ZOOTAXAISSN 1175-5326 (print edition)

ISSN 1175-5334 (online edition)Copyright © 2013 Magnolia Press

Zootaxa 3626 (1): 077–093 www.mapress.com/zootaxa/ Article

http://dx.doi.org/10.11646/zootaxa.3626.1.3http://zoobank.org/urn:lsid:zoobank.org:pub:777B69EB-9DC9-4F25-AA65-66FB1947D374

Molecular phylogenetic reconstruction of the endemic Asian salamander family Hynobiidae (Amphibia, Caudata)

DAVID W. WEISROCK1,6, J. ROBERT MACEY2,3, MASAFUMI MATSUI4, DANIEL G. MULCAHY5, & THEODORE J. PAPENFUSS3

1Department of Biology, University of Kentucky, 101 Thomas Hunt Morgan Building, Lexington, KY, 40506-02252Genomics, Department of Biology, Merritt College, 1250 Campus Drive, Oakland, CA 946193Museum of Vertebrate Zoology, 3101 Valley Life Science Building, University of California, Berkeley, CA 947204Graduate School of Human and Environmental Studies, Kyoto University, Kyoto 606-8501, Japan5Smithsonian Institution, National Museum of Natural History, 4210 Silver Hill Rd., MRC 534, Suitland, MD 20746 6Corresponding author. E-mail: dweis2@uky.edu; Phone: 859-257-2249; FAX: 859-257-1717

Abstract

The salamander family Hynobiidae contains over 50 species and has been the subject of a number of molecular phylo-genetic investigations aimed at reconstructing branches across the entire family. In general, studies using the greatest amount of sequence data have used reduced taxon sampling, while the study with the greatest taxon sampling has used a limited sequence data set. Here, we provide insights into the phylogenetic history of the Hynobiidae using both dense taxon sampling and a large mitochondrial DNA sequence data set. We report exclusive new mitochondrial DNA data of 2566 aligned bases (with 151 excluded sites, of included sites 1157 are variable with 957 parsimony informative). This is sampled from two genic regions encoding a 12S–16S region (the 3’ end of 12S rRNA, tRNAVAl, and the 5’ end of 16S rRNA), and a ND2–COI region (ND2, tRNATrp, tRNAAla, tRNAAsn, the origin for light strand replication—OL, tRNACys, tRNATyr, and the 5’ end of COI). Analyses using parsimony, Bayesian, and maximum likelihood optimality criteria produce similar phylogenetic trees, with discordant branches generally receiving low levels of branch support. Monophyly of the Hynobiidae is strongly supported across all analyses, as is the sister relationship and deep divergence between the genus Onychodactylus with all remaining hynobiids. Within this latter grouping our phylogenetic results identify six clades that are relatively divergent from one another, but for which there is minimal support for their phy-logenetic placement. This includes the genus Batrachuperus, the genus Hynobius, the genus Pachyhynobius, the genus Salamandrella, a clade containing the genera Ranodon and Paradactylodon, and a clade containing the genera Liua and Pseudohynobius. This latter clade receives low bootstrap support in the parsimony analysis, but is consistent across all three analytical methods. Our results also clarify a number of well-supported relationships within the larger Batrachu-perus and Hynobius clades. While the relationships identified in this study do much to clarify the phylogenetic history of the Hynobiidae, the poor resolution among major hynobiid clades, and the contrast of mtDNA-derived relationships with recent phylogenetic results from a small number of nuclear genes, highlights the need for continued phylogenetic study with larger numbers of nuclear loci.

Key words:

Introduction

The salamander family Hynobiidae is comprised of a diverse assemblage of over 50 species distributed entirely within continental Asia and adjacent islands. Subsets of this family have been the focus of numerous allozyme and mitochondrial DNA-based evolutionary investigations (Fu et al., 2001; Matsui et al., 1992, 2000, 2001, 2004, 2006; Matsui, Nishikawa, et al., 2007; Matsui, Tominaga, et al., 2007; Nishikawa et al., 2005; Nishikawa et al., 2001, 2007; Tominaga et al., 2003, 2005, 2006; Zeng et al., 2006). Three different studies have investigated phylogenetic relationships on a broader scale within the Hynobiidae. Zhang et al. (2006) analyzed complete mitochondrial DNA genome data (>14,000 aligned nucleotides) from 16 hynobiid taxa representing major groups within the family and a cryptobranchid outgroup (Andrias davidianus).

Accepted by M. Vences: 9 Jan. 2013; published: 12 Mar. 2013 77

The results of Zhang et al. (2006) yielded a resolved branching history and began to clarify some of the major relationships within the family. The basal split in the family between the lung-less salamanders of the genus Onychodactylus and all remaining hynobiids received strong branch support across all optimality criteria [i.e., Bayesian posterior probabilities (PP) of 1.0 and bootstrap percentages (BP) of 100]. A similar pattern of strong branch support was produced for the placement of the genera Ranodon and Paradactylodon in a clade, as well as for the individual monophyly of the genera Batrachuperus, Hynobius, Liua, and Paradactylodon. More recently, Zheng et al. (2011) used mtDNA genome data and nuclear sequence data from three loci to investigate divergence times estimated among salamander groups. This study included 25 hynobiid species with representatives of all genera and 10 species sampled from the diverse genus Hynobius. Concatenated analyses of the nuclear data produced a tree topology similar to that of the mtDNA tree. The most notable exceptions were the placement of the Pachyhynobius and Salamandrella relationships with respect to other taxa, which had alternate placements in the mtDNA and nuclear trees. Finally, as part of a large scale phylogenetic study of all amphibians, Pyron & Wiens (2011) estimated phylogenetic relationships for 49 hynobiid species using DNA sequence data (primarily mtDNA) gleaned from GenBank. The results from this study also produced a highly resolved tree, but many relationships, particularly those uniting different genera, received low maximum likelihood bootstrap branch support, despite receiving strong branch support in other studies.

These overall results represent a major advancement in our understanding of hynobiid phylogenetics; however, a number of additional aspects of hynobiid relationships remain to be clarified. The complete mtDNA genome results of Zhang et al. (2006) and Zheng et al. (2011) each provide a fully resolved tree with many branches receiving high levels of branch support; yet, there was discordance for the placement of a number of clades, particularly with the placement of Salamandrella and Pachyhynobius. Concatenated nuclear results provide yet another perspective on relationships among major clades, further complicating interpretations of hynobiid relationships. Taxon sampling has been demonstrated as an important factor in phylogenetic accuracy (DeBry, 2005; Heath et al., 2008; Hillis et al., 2003; Pollock et al., 2002; Zwickl & Hillis, 2002). Furthermore, phylogenomic studies of complete organellar DNA sequence may be prone to providing highly precise, but inaccurate reconstructions when limited taxon sampling is used (Braun & Kimball, 2002; Samuels et al., 2005; Soltis et al., 2004; Stefanovic et al., 2004). A phylogenetic analysis of the Hynobiidae using dense species sampling with a large mtDNA sequence data set may help to clarify the resolution of deeper relationships.

In this study we attempt to further address the phylogenetic history of the Hynobiidae with mtDNA sequence data from a densely sampled set of 40 hynobiid species and all three species in the sister family Cryptobranchidae. The major contribution of this work is that it includes much greater taxonomic sampling than that used in the mitogenomic and nuclear studies of Zhang et al. (2006) and Zheng et al. (2011). Furthermore, while it includes a similar taxon sampling strategy to that used in Pyron & Wiens (2011), it uses additional mtDNA data that was not included in that study. We present a mtDNA data set of nearly 2500 aligned nucleotide

positions comprised of two different contiguous regions: (1) 12S–tRNAVal–16S and (2) ND2–COI. We apply parsimony, Bayesian, and maximum likelihood (ML) optimality criteria to reconstruct phylogenetic relationships and we calculate parsimony bootstrap values (PB), decay indices (DI), Bayesian posterior probabilities (PP), and maximum likelihood bootstrap values (MLB) as heuristic measures of branch support.

Material and methods

Taxon sampling and data collection. In this study we collected DNA sequence data from 48 individuals representing 37 of the over 50 currently described hynobiid species [note: hynobiid taxonomy is in flux so we defer to further studies to stabilize the number of species in the group; i.e., Peng et al., (2010)] and all three recognized cryptobranchid species (Table 1). For four hynobiid species and the outgroup taxon Andrias japonicus we collected data from individuals representing multiple localities (Table 1). We included all three species of the family Cryptobranchidae in our study as an outgroup, because numerous lines of evidence support its sister relationship with hynobiids (Roelants et al., 2007; Weisrock et al., 2005; Wiens et al., 2005; Zheng et al., 2011).

WEISROCK ET AL.78 · Zootaxa 3626 (1) © 2013 Magnolia Press

TA

BLE

1. T

axon

sam

plin

g fo

r all

outg

roup

and

ingr

oup

sam

ples

use

d in

this

study

. M

useu

m a

bbre

viat

ions

are

as f

ollo

ws:

CA

S, C

alifo

rnia

Aca

dem

y of

Sc

ienc

es, S

an F

ranc

isco

, Cal

iforn

ia; I

NH

S, Il

linoi

s Nat

ural

Hist

ory

Surv

ey, U

nive

rsity

of I

llino

is at

Urb

ana-

Cha

mpa

ign;

KU

HE,

Gra

duat

e Sc

hool

of

Hum

an a

nd E

nviro

nmen

tal S

tudi

es, K

yoto

Uni

vers

ity; M

VZ,

Mus

eum

of V

erte

brat

e Zo

olog

y, U

nive

rsity

of C

alifo

rnia

at B

erke

ley.

Taxo

n Sp

ecim

en

Acc

essi

on

Gen

Ban

k A

cces

sion

Lo

calit

y D

escr

iptio

n

Andr

ias d

avid

ianu

s M

VZ

2042

45

12S-

16S:

AY

9159

66

ND

2-C

OI:

AY

9159

18Pu

rcha

sed

in H

ong

Kon

g m

arke

t in

1973

Andr

iasj

apon

icus

J-

35, n

o vo

uche

r 12

S-16

S: A

Y91

5967

N

D2-

CO

I: A

Y91

5919

Nab

ari-c

ity, M

ie P

refe

ctur

e, Ja

pan

Cry

ptob

ranc

hus a

llega

nien

sis

INH

S 11

236

12S-

16S:

AY

9159

68

ND

2-C

OI:

AY

9159

20C

urre

nt R

iver

, Sha

nnon

Cou

nty,

Mis

sour

i, U

SA

Batr

achu

peru

s lon

gdon

gens

is

MV

Z 20

8611

12

S-16

S: A

Y91

6005

N

D2-

CO

I: A

Y91

5957

Long

Don

g (D

rago

n C

ave)

Spr

ing,

E si

de o

f Em

ei

Shan

, Les

han

Pref

ectu

re, S

ichu

an P

rovi

nce,

Chi

na

Batr

achu

peru

s pin

chon

ii (1

) NC

AS

1952

37

12S-

16S:

AY

9160

06

ND

2-C

OI:

AY

9159

58W

a Sh

an, Y

a`an

Pre

fect

ure,

Sic

huan

Pro

vinc

e, C

hina

Batr

achu

peru

s pin

chon

ii (2

) SC

AS

1944

50

12S-

16S:

AY

9160

07

ND

2-C

OI:

AY

9159

5929

.1 k

m N

of J

osha

ng, L

iba

Shan

(Mt.)

, Ya’

an

Pref

ectu

re, S

ichu

an P

rovi

nce,

Chi

na

Batr

achu

peru

s tib

etan

us

MV

Z 21

6664

12

S-16

S: A

Y91

6004

N

D2-

CO

I: A

Y91

5956

Zhig

ou M

ount

ain,

40.

4 km

N o

f Miy

aluo

, Zan

gzu

Aut

onom

ous P

refe

ctur

e, S

ichu

an P

rovi

nce,

Chi

na

Batr

achu

peru

s yen

yuan

ensi

s (1)

M

CA

S 19

4663

12

S-16

S: A

Y91

6008

N

D2-

CO

I: A

Y91

5960

Qili

ba, L

iang

shan

(Mt.)

, Yiz

u A

uton

omou

s Pre

fect

ure,

Si

chua

n Pr

ovin

ce, C

hina

Ba

trac

hupe

rus y

enyu

anen

sis (

2)

NC

AS

1950

75

12S-

16S:

AY

9160

09

ND

2-C

OI:

AY

9159

619.

5 km

N o

f Tuo

wu,

Lia

ngsh

an (M

t.), Y

izu

Aut

onom

ous P

refe

ctur

e, S

ichu

an P

rovi

nce,

Chi

na

Batr

achu

peru

s yen

yuan

ensi

s (3)

S

CA

S 19

4483

12

S-16

S: A

Y91

6010

N

D2-

CO

I: A

Y91

5962

20.5

km

Wes

t (ai

rline

) of X

icha

ng, L

iang

shan

Yiz

u A

uton

omou

s Pre

fect

ure,

Sic

huan

Pro

vinc

e, C

hina

H

ynob

ius a

bei

KU

HE

1351

4 12

S-16

S: A

Y91

5987

N

D2-

CO

I: A

Y91

5939

Am

ino-

cho,

Kyo

to P

refe

ctur

e, Ja

pan

Hyn

obiu

s am

jiens

isC

AS

1943

76

12S-

16S:

AY

9159

80

ND

2-C

OI:

AY

9159

326.

8 km

sout

h of

Zha

ngcu

n, Ji

axin

g Pr

efec

ture

, Zh

ejia

ng P

rovi

nce,

Chi

na

Hyn

obiu

s bou

leng

eri

KU

HE

2565

5 12

S-16

S: A

Y91

5984

N

D2-

CO

I: A

Y91

5936

Kam

ikita

yam

a-m

ura,

Nar

a Pr

efec

ture

, Jap

an

cont

inue

d ne

xt p

age

Zootaxa 3626 (1) © 2013 Magnolia Press · 79HYNOBIIDAE PHYLOGENETICS

TA

BL

E 1

. (co

ntin

ued)

Taxo

n Sp

ecim

en

Acc

essi

on

Gen

Ban

k A

cces

sion

Lo

calit

y D

escr

iptio

n

Hyn

obiu

s chi

nens

is (1

) Zho

usha

n I.

CA

S 19

4377

12

S-16

S: A

Y91

5981

N

D2-

CO

I: A

Y91

5933

Zhou

shan

Isla

nd, N

ingb

o Pr

efec

ture

, Zhe

jiang

Pr

ovin

ce, C

hina

H

ynob

ius c

hine

nsis

(2) M

ainl

and

MV

Z 23

0320

12

S-16

S: A

Y91

5982

N

D2-

CO

I: A

Y91

5934

Luo

Wus

han,

Hua

ntan

Zhe

ng, 3

5 km

S (b

y ro

ad) c

ity

of X

iaos

han,

Zhe

jiang

Pro

vinc

e, C

hina

H

ynob

ius d

unni

K

UH

E 24

848

12S-

16S:

AY

9159

74

ND

2-C

OI:

AY

9159

26O

ita C

ity, O

ita P

refe

ctur

e, Ja

pan

Hyn

obiu

s for

mos

anus

M

VZ

1972

38

12S-

16S:

AY

9159

92

ND

2-C

OI:

AY

9159

44ne

ar M

eife

ng (W

usho

e), T

aiw

an

Hyn

obiu

s hid

amon

tanu

sK

UH

E 94

84

12S-

16S:

AY

9159

83

ND

2-C

OI:

AY

9159

35H

akub

a-m

ura,

Nag

ano

Pref

ectu

re, J

apan

Hyn

obiu

s hir

osei

J-

36, n

o vo

uche

r 12

S-16

S: A

Y91

5994

N

D2-

CO

I: A

Y91

5946

Shik

oku,

Jap

an

Hyn

obiu

s kim

urae

K

UH

E 22

370

12S-

16S:

AY

9159

95

ND

2-C

OI:

AY

9159

47O

tsu,

Shi

ga P

refe

ctur

e, Ja

pan

Hyn

obiu

s lee

chii

MV

Z 16

3727

12

S-16

S: A

Y91

5976

N

D2-

CO

I: A

Y91

5928

Chu

ng-J

u, S

outh

Chu

ngch

ong

Prov

ince

, Sou

th K

orea

Hyn

obiu

s lic

hena

tus

J-9,

no

vouc

her

12S-

16S:

AY

9159

88

ND

2-C

OI:

AY

9159

40H

irosa

ki C

ity, A

omor

i Pre

fect

ure,

Japa

n

Hyn

obiu

s nae

vius

K

UH

E 12

984

12S-

16S:

AY

9159

85

ND

2-C

OI:

AY

9159

37K

itaky

ushu

City

, Fuk

uoka

Pre

fect

ure,

Jap

an

Hyn

obiu

s neb

ulos

us

KU

HE

2469

8 12

S-16

S: A

Y91

5973

N

D2-

CO

I: A

Y91

5925

Isah

aya,

Nag

asak

i Pre

fect

ure,

Japa

n

Hyn

obiu

s nig

resc

ens

KU

HE

1792

4 12

S-16

S: A

Y91

5991

N

D2-

CO

I: A

Y91

5943

Miy

agi P

refe

ctur

e, Ja

pan

Hyn

obiu

s oki

ensi

s K

UH

E 18

917

12S-

16S:

AY

9159

79

ND

2-C

OI:

AY

9159

31Sa

igo-

cho,

Shi

man

e Pr

efec

ture

, Jap

an

cont

inue

d ne

xt p

age

WEISROCK ET AL.80 · Zootaxa 3626 (1) © 2013 Magnolia Press

TA

BL

E 1

. (co

ntin

ued)

Taxo

n Sp

ecim

en

Acc

essi

on

Gen

Ban

k A

cces

sion

Lo

calit

y D

escr

iptio

n

Hyn

obiu

s que

lpar

tens

is

MV

Z 23

3460

12

S-16

S: A

Y91

5978

N

D2-

CO

I: A

Y91

5930

Jang

cho-

ri, M

adan

-myo

n, H

aena

m-g

un, S

outh

Cho

lla

Prov

ince

, Sou

th K

orea

H

ynob

ius r

etar

datu

s K

UH

E 14

545

12S-

16S:

AY

9159

96

ND

2-C

OI:

AY

9159

48N

iikap

pu-c

ho, H

okka

ido

Pref

ectu

re, J

apan

Hyn

obiu

s son

ani

MV

Z 19

7249

12

S-16

S: A

Y91

5993

N

D2-

CO

I: A

Y91

5945

near

Tie

n C

hi, T

aiw

an

Hyn

obiu

s ste

jneg

eri

KU

HE

1298

4 12

S-16

S: A

Y91

5985

N

D2-

CO

I: A

Y91

5937

Gok

ase-

cho,

Miy

azak

i Pre

fect

ure,

Japa

n

Hyn

obiu

s tak

edai

K

UH

E 24

764

12S-

16S:

AY

9159

90

ND

2-C

OI:

AY

9159

42H

akui

, Ish

ikaw

a Pr

efec

ture

, Jap

an

Hyn

obiu

s tok

yoen

sis

KU

HE

1691

1 12

S-16

S: A

Y91

5989

N

D2-

CO

I: A

Y91

5941

Yok

aich

iba,

Chi

ba P

refe

ctur

e, Ja

pan

Hyn

obiu

s tsu

ensi

s K

UH

E 18

367

12S-

16S:

AY

9159

75

ND

2-C

OI:

AY

9159

27Iz

uhar

a-ch

o, N

agas

aki P

refe

ctur

e, Ja

pan

Hyn

obiu

s yan

gi

MV

Z 24

7158

12

S-16

S: A

Y91

5977

N

D2-

CO

I: A

Y91

5929

(129

° 17'

E, 3

5° 1

9'N

, alt.

45m

), H

yoam

-ri,

Jang

an-

eup,

Gija

ng-g

un, B

usan

-shi

, Sou

th K

orea

Li

ua sh

ihi

MV

Z 25

8041

12

S-16

S: A

Y91

6011

N

D2-

CO

I: A

Y91

5963

Zhu

Xia

n V

illag

e, L

ou P

ing

Tow

n, W

usha

n C

ount

y,

Sich

uan

Prov

ince

, Chi

na

Liua

tsin

paen

sis

MV

Z 23

1153

12

S-16

S: A

Y91

6012

N

D2-

CO

I: A

Y91

5964

1920

to 1

920

m, D

iao

Yu

Tai,

Huo

Zhe

n, Z

hou

Zhix

ian,

Zhe

jiang

Pro

v., C

hina

O

nych

odac

tylu

s fis

cher

i M

VZ

1637

31

12S-

16S:

AY

9159

69

ND

2-C

OI:

AY

9159

21Je

ong

Sun,

Jeon

g Su

n K

un K

ang

Won

Pro

vinc

e, S

outh

K

orea

O

nych

odac

tylu

s jap

onic

us (1

) E

MV

Z 23

1874

12

S-16

S: A

Y91

5970

N

D2-

CO

I: A

Y91

5922

1/4

mile

W. o

f Nik

ko, T

ochi

gi P

refe

ctur

e, H

onsh

u Is

land

, Jap

an

Ony

chod

acty

lus j

apon

icus

(2) W

J-

5, n

o vo

uche

r12

S-16

S: A

Y91

5971

N

D2-

CO

I: A

Y91

5923

Kaw

akam

i-mur

a, N

ara

Pref

ectu

re, J

apan

cont

inue

d ne

xt p

age

Zootaxa 3626 (1) © 2013 Magnolia Press · 81HYNOBIIDAE PHYLOGENETICS

TA

BL

E 1

. (co

ntin

ued)

Taxo

n Sp

ecim

en

Acc

essi

on

Gen

Ban

k A

cces

sion

Lo

calit

y D

escr

iptio

n

Pach

yhyn

obiu

s sha

ngch

enge

nsis

C

AS

1942

33

12S-

16S:

AY

9159

72,

ND

2-C

OI:

AY

9159

24H

ushi

ngdi

, 6 k

m fr

om B

eim

azhe

i, Lu

an P

refe

ctur

e,

Anh

ui P

rovi

nce,

Chi

na

Para

dact

ylod

on m

uste

rsi

CA

S 14

9129

12

S-16

S: A

Y91

5997

N

D2-

CO

I: A

Y91

5949

3 m

i N o

f Pag

hman

in D

arra

he P

ain

Stre

am,

Afg

hani

stan

Pa

rada

ctyl

odon

per

sicu

s M

VZ

2414

94

12S-

16S:

AY

9159

98

ND

2-C

OI:

AY

9159

505

km S

E of

Shi

raba

d, G

oles

tan

Prov

ince

, Ira

n

Pseu

dohy

nobi

us fl

avom

acul

atus

M

VZ

2311

51

12S-

16S:

AY

9160

13

ND

2-C

OI:

AY

9159

65H

anch

i, Li

chua

n C

ount

y, H

ubei

Pro

vinc

e, C

hina

Rano

don

sibi

ricu

s C

AS

1995

30

12S-

16S:

AY

9159

99

ND

2-C

OI:

AY

9159

5120

km

E o

f Tek

eli,

Tald

i-Kor

gan

Dis

trict

, Kaz

akhs

tan

Sala

man

drel

la k

eyse

rlin

gii (

1)

Mon

golia

CA

S 19

4151

12

S-16

S: A

Y91

6000

N

D2-

CO

I: A

Y91

5952

1.5

km N

E of

tow

n sq

uare

in U

laan

Baa

tar,

Mon

golia

Sala

man

drel

la k

eyse

rlin

gii (

2)

Sibe

riaM

VZ

2223

30

12S-

16S:

AY

9160

01

ND

2-C

OI:

AY

9159

53M

aisk

ii V

illag

e, n

ear K

urum

kan,

Bar

guzi

n R

iver

V

alle

y, B

urya

t Rep

ublic

, Rus

sia

Sala

man

drel

la k

eyse

rlin

gii (

3)

Hok

kaid

oK

UH

E 13

062

12S-

16S:

AY

9160

02

ND

2-C

OI:

AY

9159

54K

ushi

ro, H

okka

ido

Pref

ectu

re,H

okka

ido

Isla

nd,J

apan

Sala

man

drel

la k

eyse

rlin

gii (

4)

Ura

lsM

VZ

2484

71

12S-

16S:

AY

9160

03

ND

2-C

OI:

AY

9159

55V

icin

ity o

f Eka

terin

burg

, Ura

l Mou

ntai

ns, R

ussi

a

WEISROCK ET AL.82 · Zootaxa 3626 (1) © 2013 Magnolia Press

DNA sequence data were collected from two regions of the mitochondrial genome. One region encompassed a contiguous stretch of approximately 1100 bases of sequence spanning a 3’ region of the 12S ribosomal RNA gene, the complete tRNAVAl gene, and a 5’ portion of the 16S ribosomal RNA gene. The second mtDNA region encompassed a contiguous stretch of approximately 1450 base pairs of sequence spanning the complete ND2 gene, the tRNATrp, tRNAAla, tRNAAsn genes, the origin for light strand replication (OL), the tRNACys and tRNATyr genes, and a short 5’ portion of the COI gene.

Genomic DNA was extracted from liver or muscle using the Qiagen QIAamp tissue kit. Amplification of genomic DNA was conducted using a denaturation at 94°C for 35 sec, annealing at 50°C for 35 sec, and extension at 70°C for 150 sec with 4 sec added to the extension per cycle, for 30 cycles. Negative controls were run for all amplifications. Amplified products were purified on 2.5% Nusieve GTG agarose gels and reamplified under similar conditions. Reamplified double-stranded products were purified on 2.5% acrylamide gels (Maniatis et al., 1982). Template DNA was eluted from acrylamide passively over three days with Maniatis elution buffer (Maniatis et al., 1982). Cycle-sequencing reactions were run using the Promega fmol DNA-sequencing system with a denaturation at 95°C for 35 sec, annealing at 45–60°C for 35 sec, and extension at 70°C for 1 min for 30 cycles. Sequencing reactions were run on Long Ranger sequencing gels for 5–12 hours at 38-40°C.

Amplifications from genomic DNA were done using two main primer combinations (Table 2): L692–H3002 for the 12S–16S fragment, and L4437a–H5934 for the ND2–COI fragment. For Andrias davidianus, L4437b was used instead of L4437a, and for Cryptobranchus alleganiensis, L4160 was used instead of L4437a. Both strands were sequenced using the primers listed in Table 2. Primer numbers refer to the 3’ end on the human mitochondrial genome (Anderson et al., 1981), where L and H correspond to light and heavy strands, respectively. All sequences generated in this study are deposited in GenBank (See Table 1 for accession numbers). Specimen collection was started in 1986 and DNA sequencing initiated in 1992.

TABLE 2. Primers used in this study. Primers are designated by their 3’ ends which correspond to the position in the human mitochondrial genome (Anderson et al. 1981) by convention. H and L designate heavy and light strand primers respectively. Positions with mixed bases are labeled with their standard one-letter code: R = G and A.

Human Gene Sequence ReferencePosition

L889 12S 5’-GGGTTGGTAAATCTCGTGC-3’ Titus and Larson, 1995H1165 12S 5’-TAGAGCACCGCCAAGTCCTTTG-3’ Titus and Larson, 1995L1090 12S 5’-AAACTGGGATTAGATACCCCACTA-3’ modified from Kocher et al., 1989H1478 12S 5’-AGGGTGACGGGCGGTGTGT-3’ modified from Kocher et al., 1989L1497 12S 5’-ACACACCGCCCGTCACCCTC-3’ modified from Kocher et al., 1989H2716 16S 5’-GCTTCATAGGGTCTTCTCGTC-3’ Titus and Larson, 1995L4160 ND1 5’-CGATTCCGATATGACCARCT-3’ Kumazawa and Nishida, 1993L4437a tRNAMet 5’-AAGCTTTCGGGCCCATACC-3’ Macey et al., 1997aL4437b tRNAMet 5’-AAGCAGTTGGGCCCATRCC-3’ Macey et al., 1997cL4879 ND2 5’-TCAACTTGACAAAAACTAGC-3’ this studyL4882 ND2 5’-TGACAAAAACTAGCACC-3’ Macey et al., 1997aH4980 ND2 5’-ATTTTTCGTAGTTGGGTTTGRTT-3’ Macey et al., 1997aL5002 ND2 5’-AACCAAACCCAACTACGAAAAAT-3’ Macey et al., 1997aH5532 tRNATrp 5’-GGCTTTGAAGGCCTTTGGTC-3’ this studyL5551 tRNATrp 5’-GACCAAAGGCCTTCAAAGCC-3’ Macey et al., 1997bH5692 tRNAAsn 5’-GCGTTTAGCTGTTAACTAAA-3’ this studyL5775 tRNACys 5’-AGAAGCCCCGAAGAAAA-3’ this studyH5934 COI 5’-AGRGTGCCAATGTCTTTGTGRTT-3’ Macey et al., 1997a

Phylogenetic tree reconstruction. Alignment of protein coding and tRNA gene sequences were performed manually using amino-acid sequence translations for protein-coding genes and secondary-structural models for tRNA genes. Transfer-RNA secondary structure was determined manually using the criteria of Kumazawa & Nishida (1993) to ensure proper alignment (Macey & Verma, 1997). Protein-coding sequences were translated to amino acids using MacClade (Maddison & Maddison, 2000) for confirmation of alignment.

Zootaxa 3626 (1) © 2013 Magnolia Press · 83HYNOBIIDAE PHYLOGENETICS

Alignments of mitochondrial ribosomal RNA gene sequences were constructed based on secondary structural models. The second and fourth domains of the 12S rRNA gene were aligned to the secondary structural model of Cyprinus carpio from Van de Peer et al. (1994). The third domain of the 12S rRNA gene was aligned to the secondary structural models presented in Hickson et al. (1996). Stems were numbered as in Van de Peer et al.(1994). For the 16S rRNA gene sequences, alignments were constructed based on the secondary structural model of Xenopus leavis from Gutell & Fox (1988). Stems of the 16S rRNA gene were numbered from the 5’ end and the region sequenced corresponds to the first half of the 5’ half of the gene. Stem regions were further assessed by examination of base-pairing across all taxa. Sequences homologous to the longest stem discovered across all taxa were considered to correspond to a potential stem region. The data alignment of all DNA sequences used in this study has been deposited in the Dryad online repository (http://dx.doi.org/10.5061/dryad.g9117).

We performed phylogenetic analysis on our data using parsimony, Bayesian, and maximum likelihood criteria. Our parsimony analyses were performed on a total concatenated data set of all DNA sequence data. Parsimony analyses were performed using the program PAUP* v4.0 (Swofford, 2002). Heuristic searches were performed with 100 random-addition replicates and TBR branch swapping. All analyses gave equal weight to all character changes. To assess support for branches in parsimony trees, bootstrap percentages were calculated using 1000 bootstrap replicates with 100 random additions per replicate, and decay indices were calculated using constraint trees generated in TreeRot v2 (Sorenson, 1999) and analyzed in PAUP*.

Our Bayesian analyses of the total data set used two different data partitioning schemes: A three partition analysis using the protein coding, tRNA, and rRNA gene regions, and a seven partition analysis separating the

protein coding data into 1st, 2nd, and 3rd positions, the encoding tRNA data into stems and loops, and the rRNA encoding data into stems and loops.

Bayesian phylogenetic analysis was performed using MrBayes v3.1.2 (Huelsenbeck et al., 2001; Ronquist & Huelsenbeck, 2003). All Bayesian analyses used four Markov chains with the temperature profile at the default setting of 0.2. The best-fit evolutionary model used was determined using Akaike Information Criterion as implemented in MODELTEST v3.06 (Posada & Crandall, 1998). Flat dirichlet priors were used for the General Time-Reversible (GTR) substitution-rate parameters and for all base-frequency parameters. A flat Beta prior was used in estimating the transition/transversion substitution-rate parameter. Uniform priors were used for the gamma shape parameter and the proportion of invariant sites parameter. Unconstrained, uniform priors were used for topology and branch-length estimation. A molecular clock was not enforced. Five million generations were run with a sample taken every 1000th generation. The program TRACER v1.3 (Rambaut & Drummond, 2007) was used to determine when the sampling log likelihoods (lnL) reached a stationary distribution. In all Bayesian analyses, a stable posterior was reached within 50,000 generations; the first one million generations were discarded as burn-in. Bayesian runs were replicated in four independent analyses. Trees sampled from the posterior distribution of all four analyses were combined into a single file and parsed with the program TreeAnnotator v1.4 (Drummond & Rambaut, 2007) to generate a consensus phylogram and posterior probabilities for individual branches.

Maximum likelihood analyses were conducted using RAxML v7.0.4 (Stamatakis et al., 2005). We conducted a rapid Bootstrap analysis and searched for the best-scoring ML in a single run, with the rapid hill-climbing algorithm (Stamatakis et al., 2008), and one partition for each gene region (12S, 16S, OL, ND2 + COI, and each tRNA; note, the small COI segment was combined with the ND2 protein encoding region because RAxML will not allow for such a short partition) for a total of 11 partitions, under the GTRMIX model of nucleotide substitutions for 1000 bootstrap inferences. This model uses a GTR approximation to initiate the search, and uses a GTR + Γ model, with 25 discrete gamma rate categories, for the final tree topology.

Results

Our total mtDNA data alignment of the 12S–16S and ND2–COI gene regions contains 2566 character positions, of which 151 are excluded due to ambiguous alignment. Of the included characters, 1157 are variable with 957 of these being parsimony informative. Parsimony analysis of the total data set produces two trees of 4477 steps in length. The strict consensus of these two trees is presented (Fig. 1).

Using AIC, the three-partition character sets (protein-coding, tRNA, and rRNA) are each found to be best fit to a GTR+I+G model. The seven-partition character sets are each found to be best fit to a GTR+I+G model, except for the tRNA loops, which are best fit to an HKY+I+G model.

WEISROCK ET AL.84 · Zootaxa 3626 (1) © 2013 Magnolia Press

FIGURE 1. Consensus of two equally most parsimonious trees resulting from an analysis of the combined 12S–16S and ND2–COI mtDNA genic sequence data. Species with two or more sampled individuals are numbered as in Table 1 for sample location. Filled circles on nodes represent parsimony bootstrap values of ≥ 95. Numbers above branches are bootstrap values < 95. Bootstrap values < 50 are not presented. Bold numbers below branches represent decay indices. Filled diamonds represent branches supported with parsimony bootstrap values ≥ 90, Bayesian posterior probabilities ≥ 0.95, and maximum likelihood bootstrap values ≥ 95.

Zootaxa 3626 (1) © 2013 Magnolia Press · 85HYNOBIIDAE PHYLOGENETICS

FIGURE 2. Bayesian majority-rule consensus phylogram resulting from the posterior density of a seven-partition analysis of the combined 12S–16S and ND2–COI mtDNA genic sequence data. There was no topological difference with this tree and that derived from a three partition analysis of the data. Species with two or more sampled individuals are numbered as in Table 1 for sample location. Filled circles on nodes represent Bayesian posterior probabilities ≥ 0.95. Posterior probabilities < 0.95 are indicated on their respective branches. Filled diamonds represent branches supported with parsimony bootstrap values ≥ 90, Bayesian posterior probabilities ≥ 0.95, and maximum likelihood bootstrap values ≥ 95.

WEISROCK ET AL.86 · Zootaxa 3626 (1) © 2013 Magnolia Press

FIGURE 3. Tree resulting from a twelve-partition maximum likelihood analysis of the combined 12S–16S and ND2–COI mtDNA genic sequence data. Species with two or more sampled individuals are numbered as in Table 1 for sample location. Filled circles on nodes represent maximum likelihood bootstrap values of ≥ 95. Numbers on branches are bootstrap values < 95. Filled diamonds represent branches supported with parsimony bootstrap values ≥ 90, Bayesian posterior probabilities ≥ 0.95, and maximum likelihood bootstrap values ≥ 95.

Zootaxa 3626 (1) © 2013 Magnolia Press · 87HYNOBIIDAE PHYLOGENETICS

Bayesian analysis of the three-partition data produces a posterior density with an average lnL of –23210.0. Bayesian analysis of the seven-partition data set produces a posterior density with an average lnL of –22850.0. The topological differences between the two different partition schemes are minimal and the consensus phylogram from the seven-partition posterior density is presented (Fig. 2). Maximum likelihood analysis of a 12 partition data set (analyzing each tRNA genic region separately) produced a single tree with an lnL of –27270.3 (Fig. 3).

In general, the parsimony, Bayesian, and ML trees agree in the resolution of many of the major hynobiid clades. For example, monophyly of the specious-genus Hynobius receives high levels of branch support (PB = 94, DI = 8, PP = 1.0, MLB = 99). Branches yielding monophyly of the genera Batrachuperus, Liua, Onychodactylus,Paradactylodon, and Salamandrella receive similar or stronger levels of support (Figs. 1–3). There is also concordance across analyses in the relationships among some of these clades: the Hynobiidae is monophyletic (PB = 100, DI = 139, PP = 1.0, MLB = 100), the genus Onychodactylus is placed as sister to the remaining hynobiids (PB = 100, DI = 63, PP = 1.0, MLB = 100), and the monotypic genus Ranodon is placed in a clade with Paradactylodon (BP = 91, DI = 9, PP = 1.0, MLB = 100). All analyses also place the genus Pseudohynobius in a clade with the genus Liua, and place this clade in a larger clade with the genus Batrachuperus; however while Bayesian PPs and ML bootstrap values are high for these relationships (PP >0.98, MLB >94), parsimony support is much weaker (BP = 68 and 70, DI = 4 and 6).

Discordance among analyses is most distinctly seen among the deeper attempts at reconstructing relationships of the Hynobiidae. Bayesian and ML analyses place Salamandrella as the sister clade to all remaining hynobiids except for Onychodactylus (PP = 0.98, MLB <50) and these analyses place Hynobius, Pachyhynobius, Paradactylodon, and Ranodon in a separate clade (PP = 0.99, MLB <50). All analyses place Ranodon and Paradactylodon together with strong branch support. However, in the parsimony tree the Ranodon and Paradactylodon clade is placed as the sister clade to all remaining hynobiids except for Onychodactylus (PB <50, DI = 4), Salamandrella is placed in a clade with Pachyhynobius (PB <50, DI = 4), and Hynobius is placed sister to a clade containing Batrachuperus, Liua, and Pseudohynobius (PB <50, DI = 4).

Discussion

With taxon sampling that comprises approximately 80% of all known species and a large mtDNA data set containing over 2400 analyzed characters, our phylogenetic results resolve the Hynobiidae into a number of strongly supported clades. Nonetheless, while the deepest divergence between the genus Onychodactylus (major-group I, Fig. 4) and all remaining hynobiids (major-group II, Fig. 4) is overwhelmingly supported in all analyses, relationships among the remaining major hynobiid clades are considerably less clear. For example, Bayesian analysis places Salamandrella as the sister clade to all remaining hynobiids, excluding Onychodactylus, with a PP = 0.98. Alternatively, maximum likelihood analysis resolves this relationship with weak branch support, and parsimony analysis weakly places the Ranodon and Paradactylodon clade in this position and instead weakly supports Salamandrella as the sister clade to Pachyhynobius. Discordance across optimality criteria for phylogenetic relationships among deeply diverged groups is considerable.

In the parsimony-based analysis (Fig. 1) we identify three weakly-supported branches (each with bootstrap support <50 and decay values of four) that are not found in the Bayesian and maximum likelihood trees (Figs. 2, 3). Collapsing these branches in the parsimony tree results in a polytomy that comprises five main clades (major group II, Fig. 4): (A) Ranodon + Paradactylodon, (B) Pachyhynobius, (C) Salamandrella, (D) Pseudohynibius + Liua + Batrachuperus, and (E) Hynobius. Bayesian analysis resolves the relationships among some of these clades with apparent strong levels of branch support. These results indicate the placement of Hynobius, Pachyhynobius, and the Ranodon + Paradactylodon clade together in a larger clade with a PP = 0.99. This larger clade is then placed as the sister clade to the Pseudohynibius + Liua + Batrachuperus clade with a PP = 0.98. Similar results are seen in the ML tree, although branch support is considerably lower.

The most likely synthesis of these results are that the temporal durations of branching events during the early periods of hynobiid diversification (after the split with Onychodactylus) were very short, relative to subsequent durations of evolutionary time. Bayesian and ML analyses may be expected to better account for this type of phylogenetic history. However, these phylogenetic approaches rely heavily on proper model specification (Huelsenbeck & Rannala 2004), and in the case of Bayesian analysis, on properly dealing with prior uncertainty

WEISROCK ET AL.88 · Zootaxa 3626 (1) © 2013 Magnolia Press

(Zwickl & Holder 2004; Kolaczkowski & Thornton 2007). An inability to meet these criteria can produce inflated branch support for inaccurate topologies. Indeed, Weisrock et al. (2005) documented high Bayesian support for deep and short mtDNA branches in salamander family phylogenetics that were strongly discordant with robustly supported results from Bayesian and parsimony analyses of a more slowly evolving nuclear gene (Wiens et al., 2005). The increased level of taxon sampling in this study relative to previous studies has done well in clarifying how species are grouped into major clades, but without concordant signal across analyses the phylogenetic relationships among these five main hynobiid clades appears to be largely unresolved.

FIGURE 4. Simplified cladogram of the Hynobiidae depicting the basal split between Onychodactylus (major-group I) and all remaining hynobiids (major-group II) and the polytomy among the five major hynobiid clades (see Discussion for IIA-IIE). The polytomy is inferred based on conflicting estimates of relationships across phylogenetic optimality criteria and weak branch support in the parsimony and maximum likelihood trees.

Zootaxa 3626 (1) © 2013 Magnolia Press · 89HYNOBIIDAE PHYLOGENETICS

Interestingly, similar patterns are seen in the complete mtDNA genome results of Zhang et al. (2006). As in our results, discordance across optimality criteria exists at one of the deeper nodes in the complete mtDNA genome tree [node d in Zhang et al. (2006)] and additional deep nodes that receive Bayesian PPs = 1.0 are paired with reduced levels of parsimony bootstrap support [nodes f, j, and k in Zhang et al. (2006)]. With this in mind, it is important to consider the greater potential for systematic bias in large data sets as a result of an increased potential for convergently evolved character states. This problem can become more acute when taxon sampling is limited (Soltis et al., 2004), and has been documented to produce inaccurate topologies in a number of studies (e.g., Stefanovic et al., 2004; Samuels et al., 2005; Nishihara et al., 2007). Increased taxon sampling in the complete mtDNA genome results of Zheng et al. (2011) does not appear to improve resolution of deep hynobiid phylogenetic relationships, as their complete mtDNA genome tree conflicts substantially with a tree derived from concatenated analysis of three nuclear genes, particularly with respect to the placement of Salamandrella. Overall, we suggest that the substantial discordance seen across analyses in this study, and across data sets in other studies leaves us with a very uncertain picture of the deeper regions of hynobiid phylogenetic history. Our best interpretation is that, while the sister relationship between Onychodactylus and all remaining hynobiids is robust, the five major clades within this latter group should be best viewed as a polytomy (Fig. 4) until subsequent studies can provide new evidence that robustly disentangles their history. Future studies utilizing near-complete taxon sampling and multiple independent genes analyzed in a non-concatenated framework are expected to be necessary to more robustly reconstruct these regions of hynobiid phylogenetic relationships.

The persistence of the ancient group comprising Onychodactylus species at high-latitudes in northeastern Asia is intriguing. It is noteworthy that much of northern Asia, particularly the northeast, was never fully glaciated, unlike most of North America and Europe (McIntrye et al., 1976).

Finally, it has not escaped us that the complexity of plate tectonic development of Asia has played a major role in the diversification of the Hynobiidae. Asia is a conglomeration of plates that fused to the southern margin (reviewed in Macey et al., 1999; Macey et al., 2000). Ancient plates that currently are internal allow for more modern shifts along ancient suture zones. The North and South China Blocks fused, followed by numerous small plates making up Tibet, Iran, Afghanistan, and as deep as Mongolia. The major modern events that occur impinging these plates are: (1) 120 million years ago Southeast Asian plates conglomerate and push into Asia in a northwestern direction; (2) India collides with South Asia 50 million years ago in a northern direction that underplates Tibet; and (3) Arabia collides 18 million years ago with fused ancient plates now in southwestern Asia, and Arabia is forced in a northeastern direction by the northern movement of Africa contacting Europe. The three modern Asian plate regions are moving in different directions, impacting internal Asian plates with such force that ancient suture zones become the largest grouping of high-elevation mountain belts in the world; hence producing internal continental speciation. We know plate tectonics result in speciation among the Hynobiidae because of strict endemism in Taiwan associated with no close mainland taxa, caused by the north movement of the Philippine Plate and uplifting the island 4 million years ago (Angelier et al., 1990; Lee et al.,1997). We suggest that future studies aimed at resolving the biogeographic history of the Hynobiidae strongly consider the role of plate tectonics as an important factor in driving speciation and diversification.

Acknowledgments

We thank Zhili Fang, Tatjana N. Dujsebayeva, Natalia B. Ananjeva, H. G. Kami, and Kh. Monkhbayar for field assistance, and Eric Routman for providing the Cryptobranchus sample We thank the National Science Foundation (BSR-9106898 to Allan Larson; DEB-9318642 to Jonathan B. Losos, Kevin de Queiroz, and Allan Larson; DEB-9726064 to Allan Larson, J. R. Macey, and T. J. Papenfuss) and the National Geographic Society (4110–89 and 4872–93 to T. J. Papenfuss and J. R. Macey) for financial support. We thank Kraig Adler for facilitating contacts among scientists. We most of all thank Allan Larson for guidance, as this project was dear, and now complete — we dedicate this paper to Allan Larson who has long been on the breaking edge of salamander Genetics that is now known as Genomics.

WEISROCK ET AL.90 · Zootaxa 3626 (1) © 2013 Magnolia Press

References

Anderson, S., Bankier, A. T., Barrell, B. G., de Bruijn, M. H. L., Coulson, A. R., Drouin, J., Eperon, I. C., Nierlich, D. P., Roe, B. A., Sanger, F., Schreier, P. H., Smith, A. J. H., Staden, R., & Young, I. G. (1981). Sequence and organization of the human mitochondrial genome. Nature, 290, 457–465. http://dx.doi.org/10.1038/290457a0

Angelier, J., Bergerat, F., Chu, H.-T., & Lee, T.-Q. (1990) Tectonic analyses and the evolution of a curved collision belt: The Hsüehshan Range, northern Taiwan. in Geodynamic Evolution of the Eastern Eurasian Margin (Angelier, J., ed.). Tectonophysics, 183, 77–96.

Braun, E. L. & Kimball, R. T. (2002) Examining basal avian divergences with mitochondrial sequences: model complexity, taxon sampling, and sequence length. Systematic Biology, 51, 614–625. http://dx.doi.org/10.1080/10635150290102294

DeBry, R. W. (2005) The systematic component of phylogenetic error as a function of taxonomic sampling under parsimony. Systematic Biology, 54, 432–440. http://dx.doi.org/10.1080/10635150590946745

Drummond, A. J. & Rambaut, A. (2007) BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology, 7, 214.

Fu, J. Z., Wang, Y. Z., Zeng, X. M., Liu, Z. J. & Zheng, Y. C. (2001) Genetic diversity of eastern Batrachuperus (Caudata: Hynobiidae). Copeia, 1100–1107. http://dx.doi.org/10.1643/0045-8511(2001)001[1100:GDOEBC]2.0.CO;2

Gutell, R. R. & Fox, G. E. (1988) A compilation of large subunit RNA sequences presented in a structural format. Nucleic Acids Research, 16 Suppl, r175–269. http://dx.doi.org/10.1093/nar/16.suppl.r175

Heath, T. A., Zwickl, D. J., Kim, J. & Hillis, D. M. (2008) Taxon sampling affects inferences of macroevolutionary processes from phylogenetic trees. Systematic Biology, 57, 160–166. http://dx.doi.org/10.1080/10635150701884640

Hickson, R. E., Simon, C., Cooper, A., Spicer, G. S., Sullivan, J. & Penny, D. (1996) Conserved sequence motifs, alignment, and secondary structure for the third domain of animal 12S rRNA. Molecular Biology and Evolution, 13, 150–169. http://dx.doi.org/10.1093/oxfordjournals.molbev.a025552

Hillis, D. M., Pollock, D. D., McGuire, J. A. & Zwickl, D. J. (2003) Is sparse taxon sampling a problem for phylogenetic inference? Systematic Biology, 52, 124–126. http://dx.doi.org/10.1080/10635150390132911

Huelsenbeck, J. P., Ronquist, F., Nielsen, R. & Bollback, J. P. (2001) Bayesian inference of phylogeny and its impact on evolutionary biology. Science, 294, 2310–2314. http://dx.doi.org/10.1126/science.1065889

Kocher, T. D., Thomas, W. K., Meyer, A., Edwards, S. V., Paabo, S., Villablanca, F. X. & Wilson, A. C. (1989) Dynamics of mitochondrial DNA evolution in animals: Amplification and sequencing with conserved primers. Proceedings of the National Academy of Sciences of the United States of America, 86, 6196–6200. http://dx.doi.org/10.1073/pnas.86.16.6196

Kolaczkowski, B. & Thornton, J. W. (2007) Effects of branch length uncertainty on Bayesian posterior probabilities for phylogenetic hypotheses. Molecular Biology and Evolution, 24, 2108–2118.

Kumazawa, Y. & Nishida, M. (1993) Sequence evolution of mitochondrial transfer RNA genes and deep-branch animal phylogenetics. Journal of Molecular Evolution, 37, 380–398. http://dx.doi.org/10.1007/BF00178868

Lee, J.-C., Angelier, J., & Chu, H.-T. (1997) Polyphase history and kinematics of a complex major fault zone in the northern Taiwan mountain belt: The Lishan fault. in An Introduction to Active Collision in Taiwan (Lallemand, S. E., & Tsien, H.-H., eds.). Tectonophysics, 274, 97–115.

Macey, J. R. & Verma, A. (1997) Homology in phylogenetic analysis: alignment of transfer RNA genes and the phylogenetic position of snakes. Molecular Phylogenetics and Evolution, 7, 272–279. http://dx.doi.org/10.1006/mpev.1997.0379

Macey, J. R., Larson, A., Ananjeva, N. B., Fang, Z. L. & Papenfuss, T. J. (1997a) Two novel gene orders and the role of light-strand replication in rearrangement of the vertebrate mitochondrial genome. Molecular Biology and Evolution, 14, 91–104. http://dx.doi.org/10.1093/oxfordjournals.molbev.a025706

Macey, J. R., Larson, A., Ananjeva, N. B. & Papenfuss, T. J. (1997b) Replication slippage may cause parallel evolution in the secondary structures of mitochondrial transfer RNAs. Molecular Biology and Evolution, 14, 30–39. http://dx.doi.org/10.1093/oxfordjournals.molbev.a025699

Macey, J. R., Larson, A., Ananjeva, N. B. & Papenfuss, T. J. (1997c) Evolutionary shifts in three major structural features of the mitochondrial genome among iguanian lizards. Journal of Molecular Evolution, 44, 660–674. http://dx.doi.org/10.1007/PL00006190

Macey, J. R., Wang, Y., Ananjeva, N. B., Larson, A., & Papenfuss, T. J. (1999) Vicariant patterns of fragmentation among

Zootaxa 3626 (1) © 2013 Magnolia Press · 91HYNOBIIDAE PHYLOGENETICS

gekkonid lizards of the genus Teratoscincus produced by the Indian Collision: A molecular phylogenetic perspective and an area cladogram for Central Asia. Molecular Phylogenetics and Evolution, 12, 320–332. http://dx.doi.org/10.1006/mpev.1999.0641

Macey, J. R., Schulte II, J. A., Larson, A., Ananjeva, N. B., Wang, Y., Pethiyagoda, R., Rastegar-Pouyani, N., & Papenfuss, T. J. (2000) Evaluating trans-Tethys migration: An example using acrodont lizard phylogenetics. Systematic Biology, 49, 233–256. http://dx.doi.org/10.1093/sysbio/49.2.233

Maddison, W. P. & Maddison, D. R. (2000) MacClade4: analysis of phylogeny and character evolution. Sinauer Associates, Sunderland, MA.

Maniatis, T. E., Fritsch, F. & Sambrook, J. (1982) Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

Matsui, M., Iwasawa, H., Takahashi, H., Hayashi, T. & Kumakura, M. (1992) Invalid Specific Status of Hynobius-Sadoensis Sato: Electrophoretic Evidence (Amphibia, Caudata). Journal of Herpetology, 26, 308–315. http://dx.doi.org/10.2307/1564886

Matsui, M., Kokuryo, Y., Misawa, Y. & Nishikawa, K. (2004) A new species of salamander of the genus Hynobius from central Honshu, Japan (Amphibia, Urodela). Zoological Science, 21, 661–669. http://dx.doi.org/10.2108/zsj.21.661

Matsui, M., Misawa, Y., Nishikawa, K. & Tanabe, S. (2000) Allozymic variation of Hynobius kimurae Dunn (Amphibia, Caudata). Comparative Biochemistry and Physiology B-Biochemistry & Molecular Biology, 125, 115–125. http://dx.doi.org/10.1016/S0305-0491(99)00154-6

Matsui, M., Nishikawa, K., Misawa, Y. & Tanabe, S. (2007) Systematic relationships of Hynobius okiensis among Japanese salamanders (Amphibia: Caudata). Zoological Science, 24, 746–751. http://dx.doi.org/10.2108/zsj.24.746

Matsui, M., Nishikawa, K., Tanabe, S. & Misawa, Y. (2001) Systematic status of Hynobius tokyoensis (Amphibia: Urodela) from Aichi Prefecture, Japan: a biochemical survey. Comparative Biochemistry and Physiology B-Biochemistry & Molecular Biology, 130, 181–189. http://dx.doi.org/10.1016/S1096-4959(01)00424-9

Matsui, M., Nishikawa, K., Utsunomiya, T. & Tanabe, S. (2006) Geographic allozyme variation in the Japanese clouded salamander, Hynobius nebulosus (Amphibia: Urodela). Biological Journal of the Linnean Society, 89, 311–330. http://dx.doi.org/10.1111/j.1095-8312.2006.00676.x

Matsui, M., Tominaga, A., Hayashi, T., Misawa, Y. & Tanabe, S. (2007) Phylogenetic relationships and phylogeography of Hynobius tokyoensis (Amphibia: Caudata) using complete sequences of cytochrome b and control region genes of mitochondrial DNA. Molecular Phylogenetics and Evolution, 44, 204–216. http://dx.doi.org/10.1016/j.ympev.2006.11.031

McIntrye, A., Moore, T. C., Andersen, B., Balsam, W., Bé, A., Brunner, C., Cooley, J., Crowley, T., Denton, G., Gardner, J., Geitzenauer, K., Hays, J. D., Hutson, W., Imbrie, J., Irwing, G., Kellogg, T., Kennett, J., Kipp, N., Kukla, G., Kukla, H., Lozano, J., Luz, B., Mangion, S., Matthews, R. K., Mayewski, P., Molfino, B., Ninkovich, D., Opdyke, N., Prell, W., Robertson, J., Ruddiman, W. F., Sachs, H., Saito, T., Shackleton, N., Thierstein, H., & Thompson, P. (1976) The surface of the ice-age earth. Science, 191, 1131–1137. http://dx.doi.org/10.1126/science.191.4232.1131

Nishihara, H., Okada, N. & Hasegawa, M. (2007) Rooting the eutherian tree: the power and pitfalls of phylogenomics. Genome Biology, 8, R199.

Nishikawa, K., Matsui, M. & Tanabe, S. (2005) Biochemical phylogenetics and historical biogeography of Hynobius boulengeri and H. stejnegeri (Amphibia: Caudata) from the Kyushu region, Japan. Herpetologica, 61, 54–62. http://dx.doi.org/10.1655/03-89

Nishikawa, K., Matsui, M., Tanabe, S. & Sato, S. (2001) Geographic enzyme variation in a Japanese salamander, Hynobius boulengeri Thompson (Amphibia: Caudata). Herpetologica, 57, 281–294.

Nishikawa, K., Matsui, M., Tanabe, S. & Sato, S. (2007) Morphological and allozymic variation in Hynobius boulengeri and H. steinegeri (Amphibia: Urodela: Hynobiidae). Zoological Science, 24, 752–766. http://dx.doi.org/10.2108/zsj.24.752

Peng, R., Zhang, P., Xiong, J. L., Gu, H. J., Zeng, X. M. & Zou, F. D. (2010) Rediscovery of Protohynobius puxiongensis(Caudata: Hynobiidae) and its phylogenetic position based on complete mitochondrial genomes. Molecular Phylogenetics and Evolution, 56, 252–258. http://dx.doi.org/10.1016/j.ympev.2009.12.011

Pollock, D. D., Zwickl, D. J., McGuire, J. A. & Hillis, D. M. (2002) Increased taxon sampling is advantageous for phylogenetic inference. Systematic Biology, 51, 664–671. http://dx.doi.org/10.1080/10635150290102357

Posada, D. & Crandall, K. A. (1998) MODELTEST: testing the model of DNA substitution. Bioinformatics, 14, 817–818. http://dx.doi.org/10.1093/bioinformatics/14.9.817

Pyron, R. A. & Wiens, J. J. (2011) A large-scale phylogeny of Amphibia including over 2800 species, and a revised classification of extant frogs, salamanders, and caecilians. Molecular Phylogenetics and Evolution, 61, 543–583.

WEISROCK ET AL.92 · Zootaxa 3626 (1) © 2013 Magnolia Press

http://dx.doi.org/10.1016/j.ympev.2011.06.012Rambaut, A. & Drummond, A. J. (2007) Tracer v1.5, Available from http://beast.bio.ed.ac.uk/TracerRoelants, K., Gower, D. J., Wilkinson, M., Loader, S. P., Biju, S. D., Guillaume, K., Moriau, L. & Bossuyt, F. (2007) Global

patterns of diversification in the history of modern amphibians. Proceedings of the National Academy of Sciences of the United States of America, 104, 887–892. http://dx.doi.org/10.1073/pnas.0608378104

Ronquist, F. & Huelsenbeck, J. P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics, 19, 1572–1574. http://dx.doi.org/10.1093/bioinformatics/btg180

Samuels, A. K., Weisrock, D. W., Smith, J. J., France, K. J., Walker, J. A., Putta, S. & Voss, S. R. (2005) Transcriptional and phylogenetic analysis of five complete ambystomatid salamander mitochondrial genomes. Gene, 349, 43–53. http://dx.doi.org/10.1016/j.gene.2004.12.037

Soltis, D. E., Albert, V. A., Savolainen, V., Hilu, K., Qiu, Y. L., Chase, M. W., Farris, J. S., Stefanovic, S., Rice, D. W., Palmer, J. D. & Soltis, P. S. (2004) Genome-scale data, angiosperm relationships, and 'ending incongruence': a cautionary tale in phylogenetics. Trends in Plant Science, 9, 477–483. http://dx.doi.org/10.1016/j.tplants.2004.08.008

Sorenson, M. D. (1999) TreeRot, version 2. Boston University, Boston, MA.Stamatakis, A., Hoover, P. & Rougemont, J. (2008) A rapid bootstrap algorithm for the RAxML Web servers. Systematic

Biology, 57, 758–771. http://dx.doi.org/10.1080/10635150802429642

Stamatakis, A., Ludwig, T. & Meier, H. (2005) RAxML-III: a fast program for maximum likelihood-based inference of large phylogenetic trees. Bioinformatics, 21, 456–463. http://dx.doi.org/10.1093/bioinformatics/bti191

Stefanovic, S., Rice, D. W. & Palmer, J. D. (2004) Long branch attraction, taxon sampling, and the earliest angiosperms: Amborella or monocots? BMC Evolutionary Biology, 4, 35. http://dx.doi.org/10.1186/1471-2148-4-35

Swofford, D. L. (2002) PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Sinauer Associates, Sunderland, Massachusettes.

Titus, T. A. & Larson, A. (1995) A molecular phylogenetic perspective on the evolutionary radiation of the salamander family Salamandridae. Systematic Biology, 44, 125–151. http://dx.doi.org/10.2307/2413703

Tominaga, A., Matsui, M., Nishikawa, K. & Sato, S. (2003) Occurrence of two types of Hynobius naevius in northern Kyushu, Japan (Amphibia: Urodela). Zoological Science, 20, 1467–1476. http://dx.doi.org/10.2108/zsj.20.1467

Tominaga, A., Matsui, M., Nishikawa, K. & Tanabe, S. (2005) Phylogenetic relationships of Hynobius naevius as revealed by mitochondrial 12S and 16S rRNA genes (Amphibia: Caudata). Zoological Science, 22, 1434–1434.

Tominaga, A., Matsui, M., Nishikawa, K. & Tanabe, S. (2006) Phylogenetic relationships of Hynobius naevius (Amphibia: Caudata) as revealed by mitochondrial 12S and 16S rRNA genes. Molecular Phylogenetics and Evolution, 38, 677–684.http://dx.doi.org/10.1016/j.ympev.2005.10.014

Van de Peer, Y., Van den Broeck, I., De Rijk, P. & De Wachter, R. (1994) Database on the structure of small ribosomal subunit RNA. Nucleic Acids Research, 22, 3488–3494. http://dx.doi.org/10.1093/nar/22.17.3488

Weisrock, D. W., Harmon, L. J. & Larson, A. (2005) Resolving deep phylogenetic relationships in salamanders: analyses of mitochondrial and nuclear genomic data. Systematic Biology, 54, 758–777. http://dx.doi.org/10.1080/10635150500234641

Wiens, J., Bonett, R. & Chippindale, P. (2005) Ontogeny discombobulates phylogeny: paedomorphosis and higher-level salamander relationships. Systematic Biology, 54, 91–110. http://dx.doi.org/10.1080/10635150590906037

Zeng, X. M., Fu, J. Z., Chen, L. Q., Tian, Y. Z. & Chen, X. H. (2006) Cryptic species and systematics of the hynobiid salamanders of the Liua-Pseudohynobius complex: molecular and phylogenetic perspectives. Biochemical Systematics and Ecology, 34, 467–477. http://dx.doi.org/10.1016/j.bse.2006.01.006

Zhang, P., Chen, Y., Zhou, H., Liu, Y., Wang, X., Papenfuss, T., Wake, D. & Qu, L. (2006) Phylogeny, evolution, and biogeography of Asiatic Salamanders (Hynobiidae). Proceedings of the National Academy of Sciences of the United States of America, 103, 7360–7365. http://dx.doi.org/10.1073/pnas.0602325103

Zheng, Y., Peng, R., Kuro-o, M. & Zeng, X. (2011) Exploring patterns and extent of bias in estimating divergence time from mitochondrial DNA sequence data in a particular lineage: a case study of salamanders (order Caudata). Molecular Biology and Evolution, 28, 2521–2535. http://dx.doi.org/10.1093/molbev/msr072

Zwickl, D. J. & Hillis, D. M. (2002) Increased taxon sampling greatly reduces phylogenetic error. Systematic Biology, 51, 588–598. http://dx.doi.org/10.1080/10635150290102339

Zwickl, D. & Holder, M. (2004) Model parameterization, prior distributions, and the general time-reversible model in Bayesian phylogenetics. Systematic Biology, 53, 877–888.

Zootaxa 3626 (1) © 2013 Magnolia Press · 93HYNOBIIDAE PHYLOGENETICS