TAXONOMIC REVISION OF BRACHYMELES SAMARENSIS
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Herpetological Monographs, 25, 2011, 76–112
E 2011 by The Herpetologists’ League, Inc.
PHYLOGENY-BASED SPECIES DELIMITATION IN PHILIPPINE
SLENDER SKINKS (REPTILIA: SQUAMATA: SCINCIDAE) II:
TAXONOMIC REVISION OF BRACHYMELES SAMARENSIS AND
DESCRIPTION OF FIVE NEW SPECIES
CAMERON D. SILER1,3, ALLISON M. FUITEN1, ROBIN M. JONES1, ANGEL C. ALCALA2,
1
AND RAFE M. BROWN
1
Biodiversity Institute and Department of Ecology and Evolutionary Biology, University of Kansas, 1345 Jayhawk
Boulevard, Lawrence, KS 66045-7593, USA
2
SUAKCREM Marine Laboratory, Silliman University, Bantayan, Dumaguete City, Philippines 6200
ABSTRACT: With robust new datasets from morphology and DNA sequences, we review the limbed,
nonpentadactyl species of the Brachymeles samarensis complex (now known to include B. cebuensis, B.
minimus, and B. lukbani), and describe five new species in this highly limb-reduced, endemic Philippine
clade of scincid lizards. For more than four decades, B. samarensis has been recognized as a single
‘‘widespread’’ species. This perception of the species’ peculiar geographic range has persisted as a result of
weak sampling and similar gross morphology (body sizes, scale pigmentation) among populations. However,
previous authors have noted morphological variation between different island populations, and our new data
build on these observations and extend them to delimit new proposed species boundaries. Our data indicate
that the ‘‘widespread’’ species B. samarensis is actually a complex of six distinct lineages, some of which are
not each others’ closest relatives, and each of which is genetically unique. The taxa we define possess
allopatric geographic ranges and differ from their congeners by numerous diagnostic characters of external
morphology, and therefore should be recognized as full species in accordance with any lineage-based species
concept. Species diversity in the genus has doubled in the last 3 yr, with these six taxa increasing the total
known number of species of Brachymeles to 30.
Key words: Biodiversity; Endemism; Faunal region; Fossorial lizards; Limb reduction; Species
delimitation; Taxonomy
FEW GENERA of scincid lizards are known to B. tridactylus, and B. wrighti), and five are
possess species representing a full spectrum of entirely limbless (B. apus, B. minimus, B.
body forms, from fully limbed, pentadactyl miriamae, B. lukbani, and B. vermis).
species to limbless species (see Siler et al., Among the nonpentadactyl species, numer-
2011b; Siler and Brown, 2011). In the genus ous studies have documented a wide range of
Brachymeles, all but two of the 26 recognized limb- and digit-reduced states, from minute
species are endemic to the Philippines, with limbs with 0–3 digits (B. bonitae, B. cebuensis,
the exceptions being a single species (B. apus) B. muntingkamay, B. samarensis, B. tridacty-
from northern Borneo and another (B. lus), to moderately developed limbs with
miriamae) from Thailand (Brown and Alcala, four to five digits on the hands and feet (B.
1980; Hikida, 1982; Siler, 2010; Siler et al., elerae, B. pathfinderi, B. wright; Duméril and
2009, 2010a,b, 2011a,b,c; Siler and Brown, Bibron, 1839; Brown, 1956; Brown and Rabor,
2010, 2011). Thirteen species are pentadactyl 1967; Brown and Alcala, 1980; Taylor, 1917,
(B. bicolor, B. boholensis, B. boulengeri, B. 1918, 1925; Siler, 2010; Siler et al., 2009,
gracilis, B. kadwa, B. makusog, B. mind- 2010a, 2011a,b,c; Siler and Brown, 2011). All
orensis, B. orientalis, B. schadenbergi, B. species are semifossorial and typically found
talinis, B. taylori, B. tungaoi, and B. vindumi), in dry, rotting material inside or underneath
eight are nonpentadactyl with incompletely decaying logs or in loose soil, forest floor
developed limbs and reduced numbers of detritus, and leaf litter.
digits (B. bonitae, B. cebuensis, B. elerae, B. Although the genus was named well over
muntingkamay, B. pathfinderi, B. samarensis, 150 yr ago (Duméril and Bibron, 1839), the
rate of Brachymeles species descriptions
3
CORRESPONDENCE: e-mail, camsiler@ku.edu reached an apparent asymptotic maximum in
762011] HERPETOLOGICAL MONOGRAPHS 77
1980 (Brown and Alcala, 1980). The one diversity in the genus for 30 yr, until Taylor
exception is B. minimus, a legless species published a series of herpetofaunal descriptions
described in 1995 (Brown and Alcala, 1995). in the early 1900s (Taylor, 1917, 1918, 1922,
For more than a century, limited numbers of 1925). It would be 50 yr before Brown (1956)
specimens in museum collections, combined described B. samarensis from a single juvenile
with the similar body plans and external specimen (FMNH 44472) collected in Guiuan,
morphological features among species of Samar Island, Philippines in 1945. At the time of
Brachymeles, limited assessments of species- description, Brown (1956) hypothesized the
level diversity (Brown, 1956; Brown and species was most closely related to B. elerae
Alcala, 1980; Brown and Rabor, 1967; Taylor, due to similarities in the number of paraverte-
1917). Recent studies have revealed the bral scale rows. This single juvenile would
species-level diversity within Brachymeles to remain the only vouchered specimen of this
be drastically underestimated, and have iden- unique bidactyl species from Samar Island (Siler
tified numerous nonmonophyletic species and Brown, 2010).
complexes within the Philippines (Siler and By the time Brown and Rabor (1967)
Brown, 2010, 2011; Siler et al., 2011a). revised the genus Brachymeles, samples of
Additionally, several rare, mid-to-high eleva- specimens morphologically similar to B. sa-
tion species long represented by only a few marensis had been collected from the islands
specimens (e.g., B. bicolor, B. elerae, B. of Luzon and Leyte. Additionally, Brown and
wrighti, B. pathfinderi) have recently been Rabor (1967) reported on a second specimen
rediscovered (Siler, 2010) and redescribed as from Samar Island; however, no information
valid taxa (Siler et al., 2011a,c). Together, on where the specimen was deposited, or its
these studies coupled with increased sampling museum catalog number, was provided. Al-
throughout the Philippines and a new, robust though Brown and Rabor (1967) treated B.
molecular dataset, allow us to begin evaluating samarensis as a single widespread species,
variation across the isolated populations of they referred to the species as a ‘‘complex,’’
widespread species in the Philippines. suggesting that they suspected it contained
In a recent study, Siler and Brown (2010) multiple species. They also noted several
revised two polytypic species (B. boulengeri distinct morphological differences between
and B. schadenbergi) and one widespread island populations, including differences in
species (B. talinis), and inferred the presence fore- and hind-limb digit number and head
of 10 genetically and morphologically distinct scale patterns.
allopatric evolutionary lineages (species). Sev- Additional island populations of B. samar-
eral other species are still recognized as ensis were subsequently sampled by the time
having widespread distributions that span Brown and Alcala (1980) revised the genus,
historical faunal demarcations in the Philip- including populations from the Lapinig Group
pines (Brown and Diesmos, 2002, 2009; islands off the northeast coast of Bohol Island
Brown and Guttman, 2002; Heaney, 1985), (Fig. 1). Ross and Gonzales (1992) would later
including B. samarensis and B. bonitae report on observations of B. samarensis from
(Brown, 1956; Brown and Rabor, 1967; Brown Catanduañes Island off the northeast coast of
and Alcala, 1980). One of these species (B. the Bicol Peninsula (Fig. 1) and in 2001, RMB
samarensis) is the focus of this study. (unpublished data) recorded B. samarensis on
the southern tip of the Bicol Peninsula in the
TAXONOMIC HISTORY foothills of Mt. Bulusan.
The genus Brachymeles was first described by To date, Brachymeles samarensis remains a
Dumeril and Bibron (1839) for the small, limb- widespread species spanning islands of the
reduced species B. bonitae. Three additional Luzon and Mindanao Pleistocene Aggregate
species (Senira bicolor Gray, 1845; Eumeces Island Complexes (PAICs; Brown and Dies-
(Riopa) gracilis Fischer, 1885; E. (R.) schaden- mos, 2002; Brown and Guttman, 2002;
bergi Fischer, 1885) were transferred to the Fig. 1). Widespread distributions such as this
genus by Boettger (1886) and Boulenger (1887). have been the focus of many recent studies
These four species represented the known (Brown et al., 2000; Siler and Brown, 2010;78 HERPETOLOGICAL MONOGRAPHS [No. 25
distinct evolutionary lineages (species). In this
paper we provide a phylogenetic analysis of all
of these taxa, fully describe each species, clarify
taxonomic boundaries, and provide the first
illustrations of all included species. We also
provide information on each species’ natural
history, ecology, and geographic distribution.
MATERIALS AND METHODS
Field Work, Sample Collection, and
Specimen Preservation
Fieldwork was conducted on Catanduañes,
Lapinig Grande, Leyte, Luzon, and Samar
islands, all in the Philippines (Fig. 1) between
2001 and 2009. Specimens were collected
between 900 and 1600 h, euthanized in
aqueous chloretone, dissected for tissue sam-
ples (liver preserved in 95% ethanol or flash
frozen in liquid nitrogen), fixed in 10%
formalin, and eventually (,2 mo) transferred
to 70% ethanol. Newly sequenced specimens
are deposited in US and Philippine museum
collections (see Acknowledgments and Ap-
pendix I, Additional Specimens Examined);
voucher information corresponding to data
FIG. 1.—Map of the Philippine islands, with island
labels provided for islands with representative samples
from GenBank sequences is included in
used for this study. The five recognized major Pleistocene Table 1. Museum abbreviations for specimens
Aggregate Island Complexes (PAICs), major island examined follow Leviton et al. (1985).
groups, and additional deep-water islands are labeled for
reference. Islands of the Romblon Island Group are Taxon Sampling and Outgroup Selection for
designated by the first letter of the island name (T, Tablas Phylogenetic Analyses
Island; R, Romblon Island; S, Sibuyan Island). Current
islands in the Philippines are shown in medium grey; light Because our primary goal was to estimate
gray areas enclosed in black 120-m bathymetric contours phylogenetic relationships among the various
indicate the hypothesized maximum extent of land during populations of Brachymeles samarensis we
the mid- to late Pleistocene. sequenced only a few exemplars (2–4) per
sampled population. We included samples of
Siler et al., 2010a,b; Welton et al., 2009, Lygosoma bowringi as an outgroup based on
2010a,b), which have revealed that few relationships presented in recent phylogenetic
endemic Philippine reptiles actually possess studies of the genus Brachymeles (Siler et al.,
broad distributions spanning these regional 2011b; Siler and Brown, 2011). Additionally, we
faunistic boundaries (reviewed by Brown and included samples of B. apus, B. bonitae, B.
Diesmos, 2009). cebuensis, B. lukbani, and B. minimus to explore
The goal of the present study is to revise the the sister group relationships within the B.
taxonomy of the B. samarensis complex such samarensis complex (Siler et al., 2011b; Siler
that individual units (species) represent inde- and Brown, 2011). A total of 28 ingroup samples
pendently evolving, cohesive lineage segments were used in the phylogenetic analyses.
(sensu de Queiroz, 1998, 1999; Frost and
Hillis, 1990; Simpson, 1961; Wiley 1978). DNA Extraction, Purification,
Comprehensive examination of all recently and Amplification
collected specimens from throughout the We extracted total genomic DNA from
known range of B. samarensis results in the tissues (Table 1) using the modified guanidine
reorganization of the species complex into six thiocyanate extraction method of Esselstyn et al.2011]
TABLE 1.—Summary of specimens corresponding to genetic samples included in the study, general locality, and GenBank accession number. SP 5 Sabah Parks Reference
Collection, KU 5 University of Kansas Natural History Museum, LSUHC 5 La Sierra University Herpetological Collections, TNHC 5 Texas Natural History Collections of
the Texas Memorial Museum of the University of Texas at Austin.
GenBank accession numbers
Species Voucher Locality ND1 ND2 a-enolase PTGER4
Lygosoma bowringii LSUHC 6970 West Malaysia JN981975 JN981989 JN981962 JN982001
Lygosoma bowringii LSUHC 6998 West Malaysia HQ907328 HQ907430 HQ906969 —
Brachymeles apus SP 06915 Malaysia, Borneo, Sabah, Mt. Kinabalu National Park HQ907331 HQ907433 HQ906972 HQ907533
Brachymeles bonitae KU 307747 Philippines, Polillo Island, Municipality of Quezon JN981976 JN981990 JN981963 JN982002
Brachymeles bonitae KU 326080 Philippines, Polillo Island, Municipality of Quezon JN981977 JN981991 JN981964 JN982003
Brachymeles minimus KU 308131 Philippines, Catanduanes Island, Municipality of Gigmoto HQ907404 HQ907508 HQ907039 HQ907608
Brachymeles minimus KU 308129 Philippines, Catanduanes Island, Municipality of Gigmoto HQ907405 HQ907509 HQ907040 HQ907609
Brachymele lukbani KU 313602 Philippines, Luzon Island, Municipality of Labo HQ907407 HQ907511 HQ907042 HQ907611
Brachymele lukbani KU 313596 Philippines, Luzon Island, Municipality of Labo HQ907408 HQ907512 HQ907043 HQ907612
Brachymeles cebuensis KU 320419 Philippines, Cebu Island, Municipality of Carcar HQ907410 HQ907514 HQ907045 HQ907614
Brachymeles cebuensis KU 320421 Philippines, Cebu Island, Municipality of Carcar HQ907411 HQ907515 HQ907046 HQ907615
Brachymeles libayani2 KU 320430 Philippines, Lapinig Grande Island, Municipality of Carlos P. Garcia JN981978 — JN981965 —
Brachymeles libayani2 KU 320458 Philippines, Lapinig Grande Island, Municipality of Carlos P. Garcia JN981979 JN981992 — JN982004
Brachymeles libayani1 KU 320466 Philippines, Lapinig Grande Island, Municipality of Carlos P. Garcia JN981980 JN981993 JN981966 JN982005
Brachymeles libayani2 KU 320445 Philippines, Lapinig Grande Island, Municipality of Carlos P. Garcia JN981981 — JN981967 —
Brachymeles bicolandia1 KU 324003 Philippines, Luzon Island, Municipality of Tobaco JN981982 JN981994 JN981968 JN982006
Brachymeles bicolandia2 KU 324016 Philippines, Luzon Island, Municipality of Tobaco JN981983 JN981995 JN981969 JN982007
Brachymeles bicolandia2 KU 324010 Philippines, Luzon Island, Municipality of Tobaco JN981984 JN981996 JN981970 JN982008
Brachymeles bicolandia2 KU 324004 Philippines, Luzon Island, Municipality of Tobaco JN981985 JN981997 JN981971 JN982009
Brachymeles brevidactylus3 TNHC 62469 Philippines, Luzon Island, Municipality of Sorsogon JN981986 JN981998 JN981972 JN982010
HERPETOLOGICAL MONOGRAPHS
Brachymeles brevidactylus1 KU 324017 Philippines, Luzon Island, Municipality of Sorsogon JN981987 JN981999 JN981973 JN982011
Brachymeles cobos2 KU 324025 Philippines, Catanduanes Island, Municipality of San Miguel JN981988 JN982000 JN981974 JN982012
Brachymeles cobos2 KU 324020 Philippines, Catanduanes Island, Municipality of San Miguel HQ907349 HQ907450 HQ906989 HQ907550
Brachymeles cobos2 KU 324022 Philippines, Catanduanes Island, Municipality of San Miguel HQ907350 HQ907451 HQ906990 HQ907551
Brachymeles samarensis KU 310850 Philippines, Samar Island, Municipality of Taft HQ907416 HQ907520 HQ907051 HQ907620
Brachymeles samarensis KU 310851 Philippines, Samar Island, Municipality of Taft HQ907417 HQ907521 HQ907052 HQ907621
Brachymeles samarensis KU 310849 Philippines, Samar Island, Municipality of Taft HQ907413 HQ907517 HQ907048 HQ907617
Brachymeles paeforum2 KU 311225 Philippines, Leyte Island, Municipality of Baybay HQ907414 HQ907518 HQ907049 HQ907618
Brachymeles paeforum2 KU 311226 Philippines, Leyte Island, Municipality of Baybay HQ907418 HQ907522 HQ907053 HQ907622
Brachymeles paeforum2 KU 311227 Philippines, Leyte Island, Municipality of Baybay HQ907419 HQ907523 HQ907054 HQ907623
1
Holotype.
2
Paratopotype.
3
Paratype.
7980 HERPETOLOGICAL MONOGRAPHS [No. 25
TABLE 2.—Models of evolution selected by Akaike Information Criterion (AIC) and applied for partitioned, Bayesian
phylogenetic analyses.1
Partition AIC Model Model applied Number of characters
ND1, 1st codon position GTR + I + G GTR + G 322
ND1, 2nd codon position GTR + I + G GTR + G 322
ND1, 3rd codon position GTR + I + G GTR + G 322
ND2, 1st codon position TVM + I + G GTR + G 287
ND2, 2nd codon position GTR + I + G GTR + G 287
ND2, 3rd codon position TVM + I + G GTR + G 287
a-enolase TVMef + G GTR + G 261
PTGER4 HKY + I + G HKY + G 490
1
The model GTR + G was used for partitioned RAxMLHPC analyses.
(2008). The mitochondrial NADH dehydroge- random addition-sequence replicates and tree
nase subunit 1 (ND1), NADH dehydrogenase bisection and reconnection (TBR) branch
subunit 2 (ND2), and the protein-coding nuclear swapping. To assess heuristic support, non-
loci, a-enolase and PTGER4, were completely parametric bootstrapping was conducted us-
sequenced for nearly all samples using the ing 1000 replicates, each with 100 random
primers and protocols provided in Siler et al. addition-sequence replicates and TBR branch
(2011b) and Siler and Brown (2011). We swapping.
purified polymerase chain reaction templates Partitioned maximum likelihood (ML) anal-
with 1 mL of a 20% solution of ExoSAP-IT yses were conducted in RAxMLHPC v7.04
(US78201, Amersham Biosciences, Piscataway, (Stamatakis, 2006). The alignment was parti-
NJ). Cycle-sequencing reactions were complet- tioned into eight regions consisting of the
ed with the same primers and ABI Prism codon positions of ND1 and ND2, and the
BigDye Terminator chemistry (Ver. 3.1; Applied two nuclear loci, a-enolase and PTGER4,
Biosystems, Foster City, CA). Cycle-sequenc- following the methods of Siler et al. (2011b)
ing products were purified with Sephadex Me- and Siler and Brown (2011). Analyses that
dium (NC9406038, Amersham Biosciences) partition protein-coding genes by codon posi-
in Centri-Sep 96 spin plates (CS-961, Prince- tion have been shown to improve resulting
ton Separations, Princeton, NJ). We analyzed inferences (Brandley et al., 2005). The partitions
purified products using an ABI Prism 3130xl were run under the same generalized time-
Genetic Analyzer (Applied Biosystems), and reversible model (GTR + C) with 100 replicate
gene sequences were assembled with Sequen- best-tree inferences. Each inference was per-
cher 4.8 (Gene Codes Corp., Ann Arbor, MI). formed with a random starting tree and relied
on the rapid hill-climbing algorithm (Stamatakis,
Alignment and Phylogenetic Analysis 2006). Clade support was assessed with 1000
Initial alignments of the gene regions were bootstrap pseudoreplicates. We considered
produced in Muscle v3.7 (Edgar, 2004) and branches receiving $70% bootstrap support to
manual adjustments were made in MacClade be well supported (Wilcox et al., 2002).
4.08 (Maddison and Maddison, 2005). No The Akaike Information Criterion, as im-
instances of insertions or deletions, or ambig- plemented in jModeltest v0.1.1 (Guindon and
uously aligned regions, were observed in the Gascuel, 2003; Posada, 2008), was used to
data, and all data were used for analyses. The select the best model of nucleotide substitu-
final alignment consisted of 2570 nucleotide tion for each partition (Table 2). The best-
positions. fit model for each of the eight partitions
Phylogenetic analyses were conducted us- (Table 2) was used for Bayesian analyses
ing parsimony and likelihood optimality crite- performed in MrBayes 3.1 (Ronquist and
ria, as well as Bayesian methods. Parsimony Huelsenbeck, 2003). The same partitioning
(MP) analyses were conducted in PAUP* 4.0 strategy used for ML analyses was used for
(Swofford, 2002) with all characters weighted Bayesian inferences. Searches over tree space
equally. Most-parsimonious trees were esti- were conducted with four runs, each with four
mated using heuristic searches with 1000 chains, and were run for 2 3 107 generations.2011] HERPETOLOGICAL MONOGRAPHS 81
Trees were sampled every 1000 generations, 1978). We consider as distinct lineages those
with 4000 samples discarded as burn-in; this populations that are morphologically and
left 16,001 post–burn-in trees from each run genetically distinct, especially if allopatric.
included in the posterior distribution of Lineage-based species concepts have been
topologies. Visual inspection for chain statio- employed in the recognition of Philippine
narity and high effective sample size (ESS) biodiversity (Brown et al., 2000, 2008b, 2009;
values (equal to or greater than 800) was Brown and Diesmos, 2002; Brown and Gutt-
conducted within the program Tracer v1.4 man, 2002; Gaulke et al., 2007; Siler and
(Rambaut and Drummond, 2007). Additional- Brown, 2010; Welton et al., 2009, 2010a,b)
ly, correlations of split frequencies and cumu- due to the highly partitioned nature of the
lative split frequencies were examined using archipelago (Brown and Diesmos, 2009), and
the program AWTY (Nylander et al., 2008). We because the geological history of the islands
considered topologies with posterior probabil- has been so well documented (Hall, 2002;
ities $0.95 to be well supported (Leaché and Voris, 2000; Yumul et al., 2009). In this study
Reeder, 2002; Wilcox et al., 2002). we use an estimate of phylogenetic relation-
ships as a guide for delimiting species but
Morphological Data restrict our diagnoses of new species to those
We examined fluid-preserved specimens populations that are clearly identified by
(Appendix I) for variation in qualitative and diagnostic differences in nonoverlapping mor-
mensural characters. Sex was determined by phological character states.
gonadal inspection, and measurements were
taken to the nearest 0.1 mm with digital
RESULTS
calipers by C. D. Siler. X-rays were taken with
a company cabinet X-ray on Kodak MIN-R Phylogeny
2000 film exposed at 5 mA and 30 V for 1 min Of 2570 mitochondrial and nuclear charac-
15 s. ters, 848 were parsimony-informative. The
Meristic and mensural characters were maximum parsimony analysis inferred 10 most
chosen based on Siler et al. (2009, 2010a,b): parsimonious trees (tree length 5 2084;
snout–vent length (SVL), axilla–groin distance topology not shown; bootstrap support sum-
(AGD), total length, midbody width (MBW), marized in Fig. 2). The resulting 100 inferenc-
midbody height (MBH), tail length (TL), tail es from the partitioned RAxML ML analysis
width (TW), tail height (TH), head length show an average likelihood score 2lnL
(HL), head width (HW), head height, snout– 12011.371112, with a single inference having
forearm length (SnFa), eye diameter (ED), the highest likelihood score of 2lnL
eye–narial distance (END), snout length 12011.367644. Trees recovered from ML,
(SNL), internarial distance (IND), forelimb MP, and Bayesian analyses are topologically
length (FLL), hind limb length (HLL), mid- identical. No inferences support the monophy-
body scale-row count, paravertebral scale-row ly of Brachymeles samarensis. All analyses
count, axilla–groin scale-row count, Finger-III recover two reciprocally monophyletic clades
lamellae count, Toe-IV lamellae count, supra- within the B. samarensis complex, each char-
labial count, infralabial count, supraciliary acterized by the presence of multiple highly
count, and supraocular count. Additionally, divergent, genetically distinct lineages (Fig. 2).
we counted the number of presacral vertebrae The Leyte Island and Lapinig Group
from X-ray images of specimens. In the Islands populations were recovered as a clade,
description, ranges are followed by mean 6 sister to B. cebuensis from Cebu Island
SD in parentheses. (Fig. 2). Topotypic B. samarensis from Samar
Island were recovered as sister to a clade of
Species Concept two limbless species of Brachymeles (B.
We follow the general lineage concept of minimus and B. lukbani) and the Luzon and
species (de Queiroz, 1998, 1999) as a logical Catanduañes island populations of B. samar-
extension of the evolutionary species concept ensis (Fig. 2). Within Luzon Island, two se-
(Frost and Hillis, 1990; Simpson, 1961; Wiley, parate lineages were recovered from the Bicol82 HERPETOLOGICAL MONOGRAPHS [No. 25 FIG. 2.—Maximum likelihood (ML) estimate of combined mitochondrial and nuclear data for samples of Brachymeles used for this study (preferred ML tree, 2lnL 12011.367644; ND1, ND2, a-enolase, PTGER4). Nodes are shown with numerical values corresponding to maximum parsimony bootstrap proportions, maximum likelihood bootstrap proportions, and Bayesian posterior probabilities, respectively. Terminals are labeled with taxonomic names, fore- and hind limb digit states, and number of presacral vertebrae. Peninsula, with no support for their mono- both limbless species, nested within a didactyl phyly (Fig. 2). clade of species formerly part of B. samarensis In addition to the finding of paraphyly of B. (topotypic B. samarensis, B. sp. nov. [Catan- samarensis complex members, several other duañes Island], B. sp. nov. [Southern Bicol recognized species were recovered as part of Peninsula, Luzon Island], B. sp. nov. (Central this clade. Brachymeles cebuensis, one of only Bicol Peninsula, Luzon Island]; Fig. 2). All two recognized species to have unequal three previously recognized species (B. ce- numbers of fingers and toes (Brown and buensis, B. lukbani, B. minimus) have geo- Alcala, 1980), was supported as part of a clade graphical distributions that overlap, or are in of species (B. cebuensis + B. sp. nov. [Leyte close proximity to, the known ranges of other Island] + B. sp. nov. [Lapinig Group Islands]) species in the B. samarensis complex (Fig. 3). with three fingers and two or three toes All analyses result in the strong support (Fig. 2). Sister to this three-finger clade, all of six genetically distinct lineages within analyses recovered B. minimus and B. lukbani, the Brachymeles samarensis species complex
2011] HERPETOLOGICAL MONOGRAPHS 83 FIG. 3.—(Left) Map of the Philippine islands showing previously recognized distribution of Brachymeles samarensis (indicated by black shaded islands), and recognized distributions of other members of the B. samarensis complex (indicated by dark gray shapes). (Right) Hypothesized distributions of B. bicolandia, B. brevidactylus, B. cobos, B. libayani, B. paeforum, and B. samarensis in the eastern-central Philippines. The sampling localities are indicated by black shapes, and the hypothesized geographic range of each species is indicated by shaded islands and dashed lines, with shapes and shades of islands corresponding to the map’s key. (Fig. 2). Uncorrected pairwise sequence di- Mitochondrial sequence divergences among vergences are low within the lineages defined the other three lineages within the B. samar- here as species and high between these ensis species complex (B. samarensis [Samar lineages (Table 3). Percentage of divergences Island], B. sp. nov. [Lapinig Group islands], B. for the mitochondrial and for the nuclear data, sp. nov. (Leyte Island]) are greater than 9.2% respectively, show that the monophyletic (Table 3; Fig. 2). Intraspecific sequence diver- lineages defined by our phylogenetic analyses gences are low (0.0–1.9%) in comparison to (B. samarensis, B. sp. nov. [Leyte Island], B. divergences among monophyletic lineages. sp. nov. [Lapinig Group Islands], B. sp. nov. Additionally, moderate levels of sequence diver- [Catanduañes Island], B. sp. nov. [Southern gence are observed for nuclear sequence data Bicol Peninsula, Luzon Island], B. sp. nov. (Table 3). (Central Bicol Peninsula, Luzon Island]) are distinguished from congeners by levels of Morphology genetic divergence similar to, or greater than, Variation in morphological characters those between previously defined species— (Tables 4–6) mirrors the results observed in viz., B. bonitae, B. cebuensis, B. minimus, and phylogenetic analyses, and supports the rec- B. lukbani (Table 3; Fig. 2). The two most ognition of six B. samarensis group lineages. closely related lineages, B. sp. nov. (Catan- Characters differing among these six lineages duañes Island) and B. sp. nov. (Southern Bicol include digit number, presacral vertebrae Peninsula, Luzon Island), are separated by number, degree of digit development, head 4.1–4.7% mitochondrial sequence divergence. and body scale counts and patterns, and
84 HERPETOLOGICAL MONOGRAPHS [No. 25
paeforum, B. libayani, B. bicolandia, B. cobos, B. brevidactylus, B. bonitae, B. cebuensis, B. lukbani, and B. minimus (Fig. 2). Percentages on the diagonal represent pigmentation patterns (Tables 4–6; species
TABLE 3.—Uncorrected pairwise sequence divergence (%) for mitochondrial data (below diagonal) and nuclear data (above diagonal), for Brachymeles samarensis, B.
accounts below), all of which are typical
B. minimus
3.2–4.4
3.2–3.6
3.4–3.5
0.7–0.8
0.0
3.4
3.3
3.5
3.2
3.2
morphological diagnostic characters employed
historically by taxonomists working with this
genus (review: Brown and Alcala, 1980). We
observed no intraspecific mensural or meristic
B. lukbani
2.7–3.1
3.0–3.6
2.6–3.2
2.8–3.1
2.7–3.1
2.6–3.0
2.8–3.0
4.9–5.0
differences between the sexes of any of the six
0.1
3.0
species.
Superficially, the six lineages within the B.
samarensis complex appear morphologically
similar in overall body size and general shape;
B. cebuensis
1.4–2.8
1.1–1.6
1.4–1.6
however, upon closer inspection, three dis-
0.0
13.4
13.7
1.2
1.3
1.5
2.2
tinct body forms are observed. Among the six
lineages, two are tridactyl (B. sp. nov. [Leyte
Island] and B. sp. nov. [Lapinig Group
intraspecific genetic diversity for mitochondrial data (bolded for emphasis).
16.5–16.7
15.0–15.1
15.0–15.1
Islands]), three are bidactyl (B. samarensis,
B. bonitae
2.2–4.8
2.3–2.7
2.6–2.8
1.4
2.4
2.6
2.7
B. sp. nov. [Catanduañes Island] and B. sp.
nov. [Central Bicol Peninsula, Luzon Island]),
and one is bidactyl, but with small, highly
reduced, and nearly imperceptible claws (B.
B. brevidactylus
15.7–16.3
14.8–14.9
11.8–11.9
1.6–2.0
1.2–3.3
0.4–1.1
0.7–1.2
sp. nov. [Southern Bicol Peninsula, Luzon
1.9
11.6
1.5
Island]). Additionally, numerous nonoverlap-
ping differences were detected in meristic,
mensural, osteological, and color pattern
characters for each complex member, readily
15.7–16.2
11.0–11.2
0.0–0.2
1.6–3.6
0.5–0.9
9.3–9.5
B. cobos
defining six distinct lineages within the com-
1.6
2.0
14.7
11.2
plex (Tables 4–6). In summary, each lineage
(most of which are allopatric) possesses unique
and nonoverlapping suites of diagnostic char-
B. bicolandia
15.8–16.5
14.6–14.8
11.5–11.9
11.6–12.0
acter states of morphology, perfectly cor-
0.4–0.7
1.1–1.5
1.8–2.2
1.4–2.8
4.1–4.7
9.4–9.6
responding to the six clades defined in the
phylogenetic analyses of DNA sequence data.
Taxonomic Conclusions
14.9–15.8
14.5–14.8
14.7–15.4
15.5–16.5
12.0–12.4
14.8–15.3
13.8–14.3
B. libayani
0.0–0.1
1.8–2.8
0.2–2.8
Our estimate of phylogeny (Fig. 2); biogeo-
graphically separate ranges of island or region
endemic species; diagnostic, nonoverlapping
morphological character states; and genetic
distances between the taxa (Table 3) indicate
B. paeforum
9.2–10.5
14.1–14.5
14.4–14.6
14.6–14.9
15.0–15.6
11.6–11.7
14.1–14.4
14.0–14.2
0.0–0.7
the distinctiveness of a new species from
2.2
Catanduañes Island, two new species from the
Bicol Peninsula of Luzon Island (one from the
central Bicol region and one from the extreme
B. samarensis
14.6–15.2
14.5–14.7
14.4–14.5
14.9–15.0
16.9–17.3
southern tip of the peninsula), a new species
0.0
14.3
16.2
14.1
14.1
from the Lapinig Group Islands, and a new
species from Leyte Island (Table 3; Fig. 2).
Each of the six species of the B. samarensis
brevidactylus
complex is morphologically distinct from each
samarensis
bicolandia
paeforum
cebuensis
other and all other known species in the
minimus
libayani
lukbani
bonitae
genus, and the 11 species of Brachymeles
cobos
included in phylogenetic analyses also are
genetically distinct. All monophyletic lineages,
B.
B.
B.
B.
B.
B.
B.
B.
B.
B.2011] HERPETOLOGICAL MONOGRAPHS 85
with the exception of the two occurring on (Tables 4 and 5); from B. brevidactylus by
the Bicol Peninsula of Luzon Island, are having fewer presacral vertebrae and fewer
endemic to single islands within two isolated paravertebral scale rows (Tables 4 and 5); and
PAICs, thereby providing additional support from B. bonitae by having longer relative hind
for the distinctiveness of each clade’s evo- limb lengths, fewer presacral vertebrae, fewer
lutionary history and lineage integrity. Ac- paravertebral scale rows, six supraciliaries, five
cordingly, we recognize B. samarensis as a supraoculars, and the presence of contact
species that occurs only on Samar Island in between frontoparietals (Tables 4 and 5).
the eastern Visayan (central) Philippine islands Brachymeles samarensis can be distin-
(e.g., Mindanao PAIC; Fig. 3), and hereby guished from all limbless species of Brachy-
recognize the five additional lineages within meles (B. apus, B. lukbani, B. minimus, B.
the B. samarensis species complex as new miriamae, B. vermis) by having limbs; and
species. from all pentadactyl species of Brachymeles
(B. boholensis, B. boulengeri, B. bicolor, B.
TAXONOMIC ACCOUNTS gracilis, B. kadwa, B. makusog, B. mind-
Brachymeles samarensis Brown 1956: 6 orensis, B. orientalis, B. schadenbergi, B.
Figs. 3–4 talinis, B. taylori, B. tungaoi, B. vindumi) by
Brachymeles samarensis, Brown, 1956, having nonpentadactyl limbs, shorter adult
Type locality: Guinuan, Samar Island, Philip- forelimb lengths (less than 2.6 mm vs. greater
pines (holotype: FMNH 44472); Brown and than 5.9 mm), shorter adult hind limb lengths
Alcala, 1970, 1980; Brown and Rabor, 1967. (less than 3.1 mm vs. greater than 10.3 mm),
Diagnosis.—Brachymeles samarensis can and a narrower body (less than 6.4 mm vs.
be distinguished from congeners by the greater than 7.9 mm), and by the absence of a
following combination of characters: (1) body postnasal scale and auricular opening (vs.
size small (SVL 57.9–66.1 mm), (2) limbs presence).
bidactyl, (3) limb length short, (4) supralabials Description (based on holotype description
six, (5) infralabials six, (6) supraciliaries six, (7) and six referred specimens).—Details of the
supraoculars five, (8) midbody scale rows 19– head scalation of an adult female are shown in
22, (9) axilla–groin scale rows 66–69, (10) Fig. 5. Measurements and character states of
paravertebral scale rows 86–88, (11) pineal the holotype are provided below in square
eye spot present, (12) prefrontals not contact- brackets. Body small, slender; SVL 57.9 mm
ing on midline, (13) frontoparietals contact, for males, maximum SVL 66.1 mm for
(14) mental/first infralabial fusion absent, (15) females, [43.5 mm, juvenile] (Tables 4 and
postnasals absent, (16) enlarged chin shields 5); head weakly differentiated from neck,
in three pairs, (17) nuchal scales differentiat- nearly as wide as body, HW 7.3–9.2% (8.3 6
ed, (18) fourth supralabial below eye midline, 0.7) SVL, 91.4–117.8% (102.7 6 10.8) HL;
(19) auricular opening absent, (20) presacral HL 36.6–42.5% (38.8 6 2.1) SnFa; SnFa
vertebrae 45, and (21) uniform body color 18.8–23.5% (20.9 6 1.6) SVL; snout short,
(Tables 4 and 5). bluntly rounded in dorsal and lateral profile,
Comparisons.—Characters distinguishing SNL 50.9–55.3% (53.3 6 1.8) HL; ear
Brachymeles samarensis from all nonpenta- completely hidden by scales; eyes small, ED
dactyl, limbed species of Brachymeles are 1.3–1.6% (1.4 6 0.1) SVL, 17.0–18.7% (17.6
summarized in Tables 4 and 5. Brachymeles 6 0.6) HL, 42.6–48.0% (45.8 6 2.1) END,
samarensis most closely resembles B. bicolan- pupil subcircular; body slightly depressed,
dia, B. cobos, B. brevidactylus, and popula- nearly uniform in thickness, MBW 109.1–
tions of B. bonitae, the only other bidactyl 150.6% (130.4 6 14.9) MBH; scales smooth,
species. However, B. samarensis differs from glossy, imbricate; longitudinal scale rows at
these four taxa by having midbody scale rows midbody 19–22 [22]; paravertebral scale rows
as few as 19 and axilla–groin scale rows as few 86–88 [86]; axilla–groin scale rows 66–69;
as 66 (Table 5). Brachymeles samarensis limbs short, poorly developed, with digits
further differs from B. bicolandia by having reduced to two claws on both forelimbs and
fewer presacral vertebrae and six infralabials hind limbs, finger and toe lamellae absent;86 HERPETOLOGICAL MONOGRAPHS [No. 25
TABLE 4.—Summary of meristic and mensural characters in all known limbed, nonpentadactyl species of Brachymeles.
Sample size, body length, and total length (TotL) among males and females, and general geographical distribution
(Pleistocene Aggregate Island Complexes [PAIC], sensu Brown and Diesmos, 2002) are included for reference. SVL 5
snout–vent length, TotL 5 total length, TL 5 tail length, FLL 5 forelimb length, HLL 5 hind limb length, ToeIVlam
5 toe-IV lamellae count. (SVL, TotL, MBW, FLL, and HLL given as range over mean 6 SD; all body proportions given
as percentage over mean 6 SD).
B. samarensis (1 B. paeforum (3 males, B. libayani (10 males, B. bicolandia (6 males, B. cobos B. brevidactylus (1
male, 5 females) 9 females) 25 females) 10 females) (9 females) male, 2 females)
Lapinig Group Central Bicol Southern Bicol
Range Samar Island Leyte PAIC Islands Peninsula Catanduañes Island Peninsula
SVL (female) 62.4–66.1 47.2–61.4 52.8–66.1 46.4–67.4 54.0–64.4 54.0, 60.0
(63.4 6 1.5) (56.5 6 4.2) (58.6 6 3.3) (59.0 6 6.7) (58.7 6 3.5)
SVL (male) 57.9 62.4–66.1 52.7–57.4 56.4–66.1 — 56.8
(63.4 6 1.5) (56.0 6 1.6) (61.7 6 3.5)
TotL (female) 97.7–112.9 99.5–108.5 91.4–111.2 94.1–112.7 102.2–109.4 92.3, 95.2
(107.3 6 8.3) (102.6 6 5.1) (102.2 6 6.4) (102.1 6 8.8) (106.2 6 2.7)
TotL (male) 93.0 106.7–114.6 92.1–103.4 99.6–107.9 — 102.0
(110.6 6 5.6) (99.4 6 4.5) (104.1 6 4.2)
TL/SVL 57–81 69–79 63–84 58–93 76–96 59–80
(68 6 12) (75 6 4) (77 6 6) (75 6 12) (88 6 8) (70 6 11)
FLL 1.1–2.6 1.3–1.7 1.1–1.8 1.1–1.9 1.4–2.1 1.1–1.5
(1.8 6 0.5) (1.5 6 0.1) (1.3 6 0.2) (1.4 6 0.3) (1.7 6 0.2) (1.5 6 0.2)
FLL/SVL 2–4 2–4 2–3 2–4 3 2–3
(3 6 1) (3 6 0) (2 6 0) (2 6 0) (3 6 0) (3 6 0)
HLL 2.5–3.1 2.3–3.0 2.0–2.7 1.9–3.1 2.5–3.6 2.1– 2.7
(2.9 6 0.2) (2.6 6 0.3) (2.4 6 0.2) (2.6 6 0.3) (3.0 6 0.3) (2.5 6 0.3)
HLL/SVL 4–5 4–5 3–5 3–5 4–6 4–5
(5 6 0) (4 6 0) (4 6 0) (4 6 1) (5 6 1) (4 6 0)
ToeIVlam 0 0 0 0 0 0
FLL 2.4–5.7% (3.9 6 1.3) AGD, 1.8–3.9% anterior loreal about as long as and slightly
(2.9 6 0.9) SVL; HLL 5.3–7.2% (6.2 6 0.7) higher than posterior loreal; preocular one;
AGD, 4.0–5.0% (4.6 6 0.4) SVL [6.9]; tail not presubocular one; supraciliaries six, the ante-
as wide as body, gradually tapered towards riormost contacting prefrontal and separating
end, TW 70.2–82.6% (76.7 6 5.0) MBW, TL posterior loreal from first supraocular, poste-
56.5–80.6% (68.4 6 11.6) SVL. riormost extending to posterior edge of fifth
Rostral projecting onto dorsal snout to level supraocular; subocular scale row single, com-
in line with middle of nasal, broader than plete, in contact with supralabials; lower
high, in contact with frontonasal; frontonasal eyelid with one row of scales; supralabials
wider than long; nostril ovoid, in center of six, first twice the anteroposterior length of
single trapezoidal nasal, longer axis directed others, fourth below eye midline; infralabials
anteroventrally and posterodorsally; suprana- six (Fig. 4).
sals present, large, broadly separated; postna- Mental wider than long, in contact with first
sals absent; prefrontals moderately separated; infralabials; postmental single, enlarged, its
frontal octagonal-shaped, its anterior margin width equal to width of mental; followed by
in moderate contact with frontonasal, in con- three pairs of enlarged chin shields, first pair
tact with first two anterior supraoculars, 33 in broad medial contact, second pair wider
wider than anterior supraocular; supraoculars than first, broadly separated by single medial
five; frontoparietals moderate, in broad medi- scale, third pair separated by three medial
al contact, each frontoparietal in contact with scales (Fig. 4).
supraoculars 2–4; interparietal moderate, Scales on limbs smaller than body scales;
its length roughly equal to midline length of scales on dorsal surfaces of digits wrapping
frontoparietal, longer than wide, diamond- around lateral edges of digits; lamellae absent;
shaped, wider anteriorly; parietals broader palmar surfaces of hands and plantar surfaces
than frontoparietals, in broad contact behind of feet with several small, irregular scales,
interparietal; nuchals enlarged; loreals two, each with irregular raised anterior edges;2011] HERPETOLOGICAL MONOGRAPHS 87
TABLE 4.—Extended
B. muntingkamay B. tridactylus (9 males, B. bonitae (6 males, B. cebuensis B. elerae (2 males, B. pathfinderi B. wrighti (1
(12 females) 11 females) 7 females) (8 females) 1 female) (14 m, 23 females) male, 1 female)
Mindoro & Luzon Mindanao
Luzon Island Visayan PAIC PAICs Cebu Island Luzon Island Island Luzon Island
61.8–81.3 45.5–59.1 49.7–59.8 51.5–67.9 68.2, 71.9 55.8–68.3 113.0 1
(73.6 6 5.9) (52.1 6 5.0) (56.4 6 3.9) (62.0 6 3.4)
N/A 55.7–78.3 65.1–80.0 (61.8 6 5.3) 71.5 54.5–65.1 25.8
(68.5 6 7.4) (73.5 6 6.4) N/A (59.4 6 3.8)
107.4–136.0 102.6–154.1 93.4–150.4 104.3–128.0 109.9, 131.9 111.1–133.2 205.6
(124.0 6 8.6) (132.6 6 14.0) (126.7 6 19.9) (119.0 6 8.5) (119.7 6 8.2)
N/A 105.3–133.67 102.6–144.5 N/A 101.4–107.0 216.4
(115.9 6 15.4) (121.3 6 15.6) N/A (104.2 6 4.0)
50–79 69–112 35–93 78–115 61–84 69–95 72, 82
(65 6 10) (92 6 12) (69 6 18) (92 6 13) (72 6 16) (84 6 10)
2.4–3.0 1.5–2.5 1.0–1.5 1.1–1.8 3.3–3.5 4.4–6.9 7.5, 7.5
(2.7 6 0.2) (2.0 6 0.3) (1.3 6 0.1) (1.5 6 0.3) (3.4 6 0.1) (5.8 6 0.5)
3–4 2–3 1–2 2–3 5–5 8–11 6, 7
(4 6 0) (3 6 0) (2 6 0) (2 6 0) (5 6 0) (10 6 1)
5.3–6.0 2.6–3.6 1.3–2.0 2.3–3.0 4.3–5.4 8.4–12.9 10.9, 13.9
(5.7 6 0.2) (3.1 6 0.3) (1.6 6 0.2) (2.7 6 0.3) (5.0 6 0.6) (10.8 6 1.0)
7–9 3–6 2–3 3–5 6–8 15–21 10, 11
(8 6 1) (5 6 1) (2 6 0) (4 6 0) (7 6 1) (18 6 1)
0 0 0 0 3 5–8 4, 5
fingers equal in size; toes unequal in size, minor differences, including a dark brown
middle digit greatest in length, first and third body color and dark brown to black streaks of
digits equal in length. pigmentation.
Coloration in preservative.—Body ground Variation.—Morphometric variation of the
color medium brown, each dorsal scale light series is summarized in Table 6. We observed a
brown posteriorly, with a dark auburn streak single instance of digit variation, where one
on the anterior two-thirds to half of the scale. specimen (KU 310849) has no fingers and two
Dark streaks on each scale consist of four to toes. All specimens have two loreals with the
seven thin longitudinal lines with smudges exception of a single specimen (KU 310852),
between lines. Streaks on scales present which has a single loreal on the right side of the
around entire body, more distinct on venter, body resulting from the fusion of the two scales
where posterior ends of scales are cream, in this position. Additionally, the first and
giving greater contrast. Posterior edge of all second pairs of enlarged chin shields are equal
body scales transparent. Forelimb and hind in width among all specimens with the exception
limb scales same color as body scales. of a single specimen (KU 310850), in which the
Precloacal scale coloration matches surround- width of the second pair of enlarged chin shields
ing ventral scale coloration. Head scales is greater than the width of the first pair.
mottled light and dark brown, matching dorsal Distribution.—Brachymeles samarensis is
background coloration. Supraocular, rostral, known only from Samar Island (Fig. 3).
nasal, supranasal, and supralabials scales gray- Ecology and natural history.—Brachymeles
cream. Mental, infralabial, postmental, and samarensis occurs in primary- and secondary-
chin shields scales cream with slight brown growth forest habitats. In contrast to the other
mottling, lighter than bordering ventral scales. members of the B. samarensis complex, this
Coloration in life.—Coloration in life close- species appears to be a forest obligate, and
ly matches the coloration in preservative with was only observed within rotting logs in88 HERPETOLOGICAL MONOGRAPHS [No. 25
TABLE 5.—Summary of qualitative diagnostic characters (present, absent) in all known limbed, nonpentadactyl species of
Brachymeles. The pairs of enlarged scales posterior to the postmental scale are abbreviated as chin shield pairs with
reference to the first, second, and third pairs (when present). In cases of scale count variation within species, numbers of
individuals showing specific counts are given in parentheses. PSV 5 presacral vertebrae, MBSR 5 midbody scale-row
count, AGSR 5 axilla–groin scale-row count, PVSR 5 paravertebral scale-row count, SL 5 supralabial count, IFL 5
infralabial count, SC 5 supraciliary count, SO 5 supraocular count.
B. samarensis (1 male, B. paeforum B. libayani (10 B. bicolandia (6 B. cobos B. brevidactylus
5 females) (3 males, 9 females) males, 25 females) males, 10 females) (9 females) (1 male, 2 females)
Number of digits 2/2 3/3 3/3 2/2 2/2 2/2
(fore/hind)
PSV 45 47 47 46–49 45 47–48
MBSR 19–22 21–22 22–23 20–22 21–22 20
AGSR 66–69 71–74 72–75 68–73 68–72 73–77
PVSR 86–88 93–96 90–92 85–90 85–89 90–94
SL 6 (6) 6 (12) 6 (35) 6 (16) 6 (9) 6 (3)
IFL 6 (6) 5 (4) 6 (8) 5 (35) 5 (11) 6 (5) 6 (9) 6 (3)
SC 6 (6) 6 (12) 6 (35) 6 (16) 6 (9) 6 (3)
SO 5 (6) 5 (12) 5 (35) 5 (16) 5 (9) 5 (3)
Pineal eyespot Present Present Present Present Present Present
Prefrontal contact Absent Absent Point contact Absent Absent Absent
or Absent
Frontoparietal Present Present Present Present or Present Present
contact Absent
First chin shield Present Present or Present or Present or Present Absent
pair contact Absent Absent Absent
Thirdrd chin shield pair Present Present Present Present Present Present
Chin shield pair size 3,1#2 3,1#2 3,1,2 3,1,2 3,1#2 3,1,2
Chin shield pair 1(0); 2(1); 1(0/1); 2(1); 1(0/1); 2(1); 1(0/1); 2(1); 1(0); 2(1); 1(0); 2(1);
separation1 3(3) 3(3) 3(3) 3(3) 3(3) 3(3)
First/second loreal Present or Present or Present or Present or Absent Absent
fusion Absent Absent Absent Absent
Mental/first IFL fusion Absent Present or Present Present or Absent Absent
Absent Absent
Differentiated nuchals Present Present Present Present Present Present
Continuous subocular Present Present Present Present Present Present
scale row
Auricular opening Absent Absent Absent Absent Absent Absent
Dorsolateral stripes Absent Absent Absent Absent Absent Absent
Longitudinal rows Absent Absent Absent Absent Absent Absent
of dark spots
1
Parentheses show the number of small ventral scale rows separating each enlarged pair of chin shields.
2
Due to head damage in the nuchal region for both known specimens of B. wrighti, the presence of differentiated nuchals remains tentative.
secondary-growth forest. Three species of Other sympatric lizard species observed on
Brachymeles have been confirmed to occur Samar Island include the following: (Agamidae)
on Samar Island: B. gracilis hilong, B. orien- Bronchocela cristatella, Draco bimacula-
talis, and B. samarensis (Brown and Alcala, tus, D. ornatus, D. reticulates, Gonocephalus
1980; Siler and Brown, 2010). semperi, Hydrosaurus pustulatus; (Gekkoni-
We have evaluated this species against the dae) Cyrtodactylus annulatus, C. sumoroi,
International Union for Conservation of Gehyra mutilata, Gekko gecko, Gek. mind-
Nature (IUCN) criteria for classification, and orensis, Hemidactylus frenatus, H. platyurus,
find that it does not qualify for Critically Hemiphyllodactylus typus, Lepidodactylus aureo-
Endangered, Endangered, Vulnerable, or lineatus, L. planicaudus, Pseudogekko compressi-
Near Threatened status. Brachymeles samar- corpus; (Scincidae) Emoia atrocostata, Eutropis
ensis has been documented to have a broad multicarinata, Eu. multifasciata, Lamprolepis
geographic distribution across southern Samar smaragdina, Lipinia pulchella, Li. quadrivittata,
Island. We therefore classify this species as Sphenomorphus acutus, S. cumingi, S. fasciatus,
having Least Concern status (IUCN, 2011). S. jagori, S. cf. mindanensis, S. steerei, S.2011] HERPETOLOGICAL MONOGRAPHS 89
TABLE 5.—Extended
B. muntingkamay B. tridactylus B. bonitae B. cebuensis B. elerae (2 males, B. pathfinderi B. wrighti (1 male,
(12 females) (9 males, 11 females) (6 males, 7 females) (8 females) 1 females) (14 males, 23 females) 1 female)
3/3 3/3 0–2/0–2 3/2 4/4 5/4 4/4
42, 44 47 47–57 45 43 34 —
22–24 22–24 21–23 22–24 22–24 23–25 28, 28
65–70 70–79 73–90 65–69 63–67 44–48 85, 85
85–90 88–98 90–109 84–88 84–87 64–67 106, 108
6 (12) 6 (12) 7 (8) 6 (12) 7 (1) 6 (8) 6 (3) 6 (37) 6 (1) 7 (1)
6 (12) 6 (12) 7 (8) 5 (1) 6 (10) 7 (2) 6 (5) 7 (3) 6 (3) 6 (37) 7 (2)
6 (10) 7 (2) 5 (20) 5 (12) 6 (1) 6 (8) 5 (2) 6 (1) 5 (17) 6 (19) 6
5 (11) 6 (1) 4 (20) 4 (13) 5 (8) 4 (2) 5 (1) 5 (37) 5
Absent Present Present Present Absent Present Present
Present Absent Absent Present or Present Absent Present or
Absent Absent
Absent Absent Absent Present Present Present or Present
Absent
Absent Present or Absent Present Absent Absent Absent
Absent
Present Present Present Present Present Absent Present
3,1,2 3,1,2 3,2,1 153,2 1,3,2 1,2 2,1
1(1); 2(1); 1(0/1); 2(1); 1(1); 2(1); 1(0); 2(1); 1(1); 2(1); 1(1); 2(1) 1(1); 2(3)
3(3) 3(3) 3(3) 3(3) 3(3)
Absent Absent Absent Absent Absent Absent Absent
Absent Absent Present or Absent Absent Absent Absent
Absent
Absent Present Present Present Absent Absent Present2
Absent Absent Present Present Present Present Present
Absent Absent Absent Absent Absent Present Absent
Absent Absent Absent Absent Absent Present Absent
Present, around Present, vague Absent Absent Present, around Present, 6 Present
body to indistinct body
variegatus, Tropidophorus misaminus; (Varani- collected between 29 October and 8 Novem-
dae) Varanus cumingi samarensis. ber 2007.
Other paratypes.—One adult male (CAS-
Brachymeles paeforum sp. nov.
SU 26120), four adult females (CAS-SU
Figs. 3, 6, 7
26110, 26112, 26121–22), and two juveniles
Holotype.—PNM 9746 (CDS Field (CAS-SU 26115, 26123) collected between 1
No. 3418, formerly KU 311228), adult female, May and 4 June 1964, in Barrio Tambis,
collected under rotting logs in secondary-growth Municipality of Burauen, Leyte Province, Leyte
forest (1000 to 1230 h) on 8 November 2007 in Island, Philippines (11u009370N, 124u529190E;
the Sitio San Vicente Tree Nursery, Barangay WGS-84), by D. S. Rabor; one adult male
Pilim, Baybay City, Leyte Province, Leyte (CAS-SU 26771), two adult females (CAS-SU
Island, Philippines (10u439350N, 124u499050E; 26770, 26772), and one juvenile (CAS-SU
WGS-84), by CDS and J. Fernandez. 26773), collected between 10 June and 17 July
Paratopotypes.—One adult male (KU 1964, in the Municipality of Mahaplag, Leyte
311225), one adult female (KU 311229), and Province, Leyte Island, Philippines (10u359420N,
three juveniles (KU 311224, PNM 9747–48) 124u599130E; WGS-84), by D. S. Rabor.TABLE 6.—Summary of univariate morphological variation among mensural characters in series of Brachymeles samarensis, B. paeforum, B. libayani, B. bicolandia, B. cobos, 90
and B. brevidactylus. SVL 5 snout–vent length, AGD 5 axilla–groin distance, TotL 5 total length, MBW 5 midbody width, MBH 5 midbody height, TL 5 tail length, TW
5 tail width, TH 5 tail height, HL 5 head length, HW 5 head width, HH 5 head height, SnFa 5 snout–forearm length, ED 5 eye diameter, END 5 eye–narial distance,
SNL 5 snout length, IND 5 internarial distance, FLL 5 forelimb length, HLL 5 hind limb length.
B. brevidactylus (1
B. samarensis (1 male, 5 females) B. paeforum (3 males, 9 females) B. libayani (10 males, 25 females) B. bicolandia (6 males, 10 females) B. cobos (9 females) male, 2 females)
SVL (male) 57.9 59.7–64.1 (61.8 6 2.2) 52.7–57.4 (56.0 6 1.6) 56.4–66.1 (61.7 6 3.5) — 56.8
SVL (female) 62.4–66.1 (63.4 6 1.5) 47.2–61.4 (56.5 6 4.2) 52.8–66.1 (58.6 6 3.3) 46.4–67.4 (59.0 6 6.7) 54.0–64.4 (58.7 6 3.5) 54.0, 60.0
AGD (male) 40.3 44.1–48.2 (46.0 6 2.1) 39.5–43.6 (42.1 6 1.2) 42.9–52.0 (47.4 6 3.4) — 43.8
AGD (female) 45.3–50.3 (47.6 6 1.8) 35.6–46.5 (43.2 6 3.1) 39.4–50.6 (44.7 6 2.8) 32.9–52.3 (44.8 6 5.8) 41.0–49.1 (44.6 6 2.3) 45.5, 45.8
TotL (male) 93.0 106.7–114.6 (110.6 6 5.6) 92.1–103.4 (99.4 6 4.5) 99.6–107.9 (104.1 6 4.2) — 102.0
TotL (female) 97.7–112.9 (107.3 6 8.3) 99.5–108.5 (102.7 6 5.1) 91.4–111.2 (102.2 6 6.4) 94.1–112.7 (102.1 6 8.8) 102.2–109.4 (106.2 6 2.7) 92.3, 95.2
MBW (male) 5.7 5.0–5.3 (5.1 6 0.2) 4.0–4.7 (4.4 6 0.3) 4.2–5.1 (4.5 6 0.3) — 4.0
MBW (female) 5.2–6.4 (5.7 6 0.5) 3.7–5.0 (4.2 6 0.4) 3.4–5.9 (4.5 6 0.6) 3.8–4.8 (4.4 6 0.3) 4.2–5.4 (4.8 6 0.4) 3.2, 4.8
MBH (male) 4.0 2.9–4.5 (3.8 6 0.8) 3.2–4.7 (3.6 6 0.5) 3.0–5.1 (3.6 6 0.8) — 3.2
MBH (female) 4.3–4.8 (4.5 6 0.2) 3.0–4.7 (3.9 6 0.5) 2.9–5.3 (3.7 6 0.8) 2.9–5.0 (3.7 6 0.8) 3.4–4.4 (3.8 6 0.4) 2.5, 4.3
TL (male) 35.1 45.0–50.5 (47.8 6 3.9) 39.3–47.1 (43.6 6 3.4) 38.7–51.5 (43.2 6 7.2) — 45.2
TL (female) 35.3–50.4 (44.5 6 8.1) 41.0–47.1 (43.9 6 3.1) 38.3–50.1 (44.2 6 4.1) 32.3–52.1 (42.5 6 7.4) 45.6–53.5 (49.5 6 3.1) 38.3, 35.2
TW (male) 4.1 3.8–4.1 (4.0 6 0.2) 3.1–3.9 (3.4 6 0.2) 3.4–4.3 (3.7 6 0.3) — 3.7
TW (female) 4.1–4.6 (4.4 6 0.2) 3.3–3.8 (3.4 6 0.2) 2.8–4.6 (3.5 6 0.5) 2.8–4.2 (3.5 6 0.4) 3.6–4.5 (4.0 6 0.3) 2.4, 3.6
TH (male) 3.5 3.2–3.4 (3.3 6 0.1) 2.8–3.5 (3.1 6 0.2) 2.6–3.6 (3.0 6 0.4) — 2.6
TH (female) 3.5–3.9 (3.8 6 0.2) 2.9–3.4 (3.1 6 0.2) 2.5–4.5 (3.1 6 0.5) 2.3–3.4 (2.8 6 0.4) 2.7–3.9 (3.1 6 0.4) 2.3, 3.2
HL (male) 5.3 4.5–4.8 (4.6 6 0.2) 3.7–4.7 (4.1 6 0.4) 3.6–4.2 (3.8 6 0.2) — 3.9
HL (female) 4.8–5.2 (5.0 6 0.2) 3.5–4.4 (4.1 6 0.3) 3.5–4.5 (4.0 6 0.3) 3.5–5.3 (4.1 6 0.6) 3.8–5.5 (4.3 6 0.6) 3.1, 4.2
HW (male) 5.1 4.6–4.7 (4.6 6 0.1) 3.9–4.6 (4.2 6 0.2) 3.9–4.9 (4.2 6 0.4) — 4.0
HW (female) 4.6–5.8 (5.2 6 0.5) 3.8–4.6 (4.1 6 0.2) 3.6–5.2 (4.2 6 0.4) 3.7–5.4 (4.2 6 0.5) 4.0–5.3 (4.4 6 0.4) 3.5, 4.2
HH (male) 3.5 3.0–3.8 (3.4 6 0.4) 2.7–3.1 (2.9 6 0.2) 2.8–3.6 (3.0 6 0.3) — 2.9
HH (female) 3.4–4.2 (3.9 6 0.3) 2.9–3.4 (3.1 6 0.2) 2.5–3.9 (3.0 6 0.4) 2.5–3.9 (3.0 6 0.5) 2.9–3.9 (3.3 6 0.3) 2.5, 3.1
HERPETOLOGICAL MONOGRAPHS
SnFa (male) 13.6 11.6–12.8 (12.0 6 0.6) 10.9–12.2 (11.6 6 0.4) 11.3–12.3 (11.8 6 0.3) — 11.9
SnFa (female) 11.7–14.1 (12.9 6 0.9) 10.1–11.9 (11.2 6 0.6) 10.9–12.5 (11.6 6 0.5) 10.6–13.0 (11.7 6 0.8) 11.3–13.2 (11.8 6 0.6) 11.5, 12.1
ED (male) 0.9 1.0–1.1 (1.0 6 0.1) 1.0–1.1 (1.0 6 0.0) 0.8–0.9 (0.9 6 0.0) — 0.9
ED (female) 0.8–1.0 (0.9 6 0.1) 0.8–1.1 (1.0 6 0.1) 0.9–1.1 (1.0 6 0.1) 0.8–0.9 (0.9 6 0.1) 0.9–1.1 (1.0 6 0.1) 1.0, 1.0
END (male) 1.9 1.1–1.7 (1.5 6 0.3) 1.5–1.9 (1.7 6 0.1) 1.6–1.9 (1.7 6 0.1) — 1.6
END (female) 1.8–2.2 (1.9 6 0.1) 1.6–1.8 (1.7 6 0.1) 1.5–1.9 (1.7 6 0.1) 1.4–1.9 (1.7 6 0.2) 1.6–1.9 (1.8 6 0.1) 1.7, 1.9
SNL (male) 2.7 1.8–2.5 (2.2 6 0.4) 2.1–2.5 (2.4 6 0.1) 2.2–2.5 (2.4 6 0.1) — 2.3
SNL (female) 2.6–2.9 (2.7 6 0.1) 2.2–2.6 (2.3 6 0.1) 2.2–2.6 (2.4 6 0.1) 2.1–2.7 (2.4 6 0.2) 2.3–2.6 (2.4 6 0.1) 2.4, 2.6
IND (male) 1.4 1.2–1.4 (1.3 6 0.1) 1.2–1.4 (1.3 6 0.1) 1.1–1.5 (1.3 6 0.1) — 1.1
IND (female) 1.1–1.4 (1.3 6 0.1) 1.2–1.4 (1.3 6 0.1) 1.1–1.5 (1.3 6 0.1) 1.0–1.4 (1.3 6 0.1) 1.1–1.5 (1.3 6 0.1) 1.1, 1.3
FLL (male) 2.3 1.4–1.7 (1.6 6 0.1) 1.1–1.5 (1.3 6 0.1) 1.1–1.6 (1.3 6 0.2) — 1.4
FLL (female) 1.1–2.6 (1.7 6 0.5) 1.2–1.7 (1.4 6 0.1) 1.1–1.8 (1.3 6 0.2) 1.1–1.9 (1.5 6 0.3) 1.4–2.1 (1.7 6 0.2) 1.1, 1.5
HLL (male) 2.9 2.3–3.0 (2.6 6 0.4) 2.1–2.4 (2.3 6 0.1) 2.4–2.8 (2.6 6 0.2) — 2.6
HLL (female) 2.5–3.1 (2.8 6 0.2) 2.3–2.9 (2.6 6 0.2) 2.0–2.7 (2.4 6 0.2) 1.9–3.1 (2.7 6 0.4) 2.5–3.6 (3.0 6 0.3) 2.1, 2.7
[No. 25You can also read
