New Forest Gecko (Squamata; Gekkonidae; Genus Luperosaurus) fromMt. Mantalingajan, Southern Palawan Island, Philippines

RAFE M. BROWN,1,2 ARVIN C. DIESMOS,3 MELIZAR V. DUYA,4 HARVEY J. D. GARCIA,5 AND

EDMUND LEO B. RICO6

1Natural History Museum and Biodiversity Institute, 1345 Jayhawk Boulevard, Dyche Hall, University of Kansas,

Lawrence, Kansas 66045; E-mail: rafe@ku.edu3Herpetology Section, Zoology Division, National Museum of the Philippines, Padre Burgos Avenue, Ermita 1000,

Manila, Philippines; E-mail: arvincdiesmos@i-manila.com.ph4Conservation International Philippines, Number 6 Maalalahangin Street, Teachers Village, Diliman 1101,

Quezon City, Philippines; E-mail: mvduya@conservation.org5Conservation International Philippines, Room 207 Pacific Plaza, Rizal Avenue, Puerto Princesa City,

Palawan 5300, Philippines; E-mail: hjdgarcia@yahoo.com6Wildlife Conservation Society of the Philippines, Diliman, Quezon City, Philippines; E-mail: edmundleorico@yahoo.com

ABSTRACT.—We describe a new species of Luperosaurus from Mt. Mantalingajan, southern Palawan Island,

Philippines. The new species is distinguished from all other species of Luperosaurus by the combination of

its large body size (81.3 mm for the single male specimen), near complete absence of interdigital webbing,

absence of cutaneous expansions on limbs except for a minute flap on the posterior margins of the hind

limbs, the presence of differentiated, moderately enlarged chin shields, 40 precloacofemoral pore-bearing

scales, the limitation of scattered flattened dorsal tubercles to only the posterior portions of the torso,

absence of spinose or recurved ornamental tubercles on the head and nuchal region, and convex to

posteriorly raised tubercles clustered at the posterior margins of caudal tail annuli. Because the new species

shares features with species in both species of Luperosaurus and Gekko, we compare the new species to (and

distinguish it from) both genera. The new species is distinguished from all Southeast Asian Gekko by the

combination of its smaller body size, relatively short, stout limbs, presence of only moderately enlarged,

slightly imbricate ventral body scales, differentiated postmentals not highly elongate, dorsal body tubercles

limited to posterior trunk and not arranged in rows, absence of enlarged, spinose tubercles on the limbs and

tail, and tail encircled by small scales (enlarged subcaudals absent). The new species further emphasizes the

biogeographic distinctiveness (from Sundaland fauna) and level of vertebrate endemism of Palawan Island

and underscores the degree to which the biodiversity of the Philippines is not fully understood.

Philippine gekkonid species include 40 spe-cies in nine genera (Taylor, 1922; Brown andAlcala, 1978): Cyrtodactylus (5), Gekko (11; Ota etal., 1989; Roesler et al., 2006; Brown et al., 2008),Gehyra (1), Hemidactylus (5), Hemiphyllodactylus(1), Lepidodactylus (6), Luperosaurus (6; Gaulke etal., 2007; Brown et al., 2007), Pseudogekko (4), andPtychozoon (1; Brown et al., 2009). Brown andAlcala (1978), Russell (1979), and Brown et al.(2000, 2007) commented on variation in thecharacters distinguishing the genera Lupero-saurus and Gekko, and recent studies havecommented on the phenotypic dissimilarityand possible nonmonophyly of the genusLuperosaurus (Brown and Diesmos, 2000; Brownet al., 2000, 2007).

Currently, we refer species to Luperosaurus onthe basis of (1) small to moderate body size; (2)limbs short and stout; (3) presence in mostspecies of at least some interdigital webbing; (4)

pronounced cutaneous expansions borderingthe edges of the limbs in most species; (5)enlarged, irregular, spinose, or recurved orna-mental tubercles on dorsolateral portions of thebody and tail; (6) small, cycloid, juxtaposedscales completely encircling the tail (enlargedsubcaudals absent); (7) chin shields moderate tosmall (differentiated postmentals absent orreduced); (8) body scales generally cycloid,juxtaposed, nonimbricate, and minimally differ-entiated between dorsal and ventral surfaces(Brown et al., 2000, 2007).

There are nine endemic species of PhilippineGekko (Gekko athymus, Gekko crombota, Gekkoernstkelleri, Gekko gigante, Gekko mindorensis,Gekko palawanensis, Gekko porosus, Gekko romblon,and Gekko n. sp.; Brown and Alcala, 1978; Brownet al., 2008, 2009), two species shared with otherSoutheast Asian countries (Gekko gecko andGekko monarchus; Taylor, 1922; Manthey andGrossman, 1997), and one species (Gekko ho-kouensis) represented by a single specimen ofdoubtful origin (Ota et al., 1989) that may have2 Corresponding Author.

Journal of Herpetology, Vol. 44, No. 1, pp. 37–48, 2010Copyright 2010 Society for the Study of Amphibians and Reptiles

been included in the Philippines’ gekkonidfauna in error (Taylor, 1962; Brown and Alcala,1978; Toda et al., 2008).

Numerous additional Philippine gekkonidsawait description, including a species repre-sented so far by only a single, highly distinctspecimen from the forests of the southernPalawan Island (Fig. 1). Although the newspecies has characteristics typical of both Luper-osaurus and Gekko, we assign it to the formergenus on the basis of the number of characterstates shared with other members of the genusLuperosaurus (Brown et al., 2000, 2007) andpronounced phenotypic dissimilarity to mem-bers of the genus Gekko.

Philippine members of the genus Luperosaurusinclude six endemic species: Luperosaurus cor-fieldi, Luperosaurus cumingii, Luperosaurus joloen-sis, Luperosaurus kubli, Luperosaurus macgregori,and Luperosaurus palawanensis (Brown et al.,2000, 2007; Gaulke et al., 2007). These robust-bodied taxa are phenotypically quite differentfrom the slender, elongate-bodied non-Philip-pine species Luperosaurus browni (PeninsularMalaysia and Borneo), Luperosaurus brooksii(Sumatra), Luperosaurus iskandari (Sulawesi;Brown et al., 2000), and small-bodied Borneantaxa characterized by ornamental scalation anddermal expansions bordering the limbs and tail(Luperosaurus yasumai and Luperosaurus sorok;Ota et al., 1996; Das et al., 2008). The latter twospecies possess distinctive, serrated, denticulate

lateral dermal expansions on either side of thetail (reminiscent of the genus Ptychozoon; Brownet al., 1997, 2000). Most Philippine speciesdemonstrate a tendency toward island orPleistocene Aggregate Island Complex ende-mism (Brown and Diesmos, 2000, 2002) al-though two species (L. kubli and L. cumingii)may co-occur on different parts of Luzon Island(Brown et al., 2007; Gaulke et al., 2007). Only onBorneo are multiple species of Luperosaurusactually known to coexist in sympatry (L.yasumai and L. browni); L. sorok may occur insympatry with these species as well.

Palawan Island (considered by mammalogiststo be a biogeographic extension of Borneo;Heaney 1985; Esselstyn et al., 2004) has a singleendemic species, L. palawanensis. The typespecimens (two mature males) were collectedfrom north-central Palawan in 1961 and de-scribed in 1978 (Brown and Alcala, 1978). Nofurther reports of this species have beenforthcoming in the intervening years, and thespecies bears some resemblance to the stout-bodied, ornate species from Borneo (L. yasumaiand L. sorok; Ota et al., 1996; Das et al., 2008) andthe southern Philippines (L. joloensis; Taylor,1918; Brown and Diesmos, 2000; Brown et al.,2000).

In 2007, two major biodiversity inventoryefforts (one conducted by Conservation Inter-national, Philippines, and another led by re-searchers at the University of Kansas and theNational Museum of the Philippines) targetedMt. Mantalingajan, southern Palawan—an areaonly surveyed for herpetofauna previously byE. H. Taylor in 1923 and by members of theField Museum Philippine Zoological Expeditionin 1947 (Hoogstral, 1951; Inger, 1954). Amongthe collections resulting from nearly four con-tinuous months of fieldwork is a single speci-men of a highly distinctive undescribed speciesof Luperosaurus that cannot possibly be confusedwith any other species of Luperosaurus or anySoutheast Asian species of Gekko. Consideringthe low probability of obtaining further speci-mens, and because ignorance of regional biodi-versity resources is associated with (or perhapsindirectly contributes to) destruction of forestedareas, we are compelled to describe the newspecies on the basis of the single highlydistinctive specimen.

MATERIALS AND METHODS

We scored data from fluid-preserved speci-mens (Appendix 1) deposited in several insti-tutional collections (Acknowledgments). Sexwas determined by presence of eggs in females,the collector’s ability to evert hemipenes ofmales, and gonadal inspection when possible.

FIG. 1. Map of Palawan Island in relation to thePhilippines (inset) with the type locality of Lupero-saurus gulat (Municipality of Rizal, Mt. Mantalingajan)indicated with a shaded star. The type locality ofLuperosaurus palawanensis (Municipality of PuertoPrincesa, Thumb Peak [local name 5 ‘‘Salakot’’]) isindicated by a shaded circle.

38 R. M. BROWN ET AL.

Prominent secondary sexual characteristics con-firmed (cloacal spurs and presence of precloa-cofemoral pores in males vs. pores absent infemales; Brown et al., 1997, 2000; Brown, 1999)the determination of sex when dissection wasnot possible. Measurements (taken to thenearest 0.1 mm) were obtained with digitalcalipers following character definitions byBrown et al. (1997) and Brown (1999). Measure-ments included snout–vent length, tail length,head length, head width, head depth, snoutlength, eye diameter, eye–narial distance, tym-panic annulus diameter, internarial distance,interorbital distance, axilla–groin distance, fe-mur length, tibia length, Toe I length, Toe IVlength, tail width, and tail height. Meristiccharacters included the number of differentiat-ed precloacofemoral pore-bearing scales, supra-labials, infralabials, circumorbitals, transversemidbody scales, paravertebrals and ventrals(both counted between midpoints of limbinsertions), number of tail annuli, and subcau-dals.

TAXONOMY

Luperosaurus gulat, sp. nov.Figures 2–4

Holotype.—PNM 9282 (Field Collection num-ber ELR 1579), adult male; 1300 m above sea level,3.4 km west, 0.60 km south of Mt. Mantalingajanpeak, Barangay Ransang, area known locally as‘‘Gunob,’’ Municipality of Rizal, Palawan Prov-ince, Palawan Island, Philippines (8u48948.70N;117u38959.10E) by Uldarico Dodong Carestia andH. J. D. Garcia, 5 July 2007.

Diagnosis.—Although the new species furtherblurs the distinction between the genera Luper-osaurus and Gekko (as defined by Brown et al.,2000, 2007), we refer it to Luperosaurus by virtueof (1) its possession of a robust body and stoutlimbs; (2) vestiges of interdigital webbingbetween some fingers and toes; (3) minutecutaneous folds bordering the posterior edgeof the hind limb; (4) small scales encircling thetail (enlarged, strongly differentiated subcau-dals absent); (5) scattered, irregular flat, convex,to slightly spinose dorsal tubercles on dorsolat-eral portions of body and tail (trunk tuberclesnot arranged in rows); (6) minimal differentia-tion between dorsal and ventral body scales;and (7) postmentals not slender and elongate asin most members of genus Gekko (Brown et al.,2000, 2007, 2008, 2009). We acknowledge thephenotypic similarity between selected mem-bers of the genus Luperosaurus (L. kubli, L.macgregori, L. gulat) and members of the genusGekko. Therefore, to be clear that this new taxoncannot be confused with any known species of

Southeast Asian Gekko, we also diagnose it fromthese species (below).

The new species is readily differentiated fromall species of Luperosaurus on the basis of (1) itsrelatively large body size, (2) the near completeabsence of interdigital webbing (vs. greaterextent of webbing present in all other species),(3) the reduction of cutaneous expansionsbordering the limbs (vs. more extensive in allother species except L. kubli), (4) the presence ofmoderately enlarged postmentals (vs. morereduced or undifferentiated in other species),(5) the presence of flat to convex tail tubercleslimited to dorsolateral portions of body and tail,and (6) high numbers of precloacofemorals. Asummary of the distribution of diagnosticcharacter states in Philippine Luperosaurus ispresented in Table 1; for extensive comparisonsbetween all Luperosaurus species, see Brown etal., 2007:table 1.

Luperosaurus gulat differs from all knownPhilippine species of Gekko by (1) minimaldifferentiation between dorsal and ventral bodyscales (vs. scales on dorsum small, juxtaposed,scales on venter strongly imbricate and greatlyenlarged), (2) the presence of only moderately

FIG. 2. Lateral (A) ventral (B) and dorsal (C; withsnout tilted forward 10u to emphasize internasals)views of the head of the holotype of Luperosaurus gulat(PNM 9282).

NEW SPECIES OF PHILIPPINE LUPEROSAURUS 39

enlarged postmentals (vs. greatly enlarged,elongate, and slender in all Philippine speciesexcept G. gecko), and (3) the absence of enlargedsubcaudals (present in all Philippine Gekko).Luperosaurus gulat and G. athymus lack dorsaltubercles on anterior portions of the body; in allother Philippine Gekko, dorsal tubercles arepresent and arranged in longitudinal rows ofvarying regularity. Additionally, L. gulat has alower precloacofemoral pore-bearing scalecount than G. mindorensis (46–60), G. gigante(54–66), G. palawanensis (65–72), G. romblon (69–80), and G. porosus (74–82). See Brown et al.(2008, 2009) for recent summaries of thedistribution of diagnostic morphology in Phi-lippine members of the genus Gekko.

Comparisons.—For the recognition of the newspecies, the critical comparisons are L. palawa-nensis (the only species previously recordedfrom the Palawan faunal region), the phenotyp-ically most similar species L. kubli, and, to alesser extent, L. macgregori. Other species (Phi-lippine L. corfieldi, L. cumingii, and L. joloensis;

Indonesian L. brooksii and L. iskandari, andMalaysian L. browni and L. yasumai) are distant-ly allopatric and morphologically quite dissim-ilar from L. gulat (Table 1).

Luperosaurus gulat cannot be confused with L.palawanensis, a species that has (1) more extensivecutaneous folds bordering the limbs, (2) moreextensive interdigital webbing, (3) fewer precloa-cofemorals, (4) spinose and recurved dorsolateraltrunk and tail tubercles (5), undifferentiatedpostmentals, (6) more midbody scales, and (7) amuch smaller body size (Table 1).

Luperosaurus gulat is readily distinguished fromL. macgregori (Babuyan Islands) and L. joloensis(Mindanao and Sulu archipelago), two speciesthat have (1) fewer precloacofemorals, (2) fewerToe I and III scansors, (3) more extensiveinterdigital webbing, (4) more midbody scales,(5) undifferentiated postmentals, (6) more exten-sive cutaneous folds bordering the limbs, and (8)much smaller body size (Table 1). Additionally,L. joloensis has highly spinose and recurveddorsolateral trunk and tail tubercles, and L.macgregori lacks dorsolateral tubercles.

Luperosaurus gulat is diagnosed from thephenotypically similar L. kubli (Sierra Madremountains of Luzon) by L. kubli’s possession of(1) fewer precloacofemorals, (2) more extensiveinterdigital webbing, (3) no body tubercles, (4)undifferentiated postmentals, (5) slightly moreextensive cutaneous flaps bordering the limbs,and (6) a larger apparent body size (Table 1).

Luperosaurus cumingii (Luzon and CamiguinNorte Islands) and L. corfieldi (Panay andNegros Islands) can be distinguished from L.gulat by (1) more supralabials and infralabials,

FIG. 3. Palmar surface of right manus (A) and rightside of precloacofemoral region (B) showing 13 pore-bearing scales in the right femoral series (slightlysmaller pores), 12 pore-bearing scales in the precloacalseries (slightly larger pores), and a lack of a disruption(nonpored scaled) between the regions. Fifteen pore-bearing scales in the left femoral region not illustratedfor simplicity.

FIG. 4. Dorsal and ventral views of the body of theholotype of Luperosaurus gulat (PNM 9282; adult male,SVL 5 81.3 mm).

40 R. M. BROWN ET AL.

TA

BL

E1.

Dis

trib

uti

on

of

sele

cted

dia

gn

ost

icch

arac

ters

(+,

pre

sen

t;2

,ab

sen

t)in

Lu

per

osau

rus

gu

lat

and

the

rem

ain

ing

Ph

ilip

pin

esp

ecie

so

fL

up

eros

auru

s.S

eeB

row

net

al.

(200

7:ta

ble

1)fo

rd

istr

ibu

tio

no

fch

arac

ter

stat

esd

isti

ng

uis

hin

gth

em

orp

ho

log

ical

lysi

mil

arg

rou

po

fP

hil

ipp

ine

spec

ies

fro

mth

ere

mai

nin

g,

ph

eno

typ

ical

lyd

iver

gen

tn

on

-Ph

ilip

pin

esp

ecie

s(L

up

eros

auru

sis

kan

dar

i,L

up

eros

auru

sbr

own

i,L

up

eros

auru

sbr

ooks

i,an

dL

up

eros

auru

sy

asu

mai

).B

ilat

eral

lysy

mm

etri

cal

char

acte

rsar

ep

rese

nte

dfo

rth

eri

gh

tsi

de

of

all

spec

imen

s.M

easu

rem

ents

are

pre

sen

ted

inm

illi

met

ers

and

all

spec

imen

s(w

ith

the

exce

pti

on

of

the

Lu

per

osau

rus

jolo

ensi

sp

arat

yp

e)w

ere

mat

ure

adu

lts.

So

urc

eso

fd

ata

(all

con

firm

edb

yex

amin

atio

no

fsp

ecim

ens)

incl

ud

e(1

)B

row

net

al.,

2000

;(2

)O

taet

al.,

1996

;(3

)R

uss

ell,

1979

a;(4

)B

row

nan

dA

lcal

a,19

78;

(5)

Bro

wn

etal

.,20

07;

(6)

Gau

lke

etal

.,20

07;

and

(7)

this

stu

dy

.

gu

lat

corf

ield

icu

min

gii

mac

gre

gor

ipal

awan

ensi

sjo

loen

sis

kubl

i

N5

1N

54

N5

4N

53

N5

2N

52

N5

1

Sn

ou

t–v

ent

len

gth

81.3

70.0

–95.

061

.0–8

2.7

57.4

–58.

943

.7–5

2.0

27.5

–32.

410

5.4

Pre

clo

aco

-fem

ora

ls40

11–1

915

–20

16–1

828

–32

30–3

116

Inte

rnas

als

con

tact

ing

rost

ral

21–

21–

31–

31–

31

1S

cale

sco

nta

ctin

gn

ost

ril

54

5–6

55

55

Hea

dle

ng

th/

hea

dw

idth

1.3

1.2

1.2–

1.3

1.4

1.2

1.3–

1.4

1.2

Su

pra

lab

ials

1116

15–1

713

–15

11–1

311

–13

13In

fral

abia

ls11

13–1

413

–15

14–1

610

–11

10–1

212

Su

bri

ctal

tub

ercl

es2

2+,

22

+,2

2,

afe

w2

Tai

lh

eig

ht/

tail

wid

th0.

820.

82–0

.95

0.90

0.75

–0.8

70.

800.

500.

76N

um

ber

scan

sors

on

toe

I9

10–1

411

–14

10–1

19–

118–

912

Nu

mb

ersc

anso

rso

nto

eII

I13

14–2

013

–16

12–1

412

–13

9–13

16E

xte

nt

web

bet

wee

nd

igit

sII

Ian

dIV

of

pes

1/10

–1/

81/

3–1/

21/

2–3/

41/

5–1/

31/

5–1/

41/

5–2/

31/

6–1/

4

Au

ricu

lar

op

enin

gsu

bo

val

ob

liq

ue

larg

esu

bci

rcu

lar

ov

al,

mo

der

ate

smal

lo

val

,o

bli

qu

ela

rge

sub

circ

ula

rn

arro

wo

bli

qu

eo

val

ob

liq

ue

Pen

ult

imat

esc

anso

rsb

ow

edd

eep

lyn

otc

hed

dee

ply

no

tch

edfe

w,

div

ided

bo

wed

dee

ply

no

tch

edd

eep

lyn

otc

hed

Do

rsal

tub

ercl

esfl

atto

con

vex

con

vex

spin

ose

2sp

ino

se,

recu

rved

stro

ng

lysp

ino

se2

Ven

tro

late

ral

bo

dy

tub

ercl

esfe

w,

flat

toco

nv

exfe

w,

con

vex

toco

nic

alfe

w,

con

vex

few

,co

nv

exfe

w,

spin

ose

man

y,

spin

ose

2

Lat

eral

tail

tub

ercl

es+

2+

++

+2

Ven

tral

sla

rge,

flat

,su

b-

imb

rica

tesm

all,

gra

nu

lar,

jux

tap

ose

dsm

all,

gra

nu

lar,

jux

tap

ose

dsm

all,

gra

nu

lar,

jux

tap

ose

dla

rge,

flat

,su

bim

-b

rica

tesm

all,

flat

,su

bim

-b

rica

tela

rge,

flat

,su

bim

bri

cate

Mid

bo

dy

Sca

les

8216

4–19

415

9–17

113

5–14

699

–106

128–

133

157

An

teri

orm

ost

chin

scal

esm

od

erat

ely

enla

rged

smal

lsm

all

smal

lsl

igh

tly

enla

rged

slig

htl

yen

larg

edsl

igh

tly

enla

rged

An

teri

or

fore

lim

bex

pan

sio

ns

2w

ide

flap

sw

ide

flap

sn

arro

wfo

lds

2m

od

erat

efo

lds

2P

ost

erio

rfo

reli

mb

exp

ansi

on

s2

wid

efl

aps

wid

efl

aps

mo

der

ate

flap

s2

mo

der

ate

fold

sm

inu

tefo

lds

An

teri

or

hin

d-l

imb

exp

ansi

on

s2

wid

efl

aps

nar

row

fold

s2

2n

arro

wfo

lds

2P

ost

erio

rh

ind

-lim

bex

pan

sio

ns

min

ute

fold

sw

ide

flap

sw

ide

flap

sm

od

erat

efl

aps

wid

efl

aps

wid

efl

aps

mo

der

ate

flap

sS

ou

rce

(7)

(6)

(1,4

,6)

(4,5

,7)

(1,4

,7)

(1,4

,7)

(5)

NEW SPECIES OF PHILIPPINE LUPEROSAURUS 41

(2) fewer precloacofemorals, (3) much moreextensive interdigital webbing, (4) much greaterextent of cutaneous flaps bordering the limbs,and (5) undifferentiated postmentals. Lupero-saurus corfieldi is further distinguished from thenew species by the presence of lateral tailtubercles; L. cumingii is further distinguishedby the presence of highly spinose tuberclesthroughout the anterior portions of the body.

The non-Philippine species L. brooksii (Suma-tra), L. browni (Malaysian Peninsula and Bor-neo), and L. iskandari (Sulawesi) can be distin-guished from the new species by their (1)slender body, (2) single internasal contactingthe rostral, (3) presence of interstitial granulesbetween dorsals, (4) fewer precloacofemorals,(5) undifferentiated postmentals, (5) extensivecutaneous flaps bordering the limbs, and (6)smaller body sizes (Table 1). Luperoraurus iskan-dari further differs from the new species by thepresence of subrictal tubercles and presence ofdenticulate tail lobes.

The Bornean endemic L. yasumai can be easilydistinguished from L. gulat by (1) a greaternumber of internasals contacting the rostral, (2)fewer supralabials, (3) absence of precloacofe-morals, (4) fewer subdigital scansors under ToesI and III, (5) greater extent of interdigitalwebbing, (6) extensive cutaneous flaps border-ing the limbs, (7) smaller body size, and (8)presence of denticulate tail lobes.

Finally, the new species cannot be confusedwith Bornean L. sorok (Das et al., 2008) becauseof that species’ (1) extensive interdigital web-bing, (2) undifferentiated postmentals, (3) ex-tensive cutaneous flaps bordering the limbs, (4)spinose head and subrictal tubercles, (5) dentic-ulate tail lobes, and (6) its much smaller bodysize (Table 1).

Description of Holotype.—Adult male in excel-lent condition; hemipenes not everted (Figs. 2–4). Habitus robust, limbs stout, tail original,relatively short; head at widest point (Figs. 2Cand 4A) slightly wider than body (1.1 times) atwidest point (Fig. 4); anterior margins of alllimbs and posterior margins of forelimbssmooth, lacking cutaneous flaps or dermalfolds; posterior margins of proximal (femoral)segment of hind limbs with minute, 1.0–1.5 mmwide cutaneous expansion; distal (tibial) half ofposterior margins of hind limbs with nocutaneous flaps; cutaneous expansions withundifferentiated, minute scales on dorsal andlateral surfaces.

Head large, with hypertrophied temporal andadductor musculature; snout subelliptical,rounded at tip in dorsal and lateral aspect(Fig. 2A, C); head width 76.8% of head lengthand 22.5% of snout–vent length; snout length54.6% of head width and 42.0% of head length;

dorsal surfaces of head smooth, homogenous,with relatively indistinct postnasal, prefrontal,interoribital, and parietal concavities; transverseparietal crest absent; auricular opening anelongate, subelliptical, obliquely oriented slit,partially concealed by temporal swellings of jawclosure musculature; tympanum deeply sunk-en; orbits moderate, lacking dorsally pro-nounced supraorbital crests; palpebra onlyslightly raised above parietal surface; eyemoderate, pupil vertical, its posterior marginwavy (Fig. 2A); tympanic annulus diameter45.6% of eye diameter; limbs stout and relative-ly short, femoral segments of hind limbshypertrophied; tibia length 14.3% of snout–ventlength, 75.8% of femur length.

Rostral large, subrectangular, 1.5 times asbroad as high, with only slight dorsomedialdepression (groove absent); nostril surroundedby rostral, the first labial, an enlarged ovoidsupranasal, and moderately enlarged ovoidpostnasals, each smaller than supranasals; su-pranasals separated by two internasals, slightlylarger than surrounding snout scales; interna-sals pentagonal, flat, followed posteriorly byfour rows of similarly enlarged, convex, cycloidsnout scales (Fig. 2A, C); on both lateral sides ofsnout, the second row of snout scales followingpostnasals (i.e., postnasal–preloreal region con-cavity) are distinctively enlarged.

Supralabials 13/11 (L/R; 7–11/9 subocular),bordered dorsally by one row of very slightlydifferentiated, similarly flattened snout scalesand anteriorly (loreal region) by a secondincomplete row of elongate, flattened scales;infralabials 11/11 (last 1–2 infralabials con-cealed in postrictal pocket), bordered antero-ventrally by a single row of moderately differ-entiated chin scales (.5 times the size ofundifferentiated, minute gular scales) and pos-teroventrally by 2–3 rows of slightly enlarged,irregular scale rows (Fig. 2B); postrictal scalesslightly enlarged, approximately 2 times thesize of scales of temporal region; postrictaltubercles absent; mental very small, semicircu-lar; postmental scales moderately enlarged,followed by one pair of only slightly enlargedsecondary postmentals, undifferentiated gularscales thereafter; remainder of undifferentiatedthroat scales very small, round, nonimbricate,juxtaposed (Fig. 2B).

Dorsal cephalic scales round, nonimbricate,convex to granular, remarkably homogenoussave for enlargement on snout, lateral temporalregions, and palpebra (slightly larger than scalesin adjacent frontal region); preorbital regionlacking differentiated scales; temporal regiontubercles absent; nuchals granular, stronglyconvex, continuously grading into slightly largerdorsals; dorsals enlarged, cycloid, lacking well-

42 R. M. BROWN ET AL.

developed interstitial granules; throat and chinscales homogenous, round, convex, granular;sharply transitioning and greatly increasing insize in gular region; becoming imbricate in gularand pectoral regions, and continuing to increasein size through ventral abdomen where imbrica-tion further increases.

Ornamental cephalic scalation absent; 33/36circumorbitals, undifferentiated except for veryslight dorsolateral transverse elongation andmodification into slight, fringe-like points,projecting into orbit in posterodorsal corner oforbit; true spiny ciliaries absent; 27 interorbitalscales transversely at palpebral midpoints.

Axilla–groin distance 44.7% of snout–ventlength; undifferentiated dorsal body scalesround, convex, nonimbricate, transversely un-differentiated; granular dorsal trunk scales lackwell-developed interstitial granules; 82 trans-verse midbody scales (enlarged ventrals in 23rows across ventral surfaces of trunk); paraver-tebrals between limb insertion midpoints 124;ventrals between limb insertion midpoints 62;dorsal body tuberculation moderate, with flat toconvex enlarged scales extending over posteriortwo-thirds of body trunk, not arranged in rows.Dorsal tubercles slightly more dense in verte-bral region, becoming most dense at posteriorend of dorsum and above hind limb insertions.

Scales on dorsal surfaces of limbs flat, enlargedand imbricate on pre-brachial surfaces; small(similar to adjacent trunk scales), increasinglygranular, and convex on antebrachial surfaces;forelimb scalation otherwise homogenous, com-pletely lacking tuberculation; scalation of ventralsurfaces of forelimbs with homogenous, round,granular, minute scales that extend onto thepalmar surfaces of the manus; scales on dorsalsurfaces of femoral segment of hind limbssimilarly enlarged, flat, and imbricate on pre-brachial surfaces and granular, round, andminute on antebrachial surface (and on cutane-ous expansion posteriorly bordering the hindlimb); tibial segment of hind limb with small,granular, convex scales on dorsal surfaces, andenlarged, imbricate scales on ventral surfaces;scales of heel greatly enlarged and flat; knee withscattered enlarged scales similar in appearanceto tuberculation of dorsal trunk. Scales on dorsalsurfaces of manus and pes similar to dorsal limbscales; ventral surfaces with minute, round,convex, granular scales; digits with only aminute vestige of interdigital webbing, strongestbetween Toes III and IV.

Precloacofemoral pore-bearing scales, ar-ranged in a slightly bowed configuration,pierced by 40 pores; precloacal and femoralpore-bearing scales in continuous contact; 13femoral pore-bearing scales on right, 12 pre-cloacals (slightly larger than femorals on either

side), 15 femorals on left; precloacofemoralspreceded by two similarly enlarged but non-pored and non-dimpled scale rows; followed bysimilarly sized scales posterior to precloacals;scales posterior to femorals undifferentiated;scales posterolateral to precloacofemoral series(i.e., along ventroposterior surfaces of hindlimb) reduce in size sharply to minute scalesunder the slight cutaneous expansion of theposterior edge of the hind limb; a single, greatlyenlarged cloacal spur follows the vent on eitherside of hemipenal bulges.

Digits widely dilated and covered on palmar/plantar surfaces by wide, bowed scansors(Fig. 3A); penultimate 2 or 3 scansors moderatelynotched (not divided); digits only very slightlywebbed, with web extending 1/10–1/8 the lengthof digit from base, and ending well below thedilated hyperextensible portions of digits; vestigeof webbing only ascertainable between fingers IIIand IV and IV and V; subdigital scansors ofmanus: 10/9, 11/11, 13/13, 15/14, and 12/13 onleft/right digits I–V, respectively; pes: 12/13, 11/12, 13/13, 14/14, and 14/15 on left/right digits I–V, respectively; all digits but first (inner) clawed;inner digits of both manus and pes with enlargedpair of elongate scales in claw; remainingterminal claw-bearing phalanges compressed,with large recurved claws (claws sheathed),rising free at distal end, extending well beyonddilated distal portion of digit.

Tail relatively long, 94.6% of snout–ventlength; tail height (not including basal denticu-late lobes) 76.9% of tail width; tail not de-pressed, cylindrical, adorned throughout itsentire length with enlarged flat to convextubercle rows at the posterior margins of eachcaudal annulation (Fig. 5); 21 fracture planes/autotomy grooves (5 whorls or annuli), with 7–11 undifferentiated dorsal caudal scales perannulation; dorsal caudal scales 149, granular,convex, subcircular; subcaudals 88, small tomoderate, slightly larger than dorsal caudalscales, squarish, flat, subimbricate.

Measurements of Holotype.—Measurements inmillimeters. Snout–vent length 81.3; tail length76.9; head length 23.8; head width 18.3; headdepth 9.8; snout length 10.0; eye diameter 4.6;eye–narial distance 7.3; tympanic annulus di-ameter 2.1; internarial distance 4.2; interorbitaldistance 5.4; axilla–groin distance 36.4; femurlength 15.3; tibia length 11.6; Toe I length 4.4;Toe IV length 8.7; tail width 7.2; tail height 5.9.

Coloration in Preservative.—Dorsum dark grey,with indistinct slightly darker grey blotches,lacking any trace of transverse banding; dorsalnuchal region and head slightly darker greythan body; lateral portions of head and labialscales flat grey; postrictal region pale grey;dorsal surfaces of limbs dark grey, lacking

NEW SPECIES OF PHILIPPINE LUPEROSAURUS 43

transverse banding; dorsal surfaces of digitsdark grey with slightly lighter innermost digits;dorsal and lateral portions of tail banded,alternating dark and very dark grey (notcorresponding to tail annuli); distal portions oftail with lighter blotches.

Infralabial region and chin flat light gray;gular region pale cream, flanked by slightlydarker grey ventrolateral jaw coloration; sternalregion light cream with indistinct grey flecks;ventral body light grey with dark grey blotcheswrapping on to ventrum from flanks; ventralsurfaces of limbs light cream with grey borders;ventral surfaces of scansors of fingers and toesdark grey; precloacofemoral pore-bearing scaleseries white with pale orange pores; ventral taildark grey with indistinct white flecks and notransverse bands. Coloration in life was notrecorded, but our experience with numerousother species of Luperosaurus suggests that littlefading occurs with preservation and that liveand preserved colorations usually only differ bythe presence of faint yellowish tones on ventralsurfaces of body and limbs (Brown et al., 2000,2007; Gaulke et al., 2007).

Habitat and Natural History.—We have noinformation on microhabitat preference, abun-dance and distribution, or reproduction in thenew species. The single specimen was captured

on the trunk of a large tree. The terrain at Gunob(3.4 km west, 0.60 km south of Mt. Mantalinga-jan peak, 1,300 m) was a moderate plateau alonga ridge. The vegetation was generally a transi-tional midmontane to upper montane forest.The area is disturbed and currently being usedby several indigenous Palawanon tribal familiesas a refuge for several months after kaingin(shifting agriculture) harvest. Canopy coverheight was approximately at 15 m and tallbamboo groves were common (indicative of ahistory of disturbance), and shrublike andclimbing bamboos were common on the ridges.Undergrowth plants include ferns, orchids andgingers, and thin blade grasses were common inopen areas. Large boulders were conspicuouson the north and south ridges surrounding theapproach to Mt. Mantalingajan’s peak.

Etymology.—The specific epithet is taken fromthe Tagalog (Filipino) term gulat, meaningsomething unexpected, a surprise, or an aston-ishing finding.

DISCUSSION

In their recent description of L. kubli, Brown etal. (2007) discussed the full range of morpho-logical variation exhibited by Sundaland mem-bers of the genera Gekko and Luperosaurus; theyalso provided a revised definition of Lupero-saurus and expanded on Brown et al.’s (2000)clarification of Luperosaurus generic boundarieswith respect to the Asian members of the genusGekko. In general, recent discoveries of newspecies of Luperosaurus have forced taxonomiststo adopt an expanded definition of the genusthat is increasingly Gekko-like (Brown et al.,2008, 2009), as evinced by the shared presenceof several formerly exclusive Gekko charactersstates (e.g., larger body size, reduced bodytuberculation, reduced interdigital webbing,reduced extent of cutaneous expansions bor-dering the limbs, increased size of postmentals,increased differentiation between dorsal andventral body scalation; Brown et al., 2000, 2007,Gaulke et al., 2007).

As with other recently described Luperosaurusspecies (Brown et al., 2007; Gaulke et al., 2007),the new species further obfuscates the genericboundaries between Gekko and Luperosaurus(Brown and Alcala, 1978; Russell, 1979; Brownet al., 2000). Although the preponderance ofcharacters convinces us that the best course ofaction is to assign the new species to the genusLuperosaurus, L. gulat further stretches thegeneric definition by extending the range ofvariation observed in several Luperosauus charac-ters. First, L. gulat possesses differentiated andmoderately enlarged postmental chin shields, a

FIG. 5. Dorsolateral tail whorl scalation in (A)Luperosaurus gulat (PNM 9282; note convex tubercles)and (B) Luperosaurus palawanensis (holotype: CAS134207; note highly spinose tubercles).

44 R. M. BROWN ET AL.

markedly Gekko-like character state (not exhibitedto this extent in the other Gekko-like Luperosaurusspecies such as L. kubli but present to a lesserdegree in the miniaturized species L. joloensis). Inother respects, L. gulat and L. kubli both appearmore Gekko-like by their near complete absence ofcutaneous expansions bordering the limbs (Ta-ble 1), by increased differentiation between dor-sals (minute, juxtaposed, granular) and ventrals(moderately enlarged, imbricate), reduction ofinterdigital webbing, and large body size (Brownet al., 2007). A superficial similarity of these twospecies to the highly distinctive and distantlyallopatric Taiwan and Ryukyu Archipelago taxa(Gekko hokouensis, Gekko yakuensis, Gekko tawaensis,Gekko japonicus, Gekko shibatae, Gekko vertebralis;Ota, 1989; Toda et al., 2008) is probably indicativeof convergence rather than shared ancestry,although a phylogenetic analysis would berequired to confirm our assertion, which isadmittedly based on morphological gestalt.

The description of L. gulat brings the totalnumber of Luperosaurus species to 11, with seventaxa now known from the Philippines (Brownand Diesmos, 2000; Brown et al., 2007; Gaulke etal., 2007). Although recent years have seen anincrease in the number of species of the genusLuperosaurus, we consider 11 taxa to be a certainunderestimate of the true diversity of theseenigmatic forest geckos. By nearly all accounts,species of Luperosaurus are so rare and elusivethat any comments concerning total speciesdiversity are undoubtedly premature (Brown etal., 2000, 2007; Gaulke et al., 2007). At present, all11 species are known from a cumulative total offewer than 30 specimens that have been reportedin the literature. Nearly all new collections ofLuperosaurus specimens result in new speciesdescriptions, and to date, only a few species areknown from more than a single specimen (Brownand Diesmos, 2000; Brown et al., 2000, 2007). Alldescribed species except L. cumingii (Brown andDiesmos, 2000; Gaulke et al., 2007) and L.macgregori (R. M. Brown and C. Oliveros, unpubl.data) are known only from a cumulative total ofjust a few specimens.

Luperosaurus gulat appears phenotypicallymost similar to L. kubli, and, to a lesser extent,L. macgregori, with the principal differencesbeing L. gulat’s dense aggregation of flat toconvex posterodorsal trunk and tail tubercles(absent in L. kubli and L. macgregori), reductionin webbing and cutaneous limb flaps, moder-ately enlarged postmentals, and L. macgregori’smuch smaller body size (Table 1). Luperosaurusgulat appears morphologically quite distinctfrom the potentially sympatric L. palawanensis(and the morphologically similar but allopatricL. joloensis), a much smaller species (Table 1),with extensive interdigital webbing (absent in L.

gulat) and ornate and exceptionally spinose andelongate posteriorly facing posterodorsal andlateral tail tubercles (Fig. 5; described by Brownand Alcala [1978] as ‘‘spear-head-like;’’ dorso-lateral tail tubercles flat to convex in L. gulat),and widely expanded cutaneous flaps border-ing the limbs (absent in L. gulat). If possession ofapparently derived morphological characterstates is any indication of shared ancestry, itdoes not appear that L. palawanensis and L. gulatare sister species.

We expect future phylogenetic analyses todemonstrate that Gekko is paraphyletic withrespect to Luperosaurus, Ptychozoon, and possiblyPseudogekko and Lepidodactylus. Past studies havenoted morphological similarities between mem-bers of these genera (Brown and Alcala, 1978;Russell, 1979) and a lack of consistent morpho-logical characters distinguishing them (Brown etal., 1997, 2000, 2007), suggesting convergentevolution, common ancestry, or retention ofpleisiomorphic character states. The phylogeneticposition of the Gekko-like Luperosaurus speciesremains for now an enigmatic and interestingsystematic problem (Brown et al., 2000).

The discovery of another species of Luper-osaurus from the largely unexplored forests ofsouthern Palawan Island is surprising. Al-though this isolated mountain range is expectedto harbor undocumented endemic diversity, wedid not expect to find a second species ofLuperosaurus on Palawan. Whether the twoendemic Palawan Luperosaurus species occur insympatry has yet to be determined. Luperosauruspalawanensis is only known from north-centralPalawan (Thumb Peak, known locally as ‘‘Sa-pucoy’’; near the village of Iwahig and PuertoPrincesa), whereas L. gulat is now known fromthe southernmost portion of the island (Fig. 1).The recent discovery of new species in thisgeologically distinct and zoogeographicallyunique component of Palawan emphasizes theimmediate need for a comprehensive herpeto-logical inventory of the Palawan faunal region,particularly its long-neglected southern ex-tremes including Balabac and other islands justnorth of Borneo.

Why are Luperosaurus species so rare? Despitethe availability of new data on microhabitatpreferences of Luperosaurus species, we areunable to definitively answer this question.Having assumed for years that the rarity ofthese species was explained by a preference forthe forest canopy (Brown and Diesmos, 2000;Brown et al., 2000), we have been surprised byrecent discovery of two species in lower strataof forests (Brown et al., 2007; Gaulke et al.,2007), by high abundance of another species inshrubby vegetation of mature costal forestfragments (Brown et al., 2007, 2008; R. M.

NEW SPECIES OF PHILIPPINE LUPEROSAURUS 45

Brown and C. Oliveros, unpubl. data), and nowby the discovery of this new species at relativelyhigh elevations (1,300 m) in moist, montaneforests of central Palawan. Brown et al. (2007)surmised that the obligate canopy hypothesismay not suffice as an explanation for Luper-osaurus rarity. Instead they suggested thatLuperosaurus species may have evolved tospecialize on mature coastal forests in thePhilippines (which have been almost entirelyremoved over the last 400 years; Brown andDiesmos, 2009). This may explain why thesomewhat more common L. cumingii and L.corfieldi are most often encountered at a fewhundred meters elevation, in degraded habitatadjacent to intact lower montane forest. Thisexplanation is also consistent with the fact thatat least one species is quite common in a smallisolated fragment of coastal forests (Brown etal., 2007; R. M. Brown and C. Oliveros, unpubl.data) in the Babuyan Islands and may explainwhy remaining species of Luperosaurus are sorarely encountered and at such low densities.

Finally, as biodiversity information and newdata accumulate, the naive characterization ofPalawan Island as a simple faunal extension ofnorthern Borneo (Heaney, 1985) with a reducedbut nested subset of its fauna and flora, hasemerged as an inaccurate, mammal- and bird-biased oversimplification. All available phyloge-netic studies involving endemic amphibians andreptiles of Palawan (McGuire and Kiew, 2001;Brown and Guttman, 2002; Evans et al., 2003)have demonstrated affinities to taxa from thetruly oceanic portions of the Philippines (Brownand Diesmos, 2000, 2002, 2009). The morpholog-ical similarity of L. gulat to northern Philippinetaxa (L. kubli and L. macgregori) suggests the newspecies may be yet another lineage that defies thegross generalization of Palawan as a merebiogeographic peninsula of northern Borneo.The comparative study of the systematic rela-tionships of multiple vertebrate groups thatinclude Palawan endemics is fertile ground forfuture research (Brown and Diesmos, 2009).

Acknowledgments.—For access to institutionalcollections, we thank J. Vindum, R. Drewes, andA. Leviton (CAS); J. Rosado, J. Losos, and J.Hanken (MCZ); A. Resetar and H. Voris(FMNH); T. LaDuc and D. Cannatella (TNHC);and R. Sison (PNM; museum codes followLeviton et al., 1985). We thank the Departmentof the Environment and Natural Resources(DENR; especially Brooke’s Point Municipalauthorities), and the Manila-based ProtectedAreas and Wildlife Bureau, for facilitatingresearch permits for this and related studies.We also thank Provincial DENR authorities ofPalawan Province, the Palawan Council for

Sustainable Development, the National Com-mission on Indigenous Peoples, and PanglimaFernandez Gasang for logistical support. Field-work was conducted with support from theNational Science Foundation (DEB 0743491 toRMB), Conservation International’s CriticalEcosystem Partnership Fund, and the Universi-ty of Kansas Natural History Museum andBiodiversity Institute. Thanks are owed to M.Garfield, J. Weghorst, and C. Siler for assistancewith figures and manuscript reviews, and weare grateful to N. Antoque, J. Fernandez, N.Fernandez, B. Fernandez, R. Duya, I. Osbucan,and especially U. Carestia for their assistance inthe field.

LITERATURE CITED

BROWN, R. M. 1999. New species of parachute gecko(Squamata; Gekkonidae; genus Ptychozoon) fromnortheastern Thailand and Central Vietnam. Co-peia 1999:990–1001.

BROWN, R. M., AND A. C. DIESMOS. 2000. The lizardgenus Luperosaurus: taxonomy, history, and con-servation prospects for some of the world’s rarestlizards. Sylvatrop: Technical Journal of PhilippineEcosystems and Natural Resources 10:107–124.

———. 2001 (2002). Application of lineage-basedspecies concepts to oceanic island frog popula-tions: the effects of differing taxonomic philoso-phies on the estimation of Philippine biodiversity.Silliman Journal 42:133–162.

———. 2009. Philippines, Biology. In R. Gillespie and D.Clague (eds.), Encyclopedia of Islands, pp. 723–732.University of California Press, Berkeley.

BROWN, R. M., AND S. I. GUTTMAN. 2002. Phylogeneticsystematics of the Rana signata complex of Philip-pine and Bornean stream frogs: reconsideration ofHuxley’s modification of Wallace’s Line at theOriental-Australian faunal zone interface. Biolog-ical Journal of the Linnean Society 76:393–461.

BROWN, R. M., J. W. FERNER, AND A. C. DIESMOS. 1997.Definition of the Philippine Parachute Gecko,Ptychozoon intermedium Taylor 1915 (Reptilia: La-certilia: Gekkonidae): redescription, designation ofa neotype, and comparisons with related species.Herpetologica 53:357–373.

BROWN, R. M., J. SUPRIATNA, AND H. OTA. 2000. Discoveryof a new species of Luperosaurus (Squamata; Gekko-nidae) from Sulawesi, with a phylogenetic analysisof the genus and comments on the status of L.serraticaudus. Copeia 2000:191–209.

BROWN, R. M., A. C. DIESMOS, AND M. V. DUYA. 2007. Anew species of Luperosaurus (Squamata: Gekkoni-dae) from the Sierra Madre mountain range ofnorthern Luzon Island, Philippines. Raffles Bulle-tin of Zoology 55:153–160.

BROWN, R. M., C. OLIVEROS, C. D. SILER, AND A. C.DIESMOS. 2008. A new Gekko from the BabuyanIslands, northern Philippines. Herpetologica63:305–320.

———. 2009. Phylogeny of Gekko from the northernPhilippines, and description of a new species fromCalayan Island. Journal of Herpetology 43:620–635.

46 R. M. BROWN ET AL.

BROWN, W. C., AND A. C. ALCALA. 1978. PhilippineLizards of the Family Gekkonidae. Silliman Uni-versity Press, Dumaguete City, Philippines.

DAS, I., M. LAKIN, AND P. KANDAUNG. 2008. New speciesof Luperosaurus (Squamata: Gekkonidae) from theCrocker Range Park, Sabah, Malaysia (Borneo).Zootaxa 1719:53–60.

ESSELSTYN, J. A., P. WIDMAN, AND L. A. HEANEY. 2004.The mammals of Palawan Island, Philippines.Proceedings of the Biological Society of Washing-ton 117:271–302.

EVANS, B. J., R. M. BROWN, J. A. MCGUIRE, J. SUPRIATNA,N. ANDAYANI, A. C. DIESMOS, D. ISKANDAR, D. J.MELNICK, AND D. C. CANNATELLA. 2003. Phyloge-netics of fanged frogs: testing biogeographicalhypotheses at the interface of the Asian andAustralian faunal zones. Systematic Biology52:794–819.

GAULKE, M., H. ROSLER, AND R. M. BROWN. 2007. A newspecies of Luperosaurus (Squamata; Gekkonidae)from Panay Island, Philippines, with comments onthe taxonomic status of Luperosaurus cumingii(Gray, 1845). Copeia 2007:413–425.

HEANEY, L. R. 1985. Zoogeographic evidence formiddle and late Pleistocene land bridges to thePhilippines. Modern Quaternary Research of SEAsia 9:127–143.

HOOGSTRAL, H. 1951. Philippine Zoological Expedition,1946–1947, narrative and itinerary. Fieldiana (OldSeries) 33:1–84.

INGER, R. I. 1954. Systematics and zoogeography ofPhilippine Amphibia. Fieldiana Zoology (OldSeries) 33:182–531.

LEVITON, A. E., R. H. GIBBS JR., E. HEAL, AND C. E.DAWSON. 1985. Standards in herpetology andichthyology. Part I. Standard symbolic codes forinstitutional resource collections in herpetologyand ichthyology. Copeia 1985:802–821.

MANTHEY, U., AND W. GROSSMAN. 1997. Amphibien undReptilien Sudostasiens. Natur und Tier, Berlin,Germany.

MCGUIRE, J. A., AND A. C. ALCALA. 2000. A taxonomicrevision of the flying lizards of the PhilippineIslands (Iguania: Agamidae: Draco), with a de-scription of a new species. Herpetological Mono-graphs 14:92–145,

MCGUIRE, J. A., AND B.-H. KIEW. 2001. Phylogeneticsystematics of Southeast Asian flying lizards(Iguania: Agamidae: Draco) as inferred frommitochondrial DNA sequence data. BiologicalJournal of the Linnean Society 72:203–229.

OTA, H. 1989. A review of the geckos (Lacertilia:Reptilia) of the Ryukyu Archipelago and Taiwan.In M. Matsui, T. Hikida, and R. C. Goris (eds.),Current Herpetology in East Asia. Proceedings ofthe Second Japan-China Herpetological Sympo-sium, pp. 222–261. Herpetological Society of Japan,Kyoto.

OTA, H., K.-Y. LUE, S.-H. CHEN, AND W. C. BROWN. 1989.Taxonomic status of the Taiwanese Gekko, withcomments on the synonymy of Luperosaurusamissus Taylor. Journal of Herpetology 23:76–78.

OTA, H., S. SENGOKU, AND T. HIKIDA. 1996. Two newspecies of Luperosaurus (Reptilia: Gekkonidae)from Borneo. Copeia 1996:433–439.

ROSLER, H., C. D. SILER, R. M. BROWN, A. D. DEMEGILLO,AND M. GAULKE. 2006. Gekko ernstkelleri sp. n.—anew gekkonid lizard from Panay Island, Philip-pines. Salamandra 42:197–211.

RUSSELL, A. P. 1979. A new species of Luperosaurus(Gekkonidae) with comments on the genus. Her-petologica 35:282–288.

TAYLOR, E. H. 1918. Reptiles of the Sulu Archipelago.Philippine Journal of Science 13:233–267.

———. 1922. The Lizards of the Philippine Islands.Philippine Bureau of Science, Manila.

———. 1962. New oriental reptiles. University ofKansas Science Bulletin 43:209–263.

TODA, M., S. SENGOKU, T. HIKIDA, AND H. OTA. 2008.Description of two new species of the genus Gekko(Squamata: Gekkonidae) from the Tokara andAmami Island groups in the Ryuku Archipelago,Japan. Copeia 2008:452–466.

Accepted: 11 March 2009.

APPENDIX 1

Specimens ExaminedLuperosaurus kubli.— Philippines, Quirino Province,

Municipality of Nagtipunan, Barangay Disimungal,Sierra Madre Mountain Range, Mt. Lataan (900 m):PNM 9156 (holotype).

Luperosaurus corfieldi.—Philippines, Panay Island,Aklan Province, Municipality of Buruanga, BarangayTagosip: PNM 7919 (holotype), PNM 7920, 8489(paratypes); Negros Isl., Negros Oriental Prov.,Municipality of Valencia, Lake Balinsasayao: SUR2211; Saksak Cr., Mt. Cuernos de Negros, ‘‘CampLookout’’: CAS-SU 24394–24395; Negros Isl., LakeBalinsasayao: CAS 182570.

Luperosaurus cumingii.—Philippines, Luzon Isl., Ca-marines Sur Province, Municipality of Caramoan,Caramoan Peninsula, Anuling Mountain: UF 77829;Albay Province, Municipality of Tiwi, BarangayBanhaw, Sitio Purok 7, Mt. Malinao, 550 m abovesea level: TNHC 61910; Camiguin Norte Island,Cagayan Province, Municipality of Calayan, BarangayBalatubat KU 308023.

Luperosaurus joloensis.—Philippines, Mindanao Isl.,‘‘Cotobato Coast:’’ MCZ 26118; Jolo Isl., Siet Lake:CAS 60675 (paratype); Zamboanga Del Sur Province,Barangay Pasonanca, Sitio Canucutan, PasanoncaNatural Park: one uncataloged specimen at KU(RMB 9416).

Luperosaurus macgregori.—Philippines; Babuyan Is-lands group, Calayan Isl.: CAS-SU 6263 (paratype),USNM 36191 (holotype); Barit Isl. (near Fuga Isl.):USNM 508306, 508308.

Luperosaurus palawanensis.—Philippines, PalawanIsl., Palawan Province, Malatgaw River, southeast ofThumb Peak, approximately 3.5 km west-northwest ofIwahig (9u44940.270N; 118u379480E): CAS 134207 (ho-lotype); Thumb Peak, approximately 7 km northwestof Iwahig: (9u45942.980N; 118u369180E): CAS 136740(paratype).

Luperosaurus browni.—Malaysia, peninsular Malay-sia, Selangor, Ulu Gombak forest reserve, 35 km northof Kuala Lumpur: FMNH 185106 (holotype); Malay-

NEW SPECIES OF PHILIPPINE LUPEROSAURUS 47

sia, Sarawak (Borneo Isl.), Lambir National Park: KUZ12835 (L. serraticaudus holotype).

Luperosaurus iskandari.—Indonesia, Sulawesi Is.,Propinsi Sulawesi Tengah (Central Sulawesi Prov-ince), Kabupaten Banggai, Kecamatan Pagimana,Kampung/Desa Siuna, approximately 4 km east ofDusun Satu (Region 1), Mt. Tompotika (0u44.59S,123u01.19E): MZB Lace. 2114.

Luperosaurus yasumai.—Indonesia, Kalimantan(Borneo Isl.), Bukit Soeharto Experimental Forest,45 km south-southwest of Samarinda: KUZ 30408(holotype)

Luperosaurus brooksii.—Indonesia, Sumatra Isl., Ben-kuelen, Lebong Tandai: BMNH 1920.1.16.2 (holotype).

Gekko hokouensis.—‘‘Tablas Island’’ (presumably inerror) FMNH 17812 (L. amissus holotype).

48 R. M. BROWN ET AL.