Phylogeographic analysis of the green python, Morelia viridis , reveals cryptic diversity

Toto je HTML verze souboru http://people.uncw.edu/mccartneym/documents/Rawlingssnakepaper.pdf.
G o o g l e automaticky vytváří HTML verze dokumentů, když prohledává Net.
Pro vytvoření odkazu na stránku nebo její uložení do záložek použijte následující odkaz:

http://www.google.com/search?q=cache:UMU-PwLA4HAJ:people.uncw.edu/mccartneym/documents/Rawlingssnakepaper.pdf+(Phylogeographic+analysis+of+the+green+python+Morelia+viridis,+reveals+cryptic+diversity+&hl=cs&ct=clnk&cd=1&client=firefox-a

Google není spojen s autory této stránky ani odpovědný za její obsah.

Tyto vyhledávané výrazy byly zvýrazněny:

phylogeographic

analysis

green

python

morelia

viridis

reveals

cryptic

diversity

Page 1

Phylogeographic analysis of the green python, Morelia viridis,

reveals cryptic diversity

Lesley H. Rawlings

a,b

and Stephen C. Donnellan

Evolutionary Biology Unit and Centre for Evolutionary Biology and Biodiversity, South Australian Museum, North Terrace,

Adelaide, SA 5000, Australia

Genetics Department, University of Adelaide, Adelaide, SA 5000, Australia

Received 19 December 2001; revised 19 September 2002

Abstract

Green pythons, which are regionally variable in colour patterns, are found throughout the lowland rainforest of New Guinea

and adjacent far northeastern Australia. The species is popular in commercial trade and management of this trade and its

impacts on natural populations could be assisted by molecular identification tools. We used mitochondrial nucleotide sequences

and a limited allozyme data to test whether significantly differentiated populations occur within the species range. Phylogenetic

analysis of mtDNA sequences revealed hierarchal phylogeographic structure both within New Guinea and between New Guinea

and Australia. Strongly supported reciprocally monophyletic mitochondrial lineages, northern and southern, were found either

side of the central mountain range that runs nearly the length of New Guinea. Limited allozyme data suggest that population

differentiation is reflected in the nuclear as well as the mitochondrial genome. A previous morphological analysis did not find

any phenotypic concordance with the pattern of differentiation observed in the molecular data. The southern mitochondrial

lineage includes all of the Australian haplotypes, which forma single lineage, nested among the southern New Guinean

haplotypes.

Keywords: Mitochondrial DNA; Control region; Snake; Python; Phylogeography; New Guinea

1. Introduction

Pythons are a family of non-venomous constricting

snakes found fromAustralia, through New Guinea,

Indonesia, southern Asia to Africa. The family reaches

its greatest generic level diversity in Australia and New

Guinea. Many of the species are spectacularly coloured

and patterned. In some species these attributes show

regional variation indicating that the taxon maybe

polytypic. The green python, Morelia viridis (Schlegel,

1872), is found throughout the island of New Guinea

and its offshore islands (with the exception of the Bis-

marck Archipelago) and in a small rainforest block in

northeastern Australia (Barker and Barker, 1994;

McDowell, 1975; OÕShea, 1996) (Fig. 1). It has an alti-

tudinal range from0 to 2000 mabove sea level, inhab-

iting lowland and lower montane forests (OÕShea, 1996).

McDowell (1975) found no significant distinguishing

morphological features between populations, with the

possible exception that juveniles fromthe Sandaun

Province, Papua New Guinea are brick red rather than

the more common yellow or orange (Parker, 1982).

Anecdotal evidence suggests that colour variation and

markings may be diagnostic for some island popula-

tions, e.g., Aru and Biak Islands (F. Yuwono and

D. McCrae, pers. comm). Furthermore, the speciesÕ wide

geographic and altitudinal distribution are attributes

often indicative of cryptic species level diversity (Don-

nellan et al., 1993).

Aside fromthe taxonomic interest, understanding

evolutionary relationships among green python popu-

lations also has applications in two other areas. Firstly,

the striking appearance of green pythons makes them

popular exhibits in zoos and in the pet trade. Much of

Molecular Phylogenetics and Evolution 27 (2003) 36–44

www.elsevier.com/locate/ympev

MOLECULAR

PHYLOGENETICS

AND

EVOLUTION

Corresponding author. Fax: +61-8-8207-7222.

E-mail address: donnellan.steve@saugov.sa.gov.au (S.C. Donne-

llan).

doi:10.1016/S1055-7903(02)00396-2

Page 2

the pet trade is supplied fromIrian Jaya, although there

is now considerable success with captive breeding in the

USA and Europe fromIrian Jayan stock (D. Barker,

pers. comm; D. MacCrae, pers. comm). In contrast

there has been limited captive breeding success with

Australian green pythons (Barker and Barker, 1994) and

Australian law prohibits the unregulated capture of wild

snakes. These factors combining with its very limited

Australian range, have led to the green python being

given special status for conservation management

(Banks, 1999). In a global effort to conserve biodiversity,

zoological parks of the Australasian region established

the Australasian Species Management Program (ASMP)

to determine conservation selection criteria and appro-

priate management strategies for species at risk (Banks,

1999). These management programs are overseen by

Taxon Advisory Groups (TAGs). Two of the main goals

for the green python TAG are to clarify suggested dif-

ferences between Australian and New Guinean snakes

and to achieve captive breeding of Australian specimens.

In earlier years, before quarantine restrictions closed off

legal importation, green pythons for exhibit in Austra-

lian zoos were imported from New Guinea as well as

collected fromQueensland. Therefore, to establish a

breeding programin accordance with TAG guidelines,

the origin of green pythons already in zoos needs to be

determined.

Secondly, the rarity of the green python in Australian

herpetologistsÕ collections, and consequent high market

price, and the strict regulations on importing exotic

specimens into Australia have encouraged the continued

importation of green pythons, now an illegal activity,

into Australia (McDowell, 1997). Currently, these ille-

gal imports can easily be absorbed into the Australian

pet trade undetected. The development of molecular

makers that can be used to provenance and identify

individual animals would ensure better quarantine

management.

In the present study, we investigated the genetic

population structure of green pythons by examining

samples from representative localities in Australian and

New Guinea. We used nucleotide sequences of two

mitochondrial genes and incorporated limited allozyme

electrophoretic data to examine whether the species

shows strong phylogeographic structuring and in par-

ticular whether the Australian population could be

distinguished fromNew Guinean populations. Phylog-

eographic analyses of other snake species have provided

powerful insights into their population history and

systematics (e.g., Ashton and de Queiroz, 2001; Bur-

brink et al., 2001; Creer et al., 2001; Henderson and

Hedges, 1995; Keogh et al., 2001; Rodriguez-Robles

and De Jesus-Escobar, 2000; Zamudio and Greene,

1997).

Fig. 1. Map of northern Australia and New Guinea showing the range of M. viridis (light shading) and sample locations. The locality code is: Aru,

Aru Island; Aus, Queensland; Bia, Biak Island; Bun, Bundi; Doi, Doido; Jay, Jayapura; Mai, Maimufu; Mer, Merauke; Mor, Port Moresby; Nam,

Namasado; Nor, Normanby Island; Nru, Noru; Tim, Timika; and Wau, Wau. d, Southern populations; j, northern populations; dark shading

shows land above 2000 melevation.

L.H. Rawlings, S.C. Donnellan / Molecular Phylogenetics and Evolution 27 (2003) 36–44

Page 3

2. Materials and methods

2.1. Specimens examined

Fifty-two M. viridis were collected from16 locations

fromacross the species range in New Guinea and

northern Queensland, Australia (Fig. 1). Tissue samples

used in this study were (*

—

samples used for allozyme

electrophoresis; ABTC, Australian Biological; Tissue

Collection; AMS, Australian MuseumSydney; BPBM,

Bernice P. Bishop Museum, Hawaii; MV, Museum

Victoria, Melbourne; QM, Queensland MuseumBris-

bane; SAMA, South Australian MuseumAdelaide):

Australia

—

Aus 1–18: Iron Range, Queensland ABTC

65592, ABTC 65605, ABTC 67593-4, ABTC 67596,

ABTC 67627-34, QM J66805, MV T888, Lockhart

River, Queensland ABTC 51497-8

, McIlwraith Range,

Queensland QM CJS919; Southern PNG

—

Simbu Prov-

ince Doi 1–3: AMS R115348-50

, Nru 1–2: AMS

R115355-6

, Southern Highlands Province Nam1–2:

AMS R122363-4

; Milne Bay Province Nor 1: AMS

R129716

; Central District Mor 1: ABTC 68320; East-

ern Highlands Province Mai 1: ABTC 67151; Northern

PNG

—

Madang Province Bun 1: AMS R124531

; Mo-

robe Province Wau 1–2: BPBM 11617

, BPBM 13798

;

Southern Irian Jaya

—

Aru Island Aru 1–4: ABTC 66380-

1, ABTC 68312-3; Merauke Mer 1–6: ABTC 66384-5,

ABTC 67170, ABTC 67172, ABTC 68315-6; Timika

Tim1: ABTC 66387; Northern Irian Jaya

—

Biak Island

Bia 1–5: ABTC 66388-9, ABTC 67175-6, ABTC 67182;

Jayapura Jay 1–2: ABTC 66383, ABTC 66386; Sorong

Sor 1–5: ABTC 67173-4, ABTC 67183, ABTC 68318-9.

Morelia spilota SAMA R26878, was used as the out-

group.

2.2. Mitochondrial DNA

DNA was extracted fromliver, body scales or shed

skin of snakes using a salting out method (Miller et al.,

1988). For skin and scales, the Proteinase-K digestion at

37 °C step was extended to overnight. PCR primers

L14841 and H15149 (Kocher et al., 1989) were used to

amplify a 350 bp cytochrome b (cytb) fragment. To avoid

the amplification of a control region-like gene that is

present in snakes between the ND1 and ND2 genes

(Kumazawa and Nishida, 1996), nested PCR was used

to amplify the control region (CR). An initial amplifi-

cation used the primers L15926 and H690 (Kumazawa

et al., 1996) situated in the tRNA

Thr

and 12S rRNA genes,

respectively, with an

LONG

ASE

PCR protocol. This

product was then used as a template for a hemi-nested

amplification with primers L15926 and snake 17 (Ku-

mazawa, pers. comm.

—

-TATGTCTAACAAGCATT

AAG-3

) situated in the Conserved Sequence Block I of

the CR to amplify a $850 bp product that was used for

sequence analysis. Sequencing reactions in both direc-

tions of the CR fragment tended to stall at a C-rich

region [described by Kumazawa et al. (1998)]. Addi-

tional nested primers (Kumazawa, pers. comm.) were

used to sequence across this region in each direction.

These primers were Snake 1 (5

-CCT ATG TAT AAT

AAT ACA TTA A-3

), Snake 6 (5

-ACC CTT CCC

GTG AAA TCC-3

), and Snake 7 (5

-TGA AAG GAT

AGA GGA TTT CAC G-3

). Both strands of the PCR

amplified gene fragment were sequenced using the

PRISM Ready Reaction DyeDeoxy Terminator Cycle

Sequencing Kit on a FTS-1 thermal sequencer (Corbett

Research). The reaction products were electrophoresed

on an Applied Biosystems 373A DNA sequencer.

Sequences were aligned by eye and only 12 insertion/

deletion (indel) events were inferred in the CR align-

ment. Sequences are deposited with GenBank, Acces-

sion Nos. are AY169826–99.

The potential for the CR primers to amplify mito-

chondrial genes rather than nuclear paralogues (Zhang

and Hewitt, 1996) was tested using the parallel titration

protocol of Donnellan et al. (1999) in which the control

nuclear locus was 18S rRNA amplified with the primers

18e (Hillis and Dixon, 1991) and G59 (S. Cooper pers.

comm. 5

-GCT GGC ACC AGA CTT GCC CTC C-3

Aligned sequences were phylogenetically analysed with

maximum parsimony (MP) and maximum likelihood

(ML) and distance methods via neighbour-joining (NJ)

in PAUP

4:0b2a (Swofford, 1999). To determine whe-

ther the two data sets should be combined, the ILD test

(Farris et al., 1995) (the Partition Homogeneity Test in

PAUP) was performed in PAUP

4:0b2a. The most

appropriate model of nucleotide substitution for ML

and distance analyses was found with using Modeltest3

(Posada and Crandell, 1998). Because of the large

number of haplotypes analysed, values for parameters

of the nucleotide substitution model were estimated

using quartet puzzling by successive approximations

(Strimmer and Vonhaeseler, 1996). The one of the most

parsimonious trees was used as a starting point to esti-

mate the parameters. These parameters were then used

in a puzzling analysis to generate a tree topology for the

next round of parameter estimation. When the values of

estimated parameters no longer changed, it was con-

cluded that the best estimate of the parameters had been

reached. The estimated parameter values were then used

in a full ML analysis under the specified model.

2.3. Allozyme electrophoresis

Frozen tissues suitable for allozyme electrophoresis

were available for only 12 M. viridis due to the con-

straint of collecting many of the samples from captive

specimens. Allozyme electrophoresis was done accord-

ing to the methods of Richardson et al. (1986). Enzymes

stained are listed in Table 1 and their Enzyme

Commission Numbers and locus abbreviations are in

L.H. Rawlings, S.C. Donnellan / Molecular Phylogenetics and Evolution 27 (2003) 36–44

Page 4

Murphy et al. (1996). Phylogenetic analysis was done

using MP criterion of optimality, with loci as characters

and alleles as unordered character states. Polymor-

phisms were treated as uncertainties following the rec-

ommendations of Kornet and Turner (1999). Trees were

also constructed with NJ based on Cavalli-Sforza and

Edwards (CSE) chord distances between OTUs (Cavalli-

Sforza and Edwards, 1967).

3. Results

3.1. Mitochondrial DNA

Partial cytb and CR sequences frompurified mito-

chondrial DNA were compared with sequences ampli-

fied fromtotal cellular DNA for M. viridis AMS

R115348. The two sequences for each gene were indis-

tinguishable.

Partial sequences of the cytb and CR genes were

obtained fromone M. spilota and 36 M. viridis samples.

In addition, four specimens, Aus15, Nru1, Mer1, and

Jay1, were typed only for cytb and 14 specimens, Aus 3-

14, Sor2/6, were typed only for the CR. Among the 18

individuals sequenced only for cytb or CR, there were no

additional haplotypes that were not already observed

among the 36 individuals that had been sequenced for

both genes. From40 individuals sequenced for cytb

there were 14 haplotypes and from50 individuals se-

quenced for CR there were 33 haplotypes indicating the

higher level of variation observed in the CR data. Un-

ique haplotypes were represented in the combined data

that were used for the final analyses (Fig. 2).

Of 292 nucleotide sites of aligned cytb sequence, 64

were variable, and 42 were parsimony informative, while

786 nucleotide sites of CR sequence (47 bp of tRNA

Pro

and 739 bp of control region) had 243 variable sites, with

87 parsimony informative. The partition homogeneity

test failed to detect significant incongruence between the

two data partitions (P ¼ 0:61Þ, indicating that the two

data sets could be combined for analysis. The combined

data sets of 1078 aligned sites (307 variable, 129 parsi-

mony informative) were analyzed by MP, NJ, and ML.

For MP analysis, gaps were treated as a fifth character

state. Of a total of 12 indels that were inferred in the

alignment of the CR, all but one was a single site indel.

Single site gaps can be treated as the equivalent of a fifth

character state (Simmons and Ochoterena, 2000) while

the only multi-site indel, two sites long, was autapo-

morphic. In the MP analysis, a heuristic search found

208 equally most parsimonious trees of length 342 steps.

The model of nucleotide substitution found for the

Table 1

Allele frequencies expressed as a percentage, in seven OTUs of Morelia at 36 allozyme loci

OTU

Wau

Nor

NamDoi

Nru

Aus

M. spilota

Aat-2

–

Acoh-2

c(75)

b(75)

–

a(25)

Ada

b(50)

a(50)

Eno

b(17)

b(75)

a(83)

a(25)

Gpi

c(25)

b(75)

bð50Þ

b(50)

b(17)

a(50)

a(83)

Idh-1

c(25)

b(50)

b(75)

a(50)

PepA

b(75)

–

a(25)

PepB1

b(50)

b(83)

bð50Þ

–

a(50)

a(17)

a(50)

PepB2

–

Pgm-1

b(25)

b(83)

b(25)

a(75)

a(17)

a(75)

Pgm-2

–

Note. Alleles are designated alphabetically, with ÔaÕ being the most cathodally migrating allele. Where enzymes are encoded by more than one

locus, the loci are designated numerically in order of increasing electrophoretic mobility. Where the allele frequencies are not given, the frequency is

100. The number of individuals sampled from each population ðNÞ is given at the head of each column, except where except when fewer individuals

were successfully typed, in which case N is indicated by the number in superscript beside the first allelic frequency entry for a locus. The following loci

were invariant: Acoh-1, Acp, Ak-1, Ak-2, Ca, Est, Fbp, Fumh, Gda, Iddh, Idh-2, Lap, Ldh-1, Ldh-2, Mdh-1, Mdh-2, Ndpk, PepD, Pgam, Pgdh, and

Pgk. For Aat-1, Lgl, and Mpi, OTUs Wau-Aus had the b, a or a alleles, respectively and M. spilota had the a, b or b alleles, respectively. OTU codes

are Wau, Nor, Normanby Island; Nam, Namosado; Doi, Doido; Nru, Noru; Aus, Queensland.

L.H. Rawlings, S.C. Donnellan / Molecular Phylogenetics and Evolution 27 (2003) 36–44

Page 5

combined data set was HKY85 + C (Hasegawa et al.,

1985). Parameters, used in the ML analyses, estimated

using successive approximations for this model for

the combined data set were: nucleotide frequencies

A ¼ 0.271227, C ¼ 0.287991, G ¼ 0.131905, and T ¼

0.308877, ts=tv ratio ¼ 2.3341, and C ¼ 0.2929. Pairwise

distances between haplotypes estimated with the

HKY85 + C model were used for NJ analysis.

The ML tree is shown in Fig. 2 with bootstrap

pseudoreplicate proportions for ML, MP, and NJ

analyses indicated. Phylogenetic reconstructions for all

three methods of analysis of mtDNA showed two

monophyletic lineages supported by bootstrap pseu-

doreplicate proportions greater than 98%. One lineage,

designated the southern lineage, comprised the southern

New Guinea and Australian localities and the other,

designated the northern lineage, comprised the northern

localities. All three analyses also had the Australian

specimens forming a monophyletic clade within the

southern lineage with bootstrap pseudoreplicate support

greater than 95%.

There was polyphyly of populations within the

northern and southern lineages. Some of the western

Sorong and Biak samples clustered with the eastern

Wau and Bundi samples in the northern clade, and

amongst the southern localities some samples from

Merauke clustered with those fromAru Is., whilst other

samples from Merauke grouped with individuals from

Namasado. Relationships among haplotypes within the

northern localities were well supported, but much of the

population structure amongst the southern localities was

not supported by bootstrapping in MP and ML analy-

ses.

For the combined sequence data, there were four

synapomorphies for the Australian clade

—

three transi-

tions and an indel. One of the transitions was found in

Fig. 2. Maximum likelihood phylogram of evolutionary relationships of Australian and New Guinea populations of M. viridis. Bootstrap pseu-

doreplicates proportions are indicated for MP, ML, and NJ analyses in descending order respectively.

L.H. Rawlings, S.C. Donnellan / Molecular Phylogenetics and Evolution 27 (2003) 36–44

Page 6

the cytb, and one was in tRNA

Pro

. There was a C () T

transition unique to the southern New Guinea individ-

uals. There was also an A () G transition that was

found in all the southern New Guinea animals except

for the individual fromNormanby Island (Nor1) which

had an A at this site. Nor1 is the most divergent indi-

vidual in the southern lineage, averaging more than 3%

sequence divergence fromall the other southern ani-

mals. Whilst this sequence has a character state that is

synapomorphic for the southern lineage, it also shares

five nucleotide substitutions that would otherwise be

synapomorphies of the northern lineage, one of which is

an indel. There were 57 synapomorphies for the north-

ern lineage, (31 transitions, 23 transversions, and 3 in-

dels) 16 of the substitutions were in cytb and one in the

tRNA

Pro

. There was an average of 7.6% uncorrected se-

quence divergence between the northern and southern

lineages (range 7.0–8.9%) and 1.5% divergence between

Australia and the southern New Guinean haplotypes. In

comparisons among haplotypes within each region,

maximum uncorrected sequence divergence was 0.2%

for Australia, 1.0% for southern New Guinea, and 1.4%

for northern New Guinea.

3.2. Allozymes

Allele distributions at the 36 loci resolved are shown

in Table 1. These data were converted into a matrix of

CSE chord distances between OTUs and used to con-

struct a tree by NJ. Twelve equally most-parsimonious

trees were found using the branch and bound search

method under the MP criterion of optimality. A strict

consensus of these trees and the NJ tree both showed a

dichotomy of OTUÕs representing the northern and

southern mitochondrial lineages. Strong MP bootstrap

support (84%) was evident for monophyly of the

southern OTUÕs. Five loci, Aat-2, Ada, PepB1, PepB2,

and Pgm-2, contribute most to the differentiation of the

northern and southern lineages (Table 1). There was

little differentiation within the southern lineage, with the

Australian sample sharing alleles with the southern New

Guinean localities at the 31 loci where the Australian

OTU could be typed.

4. Discussion

4.1. Phylogeographical patterns and taxonomic implica-

tions

The discovery of two very distinct genetic lineages of

green pythons fromNew Guinea and northeastern

Australia begs the question as to whether the lineages

represent two species of green pythons. The evolution-

ary species concept (Simpson, 1951; Wiley, 1978) has

become a popular choice of species concept among in-

vestigators that use gene genealogies as part of the evi-

dence to examine whether a group of organisms is ‘‘a

phyletic lineage, i.e., an ancestral-descendent sequence

of interbreeding populations, evolving independently of

others, with its own separate and unitary evolutionary

role and tendencies.’’ Indeed de QueirozÕs (1998) further

development of this concept to the ‘‘generalised lineage

concept’’ appears to offer the best combination of

theoretical underpinning and operational criteria for

delineating species at present. Herein, we utilise infor-

mation from gene genealogies (monophyly of haplotypes

representing the group) and the degree of sequence dif-

ferentiation relative to other well accepted pairs of snake

species to indicate the presence of evolutionary species. It

is minimally desirable that a measure of differentiation of

characters other than the mtDNA sequences be avail-

able, as even deeply bifurcated single gene genealogy

may not represent population divergence but rather re-

tention of ancestral gene lineages within a single popu-

lation (e.g., Thomaz et al., 1996). We are unable to infer

reliably the geographic distribution of the taxa that we

have found as the green python has an apparently con-

tinuous distribution through New Guinea and our

sampling in likely areas of contact is not adequate.

The pattern of relationships found for mitochondrial

and nuclear genes suggests the presence of two species of

M. viridis, one present north of the central cordillera and

the other present in southern New Guinea and Australia.

Phylogenetic analyses of the mitochondrial data shows

that into two reciprocally monophyletic clades each with

very strong bootstrap support. The minimum uncor-

rected sequence divergence between the northern and

southern lineages (7.0%) is within the range of values for

minimum mitochondrial sequence divergence between

other closely related snake, 1.6–5.3% (Ashton and de

Queiroz, 2001; Burbrink et al., 2001; Keogh et al., 2001;

Zamudio and Greene, 1997) and reptile species, 2.5–18%,

mean ¼ 12% (Johns and Avise, 1998). The allozyme allelic

data suggest that the divergence apparent between the

two mitochondrial lineages is indicative of genome wide

divergence. However the small sample size of the allo-

zyme data from the northern lineage limits the strength of

this indication. Morphological analyses do not provide

any more substantive evidence of population differentia-

tion. McDowell (1975) carried out a thorough assessment

of variation in body meristics and pattern, maxillary

tooth counts, and hemipenile morphology of green py-

thons fromthroughout their geographic range. Only

body scale row counts were subject to geographic varia-

tion, but the pattern of variation was not partitioned

across the central cordillera, indeed the full range of

variation was present along the north of the island.

Moreover McDowell (1975) did not identify any broad

concordant geographic patterns across a number of

characters that would suggest morphological differentia-

tion between the northern and southern lineages. Simi-

L.H. Rawlings, S.C. Donnellan / Molecular Phylogenetics and Evolution 27 (2003) 36–44

Page 7

larly variation in colour patterns described fromliving

specimens due to the presence/absence of a vertebral

stripe and/or scattered light coloured spots shows varia-

tion within and between localities without any clear

geographical pattern (McDowell, 1975; OÕShea, 1996). A

determination of the species status of the northern and

southern lineages awaits a more thorough assessment of

divergence at nuclear genes based on wider geographic

sampling than we could achieve herein with allozymes.

Because our sampling of the eastern and western tips

of New Guinea is sparse, we are not able to define the

geographic limits of the northern and southern lineages.

There are no obvious contemporary barriers that would

maintain separate distributions for the northern and

southern lineages. The lowland to mid-montane rain-

forest habitat of green pythons apparently forms a

continuous ‘‘ring’’ around the central mountain range

or cordillera of New Guinea (Johns, 1982; Pratt, 1982).

The lowland rainforests at the western end of the cen-

tral cordillera are extensive and apparently contiguous

with similar tracts to the north and south of the central

cordillera and indeed would be contiguous with similar

habitats on the Vogelkop Peninsula. In the east, the

lowland to mid-montane habitats of the green python

are substantially constricted by the narrowing of the

island and the presence of the high mountainous spine,

but there is no evidence for exclusion of the species

fromthis area as there are several records based on

museum vouchers from the Milne Bay Province, the

most easterly part of mainland New Guinea (McDo-

well, 1975). However, the east/west limits of the dis-

tribution of the two lineages may not necessarily be at

the extreme ends of the central cordillera or the island.

In the east, the Huon Peninsula region could be a

candidate barrier to gene flow. Colgan et al. (1993)

reported a possible zoogeographic barrier for lowland

mammal species in the Huon Peninsula region and

many lowland bird species which occur throughout the

rest of New Guinea are not found in the lowlands of

the Huon Gulf and MarkhamValley even though

rainforest habitat is present (Pratt, 1982). In the

absence of physical barriers, competitive ecological ex-

clusion may maintain separate contemporary distribu-

tions of evolutionary species that had their origins in

the distant past.

Given the relatively high sequence divergence between

the northern and southern lineages it is unlikely that the

latest episodes of climate cycling of the Plio-Pleistocene

would have been responsible for initiating the diver-

gence. In any biogeographic analysis of the New Guin-

ean biota throughout the Tertiary and Quaternary, the

influence of two factors cannot be ignored: the sub-

stantial changes in the geomorphology of the area due to

complex tectonism over the entire period (Dow, 1977;

Pigram and Davies, 1987), and the dramatic climate os-

cillations of the late Tertiary and the Quaternary

(Axelrod and Raven, 1982; Haig and Medd, 1996; Read

and Hope, 1996). During this period, tectonic move-

ments due to continental and island arc collisions have

seen the formation and infilling of several major sedi-

mentary basins to produce new lowland habitats and the

rapid uplift of the central cordillera, the latter beginning

approximately 5.8 MYA (Dow, 1977; Hill and Gleadow,

1989; Hill et al., 1993; Haig and Medd, 1996). The uplift

of the cordillera has been episodic rather than a contin-

uously gradual event, with an initial major burst of uplift

at 5.8–5.3 MYA and less intense episodes of folding and

thrusting spanning the period 5.3–4.7 MYA. The degree

of sequence divergence is compatible with the uplift of

the central mountain range through the Pliocene being

the causative factor, but any further inference will re-

quire more accurate determinations of genetic divergence

and also lineage specific calibration of a local molecular

clock. Consistent with our observations, a species of bird

(pitohui) fromlowland New Guinean rainforests also

shows a north–south divergence that is estimated from

molecular clock methods at 3.5 MYA (Dumbacher and

Fleischer, 2001).

The low magnitude of sequence divergence between

haplotypes fromsouthern New Guinea and Australia

and indeed among the southern lineage haplotypes

overall is consistent with their diversification during the

late Tertiary or Quaternary. Through this period during

times of lowered sea-level, a land bridge was exposed

across the Torres Strait between Australia and New

Guinea. Because of the shallow depth of the Torres

Strait, a land bridge would have been periodically in

existence over at the least the last 500,000 years (Gal-

loway and L€ooffler, 1972) with the most recent marine

incursion severing the connection starting approxi-

mately 8000 years ago. However, opportunities for

rainforest dependent species to disperse south onto Cape

York may have existed only sporadically during the late

Tertiary and Quaternary, as the high rainfall/high hu-

midity conditions of the Miocene gave way to an in-

creasingly drier atmospheric regime (Nix and Kalma,

1972) resulting in the present-day distribution of rem-

nant rainforest patches and intervening sclerophyll for-

ests (Truswell, 1993). However fluctuating climatic

conditions may have periodically modulated the extent

of suitable rainforest on the land bridge and intervening

Cape York region allowing dispersal (Nix and Kalma,

1972; Read and Hope, 1996). Moreover, extant Aus-

tralian rainforests are similar structurally to the lower

montane forests of New Guinea not to the complex

lowland humid rainforests for which there is presently

no northern Australian equivalent (Nix and Kalma,

1972). Whilst some rainforest species with New Guinean

origins dispersed into northern Australia and continued

southwards to the Atherton Tableland region (Winter,

L.H. Rawlings, S.C. Donnellan / Molecular Phylogenetics and Evolution 27 (2003) 36–44

Page 8

1997), other rainforest dependent species, including

M. viridis, extended their distributions no further south

than the McIlwraith Range ($13°30

S) strengthening

the argument that their present distribution was

achieved during the latter part of the Tertiary or Qua-

ternary when rainforest would not have been continu-

ously distributed as far south as the Atherton Tableland

rainforest block ($16°S).

4.2. Management issues

It is clear fromthe present study that the distribution

of the green python should be considered to encompass

at least three management units. Whilst it has not been

directly tested, it is possible that the anecdotal reports of

frequent lack of breeding success in captive green py-

thons could be due to pairing of individuals fromthe

northern and southern lineages. In Australia, there are

already many specimens in captivity that are descen-

dants of New Guinean specimens and as there has also

been interbreeding with Australian individuals, the

captive population represents a mixed gene pool.

Therefore, in order for Australian zoos to conformto

the ZooTag recommendations to breed only Australian

stock, the management of the captive breeding program

will require genotyping of the present captive stock with

biparentally inherited markers to detect individuals de-

rived fromcrosses between Australian and New Guin-

ean individuals. Mitochondrial markers would provide

only a one-way test of the genetic origin of such indi-

viduals.

As poaching green pythons fromthe wild and illegal

importation of specimens from New Guinea are major

enforcement issues in Australia, the determination of

distinct mitochondrial lineages from Australia, southern

and northern New Guinea provides a basis for the en-

forcement of protective legislation. However, as sample

sizes available to us fromNew Guinea were small, the

haplotypic diversity present in each region needs further

investigation before estimates of haplotype frequencies

can be regarded as statistically robust for forensic

applications.

Acknowledgments

We thank K. Aplin, S. Irwin, W. Mannion, J. Men-

zies, D. McCrae, S. Richards, D. Storch, T. Reardon,

and F. Yuwono for supplying tissues, J. Armstrong for

laboratory assistance and T. Bertozzi for assistance with

the figures, S.N. Prijono and I. Sidik of the Bogor

Museumand H. S. Basuki for assistance in Indonesia.

Field work by LHR was supported by a D.R. Stranks

Travelling Fellowship.

References

Ashton, K.G., de Queiroz, A., 2001. Molecular systematics of the

western rattlesnake, Crotalus viridis (Viperidae), with comments on

the utility of the D-Loop in phylogenetic studies of snakes. Mol.

Phylogenet. Evol. 21, 176–189.

Axelrod, D.I., Raven, P.H., 1982. Paleobiogeography and origin of the

New Guinea flora. In: Gressitt, J.L. (Ed.), Biogeography and

Ecology of New Guinea. W. Junk, The Hague, pp. 919–941.

Banks, C., 1999. TAG Action Plan 1999. Directing the Management of

Reptiles and Amphibians in Australasian Zoos. ARAZPA, Mos-

man, Sydney.

Barker, D.G., Barker, T.M., 1994. Pythons of the World: Australia.

Advanced VivariumSystems Inc, Lakeside, CA.

Burbrink, F., Lawson, R., Slowinski, J.B., 2001. Mitochondrial DNA

phylogeography of the polytypic North American rat snake

(Elaphe obsoleta): a critique of the subspecies concept. Evolution

54, 2107–2118.

Cavalli-Sforza, L.L., Edwards, A.W.F., 1967. Phylogenetic analysis:

models and estimation procedures. Evolution 32, 550–570.

Colgan, D., Flannery, T.F., Trimble, J., Aplin, K., 1993. Elec-

trophoretic and morphological analysis of the systematics of

the Phalanger orientalis (Marsupialia) species complex in Papua

New Guinea and the Solomon Islands. Aust. J. Zool. 41, 355–378.

Creer, S., Malhotra, A., Thorpe, R.S., Chou, W.-H., 2001. Multiple

causation of phylogeographical pattern as revealed by nested clade

analysis of the bamboo viper (Trimeresurus stejnegeri) within

Taiwan. Mol. Ecol. 10, 1967–1981.

de Queiroz, K., 1998. The general lineage concept of species, species

criteria and the process of speciation. In: Howard, D.J., Berlocher,

S.H. (Eds.), Endless Forms, Species and Speciation. Oxford

University Press, Oxford, pp. 57–75.

Donnellan, S.C., Adams, M., Hutchinson, M., Baverstock, P.R., 1993.

The identification of cryptic species in the Australian herpetofa-

una

—

a high research priority. In: Lunney, D., Ayers, D. (Eds.),

Herpetology in Australia: A Diverse Discipline Transactions of the

Royal Zoological Society of N.S.W. Special Edition. Surrey Beatty

and Sons, Chipping Norton, pp. 121–126.

Donnellan, S.C., Hutchinson, M.N., Saint, K.M., 1999. Molecular

evidence for the phylogeny of Australian gekkonoid lizards. Biol. J.

Linn. Soc. 67, 97–118.

Dow, D.B., 1977. Geological synthesis of Papua New Guinea. Bureau

Mineral Resources, Geol. Geophys. Bull. 201.

Dumbacher, J.P., Fleischer, R.C., 2001. Phylogenetic evidence for

colour pattern convergence in toxic pitohuis: M€uullerian mimicry in

birds? Proc. R. Soc. London B 268, 1971–1976.

Farris, J.S., Kallersjo, M., Kluge, A.G., Bult, C., 1995. Testing

significance of incongruence. Cladistics 10, 315–319.

Galloway, R.W., L€ooffler, E., 1972. Aspects of geomorphology and soils

in the Torres Strait region. In: Walker, D. (Ed.), Bridge and

Barrier: the Natural and Cultural History of Torres Strait.

Australian National University, Canberra, pp. 11–28.

Haig, D.W., Medd, D., 1996. Latest miocene to early pliocene

bathymetric cycles related to tectonism, Puri anticline, Papuan

basin, Papua New Guinea. Aust. J. Earth Sci. 43, 451–465.

Hasegawa, M., Kishino, H., Yano, T., 1985. Dating of the human-ape

splitting by a molecular clock of mitochondrial DNA. J. Mol. Evol.

22, 160–174.

Henderson, R.W., Hedges, S.B., 1995. Origin of west Indian popula-

tions of the geographically widespread Boa Corallus enydris

inferred frommitochondrial DNA sequences. Mol. Phylogenet.

Evol. 4, 88–92.

Hill, K.C., Gleadow, A.J.W., 1989. Uplift and thermal history of the

Papuan Fold Belt, Papua New Guinea: Apatite fission track

analysis. Aust. J. Earth Sci. 36, 515–539.

L.H. Rawlings, S.C. Donnellan / Molecular Phylogenetics and Evolution 27 (2003) 36–44

Page 9

Hill, K.C., Grey, A., Foster, D., Barrett, R., 1993. An alternative

model for the Oligo-Miocene evolution of northern PNG and the

Sepik–Ramu basins. In: Carman, G.J., Carman, Z. (Eds.), Petro-

leumExploration and Development in Papua New Guinea:

Proceedings of the Second PNG PetroleumConvention, Port

Moresby, pp. 241–259.

Hillis, D.M., Dixon, M.T., 1991. Ribosomal DNA: molecular evolu-

tion and phylogenetic inference. Quart. Rev. Biol. 66, 411–453.

Johns, R.J., 1982. Plant zonation. In: Gressitt, J.L (Ed.), Monogra-

phiae Biologiae, vol. 42. W. Junk, The Hague, pp. 309–330.

Johns, G.C., Avise, J.C., 1998. A comparative summary of genetic

distances in vertebrates from the mitochondrial cytochrome b gene.

Mol. Biol. Evol. 15, 1481–1490.

Keogh, J.S., Barker, D., Shine, R., 2001. Heavily exploited but poorly

known: systematics and biogeography of commercially harvested

pythons (Python curtus group) in southeast Asia. Biol. J. Linn. Soc.

73, 113–129.

Kocher, T.D., Thomas, W.K., Meyer, A., Edwards, S.V., P€a€aabo, S.,

Villablanca, F.X., Wilson, A.C., 1989. Dynamics of mitochondrial

DNA evolution in animals: amplification and sequencing with

conserved primers. Proc. Natl. Acad. Sci. USA 86, 6196–6200.

Kornet, D J, Turner, H., 1999. Coding polymorphism for phylogeny

reconstruction. Syst. Biol. 48, 365–379.

Kumazawa, Y., Nishida, M., 1996. Variations in mitochondrial tRNA

gene organization of reptiles as phylogenetic markers. Mol. Biol.

Evol. 12, 759–772.

Kumazawa, Y., Ota, H., Nishida, M., Ozawa, T., 1996. Gene

rearrangements in snake mitochondrial genomes: highly concerted

evolution of control region-like sequences duplicated and inserted

into a tRNA gene cluster. Mol. Biol. Evol. 13, 1242–1254.

Kumazawa, Y., Ota, H., Nishida, M., Ozawa, T., 1998. The complete

nucleotide sequence of a snake (Dinodon simicarinatus) mitochon-

drial genome with two identical control regions. Genetics 150, 313–

329.

McDowell, D., 1997. Wildlife Crime Policy and the Law An Australian

Study. Australian Government Publishing Service, Canberra,

Australia.

McDowell, S.B., 1975. A catalogue of the snakes of New Guinea and

the Solomons, with special reference to those in the Bernice P.

Bishop Museum. Part II. Aniliodea and Pythoninae. J. Herp. 9, 1–

79.

Miller, S.A., Dykes, D.D., Polesky, H.F., 1988. Simple salting out

procedure for extracting DNA fromhuman nucleated cells. Nucleic

Acids Res. 16, 1215.

Murphy, R.W., Sites, J.W., Buth, D., Haufler, C.H., 1996. Proteins I:

Isozyme electrophoresis. In: Hillis, D.M., Moritz, C., Mable, B.

(Eds.), Molecular Systematics. Sinauer, Sunderland, MA, pp. 51–

120.

Nix, H.A., Kalma, J.D., 1972. Climate as a dominant control in the

biogeography of northern Australia and New Guinea. In: Walker,

D. (Ed.), Bridge and Barrier: the Natural and Cultural History of

Torres Strait. Australian National University, Canberra, pp. 61–

92.

OÕShea, M., 1996. A Guide to the Snakes of Papua New Guinea.

Independent Publishing, Independent Group Pty Ltd., Port

Moresby, PNG.

Parker, F., 1982. The Snakes of the Western Province. PNG Dept. of

Lands and Environment 82/1.

Pigram, C.J., Davies, H.L., 1987. Terranes and the accretion history of

the New Guinea orogen. BMR J. Aust. Geol. Geophys. 10, 193–

211.

Posada, D., Crandell, K.A., 1998. Modeltest: testing the model of

DNA substitution. Bioinformatics 14, 817–818.

Pratt, T.K., 1982. Biogeography of birds in New Guinea. In: Gressit,

J.L. (Ed.), Monographieae Biologicae, vol. 42. W. Junk, The

Hague, pp. 815–836.

Read, J., Hope, G., 1996. Ecology of Nothofagus forests of New

Guinea and New Caledonia. In: Veblen, T.T., Hill, R.S., Read, J.

(Eds.), The Ecology and Biogeography of Nothofagus Forests.

Yale University, pp. 200–256.

Richardson, B.J., Baverstock, P.R., Adams, M., 1986. Allozyme

Electrophoresis: A Handbook for Animal Systematics and Popu-

lation Structure. Academic Press, Sydney.

Rodriguez-Robles, J.A., De Jesus-Escobar, J.M., 2000. Molecular

systematics of New World gopher, bull, and pinesnakes (Pituophis:

Colubridae), a transcontinental species complex. Mol. Phylogenet.

Evol. 14, 35–50.

Simmons, M.P., Ochoterena, H., 2000. Gaps as characters in sequence-

based phylogenetic analyses. Syst. Biol. 49, 369–381.

Simpson, G.G., 1951. The species concept. Evolution 5, 285–298.

Strimmer, K., Vonhaeseler, A., 1996. Quartet puzzling

—

a quartet

maximum-likelihood method for reconstructing tree topologies.

Mol. Biol. Evol. 13, 964–969.

Swofford, D., 1999. PAUP*. Phylogenetic Analysis Using Parsimony.

(* and Other Methods). Sinauer, Sunderland, MA.

Thomaz, D., Guiller, A., Clarke, B., 1996. Extreme divergence of

mitochondrial DNA within species of pulmonate land snails. Proc.

R. Soc. London Ser. B 263, 363–368.

Truswell, E.M., 1993. Vegetation changes in the Australian Tertiary in

response to climatic and phytogeographic forcing factors. Aust.

Syst. Bot. 6, 533–557.

Wiley, E.O., 1978. The evolutionary species concept reconsidered.

Syst. Zool. 27, 17–26.

Winter, J.W., 1997. Responses of non-volant mammals to late

quaternary climatic changes in the wet tropics region of north-

eastern Australia. Wild Res. 24, 493–511.

Zamudio, K.R., Greene, H.W., 1997. Phylogeography of the bush-

master (Lachesis muta, Viperidae)

—

Implications for neotropical

biogeography, systematics, and conservation. Biol. J. Linn. Soc. 62,

421–442.

Zhang, D.X., Hewitt, G.M., 1996. Nuclear integrations: challenges for

mitochondrial DNA markers. TREE 11, 247–251.

L.H. Rawlings, S.C. Donnellan / Molecular Phylogenetics and Evolution 27 (2003) 36–44