Gekko canhi, Rösler, Herbert, Nguyen, Truong Quang, Doan, Kien Van, Ho, Cuc Thu, Nguyen, Tao Thien & Ziegler, Thomas, 2010

Rösler, Herbert, Nguyen, Truong Quang, Doan, Kien Van, Ho, Cuc Thu, Nguyen, Tao Thien & Ziegler, Thomas, 2010, A new species of the genus Gekko Laurenti (Squamata: Sauria: Gekkonidae) from Vietnam with remarks on G. japonicus (Schlegel), Zootaxa 2329, pp. 56-68 : 57-64

Gekko canhi sp. n.

Holotype: IEBR A.0910, adult male from Huu Lien (21o40’N, 106o20’E), Huu Lung, Lang Son Province, North Vietnam, collected in April 2004 by Doan Van Kien ( Figs. 1 A, 2, 3) .

Paratypes: ZFMK 88879, adult female from Sa Pa, Lao Cai Province, collected in between 26 June to 8 July 2001 by Ho Thu Cuc ( Fig. 1 B) ; VNMN 1001 –1002, adult females from Huu Lien, Lang Son Province, collected in June 2009 by Nguyen Thien Tao ( Fig. 4 A) .

Diagnosis. Gekko canhi sp. n. can be distinguished from all congeners on the base of the following combination of characters: A medium-sized species of the genus Gekko with <100 mm SVL; tail round, not thickened at base; supralabials 12–14; infralabials 10–13; nares in contact with rostral, internasal single, nasals 3; interorbitals 47–50; dorsal tubercle rows 10–13; scales between mental and cloacal slit 205–229; midbody scales 164–170; subdigital lamellae below first toe 13–16, below fourth toe 14–17; extensive webbing between fingers and toes lacking; shanks with tubercles; precloacal pores 5; postcloacal tubercles 2– 3; tubercles on tail present; subcaudals enlarged.

Description of holotype. Rostral rectangular, wider than high (maximum rostral width 4.2 mm, maximum rostral height 1.4 mm, RW/RH 3.0) and wider than mental, short rostral suture medially; supralabials 14/14; temporal region with several tubercles above snout; nares in contact with rostral and first supralabial; nasals 3/3; nasorostrals more or less square, double the size of supranasals; postnasals larger than nasorostrals; internasal rectangular, half as large as nasorostrals; snout medially with flat, elongate cavity; lateral snout scales oval, somewhat convex, juxtaposed, twice the size of those in midrostral region; 17–18 enlarged scales between postnasals and orbital height; 61 scales between seventh supralabials; medial snout scales oval, convex, juxtaposed; pupil vertically, with two small anterior tines and two large posterior tines; dorsal ciliary scales 2.5 times as large as medial snout scales, 5/5 spinous tubercles posteriorly; ear opening oblique/vertical, oval, about half as large as eye diameter, with skin fold above; interorbitals 49, granular, in orbital region twice the size in medial region; nuchal region scales granular, as large as medial interorbitals; tubercles in temporal and nuchal regions roundish, homogeneous, conical; mental triangular, wider than long (maximum mental width 3.4 mm, maximum mental length 2.2 mm, MW/ML 1.6), not distinctly larger than first infralabials; infralabials 12/13; postmentals 2, trapezoid, twice longer than wide and longer than length of mental, anteriorly in contact with mental and first infralabials; medial suture between postmentals longer than length of mental; gulars 7, bordering postmentals (of these, outer gulars larger than inner ones); outer postmentals twice as large than inner ones; gular scales as large as medial snout scales, round, flat, smooth, juxtaposed, in semi regular transversal rows; dorsals as large as medial snout scales round, convex, juxtaposed, in semi regular transversal rows; dorsal tubercles 2–3 times as large as adjoining dorsal scales, round to oval, convex, smooth, surrounded by 8–9 dorsal scales, in 11 semi regular longitudinal rows; lateral fold weakly developed, smooth, in width of 3 dorsal scales; ventrals between lateral folds 49; midbody scales in 170 rows; ventral scales between mental and cloacal slit 205; upper and lower arm scales slightly enlarged, smooth; forelimbs without tubercles; femoral scales anteriorly and ventrally smooth, dorsally and posteriorly granular; enlarged femorals lacking (i.e., no sharp transition between anteriorly larger and posteriorly smaller scales); tibial scales dorsally granular, ventrally flat, smooth; tibial tubercles conical, twice as large as surrounding granular scales; fingers and toes with narrow basal webbing; claws sheathed by three scales, with the dorsal scale smallest; subdigital lamellae 14/13 under first finger, 13/14 under fourth finger, 16/15 under first toe, and 15/14 under fourth toe; precloacal pores 5, in an angular series, medially separated by 5 smooth scales; enlarged scales posterior to precloacal pores in 11 rows; postcloacal tubercles 2/2, blunt, conical, ventrolkateral tubercles approximately half as the size of dorsolateral tubercles; original tail not thickened at base, with bands, and without tubercles; dorsal caudal scales as approximately twice the size of dorsal scales, squarish, flat, in regular transversal rows; third whorl in width of 11 dorsal scales; subcaudals flat, enlarged distal to second whorl; three subcaudals per whorl, with the last more distinctly enlarged; regenerated tail with distinctly broader subcaudals than beneath in orginal tail.

Color in preservative (70 % ethanol): Upper side of head, back and tail sepia (478U); snout until hind margin of eyes vermiculated, back of head with two twirl-like grey brown (481U) blotches; neck with grey brown (481U), wavy band; dorsum with narrow (1.5 mm wide), pale vertebral stripe, from neck to sacral region; dorsum traversed by 8 dark brown (4695U), strongly serrated bands, fading to blotches from by base of tail; limbs with variable light blotches and bands; tail with four broad dark violet brown (497U) and four thinner dark purple grey (4745U) rings; regenerated part paler, with very fine black brown (476U) blotches; underside of head, limbs and precloacal region buff (4745U), belly grey brown (401U).

Variation. The female paratypes differ from the male holotype by several characters summarized as follows (see also Tables 1 and 2 ): They differ from the holotype by possession of distinctly smaller postcloacal tubercles and by the lack of perforated precloacal scales; two paratypes have dorsal tail tubercles, positioned at the end of the whorls; these tubercles are conical in the first whorl, and flat in whorls 2–3/4; dorsal bands are in part distinctly reduced in the female paratypes and are built up accentuated by pairs of oval blotches, which are medially separated by the light vertebral stripe; one paratype has an original tail that has 7 blackish brown (Black 2U 2X) and 6 light grey (406U) transversal bands; regenerated tails of female paratypes are patternless.

IEBR A.0910 ZFMK 88879 VNMN 1001 VNMN 1002 Mean±SD (variation) Color in life. Color in life unrecorded by us.

Comparisons. Gekko canhi sp. n. differs from Vietnamese Gekko by the following characters (after Ngo et al. 2009; Ngo & Gamble 2009; Rösler et al. 2005): from G. badenii Szczerbak & Nekrasova (synonym G. ulikovskii Darevsky & Orlov ; see Nguyen et al. 2009; Nguyen et al. in press) by a lower maximum snout vent length (99.2 versus 108.0 mm), fewer precloacal pores (5 versus 10–15), and a greater number of interorbitals (47–50 versus 30–46), scales along underside of body from mental to the front of cloacal slit (164–170 versus 136–145), scales around the middle of the body (205–229 versus 128–137), and ventrals (46–51 versus 29– 35); from G. g e c k o (Linnaeus) by nares contacting with rostral (versus not in contact); from G. grossmanni Günther by having fewer precloacal pores (5 versus 8–14), and more interorbitals (47–50 versus 38–46), scales along underside of body from mental to the front of cloacal slit (164–170 versus 139–165), scales around the middle of the body (205–229 versus 116–135), and ventrals (46–51versus 26–31); from G.

palmatus Boulenger by having fewer precloacal pores (5 versus 24–32), more interorbitals (47–50 versus 33– 47), scales along underside of body from mental to the front of cloacal slit (164–170 versus 172–199) and scales around the middle of the body (205–229 versus 139–156), as well as by reduced webbing between fingers and toes of G. palmatus ; from G. russelltraini Ngo, Bauer, Wood & Grismer by having more interorbitals (47–50 versus 30–34), scales around the middle of the body (205–229 versus 90–107), ventrals (46–51versus 28–30), and fewer precloacal pores (5 versus 8–11); from G. scientiadventura Rösler, Ziegler, Vu, Herrmann & Böhme by the presence of 11–12 dorsal tubercle rows (versus tubercles absent).

Of the remaining Gekko species, Gekko canhi sp. n. differs as follows: from G. auriverrucosus Zhou & Liu by having snout vent length larger then 85 mm (versus <70 mm), by contact between the nares and rostral (vs. not in G. auriverrucosus ) and no cluster of large conical tubercles on upper edge of ear-opening (Zhou & Liu 1982); from G. albofasciolatus Günther ; G. s i a m e n s i s Grossmann & Ulber; G. smithii Gray ; G. verreauxi Tytler (after own data and Günther 1867, Ota et al. 1991); Gekko sp. “Tioman” (see Lim & Lim 1999, Hien et al. 2001, Grismer et al. 2002, Grismer et al. 2004, Grossmann & Tillack 2004, 2005, Koch et al. 2009) by having snout vent length less than 100 mm (versus> 150 mm) and additionally, except for G. verreauxi , which is endemic to the Andaman Islands, by contact between the nares and rostral (versus not in contact); from G. athymus Brown & Alcala ; G. melli Vogt ; G. subpalmatus Günther ; and G. t a w a e n s i s Okada by the presence (versus absence) of dorsal tubercles ( Brown & Alcala 1978; Okada 1956; Rösler et al. 2005; Rösler & Tiedemann 2007); from G. crombota Brown, Oliveros, Siler & Diesmos , G. ernstkelleri Rösler, Siler, Brown, Demegillo & Gaulke ; G. gigante Brown & Alcala ; G. kikuchii Oshima ; G. mindorensis Taylor ; G. monarchus (Schlegel) ; G. palawanensis Taylor ; and G. ro m b l o n Brown & Alcala by having fewer precloacal pores (<6 versus> 29) ( Brown et al. 2008, Brown & Alcala 1978; Oshima 1912; Rösler et al. 2006); from G. vittatus Houttuyn by absence (versus presence) tubercles on throat and lateral fold (own data); from G. hokouensis Pope ; G. liboensis Zhou & Li ; G. scabridus Liu & Zhou ; G. swinhonis Günther ; and G. taibaiensis Song by having more interorbitals (47–50 versus <41) and ventrals (46–51 versus <45) (Liu & Zhou 1982; Zhou & Li 1982; Zhao et al. 1999); from G. petricolus Taylor ; G. porosus Taylor ; G. shibatai Toda, Sengoku, Hikida & Ota ; G. similignum Smith ; and G. vertebralis Toda, Sengoku, Hikida & Ota by a greater number of midbody scales (205–229 versus <161) ( Brown & Alcala 1978; Ota et al. 1995; Taylor 1962; Toda et al. 2008); from G. yakuensis Matsui & Okada by a higher snout vent length (99.2 mm versus 72.0 mm), more lamellae below first toe (13–16 versus 10), and by relatively larger postmentals and hind leg tubercles ( Matsui & Okada, 1968); from G. japonicus (Schlegel) by a higher snout vent length (99.2 mm versus 74.0 mm), more interorbitals (47–50 versus 32–35), scales around the middle of the body (205–229 versus 130–144), ventrals (46–51 versus 39–44), lamellae below first toe (13–16 versus 10–12), and fewer precloacal pores (5 versus 6– 9), and absence of dorsal tail tubercles ( Goris & Maeda 2004; Ota et al. 1989, Zhao et al. 1999), for more detailed comparisons, see below; from G. wenxianensis Zhou & Wang by a higher snout vent length (99.2 mm versus 59.0 mm), more ventrals (47–50 versus 44), more subdigital lamellae under first toe (13–16 versus 6) and fourth toe (14–17 versus 9), separated precloacal pores (versus not separated) and smaller dorsal tubercles (3 times versus 4–6 times size of adjacent dorsals) ( Zhou & Wang 2008).

Etymology. We name this new species in honor of Professor Dr. Le Xuan Canh, director of the Institute of Ecology and Biological Resources in Hanoi, in recognition of his support and contribution towards biodiversity research in Vietnam. As common names we suggest Canh’s Gecko (English) , Tắc kè cảnh (Vietnamese), and Canhs Gecko (German) .

Distribution. The species is currently known only from northern Vietnam: Lang Son and Lao Cai provinces ( Fig. 5 ). Due to the proximity with the Chinese border, we suspect that the species additionally may occur in southern China as well.

Ecological notes. The type series were found in evergreen secondary forest at elevation at 200 m in Huu Lien ( Fig. 4 B) and 1500 m in Sa Pa.

Figures

FIGURE 1. Dorsal view of the male holotype of Gekko canhi sp. n. (IEBR A. 0910) from Lang Son Province (A) and the female paratype (ZFMK 88879) from Lao Cai Province (B). Photos T. Ziegler.

FIGURE 4. Dorsal view of the female paratypes (VNMN 1001 – 1002, upper and lower, respectively) (A – photo T. Ziegler) and habitat of Gekko canhi sp. n. in Huu Lien, Lang Son, Vietnam (B – photo T. T. Nguyen).

FIGURE 5. Known distribution of Gekko canhi sp. n. in Vietnam: 1) Huu Lien, Lang Son Province and 2) Sa Pa, Lao Cai Province.

Tables

TABLE 1. Measurements (in mm) and proportions of the type series of Gekko canhi sp. n. (for abbreviations see material and methods; * tail partially regenerated).

Sex male	female	female	female	—
SVL 91.6	99.2	85.8	92.3	92.2±5.5 (85.8–99.2)
TL 98.9*	106.7*	101.5	99.0*	—
AG 38.3	49.0	35.0	41.0	40.8±6.0 (35.0–49.0)
HL 25.5	25.7	23.2	25.2	24.9±1.2 (23.2–25.7)
HW 18.7	18.5	17.0	18.5	18.2±0.8 (17.0–18.7)
HH 10.1	11.5	9.9	10.9	10.6±0.7 (9.9–11.5)
SE 11.6	11.2	10.3	10.8	11.0±0.6 (10.3–11.6)
EE 8.8	9.0	8.0	8.8	8.7±0.4 (8.0–9.0)
RW 4.2	5.9	3.8	3.9	4.5±1.0 (3.8–5.9)
RH 1.4	1.7	1.8	1.8	1.7±0.2 (1.4–1.8)
MW 3.4	3.0	2.5	3.1	3.0±0.4 (2.5–3.4)
ML 2.2	1.8	1.7	2.2	2.0±0.2 (1.7–2.2)
SVL/TL —	—	0.9	—	—
SVL/AG 2.4	2.0	2.5	2.3	2.3±0.2 (2.0–2.5)
SVL/HL 3.6	3.9	3.7	3.7	3.7±0.1 (3.6–3.9)
HL/HW 1.4	1.4	1.4	1.4	1.4±0.0 (1.2–1.3)
HL/HH 2.5	2.2	2.3	2.3	2.4±0.1 (2.2–2.5)
SE/EE 1.3	1.2	1.3	1.2	1.3±0.0 (1.2–1.3)
RW/RH 3.0	3.6	1.8	2.2	2.7±0.7 (2.1–3.5)
MW/ML 1.6	1.7	1.5	1.4	1.5±0.1 (1.4–1.7)
RW/MW 1.3	1.9	1.5	1.3	2.3±0.6 (1.8–3.2)

TABLE 2. Scalation of the type series of Gekko canhi sp. n. (for abbreviations see material and methods). IEBR A. 0910 ZFMK 88879 VNMN 1001 VNMN 1002 Mean ± SD (variation)

SPL 14/14	14/14	12/13	13/12	13.3±0.9 (12–14)
SBL 13/12	10/12	12/11	10/11	11.4±0.9 (10–13)
N 3/3	3/3	3/3	3/3	3.0±0.0 (3)
IN 1	1	1	1	1.0±0.0 (1)
SC7SPL 61	—	57	56	58.0±1.1 (56–61)
IO 49	50	49	47	48.8±1.3 (47–50)
CS 5/5	7/7	8/9	6/6	6.6±1.4 (5–9)
PM 2	2	2	2	2.0±0.0 (2)
GP 7	5	5	7	6.0±1.2 (5–7)
DTR 11	10	10	13	11.5±1.3 (10–13)
GSDT 9	10	9	9	9.3±0.5 (9–10)
SR 170	168	169	164	167.8±2.6 (164–170)
SMC 205	227	229	218	220.0±11.0 (205–229)
V 49	48	46	50	49.0±2.2 (46–51)
LF1 14/13	13/13	14/14	13/13	14.3±1.5 (13–17)
LF4 13/14	14/14	12/15	16/15	14.4±1.3 (12–16)
LT1 16/15	13/13	15/15	14/13	14.5±1.1 (13–16)
LT4 15/14	14/14	15/15	15/17	15.6±1.2 (14–17)
PP 5	0	0	0	—
PAT 2/2	2/3	2/2	2/2	2.1±0.4 (2–3)
T1W 0	2	0	2	1.0±1.2 (0–2)
T5W 0	0	0	0	—
S3W 11	11	12	13	11.8±1.0 (11–13)