Content uploaded by Shigeki Asai
Author content
All content in this area was uploaded by Shigeki Asai on May 28, 2019
Content may be subject to copyright.
DNA barcoding reveals 24 distinct lineages as cryptic bird
species candidates in and around the Japanese Archipelago
TAKEMA SAITOH,*
1
NORIMASA SUGITA,†
1
SAYAKA SOMEYA,†YASUKO IWAMI,*†
SAYAKA KOBAYASHI,* HIROMI KAMIGAICHI,†AKI HIGUCHI,†SHIGEKI ASAI,*
YOSHIHIRO YAMAMOTO* and ISA O NISHI UMI†
*Division of Natural History, Yamashina Institute for Ornithology, 115 Konoyama, Abiko, Chiba 270-1145, Japan, †Department
of Zoology, National Museum of Nature and Science, Amakubo 4-1-1, Tsukuba, Ibaraki 305-0005, Japan
Abstract
DNA barcoding using a partial region (648 bp) of the cytochrome coxidase I (COI) gene is a powerful tool for species
identification and has revealed many cryptic species in various animal taxa. In birds, cryptic species are likely to
occur in insular regions like the Japanese Archipelago due to the prevention of gene flow by sea barriers. Using COI
sequences of 234 of the 251 Japanese-breeding bird species, we established a DNA barcoding library for species
identification and estimated the number of cryptic species candidates. A total of 226 species (96.6%) had unique COI
sequences with large genetic divergence among the closest species based on neighbour-joining clusters, genetic dis-
tance criterion and diagnostic substitutions. Eleven cryptic species candidates were detected, with distinct intraspe-
cific deep genetic divergences, nine lineages of which were geographically separated by islands and straits within
the Japanese Archipelago. To identify Japan-specific cryptic species from trans-Paleartic birds, we investigated the
genetic structure of 142 shared species over an extended region covering Japan and Eurasia; 19 of these species
formed two or more clades with high bootstrap values. Excluding six duplicated species from the total of 11 species
within the Japanese Archipelago and 19 trans-Paleartic species, we identified 24 species that were cryptic species can-
didates within and surrounding the Japanese Archipelago. Repeated sea level changes during the glacial and inter-
glacial periods may be responsible for the deep genetic divergences of Japanese birds in this insular region, which
has led to inconsistencies in traditional taxonomies based on morphology.
Keywords: cryptic species, DNA barcoding, Japanese birds, Palearctic region, taxonomy
Received 7 February 2014; revision received 1 May 2014; accepted 5 May 2014
Introduction
DNA barcoding is a useful and proven tool for species
identification. A partial region (648 bp) of the cyto-
chrome coxidase I (COI) gene in the mitochondrial
genome is used as a standard DNA barcoding region for
most animals (Hebert et al. 2003). In vertebrates, the
DNA barcoding region has been very useful in identify-
ing species due to its high interspecific and low intraspe-
cific variation (Ward et al. 2005; Hern
andez-D
avila et al.
2012), and moreover, for identifying cryptic species
(Clare et al. 2007; Lara et al. 2010).
The DNA barcoding library of bird DNA includes
several geographical regions, such as the Nearctic
(Hebert et al. 2004), South Korea (Yoo et al. 2006), the
Neotropics (Kerr et al. 2009a; Tavares et al. 2011), the
eastern Palearctic (Kerr et al. 2009b) and Scandinavia
(Johnsen et al. 2010), and is continually expanding
(Mil
aet al. 2012). DNA barcoding of distinct conspecific
genetic divergences has revealed lineages of many
cryptic species within continents (Hebert et al. 2004;
Kerr et al. 2009b; Mil
aet al. 2012) and among trans-con-
tinents (Kerr et al. 2009a; Johnsen et al. 2010; Lijtmaer
et al. 2011). Lohman et al. (2010) investigated the bar-
coding region of six resident birds in South-East Asia,
which included many small islands, and found deep
genetic divergence among the islands, suggesting that
traditional taxonomy may have overlooked endemic
species in the area.
Geographical barriers, such as mountain ranges and
seas, prevent gene flow and form definite distribution
boundaries for species and subspecies. The Japanese
Archipelago is located off the eastern coast of the
Correspondence: Takema Saitoh and Isao Nishiumi,
Fax: +81-4-7182-1106; E-mails: saitoh@yamashina.or.jp and
nishiumi@kahaku.go.jp
1
These authors contributed equally to this work.
©2014 John Wiley & Sons Ltd
Molecular Ecology Resources (2014) doi: 10.1111/1755-0998.12282
Eurasian continent, across the Sea of Japan and the East
China Sea (Fig. 1). The avifauna of Japan consists of 633
bird species, including nonbreeding birds that largely
share the eastern Eurasian continent despite the sea bar-
rier; however, the Japanese Archipelago has 11 endemic
resident bird species and six migratory bird species
breeding only in that area (The Ornithological Society of
Japan 2012). The archipelago is treated as a biodiversity
hot spot (Conservation International 2012) and a new
zoogeographic region based on the phylogenetic rela-
tionship of the birds and other vertebrates (Holt et al.
2013). The high biodiversity in Japan is due to a wide
variety of climates and ecosystems, ranging from the
humid subtropics in the Ryukyu and Ogasawara islands
to the boreal zones in northern Japan, the alpine zone at
over 3000 m above sea level, and more than 3000 islands
(Conservation International 2012). The islands and the
straits between them have separated avian species over
geographical time. Traditional taxonomists have repeat-
edly disputed whether the populations isolated on
islands and/or by the straits should be recognized as
subspecies or as separate species (Ornithological Society
of Japan 2012), as the birds form a morphologically well-
defined species group. However, molecular analysis has
determined that the straits and seas have split the Japa-
nese birds into several distinct lineages, but these have
currently only been revealed for a limited number of spe-
cies such as the Arctic Warbler Phylloscopus borealis
(Saitoh et al. 2010), the Eurasian Jay Garrulus glandarius
(Akimova et al. 2007), the Varied Tit Sittiparus varius
(Nishiumi et al. 2006) and the Ryukyu Robin Erithacus
kamodori (Seki 2006; Seki et al. 2007).
The aim of this barcoding study was to examine
whether the diversification and phylogeographic struc-
ture of Japanese birds is related to the complex arrange-
ment of islands, which extend northeast and southwest
of Japan. We focused on current Japanese-breeding bird
species. We generated a DNA barcoding library of bird
species to test whether DNA barcoding was a suitable
method for species identification and to identify the prev-
alence of cryptic species. However, the region covered by
our survey was not sufficient to fully establish a relation-
ship between genetic diversification and the Japanese
Archipelago because many Japanese birds are common
throughout Eurasia. A wider-scale investigation was
needed, and accordingly, we conducted a comprehensive
survey to assess the genetic divergence of bird species
breeding in the east Eurasian Continent and the Japanese
Archipelago using published barcoding libraries.
Materials and methods
In total, 1367 voucher specimens representing 234 Japa-
nese bird species collected throughout the Japanese
Archipelago (Appendix S1, Supporting information)
were included in this research. Blood and frozen tissue
samples (pectoral muscle) were taken from voucher
specimens held at the National Museum of Nature and
Science, Tokyo (NSMT) and the Yamashina Institute for
Ornithology (YIO). All the samples were linked to vou-
cher specimens from the NSMT (49.3%), the YIO
(40.1%), the Higashi Taisetsu Museum of Natural His-
tory (6.8%), the Botanical Garden at the Field Science
Centre for Northern Biosphere, Hokkaido University
(1.9%), the Kushiro-Shitsugen Wildlife Centre (1%) and
other institutions (0.8%). Detailed information about
these samples is accessible via ‘Birds of Japan, NSMT’
and ‘Birds of Japan, YIO’ projects on the Barcode of
Life Data Systems (BOLD) website (http://www.bold
systems.org/).
To estimate the full genetic divergence of birds in East
Asia, including the insular region of the Japanese Archi-
pelago, we also used 1737 sequences representing 142
bird species from the intersection of 437 Eurasian species
(Yoo et al. 2006; Kerr et al. 2009b) and 234 Japanese spe-
cies. The taxonomy followed Clements (2007) as recom-
mended by the ‘All Birds Barcoding Initiative’ (http://
www.barcodingbirds.org/).
DNA extraction, PCR amplification and DNA
sequencing of the COI barcoding region was conducted
at the NSMT and YIO laboratories. DNA was extracted
from blood or tissue samples using a standard phenol–
chloroform procedure at NSMT and using the DNeasy
Blood & Tissue kit (Qiagen, Hilden, Germany) at YIO.
Several pairs of primers were used as standard primers
for amplification of the COI barcoding region; Bird
F1 (50-TTCTCCAACCACAAAGACATTGGCAC-30), Bird
R1 (50-ACGTGGGAGATAATTCCAAATCCTG-30) and
Bird R2 (50-ACTACATGTGAGATGATTCCGAATCCA
G-30), alongside newly designed primers, L6697Bird
(50-TCAACYAACCACAAAGAYATCGGYAC-30) and
H7390Thrush (50-ACGTGGGARATRATTCCAAATCCT
G-30) for passerine birds. If this approach was unsuc-
cessful, an alternative forward primer, FalcoFA (50-TCA
ACAAACCACAAAGACATCGGCAC-30), or a reverse
primer, VertebrateR1 (50-TAGACTTCTGGGTGGCCAA
AGAATCA-30), was used (Kerr et al. 2007). The 25 lL
polymerase chain reaction (PCR) reaction mix com-
prised 19.2 lL ultrapure water, 1.0 U Taq polymerase
(Ex Taq, TaKaRa, Shiga, Japan), 2.5 lL PCR buffer
(Mg
2+
free), 0.3 lL of each primer (0.24 mM), 2.5 lLof
each dNTP (2.5 mM) and 0.4–1.0 lL of DNA. The
amplification protocol was as follows: 94 °C for 3 min
followed by five cycles at 94 °C for 30 s, 48 °C for 30 s,
72 °C for 1 min, then 30 cycles at 94 °C for 30 s, 51 °C
for 30 s, 72 °C for 1 min and a final 72 °C for 5 min.
The PCR products were visualized on a 1.5% agarose
gel stained with ethidium bromide and purified using
©2014 John Wiley & Sons Ltd
2T. SAITOH ET AL.
ExoSAP-IT (Amersham Biosciences, Little Chalfont,
Buckinghamshire, UK) according to the manufacturer’s
instructions. Sequencing reactions were carried out
using BigDye Terminator v1.1 (ABI, Paisley, UK), and
analysis was performed on an ABI 3130 Genetic
Analyser (ABI) at NSMT, and BigDye Terminator v3.1
(ABI) and an ABI 3100 at YIO. COI sequences were
re-covered for all 234 bird species and did not contain
insertions, deletions, or nonsense- or stop codons,
supporting the absence of nuclear pseudogene amplifi-
cation (Song et al. 2008).
Genetic distances of breeding birds in Japan were
ascertained using the Distance Summary and the Bar-
code Gap Analysis applications of the BOLD website,
with the Kimura 2-parameter model (K2P) (Ratnasing-
ham & Hebert 2007). Neighbour-joining (NJ) trees
with bootstrap values (1000 replications) were con-
structed from the COI data using the K2P model and
Mega 6.05 (Tamura et al. 2013). In cases of very low
nearest NJ distances (<1%), the Diagnostic Character
application on the BOLD website was used for
a molecular diagnostic of species assignment. We
applied >1.6% intraspecific variation as the criterion to
find candidates of cryptic species (Kerr et al. 2009b).
Similarly, genetic distances of 142 common birds from
the Japanese Archipelago and the Eurasian continents
were calculated using the applications on the BOLD
website. The mean genetic distances among popula-
tions and bootstrap values of the NJ tree (1000 repli-
cations) were ascertained using MEGA 6.05 (Tamura
et al. 2013). Bird species with a high bootstrap value
(>95%) were grouped to estimate the phylogeographic
structure.
Results
Species identification in Japanese birds
Sequence data from the COI barcoding region were
obtained for 1367 specimens of 234 bird species from the
Fig. 1 Map of the Japanese Archipelago and surrounding regions, showing the location and the names of islands and areas. The
so-called Blakiston’s Line is a biogeographical boundary that separates the Japanese bird and mammal fauna between Hokkaido and
the other areas.
©2014 John Wiley & Sons Ltd
DNA BARCODING OF JAPANESE BIRDS 3
Japanese Archipelago. Of these species, 200 were repre-
sented by multiple specimens. The average K2P genetic
distances within species and genus were 0.49% (range
0–6.13) and 8.82% (0–16.39), respectively (Fig. 2). Aver-
age K2P intraspecific distance was 0.21% based on 200
species from multiple specimens (0–6.13%). Of the 234
species, 226 (96.6%) had unique DNA sequences that did
not overlap with any other species. Ninety-six percent-
age of the species were >1.0% divergent from their near-
est neighbour, and 92.7% were >1.5% divergent. The NJ
tree of K2P genetic distances formed 192 monophyletic
clusters (96% of the species represented by multiple
specimens) that were well supported by bootstrap values
(>95%). The bootstrap value of Erithacus (94%) was
slightly lower than the reliable bootstrap value (95%) for
phylogenetic structure.
Of the 234 bird species, five pairs of sister species
showed relatively low interspecific genetic variation
(<1.0%) (Table 1). Five pairs were exhibited on the NJ
tree with unreliable bootstrap values (30–72) or on non-
monophyletic clades. For example, the Pleske’s Warbler
Locustella pleskei and the Middendorff’s Warbler Locustela
ochotensis had bootstrap values of 55% and 65%, respec-
tively, and were unable to be identified by diagnostic
subsitutions due to small sample sizes (n=2 and 6,
respectively). Only one diagnostic substitution was
observed between the Izu Thrush Turdus celaenops and
the Brown-headed Thrush T. chrysolaus. Two Guillemot
species (Cepphus carbo and C. columba) could not be dis-
criminated based on the diagnostic sequences due to
small sample sizes (n=1 and 1, respectively). The Com-
mon and Oriental cuckoos were paraphyletic and could
not be perfectly discriminated. Some Spot-billed Ducks
Anas poecilorhyncha and Mallards A. platyrhynchos shared
the same sequences, although the majority of A. poe-
cilrhyncha had diverged with a cluster supported by 96%
bootstrap values (0.79% genetic distance). Thus, 226 Jap-
anese bird species (96.6%) were identified by their COI
barcoding regions using either a distance-based criterion
or a NJ tree with bootstrap values and diagnostic
sequences in cases of low interspecific genetic distances
(<1.0%).
Deep genetic divergence within the Japanese
Archipelago
Eleven bird species showed deep intraspecific diver-
gence (>1.6% K2P) and had two or more clusters sup-
ported by high bootstrap values (>95%) (Table 2). Nine
of 11 birds had three distinct geographical splits, separat-
ing the lineages within the Japanese Archipelago. The
Tsugaru Strait splits Hokkaido and Honshu in the south
and phylogeographically separates Garrulus glandarius,
Phylloscopus borealis and the Ural Owl Strix uralensis. The
Narcissus Flycatcher Ficedula narcissina and several
Brown-eared Bulbul Ixos amaurotis populations collected
from the Ryukyu Islands had different lineages from
those collected at Kyushu in the north. Erithacus komadori
split into two distinct lineages between the Okinawa
Islands and the Amami Islands in the north. Certain pop-
ulations of the Oriental Greenfinch Carduelis sinica,Sittip-
arus varius and the Scaly Thrush Zoothera dauma, which
occupy a specific island and its neighbouring small
islands, had sequences that had deeply diverged from
sequences from other areas, although the three species
are widely distributed throughout the Japanese Archi-
pelago. Distinct lineages were found for C. sinica on the
Ogasawara Islands, S. varius on the Iriomote Island and
Z. dauma on the Amami-Oshima island. No evidence of
a geographical split was detected between the two dis-
tinct lineages of both the Oriental Turtle-Dove Streptop-
elia orientalis and the Ryukyu Scops-Owl Otus elegans.
The Japanese Scops-Owl O. semitorques formed two clus-
ters, one of which could reliably be discerned as originat-
ing from the Ryukyu Islands (>95% bootstrap value),
and the other, with a moderate cluster value (87%), from
Honshu, despite having relatively low maximum intra-
specific distances (0.9%).
(a)
(b)
Fig. 2 Frequency histogram of COI among Japanese-breeding
birds showing (a) the distribution of the average distance within
species (200 species) and (b) the average congeneric distance (47
genus).
©2014 John Wiley & Sons Ltd
4T. SAITOH ET AL.
Comparison of the Eurasian and Japanese birds
COI barcoding sequences of 1437 individuals were col-
lected from 142 species that represented multiple speci-
mens from the Japanese Archipelago and the eastern
Eurasian continent (Appendix S2, Supporting informa-
tion). The average intraspecific genetic distance (K2P)
was 0.46% (0–6.13%). The NJ tree of K2P genetic dis-
tances described 139 monophyletically distinct species
supported by high bootstrap values (>95%). Cuculus opta-
tus and C. canorus were not reciprocally monophyletic.
Motacilla flava was part of a paraphyletic relationship
between a sample from western Eurasia and samples
from Japan and eastern Eurasia.
Forty-one of the 142 species had two or more distinct
clusters supported by high bootstrap values (>95%). Of
these 41 species, 23 displayed phylogeographic patterns
that suggested genetic splits between the Eurasian conti-
nent and the Japanese Archipelago (Table 3). Of these
23 species, 19 had two or more clusters differing by at
least 1.6% K2P. Some species formed clades that
included multiple samples derived from Japanese birds,
and one sample from Eurasia (the Eurasian Skylark
Alauda arvensis, the Carrion Crow Corvus corone, the
Large-billed Crow C. macrorhynchos, the Grey-faced
Woodpeeker Picus canus and the Eurasian Magpie Pica
pica), for which it was not possible to ascertain any geo-
graphical splits between Eurasia and the Japanese
Archipelago. In addition to those mentioned above,
three species formed Japanese-specific monophyletic
clusters with moderate to relatively high bootstrap sup-
port (77–94%). A clade for the Siberian Blue Robin Lusci-
nia cyane, which included specimens form Hokkaido
and the South Kuril Islands (77% bootstrap value), was
separated from another clade from Eurasia and Honshu
(91%), differing by 1.06%. A clade for the Asian Brown
Flycatcher Muscicapa dauurica from the Japanese Archi-
pelago (94%) differed by 0.88% from a Eurasian conti-
nent clade (88%). The Rock Ptarmigan Lagopus muta
formed a Japan-specific monophyletic clade (82%) that
differed by 0.3% from another monophyletic clade
formed of Eurasian samples (68%).
Table 1 Japanese bird species with small (<1.0%) interspecific genetic COI sequence distances
Order Common name Scientific name nGenetic distance (%)
1 Anseriformes Mallard Anas platyrhynchos 40
Spot-billed Duck Anas poecilorhyncha 7
2 Passeriformes Izu Thrush Turdus celaenops 4 0.15
Brown-headed Thrush Turdus chrysolaus 12
3 Cuculiformes Common Cuckoo Cuculus canorus 5 0.3
Oriental Cuckoo Cuculus optatus 6
4 Passeriformes Middendorff’s Warbler Locustella ochotensis 6 0.63
Pleske’s Warbler Locustella pleskei 2
5 Charadriiformes Spectacled Guillemot Cepphus carbo 1 0.85
Pigeon Guillemot Cepphus columba 1
Table 2 Japanese bird species with large (>1.6%) intraspecific COI divergence, treated as candidates of cryptic species
Common name Scientific name nBootstrap
Max
intraspecific
distance (%) Collection area* Genetic break
1 Ryukyu Robin Erithacus komadori 3/4 100/100 6.13 O/A, Yk Ryukyu Is.
2 Arctic Warbler Phylloscopus borealis 1/4/3 –/89/100 5.06 Hn/H/Hn Tsugaru Strit
3 Eurasian Jay Garrulus glandarius 4/3 100/100 4.43 H/Hn Tsugaru Strit
4 Scaly Thrush Zoothera dauma 2/18 100/99 3.73 A/Y, Og, I, Hn, H Isolated Islands
5 Narcissus Flycatcher Ficedula narcissina 7/10 99/100 3.71 Y, O, A/Hn, H Ryukyu Is.
6 Brown-eared Bulbul Ixos amaurotis 1/10/33 1/99/99 3.57 Mi/M, O, A/Y, Og, I, Hn, H Ryukyu Is.
7 Oriental Greenfinch Carduelis sinica 9/1 99/–3.37 Tk, I, Hn, H/Og Isolated Islands
8 Ryukyu Scops-Owl Otus elegans 7/6 100/100 2.90 Y, O/O, A, T Sympatric
9 Varied Tit Sittiparus varius 4/18 100/100 2.82 Y/O, A, T, Yk, I, Hn, H Isolated Islands
10 Ural Owl Strix uralensis 8/13 100/99 2.81 H/Hn Tsugaru Strit
11 Oriental Turtle-Dove Streptopelia orientalis 3/5 100/99 2.42 Hn, H/I, Hn, Y Sympatric
*A =Amami Is., H =Hokkaido, Hn =Honshu (including Kyushu and Shikoku), I =Izu Is., M =Miyakojima I., Mi =Minami-Daitoj-
ima I., O =Okinawa Is., Og =Ogasawara Is., T =Tokara Is., Y =Yaeyama Is., and Yk =Yakushima I.
©2014 John Wiley & Sons Ltd
DNA BARCODING OF JAPANESE BIRDS 5
Discussion
The COI barcoding region enabled identification of 226
species from 234 Japanese-breeding birds, with distance
threshold criteria, a NJ tree with high bootstrap values
and diagnostic substitution analyses. The average conge-
neric difference (8.8%) was 18-fold that of the average
conspecific genetic difference (0.49%) (Fig. 2). The COI
barcoding region has a high resolution for identification
of Japanese bird species, making it an effective tool for
the identification of bird species inhabiting insular
regions, such as the Japanese Archipelago. DNA bar-
codes have identified 24 cryptic species candidates in the
Japanese Archipelago, suggesting that the sea and the
straits around the Japanese Archipelago act as effective
genetic barriers, despite their relatively small area.
There were five pairs of nearest-neighbour species
that had small interspecific genetic differences (<1.0%) in
the COI barcoding region (Table 1). Identical sequences
were shared between Anas platyrhynchos and Anas poe-
cilorhyncha, which has also been found for other geo-
graphical regions in previous studies (Yoo et al. 2006;
Park et al. 2011). Anas platyrhynchos is closest to A. poe-
cilorhyncha based on cytochrome band ND2 (Johnson &
Sorenson 1999), as well as on COI; thus, further taxo-
nomic re-evaluation of these species should be per-
formed. Two scenarios were proposed for the sequence
similarities of these species. An ancestral duck species
could have split relatively recently into the two species
of duck. Alternatively, the two species of duck may have
hybridized in the Russian Far East, resulting in their
sharing of the barcoding region (Kulikova et al. 2004).
Cuculus canorus and C. optatus also had a very low inter-
specific distance (0.3%) and formed paraphyletic clades;
this pair was previously grouped together as a single
clade (Kerr et al. 2009b), but hybrids have not been docu-
mented to date (Sorenson & Payne 2005). The interspe-
cific genetic difference between Cepphus carbo and
C. columba was 0.85%, which is similar to results based
on the cytochrome bgene (1.4%) (Friesen et al. 1996).
Only a very small genetic distance (0.15%) separated
T.chrysolaus, a migrant breeder in northern Japan, and
Turdus celaenops, a resident bird living in several islands
off the main islands of Japan (Ornithological Society of
Japan 2012), with only a single substitution enabling the
species to be individually identified. In addition to those
mentioned above, the sequence of the Slaty-backed Gull
Larus schistisagus that breeds in Japan completely
matched those of other gulls recorded in previous stud-
ies (Yoo et al. 2006; Kerr et al. 2009b; Lijtmaer et al. 2011)
that do not breed in Japan, but migrate to Japan as winter
visitors (Ornithological Society of Japan 2012). The COI
barcoding region could not distinguish between L. schis-
tisagus breeding in Japan and the winter visitor gulls.
We found 11 species with deep intraspecific diver-
gences (>1.6%) (Table 2) that were supported by high
bootstrap values (>95%). The patterns of these deep
intraspecific divergences were consistent with some of
the biogeographical boundaries across the Japanese
Archipelago, and these were classified into three groups
according to the sea barrier: (i) the Tsugaru Strait, (ii) the
Ryukyu Islands and (iii) other specific islands. Streptop-
elia orientalis and O. elegans, however, exhibited an
intricate pattern of haplotypic distribution. Although
haplotypes of the dove were clearly split into two clades
with high bootstrap values, no geographical structure
was observed across the Japanese Archipelago, including
the Ryukyu Islands. Otus elegans only partially separated
into two clades between Okinawa and Amami-Oshima
because two Okinawa birds belonged to the Amami-
Oshima clade. Takagi (2013) reported that the owl had a
vocal divergence among Amami-Oshima, Okinawa
Islands and Yaeyama Islands. The Okinawa population
could be an interbred population between the two
clades. Sympatric divergence of O. elegans and S.orien-
talis would reflect the admixture of multiple lineages iso-
lated over a geographical timescale (Webb et al. 2011;
Hogner et al. 2012).
The Tsugaru Strait lies between Hokkaido and
Honshu and divides the lineages of S. uralensis,Garrulus
glandarius and Phylloscopus borealis into two clades
(Table 2). In recent studies, P. borealis has been split into
two species, the Kamchatka Leaf Warbler Phylloscopus
examinandus on Hokkaido and the Japanese Leaf Warbler
Phylloscopus xanthodryas on Honshu, based on morpho-
logical, behavioural and genetic assessments (Saitoh et al.
2010; Alstr€
om et al. 2011). Strix uralensis and G. glandarius
have also been split in different subspecies in the Tsug-
aru Strait (Ornithological Society of Japan 2012). The
Tsugaru Strait (so-called Blakiston’s Line) represents one
of the biogeographical boundaries that separate Japanese
avifauna south of Honshu from typical Eurasian and
Hokkaido avifauna (Blakiston 1883).
The Ryukyu Islands have also formed boundaries that
separate some bird species and have led to deep genetic
divergences. Ficedula narcissina is classified into two sub-
species according to the breeding populations on the
main islands of Japan and the Ryukyu Islands (Ornitho-
logical Society of Japan 2012). In the Narcissus Fly-
catcher, the narcissina and owstoni subspecies are
sometimes treated as separate species, with F. narcissina
present on the main islands of Japan and F. owstoni on
the Ryukyu Islands (Brazil 2009). The I. amaurotis popu-
lation from middle Ryukyu (Okinawa and Amami
Islands) was formed as a single unique clade, supported
by high bootstrap values, but individuals on the south
Ryukyu (the Yaeyama and Miyakojima islands) belong
to the same clade as those on the Japanese main islands.
©2014 John Wiley & Sons Ltd
6T. SAITOH ET AL.
Table 3 Mean genetic distance and distinguishable lineages for birds that breed in the Palearctic region including the Japanese Archipelago. We listed the bird species with robust
divergent clusters with high bootstrap values (>95%). Superscripts show groups arranged in the NJ tree
Species nBootstrap Mean K2P (%)*
Max K2P (%)
Eurasia vs.
Japan Collection area†
Japan-specific
lineages‡
Genetic
split within
Japan
Cyptic species
in relation to
Japan
1Alauda arvensis 1
a
/6
b
/10
c
–
a
/99
b
/99
c
7.95
a-b
, 4.80
a-c
8.38 E
a
/E, S
b
/S, H, Hn
c
Yes
2Delichon dasypus 1
a
/1
b
/9
c
–
a
/–
b
/99
c
6.40
a-b
, 3.85
a-c
6.4 E
a
/S
b
/Hn, H
c
Yes Yes
3Caprimulgus indicus 4/5 99/99 5.9 6.04 E/H, Hn Yes Yes
4Phylloscopus borealis 8
a
/9
b
/4
c
/1
d
99
a
/99
b
/90
c
/–
d
3.60
a-b
, 1.78
a-c
, 3.57
a-d
5.15 E
a
/S, H
b
/Hn
c
/Hn
d
Yes Yes Yes§
5Pica pica 9/5 99/99 3.8 4.73 E/E, H Yes
6Garrulus glandarius 3
a
/12
b
/3
c
99
a
/99
b
/99
c
2.72
a-b
, 3.59
a-c
, 4.32
b-c
4.53 E
a
/E, H
b
/Hn
c
Yes Yes Yes§
7Emberiza spodocephala 8/17 99/99 3.55 3.95 E/Hn, H, S, Ss Yes Yes
8Zoothera dauma 2/25 99/99 3.63 3.73 A/E, Y, Og, I, Hn, H Yes Yes Yes§
9Ficedula narcissina 7/12 99/99 2.88 3.71 Y, O, A/Hn, H, S Yes Yes Yes§
10 Ixos amaurotis 10
a
/37
b
/1
c
99
a
/99
b
/–
c
2.53
a-b
, 3.19
a-c
3.57 E, Y, Mi, Og, I, Hn,
H
a
/O, A
b
/Mi
c
Yes Yes Yes§
11 Cettia diphone 2/35 96/99 2.93 3.43 E/Y, O, A, Og, I, Hn,
H, S, Ss
Yes Yes
12 Muscicapa sibirica 6/3 99/99 2.66 2.92 E/Y, Hn, S Yes Yes
13 Dendrocopos major 4/11 99/99 2.69 2.82 E/Hn, H, S Yes Yes
14 Strix uralensis 10/13 99/99 2.81 2.81 E, H/Hn Yes Yes Yes§
15 Corvus corone 7/7 85/98 2.12 2.45 E/E, Hn, H Yes
16 Urosphena squameiceps 4/5 99/99 2.1 2.21 E/Hn, H, S Yes Yes
17 Corvus macrorhynchos 2/23 99/78 1.55 2.13 E/E, Y, O, A, Yk, Hn, H, S Yes
18 Lanius cristatus 3/1 99/–1.91 2.01 E/Y Yes Yes
19 Uragus sibiricus 5/12 97/99 1.49 1.82 E/Hn, H, S, Ss Yes Yes
20 Picus canus 7/14 99/88 1.25 1.44 E/E, H
21 Luscinia calliope 6/9 98/68 1.03 1.34 E/Hn, H Yes
22 Tachybaptus ruficollis 1/4 –/96 1.05 1.05 E/Y, M, A, Hn Yes
23 Mergus merganser 1/2 –/98 0.59 0.59 E/Hn Yes
*Mean population genetic distance (K2P) between clades.
†E=Eurasian continent, S =Sakhalin, Ss =South Kril Is., Y =Yaeyama Is., O =Okinawa Is., M =Miyakojima I., Mi =Minami-Daitojima, A =Amami Is., Yk =Yakushima I.,
Og =Ogasawara Is., I =Izu Is., H =Hokkaido, and Hn =Honshu (including Kyushu and Shikoku).
‡Including the area of Sakhalin and S Kuril Is.
§Duplication of cryptic species candidates in Table 2.
©2014 John Wiley & Sons Ltd
DNA BARCODING OF JAPANESE BIRDS 7
Hamao et al. (2013) also reported the complex genetic
structure of the bulbuls among the Ryukyu Islands. Cli-
mate change has caused repeated emergences and disap-
pearances of land-bridges between the Ryukyu Islands,
temporarily connecting some of the islands, and even the
continent. These islands biogeographically form the
boundary between the Palearctic and the Oriental
regions.
Four species showed large genetic differences on
other islands (Table 2). For example, the sequence of
C. sinica on the Ogasawara Islands was clearly separated
from a clade clustered by other subspecies of C. sinica on
the main islands of Japan and the continental regions.
The Ogasawara Islands are oceanic islands located
1000 km south of the main islands of Japan. They contain
some endemic bird species and subspecies, including
some species which are now extinct (Ornithological Soci-
ety of Japan 2012). Sittiparus varius shows deep genetic
divergence between Iriomote Island and the other
islands. The Iriomote population belongs to the same
clade as the Taiwanese population (McKay et al. 2014).
On Amami-Oshima, Z. dauma forms a separable clade to
that on the Japanese islands. The Amami-Oshima Z. dau-
ma is often treated as a separate species of Zoothera major,
although this was not supported by genetic evidence
(Dickinson 2003; Brazil 2009). Erithacus komadori is split
into two clades with deep intraspecific divergence
(6.13%): one clade, E. k. namiyei, originates from Oki-
nawa, and the other clade, E. k. komadori, comes from
Amami-Oshima and Yakushima Islands. Our findings
closely resemble those of Seki (2006) and Seki et al.
(2007). These deep divergences between the middle/
southern Ryukyu Islands and the main islands of Japan
may have occurred during the Pliocene, when the mid-
dle Ryukyu Islands finally separated from the main
island (Kizaki & Oshiro 1980).
In total, 142 of 398 bird species that breed in the Japa-
nese Archipelago and the eastern Eurasian Continent
were analysed to reveal the genetic structure of trans-
Japanese and Eastern Eurasian birds (Appendix S2, Sup-
porting information). The average intraspecific distance
was 0.46% (0–3.32%). Ten percentage of 234 species
exhibit deep intraspecific divergence in relation to the
Japanese Archipelago. This percentage is higher than for
species in the Nearctic (2%) (Hebert et al. 2004; Kerr et al.
2007), Korea (1%) (Yoo et al. 2006), Scandinavia (1%)
(Johnsen et al. 2010), and approximately equal to the
Palearctic (10%) (Kerr et al. 2009b). Geographical
barriers, such as mountains and oceans, have divided
populations and prevented gene flow. For example, the
birds inhabiting both regions of Scandinavia and the
Nearctic had deep intraspecific divergence in 24% of spe-
cies (Kerr et al. 2007) and the birds of the Nearctic and
Argentina in 24% (Kerr et al. 2009a). Relatively high lev-
els of divergence were reported between Palearctic and
Nearctic populations of some Holarctic birds (e.g. Dro-
vetski et al. 2004; Koopman et al. 2005; Zink et al. 2006).
The avifauna of the Japanese Archipelago includes many
possible cryptic species, despite the differences in size
and geographical history of the region.
Japan has a relatively high proportion of cryptic bird
species compared with South Korea and Scandinavia,
which are similar in size to Japan, and a similar propor-
tion as Eurasia, which is geographically much larger. Sea
barriers would have led to the isolation of bird popula-
tions around Japan. The repeated southward spread of
the ice sheets to 40°N in North America during the qua-
ternary (Hewitt 2004) would have caused shallow
genetic divergence (Ball & Avise 1992; Seutin et al. 1995;
Zink 1996; Weir & Schluter 2004) and low genetic diver-
sity (Hewitt 1996; Soltis et al. 1997; Conroy & Cook 2000)
among the various taxa in the northern regions, while
the ice sheets that were limited to 50°N in the east Pale-
arctic region allowed bird populations to speciate. The
southward spread of ice did not affect the Japanese
Archipelago equally, and some areas would have been
linked to the Eurasian Continental by land-bridges. The
fluctuating sea levels during the quaternary would have
led to repeated periods of connection, isolation and sub-
mergence of the islands of the Japanese Archipelago,
which could have isolated bird populations for lengthy
periods. Such populations would have retained geo-
graphical and temporal isolation without admixture of
other populations and extinction for glacial and intergla-
cial periods. These isolated populations may lead to the
deep intraspecific genetic divergence observed in 11 Jap-
anese bird species. The distinct intraspecific divergence
observed between 23 trans-Palearctic species may reflect
the history of isolation and gene flow of these bird popu-
lations with respect to the changes in the land-bridges
over the straits between the Japanese Archipelago and
the Eurasian continent.
In this study, we constructed a barcoding library com-
prising 93.2% of Japanese-breeding bird species and
demonstrated an effective tool for the identification of
these bird species. The relatively deep genetic divergence
was related to the periodic occurrence and disappear-
ance of sea barriers. We found 24 cryptic species candi-
dates in this study, which suggests that traditional
taxonomic methods for identification of East Asian birds
do not fully reflect the divergences. For example, P. bore-
alis has been split into three species in recent studies
(Saitoh et al. 2010; Alstr€
om et al. 2011). Our study con-
tributes additional Japanese bird data to the global DNA
barcoding library and provides valuable data for future
investigation of the taxonomy of East Asian birds.
©2014 John Wiley & Sons Ltd
8T. SAITOH ET AL.
Acknowledgements
We thank the following institutions and people for providing
specimens: the Higashi Taisetsu Museum of Natural History;
the Botanical Garden, Field Science Centre for Northern Bio-
sphere, Hokkaido University; the Hokkaido Seabird Centre, the
Kushiro-Shitsugen Wildlife Centre, the Iriomote Wildlife Centre,
the Amami Wildlife Centre, the Yambaru Wildlife Conservation
Centre and Raptor Conservation Centre, the Ministry of the
Environment, Japan; Miyakojima City Museum; the Hokkaido
Research Centre, the Forestry and Forest Products Research
Institute; Okhotsk Chapter and Sapporo Chapter, the Wild Bird
Society of Japan; Shunkunitai Wild Bird Sanctuary; the Lake
Utonai Wildlife Centre; the Lake Shikaribetsu Nature Centre;
the Akkeshi Waterfowl Observation Centre; the Institute for
Raptor Biomedicine Japan; the Ushiku Nature Sanctuary; the
Ibaraki Nature Museum; the Gyotoku Wild Bird Observation
Centre; the Hachijo visitor’s centre; the Miyake Nature Centre;
the Yonagunijima Ayamihabiru Museum; the Graduate School
of Science, Osaka City University; the Graduate School of Fisher-
ies Sciences, Hokkaido University; Hiroshima City Asa Zoologi-
cal Park; the Ehime Prefectural Science Museum; Sanbonmatsu
High School; Crane Park Izimi; Amami Ornithologists’ Club; the
Ikeya Animal Hospital; and the Takizawa Animal Hospital. We
also thank Dr. Takeshi Yamasaki and Dr. Sayaka Mori for help-
ing us to obtain specimen records. This study was funded by the
research project of the National Museum of Nature and Science,
‘DNA barcoding of birds from Japan and East Asia’, JSPS
KAKENHI Grant Numbers 21370039 and 24657066, and a Grant-
in-Aid for Scientific Research (Specially Designated Research
Promotion), ‘Development and disclosure of the Yamashina Insti-
tute for Ornithology database system (2009–)’ from the Ministry
of Education, Culture, Sports, Science & Technology in Japan.
References
Akimova A, Haring E, Kryukov S, Kryukov A (2007) First insights into a
DNA sequence based phylogeny of the Eurasian Jay Garrulus glandari-
us.Russian Journal of Ornithology,356, 567–575.
Alstr€
om P, Saitoh T, Williams D et al. (2011) The Arctic Warbler Phyllosc-
opus borealis –three anciently separated cryptic species revealed. Ibis,
153, 395–410.
Ball RM, Avise JC (1992) Mitochondrial DNA phylogeographic differenti-
ation among avian populations and the evolutionary significance of
subspecies. Auk,109, 626–639.
Blakiston TW (1883) Zoological indications of ancient connection to the
Japan islands with the continent. Transactions of the Asiatic Society of
Japan,11, 126–140.
Brazil M (2009) Field Guide to the Birds of East Asia. Christopher Helm,
London.
Clare EL, Lim BK, Engstrom MD et al. (2007) DNA barcoding of Neotrop-
ical bats: species identification and discovery within Guyana. Molecular
Ecology Notes,7, 184–190.
Clements JF (2007) The Clements Checklist of Birds of the World, 6th edn.
Christopher Helm, London.
Conroy CJ, Cook JA (2000) Phylogeography of a post-glacial colonizer:
Microtus longicaudus (Rodentia: Muridae). Molecular Ecology,9, 165–
175.
Conservation International (2012) The biodiversity hotspots. Available at:
http://www.conservation.org/where/priority_areas/hotspots/
Pages/hotspots_main.aspx [Accessed 10 December 2012].
Dickinson EC (Ed.) (2003) The Howard and Moore Complete Checklist of the
Birds of the World, 3rd edn. Christopher Helm, London.
Drovetski S, Zink RM, Rohwer S et al. (2004) Complex biogeographic his-
tory of a Holarctic passerine. Proceedings of the Royal Society London Ser-
ies B,271, 545–551.
Friesen VL, Baker AJ, Piatt JF (1996) Phylogenetic relationships within
the Alcidae (Charadriiformes: Aves) inferred from total molecular evi-
dence. Molecular Biology and Evolution,13, 359–367.
Hamao S, Sugita N, Nishiumi I (2013) Geographical variation in mito-
chondrial DNA and vocalizations in two resident bird species in the
Ryukyu Archipelago, Japan. Bulletin of the National Museum of Nature
and Science, Series A (Zoology),39,51–62.
Hebert PDN, Ratnasingham S, de Waard JR (2003) Barcoding animal life:
cytochrome coxidase subunit 1 divergences among closely related spe-
cies. Proceedings of the Royal Society London Series B,270, 596–599.
Hebert PDN, Stoeckle MY, Zemlark TS, Francis CM (2004) Identification
of birds through DNA barcodes. PLoS Biology,2, e312.
Hern
andez-D
avila A, Vargas JA, Mart
ınez-m
endez N et al. (2012) DNA
barcoding and genetic diversity of phyllostomid bats from the Yucatan
Peninsula with comparisons to Central America. Molecular Ecology
Resource,12, 590–597.
Hewitt GM (1996) Some genetic consequences of ice ages, and their role
in divergence and speciation. Biological Journal of the Linnean Society,
58, 247–276.
Hewitt GM (2004) Genetic consequences of climatic oscillations in the
Quaternary. Philosophical Transactions of the Royal Society B: Biological
Sciences,359, 183–195.
Hogner S, Laskemoen T, Lifjeld JT et al. (2012) Deep sympatric mitochon-
drial divergence without reproductive isolation in the common red-
start Phoenicurus phoenicurus.Ecology and Evolution,2, 2974–2988.
Holt BG, Lessard JP, Borregaard MK et al. (2013) An update of Wallace’s
zoogeographic regions of the world. Science,339,74–78.
Johnsen A, Rindal E, Ericson PGP et al. (2010) DNA barcoding of Scandi-
navian birds reveals divergent lineages in trans-Atlantic species. Jour-
nal of Ornithology,151, 565–578.
Johnson K, Sorenson MD (1999) Phylogeny and biogeography of dab-
bling ducks (genus: Anas): a comparison of molecular and morphologi-
cal evidence. Auk,116, 792–805.
Kerr KCR, Stoeckle MY, Dove CJ et al. (2007) Comprehensive DNA bar-
code coverage of North American birds. Molecular Ecology Notes,7,
533–543.
Kerr KCR, Lijtmaer DA, Barreira AS et al. (2009a) Probing evolutionary
patterns in Neotropical birds through DNA barcodes. PLoS ONE,4,
e4379.
Kerr KCR, Birks SM, Kalyakin MV et al. (2009b) Filling the gap –COI bar-
code resolution in eastern Palearctic birds. Frontiers in Zoology,6, 29.
Kizaki K, Oshiro I (1980) The origin of the Ryukyu Islands. In: Natural
History of the Ryukyus (ed. Kizaki K), pp. 8–37. Tsukiji-shokan, Tokyo,
Japan (In Japanese).
Koopman ME, McDonald DB, Hayward GD et al. (2005) Genetic similar-
ity among Eurasian subspecies of boreal owls Aegolius funereus.Journal
of Avian Biology,36, 179–183.
Kulikova IV, Zhuravlev YN, McCracken KG (2004) Asymmetric hybrid-
ization and sex-biased gene flow between Eastern spot-billed ducks
(Anas zonorhyncha) and Mallards (Anas platyrhynchos) in the Russian
Far East. Auk,121, 930–949.
Lara A, Ponce de Le
on JL, Rodr
ıguez R et al. (2010) DNA barcoding of
Cuban freshwater fishes: evidence for cryptic species and taxonomic
conflicts. Molecular Ecology Resources,10, 421–430.
Lijtmaer DA, Kerr KCR, Barreira AS et al. (2011) DNA barcode libraries
provide insight into continental patterns of avian diversification. PLoS
ONE,6, e20744.
Lohman DJ, Ingram KK, Prawiradilaga DM et al. (2010) Cryptic genetic
diversity in ‘widespread’ Southeast Asian bird species suggests that
Philippine avian endemism is gravely underestimated. Biological Con-
servation,143, 1885–1890.
McKay BD, Mays HL, Yao C et al. (2014) Incorporating color into integra-
tive taxonomy: analysis of the varied tit (Sittiparus varius) complex in
East Asia. Systematic Biology, doi: 10.1093/sysbio/syu016. [Epub ahead
of print].
©2014 John Wiley & Sons Ltd
DNA BARCODING OF JAPANESE BIRDS 9
Mil
a B, Tavares ES, Mu~
noz Salda~
na A et al. (2012) A trans-Amazonian
screening of mtDNA reveals deep intraspecific divergence in forest
birds and suggests a vast underestimation of species diversity. PLoS
ONE,7, e40541.
Nishiumi I, Yao C, Saito DS, Lin R (2006) Influence of the last two glacial
periods and late Pliocene on the latitudinal population structure of res-
ident songbirds in the Far East. Memoirs of the National Science Museum
Tokyo,44,11–20.
Ornithological Society of Japan (2012) Check-list of Japanese Birds, 7th
revised edn. The Ornithological Society of Japan, Sanda.
Park HY, Yoo HS, Jung G, Kim CB (2011) New DNA barcodes for identi-
fication of Korean birds. Genes & Genomics,33,91–95.
Ratnasingham S, Hebert PDN (2007) BOLD: The Barcode of Life Data
System (http://www.barcodinglife.org). Molecular Ecology Notes,7,
355–364.
Saitoh T, Alstr€
om P, Nishiumi I et al. (2010) Old divergences in a boreal
bird supports long-term survival through the Ice Ages. BMC Evolution-
ary Biology,10, 35.
Seki SI (2006) The origin of the East Asian Erithacus robin, Erithacus koma-
dori, inferred from cytochrome bsequence data. Molecular Phylogenetics
and Evolution,39, 899–905.
Seki SI, Sakanashi M, Kawaji N, Kotaka N (2007) Phylogeography of the
Ryukyu robin (Erithacus komadori): population subdivision in land-
bridge islands in relation to the shift in migratory habit. Molecular Ecol-
ogy,16, 101–113.
Seutin G, Ratcliffe LM, Boag PT (1995) Mitochondrial DNA homogeneity
in the phenotypically diverse redpoll finch complex (Aves: Cardueli-
nae: Carduelis flammea–hornemanni). Evolution,49, 962–973.
Soltis DE, Gitzendanner MA, Strenge DD, Soltis PS (1997) Chloroplast
DNA intraspecific phylogeography of plants from the Pacific North-
west of North America. Plant Systematics and Evolution,206,
353–373.
Song H, Buhay JE, Whiting MF, Crandall KA (2008) Many species in
one: DNA barcoding overestimates the number of species when
nuclear mitochondrial pseudogenes are coamplified. Proceedings of
the National Academy of Sciences of the United States of America,105,
13486–13491.
Sorenson MD, Payne RB (2005) A molecular genetic analysis of cuckoo
phylogeny. In: The Cuckoos (Bird Families of the World) (ed. Payne RB),
pp. 68–94. Oxford University Press, New York City, New York.
Takagi M (2013) A typological analysis of the hoot of the Ryukyu Scops
Owl across island populations in the Ryukyu Archipelago and two
oceanic islands. The Wilson Journal of Ornithology,125, 358–369.
Tamura K, Stecher G, Peterson D, Filipski A, Kumar S (2013) MEGA6:
molecular evolutionary genetics analysis version 6.0. Molecular Biology
and Evolution,30, 2725–2729.
Tavares ES, Gonc
ßalves P, Miyaki CY, Baker AJ (2011) DNA barcode
detects high genetic structure within Neotropical bird species. PLoS
ONE,6, e28543.
Ward RD, Zemlak TS, Innes BH et al. (2005) Barcoding Australia’s fish
species. Philosophical Transactions of the Royal Society B,360, 1847–1857.
Webb WC, Marzluff JM, Omland KE (2011) Random interbreeding
between cryptic lineages of the Common Raven: evidence for specia-
tion in reverse. Molecular Ecology,20, 2390–2402.
Weir JT, Schluter D (2004) Ice sheets promote speciation in boreal birds.
Proceedings of the Royal Society London Series B: Biological Sciences,271,
1881–1887.
Yoo HS, Eah J-Y, Kim JS (2006) DNA barcoding Korean Birds. Molecules
and Cells,22, 323–327.
Zink RM (1996) Comparative phylogeography in North American birds.
Evolution,50, 308–317.
Zink RM, Drovetski SV, Rohwer S (2006) Selective neutrality of mito-
chondrial ND2 sequences, phylogeography and species limits in Sitta
europaea.Molecular Phylogenetics and Evolution,40, 679–686.
T.S. collected the specimens, analysed the data and pre-
pared the draft. N.S. provided the sequences, analysed
the data and wrote the manuscript. S.S. performed the
genetic analysis. Y.I. and S.K. collected the specimens.
H.K., A.H., S.A. and Y.Y. provided many of the
sequences and organized the sequence data. I.N.
designed and coordinated the project.
Data accessibility
DNA sequences: DDBJ accessions AB842491–AB843857.
Sequence alignment and final neighbour-joining tree
uploaded as Online Supporting Information. BOLD project
BJNSM and YIO, accessions BJNSM001-08–BJNSM842-12
and YIO001-08–YIO558-12, respectively.
Supporting Information
Additional Supporting Information may be found in the online
version of this article:
Appendix S1 List of Japanese-breeding bird species analysed in
this study.
Appendix S2 List of the bird species analysed in this study
(breeding in both Eurasia and the Japanese Archipelago).
Appendix S3 A K2P neighbour-joining tree of COI made with
1367 voucher specimens from 234 species breeding in the
Japanese Archipelago.
©2014 John Wiley & Sons Ltd
10 T. SAITOH ET AL.
