Hynobius tsuensis Abe, 1922

Niwa, Keita, Nishikawa, Kanto, Matsui, Masafumi, Kanamori, Sally & Kuro-O, Masaki, 2023, Taxonomic reassessment of salamanders (genus Hynobius) from Tsushima Islands Japan, with a resurrection of Hynobius tagoi Dunn, 1923 (Amphibia: Caudata), Zootaxa 5339 (3), pp. 201-236 : 222-226

Hynobius tsuensis Abe, 1922

(Japanese name: Tsushima-sanshou-uwo)

(English name: Tsushima salamander)

( Figs. 6 , 9A , 13A , 14 , 17A–C , 18 )

Hynobius nebulosus (part): Tago 1907, 233

Hynobius nebulosus tsuensis: Mori 1928 , 48

Hynobius tsuensis: Abe 1922 , 331

Hynobius (Hynobius) tsuensis: Nakamura & Uéno 1963 , 7

Hynobius tsuensis (part, Group A): Niwa et al. 2021, 260, Fig. 3A

Hynobius bicolor: Dunn 1923 , 28

Syntypes: two males and a female from Tsushima collected by So Takahashi in April 1909. All the syntypes were stored at Hiroshima Higher Normal School (current Hiroshima University), but are considered to have been lost because many of the collections stored at Hiroshima Higher Normal School were lost by dropping an atomic bomb on 6 August 1945 during the Second World War. Sato (1937) treated No. 1016a, an adult male, collected by So Takahashi in April 1909 (used to be stored in Hiroshima Higher Normal School ) as the type and he described that the type locality was Izuhara , Tsushima .

Neotype: KUHE 58656 , an adult male from Shimobaru (34°13' N latitude, 129°15' E longitude, alt. 183 m above sea level), Izuhara, Tsushima City ( Shimojima of Tsushima Islands ), Nagasaki Prefecture, collected by Keita Niwa on 5 February 2017. GenBank accession No. is LC760667 (cyt b).

Specimens referred in the present paper: All specimens from Tsushima Islands, Nagasaki Prefecture: KUHE 60108 , 60109 , and 60114 (three females) from Tsutsu, Izuhara ; KUHE 58656 , 58861 , 58862 , 60123–60127 (eight males), and 58863 (one female) from Shimobaru, Izuhara ; KUHE 58873 and 58874 (two males) from Sumo, Mitsushima ; KUHE 58699 , 58700 , 58869 , 58870 , 58878 , 58879 , 59019 , 60151–60153 , 60159–60161 , 60164 , 60165 (15 males), 58982, 58985, 60146–60150, and 60162–60163 (nine females) from Kin, Kamitsushima. All specimens were collected by K. Niwa in 2017–2018 .

Etymology: We consider that the specific name “ tsuensis ” probably refers to the locality name “Tsu-shima” (Tsushima Islands), where this species occurs.

Diagnosis: A large species (adult SVL 55.8–70.9 mm in males and 59.2–69.5 mm in females), breeding in lotic water; dorsal color was variable: uniformly blackish or purplish brown without markings, blackish or purplish brown with yellowish markings or spots, yellow with blackish or purplish brown markings or spots, dark brown with dark or blackish markings or spots; fore- and hindlimbs medium-sized; tips of fore- and hindlimbs adpressed on body never meeting (overlap of -3.5 to -1.0 costal folds in males and -4.0 to -2.0 in females); fifth toe long; tail long, low, and narrow; usually a distinct yellow stripe present on the upper side of tail in both sexes; lateral color of the posterior half of tail uniformly blackish or purplish brown; ova large; clutch size small; egg-sacs short and C-shaped (crescent) with distinct whiptail structure on the free end. Phylogenetically, H. tsuensis is distinct from other congeners such as H. tagoi , H. nebulosus , H. bakan sensu lato, and H. dunni .

Description of the neotype (measurements in mm): Head-body large (SVL 67.2) and robust; head oval and moderately depressed, distinctly longer (HL 16.5, 24.6%SVL) than wide (HW 11.9, 17.7%SVL); snout rounded, slightly projecting beyond lower jaw; nostril close to snout tip; labial fold absent; eye moderately large, protruded, slightly inset from edge of head in dorsal view; upper eyelid well developed (UEW 1.9, 2.8%SVL), shorter (UEL 3.9, 5.8%SVL) than snout (SL 4.0, 6.0%SVL); gular fold distinct, curving anteriorly; parotoid gland evident, extending from angle of jaw to gular fold; postorbital grooves distinct, branching posterior to angle of jaw, one short and running down to lower jaw, the other long and posterior to parotoid gland; vomerine teeth series V-shaped ( Fig. 9A ), slightly wider (VTW 4.5, 6.7%SVL) than long (VTL 4.4, 6.5%SVL), anterior margin on line connecting choanae; tongue broad, both sides free from mouth floor; fore- and hindlimbs short for the species and thick (FLL 14.1, 21.0%SVL; HLL 18.4, 27.4%SVL); CG 13; depressed limbs separated by two costal folds (LO -2.0); relative length of fingers I<IV<III<II, toes I<V<II<IV<III; fifth toe well developed (5TL 2.0, 3.0%SVL); cloaca swollen with longitudinal slit; genital tubercle on anterior cloaca absent; tail long (TAL 58.3, 86.8%SVL), cylindrical at base (BTAW 6.7, 10.0%SVL; BTAH 5.7, 8.5%SVL), gradually compressed posteriorly (MTAW 4.4, 6.5%SVL; MTAH 7.6, 11.3%SVL); tip of tail weakly pointed in lateral view ( Fig. 6E ).

Additional Measurements and counts of the neotype: MXHW (12.5, 18.6%SVL); LJL (8.9, 13.2%SVL); IND (4.1, 6.1%SVL); IOD (3.9,5.8%SVL); AGD (35.4, 52.7%SVL); TRL (50.7, 75.4%SVL); MXTAH (7.8, 11.6%SVL); 2FL (2.1, 3.1%SVL); 3FL (2.0, 3.0%SVL); 3TL (4.4, 6.5%SVL); UJTN (72); LJTN (74); VTN (56).

Color: In life, dorsum purplish brown with yellowish markings ( Fig. 6A ). Underside of body lighter than dorsum without markings ( Fig. 6B ). Throat covered with white nuptial color in males ( Fig. 6B ). Tail side of the posterior half uniformly purplish brown ( Fig. 6E ). Middorsal line of tail bright yellow ( Fig. 6A ). Midventral line of tail slightly yellow ( Fig. 6B ).

Variation: Morphometric data and body coloration data are summarized in Tables 3 and 5 , respectively. A significant sexual difference between males (n=25) and females (n=13) was not detected in body size (SVL) (P> 0.05), however, for ratio, males have significantly larger values (P <0.05) than females in head length (HL, median=24.4%SVL in males vs. 23.3%SVL in females), head width (HW, 17.1%SVL vs. 16.6%SVL), maximum head width (MXHW, 17.8%SVL vs. 16.9%SVL), snout length (SL, 6.9%SVL vs. 6.6%SVL), internarial distance (IND, 6.2%SVL vs. 6.0%SVL), interorbital distance (IOD, 6.2%SVL vs. 5.8%SVL), upper eyelid length (UEL, 5.4%SVL vs. 5.0%SVL), tail length (TAL, 81.8%SVL vs. 76.7%SVL), basal tail width (BTAW, 10.9%SVL vs. 9.7%SVL), medial tail width (MTAW, 5.8%SVL vs. 4.7%SVL), basal tail height (BTAH, 8.1%SVL vs. 7.4%SVL), medial tail height (MTAH, 8.8%SVL vs. 7.5%SVL), maximum tail height (MXTAH, 9.4%SVL vs. 7.8%SVL), forelimb length (FLL, 22.6%SVL vs. 21.5%SVL), hindlimb length (HLL, 28.5%SVL vs. 27.0%SVL), MTAW/ MTAH (64.8% vs. 56.5%), and limb overlap (LO, -2.0 vs. -3.0). Whereas, males have significantly smaller values (P <0.05) than females in axilla-groin distance (AGD, median=51.4%SVL in males vs. 52.9%SVL in females) and trunk length (TRL, 75.6%SVL vs. 76.7%SVL).

For body coloration in life, dorsum variable: uniformly blackish or purplish brown without markings ( Fig. 14B ), blackish or purplish brown with yellowish markings or spots ( Fig. 6A ), yellow with blackish or purplish brown markings or spots ( Figs. 6C , 14A ), and dark brown with dark or blackish markings or spots ( Fig. 14C ). Underside of body with ( Fig. 6D ) or without ( Fig. 6B ) spots. Tail side of the posterior half uniformly blackish or purplish brown ( Figs. 6E, F , 14 ). Middorsal line of tail bright yellow ( Figs. 6A, C , 14A, C ) although sometimes indistinct ( Fig. 14B ). Midventral line of tail yellow ( Fig. 6D ) or slightly lighter ( Fig. 6B ).

The modal dorsal color pattern was blackish or purplish brown with yellowish markings or spots (44.0% of males; 53.8% of females), but the second modal color was different between sexes: dark brown with dark or blackish markings or spots in males (32.0%); yellow with blackish or purplish brown markings or spots in females (38.5%).

Egg-sacs and eggs: Morphometric values of egg-sacs, clutch size, and unfertilized egg rate are shown in Table 4 . Egg-sacs of H. tsuensis were crescent in shape with thin envelope without any wrinkles or notable striations ( Figs. 13A , 18B ). All egg-sacs (n=13) used in the present study had a distinct whiptail structure on the free end, similar to some lotic-breeding salamanders (e.g., H. oni and H. hirosei ). The whiptail structure of H. tsuensis ranged 9.5–32 (median=17, n=13) mm in length, which was significantly longer than H. nebulosus but not significantly different from H. tagoi ( Fig. 12 ; Table 4 ). Egg-sacs of H. tsuensis were significantly shorter than those of H. tagoi and H. nebulosus ( Fig. 11 ). Clutch size of H. tsuensis ranged from 34–74 (mean±SD=52.8±10.8, n=8), which was smaller than those of relatives such as H. nebulosus (population from Fukue Island: range=111–266, mean±SD=175.9±46.9, n=9 [ Fig. 11 ; Table 4 ]; Nagasaki: 77–160, 111.7±23.9, n=12 [ Fig. 11 ; Table 4 ], 89–117, 101.7±14.2, n=3 [ Sato 1943]; all Kyushu: 41–333, 135.8±59.0, n=91 [ Matsui et al. 2019]), H. dunni (50–200, 95.1±27.5, n=55 [ Sato 1943]), and H. bakan sensu lato (56–333, 116.5±58.2, n=32 [ Matsui et al. 2019]), but not significantly differed from H. tagoi (21–62, 45.5±11.4, n=23 [ Fig. 11 ; Table 4 ]). Egg-sacs of H. tsuensis contain many unfertilized eggs (unfertilized egg rate: 63.5%, 37.5–80.8%, n=14) in nature. Ova from two females ranged from 3.3–3.5 (mean±SD=3.4±0.07, n=10) and 3.0–3.3 (mean±SD=3.2±0.11, n=10) mm in diameter. Both the animal and vegetal poles of egg were whitish cream in color.

Larvae: Views of larvae are shown in Figs. 17A–C and 18C . Larvae after hatching temporarily have a pair of balancers at the side of head. Larvae (n=5) at Stage 63 of Iwasawa and Yamashita (1991) of the first year collected from Kin, Kamitsushima on 23 June 2017 ranged from 17.7–19.8 (mean±SD=18.4±0.7) in SVL and 30.7–35.7 (32.5±1.7) mm in total length, head rounded in dorsal and lateral views ( Fig. 17A, B ); snout short and broadly rounded; eyes slightly protruded, inset from edge of head in dorsal view; labial fold distinct at half of upper jaw; external gills developed; caudal fin higher than head; dorsal fin higher than ventral fin; origin of dorsal fin in the middle portion of trunk; origin of ventral fin in vent; tail tip weakly pointed; limbs slender; claws on fingers and toes absent. In life, dorsum light brown without remarkable spots ( Fig. 17A ); venter whitish and transparent without spots ( Fig. 17C ).

The larvae of H. tsuensis tended to have smaller body size than those of H. tagoi ( Fig. 17D–F ) (17.6–28.7 [mean±SD=25.1±3.5] mm in SVL and 31.8–54.0 [47.5±7.3] mm in total length, n=7, Stage 63) collected from the same locality (stream) at the same date. In body coloration, H. tsuensis tended to have several blackish spots (markings) on lateral side of tail ( Fig. 17B ), unlike larvae of H. tagoi without any distinct blackish markings on lateral side of tail ( Fig. 17E ).

Range: Tsushima Islands (excluding the southern part of Kamijima, former Toyotama-machi), in Nagasaki Prefecture, Western Japan ( Fig. 1 ). The known localities range from 25 m to 270 m (median 110 m, n=37) above sea level.

Comparisons: Hynobius tsuensis is distinct from H. tagoi by the lateral color of the posterior half of tail: uniformly blackish or purplish brown without markings in 96.0% of male and in 69.2% of female H. tsuensis ; brown (not blackish or purplish color) with dark stipples in 96.3% of male and in 76.9% of female H. tagoi . Also, H. tsuensis is distinguishable from H. tagoi by having distinct yellow stripe(s) on the tail, having blackish (purplish) brown dorsum with yellowish markings (spots) or yellowish dorsum with blackish (purplish) brown markings (spots) or dark brownish dorsum with dark (blackish) markings (spots) (vs. lacking distinct yellow stripe(s) on the tail; having brownish dorsum with dark stipples in H. tagoi ). Hynobius tsuensis morphologically differs from H. tagoi by relatively wider and deeper vomerine teeth series (RVTW: males: 5.2%SVL vs. 5.0%SVL in H. tagoi ; females: 5.2%SVL vs. 4.8%SVL in H. tagoi ; RVTL: males: 5.5%SVL vs. 5.1%SVL in H. tagoi ; females: 5.4%SVL vs. 5.0%SVL in H. tagoi ), and smaller degree of limb overlap (LO: males: -2.0 vs. - 1.5 in H. tagoi ; females: -3.0 vs. - 2.5 in H. tagoi ). Clutch size and morphology of egg-sacs of H. tsuensis are similar to H. tagoi , but H. tsuensis is distinct from H. tagoi by having higher unfertilized egg rate (63.5% vs. 2.0% in H. tagoi ). Also, egg-sacs of H. tsuensis are different from those of H. tagoi by smaller egg-sac (MAESL: mean±SD=110.6± 10.4 mm vs. 139.2± 16.4 mm in H. tagoi ; MIESL: 43.0± 6.9 mm vs. 64.4± 8.5 mm in H. tagoi ; ESW: 12.6± 1.2 mm vs. 13.9± 1.2 mm in H. tagoi ), MIESL/MAESL (median=37.8% vs. 46.0% in H. tagoi ), ESW/ESH (103.7% vs. 111.1% in H. tagoi ).

Hynobius tsuensis is morphologically distinguished from H. nebulosus by larger body size (males: mean±SD=64.0± 3.7 mm vs. 61.4± 4.5 mm in H. nebulosus ; females: 64.0± 3.1 mm vs. 59.3± 4.7 mm in H. nebulosus ), relatively longer snout (SL: males: 6.9%SVL vs. 6.4%SVL in H. nebulosus ; females: 6.6%SVL vs. 6.0%SVL in H. nebulosus ), relatively wider internarial (IND: males: 6.2%SVL vs. 5.8%SVL in H. nebulosus ; females: 6.0%SVL vs. 5.3%SVL in H. nebulosus ) and upper eyelid (UEW: males: 3.2%SVL vs. 2.8%SVL in H. nebulosus ; females: 3.1%SVL vs. 2.7%SVL in H. nebulosus ), relatively shorter axilla-groin distance (AGD: males: 51.4%SVL vs. 52.4%SVL in H. nebulosus ; females: 52.9%SVL vs. 54.3%SVL in H. nebulosus ), relatively longer (TAL: males: 81.8%SVL vs. 77.2%SVL in H. nebulosus ; females: 76.7%SVL vs. 70.7%SVL in H. nebulosus ), narrower (BTAW: males: 10.9%SVL vs. 11.5%SVL in H. nebulosus ; females: 9.7%SVL vs. 10.7%SVL in H. nebulosus: MTAW : males: 5.8%SVL vs. 6.6%SVL in H. nebulosus ; females: 4.7%SVL vs. 6.2%SVL in H. nebulosus ), lower tail (BTAH: males: 8.1%SVL vs. 9.8%SVL in H. nebulosus ; females: 7.4%SVL vs. 8.9%SVL in H. nebulosus: MTAH : males: 8.8%SVL vs. 11.2%SVL in H. nebulosus ; females: 7.5%SVL vs. 10.2%SVL in H. nebulosus: MXTAH : males: 9.4%SVL vs. 11.7%SVL in H. nebulosus ; females: 7.8%SVL vs. 10.5%SVL in H. nebulosus ). Also, H. tsuensis is different from H. nebulosus by having blackish (purplish) brown dorsum with yellowish markings (spots) or yellowish dorsum with blackish (purplish) brown markings (spots) or dark brownish dorsum with dark (blackish) markings (spots) (vs. dark olive dorsum scattered with darker spots in H. nebulosus [ Matsui et al. 2019]).

Hynobius tsuensis is distinguished from H. dunni by smaller degree of limb overlap (LO: males: -2.0 vs. +2.0 in H. dunni ; females: -3.0 vs. +2.0 in H. dunni [ Sato 1943]), smaller clutch size (CS: mean±SD= 52.8±10.8 vs. 95.1± 27.5 in H. dunni [ Sato 1943]), and their dorsum colorations (blackish [purplish] brown dorsum with yellowish markings [spots] or yellowish dorsum with blackish [purplish] brown markings [spots] or dark brownish dorsum with dark [blackish] markings [spots] vs. yellowish brown dorsum scattered with black plastids in H. dunni [ Sato 1943]).

Hynobius tsuensis is different from H. bakan sensu lato (containing H. nagatoensis and H. nihoensis ) by larger body size (SVL: mean±SD=64.0± 3.7 mm vs. 51.8± 5.7 mm in H. bakan sensu lato [ Matsui et al. 2019]), relatively longer tail (RTAL: median 81.8%SVL vs. 74.4%SVL in H. bakan sensu lato [ Matsui et al. 2019]), relatively lower tail (RMXTAH: 9.4%SVL vs. 12.4%SVL in H. bakan sensu lato [ Matsui et al. 2019]), smaller clutch size (CS: 52.8±10.8 vs. 116.5± 58.2 in H. bakan sensu lato [ Matsui et al. 2019]), and their dorsum colorations (blackish [purplish] brown dorsum with yellowish markings [spots] or yellowish dorsum with blackish [purplish] brown markings [spots] or dark brownish dorsum with dark [blackish] markings [spots] vs. dark brown dorsum scattered with black markings in H. bakan sensu lato [ Matsui et al. 2019]).

Natural history: Hynobius tsuensis breeds in lotic water such as mountain streams ( Fig. 18D ). Sometimes, H. tsuensis breeds syntopically with H. tagoi ( Niwa et al. 2021) . Breeding occurs from late March to mid-April, which is later than that of H. tagoi (early February to early March). A female of H. tsuensis oviposits a pair of egg-sacs to stones or debris ( Ito et al. 2023) under the water in streams ( Fig. 18B ). Hatching occurs in May and larvae metamorphose in autumn of the first year, but some of the larvae once overwintered and metamorphose in April to June of the next year.

Conservation: Hynobius tsuensis has been listed as near threatened (NT) on the Red List of Ministry of the Environment Government of Japan, 2020 (Ministry of the Environment Government of Japan 2020).Also, it is listed as vulnerable (VU) on the Prefectural Red List of Nagasaki, 2022 ( Nagasaki Prefecture 2022). The distribution range of the species is small, and divided into northern (Kamijima) and southern (Shimojima) areas of Tsushima Islands ( Fig. 1 ). Thus, the northern population should be treated as a different conservation unit from the southern one.

Figures

FIGURE 1. Map of Kyushu and its adjacent islands, Japan, showing the sampling localities of H. tsuensis (triangles), Hynobius sp. (inverse triangles), and H. nebulosus (circles) used for the morphological and microsatellite (SSR) analyses (open: morphological analysis, filled: both morphological and SSR analyses). Numbers correspond to pop no. in Table 2. The colored ranges indicate the distributional range of H. tsuensis (red, solid) and Hynobius sp. (blue, horizontal lines) in Tsushima Islands, and range of H. nebulosus (yellow, solid) in Kyushu. A dashed line in Tsushima Islands indicates the boundary between the northern (Kamijima) and southern islands (Shimojima), and a star in the Shimojima shows the center of Izuhara which is considered to be the type locality of H. tsuensis. A diamond in Kyushu indicates the type locality of H. nebulosus.

FIGURE 3. Phylogenetic tree for 34 Hynobius species and out groups based on the complete cyt b gene sequences (1141 bp) of mtDNA by the Bayesian inference method. The numbers at nodes indicate Bayesian posterior probability (BPP). Asterisks (*) indicate significant supports with BPP of 0.95 or higher.

FIGURE 6. Dorsal and ventral views, and left side of the tail of a male neotype (A, B, E; KUHE 58656 from Izuhara) and a female specimen (C, D, F; KUHE 58985 from Kamitsushima) of H. tsuensis. A scale bar shows 10 mm.

FIGURE 9. Vomerine teeth series of a male neotype of H. tsuensis (A: KUHE 58656) from Izuhara, male topotypic specimens of H. tagoi (Hynobius sp.) (B: KUHE 58983) from Kamitsushima, and H. nebulosus (C: KUHE 58662) from Nagasaki City. Not to scale.

FIGURE 11. Boxplots showing maximum egg-sac length (MAESL), MAESL/ESW (egg-sac width), clutch size (CS), and unfertilized egg rate of four groups.

FIGURE 12. Histograms showing length of the whiptail structure (WTL) of egg-sacs of H. tsuensis (A), Hynobius sp. (H. tagoi) (B), H. nebulosus from Nagasaki (C) and Fukue Island (D).

FIGURE 13. A pair of egg sacs of H. tsuensis (A: KUHE 58986) from Kamitsushima, H. tagoi (Hynobius sp.) (B: KUHE 58669) from Kamitsushima, and H. nebulosus (C: KUHE 61872) from Nagasaki City. A white bar indicates 30 mm.

FIGURE 14. Dorsal views and left side of the tail of male H. tsuensis showing color variation. A yellowish dorsum with blackish spots and a bright yellowish stripe on the tail (A: KUHE 58873 from Mitsushima); a blackish dorsum without clear spots or a bright yellowish stripe on the tail (B: KUHE 58869 from Kamitsushima); a dark brownish dorsum with rough blackish spots and a bright yellowish stripe on the tail (C: KUHE 60161 from Kamitsushima). A white bar shows 10 mm.

FIGURE 17. Dorsal, lateral, and ventral views of a larva of H. tsuensis (A–C: KUHE 65029) from Izuhara and H. tagoi (D–F; KUHE 65028) from Toyotama. A white bar shows 10 mm.

FIGURE 18. An alive female (A: KUHE 60148), egg sacs (B: not captured), a larva (C: not captured), and breeding habitat (D) of H. tsuensis.