Content uploaded by Michelle S Koo
Author content
All content in this area was uploaded by Michelle S Koo
Content may be subject to copyright.
Burmese Hemidactylus (Reptilia, Squamata, Gekkonidae):
Taxonomic Notes on Tropical Asian Hemidactylus
George R. Zug
1
, Jens V. Vindum
2
, and Michelle S. Koo
3
1
Department of Vertebrate Zoology, National Museum of Natural History, Smithsonian Institution,
Washington, DC, 20560 USA.; Email: zugg@si.edu;
2
Department of Herpetology, California
Academy of Sciences, San Francisco, CA, 94103 USA; Email: jvindum@calacademy.org;
3
Museum of Vertebrate Zoology, University of California, Berkeley, CA. 94720 USA
Five species of the gecko genus Hemidactylus are commonly reported from Myanmar
(Burma). A sixth gecko, Cosymbotus platyurus, has been shown recently to be within
the tropical Asian clade of Hemidactylus species and is included in that genus here.
Not all tropical Asian species were included in that molecular study; thus we provide
a preliminary assessment of the taxonomic status of all species in the putative tropi-
cal Asian clade, with an emphasis on the species occurring within political confines of
Myanmar.
One or more species of Hemidactylus geckos occur at most localities inventoried by our Myan-
mar Herpetological Survey program (1997–2003). With a few exceptions, these geckos are readily
assigned to the five species recognized as occurring in Myanmar (Smith 1935): H. bowringii,
H. brookii, H. frenatus, H. garnotii, H. karenorum. None of these species is uniquely Burmese,
although H. karenorum has the smallest distribution of the five, occurring in Assam and Myanmar
(Smith 1935). The other taxa are more broadly distributed and two (H. frenatus, H. garnotii) have
become global via accidental human transport. A sixth species is added to the Burmese Hemidacty-
lus fauna by molecular analysis of Carranza and Arnold (2006) identifying Cosymbotus platyurus
as a member of the tropical Asian clade.
Among our biodiversity inventories, we occasionally find Hemidactylus geckos that do not
match comfortably the general characteristics that we use to recognize and differentiate the six
“typical” Burmese species. These atypical geckos are not, however, the focus of this report. Rather,
we provide a preliminary assessment of taxonomic matters affecting the known species of Burmese
Hemidactylus arising from the molecular phylogenetic study of Carranza and Arnold (2006; abbre-
viated henceforth as C&A-06). Before proceeding with that assessment, we offer a brief review of
the currently recognized species of Burmese Hemidactylus.
MATERIALS AND METHODS
The morphological data of this report derive solely from Burmese specimens vouchering the
regional inventories of the Myanmar Herpetological Survey. The specimens examined are identi-
fied in the following section. They represent only a subset of vouchers available and purposefully
derive from lowland to mid-elevation sites (< 500 m asl) of the Central Dry Zone southward to and
including the Ayeyarwaddy and Sittaung deltas. We recognize that these geographically mixed sam-
ples may cloud regional differentiation, but our goal is to provide only preliminary descriptions of
PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Fourth Series
Volume 58, No. 19, pp. 387–405, 7 figs., 1 table. September 28, 2007
387
the morphology of Burmese Hemidactylus. A limited sample size of 20 to 30 individuals for each
species is also compatible with this goal. The maps (Figs. 1–6), however, detail the distribution of
all Burmese Hemidactylus species in the CAS and USNM collections; the specific identity of these
specimens relies heavily on their identification (principally by Htun Win, J. Vindum, or G. Zug) at
time of cataloging.
The morphological data include the following characters: M
EASUREMENTS: CrusL, Crus
length; ForeaL, Forearm length; HeadL, Head length; JawW, Jaw width; SnEye, Snout-eye
length; SnForel, Snout-forelimb length; SVL, Snout-vent length; SnW, Snout width; TrunkL,
Trunk length. S
CALATION: Chin, Chin (postmental) scales; 4FingLm, Fourth finger lamellae (scan-
sors); 4FingDv, Fourth finger lamellae divided; 4ToeLm, Fourth toe lamellae; 4ToeDv, Fourth toe
lamellae paired; Inflab, Infralabials; NaInf, Naris-infralabial contact; SnS, Snout scales; Subcaud,
Subcaudal scales; Suplab, Supralabials; PoreTot, Total pores. These traits are defined in Caleb and
Zug (2007). Statistical analysis performed in SYSAT 11.
The synonymies derive from Wermuth (1965), Zug (1990), Bauer (1994), and Kluge (2002).
Kluge’s list, as the most recent of the synonymies, established the current recognition of available
and valid scientific name for these geckos. Only primary synonyms are given. Type localities are
given as in the original description, and all type localities were verified against the original descrip-
tions.
S
PECIMENS EXAMINED.— Hemidactylus bowringii: Magway Divis., Shwe-Settaw CAS 213598–599,
213603, 213619, 213779, 213782–783, 213835, 213838, 213840, 213845, 213860–861, 213876–878, 213882,
231067; Magway Divis., Shin-Ma-Taung CAS 215841, 215843–844, 215846, 215848; Sagaing Divis., mis-
cellaneous localities - Yinpaungtaing CAS 215392, Khim Aye CAS 215434, 215437, Alaungdaw Kathapa N.P.
CAS 215444, 215519, 215646, 215759, 215777, Chatthin W.S. – USNM 520552–056, 537425–426, Pale –
CAS 215434, 215437, Sweekawngan village – CAS 232243–244.
Hemidactylus brookii: Mandalay Divis., Mt. Popa CAS 213976, 214040–042, 215829, 231277,
231341–342, 231357, USNM 564914–927; Mandalay Divis., Yin Mar Bin CAS 215297; Sagaing Divis.,
Alaungdaw Kathapa NP CAS 206650 ,
Hemidactylus frenatus: Magway Divis., Shwe-Settaw WS USNM 564861–862; Sagaing Divis., Alaung-
daw Kathapa NP USNM 564889–891, Kabaing USNM564895–896, Chatthin WS USNM 520560–565,
524047–051; Sagaing Divis., Yin Bar Min CAS 215306; Yangon Divis., Yangon USNM 520558–559.
Hemidactylus garnotii: Bago Divis., Dawe USNM 564931–932; Mandalay Divis., Pyin-Oo-Lwin USNM
564933–938, 564940, Mt. Popa USNM 564945; Mon State, Kyaikhtiyo WS USNM 564941; Rakhine State.,
Gaw USNM 564944; Sagaing Divis., Alaungdaw Kathapa N.P. USNM 564942–943, Chatthin W.S. USNM
537427.
Hemidactylus karenorum: Magway Divis., Shwe-Settaw CAS 213597, 213816–817, 213833, 213836,
213846, 231136, USNM 564946–947; Mandalay Divis., Mt. Popa CAS 210670–671, 214048, 214057–058,
214078–079, 214142, 231308, 231313, 231340, USNM 564949–951.
Hemidactylus platyurus: Chin State, Nat-Ma-Taung CAS 222332; Sagaing Divis., Alaungdaw Kathapa
NP CAS 204982–983, 210135, 210152–153, 210284, 210286, 210296–297, 210308, 215559, 215585,
215600, 215679, 210308, 215559, 215585, 215600, 215679, 215734, USNM 564952–954, Htamanthi WS
CAS 232192.
SYNOPSIS OF BURMESE HEMIDACTYLUS
Hemidactylus bowringii (Gray) Bowring’s or Asian smooth gecko
Doryura Bowringii Gray 1845:156. Type locality, syntypes without locality data; restricted to “Hong-kong or
neighbourhood” by Smith (1935:99).
Leiurus berdmorei Blythe 1853:646. Type locality, “Mergui” [= Myeik Beik, Tanintharyi Divis., Myanmar].
DESCRIPTION.— Adults 34–51 mm SVL; adult SVL not sexually dimorphic: females average
388 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Fourth Series, Volume 58, No. 19
41.8±4.36 mm SVL (35.6–50.7 mm); male average 39.7±3.95 mm SVL (34.5–49.0 mm). Moder-
ately built, somewhat flattened lizards. Proportions (female – mean, min.-max.; male – same):
SnForel/SVL 38, 34–41%; 38, 35–40%; TrunkL/SVL 47, 42–51%; 44, 34–51%; ForeaL/SVL 12,
10–14%; 12, 11–14%; CrusL/SVL 14, 12–16%; 14, 12–15%. Tail length (unregenerated) slightly
longer than SVL and equal in females and males. Tail flattened, broader than high (oblong in cross-
section), and tapering gradually to thin tip.
Head moderately large and broad, distinct from neck, flattened and conical to obovate in dor-
sal outline. HeadL/SVL 25, 22–27%; 26, 24–29%; JawW/HeadL 64, 57–72%; 62, 51–69%; SnEye/
HeadL 40, 35–43%; 40, 35–45%; SnW/HeadL 7, 4–9%; 6, 4–9%. All digits of fore- and hindfeet
with digital pads. Digital pads oblong in shape with distal end only slightly wider than proximal
end; slender terminal portion (ultimate and penultimate phalanges) of digit arising from within and
free of digital pad of all digits. Claws on all digits; slight or no webbing at base of digits.
Head, body, and tail scalation of small, equal-sized, juxtaposed tubercles dorsally and lateral-
ly; ventrally slightly overlapping scales, >5× dorsal tubercles, from base of neck to pelvic area;
transition from ventral scales to tubercles ventrolaterally; no ventrolateral skin fold on trunk. Head
dorsally and laterally with scales on snout and lips, elsewhere small tubercles. Rostral scale large,
rectangular with middorsal cleft; nares bordered by rostral, first supralabial, and 3 nasal scales,
supranasal largest. Supralabials 7–12, infralabials 6–10. Broad triangular mental scale ventrally,
bordered posteriorly by 2 large anterior chin scales, usually touching one another midventrally; pos-
terior chin scales about half size of anterior ones and rarely
touching medially.
Limbs scaled above and below, except for tubercles on
posterior surface of thigh. Subdigital lamellae on pad: 6–9
on 4
th
finger, distal lamella undivided, subsequent 3–7 divid-
ed; 7–11 on 4
th
toe, distal lamella undivided, subsequent 3–7
divided. Bilateral series of 18–27 (total) precloacal-femoral
pores in males, left and right sides separated at midpelvic by
1–4 nonpore-bearing scales. Tail midventrally with rectan-
gular, slightly overlapping, smooth-surfaced plates from
vent to at least mid-length; large smooth scales ventrolater-
ally quickly grading into tubercles. Tail distinctly segment-
ed, each segment 8–10 scales long; pair of small erect scales
ventrolaterally at posterior edge of each segment.
Hemidactylus geckos can lighten or darken their skin
tones. Adults in dark phase, dorsal ground color from head to
base of tail medium brown to tan with diffuse, small dark-
brown smudges dorsally and dorsolaterally and occasionally
with faint white spots to nearly uniform; laterally often series
of white spots highlighted by dark-brown smudge in front of
each spot. Some individuals with dark-brown postorbital
stripe from eye, above ear-opening, to axilla. Dorsal and lat-
eral pattern of white spots bordered by dark marks sharply
defined in some population, e.g., Shwe-Settaw. In light
phase, dorsum nearly uniform beige or with very faded
markings on light background. Ventrally immaculate cream
to light yellow from chin through belly; pelvic area and
underside of tail light orangish beige.
ZUG ET AL.: TAXONOMIC NOTES ON TROPICAL ASIAN HEMIDACTYLUS 389
FIGURE 1. Distribution of Hemidactylus
bowringii in Myanmar based on voucher
specimens of the CAS/NWCD/SI Myanmar
Herpetological Survey.
DISTRIBUTION.— Bowring’s gecko occurs broadly in Burma (Fig. 1) from the Shan Plateau and
adjacent China (Yunnan [CAS specimens] and Sichuan [Zhao 2003]) southward to the Myanmar
coast from the western Rakhine to southern Tanintharyi. To the west, its distribution includes north-
ern India — Godavari Valley, Sikkim, Darjeeling — and Bangladesh (Smith 1935), North [=West]
Bengal (Tikader and Sharma 1992), and the terai of Nepal (Kästle 2002). This gecko has been
reported only thrice from Indochina: three juveniles from Laos (Bourett 1939), one individual from
Saigon, Vietnam (Bobrov 1992), and apparently one or a few from northern Vietnam (Darevsky et
al. 1984). H. bowringii does occur abundantly in some localities in southern China from Hainan
westward to Taiwan (Zhao and Adler 1993) and the Ryukyu Islands (Ota 1989).
Because of this gecko’s near absence in Indochina, we question the naturalness of its Oriental
distribution. Is it an invasive or does the disjunct distribution denote the presence of eastern and
western species? Resolving this matter is beyond the scope of this study; however for the moment,
we favor the invasive-species explanation. Published evidence (e.g., Karsten et al. 1986; Ota 1989;
Lazell 2002) notes its occurrence predominantly or exclusively as a human commensal. Only Lazell
(2002) reported it as occurring abundantly in woodlands (disturbed); however, he stated (in litt,
9&10 Aug. 2006) that it did not occur in woodlands when Hemiphyllodactylus chapaensis was
present, and furthermore, H. bowringii was abundant and widespread in and around human edifices.
N
ATURAL HISTORY.— Burmese Hemidactylus bowringii is a forest-floor resident. During the
day, it occurs beneath leaf litter, logs, and bark. At night, it forages on and beneath the leaf litter.
Gravid females from central Myanmar samples were taken in May, July, August, and October.
Hemidactylus brookii (Gray) Asian spiny gecko
Hemidactylus Brookii Gray 1845:153. Type locality, “Borneo.” “Australia.”; restricted to “Borneo” by Smith
(1935:89).
Gecko Tytleri Tytler 1865:547. Type locality, “Moulmein” [= Mawlamyaing, Mon State, Myanmar].
Hemidactylus kushmorensis Murray 1884:109. Type locality, “Bhaner, Upper Sind frontier” [presumably =
Bhanar, Sindh, Pakistan].
Hemidactylus Gleadowi Murray 1884:360, Type locality, “Rantha forest in Sind, (Jerruck division)” [= Jerruck
region, Pakistan].
Hemidactylus gleadovii Boulenger 1885:129. Amended spelling.
Hemidactylus Murrayi Gleadow 1887:49. Type locality, “Pimpri and Garvi, in the “Dangs”, [Gujarat State,
India].
Hemidactylus Tenkatei Lidth de Jeude 1895:121. Type locality, “Rotti” [= Pulau Roti (Lesser Sundas), Indone-
sia].
Hemidactylus subtriedroides Annandale 1905:29. Type locality, “Tsagain, Upper Burma” [= Sagaing, Sagaing
Divis., Myanmar].
Hemidactylus brookii parvimaculatus Deraniygala 1953:45. Type locality, “Colombo” [Sri Lanka].
DESCRIPTION.— Adults 45–65 mm SVL; adults sexually dimorphic: females average
53.7±6.28 mm SVL (45.06–61.7 mm); male average 59.6±4.31 mm SVL (50.2–65.0 mm). Moder-
ately robust, slightly flattened lizards. Proportions (female – mean, min.-max.; male – same):
SnForel/SVL 38, 35–41%; 37, 33–40%; TrunkL/SVL 41, 36–48%; 42, 39–46%; ForeaL/SVL 12,
11–14%; 12, 10–13%; CrusL/SVL 14, 12–15%; 13, 12–14%. Tail length (unregenerated) slightly
longer than SVL and equal in females and males. Tail broader than high (oblong in cross-section),
distinctly spiny, and tapering gradually to thin tip.
Head moderately large and broad, distinct from neck, flattened and conical to obovate in dor-
sal outline. HeadL/SVL 25, 22–27%; 26, 24–29%; JawW/HeadL 64, 57–72%; 62, 51–69%; SnEye/
HeadL 40, 35–43%; 40, 35–45%; SnW/HeadL 7, 4–9%; 6, 4–9%. All digits of fore- and hindfeet
390 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Fourth Series, Volume 58, No. 19
with digital pads. Digital pads obovate with distal end slightly wider than proximal base; slender
terminal portion (ultimate and penultimate phalanges) of digit arising from within and free of dig-
ital pad of all digits. Claws on all digits; slight or no webbing at base of digits.
Head, body, and tail scalation of small, equal-sized, juxtaposed tubercles dorsally and lateral-
ly between multiple (14–16) longitudinal rows of enlarged, slightly keeled, conical tubercles; tuber-
cles of parasagittal 2–3 rows slightly smaller and flattened; ventrally, slightly overlapping scales,
> 5× dorsal tubercles, from base of neck to pelvic area; transition from ventral scales to tubercles
ventrolaterally; no ventrolateral skin fold on trunk. Head dorsally and laterally with scales on snout
and lips, elsewhere small tubercles. Rostral scale large, rectangular with middorsal cleft; nares bor-
dered by rostral, first supralabial, and 3 nasal scales, supranasal largest. Supralabials 8–11, infra-
labials 8–10. Broad triangular mental scale ventrally, bordered posteriorly by 2 large anterior chin
scales, broadly touching one another medially; posterior chin scales about two-thirds size of ante-
rior ones and not in contact medially.
Limbs scaled above and below, some large tubercles on anterior and dorsal surface of forearm,
more numerous enlarged tubercles on dorsal surface of thigh and crus. Subdigital lamellae on pad:
6–8 on 4
th
finger, distal lamella undivided, subsequent 5–7 divided; 7–8 on 4
th
toe, distal lamella
undivided, subsequent 5–7 divided. Bilateral series of 11–16 (total) precloacal-femoral pores in
males, left and right sides separated at midpelvic by 4–7 nonpore scales. Tail midventrally with rec-
tangular, slightly overlapping, smooth-surfaced plates from vent to at least mid-length; large
smooth scales ventrolaterally quickly grading into tubercles. Tail distinctly segmented, each seg-
ment 8–10 dorsal scales long; pair of small erect scales ven-
trolaterally at posterior edge of each segment; dorsally tail
distinctly segmented, with middle of each segment bearing
roseate of 6 large, keeled conical tubercles.
Adults in dark phase, two-toned brown from head to
base of tail, top of head and middorsum onto tail lighter
brown than dorsolaterally and sides; sides medium dark-
brown to tan; shape and width of lighter middorsal area nar-
row to broad with broadly scalloped edge; typically in small-
er adults and juveniles small dark spots/dashes scattered in
both light and dark area; in some individuals, laterally
enlarged tubercles lighter than surrounding granular tuber-
cles creating a spotting effect. Many individuals with orbital
stripe from snout through eye to ear; broad whitish stripe on
snout border above and below by dark border, beyond eye
only white stripe and ventral dark border. Ventrally immacu-
late white to cream from chin to underside of tail. In light
phase, two toned dorsum faded but still evident as is faded
orbital stripe behind eye. Ventrally immaculate white to
cream to light yellow from chin through belly; pelvic area
and underside of tail light orangish beige.
D
ISTRIBUTION.— Asian Hemidactylus brookii occurs
widely, but not abundantly, in low elevation areas of north-
central Myanmar southward to southern Thaninthary
(Fig. 2). Kästle (2002) recognized the taxon H. brookii sub-
treidroides and depicted the latter’s occurrence in northern
Myanmar and adjacent Arunachal Pradesh. Our survey of
ZUG ET AL.: TAXONOMIC NOTES ON TROPICAL ASIAN HEMIDACTYLUS 391
FIGURE 2. Distribution of Hemidactylus
brookii in Myanmar based on voucher speci-
mens of the CAS/NWCD/SI Myanmar Her-
petological Survey.
sites in northern Sagaing Division has not found H. brookii in the northernmost reaches of Myan-
mar.
The usual description of a nearly pan-tropical distribution for H. brookii is applicable only to
the paraphyletic concept of this “species.” C&A-06’s Asian brookii still has a broad distribution,
but one that naturally lies entirely east of the Indus R. valley. It is a common house gecko in penin-
sular India and Sri Lanka according to Smith (1935) and others (Sri Lanka, Deraniyagala 1953;
India, Daniel 1983). Tikader and Sharma (1992) mapped its occurrence thoughout India. Das
(2002), however, reported only a northern India occurrence, and Khan (2002) listed it as a common
gecko of the plains of Pakistan and absent from the northern mountains. Kästle (2002) showed a
broad terai occurrence and a few records from mid-montane elevations in Nepal. Bauer and Gün-
ther (1992) reported a single specimen from the Bhutan-Indian border. Pawar and Birand (2001)
found it as human commensal in only two (Nameri National Park, western Arunachal Pradesh, and
Balphakaram NP, central Meghalaya) of eight nature reserves surveyed in Northeast India.
As for H. bowringii, H. brookii is largely absent from Indochina and is excluded from recent
reptile field guides (e.g., Cox et al. 1998, Manthey and Grossman 1997) to this area. De Rooij
(1915) reports H. brookii from Singapore, Sarawak and several Lesser Sunda islands. More recent
surveys (Dunn 1927; Darevsky 1964; Auffenberg 1980; Lian 1993) report H. frenatus, H. garnotii,
and/or H. platyurus from the latter areas but not H. brookii. Denzer and Manthey (1991) noted that
the Singapore record was doubtful. Das and Sukumaran (2006) recently documented a single breed-
ing population of H. brookii in Borneo.
Southern and eastern China records are limited to Hong Kong and Macau (Karsten et al. 1986)
and Zhejiang (Zhao and Adler 1993). Karsten and co-authors stated that it is an introduced species.
The Zhejiang record requires investigation, but owing to its presence near Shanghai, it seems like-
ly a record of an introduced population. A similar explanation of introduction is proposed also for
the Philippine populations of H. brookii; even though it occurs broadly in the Philippines. Brown
and Alcala (1978:30) stated: “limited or nearly so to habitats associated with man.”
N
ATURAL HISTORY.— Hemidactylus brookii is largely a commensal species in Myanmar,
occurring on assorted constructs or on vegetation in the immediate vicinity of manmade structures.
Gravid females occur in central Myanmar samples from February and March.
Hemidactylus frenatus Duméril and Bibron Indo-Pacific house gecko
Hemidactylus javanicus Fitzinger 1826:46. Nomen nudum.
Hemi-dactylus frenatus Kuhl and Van Hasselt in Schlegel 1827:290. Nomen nudum.
Hemidactylus frenatus Duméril and Bibron 1836:366. Type locality, “l’Afrique australe, et . . . tout l’archipel
des grandes Indes”; restricted to “Java” by Loveridge (1947:127).
Hemidactylus Bojeri Fitzinger 1843:106. Type locality, “Africa. Madagascar. Mauritus.”
Hemidactylus(Pnoëpus) javanicus (Cuv.) Fitzinger 1843:106. Type locality, “Asia. India. Bengala. Ceylon.
Java. Timor. Amboina. Ins. Mariannae.”; restricted to “Java” by Loveridge (1947:127).
Hemidactylus vittatus Gray 1845:155. Type locality, “Borneo.”
Hemidactylus punctatus Jerdon 1853:467. Type locality, “Tellicherry” [= Thalassery, Kerala State, India].
Hemidactylus inornatus Hallowell 1861:469. Type locality, “Loo-Choo” [= Ryukyu Ids.].
Hemidactylus pumilus Hallowell 1861:502. Type locality, “Hong Kong.”
Gecko chaus Tytler 1865:547. Type locality, “Moulmein and Rangoon”; restricted to “Rangoon, Burma” by
Loveridge (1947:128).
Gecko caracal Tytler 1865:547. Type locality, “Rangoon” [= Yangon, Yangon Divis., Myanmar].
Hemidactylus longiceps Cope 1869:320. Type locality, “Manilla” [= Manila, Luzon Isl., Philippines].
Hemidactylus hexaspis Cope 1869:320. Type locality, “Madagascar.”
Peripia papuensis Macleay 1878b:97. Type locality, “Katow” [= Katau, Western Dist., Papua New Guinea].
392 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Fourth Series, Volume 58, No. 19
Hemidactylus tristis Sauvage 1879:49. Type locality, “le nord de la Nouvelle-Guinée.”
Hemidactylus nigriventris Lidth de Jeude 1905:188. Type locality, “Sintang [Kalimantan Barat, Indonesia]”.
Hemidactylus fragilis Calabresi 1915:236. Type locality, “Bur Meldac” [Somalia].
Hemidactylus vandermeer-mohri Brongersma 1928:1. Type locality, “Pulu Berhala” [= Palau Berhala – there
are two].
Hemidactylus okinawensis Okada 1936:271. Type locality, “Okinawa-jima” [Ryukyu Ids.].
Hemidactylus auritus Poeppig in Obst 1977:182. Nomen nudum.
DESCRIPTION.— Adults 42–59 mm SVL; adult SVL sexually dimorphic: females average
45.7±1.98 mm SVL (42.5–49.1 mm); male average 51.1±3.12 mm SVL (47.8–58.6 mm). Moder-
ately built, somewhat flattened lizards. Proportions (female – mean, min.-max.; male – same):
SnForel/SVL 36, 33–39%; 38, 35–40%; TrunkL/SVL 46, 39–53%; 43, 40–48%; ForeaL/SVL 12,
10–13%; 12, 10–13%; CrusL/SVL 13, 12–15%; 13, 12–14%. Tail length (unregenerated) slightly
longer than SVL and proportionally equal in females and males. Tail round to oblong (broader than
high) in cross-section, and tapering gradually to thin tip.
Head moderately large, distinct from neck, flattened and elongate conical (blunt triangular) in
dorsal outline, yielding a pointed-snout appearance. HeadL/SVL 25, 24–28%; 25, 25–26%; JawW/
HeadL 65, 61–67%; 67, 61–72%; SnEye/HeadL 43, 41–45%; 44, 41–47%; SnW/HeadL12,
12–16%; 13, 12–14%. All digits of fore- and hindfeet with digital pads. Digital pads oblong in
shape with distal end only slightly wider than proximal end; slender terminal portion (ultimate and
penultimate phalanges) of digit arising from within and free of digital pad of all digits. Claws on all
digits; slight or no webbing at base of digits.
Head, body, and tail scalation of small, equal-sized, juxtaposed tubercles dorsally and lateral-
ly with several longitudinal rows (about 6) of widely spaced enlarged tubercles (flatten cones, usu-
ally unkeeled); ventrally, slightly overlapping scales, 4–5× dorsal tubercles, from base of neck to
pelvic area; transition from ventral scales to tubercles ventrolaterally; no ventrolateral skin fold on
trunk. Head dorsally and laterally with scales on snout and lips, elsewhere small tubercles. Rostral
scale large, rectangular with middorsal groove; nares bordered by rostral, first supralabial, and
3 nasal scales, supranasal largest. Supralabials 9–12, infralabials 7–10. Triangular mental scale ven-
trally, bordered posteriorly by 2 large anterior chin scales, in contact with one another midventral-
ly; posterior chin scales from subequal to about half size of anterior ones and not touching medial-
ly.
Limbs scaled above and below, except for tubercles on posterior surface of fore- and hindlimbs.
Subdigital lamellae on pad: 7–9 on 4
th
finger, distal lamella undivided, subsequent 5–7 divided;
8–11 on 4
th
toe, distal lamella undivided, subsequent 5–8 divided. Bilateral series of 23–34 (total,
usually ≥29) precloacal-femoral pores in males, left and right sides separated at midpelvic by 0–2
nonpore scales. Tail midventrally with rectangular, slightly overlapping, smooth-surfaced plates
from 8–12 scale rows behind vent to at least mid-length; 2–3 rows of large smooth scales ventro-
laterally grading into tubercles. Tail distinctly segmented, each segment 10–12 scales long; each
segment with 6 (usually) enlarged flattened cone-shaped scales projecting above surface; project-
ing scales inset about 2 small tubercle rows from rear edge of segment and ventrolateral pair largest
in each spiral.
Adults in dark phase, dorsal ground color from head to base of tail dusky brown dorsally,
lighter brown stripe from snout through eye above ear to trunk fading thereafter, and dark brown
stripe from snout on upper lip through shoulder and distinct to hindlimbs. Preceding bold pattern
uncommonly seen; often only marking is faded lateral stripe from snout to shoulder. In light phase,
dorsum uniform whitish gray without markings. Ventrally whitish to light beige in all color phases
from chin onto underside of tail.
ZUG ET AL.: TAXONOMIC NOTES ON TROPICAL ASIAN HEMIDACTYLUS 393
DISTRIBUTION.— Hemidactylus frenatus occurs broadly
throughout Myanmar (Fig. 3) from north-central Sagaing
Division and central Kachin State southward to the coast of
Tanintharyi, Mon, Yangon, Ayeyarawady, and Rakhine. It is
a common house gecko from India eastward through tropical
Asia and Indoaustralia to the central Pacific. Sharma (2002:
map 29) excluded H. frenatus from India north of 24°N.
N
ATURAL HISTORY.— Our surveys reveal that Hemi-
dactylus frenatus is always associated with man-made struc-
tures. Gravid females occur in central Myanmar samples
from May and July.
Hemidactylus garnotii Duméril and Bibron
Fox gecko
Hemidactylus peruvianus Wiegmann 1835:240. Type locality,
“Peru, bei Tacna”; probably in error, see taxonomic comments
in Bauer (1994).
Hemidactylus Garnotii Duméril and Bibron 1836:368. Type locali-
ty, “l’île de Taíti” [= Tahiti, French Polynesia].
Doryura vulpecula Girard 1857:197. Type locality, “Sandwich
Islands” [= Hawaiian Islands].
Hemidactylus Ludekingii Bleeker 1859a:27. Type locality, “Agam,
Padangsche bovenlande” [= Agam, Sumatra].
D.[oryura] gaudama Theobald 1868:30. Type locality, “Tonghu
(valle Sittangensi).” [= Taungoo (Sittaung R. valley), Bago
Divis., Myanmar].
H.[emidactylus] Mortoni Theobald 1868:32. Type locality, “Teik-
gyie” [= Taikkyi, Yangon Divis., Myanmar].
Hemidactylus (Doryura) mandellianus Stoliczka 1872:101. Type
locality, “Pankabari [Sikkim, India], just above the Sikkim Terai, and . . . the Rungnu and Tístá valleys”.
Hemidactylus blanfordii Boulenger 1885:141. Type locality, “Himalayas.” “Darjeeling [West Bengal State,
India].”
DESCRIPTION.— Adults 49–66 mm SVL, average 56.5±5.13 mm; female only species. Moder-
ately built, somewhat flattened lizards. Proportions (mean, min.-max.): SnForel/SVL 39, 37–43%;
TrunkL/SVL 43, 39–46%; ForeaL/SVL 11, 10–13%; CrusL/SVL 13, 11–14%. Tail length (unregen-
erated) distinctly longer than SVL, about 1.2×. Tail flattened, broader than high (oblong in cross-
section), and tapering gradually to thin tip.
Head moderately large and broad, distinct from neck, flattened and elongate triangular and
snout truncate in dorsal outline. HeadL/SVL 25, 24–26%; JawW/HeadL 63, 60–72%; SnEye/
HeadL 46, 40–47%; SnW/HeadL 12, 10–14%. All digits of fore- and hindfeet with digital pads.
Digital pads oblong in shape with distal end only slightly wider than proximal end; slender termi-
nal portion (ultimate and penultimate phalanges) of digit arising from within and free of digital pad
of all digits. Claws on all digits; slight or no webbing at base of digits.
Head, body, and tail scalation of small, equal-sized, juxtaposed tubercles dorsally and lateral-
ly; slightly overlapping scales, >5× dorsal tubercles, ventrally from base of neck to pelvic area;
transition from ventral scales to tubercles ventrolaterally; no ventrolateral skin fold on trunk. Head
dorsally and laterally with scales on snout and lips, elsewhere small tubercles. Rostral scale large,
rectangular with middorsal cleft; nares bordered by rostral, first supralabial, and 3 nasal scales,
394 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Fourth Series, Volume 58, No. 19
FIGURE 3. Distribution of Hemidactylus
frenatus in Myanmar based on voucher spec-
imens of the CAS/NWCD/SI Myanmar Her-
petological Survey.
supranasal largest. Supralabials 8–13, infralabials 7–10. Triangular mental scale ventrally, bordered
posteriorly by 2 moderately large anterior chin scales, in contact midventrally; posterior chin scales
two-thirds or less size of anterior ones, rarely touching medially, and regularly separated from
infralabials by row of smaller gular scales.
Limbs with moderate-size tubercles above and smooth scale below. Subdigital lamellae on pad
extending to base of digit and slightly on to palm/sole: 8–13 on 4
th
finger, distalmost lamella undi-
vided, subsequent 5–8 divided; 10–15 on 4
th
toe, distalmost lamella undivided, subsequent 5–8
divided. No precloacal-femoral pores. Tail midventrally with rectangular, slightly overlapping,
smooth-surfaced plates from vent to at least mid-length; large smooth scales ventrolaterally quick-
ly reducing in size to small elongate conical scales on ventrolateral edges. Tail distinctly segment-
ed, each segment 8–10 scales long; ventrolaterally sawblade-like with four conical “spine” scales
in each segment, posteriormost one at edge of each segment
double the size of preceding three.
Adults in dark phase, dorsal ground color from head to
base of tail grayish or yellowish tan to medium brown with
five longitudinal rows (middorsal, dorsolateral, and lateral)
of whitish spots from nape/neck to rump; on lighter back-
ground spots in diffuse brown longitudinal stripes. Only
middorsal spots on tail, 2–3× larger than on trunk and often
dark edged anteriorly. Most individuals, light or dark phase,
with dark-brown postorbital stripe from eye to above ear-
opening; this stripe occasionally in front of eye to snout
and/or continuous with white spot-bearing lateral stripe. In
light phase, dorsum nearly uniform beige or with faded
markings on light background. Ventrally immaculate cream
to light yellow from chin onto underside of tail.
D
ISTRIBUTION.— Hemidactylus garnotii occurs widely,
although nowhere abundantly, throughout Myanmar (Fig. 4),
largely a low elevation distribution from southern Sagaing
Division and central Kachin State southward to the Martaban
coast from Rakhine to Mon States.
Its extralimital distribution is broad although spotty with
populations established in the Bahamas and southern Florida
(USA), throughout the Pacific, and Islands Asia, East Asia
westward to Pakistan, and southward to the Seychelles.
Tikader and Sharma (1992: map 8) showed its presence in
India restricted to Sikkim eastward into central Assam. The
recognition of two karyotypically different species (H. viet-
namensis, Vietnam; H. stejnegeri, Taiwan, Ryukyu Islands
and likely all of the Philippine Islands [Ota and Hikida, 1989]) suggests a more careful assessment
of individual Asian populations before a broad-brush assignment to H. garnotii.
N
ATURAL HISTORY.— The few H. garnotii captured were either on buildings or immediately
adjacent to human constructs. Gravid females occur in central Myanmar samples taken only in Feb-
ruary.
ZUG ET AL.: TAXONOMIC NOTES ON TROPICAL ASIAN HEMIDACTYLUS 395
FIGURE 4. Distribution of Hemidactylus
garnotii in Myanmar based on voucher spec-
imens of the CAS/NWCD/SI Myanmar Her-
petological Survey
Hemidactylus karenorum (Theobald) Burmese spotted gecko
D.[oryura] Karenorum Theobald 1868:30. Type locality, “Karen-choung, prope Toungoo, valle Sittangensi” [=
Karenchaung, above Taungoo, Sittang R. valley, Bago Divis., Myanmar].
DESCRIPTION.— Adults 38–56 mm SVL; adult SVL not sexually dimorphic: females average
50.2±3.56 mm SVL (45.7–56.1 mm); male average 47.2±5.68 mm SVL (38.3–55.8 mm). Moder-
ately built, somewhat flattened lizards. Proportions (female – mean, min.-max.; male – same):
SnForel/SVL 38, 36–42%; 38, 36–41%; TrunkL/SVL 41, 38–43%; 42, 39–48%; ForeaL/SVL 12,
11–13%; 12, 11–14%; CrusL/SVL 13, 12–14%; 14, 13–15%. Tail length (unregenerated) approxi-
mately equal to SVL in both females and males. Tail flattened, broader than high (semicircular to
spindle-shape in cross-section), and tapering gradually to thin tip.
Head moderately large and broad, distinct from neck, flattened and conical to obovate in dor-
sal outline. HeadL/SVL 25, 24–26%; 26, 25–28%; JawW/HeadL 65, 62–70%; 63, 58–71%; SnEye/
HeadL 43, 42–45%; 43, 42–46%; SnW/HeadL 13, 12–15%; 13, 11–15%. All digits of fore- and
hindfeet with digital pads. Digital pads oblong to obovate in shape with distal end usually wider
than proximal end; slender terminal portion (ultimate and penultimate phalanges) of digit arising
from within and free of digital pad of all digits. Claws on all digits; slight or no webbing at base of
digits.
Dorsal and lateral scalation of head and body of small, equal-sized, juxtaposed tubercles inter-
spersed with numerous smooth-surfaced cone-shaped tubercles (area of each about 3 smaller back-
ground tubercles); usually cone-shaped tubercles densely packed on body and without a longitudi-
nal arrangement into rows; dorsally fewer enlarged tubercles on tail and more longitudinally; ven-
tral scales slightly overlapping, about 5× smaller dorsal tubercles; transition from ventral scales to
tubercles ventrolaterally; no ventrolateral skin fold on trunk. Head dorsally and laterally with scales
on snout and lips, elsewhere small tubercles. Rostral scale large, rectangular with long median cleft,
about two-thirds height of rostral; nares bordered by rostral, first supralabial, and 3 nasal scales,
supranasal largest. Supralabials 10–12, infralabials 7–11. Broad triangular mental scale ventrally,
bordered posteriorly by 2 large anterior chin scales, in contact medially; posterior chin scales equal
size of anterior ones and not touching medially.
Limbs uniformly scaled above and below, except for small tubercles on posterior surface of
thigh. Subdigital lamellae on pad: 7–9 on 4
th
finger, distal lamella undivided, subsequent 4–7 divid-
ed; 8–10 on 4
th
toe, distal lamella undivided, subsequent 5–7 divided. Bilateral series of 26–38
(total and most >34) precloacal-femoral pores in males, left and right sides separated at midpelvic
by 1–3 nonpore scales. Tail midventrally with rectangular, slightly overlapping, smooth-surfaced
plates from vent to at least mid-length; large smooth scales ventrolaterally quickly grading into lat-
eral fringe scales. Tail distinctly segmented, each segment 8–10 scales long; dorsal scale small,
smooth and slightly overlapping; lateral edge of tail fringed with modestly enlarged triangular
scales, 3–4 per segment, usually posterior-most one of each segment larger than other, yielding a
ragged saw-like fringe.
Adults in dark phase, dorsal ground color from snout onto tail medium brown to grayish khaki;
head uniform to diffusely mottled in light and medium brown; dark irregular rectangular spots on
dorsum from nape to base of tail; usual spot pattern of dorsum, row of about six spots on midline
from nape onto tail, dorsolateral row of spots, and lateral row on trunk often fused into nearly con-
tinuous stripe. Most dark phase individuals with dark-brown postorbital stripe from snout through
eye and ear to axilla continuous with lateral trunk stripe. In light phase, dorsum nearly uniform
beige or with faded markings on light background. Ventrally immaculate cream to light yellow from
chin onto tail.
396 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Fourth Series, Volume 58, No. 19
DISTRIBUTION.— Our survey data suggest that Hemi-
dactylus karenorum may be a Burmese endemic (Fig. 5),
although it might reach the eastern mountains in Bangladash.
Our specimen-vouchers limit this species’ occurrence to the
central and north-central portion of the Ayeyarwady River
valley, even absent from the northern portion of the Sittaung
River drainage. We have not yet had the opportunity to
examine the specimen reported (Smith 1935) from Assam,
but the absence of this species from our survey sites in the
northern half of the Chindwin River basin suggests a
misidentification of the Assam specimen. Tikader and Shar-
ma (1992) listing of H. karenorum from Assam appears to be
a repeat of Smith’s (1935) earlier report. As yet, we are
unable to locate the specimen [“Cachar in Assam”] reported
by Smith. The BMNH has only four karenorum specimens,
and they represent two Burmese localities.
Natural history.— Hemidactylus karenorum is a forest
gecko, commonly found in leaf litter along streams or the
base of trees.
Gravid females were found in central Myanmar samples
taken in February and March.
Hemidactylus platyurus (Schneider)
Asian flat-tailed gecko
Stellio platyurus Schneider 1792:30[62]. Type locality, none given.
Lacerta Schneideriana Shaw 1802:278. Subsitute name for Stellio
platyurus Schneider.
G[ecko] marginata Cuvier 1829:55. Type locality, “Bengale.” [original description not seen]
Nycteridium schneideri Günther 1864:111. Subsitute name for Stellio platyurus Schneider.
Nycteridium Himalayanum Anderson 1871:15. Type locality, “Darjeeling; 3000 feet” [West Bengal State,
India].
Hemidactylus nepalensis Annandale 1907:151. Type locality, “Kathmandu, Nepal: altitude 4,500 feet.”
DESCRIPTION.— Adults 47–58 mm SVL; adult SVL and not sexually dimorphic in any traits:
females average 53.2±2.94 mm SVL (47.5–50.7 mm); male average 51.8±3.63 mm SVL (49.2–58.1
mm). Moderately built, somewhat flattened lizards. Proportions (female – mean, min.-max.; male
– same): SnForel/SVL 36, 34–38%; 36, 33–38%; TrunkL/SVL 46, 40–51%; 45 44–46%; ForeaL/
SVL 11, 11–12%; 12, 11–12%; CrusL/SVL 12, 11–14%; 13, 11–15%. Tail length (unregenerated)
subequal to or slightly longer than to SVL and equal in females and males. Tail strongly flattened,
much broader than high, and elongate dagger-like from base to pointed tip, lateral edges serrated
with flatten, pointed scales.
Head moderately large and broad, distinct from neck, flattened and broad-triangular to pentag-
onal in dorsal outline. HeadL/SVL 24, 23–25%; 24, 22–25%; JawW/HeadL 67, 63–74%; 68,
63–73%; SnEye/HeadL 42, 40–46%; 42, 40–47%; SnW/HeadL 15, 14–17%; 14, 14–16%. All dig-
its of fore- and hindfeet with digital pads. Digital pads obovate in shape with distal end slightly
wider than proximal end; slender terminal portion (ultimate and penultimate phalanges) of digit
arising from within and free of digital pad of all digits. Claws on all digits; modest webbing, about
one-third length of digits on fore- and hindfeet.
ZUG ET AL.: TAXONOMIC NOTES ON TROPICAL ASIAN HEMIDACTYLUS 397
FIGURE 5. Distribution of Hemidactylus
karenorum in Myanmar based on voucher
specimens of the CAS/NWCD/SI Myanmar
Herpetological Survey.
Head, body, and tail scalation of small, equal-sized, juxtaposed tubercles dorsally and lateral-
ly, including the trunk skin-fold; slightly overlapping smooth scales, >5× dorsal tubercles, ventral-
ly from base of neck to pelvic area; transition from ventral scales to tubercles ventrolaterally, 5–6
scale rows from base of skin-fold; distinct ventrolateral skin fold on trunk, posterior edge of thigh
and crus, and often on anterior edge of upper arm. Head dorsally and laterally with scales on snout
and lips, elsewhere small tubercles. Rostral scale large, rectangular with middorsal cleft; nares bor-
dered by rostral, first supralabial, and 3 equal-sized nasal scales. Supralabials 9–13, infralabials
8–11. Broad triangular mental scale ventrally, bordered posteriorly by 2 large anterior chin scales,
usually broadly in contact medially; posterior chin scales half or less the area of anterior ones and
broadly separated medially.
Limbs covered above and below with granular scales (about 2× dorsum scales), except for
smooth scales on posteroventral surface of thigh. Subdigital lamellae on pad: 7–9 on 4
th
finger, dis-
tal lamella undivided, subsequent 5–7 divided; 6–9 on 4
th
toe, distal lamella undivided, subsequent
5–7 divided. Bilateral series of 36–40 (total) precloacal-femoral pores in males, left and right sides
separated at midpelvic by 1–4 nonpore scales. Tail midventrally with rectangular, slightly overlap-
ping, smooth-surfaced plates from vent to near tip; large smooth scales ventrolaterally quickly grad-
ing into tubercles. Tail indistinctly segmented, evident only by enlarged lateral-edge spine at poste-
rior edge of each segment, segments 8–9 scales long; ventrolateral edge with series of spine-like
scales, last one largest on each segment.
Adults in dark phase, dorsal ground color from head to
base of tail medium brown to dusky tan with scattered small,
elongate dark-brown spots dorsally, occasionally dorsolater-
al spots coalesced into dorsolateral stripe from posterior cor-
ner of eye (rarely from tip of snout) to shoulder, laterally
broad dark-brown stripe from loreal area to inguen. Some
individuals with beige stripe between dark-brown stripes
from snout through eye to ear-opening. Dorsum of tail vari-
ably banded with dark and light brown, dark bands subequal
to twice width of light ones.
In light phase, dorsum nearly uniform beige, rarely with
diffused mottling on light background. Ventrally immaculate
cream to light yellow from chin onto tail.
C
OMMENT.— Our survey has collected a few individu-
als from southern Tanintharyi, two of which are males — a
subadult and an adult. Neither of these individuals displays
precloacal-femoral pores or the precursor pits. None of the
other traits are strikingly different from the central Myanmar
H. platyurus sample. This observation and those of
M. Smith’s (1935) suggest the potential of regional differen-
tiation. The inclusion of images of both H. craspedotus and
H. platyurus in the Cox et al. (1998) H. platyurus account
potentially exaggerates our interpretation of pattern variation
in the latter species.
D
ISTRIBUTION.— Even though Hemidactylus platyurus
occurs from about 26°N to the southern most tip of Myan-
mar (~10°N), our survey records (Fig. 6) show a very spotty
occurrence and, at least for central and northern Myanmar,
398 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Fourth Series, Volume 58, No. 19
FIGURE 6. Distribution of Hemidactylus
platyurus in Myanmar based on voucher
specimens of the CAS/NWCD/SI Myanmar
Herpetological Survey.
no populations in the Ayeyarwady valley or in the foothills and mountains to the east. Tikader and
Sharma (1992) also reported a limited distribution (Sikkim, Darjeeling District) in India; further-
more, their statement of its commensal habits suggests a human introduction into the Darjeeling
area. Ulber and Ulber (1991) provided a general distribution map of H. platyurus, showing a broad
distribution from Sri Lanka and eastern India through South Asia to the Philippines and New
Guinea. They further noted that this species occurs exclusively in the vicinity of human habitation.
Our natural history note below offers a slight contradiction to this exclusivity; however, its high
human commensalism likely explains its broad distribution via accidental transport.
N
ATURAL HISTORY.— Most individuals were captured in or on buildings, otherwise two indi-
viduals from a sandstone wall of a stream-cut and one from the forest floor.
Gravid females were found in central Myanmar samples taken in June and July, although other
adults from these samples were only in early to mid-vitellogenesis.
TAXONOMIC NOTES ON TROPICAL ASIAN HEMIDACTYLUS
The results of Carranza and Arnold (2006;
abbrev. C&A-06) encouraged the current inves-
tigation of Burmese Hemidactylus. The C&A-
06 molecular (mtDNA) phylogeny demonstrat-
ed that Hemidactylus is monophyletic when
Cosymbotus platyurus is returned to Hemi-
dactylus as proposed by Boulenger (1885).
Additionally, their analysis revealed five
clades, each of which represents a distinct geo-
graphic and evolutionary arena. Their tropical
Asian clade ((platyurus (bowringii (karenorum,
garnotii)) (flavivirdis (brookii, frenatus)))
(Fig. 7) contains seven of the approximately 20
species know to occur naturally from the Indus
River eastward through tropical Asia. We
assume for the following discussion that all Asian Hemidactylus are members of the C&A-06’s
tropical Asian clade. We have not attempted a morphological comparison to locate a synapomor-
phic trait that supports this assumption. Such a comparison would be useful, and critically, a broad-
er DNA sampling of tropical Asian Hemidactylus is essential to test the monophyly of this putative
Asian clade.
The monophyly of tropical Asian Hemidactylus creates a taxonomic unit that contains about
three times as many species (Table 1) as studied by C&A-06. We share C&A-06’s unwillingness
[implied] to assign formal taxonomic names to this or any of the other identified clades. Formal
name assignment of generic or subgeneric names for each of the clades would be easy owing to the
surfeit of synonyms for the genus Hemidactylus. A partitioning of the genus, however, without a
further testing of the monophyly of the five clades is premature and will only obfuscate the affini-
ties of Hemidactylus geckos, particularly when less than a quarter of the known species were
included in C&A-06’s phylogenetic analysis.
Our only exception to a change in generic assignment is reverting to Boulenger’s 1885 usage
of Hemidactylus platyurus. We believe this usage is now necessary in any discussion of Asian
Hemidactylus. Shifting platyurus to Hemidactylus restores the combination H. craspedotus Moc-
quard, 1890. This latter taxon became Mimetozoon craspedotus when de Rooij (1915) assigned the
ZUG ET AL.: TAXONOMIC NOTES ON TROPICAL ASIAN HEMIDACTYLUS 399
FIGURE 7. Stylized dendrogram displaying the proposed
relationships among members of the tropical Asian clade of
Hemidactylus. The dendrogram derives from the maximum
likelihood tree of Carrenza and Arnold (2006: fig. 1).
type species Mimetozoon floweri Boulenger, 1897 to the synonymy H. craspedotus and retained
Mimetozoon for this gecko with large bilateral body folds and well-webbed digits. She retained the
combination Hemidactylus platyurus. Subsequently, Smith (1935) synonymized Mimetozoon
Boulenger, Nycteridium Günther, 1864, and Cosymbotus Fitzinger, 1843 with Platyurus Oken,
1836 (see Taylor, 1963, for additional history). Thereafter, these two species regularly shifted
between the latter two genera and only with Wermuth’s 1965 checklist attained stability in assign-
ment to Cosymbotus.
Nowhere in this history of generic reassignments did anyone examine the affinities of craspe-
dotus and platyurus to one another or to other Hemidactylus species. The body folds and digital
webbing kept these two taxa closely linked for the last 70 years. C&A-2006’s study does not
address the relationship of these two species, but it does eliminate the implied relationships of
shared generic assignment outside of the genus Hemidactylus. It is such implied relationship that
argues against the use of formal group names for the various Hemidactylus geographic clades.
As previously noted, the C&A-2006 phylogram (Fig. 7) addresses the relationships of only a
400 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Fourth Series, Volume 58, No. 19
TABLE 1. Putative species of the tropical Asian clade of Hemidactylus. Species
names are arranged chronological. Only those taxa currently recognized as valid by
Kluge (2001) and Bauer (1994) are listed. Type localities are in brackets and general-
ized; species occurring in Burma in bold.
1792 Stellio platyurus Schneider [not stated]
1802 Gecko triedrus Daudin [not stated]
1835
1
Hemidactylus flavivirdis Rüppell [Massaua Isl. Eritrea]
1836 Hemidactylus frenatus Duméril & Bibron
2
[Java]
1836 Hemidactylus garnotii Duméril & Bibron [Tahiti]
1836 Hemidactylus leschenaultii Duméril & Bibron [Sri Lanka]
1836 Hemidactylus maculatus Duméril & Bibron [Bombay Pres., India]
1842 Hemidactylus depressus Gray [Madagascar]
1845 Doryura bowringii Gray [Hong Kong]
1845 Hemidactylus brookii Gray [Borneo]
1854 Hemidactylus subtriedus
3
Jerdon [Madras Pres., India]
1861 Hemidactylus marmoratus
4
Hallowell [Ryukyu Ids.]
1868 Doryura karenorum Theobald [Myanmar]
1870 Hemidactylus reticulatus Beddome [eastern Karnataka, India]
1870 Hemidactylus gracilis Blanford [Central Prov., India]
1871 Hemidactylus giganteus Stoliczka [Godavari Valley, India]
1875 Gecko anamallensis Günther [Western Ghats, India]
1890 Hemidactylus craspedotus Mocquard [Borneo]
1906 Tetratolepis scabriceps Annandale [southern Tamil Nadu, India]
1935 Hemidactylus prashadi M.A. Smith [Bombay Pres., India]
1981 Hemidactylus porbandarensis Sharma [Gujarat, India]
1984 Hemidactylus mahendrai Shukla [Uttar Pradesh, India]
1984 Hemidactylus vietnamensis Darevsky & Kupriyanova [northern Vietnam]
1989 Hemidactylus stejnegeri Ota & Hikida [Taiwan]
1
Kluge gives dates as 1840, presumably because the latter sections of this publication appeared
over several years.
2
Kluge credits authorship to Schlegel 1836.
3
Smith (1935) considered H. subtriedus questionably distinct from H. triedrus. Its absence from
Das’ (2002) pocket guide indicates a similar opinion.
4
Kluge recognized H. marmoratus as a valid taxon, yet Ota (1989) does not include it as a mem-
ber of the Ryukyu gecko fauna. Smith (1935) synonymized it with H. leschenaultii.
third of the members of the putative tropical Asian clade. The phylogram indicates two lineages or
clades (bowringii-garnotii-karenorum-platyurus and brookii-flaviviridis-frenatus). [Hereafter, we
use complex for the bowringii and brookii clades as a phylogenetically neutral label.] Two ques-
tions immediately derive from this proposed relationship: 1) Do these two complexes encompass
all the tropical Asian Hemidactylus or, as additional Asian taxon are examined, will new clades
emerge and/or change the composition of current ones? 2) Do the members of each of these clades
display a set of morphological traits that permit the visual differentiation of the complexes and
assignment of molecularly untested Asian species to one or the other of the complexes? We cannot
address the first question as a solution requires a broader molecular sampling of Asian species. This
question, however, draws attention to a serious flaw of the C&A-06 study, i.e., inadequate speci-
men vouchering of their molecular samples. Only about a quarter of the C&A-06 DNA-samples
have voucher specimens on which species identification can be re-examined and confirmed. An
example of the problems arising from unvouchered samples is the identity of their Thiruvanan-Indi-
an “Hemidactylus frenatus.” It is likely not a H. frenatus as currently conceived throughout the ‘fre-
natus’ broad invasive-distribution. The specimens from recently invaded areas (Colombia and
Hawaii) are genetically identical to one another and paired with Burmese specimens (C&A-06:
fig. 1; our Fig. 7). The Indian H. frenatus is genetically distinct from the Burmese-Invasive lineage,
thereby hinting that Hemidactylus punctatus Jerdon, 1854 may be a valid taxon, because the C&A-
06 sample derived from Thiruvananthapuram, Kerala, about 350 km south of the type locality (Tel-
licherry) of Jerdon’s species. This possibility cannot be examined further without a voucher speci-
men for the molecular data.
Returning to the second question, a superficial morphological survey of C&A-06’s tropical
Asian species identified a set of traits (trunk scalation, tail ornamentation, digital lamellae, and pre-
cloacal-femoral pores) that generally delimits the two Asian complexes. The bowringii complex
either lack ornamentation on the tail or the spines are confined to the ventrolateral margin as a sin-
gle scale-spine or a fringe of unequal sized scale-spines on each side of each caudal segment. Tail
ornamentation in the brookii complex consists of a circumferential row or rows of scale-spines on
each segment. The number of rows, numbers of spines, and relative size of the spines vary between
the species, with H. brookii having the most spinose tail. Trunk scalation (back and sides) consists
of uniformly small juxtaposed (granular) scales in the bowringii complex, except for H. karenorum
with its irregular longitudinal rows of small tubercles. The brookii complex members have numer-
ous longitudinal rows of variously developed tubercles on the trunk, except for H. flaviviridis,
which lacks tubercles. A third to half of the digital lamellae on each digit of bowringii complex are
divided, in contrast to only the terminal lamellae of the brookii complex are undivided. Precloacal-
femoral pores occur only in males and are typically continuous across the hindlimb-pelvic junction.
Members of the bowringii complex regularly possess a total of 26 pores or more, whereas brookii
members usually have a total of 18 or fewer, except for H. frenatus commonly ≥ 24 pores. With
more pores on each side, bowringii members have fewer (0–3) nonpore scales separating the left
and right rows of pore-scales, and brookii members ≥ 4 nonpore scales, except for H. frenatus
(0–1).
These traits allow a hypothetical assignment of the molecularly untested tropical Asian Hemi-
dactylus to the two complexes: bowringii complex — anamallensis, craspedotus, depressus, gigan-
teus,
leschenaultii, stejnegeri, and vietnamensis; brookii complex — maculatus, prashadii, reticu-
latus, and triedrus. Several taxa are not assigned: 1) marmoratus and subtriedus are not considered
valid taxa (see footnotes of Table 1); 2) gracilis has 10–12 longitudinal rows of well-developed
tubercles on the trunk and a tail oblong in cross-section without ornamentation, sharing a major trait
of each complex, but more critically only precloacal pores [a trait shared only with H. porbandaren-
ZUG ET AL.: TAXONOMIC NOTES ON TROPICAL ASIAN HEMIDACTYLUS 401
sis among tropical Asian Hemidactylus, although approached in some brookii complex members];
3) porbandarensis [possibly a brookii complex member] has 16–17 longitudinal rows of enlarged,
keeled tubercles on the trunk dorsum, moderately compressed tail, presumably with transverse rows
of six spines at edge of each segment, and a short row of precloacal pores; 4) Annandale’s descrip-
tion of scabriceps reported dorsal trunk scales imbricate and equal in size to the belly scales [a trait
not known elsewhere in tropical Asian Hemidactylus]. Shukla’s description (1984) of H. mahendrai
is based on a few juvenile and female specimens. Further, the description lacks a statement differ-
entiating H. mahendrai from H. brookii, although diagnostic traits are provided for other Indian
Hemidactylus species. Shukla’s description match the traits of female Asian H. brookii, thus we ten-
tatively suggest that mahendrai is a synonym of the latter species. The uncertainty of assignment
for the latter two taxa urges their inclusion in the next molecular phylogenetic analysis.
The preceding taxon assignment derived mainly from the states of trunk scalation, tail orna-
mentation, and precloacal-femoral pore number. Hemidactylus leschenaultii has some enlarged dor-
sal tubercles, but tail ornamentation and precloacal-femoral pores are strongly bowringii complex.
Hemidactylus depressus similarly has numerous rows of dorsal tubercles and otherwise has
bowringii complex traits. We note that H. karenorum also has numerous tubercles, although small-
er than the brookii complex condition, smooth and rounded, and not arranged in longitudinal rows.
Ignoring the high number of pores in H. depressus, bowringii complex assignments were unequiv-
ocal.
Hemidactylus stejnegeri and H. vietnamensis are all-female species and without karyotypic
data would have remained unrecognized as distinct genetic lineages from H. garnotii. Hemidacty-
lus vietnamensis was recognized by Darevsky and colleagues (1984) when they obtained kary-
otypes from northern Vietnam H. garnotii. Contrary to Kluge and Eckardt’s karyotypic results
(3n = 70; Kluge and Eckardt 1969) for Floridian and Hawaiian H. garnotii, the karyotype of the
Vietnamese “garnotii” specimens was 3n = 60. Darevsky et al. (1984: table 2) contrasted seven sca-
lation traits of their Vietnamese specimens (n = 17) with a subset (n = 28) of Kluge and Eckardt’s
Hawaiian sample (n = 66) and found significant differences in 6 traits. The scalation and karyotyp-
ic differences led them to recognize the Vietnamese population as a distinct species. These scala-
tion differences are slight and might result from differences in trait definition and data-collection;
however, Ota et al. (1986) examined a set of 7 traits in the Kluge-Eckardt Hawaiian, Darevsky et
al. Vietnamese, and 3 Chinese (Yunnan, Hainan, Taiwan) samples. This analysis confirmed the dis-
tinctiveness of the Hawaiian and Vietnamese samples from one another. The Hainan and Yunnan
samples were not significantly different from one another, but combined they differed modestly
from the Hawaiian and Taiwan samples, and strongly differentiated from the Vietnamese one.
Subsequently, Ota and Hikida (1989b) examined the karyotypes of Taiwanese “garnotii” and
discovered another karyotype (3n = 56) and recognized this population as H. stejnegeri. In a series
of research articles, Ota and collaborators examined karyotypic evolution and the origin of
parthenogenesis in unisexual Hemidactylus. Their general conclusion is that there are two partheno-
genetic groups: the H. garnotii-vietnamensis complex and H. stejnegeri. Both groups arose by
hybridization, and presently, evidence identifying the parental species is not available. They also
discussed the likelihood of karyotypic evolution without concomitant morphological differentiation
within a unisexual species. We lack data and expertise to support or negate their hypothesis of rela-
tionships and accept their assessment of two clonal lineages. The C&A06 molecular data show
genetic uniformity within Burmese H. garnotii and between Burmese and Floridian specimens.
Moritz et al. (1993) showed a uniformity of H. garnotii karyotype number (3n = 63) among Pacif-
ic island populations, including Hawaii, and the Florida population. They also noted a low diversi-
ty of allozymes and mtDNA among their samples. Combing there results with those of C&A-06
402 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Fourth Series, Volume 58, No. 19
indicates that the Burmese populations is H. garnotii. Nevertheless until there is a molecular analy-
sis among the various Asian populations of the parthenogens, we urge caution on assigning the
name garnotii.
ACKNOWLEDGMENTS
The Myanmar Herpetological Survey is a country-wide species inventory designed to uncover
the full diversity of the Burmese herpetofauna by surveying broadly but focusing on the forest and
wildlife reserves of the Myanmar Forestry Department, Ministry of Forestry. The Survey owes it
success to the administrative and logistic support of U Shwe Kyaw, Director General, Forestry
Department, and U Khin Maung Zaw, Director, and U Tin Tun, Deputy Director, of the Nature and
Wildlife Conservation Division (NWCD, Forestry Dept.), and critically to the NWCD staff (U Htun
Win, Daw Thin Thin, U Kyi Soe Lwin, U Awan Khwi Shein, U San Lwin Oo, U Sai Wunna Kyi)
who have been the core of our survey team. The Survey is a collaborative effort among the staff of
NWCD, the California Academy of Sciences, and the Smithsonian’s National Museum of Natural
History (SI-NMNH). The latter two institutions have provided a variety of administrative and finan-
cial support. We thank all the above individuals and institutions for their past and ongoing support.
Financial support for the Survey and assorted inventory-monitoring programs derived from the
National Science Foundation (DEB-9971861: J.B. Slowinski and G.R. Zug) as the primary source,
CAS donations, and the SI-NMNH’s Biodiversity and Survey Program. The present research
received support from NSF (DEB-0451832: A. Leviton, J. Vindum, and G. Zug).
We thank the collections management staff of our respective museum for their assistance with
the examination of specimens. Robert F. Fisher and Aaron Bauer reviewed the manuscript; we
thank them for their efforts on our behalf. Caleb McMahan provided the morphological data for the
H. bowringii account. We thank Patricia Zug for data input and assistance with data collection.
LITERATURE CITED
N.B. We have not listed citations for authors used solely as part of a
scientific name to denote authorship and date of publication.
A
NNANDALE, N. 1906. Notes on the fauna of a desert tract in southern India. Part I.—Batrachians and reptiles,
with remarks on the reptiles of the desert region of the North-West Frontier. Memoirs of the Asiatic Socie-
ty of Bengal 1(10):183–202.
A
UFFENBERG, W. 1980. The herpetofauna of Komodo, with notes on adjacent areas. Bulletin of the Florida
State Museum, Biological Sciences 25:39–156.
BAUER, A.M. 1994. Familia Gekkonidae (Reptilia, Sauria). Part I Australia and Oceania. Das Tierreich
109:1–306.
B
AUER, A.M., AND R. GÜNTHER. 1992. A preliminary report on the reptile fauna of the Kingdom of Bhutan with
the description of a new species of scincid lizard (Reptilia: Scincidae). Asiatic Herpetological Research
4:23–36.
BOBROV, V.V. 1992. Notes on lizards (Reptilia, Sauria) from southern Vietnam. Journal of Bengal Natural His-
tory Society n.s. 11:17–24.
B
OULENGER, G.A. 1885. Catalogue of the Lizards in the British Museum (NaturalHistory), Second edition..
Volume 1. Geckonidae, Eublepharidae, Uroplatidae, Pygopodidae, Agamidae. Taylor and Francis, London,
UK. xii + 436, 32 plates.
B
OURRET, R. 1939. XVII. Reptiles et batraciens reçus au Laboratoire des Sciences Naturelles de l’Université
au cours de l’année 1938. Descriptions de trois espèces nouvelle. Bulletin général de l’Instruction publique
[Hanoi] (6):13–34.
B
ROWN, W.C., AND A.C. ALCALA. 1978. Philippine lizards of the family Gekkonidae. Silliman University Nat-
ural Science Monograph Series 1:1–146.
ZUG ET AL.: TAXONOMIC NOTES ON TROPICAL ASIAN HEMIDACTYLUS 403
CARRANZA, S, AND E.N. ARNOLD. 2006. Systematics, biogeography, and evolution of Hemidactylus geckos
(Reptilia: Gekkonidae) elucidated using mitochondrial DNA sequences. Molecular Phylogenetics and Evo-
lution 38:531–545.
C
OX, M.J., P.P. VA N DIJK, J. NABHITABHATA, AND K. KHIRAKHUPT. 1998. A Photograpahic Guide to Snakes and
Other Reptiles of Peninsular Malaysia, Singapore and Thailand. Ralph Curtis Books, Sanibel Island, Flori-
da, USA. 144 pp.
D
ANIEL, J.C. 1983. The Book of Indian Reptiles. Bombay Natural History Society, Bombay, India. x + 141 pp.,
52 unnumbered plates.
D
AREVSKY, I.S. 1964. Die Reptilien der Inseln Komodo, Padar un Rintja im Kleinen Sunda-Archipel, Indone-
sien. Senckbergiana biologica 45:563–576.
D
AREVSKY, I.S., L.A. KUPRIYANOVA, AND V. V. R OSHCHIN 1984. A new all-female triploid species of gecko and
kayological data on the bisexual Hemidactylus frenatus from Vietnam. Journal of Herpetology
18:277–284.
D
AS, I. 2002. A Photographic Guide to Snakes and Other Reptiles of India. New Holland Publishers (UK) Ltd.,
London, UK. 144 pp.
DAS, I., AND J. SUKUMARAN. 2006/7. A recent record of Hemidactylus brookii Gray, 1845 from Borneo. Gekko
5: in press.
D
ENZER, W., AND U. MANTHEY. 1991. A nominal checklist of the lizards inhabiting peninsular Malaysia and
Singapore. Raffles Bulletin of Zoology 39:309–322.
D
ERANIYAGALA, P.E.P. 1953. A Colored Atlas of Some Vertebrates from Ceylon. Volume Two. Tetrapod Reptil-
ia. Ceylon National Museums Publication -Ceylon Government Press, Colombo, Ceylon [Sri Lanka]. xvii
+ 101 pp., 35 pl.
DE ROOIJ, N. 1915. The Reptiles of the Indo-Australian Archipelago. I. Lacertilia, Chelonia, Emydosauria.
E.J. Brill, Ltd., Leiden, Netherlands. xiv + 384 pp.
D
UNN, E.R. 1927. Results of the Douglas Burden Expedition to the island of Komodo. III.— Lizards from the
East Indies. American Museum Novitates 288:1–13.
G
LEADOW, F. 1887. Description of a new lizard from the Dangs. Journal of the Bombay Natural History Soci-
ety 2:49–51.
K
ARSTEN, S.J., M. W-N. LAU, AND A. BOGADEK. 1986. Hong Kong Amphibians and Reptiles. Urban Council,
Hong Kong. 136 pp.
K
ÄSTLE, W. 2002. Family Gekkonidae (geckos). Pages 674–691, 697–722 in H.H. Schleich and W. Kästle,
eds., Amphibians and Reptiles of Nepal. Biology, Systematics, Field Guide. A.R. G. Gantner Verlag Kom-
manditgesellschaft, Ruggell, Germany.
K
HAN, M.S. 2002. Key and checklist to the lizards of Pakistan. Herpetozoa 15(3/4):99–119.
K
LUGE, A.G. 2001. Gekkotan lizard taxonomy. Hamadryad 26:1–209.
K
LUGE, A.G., AND M.J. ECKARDT. 1969. Hemidactylus garnotii Duméril and Bibron, a triploid all-female
species of gekkonid lizard. Copeia 1969:651–664.
L
AZELL, J. 2002. The herpetofauna of Shek Kwu Chau, South China Sea, with descriptions of two new colu-
brid snakes. Memoirs of the Hong Kong Natural History Society 25:1–82.
L
IAN, T.F. 1993. Checklist of lizards of Sabah. Sabah Parks Trustees, Kota Kinabalu. 18 pp.
M
ANTHEY, ULRICH, AND WOLFGANG GROSSMANN. 1997. Amphibien & Reptilien Südostasiens. Natur u. Tier
Verlag, Münster, Germany. 512 pp.
M
ORITZ, C., T.J. CASE, D.T. BOLGER, AND S. DONNELLAN. 1993. Genetic diversity and the history of pacific
[sic] island house geckos (Hemidactylus and Lepidodactylus). Biological Journal of the Linnean Society
48:113–133.
O
TA, H. 1989. A review of the geckos (Lacertilia: Reptilia) of the Ryukyu Archipelago and Taiwan. Pages
222–261 in M. Matsui, T. Hikida, and R.C. Goris, eds., Current Herpetology in East Asia. Herpetological
Society of Japan, Kyoto, Japan.
O
TA, H., I.S. DAREVSKY, L.A. KUPRIYANOVA, T. HIKIDA, K.-Y. LUE, S.-H. CHEN, AND T. HAYASHI. 1993. Geo-
graphic variation in the parthenogenetic lizard, Hemidactylus stejnegeri Ota & Hikida (Gekkonidae Rep-
tilia), with comments on recently discovered male phenotypes. Tropical Zoology 6:125–142.
O
TA, H., T. HIKIDA, AND E. ZHAO. 1986. Notes on members of the Hemidactylus garnotii-vietnamensis species
404 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Fourth Series, Volume 58, No. 19
complex from Hainan and Yunnan of China. Japanese Journal of Herpetology 11(3):79–85.
O
TA, H., AND T. HIKIDA. 1989a. A record of a triploid gecko, Hemidactylus stejnegeri from the northern Philip-
pines. Japanese Journal of Herpetology 13(2):35–39.
OTA, H., AND T. HIKIDA. 1989b. A new triploid Hemidactylus stejnegeri (Gekkonidae: Sauria) from Taiwan,
with comments on morphological and karyological variation in the H. garnotii-vietnamensis complex.
Journal of Herpetology 23:50–60.
PAWAR, S., AND A. BIRAND. 2001. A Survey of Amphibians, Reptiles, and Birds in Northeast India. CERC Tech-
nical Report #6. Center for Ecological Research and Conservation, Mysore, India. 180 pp.
S
HARMA, R.C. 2002. The Fauna of India and the Adjacent Countries. Reptilia, Volume II (Sauria). Zoological
Survey of India, Kolkata, India. xxv + 430 pp.
S
HUKLA, V.N. 1984. A new species of the lizard Hemidactylus from Kanpur (India). Indian Journal of Zooto-
my 24:81–83.
S
MITH, M. A. 1935. The Fauna of British India, Including Ceylon and Burma. Reptilia and Amphibia. Vol. II—
Sauria. Taylor & Francis Ltd., London, UK. xii + 445 pp.
TAYLOR, E. H. 1963. The lizards of Thailand. University of Kansas Science Bulletin 44(14):687–1077.
T
IKADER, B.K., AND R.C. SHARMA. 1992. Handbook Indian Lizards. Zoological Survey of India, Calcutta,
India. xv + 250 pp.
ULBER, T., AND E. ULBER. 1991. Cosymbotus platyurus (Schneider). Sauria, Supplement 13:201-204.
W
ERMUTH, H. 1965. Liste der rezenten Amphibien und Reptilien. Gekkonidae, Pygopodidae, Xantusiidae. Das
Tierreich 80:1-246.
ZHAO, E., AND K. ADLER. 1993. Herpetology of China. Society for Study of Amphibians and Reptiles, Ithaca,
New York, USA. 522 pp.
Z
HAO, E., ED. 2003. Coloured Atlas of Sichuan Reptiles. Sichuan Provincial Bureau of Forestry <<actual pub-
lisher is unclear>> 292 pp.
Z
UG, G.R. 1990. Lizards of Fiji: natural history and systematics. Bishop Museum Bulletin in Zoology 2:1–136.
ZUG ET AL.: TAXONOMIC NOTES ON TROPICAL ASIAN HEMIDACTYLUS 405
Copyright © 2007 by the California Academy of Sciences
San Francisco, California, U.S.A.
