Molecular Phylogenetics and Evolution 55 (2010) 952–967
Contents lists available at ScienceDirect
Molecular Phylogenetics and Evolution
journal homepage: www.elsevier.com/locate/ympev
Phylogeny of long-tailed tits and allies inferred from mitochondrial and nuclear markers (Aves: Passeriformes, Aegithalidae) q
Martin Päckert a,*, Jochen Martens b, Yue-Hua Sun c,* a Senckenberg Naturhistorische Sammlungen, Museum für Tierkunde, Königsbrücker Landstraße 159, 01109 Dresden, Germany b Institut für Zoologie, Johannes-Gutenberg-Universität, Saarstraße 21, 55099 Mainz, Germany c Institute of Zoology, Chinese Academy of Science, 100080 Beijing, China article info abstract
Article history: In this paper we provide a molecular phylogeny based on three mitochondrial and three nuclear markers Received 31 August 2009 for all long-tailed tit species of the genus Aegithalos including several doubtful subspecies (17 taxa) plus Revised 18 January 2010 three close allies of SE Asian Leptopoecile and North American Psaltriparus. Genus Aegithalos is divided Accepted 21 January 2010 into three major clades, two of them showing only minor differentiation. Separation of two mitchondrial Available online 25 January 2010 haploytpe clusters in the N Palearctic Long-tailed Tit, Ae. caudatus, was dated back to the Late Pleistocene, however, descendants from both lineages underwent a rapid post-Pleistocene range expansion and lar- Keywords: gely mixed over the entire distribution area. The Chinese populations of the glaucogularis subspecies Aegithalos group represent a slightly earlier Pleistocene split from the Ae. caudatus clade. Cytochrome b Nuclear markers Genetic differentiation among several doubtful SE Asian species taxa on the sister clade of the latter N Molecular clock Palearctic/Chinese clade matches the intraspecific differentiation within Ae. caudatus. Unexpectedly, Pleistocene cytochrome-b distances among Himalayan Ae. iouschistos (including the subspecies bonvaloti from China Himalayas and sharpei from Myanmar) and the Chinese endemic Ae. fuliginosus range at approximately 0.5% and apparently all these extant populations separated only very recently during late Pleistocene times, too. W Himalayan Ae. niveogularis clearly appeared as the sister species of the latter taxon assemblage. Unlike the two latter major clades, Ae. concinnus shows strong intraspecific differentiation with cyt-b distances as high as 6% among two Himalayan populations of ssp. iredalei, ssp. manipurensis from Myan- mar and a fourth lineage from SW and SC China including ssp. talifuensis and nominate concinnus. A sis- ter-group relationship between all Ae. concinnus and Ae. leucogenys was strongly supported. N American bushtits of genus Psaltriparus represent the sister clade to Palearctic genus Aegithalos, including a clear split between the minimus and the plumbeus subspecies group which was again dated back to Pleistocene times. The two tit-warbler species of genus Leptopoecile are strongly differentiated from one another and represent an early split from the Aegithalidae tree. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction 1925-1930), sometimes ranked as a separate subfamily by differ- ent authors (Stresemann, 1921; Mayr and Amadon, 1951), until Long-tailed tits and allies (Aegithalidae) are small bush- and both Aegithalidae and Remizidae were attributed family status tree-dwelling passerines with relatively short wings and dispro- by Vaurie (1957a). Subsequently their phlyogenetic relationships portionately long tails at least in one species. They inhabit decidu- to other passerine families were highly disputed until molecular ous and mixed mountain forests of the holarctic region; in some studies almost unanimously placed Aegithalidae within a super- Asian species the maximum vertical extent of breeding areas family Sylvioidea, but with conflicting results concerning their reaches up to 3600 m asl in the Himalayas and even higher, up closest relatives (Barker et al. 2002: close relations to Alaudidae to 4300 m asl, in West and Southwest China (Harrap and Quinn, and Cettia; Barker et al., 2004: close relations to Hirundinidae; Eric- 1996; del Hoyo et al., 2008). With the exception of the Sino-Hima- son et al., 2000; Ericson and Johansson, 2003; Jønsson and Fjeldså, layan tit warblers of the genus Leptopoecile, long-tailed tits were 2006; Alström et al., 2006: close relation to Phylloscopus by for- traditionally placed within the Paridae (Hartert, 1910; La Touche, mally recognizing Cettiidae as a separate clade of family status). The Aegithalidae are a rather small passerine family comprising four genera only. The Bushtit, Psaltriparus minimus, is the only New q Results of the Himalaya Expeditions of J. Martens No. 268. - For No. 267 see: World species with a large breeding distribution range in western Invertebrate Systematics 23, 452–505, 2009. * Corresponding authors. Fax: +49 351 8926327 (M. Päckert). North America from the Pacific Coast of British Columbia in the E-mail address: [email protected] (M. Päckert). north along the entire Rocky Mountain chain to Central America
1055-7903/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2010.01.024 M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967 953 in the south, with some small isolated populations in Mexico and cent Tibet and Yunnan northeastwards to Sichuan and Guizhou (for Guatemala (Sloane, 2001). Two species of the genus Leptopoecile species rank of the latter see also Dickinson (2003)). Based on mor- are restricted to the mountainous forests of the Sino-Himalayan re- phological character variation, these two taxa were formerly either gion: L. elegans and L. sophiae and adjacent areas in central Asia in treated as subspecies of the Black-browed Tit, Ae. iouschistos, from the latter species. L. sophiae was repeatedly placed in a separate the eastern Himalayas (Harrap and Quinn, 1996) or as subspecies (sub)genus Lophobasileus (Pleske, 1890; Neufeldt and Wunderlich, of the White-throated Tit, Ae. niveogularis, from the Western Hima- 1983; Mayr et al., 1986, p. 294) a distinction which was already re- layas and adjacent Kashmir (Eck and Martens, 2006; see also Ta- jected by Vaurie (1957b). In general for most of the 20th century ble 1, this paper). both Leptopoecile species were not affiliated with Aegithalidae Three further Asian taxa are unanimously treated as species on but with Sylviidae assuming close relationships to Phylloscopus their own with unknown phylogenetic affinities to other Asian and Regulus. Apart from morphological characters, similarities populations. The Black-throated Tit, Ae. concinnus, is the most and differences in behaviour and ecological features among king- widely distributed of the three exclusively Asian species, with a lets, parids, penduline tits and long-tailed tits were discussed large breeding range from the western Himalayas (ssp. iredalei) (Löhrl, 1964) until molecular studies by Sturmbauer et al. (1998) to Myanmar and southeastern China including populations on Tai- rejected a suspected sister-group relationship between Regulus wan (concinnus subspecies group including ssp. manipurensis and and Leptopoecile in favour of a close relationship of the latter to talifuensis) and the distinctly grey-capped populations of southern Aegithalos (later confirmed by Alström et al. (2006)). Indochina (annamensis group; compare Harrap and Quinn, 1996; The only Indomalayan representative of the family is the small- Eck and Martens, 2006). The White-cheeked Tit, Ae. leucogenys,is est member of the Aegithalidae, one of the smallest passerines a rare endemic of central Asian dry scrubs and woodlands in parts worldwide: the Pygmy Tit, Psaltria exilis. It is endemic to Java and of Afghanistan and western Pakistan . The Chinese Sooty Tit, Ae. so far data on the breeding biology, ecology and behaviour of this fuliginosus, has a patchy distribution with small isolated mountain species are quite scarce.The Palearctic genus Aegithalos is the most populations in W China: Qinglin range, namely Taibai Shan and species-rich genus of the family. Depending on the authors and Daba Shan in S Shaanxi, Quionglai Shan in Sichuan and Gansu Prov- species concepts applied, six species (Harrap and Quinn, 1996; inces (Vaurie, 1957a,b; Cheng 1987, map on p. 901; Harrap and Eck and Martens, 2006), seven species (Dickinson, 2003) or even Quinn 1996, map on p. 436, see also Eck and Martens(2006)). nine species (Harrap in del Hoyo et al., 2008) are distinguished (Ta- So far taxonomy and systematics of the genus are entirely based ble 1). The Long-tailed Tit (Aegithalos caudatus) is the most wide- on external morphological character variation. Since all aegithalid spread species, covering a breeding area from the British Isles species have a rather reduced vocal repertoire and do not perform and Western Europe over the entire Eurasian taiga belt to Far East elaborate territorial songs, there is no reliable additional marker Siberia, Japan and all over Northeast to Southwest China with a system at hand from comparative bioacoustic studies such as are number of isolated area splinters in the southwest Palearctic. used for resolving taxonomic problems in other bird passerine Based on distinct differences in head pattern Ae. caudatus is tradi- groups (general review in Alström and Ranft, 2003). tionally divided into three subspecies groups: (a) the caudatus From previous molecular studies there are only preliminary group with pure white head lacking lateral crown stripes, (b) the data available. Zink et al. (2008) found two mitochondrial lineages europaeus group with distinct black crown stripes, otherwise as of Ae. caudatus, but these did not reveal any phylogeographic struc- in caudatus group, and (c) the alpinus group with head as in euro- ture throughout the entire Russian breeding range (only popula- paeus group, but cheek patches streaked darker, with dull mantle, tions of the caudatus subspecies group were included in this dark brown bib and dark eye lores (Vaurie, 1957a; Harrap and study except for one population from the Russian northwestern Quinn, 1996). The Chinese populations belonging to the third sub- Caucasus, alpinus-group). Some notable intraspecific mitochondrial species group, glaucogularis and vinaceus, were occasionally ranked differentiation was reported for Ae. concinnus with genetic cyto- as a separate species (or distinct fourth subspecies group): Ae. glau- chrome-b distances up to 6% between populations from the Hima- cogularis, ranging across the country from eastern Nei Mongol to layas (ssp. iredalei), W Myanmar (ssp. manipurensis) and SW China Hubei and Zheijang Provinces (del Hoyo et al., 2008; compare Ta- (ssp. talifuensis; Päckert et al., 2005; Eck and Martens, 2006). No ble 1, this paper). All other taxa of Aegithalos inhabit the Sino- further information on intra- and intergeneric differentiation and Himalayan mountain forests and adjacent regions like the Pamir phylogenetic relationships is available so far. massif in the West, across the Himalayan chain southwards to In this study we present a molecular phylogeny based on three Myanmar. To date there is a lively debate on the species status of mitochondrial and three nuclear marker genes for nearly all spe- some narrowly distributed SE Asian taxa as well as on their phylo- cies taxa of the family Aegithalidae distinguished by del Hoyo genetic relationships to other congeners. In the Handbook of the et al. (2008). Nevertheless, for consistency with our own results Birds of the World (del Hoyo et al., 2008, vol. 13) two of these were in this paper we follow the taxonomy according to Harrap and ranked as species of their own: The Burmese Tit, Ae. sharpei, is en- Quinn (1996)—compare Table 1, this paper—particularly with re- demic to Mount Victoria, SW Myanmar, and Père Bonvalot’s Tit, Ae. spect to the treatment of the forms bonvaloti and sharpei as subspe- bonvaloti, inhabits a small breeding range from NE Burma and adja- cies of the Black-browed Tit, Ae. iouschistos (Table 1).
Table 1 Taxonomy of long-tailed tits of genus Aegithalos according to different authors and species concepts ranging from six to nine species.
Taxon Dickinson (2003) Eck and Martens (2006) Harrap and Quinn (1996) del Hoyo et al. (2008) caudatus Ae. caudatus Ae. caudatus Ae. caudatus Ae. caudatus glaucogularis Ae. c. glaucogularis Ae. c. glaucogularis Ae. c. glaucogularis Ae. glaucogularis concinnus Ae. concinnus Ae. concinnus Ae. concinnus Ae. concinnus iouschistos Ae. iouschistos Ae. iouschistos Ae. i. iouschistos Ae. iouschistos sharpei Ae. b. sharpei Ae. n. sharpei Ae. i. sharpei Ae. sharpei bonvaloti Ae. b. bonvaloti Ae. n. bonvaloti Ae. i. bonvaloti Ae. bonvaloti niveogularis Ae. niveogularis Ae. n. niveogularis Ae. niveogularis Ae. niveogularis fuliginosus Ae. fuliginosus Ae. fuliginosus Ae. fuliginosus Ae. fuliginosus leucogenys Ae. leucogenys Ae. leucogenys Ae. leucogenys Ae. leucogenys 954 M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967
2. Materials and methods [fib7], transforming growth factor-beta 2 [TGFB2] and ornithine decarboxylase introns 6–7 [ODC]). Cytochrome-b sequencing was 2.1. Sampling and DNA extraction done for all samples available; for amplification and sequencing of the five additional genes we chose representative samples of For molecular analyses 71 samples from seventeen taxa of the each major mitochondrial lineage. Aegithalidae were available (origin of samples given in Table 2). Sample loans of North American Bushtits, P. minimus, and of Rus- 2.2. PCR settings and sequencing sian Long-tailed Tits, Ae. caudatus, were kindly provided by the Burke Museum of Natural History and Culture, Washington USA. Cytochrome b: A 1453-bp-long fragment comprising the whole Four of the latter Long-tailed Tit samples had already been used cytochrome-b gene and parts of ND2 gene was amplified in a dou- for the study by Zink et al. (2008) and could be referred to either ble-stranded PCR using the primer combination L14464-Cyt-b (50- of the two respective mitochondrial clades within Ae. caudatus (S. CTW GGC AGC ATT AYA GCA GG-30) and H15917-Cyt-b (50-TAG Rohwer, S. Birks, pers. commun.). ND2 sequences by Zink et al. TTG GCC AAT GAT GAT GAA TGG GTG TTC TAC TGG TT-30), PCR set- (2008) were available at GenBank and accordingly added to the tings followed Dietzen et al. (2003), as additional sequencing ND2 alignment. Because no fresh samples were available from primers we used L15087-Cyt-b, H15149-Cyt-b and H15547-Cyt-b Ae. niveogularis, Ae. i. iouschistos, Ae. i. sharpei and Ae. leucogenys (Edwards et al., 1991; Kocher et al., 1989). A fragment of approxi- we used tissue samples from museum specimens of these five taxa mately 2050 bp (varying with the number of indels) encompassing for DNA extraction and PCR (Table 2). When a specimen was sam- the entire 16S ribosomal RNA gene and partial t-RNA-Valine gene pled for a second time, small skin samples were taken from bare was amplified using the primer combination 12Sa (50-AAA CTG parts of the ventral skin edges of whole skins instead of toe pads GGA TTA GAT ACC CCA CTA T-30; Palumbi et al., 1991) and 16Sbr from the first sampling. When comparing PCR and sequencing re- (50-CCG GTC TGA ACT CAG ATC ACG T-30; Palumbi et al., 1991). sults yielded from toe pad and skin patch samples from the same Amplification of the 16S rRNA fragment was carried out according specimens, no apparent differences in the strength of PCR-bands to the protocol of Spicer and Dunipace (2004) with 94 °C for 45 s or quality of sequences could be observed. and 30 cycles of 92 °C for 45 s, 50 °C for 1 min and 72 °C for To our cytochrome-b data set we added fourteen further se- 2 min. Along with the PCR primers 12Sa and 16Sbr two additional quences from GenBank: Ae. caudatus trivirgatus from Japan internal primers 16S500 and 16Sa were used for sequencing. For (AB159169, -70) Ae. caudatus magnus from Korea (AB159171, -72), amplification of ND2 we used a newly designed forward primer Ae. caudatus ssp. (DQ792802, -03, DQ119539, -40, AY228044), Ae. (ND2 sequences from Zink et al. (2008)) AegND2_L19 (50-ATG concinnus iredalei from West Benghal (DQ008519), P. minimus AAC CCC CAA GCA AAA C-30) and a reverse primer H1064 given (AF074597), Leptopoecile sophiae (DQ008518) and Remiz pendulinus in Drovetski et al. (2004). Amplification and of beta-fibrinogene-7 (AY228081). For hierarchical outgroup rooting we used sequences intron was performed with the primer combination Fib-17S-L of Remizidae (Remiz pendulinus) and Timaliidae (Garrulax ellioti, and Fib-17S-U (van der Meij et al., 2005), however, sequencing Pnoepyga pusilla). was carried out using two newly designed forward and reverse Apart from material from our own collections we received fur- primers (due to long runs of single nucleotides at both the 30- ther specimens for morphological comparison from Museum für and the 50-end of the sequences): FibL121 (50-CCT GCC MRT GTA Naturkunde Berlin (ZMB) of the following taxa: Ae. i. iouschistos, CTG AAA TTT TC-30) and FibH836 (50-GTC AAA ATA TGT WAT RTC n = 26, Ae. i. bonvaloti (incl. ssp. sharpei), n = 20, Ae. fuliginosus, CCT GC-30). External primers OD6 (50-GACTCCAAAGCAGTTTGTC n =6,Ae concinnus, n = 38. For Palearctic Aegithalos species coordi- GTCTCAGTGT-30) and OD8r (50-TCTTCAGAGCCAGGGAAGCCACCAC nates of collection sites (genetic material and specimens) were im- CAAT-30) were used for PCR of ODC according to Allen and Omland ported to a shapefile in Diva-GIS 5.2.0.2 (Hijmans et al. 2005). (2003). External primers TGFB2-5f (50-GAAGCGTGCTCTAGAT Fig. 1 shows the distribution of all sample sites of fresh genetic GCTG-30) and TGFB2-6r (50-AGGCAGCAATTATCCTGCAC-30) were material (Palearctic Aegithalos species, locality data of GenBank se- used for PCR of TGFB2 according to Bures et al. (2002). quences included) with the most widespread breeding area of Ae. In order to avoid any kind of contamination, DNA extraction and caudatus (including the Chinese populations of the glaucogularis PCR from toe pad or whole skin samples was carried out in a sep- subspecies group) indicated. Because breeding areas of several SE arate clean lab and each step of analysis (sampling, extraction and Asian species are poorly known and regions of secondary contact PCR) was done under separate sterile benches. DNA from museum are under discussion we compiled the field records (breeding sea- specimens was extracted using a special forensic kit AGOWA-DNA- son) from the literature, from photographs provided at the Orien- easy-extraction kit according to the manufacturers’ instructions tal Bird Club Image Database (www.orientalbirdimages.org) and with small modifications; elution volume of 75 ll instead of from the specimens and samples available to us for the following 200 ll, digestion overnight). Using the sequence alignment that re- taxa (Fig. 2): Ae. iouschistos (coordinates inferred from Wunderlich sulted from DNA-analysis of fresh samples we designed several (1991b)), Ae. i. bonvaloti (and ssp. sharpei, coordinates inferred primer combinations for amplification of shorter gene fragments from Wunderlich (1991a)) and Ae. fuliginosus (breeding distribu- from toe pad DNA extracts (112–231 bp, depending on the quality tion according to Harrap and Quinn (1996, p. 436, fig. 108.1)). De- and age of template DNA). Amplification primer combinations for spite the assumptions that breeding ranges of these SE long-tailed short cytochrome-b fragments and further aDNA primer combina- tit taxa ‘‘overlap minimally or not at all” (del Hoyo et al., 2008) lo- tions for PCR of other marker genes are given in Appendix/Supple- cal contact zones were described between iouschistos and bonvaloti mentary material. Each PCR was performed in a 25 ll volume using east of the Brahmaputra bent at 96°E(Martens and Eck, 1995, 1–2 ll eluted DNA as template (4–6 ll of eluted aDNA), 3 ll of PCR p. 314; Harrap and Quinn 1996, p. 434, Fig. 107.1) and between buffer (including Mg), 1 ll of each forward and reverse primer, 1 ll the latter and fuliginosus at the Tibet-Sichuan borderland and in DNTp and 0.2 ll Taq-Polymerase (filled up to a reaction volume of large parts of Sichuan itself (Cheng, 1987; compare Harrap and 25 ll). PCR products were purified using ExoSap-IT (GE-Health- Quinn (1996)). care; adding 0.1 ll ExoSap-IT Solution in 4 ll H2O dest. to each DNA extraction from blood and tissue samples was performed sample; cycling program: 37 °C for 30 min and 94 °C for 15 min) in a chloroform-isoamyl isolation. From fresh samples we ampli- and sequenced in both directions on an ABI3700 DNA sequencer. fied three mitochondrial genes (cytochrome b, 16S rRNA and Sequencing of the PCR products was performed with BigDyeTM ND2) and three nuclear gene regions (beta-fibrinogen intron 7 v. 3.0 and v. 3.1 Dye Terminator Cycle Sequencing Kits (Applied M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967 955
Table 2 Blood and tissue samples used for molecular analysis; museum acronyms: MTD = Senckenberg Natural History Collections, Museum für Tierkunde Dresden, Germany; ZFMK = Zoologisches Forschungsmuseum Alexander Koenig, Bonn, Germany; UWBM = Burke Museum of Natural History and Culture, Washington, USA; cyt-b: number of haplotype (1–45); GenBank Accession Nos. listed for cyt-b sequences, further Accession Nos. for other genes (16S, ND2, fib7, ODC, TGFB2) listed in the text; if not stated otherwise numbers (No) refer to tissue collection J.M. at Mainz University.
No Species Subspecies Country Region Locality Voucher cyt-b Acc. No. further genes 98 Aegithalos caudatus Czech Bohemia, Šumava Nová Pec, SSE MTD 4 GU244453 (16S, ND2, caudatus Republic Volary C63024 Fib7) 96 europaeus Czech Bohemia, Šumava Nová Pec, SSE MTD 4 GU244448 Republic Volary C63017 1911 europaeus Czech Bohemia, Šumava Nová Pec, SSE 2 GU244466 Republic Volary 2412 europaeus Germany Baden-Württemberg Kehl-Straßbourg 1 GU244467 5189, 5191, europaeus Germany Rheinland-Pfalz Eich am Rhein 1 GU244471–74 5194, 5195 3058 europaeus Germany Baden-Württemberg Grafenhausen 2 GU244469 2819, 2820 europaeus Germany Rheinland-Pfalz Mainz- 5 GU244454, GU244468 Bretzenheim 3057 europaeus Germany Baden-Württemberg Grafenhausen 8 GU244457 5617 europaeus Germany Rheinland-Pfalz Eich am Rhein 10 GU244459 6197 europaeus Germany Rheinland-Pfalz Eich am Rhein 2 GU244475 540 Aegithalos taiti France Pyrénées-Orientales Serralongue MTD 7 GU244456 (five) caudatus C56788 541 taiti France Pyrénées-Orientales Serralongue MTD 1 GU244450 C56789 676, 677, 678, taiti France Cévennes Tarnon Valley 1 GU244449, 680, 518 GU244462–65 673, 675 taiti France Cévennes Tarnon Valley 6 GU244455, GU244461 3454 Aegithalos irbii France Corsica La Chiappa 11 GU244460 caudatus 3455 irbii France Corsica La Chiappa 1 GU244470 52653 Aegithalos caudatus Russia Magadanskaya Oblast UWBM 1 GU244447 caudatus 52653 60007 caudatus Mongolia Töv Aymag Ulanbataar UWBM 3 GU244452 (ND2, Fib7) 60007 64642 caudatus Russia Krasnordarskyi Kray UWBM 9 GU244458 64642 1152 caudatus Russia Primorskij Krai Ochotnicij 2 GU244451 (16S, ND2, Fib7) GenBank Aegithalos trivirgatus Japan Honshu Saitama 1 AB159170 caudatus GenBank trivirgatus Japan Honshu Hiroshima 2 AB159169 GenBank Aegithalos magnus Korea Gyenoggi-Do 2 AB159171 caudatus GenBank magnus Korea Gyenoggi-Do 2 AB159172 GenBank Aegithalos 1 AY228044 caudatus GenBank Aegithalos 2 DQ119539 caudatus GenBank Aegithalos 2 DQ792802 caudatus GenBank Aegithalos 3 DQ119540 caudatus GenBank Aegithalos 12 DQ792803 caudatus 7810 Aegithalos glaucogularis China Henan Dongzhai Nat 13 GU244477 caudatus Reserve 7809 glaucogularis China Henan Dongzhai Nat 14 GU244478 Reserve 7811 glaucogularis China Henan Dongzhai Nat 15 GU244479 Reserve 7813 glaucogularis China Henan Dongzhai Nat 16 GU244476 Reserve 7812 glaucogularis China Henan Dongzhai Nat 13 GU244481 (ND2) Reserve 7814 glaucogularis China Henan Dongzhai Nat 13 GU244480 (five) Reserve 2068 Aegithalos talifuensis China Sichuan Omei Shan: MTD 17 AY755556 (five) concinnus Hongchun C59795 3292 talifuensis China Sichuan Longxi-Hongkou 17 GU244442 (16S, ND2, reserve Fib7) 7802 Aegithalos concinnus China Henan Dongzhai Nat 18 GU244421 concinnus Reserve
(continued on next page) 956 M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967
Table 2 (continued)
No Species Subspecies Country Region Locality Voucher cyt-b Acc. No. further genes 7803 concinnus China Henan Dongzhai Nat 19 GU244422 Reserve 7804 concinnus China Henan Dongzhai Nat 20 GU244423 Reserve 7806 concinnus China Henan Dongzhai Nat 21 GU244424 Reserve 7808 concinnus China Henan Dongzhai Nat 22 GU244425 Reserve 7805, 7807 concinnus China Henan Dongzhai Nat 17 GU244440, GU244441 Reserve 4101 Aegithalos iredalei Nepal Rasuwa District W Syabrubesi, 1. 23 GU244427 concinnus camp 4128 iredalei Nepal Rasuwa District W Syabrubesi, 1. 23 GU244426 (five) camp 2702 iredalei Nepal Bhojpur District Irkuwa Khola 24 AY755555 (ND2) GenBank iredalei India W Benghal NRM 26 DQ008519 20046818 4722 Aegithalos manipurensis Myanmar Chin State Natmataung 25 GU244428 (five) concinnus National Park 1974 Aegithalos Aviary 17 GU244419 concinnus 3087 Aegithalos China Aviary, import, exact MTD 17 GU244420 concinnus locality unknown C63026 706 Aegithalos China Shaanxi Taibai Shan, MTD 27 GU244429 (16S, ND2, fuliginosus Houzhenzi C56721 Fib7) 809 China Shaanxi Taibai Shan, 27 GU244444 (five) Houzhenzi 814 China Shaanxi Taibai Shan, 28 GU244430 Houzhenzi 808, 812, 813 China Shaanxi Taibai Shan, 27 GU244443, Houzhenzi GU244445–46 4814 Aegithalos iouschistos Nepal Rasuwa Distr. Gosainkund, Syng MTD 29 GU244431 iouschistos Gyang C58573 4815 iouschistos Nepal Ramechap Distr. Chordung Mt. near MTD 30 GU244432 (16S, Fib7) Jiri C58574 3232 Aegithalos bonvaloti China Yunnan Jizu Shan MTD 31 GU244433 (five) iouschistos C63550 56561 Aegithalos sharpei Myanmar Mt. Victoria Pakokku MTD 32 GU244434 iouschistos C56561 50885 sharpei Myanmar Mt. Victoria Pakokku MTD 33 GU244435 (ODC) C50885 3200 Aegithalos Nepal Dolpo Distr. Gompa near ZFMK 34 GU244436 niveogularis Tarakot 71.1075 3951 Aegithalos Nepal Dolpo Distr. Ringmo/Lake ZFMK 35 GU244437 (16S, ND2, niveogularis Phoksumdo 71.1074 Fib7, ODC) 3953 Aegithalos Afghanistan Sinsoi/Nuristan ZFMK 36 GU244438 leucogenys Nieth. 3955 Aegithalos Afghanistan Dar-e-Nur ZFMK 37 GU244439 (16S, ND2, leucogenys Nieth. Fib7, ODC) 56377 Psaltriparus plumbeus USA Colorado Fremont C., Camon UWBM 38 GU244416 (five) minimus City 56377 77637 plumbeus USA New Mexico Hidalgo C., Animas UWBM 39 GU244417 77637 77647 plumbeus USA Arizona Cochise C., Portal UWBM 40 GU244418 77647 79247 Psaltriparus minimus USA Washington Pacific C., UWBM 41 GU244415 (five) minimus Grayland 79247 GenBank minimus USA 41 AF074597 5767 Leptopoecile China Sichuan Bang Ba village 42 GU244414 (five) elegans 2117 Leptopoecile Aviary 43 GU244482 (five) sophiae GenBank Leptopoecile 44 DQ008518 sophiae 5636 Remiz Germany Rheinland-Pfalz Eich am Rhein 45 GU244483 (five) pendulinus
Biosystems) according to the manufacturers’ instructions. used for the analysis were deposited at GenBank under the Acces- Reactions were electrophoresed with the ABI 377 automatic sion Nos. GU244413–GU244483 (cytochrome b, see Table 1), sequencer. GU433965–GU433986 (16S rRNA), GU433942–GU433964 (ND2) The sequences were aligned by ClustalW using MEGA 3.1 GU433987–GU434005 (fib7), GU434021–GU434035 (ODC) and (Tamura et al., 2007) and slightly adjusted by eye. All sequences GU434006–GU434020 (TGFB). M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967 957
Fig. 1. Origin of blood and tissue samples used for genetic analysis (including cyt-b-sequences drawn from GenBank from four SE Asian localities); breeding area of North Palearctic Aegithalos caudatus including the Chinese vicariant subspecies group glaucogularis indicated by dotted lines.
Fig. 2. Breeding records of Sino-Himalayan (species) taxa of the Ae. iouschistos group sensu Harrap and Quinn (1996), dashed line, and of Ae. fuliginosus, dotted line according to Harrap and Quinn (1996, p. 436, Fig. 108.1) symbols: triangles = Ae. i. iouschistos; circles = Ae. i. bonvaloti (incl. sharpei); squares = Ae. fuliginosus; data sources, colour: light grey = data drawn from Wunderlich (1991a,b) (Chinese race obscuratus was included in iouschistos) and Harrap and Quinn (1996) for fuliginosus; dark grey = field records from orientalbirdclub.images.com; black = voucher specimens and genotyped samples, white symbols = winter birds.
2.3. Data analysis (Nylander, 2004). According to the Akaike information criterion the best fit model for the cytochrome-b sequence data set was The appropriate substitution model for each of the six sequence GTR + I + G with the following likelihood settings (the same set- data sets was estimated for each of the six genes analysed using tings were estimated for likelihood and Bayesian analysis): empir- Modeltest 3.04 (Posada and Crandall, 1998) and MrModeltest ical base frequencies: pA = 0.2777, pC = 0.3976, pG = 0.1236, 958 M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967 pT = 0.2011; proportion of invariable sites I = 0; gamma shape We performed two different runs with the cytochrome-b data parameter a = 0.1652; rate matrix: R(a)[A-C] = 2.1260, R(b)[A- set: in the first one we applied a mean substitution rate of 2.1% be- G] = 12.3488, R(c)[A-T] = 1.7314, R(d)[C-G] = 0.1830, R(e)[C- tween lineages per my (r = 0.0105 according to Weir and Schluter T] = 25.4000, R(f)[G-T] = 1.0000. For all five other gene data sets (2008)), who re-evaluated about 90 different avian sequence data the GTR model was estimated the best fit model, too (for model sets) to our cyt-b data without constraining the age of any node. settings see Supplementary material). In a second run an age constraint was assigned to the node defining Phylogenetic trees were reconstructed using Maximum Parsi- the genetic lineage split between Nearctic Psaltriparus and Palearc- mony (MP) and Maximum Likelihood as implemented in PAUP* tic Aegithalos referring it to a faunal interchange via the Bering 4.0b10 (Swofford, 2002) and Bayesian inference of phylogeny with Land Bridge. When genetic lineage splits among taxa from both MrBayes 3.1.2 (Huelsenbeck and Ronquist, 2001). MP and ML anal- sides of the land bridge were referred to a relatively narrow time yses were performed in a heuristic search with TBR branch swap- constraint of the last Pliocene opening of the Bering Strait (5.5– ping option (104 rearrangements) and gaps coded as fifth character 4.7 my BP: Marincovich et al., 2002), age estimates were always state in ML. Clade support in MP analysis was estimated by 1000 associated with the upper age constraint (Päckert et al., 2006, bootstrap replicates (Felsenstein, 1985) in a fast heuristic search 2007). Assuming that according to these estimates the split be- with all characters unordered and equally weighted and gaps trea- tween avian Nearctic and Palearctic sister clades was not caused ted as fifth character state. ML bootstrap support was obtained by by a disruption of a formerly widespread distribution area of a 100 bootstrap replicates in GARLI 0.951 (Zwickl, 2006) with the common ancestor but presumably occurred via dispersal during model parameter raw string defined within a separate garli nexus times when the Bering Land Bridge still existed, we used a broader block file. By default settings, branch length optimization was per- time constraint in this study. For the Psaltriparus-Aegithalos node formed with a threshold of 2 104 generations per bootstrap rep- we set the prior distribution = uniform with an upper bound = 10.0 licate. However, constant ln L values were reached already after a my referring to the beginning of a major faunal interchange via few hundred generations and the allowed minimum value of opti- Beringia (Hopkins, 1967) and a lower bound = 4.7 my referring to mization precision was reached after approximately 5500 genera- the most recent estimate for the Pliocene opening of the land tions in different replicates. Therefore, the parameter bridge, i.e. the interruption of the faunal interchange (Marincovich
The Bayesian haplotype tree is shown in Fig. 3 with each haplo- atus2: n = 8) is spread over the entire distribution area from wes- type represented once and with bootstrap support from ML and tern Europe to eastern Russia, and each of these represents the MP reconstructions indicated for all relevant nodes. The deeper centre of a star-shaped haplotype group (Fig. 4A). The two central splits among the three genera are poorly resolved by cyt-b se- haplotypes of each cluster are separated by two mutations only. A quence data alone, only Bayesian posterior probabilities weakly third haplotype cluster representing the E Chinese Silver-throated support a sister-group relationship of Psaltriparus and Aegithalos. Tit, Ae. c. glaucogularis, is separated by eleven substitutions from Monophyly of genus Aegithalos itself was strongly supported by haplotype caudatus2(Fig. 4A). Phylogenetic relationships between Bayesian analysis but only moderately/poorly supported by ML/MP the two Ae. caudatus clades and the glaucogularis haplotype cluster analysis. are not resolved regardless of the method of tree reconstruction. The long-tailed tit clade (genus Aegithalos) is trifurcated, com- Pairwise comparisons between haplotypes of Ae. caudatus range prising two Sino-Himalayan and one N Palearctic subclade. Within between 0.01 and 1% (mean = 0.4%), those between N Palearctic the latter (clade 1, Fig. 3) only slight genetic variation was found caudatus and Chinese glaucogularis range from 1.6% to 2.0%. The among 12 cytochrome-b haplotypes of N Palearctic Ae. caudatus. pooled data set of ND2 sequences by Zink et al. (2008) available Each of the two most common haplotypes (caudatus1: n = 15, caud- from GenBank and our own ND2 sequences confirms the phyloge-
Fig. 3. Molecular phylogeny of Aegithalidae based on 838 bp cytochrome b; Bayesian inference of phylogeny, 1,000,000 generations, each haplotype represented by one sequence only; node support from Bayesian posterior probabilities and from MP and ML bootstrap support indicated at the according nodes (full node support of 1.00/100/ 100 indicated by asterisk). 960 M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967
Fig. 4. Cytochrome-b haplotype networks of (A) Ae. caudatus incl. Chinese subspecies group glaucogularis, 983 bp cyt-b; (B) Sino-Himalayan Rufous-fronted Tits, White- throated Tits and Sooty Tits (Ae. i. iouschistos, Ae. i. bonvaloti, Ae. i. sharpei, Ae. niveogularis and Ae. fuliginosus), 940 bp cyt-b.
ographic structure of the N Palearctic long-tailed tit group: Two all members of this clade is shown in Fig. 4B). The most common common and widespread haplotypes (n = 47 and 23) are separated SE Asian haplotype (27), found in four specimens of Ae. fuliginosus from each other by seven substitutions and each of them from one from Shaanxi, China, is the centre of a starlike haplotype cluster, glaucogularis haplotype by eleven and sixteen substitutions respec- with all other haplotypes of the three related taxa differing by tively (network not shown). three to six base substitutions only from the central one. Pairwise In all tree reconstructions Ae. caudatus represents the sister comparisons range at lowest distances of 0.5% between Ae. fuligino- group to another clade comprising five different south-east Asian sus and all Ae. iouschistos ssp., between group distances among the taxa (Fig. 3, clade 2). Within the latter clade Ae. niveogularis is sister latter group and Ae. niveogularis range from 2.6% to 3.1% (Table. 3; to a strongly supported monophyletic group comprising all haplo- 27 substitutions in the haplotype network, Fig. 4B). types of Ae. fuliginosus, Ae. i. iouschistos, Ae. i. bonvaloti and Ae. i. Clade 3 of the Aegithalos tree unites the two sister taxa Ae. con- sharpei. Phylogenetic relationships among the latter four are not cinnus and Ae. leucogenys (Fig. 3). Unlike clades 1 and 2, Ae. concin- well resolved in any reconstruction. The haplotype network for nus shows a strong phylogeographic structure comprising four
Table 3 Uncorrected p-distances based on 838 bp of the cytochrome-b gene between taxa of Aegithalidae; values for members of the Ae. iouschistos clade marked with grey shade.
*Ae. c. concinnus includes the sequences of SW Chinese ssp. talifuensis. M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967 961 separate and deeply split cyt-b lineages. Populations of the Hima- 11.4%, and even within North American Psaltriparus two clades layas (ssp. iredalei) appear to be clearly separated into an eastern representing minimus and plumbeus subspecies groups are consid- and a western clade, monophyly of the Himalayan populations is erably diverged (genetic distance: 3.0%; Table 3, Fig. 3). only moderately supported (BI: 0.89, ML: 73) and poorly supported In two nuclear genes several insertions and deletions could be by MP analysis (<50). The Himalayan haplotype lineage is sister to observed in single lineages only. ODC: one 16-bp long insertion in one deeply separated haplotype from Myanmar (ssp. manipurensis) Ae. leucogenys and another one of the same length in Ae. c. concin- in all reconstructions, although with likewise poor support. This nus, another 17-bp long insertion in L. sophiae plus a 20-bp long entire Himalayan-Burmese clade is sister to a fourth haplotype deletion shared by Ae. c. iredalei and Ae. c. manipurensis (though lineage from W and SC China (ssp. talifuensis and nominate concin- not found in nominate concinnus!). TGFB: one 36-bp long inser- nus; haplotype 17 shared by these two subspecies). Monophyly of tion in L. sophiae. Two nuclear gene networks (fibrinogen intron7 all Ae. concinnus haplotypes receives poor support from Bayesian and TGFB) for all Aegithalidae are shown in Fig. 5. In both net- posterior probabilities and moderate support from ML and MP works the three aegithalid genera are separated from each other bootstrap. Between group genetic distances range from 4.3% be- by 16–27 substitutions (fib7) and 25–62 substitutions (TGFB2), tween the two Himalayan clades to 5.4–5.8% between ssp. iredalei respectively. The three main mitochondrial clades of Aegithalos and ssp. manipurensis and up to 5.3–5.8% between the latter are well reflected by distinct clusters in the fib7 network showing two and the talifuensis–concinnus clade (Table 3). Sister-group a close relationship between the Palearctic species (Ae. caudatus) relationship between western Himalayan Ae. leucogenys and all and the subalpine Sino-Himalayan species (Ae. iouschistos, Ae. members of Ae. concinnus receives strong support in all reconstruc- fuliginosus and Ae. niveogularis). Within the latter group Ae. iou- tions. schistos and Ae. niveogularis share the same TGFB and OCD geno- The two Sino-Himalayan Tit-warbler species, Leptopoecile so- type and only Ae. fuliginosus stands out having a distinct genotype phiae and L. elegans, are separated by a large genetic distance of of these two nuclear genes. The relationships between Ae. concin-
Fig. 5. Nuclear gene networks for the aegithalid genera Aegithalos, Psaltriparus and Leptopoecile generated with TCS 2.1; (A) fibrinogen intron7, 444 bp; (B) TGFB2, 567 bp, one 36-bp long indel in L. sophiae was deleted from the alignment prior to analysis; caudatus includes glaucogularis (same fib7 genotype as cau4); iouschistos includes sharpei, bonvaloti and niveogularis (all have the same TGFB2 genotype). 962 M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967 nus and Ae. leucogenys are ambiguous in the TGFB and ODC net- 3.2. Age dating works due to the effect of large indels mentioned above (Fig. 5). In the ODC network (not shown) Ae. leucogenys represents the Mean ages estimates for Aegithalos lineage splits inferred from closest relative only of Chinese Ae. concinnus (ssp. talifuensis and two independent runs with the cty-b data set using either a fixed concinnus) separated by 35 substitutions) and both differ from mean substitution rate or fixed node ages for calibration differed Himalayan-Burmese Ae. concinnus (ssp. iredalei and manipurensis) considerably. Estimates based on a mean substitution rate of by 50–55 substitutions. Bayesian inference of phylogeny based on 0.0105 yielded older tmrca values especially for the deeper splits three concatenated nuclear intron sequences (1662 bp) shows of the aegithalid tree compared to those estimates based on a sin- largely the same branching pattern as the cyt-b tree, however, gle calibration point from runs with the cyt-b and the concatenated with weaker support for the younger nodes but with strong sup- sequence data respectively (confidence intervals shown in Figs. 6 port for the basal splits (tree not shown). Monophyly of Aegithalos and 7). Nevertheless, the two time scenarios suggest similar suc- and sister-group relationship to Psaltriparus is strongly sup- cessive separation events triggered by Pleistocene climate on the ported (posterior = 1.00). N Palearctic caudatus (with sister clade one hand and earlier Pliocene lineage splits on the other. Three glaucogularis) form a well supported clade with Sino-Himalayan most recent separation events were dated to the late Pleistocene niveogularis-iouschistos, however, with poor interior support for era in both calibrations: (i) differentiation among all popluations sister-group relationship of the latter two taxa. Unlike in the of N Palearctic Ae. caudatus; (ii) separation of ancestral stock of cyt-b tree the nuclear phylogeny does not confirm the monophyly caudatus and E Chinese glaucogularis subspecies groups; (iii) sepa- of Ae. concinnus, but strongly supports a sister-group relationship ration of SE Asian populations from the Ae. iouschistos haplotype of Ae. leucogenys and Ae. concinnus iredalei (posterior = 1.00). The cluster (including the forms bonvaloti, sharpei and even fuliginosus). entire Ae. concinnus/Ae. leucogenys clade receives strong support Confidence intervals of tmrca for three further splits range at the (posterior = 1.00), however, the interior topology of this clade early Pleistocene-Pliocene boundary: (i) East–West split within with respect to the position of ssp. manipurensis and concinnus re- the Himalayan range between the entire Ae. iouschistos clade and mains unresolved. W Himalayan Ae. niveogularis; (ii) a second E–W Himalayan split The concatenated sequence data set of 3995 bp included frag- between populations of Ae. concinnus iredalei from Nepal and from ments of all six genes investigated (cytb: 983 bp; 16S rRNA: 734; W Benghal; (iii) the split between NW and SC Nearctic Psaltriparus ND2: 616; fib7: 371 bp; ODC: 688 bp; TGFB: 603 bp). Like in the of the minimus and the plumbeus subspecies groups. Three further nuclear tree monophyly of all members of Aegithalidae is strongly splits were dated to the Pleistocene–Pliocene boundary in the age- supported and phylogenetic relationships among the three genera constraint calibration and to Pliocene and even earlier in the rate- are well resolved by the multilocus data set (Fig. 6). Aegithalos and constraint calibration: (i) the split among all subspecies of SE Asian Psaltriparus are sister taxa with strong support and Leptopoecile is Ae. concinnus; (ii) the split between all members of the N Palearctic sister of the latter two representing an old and basal split from Aegithalos clade (Ae. caudaus incl. glaucogularis group) and the the aegithalid clade. Other than in the nuclear tree, monophyly Sino-Himalayan sister clade (Ae. iouschistos group and Ae. niveogu- of all Ae. concinnus and sister-group relationship of these to Ae. leu- laris); (iii) the split between C Asian Ae. leucogenys and SE Asian Ae. cogenys receives strong support. Intraspecific phylogenetic rela- concinnus (the latter representing the oldest age estimate for the tionships among subspecies groups of Ae. concinnus are still separation of Aegithalos species pairs, confidence interval based poorly resolved in the six-genes tree. on mean rate of 0.0105, CI = [5.9–8.0 my BP]).
Fig. 6. Molecular phylogeny of Aegithalidae based on concatenated partial mitochondrial and nuclear gene sequences (cytb, 16S rRNA, ND2, Fib7, ODC, TGFB2: 3995 bp); Bayesian inference of phylogeny, model parameters unlinked across all six partitions (GTR model, separate estimates for each partition); relaxed clock calibration performed with BEAST, confidence intervals for age estimates indicated by grey bars at according nodes. M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967 963
Fig. 7. Times intervals for the most recent common ancestor (tmrca) of aegithalid taxon pairs based on cytochrome-b data (each mtc lineage represented by one sequence; 95% confidence intervals estimated with BEAST, relaxed clock); black lines: estimates based on an empiric substitution rate of 0.0105 substitutions per site per lineage per my; grey bold lines: estimates based on one calibration point for the split between Nearctic Psaltriparus and Palearctic Aegithalos (prior uniform with upper bound = 10.0 my and lower bound = 4.7 my); Pliocene–Pleistocene boundary indicated by horizontal line at 2.4 my BP according to West (1988).
4. Discussion tain systems of south-western China and Myanmar (extant populations of bonvaloti and sharpei) as well as into the Himalayas 4.1. Phylogeography and evolutionary time scale (extant iouschistos). There, throughout its distributional range Ae. iouschistos established secondary contact (locally in vertical parap- Our phylogenetic and phylogeographic reconstructions reveal atry) with Ae. concinnus (ssp. iredalei) who then had already settled that a large number of morphologically diagnosable taxa within the Himalayas since the Early/Mid Pliocene (for horizontal allopa- Palearctic Aegithalidae actually diversified only recently during try and a distribution gap in mid-west Nepal between ranges of Ae. Pleistocene times. At an undoubtedly intraspecific level our mito- niveogularis and Ae. iouschistos, see Martens and Eck (1995, chondrial and nuclear sequence data confirm the recent, presum- Fig. 100)). According to our molecular dating and regardless of ably Late Pleistocene separation of two ancestral populations of the method applied, the (sub)tropical faunal element of this pas- extant Palearctic Ae. caudatus, as already suggested by Zink et al. serine group, Ae. concinnus, invaded the Himalayas considerably (2008). Presumably, at least two Late Pleistocene refuges existed earlier than their Palearctic counterparts (Ae. iouschistos and Ae. from which Long-tailed Tits rapidly dispersed throughout entire niveogularis) and even underwent noticeable East–West diversifi- Eurasia and completely mixed, but a localization of these hypoth- cation within the Himalayas (split within Ae. concinnus iredalei). esized refuges based on the genetic data available remains crucial. These findings are consistent with the phylogeographic recon- A likewise complete dissolution of former genetic differentiation structions by Johansson et al. (2007) for SE Asian Old World War- without any extant phylogeographic structure is a rare case in blers (Phylloscopus, Seicercus) suggesting that the present patterns Palearctic passerines. Several studies confirm a strong phylogeo- of sympatry or vertical allopatry in the Himalayas did not result graphic structure within N Palearctic passerine species including from diversification/speciation within the mountain range but several marginal disjunct populations from C, SE or FE Asia (Willow from successive dispersal into the Himalayas from SE Asian refuges Tits, Poecile montana, Great Tits, Parus major: Kvist et al. (2001, (compare Price et al. (2003)). Their inference might be generalized 2003); Rosefinches, Carpodacus erythrinus: Pavlova et al. (2005); for sympatry of closely related Himalayan passerines from several Winter Wrens, Troglodytes troglodytes, Drovetski et al. (2004)). families. However, speciation in situ in the Himalayas is not a rare Slightly prior to the separation of the two genetic clusters of Ae. event and resulted in closely related allopatric species or strongly caudatus during Mid to Early Pleistocene times a S Chinese refuge differentiated subspecies (Martens et al., 2010; Päckert et al., area might have harboured founders of extant glaucogularis popu- 2009). lations. At a similar level of genetic differentiation, close phyloge- On the North American continent genetic differentiation pre- ographic affinities to Far East Russian populations were already sumably driven by glacial cycles was found between NW minimus described for several Old World warblers from the NW Chinese subspecies group and SW plumbeus subspecies group of the Bush- mountain systems, e.g. Phylloscopus trochiloides obscuratus (Irwin tit. Based on our preliminary results further genetic differentiation et al., 2001), Phylloscopus kansuensis (Martens et al., 2004) and would be expected with with respect to isolated or marginal pop- Phylloscopus fuscatus robustus (Martens et al., 2008). ulations, e.g. the morphologically distinct ‘‘Black-eared Bushtits” of The nowadays patchily distributed populations of the Sino- the melanotis subspecies group from Mexico and Guatemala Himalayan iouschistos group apparently originated from a single (Harrap and Quinn, 1996; del Hoyo et al., 2008) and thus further and presumably very restricted late Pleistocene refuge area that sampling and investigation is necessary in order to fully evaluate according to the cyt-b haplotype network was located in SW China the intraspecific differentiation of P. minimus. However, similar ge- (area largely that of extant Ae. fuliginosus). From there a rapid late- netic divergence dating back to separate Pleistocene refuges was pleistocene expansion (along with short phases of geographic iso- also found in other North American passerines like the Winter lation during late glacial cycles) took place into the adjacent moun- Wren, Troglodytes troglodytes (Toews and Irwin, 2008), Carolina 964 M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967 chickadees, Poecile carolinensis (Gill et al., 2005), the Golden- species group together with similarly dull forms showing a dis- crowned Kinglet, Regulus satrapa (Päckert et al., 2008) and a num- tinctly contrasting black bib, such as the Near East Asian ber species pairs from different passerine families (Weir and Sch- tephronotus from Turkey and passeki from western Iran (Vaurie, luter, 2004; Smith et al., 2005). 1957aHarrap and Quinn, 1996; colour plates in the latter work and in del Hoyo et al. (2008)). 4.2. Intra- and interspecific variation: molecular genetics vs. morphology 4.2.2. SE Asian Rufous-fronted Tits (Ae. iouschistos sensu Harrap and Quinn, 1996) and allies (clade 2) Molecular phylogenies do often conflict with traditional avian With respect to their external appearance all these taxa share a systematics particularly when a focus on morphological diagnos- common head pattern with dark eye lores merging with the broad ability of species taxa had apparently led to an over-splitting of lateral crown stripes, brownish ear-coverts, black moustachial widely distributed Palearctic bird species on the one hand or estab- stripes contrasting with washed greyish bib and brownish-cinna- lished paraphyletic species taxa on the other hand. For instance, mon breast, flanks and underparts (only bonvaloti with whitish the tits and chickadees (Paridae) provide a perfect example for re- underparts; compare Harrap and Quinn (1996), del Hoyo et al. peated misinterpretation of morphological character variation (2008)). Surprisingly, our specimens of Ae. fuliginosus from Shaanxi within and among closely related species (review in Eck and Mar- Province fall into the Ae. iouschistos haplotype cluster, too, and tens, 2006, pp. 23–28; Päckert and Martens, 2008). Within the N notably, these specimens were recorded far northward of the Palearctic range the situation as we found in Ae. caudatus is best breeding range of bonvaloti and even northward of the northern paralleled by the Willow Tits (Parus montanus): Despite a high hap- range boundary of fuliginosus given in Cheng (1987); see also Eck lotype diversity there is no marked genetic differentiation among and Martens (2006). Remarkably, the genetic differentiation within all morphologically distinctive subspecies groups of the N Palearc- this E-Himalayan and SW Chinese taxon group is similar to that be- tic , while only the songarus subspecies group constitutes even two tween the two haplotype clusters of Ae. caudatus and is likewise separate C Asian and C Chinese lineages (Salzburger et al., 2002; considerably lower than any interspecific differences among cur- Kvist et al., 2001). Similarly, patterns of genetic and morphological rently accepted avian sister species (Tavares and Baker, 2008). variation are not congruent among species of North American rosy- Even a hybrid zone between fuliginosus and bonvaloti in the wild finches of genus Leucosticte (Drovetski et al., 2009), among subspe- was already postulated on grounds of a series of voucher speci- cies of Motacilla flava and M. citreola (Pavlova et al., 2003) and mens (Kleinschmidt and Weigold, 1922; Birckhead, 1937). Espe- among subspecies of Palearctic Bluethroat, Luscinia svecica (Zink cially two bonvaloti specimens from Wenchuan were reported to et al., 2003; compare Zink (2004)). Most species of Aegithalidae, represent a transiton form (‘‘...bildet einen Übergang...”) with especially those of the genus Aegithalos itself, are morphologically fuliginosus, morphological details were described and taxonomic diverse in such a way that taxonomic importance of characters of consequences were tentatively drawn (Kleinschmidt and Weigold, plumage colouration and/or colour pattern might easily be overes- 1922). We revisited one of the named specimens from Wenchuan timated with respect to species status of or phylogenetic affinities (MTD C23887) and confirm its intermediate phenotype: its overall between morphologically distinct taxa. plumage colouration and black lateral crown stripes of the male bird matches the characters of bonvaloti, however, its silver-grey 4.2.1. North Palearctic Long-tailed Tits (clade 1) facial patches and ashy-grey throat without any black moustachial Ae. caudatus shows a highly variable plumage pattern through- stripe rather match the fuliginosus phenotype. From the same loca- out its range with respect to black lateral crown stripes (e.g. these tion also one specimen of clear fuliginosus phenotype specimen are not present in any member of the caudatus subspecies group), was available to us (MTD C23882). pinkish colouration of shoulder patches and underparts, contrast of Moreover, in the named region the doubtful subspecies Ae. iou- cheek patches and bib (for colour plates see Harrap and Quinn schistos obscuratus occurs more or less sympatrically with bonvaloti (1996, p. 101, plate 33), del Hoyo et al. (2008, p. 95, plate 2)). Strik- and fuliginosus, i.e. (described by Mayr (1940) from northern Sich- ingly, morphologically distinct subspecies groups of Ae. caudatus uan, Sungpan Distr., Chengou Forks; see Fig. 2, this paper). Due to do not represent distinct genetic units, because each of the two its intermediate rather dull and dark colouration the latter form haplotype clusters comprises individuals from the europaeus and was sometimes regarded as approaching Ae. fuliginous in some re- caudatus group (and even from the alpinus-group; two specimens spects or as intermediate between the latter and bonvaloti (del from Corsica) and these findings are paralleld by the geographic Hoyo et al., 2008). Details on the situation in the field, however, variation of nuclear genes, too. Vaurie (1957a) described zones of are still lacking and should be further investigated. However, based intergradation between the caudatus and the europaeus groups in on his own comparative analyses, Vaurie already formally agreed E Europe and for the alpine foothills of northern Italy between with Snow’s diagnosis (cited in Vaurie, 1957a) that iouschistos, the europaeus and the alpinus groups. So far, there is no plausible sharpei, bonvaloti (and niveogularis!) ‘‘replace one another geo- explanation why and how these narrow zones of intergradation graphically (...) and the differences between them, although strik- between subspecies groups remained stable over time and how ing, are not much more striking than in the three groups of distinct phenotypes retain their integrity against each other, subspecies in Ae. caudatus”. though there is obviously considerable gene flow between them. A similar situation was found in the Carrion Crow and the Hooded 4.2.3. SE Asian Black-throated Tits and allies (clade 3) Crow (Corvus c. corone, C. c. cornix) where bidirectional mitochon- Surprisingly Ae. concinnus is the only SE Asian species showing drial introgression extends far beyond the E and W boundaries of high intraspecific phylogeographic differentiation—up to 5.9% ge- the relatively narrow hybrid zone of distinct differently coloured netic distance (cytb) between Ae. c. manipurensis (white-bellied) morphotypes (Haring et al., 2007). from Myanmar and populations from SW China (Ae. c. talifuensis Among Long-tailed Tits only the Chinese glaucogularis group and Ae. c. concinnus: buff-bellied; all three subspecies rufous-red- (including ssp. vinaceus) was occasionally considered highly dis- dish crowned; compare Harrap and Quinn (1996), del Hoyo et al. tinctive for having a dark sooty bib, dark brown eye lores and dull (2008)). Compared to other Aegithalos species pairs the intraspe- beige-brown breast and underparts—and its distinctiveness is well cific differentiation within Ae. concinnus is three times as high as corroborated by our genetic findings. So far, these two Chinese taxa the genetic differentiation among West and East Himalayan Ae. were often included in the otherwise West Palearctic alpinus sub- niveogularis and Ae. i. iouschistos on the one hand and as that M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967 965 between all haplotypes of N Ae. c. caudatus and Chinese Ae. c. glau- suggest a treatment of bonvaloti and sharpei as subspecies of the cogularis on the other (about 2% genetic distance for the latter two Rufous-fronted Tit, Ae. iouschistos (Blyth, 1844) in accordance with taxon pairs; morphological diagnosability of taxa given in all Harrap and Quinn (1996) and Clements (2007). Genetic distances cases). So far, no genetic data are available for the distinctive between the latter and Ae. fuliginosus range considerably below grey-headed taxa of the annamensis subspecies group from Thai- species level and probably all these taxa in question represent just land, Vietnam and Cambodia. Only further studies and sampling distinct morphotypes at an intraspecific differentiation level (just in the southern range of this species will complete its phylogeo- like in the case of Ae. caudatus s. str.). graphic pattern. The close relationship of Ae. concinnus and Ae. leu- We do not give explicit taxonomic recommendations for Ae. cogenys as confirmed by our molecular data was already suggested concinnus, because genetic data for the SE Asian annamensis-group on grounds of morphological characters by Eck (1996). are still missing. However, we point out that genetic differentiation Aegithalos vocalisations are little documented, because they are between Himalayan, Burmese and Chinese subspecies of Ae. con- low, inconspicuous and there is no far-ranging territorial song. cinnus is large and equals or even notably exceeds that between However, calls so far recorded seem to be surprisingly similar even other species pairs within Aegithalos listed above. We furthermore among taxa not closely related according to the information gained point out that for the populations of the E Himalayas the name from molecular genetic analyses. Contact calls of Ae. leucogenys rubricapillus (Aegithaliscus concinna rubricapillus Ticehurst, 1925; from Afghanistan (Löhrl and Thielcke, 1969), Ae. iouschistos and type locality: Sikkim) is available. For a presumed lack of distinc- Ae. concinnus (both from Nepal: Martens and Eck, 1995) are trill- tiveness this taxon was later subsumed under the subspecies ireda- like series of 4–10 inverted ‘v’-like notes with steep up- and down- lei which was until then restricted to the W and C Himalayas (type strokes. Similar though slightly differing notes are also produced locality restricted to Shimla by Baker (1920); compare Dickinson by Ae. caudatus (Bergmann et al., 2008; Perrins, 1993). Common et al. (2006) and Eck and Martens (2006); for recognition of rubri- to Ae. iouschistos, Ae. concinnus and Ae. caudatus are also narrowly capillus and slight morphological distinction from iredalei see Rand spaced click-like notes with a broad frequency range, up to 7 kHz. and Fleming (1957)). This phylogeographic differentiation pattern between Easternmost Himalayan populations and those from Ne- 4.3. Taxonomic conclusions pal and adjacent Western Himalayas at a considerable genetic dis- tance of about 4.3% even exceeds the differences found between The criterion of diagnosability of species by genetic markers coal tits and spot-winged tits Periparus ater aemodius and P. a. mel- and/or morphological, behavioural and other characters (Helbig anolophus from the same region (about 2.3%; Gill et al., 2005; Mar- et al., 2002) has recently led to an increase of ‘diagnosable’ spe- tens et al., 2006). cies-level taxa. In the light of our genetic study, a splitting of SE Asian long-tailed tit species as executed by Harrap in HBW 13 (del Hoyo et al. 2008) was certainly inspired by two major miscon- Acknowledgements ceptions: (i) the extreme morphological differentiation in Ae. caud- atus on the one hand and the parallel situation in the Ae. iouschistos Over the years J.M. was sponsored by field research grants from group on the other hand were interpreted in a different way; (ii) Deutscher Akademischer Austauschdienst, Deutsche Forschungs- geographical separation of diagnosable taxa within the iouschistos gemeinschaft, Feldbausch-Stiftung and Wagner-Stiftung, the last group was taken for granted and equated with reproductive isola- two at Fachbereich Biologie of Mainz University. In addition, Vere- tion in the sense of the Mayrian biospecies concept. According to inigung ‘‘Freunde der Universität Mainz”, Deutsche Ornithologen- our molecular data we come to the following conclusions. Gesellschaft (East Asia grants to A. Gebauer, M. Kaiser and J.M.), Species status of Ae. glaucogularis seems justified for its marked and Gesellschaft für Tropenornithologie provided travel funds. Tis- genetic differentiation from all subspecies of Ae. caudatus studied sue samples were kindly provided by S. Rohwer and S. Birks from here. Within Ae. caudatus s. l. Chinese glaucogularis is the only tra- the Burke Museum of Natural History Washington, USA, Prof. Dr. ditional subspecies group that is equally morphologically distinc- Zhang Zhengwang and Jianqiang Li from Beijing Normal University, tive and represents a genetic unit of its own at the same time. China, R. van den Elzen from Zoologisches Forschungsmuseum The genetic distance between the SC Chinese population studied Alexander Koenig, Bonn, Germany, R. Pfeifer, Bayreuth and D.T. Tie- and all subspecies of Ae. caudatus studied of 1.6–2.0% ranges tze. S. Frahnert from Museum für Naturkunde Berlin, Germany, slightly below the values among currently accepted aegithalid sis- facilitated collection work with further SE Asian voucher speci- ter species, e.g. Ae. niveogularis and Ae. iouschistos (the latter mens. This study was supported by grants of the National Natural including the Chinese and Burmese forms discussed in the Science Foundation of China to Sun Y.-H. (30620130110). Many following). cordial thanks are due to all friends, colleagues and organizations The phylogeographic pattern of intra- and interspecific differen- mentioned. tiation of molecular markers and morphological characters in Ae. caudatus s. l. (clade 1) is exactly paralleled in its sister group (clade Appendix A. Supplementary data 2): a single geographically restricted taxon of a distinct phenotype (Ae. niveogularis) is clearly separated from a morphologically highly Supplementary data associated with this article can be found, in diverse group which shows only little genetic differentiation the online version, at doi:10.1016/j.ympev.2010.01.024. among the phenotypically distinct forms iouschistos, bonvaloti, sharpei and fuliginosus. Genetic distances among the latter four equal the intraspecific differentation level found within Ae. cauda- References tus and clear geographic separation as postulated in DelHoyo et al. (2009) is only given in the case of sharpei from Mt Victoria, Myan- Allen, E.S., Omland, K.E., 2003. Novel intron phylogeny (ODC) supports plumage mar. Moreover, there is some evidence for a zone of intergradation convergence in orioles (Icterus). The Auk 120, 961–969. Alström, P., Ranft, R., 2003. The use of sound in bird systematics and the importance between fuliginosus and bonvaloti, but the situation in the field is of bird sound archives. Bull. Brit. Ornithol. Club 123A, 114–135. poorly known and further research is required. However, we Alström, P., Ericson, P.G.P., Olsson, U., Sundberg, P., 2006. Phylogeny and clearly state that genetic differentiation among the taxa of clade classification of the avian superfamily Sylvioidea. Mol. Phylogenet. Evol. 38, 381–397. 2 (when excluding niveogularis!) ranges at an extremely low level Baker, E.C.S., 1920. New names for certain birds on the Indian list and certain new rather found among populations of the same species. Therefore, we species and subspecies. Bull. Brit. Orn. Cl. 41, 8–9. 966 M. Päckert et al. / Molecular Phylogenetics and Evolution 55 (2010) 952–967
Barker, F.K., Barrowclough, G.F., Groth, J.G., 2002. A phylogenetic hypothesis for Johansson, U.S., Alström, P., Olsson, U., Ericson, P.G.P., Sundberg, P., Price, T.D., 2007. passerine birds: taxonomic and biogeographic implications of an analysis of Build-up of the Himalayan avifauna through immigration: a biogeographical nuclear DNA sequence data. Proc. R. Soc. Lond. B 269, 295–308. analysis of the Phylloscopus and Seicercus Warblers. Evolution 61, 324–333. Barker, F.K., Cibois, A., Schikler, P., Feinstein, J., Cracraft, J., 2004. Phylogeny and Jønsson, K.A., Fjeldså, J., 2006. Determining biogeographical patterns of dispersal diversification of the largest avian radiation. Proc. Natl. Acad. Sci. 101, 11040– and diversification in oscine passerine birds in Australia, Southeast Asia and 11045. Africa. J. Biogeogr. 33, 1155–1165. Bergmann, H.-H., Helb, H.-W., Baumann, S., 2008. Die Stimmen der Vögel Europas. Kleinschmidt, O., Weigold, H., 1922. Zoologische Ergebnisse der W. Stoetzner´schen Aula-Verlag GmbH, Wiebelsheim. Expeditionen nach Szetschwan, Osttibet und Tschili aufgrund der Sammlungen Birckhead, H., 1937. The birds of the sage western China expedition. Am. Mus. Novit. und Beobachtungen Dr. H. Weigolds–Corvidae, Certhiidae, Sittidae, Paridae, 966, 1–17. Cinclidae. 1. Teil. Abh. Ber. Zool. und Anthr.-Ethn. Mus. Dresden 15, 1–18. Bures, S., Dvornik, P., Saetre, G.-P., 2002. Hybridization and apparent hybridization Kocher, T.D., Thomas, K.W., Meyer, A., Edwards, S.V., Pääbö, S., Villablanca, F.X., between meadow pipit (Anthus pratensis) and water pipit (A. spinoletta). Wilson, A.C., 1989. Dynamics of mitochondrial DNA evolution in animals: Hereditas 136, 254–256. amplification and sequencing with conserved primers. Proc. Natl. Acad. Sci. USA Cheng, T.-H., 1987. A Synposis of the Avifauna of China. Science Press/Paul Parey 86, 6196–6200. Scientific Publishers, Beijing/Hamburg, Berlin. Kvist, L., Martens, J., Ahola, A., Orell, M., 2001. Phylogeography of a Palaearctic Clement, M., Posada, D., Crandall, K.A., 2000. TCS: a computer program to estimate sedentary passerine, the willow tit (Parus montanus). J. Evol. Biol. 14, 930–941. gene genealogies. Mol. Ecol. 9, 1657–1660. Kvist, L., Martens, J., Orell, M., Higuchi, H., 2003. Evolution and genetic structure of Clements, J.F., 2007. Clements Checklist of Birds of the World, sixth ed. Cornell the great tit (Parus major complex). Proc. R. Soc. Lond. B 270, 1447–1454. University Press. La Touche, J.D.D., 1925-1930. A Handbook of the Birds of Eastern China, vol. 1. del Hoyo, J., Elliot, A., Christie, D. (Eds.), 2008. Handbook of the Birds of the World: Taylor and Francis, London. p. 500. Penduline Tits to Shrikes, vol. 13. Lynx Edicions, Barcelona. Löhrl, H., 1964. Verhaltensmerkmale der Gattungen Parus (Meisen), Aegithalos Dickinson, E.C., 2003. The Howard and Moore Complete Checklist of the Birds of the (Schwanzmeisen), Sitta (Kleiber), Tichodroma (Mauerläufer) und Certhia World. University Press, Oxford, Princeton. (Baumläufer). J. Ornithol. 105, 153–181. Dickinson, E.C., Loskot, V.M., Morioka, H., Somadikarta, S., van den Elzen, R., 2006. Löhrl, H., Thielcke, G., 1969. Zur Brutbiologie, Ökologie und Systematik einiger Systematic notes on Asian birds 50. Types of the Aegithalidae, Remizidae and Waldvögel Afghanistans. Bonner Zool. Beitr. 20, 85–98. Paridae. Zool. Med. Leiden 80, 65–111. Lovette, I.J., 2004. Mitochondrial dating and mixed support for the ‘‘2%” rule in Dietzen, C., Witt, H.-H., Wink, M., 2003. The phylogeographic differentiation of the birds. The Auk 121, 1–6. European robin Erithacus rubecula on the Canary Islands revealed by Marincovich Jr., L., Barinov, K.B., Oleinik, A.E., 2002. Astartae (Bivalvia: Astartidae) mitochondrial sequence data and morphometrics: evidence for a new robin that document the earliest opening of the Bering Strait. J. Paleontol. 76, 239– taxon on Gran Canaria? Avian Sci. 3, 115–131. 245. Drovetski, S.V., Zink, R.M., Rohwer, S., Fadeev, I.V., Nesterov, E.V., Karagodin, I., Martens, J., Eck, S., 1995. Towards an ornithology of the Himalayas. Systematics, Koblik, E.A., Red´ kin, Y.A., 2004. Complex biogeographic history of a holarctic ecology and vocalizations of Nepal birds. Bonn. Zool. Monogr. 38, 1–445. passerine. Proc. R. Soc. Lond. B 271, 545–551. Martens, J., Tietze, D.-T., Eck, S., Veith, M., 2004. Radiation and species limits of the Drovetski, S.V., Zink, R.M., Mode, N.A., 2009. Patchy distributions belie Asian Pallas’s warbler complex (Phylloscopus proregulus s.l.). J. Ornithol. 145, morphological and genetic homogeneity in rosy-finches. Mol. Phyl. Evol. 50, 206–222. 437–445. Martens, J., Tietze, D.T., Sun, Y.H., 2006. Molecular phylogeny of Parus (Periparus), a Drummond, A.J., Rambaut, A., 2007. BEAST: Bayesian evolutionary analysis by Eurasian radiation of tits (Aves: Passeriformes: Paridae). Zool. Abh. Mus. Tierkd. sampling trees. BMC Evol. Biol. 7, 214. Dresden 55, 103–120. Drummond, A.J., Ho, S.Y.W., Phillips, M.J., Rambaut, A., 2006. Relaxed phylogenetics Martens, J., Sun, Y.-H., Wei, L., Päckert, M., 2008. Intraspecific differentiation of Sino- and dating with confidence. PLoS Biol. 4, e88. Himalayan bush-dwelling Phylloscopus leaf warblers, with description of two Eck, S., 1996. Die Paläarktischen Vögel–Geospezies und Biospezies. Zool. Abh. Mus. new taxa (P. Fuscatus, P. fuligiventer, P. affinis, P. armandii, P. subaffinis). Vert. Tierkd. Dresden 49 (Suppl.), 1–103. Zool. 58, 233–266. Eck, S., Martens, J., 2006. Systematic notes on Asian birds. 49. A preliminary review Martens, J., Tietze, D.T., Päckert, M., 2010. Speciation of passerine birds in the Sino- of the Aegithalidae, Remizidae and Paridae. Zool. Meded. Leiden 80, 1–63. Himalayan region – genetics, acoustics and biogeography. The Auk. Edwards, S.V., Arctander, P., Wilson, A.C., 1991. Mitochondrial resolution of a deep Mayr, E., Stanford, J.K., Mayr, E., 1940. The vernay cutting expedition to Northern branch in the genealogical tree for perching birds. Proc. R. Soc. Lond. B 243, 99– Burma, part I. Ibis 4 (14), 679–711. 107. Mayr, E., Amadon, D., 1951. A classification of recent birds. Amer. Mus. Novit. 1496, Ericson, P.G.P., Johansson, U.S., Parsons, T.J., 2000. Major divisions in oscines 1–42. revealed by insertions in the nuclear gene c-myc: a novel gene in avian Mayr, E., Traylor, M.A., Watson, G.W., 1986. Peters´ Checklist of Birds of the World, systematics. The Auk 117, 1069–1078. vol. XI. Museum of Comparative Zoology, Cambridge Mass. p. 638. Ericson, P.G.P., Johansson, U., 2003. Phylogeny of Passerida (Aves: Passeriformes) Nylander, J.A.A., 2004. MrModeltest v2 [Computer software and manual]. based on nuclear and mitochondrial sequence data. Mol. Phylogenet. Evol. 29, Evolutionary Biology Centre, Uppsala University. 126–138. Neufeldt, I.A., Wunderlich, K., 1983. Lophobasileus elegans (Przewalski). In: Dathe, H., Felsenstein, J., 1985. Confidence limits on phylogenies: an approach using the Neufeldt, W.M. (Eds.), Atlas der Verbreitung Paläarktischer Vögel, vol. 17. bootstrap. Evolution 39, 783–791. Akademie Verlag Berlin, Lieferung. Berlin. Fleischer, R.C., McIntosh, C.E., Tarr, C., 1998. Evolution on a volcanic conveyor belt: Päckert, M., Martens, J., 2008. Taxonomic pitfalls in tits – comments on the chapter using phylogeographic reconstructions and K–Ar-based ages of the Hawaiian Paridae of the Handbook of the Birds of the World (A.G. Gosler, P. Clement, HBW Islands to estimate molecular evolutionary rates. Mol. Ecol. 7, 533–545. vol. 12). Ibis 150, 829–831. Fleischer, R.C., Kirchman, J.J., Dumbacher, J.P., Bevier, L., Dove, C., Rotzel, N.C., Päckert, M., Martens, J., Nazarenko, A.A., Valchuk, O., Petri, B., Eck, S., Veith, M., 2005. Edwards, S.C., Lammertink, M., Miglia, K.J., Moore, W.S., 2006. Mid-Pleistocene The great tit, Parus major, a misclassified ring species. Biol. J. Linn. Soc. 86, 153– divergence of Cuban and North American ivory-billed woodpeckers. Biol. Lett. 2, 174. 466–469. Päckert, M., Martens, J., Dietzen, C., Wink, M., Kvist, L., 2006. Radiation of goldcrests García-Moreno, J., 2004. Is there a universal mtDNA clock for birds? J. Avian Biol. 35, (Regulus regulus) on the Atlantic Islands: evidence of a new taxon from the 465–468. Canary Islands. J. Avian Biol. 37, 364–380. Gill, F.B., Slikas, B., Sheldon, F.H., 2005. Phylogeny of titmice (Paridae): II. Species Päckert, M., Martens, J., Tietze, D.T., Dietzen, C., Wink, M., Kvist, L., 2007. Calibration relationships based on sequences of the mitochondrial cytochrome-b gene. The of a molecular clock in tits (Paridae) – do nucleotide substitution rates of Auk 122, 121–143. mitochondrial genes deviate from the 2% rule? Mol. Phylogenet. Evol. 44, 1–14. Haring, E., Gammauf, A., Kryukov, A., 2007. Phylogeographic patterns in widespread Päckert, M., Martens, J., Severinghaus, L.L., 2008. The Taiwan Firecrest (Regulus corvid birds. Mol. Phyl. Evol. 45, 840–862. goodfellowi) belongs to the Goldcrest assemblage (Regulus regulus s. l.) – Harrap, S., Quinn, D., 1996. Tits Nuthatches and Treecreepers. Christopher Helm, evidence from mitochondrial DNA and territorial song of the Regulidae. J. London. Ornithol. 150, 205–220. Hartert, E., 1910. Die Vögel der Paläarktischen Fauna, vol. 1. Springer, Berlin. p. 832. Päckert, M., Martens, J., Sun, Y.-H., Tietze, D.T., 2009. Phylogeography and the Helbig, A.J., Knox, A.G., Parkin, D.T., Sangster, G., Collinson, M., 2002. Guidelines for evolutionary time-scale of passerine radiations in the Sino-Himalaya. In: assigning species rank. Ibis 144, 518–525. Hartmann, M., Weipert, J. (Eds.), Biodiversity and Natural Heritage of the Hijmans, R.J., Guarino, L., Jarvis, A., O’Brien, R., Mathur, P., Bussink, C., Cruz, M., Himalaya, vol. III. Verein der Freunde und Förderer des Naturkundemuseums, Barrantes, I., Rojas, E., 2005. DIVA-GIS Version 5.0. Software. Available from: Erfurt, pp. 71–80.
Perrins, C.M. (Ed.), 1993. Handbook of the Birds of Europe the Middle East and North Tamura, K., Dudley, J., Nei, M., Kumar, S., 2007. Mega4: Molecular Evolutionary Africa. The Birds of the Western Palaearctic, vol. 7. Oxford University Press, Genetics Analysis (mega) software version 4.0. Mol. Biol. Evol. 24, 1596–1599. Oxford, New York. Tavares, E.S., Baker, A.J., 2008. Single mitochondrial gene barcodes reliably identify Pleske, T., 1890. Wissenschaftliche Resultate der von N.M. Przewalski nach Central- sister-species in diverse clades of birds. BMC Evol. Biol. 8, 18. Available from: Asien unternommenen Reisen auf Kosten einer von seiner Kaiserlichen Hoheit
学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，
提供一站式文献检索和下载服务”的24 小时在线不限IP 图书馆。 图书馆致力于便利、促进学习与科研，提供最强文献下载服务。
图书馆导航：
图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具