J Ornithol (2007) 148:417–426

DOI 10.1007/s10336-007-0150-4

ORIGINAL ARTICLE

Molecular phylogeny of the yuhinas (Sylviidae: Yuhina):

a paraphyletic group of babblers including Zosterops
and Philippine Stachyris
Shuxia Zhang Æ Lan Yang Æ Xiaojun Yang Æ
Junxing Yang

Received: 11 September 2006 / Revised: 2 February 2007 / Accepted: 21 March 2007 / Published online: 19 June 2007

Dt. Ornithologen-Gesellschaft e.V. 2007

Abstract Mitochondrial sequences (2,379 bp) from also implies that an ancestor of Zosterops/Philippine
cytochrome b, ND3, 12s and 16s rRNA were analyzed in Stachyris derived the ability to disperse over long distances,
order to reconstruct the phylogenetic relationships within so that it could fly over the sea and arrive at the Philippines.
the yuhinas (Yuhina), including the chestnut-faced babbler
Stachyris whiteheadi which is endemic to the Philippines, Keywords Phylogenetic analysis
Mitochondrial DNA

the Japanese white-eye Zosterops japonicus, the chestnut- Yuhina
Zosterops
Stachyris
flanked white-eye Z. erythropleurus, and the oriental white-
eye Z. palpebrosus. The results showed strong support for
the idea that S.whiteheadi and three white-eye species form Introduction
a clade embedded within the Yuhina clade. The robustness
of the Yuhina/Philippine Stachyris/Zosterops clade rejected Babblers (Timaliini) comprise more than 200 highly mor-
the validity of the monotypic genus Staphida, which was phologically diversified species in South-East Asia and
proposed for the striated yuhina Yuhina castaniceps based Africa. Although both morphological and molecular data
on its peculiar morphological traits. Sister group relation- have been used to assess the systematic relationships among
ships between the striped-throated yuhina Yuhina gularis major groups of Timaliini (Delacour 1946, 1950; Cibois
and the rufous-vented yuhina Yuhina occipitalis and be- et al. 1999, 2002; Cibois 2003), phylogenetic relationships
tween the whiskered yuhina Yuhina flavicollis and the of the babblers within genus are still poorly understood.
white-naped yuhina Yuhina bakeri were discovered. The Here we used mitochondrial DNA sequence data to re-
sympatric patterns of the above two sister groups in evaluate the phylogenetic relationships within a group of
the Himalayas is most likely due to secondary contact. The Asia endemic birds, the yuhinas (Yuhina). The genus Yuh-
molecular phylogeny also suggests that crestlessness was ina includes eleven species that are distributed in the
derived just once for the Zosterops/Philippine Stachyris Himalayas, Indochina and the Sunda region (Sibley and
clade. The Zosterops/Philippine Stachyris clade in our study Monroe 1990). Yuhinas inhabit broad-leaved, coniferous or
bamboo forests. Their distributions range in altitude from
several hundred meters to 4,000 m high (Sibley and Monroe
Communicated by M. Wink. 1990). Yuhinas are omnivorous, feeding on insects, flower
honey, small berries and seeds (Ali and Ripley 1987).
S. Zhang
L. Yang (&)
X. Yang
J. Yang Recent molecular data had suggested that the genus
Kunming Institute of Zoology,
Yuhina is not a monophyletic group; three species of
The Chinese Academy of Sciences,
32 jiaochang donglu, Kunming, Stachyris endemic to the Philippines are embedded within
Yunnan Province 650223, China the yuhina clade (S. whiteheadi, S. striata, S. dennistouni),
e-mail: [email protected] and are not related to other continental species of Stachyris
(Cibois et al. 2002). Furthermore, the white-bellied yuhina
S. Zhang
Graduate School, Chinese Academy of Sciences, Yuhina zantholeuca, which has traditionally been placed
Beijing 100039, China among the yuhinas (Deignan 1964; Sibley and Monroe

123
418 J Ornithol (2007) 148:417–426

1990; Inskipp et al. 1996; Clements 2000), is not even a samples, DNA was isolated by the standard phenol/chlo-
babbler (Cibois et al. 2002). Instead, it is closely related to roform extraction method (Sambrook et al. 1989); for toe
the Vireonidae (Barker et al. 2004). Molecular data have pad samples, Puregene DNA purification kits (Gentra
also revealed that the Japanese white-eye, Zosterops Systems, Minneapolis, MN, USA) was used; the manu-
japonicus, appears to be closely related to the yuhinas facturer’s recommendations were followed. All primer
(Cibois 2003). sequences are shown in Table 2. For the historical DNA
In this study, by including two more Yuhina species (the template of Y. nigrimenta, four primer pairs were used to
white-naped yuhina Yuhina bakeri, and the striated yuhina amplify overlapping fragments of the cytochrome b gene.
Yuhina castaniceps), we aimed to elucidate the phyloge- We were careful not to contaminate the historical DNA
netic relationship within Yuhina in more detail by during DNA isolation and PCR (polymerase chain reac-
sequencing four mitochondrial DNA fragments (2,379 bp tion). PCR amplification was carried out using the fol-
in total). By adding three species of Zosterops and one lowing parameters: 94 C for 5 min, followed by 35 cycles
species of Stachyris, which is endemic to the Phillippines, of 94 C (30 s), 45–52 C (40 s), 72 C (40 s), and final
we also attempted to elucidate their relationships to the extension at 72 C (5 min). A touchdown PCR was used to
yuhinas. While typical yuhinas are crested, the Philippine amplify four fragments of the cytochrome b gene of
Stachyris and Zosterops are crestless. Molecular phylogeny Y. nigrimenta. Annealing temperature ranged from 64 to
of these birds was used to infer the evolution trend of the 57 C for the two terminal fragments and 62–55 C for
crest. Since some species sampled in the study are endemic the two internal fragments. Primer pairs L1549/H1991
to Southeast Asia, we also hoped that molecular phylogeny (Desjaidin and Morais 1990), L3214 (Hedges 1994)/H3783
could provide a better basis for interpreting the origins and (Hedges and Sibley 1994) and L10755/H11151 (Chesser
diversification of birds in Southeast Asia, which is still 1999) were used to amplify the 12s, 16s rRNA, and ND3
poorly understood (Moyle et al. 2005). genes. The same PCR amplification parameters used for
cytochrome b were used to amplify the above fragments,
except that the annealing temperature was changed to
Methods 50–65 C. PCR products were purified using a DNA puri-
fication kit (Sangon Inc., Shanghai, China) and sequencing
Taxa sampling, amplification, and sequencing of purified PCR products was performed with BigDye
Terminator Cycle Sequencing Kits (Applied Biosystems,
Due to the difficulty involved in choosing an appropriate Foster City, CA, USA) on an ABI Prism 3100 automated
outgroup, five species of Timaliini, the red-tailed minla DNA sequencer (Perkin-Elmer Applied Biosystems).
Minla ignotincta, the blue-winged laughingthrush Garrulax
squamatus, the grey-throated babbler Stachyris nigriceps, Phylogenetic analysis
the rufous-capped babbler Stachyris ruficeps, the golden
babbler Stachyris chrysaea, and a species of Acrocephali- DNA sequences were proofread by eye and then assembled
nae, the brownish-flanked bush-warbler Cettia fortipes, using the software DNAstar (DNAstar Inc., Madison, WI,
were included as a composite outgroup. All Yuhina species USA). All cytochrome b and ND3 sequences were aligned
described, except for the Burmese yuhina Yuhina humilis, by using Clustal X with default parameters (Thompson
were sampled. However, Y. humilis is considered to be et al. 1997). To avoid the alignment difficulties introduced
closely related to the whiskered yuhina Y. flavicollis, and is by indels, we used Gblock 0.91b (Castresana 2000) to
treated as a subspecies of Y. flavicollis by some authors delete gaps within certain regions of the 12s and 16s rRNA.
(Deignan 1964; Dickinson 2003). Three species of Pairwise distances (Tamura–Nei distance) and other char-
Zosterops (the Japanese white-eye Zosterops japonicus, the acteristics of the sequence were calculated using MEGA3.0
chestnut-flanked white-eye Z. erythropleurus, and the or- (Kumar et al. 2004). A partition homogeneity test (ILD)
iental white-eye Z. palpebrosus) and one species of and a test of stationary nucleotide composition across taxa
Stachyris endemic to the Philippines (the chestnut-faced were conducted in PAUP*4.0b10 (Swofford 2003).
babbler Stachyris whiteheadi) were also included. The taxa Phylogenetic analysis was performed using the maxi-
sampled in this study are summarized in Table 1. mum parsimony (MP) and maximum likelihood (ML)
Four mitochondrial fragments (cytochrome b, ND3, 12s algorithms implemented in PAUP*4.0b10. For the MP
and 16s rRNA) were used to infer the molecular phylogeny analysis, a weighted parsimony method was used by
of Yuhina and its allies. Total genomic DNA was isolated removing the transitions at the third codon position in both
from 95% alcohol-preserved tissues (muscle, liver, and cytochrome b and ND3 genes to avoid the saturation effect.
blood) or the toe pads of museum skin specimens (for A heuristic search, with 1,000 replicates of random addi-
the black-chinned yuhina Yuhina nigrimenta). For tissue tion sequences of taxa and tree bisection reconnection

123
J Ornithol (2007) 148:417–426 419

Table 1 Samples and sequences used in this study

Taxa Museum, tissue number Collecting locality GenBank accession no.
Cytb ND3 12s 16s

Zosterops erythropleurus KIZ, GLGS1443 Yunnan, China DQ837524a DQ837492a DQ837507a DQ837515a
a a a
Zosterops palpebrosus KIZ, GLGS1816 Yunnan, China DQ837522 DQ837493 DQ837509 DQ837517a
Zosterops japonicus KIZ, GLGS2247 Yunnan, China DQ837523a DQ837491a DQ837508a DQ837516a
a a a
Cettia fortipes KIZ, 05353 Yunnan, China DQ837512 DQ837490 DQ837512 DQ837518a
a a b
Garrulax squamatus KIZ, GLGS1833 Yunnan, China DQ092872 DQ837502 AF484900 AF484357b
a a b
Minla ignotincta KIZ, GLGS091 Yunnan, China DQ092874 DQ837503 AF484916 AF484373b
a a b b
Stachyris ruficeps KIZ, GLGS494 Yunnan, China DQ092876 DQ837505 AF376925 AF376905
Stachyris chrysaea KIZ, 04035 Yunnan, China DQ092875a DQ837506a AF376922 b
AF376902 b

b b b
Stachyris whiteheadi AY124542 AF376928 AF094661
a a b b
Stachyris nigriceps KIZ, GLGS1854 Yunnan, China DQ092873 DQ837495 AF376924 AF376904
a a a a
Yuhina castaniceps KIZ, 99172 Guangxi, China DQ092879 DQ837488 DQ837510 DQ837514
Yuhina everetti LSUMNS, B36457 Sabah, Malaysia DQ092881a DQ837489a AF376917b AF376898b
Yuhina bakeri KIZ, 04323 Yunnan, China DQ092877a DQ837499a DQ837511a DQ837513a
Yuhina flavicollis KIZ, GLGS1424 Yunnan, China DQ092882a DQ837497a AF376916b AF376897b
a a b
Yuhina gularis KIZ, GLGS1248 Yunnan, China DQ092883 DQ837498 AF376918 AF094659b
a a b
Yuhina diademata KIZ, 03026 Yunnan, China DQ092880 DQ837496 AF376915 AF376896b
a a b
Yuhina occipitalis KIZ, 03031 Yunnan, China DQ092885 DQ837494 AF376920 AF376900b
a a b
Yuhina brunneiceps TNU, T0620 Taiwan, China DQ092878 DQ837501 AF376914 AF376895b
Yuhina nigrimenta KIZ, 012777 Yunnan, China DQ092884a DQ837500a AF376919b AF376899b
a a b
Yuhina zantholeuca TNU, T0325 Taiwan, China DQ092886 DQ837504 AF376921 AF376901b
Taxonomic denomination followed Sibley and Monroe (1990)
a
Sequences new to this study
b
Sequences taken from GenBank database, published by Barhoum and Burns (2002), Cibois et al. (2002), Cibois (2003)
KIZ Kunming Institute of Zoology, the Chinese Academy of Sciences; TNU Department of Life Science, Taiwan Normal University; LSUMNS
Louisiana State University Museum of Natural History

Table 2 Primers used in this study

Gene Primer name Primer sequence (5¢-3¢) References

Cyt b External primers L14827a CCACACTCCACACAGGCCTAATTAA Helm-Bychowski and Cracraft (1993)

H16065a GGAGTCTTCAGTCTCTGGTTTACAAGAC Helm-Bychowski and Cracraft (1993)
Internal primers L15206a CACATCGGCCGAGGAATCTACTA Cibois et al. (1999)
H15224a GTACGACTCCGATGTTTCAGGTTTC Designed in the study
L15407a TGAGGTGGATTCTCAGTAGAC Designed in the study
H15487a GATCCTGTTTCGTGGAGGAAGGT Cibois et al. (1999)
L15664a CTCACTAGGAGACCCAGAAAACTTCAC Designed in the study
H15709a GTAGGCAAATAGGAAGTATC Designed in the study
L15383 GGACAAACACTAGTAGAATG Cibois et al. (1999)
12s External primers L1549 GGGTTGGTAAATCCTGTGCCAGCCA Desjaidin and Morais (1990)
H1991 GCTATACCTTGACCTGTCTT Desjaidin and Morais (1990)
16s External primers L3214 CGCCTGTTTATCAAAAACAT Hedges (1994)
H3783 CCGGTCTGAACTCAGATCACGT Hedges and Sibley (1994)
ND3 External primers L10755 GACTTCCAATCTTTAAAATCTGG Chesser (1999)
H11151 GATTTGTTGAGCCGAAATCAAC Chesser (1999)
Note: the letters L and H refer, respectively, to the light and heavy strands, and the number refers to the base position at the 3¢ end of the primer in
the complete chicken mtDNA sequences established by Desjardins and Morais (1990)
a
Signifies primers for the specimen DNA

123
420 J Ornithol (2007) 148:417–426

(TBR) branch swapping, was executed to obtain the MP mitochondrial partitions. Therefore, we combined se-
tree. We assessed the robustness of the MP tree with 1,000 quences of four mitochondrial fragments into a larger data
bootstrap replicates. set (2,379 bp). The characteristics of these sequences are
We used the program Modeltest 3.6 (Posada and summarized in Table 3. The pattern of base compositional
Crandall 1998) to choose an appropriate substitution model bias was similar to that found in previous avian studies
for the ML analysis. The model was subsequently used in (Barhoum and Burns 2002; Perez-Eman 2005). The result
PAUP*4.0b10 to search the ML tree using a heuristic search of a chi-square test (X2 = 13.24, P = 1.00, df = 57) sup-
with 1,000 replicates of random addition sequences for taxa ported the notion that nucleotide composition is stationary
and TBR branch swapping. Reliability of the phylogenetic across taxa.
relationships was evaluated by performing 100 replicates of Because data on avian cytochrome b gene is commonly
bootstrap analysis with taxa added ‘‘as is.’’ available, we calculated its divergence between our taxa
In addition, Bayesian inference of phylogeny was per- for the purposes of comparison (Table 4). The average
formed using the program Mrbayes 3.0b4 (Huelsenbeck sequence divergence between the Yuhina clade and out-
and Ronquist 2001). The substitution model selected by the group was 16.18%, but it was only 12.12% for Yuhina and
program Modeltest 3.6 was likewise used in the Bayesian Zosterops, 11.85% for Yuhina and the Philippine Stachyris,
method. The Bayesian analysis started with randomly and 10.77% for the Philippine Stachyris and Zosterops. The
generated trees; four Markov chains under default heating average sequence divergence between Yuhina species was
values were run for four million generations and sampled about 11.10% (ranging from 3.25% for Y. castaniceps to
every 100 generations. The ‘‘burn-in’’ was determined by Y. everetti to 13.75% for Y. castaniceps to Y. nigrimenta).
checking for the likelihood of being stationary. Posterior The average sequence divergence among Zosterops species
probabilities were estimated to assess the robustness of was only 5.19%.
the tree.
Phylogenetic analysis

Results Rooting the tree with C. fortipes, parsimony analysis

without transitions of the third codon position yielded only
Sequence variation one MP tree (branch length = 1,277, consistency in-
dex = 0.5043, retention index = 0.5149, rescaled consis-
We obtained a total of 50 sequences, and their GenBank tency index = 0.2597) (Fig. 1). In the MP analysis, the
accession numbers, together with those of 29 sequences Yuhina clade was robust for the exclusion of Y. zantholeuca
published in Barhoum and Burns (2002), Cibois et al. (97% bootstrap values). S. whiteheadi and three species of
(2002) and Cibois (2003), are listed in Table 1. We ob- Zosterops formed a clade embedded within the Yuhina
tained complete cytochrome b (1,143 bp) and ND3 clade (99% bootstrap values). The white-collared yuhina
(351 bp) sequences for all species sampled except for the Yuhina diademata was found to be a sister group to the rest
ND3 sequences of S. whiteheadi. We also obtained se- of the yuhinas (90% bootstrap values). Other terminal
quences of partial mitochondrial 12s (about 400 bp) and clades were also strongly supported: Y. castaniceps was a
16s (about 530 bp) rRNA genes for species omitted from sister group to Y. everetti (100% bootstrap values), and the
previous studies (Cibois et al. 2002; Cibois 2003). After Taiwan yuhina Yuhina brunneiceps was a sister group to
removing the gaps, 386 bp of 12s rRNA and 499 bp of 16s Y. nigrimenta (94% bootstrap values). A sister group
rRNA was included in the subsequent phylogenetic relationship between the stripe-throated yuhina Yuhina
analysis. gularis and the rufous-vented yuhina Yuhina occipitalis
The ILD test revealed that phylogenetic signals did was discovered with strong support (96% bootstrap values).
not conflict significantly (P = 0.896) among the four Z. japonicua, Z. palpebrosus and Z. erythropleurus were

Table 3 Molecular characterization of the mitochondrial genes used in this study

Genes Total sites Variable sites Informative sites Variable sites by codon position Nucleotide frequencies
First Second Third %A %C %G %T

Cytochrome b 1143 419 (36.7%) 338 (29.6%) 84 (20.0%) 21 (5.0%) 314 (75.0%) 28.5 33.4 13.2 24.9
ND3 351 157 (44.7%) 132 (37.6%) 44 (28.0%) 15 (9.6%) 98 (62.4%) 29.4 32.4 12.3 25.9
12s 386 81 (21.0%) 44 (11.4%) 33.1 25.7 19.4 21.8
16s 499 92 (18.4%) 53 (10.6%) 30 25.7 23.1 21.1

123
 Table 4 Pairwise sequence divergence (Tamura–Nei distances) in cytochrome b
J Ornithol (2007) 148:417–426

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

1.Y. gularis
2.Y. occipiotalis 0.0869
3.Y. flavicollis 0.0944 0.0840
4.Y. bakeri 0.0964 0.0839 0.0811
5.Z. palpebrosus 0.1225 0.1121 0.1134 0.1038
6.Z. japonicus 0.1168 0.1126 0.1032 0.1072 0.0505
7.Z. erythropeurus 0.1116 0.1099 0.1046 0.1028 0.0544 0.0491
8.S. whiteheadi 0.1193 0.1148 0.1143 0.1121 0.1147 0.1083 0.1000
9.Y. nigrimenta 0.1176 0.1181 0.1212 0.1151 0.1367 0.1407 0.1322 0.1129
10.Y. brunneiceps 0.0983 0.0876 0.0908 0.0868 0.1097 0.1140 0.1032 0.1045 0.0967
11.Y. everetti 0.1134 0.1079 0.1080 0.1166 0.1263 0.1372 0.1307 0.1248 0.1372 0.1062
12.Y. castaniceps 0.1128 0.1070 0.1007 0.118 0.1285 0.1309 0.1267 0.1253 0.1375 0.1012 0.0325
13.Y. diademata 0.1317 0.1174 0.1228 0.1054 0.1379 0.1463 0.1331 0.1385 0.1310 0.1294 0.1322 0.1326
14.M. ignortincta 0.1495 0.1502 0.1479 0.1601 0.1664 0.1682 0.1681 0.1679 0.1611 0.1492 0.1500 0.1436 0.1472
15.G. squamatus 0.1241 0.1467 0.1357 0.1315 0.1568 0.1609 0.1475 0.1515 0.1416 0.1307 0.1493 0.1505 0.1537 0.1252
16.C. forticeps 0.1498 0.1553 0.1465 0.1560 0.1701 0.1635 0.1650 0.1645 0.1689 0.1525 0.1654 0.1711 0.1509 0.1665 0.1727
17.Y. zantholeuca 0.1611 0.1653 0.1636 0.1634 0.1732 0.1797 0.1734 0.1684 0.1534 0.1701 0.1787 0.1826 0.1676 0.1855 0.1690 0.1742
18.S. ruficeps 0.1506 0.148 0.1606 0.1606 0.1842 0.1772 0.1761 0.1826 0.1449 0.1645 0.1676 0.1641 0.1732 0.1766 0.1570 0.1867 0.1652
19.S. chrysaea 0.1583 0.1425 0.1677 0.1604 0.1689 0.1717 0.1742 0.1667 0.1688 0.1643 0.1705 0.1692 0.1681 0.1744 0.1639 0.1904 0.1678 0.0931
20.S. nigriceps 0.1717 0.16 0.1757 0.1717 0.1807 0.1929 0.1790 0.1754 0.1248 0.1718 0.1825 0.1778 0.1648 0.1718 0.1690 0.1910 0.1787 0.1408 0.1527
421

123
422 J Ornithol (2007) 148:417–426

Fig. 1 Phylogenetic interspecies relationships of Yuhina based on cyt Fig. 2 Phylogenetic relationships of Yuhina based on cytb, ND3, 12s
b, ND3, 12s, and 16s sequences; maximum parsimony tree is shown and 16s sequences obtained with maximum likelihood and Bayesian
with transitions of third position removed. Numbers represent analysis. Numbers represent bootstrap values/posterior probabilities,
bootstrap values; only those >70% are shown above the branches and are only shown when bootstrap values >70% and posterior
probabilities >95%

found to form a clade (100% bootstrap values). Z. palpe- probability values). The clade of Y. bakeri and Y. flavicollis
brosus was a sister group to the clade made up of Z. ery- shared a common ancestor with Y. gularis/Y. occipitalis,
thropleurus and Z. japonicus, and the relationship obtained and obtained strong support in Bayesian analysis (97%
strong support (77% bootstrap values). posterior probability values). The topological difference
Modeltest 3.6 selected the GTR + I + G model as a between the MP and ML trees was within the Zosterops
substitution model for our combined data set. The param- clade; in ML and Bayesian analysis, Z. erythropleurus was
eter values were estimated as follows: base frequencies a sister group to the clade comprising Z. palpebrosus and
A = 0.3128, C = 0.3459, G = 0.1496, T = 0.1917; proba- Z. japonicus, but this did not obtain strong support (with
bilities for the six different substitution types Rmatrix = 59% bootstrap values in ML analysis and 80% posterior
1.8632, 6.6449, 2.3026, 0.2447, 17.5067, 1, with 0.5857 of probability values in Bayesian analysis, not shown).
sites invariable, and the substitution rates for variable sites The hypothesis of constant rate within the alignment
followed a gamma distribution with a shape parameter of was rejected by a likelihood ratio test (LRT test) (Swofford
0.7868. The ML and Bayesian trees were identical (Fig. 2), et al. 1996) in PAUP*4.0b10. Under the GTR + I + G
and they were almost identical to the MP tree, with the model of evolution, with no molecular clock enforced, the
aforementioned nodes gaining strong support. Furthermore, likelihood (log L = –13171.3846) was significantly better
the sister group relationship of Y. flavicollis and Y. bakeri at the 5% level than when the molecular clock was en-
was strongly supported in Bayesian analysis (97% posterior forced (log L = –13199.7564) (P = 0.000007).

123
J Ornithol (2007) 148:417–426 423

Discussion Yuhina castaniceps and Y. everetti formed a well-sup-

ported sister group in our study, while a debate regarding
The likely close relationship of Zosterops and babblers the species status of Y. everetti exists (Inskipp et al. 1996).
In light of the phylogenetic species concept (PSC) (Cra-
Our results revealed that the crestless Zosterops and the craft 1983; Nixon and Wheeler 1990), diagnosability from
Philippine Stachyris formed a clade embedded within the molecular or morphological characters could justify the
yuhinas with strong support (Figs. 1, 2). Zosterops is a species status of Y. everetti. When we took all five sub-
large genus including 76 species, which are mostly dis- species of Y. castaniceps together with Y. everetti into
tributed in the old world (Sibley and Monroe 1990), and account, we found that the morphological differences
few studies of phylogenetic relationships within Zosterops among the six taxa were obvious. The chestnut area on the
have been reported (Warren et al. 2006). A close rela- head showed a trend for enlargement only in the ear-co-
tionship between Zosterops and babblers has not previously verts for Y. c. plumbeiceps; in the ear-coverts and a patch
been proposed by taxonomists on the basis of morpho- above the posterior supercilium for Y. c. rufigensis; in the
logical or ecological characteristics (Cibois 2003). How- ear-coverts and the sides of the neck in Y. c. striata; and in
ever, more and more molecular studies have suggested that the ear-coverts, on the sides of the neck and the nape in
some species of Zosterops are members of the Timaliini Y. c. torqueola. Y. c. castaniceps is most similar to
(Cibois 2003; Ericson and Johansson 2003; Barker et al. Y. everetti in appearance, but it is scalloped with pale grey
2004; Voelker and Spellman 2004; Alstrom et al. 2006; on its forehead (Ali and Ripley 1987; Grimmett et al. 1999;
Beresford et al. 2005; Fuchs et al. 2005). Four species of Robson 2000). In addition, the Tamura–Nei distance of the
Zosterops have been proved to be members of the Timaliini mitochondrial cytochrome b gene between Y. castaniceps
by these studies, namely the African yellow white-eye and Y. everetti is 3.25%; the divergence level is comparable
Z. senegalensis, the pale white-eye Z. pallidus, and the to that of well accepted sister groups in North American
yellowish white-eye Z. nigrorum and Z. japonicus. Our birds (Johnson and Cicero 2004).
study proved that another two species of Zosterops, Our study discovered a sister relationship between
Z. palpebrosus and Z. erythropleurus, are closely related to Y. gularis and Y. occipitalis, with strong support in all
the Timaliini. However, a larger sampling of Zosterops is trees. This is consistent with their morphological similarity;
needed to verify the likely close relationship with babblers. they share the same color pattern on the under-parts, except
Within Zosterops, less average sequence divergence (5.2%) for some black stripes on the throat of Y. gularis. Inter-
was observed among Z. japonicus, Z. palpebrosus and estingly, the current ranges of this species pair are highly
Z. erythropleurus, which is consistent with their extremely sympatric; both are distributed in northern India, south-
similar plumage patterns. eastern Tibet, northeastern Burma, and southwestern
China, with Y. gularis also extending into northern Laos
Interspecies relationships among yuhinas and Vietnam (Sibley and Monroe 1990). A sympatric sister
species pattern was also found for the Y. flavicollis/
The middle nodes of the Yuhina clade have been poorly Y. bakeri clade. These are both distributed in northern
resolved in a previous study (Cibois et al. 2002). How- India, northern Burma and southwestern China, with
ever, even when the length of the sequence used was Y. flavicollis also extending into northwestern Thailand,
doubled in the current study, these nodes were still dif- northern Laos and northern Vietnam (Sibley and Monroe
ficult to resolve. This may be caused by the short branch 1990). Although the four closely related species are cur-
lengths of the nodes (rapid diversification). Similar cases rently sympatric in the Himalayas, it is hard to infer the
have been reported for numerous passerine birds (re- cause of their distribution pattern. Due to their high
viewed by Perez-Eman 2005). Adding more informative mobility, the distributions of these birds would have easily
characters, especially nuclear loci, may shed light on this adapted to the historical shift in vegetation (Losos and Glor
problem. 2003). However, their profound molecular divergence
Systematic treatment of Y. castaniceps used to be con- indicates independent evolutionary histories (8.11% for
troversial. Harrison (1986a, 1986b) proposed removing Y. flavicollis/Y. bakeri, 8.69% for Y. gularis/Y. occipitalis).
Y. castaniceps from the genus Yuhina and resurrecting its The sympatric distribution patterns of these species were
original monotypic genus Staphida (Gould 1871), as its tail most likely caused by secondary contact.
is round-ended, which is different from the typical yuhina
square-ended tail. Our data does not support this proposi- Crest evolution
tion; molecular phylogeny showed that all yuhinas (except
Y. zantholeuca), along with Zosterops and the Philippine Including the results from the current study, the molecular
Stachyris, formed a robust clade. phylogeny of the Yuhina and its allies has proved that the

123
424 J Ornithol (2007) 148:417–426

contentious species Y. zantholeuca is not a babbler but is Philippine Stachyris (Rand 1970, but see Cibois et al.
instead closely related to the Vireonidae (Cibois et al. 2002).
2002; Barker et al. 2004). This suggests that the morpho-
logical similarities between Y. zantholeuca and other
yuhinas, especially the presence of a crest, are probably Zusammenfassung
caused by convergent evolution. In fact, when we carefully
examined the crest from specimens of Y. zantholeuca and Molekulare Phylogenie der Yuhinas (Sylviidae:
another seven Yuhina species kept by the Kunming Insti- Yuhina), einer paraphyletischen Gruppe von
tute of Zoology of the Chinese Academy of Sciences, we Brillenvögeln aus Zosterops und Philippinischer
found that the crest of Y. zantholeuca is merely composed Stachyris
of crown feathers, while the crest of the other Yuhina
species comprises both crown and occipital feathers. It is Um die phylogenetischen Zusammenhänge innerhalb der
the convergent crest characteristic that has caused this Yuhinas (Yuhina) zu rekonstruieren, wurden Mitochond-
contentious species to be misclassified into the Yuhina. riensequenzen (2379 bp) von Cytochrom-b Genen, ND3,
Similarly, the broad-billed sapayoa, Sapayoa aenigma, and 12s und 16s rRNA analysiert. Die Yuhinas beinhalten die
the Tibetan ground-jay, Pseudopodoces humilis, have re- Brillentimalie Stachyris whiteheadi, einen Endemiten der
cently been proven by molecular systematics to have been Philippinen, den Japanbrillenvogel Zosterops japonicus,
misclassified into different families (Fjeldså et al. 2003; den Rotflanken-Brillenvogel Z. erythropleurus, sowie den
James et al. 2003; but see Fjeldså et al. 2005). Overall Indischen-oder Ganges-Brillenvogel Z. palpebrosus. Die
morphological similarities could not reflect the true rela- Ergebnisse unterstützen stark die Hypothese, wonach
tionships of these contentious birds. S. whiteheadi und drei Brillenvogelarten eine phyletische
Our phylogeny also suggested that crestlessness was a Einheit innerhalb des Yuhina-Zweigs bilden. Die Robust-
derived character of the Zosterops/Philippines Stachyris heit des Zweiges Yuhina/Philippinische Stachyris/Zoster-
clade, and was only derived once. The general evolutionary ops lehnt die Gültigkeit der monotypischen Gattung
trend of the crest showed that the crest was not stable in the Staphida ab, die für die Rotohr-Yuhina Yuhina castaniceps
true Yuhina clade, which made Yuhina a paraphyletic auf Grund besonderer morphologischer Merkmale vorg-
group. It has been suggested that, once evolved, similar eschlagen worden war. Eine Neuentdeckung stellt das
color patterns can easily disappear and re-appear inde- Schwesterntaxa-Verhältnis zwischen Kehlstreifen-Yuhina
pendently in the leaf warblers, Phylloscopus (Price and Yuhina gularis und Roststeiß-Yuhina Yuhina occipitalis
Palvelka 1996). Therefore, extreme caution should always dar, ebenso wie das zwischen Gelbnacken-Yuhina Yuhina
be taken when applying any morphological traits to avian flavicollis und Rotkopf-Yuhina Yuhina bakeri. Das sym-
systematics, otherwise the conclusions could be totally patrische Verbreitungsmuster der beiden oben genannten
misleading. Schwesterngruppen in der Himalaya-Region entstand
höchstwahrscheinlich im Zuge eines sekundären Kontakts.
The dispersal route for yuhinas and their allies Unsere molekulare phylogenetische Systematik zeigt auch,
dass das Fehlen einer Haube ein einmalig abgeleitetes Ei-
Geologically, the Philippine archipelago had never been genmerkmal für den Zweig aus Zosterops und Philippini-
connected to the Asian continent or to Sunda land directly scher Stachyris darstellt. Die monophyletische Einheit aus
(Hall 1998). In addition, no species of Yuhina is known for Zosterops und Philippinischer Stachyris in unserer Arbeit
its long distance dispersal or migration. Therefore, the bedeutet außerdem, dass ein gemeinsamer Vorfahr von
origin of the Philippine Stachyris is not straightforward. Zosterops und Philippinischer Stachyris als Autapomorphie
Seasonal migration and long distance dispersal are, how- die Fähigkeit zur Langstrecken-Dispersion entwickelt hat,
ever, well documented for several Zosterops species. For was ihm ermöglichte, über See zu fliegen und die Philip-
example, Z. lateralis have been found to frequently dis- pinen zu erreichen.
perse from New Zealand to its nearby islands (Mees 1969).
Acknowledgments We thank the Comparative Genomics Group,
Zosterops japonicus and Z. erythropleurus were found to
which is directed by Dr. Bing Su (Kunming Institute of Zoology, the
migrate between northern and southeastern Asia seasonally Chinese Academy of Sciences), for providing their laboratory and
(Sibley and Monroe 1990). It is possible that an ancestor of kindly helping with the experiment. We would also like to thank Dr.
Zosterops/Philippines Stachyris was able to disperse from Shou-Hsien Li and Carol Yeung (Department of Life Science, Taiwan
Normal University) for providing tissue samples and providing
its Indochinese or Himalayan origins into the Sunda region
helpful comments on the manuscript. We also thank Donna L. Ditt-
and subsequently into the Philippines. This is also consis- mann (Museum of Natural Science, Louisiana State University) for
tent with the Borneo origin hypothesis proposed for the providing tissue samples. Moe Flannery (California Academy of

123
J Ornithol (2007) 148:417–426 425

Sciences) kindly checked old references for us and facilitated the Fuchs J, Fjeldsa J, Bowie RCK, Voelker G, Pasquet E (2005) The
tissue loan from the Museum of Natural Science, Louisiana State African warbler genus Hyliota as a lost lineage in the oscine
University. This work was supported by the Chinese Academy of songbird tree: molecular support for an African origin of the
Sciences Innovation Program (KSCX2-1-09). The work was in Passerida. Mol Phylogenet Evol 39: 186–197
agreement with all of the laws of the countries in which the samples Grimmett R, Inskipp C, Inskipp T (1999) Birds of India, Pakistan,
were taken. Nepal, Bangladesh, Bhutan, Sri Lanka, and the Maldives.
Princeton University Press, Princeton, NJ, pp 316
Gould S (1871) Birds Asia 4:23
Hall R (1998) The plate tectonics of Cenozoic SE Asia and the
References distribution of land and sea. In: Hall R, Holloway D (eds)
Biogeography and geological evolution of SE Asia. Backhuys,
Ali S, Ripley SD (1987) Compact handbook of the birds of India and Leiden
Paskistan, 2nd edn. Oxford University Press, Oxford, 7:101–109 Harrison CJO (1986a) A re-assignment of the affinities of some small
Alstrom P, Ericson PGP, Olsson U, Sundberg P (2006) Phylogeny and oriental babblers Timaliidae. Forktail 1:81–83
classification of the avian superfamily Sylvioidea. Mol Phylo- Harrison CJO (1986b) A re-assignment of two small babblers at
genet Evol 38:381–397 present in the genus Yuhina. J Bombay Nat Hist Soc 83:202–206
Barker FK, Cibois A,Schikler P, Feinstein J, Cracraft J (2004) Hedges SB (1994) Molecular evidence for the origin of birds. Proc
Phylogeny and diversification of the largest avian radiation. Proc Natl Acad Sci USA 91:2621–2624
Natl Acad Sci USA 101:11040–11045 Hedges SB, Sibley CG (1994) Molecules vs morphology in avian
Barhoum DN, Burns KJ (2002) Phylogenetic relationships of the evolution: The case of the ‘‘pelecaniform’’ birds. Proc Natl Acad
wrentit based on mitochondrial Cytochrome b sequences. Sci USA 91:9661–9865
Condor 104:740–749 Helm-Bychowski K, Cracraft J (1993) Recovering phylogenetic
Beresford P, Barker FK, Ryan PG, Crowe TM (2005) African signal from DNA sequences: relationships within the Corvine
endemics span the tree of songbirds (Passeri): molecular assemblage (Class Aves) as inferred from complete sequences of
systematics of several evolutionary ‘‘enigmas’’. Proc R Soc B the mitochondrial DNA cytochrome-b gene. Mol Biol Evol
272:849–858 10:1196–1214
Castresana J (2000) Selection of conserved blocks from multiple Huelsenbeck JP, Ronquist FR (2001) MRBAYES: Bayesian inference
alignments for their use in phylogenetic analysis. Mol Biol Evol of phylogenetic trees. Bioinformatics 17:754–755
17:540–552 Inskipp T, Lindsey N, Duckworth W (1996) An annotated checklist of
Chesser RT (1999) Molecular systematics of the rhinocryptid genus the birds of the Oriental Region. Oriental Bird Club, Sandy, UK,
Pteroptochos. Condor 101:439–446 pp 191–192
Cibois A (2003) Mitochondrial DNA phylogeny of babblers (Timal- James HF, Ericson PGP, Slikas B, Lei FM, Gill FB, Olson SL (2003)
iidae). Auk 120:35–54 Pseudopodoces humilis, a misclassified terrestrial tit (Aves:
Cibois A, Kalyakin MV, Han LX, Pasquet E (2002) Molecular Paridae) of the Tibetan Plateau: evolutionary consequences of
phylogenetics of babblers (Timaliidae): revaluation of the genera shifting adaptive zones. Ibis 145:185–202
Yuhina and Stachyris. J Avian Biol 33:380–390 Johnson NK, Cicero C (2004) New mitochondrial DNA data affirm
Cibois A, Pasquet E, Schulenberg TS (1999) Molecular systematics of the importance of Pleistocene speciation in North American
the Malagasy babblers (Passeriformes: Timaliidae) and warblers birds. Evolution 58:1122–1130
(Passeriformes: Sylviidae), based on Cytochrome b and 16s Kumar S, Tamura K, Nei M (2004) MEGA3: Integrated software for
rRNA sequences. Mol Phylogenet Evol 13:581–595 molecular evolutionary genetics analysis and sequence align-
Clements JF (2000) Birds of the world: a checklist, 5th edn. Ibis, ment. Brief Bioinform 5:150–163
Vista, CA, pp 520 Losos JB, Glor RE (2003) Phylogenetic comparative methods and the
Cracraft J (1983) Species concepts and speciation analysis. Curr geography of speciation. Trends Ecol Evol 18:220–227
Ornithol 1:159–187 Mees GF (1969) A systematic review of the Indo-Australian
Deignan HG (1964) Timaliinae, Panurinae, Picathartinae. In: Mayr E, Zosteropidae (Part III). Zoologische Verhandelingen 102:1–390
Paynter RAJr (eds) A checklist of birds of the world: a Moyle RG, Schilthuizen M, Rahman MA, Sheldon FH (2005)
continuation of the work of James L. Peters, vol 10. Museum Molecular phylogenetic analysis of the white-crowned forktail
of Comparative Zoology, Cambridge, pp 240–442 Enicurus leschenaultia in Borneo. J Avian Biol 36:96–101
Delacour J (1946) Les Timaliines. L’Oiseau 16:7–36 Nixon KC, Wheeler QD (1990) An amplication of the phylogenetic
Delacour J (1950) Les Timaliines: additions et modifications. species concept. Cladistics 6:211–223
L’Oiseau 20:186–191 Perez-Eman JL (2005) Molecular phylogenetics and biogeography of
Desjardins P, Morais R (1990) Sequence and gene organization of the the Neotropical redstarts (Myioborus: Aves, Parulinae). Mol
chicken mitochondrial genome. J Mol Biol 212:599–634 Phylogenet Evol 37:511–528
Dickinson EC (2003) The Howard and Moore complete checklist of Posada D, Crandall KA (1998) Modeltest: testing the model of DNA
the birds of the world, 3rd edn. Princeton University Press, substitution. Bioinformatics 14:817–818
Princeton, NJ, pp 621–622 Price T, Palvelka M (1996) Evolution of colour pattern: history,
Ericson PGP, Johansson US (2003) Phylogeny of passerida (Aves: development, and selection. J Evol Biol 9:451–470
Passeriformes) based on nuclear and mitochondrial sequence Rand AL (1970) Species formation in the blue monarch flycatchers
data. Mol Phylogenet Evol 29:126–138 genus Hypothymis. Nat Hist Bull Siam Soc 23:353–365
Fjeldså J, Zuccon D, Irestedt M, Johansson US, Ericson PG (2003) Robson C (2000) A guide to the birds of Southeast Asia. Princeton
Sapayoa aenigma: a New World representative of ‘‘Old World University Press, Princeton, NJ
suboscines’’. Proc R Soc Lond B 270:S238–S241 Sambrook E, Fritsch F, Maniatis T (1989) Molecular cloning. Cold
Fjeldså J, Irestedt M, Ericson PGP (2005) Molecular data reveal some Spring Harbor Press, Cold Spring Harbor, NY
major adaptational shifts in the early evolution of the most Sibley CG, Monroe BLJr (1990) Distribution and taxonomy of birds
diverse avian family, the Furnariidae. J Ornithol 146:1–13 of the world. Yale University Press, New Haven, CT

123
426 J Ornithol (2007) 148:417–426

Swofford DL (2003) PAUP*. Phylogenetic analysis using parsimony Voelker G, Spellman GM (2004) Nuclear and mitochondrial DNA
(*and other methods), v 4.0b10. Sinauer, Sunderland, MA evidence of polyphyly in the avian superfamily Muscicapoidea.
Swofford DL, Olsen GJ, Wadell PJ, Hillis DM (1996) Phylogenetic Mol Phylogenet Evol 30:386–394
inference. In: Hillis DM, Moritz C, Mable BK (eds) Molecular Warren BH, Bermingham E, Prys-Jones RP, Thébaud C (2006)
systematics, 2nd edn. Sinauer, Sunderland, MA, pp 407–514 Immigration, species radiation and extinction in a highly diverse
Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG songbird lineage: white-eyes on Indian Ocean islands. Mol Ecol
(1997) The clustalx windows interface: flexible strategies for 15:3769–3786
multiple sequence alignment aided by quality analysis tools.
Nucleic Acid Res 24:4876–4882

123