Gehyra variegata ( Duméril & Bibron, 1836 )

Hutchinson, Mark N., Sistrom, Mark J., Donnellan, Stephen C. & Hutchinson, Rhonda G., 2014, Taxonomic revision of the Australian arid zone lizards Gehyra variegata and G. montium (Squamata, Gekkonidae) with description of three new species, Zootaxa 3814 (2), pp. 221-241 : 222-229

Gehyra variegata ( Duméril & Bibron, 1836)

Figs. 1 , 2A , 3 , 4A , 5 , 6 .

Hemidactylus variegatus Duméril & Bibron, 1836: p. 353 . Lectotype: MNHN 2295 from Shark Bay, Western Australia. Lectotype designation Wells & Wellington (1983)

Specimens scored for morphology (n=29; all genotyped as “ variegata clade”): SAMA: R 26487– 88, 68 km N Colona HS, SA (30º 51’S, 132º 09’E) , R 31997– 98, 5 km S Mitcherie Rockhole , SA (31º 30’S, 132º 50’E) , R 32175, 8.5 km SW Maralinga , SA (30º 29’S, 131º 31’E) , R 32281, 42.5 km N Muckera Rockhole , SA (29º 42’S, 130º 07’E) , R57176–77 , 48.7 km S Vokes Hill Corner , SA (28º 51’S, 130º 29’E) , R 58995, 47.4 km W Oak Valley , SA (29º 31’S, 130º 15’E) , R 59074, Mt Gibson camp, WA (29º 45’S, 117º 09’E) , R63255–56 K, Eyre Highway at Fraser Range, WA (32º 01’S, 122º 49’E) , R63281–83 K, 57 km ENE Balladonia Rock , WA (32º 24’S, 124º 28’E) , R 65764 K,5.6km ENE Maralinga (30º 08’S, 131º 38’E) , R 65769 K, 3.8 km E Maralinga (30º 09’S, 131º 37’E) , R 65782 K, 31.8 km NNW Pidinga Tank (30º 36’S, 132º 01’E , R 65790 K, 3.9 km E Maralinga (30º 10’S, 136º 11’E) . WAM: R100002 , Kwolyin, WA (31º 55’S, 117º 46’E) , R114039 , Peron homestead, WA (25º 50’S, 113º 33’E) , R117025 , Bush Bay (25°09’S, 113°47’E) , R 126810, 10km ESE Mardathuna homestead (24°30’S, 114°38’E) , R136313 , Muggon Station (26°31’S, 115°31’ 115°31`E) , R141460 , R141467 , Faure Island , Shark Bay, WA (25º 50’S, 113º 54’E) , R141662 , Cape Rose, WA (25º 45 S’, 113º 39’E) , R156674 , North-West Coastal Highway, near Billabong Roadhouse (26°49’S, 114°36`E) , R165160 , 8.5 km WSW Yanyare River mouth (20º 50’S, 116º 22’E) . WAM R117025 and 141662 are both sexually immature juveniles ( SVL 27 and 28 mm respectively) and were excluded from adult body measurements .

Non-sequenced specimens examined. Paralectotypes: MNHN: 254, 254 A, 254 B, “Van Diemen’s Land”. Additional topotypic specimens from Peron Peninsula, Shark Bay Western Australia ): WAM: R 54636, 1 km S Monkey Mia, Shark Bay , WA (25º 48’S, 113º 42’E) , R54639–40 , Peron Homestead , Peron Peninsula, WA (25º 50’S, 113º 33’E) , R 54654, Monkey Mia, Shark Bay , WA (25º 48’S, 113º 43’E) , R 54717, Peron Hills, 28 km NW Denham, WA(25º 43’S, 113º 26’E) , R 54638, 54806–07 , 10 km NE Denham, Shark Bay , WA (25º 52’S, 113º 36’E) , R 54873, 13 km S Nanga Peron Peninsula , WA (26º 20’S, 113º 54’E) , R 55003, 55067 , 18km SE Nanga, Peron Peninsula , WA (26º 20’S, 113º 58’E) , R 55069- 72, 8 km SE Nanga, Peron Peninsula , WA (26º 17’S, 113º 51’E) , R 60428, 3 km N Peron Homestead , WA (25º 48’S, 113º 33’E) , R 123646, 3 km NW Peron Homestead , WA (25º 49 S’, 113º 33’E) .

Diagnosis. Distinguished from other Australian Gehyra (except G. versicolor ) by a combination of 7 or 8 divided scansors under the expanded portion of the fourth toe, moderate size, generally two pairs of enlarged chin shields, second infralabial notched and a dorsal pattern in which dark lines and white markings coordinate to produce a pattern of dark lines and bars with white trailing edges. Not readily distinguishable by external morphology from G. versicolor sp. nov. (see below), but distinguished karyotypically by the unique 2n=40b arrangement ( King 1979). Otherwise most similar to G. montium but distinguished by grey to brown rather than more rufous dorsal colouring, with white markings that form a posterior highlight or margin on the trailing edge of the dark dorsal lines, rather than small poorly contrasting dots that are not coordinated with the dark markings.

This diagnosis applies to populations of Gehyra genetically assignable to the “ variegata clade” of Sistrom et al. (2013).

Description (based on the lectotype and recent, genotyped material) Adult snout-vent length 34–49 mm (mean = 44.2 mm, n = 27). Length of original tail 43-48 mm (mean 107% of SVL, n=3). Sarre (1998) recorded that individuals in some Western Australian Wheatbelt populations can reach 60 mm SVL.

Nostril bordered by rostral, first supralabial, supranasal and two subequal post nasals. Most frequently, a single moderate internasal separates the supranasals above the rostral, but supranasals often (f = 0.41) in medial contact. Supralabials 8– 10 (mode 9). Infralabials 7–10 (mode 9). Usually two pairs chin shields (four specimens with 3 pairs), anterior pair in contact with only the first infralabial. Chin shields separated from the third and succeeding infralabials by the interpolation of a series of enlarged scales (parinfralabials, King 1982) that margin the ventral edge of the infralabials. Second infralabial notched where this parinfralabial scale row starts ( Fig 4A ). Scansors under pad of fourth toe divided, 6–8 (mode 8). Precloacal pores in males 10–15 (mean = 12.5, n = 11) arranged in a chevron with median pore anterior most.

The karyotype shown in Fig. 2A is 2n=40b ( King 1979, Moritz 1986).

In life, dorsally light to medium grey or brown, generally with a complex pattern of white-edged black lines ( Figs 1B , 5 A-B). These usually include several temporal streaks and often form continuous paravertebral and dorsolateral irregular lines, often with short cross bars that may form a vaguely ladder-like pattern. A trend is for specimens from rock-dwelling populations to have bolder and more contrasting markings, especially the white highlights which are expressed as spots ( Fig. 5A ), while some animals from arboreal populations may have only weakly contrasting patterns ( Fig. 5B ). However, colour pattern and intensity is variable both within and between populations, and individual animals can change colour intensity from pale to very dark in the course of a day. In preservative, the colour pattern is often greatly reduced in contrast and can be hard to discern. These patterns of extensive interpopulation, intrapopulation and temporal colour variation are characteristic of most species of Australian Gehyra .

Distribution. Widespread through the southern half of Western Australia, from the west coast as far north as the Pilbara, east to the Central Ranges and southeast to the western interior of South Australia ( Fig. 3 ).

Comments. General. The combination Gehyra variegata has long been applied Australia-wide to populations of morphologically similar Gehyra species with similarly generalist habits. Sistrom et al. (2013) clearly distinguished two major genetic clades within these populations, and confirmed that the karyotypic differences between western and eastern populations first noted by King (1979) reflect the existence of at least two distinct species. Indeed, the two are not even sister lineages. The two populations have generally similar colour patterns but are subtly different in proportions. Sistrom et al. (2013) could distinguish samples of the two species using multivariate analysis but extensive overlap of character variation is such that identification of an individual from morphology alone is problematic. Allocation of the name variegata to the western species centres on the correct identification of the type locality and evidence that specimens from the vicinity of this locality are conspecific with the more widespread western populations.

Type locality. The syntypes of Hemidactylus variegatus in the MNHN collection are from two localities, Shark Bay, Western Australia (one specimen, MNHN 2295) and Tasmania (three specimens, MNHN 254, 254A and 254B). Duméril & Bibron (1836) reported that the collectors of MNHN 2295 were [Jean-René] Quoy and [Joseph Paul] Gaimard, naturalists who travelled with Freycinet’s 1817–1820 expedition. The Shark Bay region (specifically the Peron Peninsula and Dirk Hartog Island) was the only location where the expedition spent any time (a week in September 1818) collecting in an area of Australia that would have yielded specimens of Gehyra ( Arago, 1823) . By contrast, the paralectotypes, MNHN 254, 254A and 254B, cannot be associated with a collecting locality. The given locality of Van Diemen’s Land (Tasmania) is in error as no Gehyra species (in fact, no gekkotans) occur in Tasmania, Hutchinson et al. 2001). The three specimens were collected on the Baudin expedition (1800–1803), which visited several localities along the west coast of Western Australia, including Shark Bay, where specimens of Gehyra might have been obtained ( Bonnemains et al. 1988), but there is no more detailed information that would allow the locality to be identified. All four syntypes are in poor condition, very faded, and soft; MNHN 254B is almost macerated. Without seeing any of these specimens, and under the misapprehension that all were from Shark Bay, Wells & Wellington (1983) named “the largest” specimen registered under MNHN 2295 as the lectotype. In fact there is only a single specimen registered under this number.

The lectotype is distinctive in being very small ( SVL 35 mm) for an apparently adult male (highly active precloacal pores). A series of 16 Western Australian Museum specimens from the Peron Peninsula was examined and revealed only one Gehyra morphotype, similar to the “ variegata clade” in colour pattern. All specimens are noticeably small in size, the three largest individuals ( WAM R 54640, 54654 and 55070) being 40 mm SVL. The smallest male with active precloacal pores ( WAM R 54636) measured 34 mm SVL; the smallest females with an enlarging ovarian follicle or a fully yolked egg measured 38 mm SVL. By comparison, in our random sample of genetically typed specimens from across the species range the smallest male with active pores and smallest ovigerous female both measured 41 mm SVL and the largest animals were 49 mm SVL. Thus the Shark Bay specimen ( MNHN 2295) is consistent in its small adult size with recently collected specimens from the same area, and we are confident that the type locality for Hemidactylus variegatus is the Peron Peninsula and that MNHN 2295 is correctly designated as lectotype.

Conspecifity of topotypic and “ variegata clade” populations. Molecular data ( Sistrom et al. 2013) show that all of the samples that have been DNA sequenced from the vicinity of Shark Bay (specimens from Peron homestead, Cape Rose, Babbage Island, Bush Bay, Faure Island, Baudin Island, and the North-west Highway near ‘Billabong’ roadhouse) are members of the “ variegata clade”. Our two sequenced Peron Peninsula specimens (Cape Rose and Peron homestead) are closely related, and some other Shark Bay area specimens are relatively close genetically to them (e.g., the Carnarvon area specimens from Babbage Island and Bush Bay), while others are more distant (e.g., the Faure Island specimens). Equally, some geographically remote specimens are close genetic relatives of Shark Bay area specimens; the nearest relative of the Peron Peninsula specimens is a specimen from near Dampier (mouth of the Yanyare River) more than 600 km to the northwest, while the Carnarvon specimens are even more closely related to Pilbara specimens (Pannawonica, Muggon Station), than to the Peron Peninsula specimens, and one of the Faure Island specimens is closest to one from Neale Junction on the edge of the Great Victoria Desert, over 1200 km distant. These patterns of genetic relatedness are consistent with a single species with subpopulations that have undergone dispersal and introgression subsequent to a period of isolation and differentiation. Genetic studies in the Western Australian wheatbelt have suggested that this species of Gehyra can disperse widely over relatively short time scales ( Hoehn et al. 2007).

However the species varies in body size and this morphological variation may be indicating very recent differentiation not reflected in the initial genetic survey. In contrast to the dwarfish populations on the Peron Peninsula, populations in the Western Australian Wheatbelt, can grow to large body sizes. How et al. (1988) and Sarre and coworkers ( Sarre et al. 1995; Sarre 1996, 1998) who worked on these populations reported the smallest breeding female at 44 mm SVL and the maximum recorded SVL as much as 61mm. As with the Peron Peninsula ‘dwarfs’ these Wheatbelt ‘giants’ are nested genetically well within the “ variegata clade”. We note that body size alone need not be an indicator of speciation in Gehyra , as our data in this study on G. moritzi sp. nov. (q.v.), as well as striking size variation in G. lazelli ( Sistrom et al. 2012) show that intraspecific selection for local differences in body size has happened more than once in the Australian Gehyra radiation. Consequently, we conclude that the observed body size variation alone does not overturn the idea of only a single species across these populations. Further work, integrating additional genetic and morphological data, will be necessary to better understand the variation seen among populations of this species as we have construed it.

Contact between G. variegata and G. versicolor , sp. nov. The contact between our restricted concept of G. variegata and the eastern G. versicolor sp. nov. (2n=40a/Clade 5 of Sistrom et al. 2013) appears to occur at the eastern margins of the Nullarbor Plain ( Fig.6 ). King (1979) mapped 2n=40a individuals as far west as Bates on the transcontinental railway line. Sistrom et al. (2013 and unpublished) found only true G. variegata (DNA data from nine individuals from the far west of South Australia), and these findings have been supported by seven more individuals that we karyotyped from around Maralinga (40 km NW of Bates) and a site on the northeast Nullarbor Plain (23 km SE of Bates) which all proved to have the 2n=40b karyotype. Thus our data, when combined those of King, appear to show an abrupt changeover in this area from G. variegata to the eastern species. It is noteworthy that we have found only G. variegata (i.e,. 2n=40b) in this area while King found only G. versicolor sp. nov. (2n=40a), one from Bates, another five from several localities approximately 160–180 km further east. Further detailed sampling in this area with both karyotypic and DNA sequence data would be very useful in exploring whether there is any gene flow between the two.

Figures

FIGURE 1. A. Lectotype of Hemidactylus variegatus Duméril & Bibron, 1836, MNHN 2295, adult male from Shark Bay, Western Australia. Inset shows the active precloacal pores. B. Live male Gehyra variegata from 8 km NE of Denham, Peron Peninsula, Western Australia (Photo: Ryan Ellis). This specimen with well-developed dark longitudinal lines; see Fig 5 for some other colour pattern variations.

FIGURE 2. Karyotypes which are species-specific. A, 2n=40b, Gehyra variegata, SAMA R65764 from the Maralinga area, South Australia. Note submetacentric pairs 3 and 6. B, 2n=42b, Gehyra pulingka sp. nov, SAMA R51574 from the Amata area, South Australia. The submetacentric pair 6 with a secondary constriction is characteristic. C, 2n=40a, Gehyra versicolor sp. nov, SAMA R51790 from Yudnamutana Bore, Flinders Ranges, South Australia, with submetacentric pairs 5 and 7. B. 2n=38a karyotype, G. versicolor sp. nov, SAMA R65948 from Native Gap, SSE of Aileron, NT, with submetacentric pairs 5, 7 and 8.

FIGURE 3. Distribution of the species of the Gehyra montium - variegata complex, based on populations identified from DNA sequences. Inset shows the detailed distributions of species in the Central Ranges area.

FIGURE 4. Chin shields of Gehyra showing features used in species descriptions. Numbers indicate chin shield pairs. Grey shaded scales are the notched infralablial and the parinfralabial scale series. A. Gehyra variegata (SAMA R59074), with two pairs of chin shields and the second infralabial notched by the parainfralabials. B. Gehyra pulingka sp. nov (SAMA R51574), with three pairs of chin shields and the third infralabial notched by the parainfralabials.

FIGURE 5. Gehyra variegata Duméril & Bibron, 1836. A. Adult female, SAMA R63256, from a rocky microhabitat from Fraser Range, Western Australia, showing a complex reticulate pattern, with strongly contrasting dark markings strongly margined with white. B. Adult female, SAMA R65740, from an arboreal microhabitat from Maralinga South Australia, showing a less well-developed pattern with weaker pale margins.

FIGURE 6. Map showing the documented occurrence of G. variegata (dark grey) and G. versicolor sp. nov. (light grey). The inset shows the western part of the state of South Australia where distributions of the two species approach one another. Dashed line indicates the transcontinental railway line.