7. K. E. Shamberger et al., Geophys. Res. Lett. 41, 499–504 (2014).8. K. E. Fabricius et al., Nat. Clim. Change 1, 165–169 (2011).9. L. W. Smith, D. Barshis, C. Birkeland, Coral Reefs 26, 559–567

(2007).10. L. W. Smith, H. H. Wirshing, A. C. Baker, C. Birkeland, Pac. Sci.

62, 57–69 (2008).11. J. T. Ladner, S. R. Palumbi, Mol. Ecol. 21, 2224–2238 (2012).12. D. J. Barshis et al., Proc. Natl. Acad. Sci. U.S.A. 110, 1387–1392

(2013).13. P. De Wit et al., Mol. Ecol. Resour. 12, 1058–1067 (2012).14. M. Karin, E. Gallagher, Immunol. Rev. 228, 225–240 (2009).15. H. M. Shen, S. Pervaiz, FASEB J. 20, 1589–1598 (2006).16. T. A. Oliver, S. R. Palumbi, Coral Reefs 30, 241–250 (2011).17. J. T. Ladner, D. J. Barshis, S. R. Palumbi, BMC Evol. Biol. 12,

217 (2012).18. D. J. Barshis, J. T. Ladner, T. A. Oliver, S. R. Palumbi, Mol. Biol.

Evol. (2014).19. J. Endler, Natural Selection in the Wild (Princeton Univ. Press,

Princeton, NJ, 1986).

20. S. L. Coles, P. L. Jokiel, C. R. Lewis, Pac. Sci. 30, 155 (1976).21. P. Jokiel, S. Coles, Coral Reefs 8, 155–162 (1990).22. C. A. Logan, J. P. Dunne, C. M. Eakin, S. D. Donner, Glob.

Change Biol. 20, 125–139 (2014).23. P. Craig, C. Birkeland, S. Belliveau, Coral Reefs 20, 185–189

(2001).24. J. H. Stillman, Science 301, 65 (2003).25. J. E. Carilli, R. D. Norris, B. A. Black, S. M. Walsh, M. McField,

PLOS ONE 4, e6324 (2009).26. C. Parmesan et al., Nature 399, 579–583 (1999).27. C. Parmesan, G. Yohe, Nature 421, 37–42 (2003).28. T. L. Root, D. P. MacMynowski, M. D. Mastrandrea,

S. H. Schneider, Proc. Natl. Acad. Sci. U.S.A. 102, 7465–7469 (2005).29. C. Parmesan, Annu. Rev. Ecol. Evol. Syst. 37, 637–669 (2006).30. O. Honnay et al., Ecol. Lett. 5, 525–530 (2002).

ACKNOWLEDGMENTS

The transcriptome data are archived at NCBI’s Gene ExpressionOmnibus database (series record GSE56278: www.ncbi.nlm.nih.

gov/geo/query/acc.cgi?acc=GSE56278). Funding was providedby The Gordon and Betty Moore Foundation through grantGBMF2629, the Schmidt Ocean Institute, and a fellowshipfrom the NSF to N.T.K. We thank M. van Oppen, I. Baums,M. De Salvo, L. Bay, M. Morikawa, and O. Hoegh-Guldberg forcomments on earlier versions of the manuscript. We also thankthe U.S. National Park of American Samoa for permission to workon Ofu reefs and C. Caruso for logistical and research help.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/344/6186/895/suppl/DC1Materials and MethodsFigs. S1 and S2Tables S1 to S4References (31–35)

27 January 2014; accepted 1 April 2014Published online 24 April 2014;10.1126/science.1251336

EVOLUTION

Ancient DNA reveals elephant birdsand kiwi are sister taxa and clarifiesratite bird evolutionKieren J. Mitchell,1 Bastien Llamas,1 Julien Soubrier,1 Nicolas J. Rawlence,1*Trevor H. Worthy,2 Jamie Wood,3 Michael S. Y. Lee,1,4 Alan Cooper1†

The evolution of the ratite birds has been widely attributed to vicariant speciation,driven by the Cretaceous breakup of the supercontinent Gondwana. The early isolation ofAfrica and Madagascar implies that the ostrich and extinct Madagascan elephant birds(Aepyornithidae) should be the oldest ratite lineages. We sequenced the mitochondrialgenomes of two elephant birds and performed phylogenetic analyses, which revealedthat these birds are the closest relatives of the New Zealand kiwi and are distant fromthe basal ratite lineage of ostriches. This unexpected result strongly contradictscontinental vicariance and instead supports flighted dispersal in all major ratite lineages.We suggest that convergence toward gigantism and flightlessness was facilitatedby early Tertiary expansion into the diurnal herbivory niche after the extinction ofthe dinosaurs.

Despite extensive studies, the evolutionaryhistory of the giant flightless ratite birdsof the Southern Hemisphere landmassesand the related flighted tinamous of SouthAmerica has remained a major unresolved

question. The ratites and tinamous, termed“palaeognaths” due to their shared basal palatestructure, form the sister taxon to all other livingbirds (neognaths). The living ratites are one ofthe few bird groups composed largely of giantterrestrial herbivores and include: the emu andcassowary in Australia and New Guinea, thekiwi in New Zealand, the ostrich in Africa, and

the rhea in South America. In addition, two re-cently extinct groups included the largest birdsknown: the moa from New Zealand (height upto 2 to 3 m, 250 kg in weight) (1) and elephantbirds from Madagascar (2 to 3 m in height, upto 275 kg in weight) (2, 3). Ratites have beenbelieved to have originated through vicariantspeciation driven by the continental breakup ofthe supercontinent Gondwana on the basis ofcongruence between the sequence of continentalrifting and the presumed order of lineage diver-gence and distribution of ratites (4, 5).New Zealand is the only landmass to have sup-

ported two major ratite lineages: the giant her-bivorous moa and the chicken-sized, nocturnal,omnivorous kiwi. Morphological phylogeneticanalyses initially suggested that these two groupswere each other’s closest relatives (6, 7), presum-ably diverging after the isolation of an ancestralform following the separation of New Zealandand Australia in the late Cretaceous ~80 to 60million years ago (Ma) (8). However, subsequentstudies suggest that kiwi are more closely related

to the Australasian emu and cassowaries (9, 10),whereas the closest living relatives of the giantmoa are the flighted South American tinamous(11–14). The latter relationship was completely un-expected on morphological grounds and sug-gests a more complex evolutionary history thanpredicted by a model of strict vicariant specia-tion. By rendering ratites paraphyletic, the rela-tionship between the moa and tinamous alsostrongly suggests that gigantism and flightless-ness have evolved multiple times among palae-ognaths (12, 13).Perhaps the most enigmatic of the modern

palaeognaths are the recently extinct giant Mad-agascan elephant birds. Africa and Madagascarwere the first continental fragments to rift fromthe supercontinent Gondwana, separating fromthe other continents (and each other) completelyduring the Early Cretaceous (~130 to 100 Ma)(15). Consequently, the continental vicariancemodel predicts that elephant birds and os-triches should be the basal palaeognath lin-eages (16). Most molecular analyses recover theostrich in a basal position, consistent with a vi-cariant model. However, the phylogenetic po-sition of the elephant birds remains unresolved,as cladistic studies of ratite morphology are sen-sitive to character choice and may be confoundedby convergence (17), whereas DNA studies havebeen hampered by the generally poor molecularpreservation of elephant bird remains (18).We used hybridization enrichment with in-

solution RNA arrays of palaeognath mitochon-drial genome sequences and high-throughputsequencing to sequence near-complete mitochon-drial genomes from both elephant bird genera:Aepyornis and Mullerornis. Phylogenetic analysesplaced the two taxa, Aepyornis hildebrandti(15,547 base pairs) andMullerornis agilis (15,731base pairs), unequivocally as the sister taxa tothe kiwi (Fig. 1 and fig. S1). This result was con-sistently retrieved, regardless of phylogeneticmethod or taxon sampling, and was strongly sup-ported by topological tests (19). To our knowl-edge, no previous study has suggested thisrelationship, probably because of the disparatemorphology, ecology, and distribution of the twogroups. Elephant birds were herbivorous, almostcertainly diurnal, and among the largest birds

1Australian Centre for Ancient DNA, School of Earth andEnvironmental Sciences, University of Adelaide, NorthTerrace Campus, South Australia 5005, Australia. 2School ofBiological Sciences, Flinders University, South Australia5001, Australia. 3Landcare Research, Post Office Box 40,Lincoln 7640, New Zealand. 4South Australian Museum,North Terrace, South Australia 5000, Australia.*Present address: Allan Wilson Centre for Molecular Ecology andEvolution, Department of Zoology, University of Otago, Dunedin,New Zealand. †Corresponding author. E-mail: alan.cooper@adelaide.edu.au

RESEARCH | REPORTS

898 23 MAY 2014 • VOL 344 ISSUE 6186 sciencemag.org SCIENCE

known, whereas kiwi are highly derived omni-vores, nocturnal, and about two orders of mag-nitude smaller. Elephant birds more closelyresemble the moa, and analyses of morphologyhave suggested a close relationship between thesetaxa (17). However, adding morphological char-acters to our molecular data set increased sup-port for the relationship between elephant birdsand kiwi (figs. S2 and S3) and allowed for iden-tification of several distinctive character statesthat diagnose this clade [see list in (19)].Speciation by continental vicariance provides

a poor explanation of the close relationship be-tween elephant birds and kiwi. Madagascar andNew Zealand have never been directly connected,and molecular dates calculated from the geneticdata suggest that kiwi and elephant birds di-verged after the breakup of Gondwana (Fig. 1 andfig. S4). However, mean node age estimates amongpalaeognath lineages are sensitive to taxon sam-pling (Fig. 2), so molecular dating provides

limited power for testing hypotheses aboutratite biogeography. Depending on taxonsampling, estimates for the basal divergenceamong palaeognaths are equally consistent withthe separation of Africa ~100 Ma (15) and theCretaceous-Tertiary (KPg) boundary (~65 Ma)(Fig. 2). Thus, topological comparisons may bea more robust tool to test hypotheses of vicar-iance and connection.The phylogenetic placement of the elephant

bird as sister to the kiwi creates a marked discor-dance between the order of continental breakup(Fig. 3, A and B) and the sequence of palaeognathdivergences (Fig. 3C). Instead, it appears thatthe common ancestor of elephant birds and kiwiswas probably flighted and capable of long-distancedispersal, which is supported by a small, pos-sibly flighted kiwi relative from the Early Mioceneof New Zealand (20). Together, the phylogeneticposition of the flighted tinamous and apparentflighted ancestor of the kiwi and elephant bird

imply that every major ratite lineage indepen-dently lost flight (Fig. 1). We suggest that flighteddispersal was the primary driver of the distribu-tion of palaeognath lineages and that the dis-cordance between distribution and phylogenyis more consistent with lineage turnover in aphylogenetically diverse, flighted, and wide-spread clade. Early Tertiary palaeognaths werecapable of long-distance dispersal, with remainsfound well outside the range of modern ratites,including the flighted lithornithids in NorthAmerica and Europe and the flightless Palaeotisand Remiornis in Europe (Fig. 3A) (21). Rapiddiversification through flighted dispersal alsoprovides an explanation for the short and oftenpoorly supported internodes amongst basalextant ratite lineages (13, 14).Early ratite evolution appears to have been

dominated by flighted dispersal and parallelevolution, with flightlessness evolving a mini-mum of six times and gigantism a minimum offive (Fig. 1) (22), suggesting that adaptationsfor cursoriality may have confounded phyloge-

Taxon Sets

Est

imat

ed t

ime

of

div

erg

ence

(M

a)

B CA

5060

7080

9010

0 MitochondrialNuclear

D

Fig. 2. Sensitivity of palaeognath age estima-tion to taxon sampling and genetic loci used.Mean and 95% HPD intervals are displayed forthe age (basal divergence) of crown palaeognaths,as inferred under several data set permutations.They axis represents time before the present in mil-lions of years, whereas age estimates for individualdata sets are arrayed on the x axis for bothmitochon-drial (circles) and nuclear loci (triangles). Resultsare presented for the following taxon sets: A, a fulltaxon set including ratites, tinamous, neognaths,and crocodilians; B, ratites, tinamous, and neognathsonly; C, ratites, neognaths, and crocodilians only;and D, ratites and neognaths only.Taxon sets A andC are each calibrated with six fossil node constraints,whereas taxon sets B and D are calibrated with asubset of four relevant constraints (table S5). Taxonsets are represented visually with silhouettes of anostrich (ratites), a flying tinamou (tinamous), a duck(neognaths), and an alligator (crocodilians). Analysesexcluding the rate-anomalous tinamous (taxon sets Cand D) retrieve a young age near the KPg boundary.

050 2575Ostrich (1)

Rhea (2)

Tinamou (47)

Moa (9)

Emu (1)

Elephant bird (7 )

Kiwi (5)

Ma

68.1

7.7

57.9

17.265.3

41.7

4.3

32.1

3.4

26.8

58.1

72.8

2.9

8.6

50.0

8.6

1/96

1/63

1/59

1/87

Cassowary (3)

*

*

*

*

*

*

*

*

*

*

*

1/95

Fig. 1. Phylogenetic position of the elephant birds from mitochondrial sequence data. Bayesianposterior probabilities and maximum likelihood bootstrap are presented in black below each branch;asterisks denote branches that received maximum possible support (bootstrap = 100%, Bayesianposterior probability = 1.0). Divergence dates [blue numbers above branches; blue bars represent95% highest posterior density (HPD) intervals] were inferred with six well-supported node age con-straints (table S5). Blue arrows mark the minimum date for the evolution of flightlessness in lineagesfor which fossil evidence is available (21, 22). The scale is given in millions of years before the present.Silhouettes indicate the relative size of representative taxa. Species diversity for each major clade ispresented in parentheses, with extinct groups shown in red.The dagger symbol (†) indicates that thenumber of elephant bird species is uncertain.

RESEARCH | REPORTS

SCIENCE sciencemag.org 23 MAY 2014 • VOL 344 ISSUE 6186 899

netic inference. Elsewhere, avian gigantism andflightlessness are almost exclusively observed inisland environments in the absence of mam-malian predators and competitors (e.g., thedodo). However, each of the landmasses occu-pied by ratites (excluding New Zealand) is nowhome to a diverse mammalian fauna. We sug-gest that the initial evolution of flightless ra-tites began in the ecological vacuum after theKPg mass-extinction event and the extinctionof the dinosaurs (12, 21). Most mammals appearto have remained relatively small and unspe-cialized for up to 10 Ma after the KPg extinc-tion (23), potentially providing a window ofopportunity for the evolution of large flight-less herbivores in continental bird lineages.The early Tertiary fossil record supports thisinterpretation, with geographically widespreadflighted palaeognath fossils (Fig. 3A) (22) andthe appearance of other flightless avian her-bivores such as gastornithids in Europe andNorth America, dromornithids in Australia, andBrontornis from South America (21). After theearly Tertiary, the increasing prevalence ofmorphologically diverse mammalian compet-itors is likely to have prevented flightlessnessfrom developing in other continental birdlineages.The kiwi and tinamous are the only recent

palaeognath lineages to not exhibit gigantism,and both taxa co-occur with a second palae-ognath lineage (moa and rhea, respectively) thatis both much larger and not their closest rela-tive. We suggest that the disparity in size betweenco-occurring lineages may be a result of therelative timing of arrival of ancestral flightedpalaeognaths coupled with competitive exclu-sion: The first palaeognath to arrive on each land-mass monopolized the available niche space

for large flightless herbivores and omnivores,forcing subsequent arrivals to adopt an alter-native role and remain much smaller. For ex-ample, the South American ancestors of therhea lineage (Diogenornis) were already largeand flightless at 55 Ma (21) when the tinamoulineage originated. The absence of sympatriclineages of small palaeognaths on other land-masses in the recent past may reflect un-availability of alternative niches upon arrival(e.g., due to diversification of herbivorous mam-mals during the early Tertiary) or subsequentcompetition with mammals and/or neogna-thous birds. It is presumably the latter that hasnecessitated the maintenance of flight in thetinamous.

REFERENCES AND NOTES

1. M. Bunce et al., Proc. Natl. Acad. Sci. U.S.A. 106, 20646–20651(2009).

2. T. H. Worthy, R. N. Holdaway, The Lost World ofthe Moa (Indiana Univ. Press, Bloomington, IN,2002).

3. J. P. Hume, M. Walters, Extinct Birds (Bloomsbury, London,2012).

4. J. Cracraft, J. Zool. 169, 455–543 (1973).5. K. Lee, J. Feinstein, J. Cracraft, in Avian Molecular Evolution

and Systematics, D. Mindell, Ed. (Academic Press, New York,1997), pp. 173–208.

6. T. J. Parker, Trans. Zool. Soc. London 13, 373–431(1895).

7. J. Cracraft, Ibis 116, 494–521 (1974).8. W. P. Schellart, G. S. Lister, V. G. Toy, Earth Sci. Rev. 76,

191–233 (2006).9. A. Cooper et al., Proc. Natl. Acad. Sci. U.S.A. 89, 8741–8744

(1992).10. A. H. Bledsoe, Ann. Carnegie Mus. 57, 73 (1988).11. O. Haddrath, A. J. Baker, Proc. Biol. Sci. 279, 4617–4625

(2012).12. M. J. Phillips, G. C. Gibb, E. A. Crimp, D. Penny, Syst. Biol. 59,

90–107 (2010).13. J. Harshman et al., Proc. Natl. Acad. Sci. U.S.A. 105,

13462–13467 (2008).

14. J. V. Smith, E. L. Braun, R. T. Kimball, Syst. Biol. 62, 35–49(2013).

15. J. R. Ali, D. W. Krause, J. Biogeogr. 38, 1855–1872(2011).

16. P. Johnston, Zool. J. Linn. Soc. 163, 959–982 (2011).17. T. H. Worthy, R. P. Scofield, N.Z. J. Zool. 39, 87–153

(2012).18. A. Cooper et al., Nature 409, 704–707 (2001).19. For details, see materials and methods on Science Online.20. T. H. Worthy et al., in Paleornithological Research

2013 - Proceedings of the 8th International Meetingof the Society of Avian Paleontology and Evolution,U. B. Göhlich, A. Kroh, Eds. (Natural HistoryMuseum Vienna, Vienna, Austria, 2013),pp. 63–80.

21. G. Mayr, in Paleogene Fossil Birds (Springer, Berlin, 2009),pp. 25–34.

22. See supplementary text on Science Online.23. K. Black, M. Archer, S. Hand, H. Godthelp, in Earth

and Life, J. Talent, Ed. (Springer Netherlands, 2012),pp. 983–1078.

ACKNOWLEDGMENTS

This study was funded by the New Zealand Marsden Fundand the Australian Research Council. Grid computing facilitieswere provided by eRSA (e-research South Australia) andCIPRES (Cyberinfrastructure for Phylogenetic Research).We thank the Museum of New Zealand Te Papa Tongarewa(P. Millener, S. Bartle, A. Tennyson), Natural HistoryMuseum Oslo (N. Heintz), and National Museum of NaturalHistory Paris (C. Lefèvre) for samples. We acknowledgeD. Penny, M. Phillips, D. Burney and R. Ward forvaluable advice and assistance. All data associatedwith this study are available on GenBank (accessionnos. KJ749824 and KJ749825) and DRYAD(doi:10.5061/dryad.2727k).

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/344/6186/898/suppl/DC1Materials and MethodsSupplementary TextFigs. S1 to S9Tables S1 to S10References (24–76)

10 February 2014; accepted 21 April 201410.1126/science.1251981

Fig. 3. Conflict between inferred palaeognathphylogeny and the topology predicted by conti-nental vicariance. (A) Relative position of con-tinents during the Late Cretaceous and Tertiary.Continental landmasses are colored according toorder of severance from the remaining Gondwananlandmass: Africa and Madagascar first (dark gray;100 to 130 Ma), followed by New Zealand (red; 60to 80 Ma), then finally Australia, Antarctica, andSouth America (green; 30 to 50 Ma). Palaeognath-bearing fossil localities from the late Palaeoceneand Eocene (21, 22) are represented by circles(flighted taxa) and triangles (flightless taxa). (B)Predicted phylogeny of ratites under a model ofspeciation governed solely by continental vicar-iance. (C) Palaeognath phylogeny as inferred in thepresent study (see Fig. 1).

160 Ma

120 Ma

50 Ma

80 Ma

RESEARCH | REPORTS

900 23 MAY 2014 • VOL 344 ISSUE 6186 sciencemag.org SCIENCE

www.sciencemag.org/content/344/6186/898/suppl/DC1

Supplementary Materials for

Ancient DNA reveals elephant birds and kiwi are sister taxa and

clarifies ratite bird evolution

Kieren J. Mitchell, Bastien Llamas, Julien Soubrier, Nicolas J. Rawlence, Trevor H.

Worthy, Jamie Wood, Michael S. Y. Lee, Alan Cooper*

*Corresponding author. E-mail: alan.cooper@adelaide.edu.au

Published 23 May 2014, Science 344, 898 (2014)

DOI: 10.1126/science.1251981

This PDF file includes:

Materials and Methods

Supplementary Text

Figs. S1 to S9

Tables S1 to S10

Full Reference List

 

Ancient DNA reveals elephant birds and kiwi are sister taxa and clarifies ratite bird evolution

Supporting Online Material

Materials and Methods

Extraction

The two elephant bird specimens were Mullerornis agilis (MNZ S38300, tibiotarsus from

Beloha, Madagascar) and Aepyornis hildebrandti (MNZ S38301, femur from central

Madagascar). The extraction procedure was performed as outlined in a previous study, which

also analyzed MNZ S38300 (18). A Dremel tool and disposable carborundum disks were used

to remove approximately 1 mm of the exterior surface of each bone, and a 0.1 g fragment of

cortical bone was collected. Each bone fragment was powdered using an 8 mm tungsten ball

bearing in a Braun Mikrodismembrator U (B. Braun Biotech International, Germany) for 30 s

at 2000 rpm in a sterilized stainless steel container. The powder was decalcified overnight in

10-30 vol of 0.5 M EDTA (pH 8.0) at room temperature, along with a negative control (no

bone powder). The resulting sediment was collected by centrifugation and digested with

proteinase-K/DTT overnight at 50-55 °C, then extracted twice with Tris-saturated phenol and

once with chloroform. The DNA was desalted using Centricon-30 filter units (Millipore) and

concentrated to approximately 100-150 µL.

Library Preparation

DNA templates were enzymatically repaired and blunt-ended, and had custom adapters (Table

S1) ligated following the protocol of Meyer and Kircher (24). The 5’ adapter featured a

diagnostic index sequence to allow identification and exclusion of any contamination resulting

from the downstream hybridization and sequencing reactions. In order to produce the quantity

of DNA required for hybridization-enrichment (~200 ng) each library was subjected to a series

of short PCR amplifications using primers complementary to the adapter sequences (Table

S2). Cycle number was kept low and each round of amplification was split into 8 separate

PCRs in order to minimize PCR bias. Each individual PCR (25 µL) contained 1 × PCR buffer,

2.5 mM MgCl2, 1 mM dNTPs, 0.5 mM each primer (Table S2), 1.25 U AmpliTaq Gold and 2

µL DNA library. Cycling conditions were as follows: 94 °C for 12 min; 12 cycles of 94 °C for

30 s, 60 °C for 30 s, 72 °C for 40 + (2 x cycle number) s; and 72 °C for 10 min. PCR products

were pooled and purified using AMPure magnetic beads (Agencourt), eluting in 30 µL TLE

buffer.

 

Hybridization enrichment

Biotinylated 80-mer RNA baits, synthesized commercially by MYcroarray (MI, USA) were

used to enrich the target library for avian mitochondrial DNA (mtDNA). Baits were designed

using published whole mitochondrial genome sequences (excluding D-loop) and included the

palaeognaths: Anomalopteryx didiformis (little bush moa; NC_002779), Dinornis robustus

(giant moa; NC_002672), Emeus crassus (eastern moa; NC_002673), Apteryx owenii (little

spotted kiwi; NC_013806), Apteryx haastii (great spotted kiwi; NC_002782), Rhea americana

(greater rhea; NC_000846), Struthio camelus (ostrich; NC_002785) and Tinamus major (great

tinamou; NC_002781); see Table S3 for a complete list of probe taxa. One hybridization

reaction was performed for each of the two elephant bird libraries. Each reaction was

performed according to manufacturer’s instructions in a final volume of 26 µL: 5.2 × SSPE,

5.2 × Denhardt’s, 5 mM EDTA, 0.1% SDS. This solution was incubated for 44 hr (3 hr at 60

°C, 12 hr at 55 °C, 12 hr at 50 °C, 17 hr at 55 °C) with a total of 200 ng of library DNA. After

incubation the RNA baits were immobilized on magnetic streptavidin beads and washed once

with 1 × SSC + 0.1% SDS for 15 min at room temperature, and then twice with 0.1 × SSC +

0.1% SDS for 10 min at 50 °C. The beads were subsequently resuspended in 0.1M NaOH pH

13.0, which destroyed the RNA baits and released the captured DNA into solution. The

captured DNA molecules were purified using a Minelute spin-column (Qiagen), and eluted in

20 µL EB buffer.

Sequencing

The overall quantity of DNA remaining after enrichment was extremely small. To increase the

DNA yield, we performed a post-hybridization round of amplification for each library: 8 × 25

µL: 1 × PCR buffer, 2.5 mM MgCl2, 1 mM dNTPs, 0.5 mM each primer (Table S2), 1.25 U

AmpliTaq Gold and 2 µL DNA library (Cycling: 94 °C for 12 min; 12 cycles of 94 °C for 30

s, 60 °C for 30 s, 72 °C for 40 + [2 x cycle number] s; and 72 °C for 10 min). To add

IonTorrent recognition sequences to the sequencing libraries, we performed a short round of

PCR with fusion primers: 5 × 25 µL reaction: 1 × PCR buffer, 2.5 mM MgCl2, 1 mM dNTPs,

0.5 mM each primer (Table S2), 1.25 U AmpliTaq Gold and 1 µL DNA library (Cycling: 94

°C for 12 min; 7 cycles of 94 °C for 30 s, 60 °C for 30 s, 72 °C for 40 + [2 x cycle number] s;

and 72 °C for 10 min). PCR products were purified using AMPure magnetic beads

(Agencourt), eluting in 30 µL TLE buffer. The two libraries were then pooled in preparation for

sequencing.

The pooled library was diluted to 11.6 pM, linked to proprietary micron-scale beads (Ion

 

Sphere Particles; Life Technologies), and clonally amplified according to manufacturer’s

protocols via emulsion PCR on an Ion OneTouch system (Life Technologies) using the Ion

OneTouch 200 Template Kit v2 DL (Life Technologies). The template beads were then loaded

onto a 316 chip and sequenced on an IonTorrent PGM (Life Technologies) using the 200 bp

sequencing chemistry. In order to increase read coverage and ensure reproducibility, this

template preparation and sequencing protocol was performed twice. The resulting reads were

pooled for downstream analysis.

Sequence processing and quality filtering

Following sequencing, base calling was performed using Torrent Suite v3.2.1 (Life

Technologies). Sequencing reads were immediately demultiplexed according to 5’ Index II

(Table S1) using the fastx_barcode_splitter tool (FASTX-toolkit v0.0.13;

http://hannonlab.cshl.edu/fastx_toolkit) allowing no mismatches in the index sequences (--

mismatches 0). Demultiplexed reads (14,192,512) were trimmed with cutadapt v1.1 (25),

using the short custom adapter sequences (Table S1). Trimming parameters included a

maximum error rate of 0.333 (-e 0.333), a minimum read length of 30 nucleotides (-m 30), a

maximum read length of 200 nucleotides (-M 200), a minimum read quality Phred score of 20

(-q 20), a minimum of three-nucleotides overlap between the read and the adapter (-O 3), and

five iterations of trimming (-n 5). Read quality was visualized using fastQC v0.10.1

(http://www.bioinformatics.bbsrc.ac.uk/projects/fastqc) before and after trimming to make

sure the trimming of adapters was efficient. Trimmed reads were then sorted according to 5’

Index I (Table S1) using the fastx_barcode_splitter tool (FASTX-toolkit v0.0.13;

http://hannonlab.cshl.edu/fastx_toolkit) allowing no mismatches in the index sequences (--

mismatches 0). The removal of the five bp 5’ Index I resulted in a pool of reads between 25 -

195 bp in length, which were subsequently used for mapping and consensus assembly.

Read mapping and consensus sequence assembly

As we were sequencing elephant bird mitochondrial genomes de novo, we had no established

reference genome to map against. Previous studies have overcome this limitation by mapping

iteratively to a phylogenetically distant reference sequence (26, 27). In the current study we

followed this methodological principle using TMAP (https://github.com/nh13/TMAP), which

is an assembly tool optimized for mapping IonTorrent data. In the initial round of mapping,

only reads from relatively conserved regions of the mitochondrial genome are mapped

successfully due to the divergence between the target and the reference. However, the

 

information from these reads is taken into account in the reference for each subsequent round

of mapping and acts as a seed for the mapping of reads in more divergent regions. Over

multiple iterations, the number of mapped reads increases and the assembly grows outwards

from the original seeds. Iterative mapping continues until either the mitochondrial genome has

been completed, or the number of reads added fails to increase.

For Mullerornis agilis, a total of 2,862,460 deconvoluted and trimmed reads

(doi:10.5061/dryad.2727k) were mapped to the ostrich (Struthio camelus) mitochondrial genome

(NC_002785) using TMAP v3.2.2 (https://github.com/nh13/TMAP) with the default parameters for the

mapall command. Reads with a mapping quality Phred score >30 were selected using the SAMtools v1.4

(28) view command (-q 30), and duplicate reads were discarded using the MarkDuplicates.jar tool from

Picard tools v1.79 (http://picard.sourceforge.net). A 50% consensus was generated from the

mapped reads, and used as the reference for an additional round of mapping. This process was

iterated seven times until no additional reads could be mapped and the quality of the

consensus ceased to improve. A final 75% consensus sequence for M. agilis (15,731

contiguous bases) was generated from 5,482 unique reads and checked by eye in Geneious

v6.1.2 (Biomatters; http://www.geneious.com) before being deposited in GenBank (KJ749825).

The same mapping protocol was followed for A. hildebrandti: 853,102 deconvoluted and

trimmed reads (doi:10.5061/dryad.2727k) were mapped to the M. agilis mitochondrial genome

(KJ749825) for seven iterations. A final 75% consensus sequence for A. hildebrandti (15,547 contiguous

bases) was generated from 2,704 unique reads and deposited in GenBank (KJ749824). At least one read

covered 99.8% of the final M. agilis consensus sequence and 92.6% of the A. hildebrandti

consensus sequence (sites that received no coverage or where there was not a 75% majority

base call were coded as IUPAC ambiguities). Mean length of individual reads was 70.6 bp

(standard deviation = 26.8) while mean read-depth across the consensus was 24.7x (standard

deviation = 8.1) for M. agilis; for A. hildebrandti, mean length of individual reads was 79.8 bp

(standard deviation = 24.9) while mean read-depth across the consensus was 14.5x (standard

deviation = 16.3).

Sequence authenticity

We consider the unambiguous, united position of our two elephant bird mitogenomes on the

palaeognath phylogeny to be strong evidence for their authenticity. Further, during consensus

sequence assembly we observed no evidence of multiple sequence variants (non-duplicate

reads that agree with each other but conflict with the consensus, which may indicate

contamination) in the final read pile-up for either taxon. However, to more comprehensively

 

test the validity of our consensus sequences and phylogenetic results we: examined the

damage profile of mapped reads; compared our consensus sequences to previously published

Sanger sequencing data; subjected all mapped reads to a BLAST search; and re-mapped all

reads using a different algorithm at even higher stringency and monitored for any change in

phylogenetic signal.

We used MapDamage v0.3.6 (29) to assess patterns of damage across all mapped reads.

Analyses were performed for each taxon on the first (Fig. S5, S7) and final iterations (Fig. S6,

S8) of TMAP mapping. In all cases, patterns observed were consistent with degraded ancient

DNA (elevated 5’ C-to-T and 3’ A-to-G, and depurination at the position preceding the

beginning of the reads), suggesting that contamination of the mapped read pool by modern

DNA is negligible. Notably, after the first iteration of mapping elevated levels of all

substitutions are observed (Fig. S5, S7). However, this is an artifact resulting from mapping to

a phylogenetically distant reference genome. Subsequent iterations of mapping see a marked

decrease in this artifact, and it is not observed at all in the final iteration (Fig. S6, S8).

Our consensus sequence for Mullerornis agilis is 100% consistent with mitochondrial

sequence fragments (AY016018, 391 bp; AY016019, 225 bp) previously published for this

same sample (MNZ S38300) (18). Our consensus sequence differs from a third published

fragment (AY016017, 385 bp) by the presence of two adenosine insertions (sites 7841 and

7851 of our consensus), however these insertions are present in 34/34 (7841) and 31/32 (7851)

of mapped reads and may therefore represent sequencing errors or DNA damage in the

previously published data.

All reads used to construct the final consensus sequences for Aepyornis and Mullerornis were

subjected to a BLAST search (blastn as part of blastall 2.2.17 against the full NCBI nucleotide

database as of August 2012). The BLAST outputs (20 top hits) were processed with

Metagenome Analyzer (MEGAN5)(30) using default parameters. As shown on Figure S9, all

reads used in the final consensus either match a ratite, a higher taxonomic rank within Aves

(Neognathae or Passeriformes: likely reflecting regions of the mitochondrial genome highly

conserved across all birds), or a higher taxonomic rank in which Palaeognathae is nested. This

result indicates that each consensus is built from genuine sequences from the target organism

rather than non-specific mapping of exogenous environmental or bacterial DNA. In the

absence of established elephant bird reference genomes, it is impossible to conclusively test

for contamination by palaeognath taxa, but mapDamage results suggest that any contamination

by modern samples must be low. Further, it is highly unlikely that low levels of contamination

by ancient palaeognath sequences would be sufficient to mislead our assembly.

 

To further assess the validity of the consensus sequences of both our reconstructed elephant

bird mitochondrial genomes, all reads were re-mapped iteratively to the ostrich mitochondrial

genome using a conservative algorithm and higher stringency quality thresholds. Specifically,

all reads shorter than 30 bp were discarded (instead of 25 bp with TMAP), and reads were

mapped with BWA v0.6.2 (31) with published modified parameters for seed length (-l 1024),

edit distance (-n 0.01) and gap openings (-o 2) (32, 33). Seven iterations of mapping (building

consensus sequences and calling bases as for TMAP above) were performed for Aepyornis

(resulting in 896 reads covering 32.3% of the consensus; consensus length = 11,220 bp), while

nine iterations were performed for Mullerornis (resulting in 2,541 reads covering 59.2% of the

consensus; consensus length = 14,849 bp). Using BWA, mean length of individual reads was

67.5 bp (standard deviation = 24.8) while mean read-depth across the consensus was 11.7x

(standard deviation = 12.0) for M. agilis; for A. hildebrandti mean length of individual reads

was 70.5 bp (standard deviation = 23.1) while mean read-depth across the consensus was 5.5x

(standard deviation = 12.9).

Following this more stringent protocol, it is likely that real endogenous reads were discarded

as BWA is not ideal for mapping to a phylogenetically distant references and is not optimized

for the error profile of IonTorrent data. Consequently, the consensus sequences generated

using BWA represent the most conserved regions of the mitochondrial genome. Phylogenetic

analyses including the BWA consensus sequences were performed in parallel to analyses

using the TMAP consensus sequences (see below) following the exact same protocol. Results

are presented alongside the main results in Figure S1. The tree topology is identical to that

obtained using the TMAP consensus sequences, and statistical support measures do not differ

substantially. Constant phylogenetic signal between analyses confirms that the extra reads

mapped using TMAP are most likely endogenous.

Molecular phylogenetics and divergence dating

We added our new mitogenomic data to existing mitogenome sequences from other

palaeognaths plus a broad sampling of neognath and crocodilian outgroups (Table S4). Alignment

was performed using the ClustalW algorithm in Geneious v6.1.2 (Biomatters;

http://www.geneious.com) and checked by eye. Partitionfinder v1.1.1 (34) was used to select

the partitioning scheme and substitution models for downstream analysis in RAxML (35, 36),

MrBayes (37) and BEAST (38). Each nucleotide partition was assessed for signal saturation using an

information entropy index (40, 41) as implemented in DAMBE v5.2.65 (42). This method uses computer

simulation to estimate the critical value of saturation at which sufficient signal is lost to mislead

 

phylogenetic inference, and the level of saturation estimated from the data is compared to this critical

value. We performed 100 replicates of taxon subsampling, for sets of 4, 8, 16 and 32 taxa. Proportion of

invariant sites was estimated for each partition in DAMBE. To be conservative, gaps were treated as

unknown states (which reduces the power of the test to detect phylogenetic signal), although taxa for

which no data were available for a given partition were omitted. Significant saturation could be rejected

(p < 0.01) for each partition, assuming a symmetrical tree. In any case, we took a cautious approach

(following 12) and RY-coded mitochondrial third codon positions to reduce any possibility of bias arising

as a result of saturation.

Tree topology (undated) was estimated under maximum likelihood and Bayesian frameworks

using RAxML v7.2.8 (35, 36) and MrBayes v3.2.2 (37), respectively. Our RAxML analysis

comprised a maximum likelihood (ML) search for the best-scoring tree (Fig. S1) from 1,000

bootstrap replicates performed on six individually-modeled partitions: H-strand 1st codon

positions (GTR+G), H-strand 2nd codon positions (GTR+G), H-strand RY-coded 3rd codon

positions (GTR+G), RNA stems (GTR+G), RNA loops (GTR+G) and ND6 (GTR+G). Our

MrBayes analysis employed four Markov chains (one cold and three incrementally heated)

with default priors. Data was modeled as six separate partitions: H-strand 1st codon positions

(GTR+I+G), H-strand 2nd codon positions (GTR+I+G), H-strand RY-coded 3rd codon

positions (F81), RNA stems (SYM+I+G), RNA loops (GTR+I+G) and ND6 (GTR+I+G).

Each chain ran for 107 generations, sampling every 1,000. Convergence in topology was

assessed using the average standard deviation of clade (split) frequencies (<0.02), while convergence in

individual parameter values was assessed through potential scale reduction

factors approaching 1, and through broadly overlapping distributions and effective sample sizes >>200 in

Tracer v1.5 (39). Sampled trees were summarized as a majority-rule consensus tree by MrBayes after

discarding the first 25% of trees as burnin. Bayesian posterior probabilities for clades (labeled on the

RAxML tree) are presented in Figure S1. The data matrix for this analysis, and the resulting tree, is

available on DRYAD (doi:10.5061/dryad.2727k).

Our molecular dating analysis was performed using BEAST v1.7.5 (38) using six fossil-based

node constraints (Table S5). The branching order was constrained to the maximum likelihood tree

topology (Fig. S1); fixing topology facilitated convergence under the complex partitioned clock models

used here. Nucleotide substitution models were applied independently to six data partitions: H-strand 1st

codon positions (GTR+I+G), H-strand 2nd codon positions (TVM+I+G), H-strand RY-coded

3rd codon positions (F81), RNA stems (SYM+I+G), RNA loops (GTR+I+G) and ND6

(TIM+G). The hypothesis of a global clock was tested and rejected for each data partition so

independent relaxed lognormal clock models were used for each partition. Three independent

 

BEAST MCMC chains were run for 2x108 iterations, sampling every 104 for a total of 6x105

sampled trees; the first 10% of trees from each chain were discarded as burnin before

summarizing to create a maximum clade credibility tree in TreeAnnotator (Fig. S4). Parameter values

were monitored in Tracer v1.5 (39) to ensure that all parameters converged and had ESSs > 200.

The data matrix for this analysis, and the resulting tree, is available on DRYAD

(doi:10.5061/dryad.2727k).

Testing Sensitivity of Molecular Divergence Dates (mitochondrial and nuclear data)

Palaeognath divergence dates estimated in several recent studies differ substantially. Because

the timeline of palaeognath evolution is so important to our understanding of their evolution

and biogeography, we sought to explore how taxon sampling and sequence data source

(mitochondrial vs. nuclear) contribute to differences in inferred node ages.

We selected taxa from our mitochondrial matrix and a previously published, well sampled 10-locus

nuclear dataset such that each mitochondrial and nuclear dataset contained equivalent lineages (Table

S6). This necessarily meant that our elephant bird mitochondrial genomes were excluded, as

no nuclear loci are available for these taxa. We further subsampled these datasets to exclude: i)

crocodilians (due to their huge divergences from all other sampled taxa), ii) tinamous (due to their

anomalous evolutionary rates), and iii) tinamous and crocodilians. This resulted in four equivalent

taxon sets for both the nuclear and mitochondrial dataset: A) ratites, tinamous, neognaths and

crocodilians; B) ratites, tinamous and neognaths (crocodilians excluded); C) ratites, neognaths and

crocodilians (tinamous excluded); and D) ratites and neognaths (tinamous and crocodilians excluded).

The nuclear and mitochondrial data were analyzed separately under each taxon set.

We performed molecular dating on each of the taxon sets using BEAST v1.7.5 (38), with the topology

constrained to be compatible with Fig. 1 (allowing for different taxon sampling). Partitioning scheme and

model choice were performed for each taxon set using Partitionfinder v1.1.1 (Table S7) (34). Independent

relaxed lognormal clock models were applied to each partition to estimate branch lengths; the hypothesis

of a global clock was tested and rejected for each partition. Calibrations were implemented as above

(Table S5), with the exception that the Alligatorinae and Archosauria constraints were omitted from

analyses of taxon sets excluding crocodilians (B and D). For each taxon set, between three and five

independent BEAST MCMC chains were each run for 108 iterations, sampling every 104 for a total of

3x104 to 5x104 sampled trees. Parameter values were monitored in Tracer v1.5 (39) to ensure

that all parameters converged between chains and had combined ESSs > 200. Mean node ages

and 95% highest posterior diversities were calculated for each taxon set by combining the

sampled trees from each chain (Fig. S2, Table S8). The first 10% of sampled trees from each

 

chain were discarded as burnin. Data matrices for these analyses, and resulting trees, are available on

DRYAD (doi:10.5061/dryad.2727k).

Testing Sensitivity of Tree Topology (Mitochondrial, nuclear and morphological data)

To further test the robustness of tree topology, we performed analyses using parsimony,

maximum likelihood and Bayesian methods on concatenated mitochondrial (Table S4) and nuclear data

(Table S6: taxon set A), with and without a large recent morphological dataset. Further, to test the

robustness of our phylogenetic results to taxon sampling, we analyzed each combined matrix (including

and excluding morphology): i) without tinamous, ii) without any crocodilian and bird outgroups (i.e.

effectively unrooted), and iii) without tinamous or outgroups. Data matrices for these analyses, and

resulting trees, are available on DRYAD (doi:10.5061/dryad.2727k). All analyses strongly retrieved the

kiwi-elephant bird clade/split (Table S9).

The 34 taxa (17 palaeognaths and 17 outgroups; Table S4) in the concatenated mitochondrial

and nuclear datasets were scored for morphological data using a dataset derived from Worthy

and Scofield (17), with additional characters from Johnston (16). Fifteen of the seventeen

palaeognath taxa could be scored for morphology (only Apteryx haastii and Tinamus major

lacked morphological data), along with three outgroups (Table S4). The addition of this

morphological data permitted evaluation of the affinities of several fossil taxa (without any

molecular data); addition of these taxa did not change relationships among the “core” taxa, so

the discussion first focuses on analyses restricted to taxa scored for both molecules and morphology.

The 243 characters were added to the molecular dataset as an additional partition (characters

24901-25143; see data matrices available on DRYAD doi:10.5061/dryad.2727k). They are as follows:

- Characters 1-179 are directly from Worthy and Scofield (17).

- Characters 180-202 are characters 2-24 as scored by Johnston (16), except Aepyornis was

scored for character 202 by Worthy (this study).

- Characters 203-208 are characters 26-31 as scored by Johnston (16), except moa were scored

for character 207 by Worthy (this study).

- Characters 209-219 are characters 33-43 as scored by Johnston (16).

- Character 220 is character 55 from Johnston (16), except Struthio, Casuarius and moa

were scored by Worthy (this study).

- Characters 221-243 are characters 63-85 from Johnston (16).

Twenty-five of the multistate characters formed morphoclines and were ordered (24 28 32 35

37 38 39 43 49 51 60 66 80 85 88 96 98 99 100 104 111 137 153 167 171); all other characters

were unordered. All parsimony analyses used PAUP* (43), with heuristic searches involving

 

100 random addition replicates. Branch support was calculated using nexus batch files

constructed with the aid of TreeRot v3 (44), and bootstrap frequencies used 200 bootstraps.

The combined data yielded the same single most-parsimonious tree, whether morphology was

included (treelength = 38943), or excluded (treelength = 38288). The bootstrap majority-rule

consensus was also identical to this tree. Bremer and Bootstrap support for relationships

among palaeognath taxa are shown in Figure S2. Addition of morphological traits to the

molecular dataset improves Bremer support for six nodes within palaeognaths (Fig. S2),

including the kiwi-elephant bird clade: Bremer support increases from 27 to 28 for this clade,

while bootstrap support is ~99% in both instances. Previous analyses of the morphological

data in isolation grouped elephant birds with moas (17): the combined analysis here reveals

that there is a secondary signal in the morphological data that unites elephant birds and kiwis,

congruent with the molecular data (see 45). There are 10 unequivocal (optimization independent)

synapomorphies diagnosing the elephant bird + kiwi clade, of which the following are well-sampled

across palaeognaths, and relatively conservative:

60: palatine roofing the enlarged choanal fossa (unique within palaeognaths).

80: wide sternum (occurs also in ostriches).

81: sternum with a shallowly concave dorsal surface (occurs in some moa).

171: foramen vasculare distale vestigial (a more extreme state - total loss - occurs in moa).

All clock-free Bayesian analyses used MrBayes 3.2.2 (37), on the e-research South Australia

computer grid and via the CIPRES (Cyberinfrastructure for Phylogenetic Research) Science

Gateway. The Bayesian Information Criterion as employed by Partitionfinder (34) was used

to choose the optimal partitioning scheme for the concatenated molecular data, and models for

each molecular partition. A 17-partition model for the molecular data was chosen (Table S10). The

morphological data was included as an 18th data partition and analyzed using the Lewis

(46) stochastic model; Bayes Factors were slightly higher/better when the gamma parameter for rate

variability was included (ΔBF = 3.3 for rooted analyses excluding taxa scored only for morphology;

ΔBF = 5.5 for rooted analyses including taxa scored only for morphology). However, inclusion or

exclusion of this parameter for morphology did not change topology, and posterior

probabilities remained essentially identical. The results shown in Figure S2 are for rooted analyses of the

full dataset including the gamma parameter.

Branch proportions (“lengths”) were linked across nuclear molecular partitions, and

mitochondrial molecular partitions, in both the Partitionfinder tests and the MrBayes analyses.

For analyses including only taxa scored for both molecules and morphology, four runs of four

chains were used, with default heating parameters. Each run was 2.5 x 107 steps long,

 

sampling every 2 x 104 steps, with a burnin of 20%. Longer runs and burnin were required for

analyses that also included taxa scored only for morphology: four runs of four chains were

used, with the heating parameter (temp) lowered to 0.08. Each run was 108 steps long,

sampling every 8 x 104, with a burnin of 30%. Separate branch lengths were allowed for the

combined nuclear, combined mitochondrial, and morphological data; different partitions

within each of these three datasets (e.g. different nuclear genes) were permitted different overall rates

while sharing the same relevant branch proportions (e.g. the nuclear branch proportions). Convergence in

topology was assessed by similarity in split (clade) frequencies across runs (<0.05), convergence in

numerical parameters assessed using potential scale reduction factor (~1, meaning variance between and

within runs is similar) and broadly overlapping parameter samples in Tracer v1.5 (39). A majority-rule

consensus tree was constructed from the post-burnin samples of all four runs from each analysis.

Maximum likelihood analyses were performed using RAxML v7.2.8 (35, 36). The Bayesian

Information Criterion as employed by Partitionfinder (34) was used to choose the optimal partitioning

scheme for the concatenated molecular data, and models for each molecular partition. A 15-partition

model for the molecular data was chosen (Table S10). The morphological data was included as a 16th

partition, using the MK model option in RAxML. Due to constraints in RAxML, all characters had to be

treated as unordered and all polymorphic characters as unresolved for maximum likelihood analyses.

Each RAxML analysis comprised a maximum likelihood search for the best-scoring tree from 1,000

bootstrap replicates. Bootstrap support values are shown in Figure S2 for the rooted analyses.

All analyses (both rooted and unrooted) returned very similar topologies. The main difference

was the position of the rhea: as sister taxa to all other palaeognaths excluding ostriches in rooted Bayesian

analyses (Bayesian posterior probability [BPP] = 1.0, with or without morphology), sister taxa to the

casuariids in rooted maximum likelihood analyses (including morphology = 94, excluding morphology =

66), and as sister taxa to the clade comprising moa and tinamous in the parsimony analyses (including

morphology: bootstrap = 75, Bremer = 20; excluding morphology bootstrap = 93, Bremer = 23). Using

the unrooted dataset, rhea grouped together with the casuariids in the maximum likelihood analysis

(including morphology: bootstrap = 83; excluding morphology: bootstrap = 74) and the Bayesian analysis

including morphology (BPP = 0. 74), versus the clade comprising moa and tinamous in the parsimony

analyses (including morphology: bootstrap = 83; excluding morphology: bootstrap = 74,) and Bayesian

analysis excluding morphology (BPP = 0.96). However, elephant birds grouped with kiwi in all these

analyses (Table S9).

Adding eight fossil palaeognath taxa scored for morphology by Worthy and Scofield

(17), but with no DNA data (Pachyornis australis, Pachyornis elephantopus, Pachyornis

 

geranoides, Megalapteryx, Euryapteryx curtus, Dromaius baudinianus, Emuarius, Lithornis) to the

Bayesian and parsimony rooted analyses did not change topology among the remaining (core) taxa (Fig.

S3). However, the missing data for these fossil taxa greatly reduced Bremer support. Despite this,

Lithornis emerges as sister to tinamous with moderate support (Bremer 8, bootstrap 87%, PP = 0.9). In

Worthy and Scofield (17) Lithornis was found to be sister to remaining palaeognaths but also with some

character support for affinities with tinamous. In the present analysis, where molecular data largely

constrain relationships of living palaeognaths to a very different topology, the morphological data

strongly place Lithornis as a sister group to tinamous. The close relationship between Lithornis and

tinamous found here was first advocated by Houde and Olson (47) based on morphology and was later

supported by analyses of Lithornis eggshell (48, 49), although on morphology (48) the

similarities of tinamous and lithornithids were considered to be plesiomorphic within

palaeognaths. It is therefore likely that lithornithids are related, or perhaps even ancestors

(paraphyletic) to tinamous.

Testing the elephant bird + kiwi relationship against alternative nuclear topologies

Within extant ratites, the greatest phylogenetic uncertainty concerns the position of the rhea. To

test whether alternative assumptions regarding the position of the rhea affect the strength of our elephant

bird+kiwi clade, we ran additional analyses using our mitochondrial+nuclear matrices (with and without

morphology) in which we constrained the position of the rhea to be consistent with the results of recent

nuclear studies: rhea+casuariids (Haddrath and Baker’s 10 locus phylogeny (11)), rhea+kiwi+casuariids

(Haddrath and Baker’s 27 locus phylogeny (11)) and rhea+ tinamous (Smith et al.’s 40 locus

concatenated phylogeny (14); Smith's 40 locus coalescent phylogeny is consistent with our preferred trees

Fig. S2, S3). Data matrices for our analyses, and resulting trees, are available on DRYAD

(doi:10.5061/dryad.2727k). Analyses were run as described above in RAxML v7.2.8, PAUP* and

MrBayes v3.2.2 with the addition of backbone topological constraints (see data matrices on DRYAD).

Assuming these alternative positions of the rhea still results in the elephant bird+kiwi clade being

recovered by all analyses, with high support in the vast majority (Table S9).

Testing the basal position of African/Madagascan ratites

The most concordant position of the elephant birds in regard to vicariant biogeography is a basal

position (along with ostriches) in the palaeognath tree (as Madagascar and Africa separate early from all

other Gondwanan landmasses). We specifically tested for this phylogenetic topology by running

additional analyses using our mitochondrial+nuclear matrices (with and without morphology) in which

we constrained the ostriches and elephant birds to fall outside a clade consisting of all other (non-

 

AfroMadagascan) palaeognaths. Data matrices for these analyses, and resulting trees, are available on

DRYAD (doi:10.5061/dryad.2727k). Rooted and unrooted analyses were run in RAxML v7.2.8, PAUP*

and MrBayes v3.2.2 as described above with the addition of topological constraints (see data matrices on

DRYAD). The tree resulting from each constrained analysis was compared to the tree resulting from the

equivalent unconstrained analysis using the Templeton test (parsimony), SH test (maximum likelihood)

and Bayes factors (Bayesian). The unconstrained tree (which grouped elephant bird with kiwi) was

significantly better in each case (see Table S9 for value of statistics rejecting basal elephant birds).

Testing for base composition bias

Similarities in nucleotide frequency can potentially cause phylogenetic analyses to spuriously

group together otherwise distantly related taxa (e.g. 50, 51). In order to test whether elephant bird and

kiwi group together as a result of shared base frequency we analyzed our combined

mitochondrial+nuclear matrix (including all taxa) using the LogDet distance model (52) in PAUP*. The

LogDet transformation explicitly accounts for base composition bias. Neighbor-joining was used to

construct a best tree, and 100 bootstrap replicates were performed (see data matrix on DRYAD

doi:10.5061/dryad.2727k). This analysis still strongly supported a sister-taxon relationship between

elephant bird and kiwi (bootstrap = 96%).

 

Supplementary Text

Kiwi egg

The characteristically large eggs of the ratites now appear to be one of the many morphological

characters that have evolved independently in each major ratite lineage. Startlingly, this is apparently true

even for the kiwi, which has one of the largest eggs of any bird for its body size. The kiwi’s huge egg has

long been thought to be a trait inherited from a much larger, flightless ancestor (53, 54). However, our

results instead suggest that it is most likely an adaptation to predation pressure (increasing chick

precociousness) after the arrival of small, flighted ancestral kiwi in New Zealand (20).

Early Tertiary distribution and dispersal

Our hypothesis predicts that during the early Tertiary small flighted palaeognaths should have

been present on the southern landmasses where modern ratites are distributed. These flighted

taxa would have independently given rise to modern flightless lineages, before eventually

going extinct (with the exception of tinamous). However, the only known early Tertiary

flighted palaeognath fossils are from North America and Europe (47, 48). While the lack of

flighted palaeognaths in the ex-Gondwanan landmasses initially appears contrary to

expectation, it may be more a reflection of taphonomic/sampling bias than of true distribution.

North America and Europe have the best-sampled and most comprehensive early Tertiary

fossil record, while the southern continents are relatively depauperate (21). Only two

unequivocal early Tertiary palaeognath taxa are known from the ex-Gondwanan landmasses: a

putative stem-rheid (Diogenornis fragilis) from the late Palaeocene of Itaborai, Brazil, South

America (55); and Eremopezus eocaenus from the late Eocene of Egypt, Africa (56). Both

taxa were flightless, but the relationship of E. eocaenus to modern palaeognaths is unclear.

Remiornis (late Palaeocene of France, Europe (57)) and Palaeotis (middle Eocene of

Germany, Europe (58)) were also flightless and, like Eremopezus, are of uncertain affinity.

The ages of these flightless taxa are consistent with our hypothesis of widespread parallel loss

of flight following the KPg mass extinction (in the early Palaeocene). Further, the presence of

one particular flighted taxon in both North America and Europe contemporaneously

(Pseudocrypturus cercanaxius, early Eocene (48)) indicates a capacity for inter-continental

dispersal in early Tertiary flighted palaeognaths. A virtue of our hypothesis is that it offers a

more satisfying explanation for the presence of flightless palaeognaths in Europe than does the

model of continental vicariance.

Conserved biogeographical history is implicit in the hypothesis of diversification driven by

continental vicariance. However, if the distribution of modern palaeognaths is primarily a

 

result of flighted (frequent and potentially long-range) dispersal, then inferring historical

biogeography based only on extant taxa is likely to be misleading in the absence of dense

fossil data. Unfortunately, the paucity of Palaeocene fossils likely makes this problem

intractable for the present. Antarctica is a promising avenue of research that is currently

underexplored. While no unequivocal fossil palaeognaths are known from Antarctica (21) (but

see Tambussi et al. (59)), its close proximity to the other ex-Gondwanan continents make it

likely that at least some taxa were present during the early Tertiary. It is also possible that

Antarctica was the source (or one of several sources) of dispersing flighted palaeognath

waves.

Supplementary Figures

Fig. S1: Best maximum likelihood tree topology. Branch lengths in number of changes per site. Numbers associated with nodes

are: Bayesian posterior probabilities (BPP) from analyses of the TMAP / BWA consensus sequences followed by bootstrap

percentages (BS) from 1000 replicates for the TMAP / BWA consensus sequences. Nodes receiving maximum support in all

analyses (BPP = 1, BS = 100%) are indicated by an asterisk (*). Dashes indicate nodes that do not appear in the Bayesian highest

posterior probability tree; MrBayes analyses alternatively favour a clade comprising Arenaria interpres, Haematopus ater, Gavia

stellata, Pterodroma brevirostris, Ciconia ciconia and Pygoscelis adeliae to the exclusion of Apus apus and Archilochus colubris

(BPP = 1 TMAP / 1 BWA), and a clade comprising Pterodroma brevirostris and Gavia stellata (BPP = 0.87 TMAP / 0.71 BWA).

Fig. S2: Relationships among palaeognath birds from parsimony, maximum likelihood and Bayesian analyses of nuclear,

mitochondrial and morphological data. Topology and branch lengths (scale=changes per site) are those found in the majority-rule

consensus from Bayesian analyses. Values at nodes represent support values for combined / molecular only analyses, in this vertical

order: Bremer support (parsimony), parsimony bootstrap, posterior probability (Bayesian inference), maximum likelihood

bootstrap. Dash (-) indicates a clade is not present in the parsimony and/or maximum likelihood tree.

Fig. S3: Relationships among palaeognath birds from parsimony and Bayesian analyses of combined nuclear, mitochondrial

and morphological data, with inclusion of extinct taxa only scored for morphology (red taxon names). Topology is that

returned from the Bayesian (MrBayes) analyses; branch lengths are from the Bayesian analysis of the combined molecular and

morphological data. Scale is in changes per site. Values at nodes represent support values for combined analyses, in this vertical

order: posterior probability (Bayesian inference), bootstrap frequency / Bremer support (parsimony). Dash (-) indicates a clade is

not present in the parsimony tree.

Fig. S4: BEAST time-calibrated phylogeny based on the topology from Figure S1. Time scale is in millions of years.

Numbers associated with nodes are mean divergence dates; 95% HPDs are represented by blue bars. Divergence dates among

palaeognath taxa are detailed in Table S8 for all analyses.

0.050.0100.0150.0200.0

Dromaius novaehollandiae

Taeniopygia guttata

Crypturellus tataupa

Eudromia elegans

Tinamus major

Caiman crocodilus

Alligator mississippiensis

Anomalopteryx didiformis

Ciconia ciconia

Pterodroma brevirostris

Apteryx owenii

Dinornis robustus

Pterocnemia pennata

Anseranas semipalmata

Acanthisitta chloris

Pygoscelis adeliae

Mullerornis agilis

Emeus crassus

Apteryx mantelli

Alectura lathami

Haematopus ater

Anser albifrons

Gavia stellata

Apus apus

Anas platyrhynchos

Aepyornis hildebrandti

Arenaria interpres

Rhea americana

Casuarius casuarius

Archilochus colubris

Gallus gallus

Struthio camelus

Apteryx haastii

Casuarius bennetti

71.0

73.3

72.9

64.0

7.7

41.9

62.7

65.3

68.8

3.4

26.4

4.3

52.5

57.5

17.2

8.6

68.1

61.3

66.8

8.6

63.4

50.0

73.0

88.9

58.0

72.0

57.9 2.9

244.5

32.1

101.0

77.1

26.8

Ma

Fig. S5: MapDamage report for the first round of mapping of Mullerornis reads against

the ostrich mitochondrial genome. The four top panels show the characteristic high

frequency of purines immediately before the reads. The two lower panels show the

characteristic accumulation of 5’ C-to-T (red) and 3’ G-to-A (blue) misincorporations, and an

atypical elevated frequency of T-to-C (orange) and A-to-G (black) across the reads. Other

substitutions are shown in grey; insertions and deletions are shown in purple and green,

respectively.

A

0.0

0.1

0.2

0.3

0.4

0.5

Fre

qu

en

cy

●

● ●●●● ● ● ●

●

●

● ●

●

●●

● ●● ●

A

●

● ●

●● ●

●●

●

●

●

●

●

●

●

●

●

●● ●

plot

C

● ● ● ● ●

●● ●

●●

●

●

● ●●●●● ● ●

C

0.0

0.1

0.2

0.3

0.4

0.5

●

● ●●●●●● ●

●

●

● ●

● ●● ●

● ●●

G

0.0

0.1

0.2

0.3

0.4

0.5

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

Fre

qu

en

cy

●● ● ●

●

● ●

●●

●

●

●●●

●

●●

● ● ●

G

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

●●●●

●●●

●●

●

●

●

● ●

● ● ●● ●

●

T

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

● ● ●●

●

●●

●●

●

● ●

●●

●

● ● ●●●

T

0.0

0.1

0.2

0.3

0.4

0.5

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

●●●

● ●● ●

●

●

●

●

●

●

●

●●

●

●● ●

1 2 3 4 5 6 7 8 9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Fre

qu

en

cy

−25

−24

−23

−22

−21

−20

−19

−18

−17

−16

−15

−14

−13

−12

−11

−10

−9

−8

−7

−6

−5

−4

−3

−2

−1

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Fig. S6: MapDamage report for the final round of mapping of Mullerornis reads against

the final 75% consensus sequence for Mullerornis agilis. The four top panels show the

characteristic high frequency of purines immediately before the reads. The two lower panels

show the characteristic accumulation of 5’ C-to-T (red) and 3’ G-to-A (blue)

misincorporations. T-to-C and A-to-G transitions are shown in orange and black, respectively;

other substitutions are shown in grey; insertions and deletions are shown in purple and green,

respectively.

A

0.0

0.1

0.2

0.3

0.4

0.5

Fre

qu

en

cy

● ● ● ●●● ● ● ●

●

●

●●●● ●

● ● ● ●

A

●● ●

●● ●

●●

●

●

● ●

●

●

●●●

●●

●

plot

C

● ● ● ●●● ● ●

● ●

●●

●● ● ●

●● ● ●

C

0.0

0.1

0.2

0.3

0.4

0.5

● ● ● ●●●●●●

●

●

● ●

●●

● ●● ●

●

G

0.0

0.1

0.2

0.3

0.4

0.5

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

Fre

qu

en

cy

● ● ●● ●

●●●●

●

●

●●●●

●●● ● ●

G

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

●● ● ●

● ●●

●●

●

●

●

● ●

●●

● ● ● ●

T

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

● ● ●● ●

● ●● ●

●

● ●

● ●

●

● ● ● ● ●

T

0.0

0.1

0.2

0.3

0.4

0.5

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

● ● ●

● ●

● ●

●

●

●

●

●

●

●

● ●

●

●●●

1 2 3 4 5 6 7 8 9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Fre

qu

en

cy

−25

−24

−23

−22

−21

−20

−19

−18

−17

−16

−15

−14

−13

−12

−11

−10

−9

−8

−7

−6

−5

−4

−3

−2

−1

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Fig. S7: MapDamage report for the first round of mapping of Aepyornis reads against

the ostrich mitochondrial genome. The four top panels show the characteristic high

frequency of purines immediately before the reads. The two lower panels show the

characteristic accumulation of 5’ C-to-T (red) and 3’ G-to-A (blue) misincorporations, and an

atypical elevated frequency of T-to-C (orange) and A-to-G (black) across the reads. Other

substitutions are shown in grey; insertions and deletions are shown in purple and green

respectively.

A

0.0

0.1

0.2

0.3

0.4

0.5

Fre

qu

en

cy

●

●

●● ● ●

●● ●

●

●

●

●● ● ●

●

● ● ●

A

●

● ●●●●

●

●

●

●

●

●

●

●

● ●● ●

●

●

plot

C

●

●●● ● ● ● ●

●

●

●●● ●

●●

●

● ●●

C

0.0

0.1

0.2

0.3

0.4

0.5

● ● ●●● ●

● ● ●

●

●

●●

●●● ● ●

●●

G

0.0

0.1

0.2

0.3

0.4

0.5

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

Fre

qu

en

cy

●●●

● ● ● ● ●

●

●

● ●

●

● ● ● ●●● ●

G

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

●

● ●●●

●

●

● ●

●

●

●

● ●

●●

●●●

●

T

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

●

●● ● ● ●

●

●●

●

●

●

●●

●● ●

● ●●

T

0.0

0.1

0.2

0.3

0.4

0.5

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

● ● ●

●●

●●

●

●

●

●

●

●

●

● ●●

●

●

●

1 2 3 4 5 6 7 8 9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

0.22

0.24

0.26

0.28

0.30

Fre

qu

en

cy

−25

−24

−23

−22

−21

−20

−19

−18

−17

−16

−15

−14

−13

−12

−11

−10

−9

−8

−7

−6

−5

−4

−3

−2

−1

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

0.22

0.24

0.26

0.28

0.30

Fig. S8: MapDamage report for the final round of mapping of Aepyornis reads against

the final 75% consensus sequence for Aepyornis hildebrandti. The four top panels show the

characteristic high frequency of purines immediately before the reads. The two lower panels

show the characteristic accumulation of 5’ C-to-T (red) and 3’ G-to-A (blue)

misincorporations. T-to-C and A-to-G transitions are shown in orange and black, respectively;

other substitutions are shown in grey; insertions and deletions are shown in purple and green,

respectively.

A

0.0

0.1

0.2

0.3

0.4

0.5

Fre

qu

en

cy

●

●

●● ● ●

●●●

●

●

●

● ● ● ●

●

● ● ●

A

●

● ●●

●●●

●

●

●

●

●

●

●

● ●

● ● ●

●

plot

C

●

● ●● ● ● ●

●

●

●

●

●●●

● ●

●

● ●●

C

0.0

0.1

0.2

0.3

0.4

0.5

● ●●

● ●●

● ● ●

●

●

●●

● ●●● ●

●●

G

0.0

0.1

0.2

0.3

0.4

0.5

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

Fre

qu

en

cy

● ● ●

● ● ● ●●

●

●

●●

●

●●● ● ●

● ●

G

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

●

● ●●●

●

●● ●

●

●

●

●●

●●

●●

●

●

T

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

●●● ● ● ●

●

●

●

●

●

●●●

●●● ● ●

●

T

0.0

0.1

0.2

0.3

0.4

0.5

−1

0−

9−

8−

7−

6−

5−

4−

3−

2−

1 0 1 2 3 4 5 6 7 8 910

●●●

●

●

●●

●

●

●

●

●

●

●

● ●

●●

●

●

1 2 3 4 5 6 7 8 9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

0.22

0.24

0.26

0.28

0.30

Fre

qu

en

cy

−25

−24

−23

−22

−21

−20

−19

−18

−17

−16

−15

−14

−13

−12

−11

−10

−9

−8

−7

−6

−5

−4

−3

−2

−1

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

0.22

0.24

0.26

0.28

0.30

Fig. S9: Taxonomy profiles for sequencing reads used to build the Aepyornis and Mullerornis consensus sequences, as

determined using BLAST and MEGAN5. Circles on each cladogram indicate relative frequency of reads assigned to taxa (taxon

labels shown only for taxa receiving at least 50 hits). The bar graph for each consensus sequence displays relative frequencies of

reads that could be assigned to a “tip” (colored circles).

Low complexity

Not assigned - 61

Mullerornis agilis - 78

Casuarius - 69

Apteryx - 105

Palae

ogna

thae

- 28

1

Neognathae - 626

Aves - 930

Euteleostom

i - 220

Am

niota - 77

Casuariiformes - 53

0

100

200

300

400

500

600N

um

be

r o

f re

ad

s

Ne

og

na

thae

Apte

ryx

Ca

suari

us

Mu

llero

rnis

agili

s

Low complexity - 142

No hits

Not assigned - 300

Mullerornis agilis - 226

Dinornithiformes - 100

Dromaius novaehollandiae - 83

Casuarius - 103

Casuariiformes - 75

Apteryx - 340

Palaeognathae - 506

Passeriformes - 118

Aves - 1,578

Taxonomy profile for Mullerornis

0

50

100

150

200

250

300

350

Nu

mbe

r o

f re

ads

Pa

sse

rifo

rme

s

Ap

tery

x

Ca

sua

riu

s

Dro

ma

ius n

ova

eh

olla

nd

iae

Din

orn

ith

ifo

rme

s

Mu

llero

rnis

ag

ilis

Testudines + Archosauria - 60

Gnathostom

ata- 62

Sauria - 65

Bilateria- 80

Am

niota- 187

Tetrapoda- 83

Euteleostom

i- 358

Taxonomy profile for Aepyornis

Neognathae - 942

Supplementary Tables

Table S1: Structure of DNA sequencing libraries. Components and nucleotide sequences

listed in order 5’ to 3’. Component Sequence (5’ - 3’)

5’ IonTorrent Primer Bind Site CCATCTCATCCCTGCGTGTCTCCGACTCAG

5’ Index II TACGTCG (A. hildebrandti) / TGACGTG (M. agilis)

5’ Adapter ACACTCTTTCCCTACACGACGCTCTTCCGATCT

5’ Index I AAGTT (A. hildebrandti) / AGGAC (M. agilis)

DNA fragment Variable

3’ Adapter AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC

3’ IonTorrent Adapter ATCACCGACTG CCCATAGAGAGG

Table S2: Primer sequences used in library amplification and sequencing preparation. Primer Sequence (5’ – 3’)

Library amp. Fwd ACACTCTTTCCCTACACGAC

Library amp. Rev GTGACTGGAGTTCAGACGTGT

IonTorrent fusion Fwd CCATCTCATCCCTGCGTGTCTCCGACTCAGTACGTCGACACTCTTTCCCTA

CACGACGCTCTTCCGATCT (Aepyornis hildebrandti) /

CCATCTCATCCCTGCGTGTCTCCGACTCAGTGACGTGACACTCTTTCCCTA

CACGACGCTCTTCCGATCT (Mullerornis agilis)

IonTorrent fusion Rev CCTCTCTATGGGCAGTCGGTGATGTGACTGGAGTTCAGACGTGTGCTCTTC

CGATCT

Table S3: Taxonomy and NCBI accession numbers for data used to create in-solution

RNA probe array.

Species Common name Family Accession

Acanthisitta chloris Rifleman Acanthisittidae AY325307

Aegotheles cristatus Owlet nightjar Aegothelidae EU344979

Anas formosa Baikal teal Anatidae JF730435

Anas platyrhynchos Mallard Anatidae NC_009684

Anomalopteryx didiformis Little bush moa Emeidae NC_002779

Anser anser Domestic goose Anatidae NC_011196

Apteryx haastii Great spotted kiwi Apterygidae NC_002782

Apteryx owenii Little spotted kiwi Apterygidae NC_013806

Branta canadensis Canada goose Anatidae NC_007011

Cacatua moluccensis Salmon-crested cockatoo Cacatuidae NC_020592

Corvus frugilegus Rook Corvidae NC_002069

Dendrocygna javanica Lesser whistling duck Anatidae NC_012844

Dinornis robustus Giant moa Dinornithidae NC_002672

Emeus crassus Eastern moa Emeidae NC_002673

Eudyptes chrysocome Rockhopper penguin Spheniscidae NC_008138

Eudyptula minor Little blue penguin Spheniscidae NC_004538

Gallinula chloropus Common moorhen Rallidae NC_015236

Gallirallus okinawae Okinawa rail Rallidae NC_012140

Pica pica Common magpie Corvidae NC_015200

Porphyrio hochstetteri South Island takahe Rallidae NC_010092

Procellaria cinerea Grey pertrel Procellariidae AP009191

Pterodroma brevirostris Kerguelen petrel Procellariidae NC_007174

Rallina eurizonoides Slake Rallidae NC_012142

Rhea americana Greater rhea Rheidae NC_000846

Strigops habroptilus Kakapo Strigopidae NC_005931

Struthio camelus Ostrich Struthionidae NC_002785

Tinamus major Great tinamou Tinamidae NC_002781

Table S4: NCBI accession numbers for mitochondrial data used in analyses to determine

the phylogenetic placement of the elephant birds. Species Common name Family Accession Aepyornis hildebrandti* - Aepyornithidae KJ749824 Mullerornis agilis* - Aepyornithidae KJ749825 Alligator mississippiensis American alligator Alligatoridae NC_001922 Anomalopteryx didiformis* Little bush moa Emeidae NC_002779 Apteryx mantelli* North Island brown kiwi Apterygidae AY016010 Apteryx haastii Great spotted kiwi Apterygidae NC_002782 Apteryx owenii* Little spotted kiwi Apterygidae NC_013806 Alectura lathami* Australian brush-turkey Megapodiidae NC_007227 Caiman crocodilus Spectacled caiman Alligatoridae NC_002744 Casuarius bennetti* Dwarf cassowary Casuariidae AY016011 Casuarius casuarius* Southern cassowary Casuariidae AF338713 Gallus gallus* Chicken Phasianidae NC_001323 Crypturellus tataupa* Tataupa tinamou Tinamidae AY016012 Dinornis robustus* Giant moa Dinornithidae NC_002672 Dromaius novaehollandiae* Emu Casuariidae AF338711 Emeus crassus* Eastern moa Emeidae NC_002673 Eudromia elegans* Elegant crested-tinamou Tinamidae NC_002772 Archilochus colubris Ruby-throated hummingbird Trochilidae NC_010094 Gavia stellata Red-throated loon Gaviidae NC_007007 Anseranas semipalmata* Magpie goose Anseranatidae NC_005933 Anas platyrhynchos Mallard Anatidae NC_009684 Haematopus ater Blackish oystercatcher Haematopodidae NC_003713 Pygoscelis adeliae Adelie penguin Spheniscidae NC_021137 Pterodroma brevirostris Kerguelen petrel Procellariidae NC_007174 Pterocnemia pennata* Darwin’s rhea Rheidae NC_002783 Rhea americana* Greater rhea Rheidae NC_000846 Acanthisitta chloris Rifleman Acanthisittidae AY325307 Ciconia ciconia White stork Ciconiidae NC_002197 Struthio camelus* Ostrich Struthionidae NC_002785 Apus apus Common swift Apodidae NC_008540 Tinamus major Great tinamou Tinamidae AF338707 Arenaria interpres Ruddy turnstone Scolopacidae NC_003712 Anser albifrons White-fronted goose Anatidae NC_004539 Taeniopygia guttata Zebra finch Estrildidae NC_007897 *Scored for morphology

Table S5: Fossil node age constraints used in molecular dating analyses. Node Min. Age

(Ma)

Max.

Age

(Ma)

Distribution Comments on Min. Age Comments on Max. Age.

Archosauria

crown

235 250 Normal

(soft 2.5%

minimum, soft

97.5%

maximum)

Several stem-crocodiles (Bromsgroveia,

Wangisuchus, Fenhosuchus, Vjushkovisaurus,

Stagonosuchus and Mandasuchus) are known

from the Anisian (60). Thus we use the end of

the Anisian as a minimum constraint,

following Benton and Donoghue (61).

Absence of Archosauria from numerous Olenekian fossil

sites which yield archosaur outgroups:

Proterochampsidae, Euparkeriidae, Erythrosuchidae and

Proterosuchidae (62-64). Thus, we use the beginning of

the Olenekian as a maximum constraint, following

Benton and Donoghue (61).

Alligatorinae

crown

64 80 Lognormal

(soft 97.5%

maximum)

Presence of Navajosuchus mooki in the Early

Palaeocene (65). Navajosuchus is a stem-

alligatorine.

No alligatorids are known from earlier than the

Palaeocene. However, Alligatoroidea (a close outgroup)

is represented by several Campanian taxa (66-69). Thus,

we use the early to mid-Campanian as a maximum bound

(65).

Aves crown 66 133 Uniform Presence of Vegavis iaai in the Late

Cretaceous (70). Vegavis has been interpreted

as a crown-anseriform and most closely

related to Anatidae.

Conservative upper bound based on independent

mitochondrial clock estimates, following Haddrath and

Baker (11). Other studies have used similar constraints

(e.g. Phillips et al. (12): ~121 Ma) based on the absence

of Neornithes in the avian bearing beds within the Jehol

biota (71). Our constraint allows for the possibility of an

earlier origin of neoaves in the southern hemisphere.

Galloanserae

crown

66 86 Uniform Presence of Vegavis iaai in the Late

Cretaceous (70). Vegavis has been interpreted

as a crown-anseriform and is most closely

related to Anatidae.

Absence of Neornithes from ~86 Ma facies that reflect

the depositional environment from which Vegavis is

known (shallow marine/coastal belt) and are rich in

outgroup specimens: Ichthyornithiformes and

Hesperornithiformes (61).

Spheniscidae

stem

61 74 Lognormal

(soft 97.5%

maximum)

Presence of the sphenisciform Waimanu

manneringi in the Palaeocene (72).

Reflects an expectation that seabirds radiated following

the KPg after the extinction of numerous stem lineages.

Allows for the possibility that modern seabirds evolved

in the southern hemisphere during the Campanian to late

Maastrichtian hiatus in that region’s fossil record (see

Phillips et al. (12)).

Casuariidae

crown

25 35 Uniform Presence of Emuarius gidju (73) in the

Oligocene. Emuarius is a crown casuariid:

sister taxon to Dromaius, to the exclusion of

Casuarius (74).

Reflects the assertion that Emuarius is “probably close to

the mutual ancestor of emus and cssowaries” (75). Our

maximum constraint of 35 Ma follows several previous

studies (11, 12, 18, 76).

Table S6: NCBI accession numbers and taxon sets for the mitochondrial data and the ten nuclear loci used in comparative

molecular dating analyses for testing the sensitivity of inferred divergence times to taxon sampling.

Taxon Mito. BACH1 BMP2 CMOS DNAH3 IRBP NT3 PNN PTPN RAG2 TRAF6 Taxon

Sets

Acanthisitta

chloris AY325307 JX533012 JX533043 JX533074 JX533122 JX533214 JX533265 JX533296 JX533327 JX533380 JX533411

A, B,

C, D

Alligator

mississippiensis NC_001922 JX533010 JX533041 JX533072 JX533120 JX533212 JX533263 JX533294 JX533325 JX533378 JX533409 A, B

Anomalopteryx

didiformis NC_002779 JX533038 JX533069 JX533100 JX533148 JX533240 JX533291 JX533322 JX533353 JX533406 JX533437

A, B,

C, D

Anser albifrons NC_004539 JX533023 JX533054 JX533085 JX533133 JX533225 JX533276 JX533307 JX533338 JX533391 JX533422

A, B,

C, D

Anseranas

semipalmata NC_005933 JX533024 JX533055 JX533086 JX533134 JX533226 JX533277 JX533308 JX533339 JX533392 JX533423

A, B,

C, D

Apteryx mantelli AY016010 JX533033 JX533064 JX533095 JX533143 JX533235 JX533286 JX533317 JX533348 JX533401 JX533432

A, B,

C, D

Apteryx haastii NC_002782 JX533031 JX533062 JX533093 JX533141 JX533233 JX533284 JX533315 JX533346 JX533399 JX533430

A, B,

C, D

Archilochus

colubris NC_010094 JX533022 JX533053 JX533084 JX533132 JX533224 JX533275 JX533306 JX533337 JX533390 JX533421

A, B,

C, D

Caiman

crocodilus NC_002744 JX533011 JX533042 JX533073 JX533121 JX533213 JX533264 JX533295 JX533326 JX533379 JX533410 A, B

Casuarius

casuarius AF338713 JX533036 JX533067 JX533098 JX533146 JX533238 JX533289 JX533320 JX533351 JX533404 JX533435

A, B,

C, D

Apodidaea

NC_008540 JX533021 JX533052 JX533083 JX533131 JX533223 JX533274 JX533305 JX533336 JX533389 JX533420 A, B,

C, D

Gallus gallus NC_001323 Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl

A, B,

C, D

Dromaius

novaehollandiae AF338711 JX533037 JX533068 JX533099 JX533147 JX533239 JX533290 JX533321 JX533352 JX533405 JX533436

A, B,

C, D

Ciconiidaeb

NC_002197 JX533016 JX533047 JX533078 JX533126 JX533218 JX533269 JX533300 JX533331 JX533384 JX533415 A, B,

C, D

Eudromia elegans

NC_002772 JX533025 JX533056 JX533087 JX533135 JX533227 JX533278 JX533309 JX533340 JX533393 JX533424 A, C

Gaviidaec

NC_007007 JX533013 JX533044 JX533075 JX533123 JX533215 JX533266 JX533297 JX533328 JX533381 JX533412 A, B,

C, D

Haematopus ater NC_003713 JX533018 JX533049 JX533080 JX533128 JX533220 JX533271 JX533302 JX533333 JX533386 JX533417

A, B,

C, D

Charadriiformesd

NC_003712 JX533019 JX533050 JX533081 JX533129 JX533221 JX533272 JX533303 JX533334 JX533387 JX533418

A, B,

C, D

Procellariiformese

NC_007174 JX533017 JX533048 JX533079 JX533127 JX533219 JX533270 JX533301 JX533332 JX533385 JX533416 A, B,

C, D

Pterocnemia

pennata NC_002783 JX533030 JX533061 JX533092 JX533140 JX533232 JX533283 JX533314 JX533345 JX533398 JX533429

A, B,

C, D

Spheniscidaef NC_021137 JX533015 JX533046 JX533077 JX533125 JX533217 JX533268 JX533299 JX533330 JX533383 JX533414 A, B,

C, D

Rhea americana NC_000846 JX533029 JX533060 JX533091 JX533139 JX533231 JX533282 JX533313 JX533344 JX533397 JX533428

A, B,

C, D

Struthio camelus NC_002785 JX533028 JX533059 JX533090 JX533138 JX533230 JX533281 JX533312 JX533343 JX533396 JX533427

A, B,

C, D

Taeniopygia

guttata NC_007897 Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl Ensembl

A, B,

C, D

Tinamus major

AF338707 JX533026 JX533057 JX533088 JX533136 JX533228 JX533279 JX533310 JX533341 JX533394 JX533425 A, C

a Mitochondrial = Apus apus; Nuclear = Chaetura pelagica

b Mitochondrial = Ciconia ciconia; Nuclear = Ephippiorhynchus senegalensis

c Mitochondrial = Gavia stellata; Nuclear = Gavia immer

d Mitochondrial = Arenaria interpres; Nuclear = Jacana jacana

e Mitochondrial = Pterodroma brevirostris; Nuclear = Oceanites oceanicus

f Mitochondrial = Pygoscelis adeliae; Nuclear = Pygoscelis papua

Table S7: Partitioning schemes and substitution models for comparative molecular

dating analyses for testing the sensitivity of inferred divergence times to taxon sampling.

Taxonset Partition

Number

Composition Model

Mitochondrial

(A)

All taxa

1 Mitochondrial ND6 locus K81uf+G

2 Mitochondrial 1st codon positions GTR+I+G

3 Mitochondrial 2nd codon positions TVM+I+G

4 Mitochondrial 3rd codon positions F81

5 Mitochondrial RNA loop positions GTR+I+G

6 Mitochondrial RNA stem positions SYM+I+G

Mitochondrial

(B)

No tinamous

1 Mitochondrial ND6 locus K81uf+G

2 Mitochondrial 1st codon positions GTR+I+G

3 Mitochondrial 2nd codon positions TVM+I+G

4 Mitochondrial 3rd codon positions F81

5 Mitochondrial RNA loop positions GTR+I+G

6 Mitochondrial RNA stem positions SYM+I+G

Mitochondrial

(C)

No crocodilians

1 Mitochondrial ND6 locus K81uf+G

2 Mitochondrial 1st codon positions GTR+I+G

3 Mitochondrial 2nd codon positions TVM+I+G

4 Mitochondrial 3rd codon positions F81

5 Mitochondrial RNA loop positions GTR+I+G

6 Mitochondrial RNA stem positions SYM+I+G

Mitochondrial

(D)

No tinamous or

crocodilians

1 Mitochondrial ND6 locus K81uf+G

2 Mitochondrial 1st codon positions GTR+I+G

3 Mitochondrial 2nd codon positions TVM+I+G

4 Mitochondrial 3rd codon positions F81

5 Mitochondrial RNA loop positions GTR+I+G

6 Mitochondrial RNA stem positions SYM+I+G

Nuclear

(A)

All taxa

1 BACH1_1, PNN_3 TrN+G

2 BACH1_2, PTPN_2 TVM+G

3 BACH1_3, DNAH3_3, NT3_3, PTPN_3, RAG2_3 K81uf+G

4 BMP2_1, BMP2_2 K80+I

5 BMP2_3, CMOS_2, TRAF6_1 K80+G

6 CMOS_1, DNAH3_1, IRBP_1, PTPN_1, RAG2_1 GTR+I+G

7 CMOS_3 HKY+G

8 DNAH3_2, IRBP_2, NT3_1, NT3_2, RAG2_2, TRAF6_2 GTR+I+G

9 IRBP_3, TRAF6_3 TVM+G

10 PNN_1 TrN+G

11 PNN_2 GTR+G

Nuclear

(B)

No Tinamous

1 BACH1_1, CMOS_1, DNAH3_1, IRBP_1, PTPN_1,

RAG2_1 GTR+G

2 BACH1_2, PTPN_2 TVM+G

3 BACH1_3, DNAH3_3, NT3_3, RAG2_3 HKY+G

4 BMP2_1, BMP2_2 K80+I

5 BMP2_3, CMOS_2, TRAF6_1 K80+G

6 CMOS_3 HKY+G

7 DNAH3_2, IRBP_2, NT3_1, NT3_2, RAG2_2, TRAF6_2 TVM+I+G

8 IRBP_3, TRAF6_3 TVM+G

9 PNN_1 TrN+G

10 PNN_2 GTR+G

11 PNN_3, PTPN_3 K81+G

Nuclear

(C)

No Crocodilians

1 BACH1_1, PNN_3 TrN+G

2 BACH1_2, PTPN_2 TVM+G

3 BACH1_3, DNAH3_3, IRBP_3, NT3_3, PTPN_3,

RAG2_3, TRAF6_3 K80+G

4 BMP2_1, BMP2_2 K80+I

5 BMP2_3, CMOS_2, DNAH3_2, IRBP_2, NT3_2, RAG2_2,

TRAF6_2 GTR+I+G

6 CMOS_1, DNAH3_1, IRBP_1, NT3_1, PNN_1, PTPN_1,

RAG2_1, TRAF6_1 GTR+I+G

7 CMOS_3 HKY+G

8 PNN_2 GTR+G

Nuclear

(D)

No tinamous or

crocodilians

1 BACH1_1, CMOS_1, DNAH3_1, IRBP_1, PTPN_1,

RAG2_1 TVM+G

2 BACH1_2, PTPN_2 HKY+G

3 BACH1_3, DNAH3_3, NT3_3, RAG2_3 HKY+G

4 BMP2_1, BMP2_2 K80+I

5 BMP2_3, CMOS_2, NT3_1, TRAF6_1 K80+G

6 CMOS_3 HKY+G

7 DNAH3_2, IRBP_2, NT3_2, RAG2_2, TRAF6_2 GTR+I+G

8 IRBP_3, PNN_3, PTPN_3 K81+G

9 PNN_1 TrN+G

10 PNN_2 GTR+G

11 TRAF6_3 K80+G

Table S8: Node ages and 95% HPDs for palaeognath taxa, estimated from molecular dating analyses. Node Mito All Mito. A Mito. B Mito. C Mito. D Nuclear A Nuclear B Nuclear C Nuclear D

Struthionidae|Rheidae+Casuariidae+Aepyornit

hidae+Apterygidae+Tinamidae+Dinornithifor

mes

72.9

(62.6-84.2)

76.0

(63.5-89.1)

79.8

(67.2-92.9)

63.3

(52.3-74.8)

63.4

(54.7-73.0)

88.9

(74.5-104.3)

82.3

(66.2-99.8)

69.4

(56.6-83.3)

62.4

(50.6-74.5)

Rheidae|Casuariidae+Aepyornithidae+Apteryg

idae+Tinamidae+Dinornithiformes

68.1

(58.6-78.7)

70.4

(58.9-83.3)

74.7

(63.0-87.3)

56.4

(47.1-66.8)

57.3

(49.5-65.9)

77.8

(64.3-92.9)

72.6

(58.3-89.4)

55.0

(44.6-65.0)

50.0

(40.5-60.0)

Rhea|Pterocnemia 8.6

(5.6-12.0)

8.8

(5.8-12.1)

9.5

(6.7-12.9)

8.7

(5.9-11.7)

9.5

(7.0-12.3)

3.2

(1.5-5.1)

3.1

(1.4-5.0)

3.5

(1.8-5.4)

3.2

(1.7-4.9)

Casuariidae+Aepyornithidae+Apterygidae|

Tinamidae+Dinornithiformes

65.3

(56.2-75.8)

67.4

(56.2-80.0)

64.9

(59.9-83.5)

52.1

(43.4-62.0)

52.9

(45.6-61.2)

77.5

(64.3-92.9)

72.3

(58.2-89.2)

54.7

(44.3-66.1)

49.7

(40.1-59.3)

Casuariidae|Aepyornithidae+Apterygidae 57.9

(48.1-68.6)

60.6

(49.1-73.8)

71.5

(53.5-77.4)

47.7

(39.7-57.0)

48.8

(41.7-56.6)

76.3

(63.2-92.1)

71.1

(56.3-87.5)

53.7

(43.2-65.0)

48.8

(39.4-58.6)

Dromaius|Casuarius 26.8

(25.0-30.1)

27.4

(25.0-31.6)

27.5

(25.0-31.7)

26.6

(25.0-29.5)

26.4

(25.0-29.0)

27.0

(25.0-30.9)

27.1

(25.0-31.2)

26.7

(25.0-30.0)

26.3

(25.0-29.0)

Casuarius bennetti|C. casuarius 4.3

(2.5-6.4)

- - - - - - - -

Aepyornithidae|Apterygidae 50.0

(40.1-61.5)

- - - - - - - -

Aepyornis|Mullerornis 17.2

(11.3-24.0)

- - - - - - - -

Apteryx mantelii|A. haasti+A. owenii 8.6

(5.6-11.9)

8.1

(4.6-11.7)

9.7

(6.2-13.6)

8.0

(4.8-11.4)

9.6

(6.7-13.0)

4.0

(2.0-6.2)

4.0

(2.0-6.5)

4.2

(2.1-6.5)

3.9

(2.0-5.8)

A. haasti |A. owenii 2.9

(1.9-4.1)

- - - - - - - -

Tinamidae|Dinornithiformes 58.0

(49.5-68.1)

59.0

(48.6-71.3)

62.6

(52.1-74.0)

- - 62.3

(49.8-75.9)

59.0

(46.2-74.1)

- -

Eudromia|Tinamus+Crypturellus 41.9

(34.2-50.7)

39.4

(30.1-50.1)

41.9

(33.2-51.1)

- - 36.8

(27.5-47.4)

34.9

(25.4-46.2)

- -

Tinamus|Crypturellus 32.1

(25.2-40.4)

- - - - - - - -

Dinornis|Emeus+Anomalopteryx 7.7

(5.5-10.1)

- - - - - - - -

Emeus| Anomalopteryx 3.4

(2.2-4.7)

- - - - - - - -

Table S9: Comparison of statistical support for elephant bird + kiwi between total evidence analyses

Support for elephant bird + kiwi (all data / molecular only)

Effect of tinamous and outgroups MP bootstrap ML Bootstrap Bayesian PP

Ratites + Outgroups (i.e. rooted) 99 / 99 94 / 99 1 / 1

Ratites + Outgroups (i.e. rooted), no tinamous 96 / 99 60 / 100 1 / 1

Ratites only (i.e. unrooted) 90 / 96 81 / 95 1 / 1

Ratites only (i.e. unrooted), no tinamous 84 / 97 57 / 93 1 / 1

Effect of enforcing nuclear constraints (backbone/scaffold) MP bootstrap Bayesian PP

Rhea+ tinamous 99 / 99 87 / 100 1 / 1

Rhea+kiwi+casuariids 98 / 95 95 / 99 1 / 1

Rhea+casuariids 99 / 100 56 / 63 1 / 1

Topology tests: optimal topology vs. Africa basal MP Templeton

(p–value)

ML SH test

(p-value)

Δ Bayes factor

Ratites + Outgroups (i.e. rooted) 0.005 / 0.0001 0.01 / 0.01 169.20 / 161.80

Ratites only (i.e. unrooted) 0.01 / 0.005 0.02 / 0.02 52.04 / 643.00

Table S10: Partitioning schemes and substitution models for combined nuclear

and mitochondrial data used in total evidence Bayesian and maximum likelihood

analyses. Program Partition

Number

Composition Model

MrBayes 1 Mitochondrial ND6 locus GTR+I+G

MrBayes 2 Mitochondrial 1st codon positions GTR+I+G

MrBayes 3 Mitochondrial 2nd codon positions GTR+I+G

MrBayes 4 Mitochondrial 3rd codon positions F81

MrBayes 5 Mitochondrial RNA loop positions GTR+I+G

MrBayes 6 Mitochondrial RNA stem positions SYM+I+G

MrBayes 7 BACH1_1, CMOS_1, DNAH3_1, IRBP_1, PTPN_1, RAG2_1 GTR+G

MrBayes 8 BACH1_2, PTPN_2 GTR+G

MrBayes 9 BACH1_3, DNAH3_3, NT3_3, RAG2_3 HKY+G

MrBayes 10 BMP2_1, BMP2_2 K80+I

MrBayes 11 BMP2_3, CMOS_2, TRAF6_1 K80+G

MrBayes 12 CMOS_3 HKY+G

MrBayes 13 DNAH3_2, IRBP_2, NT3_1, NT3_2, RAG2_2, TRAF6_2 GTR+I+G

MrBayes 14 IRBP_3, PTPN_3, TRAF6_3 K80+G

MrBayes 15 PNN_1 HKY+G

MrBayes 16 PNN_2 GTR+G

MrBayes 17 PNN_3 K80+G

RAxML 1 BACH1_1, PNN_3 GTR+G

RAxML 2 BACH1_2, PTPN_2 GTR+G

RAxML 3 BACH1_3, DNAH3_3, NT3_3, PTPN_3, RAG2_3 GTR+G

RAxML 4 BMP2_1, BMP2_2, NT3_2 GTR+G

RAxML 5 BMP2_3, CMOS_2, DNAH3_2, IRBP_2, RAG2_2, TRAF6_2 GTR+G

RAxML 6 CMOS_1, DNAH3_1, IRBP_1, NT3_1, PNN_1, PTPN_1, RAG2_1,

TRAF6_1 GTR+G

RAxML 7 CMOS_3 GTR+G

RAxML 8 IRBP_3, TRAF6_3 GTR+G

RAxML 9 PNN_2 GTR+G

RAxML 10 Mitochondrial ND6 locus GTR+G

RAxML 11 Mitochondrial 1st codon positions GTR+G

RAxML 12 Mitochondrial 2nd codon positions GTR+G

RAxML 13 Mitochondrial 3rd codon positions GTR+G

RAxML 14 Mitochondrial RNA loop positions GTR+G

RAxML 15 Mitochondrial RNA stem positions GTR+G

References and Notes

1. M. Bunce, T. H. Worthy, M. J. Phillips, R. N. Holdaway, E. Willerslev, J. Haile, B. Shapiro,

R. P. Scofield, A. Drummond, P. J. Kamp, A. Cooper, The evolutionary history of the

extinct ratite moa and New Zealand Neogene paleogeography. Proc. Natl. Acad. Sci.

U.S.A. 106, 20646–20651 (2009). Medline doi:10.1073/pnas.0906660106

2. T. H. Worthy, R. N. Holdaway, The Lost World of the Moa (Indiana Univ. Press,

Bloomington, IN, 2002).

3. J. P. Hume, M. Walters, Extinct Birds (Bloomsbury, London, 2012).

4. J. Cracraft, Continental drift, paleoclimatology, and the evolution and biogeography of birds.

J. Zool. 169, 455–543 (1973). doi:10.1111/j.1469-7998.1973.tb03122.x

5. K. Lee, J. Feinstein, J. Cracraft, in Avian Molecular Evolution and Systematics, D. Mindell,

Ed. (Academic Press, New York, 1997), pp. 173–208.

6. T. J. Parker, On the cranial osteology, classification and phylogeny of Dinornithidae. Trans.

Zool. Soc. London 13, 373–431 (1895). doi:10.1111/j.1096-3642.1895.tb00044.x

7. J. Cracraft, Phylogeny and evolution of the ratite birds. Ibis 116, 494–521 (1974).

doi:10.1111/j.1474-919X.1974.tb07648.x

8. W. P. Schellart, G. S. Lister, V. G. Toy, A Late Cretaceous and Cenozoic reconstruction of the

Southwest Pacific region: Tectonics controlled by subduction and slab rollback

processes. Earth Sci. Rev. 76, 191–233 (2006). doi:10.1016/j.earscirev.2006.01.002

9. A. Cooper, C. Mourer-Chauviré, G. K. Chambers, A. von Haeseler, A. C. Wilson, S. Pääbo,

Independent origins of New Zealand moas and kiwis. Proc. Natl. Acad. Sci. U.S.A. 89,

8741–8744 (1992). Medline doi:10.1073/pnas.89.18.8741

10. A. H. Bledsoe, A phylogenetic analysis of postcranial skeletal characters of the ratite birds.

Ann. Carnegie Mus. 57, 73 (1988).

11. O. Haddrath, A. J. Baker, Multiple nuclear genes and retroposons support vicariance and

dispersal of the palaeognaths, and an Early Cretaceous origin of modern birds. Proc. Biol.

Sci. 279, 4617–4625 (2012). Medline doi:10.1098/rspb.2012.1630

12. M. J. Phillips, G. C. Gibb, E. A. Crimp, D. Penny, Tinamous and moa flock together:

Mitochondrial genome sequence analysis reveals independent losses of flight among

ratites. Syst. Biol. 59, 90–107 (2010). Medline doi:10.1093/sysbio/syp079

13. J. Harshman, E. L. Braun, M. J. Braun, C. J. Huddleston, R. C. Bowie, J. L. Chojnowski, S.

J. Hackett, K. L. Han, R. T. Kimball, B. D. Marks, K. J. Miglia, W. S. Moore, S. Reddy,

F. H. Sheldon, D. W. Steadman, S. J. Steppan, C. C. Witt, T. Yuri, Phylogenomic

evidence for multiple losses of flight in ratite birds. Proc. Natl. Acad. Sci. U.S.A. 105,

13462–13467 (2008). Medline doi:10.1073/pnas.0803242105

14. J. V. Smith, E. L. Braun, R. T. Kimball, Ratite nonmonophyly: Independent evidence from

40 novel Loci. Syst. Biol. 62, 35–49 (2013). Medline doi:10.1093/sysbio/sys067

15. J. R. Ali, D. W. Krause, Late Cretaceous bioconnections between Indo- Madagascar and

Antarctica: Refutation of the Gunnerus Ridge causeway hypothesis. J. Biogeogr. 38,

1855–1872 (2011). doi:10.1111/j.1365-2699.2011.02546.x

16. P. Johnston, New morphological evidence supports congruent phylogenies and Gondwana

vicariance for palaeognathous birds. Zool. J. Linn. Soc. 163, 959–982 (2011).

doi:10.1111/j.1096-3642.2011.00730.x

17. T. H. Worthy, R. P. Scofield, Twenty-first century advances in knowledge of the biology of

moa (Aves: Dinornithiformes): A new morphological analysis and moa diagnoses

revised. N.Z. J. Zool. 39, 87–153 (2012). doi:10.1080/03014223.2012.665060

18. A. Cooper, C. Lalueza-Fox, S. Anderson, A. Rambaut, J. Austin, R. Ward, Complete

mitochondrial genome sequences of two extinct moas clarify ratite evolution. Nature 409,

704–707 (2001). Medline doi:10.1038/35055536

19. For details, see materials and methods on Science Online.

20. T. H. Worthy, J. P. Worthy, A. J. D. Tennyson, S. W. Salisbury, S. J. Hand, R. P. Scofield, in

Paleornithological Research 2013 - Proceedings of the 8th International Meeting of the

Society of Avian Paleontology and Evolution, U. B. Göhlich, A. Kroh, Eds. (Natural

History Museum Vienna, Vienna, Austria, 2013), pp. 63–80.

21. G. Mayr, in Paleogene Fossil Birds (Springer, Berlin, 2009), pp. 25–34.

22. See supplementary text on Science Online.

23. K. Black, M. Archer, S. Hand, H. Godthelp, in Earth and Life, J. Talent, Ed. (Springer

Netherlands, 2012), pp. 983–1078.

24. M. Meyer, M. Kircher, Illumina sequencing library preparation for highly multiplexed target

capture and sequencing. Cold Spring Harbor Protoc. 2010, 10.1101/pdb.prot5448

(2010). Medline doi:10.1101/pdb.prot5448

25. M. Martin, Cutadapt removes adapter sequences from high-throughput sequencing reads.

Bioinformatics Action 17, 10 (2012).

26. C. Hahn, L. Bachmann, B. Chevreux, Reconstructing mitochondrial genomes directly from

genomic next-generation sequencing reads—A baiting and iterative mapping approach.

Nucleic Acids Res. 41, e129 (2013). Medline doi:10.1093/nar/gkt371

27. R. E. Green, A. S. Malaspinas, J. Krause, A. W. Briggs, P. L. Johnson, C. Uhler, M. Meyer,

J. M. Good, T. Maricic, U. Stenzel, K. Prüfer, M. Siebauer, H. A. Burbano, M. Ronan, J.

M. Rothberg, M. Egholm, P. Rudan, D. Brajković, Z. Kućan, I. Gusić, M. Wikström, L.

Laakkonen, J. Kelso, M. Slatkin, S. Pääbo, A complete Neandertal mitochondrial genome

sequence determined by high-throughput sequencing. Cell 134, 416–426 (2008). Medline

doi:10.1016/j.cell.2008.06.021

28. H. Li, B. Handsaker, A. Wysoker, T. Fennell, J. Ruan, N. Homer, G. Marth, G. Abecasis, R.

Durbin, The Sequence Alignment/Map (SAM) format and SAMtools. Bioinformatics 25,

2078–2079 (2009). doi:10.1093/bioinformatics/btp352

29. A. Ginolhac, M. Rasmussen, M. T. P. Gilbert, E. Willerslev, L. Orlando, mapDamage:

Testing for damage patterns in ancient DNA sequences. Bioinformatics 27, 2153–2155

(2011). Medline doi:10.1093/bioinformatics/btr347

30. MEGAN 5 (Metagenome Analyzer), written by D. H. Huson, original design by D. H. Huson

and S. C. Schuster, with contributions from S. Mitra, D. C. Richter, P. Rupek, H.-J.

Ruscheweyh, R. Tappu, and N. Weber; http://ab.inf.uni-tuebingen.de/software/megan5/.

31. H. Li, R. Durbin, Fast and accurate short read alignment with Burrows-Wheeler transform.

Bioinformatics 25, 1754–1760 (2009). Medline doi:10.1093/bioinformatics/btp324

32. M. Kircher, in Ancient DNA: Methods and Protocols, B. Shapiro, M. Hofreiter, Eds. (Springer, New York, 2012), vol. 840, pp. 197–228.

33. M. Schubert, A. Ginolhac, S. Lindgreen, J. F. Thompson, K. A. Al-Rasheid, E. Willerslev, A.

Krogh, L. Orlando, Improving ancient DNA read mapping against modern reference

genomes. BMC Genomics 13, 178 (2012). Medline doi:10.1186/1471-2164-13-178

34. R. Lanfear, B. Calcott, S. Y. W. Ho, S. Guindon, PartitionFinder: Combined selection of

partitioning schemes and substitution models for phylogenetic analyses. Mol. Biol. Evol.

29, 1695–1701 (2012). Medline doi:10.1093/molbev/mss020

35. A. Stamatakis, P. Hoover, J. Rougemont, A rapid bootstrap algorithm for the RAxML Web

servers. Syst. Biol. 57, 758–771 (2008). Medline doi:10.1080/10635150802429642

36. A. Stamatakis, RAxML-VI-HPC: Maximum likelihood-based phylogenetic analyses with

thousands of taxa and mixed models. Bioinformatics 22, 2688–2690 (2006). Medline

doi:10.1093/bioinformatics/btl446

37. F. Ronquist, M. Teslenko, P. van der Mark, D. L. Ayres, A. Darling, S. Höhna, B. Larget, L.

Liu, M. A. Suchard, J. P. Huelsenbeck, MrBayes 3.2: Efficient Bayesian phylogenetic

inference and model choice across a large model space. Syst. Biol. 61, 539–542 (2012).

Medline doi:10.1093/sysbio/sys029

38. A. J. Drummond, A. Rambaut, BEAST: Bayesian evolutionary analysis by sampling trees.

BMC Evol. Biol. 7, 214 (2007). Medline doi:10.1186/1471-2148-7-214

39. A. Rambaut, A. J. Drummond, Tracer v1.4, available from

http://tree.bio.ed.ac.uk/software/tracer/ (2007).

40. X. Xia, Z. Xie, M. Salemi, L. Chen, Y. Wang, An index of substitution saturation and its

application. Mol. Phylogenet. Evol. 26, 1–7 (2003). Medline doi:10.1016/S1055-

7903(02)00326-3

41. X. Xia, P. Lemey, in The Phylogenetic Handbook: A Practical Approach to Phylogenetic

Analysis and Hypothesis Testing, P. Lemey, M. Salemi, A. Vandamme, Eds. (Cambridge

Univ. Press, Cambridge, 2009).

42. X. Xia, Z. Xie, DAMBE: Software package for data analysis in molecular biology and

evolution. J. Hered. 92, 371–373 (2001). Medline doi:10.1093/jhered/92.4.371

43. D. L. Swofford, PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods)

(Sinauer Associates, Sunderland, MA, 2002).

44. M. D. Sorenson, E. A. Franzosa, TreeRot, version 3 (Boston Univ., Boston, 2007).

45. J. Gatesy, G. Amato, M. Norell, R. DeSalle, C. Hayashi, Combined support for wholesale

taxic atavism in gavialine crocodylians. Syst. Biol. 52, 403–422 (2003). Medline

doi:10.1080/10635150390197037

46. P. O. Lewis, A likelihood approach to estimating phylogeny from discrete morphological

character data. Syst. Biol. 50, 913–925 (2001). Medline

doi:10.1080/106351501753462876

47. P. Houde, S. L. Olson, Paleognathous carinate birds from the early tertiary of North America.

Science 214, 1236–1237 (1981). Medline doi:10.1126/science.214.4526.1236

48. P. W. Houde, Palaeognathous birds from the early Tertiary of the Northern Hemisphere.

Publ. Nuttall Ornithol. Club 22, 1 (1988).

49. G. Grellet-Tinner, G. J. Dyke, The eggshell of the Eocene bird Lithornis. Acta Palaeontol.

Pol. 50, 831 (2005).

50. P. G. Foster, D. A. Hickey, Compositional bias may affect both DNA-based and protein-

based phylogenetic reconstructions. J. Mol. Evol. 48, 284–290 (1999). Medline

doi:10.1007/PL00006471

51. M. J. Phillips, D. Penny, The root of the mammalian tree inferred from whole mitochondrial

genomes. Mol. Phylogenet. Evol. 28, 171–185 (2003). Medline doi:10.1016/S1055-

7903(03)00057-5

52. P. J. Lockhart, M. A. Steel, M. D. Hendy, D. Penny, Recovering evolutionary trees under a

more realistic model of sequence evolution. Mol. Biol. Evol. 11, 605–612 (1994).

Medline

53. S. J. Gould, Of Kiwi eggs and the Liberty Bell. Nat. Hist. 95, 20 (1986).

54. S. J. Gould, in Bully for Brontosaurus, S. J. Gould, Ed. (W. Norton, New York, 1991), pp.

109–123.

55. H. M. F. Alvarenga, Uma ave ratitae do Paleoceno Brasileiro: Bacia calcária de Itaboraí,

Estado do Rio de Janeiro, Brasil. Bol. Mus. Nac. Geol. 41, 1 (1983).

56. C. W. Andrew, On the pelvis and hind-limb of Mullerornis betsilei M.-Edw. & Grand.; with

a note on the occurrence of a ratite bird in the Upper Eocene beds of the Fayum, Egypt.

Proc. Zool. Soc. Lond. 74, 163–171 (1904).

57. L. D. Martin, in Papers in Avian Paleontology Honoring Pierce Brodkorb, in Natural

History Museum of Los Angeles County Science Series 36, K. E. Campbell, Ed. (Natural

History Museum of Los Angeles County, Los Angeles, 1992), pp. 97–125.

58. P. W. Houde, H. Haubold, Palaeotis weigelti restudied: A small middle Eocene ostrich

(Aves: Struthioniformes). Palaeovertebrata 17, 27 (1987).

59. C. P. Tambussi, J. I. Noriega, A. Gaździcki, A. Tatur, M. A. Reguero, S. F. Vizcaino, Ratite

bird from the Paleogene La Meseta Formation, Seymour Island, Antarctica. Polish Polar

Res. 15, 15–20 (1994).

60. M. J. Benton, in The Fossil Record 2, M. J. Benton, Ed. (Chapman & Hall, London, 1993),

pp. 681–715.

61. M. J. Benton, P. C. J. Donoghue, Paleontological evidence to date the tree of life. Mol. Biol.

Evol. 24, 26–53 (2007). Medline doi:10.1093/molbev/msl150

62. J. Gauthier, Saurischian monophyly and the origin of birds. Mem. Calif. Acad. Sci. 8, 1

(1986).

63. P. C. Sereno, Basal archosaurs: Phylogenetic relationships and functional implications. Soc.

Vertebr. Paleontol. Mem. 2, 1–53 (1991). doi:10.2307/3889336

64. M. J. Benton, Scleromochlus taylori and the origin of dinosaurs and pterosaurs. Philos.

Trans. R. Soc. London Ser. B 354, 1423–1446 (1999). doi:10.1098/rstb.1999.0489

65. C. A. Brochu, Patterns of calibration age sensitivity with quartet dating methods. J.

Paleontol. 78, 7–30 (2004). doi:10.1666/0022-

3360(2004)078<0007:POCASW>2.0.CO;2

66. T. E. Williamson, Brachychampsa sealeyi, sp nov., (Crocodylia, Alligatoroidea) from the

Upper Cretaceous (lower Campanian) Menefee Formation, northwestern New Mexico. J.

Vertebr. Paleontol. 16, 421–431 (1996). doi:10.1080/02724634.1996.10011331

67. A. D. Buscaloni, F. Ortega, D. Vasse, New crocodiles (Eusuchia: Alligatoroidea) from the

Upper Cretaceous of southern Europe. Comptes Rendus l’Acad. Sci. Paris. Sci. Terre

Plandtes 325, 525 (1997).

68. X.-C. Wu, A. P. Russell, D. B. Brinkman, A review of Leidyosuchus canadensis Lambe,

1907 (Archosauria: Crocodylia) and an assessment of cranial variation based upon new

material. Can. J. Earth Sci. 38, 1665–1687 (2001). doi:10.1139/e01-059

69. D. R. Schwimmer, King of the Crocodylians: The Paleobiology of Deinosuchus. (Indiana

Univ. Press, Bloomington, IN, 2002).

70. J. A. Clarke, C. P. Tambussi, J. I. Noriega, G. M. Erickson, R. A. Ketcham, Definitive fossil

evidence for the extant avian radiation in the Cretaceous. Nature 433, 305–308 (2005).

Medline doi:10.1038/nature03150

71. Z. Zhonghe, Evolutionary radiation of the Jehol Biota: Chronological and ecological

perspectives. Geol. J. 41, 377–393 (2006). doi:10.1002/gj.1045

72. K. E. Slack, C. M. Jones, T. Ando, G. L. Harrison, R. E. Fordyce, U. Arnason, D. Penny,

Early penguin fossils, plus mitochondrial genomes, calibrate avian evolution. Mol. Biol.

Evol. 23, 1144–1155 (2006). Medline doi:10.1093/molbev/msj124

73. C. Patterson, P. V. Rich, The fossil history of the emus, Dromaius (Aves: Dromaiinae). Rec.

South Austral. Mus. (Adelaide) 21, 85 (1987).

74. T. H. Worthy, S. J. Hand, M. Archer, Phylogenetic relationships of the Australian Oligo-

Miocene ratite Emuarius gidju Casuariidae. Integr. Zool. 9, 148–166 (2014). Medline

doi:10.1111/1749-4877.12050

75. W. E. Boles, “Revision of Dromaius gidju Patterson and Rich 1987 from Riversleigh

Northwestern Queensland Australia with a Reassessment of its generic position.” in

Natural History Museum of Los Angeles City Science Series 195 (Natural History

Museum of Los Angeles City, Los Angeles, 1992).

76. O. Haddrath, A. J. Baker, Complete mitochondrial DNA genome sequences of extinct birds:

Ratite phylogenetics and the vicariance biogeography hypothesis. Proc. R. Soc. London

Ser. B 268, 939–945 (2001). Medline doi:10.1098/rspb.2001.1587