Content uploaded by George Zug
All content in this area was uploaded by George Zug on Dec 12, 2014
Content may be subject to copyright.
Pythons in Burma: Short-tailed python (Reptilia: Squamata)
George R. Zug,* Steve W. Gotte, and Jeremy F. Jacobs
(GRZ, JFJ) Department of Vertebrate Zoology, National Museum of Natural History,
Smithsonian Institution, Washington, D.C. 20013, U.S.A., e-mail: (GRZ) email@example.com;
(SWG) U.S. Geological Survey, Patuxent Wildlife Research Center, National Museum of
Natural History, Museum Support Center, Suitland, Maryland 27046, U.S.A.
Abstract.—Short-tailed pythons, Python curtus species group, occur
predominantly in the Malayan Peninsula, Sumatra, and Borneo. The
discovery of an adult female in Mon State, Myanmar, led to a review of the
distribution of all group members (spot-mapping of all localities of
confirmed occurrence) and an examination of morphological variation in
P. brongersmai. The resulting maps demonstrate a limited occurrence of
these pythons within peninsular Malaya, Sumatra, and Borneo with broad
absences in these regions. Our small samples limit the recognition of
regional differentiation in the morphology of P. brongersmai populations;
however, the presence of unique traits in the Myanmar python and its
strong allopatry indicate that it is a unique genetic lineage, and it is
described as Python kyaiktiyo new species.
Keywords: biogeography, herpetofauna, morphometrics, scalation
In mid 1997, the herpetological sections
of the California Academy of Sciences
and the National Museum of Natural
History began an intensive survey of the
Burmese herpetofauna. The survey was
initially limited to a few areas in the
central part of Myanmar. Then in late
1999, the survey became countrywide in
collaboration with the Myanmar Nature
and Wildlife Conservation Division. A
striking feature of our survey has been the
dearth of python vouchers. From 1997
through 2008, we have observed fewer
than a dozen pythons. Of these observa-
tions, a single adult female short-tailed
python was captured in the Kyaiktiyo
area in 2003. This female is the first
vouchered report of a member of the
Python curtus species group in Myanmar
and is ,600 km north of the northern-
most Thailand locality of Python bron-
The Burmese female has the general
appearance of Python brongersmai and
might have been assigned reflexively to
this taxon and considered an anomalous
occurrence, perhaps even an introduction.
Because the specimen was gravid, sug-
gesting a breeding population in the
Kyaiktiyo area, and somewhat atypical
in general appearance, we offer the
following morphological analysis to test
the uniqueness of the Kyaiktiyo female
and to examine geographic variation in
the morphology of P. brongersmai.
Materials and Methods
Thousands of short-tailed pythons en-
ter the pet-trade each year and many
more are harvested for the leather, meat,
and folk-medicine trade (Groombridge &
Luxmoore 1991, Shine et al. 1998a,
Keogh et al. 2001). In spite of the number
of individuals harvested, the number of
specimens with reliable locality data in
* Corresponding author.
PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
museum collections is miniscule. We have
examined 32 specimens of the Python
curtus species group with trustworthy
locality data (see Appendix 2), the ma-
jority of which are P. brongersmai. There
are dozens more P. curtus group speci-
mens in museums accessible to us, but
these specimens derive from the pet-trade
and possess no or questionable locality
data. We were unable to examine speci-
mens in the Naturalis, Leiden and other
European museums, and those specimens
recently donated to the Australian Muse-
um, Sydney. We have, however, used data
from these specimens that have appeared
in publications. Specimens examined de-
rived from eight museums: AMNH,
American Museum of Natural History;
BMNH, Natural History Museum, Lon-
don; CAS, California Academy of Sci-
ences; CUB, Chulalongkorn Museum of
Natural History, Chulalongkorn Univer-
sity; MCZ, Museum of Comparative
Zoology, Harvard University; UCM,
University of Colorado Museum – Zool-
ogy Section; USNM, National Museum
of Natural History, Smithsonian Institu-
tion; ZRC, Zoological Reference Collec-
tion, National University of Singapore.
We developed a character set of 15
measurements, 19 features of scalation,
and 9 coloration traits (see Appendix 1.
Definition of characters). Although these
characters derived largely from previous
studies of short-tailed pythons, they were
largely undefined, and owing to consid-
erable variation (especially in head scala-
tion), we defined each character to ensure
a consistence in our data-gathering. We
examined variation and relationships
among the characters and the samples
with basic parametric statistics of means
(measurements), medians (counts and
color coding – integers), standard devia-
tions (SD) and coefficients of variation
(V). We used Vas the initial assessment
for recognition of mixed samples, i.e., an
increase of Vfrom a localized sample to a
more regional one suggested different
means-medians for populations within
the region. Our sample sizes were too
small to perform statistical tests on the
variation and relationships among the
character sets; however, the sign test
allowed us to examine the symmetry of
bilateral head scalation. SYSTAT version
11 was used for all statistical analyses.
Distribution – Patterns and Peculiarities
What is the natural distribution of
Python curtus species group members?
This datum is critical if we are to assess
accurately the evolution of members of
this complex. Mapping the localities of
voucher specimens collected over the last
several decades and even the past two
centuries presumably will yield the core of
the original distribution, but for commer-
cially important species, such as short-
tailed pythons, individuals have been
transported widely and some undoubted-
ly have escaped or have been released.
Below we examine the distributional
records and address the ‘‘naturalness’’
issue. In that respect, it is worth repeating
a comment of Keogh et al. (2001:14, 15):
‘‘There is also a strong possibility of range
modification due to escape, because
snakes collected for commercial trade
are often moved long distances from the
point of collection to the slaughterhouses.
For example, slaughter houses in the
Medan area (within the range of P.
brongersmai) often receive shipments of
live P. curtus from the west coast (Shine et
al. 1998a, 1999).’’ Auliya (2006) reviewed
the distribution of Python curtus group
members in association with his review of
P. breitensteini biology, and we have
drawn heavily from his review in our
distributional assessment. Groombridge
& Luxmoore (1991) provided the first
assessment of Python occurrences and
populations’ status (P. curtus group, P.
molurus,P. reticulatus) in South-East
Asia with an evaluation of occurrence
records for the past century.
VOLUME 124, NUMBER 2 113
Python brongersmai has the broadest
distribution (Fig. 1) of the P. curtus
group, occurring from northern peninsu-
lar Thailand to Bangka and Belitung
islands off the southeast coast of Suma-
tra. This taxon apparently has not crossed
the Selat Sunda into Java. Although
broadly distributed, its occurrence is
peculiarly disjunct both within its core
range south of the Isthmus of Kra and in
Sumatra. Core range refers to the south-
ern Malayan peninsula and Sumatra from
which most museum vouchers derive and
in which an active harvest for leather,
food, and pet trade presumably continues.
Northern disjunct populations of P.
‘brongersmai’ exist in Mon State, Myan-
mar, Phetchaburi Province, Thailand,
and southern Vietnam. The Vietnam
population was first reported by Tirant
(1885). We assume that this record
represents, P. brongersmai; however, it
preceded the description of this taxon as
well as its recognition as a separate
species. Several researchers have subse-
quently commented on this record. No-
tably, Bourret (1936) accepted the Tirant
record without questioning it and includ-
ed it in his identification key and gave a
full description (not based on Vietnamese
specimens) in his species accounts and
noted its extreme rarity. Campden-Main
(1970) similarly accepted the record as
valid. In contrast, Groombridge & Lux-
moore (1991) questioned the validity of
the southern Vietnam record, and further
proposed that if the record was based on
a correct species identification, the py-
thons had likely been introduced by
humans. More recently, Nguyen et al.
Fig. 1. Distribution of members of the Python curtus species group. Solid symbols denote museum
specimens examined by us; open symbols are literature records or unexamined museum specimens; question
marks denote doubtful records reported in the literature and lacking voucher specimens. A, P. brongersmai
records for Sumatra and mainland Southeast Asia, and P. breitensteini for Borneo; B, P. curtus records for
Sumatra. See Appendix 3 for mapping data and for information on non-occurrence; USNM 103513 is
mapped twice; once at its corrected locality in Sumatra and also at its erroneous locality [?] in Ceram.
114 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
(2009) gave three localities, two new
provincial ones and Ho Chi Minh city
(5Saigon). We assume the latter repre-
sents the original Tirant record. They
neither confirmed nor questioned this
record. They added Binh Thuan and Ca
Mau without comment. The latter two
localities derive from Orlov et al.’s report
(2003) of short-tailed pythons in Saigon
markets and the dealers’ reference to the
pythons’ origins from those two provinc-
es. Orlov et al. (2003) identified the
specimens, although no vouchers were
retained, as P. curtus brongersmai.We
have included all three Vietnamese local-
ities on the map (Fig. 1A) but as ques-
tionable records. This Mekong Delta
population may be a natural isolate, an
introduced population, or a false report
of the origin of imported specimens.
Cox (1991) noted that P. brongersmai
was the least common python in Thailand
and was confined to southern Thailand.
Cox also reported this species’ presence in
Myanmar, Cambodia, and Vietnam, but
he provided no evidence supporting these
latter three localities. Barker & Barker
(1994) similarly included Cambodia and
southern Vietnam within the range of P.
brongersmai. They (D. Barker in litt.,
June 2010) subsequently learned that their
source made false statements on the
origins of numerous pet-trade animals,
hence their Cambodia record is most
likely incorrect. No voucher specimens
exist for Indochina, and recent surveys in
Cambodia (see Appendix 3) have found
no short-tailed pythons.
The most northern Thailand record is
Kaeng Krachen (Fig. 1A). A short-tailed
python was observed and photographed
by an ornithologist and subsequently
identified by T. Chan-ard (in litt., June
2010). Although this record is about
450 km north of other Thai occurrences
(all south of the Isthmus of Kra), we
accept it as a natural occurrence because it
is based on a photograph of a specimen in
a nature reserve by a field biologist. We
note, however, the nearness of this locality
to Bangkok, hence the possibility of its
intentional release into the wildlife reserve,
e.g., Mark O’Shea noted (in litt., Decem-
ber 2010) ‘‘The Thai authority’s policy is
to release live snakes from such raids [of
illegal reptile butcher-factories] into the
nearest national park area.’’ The few
museum vouchers and published records
confirm Cox’s statement on the rarity of
short-tailed pythons in Thailand. Groom-
bridge & Luxmoore (1991) reported the
natural distribution as extreme southern
peninsular Thailand and proposed that the
occasional reports of more northern local-
ities derived from escaped pythons.
Python brongersmai appears to be
somewhat more common in peninsular
Malaysia, although the available records
still indicate that they are uncommon
snakes. Groombridge & Luxmoore (1991)
provided evidence that P. brongersmai has
always been relatively uncommon com-
pared to P. reticulatus. The peninsular
Malaysian distribution (Fig. 1) shows an
absence in northeastern and south-central
Malaysia. We tentatively accept these
absences as the result of the sporadic
nature of faunal surveys and specimen
vouchering in association with the crypsis
of this species. Vouchers are available for
southern-most Malaysia and Singapore.
The most recent vouchers for Singapore
were collected in 1901 and 1903 (ZRC
2.3090, 2.3091, respectively). Barker &
Barker (1994) reported P. brongersmai
from the string of islands (Riau group,
Linnga, Singkep), extending from Singa-
pore to south-central Sumatra, and also
from Bangka and Belitung islands. These
records presumably derive from animal
exporters. We tentatively accept these
locations as natural occurrences; we have
seen vouchers (UCM) only from Bankga;
these vouchers derived from an animal-
dealer’s direct import from Bangka (M.
Kageya´ma in litt., April 2006).
Sumatra, like Malaysia, shows a large
lacuna in the distribution of P. brongers-
VOLUME 124, NUMBER 2 115
mai (Fig. 1A). Records are concentrated
in the northeast and north-central region
(Sumatera Utara and northern Riau
province), and apparently absent from
the broad lowlands of central and south-
ern Sumatra with the exception of the
Palembang leather-market record (Shine
et al. 1998a). The ‘Utara’ region was the
major rubber-growing area (Babcock
1966, Yacob 2007) for much of the
twentieth century. Now much of the area
is occupied by oil-palm plantations. The
plantation habitat fosters large rodent
populations supporting denser popula-
tions of P. brongersmai (Abel, 1998),
and possibly more people tending planta-
tions find more snakes. Nonetheless, the
absence of voucher specimens or records
from the large central and southern area
is striking and suggestive of a real
absence. There are a few scattered records
of P. brongersmai from the western
margin of Sumatra. Two localities, Kor-
ninchi and Siulakderas, are located in a
mountain valley in western Sumatra, but
the valley’s drainage is eastward. These
specimens of P. brongersmai are museum
vouchers and apparently deposited by the
collectors. We, thus, interpret these local-
ities as natural occurrences of P. bron-
gersmai. As noted above, the larger
islands (Riau Archipelago, Lingga, Sing-
kep, Bangka, Belitung) off the east coast
of southern Sumatra have reliable reports
of P. brongersmai.
The Palembang records (Shine et al.
1998a) of short-tailed pythons are puz-
zling. First, were these specimens P.
two species? Second, were these pythons
collected locally, i.e., within a 100 km
radius, or transported a long distance to
Palembang? The small number of short-
tailed pythons (n535) versus the large
number of reticulated pythons (n51067)
suggests long distance transport, possibly
even from Bangka. Shine et al. (1998a)
identified them as P. curtus and also all
specimens from Medan as P. curtus,
although in subsequent publications
(Shine et al. 1998b, 1999), they reported
both P. brongersmai and P. curtus from
the Medan leather market with P. curtus
noted as coming from the Sibolga area.
We suspect the single species Palembang
identification resulted from a manuscript
overly long in press. Owing to the
preceding uncertainty on local origin
and specific identity, we consider Palem-
bang a questionable locality for either
species. Shine (in litt., September 2010)
agreed with our interpretation: ‘‘The
locality (and species) are unreliable.’’
We have found only two localities
(Appendix 3) with specimen-based rec-
ords of P. curtus, which do not derive
from recent leather-market hubs. These
localities are Mt. Kabor, and Kaba
Wetan. Both are montane localities and
are interpreted as natural occurrences.
Other P. curtus localities are less easily
confirmed, although without contrary
evidence, we accept them to represent
the area of natural occurrence. The
northern-most locality for P. curtus is
Sibolga. It was mentioned first by Shine
et al. (1998b) as the source for specimens
arriving by truck at leather markets in
north-central Sumatra. A Padang record
represents a mid-west coast record. Be-
cause Padang is a port city, we assume
Padang represents a purchased specimen,
possibly captured locally. Barker & Bark-
er (1994) reported these pythons from
southern Sumatra and pictured a speci-
men from vicinity of Bandar Lampung.
Python curtus is also reported from
Bengkulu (Keogh et al. 2001). These data
suggest a linear range of more than a
1000 km along the western front of the
mountain range to southern ‘tip’ of
Sumatra for P. curtus. Because of the
leather-market nature of the Sibolga
record, the possible sympatry of P.
brongersmai and P. curtus in this area
cannot be verified.
A possible exception to this proposed
distribution is the Siberut (Mentawai
116 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
Islands) record of Dring et al. (1990).
Both reticulated and short-tailed pythons
were observed, although the latter was
represented only by a sight record without
an examination to confirm species identi-
ty. We tentatively accept the Siberut
population as P. curtus and propose that
it represents a relict population that
arrived from mainland Sumatra during
the lower sea level of the last glacial
With exception of one specimen, short-
tailed pythons do not occur in Java or
southward. The P. brongersmai specimen
(USNM 103513) from Ceram has very
precise locality data, yet it is so distant
from its conspecifics as to signal a data
mix-up. No published checklists (e.g., de
Rooij 1917, de Haas 1950, or in den
Bosch 1985) of Lesser Sunda snake
faunas confirm the presence of any
Python curtus group member in this
region. The specimen (USNM 103513)
was received from the National Geo-
graphic-Smithsonian Institution East In-
dies Expedition through the National
Zoological Park. This expedition was
primarily to collect live animals for the
zoo, though a few preserved specimens
for the museum were collected (Mann
1938). The party assembled in early
March 1937 at Siantar in northern
Sumatra from which they collected for
5 months, obtaining most of their speci-
mens from local collectors. Members of
the expedition made several trips to the
mainland and ‘‘down the Archipelago as
far as Amboina in the Moluccas,’’ ‘‘Dr.
Coenraad … made a hurried trip to
Sorong … while the rest of the party
stayed at Piroe on the island of Ceram.’’
(Mann 1938:35). The only Ceram reptile
mentioned by Mann was an amethystine
python from the Moluccas. The National
Zoological Park specimen transfer cards
in our Division of Amphibians and
Reptiles indicate that the expedition
brought back at least 17 pythons to the
National Zoo (two Morelia amethistina,
eight P. curtus,threeP. molurus, and four
P. reticulatus), none of which are listed as
being cataloged at the USNM. Only the
P. curtus specimens have locality data
on the cards, and the data for all are
‘‘Vicinity, Siantar, Sumatra.’’ We propose
that zoo personnel accidentally assigned
the amethystine python data to the P.
curtus. Additionally, other USNM
‘Ceram’ reptiles have proved to be Suma-
tran specimens. These include a Calliophis
gracilis, whose locality datum was ques-
tioned by both Alan Leviton (USNM
ledger notation, not dated) and Eric Smith
(in litt., December 2006). Calliophis grac-
ilis occurs only on the Malay Peninsula
and northern Sumatra, a distribution
consistent with the Siantar base camp.
Python breitensteini is outside our
specific research focus; nonetheless, we
considered it valuable to plot the known
occurrence of this species for comparison
to P. brongersmai. Our literature and
specimen search confirms the absence of
Brunei specimens. As for P. brongersmai,
the distribution of P. breitensteini (Fig. 1)
has large areas of Borneo without rec-
ords, reflecting both the inadequacy of
the biological survey of the Bornean
herpetofauna and the cryptic behavior
of short-tailed pythons.
While the absence of a literature or
specimen record is interpreted here as the
absence of a short-tailed python popula-
tion, these python are quite cryptic in
spite of their large adult size, thus the lack
of a record is not assurance of actual
absence. Further, our proposed distribu-
tions (Fig. 1) rest on over a century and a
half of records and are, in our opinion, an
optimistic assessment of the present day
occurrence of the Python curtus group
members. Even though P. reticulatus
persists in some urban and suburban
areas, short-tailed pythons do not; other-
wise, more recent locality records would
be denser. Singapore’s most recent record
(1903) for P. brongersmai is now more
than a century old, despite that city-
VOLUME 124, NUMBER 2 117
state’s aggressive protection of natural
areas. In contrast, some oil palm planta-
tions seem to provide an exceptional
match to the biological needs of these
pythons and have resulted in extraordi-
nary densities in some areas, judging from
their harvest for the leather and pet
trades. The paucity of Thai localities
and the absence of Brunei records suggest
that short-tailed pythons are easily erad-
icated from an area. This conclusion is
seemingly at odds with the continuing
harvesting of these pythons from north-
central Sumatra and Sabah. What allows
these populations to survive, seemingly
even thrive, in some oil-palm plantation
areas and yet not in others?
Data Review – Morphometrics
Because our samples are small and
from only a few localities, we include
published data as well as our data and
analyses in the following analytical sec-
tions; however, the former is strictly data
presentation. Interpretation of our and
published data is in the Patterns section.
Our data do not allow us to test for
sexual dimorphism among any of our
samples. Aside from the small size of the
samples, we were able to determine the
sex confidently on only a few specimens,
owing to strongly contorted specimens
and evisceration of others preventing sex
determination on most specimens. Shine
et al. (1998b) examined body size in
different color morphs (brown, orange,
red, yellow) of P. brongersmai from four
leather-markets in northern Sumatra
(Medan to Cikampak) and found adult
females of each morph to be significantly
larger than the adult males of the same
color; differences of means range from 11
to 16 cm SVL (snout-vent length). Their
size data did not include ranges for
individual color morphs or for the total
sample (n5730, 1038; females, males
respectively). They also did not explicitly
provide SVL at sexual maturity, although
their size-distribution graphs (Shine et al.
1999: fig. 1) indicate a minimum adult
SVL of 85 cm for both females and males.
The grand means (our calculations) for
adult SVL are 143.9 cm (85–185 cm)
females and 132.5 cm (85–175 cm) males.
Even though they tested relative tail
lengths among the color morphs and
found significant differences, they did
not test differences between the sexes. In
their comparison of morphology and
ecological traits (Shine et al. 1998b), they
also included P. curtus (presumably from
the Sibolga area); female P. curtus were
larger than males (means 130 cm (90–
150 cm), 121 cm (90–145 cm) SVL, n5
54, 99, respectively).
The largest P. brongersmai (167 cm
SVL) in our sample was a Thai male of
uncertain locality. The Singapore sample
had five adults ranging from 78–146 cm
SVL. Although we question our assess-
ment of maturity, the small specimen (the
holotype of P. brongersmai) was consid-
ered an adult male when first described.
We also considered several 90–100 cm
long individuals in our samples as mature.
Of the two P. curtus specimens examined,
the 67 cm SVL one was considered
mature. The Mon-Burmese specimen
was a gravid adult female at 117 cm
SVL. Boulenger (1912:109) reported that
the length of Malaysian short-tailed
python ‘Grows to 9 ft,’ a total length of
274 cm; however, the largest measure-
ment that Boulenger (1893:90) provided
was 137 cm TotalL. Concerning this
specimen, which was not identified specif-
ically in the catalog, C. McCarthy (in litt.
September 2010) wrote: ‘‘I imagine that
it is based on the largest specimen
available to him at the time (and perhaps
may even be based on the stuffed specimen
[BMNH 18220.127.116.11 from Singapore]).’’
Brongersma (1947) gave 152 cm SVL and
12 cm TailL for the largest Sumatran P.
brongersmai. The maximum SVL reported
by Shine et al. (1999) was ,185 cm for P.
brongersmai and ,150 cm for P. curtus.
118 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
Keogh et al. (2001) said that P. brongers-
mai reach 260 cm TotalL.
Body proportions permit the compari-
son of python morphometrics (Table 1)
among our small regional samples. The
Bangka sample (n59) consists entirely of
juveniles with the largest individual a
possible subadult male. A noteworthy
feature of the Bangka pythons is the low
variation for all proportions as estimated
by the Coefficient of Variation (V,1.8–
11.3%). Two proportions (SnEye/HeadL,
Intorb/HeadL) have strikingly low varia-
tion, i.e., 1.8%and 3.4%; respectively.
The others (TailL/SVL, HeadL/SVL
HeadW/SVL, HeadW/HeadL, OrbD/
HeadL and Intnar/HeadL) are 8–11%,a
typical range for morphometric samples.
The total Sumatran sample (n513) has a
similar pattern of variation as the Bangka
sample, which is included therein but
shows a distinct elevation in the Vs
for TailL/SVL, HeadL/SVL HeadW/
SVL, and OrbD/HeadL range 13.3%to
17.0%. Relative tail length has the great-
est variation. The Singapore sample (n5
5) is only slightly more variable (4.6–10%)
for most proportions, except for tail and
head length and head width (27.5%,
23.7%,29.9%, respectively). Combining
all regional samples of P. brongersmai (n
526) resulted in a significant jump in
variation for TailL/SVL, HeadL/SVL,
HeadW/SVL (20, 22, 23%) and only a
slight increase for the others. Body
proportions, other than relative TailL,
have not been reported by previous
researchers, and the tail data were not
given as metric proportions.
The large northern Sumatran sample of
P. brongersmai (Shine et al. 1999) showed
Vs of 8.2 and 9.0%(our calculations) for
adult female and male SVL samples (n5
732, 1037), and a sample of hatchlings (n
518) from a single egg clutch (Abel 1998)
had Vranging from 1.5%(SVL) to 5.6%
(TailL/SVL): mean 6SD 412.6 6
6.33 mm SVL; 34.8 61.58 mm TailL;
31.4 60.62 mm HeadL; 11.7 60.48 mm
Table 1.—Morphometrics of regional samples of Python brongersmai and P. curtus. The values are mean 6standard deviation and minimum, maximum.
Character names (abbreviations) explained in Appendix 1.
Taxon/Location SVL (mm) TailL/SVL HeadL/SVL HeadW/SVL HeadW/HeadL OrbD/HeadL
Singapore, n55 1180.6 6251.92 0.086 60.024 0.056 60.013 0.031 60.009 0.562 60.045 0.091 60.006
779–1460 0.067–0.126 0.043–0.078 0.023–0.048 0.517–0.613 0.084–0.098
Sumatra, n513 674.1 6291.76 0.090 60.015 0.072 60.010 0.041 60.006 0.574 60.049 0.108 60.014
353–1189 0.074–0.124 0.054–0.084 0.029–0.050 0.509–0.646 0.087–0.131
Bangka Island, n58503.86183.03 0.082 60.007 0.077 60.007 0.043 60.005 0.560 60.046 0.117 60.010
353–915 0.074–0.092 0.067–0.084 0.036–0.049 0.509–0.646 0.103–0.131
West Sumatra, n52596.56112.43 0.134 60.022 0.081 60.005 0.046 60.001 0.563 60.026 0.103 60.009
517–676 0.118–0.149 0.078–0.085 0.045–0.046 0.545–0.582 0.097–0.100
Myanmar n. sp. 1166 0.089 0.058 0.034 0.589 0.098
Siantar/Ceram 354 0.105 0.088 0.042 0.474 0.141
VOLUME 124, NUMBER 2 119
OrbD; 8.3 60.5%TailL/SVL; 7.6 6
0.2%HeadL/SVL; 37.1 61.5%OrbD/
HeadL. This latter sample (our calcula-
tions) is useful in providing a baseline for
variation within P. brongersmai,owingto
its high genetic homogeneity.
Some regional differentiation is sug-
gested by the proportions (Table 1).
TailL/SVL among P. brongersmai sam-
ples range between 6.7%and 12.6%,
mean 9%. The tail is strikingly longer in
P. curtus,13%(Table 1) than in P.
brongersmai. Relative head length and
width average less in the Singapore
sample and the Mon female. Head shape
(HeadW/HeadL) is similar throughout all
samples and the Mon female but distinct-
ly narrower (47%) for the Siantar/Ceram
juvenile. Among the within head propor-
tions, relative orbit or eye diameter
displays the greatest differences among
samples (Table 1). Singapore and Mon
pythons average the smallest eyes. Both
P. brongersmai and P. curtus share
intermediate relative eye diameters, and
the Bangka and Siantar/Ceram pythons
have the largest relative eye dimensions.
The other three head proportions are
strikingly similar in means for all six of
our samples: SnEye/HeadL 38–39%;Int-
orb/HeadL 32–35%; Intorb/HeadL, 16–
Data Review – Scalation
Our scalation data examines 19 discrete
characters. Ten of them are bilateral traits
on which we gathered data from both left
and right sides. We tested bilateral
symmetry in the total Sumatran sample
(n514) and the Bangka sample (n58)
of P. brongersmai using the Sign test. This
test examines only whether the values, if
different, from the right and left sides
vary in the same direction. The test
ignores the number of times a trait is
identical on the two sides. For the
Sumatran and Bangka samples, the num-
ber of scales on the left and right sides
were identical for all specimens in Suboc
and Supoc. The Bangka sample had
$75%of the individuals with identical
left and right counts for Suplab, Suplab-
Eye, and TemporG, and 50–75%of
individuals for the Preoc and Postoc
counts. For the total Sumatran sample,
most ($75%) individuals had Suplab,
Preoc, SuplabEye, Postoc, and TemporG
counts that were the same on the two
sides. Asymmetry occurs for Inflab and
LorGrv in both samples. In these instanc-
es of asymmetry, the left side had
significantly more scales than the right.
Even in those traits that were more
symmetrical ($50%), the left side usually
had more scales than the right side,
although this difference was significant
only for Preoc and in both samples.
Supoc is symmetrical (10 of 13 individu-
als, 77%)inP. brongersmai and where
asymmetrical, two individuals had more
scales on the right side. Brongersma’s
tables (1947:Tables 1, 2) provided a
comparison for the left and right Supoc.
All three of Brongersma’s P. curtus were
symmetrical with a single Supoc on each
Our head scale characters (list and
definitions in Appendix 1) contain 15
traits that are counts and five that are
measurements. The latter traits are mid-
line lengths of the large dorsal head
scales. Because size of the frontoparietals
varies among the curtus-group taxa, we
wished to determine whether other head
shields had size/shape differences. Again
owing to our small samples and the
disparity of body size in the samples, we
examine this aspect through proportions.
Most scalation characters derive from
the side of the head; the subsequent
examination of these characters uses only
values from the right side. Head scalation
is moderately variable (Table 2). For
example, Vvaries from 7–13%in Suplab
and Inflab; however, Vjumps to 20%or
higher for many other characters. These
high Vcharacters are ones with narrow
120 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
ranges of values, such as Preoc and
Postoc (1–3 scales each, Bangka sample),
and a predominance of one state (i.e., 2
scales). Some characters (InflCAL,
InflCPL, LorGrv, TemporG) are more
variable, that is a moderately broad range
of counts with no counts concentrated
at the mean/median, hence yielding V
.20%. Table 2 offers a portrait of the
statistics for a subset of head scalation
characters and demonstrates the broad
overlap among the P. brongersmai sam-
ples and between those samples and the
P. curtus sample. The ranges of characters
in P. brongersmai not displayed are: 0–2,
1–3, 1–2 InflCAnt, 0–5, 3–7, 3–7 Infl-
CAL; 7–13, 11–13, 11–13 InflCPo; 4–5,
4–8, 3–7 InflCPL; 0, 0–4, 1–3 Postoc; 10–
11, 10–13,10–13 Suplab; 2, 1–2, 1–2
SuplabEye; 3–4, 1–7, 2–6 TemporG,
respectively for Singapore, Sumatra, and
Keogh et al. (2001) provided basic
statistics for scalation traits within his P.
brongersmai description. We have extract-
ed those data and present some in
Table 2. They generally match our values,
but where there are differences, such as
LorGrv and Supoc, it seems evident that
Keogh and colleagues defined the traits
differently than we did. In addition to
those presented in Table 2, they reported
1–3 Postoc, 9–13 Suplab, and 1–2 Suplab-
The number of supraoculars (Supoc) is
regularly reported as a single Supoc
dorsally touching the eye in P. curtus
and P. breitensteini, and two in P.
brongersmai (e.g., Stull 1938, Keogh et
al. 2001). We found the situation more
variable (Table 2) in part owing to our
use of a standard definition of squamate
head scalation (e.g., Gauthier et al. 2008;
also see our scalation definitions in
Appendix 1 and Fig. 2). First, all three
curtus group taxa commonly have three
Supoc in contact with the eye dorsally.
The middle one is the largest and broadly
borders the first Supoc, frontal, and
frontoparietal. The anterior or first Supoc
is also large, roughly half the area of the
middle Supoc and dorsally borders the
prefrontal (Prefron). In past usage for
short-tail pythons, the anterior Supoc has
been labeled a preocular, thus in our
usage, the number of Preoc in an
individual is reduced by one scale; how-
ever, the range (variation) of Preoc
(Table 2) matches the values reported by
previous authors. The posterior Supoc is
small, roughly 10–15%the area of the
middle Supoc. It is in broad contact with
the middle Supoc and supratemporal,
rarely with a point contact with a
frontoparietal. The third Supoc has been
variously labeled a Supoc or Postoc.
The paired, midline abutting head
shields are little mentioned, except for
the frontoparietals (called parietals by
Keogh et al. 2001). The frontoparietals
are broadly in contact medially in P.
breitensteini and P. brongersmai, and not
or barely in contact in P. curtus. As noted
above, we have attempted to quantify the
nature of these dorsal head scales by
measuring their lengths along the dorsal
midline (Appendix 1). The frontoparietals
were not included in this set of measure-
ments owing to a high frequency of
bilateral asymmetry, that is, either the
left or right frontoparietal is strikingly
larger than it’s opposite. The scales
posterior to the frontoparietals are slight-
ly enlarged and irregularly paired; usually
within four or five rows, they grade into
the dorsal scales. Thus like most previous
researchers, we did not attempt to label
the parietals and other posterior head
scales. We also did not examine them
rigorously to observe any morphological
patterns among the regional samples.
Again to reduce the effect of size, we
converted our scale length measurements
to proportions (Table 3). Variation rang-
es from low to high in all regional
samples, e.g., Vs from 2.7%(MidHead/
HeadL) to 21%(Frntnas/HeadL) in
Singapore sample. This pattern of varia-
VOLUME 124, NUMBER 2 121
tion was consistent across all samples
with relative mid-head length (MidHead/
HeadL) the least variable and relative
prefrontal length (Prefr/HeadL) the most
variable, regularly .18%. The other
proportions had Vsof9%to 16%.As
with their variation, the means and ranges
of the proportions are similar among the
samples (Table 3). We note only the low
relative Intnas mean for P. curtus,low
relative Prefron of Siantar/Ceram, and
low Front of Myanmar. Of these low
means, all but the P. curtus Intnas/HeadL
lie within the ranges of the other samples.
Body scalation (Table 4) shows modest
variation with Vs ranging from 2%to
11%. Ventrals are consistently the least
variable and Subcaudals the most vari-
able among our P. brongersmai subsam-
ples: Bangka, Singapore, and total Suma-
tran. The number of dorsal scale rows
(DorsAnt, DorsMidb, DorsPost) shows a
consistent pattern for all our samples at
midbody and posteriorly, with similar
means and ranges for P. brongersmai
Table 2.—Head scales (right side) of regional samples of Python brongersmai and P. curtus. The values are
median 6standard deviation and minimum, maximum; *data from Keogh et al. (2001:7, 8). Character
names (abbreviations) explained in Appendix 1.
Taxon/Location Inflab InflabEn LorGrv Supoc Preoc Suboc
Singapore, n551762.17 9 60.82 5 60.89 4 60.45 1 60.054 0 60
14–18 8–10 5–7 3–4 1–2 0
Sumatra, n514 18 61.07 9 61.19 5 61.83 3 60.58 1 60.65 0 60
16–20 6–11 1–8 2–4 1–3 0
Bangka Island, n591860.93 9 61.20 5 61.67 3 60.50 2 60.71 0 60
16–19 6–10 4–9 2–4 1–3 0
West Sumatra, n5216608.560.71 6.5 62.12 3 601.562.12 1 60
16 8–9 5–8 3 0–3 1
Myanmar n. sp. 19 8 3 3 1 0
Siantar/Ceram 17 10 8 2 1 0
19.0 61.31 — 8.4 62.41 1.8 60.48 2.1 60.48 0 60
17–22 4–14 1–2? 1–3 0
Fig. 2. Diagrammatic views of the head of a
Python kyaiktiyo from Mon State, Myanmar
(USNM 572046). A, Dorsal; B, Lateral.
122 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
and P. curtus (Table 4). For DorsAnt, the
Singapore sample has a lower mean and a
barely overlapping range with those of
other samples. The Myanmar and Sian-
tar/Ceram pythons match the others in
DorsMidb and DorsPost. The Siantar/
Ceram specimen has DorsAnt within the
range of the Singapore sample and the
Myanmar specimen at the upper end of
the range for all other samples (Table 4).
For Ventral, P. curtus has the lowest
mean and range, and the Singapore
sample the lowest mean Ventral of the
P. brongersmai samples, although its
range mostly overlaps the ranges of those
samples (Table 4). The Myanmar Ventral
count is higher than any other short-
tailed pythons examined, and the Siantar/
Ceram Ventral is at the upper end of the
P. brongersmai ranges. Subcaud is uni-
form across all samples, including the
Data from a larger sample of Sumatran
P. brongersmai (Keogh et al. 2001)
yielded similar means and ranges (Ta-
ble 4) as our smaller P. brongersmai
samples, although their Subcaud mean is
higher and its range slightly broader.
Brongersma (1947:Table 2) presented
data on body scalation of three P. curtus.
His median and mean values (Table 3)
match our sample for DorsAnt, Dors-
Midb, and Ventral and are slightly
different for DorsPost (32) and Subcaud
Data Review – Coloration
Coloration in short-tailed pythons is
presented usually as background color
(red, orange, yellow, brown) and a
general description of head and body
stripes and blotches. Because our sample
contained only preserved specimens and
most having been stored in ethanol for
decades, we were reluctant to assign our
specimens to background colors classes.
We attempted to quantify color patterns
by establishing binary or multistate char-
Table 3.—Dorsal head scalation of regional samples of Python brongersmai and P. curtus. The values are mean 6standard deviation and minimum, maximum
of relative lengths. Character names (abbreviations) explained in Appendix 1.
Taxon/Location MidHead/SVL MidHead/HeadL Intnas/HeadL Frtnas/HeadL Prefron/HeadL Front/HeadL
Singapore, n55 0.025 60.006 0.452 60.012 0.070 60.008 0.124 60.018 0.284 60.060 0.158 60.019
0.019–0.035 0.442–0.472 0.063–0.083 0.099–0.140 0.214–0.377 0.140–0.187
Sumatra, n513 0.033 60.006 0.453 60.031 0.071 60.009 0.138 60.015 0.201 60.081 0.148 60.024
0.023–0.043 0.423–0.529 0.059–0.085 0.110–0.160 0.100–0.341 0.115–0.197
Bangka Island, n59 0.034 60.004 0.446 60.024 0.069 60.011 0.140 60.018 0.147 60.030 0.144 60.017
0.028–0.039 0.423–0.489 0.059–0.085 0.110–0.160 0.100–0.200 0.127–0.180
West Sumatra, n52 0.036 60.003 0.443 60.014 0.050 60.007 0.114 60.009 0.248 60.042 0.166 60.033
0.034–0.038 0.433–0.453 0.046–0.055 0.108–0.120 0.218–0.277 0.143–0.190
Myanmar n. sp. 0.025 0.422 0.065 0.154 0.236 0.128
Siantar/Ceram 0.042 0.481 0.080 0.170 0.105 0.157
VOLUME 124, NUMBER 2 123
acters for nine of the commonly men-
tioned pattern traits (five cephalic and
four body ones; see definitions in Appen-
dix 1). Rostral color (Rostrl) is light in all
immature individuals of P. brongersmai
and P. curtus, and light in an adult P.
breitensteini. The transverse parietal stripe
(PaStrp) is absent in all Singapore py-
thons and most mainland individuals.
PaStrp occurs in about half of the
Sumatran specimens. The postocular
stripe (PostocStr) occurs in all our short-
tailed pythons, and its anterior edge lies
about equally on the seventh or eighth
Suplab. The PostocStr distinctly shifted
posterior in one Sumatran P. brongersmai
and the P. breitensteini. It is strongly
anterior (on fifth Suplab) in our single
southern Malaysian specimen. In our
preliminary examination of coloration,
we observed that the alignment of the
postocular stripe with ventrolateral one
was not continuous in some individuals.
This misalignment occurs in a third of our
total sample, and no geographic concen-
tration is evident in the data set. The
Mental is dark in 90%of the P. bron-
gersmai and one of two P. curtus.The
width of the dark lateral stripe (Body-
LatW) varies from six to ten dorsal scale
rows, with widths of 6, 7, and 8 rows of
near equal frequency. We see no regional
pattern in the different states of Body-
LatW. MiddStrp addresses the condition
(entire or interrupted) of the middorsal
stripe posteriorly but not its linearity or
constancy of width. MiddStrp is entire in
more than 70%of our total sample and
again without evidence of regional con-
centrations, although both our P. curtus
have interrupted stripes. Most (.70%)P.
brongersmai have dark venters; our P.
breitensteini and P. curtus specimens have
light venters. SubcaudC displays the
reverse pattern with the latter two taxa
with dark subcaudals and most P. bron-
gersmai with light ones.
Patterns of Variation in Morphology
Mensural features.—Short-tailed py-
thons display size dimorphism between
adult females and males. The large data
set of northeastern Sumatra P. brongers-
Table 4.—Body scalation of regional samples of Python brongersmai and P. curtus. The values are median
6standard deviation and minimum, maximum; *data from Keogh et al. (2001:8). Character names
(abbreviations) explained in Appendix 1.
Taxon DorsAnt DorsMidb DorsPost Ventral Subcaud
Singapore, n554164.04 55 65.15 33 63.44 169 63.36 28 62.59
34–45 44–57 29–37 165–174 26–32
Sumatra, n514 48 62.69 57 62.09 34.5 63.21 172 63.39 28 62.99
44–53 53–59 31–43 167–177 22–34
Bangka Island, n594762.83 56 61.81 34 62.00 171 63.71 27.5 63.18
44–53 53–58 31–37 167–177 24–34
West Sumatra, n5247.564.95 57 64.24 34.5 62.12 157 62.83 28 61.41
44–51 54–60 33–36 155–159 27–29
Myanmar n. sp. 51 58 36 184 27
Siantar/Ceram 43 53 33 178 28
49.6 62.2 56.7 61.82 33.7 61.72 171.8 62.86 30.0 62.64
45–53 53–61 32–38 167–178 24–36
124 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
mai (Shine et al. 1998b) provides evidence
of size dimorphism. These data record
SVL for four color morphs from four
samples derived from an area lying within
an ,200 km radius south and southeast
of Medan. Each color morph (brown,
orange, red, and yellow) has different
average SVLs, and in each one, females
are significantly larger (,10–12 cm lon-
ger) than males. The sample sizes within
each morph are adequate to support the
dimorphism and different maximum
adult lengths of the different morphs.
The same researchers (Shine et al.
1999:Table 1) demonstrated size dimor-
phism in adult P. curtus by noting slightly
different mean SVLs for the two sexes.
These data are not available for P.
breitensteini, although Auliya (2006) not-
ed a difference in relative tail length
among adults; he proposed this difference
as sexual dimorphism. Neither our data
nor Shine et al.’s permit us to confirm tail
length dimorphism in P. brongersmai or
P. curtus, although the high variation (V)
of relative TailL, HeadL, and HeadW in
the adult sample from Singapore and the
total P. brongersmai is strongly suggestive
of dimorphism, i.e., increased variation
owing to mixed samples.
Maximum adult size (TotalL) of P.
brongersmai has been reported as 270 cm
(Boulenger 1912) and nearly that by
Keogh et al. (2001). These maximum
sizes are strikingly higher than our largest
specimen (Thailand) at 179 cm TotalL.
This maximum length matches the data
from the north-central Sumatran leather
markets (Shine et al. 1998a, 1998b, 1999)
in which no specimen exceeded 200 cm
TotalL. There is no evidence that Bou-
lenger’s 270 cm length derived from an
actual measurement of a specimen.
Hence, we propose that two meters is
the maximum for wild P. brongersmai at
the end of the twentieth century. This
maximum size is also likely for P. curtus.
Relative head length and width aver-
ages are strikingly smaller in the Singa-
pore sample and the Mon female. In
contrast, head shape (HeadW/HeadL) is
similar throughout all samples, including
the Mon female but distinctly narrower
(47%) for the Siantar/Ceram juvenile.
Bangka and Siantar/Ceram pythons also
have the largest relative eye dimensions.
We expect the larger size in these two
samples reflects allometry, that is, juve-
niles have proportionately larger eyes.
Are the shorter and narrower heads of
the Mon and Singapore pythons also the
result of allometry or a hint at regional
differentiation between mainland and
insular populations? Our limited data set
cannot satisfactorily answer this question.
Because the mainland samples contain
predominantly adults and the insular ones
juveniles, allometry is a possibility.
Scalation traits.—Not surprisingly, bi-
lateral cephalic scalation is predominantly
symmetrical, although the traits display-
ing asymmetry (Inflab, LorGrv) regularly
have more scales (higher counts) on the
left side. Further, scalation traits that are
not statistically asymmetrical also had a
preponderance of individuals with more
scales on the left. We cannot explain the
sinistral dominance, but we suspect that it
has a biological explanation and is not a
There is not a strong regional signal in
the lateral cephalic scales (Table 2). The
ranges of each trait overlap broadly
among the regions and median values
are identical in most cases. Only in
LorGrv do the two geographic outliers
deviate from the median values of the
other samples, although both are within
the cumulative range of the samples. The
P. curtus sample matches the medians of
all P. brongersmai traits with the single
exception of possessing a Suboc.
In body scalation, DorsMid, DorsPost,
and Subcaud are similar throughout all
samples, although DorsMidb is slightly
lower in the Singapore sample. A lower
median for DorsAnt characterizes the
Singapore pythons and does not appear
VOLUME 124, NUMBER 2 125
to be a sampling artifact owing to the
barely overlapping ranges. The DorsAnt
of the Siantar/Ceram juvenile is also less
than the insular samples of P. brongers-
mai, and the DorsAnt of the Mon-
Myanmar female lies at the upper limit
of insular P. brongersmai.InP. curtus,the
medians of the dorsal scale row characters
match those of the insular brongersmai
All samples share the same or near
identical medians for Subcaud, and this
body scalation trait is the only one in
which the mean of Keogh et al. (2001) is
strikingly different even though the range
is the same. We note again the high
variation (V) of Subcaud and suggest that
this variation might indicate that Subcaud
Regional differences are evident in
Ventral medians. Python curtus has the
fewest Ventral, and in our small sample,
the range is well below the minimum for
any P. brongersmai sample. The median
for Singapore P. brongersmai is lower
than the Sumatran one, although the
ranges broadly overlap. The Keogh mean
and range match our Sumatran samples.
The Ventral of both outliers are higher
than those of other P. brongersmai.The
Siantar/Ceram juvenile is at the maximum
end of the range of total Sumatran
sample, and the Mon-Myanmar python
has more Ventral than previously report-
ed for P. brongersmai.
Coloration.—Our quantification of col-
or pattern revealed a few regional and
specific differences. Rostral color (Rostrl)
darkens with maturity in P. brongersmai
and P. curtus, and remains light in the
Mon-Myanmar python and presumably
adult P. breitensteini. The PaStrp is
typically absent in mainland P. brongers-
mai and the Mon python and present in
most (.50%) Sumatran specimens and
the Siantar/Ceram juvenile. MiddStrp is
usually entire in P. brongersmai and the
Mon and Siantar/Ceram specimens, and
broken in P. curtus.
Ventral coloration differs for P. brei-
tensteini and P. curtus as compared to P
brongersmai and the Mon female. The
former populations have light VentralC
and dark SubcaudC. The reverse pattern
exists for most P. brongersmai and the
Siantar/Ceram juvenile. The Mon speci-
men has dark VentralC and SubcaudC.
We note that our ventral coloration
characters measure the extent of dark
pigmentation on the ventral surface.
None of the P. curtus group members
have immaculate venters, although some
P. breitensteini and curtus individuals
have solid black undersides of their tails.
Both Barker & Barker (1994) and
Keogh et al. (2001) provide details of
coloration in living specimens. We only
abstract a few of their coloration obser-
vations here. P. brongersmai changes the
darkness of its head from a usual darker
shade to a lighter shade; for black-headed
individuals, the head lightens to pale
silver gray (Keogh et al. 2001). The
lightening occurs relatively quickly (a
few hours) and reverts to the darker
shade within a day or less. The preceding
authors do not mention that the head
darkens in P. breitensteini and P. curtus,
and Auliya (2006) described black-headed
P. breitensteini although not darkening-
The lateral pattern of short-tailed
pythons is quite variable: however, the
description of the preceding authors and
our observations suggest the following
generalization. At mid-trunk, the sides of
adult P. brongersmai are dark brown on
the dorsal half and irregularly blotched
with light shades of brown to yellow on
the ventral half; these blotches have small
dark centers. The sides of adult P.
breitensteini and P. curtus have a reverse
pattern of lighter shades of brown dor-
sally extending to the venter as the
ground color; irregularly shaped and
sized dark blotches occur laterally and
ventrolaterally; and these blotches are
white to cream edged.
126 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
The final coloration feature to note is
tail coloration. From Keogh et al. (2001),
we generalize that all three species have
the tails dark dorsally. Subcaudals are
dark in P. brongersmai and P. curtus,and
light in P. breitensteini. The latter’s light
subcaudal coloration conflicts with our
observation on a single specimen. Auliya
(2006) does not mention the color of the
underside of the tail; however, his obser-
vation on a range of ground color
variation matching P. brongersmai is
noteworthy. The Mons python has three
distinct black spots on the subcaudals.
Geographic and Taxonomic Observations
The overall distribution of the Python
curtus species group (Fig. 1A, B) is a
peculiar one for Southeast Asian reptiles
and further contains disjunct occurrences
not observed in any other Asian squa-
mate. A few Asian frogs and reptiles
share the Borneo, Sumatra, and Malaya
Peninsula (south of Isthmus of Kra) with
the curtus group. Among the frogs, these
taxa are Pseudobufo and Metaphrynella.
Leptophryne and Pedostibes occur in the
preceding three areas and Java. Among
reptiles, only a few snakes (Anomochilus,
Dryocalamus subannulatus) and turtles
(Batagur borneoensis,Orlitia borneensis)
share the curtus group pattern. None of
the preceding, however, shares the specific
occurrence pattern in any of the three
areas. The spotty or disjunct occurrences
are unique to the curtus group taxa. The
absence of these pythons from large areas
of lowland Borneo likely results from
inadequate surveys of appropriate habi-
tats. Their absence from much of central
and south-central Sumatra, speculatively,
derives from human-mediated habitat
change and predation. This conjecture
might also explain their low abundance
and spotty occurrence in Thailand and
their absence in Brunei. We further note
that the three named populations of the
curtus group are allopatric. The potential
sympatry of P. brongersmai and P. curtus
in the Sibolga area derive from leather-
market specimens that were examined in
the Medan area slaughter houses and
whose origins were provided by animal
dealers, who had transported them from
the west coast of Sumatra and unlikely to
have captured them in the wild but rather
to have assembled (purchased) them from
the original collectors.
Our P. breitensteini and P. curtus sam-
ples are too small to draw any conclusions
of regional differentiation in morphology.
Our data simply confirm that the speci-
mens examined lie within the morpholog-
ical parameters established by Keogh et al.
(2001). The samples from the ‘brongers-
mai’ group are suggestive of regional
differentiation, although not conclusive;
nonetheless, we propose that the Mon-
Myanmar individual represents a unique
population and is described following our
discussion of morphological variation.
The largest P. brongersmai in our
sample is 179 cm TotalL (167 cm SVL)
from Thailand. The largest individuals
from the leather-market sample (Shine et
al. 1999) were ,2 m TotalL (185 cm
SVL). These data suggest that 2 m TotalL
is the maximum for P. brongersmai. While
P. brongersmai might have reached larger
sizes in the nineteenth century and earlier,
there are no published measurements, nor
even anecdotal ones, to support greater
Leather-market data (Shine et al.
1998b, 1999) demonstrate sexual dimor-
phism. Females average larger, usually in
the neighborhood of 10 cm.
Because of variably sized samples and
ones comprised mainly of juveniles, body
proportions offer theonly means to search
for morphometric differences among sam-
ples. Our data hint at allometry in two
proportions OrbD/HeadL and HeadL/
SVL. The juveniles from Bangka and
Siantar/Ceram have proportionately larg-
er eyes than adults (compare to Singapore
sample, Table 1). The Siantar/Ceram ju-
VOLUME 124, NUMBER 2 127
venile has a longer head (HeadL/SVL)
than any other P. brongersmai specimens,
including similar sized Bangka juveniles.
All proportions of the Myanmar python
lie within the ranges of the Singapore and
Sumatran samples. With the exception of
the proportionately longer tails, P. curtus
proportions strongly overlap those of P.
brongersmai. Given the significant body
length differences among the four color
morphs of Sumatran P. brongersmai
(Shine et al. 1998b), we predict that there
are additional differences in proportions
among these morphs, and that any re-
gional differences in our small samples are
a sampling artifact.
Lateral head scalation is similar among
the Singapore, Sumatra, Siantar/Ceram,
and Myanmar samples (Table 2). The
only striking difference is the low LorGrv
count in the Myanmar specimen; al-
though it lies within the range of Suma-
tran P. brongersmai, the latter sample has
a higher average as do all other samples,
including Keogh’s sample (Table 2). The
Myanmar specimens also have more
Inflab than the other samples, and its
value is at the maximum for that of the
other samples; however, note the higher
average and maximum of the Keogh
sample likely result from a different
definition for Inflab counts. All P.
brongersmai lack Suboc and the eye
contacts the supralabials; this is the
condition of the Myanmar and Siantar/
Ceram individuals. P. curtus and P.
breitensteini have a single modest-sized
Suboc separating the eye and suprala-
bials; one or two tiny scales may be
present in the subocular space.
Dorsal head scalation (Fig. 2) consists
of series of large paired shields lying
between the rostrum and an aggregate of
smaller and commonly irregular-shaped
scales in the parietal region. Using
measurements and their proportions (Ta-
ble 3) permits a quantitative examination
of size of these scales. The total length of
the series (MidHead) is proportionately
shorter (MidHead/SVL) in Singapore and
Myanmar pythons, but this proportionate
length difference among samples disap-
pears with MidHead/HeadL comparison.
Other proportional differences exist. The
relative length of the internasals (Intnas/
HeadL) is shorter in P. curtus than in P.
brongersmai. The frontonasal is propor-
tionately longer (Frtnas/HeadL) in the
Myanmar specimen than individuals and
samples of P. brongersmai and P. curtus.
The Myanmar individual’s frontal is
proportionately shorter (Front/HeadL)
than in P. curtus and P. brongersmai,
although the minimum range of some
brongersmai samples encompasses the
Myanmar proportion. In the latter in-
stance, the minima are of juvenile P.
brongersmai and the Myanmar individual
is an adult.
Body scalation displays both uniformi-
ty and populational differences. Dors-
Midb, DorsPost, and Subcaud are similar
in mean and range across P. curtus, the
brongersmai samples, Siantar/Ceram and
Myanmar individuals. Ventral (counts) is
lowest in P. curtus (155–159) and P.
breitensteini (150–169; Auliya 2006) and
contrast with our P. brongersmai samples
(165–178; 167–178 Keogh’s Sumatra,
Table 4). The Myanmar specimen has
184 Ventral. There is a potential regional
difference of DorsAnt. If the Singapore
sample (34–45) is characteristic of main-
land P. brongersmai, the mainland popu-
lations average fewer DorsAnt than the
Sumatran ones (44–53, Keogh’s 45–53).
The Myanmar DorsAnt is at the maxi-
mum of the Sumatran range.
Regional morphological variation ex-
ists among the P. brongersmai population.
Our small samples hint at these patterns
but cannot conclusively demonstrate
them. The molecular analyses of Keogh
et al. (2001:Fig. 1) demonstrated broad
genetic differences between P. brongers-
mai and the sister pair P. breitensteini and
P. curtus. Within P. brongersmai,they
demonstrate the absence of genetic differ-
128 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
entiation between the Sumatran color
morphs and provide a hint of slight
genetic differentiation of Sumatra and
mainland populations. The few morpho-
logical differences and the strong allopa-
try of the Mon/Myanmar python con-
vince us that this population represents a
divergent phylogenetic lineage.
Python kyaiktiyo, new species
Mon short-tailed python
Figs. 2, 3
Holotype.—USNM 572046, an adult
female from Myanmar, Mon State, in
Kyaiktiyo Wildlife Sanctuary at Yetagon
Myaung (17u269380N,97u059580E; ,390 m
elevation), approximately 5 km NE Kin-
mun, collected 11 Mar 2002 by Sai Wunna
Kyi, Win Zaw Lhon, and Hla Naing.
Diagnosis.—A member of the Python
curtus species group, differing from P.
breitensteini and P. curtus by the presence
of a subocular scale preventing eye-
supralabial contact, and from P. breiten-
steini,P. brongersmai,andP. curtus by
180 or more ventral scales and by the
presence of tan ocelli anteriorly in the
trunk’s middorsal stripe.
Description of holotype.—An adult
female, 116.6 cm SVL, 10.4 cm TailL
(140 cm SVL, 11.8 cm TailL in life),
25.9 mm SnEye, 20.6 mm NarEye,
6.7 mm OrbD, 11.1 mm Intnar, and
24.0 mm Interorb; proportions, TailL/
SVL 8.9%, HeadL/SVL 5.8%,HeadW/
SVL 3.4%, SnEye/SVL 38%,OrbD/
HeadL 9.8%, Intorb/HeadL 35.2%,Int-
nar/HedL 16.3%. Scalation right side for
bilateral traits: modest-sized semilunate
rostral posteriorly contacting paired in-
ternasals dorsally and laterally large nasal
and first supralabial, dorsal edge of
rostral with elongate crescent-shaped
sensory pit on each side; dorsally, pair
each of rectangular internasals, triangular
frontonasals, trapezoidal prefrontals,
semicircular frontals and irregular pen-
tagonal frontoparietals, frontonasals larg-
est 37%of total length of MidHead
suture; irregularly arranged scales behind
frontoparietals; four supraoculars above
right eye, three above left eye; large
triangular nasal with large naris dorsally
abutting posteriorly large anterior loreal,
then medium-sized posterior loreal sepa-
rated from eye by preocular, these four
lateral scales contact supralabials ventral-
ly, dorsally is a narrow loreal groove with
three small, elongate scales; eye bordered
posteriorly by two postoculars, dorsal one
largest; temporal groove with four small
scales between posterior supralabials and
anterior and posterior temporals; 12
Fig. 3. Dorsolateral view of the holotype of Python kyaiktiyo (USNM 572046) in life.
VOLUME 124, NUMBER 2 129
Suplab, sixth and seventh touching eye,
seventh Suplab largest, heat-sensory pits
on dorsoposterior corners of first and
second Suplab; 19 Inflab, first 8 Inflab
large (height 2 times width), remainder
half or less size of anterior Inflab; left and
right Inflab rows separate anteromedially
by small triangular mental. On body,
numerous small, smooth dorsal scales in
51 DorsAnt, 58 DorsMidb, and 36 Dors-
Post rows; 184 Ventral and 27 Subcaud;
precloacal scale entire (single).
Coloration in life (Fig. 3): general
impression is tan anteriorly becoming
rusty light brown on posterior two thirds
of trunk; dorsally head uniform yellowish
tan, loreals and adjacent supralabials dif-
fuse brown, temporal area dark brown
separated by white diagonal postocular
stripe from brown of snout; brown tempo-
ral area continuous to anterior trunk as
broad lateral stripe, separated from venter
by white border; stripe becoming series of
lateral dark brown blotches gradually
becoming larger and rusty brown, then at
third body length decreasing in size and
developing dusky brown halo, and at mid-
body becoming distinctly smaller, widely
spaced and confined ventrolaterally; dor-
sally, uniform tan of head clefted middor-
sally by rusty brown stripe becoming
progressively broader posteriorly compress-
ing the tan into narrow dorsolateral stripes
at midbody and posteriorad, anteriorly
middorsal stripe with tan ocelli, which soon
fuse into irregular edged and diffuse dusky
stripe; tail uniformly dark brown dorsally
and laterally. Anteriorly trunk white ven-
trolaterally; ventrally chin and throat white,
remainder of venter cream progressively
filled with dusky brown mottling; underside
of tail cream with three dark brown spots.
Coloration in preservative: still boldly
colored although muted by preservation.
Pattern of dark and light markings
unchanged. Dorsal ground color muted
tannish brown, ventrally tannish white
and elongate dorsal and lateral spots also
Distribution.—Presently known only
from the type locality; presumably an
isolated population occurring on the
western face of the Tenghyo Range.
Etymology.—The specific epithet de-
rives from the Kyaiktiyo Pagoda (Golden
Rock) area. Pronunciation sounds like
Jack-T-Yo. The specific epithet is pro-
posed as a noun in apposition. Legend
says that the balancing rock owes its
stability to a strand of Buddha’s hair,
which Buddha gave to the hermit Taik
Tha during one of the Bodhisattva’s
many visits to earth. The hair is enshrined
in a miniature pagoda built on top of the
Natural history notes.—The female
contained six flexibly shelled eggs (,9cm
36 cm) in her oviducts. She was
discovered in a small dry streambed in
an area of low secondary growth scrub
near a small betel palm plantation. Her
live TotalL was 152 cm, and she weighed
Key to Python curtus group species
1. Usually one or more subocular scales
separating eye from supralabial scales;
fewerthan166ventralscales ....... 2
– No subocular scale, eye contact with
supralabial scales; 165 or more ven-
tral scales ...................... 3
2. Frontoparietal scales not or barely in
contact along midline ...... P. curtus
– Frontoparietal scales broadly in con-
tact along midline . . . . P. breitensteini
3. Ventral scales range 165 to 178;
usually underside of tail light with
some dark mottling . . . P. brongersmai
– Ventral scales more than 180; underside
oftailwith fewdarkspots .......
....................... P. kyaiktiyo
We purposefully retain the reticulated
python in the genus Python. This conser-
vative position derives from our past
experience with the instability of Austra-
lian python generic names that required a
decade to reach a consensus of use and
130 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
from the likelihood that further molecular
studies with a variety of genes will reveal
differences in phylogenetic relationships
(see Wiens et al. 2010 for a discussion of
the fluctuation in relationship estimates
with different DNA sequence data sets).
The collection of voucher locality
records and the examination of specimens
relied on the cordial assistance of many
collections managers and curatorial assis-
tants. We sincerely thank them all: D.
Kizirian & R. Pascocello, American
Museum of Natural History; R. Stoelting
& J. V. Vindum, Californian Academy of
Sciences; M. Kageya´ma, University of
Colorado; J. Rosado, Museum of Com-
parative Zoology, Harvard; K. Thira-
khupt, Natural History Museum – Chu-
lalongkorn University; C. J. McCarthy,
Natural History Museum, London; K. K.
P. Lim, Raffles Museum of Biodiversity
Research, Zoological Reference Collec-
tion, National University of Singapore.
We offer a special thank you to D.
Barker for promptly providing locality
data and comments on short-tailed py-
thon distributions, to T. Chan-ard for his
information on the northernmost Thai-
land record of P. brongersmai,toL.L.
Grismer for sharing his field observations
on these pythons in Malaysia, to T.
Nguyen for identifying the origin of the
recent Vietnamese records, and to R.
Shine for promptly clarifying his earliest
Sumatran market locality. We are always
indebted to C. J. McCarthy, Natural
History Museum, London, for his whole-
hearted support in correspondence and
specimen examination during visits to the
BMNH collection. A number of col-
leagues have reviewed various drafts of
this manuscript; we thank the following
individuals R. McDiarmid, M. O’Shea, R.
Reynolds, G. Rodda, G. Vogel, and P.
Zug, and appreciated their advice thereby
improving the clarity of the manuscript.
We also gratefully acknowledge and thank
A. Wynn for distribution map production
and M. D. Griffin for the illustration of
the new python’s head scales.
The Myanmar Herpetological Survey
has received major support from the
Biodiversity program of the National
Sciences Foundation (DEB-9971861,
DEB-0451832). Additional support for
our fieldwork in the Kyaiktiyo area came
from the Smithsonian-National Museum
of Natural History’s Biodiversity Pro-
gram. The Myanmar Nature and Wildlife
Conservation Division provided logistic
and personnel support.
Abel, F. 1998. Status, population biology and
conservation of the water monitor (Varanus
salvator), the reticulated python (Python
reticulatus), and the blood python (Python
curtus) in Sumatra and Kalimanatan, Indo-
nesia: Project report North Sumatra.
Pp. 111–117 in W. Erdelen, ed., Conservation,
Trade and Sustainable Use of Lizards and
Snakes in Indonesia. Mertensiella 9, Rhein-
Auliya, M. A. 2006. Taxonomy, life history and
conservation of giant reptiles in West Kali-
mantan. Natur und Tier Verlag, Mu¨nster,
Babcock, G. D. 1966. History of the United States
Rubber Company: a case study in corporation
management (Indiana Business Report 39).
Bureau of Business Research, Graduate School
of Business, Indiana University, 477 pp.
Barker, D., & T. Barker. 1994. The blood python and
other subspecies of the short-tailed python
Python curtus.—The Vivarium 6(3):30–35.
Bezuijen, M. R., B. Vinn, & L. Sieng. 2009. A
collection of amphibians and reptiles from the
Mekong River, north-eastern Cambodia.—
Boulenger, G. A. 1893. Catalog of the snakes in the
British Museum (Natural History). vol. 1.
Typhlopidae, Glauconiidae, Boidae, Ilysiidae,
Uropeltidae, Xenopeltidae, and Colubridae
Aglyphae, part. Trustees of the British Muse-
um (Natural History), London, xiii +448 pp.
———. 1912. A vertebrate fauna of the Malay
Peninsula from the Isthmus of Kra to
Singapore including the adjacent islands.
Reptilia and Batrachia. Taylor and Francis,
London, xiii +286 pp.
VOLUME 124, NUMBER 2 131
———. 1920. Reptiles and batrachians collected in
Korinchi, West Sumatra by Messrs. H. C.
Robinson and C. Boden Kloss.—Journal of
the Federated Malay States Museums
Bourret, R. 1936. Les Serpents de l’Indochine. Tome
II. Catalog Syste´matique Descriptif. Henri
Basuyau & Cie, Toulouse, 505 pp.
Brongersma, L. D. 1947. On the subspecies of Python
curtus Schlegel occurring in Sumatra.—Pro-
ceedings of the Koninklijke Nederlandsche
van Akademie Wetenschappen 50(6):666–671.
Campden-Main, S. M. 1970. A Field Guide to the
Snakes of South Vietnam. Division of Rep-
tiles and Amphibians, U.S. Natural Museum,
Smithsonian Institution, Washington, DC,
Chan-ard, T., W. Grossman, A. Gumprecht, &
K.-D. Schulz. 1999. Amphibians and reptiles
of peninsular Malaysia and Thailand: an
illustrated checklist. Bushmaster Publications,
Wu¨ rselen, Germany, 240 pp.
Cox, M. J. 1991. The snakes of Thailand and their
husbandry. Krieger Publishing Co., Malabar,
Florida, 526 pp.
Crampton, W., A. Goodwin, A. Lockett, & S.
Sinkins. 1990. Oxford University herpetolog-
ical expedition to Ujung Kulon National
Park, West Java 1990. Preliminary report
[unpublished], Bogor, 25 pp.
David, P., & G. Vogel. 1996. The snakes of
Sumatra. An annotated checklist and key
with natural history notes. Edition Chimaira,
Frankfurt am Main, 260 pp.
de Haas, C. P. J. 1950. Checklist of the snakes of the
Indo-Australian Archipelago (Reptiles, Ophid-
de Rooij, N. 1917. The reptiles of the Indo-
Australian Archipelago. II. Ophidia. E. J.
Brill, Leiden, 334 pp.
Dring, J. C. M., C. J. McCarthy, & A. J. Whitten.
1990. The terrestrial herpetofauna of the
Mentawai Islands, Indonesia.—Indo-Malay-
an Zoology 6(1989):119–132.
Gauthier, J., M. Kearney, & R. L. Bezy. 2008.
Homology of cephalic scales in xantusiid
lizards, with comments on night lizard
phylogeny and morphological evolution.—
Journal of Herpetology 42(4):708–722.
Grismer, L. L., & P. K. Aun. 2008. Diversity,
endemism, and conservation of the amphib-
ians and reptiles of southern peninsular
Malaysia and its offshore islands.—Herpeto-
logical Review 39(3):270–281.
———, N. Thy, C. Thou, & J. L. Grismer. 2008.
Checklist of the amphibians and reptiles of
the Cardamom region of southwestern Cam-
bodia.—Cambodian Journal of Natural His-
Groombridge, B., & R. Luxmoore. 1991. Pythons in
South East Asia: a review of distribution,
status and trade in three selected species. A
report to CITES Secretariat August 1990.
Secretariat of the Convention on Internation-
al Trade in Endangered Species of Wild
Fauna and Flora, Lausanne, 127 pp.
in den Bosch, H. A. J. 1985. Snakes of Sulawesi:
checklist, key and additional biogeographical
remarks.—Zoologische Verhandelingen 217:
Inger, R. F., & T. F. Lian. 1996. The Natural
History of Amphibians and Reptiles in
Sabah. Natural History Publications (Bor-
neo), 101 pp.
Keogh, J. S., D. G. Barker, & R. Shine. 2001.
Heavily exploited but poorly known: system-
atics and biogeography of commercially
harvested pythons (Python curtus group) in
Southeast Asia.—Biological Journal of the
Linnean Society 73(1):113–129.
Lim, B. L. 1955. Snakes collected near Kuala
Lumpur.—Malayan Nature Journal 9:122–
J. Kosuch. 2002. Amphibians & reptiles of
Mount Kinabalu (North Borneo). A. R. G.
Gantner Verlag, Ruggell, Liechtenstein, 424 pp.
Mann, W. M. 1938. The National Geographic
Society-Smithsonian Expedition to the Dutch
East Indies.—Explorations and Field-Work
of the Smithsonian Institution in 1937,
Manthey, U., & W. Grossmann. 1997. Amphibien &
Reptilien Su¨ dostasiens. Natur und Tier Ver-
lag, Mu¨ nster, Germany, 512 pp.
Mertens, R. 1957a. Amphibien und Reptilien aus
dem a¨ ussersten Westen Javas und von be-
nachbarten Eilanden.—Treubia 24(1):83–105.
———. 1957b. Zur Herpetofauna von Ostjava und
Bali.—Senckenbergiana biologica 38(1/2):
Murphy, J. C., H. K. Voris, & D. R. Karns. 1994. A
field guide and key to the snakes of the
Danum Valley, a Bornean tropical forest
ecosystem.—Bulletin of the Chicago Herpe-
tological Society 29(7):133–151.
Nguyen, V. S., T. C. Ho, & Q. T. Nguyen. 2009.
Herpetofauna of Vietnam. Edition Chimaira,
Frankfurt am Main, 768 pp.
Noor, H. M. 1964? Blood python (Python curtus): a
brooding affair.—Nature Malaysia 18:78–79.
[not seen; from Auliya (2006)]
Orlov, N. L., S. A. Ryabov, V. S. Nguyen, & Q. T.
Nguen [sic]. 2003. New records and data on
the poorly known snakes of Vietnam.—
Russian Journal of Herpetology 10(3):217–
132 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
Pauwels, O. S. G., O.-A. Laohawat, P. David, R.
Bour, P. Dangsee, C. Puangjit, & C. Chim-
sunchart. 2000. Herpetological investigations
in Phang-Nga Province, southern peninsular
Thailand, with a list of reptile species and
notes on their biology.—Dumerilia 4(2):123–
Riquier, M. A. 1998. Status, population biology and
conservation of the water monitor (Varanus
salvator), the reticulated python (Python
reticulatus), and the blood python (Python
curtus) in Sumatra and Kalimantan, Indone-
sia – Project Report Kalimantan. Pp. 119–129
in W. Erdelen, ed., Conservation, Trade and
Sustainable Use of Lizards and Snakes in
Indonesia. Mertensiella 9, Rheinbach.
Shine, R., P. Harlow, Ambariyanto, Boeadi, Mum-
puni, & J. S. Keogh. 1998a. Monitoring
monitors: a biological perspective on the
commercial harvesting of Indonesian reptiles.
Pp. 61–68 in W. Erdelen, ed., Conservation,
Trade and Sustainable Use of Lizards and
Snakes in Indonesia. Mertensiella 9, Rhein-
———, Ambariyanto, P. S. Harlow, & Mumpuni.
1998b. Ecological divergence among sympat-
ric color morphs in blood pythons, Python
———, ———, ———, & ———. 1999. Ecological
attributes of two commercially-harvested
Python species in northern Sumatra.—Jour-
nal of Herpetology 33(2):249–257.
Steindachner, F. 1880. U
¨ber eine neue Pythonart
(Python Breitensteini) aus Borneo.—Sitzungs-
berichte der mathematisch-naturwissenschaft-
lichen classe der kaiserlichen Akademie der
Wissenschaften, I Abteilung 82:267–268.
Stuart, B. L. 1999. Amphibians and reptiles.
Pp. 43–67 in J. W. Duckworth, R. E. Salter
and K. Khounboline, eds., Wildlife in Lao
PDR: 1999 Status Report. IUCN-The World
Conservation Union/Wildlife Conservation
Society/Center for Protected Areas and Wa-
tershed Management, Vientiane, 275 pp.
———, K. Sok, & T. Neang. 2006. A collection of
amphibians and reptiles from hilly eastern
Cambodia.—The Raffles Bulletin of Zoology
Stull, O. G. 1938. Three new subspecies of the family
Boidae.—Occasional Papers of the Boston
Society of Natural History 8:297–300.
Taylor, E. H. 1965. The serpents of Thailand and
adjacent waters.—University of Kansas Sci-
ence Bulletin 45(9):609–1096.
Teynie´ , A., P. David, A. Ohler, & K. Luanglath.
2004. Notes on a collection of amphibians
and reptiles from southern Laos, with a
discussion of the occurrence of Indo-Malayan
Tirant, G. 1885. Notes sur les reptiles et les
batraciens de la Cochinchine et du Cam-
bodge. Pp. 209–246 in Cochinchine franc¸aise:
excursions et reconnaissances. Imprimerie du
Gouvernement, Saigon. [not seen; from Au-
Tran, K., H. P. Nguyen, K. H. Nguyen, & V. S.
Nguyen. 1992. Phan III. Bo sat, luong cu.
Pp. 180–237 in D. H. Tuˆ , ed., Red Data Book
of Vietnam, vol. 1: Animals. Science and
Technics Publishing House, Hanoi, 396 pp.
[not seen; from Auliya (2006)]
Werner, F. 1900. Reptilien und Batrachier aus
Sumatra, gesammelt von Herrn Gustav
Schneider jr. im Jahre 1897–1998.—Zoolo-
gische Jahrbu¨ cher, Abtheilung fu¨ r Systema-
tik, Gographie und Biologie der Thiere
———. 1927. U
¨ber Schlangen von Medan, Suma-
tra’s Ostku¨ste.—Miscellanea Zoologica Su-
Westermann, J. H. 1942. Snakes from Bangka and
Wiens, J. J., C. A. Kuczynski, & P. R. Stephens.
2010. Discordant mitochondrial and nuclear
gene phylogenies in emydid turtles: implica-
tions for speciation and conservation.—Bio-
logical Journal of the Linnean Society
Yacob, S. 2007. Model of welfare capitalism? The
United States Rubber Company in Southeast
Asia, 1910–1942.—Enterprise & Society: The
International Journal of Business History
Associate Editor: Robert P. Reynolds.
Definition of characters
Mensural.—All bilateral measurements recorded
from right side; in mm. (15 characters)
Front 5Frontal length: midline distance from
anterior to posterior end of frontal.
Frtnas 5Frontonasal length: midline distance from
anterior to posterior end of frontonasal.
HeadL 5Head length: straight-line, horizontal
distance from tip of snout to posterior corner
of jaw, external and posterior to quadrate-
HeadW 5Head width: straight-line, transverse
distance from left to right edges of head at
posterior edges of last supralabials.
Intnar 5Internarial distance: transverse distance
between left and right nares (anterodorsal
VOLUME 124, NUMBER 2 133
Intnas 5Internasal length: midline distance from
anterior to posterior end of internasal scale.
Intorb 5Interorbital distance: transverse distance
between left and right anterodorsal edges of
MidHead 5Midline suture length: distance from
the rostral-internasal suture notch to the
posterior end of frontal scale. All suture
lengths measured along scales of right side.
NarEye 5Naris-eye distance: distance between
naris (anterior edge) and anterior corner of
OrbD 5Orbit diameter: distance from anteromedial
to posteromedial edge of orbit (maximum
Prefron 5Prefrontal length: midline distance from
anterior to posterior end of prefrontal scale.
SnEye 5Snout-eye distance: distance between
middle/tip of snout to anterior corner of orbit.
SVL 5Snout-vent length: distance from tip of snout
to vent, posteromedial edge of anal scale.
TailL 5Tail length: distance from vent (postero-
medial edge of anal scale) to tip of tail.
TotalL 5Total length: distance from tip of snout to
tip of tail, obtained by adding SVL and
Scalation.—Asterisk denotes bilateral characters
recorded on both left and right sides, except
infralabial creases (right side only). (19 characters)
DorsAnt 5Rows of dorsal scales anteriorly:
number of rows dorsal scales at level of tenth
ventral scale behind head.
DorsMidb 5Rows of dorsal scales at midbody:
number of dorsal scale rows at level of
eightieth ventral scale behind head.
DorsPost 5Rows of dorsal scales posteriorly:
number of dorsal scale rows at level of tenth
ventral scale in front of vent.
Inflab 5Infralabials*: number of scales edging
mouth (lower mandible) from first touching
mental to last enlarged scale below last
InflabEn 5Infralabials enlarged: number of ante-
rior infralabials higher than long.
InflCAnt 5Infralabial anterior crease: anteriormost
infralabial scale on which anterior crease
InflCAL 5Infralabial anterior crease length: num-
ber of anterior infralabial scale within crease.
InflCPo 5Infralabial posterior crease: anteriormost
infralabial scale on which posterior crease
InflCPL 5Infralabials: number of infralabial scale
LorGrv 5Loreal groove scales*: number of scales
in loreal groove from nasal to preocular scale
in contact with dorsal head plates.
Preoc 5Preoculars*: number of scales touching eye
between anterior supraocular and suprala-
Postoc 5Postoculars*: number of scales touching
eye between posterior supraocular scale and
Subcaud 5Subcaudals: number of subcaudal scales,
excluding tip scale.
Suboc 5Suboculars*: number of scales touching
eye and supralabial scale(s).
Suplab 5Supralabials*: number of scales edging upper
mandible from first touching rostral to last
enlarged scale toward posterior edge of mouth.
SuplabEye 5Supralabials in eye: number of scales
Supoc 5Supraoculars*: number of scales touching
eye and medial head plates.
TemporG 5Temporal groove*: number of scales in
temporal groove touching supralabials from
postocular scale to suture between the second
and third supralabials after supralabial
Ventral 5Ventrals: number of ventral scales
(transverse width .midline length) from
neck to anal scale.
Color pattern.—All observations on color pattern
derive from preserved specimens and recorded from
right side if pattern is bilateral. (9 characters)
BodyLatW 5Lateral stripe width: width of lateral
dark stripe at level of first ventral scale,
measured by number of dorsal scales in scale
row traversing stripe (counted diagonally
from ventral edge).
Mental 5Mental: mental scale light, less than 40%
pigmented  or dark, more than 60%
MiddStrp 5Middorsal stripe: complete or inter-
rupted light middorsal strip on posterior half
of trunk, absent (0) or present (1).
PaStrp 5Parietal stripe: narrow dark transverse
strip in parietal area, absent or barely evident
 or present .
PostocStr 5Postocular stripe: supralabial on which
anterior edge of white stripe begins.
PostocVntrl 5Postocular-ventrolateral stripe align-
ment: alignment of stripes at juncture on
labial scales, parallel and continuous  or
Rostrl 5Rostral: color of rostral scale, light  or
darkly pigmented, occasionally with a light
SubcaudC 5Color of subcaudal scales: subcaudal
scales dark throughout  or with light areas
on posterior third of tail .
VentralC 5Color of ventrum: ventral scales light or
minimally pigmented (,10%)  or dusky
134 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
Borneo: Kalimantan. Kalimantan Barat – ZRC
Thailand: no specific locality CUB R08.11.1.
This specimen is assumed to derive from
Thailand although the available museum
records cannot confirm this assumption.
Malaysia: Malacca [5Malaysia] BMNH 1891.7.
24.1 no specific locality; Selangor – CAS
8471, USNM 53427 Kuala Lumpur; Johor –
ZRC 2.3087 Kota Tinggi.
Singapore: BMNH 1906.2.28.18, CAS 16748,
MCZ R29779, ZRC 2.3089–2.3091 Singa-
pore, various localities therein.
Sumatra: no specific locality USNM 89399-400,
94431; Sumatera Barat – BMNH 1915.
10.2.31 Korninchi [Kerinci]; Sumatera Utara
– AMNH 50823, 50993-994 US rubber
planations [probably near Kisaran southeast
of Medan], UCM 59069, USNM 103576
Siantar; Bangka-Belitung UCM 57837,
58598-602, 58628, 58768 Bangka Island;
Indonesia: – Riau MCZ R34127 state only.
Sumatra: Sumatera Barat – USNM 70942 Mt.
Kabor [likely northeast of Fort de Kock];
Sumatera Bengkulu – USNM 70943 Kaba
Wetan [near Kapahiang].
Myanmar: Mon – USNM 572046 Kinum.
Molluca: Ceram – USNM 103513 Piroe [Piru];
actually Siantar, Sumatra, see above and
explanation in Distribution section.
Map resources for Fig. 1
The geo-coordinate data are often approxima-
tions because of imprecision in original locality data.
Arrangement is within country north to south, then
east to west.
Myanmar: P. kyaiktiyo – Mon State (17u269390N,
97u059580E), see Appendix 2. Mentioned
without data-support – Cox 1991.
Laos: no P. curtus sp group taxa reported – Stuart
1999, Teynie´ et al. 2004.
Cambodia: no P. curtus sp group taxa reported;
Bezuijen et al., 2009, Grismer et al. 2008,
Stuart et al. 2006. Mentioned without
data support – Cox 1991, Barker & Barker
Vietnam: P. brongersmai – Saigon area (Ho Chi
Minh City 10u459N, 106u409E) Tirant 1885,
Tran et al. 1992, Nguyen et al. 2009; Binh
Thuan (,11uN, 108uE) Nguyen et al. 2009;
Ca Mau (,9u199N, 105u069E) Nguyen et al.
Thailand: P. brongersmai – Kaeng Krachen Natl.
Park (12u549N, 99u369E) photo-record, Chan-
ard, in litt. Jun 2010; Phang-Nga (8u289N,
98u329E) Pauwels et al. 2000; Krabi (8u049N,
98u559E) Chan-ard et al. 1999; Pattani
(6u509N, 101u189E) Taylor 1965.
Malaysia: P. brongersmai – Pinang Island (Pulau
Pinang; 5u249N, 100u149E; there are four
Pulau Pinang in Malaysia) Keogh et al.
2001, Chan-ard et al. 1999; Ipoh (4u359N,
101u059E) Chan-ard et al. 1999; Tasik Chini
(3u239N, 102u539E) L. Grismer, in litt. Jun
2010; Selangor (3u219N, 101u159E) Noor,
1964?, see Appendix 2; Sungei Buloh
(3u159N, 101u189E) Lim 1955; Gombak
(3u159N, 101u339E) L. Grismer, in litt. Jun
2010; Kuala Lumpur (3u109N, 101u409E)
Keogh et al. 2001, see Appendix 2; Bangi
(2u559N, 101u479E) L. Grismer, in litt. Jun
2010; Port Dickson (2u319N, 101u489E)
Brongersma 1947; Johor Baharu (1u289N,
103u459E) Keogh et al. 2001; Kota Tinggi
(1u439N, 103u549E) see Appendix 2. Not
reported from Seribuat Archipelago, Grismer
& Aun 2008.
Singapore: P. brongersmai – miscellaneous localities
throughout island (1u209N, 103u509E), de
Rooij 1917, Stull 1938; see Appendix 2.
Sumatra: P. brongersmai – Kuala Simpang (now
Kulasimpang 4u169460N, 98u039500E) Bron-
gersma 1947, David & Vogel 1996; Deli (now
Labuhandeli 3u449590N, 98u419000E) Bron-
gersma 1947, David & Vogel 1996; Medan
(3u359N, 98u409E) de Rooij 1917, Werner
1927, Brongersma 1947, David & Vogel 1996,
Shine et al. 1998b, 1999, Keogh et al. 2001;
Seisuka (3u259N, 99u279E) Shine et al. 1998b,
1999; Berastagi & Pispis (3u109N, 99u019E)
David & Vogel 1996; Sungaibejangkar
(3u079N, 99u319E) David & Vogel 1996;
U.S. rubber plantations (probably near Ki-
saran southeast of Medan) (3u009N, 99u359E)
see Appendix 2; Siantar (3u009N, 99u109E) see
Appendix 2; Rantauprapat (2u069N, 99u509E)
Shine et al. 1998b, 1999, Keogh et al. 2001;
Surbo Dolock (now Seribudolok 2u569N,
98u379E) Werner 1900, de Rooij 1917, David
& Vogel 1996; Assahan (now Aek-kanan
1u579580N, 99u479590E) Brongersma 1947;
Sibolga (1u459N, 98u489E) Keogh et al.
2001; Cikampak (1u439N, 100u159E) Shine
et al. 1998b, 1999; Aek Batu (1u409N,
101u109E) Abel 1998; Pinang and Riau group
VOLUME 124, NUMBER 2 135
(0u479N, 104u339E) Barker & Barker 1994;
Linnga (0u129S, 104u339E) Barker & Barker
1994; Singkep (0u299390S, 104u259570E) Bark-
er & Barker 1994; Korninchi (Kerinci;
1u429S, 101u169E), see Appendix 2, also
Boulenger 1920; Siulakderas (5Siolak Daras,
Brongersma 1947) (1u559S, 101u189E) Bou-
lenger 1920, David & Vogel 1996; Bangka
Island (2u209S, 106u059E) Westermann 1942,
Barker & Barker 1994, Keogh et al. 2001,
see Appendix 2; Belitung Island (2u509000S,
107u559000E) Barker & Barker 1994;
Palembang (2u559S, 104u459E) Shine et al.
P. curtus – Sibolga (1u459N, 98u489E) Shine et al.
1999, Keogh et al. 2001; Mt. Kabor (NE of
Fort-de-Kock, 0u199S, 100u229E) Brongersma
1947, David & Vogel 1996; Padang (27
Padangs on Sumatra proper, presumed one
is 0u579S, 100u219E) David & Vogel 1996;
Siberut, Mentawai Islands (1u369S, 99u119E)
Dring et al. 1990; Kaba Wetan (3u399020S,
102u339460E) Brongersma 1947, David &
Vogel 1996, see Appendix 2; Bengkulu
(3u489S, 102u169E) Keogh et al. 2001; Palem-
bang (2u559S, 104u459E) Shine et al. 1998a;
Bandar Lampung (5u279S, 105u169E) Barker
& Barker 1994, Keogh et al. 2001.
Dipung (geo-coordinates not located) Keogh et al.
Java: no P. curtus sp group taxa reported – de Rooij
1917, Mertens 1957a, 1957b, Crampton et al.
1990, Manthey & Grossmann 1997.
Borneo: P. breitensteini – Kalimantan: Sambas
(1u209N, 109u159E) Barker & Barker 1994,
Riquier 1998, Keogh et al. 2001; Muarateweh
(0u579N, 114u539E) Steindachner 1880; Pon-
tianak (0u019N, 109u209E) Keogh et al. 2001,
see Appendix 2; Sambas (2u009N, 109u219E)
Riquier 1998; Telang (2u079S, 115u009E) Tay-
lor 1965; Banjarmasin (3u209S, 114u359E),
Barker & Barker 1994, Keogh et al. 2001.
Sarawak: Lawas (4u519N, 115u249E) de Rooij
1917; Sibu (2u189N, 111u499E) de Rooij 1917;
Kuching area (1u339N, 110u209E) de Rooij
1917, Keogh et al 2001; Telang (? 5Tekalong
1u129N, 111u339E) de Rooij 1917. Brunei:
Lawas (in error; actually in Sarawak) de Rooij
1917; no specimens reported. Sabah: Danum
Valley (5u129N, 117u509E) Murphy et al. 1994;
Poring (Kampong Paring, 5u599N, 116u149E)
Malkmus et al. 2002; without specific locality
Inger & Lian 1996.
Molluca: Python brongersmai – Ceram, Piroe
(5Piru, 3u209S, 128u209E), incorrect locality
data, see Appendix 2.
136 PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON