Correlophus belepensis Bauer, Whitaker, Sadlier & Jackman

Bauer, Aaron M., Jackman, Todd R., Sadlier, Ross A. & Whitaker, Anthony H., 2012, Revision of the giant geckos of New Caledonia (Reptilia: Diplodactylidae: Rhacodactylus), Zootaxa 3404, pp. 1-52: 25-32

publication ID

http://doi.org/ 10.5281/zenodo.211734

persistent identifier

http://treatment.plazi.org/id/5C4F87E1-FFA6-FFB9-088F-FF16D643B788

treatment provided by

Plazi

scientific name

Correlophus belepensis Bauer, Whitaker, Sadlier & Jackman
status

sp. nov.

Correlophus belepensis Bauer, Whitaker, Sadlier & Jackman  sp. nov.

( Figs. 12–15View FIGURE 12View FIGURE 13View FIGURE 14View FIGURE 15)

Holotype. MNHN 2011.1100 (formerly AMS 161281), adult male ( Figs. 12–15View FIGURE 12View FIGURE 13View FIGURE 14View FIGURE 15). New Caledonia, Province Nord, Îles Belep, Île Art, 2 km E Waala, Wênè Côgat, 19 ° 42 ' 46.9 " S, 163 ° 39 ' 37.7 " E, 230 m. Collected 23 May 2002 by A.H. Whitaker and V.A. Whitaker.

Paratypes. AMS R 161282 – 283, CAS 250865 (formerly AMS R 161284) ( Fig. 15View FIGURE 15). Data as for holotype.

Diagnosis. Correlophus belepensis  sp. nov. is a large (to 100 mm SVL) diplodactylid. It differs from C. sarasinorum  in possessing a prominent crest of spinose scales extending from behind the orbit to the shoulder region. It is most similar to its sister taxon C. ciliatus  , but may be distinguished from it by its homogeneous dorsolateral trunk scalation ( Figs. 12View FIGURE 12, 13View FIGURE 13 A) (versus a raised and enlarged series of scales extending posteriorly from the end of the spinose crest, Fig. 10View FIGURE 10), and (in three of the four types) its possession of a series of small whitish tubercles on the lower back and/or tail base ( Figs. 12View FIGURE 12, 13View FIGURE 13 A, 14 A, 15) (absent in C. ciliatus  ).

Description. (data from adult male holotype, MNHN 2011.1100). Specimen fixed with mouth open wide; abdominal incision for removal of liver sample for DNA. SVL 95.6 mm; TailL 11.6 (broken with minimal regeneration); TrunkL 39.4 mm; HeadL 27.4 mm; HeadW 21.6 mm; SnEye 11.2 mm; OrbD 7.1 mm; EyeEar 7.8 mm. Body moderately long (TrunkL = 41 % SVL), slender, slightly depressed. Head triangular, large (HeadL = 29 % SVL), very wide (HeadW = 79 % HeadL), very well demarcated from neck ( Fig. 12View FIGURE 12); nasofrontal region depressed, dorsal orbital rims raised; canthus distinct; snout relatively long (SnEye = 41 % HeadL), much longer than eye diameter (OrbD = 64 % SnEye). Scales on dorsum of snout approximately two to five times the diameter of those on occipital region, largest along canthus and immediately posterior to supranasals. Eye large (OrbD = 26 % HeadL); pupil oval, margins faintly crenellated. Ear opening approximately two to three times longer than high, canted posterodorsally to anteroventrally at <45 ° to the horizontal; eye to ear distance longer than diameter of eyes (EyeEar = 110 % OrbD). Rostral rectangular, much broader (4.1 mm) than high (2.2 mm), without rostral crease, contacted posteriorly by three internasals — median smaller and pentagonal, lateral ones larger and hexagonal — and two large supranasals, each somewhat larger than the lateral internasals; contacted posteroventrally by first supralabial. Nostrils oval, laterally oriented, surrounded by eight circumnarial scales, including enlarged supranasal, and broadly contacted by first supralabial. Mental subtriangular, deeper (2.3 mm) than broad (1.7 mm). Two somewhat enlarged (6–8 times size of throat granules) postmental scales separate first infralabials from one another. Postmentals bordered posterolaterally and posteriorly by series of five smaller (~ 50 % size of postmentals) scales. Scales in two to five rows medial to infralabials somewhat enlarged and elongate. 13 (left)– 15 (right) enlarged supralabial scales, posteriormost only about 3 times size of rictal scales, 10 (left) or 12 (right) supralabials to midpoint of orbit; 12 (left)– 13 (right) enlarged infralabial scales (right infralabials 8–11 fragmented into two rows of scales); 38 scale rows between supraciliaries, 18 scale rows across frontal bones at midpoint of orbit. Supraciliary scales forming 17 (L) and 18 (R) large pointed spines, smaller anteriorly, largest posterodorsally and posteriorly.

Dorsal scales small, mostly homogeneous ( Figs. 12View FIGURE 12, 13View FIGURE 13 A), conical; ventral scales 1.3–2 times size of dorsals, smooth, flattened, oval, juxtaposed to subimbricate, enlarged under sternum, midventrally on posterior abdomen, and in precloacal region. Sternal and posterior abdominal scales rounded. Approximately 154 scale rows around mid-body. A dorsolateral fold of skin from anterior to ear to approximately level of adpressed elbow, strongly projecting above ear and on neck, continuing posteriorly as a slightly raised crest. Entire fold/crest bearing elongate spinose scales, becoming smaller above shoulder; a few small, scattered spinose scales between anterior margin of fold and posteriormost spinose supraciliaries. Lumbosacral region with six scattered, enlarged, semi-erect conical white tubercles, with their apices directed posteriorly or posterolaterally. Another two such tubercles at tail base, followed posteriorly by several slightly enlarged and irregularly-shaped non-tuberculate scales.

Skin folds small but distinct on ventrolateral trunk between limb insertions, on ventrolateral border of neck, and on posterior margins of forelimbs. Well developed folds along postaxial border of hindlimbs, forming deep popliteal pockets behind the knees ( Fig. 13View FIGURE 13 B). Scales on postaxial hindlimb fold as far as base of digit V of pes with scattered, white, conical to weakly spinose tubercles. Scales on limbs granular to low and conical, subimbricate to juxtaposed. Scales on palms and soles smooth, flattened. Fore- and hindlimbs, exclusive of skin folds, moderately long and slender (ForeaL = 14 % SVL; CrusL = 18 % SVL), axillary pocket weakly developed.

Digits moderately long, all bearing claws, those on digit I of both manus and pes reduced and partially sheathed, remaining claws long and strongly recurved; relative length of digits of manus: IV>III>II>V>I, and of pes: IV>III>V>II>I; digits well webbed; digits III and IV of pes tightly bound along length of elongate metatarsals. Subdigital lamellae all unpaired, somewhat bowed, with lateral margins gently angled distally (except for proximalmost lamellae, which are straight). Claw of digit I of manus and pes, lies between a smaller lateral and a larger (twice size of lateral) medial apical scansor. Lamellar counts from right (and left) sides 10-13 - 17-18 - 15 (10- 15 - 18 - 17 - 15) manus and 12-15 - 17-20 - 18 (12-14 - 18-19 - 17) pes (excludes apical scansors of digit I).

Small precloacal pores in a patch of somewhat enlarged scales, arranged in three rows (anterior to posterior), each divided by two small poreless scales medially, of 15 (L) + 15 (R), 12 (L) + 10 (R), and 3 (L) + 4 (R). Pores restricted to proximal 40 % of thighs ( Fig. 13View FIGURE 13 B). Hemipenial bulge large; a single large, smooth, flattened, conical cloacal spur comprising a much larger dorsal scale and a small posterolateral scale subtending it on each side of tail base. Tail broken at base and supporting a rudimentary regenerate of narrow diameter (see Discussion).

Color in preservative (based on holotype): Dorsum, lateral surfaces of trunk, side of neck and head posterior to orbit, and limbs more-or-less a uniform dark chocolate brown. Snout, dorsum of head and middorsal region between dorsolateral spinose crests a lighter brown with a tinge of brick red ( Fig. 12View FIGURE 12). Posterior infralabial scales with whitish ventral borders. Midvertebral region between end of spinose crests and white tubercles of lumbar region mottled with some hint of alternating lighter and darker brown bands. Venter light brown with darker mottling, especially on anterior chest, mid- to posterior abdomen, in precloacal region, and under hemipenial bulge. Soles and palms pale grayish brown. A white line along the postaxial margin of the hindlimb. Tail base bearing a cream “Y”-shaped marking dorsally; vestigial regenerated tail beige with dark pigment around its base.

In life the dorsal coloration of the types was an orangey-brown with a slightly greenish suffusion, with darker diffuse irregular reticulations on the flanks. Crown of the head, area between denticulate crests, and anterior middorsal region more distinctly orangey to russet than adjacent flanks, fading either abruptly or gradually posterior to the shoulders. Anterior of head with yellowish highlights; orbital rim and ventrolateral surfaces of supraciliary scales yellow. Iris silvery to whitish ( Fig. 14View FIGURE 14). Venter reddish-brown, with diffuse, darker gray-brown irregular markings, particularly along the ventrolateral margins and with scattered isolated greenish-yellow scales. Tongue and interior of mouth unpigmented.

Osteology. Vertebral counts are typical for diplodactylid geckos, with 26 presacral and 2 sacral vertebrae. The first three cervical vertebrae are without ribs, as is the last presacral (lumbar) vertebra. Also typical for the family, the caudal skeleton includes 5 pygal vertebrae. The holotype and all paratypes have the tail broken within the first postpygal vertebra. The phalangeal formulae of the manus and pes are unreduced, 2-3 - 4-5 - 3 and 2-3 - 4-5 - 4, respectively. Total tooth loci in upper jaw of holotype 85, of which 9 in premaxilla; total mandibular tooth loci 78. A single pair of crescentic cloacal bones is present in the holotype and in the male paratype AMS R 161282. In both female paratypes the endolymphatic system is visible in x-rays by its radio-opaque calcium content and can be seen to be entirely intracranial.

Variation. Comparative mensural and meristic data for the holotype and paratypes are given in Table 2. Meristic characters of paratypes are mostly similar to those of the holotype, and are mentioned hereafter if only they differ. All paratypes had only a single postmental scale. AMS R 161283 had seven circumnarial scales and CAS 250865 had seven around the left naris and eight around the right. Precloacal pores are arranged in rows of 18 (L) + 15 (R), 10 (L) + 10 (R), and 3 (L) + 2 (R) in the male paratype ( AMS R 161282) and are absent in the female paratypes. In the male paratype the posterolateral scale of the cloacal spur is separate from the larger scale, resulting in two discrete cloacal spurs on each side of the tail base. In the females the cloacal spurs are represented by one to three slightly enlarged smooth conical scales.

MNHN 2011.1100 AMS R 161282 AMS R 161283 CAS 250865 holotype paratype paratype paratype

Sex male male female female

SVL 95.6 88.6 99.7 99.4

ForeaL 13.4 12.1 13.0 12.3

CrusL 17.2 15.3 16.7 16.4

TailL* 11.6 12.9 15.0 16.4

TrunkL 39.4 34.5 40.6 41.0

HeadL 27.4 25.4 29.6 27.8

HeadW 21.6 19.8 22.4 22.0

OrbD 7.1 6.8 7.6 7.0

EyeEar 7.8 7.5 8.8 8.0

SnEye 11.2 11.1 11.8 11.7

NarEye 8.0 8.2 8.5 8.5

Interorb 11.3 10.3 13.2 13.9

EarL 2.2 1.3 1.9 occluded

Internar 3.3 3.2 3.4 3.7

SupraL L/R 13 (10)/ 15 (12) 15 (12)/ 16 (13) 14 (11)/ 14 (11) 15 (12)/ 14 (11)

InfraL L/R 12 / 13 13 / 13 13 / 14 13 / 13

LamManus L/R 10-15 - 18 - 17 - 15 / 10- 14 - 15-16 - 19 - 15 / 13-14 - 11-14 - 17-19 - 15 / 10 - 9-13 - 17 -20- 14 / 11- 13 - 17-18 - 15 16-19 - 15 14-18 - 18 - 15 13-16 - 17 - 15

LamPes L/R 12-14 - 18-19 - 17 / 12- 13 - 16-18 - 19 - 18 / 13-15 - 10-13 - 16-19 - 17 / 10 - 8-14 - 19 - 19 - 17 / 10- 15 - 17-20 - 18 17-21 - 17 15-19 - 21 - 17 15-18 - 21 - 17 Color pattern is generally similar among all specimens examined, although in AMS R 161282 and CAS 250865 the dark chocolate dorsal and lateral coloration is more mottled and less uniform than in the holotype. The number of white tubercles on the dorsum is also variable ( Fig. 15View FIGURE 15). AMS R 161283 has three tubercles on the lumbosacral region, two to the left of the midline and one to the right. In AMS R 161282 there is a single such scale anterior to the hindlimb insertion and left of the dorsal midline and several irregular scales, including one small white tubercle at the tail base, but neither tubercle is conical as in the holotype. In CAS 250865 there are no lumbosacral tubercles. Venter in life yellowish ( CAS 250865) to reddish-brown; females lacking the scattered greenish-yellow ventral scales of males.

Etymology. The specific epithet refers to the Îles Belep, to which this species is apparently restricted.

Distribution. Correlophus belepensis  has been recorded only from Île Art, in the Îles Belep group more than 40 km north-west of Grande Terre ( Whitaker et al 2004; Fig. 11View FIGURE 11)). Two discrete sub-populations occur on Île Art, 4.5 km apart, one on the Tolé Munu plateau (c. 700 ha) and the other on the Kalidan plateau (c. 220 ha). The type series was collected on the former; specimens were observed, but not collected, at Puröbi (19 ° 45 '06.1"S, 163 ° 40 '29.0"E, 230 m) on the latter. These plateaux comprise an undulating ultramafic substrate between 220–250 m in elevation that is capped by a rocky cuirasse surface covered with closed forest or paraforestier. They are separated by a wide, low saddle (<90 m) with sparse fire-induced savannah woodland and grassland. A small area (c. 90 ha) of similar habitat to that on the Tolé Munu and Kalidan plateaux occurs at lower elevation (140–150 m) on Île Pott, the only other large island in the Belep group, but it has not been surveyed.

Natural History. Correlophus belepensis  has only been found in old-growth, low-stature (<10m) closed humidforest and paraforestier habitat (<8 m) on bouldery cuirasse surfaces ( Fig. 16View FIGURE 16). The Tolé Munu plateau has a diverse flora (> 170 species) including several rare and threatened taxa, several of which are endemic to that site ( Munzinger et al. 2009). Dominant canopy trees in this forest, with large trunks with holes and crevices as retreat sites for geckos, are Ficus prolixa G. Forst.  (Fam. Moraceae  ), Alphandia resinosa Baill.  (Fam. Euphorbiaceae  ), Iteiluma  sp. (Fam. Sapotaceae  ), Planchonella wakere (Pancher & Sebert) Pierre  (Fam. Sapotaceae  ), Mimusops elengi  L. (Fam. Sapotaceae  ), and Piliocalyx laurifolius Brongn. & Gris  and Piliocalyx  sp. (Fam. Myrtaceae  ). Fortuitously, browsing ungulates and pigs are absent from the Belep Islands so this forest is in reasonably good condition with thick undergrowth. All observations of C. belepensis  have been at night while they are active and it appears to be an exclusively arboreal species. Unlike Mniarogekko jalu  n. sp., with which it is syntopic, C. belepensis  was found at all levels within the forest from the upper surface of the canopy to low in sub-canopy and forest margin shrubs. Perches used were similarly varied and included foliage, twigs, branches and trunks. The new species is particularly agile when moving through the vegetation and frequently leaps between branches, as much as 30 cm vertically upwards and> 40 cm horizontally. When threatened they jump from the canopy or other vegetation and fall, spread-eagled and horizontal, to lower plants or the forest floor. If pursued on the ground they move in a series of frog-like leaps, up to 12 cm high and covering 30–40 cm in each bound. A similar tendency to jump has been documented in its sister species C. ciliatus ( Vosjoli et al. 2003)  .

On Île Art C. belepensis  is syntopic with Mniarogekko jalu  n. sp., Eurydactylodes agricolae  and Dierogekko insularis  . Species sympatric within its habitat include Kanakysaurus viviparus Sadlier, Smith, Bauer & Whitaker, 2004  , Lioscincus nigrofasciolatus ( Peters, 1869)  , Caledoniscincus atropunctatus ( Roux, 1913)  , C. austrocaledonicus ( Bavay, 1869)  , and C. haplorhinus ( Günther, 1872)  , and Bavayia aff. cyclura  , Hemidactylus frenatus Schlegel  in Duméril & Bibron, 1836, Lepidodactylus lugubris ( Duméril & Bibron, 1836)  , and Phoboscincus garnieri ( Bavay, 1869)  were recorded in adjacent savannah and secondary shrubland.

Trombiculid mites are present between the subdigital lamellae and in one or both popliteal pockets in all of the specimens examined.

Conservation status. Correlophus belepensis  has been recorded only on Île Art, in the Belep archipelago, where it occurs as two sub-populations within a single locality. It is expected that the species may also occur in similar habitat on nearby Île Pott. Whether its entire distribution is confined to the Belep archipelago is unclear but at present there is no evidence of populations on the mainland.

Within the known extent of occurrence for C. belepensis  , the closed forest and paraforestier formations on cuirasse surfaces are surprisingly intact given the high level of habitat modification that has occurred elsewhere on Île Art. Both plateaux are still largely covered by forest, but as a result of repeated burning and clearance for subsistence gardening since colonization by Melanesians, the surrounding slopes and remainder of the island have been modified and are characterized by depleted niaouli ( Melaleuca quinquenervia (Cav.) Blake, Fam.  Myrtaceae  ) savannah, secondary shrublands and grassland. This is assumed to have resulted in a contraction of the area of occupation for this species.

The sub-populations of C. belepensis  on Île Art are at risk to several direct and indirect threats. Wildfires regularly affect large areas of the grassland, savannah and shrubland, and these inevitably encroach on the remaining forest, reducing it in extent. There is also sporadic timber cutting for local use but to date this has had only a very localized impact. Introduced rats ( Rattus  spp.) and feral cats are present throughout the habitat. Both are known to be predators of lizards but no direct evidence of predation on C. belepensis  was obtained. Little red fire ants, Wasmannia auropunctata ( Roger, 1863)  are also present in the forest on the plateaux. These highlyinvasive ants are known to have a severe detrimental impact on lizard populations ( Jourdan et al. 2000, 2001). Hemidactylus frenatus  has colonized the Îles Belep relatively recently and was detected in the township of Waala, as well as in littoral vegetation, garden sites and savannah woodlands. If it spreads into forest habitat, as is expected, it is unclear what impact this invasive gecko  would have on giant gecko  species but elsewhere it is known to result in competitive exclusion and local extirpation of smaller indigenous geckos (e.g., Case et al. 1994; Petren et al. 1993; Petren & Case 1996; Cole et al. 2005; Rivas et al. 2005).

The ultramafic plateaux on both Île Art and Île Pott have been subjected to extensive prospecting in the past and are criss-crossed by old access tracks and drill sites. Although these areas are held under current mining licenses there appear to be no immediate plans to exploit the areas. However, should it occur, mining would almost certainly remove a significant proportion—if not all—of the available habitat for this species.

No quantitative data on population size and trends are available for C. belepensis  . In 2001 it was relatively abundant in the forest on the Tolé Munu plateau, with encounter rates of 0.57–0.83 /person hour, implying a population density approximately half that of the syntopic Mniarogekko jalu  n. sp. ( Whitaker et al. 2004).

Because of its extremely limited extent of occurrence (<25 km ²) and area of occupation (<10 km ²), presence at a single locality, the threats to its habitat (wildfires, and perhaps mining), the presence of mammalian predators (rats, cats) and the impacts of fire ants, Correlophus belepensis  is assessed as Critically Endangered (B 1 a, b[i –iii, v]; B 2 a, b[i –iii, v]) ( IUCN 2001).

Remarks. Correlophus ciliatus  occurs in southern Grande Terre and on the Île des Pins, with a northern known limit of range around 300 km south-east of the Îles Belep (Bauer & Sadlier 2000; Fig. 11View FIGURE 11). The presence of these two taxa at either end of Grande Terre suggests the occurrence of C. belepensis  is relictual and that it is not a primary endemic in the Belep group, but extensive fieldwork in the northern Grande Terre has not revealed any Correlophus  .

Although the diagnostic heterogeneous paravertebral trunk scalation of C. ciliatus  is not illustrated in the type illustration of that species ( Guichenot 1866), it is clearly visible in the lectotype specimens ( Fig 10View FIGURE 10) and in published photos of the species (e.g., Seipp & Klemmer 1994; Gerard 1999; Seipp & Henkel 2001, 2011; Bauer & Sadlier 2000; de Vosjoli et al. 2003; Bach 2006; Henkel & Schmidt 2007; Hamper 2003; Wirth & Peukert 2009; Sommer 2009).

MNHN

Museum National d'Histoire Naturelle

CAS

California Academy of Sciences

DNA

Department of Natural Resources, Environment, The Arts and Sport