The historical biogeography of Mammalia€¦ · Research The historical biogeography of Mammalia Mark S. Springer1,*, Robert W. Meredith1, Jan E. Janecka2 and William J. Murphy2 1Department

doi 101098rstb20110023 2478-2502366 2011 Phil Trans R Soc B

Mark S Springer Robert W Meredith Jan E Janecka and William J Murphy The historical biogeography of Mammalia

References

httprstbroyalsocietypublishingorgcontent36615772478fullhtmlrelated-urls Article cited in

httprstbroyalsocietypublishingorgcontent36615772478fullhtmlref-list-1

This article cites 165 articles 53 of which can be accessed free

Subject collections

(45 articles)taxonomy and systematics (31 articles)palaeontology

(535 articles)evolution Articles on similar topics can be found in the following collections

Email alerting service hereright-hand corner of the article or click Receive free email alerts when new articles cite this article - sign up in the box at the top

httprstbroyalsocietypublishingorgsubscriptions go to Phil Trans R Soc BTo subscribe to

on October 29 2012rstbroyalsocietypublishingorgDownloaded from

Phil Trans R Soc B (2011) 366 2478ndash2502

doi101098rstb20110023

Research

One conmamma

The historical biogeography of MammaliaMark S Springer1 Robert W Meredith1 Jan E Janecka2

and William J Murphy2

1Department of Biology University of California Riverside CA 92521 USA2Department of Veterinary Integrative Biosciences College of Veterinary Medicine and Biomedical Sciences

Texas AampM University College Station TX 77843 USA

Palaeobiogeographic reconstructions are underpinned by phylogenies divergence times and ances-tral area reconstructions which together yield ancestral area chronograms that provide a basis forproposing and testing hypotheses of dispersal and vicariance Methods for area coding includemulti-state coding with a single character binary coding with multiple characters and stringcoding Ancestral reconstruction methods are divided into parsimony versus Bayesianlikelihoodapproaches We compared nine methods for reconstructing ancestral areas for placental mammalsAmbiguous reconstructions were a problem for all methods Important differences resulted fromcoding areas based on the geographical ranges of extant species versus the geographical provenanceof the oldest fossil for each lineage Africa and South America were reconstructed as the ancestralareas for Afrotheria and Xenarthra respectively Most methods reconstructed Eurasia as the ances-tral area for Boreoeutheria Euarchontoglires and Laurasiatheria The coincidence of moleculardates for the separation of Afrotheria and Xenarthra at approximately 100 Ma with the plate tec-tonic sundering of Africa and South America hints at the importance of vicariance in the earlyhistory of Placentalia Dispersal has also been important including the origins of Madagascarrsquosendemic mammal fauna Further studies will benefit from increased taxon sampling and theapplication of new ancestral area reconstruction methods

Keywords ancestral areas dispersal historical biogeography Mammalia vicariance

1 INTRODUCTIONClass Mammalia is impressive for its taxonomic ecologi-cal and morphological diversity [1] A fundamental goalof mammalian palaeobiogeography is to reconstruct theunderlying history of vicariant and dispersal events thathave shaped this diversity Here we highlight theimportance of phylogeny reconstruction ancestral areareconstruction and molecular dating for producingancestral area chronograms We compare differentapproaches for reconstructing ancestral areas andillustrate similarities and differences between theseapproaches using a dataset for placental mammals Weconclude with a review of selected topics in placentalmammal palaeogeography that illustrates how phyloge-nies ancestral area reconstructions molecular datesand palaeographic histories have reshaped our views onmammalian historical biogeography Finally we identifyimportant areas for future inquiry

2 PHYLOGENY RECONSTRUCTIONPhylogeny reconstruction begins with character dataLarge molecular datasets have yielded robust phyloge-nies for many groups thereby reducing the number of

r for correspondence (markspringerucredu)

tribution of 12 to a Theme Issue lsquoGlobal biodiversity oflsrsquo

phylogenetic hypotheses that must be considered whenformulating ancestral area chronograms The inclusionof morphological data from fossils allows for taxono-mically richer phylogenies while also providing keydata points that bear on the geographical provenanceof a taxonomic group In the words of Simpson [2]fossils lsquoare the historical documents of animal distri-butionrsquo Fossils are also more difficult to place withconfidence in a phylogenetic framework owing to miss-ing (molecular) data and the inability of currentmethods to separate homology and homoplasy withsome morphological datasets [34]

Maximum parsimony (MP) and maximum likeli-hood (ML) yield a single best tree(s) whereasBayesian methods yield a sampling of trees from pos-terior probability space ML and Bayesian methodshave the advantage of incorporating models ofsequence evolution and yield trees with branchlengths Some ancestral area reconstruction methodssuch as those implemented in SIMMAP [5] cantake advantage of trees with branch lengths as wellas multiple trees from posterior probability space

Phylogeny reconstruction is usually the first step inconstructing an ancestral area chronogram followedby the estimation of divergence times at each of thenodes However BEAST [6] allows for simultaneousestimation of branching relationships and divergencetimes After reconstructing a phylogeny molecular

This journal is q 2011 The Royal Society

palaeobiogeographicreconstructions

phylogenies

paleontologicaldata

chronograms

ancestral areachronograms

moleculardating

palaeogeographicreconstructions

geophysicsclimatologysedimentologystratigraphygeology

phylogenyreconstruction

ancestral areareconstruction

neontologicaldata

general area chronogram

Figure 1 A flowchart of the approach used for incorporating different types of data in conjunction with methods in phylogenyreconstruction molecular dating and ancestral area reconstruction for inferring ancestral area chronograms and palaeobiogeo-graphic history

Mammalian historical biogeography M S Springer et al 2479

dating analyses and ancestral area reconstructions canbe performed in parallel or in series and thenintegrated to yield an ancestral area chronogram(figure 1)

3 MOLECULAR DATING ANALYSESMolecular clocks were introduced by Zuckerkandl ampPauling [7] but have fallen out of favour owing tothe prevalence of lineage-specific rate variation Theemergence of relaxed molecular clock methods haspromoted a resurgence of studies that have examinedboth interordinal and intraordinal divergence timesin Mammalia [8ndash23] Relaxed clock methods includepenalized likelihood approaches [2425] and BayesianMarkov chain Monte Carlo methods such as multidiv-time [26] BEAST [6] and mcmctree [27] It is useful tocompare both the results of different programs and theresults of the same program under different model andparameter settings [28ndash30]

An important difference between BEASTand multi-divtime is that BEAST allows rates to vary randomlyover lineages in a phylogeny whereas multidivtimeassumes autocorrelated rates In simulation studiesBEAST performed poorly when rates were autocorre-lated whereas multidivtime performed poorly whenthere was uncorrelated rate variation [28] Giventhese results Battistuzzi et al [28] recommended com-posite 95 credibility intervals

Relaxed clock methods allow for multiple calibra-tions including minimum and maximum constraintson individual nodes Multidivtime only allows forlsquohardrsquo constraints whereas BEAST and mcmctree pro-vide other options including lsquosoftrsquo constraints thatpermit specification of a given percentage (eg 95)of the normal distribution between the minimumand maximum with half of the remainder (eg25) allocated to each tail

Phil Trans R Soc B (2011)

4 ANCESTRAL AREA RECONSTRUCTIONMethods for reconstructing historical biogeographyinclude dispersalism phylogenetic biogeography pan-biogeography parsimony analysis of endemicity andcladistic biogeography [31] Early reconstructions ofmammalian historical biogeography were based on dis-persalism and land bridges [232] and pre-date thegeneral acceptance of plate tectonic theory Sub-sequently cladistic biogeography emphasized vicarianceas the most important factor in diversification bydiscovering dichotomous area relationships (area clado-grams) from taxon cladograms In response to thisparadigm which paid little regard to dispersal andextinction Ronquist [33] proposed dispersalndashvicarianceanalysis (DIVA) for reconstructing patterns of historicalbiogeography [34] DIVA infers ancestral areas by mini-mizing the number of dispersal and extinction eventsRecent methods that build on Ronquistrsquos work includeBayes-DIVA [35] and dispersalndashextinction cladogenesis(DEC) [3637]

A fundamental issue in ancestral area reconstructionis area coding Areas are usually coded to include theentire geographical range of each species Other optionsinclude coding the entire area of the monophyletic cladethat is represented by the species or the geographicalarea of the oldest fossil belonging to each lineage Anadditional topic worthy of investigation is the problemof coding geographical areas for taxa from the geologicalpast versus the present given that areas as well as theirboundaries and physical relationships to each othercan fluctuate over time Parametric methods such asDEC which allow for changing dispersal probabilitiesover time provide a mechanism to accommodate theimpact of continental fragmentation and suturing on his-torical biogeography

Three general approaches are available for codingareas (box 1) whether for living species fossils orlarger monophyletic groups The first method is

Box 1 Methods for coding areas and analysing area-coded data matrices

I Area coding1 single multi-state character coding Individual character states are non-overlapping and correspond to a single areadisadvantages ranges are limited to a single area (character state) unless they are coded as polymorphic2 binary character coding with multiple characters Each binary character corresponds to the presenceabsence of a taxonin a single areaadvantages allows for the occupation of multiple areasdisadvantages ancestral areas may receive no state assignments3 string character coding (frac14polymorphism coding)advantages individual character states may include one or more areasdisadvantages the number of character states becomes intractable when there are too many individual areas

II Ancestral area reconstruction1 monomorphic ancestral area reconstruction methods These methods are used in conjunction with area data that havebeen coded as a single multi-state charactera Fitch parsimony (eg MACCLADE)b stochastic mapping (eg SIMMAP)advantages stochastic mapping allows for branch lengths and multiple treesdisadvantages methods in this category implicitly assume that different character states (areas) are homologous toeach other and attempt to find a single ancestral area (character state) at each node2 polymorphic ancestral area reconstruction methods These methods allow for ancestral areas that encompass more thanone area and employ either binary character data for multiple characters or string character dataa Fitch parsimony (eg MACCLADE) with multiple binary charactersb stochastic mapping (eg SIMMAP) with multiple binary charactersc dispersalndashvicariance (DIVA)d Bayes-DIVAe dispersalndashextinction cladogenesis (DEC)f minimum area change (MAC) parsimonyadvantages all methods in this category allow for reconstructions that include multiple areas per node Stochasticmapping and DEC incorporate branch lengths stochastic mapping and Bayes-DIVA allow for multiple treesdisadvantages methods that employ multiple binary characters can result in empty ancestral area reconstructionsFitch parsimony MAC parsimony and DIVA ignore branch length information DIVA Bayes-DIVA and DEC arebiased towards ancestral reconstructions that include numerous individual areas

2480 M S Springer et al Mammalian historical biogeography

single character multi-state coding with non-overlap-ping character states The second method is binarycharacter coding with multiple characters whereeach character represents the presence or absence ofa taxon in a single area In contrast to the firstmethod this approach allows ancestral nodes toencompass more than one area Ancestral area recon-structions are simply the sum of the individual areareconstructions A disadvantage of this approach isthat ancestral areas may receive lsquono-statersquo assignmentswhich imply empty ancestral areas No-state assign-ments are an artefact of the character independenceassumption [38] Finally string character coding[37] allows individual character states to include oneor more geographical areas Specifically the geo-graphical range of a species is coded as a stringdenoting its presenceabsence in a set of individualareas Ree amp Smithrsquos [37] string character coding isequivalent to Maddison amp Maddisonrsquos [39] poly-morphism coding

Ancestral reconstruction methods can be dividedinto parsimony versus Bayesianlikelihood approaches[40] Only the latter takes advantage of branch lengthsAnother useful distinction is between methods thatreconstruct ancestral nodes as monomorphic characterstates versus those that allow for range expansion andcontraction

MP and ML methods employ discrete-state tran-sition models and reconstruct ancestral nodes asmonomorphic Monomorphic methods for character

state reconstruction assume that different characterstates are homologous to each other as is the casefor characters that pass Pattersonrsquos [41] conjunctiontest which states that two structures that are foundin the same organism cannot be homologous How-ever this test is nonsensical when applied togeographical areas because the presence of a speciesin one area does not rule out its presence in anotherarea

Other ancestral range reconstruction methods havethe advantage of allowing for polymorphic ancestralstates and thereby accommodating range expansionand contraction (box 1) [40] Polymorphic reconstruc-tions can be achieved using (i) monomorphic methodswith multiple binary characters each of which codesfor the presenceabsence of a taxon in one area and(ii) polymorphic methods that allow ancestral nodesto include one or more areas

Fitch parsimony and stochastic mapping can beused to reconstruct ancestral nodes for multiplebinary characters and then summed over all characterreconstructions to obtain the complete set of areas foreach ancestral node One difficulty is ancestral nodeswith no-state assignments In these instances multipleinterpretations are possible including vicariance of anancestral area that was not included in the originalanalysis If there is geological evidence for formerlycontiguous areas this information may be incorpor-ated into ancillary characters to assist ancestral areareconstructions

A B C D E F G H I J K L M N O A mdash 2 2 2 1 1 1 3 3 3 2 2 2 4 3 B 2 mdash 2 2 1 3 3 1 1 3 2 2 4 2 3 C 2 2 mdash 2 3 1 3 1 3 1 2 4 2 2 3 D 2 2 2 mdash 3 3 1 3 1 1 4 2 2 2 3 E 1 1 3 3 mdash 2 2 2 2 4 1 1 3 3 2 F 1 3 1 3 2 mdash 2 2 4 2 1 3 1 3 2 G 1 3 3 1 2 2 mdash 4 2 2 3 1 1 3 2 H 3 1 1 3 2 2 4 mdash 2 2 1 3 3 1 2 I 3 1 3 1 2 4 2 2 mdash 2 3 1 3 1 2 J 3 3 1 1 4 2 2 2 2 mdash 3 3 1 1 2 K 2 2 2 4 1 1 3 1 3 3 mdash 2 2 2 1 L 2 2 4 2 1 3 1 3 1 3 2 mdash 2 2 1 M 2 4 2 2 3 1 1 3 3 1 2 2 mdash 2 1 N 4 2 2 2 3 3 3 1 1 1 2 2 2 mdash 1 O 3 3 3 3 2 2 2 2 2 2 1 1 1 1 mdash

A = Africa B = Eurasia C = North America D = South America E = Africa + Eurasia F = Africa + North America G = Africa + South America H = Eurasia + North America I = Eurasia + South America

J = North America + South America

K = Africa + Eurasia + North America

L = Africa + Eurasia + South America

M = Africa + North America + South America

N = Eurasia + North America + South America

O = Africa + Eurasia + North America + South America

Figure 2 Example of a step matrix for minimum area change (MAC) parsimony MAC parsimony assigns equal cost to allgains and losses of an area For example a change in area from A (Africa) to G (Africa thorn South America) requires onestep (gain South America) whereas a change from A to H (Eurasia thornNorth America) requires three steps (Africa loss Eurasiagain North America gain) The step matrix is fully symmetrical

In contrast to methods that were co-opted fromphylogenetics DIVA [33] and DEC [363742] weredeveloped explicitly for historical biogeographicreconstruction DIVA assigns no cost to widespreadancestral areas that are subdivided by vicariance butassigns a cost to dispersal and local extinctionevents DIVA ignores branch lengths DEC uses a con-tinuous time model for geographical range evolutionand employs string character coding to accommodatepolymorphic areas DEC permits range expansionthrough dispersal events and range contractionthrough local extinction events DEC also allowsareas of implausible distribution to be excludedsuch as those that are geographically discontinuous[43] DIVA and DEC are prone to reconstructingancestral areas that include too many individualareas especially towards the root of the tree Howeverboth programmes have options for limiting thenumber of ancestral areas

An additional approach that we introduce is mini-mum area change (MAC) parsimony which usespolymorphic character coding [39] and Sankoff optim-ization and can be implemented with MESQUITE [44]MAC parsimony requires a step matrix (figure 2) Incontrast to DIVA MAC parsimony assigns equal costto all gains and losses of an area whether through dis-persal local extinction or vicariance An advantage ofthis approach is that it should be less prone than DIVAto reconstructing ancestral areas that are too broadrelative to terminal taxa

Another recent approach that builds on earlier cla-distic biogeography methods is phylogenetic analysisof comparing trees (PACT) [45ndash47] Unlike earlier

cladistic biogeography methods PACT explicitlyincorporates molecular dates into general areacladograms

5 ANCESTRAL AREA CHRONOGRAMS ANDPALAEOGEOGRAPHYAncestral area chronograms are similar to ancestral areacladograms but additionally incorporate temporal infor-mation into their framework Alternate approaches forreconstructing phylogeny estimating divergence timesand reconstructing ancestral areas may yield differentancestral area chronograms each of which may beinterpreted in the context of geology-based palaeogeogra-phical hypotheses (figure 1) Ancestral area chronogramsin conjunction with geology-based palaeogeographicalreconstructions provide a framework for proposingtesting and refining palaeobiogeographic hypothesesAncestral area chronograms when interpreted in thecontext of palaeogeographical hypothesis yield insightsinto dispersal vicariance and area extinctions all ofwhich are incorporated into palaeobiogeographichypotheses (figure 1)

Ancestral area chronograms are taxon-specific butancestral area chronograms for multiple taxa that co-occur in the same region can yield general area chron-ograms General area chronograms are similar togeneral area cladograms but include temporal infor-mation that is absent from general area cladogramsThe fundamental idea behind cladistic biogeographyis that broad patterns which are revealed through gen-eral area cladograms demand comprehensive causalexplanations However general area cladograms

ignore temporal information and may result frompseudo-congruence when taxonomic groups with thesame area relationships have different divergencetimes and presumably different underlying causes[48] Temporal information is critical for discriminat-ing between groups that diversified during the sametime period and therefore may have experienced thesame causal events and groups that diversifiedduring different time periods and require differentcausal explanations [48]

Just as there may be multiple ancestral area chro-nograms for a taxonomic group there may also bemultiple palaeogeographical hypotheses regarding thehistory of connections of formerly connected land-masses For example the lsquopan-Gondwananrsquo andlsquoAfrica-firstrsquo hypotheses represent alternate scenariosfor the breakup of Gondwana [49] Both hypothesesagree that the initial rift was between the African com-ponent of West Gondwana (Africa South America)and the Indo-Madagascar component of EastGondwana although connections between Africa andIndo-Madagascar were maintained via South Amer-icandashAntarctica Subsequent to this initial rift the pan-Gondwanan hypothesis [50] postulates that three vicar-iant separations South America from Africa SouthAmerica from Antarctica and Antarctica from Indo-Madagascar all occurred during a narrow timewindow (100ndash90 Ma) The Africa-first hypothesis inturn suggests that Africa was the first Gondwanan con-tinent to become completely separated from otherGondwanan landmasses when it separated fromSouth America by approximately 100 Ma Indo-Mada-gascar separated from AntarcticandashAustralia atapproximately 130ndash110 Ma but maintained subaerialconnections with Antarctica via the Kerguelen Plateauand possibly the Gunnerus Ridge to the west well intothe Late Cretaceous (approx 80 Ma) The final separ-ation was between the Antarctica Peninsula and the tipof South America in the Eocene

Krause et al [49] compared Cretaceous vertebratefaunas from different Gondwanan landmasses and con-cluded that palaeontological data are most compatiblewith a modified version of the Africa-first hypothesisKrause et alrsquos [49] work also illustrates how biogeographichypotheses based on fossils can be compared withgeology-based palaeogeographical hypotheses in anarena that allows for reciprocal illumination Thus ances-tral and general area chronograms provide a frameworkfor evaluating competing geology-based palaeogeo-graphical reconstructions just as geology-basedpalaeogeographical reconstructions provide a frameworkfor evaluating alternate ancestral area chronograms(figure 1) Krause et al [49] noted that there is no apriori reason to assume that geological data trumppalaeontological data or vice versa insofar as each typeof data can be used to reveal large-scale biogeographicpatterns

6 PLACENTAL PHYLOGENY AND ACOMPARISON OF DIFFERENT ANCESTRALAREA RECONSTRUCTION METHODSMost placental orders have first fossil occurrences andprobable origins in Laurasia but there are also orders

with Gondwanan origins based on first fossil occur-rences in South America (Xenarthra) or Africa (mostafrotherian orders) Traditional morphologicalphylogenies [5152] have suggested close relationshipsbetween Laurasian and Gondwanan orders egEdentata (Xenarthra (Gondwanan) thorn Pholidota(Laurasian)) By contrast molecular phylogenies haverecovered three superordinal groups AfrotheriaLaurasiatheria and Euarchontoglires [353ndash63] thatwere not recovered on morphological trees Thesethree groups plus Xenarthra comprise the four majorclades of placental mammals There is also robust mol-ecular support for Boreoeutheria (EuarchontogliresthornLaurasiatheria) [60ndash6264] This overhaul of placentalphylogeny in conjunction with the results of moleculardating analyses laid the foundation for newbiogeographic hypotheses We discuss these in sect7 afterfirst comparing the results of different ancestralarea reconstruction methods in the remainder ofthis section

Ancestral area chronograms were reconstructed for43 fully terrestrial placental taxa from Springer et al[3] Chiropterans and fully aquatic forms wereexcluded because of their different modes of dispersal(ie flight swimming) and also because most fullyaquatic taxa inhabit areas (ie oceans) that are notcontained in the four-area scheme used in our analyses(see below) Ancestral area chronograms were recon-structed using a ML phylogram obtained withRAXML [65] molecular divergence dates estimatedwith BEAST [6] and ancestral areas reconstructedwith a variety of methods

Four areas (Africa Eurasia North America andSouth America) were recognized and two methodswere used to code areas for terminal taxa Firstareas were coded based on the geographical ranges ofextant species Second areas were coded based onthe geographical provenance of the oldest fossil foreach lineage The step matrix that was used in MACparsimony analysis is shown in figure 2 Given thatthe number of character states that are chosen for geo-graphical range subdivision is arbitrary it may beinstructive to compare the results of analyses withcoarser (eg Gondwana versus Laurasia) and finer(eg Europe and Asia instead of Eurasia) scales forarea coding although the analyses reported here areconfined to the four areas listed above

We reconstructed ancestral areas using ninemethods (i) MAC parsimony (ii) Fitch parsimonywith multiple binary characters (FP-MBC) (iii) Fitchparsimony with a single multi-state character (FP-SMC) (iv) DIVA with no constraints on the maximumnumber of areas per node (v) DIVA with a maximumof two areas per node (DIVA-2) (vi) DEC with no con-straints on the maximum number of areas per node(vii) DEC with a maximum of two areas per node(DEC-2) (viii) stochastic mapping with multiplebinary characters (SM-MBC) and (ix) stochastic map-ping with a single multi-state character (SM-SMC)Ancestral area chronograms (MAC parsimony) basedon the geographical ranges of extant species and fossillineages are shown in figures 3 and 4 respectivelyTables 3 and 4 summarize the results of analyses withall nine methods

0204060

NeogeneQuaternary

CenozoicPaleogene

Upper Miocene PP

Paleoc

OligEoceneCretaceousMesozoic

Choloepus

Tamandua

MyrmecophagaEuphractus

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

MacroscelidesElephantulus

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

Africa + EurasiaAfrica + North America

Figure 3 Ancestral area chronogram for 43 placental taxa from Springer et al [3] with area coding based on extant ranges forterminal taxa RAxML was used to infer phylogenetic relationships BEAST was used to infer divergence times MAC parsi-mony was used to infer ancestral areas with the step matrix in figure 2 We employed soft constraints (nodes 3 8 10 16 19

21 32 34 36 38 41) that followed a normal distribution with 95 of the normal distribution between the specified mini-mum and maximum constraints (table 1) Areas for extant taxa are enumerated in table 2 and are colour-coded as followsAfrica blue Eurasia green North America brown South America red Multi-coloured names denote taxa that occur inmore than one area (table 2) Nodes with unambiguous ancestral area reconstructions are shown with a single colouredcircle nodes with ambiguous reconstructions are shown with two or more circles and each coloured circle corresponds to

a different reconstruction

Table 1 Fossil constraints Minimum ages are based on the age of the oldest unequivocal fossils belonging to the clade

Maximum ages are based on the maximum of stratigraphic bounding [66] phylogenetic bracketing [6768] and phylogeneticuncertainty Stratigraphic bounding encompassed two successive underlying fossil-bearing deposits that did not contain anyfossils from the lineage of interest phylogenetic bracketing encompassed the age of the oldest fossils that were up to twonodes below the divergence event and phylogenetic bracketing allowed for the possibility that taxa of uncertain phylogeneticaffinities belong to the crown clade first outgroup or second outgroup Dates used in stratigraphic bounding are from

Gradstein et al [69] We recognized the following chronological units in succession from youngest to oldest PleistocenePliocene Late Miocene Middle Miocene Early Miocene Late Oligocene Early Oligocene Late Eocene Middle EoceneEarly Eocene Late Palaeocene Middle Palaeocene Early Palaeocene Maastrichtian and Campanian

node numbera

fossil constraints (Ma)

oldest fossil for minimum reference(s)minimum maximum

3 556 712 Eritherium [70]8 585 712 Riostegotherium [6671]

10 338 655 Antarctic specimenb [7273]16 611 842 Adunator [74]19 371 658 Hesperocyon gregarious [75ndash77]21 555 611 Hyracotherium [78]

32 484 611 leporid tarsals [79]34 484 611 Eogliravus [80]36 338 56 Gaudeamus [8182]38 118 34 Prodolichotis [83]41 524 611 Mattimys [84]

aNode numbers refer to figures 3 and 4bThe Eocene Antarctic specimen is an ungual phalanx that Carlini et al [72] identified as a megatheroid sloth Marenssi et al [85] revisedthe identification of the phalanx to include either Tardigrada (sloths) or Vermilingua (anteaters) Subsequently Vizcaıno amp Scillato-Yane[73] described a fragmentary tooth from the Eocene of Antarctica and referred this tooth to Tardigrada but MacPhee amp Reguero [86]reinterpreted this tooth fragment as Mammalia incertae sedis based on histological evidence

Ambiguous ancestral area reconstructions were aproblem for all methods and the number of nodeswith equivocal reconstructions ranged from four(SM-SMC with extant coding) to 26 (DEC-2 withextant coding) For some methods the number ofambiguous nodes was higher with extant coding thanwith fossil coding (FP-MBC FP-SMC MAC parsi-mony DIVA DIVA-2 DEC DEC-2) but in othercases this pattern was reversed (SM-MBC SM-SMC) Ancestral areas for Placentalia Exafroplacenta-lia (frac14Boreoeutheria thorn Xenarthra) and several nodeswithin Rodentia were reconstructed as ambiguous bynearly all methods Other nodes were consistentlyreconstructed with unambiguous ancestral areasincluding clades with ancestral areas in Africa(Afrotheria and its internal nodes) Eurasia (Euarch-onta Paraprimates [frac14Dermoptera thorn Scandentia]Muridae) North America (Erinaceidae thorn Soricidae)and South America (Xenarthra and its internalnodes Cavioidea) Most analyses reconstructedEurasia as the ancestral area for BoreoeutheriaLaurasiatheria and Euarchontoglires This finding isdiscussed below

The importance of fossils is illustrated by recon-structions for Lagomorpha (tables 3 and 4) Allmethods returned North America as the ancestralarea when extant taxa were used for area coding butidentified Eurasia with fossil coding

DIVA and DEC analyses reconstructed more nodeswith multiple areas than did the other methodsAnalyses with DEC reconstructed 17ndash20 nodes withtwo or more areas and four to six nodes with threeor more areas DIVA analyses resulted in 15ndash18

nodes with at least two areas and five to six nodeswith three or more areas None of the other methodsreconstructed ancestral nodes to include three ormore areas in a single reconstruction although threeor four areas were sometimes represented by thefull complement of alternate reconstructions for agiven node

FP-MBC returned nine empty nodes with extantcoding and five empty areas with fossil coding SM-MBC with extant coding resulted in three or fourempty nodes with extant coding and four emptynodes with extinct coding (table 5)

7 PLACENTAL BIOGEOGRAPHYAfrotheria (Afrosoricida Hyracoidea MacroscelideaProboscidea Sirenia Tubulidentata) was the first ofthe new superordinal groups to receive robust molecu-lar support [535556] With the exception of Sireniaall afrotherian orders have first fossil occurrences inAfrica and two orders (Macroscelidea Afrosoricida)have evolutionary histories that are restricted to theAfro-Malagasy region Springer et al [53] suggestedthat interordinal separation of afrotherian orders com-menced during a window of isolation that began in theCretaceous after Africa separated from South Amer-ica and lasted until the early Cenozoic when Africadocked with Europe Consistent with this scenarioAfrica was unambiguously reconstructed as the ances-tral area for Afrotheria (figures 3 and 4) Thishypothesis contrasts with traditional views whereinthe African mammal fauna arrived from the northincluding a condylarth stock that arrived in Africa

Table 2 Geographical area of extant taxa and oldest fossils used in ancestral area reconstruction

taxona area of extant species area of oldest fossilb

Choloepus didactylus SA SA Megalonychidae Miocene [87]Tamandua tetradactyla SA SA Tamandua Pleistocene [87]

Myrmecophaga tridactyla SA SA Neotamandua Miocene [8788]Euphractus sexcinctus SA SA Zaedyus Pliocene [8789]Chaetophractus villosus SA SA Chaetophractus Pliocene [90]Erinaceus europaeus Eurasia NA Adunator Palaeocene [74]Talpa altaica Eurasia Eurasia Eotalpa Eocene [91]

Sorex araneus Eurasia NA Domnina Eocene [92]Echinops telfairi Africa Africa Widanelfarasia Eocene [93]Amblysomus hottentotus Africa Africa Eochrysochloris Oligocene [93]Procavia capensis Africa Africa Seggeurius Eocene [94]

Loxodonta africana Africa Africa Eritherium Palaeocene [70]Macroscelides proboscideus Africa Africa Macroscelides Pliocene [95]Elephantulus rufescens Africa Africa Elephantulus Pliocene [95]Orycteropus afer Africa Africa Orycteropus Miocene [96]Tamias striatus NA NA Spurimus Eocene [97]

Muscardinus avellanarius Eurasia Eurasia Eogliravus Eocene [80]Mus musculus Eurasia Eurasia Progonomys Miocene [74]Rattus norvegicus Eurasia Eurasia Karnimata Miocene [74]Pedetes capensis Africa Africa Pondaungimys Eocene [98]Hystrix brachyurus Eurasia Africa Gaudeamus Eocene [81]

Castor canadensis NA NA Mattimys Eocene [84]Dipodomys merriami NA NA Proheteromys Oligocene [99]Cavia porcellus SA SA Prodolichotis Miocene [83100]Hydrochaeris hydrochaeris SA SA Cardiatherium Miocene [101]Erethizon dorsatum NA SA Eopululo Eocene [102]

Sylvilagus floridanus NA SA Eurasia tarsal elements Eocene [79]Ochotona princeps NA Eurasia Sinolagomys Oligocene [103104]Cynocephalus variegatus Eurasia Eurasia Dermotherium Eocene [105]Tupaia minor Eurasia Eurasia Eodendrogale Eocene [106]Lemur catta Africa Africa Pachylemur Quaternary [107]

Homo sapiens Eurasia NA SA Africa Eurasia Anthrasimias Palaeocene [108]Tarsius syrichta Eurasia Eurasia Tarsius Eocene [109]Hippopotamus amphibius Africa Africa Morotochoerus Miocene [110]Lama glama SA NA Poebrodon Eocene [111]

Tragelaphus eurycerus Africa Eurasia Archaeomeryx Eocene [112]Sus scrofa Eurasia Africa Eurasia Eocenchoerus Eocene [113]Equus caballus Eurasia Eurasia NA Hyracotherium Eocene [78114115]Ceratotherium simum Africa NA Hyracodontidae Eocene [116]Tapirus indicus Eurasia NA Helaletes Eocene [117]

Felis catus Africa Eurasia Stenoplesictis Eocene [118119]Canis familiaris Eurasia NA Hesperocyon Eocene [120]Manis pentadactyla Eurasia Eurasia Eomanis Eocene [121]

aIn cases of chimeric taxa we used the most common species from Springer et alrsquos [3] concatenated supermatrix NA North AmericaSA South AmericabArea of the oldest stem fossil belonging to the terminal branch represented by each living taxon

from Europe in the early Cenozoic and insectivoresthat arrived in the Neogene [124]

Asher et al [125] Zack et al [126] and Tabuce et al[127] suggested that the geographical distributions ofliving afrotherians are not representative of the histori-cal geographical distribution of this clade and thatAfrotheria is Holarctic in origin based on the place-ment of extinct taxa from the Palaeocene of Laurasiawithin or at the base of Afrotheria However pseu-doextinction tests call into question the reliability ofthe placement of fossil taxa in morphological cladisticanalyses [3]

The oldest xenarthran fossils are scutes from thePalaeocene of South America [71] Living membersof Xenarthra (anteaters sloths armadillos) are

restricted to South and Central America with theexception of the nine-banded armadillo whose ances-tors dispersed to North America during the GreatAmerican Interchange [128] Simpson [129130] sup-ported the view that South American xenarthransevolved in situ during South Americarsquos isolation fromother continents in the early Tertiary All of our ana-lyses are consistent with the hypothesis that SouthAmerica was the ancestral area for Xenarthra (figures3 and 4)

The remaining placental orders are placed in Laur-asiatheria (Eulipotyphla Chiroptera PerissodactylaCetartiodactyla Carnivora Pholidota) and Euarchon-toglires (Primates Dermoptera Scandentia RodentiaLagomorpha) With the exception of bats these orders

0204060

NeogeneQuaternary

Cenozoic

Paleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

Africa + EurasiaAfrica + North AmericaEurasia + North America

Figure 4 Ancestral area chronogram for 43 placental taxa from Springer et al [3] with area coding based on the oldest fossil foreach lineage RAxML was used to infer phylogenetic relationships BEAST was used to infer divergence times and MAC par-simony was used to infer ancestral areas with the step matrix in figure 2 Areas for the oldest fossil lineage are enumerated intable 2 and are colour-coded as follows Africa blue Eurasia green North America brown South America red Nodes with

unambiguous ancestral area reconstructions are shown with a single coloured circle nodes with ambiguous reconstructions areshown with two or more circles and each coloured circle corresponds to a different reconstruction

have first fossil occurrences that are exclusively Laura-sian Our reconstructions provide support for Eurasiabut not North America as the ancestral area for theseclades (figures 3 and 4) These results are consistent

with previous suggestions that Cretaceous zhelestidsand zamlambdalestids from Asia are members ofcrown Placentalia [131132] Further the fossilrecord suggests that Eutheria were dominant in

chpars

salndash

lndashvic

salndash

lndashex

o(lsquo

splitrsquo

inhabit

iphilia

ulipoty

orexthorn

Susthorn

Bosthorn

toglire

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Table 5 Comparison of different methods for reconstructing ancestral areas NA1 not applicable for monomorphic

reconstruction methods NA2 not applicable when the maximum number of areas is set at two NA2 not applicable formethods that employ single multistate charactersa

FP-MBC FP-SMCMACParsimony DIVA DIVA-2 DEC DEC-2 SM-MBC SM-SMC

nodes with ambiguous 75 129 128 1211 107 2323 2623 1617 614reconstructionsb 1920 1718 1012 410

nodes with 2 areasc 33 NA1 46 1618 1516 1820 2019 77 NA11720 1717 46

nodes with 3 areasd 00 NA1 00 65 NA2 66 NA2 00 NA145 00

empty nodese 95 NA3 NA3 NA3 NA3 NA3 NA3 34 NA344

aNumbers before slashes are based on analyses with area coding for extant taxa and numbers after slashes are based on analyses with areacoding for the oldest fossil See table 3 for abbreviationsbFor FP-MBC nodes were considered ambiguous if at least one area was reconstructed as (01) For SM-MBC and SM-SMC nodes wereconsidered ambiguous if the posterior probability (PP) of at least one area was 01 PP 09 (top line) or 02 PP 08 (bottom line)For DEC and DEC-2 nodes were considered ambiguous if the frequency ( f ) of at least one area was 01 f 09 (top line) or 02 p 08 (bottom line)cAt least two areas in at least one of the alternate resolutions for an ancestral node For FP-MBC each occurrence of 1 or (01) was takento include an ancestral area For SM-MBC areas were counted as present at a node if posterior probabilities were 010 (top line) or020 (bottom line) For DEC and DEC-2 areas were counted as present at a node if frequencies were 01 (top line) or 02 (bottomline)dAt least three areas in more than one of the alternate resolutions for an ancestral node For FP-MBC each occurrence of 1 or (01) wastaken to include an ancestral area For SM-MBC areas were counted as present at a node if posterior probabilities were 010 (top line)or 020 (bottom line) For DEC and DEC-2 areas were counted as present at a node if frequencies were 01 (top line) or 02(bottom line)eFor FP-MBC nodes were considered empty if all areas were reconstructed as 0 For SM-MBC nodes were considered empty if posteriorprobabilities were 010 (top line) or 020 (bottom line) for all four areas

Eurasia throughout the Cretaceous but were absentfrom North America through much of the Late Cre-taceous and only attained appreciable diversity thereduring the last approximately 10 Myr of the period[133134] Boyer et al [135] concluded that theIndian subcontinent Eurasia and Africa are morelikely places of origin for Euarchonta than is NorthAmerica This agrees with our ancestral areareconstructions (figures 3 4 and tables 3 4)

Although there is robust support for the monophylyof Xenarthra Afrotheria and Boreoeutheria relation-ships among these three groups and the root of theplacental tree remain contentious [105460ndash63136]Murphy et al [62] and Springer et al [10] suggested acausal relationship between the sundering of Africa andSouth America and basal cladogenesis among crown-group placental mammals given the coincidence ofmolecular dates for the base of placentals and thevicariant separation of Africa and South Americaapproximately 100ndash120 Ma

Asher et al [125] analysed a combined matrix andrecovered Afrotheria in a nested position within Pla-centalia which contradicts the hypothesis that theplate tectonic separation of Africa and South Americaplayed a causal role in the early cladogenesis of placen-tal mammals However the nested position forAfrotheria resulted from the paraphyly of Euarchonto-glires Glires and Rodentia Rare genomic changesconfirm the monophyly of Xenarthra [137] Afrotheria[138ndash142] Euarchontoglires [139141142] Laura-siatheria [139141142] and Boreoeutheria[139141142] and preclude a nested position forAfrotheria in the placental tree

Rare genomic changes have also been used toexamine the position of the placental root Kriegs

et al [139] reported LINE insertions that are sharedby Epitheria whereas Murphy et al [16] discoveredrare genomic changes that support AtlantogenataNishihara et al [142] performed genome-wide retro-poson analyses and found 22 25 and 21 LINEinsertions for Exafroplacentalia Epitheria and Atlan-togenata respectively Based on these resultsNishihara et al [142] concluded that XenarthraAfrotheria and Boreoeutheria diverged from oneanother nearly simultaneously They also suggested anew palaeogeographical model for the breakup of Pan-gaea and Gondwana in which Africa becomes isolatedfrom both South America and Laurasia at approxi-mately 120 Ma and argued that these coeval platetectonic events provide an explanation for the simul-taneous divergence of Afrotheria Xenarthra andBoreoeutheria However relaxed clock dates for thebase of Placentalia are closer to 100 Ma than to120 Ma (figures 3 and 4) A second difficulty concernsthe opening of the South Atlantic Nishihara et al[142] suggested that the Brazilian Bridge which rep-resented the last connection between Africa andSouth America was severed at approximately120 Mya but other recent reconstructions suggestthat the connection between the South Atlantic andCentral Atlantic was not established until lateAptianmid-Albian times (approx 110ndash100 Ma)[143144]

8 THE IMPORTANCE OF DISPERSALIn the context of pre-plate tectonic views of theEarth Simpson [2] proposed three types ofmigration routes to describe the movement of ani-mals corridors filter bridges and sweepstakes

Africa

Tanzania

Mozambique

100 mi

Figure 5 Present day surface ocean currents in the Mozam-bique Channel (solid arrows) are southndashsouthwest andwould not have facilitated west to east transoceanic dispersalfrom Africa to Madagascar [153] By contrast westerly sur-face ocean currents in the Eocene (dashed arrows) would

have facilitated dispersal across the Mozambique Channelfrom Africa to Madagascar especially during tropicalstorms [154] The outline of Madagascar with dashed linesshows its approximate position relative to Africa during theEocene

dispersal Corridors connect two areas and are per-meable to all animals filter bridges impose selectivebarriers that affect some but not all animals andsweepstakes dispersal is required when there arestrong barriers to migration such as high mountainbarriers or oceans

Simpson [2] suggested that Madagascarrsquos livingmammals were the product of sweepstakes dispersalfrom Africa to Madagascar Sweepstakes dispersalhypotheses fell out of favour with the validation ofplate tectonic theory and were summarily dismissedas lsquomiraculousrsquo hypotheses with no scientific basis[145] However it has become apparent that some dis-tributional patterns can only be explained bysweepstakes dispersal [146] Observational data alsoprovide support for long-distance vertebrate dispersal[147] Examples of low probability sweepstakes disper-sal involving mammals include the origins of theendemic mammal fauna in Madagascar and theoccurrence of caviomorph rodents and platyrrhine pri-mates in South America

Madagascarrsquos strictly terrestrial extant mammal faunaincludes endemic lineages from four placental orderstenrecs (Afrosoricida) euplerids (Carnivora) neso-myines (Rodentia) and lemurs (Primates) In eachlineage Madagascar endemics comprise monophyleticassemblages with closest living relatives in Africa[148149] Madagascar separated from Africa approxi-mately 165 Ma but maintained its connection withAntarctica via the Kerguelen Plateau until as late as80 Ma at which time it became permanentlyseparated from other Gondwanan landmasses This his-tory suggests that Madagascarrsquos terrestrial endemicmammals are either the ancient descendants of vicariantevents that occurred prior to 80 Ma or reached Mada-gascar via transoceanic sweepstakes dispersal at a latertime Another possibility is that a land bridge connectedAfrica and Madagascar between 45 and 26 Ma [150]

Molecular divergence dates suggest that all fourendemic lineages last shared a common ancestorwith their African sister group in the Cenozoic[148149151152] Poux et al [148] concluded thatdispersal by lemurs rodents and carnivorans musthave occurred by transoceanic dispersal rather thanland bridge dispersal based on molecular dates forthe colonization of Madagascar that were outside ofthe land bridge window ie 60ndash50 Ma for lemurs26ndash19 Ma for carnivorans and 24ndash20 Ma for rodentsHowever present ocean currents allow for dispersalfrom Madagascar to Africa but oppose reciprocal dis-persal from Africa to Madagascar across theMozambique Channel If ocean currents were thesame for most of the Cenozoic as they are todaythey would not have facilitated west to east transocea-nic dispersal across the Mozambique Channel becauseof the strong southndashsouthwest flow of the Mozambi-que Current [153]

Ali amp Huber [154] addressed this problem by simu-lating surface ocean currents in the Indian Oceanduring the Eocene They concluded that large-scaleocean current systems in the Eocene were profoundlydifferent from modern observed circulatory patternsand that the flow along the African coast was eastwardtowards Madagascar instead of southward through the

Mozambique Channel (figure 5) Ali amp Huber [154]further suggested that dispersal probabilities wereenhanced by tropical storms that (i) generated largefloating tree islands that would have allowed for asuccessful oceanic voyage and (ii) acceleratedtransportation rates from Africa to Madagascar thatwould have allowed for complete crossing of theMozambique Channel in 25ndash30 days

The dispersal of four groups of fully terrestrialmammals from Africa to Madagascar at a time whenthere was no land bridge is a testament to the impor-tance of rare sweepstakes events in the evolutionaryhistory of Placentalia Even more remarkable is theoccurrence of two different groups of placental mam-mals hystricognath rodents and anthropoid primatesin Africa and South America

Hystricognathi includes Hystricidae (Old World por-cupines) and Phiomorpha (eg cane rats dassie rats)from the Old World and Caviomorpha (eg porcupineschinchillas) from the New World The oldest hystricog-naths are from the late Eocene Egypt and have beendated at approximately 37 Ma [81] Old World hystri-cognaths are paraphyletic usually with phiomorphshaving closer phylogenetic affinities to South Americancaviomorphs than to hystricids [14155156] Relaxedclock dates suggest that South American caviomorphslast shared a common ancestor with phiomorphsbetween 45 and 36 Ma [81155157] The most recentcommon ancestor of Caviomorpha has been dated at45ndash31 Ma [81155157158]

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

Figure 6 Alternate hypotheses for the dispersal of platyrrhine and caviomorph ancestors respectively from AfricaAsia toSouth America Hypothesis 1 transoceanic dispersal (1a) from Africa to South America possibly with an earlier dispersalfrom Asia to Africa (1b) if origination occurred in Asia Hypothesis 2 dispersal from Asia through North America to

South America Hypothesis 3 dispersal from Asia to South America via Australia and Antarctica after two transoceanic cross-ings Middle Eocene world map based on Palaeomap Project (httpwwwscotsecomnewpage9htm)

Among anthropoids Old World catarrhines (egmacaques apes) and South American platyrrhines(eg marmosets capuchins spider monkeys) are reci-procally monophyletic sister taxa The oldestanthropoid fossils are from the Old World althoughwhether the most recent common ancestor of Anthro-poidea is African or Asian is uncertain [108159160]Poux et al [155] dated the split between catarrhinesand platyrrhines at approximately 37 Ma and thebase of Platyrrhini at approximately 17 Ma

The vicariant separation of Africa and South Amer-ica (110ndash100 Ma) is too old to explain the separationof either Phiomorpha and Caviomorpha or Catarrhiniand Platyrrhini Similarly Arnason et alrsquos [161]hypothesis of land bridge dispersal during the LateCretaceousndashEarly Palaeocene is too old for relaxedclock dates which instead rule out the colonizationof South America by Caviomorpha and Platyrrhiniprior to the Eocene Other hypotheses for the coloni-zation of South America by caviomorphs andorplatyrrhines include (i) trans-Atlantic dispersal fromAfrica to South America [162] (ii) dispersal fromAsia through North America to South America[163164] and (iii) dispersal from Asia to SouthAmerica via Australia and Antarctica after two oceancrossings (figure 6) [165]

Most workers favour transoceanic dispersal fromAfrica to South America for both Caviomorpha and Pla-tyrrhini Dispersal through Asia and North America is anintriguing possibility but palaeontological data provideno support for migrations through North America Simi-larly dispersal from Asia to South America throughAustralia and Antarctica lacks palaeontological supportrequires multiple transoceanic dispersals and becomeseven less likely after the Eocene because of the severedconnection between Antarctica and South Americaand climatic deterioration in Antarctica associated withthe opening of the Drake Passage In view of

phylogenetic geological palaeontological and moleculardata trans-Atlantic dispersal is the most likely scenariofor colonization of South America by caviomorphs andplatyrrhines

9 BAT BIOGEOGRAPHYIn contrast to other mammals bats are capable ofpowered flight which has profoundly enhanced theirdispersal capabilities The occurrence of seven differ-ent families of extant bats in Madagascar includingthe endemic sucker-footed bats (Family Myzopodi-dae) and of another family in New Zealand theshort-tailed bats (Family Mystacinidae) providesabundant evidence of the dispersal capabilities ofbats [166]

The oldest bat fossils are from the Early Eocene ofNorth America [167168] Early Eocene bats are alsoknown from Europe Africa and Australia [167] Theprevalent view is that bats originated in Laurasia buta minority view holds that bats originated in Gond-wana [169170] Teeling et al [13] reconstructedancestral areas for bats with (i) multistate-coded datafor the current global distribution of each lineagewith nine different character states (Europe AfricaAsia Madagascar Australia New Zealand NorthAmerica Central thorn South America and West Indies)and (ii) binary-coded data for the earliest fossil occur-rence for each lineage (Laurasia versus Gondwana)Teeling et alrsquos [13] results suggested North Americaor Laurasia as the ancestral area for bats and AsiaEurope or Laurasia as the ancestral area for both Yinp-terochiroptera and Yangochiroptera Eick et al [12]used DIVA [33] to estimate ancestral areas for Chirop-tera and its subclades and coded areas based oncurrent distributions for each family Seven areas(Africa Asia Australia Europe North AmericaSouth America and New Zealand) were recognized

Table 6 A comparison of ancestral area reconstructions for bats based on DIVA analyses Eick et al [12] coded the presence

or absence of extant bat families in seven different areas and performed DIVA analyses with no constraints on the maximumnumber of areas We re-analysed Eick et alrsquos [12] dataset with DIVA using the same settings reported by these authorsAfrica A Asia B Australia C Europe D North America E South America F New Zealand G

node number (figure 7) Eick et al [12] re-analysis

1 A ABCEF ABDEF ABCDEF ABEFG ABCEFG ABDEFG ABCDEFG2 A ACEF BCEF ABCEF DEF ADEF BDEF ABDEF ACDEF BCDEF

ABCDEF AEFG ABEFG ACEFG BCEFG ABCEFG DEFGADEFG BDEFG ABDEFG ACDEFG BCDEFG ABCDEFG

3 AE AF E AE BE CE ACE BCE ABCE F AF BF CF ACF BCF ABCFCEF ACEF BCEF ABCEF AG BG CG ACG BCG ABCG AEGBEG CEG ACEG BCEG ABCEG AFG BFG CFG ACFG BCFGABCFG AEFG BEFG CEFG ACEFG BCEFG ABCEFG

4 E F E F EG FG EFG5 E F E F6 E F E F7 E F E F8 EG FG EFG EG FG EFG

9 A A B AC BC ABC AE BE ABE ACE BCE ABCE AF BF ABF ACFBCF ABCF AEF BEF ABEF ACEF BCEF

10 A A AC AD ACD ABCD ACE ADE ACDE ABCDE ACF ADFACDF ABCDF ACEF ADEF ACDEF ABCDEF ACDEG ABCDEGACDFG ABCDFG ACDEFG ABCDEFG

11 AE AF AEF AE CE DE CDE ACDE BCDE ABCDE AF CF DF CDF ACDFBCDF ABCDF AEF CEF DEF CDEF ACDEF BCDEF ABCDEFCDEG ACDEG BCDEG ABCDEG CDFG ACDFG BCDFGABCDFG CDEFG ACDEFG BCDEFG ABCDEFG

12 A A C D CD ACD BCD ABCD CDE ACDE BCDE ABCDE CDF

ACDF BCDF ABCDF CDEF ACDEF BCDEF ABCDEF CDGACDG BCDG ABCDG CDEG ACDEG BCDEG ABCDEGCDFG ACDFG BCDFG ABCDFG CDEFG ACDEFG BCDEFGABCDEFG

13 A A B C D AG BG ABG CG ACG BCG ABCG DG ADG BDG

ABDG CDG ACDG BCDG ABCDG AEG BEG ABEG CEGACEG BCEG ABCEG DEG ADEG BDEG ABDEG CDEGACDEG BCDEG ABCDEG AFG BFG ABFG CFG ACFGBCFG ABCFG DFG ADFG BDFG ABDFG CDFG ACDFG

BCDFG ABCDFG AEFG BEFG ABEFG CEFG ACEFG BCEFGABCEFG DEFG ADEFG BDEFG ABDEFG CDEFG ACDEFGBCDEFG ABCDEFG

14 A A B C AC BC ABC15 A A B C AC BC

16 A A B C17 A A B AC BC ABC

and Africa was reconstructed as the ancestral area forthe most recent common ancestors of ChiropteraYinpterochiroptera and Yangochiroptera Lim [47]used parsimony to reconstruct ancestral areas andalso recovered Africa as the ancestral area forYangochiroptera and its deepest nodes

We recovered more inclusive ancestral areas forChiroptera Yinpterochiroptera and Yangochiropterawhen we performed analyses with DIVA using thesame data and settings that were reported by Eicket al [12] (figure 7 and table 6) The reconstructionfor the base of Chiroptera was equivocal and includedseven different possibilities all of which were equallyparsimonious based on DIVArsquos criteria for minimiz-ing dispersal and extinction (figure 7 and table 6)Each of these reconstructions included at least fiveareas and four areas (Africa Asia North Americaand South America) were common to all sevenreconstructions

Among the most comprehensive studies in mamma-lian historical biogeography are Limrsquos [4647] analysesof South American bats Ancestral reconstructions pro-vided evidence for multiple dispersals from Africa toSouth America One dispersal occurred in Noctilionoi-dea (Eocene approx 42 Ma) and another occurred inEmballonuroidea (Oligocene approx 30 Ma) Vesperti-lionoidea have a more complex history that involvesnumerous independent dispersals from Africa(Eocene earliest event approx 50 Ma) as well asfrom North America Lim [46] used PACT to examineevolutionary processes that have been important in thediversification of South American emballonurids Hisgeneral area cladogram revealed a complex historywith multiple vicariant within-area and dispersalevents all playing a role Within-area speciation duringthe Miocene particularly in the northern Amazonarea was the most important diversification process inthis group Lim [47] correlated Miocene speciation

FuripteridaeNoctilionidaeMormoopidaePhyllostomidaeThyropteridaeMystacinidaeEmballonuridaeNycteridaeMyzopodidaeMolossidaeMiniopteridaeVespertilionidaeNatalidaeHipposideridaeRhinolophidaeRhinopomatidaeMegadermatidaePteropodidae

EFEFEFEFEFGABCEFABAABCDEFABCDABCDEFGEFABCABCDABABCABC

Figure 7 Eick et alrsquos [12] phylogeny and area coding forextant bat families Ancestral area reconstructions based onDIVA analyses are shown in table 6 for nodes 1ndash17Africa A Asia B Australia C Europe D North AmericaE South America F New Zealand G

with contemporaneous climatic and habitat changes thatoccurred in the Amazon Basin Construction of anancestral area cladogram for all bat species will providean unprecedented opportunity to examine the impor-tance of transoceanic dispersal in promotingtaxonomic diversity in this highly successful group ofmammals

10 MARSUPIAL BIOGEOGRAPHYThe oldest metatherian is Sinodelphys from China[171] Cretaceous marsupial fossils are also knownfrom Europe [172173] and North America [174ndash178] The consensus is that metatherians originatedin Asia and subsequently dispersed to North Americaand Europe [173]

In contrast to the Cretaceous record of Metatheriaalmost all living metatherians have geographical distri-butions that are entirely Gondwanan Case et al[179] suggested that the ancestor of living marsupialsdispersed to South America in the Late Cretaceous orearly Palaeocene The South American marsupialcohort Ameridelphia which includes Paucituberculata(shrew opossums) and Didelphimorphia (opossums)is paraphyletic at the base of Australidelphia whichincludes the South American order Microbiotheria(monito del monte) and the Australasian orders Dipro-todontia (eg wombats kangaroos) Dasyuromorphia(eg quolls numbats) Peramelemorphia (eg bandi-coots bilbies) and Notoryctemorphia (marsupialmoles) [1721180ndash182]

Subsequent to Kirsch et alrsquos [183] single-copyDNA hybridization study of marsupials whichplaced South American microbiotheres within Austra-lidelphia marsupial biogeographers have focused onthe monophyly or paraphyly of Australasian taxa Aus-tralasian monophyly is consistent with a singledispersal from South America to Australia via Antarcticabut Australasian paraphyly requires either multiple dis-persals to Australia or dispersal to Australia followedby back dispersal to South America [183ndash185] Molecu-lar phylogenies based on concatenated nuclear gene

sequences [21182] and retroposon insertions [186] sup-port the monophyly of Australasian marsupials andsuggest that Australasian marsupials last shared acommon ancestor with microbiotheres between 65 and58 Ma This phylogeny is compatible with a single dis-persal event from South America to Australia viaAntarctica [21] This dispersal would have been overlandif it occurred prior to the complete submergence of theSouth Tasman Rise approximately 64 Ma [187]

In contrast Beck et al [181] analysed a datasetcomprising living and fossil taxa including the earlyEocene genus Djarthia from Australia and recovereda sister-group relationship between Djarthia andliving australidelphians Beck et alrsquos [181] topologysuggest that South American microbiotheres back-dis-persed from eastern Gondwana to South America eventhough living Australasian marsupials comprise amonophyletic taxon However the decay index thatassociates crown Australidelphia to the exclusion ofDjarthia is only one step This result highlights thepotential importance of fossils for inferring biogeo-graphic history and the precarious nature ofconclusions based on a fragmentary fossil record

11 MONOTREME BIOGEOGRAPHYLiving monotremes include the semi-aquatic platypus(Ornithorhynchus) which occurs in Australia and Tas-mania and echidnas which occur in Australia(Tachyglossus) and New Guinea (Zaglossus) Theoldest monotreme is Teinolophos (121ndash1125 Ma) ofAustralia Rowe et al [188] suggested that Teinolophosis a crown monotreme based on cladistic analyses

In contrast to this ancient fossil record relaxedclock estimates for the platypus-echidna divergencerange from 889 to 277 Ma [188ndash191] and are tooyoung to accommodate Teinolophos in crown-groupMonotremata Rather these dates suggest that Teinolo-phos lies on the monotreme stem branch Youngermonotreme fossils whether stem or crown are exclu-sively from the Southern Hemisphere Luo et al[192193] and Kielan-Jaworowska et al [194]suggested that Monotremata belongs to the moreinclusive Gondwanan clade Australosphenidaalthough other studies place these Mesozoic taxacloser to Theria than to Monotremata [188189]

With or without these Mesozoic taxa it appears thatthe entire evolutionary history of Monotremata isrestricted to Gondwana Details of this history are diffi-cult to reconstruct owing to Monotrematarsquos depauperatetaxonomic diversity and meagre fossil record Futurefossil discoveries and more robust phylogenetic analysesare essential for revealing the full biogeographic range ofancient monotremes in Gondwana The occurrence ofmonotremes in Australia and South America suggeststhat their ancestral distribution may have includedother fragments of Gondwana such as Africa AntarcticaIndia and Madagascar

12 CONCLUSIONSContemporary methods for deciphering palaeobiogeo-graphy are underpinned by phylogenies divergencetimes and ancestral area reconstructions which

together yield ancestral area chronograms that providea powerful framework for proposing and testinghypotheses of dispersal and vicariance when evaluatedin the context of palaeographic hypotheses The toolkitfor unravelling historical patterns of vicariance anddispersal that have moulded the evolutionary historyof Mammalia now includes molecular data fossilsreconstructions of palaeogeography and palaeo-oceancurrents and a burgeoning array of methods in phylo-geny reconstruction molecular dating and ancestralarea reconstruction Larger and taxonomically morecomplete molecular datasets new fossil discoveriesand the application of new techniques will lead to sig-nificant advances in our understanding of thehistorical biogeography of Mammalia

MSS and WJM acknowledge support from NSF Threeanonymous reviewers provided helpful comments on anearlier draft of this manuscript We thank Kate Jones forinviting us to contribute to this volume

REFERENCES1 Jones K E amp Safi K 2011 Ecology and evolution of

mammalian biodiversity Phil Trans R Soc B 366

2451ndash2461 (doi101098rstb20110090)2 Simpson G G 1940 Mammals and land bridges

J Washington DC Acad Sci 30 137ndash1633 Springer M S Burk-Herrick A Meredith R

Eizirik E Teeling E OrsquoBrien S J amp Murphy W J2007 The adequacy of morphology for reconstructingthe early history of placental mammals Syst Biol 56673ndash684 (doi10108010635150701491149)

4 Springer M S Meredith R W Eizirik E Teeling

E amp Murphy W J 2008 Morphology and placentalmammal phylogeny Syst Biol 57 499ndash503 (doi10108010635150802164504)

5 Bollback J P 2006 SIMMAP stochastic charactermapping of discrete traits on phylogenies BMC Bioin-form 7 88 (doi1011861471-2105-7-88)

6 Drummond A J Ho S Y W Phillips M J amp Ram-baut A 2006 Relaxed phylogenetics and dating withconfidence PLoS Biol 4 e88 (doi101371journalpbio0040088)

7 Zuckerkandl E amp Pauling L 1962 Molecular diseaseevolution and genetic heterogeneity In Horizons in bio-chemistry (eds M Kasha amp B Pullman) pp 189ndash225New York NY Academic Press

8 Douady C J amp Douzery E J P 2003 Molecular esti-mation of eulipotyphlan divergence times and theevolution of lsquoInsectivorarsquo Mol Phylogenet Evol 28285ndash296 (doi101016S1055-7903(03)00119-2)

9 Douady C J Catzeflis F Raman J Springer M S

amp Stanhope M J 2003 Molecular evidence for theSahara as a vicariant agent and the role of Miocene cli-matic events in the diversification of the mammalianorder Macroscelidea (elephant shrews) Proc NatlAcad Sci USA 100 8325ndash8330 (doi101073pnas

0832467100)10 Springer M S Murphy W J Eizirik E amp OrsquoBrien

S J 2003 Placental mammal diversification and theCretaceousndashTertiary boundary Proc Natl Acad SciUSA 100 1056ndash1061 (doi101073pnas0334222

100)11 Delsuc F Vizcaıno S F amp Douzery E J P 2004

Influence of Tertiary paleoenvironmental changes onthe diversification of South American mammals a

relaxed molecular clock study within xenarthransBMC Evol Biol 4 11 (doi1011861471-2148-4-11)

12 Eick G N Jacobs D S amp Matthee C A 2005 Anuclear DNA phylogenetic perspective on the evolutionof echolocation and historical biogeography of extant

bats (Chiroptera) Mol Biol Evol 22 1869ndash1886(doi101093molbevmsi180)

13 Teeling E C Springer M S Madsen O Bates POrsquoBrien S J amp Murphy W J 2005 A molecular phy-logeny for bats illuminates biogeography and the fossil

record Science 307 580ndash584 (doi101126science1105113)

14 Huchon D Chevret P Jordan U Kilpatrick C WRanwez V Jenkins P D Brosius J amp Schmitz J

2007 Multiple molecular evidences for a living mamma-lian fossil Proc Natl Acad Sci USA 104 7495ndash7499(doi101073pnas0701289104)

15 Janecka J E Miller W Pringle T H Wiens FZitzmann A Helgen K M Springer M S amp

Murphy W J 2007 Molecular and genomic data ident-ify the closest living relative of Primates Science 318792ndash974 (doi101126science1147555)

16 Murphy W J Pringle T H Crider T A SpringerM S amp Miller W 2007 Using genomic data to unravel

the root of the placental mammal tree Genome Res 17413ndash421 (doi101101gr5918807)

17 Beck R M 2008 A dated phylogeny of marsupialsusing a molecular supermatrix and multiple fossil con-straints J Mammal 89 175ndash189 (doi10164406-

MAMM-A-4371)18 Meredith R W Westerman M amp Springer M S

2008 A timescale and phylogeny for lsquobandicootsrsquo (Pera-melemorphia Marsupialia) based on sequences for five

nuclear genes Mol Phylogenet Evol 47 1ndash20 (doi101016jympev200801002)

19 Meredith R W Westerman M amp Springer M S2008 Phylogeny and timescale for the living genera ofkangaroos and kin (Macropodiformes Marsupialia)

based on nuclear sequences Aust J Zool 56 395ndash410 (doi101071ZO08044)

20 Meredith R W Westerman M amp Springer M S 2009A phylogeny of Diprotodontia (Marsupialia) based onsequences for five nuclear genes Mol Phylogenet Evol51 554ndash571 (doi101016jympev200902009)

21 Meredith R W Krajewski C Westerman W ampSpringer M S 2009 Relationships and divergencetimes among the orders and families of marsupialsMus N Ariz Bull 65 383ndash406

22 Chatterjee H J Ho S W Y Barnes I amp Groves C2009 Estimating the phylogeny and divergence times ofprimates using a supermatrix approach BMC EvolBiol 9 259 (doi1011861471-2148-9-259)

23 Eizirik E Murphy W J Koepfli K P JohnsonW E Dragoo J W Wayne R K amp OrsquoBrien S J2010 Pattern and timing of diversification of the mam-malian order Carnivora inferred from multiple nucleargene sequences Mol Phylogenet Evol 56 49ndash63

(doi101016jympev201001033)24 Sanderson M J 1997 A nonparametric approach to

estimating divergence times in the absence of rate con-stancy Mol Biol Evol 14 1218ndash1231

25 Sanderson M J 2002 Estimating absolute rates of mol-

ecular evolution and divergence times a penalizedlikelihood approach Mol Biol Evol 19 101ndash109

26 Thorne J L amp Kishino H 2002 Divergence time andevolutionary rate estimation with multilocus data SystBiol 51 689ndash702 (doi10108010635150290102456)

27 Yang Z amp Rannala B 2006 Bayesian estimation ofspecies divergence times under a molecular clockusing multiple fossil calibrations with soft boundsMol Biol Evol 23 212ndash226 (doi101093molbevmsj024)

28 Battistuzzi F U Filipski A Hedges S B amp KumarS 2010 Performance of relaxed-clock methods in esti-mating evolutionary divergence times and their

credibility intervals Mol Biol Evol 27 1289ndash1300(doi101093molbevmsq014)

29 Brown R P amp Yang Z 2010 Bayesian dating of shal-low phylogenies with a relaxed molecular clock SystBiol 59 119ndash131 (doi101093sysbiosyp082)

30 Inoue J Donoghue P C J amp Yang Z 2010 Theimpact of the representation of fossil calibrations onBayesian estimation of species divergence times SystBiol 59 74ndash89 (doi101093sysbiosyp078)

31 Morrone J J amp Crisci J V 1995 Historical biogeogra-phy introduction to methods Annu Rev Ecol Syst 26373ndash401 (doi101146annureves26110195002105)

32 Simpson G G 1965 The geography of evolution collectedessays PhiladelphiaNew York PANY Chilton Books

33 Ronquist F 1997 Dispersalndashvicariance analysis a newapproach to the quantification of historical biogeogra-phy Syst Biol 45 195ndash203 (doi101093sysbio461195)

34 Wen J Xiang Q-Y Qian H Li J Want X-W amp

Ickert-Bond S M Intercontinental and intracontinen-tal biogeographymdashpatterns and methods J Syst Evol4 327ndash329

35 Nylander J A A Olsson U Alstrom P amp Sanmar-tın I 2008 Accounting for phylogenetic uncertainty

in biogeography a Bayesian approach to dispersalndashvicariance analysis of the thrushes (Aves Turdus) SystBiol 57 257ndash268 (doi10108010635150802044003)

36 Ree R H Moore B R Webb C O amp Donoghue

M J 2005 A likelihood framework for inferring theevolution of geographic range on phylogenetic treesEvolution 59 2299ndash2311

37 Ree R H amp Smith S A 2008 Maximum likelihoodinference of geographic range evolution by dispersal

local extinction and cladogenesis Syst Biol 57 4ndash14 (doi10108010635150701883881)

38 Hardy C R amp Linder H P 2005 Intraspecific varia-bility and timing in ancestral ecology reconstruction atest case from the Cape flora Syst Biol 54 299ndash316

(doi10108010635150590923317)39 Maddison W P amp Maddison D R 1992 MacClade

version 3 analysis of phylogeny and character evolutionSunderland MA Sinauer Associates

40 Clark J R Ree R H Alfaro M E King M G

Wagner W L amp Roalson E H 2008 A comparativestudy in ancestral range reconstruction methodsretracing the uncertain histories of insular lineagesSyst Biol 57 693ndash707 (doi101080106351508

02426473)41 Patterson C 1982 Morphological characters and hom-

ology In Problems of phylogenetic reconstruction (eds K AJoysey amp A E Friday) pp 21ndash74 London UK Aca-demic Press

42 Ree R H amp Sanmartın I 2009 Prospects and chal-lenges for parametric models in historicalbiogeographical inference J Biogeogr 36 1211ndash1220(doi101111j1365-2699200802068x)

43 Lamm K S amp Redelings B D 2009 Reconstructing

ancestral ranges in historical biogeography propertiesand prospects J Syst Evol 47 369ndash382 (doi101111j1759-6831200900042x)

44 Maddison W P amp Maddison D R 2009 Mesquite amodular system for evolutionary analysis version 272See httpmesquiteprojectorg

45 Wojcicki M amp Brooks D R 2005 PACT an efficientand powerful algorithm for generating area cladogramsJ Biogeogr 32 755ndash774 (doi101111j1365-2699200401148x)

46 Lim B K 2008 Historical biogeography of New Worldemballonurid bats (Tribe Diclidurini) taxon pulsediversification J Biogeogr 35 1385ndash1401 (doi10

1111j1365-2699200801888x)47 Lim B K 2009 Review of the origins and biogeogra-

phy of bats in South America Chiroptera Neotropical15 391ndash410

48 Donoghue M J amp Moore B R 2003 Toward an inte-

grative historical biogeography J Int Comp Biol 43261ndash270 (doi101093icb432261)

49 Krause D W OrsquoConnor P M Rogers K C Samp-son S D Buckley G A amp Rogers R R 2006 Late

Cretaceous terrestrial vertebrates from Madagascarimplications for Latin American biogeography AnnMo Bot Gard 93 178ndash208 (doi1034170026-6493(2006)93[178LCTVFM]20CO2)

50 Sereno P C Wilson J A amp Conrad J L 2004 New

dinosaurs link southern landmasses in the Mid-Cretac-eous Proc R Soc Lond B 271 1325ndash1330 (doi101098rspb20042692)

51 Novacek M J 1992 Mammalian phylogeny shakingthe tree Nature 356 121ndash125 (doi101038

356121a0)52 Novacek M J 1993 Reflections on higher mammalian

phylogenetics J Mamm Evol 1 1064ndash755453 Springer M S Cleven G C Madsen O de Jong

W W Waddell V G Amrine H M amp Stanhope

M J 1997 Endemic African mammals shake the phylo-genetic tree Nature 388 61ndash64 (doi10103840386)

54 Springer M S Murphy W J Eizirik E amp OrsquoBrienS J 2005 Molecular evidence for major placental

clades In The rise of placental mammals origins andrelationships of the major extant clades (eds K D Roseamp J D Archibald) pp 37ndash49 Baltimore MD JohnsHopkins University Press

55 Stanhope M J Madsen O Waddell V G Cleven

G C de Jong W W amp Springer M S 1998 Highlycongruent molecular support for a diverse superordinalclade of endemic African mammals Mol PhylogenetEvol 9 501ndash508 (doi101006mpev19980517)

56 Stanhope M J Waddell V G Madsen O de Jong

W W Hedges S B Cleven G C Kao D ampSpringer M S 1998 Molecular evidence for multipleorigins of Insectivora and for a new order of endemicAfrican insectivore mammals Proc Natl Acad SciUSA 95 9967ndash9972 (doi101073pnas95179967)

57 Waddell P Okada N amp Hasegawa M 1999 Towardsresolving the interordinal relationships of placentalmammals Syst Biol 48 1ndash5 (doi101093sysbio4811)

58 Waddell P J Kishino H amp Ota R 2001 A phyloge-netic foundation for comparative mammaliangenomics Genome Inform 12 141ndash154

59 Eizirik E Murphy W J amp OrsquoBrien S J 2001 Mol-ecular dating and biogeography of the early placental

mammal radiation J Hered 92 212ndash219 (doi101093jhered922212)

60 Madsen O et al 2001 Parallel adaptive radiations intwo major clades of placental mammals Nature 409610ndash614 (doi10103835054544)

61 Murphy W J Eizirik E Johnson W E Zhang Y PRyder O A amp OrsquoBrien S J 2001 Molecular phyloge-netics and the origins of placental mammals Nature409 614ndash618 (doi10103835054550)

62 Murphy W J et al 2001 Resolution of the early placen-

tal mammal radiation using Bayesian phylogeneticsScience 294 2348ndash2351 (doi101126science1067179)

63 Scally M Madsen O Douady C J de Jong W WStanhope M J amp Springer M S 2001 Molecular

evidence for the major clades of placental mammalsJ Mamm Evol 8 239ndash277 (doi101023A1014446915393)

64 Springer M S amp de Jong W W 2001 Which mamma-lian supertree to bark up Science 291 1709ndash1711

65 Stamatakis A 2006 RAxML-VI-HPC maximum like-lihood-based phylogenetic analysis with thousands oftaxa and mixed models Bioinformatics 22 2688ndash2690

(doi101093bioinformaticsbtl446)66 Benton M J amp Donoghue P C J 2007 Paleontologi-

cal evidence to date the tree of life Mol Biol Evol 2426ndash53 (doi101093molbevmsl150)

67 Reisz R R amp Muller J 2004 Molecular timescales andthe fossil record a paleontological perspective TrendsGenet 20 237ndash241 (doi101016jtig200403007)

68 Muller J amp Reisz R R 2005 Four well-constrainedcalibration points from the vertebrate fossil record for

molecular clock estimates BioEssays 27 1069ndash1075(doi101002bies20286)

69 Gradstein F M amp Ogg J G 2009 The geologic timescale In The timetree of life (eds S B Hedges amp SKumar) pp 26ndash34 Oxford UK Oxford University Press

70 Gheerbrant E 2009 Paleocene emergence of elephantrelatives and the rapid radiation of African ungulatesProc Natl Acad Sci USA 106 10717ndash10721(doi101073pnas0900251106)

71 Bergqvist L P Abrantes E A L amp Avilla L D S

2004 The Xenarthra (Mammalia) of Sao Jose de Ita-boraı Basin (upper Paleocene Itaboraian) Rio deJaneiro Brazil Geodiversitas 26 323ndash337

72 Carlini A A Pascual R Reguero M A Scillato-

Yane G J Tonni E P amp Vizcaıno S F 1990 Thefirst Paleogene land placental mammal from Antarcticaits paleoclimatic and paleobiogeographical bearings InAbstracts IV International Congress of Systematic and Evol-utionary Biology (eds B Cox amp J Reveal) 325 p

Baltimore MD University of Maryland73 Vizcaıno S F amp Scillato-Yane G J 1995 An Eocene

Tardigrada (Mammalia Xenarthra) from SeymourIsland Antarctica Antarctic Sci 7 407ndash408

74 Benton M J Donoghue P C J amp Asher R J 2009

Calibrating and constraining molecular clocks In Thetimetree of life (eds S B Hedges amp S Kumar) pp35ndash86 Oxford UK Oxford University Press

75 Flynn J J 1996 Carnivoran phylogeny and rates ofevolution morphological taxonomic and molecular

In Carnivore behavior ecology and evolution vol 2 (edJ L Gittleman) pp 542ndash581 Ithaca NY CornellUniversity Press

76 Hunt Jr R M amp Tedford R H 1993 Phylogenetic

relationships within the aeluroid Carnivora and impli-cations of their temporal and geographic distribution InMammal phylogeny placentals vol 2 (eds F S SzalayM J Novacek amp M C McKenna) pp 53ndash74 BerlinGermany Springer

77 Wesley-Hunt G D amp Flynn J J 2005 Phylogeny ofthe Carnivora basal relationships among the carnivora-morphans and assessment of the position oflsquoMiacoidearsquo relative to Carnivora J Syst Palaeontol 31ndash28 (doi101017S1477201904001518)

78 Woodburne M O Gunnell G F amp Stucky R K2009 Climate directly influences Eocene mammalfaunal dynamics in North America Proc Natl AcadSci USA 106 13 399ndash13 403 (doi101073pnas0906802106)

79 Rose K D DeLeon V B Missiaen P Rana R SSahni A Singh L amp Smith T 2008 Early Eocenelagomorph (Mammalia) from Western India and theearly diversification of Lagomorpha Proc R Soc B275 1203ndash1208 (doi101098rspb20071661)

80 Storch G amp Seiffert C 2007 Extraordinarily preservedspecimen of the oldest known glirid from the middleEocene of Messel (Rodentia) J Vertebr Paleontol 27

189ndash194 (doi1016710272-4634(2007)27[189EPSOTO]20CO2)

81 Sallam H M Seiffert E R Steiper M E amp SimonsE L 2009 Fossil and molecular evidence constrainscenarios for the early evolutionary and biogeographic

history of hystricognathous rodents Proc Natl AcadSci USA 106 16 722ndash16 727 (doi101073pnas0908702106)

82 Hartenberger L 1998 Description of the radiation of

the Rodentia (Mammalia) from the Late Paleocene tothe Miocene phylogenetic consequences C R AcadSci II A 326 439ndash444

83 Vucetich M G Verzi D H amp Hartenberger L 1999Review and analysis of the radiation of the South

American Hystricognathi (Mammalia Rodentia)C R Acad Sci II A 329 763ndash769

84 Flynn L J amp Jacobs L L 2008 Castoridea In Evolution oftertiary mammals of North America small mammals xenar-thrans and marine mammals (eds C M Janis G F

Gunnell amp M D Uhen) vol 2 pp 391ndash405 CambridgeUK Cambridge University Press

85 Marenssi S A Reguero M A Santillana S N ampVizcaıno S F 1994 Eocene land mammals from Sey-mour Island Antarctica paleobiogeographical

implications Antarctic Sci 6 3ndash15 (doi101017S0954102094000027)

86 MacPhee R D E amp Reguero M A 2010 Reinterpre-tation of a middle Eocene record of Tardigrada (Pilosa

Xenarthra Mammalia) from La Meseta FormationSeymour Island West Antarctica Am Mus Novit3689 1ndash21 (doi1012067031)

87 McKenna M C amp Bell S K 1997 Classification ofmammals above the species level New York NY Colum-

bia University Press88 Gaudin T J amp Branham D G 1998 The phylogeny of

the Myrmecophagidae (Mammalia Xenarthra Vermi-lingua) and relationship of Eurotamandua to theVermilingua J Mamm Evol 5 237ndash265 (doi10

1023A1020512529767)89 Galliari F C Carlini A A amp Sanchez-Villagra M R

2010 Evolution of the axial skeleton in armadillos(Mammalia Dasypodidae) Mamm Biol 75 326ndash333 (doi101016jmambio200903014)

90 Poljak S Confalonieri V Fasanella M Gabrielli Mamp Lizarralde M S 2010 Phylogeography of the arma-dillo Chaetophractus villosus (Dasypodidae Xenarthra)post-glacial range expansion from Pampas to Patagonia

(Argentina) Mol Phylogenet Evol 55 38ndash46 (doi101016jympev200912021)

91 Sige B Crochet J-Y amp Insole A 1977 Les plusvielles taupes Geobios Mem Spec 1 141ndash157(doi101016S0016-6995(77)80014-4)

92 Gunnell G F Bown T M Hutchinson J H ampBloch J I 2008 Lipotyphla In Evolution of Tertiarymammals of North America small mammals xenarthransand marine mammals vol 2 (eds C M Janis G FGunnell amp M D Uhen) pp 89ndash125 Cambridge

UK Cambridge University Press93 Seiffert E R Simons E L Ryan T M Bown T M

amp Attia Y 2007 New remains of Eocene and OligoceneAfrosoricida (Afrotheria) from Egypt with implicationsfor the origin(s) of afrosoricid zalambdodonty J VertebrPaleontol 27 963ndash972 (doi1016710272-4634(2007)27[963NROEAO]20CO2)

94 Tabuce R Asher R J amp Lehmann T 2008 Afrother-ian mammals a review of current data Mammalia 722ndash14 (doi101515MAMM2008004)

95 Butler P M 1995 Fossil Macroscelidea Mammal Rev25 3ndash14 (doi101111j1365-29071995tb00432x)

96 Milledge S 2003 Fossil aardvarks from the Lothagam

beds In Lothagam the dawn of humanity in EasternAfrica (eds J Leakey amp J Harris) pp 363ndash368New York NY Columbia University Press

97 Flynn L J amp Jacobs L L 2008 Aplodontia In Evol-ution of tertiary mammals of North America smallmammals xenarthrans and marine mammals vol 2(eds C M Janis G F Gunnell amp M D Uhen) pp377ndash390 Cambridge UK Cambridge UniversityPress

98 Marivaux L Ducrocq S Jaeger J-J Marandat BSudre J Chaimanee Y Tun S T Htoon W ampSoe A N 2005 New remains of Pondaungimysanomaluropsis (Rodentia Anomaluroidea) from thelatest middle Eocene Pondaung Formation of

Central Myanmar J Vertebr Paleontol 25 214ndash227(doi1016710272-4634(2005)025[0214NROPAR]20CO2)

99 Flynn L J Lindsay E H amp Martin R A 2008Geomorpha In Evolution of Tertiary mammals of NorthAmerica small mammals xenarthrans and marine mam-mals vol 2 (eds C M Janis G F Gunnell amp M DUhen) pp 428ndash455 Cambridge UK CambridgeUniversity Press

100 Antoine P-O et al 2007 The middle Miocene (Laven-

tan) Fitzgcarrald Fauna Amazonian Peru In Proc 4thEur Meeting on the Palaeontology and Stratigraphy ofLatin America (eds E Dıaz-Martınez amp I Rabano)pp 19ndash24 Madrid Spain Instituto Geologico y

Minero de Espana101 Deschamps C M Olivares A I Vieytes E C amp

Vucetich M G 2007 Ontogeny and diversity of theoldest capybaras (Rodentia Hydrochoeridae late Mio-cene of Argentina) J Vertebr Paleontol 27 683ndash692

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

102 Frailey C D amp Campbell K E 2004 The rodents ofthe Santa Rosa Local Fauna In The Paleogene mamma-lian fauna of Santa Rosa Amazonian Peru (ed K E

Campbell Jr) pp 1ndash130 Los Angeles CA NaturalHistory Museum of Los Angeles County ScienceSeries 40

103 Martin T 2004 Evolution of incisor enamel micro-structure in Lagomorpha J Vertebr Paleontol 24

411ndash426 (doi1016712513)104 Lopez-Martinez N 2008 The lagomorph fossil

record and the origin of the European rabbit InLagomorph biology evolution ecology and conservation(eds P C Alves N Ferrand amp K Hacklander)pp 26ndash47 Amsterdam The Netherlands Springer

105 Marivaux L Bocat L Chaimanee Y Jaeger J-JMarandat B Srisuk P Tafforeau P Yamee C ampWelcomme L 2006 Cynocephalid dermopterans

from the Palaeogene of South Asia (Thailand Myan-mar and Pakistan) systematic evolutionary andpalaeobiogeographic implications Zool Scripta 35395ndash420 (doi101111j1463-6409200600235x)

106 Tong Y 1988 Fossil tree shrews from the Eocene

Hetaoyuan Formation of Xichuan Henan VertebrataPalasiatica 26 214ndash220

107 Godfrey L R amp Jungers W L 2002 Quaternary fossillemurs In The primate fossil record (ed W C Hartwig)pp 97ndash121 Cambridge UK Cambridge University

Press108 Bajpai S Kay R F Williams B A Das D P

Kapur V V amp Tiwari B N 2008 The oldest Asianrecord of Anthropoidea Proc Natl Acad Sci USA105 11 093ndash11 098 (doi101073pnas0804159105)

109 Beard K C Qi T Dawson M R Wang B amp Li C1994 A diverse new primate fauna from middle Eocenefissure-fillings in southeastern China Nature 368

604ndash609 (doi101038368604a0)110 Orliac M Boisserie J-R MacLatchy L amp Lihoreau

F 2010 Early Miocene hippopotamids (Cetartiodac-tyla) constrain the phylogenetic and spatiotemporalsettings of hippopotamid origin Proc Natl Acad SciUSA 107 11 871ndash11 876 (doi101073pnas1001373107)

111 Honey J G Harrison J A Prothero D R ampStevens M S 1998 Camelidae In Evolution of Tertiarymammals of North America terrestrial carnivoresungulates and ungulatelike mammals vol 1 (edsC M Janis K M Scott amp L L Jacobs) pp439ndash462 Cambridge UK Cambridge University Press

112 Metais G amp Vislobokova I 2008 Basal ruminants In

The evolution of artiodactyls (eds D R Prothero amp S EFoss) pp 189ndash212 Baltimore MD Johns HopkinsUniversity Press

113 Harris J M amp Li-Ping L 2008 Superfamily SuoideaIn The evolution of artiodactyls (eds D R Prothero amp S

E Foss) pp 130ndash150 Baltimore MD Johns HopkinsUniversity Press

114 Bowen G J Clyde W C Koch P L Ting SAlroy J Tsubamoto T Wang Y amp Wang Y 2002Mammalian dispersal at the PaleoceneEocene bound-

ary Science 295 2062ndash2065 (doi101126science1068700)

115 Smith T Rose K D amp Gingerich P D 2006 RapidAsia-Europe-North America geographic dispersal of

earliest Eocene primate Teilhardina during the Paleo-cenendashEocene thermal maximum Proc Natl Acad SciUSA 103 11223ndash11227 (doi101073pnas0511296103)

116 Prothero D R 1998 Hyracodontidae In Evolutionof Tertiary mammals of North America terrestrial carni-vores ungulates and ungulatelike mammals vol 1 (edsC M Janis K M Scott amp L L Jacobs) pp 589ndash594 Cambridge UK Cambridge University Press

117 Colbert M W amp Schoch R M 1998 Tapiroidea and

other moropomorphs In Evolution of Tertiary mammalsof North America terrestrial carnivores ungulates andungulatelike mammals vol 1 (eds C M Janis K MScott amp L L Jacobs) pp 569ndash582 Cambridge UKCambridge University Press

118 Dashzeveg D 1996 Some carnivorous mammals fromthe Paleogene of the Eastern Gobi Desert Mongoliaand the application of Oligocene carnivores to strati-graphic correlation Am Mus Novit 3179 1ndash14

119 Hunt Jr R M 1998 Evolution of the aeluroid Carni-vora diversity of the earliest aeluroids from Eurasia(Quercy Hsanda-Gol) and the origin of felids AmMus Novit 3252 1ndash65

120 Spaulding M amp Flynn J J 2009 Anatomy of the post-

cranial skeleton of lsquoMiacisrsquo uintensis (MammaliaCarnivoramorpha) J Vertebr Paleontol 29 1212ndash1223 (doi1016710390290408)

121 Storch G 2003 Fossil Old World lsquoedentatesrsquo InMorphological studies in fossil and extant Xen-arthra (Mammalia) Senckenbergiana biologica 83 (edsR A Farina S F Vizcaıno amp G Storch) pp 51ndash60Germany Schweizerbart Science Publishers

122 Huelsenbeck J P amp Ronquist F 2001 MrBAYESBayesian inference of phylogenetic trees Bioinformatics17 754ndash755 (doi101093bioinformatics178754)

123 Ronquist F amp Huelsenbeck J P 2003 MrBayes 3Bayesian phylogenetic inference under mixed modelsBioinformatics 19 1572ndash1574 (doi101093bioinfor-maticsbtg180)

124 Carroll R L 1988 Vertebrate paleontology and evolutionNew York NY W H Freeman and Company

125 Asher R J Novacek M J amp Geisler J G 2003

Relationships of endemic African mammals and theirfossil relatives based on morphological and molecularevidence J Mamm Evol 10 131ndash194 (doi101023A1025504124129)

126 Zack S P Penkrot T A Bloch J I amp Rose K D

2005 Affinities of lsquohyopsodontidsrsquo to elephant shrewsand a Holarctic origin of Afrotheria Nature 434 497ndash501 (doi101038nature03351)

127 Tabuce R Marivaux L Adaci M Bensalah M

Hartenberger J-L Mahboudi M Mebrouk F Taf-foreau P amp Jaeger J 2007 Early Tertiary mammalsfrom North Africa reinforce the molecular Afrotheriaclade Proc R Soc Lond B 274 1159ndash1166 (doi101098rspb20060229)

128 Marshall L G Webb S D Sepkowski J J amp RaupD M 1982 Mammalian evolution and the Great Amer-ican Interchange Science 215 1351ndash1357 (doi101126science21545381351)

129 Simpson G G 1951 History of the fauna of Latin

America In Science in progress 7th series (ed G A Bait-sell) pp 369ndash408 New Haven CT Yale UniversityPress

130 Simpson G G 1978 Early mammals in South Amer-ica fact controversy and mystery Proc Am PhilSoc 122 318ndash328

131 Archibald J D 1996 Fossil evidence for a Late Cretac-eous origin of lsquohoofedrsquo mammals Science 272 1150ndash1153 (doi101126science27252651150)

132 Archibald J D Averianov A O amp Ekdale E G 2001Oldest relatives of rabbits rodents and other extanteutherian mammals Nature 414 62ndash65 (doi10103835102048)

133 Cifelli R L amp Davis B M 2003 Marsupial origins

Science 302 1899ndash1900 (doi101126science1092272)134 Davis B M Cifelli R L amp Kielan-Jaworowska Z

2008 Earliest evidence of Deltatheroida (MammaliaMetatheria) from the Early Cretaceous of North Amer-ica In Mammalian evolutionary morphology a tribute toFrederick S Szalay (eds E J Sargis amp M Dagosto)pp 3ndash24 Amsterdam The Netherlands Springer

135 Boyer D M Prasad G V R Krause D W Godi-not M Goswami A Verma O amp Flynn J J 2010New postcrania of Deccanolestes from the Late Cretac-

eous of India and their bearing on the evolutionaryand biogeographic history of euarchontan mammalsNaturwissenschaften 97 365ndash377 (doi101007s00114-010-0648-0)

136 Nikolaev S Montoya-Burgos J I Margulies E HProgram N C S Rougemont J Nyffeler B amp Anto-narakis S E 2007 Early history of mammals iselucidated with the ENCODE multiple species sequen-cing data PLoS Genet 3 e2 (doi101371journal

pgen0030002)137 van Dijk M A M Paradis E Catzeflis F amp de Jong

W W 1999 The virtues of gaps xenarthran (edentate)monophyly supported by a unique deletion in aA-crys-tallin Syst Biol 48 94ndash106 (doi101080106351599

260463)138 Nikaido M Nishihara H Hukumoto Y amp Okada

N 2003 Ancient SINEs from African endemic mam-mals Mol Biol Evol 20 522ndash527 (doi101093molbevmsg052)

139 Kriegs J O Churakov G Kiefmann M Jordan UBrosius J amp Schmitz J 2006 Retroposed elements asarchives for the evolutionary history of placental mam-mals PLoS Biol 4 e91 (doi101371journalpbio0040091)

140 Nishihara H Satta Y Nikaido M ThewissenJ G M Stanhope M J amp Okada N 2005 A retropo-son analysis of afrotherian phylogeny Mol Biol Evol22 1823ndash1833 (doi101093molbevmsi179)

141 Nishihara H Hasegawa M amp Okada N 2006 Pega-soferae an unexpected mammalian clade revealed bytracking ancient retroposon insertions Proc NatlAcad Sci USA 103 9929ndash9934 (doi101073pnas

0603797103)142 Nishihara H Maruyamab S amp Okada N 2009 Ret-

roposon analysis and recent geological data suggestnear-simultaneous divergence of the three superorders

of mammals Proc Natl Acad Sci USA 106 5235ndash5240

143 Eagles G 2007 New angles on South Atlantic openingGeophys J Int 166 353ndash361 (doi101111j1365-246X200603206x)

144 Torsvik T H Rousse S Labails C amp Smethurst MA 2009 A new scheme for the opening of the SouthAtlantic and the dissection of an Aptian salt basinGeophys J Int 177 1315ndash1333 (doi101111j1365-246X200904137x)

145 Nelson G 1978 From Candolle to Croizat commentson the history of biogeography J Hist Biol 11 269ndash305 (doi101007BF00389302)

146 de Queiroz A 2005 The resurrection of oceanic disper-sal in historical biogeography Trends Ecol Evol 20 68ndash

73 (doi101016jtree200411006)147 Censky E J Hodge K amp Dudley J 1998 Over-water

dispersal of lizards due to hurricanes Nature 395 556(doi10103826886)

148 Poux C Madsen O Marquard E Vieites D R deJong W W amp Vences M 2005 Asynchronous coloniza-tion of Madagascar by the four endemic clades ofprimates tenrecs carnivores and rodents as inferredfrom nuclear genes Syst Biol 54 719ndash730 (doi10

108010635150500234534)149 Yoder A D amp Nowak M D 2006 Has vicariance or

dispersal been the predominant biogeographic force inMadagascar Only time will tell Annu Rev EcolEvol Syst 37 405ndash431 (doi101146annurevecolsys

37091305110239)150 McCall R 1997 Implications of recent geological

investigations of the Mozambique Channel for themammalian colonization of Madagascar Proc R SocLond B 264 663ndash665 (doi101098rspb19970094)

151 Yoder A D Burns M M Zehr S Delefosse TVeron G Goodman S M amp Flynn J J 2003 Singleorigin of Malagasy Carnivora from an African ancestorNature 421 734ndash737 (doi101038nature01303)

152 Poux C Madsen O Glos J de Jong W W ampVences M 2008 Molecular phylogeny and divergencetimes of Malagasy tenrecs influence of data partitioningand taxon sampling on dating analyses BMC Evol Biol8 102 (doi1011861471-2148-8-102)

153 Stankiewicz J Thiart C Masters J C amp de WitM J 2006 Did lemurs have sweepstake tickets Anexploration of Simpsonrsquos model for the colonization ofMadagascar by mammals J Biogeogr 33 221ndash235(doi101111j1365-2699200501381x)

154 Ali J R amp Huber M 2010 Mammalian biodiversity onMadagascar controlled by ocean currents Nature 463653ndash656 (doi101038nature08706)

155 Poux C Chevret P Huchon D de Jong W W ampDouzery E J P 2006 Arrival and diversification of

caviomorph rodents and platyrrhine primates in SouthAmerica Syst Biol 55 228ndash244 (doi10108010635150500481390)

156 Blanga-Kanfi S Miranda H Penn O Pupko TDeBry R W amp Huchon D 2009 Rodent phylogeny

revised analysis of six nuclear genes from all majorrodent clades BMC Evol Biol 9 71 (doi1011861471-2148-9-71)

157 Hasegawa M Thorne J L amp Kishino H 2003 Timescale of eutherian evolution estimated without assuminga constant rate of molecular evolution Genes Gen Syst78 267ndash283 (doi101266ggs78267)

158 Rowe D L Dunn K A Adkins R M amp Honeycutt

R L 2010 Molecular clocks keep dispersal hypothesesafloat evidence for trans-Atlantic rafting by rodentsJ Biogeogr 7 305ndash324

159 Beard K C Wang B Dawson M Huang X amp

Tong Y 1996 Earliest complete dentition of an anthro-poid primate from the late middle Eocene of ShanxiProvince China Science 272 82ndash85 (doi101126science272525882)

160 Beard K C amp Wang J 2004 The eosimiid primates

(Anthropoidea) of the Heti Formation Yuanqu BasinShanxi and Henan Provinces Peoplersquos Republic ofChina J Hum Evol 46 401ndash432 (doi101016jjhevol200401002)

161 Arnason U Gullberg A Schweizer B A amp Janke A

2000 Molecular estimates for primate dispersal and theorigin of modern humans Hereditas 133 217ndash228(doi101111j1601-5223200000217x)

162 Lavocat R 1969 La systematique des rongeurs hystri-comorphes et la derive des continents C R AcadSci Ser D 269 1496ndash1497

163 Hussain S T de Bruijn H amp Leinders J M 1978Middle Eocene rodents from the Kala Chitta Range(Punjab Pakistan) (III) Proc Kon Ned AkadWetensch Ser B 81 101ndash112

164 Wood A E 1985 The relationships origin and disper-sal of the hystricognathous rodents In Evolutionaryrelationships among rodents a multidisciplinary analysis(eds W P Luckett amp J-L Hartenberger) pp 475ndash

513 New York NY Plenum165 Houle A 1999 The origin of platyrrhines an evalu-

ation of the Antarctic scenario and the floating islandmodel Am J Phys Anthropol 109 541ndash559 (doi101002(SICI)1096-8644(199908)1094541AID-

AJPA930CO2-N)166 Tattersall I 2005 Mechanisms of faunal origin and

diversity in island environments the case of Madagas-carrsquos mammals Hellenic J Geosci 41 35ndash46

167 Gunnell G F amp Simmons N B 2005 Fossil evidence

and the origin of bats J Mamm Evol 12 209ndash246(doi101007s10914-005-6945-2)

168 Simmons N B Seymour K L Habersetzer J ampGunnel G F 2008 Primitive early Eocene bat from

Wyoming and the evolution of flight and echolocationNature 451 818ndash822

169 Sige B 1991 Rhinolophoidae et Vespertilionoidea(Chiroptera) du Chambi (Eocene inferieur de Tunisie)Aspects biostratigraphique biogeographique and

paleoecologique de lrsquoorigine des chiropters modernesNeues Jahrb Geol Palaontol Abh 182 355ndash376

170 Hand S J amp Kirsch J A W 1998 A southern originfor the Hipposideridae (Microchiroptera) Evidencefrom the Australian fossil record In Bats phylogenymorphology echolocation and conservation biology (edsT H Kunz amp P A Racey) pp 72ndash90 WashingtonDC Smithsonian Institution Press

171 Luo Z-X Ji Q Wible J R amp Yuan X 2003 AnEarly Cretaceous tribosphenic mammal and metather-

ian evolution Science 302 1934ndash1939 (doi101126science1090718)

172 Martin J E Case J A Jagt J W M Schulp A S ampMulder E W A 2005 A new European marsupial indi-cates a Late Cretaceous high-latitude transatlantic

dispersal route J Mamm Evol 12 495ndash511 (doi101007s10914-005-7330-x)

173 Vullo R Gheerbrant E de Muizon C amp Neraudeau

D 2009 The oldest modern therian mammal fromEurope and its bearing on stem marsupial paleobiogeo-graphy Proc Natl Acad Sci USA 106 19 910ndash19 915

174 Cifelli R L 1990 Cretaceous mammals of SouthernUtah II marsupials and marsupialndashlike mammals

from the Wahweap Formation (Early Campanian)J Vertebr Paleontol 10 320ndash331 (doi10108002724634199010011817)

175 Cifelli R L 1993 Theria of metatherianndasheutherian

grade and the origin of marsupials In Mammal phylo-geny Mesozoic differentiation multituberculatesmonotremes early therians and marsupials vol 1 (edsF S Szalay M J Novacek amp M C McKenna) pp205ndash215 Berlin Germany Springer

176 Cifelli R L 2004 Marsupial mammals from the Albian-Cenomanian (Early-Late Cretaceous) boundary UtahBull Am Mus Nat Hist 285 62ndash79 (doi1012060003-0090(2004)2850062C20CO2)

177 Wible J R 1990 Late Cretaceous marsupial petrosal

bones from North America and a cladistic analysis ofthe petrosal in therian mammals J Vertebr Paleontol10 183ndash205 (doi10108002724634199010011807)

178 Cifelli R L amp de Muizon C 1997 Dentition and jawof Kokopellia juddi a primitive marsupial or near-marsu-

pial from the medial Cretaceous of Utah J MammEvol 4 241ndash258 (doi101023A1027394430433)

179 Case J A Goin F J amp Woodburne M O 2005lsquoSouth Americanrsquo marsupials from the Late Cretaceous

of North America and the origin of marsupial cohortsJ Mamm Evol 11 223ndash255 (doi101023BJOMM00000473393963082)

180 Amrine-Madsen H Scally M Westerman MStanhope M J Krajewski C amp Springer M S

2003 Nuclear gene sequences provide evidence for themonophyly of australidelphian marsupials MolPhylogenet Evol 28 186ndash196 (doi101016S1055-7903(03)00122-2)

181 Beck R M Godthelp H Weisbecker V Archer M

amp Hand S J 2008 Australiarsquos oldest marsupial fossilsand their biogeographical implications PLoS ONE 3e1858 (doi101371journalpone0001858)

182 Springer M S Krajewski C amp Meredith R W 2009Marsupials (Metatheria) In The timetree of life (eds S B

Hedges amp S Kumar) pp 466ndash470 Oxford UKOxford University Press

183 Kirsch J A W Dickerman A W Reig O A ampSpringer M S 1991 DNA hybridization evidence

for the Australasian affinity of the American marsupialDromiciops australis Proc Natl Acad Sci USA 8810 465ndash10 469 (doi101073pnas882310465)

184 Kirsch J A W Lapointe F-J amp Springer M S 1997DNA-hybridisation studies of marsupials and their

implications for metatherian classificationAust J Zool 45 211ndash280 (doi101071ZO96030)

185 Springer M S Westerman M Kavanagh J R BurkA Woodburne M O Kao D amp Krajewski C 1998The origin of the Australasian marsupial fauna and the

phylogenetic affinities of the enigmatic monito delmonte and marsupial mole Proc R Soc Lond B 2652381ndash2386 (doi101098rspb19980587)

186 Nilsson M A Churakov G Sommer M Tran NV Zemann A Brosius J amp Schmitz J 2010 Tracking

marsupial evolution using archaic genomic retroposoninsertions PLoS Biol 8 e1000436 (doi101371jour-nalpbio1000436)

187 Woodburne M O amp Case J A 1996 Dispersal vicar-iance and the Late Cretaceous to early Tertiary land

mammal biogeography from South America to Austra-lia J Mamm Evol 3 121ndash161 (doi101007BF01454359)

188 Rowe T Rich T H VickersndashRich P Springer MS amp Woodburne M O 2008 The oldest platypusand its bearing on divergence timing of the platypusand echidna clades Proc Natl Acad Sci USA 1051238ndash1242 (doi101073pnas0706385105)

189 Woodburne M O Rich T H amp Springer M S 2003The evolution of tribospheny and the antiquity of mam-malian clades Mol Phylogenet Evol 28 360ndash385(doi101016S1055-7903(03)00113-1)

190 Phillips M J Bennetta T H amp Lee M S Y 2009Molecules morphology and ecology indicate a recentamphibious ancestry for echidnas Proc Natl Acad

Sci USA 106 17 089ndash17 094 (doi101073pnas0904649106)

191 Springer M S amp Krajewski C 2009 Monotremes

(Prototheria) In The timetree of life (eds S B Hedges ampS Kumar) pp 462ndash465 Oxford UK OxfordUniversity Press

192 Luo Z-X Cifelli R L amp Kielan-Jaworowska Z 2001Dual origin of tribosphenic mammals Nature 409

53ndash57 (doi10103835051023)193 Luo Z-X Cifelli R L amp Kielan-Jaworowska Z 2002

In quest for a phylogeny of Mesozoic mammals ActaPalaeont Polon 47 1ndash78

194 Kielan-Jaworowska Z Cifelli R L amp Luo X 2004Mammals from the age of dinosaursmdashorigins evolutionand structure New York NY Columbia University Press

The historical biogeography of Mammalia

Introduction
Phylogeny reconstruction
Molecular dating analyses
Ancestral area reconstruction
Box 1
Ancestral area chronograms and palaeogeography
Placental phylogeny and a comparison of different ancestral area reconstruction methods
Placental biogeography
The importance of dispersal
Bat biogeography
Marsupial biogeography
Monotreme biogeography
Conclusions
MSS and WJM acknowledge support from NSF Three anonymous reviewers provided helpful comments on an earlier draft of this manuscript We thank Kate Jones for inviting us to contribute to this volume
REFERENCES

Phil Trans R Soc B (2011) 366 2478ndash2502

doi101098rstb20110023

Research

One conmamma

The historical biogeography of MammaliaMark S Springer1 Robert W Meredith1 Jan E Janecka2

and William J Murphy2

1Department of Biology University of California Riverside CA 92521 USA2Department of Veterinary Integrative Biosciences College of Veterinary Medicine and Biomedical Sciences

Texas AampM University College Station TX 77843 USA

Palaeobiogeographic reconstructions are underpinned by phylogenies divergence times and ances-tral area reconstructions which together yield ancestral area chronograms that provide a basis forproposing and testing hypotheses of dispersal and vicariance Methods for area coding includemulti-state coding with a single character binary coding with multiple characters and stringcoding Ancestral reconstruction methods are divided into parsimony versus Bayesianlikelihoodapproaches We compared nine methods for reconstructing ancestral areas for placental mammalsAmbiguous reconstructions were a problem for all methods Important differences resulted fromcoding areas based on the geographical ranges of extant species versus the geographical provenanceof the oldest fossil for each lineage Africa and South America were reconstructed as the ancestralareas for Afrotheria and Xenarthra respectively Most methods reconstructed Eurasia as the ances-tral area for Boreoeutheria Euarchontoglires and Laurasiatheria The coincidence of moleculardates for the separation of Afrotheria and Xenarthra at approximately 100 Ma with the plate tec-tonic sundering of Africa and South America hints at the importance of vicariance in the earlyhistory of Placentalia Dispersal has also been important including the origins of Madagascarrsquosendemic mammal fauna Further studies will benefit from increased taxon sampling and theapplication of new ancestral area reconstruction methods

Keywords ancestral areas dispersal historical biogeography Mammalia vicariance

1 INTRODUCTIONClass Mammalia is impressive for its taxonomic ecologi-cal and morphological diversity [1] A fundamental goalof mammalian palaeobiogeography is to reconstruct theunderlying history of vicariant and dispersal events thathave shaped this diversity Here we highlight theimportance of phylogeny reconstruction ancestral areareconstruction and molecular dating for producingancestral area chronograms We compare differentapproaches for reconstructing ancestral areas andillustrate similarities and differences between theseapproaches using a dataset for placental mammals Weconclude with a review of selected topics in placentalmammal palaeogeography that illustrates how phyloge-nies ancestral area reconstructions molecular datesand palaeographic histories have reshaped our views onmammalian historical biogeography Finally we identifyimportant areas for future inquiry

2 PHYLOGENY RECONSTRUCTIONPhylogeny reconstruction begins with character dataLarge molecular datasets have yielded robust phyloge-nies for many groups thereby reducing the number of

r for correspondence (markspringerucredu)

tribution of 12 to a Theme Issue lsquoGlobal biodiversity oflsrsquo

phylogenetic hypotheses that must be considered whenformulating ancestral area chronograms The inclusionof morphological data from fossils allows for taxono-mically richer phylogenies while also providing keydata points that bear on the geographical provenanceof a taxonomic group In the words of Simpson [2]fossils lsquoare the historical documents of animal distri-butionrsquo Fossils are also more difficult to place withconfidence in a phylogenetic framework owing to miss-ing (molecular) data and the inability of currentmethods to separate homology and homoplasy withsome morphological datasets [34]

Maximum parsimony (MP) and maximum likeli-hood (ML) yield a single best tree(s) whereasBayesian methods yield a sampling of trees from pos-terior probability space ML and Bayesian methodshave the advantage of incorporating models ofsequence evolution and yield trees with branchlengths Some ancestral area reconstruction methodssuch as those implemented in SIMMAP [5] cantake advantage of trees with branch lengths as wellas multiple trees from posterior probability space

Phylogeny reconstruction is usually the first step inconstructing an ancestral area chronogram followedby the estimation of divergence times at each of thenodes However BEAST [6] allows for simultaneousestimation of branching relationships and divergencetimes After reconstructing a phylogeny molecular

This journal is q 2011 The Royal Society

phylogenies

paleontologicaldata

chronograms

moleculardating

neontologicaldata

0204060

NeogeneQuaternary

CenozoicPaleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

node numbera

0204060

NeogeneQuaternary

Cenozoic

Paleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

chpars

salndash

lndashvic

salndash

lndashex

o(lsquo

splitrsquo

inhabit

iphilia

ulipoty

orexthorn

Susthorn

Bosthorn

toglire

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

phylogenies

paleontologicaldata

chronograms

moleculardating

neontologicaldata

0204060

NeogeneQuaternary

CenozoicPaleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

node numbera

0204060

NeogeneQuaternary

Cenozoic

Paleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

chpars

salndash

lndashvic

salndash

lndashex

o(lsquo

splitrsquo

inhabit

iphilia

ulipoty

orexthorn

Susthorn

Bosthorn

toglire

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

0204060

NeogeneQuaternary

CenozoicPaleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

node numbera

0204060

NeogeneQuaternary

Cenozoic

Paleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

chpars

salndash

lndashvic

salndash

lndashex

o(lsquo

splitrsquo

inhabit

iphilia

ulipoty

orexthorn

Susthorn

Bosthorn

toglire

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

0204060

NeogeneQuaternary

CenozoicPaleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

node numbera

0204060

NeogeneQuaternary

Cenozoic

Paleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

chpars

salndash

lndashvic

salndash

lndashex

o(lsquo

splitrsquo

inhabit

iphilia

ulipoty

orexthorn

Susthorn

Bosthorn

toglire

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

0204060

NeogeneQuaternary

CenozoicPaleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

node numbera

0204060

NeogeneQuaternary

Cenozoic

Paleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

chpars

salndash

lndashvic

salndash

lndashex

o(lsquo

splitrsquo

inhabit

iphilia

ulipoty

orexthorn

Susthorn

Bosthorn

toglire

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

0204060

NeogeneQuaternary

CenozoicPaleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

node numbera

0204060

NeogeneQuaternary

Cenozoic

Paleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

chpars

salndash

lndashvic

salndash

lndashex

o(lsquo

splitrsquo

inhabit

iphilia

ulipoty

orexthorn

Susthorn

Bosthorn

toglire

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

node numbera

0204060

NeogeneQuaternary

Cenozoic

Paleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

chpars

salndash

lndashvic

salndash

lndashex

o(lsquo

splitrsquo

inhabit

iphilia

ulipoty

orexthorn

Susthorn

Bosthorn

toglire

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

0204060

NeogeneQuaternary

Cenozoic

Paleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

chpars

salndash

lndashvic

salndash

lndashex

o(lsquo

splitrsquo

inhabit

iphilia

ulipoty

orexthorn

Susthorn

Bosthorn

toglire

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

0204060

NeogeneQuaternary

Cenozoic

Paleogene

Upper Miocene PP

Paleoc

Choloepus

Tamandua

Chaetophractus

Erinaceus

Echinops

Amblysomus

Procavia

Loxodonta

Orycteropus

Tamias

Muscardinus

Rattus

PedetesHystrix

Castor

Dipodomys

CaviaHydrochaeris

Erethizon

SylvilagusOchotona

Cynocephalus

TupaiaLemur

Tarsius

Hippopotamus

Tragelaphus

Ceratotherium

TapirusFelis

Africa

South America

Eurasia

North America

chpars

salndash

lndashvic

salndash

lndashex

o(lsquo

splitrsquo

inhabit

iphilia

ulipoty

orexthorn

Susthorn

Bosthorn

toglire

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

chpars

salndash

lndashvic

salndash

lndashex

o(lsquo

splitrsquo

inhabit

iphilia

ulipoty

orexthorn

Susthorn

Bosthorn

toglire

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

tinued

ethorn

hathorn

topolo

replica

elpara

lsquo-f

arsquo

esboots

illion

million

babilit

(lsquosp

rsquo)at

splits

follow

siafrac14

icafrac14

siathorn

icafrac14

splits

right-

icathorn

right-

splits

athorn

babilit

4thorn

0frac14

follow

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

iphilia

lipoty

orexthorn

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

tinued

Susthorn

Bosthorn

toglire

ethorn

hathorn

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

Africa

Tanzania

Mozambique

100 mi

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

PacificOcean

NorthAtlanticOcean

SouthAtlantic Ocean

IndianOcean

NorthAmerica

SouthAmerica

Africa

Antarctica

Australia

Europe

PacificOcean1b

Middle Eocene

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

(doi1016710272-4634(2007)27[683OADOTO]20CO2)

Introduction
Box 1
Bat biogeography
Conclusions
REFERENCES

The historical biogeography of Mammalia€¦ · Research The historical biogeography of Mammalia Mark S. Springer1,*, Robert W. Meredith1, Jan E. Janecka2 and William J. Murphy2 1Department

Documents

CICAVCE (MAMMALIA)

Biogeography and Humans March 24, 2014. Humans Kingdom:...

Molecular phylogeny of treeshrews (Mammalia: Scandentia...

Island biogeography - Ecology...

Herp Biogeography

Mammalia Species

Ordem Insectivora (Mammalia)

Biogeography Notes

UNIVERSIDAD DE SAN CARLOS DE GUATEMALA FACULTAD … · Filo...

Gambar praktikum mammalia

Kindly Pay Cash Or Furnish Good Security Kingdom Phylum...

BEMO T03...

Topic 9: Biogeography What is biogeography?

Fichas Clase Mammalia

Cladistic Biogeography

Clado mammalia