanese specimens that is absent in other material. Compagno et al. (2005b) also noticed variations in this species when examining material from the Philippines and considered it to be a separated species (Squalus sp. 2) due to differences with S. japonicus and S. nasutus. In contrast, Last et al. (2007b) considered it to be con-specific with S. nasutus as the morphological differences pointed in Compagno et al. (2005b) and in their study represent intraspecific variations. Whether the Philippine form represents an undescribed species is still unclear, but we consider it here to be con-specific with S. nasutus for nomenclature stability.
Preoral length somewhat equal to mouth width, corresponding to 1.1 (1.1-1.2) times the latter. Mouth markedly arched and narrow, its width 2.3 (1.8-2.3) times broader than internarial space; upper labial furrow small, its length 2.6% (2.0%-2.7% TL) with thin fold; lower labial furrow also short with fold subdivided into three small inter-digits. Teeth tiny and unicuspid, labial-lingually flattened and alternate, similar in both jaws, lower teeth slightly wider than the upper teeth; cusp short and oblique; mesial cutting edge convex; both distal and mesial heels rounded; apron elongate in upper and lower teeth (Fig. 3G, H). Three and two series of functional teeth on upper and lower jaws for holotype (two series for non-type specimens); 12-0-13 (12-0-12) teeth rows in upper jaw; 10-0-10 (10-0-10) teeth rows in lower jaw.
Pre-first dorsal length 1.4 (1.3-1.6) times prepectoral length. First dorsal fin short, its length 1.7 (1.5-1.9) times its height; first dorsal-fin low, its height 1.3 (1.2-1.5) times greater than its inner margin length; first dorsal-fin anterior margin concave and posterior margin straight; first dorsal-fin free, rear tip pointed; first dorsal-fin apex rounded and slender at fin web (Fig. 5G, K). Origin of first dorsal-fin spine anterior to vertical traced at free rear tips of pectoral fins; first dorsal-fin spine short (broken in the holotype), never reaching first dorsal-fin apex, its length 0.3-0.6 times first dorsal-fin height. Interdorsal space 1.1 (1.1-1.3) times prepectoral length and 2.4 (2.2-2.6) times larger than dorsal-caudal space. Pre-second dorsal length 2.8 (2.7-3.3) times greater than prepectoral length and 3.0 (2.8-3.2) times the length of dorsal caudal margin. Second dorsal fin very short, its length 0.9 (0.8-1.0) times first dorsal-fin length; second dorsal fin low, its height 1.2 (0.8-1.4) times length of second dorsal-fin inner margin; second dorsal-fin anterior margin concave and posterior margin markedly falcate; second dorsal-fin apex rounded and evidently slender at fin web (Fig. 5H, L). Second dorsal-fin spine thick and elongate, although not reaching the apex, its length 0.8 (0.8-1.1) times the fin height; second dorsal-fin spine length 1.3-2.6 times greater than length of first dorsal-fin spine.
Caudal keel prominent, reaching from forward second dorsal fin insertion to behind origin of caudal fin. Caudal fin slightly slender in the dorsal caudal lobe; dorsal caudal and upper postventral caudal margins convex; dorsal caudal tip rounded (Fig. 5J, N); dorsal caudal margin 0.9 (0.7-1.0) times head length and 2.0 (1.5-2.0) times larger than preventral caudal margin; lower postventral caudal margin convex (broken in holotype); preventral caudal margin strongly convex and short, its length 1.7 (1.6-2.3) times greater than length of pelvic-fin inner margin; ventral caudal tip rounded (broken in holotype); caudal fork discontinuous between lobes, its width 6.6% (6.6%-7.6%) TL. Squamation (Fig. 7I, J). Dermal denticles unicuspid and lanceolate, not imbricated; cusp rounded and posterior; denticles conspicuously slender at crown, although markedly broad medial-anteriorly at its base; denticles with length much greater than its width; median ridge tall and thick, bifurcated anteriorly with profound anterior furrow; lateral ridges tall and thin, shorter than median ridge. Colouration (Fig. 5). Body brownish-grey dorsally, white ventrally and latero-posteriorly Pectoral fins grey with pectoral-fin posterior and inner margins white. Pelvic fins white dorsally and ventrally. First dorsal fin brownish-grey, slightly white at first dorsal-fin posterior margin, free rear tip and fin base; second dorsal fin light brown, whitish at the fin base, somewhat white at second dorsal-fin apex, posterior margin and freer rear tip. Dorsal-fin spines grey, dark grey anteriorly and white at Geographical distribution. This species occurs in tropical and subtropical areas of the North-western Pacific Ocean from Southern Japan, China and Taiwan on continental shelves and upper continental slopes between 50-342 m depth (Fig. 8D). It is also recorded from the Philippines in Manjaji-Matsumoto (2017).

Remarks.
Morphological variations in the shape of pectoral and dorsal fins, dermal denticles and colour of caudal fin are observed within S. brevirostris. First dorsal-fin posterior margin is somewhat straight near its apex like in the holotype, although other specimens exhibit first dorsal-fin posterior margin concave. Pectoral fins are slightly broad at posterior margin in specimens from the Central Western Pacific Ocean (except Taiwan), while it is narrow in others like that observed for the holotype. Pelvic fins are nearer to first dorsal fin, although it may be in the midline between the two dorsal fins in a few specimens. Caudal fin has postventral caudal margins white, although narrowly white at caudal fork in specimens from Kagoshima. Thickness of the dermal denticles varies from very slender to conspicuously broad at the crown in few specimens from Kagoshima. These variations are possibly not related to dimorphism or ontogeny and apparently it is scattered, as it has an unclear morphological pattern throughout the distributional range of this species in the North and Central Western Pacific Ocean. Duffy and Last (2007a) previously noticed differences in the dermal denticles, which was suggested to be due to ontogeny. Further comparative investigations are required to include more specimens from Taiwan and South Korea in order to better understand these variations within the species.
Squalus brevirostris exhibits low vertebral counts that are usually overlapped to those of morphologically similar species occurring elsewhere, such as S. megalops, S. acutipinnis and S. lobularis. However, it can be separated from other similar species by precaudal vertebrae with S. hemipinnis, S. notocaudatus and S. formosus, total vertebrae with S. hemipinnis, S. notocaudatus, S. formosus and S. albifrons and monospondylous vertebrae with S. notocaudatus, S. formosus, S. albifrons and S. bucephalus.
Remarks. Molecular genetic data has supported the taxonomic separation between the morphologically similar species S. acanthias Linnaeus, 1758, that bears circumglobal distribution and S. suckleyi (e.g. Ward et al. 2007;Veríssimo et al. 2010;Ebert et al. 2010;Naylor et al. 2012;Lee et al. 2016). The latter species is a regional endemic to the North Pacific Ocean with distribution ranging from Korea to Japan (western side) and from California to Canada (east side). Morphological separation between these two species is still unreliable as no efficient diagnostic features were provided as yet. Lindberg and Legeza (1956) suggested that the ratio pelvic-fin midpoint to first dorsal-fin insertion (PDI) and pelvic-fin midpoint to second dorsal-fin insertion (PDO) comprise diagnostic characters as well as the first dorsal-fin midpoint to pectoral-fin insertion (DPI) and first dorsal-fin midpoint to pelvic-fin origin (Ebert et al. 2010); however, these parameters are subject to bias due to the midpoint definition that relies on subjectivity. Squamation, colouration and vertebral counts, however, separate S. acanthias and S. suckleyi (Table 6). The latter species exhibits dermal denticles with a median ridge forming an acute angle or right angle when perpendicular to the body axis (vs. in parallel for S. acanthias). It is easily distinguished from S. acanthias by having body conspicuously dark grey dorsally with very few white spots distributed dorsally in a single row on each side of the body (vs. light grey body with many white spots in two rows). Specimens of S. suckleyi from Japan are also distinct from S. acanthias occurring elsewhere by having smaller number of vertebrae (monospondylous, diplospondylous, precaudal and total vertebrae; see Springer & Garrick, 1964 for S. acanthias), corroborating Ebert et al. (2010). Additional characteristics of S. suckleyi include: dermal denticles arrow-shaped and unicuspid, small and not imbricated with a single ridge (lateral ridges absent); ridge very prominent, narrower distally than proximally with furrow anterior and profound; ridge very tall and convex, forming 45 degrees angle with horizontal axis of the body; crown base strongly broad and diamond-like with four prominent pedicels; dermal denticles conspicuously expanded laterally at the crown (Fig. 7K, L). A single adult male shows few dermal denticles with weak lateral cusps, although lateral ridges are still absent. Colouration of Japanese specimens consists of body dark grey dorsally with very few and inconspicuous white spots (1-6 pairs), rounded, distributed symmetrically in a single row on each side and laterally at the body; white spots commonly absent in some large specimens, although clearly evident in young juveniles; white to greyish-white ven-trally. Both dorsal fins grey, brownish in the apex, slightly darker from the apex tip to the midline of the posterior margin, white distally at posterior margin and free rear tips; fin base slightly white; dorsal spines dark brown, white at the tip. Pectoral fins dark grey dorsally and ventrally with white posterior margin, although not uniform. Pelvic fins also dark grey with posterior and inner margins slightly white. Caudal fin grey, darker at the tip of the dorsal lobe and in the lower caudal lobe, white over vertebral column; proximal end of dorsal caudal margin white; postventral caudal margins narrowly white; preventral caudal margin slightly white; black caudal stripe conspicuous above vertebral column.
Variations of diet and reproduction, as well as migration pattern, spatial distribution and size composition, were noticed in populations of S. suckleyi between the East and West sides of the North Pacific Ocean (Orlov et al. 2012a,b;Yano et al. 2017), suggesting that this species may comprise more than one fishing stock rather than a single one as supported in Ebert et al. (2010). In contrast to S. acanthias, this species has low migration movements and usually inhabits the pelagic zones up to 200 m depth with two possibly distinct populations observed throughout the North Pacific Ocean (Yano et al. 2017). Emperor Seamout Chain works as a biogeographic barrier between the East and West sides whose belt-shaped pelagic zone exhibits low probability of presence of S. suckleyi, according to Yano et al. (2017).
A single nominal species of spotted spiny dogfish, Squalus wakiyae Tanaka, 1917, was described from the North-western Pacific Ocean and it has been placed under synonymy with S. acanthias and S. suckleyi in Fowler (1941), White et al. (2007c), Ebert et al. (2010) and the present study, as no apparent differences in morphology and genetic data are noticed between the west and east populations of S. suckleyi. Morphometric variations, however, are noticed when comparing the neotype (taken from Ebert et al. 2010) to specimens from off Japan and Russia: greater prepelvic length than in specimens from the NWPO (51.5% TL for neotype vs. 47.2%-50.6% TL); greater pre-vent length (54.9% TL vs. 50.1%-53.6% TL); narrower mouth (6.2% TL vs. 6.9%-8.2% TL); lower head height (6.5% TL vs. 7.8%-10.6% TL); pectoral fin broader at the base (6.4% TL vs. 3.5%-4.9% TL); lower second dorsal fin in S. suckleyi (3.0% TL vs. 3.6%-5.0% TL). A larger number of comparative representatives are required to re-evaluate these variations between populations of S. suckleyi and should include additional characteristics, such as dentition and vertebral counts.

Discussion
Despite the fact that the taxonomy of Japanese dogfish sharks has been previously well discussed (e.g. Chen et al. 1979;Shirai 1989) and the species accounts constantly updated (e.g. Shinohara et al. 2014), the identification and validation of certain species were still unclear until the present. This study provides a revisited evaluation of Squalus species occurring in Japan and adjacent areas, following recent trends in the Systematics of the group with description of a novel species.
Characteristics of the external morphology, morphometric and meristic data support the separation and identification of the Japanese species which is congruent with previously similar work on the genus (e.g. Last et al. 2007f;Viana and Carvalho 2018b). Squalus shiraii sp. nov. bears many characteristics that distinctly separates it from its regional congeners as seen in the differential diagnosis provided. These characteristics are mostly related to external measurements of the body (e.g. height of dorsal fins), body colouration (e.g. dorsal and ventral caudal tips broadly white; body markedly dark grey in colour) and vertebral counts (e.g. precaudal and total vertebrae). Our results reveal that Squalus shiraii is morphologically similar to species of the "highfin megalops group" (S. albifrons, S. altipinnis, S. formosus and S. notocaudatus) in having dorsal fins conspicuously tall and upright and dorsal-fin spines robust and elongate. S. shiraii has postventral caudal margins not uniformly white with conspicuous black caudal bar, dermal denticles unicuspidate and lanceolate and snout markedly elongate and pointed at tip which are distinct from these species (vs. postventral caudal margins uniformly white without black caudal bar; denticles tricuspidate and rhomdoid, see Fig. 7C, D; snout small and rounded at tip). It lacks a white upper ocular margin and white bar at anterior margin of first dorsal fin that is diagnostic for S. albifrons and S. formosus, respectively. The latter species has been previously misidentified with S. blainvillei in Japan but White and Iglésias (2011) noticed that the characteristics in Chen et al. (1979) overlap those of the Taiwanese species, especially regarding having body greyish-brown dorsally and pale ventrally, first dorsal fin with anterior margin conspicuously white and vertebral counts. With the exception to vertebrae, Squalus shiraii has a variety of characteristics that separates it from S. formosus besides those listed here, including larger snout with nostrils closer to mouth and snout tip, smaller and lower dorsal fins and smaller pelvic fins as noticed in the diagnosis and remarks above. The distribution of S. formosus in Japanese waters is still uncertain as no specimen data were found in the ichthyological collections visited and, thus, this species was not included in the present revision.
Squalus shiraii has also been confused with S. mitsukurii in the area in previous taxonomic accounts because they share body robust, elongate snout and dark caudal bar and/or black upper caudal blotch, evident in at least one stage of maturity. The current analysis shows that they are non-conspecific by having clear morphological separation, based on vertebral counts (precaudal and total vertebrae), body colouration and morphometrics. Squalus mitsukurii is supported here as a regional endemic species to the North-western Pacific Ocean, which is in disagreement with Schroeder (1948, 1957), Bass et al. (1976), Compagno (1984) and Compagno et al. (2005a) and which was previously known as cosmopolitan. This result is congruent with recent taxonomic investigations on this species of Viana et al. (2017b), Veríssimo et al. (2017), Pfleger et al. (2018) and Daly-Engel et al. (2018) who refuted the occurrence of S. mitsukurii in the Western Indian, Western and Eastern Atlantic Oceans. The present analysis also rejects the occurrence of S. mitsukurii in Hawaiian waters given the validity of S. hawaiiensis, as supported in Daly-Engel et al. (2018) and in the diagnosis and remarks above. The recognition of S. hawaiiensis and S. boretzi in the North Pacific Ocean, as well as S. shiraii sp. nov. that exhibit closely sympatric distribution with S. mitsukurii, represent the still ongoing taxonomic confusion behind this species even in the surrounding topotypic region. Single or combined taxonomic evidence (e.g. Table 6. Meristic data for species of Squalus from Japan. H: holotype; P: paratypes; NT: neotype; N: number of specimens; x; mean; m: mode; SD: standard deviation.  (2019) and Nión and Meneses (2016) require taxonomic clarification. It is more likely that it has been confused either with a variety of the latest Squalus species that were described or resurrected in the past 10 years or may represent undescribed species with a possible degree of regionalism/endemism. Another nominal species, Squalus acutirostris, was also misidentified with S. mitsukurii, but it is regarded here as a valid species with occurrences in China, Japan, Taiwan and Vietnam. Squalus acutirostris seems to be rarely caught in Japanese waters once the availability of specimen data in the local ichthyological collections that were visited is extremely low. Due to limited availability of specimen data for this species, a tentative diagnosis was provided.
A detailed re-description of this nominal species is urgently required, based on new material from the areas in which this species has been reported in order to improve its morphological characterisation and better comprehend the separation with its regional congeners. Investigations on other aspects of S. acutirostris (e.g. vertebral counts, dentition, dermal denticles) is crucial. Squalus japonicus is endemic to the Indo-West Pacific Ocean from Japan and Taiwan. Its occurrence in the Eastern Pacific Ocean, off waters from Australia, New Zealand, New Caledonia and Vanuatu are not observed in the present study. These findings are in congruence with Compagno (1984Compagno ( , 2000, Compagno and Niem (1998), Compagno et al. (2005a), Last et al. (2007b), Yamashita et al. (2012), Motomura et al. (2015) and Viana and Carvalho (2018a). Squalus japonicus is very similar in morphology to S. nasutus and S. melanurus for sharing a body conspicuously thin from head to tail, head elongate, snout markedly elongate and obtuse, mouth narrow and pectoral fins tapered. These three species exhibit the greatest values of some head measurements (e.g. prebranchial length, preorbital length, preoral length, prenarial length) when compared to all other Squalus congeners, including S. shiraii.
Out of the six species examined in the present study, S. brevirostris appears to be the one with highest degree of taxonomic confusion. Difficulties in adequately identifying this species were widely reported (e.g. Garrick 1960;Chen et al. 1979;Compagno 1984;Last et al. 2007a;Straube et al. 2013) and were most likely due to intraspecific variations that were noticed here in the shape of denticles and pectoral fins, as well as caudal colouration. These variations, however, are not uniform and no apparent morphological pattern has been detected throughout the overall distribution of this species, even when maturity and sex are considered. Squalus brevirostris was thought to extend its occurrence to Australian waters in Chen et al. (1979), but it was refuted in Last et al. (2007) which is corroborated here. This species differs from S. megalops by having the first dorsal fin raked with posterior margin concave near the apex, body brownish-grey, postventral caudal margins broadly white, while S. megalops shows first dorsal fin not raked with posterior margin straight, body grey and postventral caudal margins narrowly white. It can be further separated from the Australian species and other closely-related species, including S. acutipinnis, S. crassispinus, S. albifrons, S. notocaudatus, S. raoulensis, S. hemipinnis and S. bucephalus by having pectoral-fin free rear tips markedly pointed (vs. pectoral-fin free rear tips rounded). Squalus brevirostris has small pectoral fins, never transcending the trunk height, while in S. raoulensis, the pectoral fins are conspicuously broad, always transcending the trunk height when adpressed on body. Dorsal fins are low and oblique in S. brevirostris while S. albifrons, S. crassispinus, S. formosus, S. notocaudatus, S. raoulensis, S. hemipinnis and S. bucephalus exhibit conspicuously high and upright dorsal fins. S. brevirostris lacks whitish eyebrow, a diagnostic characteristic of S. albifrons and it has very thin dorsal-fin spines (vs. heavy dorsal-fin spines in S. crassispinus). These findings support the fact that the distribution of S. brevirostris is restricted to the North and Central Western Pacific Oceans and is in congruence with Duffy and Last (2007a), Yamashita et al. (2012), Naylor et al. (2012), Straube et al. (2013) and Shinohara et al. (2014).
Species delimitation through molecular data is still unavailable for Squalus from Japan and surrounding countries as genetic information on the species through topotypic voucher specimens are still uncommon, except for S. brevirostris and S. suckleyi (e.g. Ward et al. 2007;Naylor et al. 2012;Straube et al. 2013;Veríssimo et al. 2017) and S. mitsukurii and S. japonicus (e.g. Pfleger et al. 2018;Daly-Engel et al. 2018) that have been barcoded in recent years. Thus, molecular analysis of mitochondrial genes for S. shiraii and S. acutirostris are lacking. Surprisingly, studies that have included molecular data of Japanese species did not suggest occurrence of possible undescribed or unidentified species in the region, indicating that insufficient data-collecting effort throughout the species distribution possibly do not cover the local diversity. The recognition of S. shiraii as a novel species and the present taxonomic revision represent the latest comprehensive evaluation of Japanese Squalus species to date and highlights that further efforts in elucidating the complexity behind the genus in other countries in Asia (e.g. South Korea, Philippines, Vietnam, China) are still required. Aspects of the biology, distribution, ecology and population size of Squalus species from Japan are still poorly acknowledged with only two species, S. mitsukurii and S. suckleyi, usually targeted and more rarely S. japonicus (e.g. Chen et al. 1981;Orlov et al. 2011Orlov et al. , 2012Cotton et al. 2011;Yano et al. 2017). Understanding the related threats and population trends are paramount as these species are historically under fishing pressures in Japanese waters.

Conclusions
Six species of dogfish sharks occur in Japan according to the present study: S. mitsukurii, S. japonicus, S. brevirostris, S. acutirostris, S. suckleyi and S. shiraii sp. nov (Fig.  8). Occurrence of S. formosus is still doubtful as no specimen caught from the country was found in the ichthyological collections visited. One or more of these species are sympatric in the area, as well as in waters surrounding China, South Korea, Taiwan, Philippines, Vietnam and Russia as seen in the above results. The previous hypothesis on the occurrence of S. megalops and S. blainvillei in Japan is thus rejected herein. Morphological, meristics and morphometric data support the separation between the species included in the analysis and it is congruent with the latest studies on the Squalus taxonomy in other regions of the world. The present analysis furthermore points out that two species, S. acutirostris and S. shiraii, were not previously recognised in Japan in neither morphological and/or combined molecular taxonomic accounts which warrants the need of improvements in the availability of specimen data.