Amber fossils demonstrate deep-time stability of Caribbean lizard communities

18 Pages·2015·4.35 MB·English

Amber fossils demonstrate deep-time stability of
Caribbean lizard communities
EmmaSherratta,b,1,MaríadelRosarioCastañedab,2,Russell J.Garwoodc,D.LukeMahlerd,3,ThomasJ.Sangere,
AnthonyHerrelf,g,KevindeQueirozh,4,andJonathanB.Lososb,4
aSchoolofEnvironmentalandRuralScience,UniversityofNewEngland,Armidale,NSW2351,Australia;bDepartmentofOrganismicandEvolutionary
BiologyandMuseumofComparativeZoology,HarvardUniversity,Cambridge,MA02138;cSchoolofEarth,AtmosphericandEnvironmentalSciences,
UniversityofManchester,ManchesterM139PL,UnitedKingdom;dCenterforPopulationBiology,UniversityofCalifornia,Davis,CA95616;eDepartmentof
MolecularGeneticsandMicrobiology,UniversityofFlorida,Gainesville,FL32610;fDépartementd’EcologieetdeGestiondelaBiodiversité,UMR7179,
MuséumNationald’HistoireNaturelle,Paris,France;gEvolutionaryMorphologyofVertebrates,GhentUniversity,B-9000Gent,Belgium;andhDepartment
ofVertebrateZoology,NationalMuseumofNaturalHistory,SmithsonianInstitution,Washington,DC20560
EditedbyDavidM.Hillis,TheUniversityofTexasatAustin,Austin,TX,andapprovedJune19,2015(receivedforreviewApril1,2015)
Whether the structure of ecological communities can exhibit Amber, which is fossilized tree resin, provides a unique op-
stabilityovermacroevolutionarytimescaleshaslongbeendebated. portunitytoexamineancientecosystemsbyofferingunparalleled
ThesimilarityofindependentlyevolvedAnolislizardcommunitieson preservation of biological material. In the Greater Antilles,
environmentallysimilarGreaterAntilleanislandssupportsthenotion amber has been found on a few islands in small quantities, but
that community evolution is deterministic. However, a dearth of amber from the Dominican Republic stands out due to its
CaribbeanAnolisfossils—onlythreehavebeendescribedtodate— quantityandqualityoffossilinclusions(10–12).Todate,amber
fossils have been recovered for a variety of vertebrate groups
hasprecludeddirectinvestigationofthestabilityofanolecommu-
(13–16), but most are presently known from only one or a few
nitiesthroughtime.Herewereportonanadditional17fossilanoles
inDominicanamberdatingto15–20Mybeforethepresent.Using scpluesciiomnesnasb,owuhtitchheleimvoitlsuttihoenpoofttehnetsiealgfroorupdsr.awing general con- ON
datacollectedprimarilybyX-raymicrocomputedtomography(X-ray The pre-Pleistocene fossil record for anoles consists solely of UTI
micro-CT), we demonstrate that the main elements of Hispaniolan amber fossils (15, 17–19). Previous authors have reported three OL
anoleecomorphologicaldiversitywereinplaceintheMiocene.Phy- anole fossils from Dominican amber, all dated within the Mio- EV
logeneticanalysisyieldsresultsconsistentwiththehypothesisthat cene,15–20Mya(20)andpotentiallymembersofthesamespecies
the ecomorphs that evolved in the Miocene are members of the (15, 18, 19). We vastly expanded this sample by examining an
same ecomorph clades extant today. The primary axes of ecomor- additional 35 fossils (as well as reexamining the original 3), in-
phologicaldiversityintheHispaniolananolefaunaappeartohave cluding all but one specimen known to us. Of these, we present
changed little between the Miocene and the present, providing dataon20fossilsthatpreservesubstantialskeletalmaterialorsoft
evidence for the stability of ecological communities over macro- tissue to provide information on ecomorphological variation
evolutionarytimescales. (Table S1). The fossils vary in state of preservation and com-
pletenessoftheskeleton,fromfullskeletonstoisolatedlimbsand
adaptiveradiation|ecomorph|Hispaniola|Miocene|Anolis heads (Fig. 1 A–D and Movie S1). Extraordinarydetail of the
Ecologists and evolutionary biologists have long debated the Significance
temporal stability of community structure. In recent years,
many comparisons of glacial and postglacial communities have Anunresolvedquestioninecologyiswhetherthestructureof
shown great dissimilarities—species have responded to environ- ecologicalcommunitiescanbestableoververylongtimescales.
mental change individualistically, indicating that the structure Herewedescribeawealthofnewamberfossilsforanancient
and function of assemblages can change substantially over even radiationofHispaniolanlizardsthat,untilnow,hashadavery
short geological timescales (e.g., several thousand to a few mil- poorfossil record. These fossilsprovideanimportantand pre-
lionyears)(1–3).Conversely,someresearchershavearguedfor viously unavailable perspectiveonan ecologically well-studied
stabilityincommunitystructure(4),leadingtocommunitiesthat groupandindicatethatanolelizard communitiesoccurringon
may be composed of stable sets of ecological specialists for Hispaniola20Myaweremadeupofthesametypesofhabitat
millionsofyears(5,6). specialistspresentinthisgrouptoday.Thesedataindicatethat
CaribbeanAnolislizardsareacontemporarymodelofsimilarity theecologicalprocessesimportantinextantanolecommunities
in community structure. Replicated adaptive radiations have oc- havebeenoperativeoverlongperiodsoftime.
curredindependentlyoneachislandintheGreaterAntilles,pro-
ducing a similar set of habitat specialists, termed ecomorphs, on
Authorcontributions:E.S.,K.d.Q.,andJ.B.L.designedresearch;E.S.,M.d.R.C.,R.J.G.,D.L.M.,
each island (7, 8). This similarity in community structure across K.d.Q.,andJ.B.L.performedresearch;T.J.S.andA.H.contributednewreagents/analytic
islandssuggeststhatanolecommunitiesaremorethanephemeral tools;E.S.,M.d.R.C.,R.J.G.,andD.L.M.analyzeddata;andE.S.,M.d.R.C.,R.J.G.,D.L.M.,
and haphazard sets of species that happen to occupy the same T.J.S.,K.d.Q.,andJ.B.L.wrotethepaper.
placeatthesametime(9).However,intheabsenceofasubstantial Theauthorsdeclarenoconflictofinterest.
fossilrecord, it hasnot previouslybeenpossibleto directly assess ThisarticleisaPNASDirectSubmission.
whether similarly structured communities are a recent phenome- Datadeposition:Rawmorphometricandphylogeneticdataandhigh-resolutionim-
non or whether ecological stability has been a longstanding char- agesofthefossilshavebeenarchivedinZenodo,https://zenodo.org/record/17442
acteristicoftheseislandanolefaunas. (doi:10.5281/zenodo.17442).
Here we report on 20 remarkableAnolis fossils in amber from 1Towhomcorrespondenceshouldbeaddressed.Email:[email protected].
theislandofHispaniola.Using3DreconstructionsbasedonX-ray 2Presentaddress:SecciónGenética,DepartamentodeBiología,UniversidaddelValle,A.A.
microcomputed tomography (X-ray micro-CT) and detailed 25360Cali,ValledelCauca,Colombia.
morphometricanalysisofthesefossils,weshowthatthemainele- 3Presentaddress:DepartmentofEcology&EvolutionaryBiology,UniversityofToronto,
ments of Hispaniolan anole ecomorphological diversity were in Toronto,ON,M5S3B2,Canada.
place in the Miocene, confirming the antiquity of the island radi- 4K.d.Q.andJ.B.L.contributedequallytothiswork.
ation. This continuity emphasizes the stability of ecology speciali-
Thisarticlecontainssupportinginformationonlineatwww.pnas.org/lookup/suppl/doi:10.
zationsubsequenttoinitialadaptivediversification. 1073/pnas.1506516112/-/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1506516112 PNASEarlyEdition | 1of6
squamation is preserved in many fossils (Fig. 1D), revealing intermediate-length forelimbs, and narrow toepads with few to
thesubdigitallamellae(transverselywidenedscalescomprising intermediate numbers of lamellae (on toes of the fore- and
thetoepad),acharacterthatinformsonthearborealbehavior hindfeet, respectively). Two fossils are assigned with high prob-
of these lizards (21) (Fig. 1E and Fig. S1). abilitytothetrunk-groundecomorph:onerelativelylong-limbed
WeusedX-raymicro-CT,lightmicroscopy,andphotographsto partial skeleton with an intermediate number of lamellae (M)
examinetheamberfossils(TableS1)andpreservedspecimensof (Fig.2BandCandFigs.S1andS4),andasecond(O)composed
extantHispaniolanspeciesrepresentingdifferentecomorphs.We ofabroadskullassociatedwithanintermediate-lengthforelimb
recorded linear measurements of the cranial and postcranial (Fig.2CandFigs.S2–S4).
skeletonandcountsofthenumberoftoepadlamellae(TableS2), Members of the trunk ecomorph are found on broad tree
which are known to correlate with microhabitat use in extant trunksandhavebroad,shortheads,longhindlimbsandforelimbs,
anoles (21), to test the hypothesis that Miocene Hispaniolan and hindfoot toepads with intermediate lamella numbers. Two
anoles filled the same ecological niches as their extant counter- fossils (E and L) have similarly long hindlimbs (with femur and
parts. Discriminant function analysis (DFA) was used to assign tibia proportionally longer than metatarsal length) and in-
fossils to ecomorph category using body size-corrected morphom- termediatenumbersoflamellae(Fig.2BandCandFigs.S1and
etricdataandlamellanumberforextanttaxa,withfossilsassigneda S4)andareassignedwithhighprobabilitytothetrunkecomorph.
posteriori. Then we used the reconstructions and direct observa- Last, the fossil with the smallest body size, a headless skeleton
tions to score the amber fossils for morphological characters tra- with preserved lamellae (D), is assigned to the twig ecomorph.
ditionallyusedinanolephylogenetics;thesecharactersdifferfrom Twiganoleshaveveryshortlimbsanduseverynarrowbranches.
thoseusedinecomorphologicalanalysesanddonotindicatethat Theyalsohavetoepadswithfewlamellae.FossilDhassimilarly
members of the same ecomorph class form a clade (22). We shortlimbsandfewlamellae(Fig.2BandCandFigs.S1andS4).
inferred the phylogenetic relationships of the fossils to extant Theremainingsixfossilsthatpreservesubstantialskeletalmaterial
species to test whether the fossils are members of clades repre- orsofttissuearenotassignedtoanyecomorphwithhighprobability
sentingthe sameecomorphspresentonHispaniolatoday. (Table S3). Two isolated skulls are assigned with probabilities
between 0.80 and 0.90 to the trunk-crown (N) and trunk (K)
ResultsandDiscussion ecomorphs,respectively,andacompleteskeletonwithpreserved
Thefossilsvarysubstantiallyinnumberoflamellaeandbodypro- lamellae(G)isassignedtothetrunk-crownecomorph,butwith
portionsandrepresentalargeportionofmodernecomorphological only an intermediate probability (0.69). In these three cases, as
diversity(Fig. 2 and Figs. S1–S4). Fourteen fossils are assigned to with fossils with higher DFA probabilities, the individual mor-
anecomorphclasswithhighprobability(greaterthan0.90;Fig.2A phologicalcharactersareconsistentwiththeirecomorphassign-
and Table S3). Of these, nine are assigned to the trunk-crown ments(Fig.2andFigs.S3andS4).Threeadditionalfossilsrep-
ecomorph: seven complete skeletons, five with preserved la- resented by isolated fragments of forelimbs with toepads (P, Q,
mellae (A, C, F, H, I) and two without (B and J); one headless and S) are assigned to ecomorph classes with low probability.
body(T);andafragmentaryfossilwithforelimblamellaepreserved Giventhatoverlapexistsamongtheecomorphsinlamellanumber
(R).Aswithmodern-daytrunk-crownspecies,whicharefoundhigh (Fig.2B),ourinabilitytoassignthesefossilstoasingleecomorphis
ontreetrunksandinthecanopy,thesefossilshaverelativelyshort not surprising, although the observed variation in lamella number
limbs and long, narrow heads. Extant Hispaniolan trunk-crown indicatesthatmembersofmorethanoneecomorphclassmaybe
anoles exhibit considerable variation in toepad lamella number presentinthistrio.
(23), with smaller-bodied species having fewer lamellae than In summary, morphometric analysis places fossils in four eco-
larger-bodiedspecies(Fig.2B).Ourresultssuggestthatspeciesof morphswithhighprobability.Completenessofmanyofthetrunk-
bothsizeclassesarepresentintheamberfossilrecord,thesmaller crownandthetwotrunk-groundfossilsprovidesconfidenceintheir
beingmorecommoninoursample(e.g.,A,F,H,I,andR). ecomorphassignments.Thepresenceofkeyskeletalelementsand
Moderntrunk-groundspeciesaregenerallyfoundaroundthe thedistinctivemorphologyoftwiganolesalsorendersplausiblethe
bases of trees and have broad, short heads, long hindlimbs, assignment of fossil D to that class. By contrast, the two fossils
Fig.1. Fossilanolelizardspreservedinamber.Someofthefossilsinthisstudyareexceptionallywellpreserved(A–C).FromX-raymicro-CTscanning,the
skeletoncanbereconstructedin3D,revealingcompleteskeletons(B),fullyarticulatedskulls(C),andfragments(D).Theexternalsurfaceofthelizardis
sometimesoutlinedintheamberbyair-filledvoids,whichwhenreconstructedin3Drevealdetailsofthebodyscales(D)andsubdigitallamellae(E,Center).
Scaledetailisalsovisiblethroughtheamber(E,LeftandRight).TheamberfossilsarerepresentedbyletterscorrespondingtothoseinFig.2.Animationsof
the3DreconstructedfossilsareinMovieS1.
2of6 | www.pnas.org/cgi/doi/10.1073/pnas.1506516112 Sherrattetal.
A B
Fig. 2. Amber fossils exhibit morphometric and
meristic variation, which corresponds to variation
among the modern ecomorphs. (A) Results of dis-
criminant function analyses (DFA) conducted sepa-
rately on each fossil. Columns indicate variables
present in each fossil: estimated size (snout-vent
length), cranial (Cran.), and postcranial (Postc.)
measurements and lamella number (Lam.); addi-
tionaldetailsaregiveninTableS2.Asubsetofthe
variablesaredisplayedtoillustratevariationamong
the fossils and how it corresponds with variation
amongtheecomorphs(BandC).Amberfossilsare
colored according to the DFA results (if assigned
withgreaterthan0.90probability).(B)Variationin
thenumberofsubdigitallamellaeonthefore-and
hindfeetoftheamberfossilsisshownalongsidethe
C range (bars) for each ecomorph. Large and small
trunk-crown species also differ in lamella number
(Fig.S1).(C)Size-correctedresidualsofulnaandfe-
murlengthandheadwidth.TheDFAresultscanbe
understood by examining character variation. For
example, for amber fossil C, the trunk-crown eco-
morphistheonlyoneforwhichallmeasurements N
O
fallwithintherangeforthepresent-dayspecies.All TI
cranial and postcranial variables are illustrated in LU
O
Figs.S3andS4,whichshouldbeconsultedtounderstand V
E
fullytheDFAclassificationofsomeofthefossils.
assignedtothetrunkecomorpharefragmentary,comprisingonly placementsoftwoofthesetrunk-crownfossils(BandT)included
hindquarters,andthustheirassignmentislesscertain.Giventhat closerelationshipswiththechlorocyanusclade(Fig.S5).
someextantHispaniolananolesdonotconformtoanyecomorph Phylogenetic results were inconclusive for fossils identified
classandaresimilartotrunkanolesinhindlimbdimensions(21, with high probability as trunk-ground, trunk, and twig anoles.
24),wecannotruleoutthepossibilitythatthesefossils,although However,inallbutonecase(L),Bayesianandparsimonyresults
most similar among the ecomorphs to trunk anoles, are not both indicated the possibility of close relationships between
members of any ecomorph class. Regardless of whether these fossils and Hispaniolan clades to which their extant ecomorph
fossils are trunk anoles, the data indicate that at least four eco- counterpartsbelong.Forthetwotrunk-groundfossils(MandO),
morphologically distinct species occurred in Hispaniola in the Bayesian results included membership in the trunk-ground
Miocene.Moreover,basedonthenumbersoftoepadlamellae,our cybotesseriesamongpossibleplacements(Fig.4andTableS4),a
datasuggesttheexistenceoftwotrunk-crownspecies,likelypar- relationshipthatwasevenmorestronglysupportedinparsimony
allelingthelargeandsmallmodern-daymembersofthatecomorph analyses(cybotesseriesmembershipoffossilsMandOin1/3and
(23).Insummary,althoughtodateonlyoneanolespecieshasbeen 1/1 most parsimonious trees, respectively; Fig. S5). Likewise,
described from Dominican amber (15), our data reveal the exis- trunk ecomorph fossils (E and L) were inferred in the trunk
tenceofminimallyfouradditionalspecies. specialist distichus series, and the twig ecomorph fossil (D) was
TheexistenceoffourecomorphsinHispaniolaintheMioceneis inferred as sister to the twig specialist insolitus, in some of the
consistent with molecular dating analyses, which indicate the exis- many credible phylogenetic trees (Table S4; although fossil L
tenceofatleastfourandpossiblysixpresent-dayecomorphsatthat only grouped with the distichus clade in the Bayesian analysis).
time(Fig.3).Nonetheless,giventhateachoftheanoleecomorphs The relatively few available morphological characters (n = 91)
has evolved multiple times (albeit primarily on different islands), usedinanolephylogeneticsandthepartialnatureofthecharacter
their fossil representatives might not be members of the same data for these fragmentary fossils both likely contribute to the
ecomorph clades present on Hispaniola today; an alternative pos- uncertain phylogenetic placement of many of these specimens.
sibilityisthatthefossilsrepresentadditionalinstancesofevolution Overall,thephylogeneticresultsarecongruentwiththeclosere-
oftheseecomorphsincladesthatdidnotpersisttothepresent.We lationships of the fossils to extant species representing the same
testedthishypothesisbyusingmorphologicalcharacterstraditionally ecomorphs. These results support the hypothesis that the anole
used in anole phylogenetics (22) to infer the relationships of the ecomorphsevolvedearlyintheCaribbeanadaptiveradiationand
fossilstoextantspecies. thatanolecommunitystructurehasbeenstableeversince.
In Bayesian phylogenetic analyses, all 14 fossils with DFA as- Justasinterestingastheecomorphspresentintheamberfauna
signments greater than 0.90 were inferred within crown group arethosethatareabsent.Crown-giantshatchatasizeconsiderably
Anolis(Fig.4).Parsimonyanalysesyieldedsimilarresults,although largerthanotheranoles,andthelackoflargefossilswithjuvenile
onefossilwithahighDFAassignment(B)wasnotinferredwithin characteristics(18)(Fig.S6B)indicatesthatnocrown-giantanoles
crowngroupAnolis(Fig.S5).Fivefossilsidentifiedastrunk-crown werepresentinoursample.Giventhatthesimilarlyarborealtrunk-
anoleswith high-probability(A,F,H,I,andJ;Fig.2AandTable crownanoleswerepreservedingreatnumbers,habitatusedoesnot
S3),wereplacedwiththechlorocyanusseriesofextanttrunk-crown explain the absence of crown-giants. One possibility, of course, is
anoleswithstrongsupportinBayesiananalysesandunambiguously thatcrown-giantsaretoolargetobecometrappedinresin.Inad-
inparsimonyanalyses(Fig.4,Fig.S5,andTableS4).Otherfossils dition,crown-giantstendtobelesscommonthanmembersofother
identified as trunk-crown anoles (B, C, R, and T) were placed ecomorphs, and thus their absence may be a reflection of their
ambiguously in the phylogeny, revealing little about their phylo- abundance. A last possibility, however, is that crown-giant anoles
genetic relationships to their extant ecomorph counterparts. maynothaveoccurredonHispaniolawhentheamberfossilswere
However,inbothBayesianandparsimonyanalyses,thealternative formed; molecular dating (25) places the stem age of the ricordii
Sherrattetal. PNASEarlyEdition | 3of6
Amber horizon However,untilnowtherehasbeenaholeinourunderstandingof
this system because of a dearth of fossil evidence of their evolu-
tionarypast.OuranalysisofthediversityoftheMiocenefaunaof
Hispaniola reveals that anole community structure has remained
muchthesameforatleastthelast20My,providingstrongevidence
forthestabilityofcommunitiesacrosslongperiodsoftimeduring
which substantial environmental change has occurred. Whether
theseresultsareparalleledinotherCaribbeangroupsremainstobe
Twig seen;emergingfossils(16,31)ofaradiationofgekkonidlizardsmay
Trunk proveaninterestingcomparison.
Grass-bush
Crown-giant
Trunk-crown Methods
Trunk-ground
Specimens.Thefossilsincludedinthisstudyareidentifiedasbelongingtothe
Anolis clade by the presence of one or more of the following derived
Eocene Oligocene Miocene PO PS
character states: coronoid labial blade, slender clavicles, reduction or ab-
50.0 40.0 30.0 20.0 10.0 0.0 senceofthesplenialandangularbones,noribsoncervicalvertebrae3and
4,threesternalribs,noribsonthreeormore(uptoseven)posteriorpre-
Fig. 3. Ancient divergence of anole ecomorphs. Six habitat specialists, sacralvertebrae,transverseprocessesofcaudalvertebraelocatedposterior
called ecomorphs, occur on islands in the Greater Antilles today: twig, tofractureplanes,andexpandedsubdigitalscalesthatformapadunder(at
grass-bush, trunk, trunk-ground, trunk-crown, and crown-giant, named least)theantepenultimatephalangesofdigitsII–V(32).
fortheircharacteristicmicrohabitatsanddifferingmorphology,ecology,
Thirty-eightamberfossilswereavailableforstudy(3previouslydescribedand
andbehavior.AchronogramofAnolislizarddiversification(55)indicates
35new).Toqualifyforinclusionintheecomorphologicalanalysis,thefossil
that Hispaniolan ecomorph clades originated long ago and that most
hadtohaveatleasttwoofthethreedatatypes(cranial,postcranial,lamella
predatetheoriginofHispaniolanamber,whichwasformedatsomepoint
number),or,ifonlyonetypewasavailable,atleasttwotoeswithpreserved
within the range indicated by the amber horizon. Hispaniolan clades
lamellae, or three skeletal elements and an estimated snout-to-vent length
(highlighted) are colored according to ecomorph, with wide lines in-
(SVL)withwhichtosize-correctthedata.Therefore,20amberfossils(TableS1)
dicatingcrowncladesforthosecontainingmorethanonespecies.Clades
wereavailabletocomparewith15speciesrepresentingthemaincladesand
ofnon-Hispaniolanspecieshavebeencollapsed(forafullsummary,see
ecomorphsextantonHispaniola:A.aliniger(trunk-crown,TC),A.chlorocyanus
ref. 56). The Hispaniolan crown-giants are closely related to the crown-
(TC),A.coelestinus(TC),A.singularis(TC),A.brevirostris(trunk,TR),A.distichus
giantsofPuertoRico,aswellassomenon–crown-giants,suggestingthe
(TR),A.marcanoi(trunk-ground,TG),A.cybotes(TG),A.sheplani(twig,TW),
possibilityofanearlieroriginofthecrown-giantconditionthanindicated
A.placidus(TW),A.insolitus(TW),A.bahorucoensis(grass-bush,GB),A.doli-
here.Po,Pliocene;Ps,Pleistocene.
chocephalus(GB),A.hendersoni(GB),andA.olssoni(GB)(100specimens).For
eachextantspecies,malesandfemalesspanningthesizerangefromjuvenileto
adultweresampledfromthealcohol-preservedcollectionattheMuseumof
speciesgroup,whichcontainsextantHispaniolancrown-giants,at17
ComparativeZoology(Cambridge,MA).
My,possiblyyoungerthantheformationofDominicanamber(Fig.
3). By contrast, molecular divergence time estimates indicate that
X-Ray Microcomputed Tomography. The amber fossils and modern species
the other five ecomorphs clearly evolved substantially earlier than
wereexaminedusingX-raymicro-CT.ScansweremadeusingNikon(Metris)
the formationof the amberdeposits(Fig.3). In this light, the ab- X-Tek HMXST 225 machines in Harvard University’s Center for Nanoscale
sence of grass-bush anoles is somewhat surprising. Small and oc-
Systems[amemberoftheNationalNanotechnologyInfrastructureNetwork
curring low in the vegetation, we might have expected them to (NNIN), which is supported by the National Science Foundation] and The
become entrapped in amber. On the other hand, as their name NaturalHistoryMuseum,London,andsupplementedwithdatafromother
implies,theselizardsusuallyarefoundonsurfacesotherthantree systems,detailsofwhicharegiveninTableS1.TheX-raymicro-CTdatawere
trunks, and thus habitat use may account for their absence in processedusingVGStudioMAXv2.2(VolumeGraphicsGmbH).Thedifferent
oursample. elements in the fossils (i.e., bone, amber, air) and modern specimens (i.e.,
Evolutionary interpretations from phylogenetic analyses of bone,softtissue)werereconstructedin3Dbyusingdifferentthresholdson
extantspeciesarealwayshauntedbythespecterofextinction,but theslicesforgrayvaluesspecifictotheseelements.Insomeamberpieces,air-
ourdataprovidelittleevidencefordimensionsofanolediversity filledvoidsoutlinedthebodyofthelizardinclusion(andthesofttissuewas
not suggested by analyses onthe contemporary fauna. Quite the mostlydegraded),makingitpossibletoexaminescalepatterns,particularlyof
contrary,thefossildataandphylogeneticcomparativeanalysesare thetoepads.
entirely congruent in suggesting that the ecomorph radiation is
ancientandthattheecomorphsevolvedearlyandhaveremained BodySizeEstimation.SVL,usuallymeasuredasthedistancefromthetipofthe
in place ever since (9). This is not to say that no adaptive di- snouttotheanteriorendofthevent,iscommonlyusedasaproxyforbodysizein
lizards.Onaskeleton,thedistanceistakenfromtheanterioredgeofthepre-
versification has occurred in anoles since the Miocene. Rather,
maxillaonthemidlinetothearticulationbetweenthesecondsacralvertebraand
there has been considerable subdivision of ecomorph niches as
species specialize for different thermal microhabitats (26)— first caudal vertebra. SVL was measured from 3D-rendered skeletons in all
modernspecimensandonamberfossilswherepossible,usingthepolylinetoolin
unfortunately,thescalationcharactersthattendtocorrelatewith
VGStudioMAXv2.2(VolumeGraphics),whichallowsmoreaccuratemeasure-
thermalbiology(27)werenotpreservedinmostfossils,preventing
mentofdistortedspecimensbyusingcumulativepointspositionedalongaline.
inferencesabouttheevolutionofthermalspecialization.Ourdata
SVLcouldnotbemeasureddirectlyonamberfossilsthataremissingthe
do confirm, however, that the subsequent division of the trunk-
skullorpartsoftheaxialskeleton.Therefore,SVLwasestimatedfromthe
crown anole niche into small and large species, presumably to
lengthofotherbodypartsthatwefoundtohaveminimalvariationamong
partitionpreyresources(21),alsohasanancientorigin,againin ecomorphs using a regression model from the 100 modern specimens,
agreementwithmoleculardivergenceestimates. summarizedinTableS2.Specifically,weusedfourapproaches:
Caribbean islands are a hotspot of biological diversity, and
i) Thelength of a lumbar vertebra (Lumbar), calculated asthe average
researchontheirfloraandfaunahascontributedimmenselytoour
lengthofthelasttwopresacralvertebrae,scalesverystronglywithSVL
understanding of evolutionary processes. The abundance of in-
(r2=0.96):logLumbar=−1.58223+1.0032746×logSVL.
vertebrates preserved in amber from the Caribbean and other
ii) Thelengthoftheilium(Ilium),measuredfromtheposteriortipofthe
geographical regions has led to macroevolutionary and macro-
shaft to the anterior tip of the anterior iliac process, also scales very
ecologicalinsightsaboutsomegroups(28–30),butsuchapproaches stronglywithSVL(r2=0.96):logIlium=−1.48835+1.23261×logSVL.
have not previously been possible for vertebrates. The Caribbean iii) Bodylength(BL),measuredfromtheanterioredgeoftheatlastothe
island Anolis lizards have contributed to our understanding of posterioredgeofthelastsacralvertebramediallyalongtheventralside
evolutionaryprocessesasawell-knowncaseofadaptiveradiation. oftheexternalbodyfromthenarrowestpointofneckregiontothe
4of6 | www.pnas.org/cgi/doi/10.1073/pnas.1506516112 Sherrattetal.
Outgroups insomeamberfossilsfromreconstructedsofttissueusingX-raymicro-CT;la-
0.50 DDactyloa clade mellaeofotheramberfossilsandalcohol-preservedmodernspecimenswere
occultus countedusingstereomicroscopes.Variablesmeasuredforeachamberfossilare
darlingtoni given in Table S2. Cranial shape variation was captured using ten linear
equestris species group
vermiculatus s}pecies group measurements(Fig.S6A):skullwidth,frontalwidth,prefrontalwidth,snout
monticola length,premaxillawidth,jugalheight,orbitdiameter,maxillalength,man-
hendersoni series
coelestinus diblewidth,andmandiblelength.Thesemeasurementsreflectmuchofthe
00..6634JF cranialvariationamongecomorphs(33,34).Postcranialmeasurementsreflect
0.64 G thoseknowntovaryamongecomorphs,theadaptivesignificanceofwhichis
00..5568 HK chlosreorcieysa nus wellunderstood(24,34):lengthsofthehumerus,ulna,femur,tibia,fourth
0.40 N metatarsal, third and fourth toesof the fore- and hindfeet, widths of the
0.68 aIliniger sternum, pelvis and sacrum, and pubis length. Numbers of lamellae were
0.77A countedforthethirdandfourthtoesofthefore-andhindfeet.
chlorocyanus
etheridgei Cranialandpostcranialmeasurements(TableS2)werecorrectedforbody
insolitus sizebyusingtheresidualsfromseparatelinearregressionsoflog-transformed
fowleri
barbouri cranial and postcranial metrics on log-transformed SVL. We evaluated the
semilineatus series probabilityofassignmentofeachfossiltotheecomorphclassesusingDFA.
0.52 M
0.46 O Becauseofvariationinwhichelementswerepreservedinthefossils,DFAs
0.50 P wereperformedseparatelyforeachfossil(TableS3).Ineachcase,aDFAwas
lucius species group
christophei performedusingvariablesfromthemodernspecimens,whichincludedonly
cuvieri thevariablesthatwerealsoavailableforthefossil,andtheassignmentofa
roosevelti
Chamaeleolis clade fossiltoanecomorphwasdeterminedbasedonthissubsetofvariablesfrom
ricordii spec}ies group modernspecimens.DFAwasperformedusingacommon(within-)covariance
0.49 B
marcanoi matrixforallgroups(linearDA)inJMPProv.10(SASInstituteInc.).
strahmi
wlohnigteitmibaianliis csyebroietes s PhylogeneticAnalysis.Wescoredthe20amberfossilsforthemorphological
cybotes charactersusedbyPoe(22)toinferanolephylogeny,usingX-raymicro-CT N
armouri O
shrevei scans,photographs,anddirectobservations.Morphometricmeasurements TI
0.38 E (characters1–9inref.22)werenotincludedinthedatamatrixbecausethe LU
0.46 Cbimaculatus series absence of minimum and maximum values in Poe’s (22) character de- EVO
distichus series scriptions precluded the addition of new information. Data for 13 mor-
0.46 cQristatellus series phologicalcharacters(characters12,55–57,59,65–66,71,74–75,78–79,and
Twig 0.51 S 90inref.22)werenotincludedforfossils,exceptC,theonlylikelyadult,
Trunk 0.53 T
Grass-bush sahluetapclaenui ss eserieriess banecdatuhsuesawtoleualsdtnoontehoafvtehebesteantepsrefosernthtefoserfcohsasrilascotethrsedrethvealnopCs.Tlahteesiendoanttaowgeenrey
Crown-giant angusticeps subgroup
Trunk-crown carolinensis subgroup combinedwithmorphological(22)andmoleculardata[ND2,COI,andRAG1
Trunk-ground 0.57 Largillaceus species group genes,5transferRNAs,andtheoriginforlight-strandreplication(35–45)]for
0.35 R extantspecies.MoleculardatawerealignedusingClustalX(46)andtranslated
Norops clade intoaminoacidsusingMacCladev.4.07(47)toconfirmthecorrecttranslation
frame. Sequence coding for transfer RNAs was aligned manually following
Fig.4. Summarytreeshowingthephylogeneticrelationshipsoftheamber
KumazawaandNishida’s(48)modeloftRNAsecondarystructure.Onehundred
fossilsinferredbyBayesiananalyses.NamesofAnolissubcladesaregivenon
basescorrespondingtosectionsofthe tRNAsandtheoriginoflight-strand
the right. The amber fossils are represented by letters corresponding to
replicationwereexcludedfromtheanalysesduetoambiguousalignment.The
those in Fig. 2, and their relationships are indicated by yellow-orange
finalmatrixanalyzedincluded4,873bases,91morphologicalcharacters,and
branches(seeMethodsfordetails).Bayesianposteriorprobabilitiesforfossil
181extanttaxa(174Anolisand7outgroupspecies).All181specieswerescored
placementsareshownabovebranches.DFAclassificationsforthefossilsare
formorphologicalcharacters;140alsohadmoleculardata.Theresultingdata
indicatedbypiecharts,coloredbyecomorph(onlyprobabilitiesgreaterthan
matrixwasanalyzedusingparsimonyandBayesianmethods.
0.1areshown).
To reduce the computational time that results from large amounts of
missingdataintheamberfossils,weinferredatopologyfortheextantspecies
vent,scalesthemoststronglywithSVL(r2=0.99):logBL=−0.299015+ andthenusedthattopologyasabackbonetopologicalconstraint(whichallows
1.0836932×logSVL. speciesnotincludedintheconstrainttreetobeplacedfreelyinthetopology)to
iv) Mandiblelength(MandL),measuredfromtheposteriormostpointon inferthephylogeneticpositionoftheamberfossils.Thetopologyforthe
theretroarticularprocesstotheanteriormostpartofthedentaryat extantspecieswasinferredwithPAUP*v.4.0b10(49)usingparsimonyonthe
the symphysis, scales very strongly with SVL (r2 = 0.96): logMandL = combined dataset (181 species: 91 morphological characters and 4,873 DNA
−0.521892+1.0051243×logSVL. bases).Weusedequalcostsforstatetransformations,exceptformultistateor-
deredmorphologicalcharacters,whichwereweightedsuchthattherangeof
Formanyofthefossils,SVLwasmeasuredorestimatedbyseveraldifferent each character equals 1. We performed 2,000 replicates of random stepwise
methods(TableS2).TheestimatedSVLusedforsizecorrectionwasassessed addition,withdefaultsettingsforallotheroptions.
on a fossil-by-fossil basis, using the most reliable measurement given the Wethenperformedaseparateparsimonyanalysisforeachamberfossilby
stateofpreservationofthespecimen.Forthemorefragmentaryfossils(K,N, includingeachfossilinthecombinedmatrixfortheextantspecies,imposing
O,P,andS),forwhichapproachesi–iiiabovecouldnotbeused,wealsoprovide thetopologicalconstraint,andusingasinglereplicateofrandomstepwise
a range of SVL estimates (Table S2). The ranges were estimated from the addition(otheroptionsleftondefaultsettings).Wecombinedtheresultsfor
distribution of ecomorph-specific slopes constituting various body measure- allofthefossilsinasinglesummarytree(Fig.S5).Forfossilsforwhichasingle
ments.Forexample,infossilswithisolatedheads(K,N,andO),therangeof most parsimonious tree (MPT) was inferred, the fossil was placed in that
possibleSVLswastakenfromtheecomorphslopeswiththelowestandhighest positioninthesummarytree.ForfossilsforwhichmorethanoneMPTwas
interceptinregressionsofmandiblelength,mandiblewidth,and/orskullwidth. inferred,weusedapruneandregraftconsensusmethod(50),placingeach
ForfossilsPandS,therangeofSVLswerepredictedusingulnalength(r2=0.84; fossilatthebasalnodeoftheleastinclusivecladeinthebackboneconstraint
logUlna=−1.195832+1.205887×logSVL).FossilPhasawell-defineddewlap, treethatincludedallofthealternativeplacementsofthefossil.
indicatingmaturity,andwouldsuggestalargeranimalofanecomorphmadeup ToperformBayesiananalyses,wefirstrescoredthemorphologicalcharacters
predominantlyofspeciesofsmallbodysize,suchastwiganoles(44.7mmusing inthecombinedmatrixintoorderedorunorderedcharactersconsistingof6or10
twig-specificslope).FossilShasunfusedepiphyses,suggestingajuvenile. character states, respectively [following Poe’s (22) character descriptions]. We
thenfollowedasimilarapproachtothatusedintheparsimonyanalyses.First,
Morphometric Analysis. To compare the morphology of extant and fossil weestimatedatopologyofextantspecies(tobeusedastheequivalentofa
anoles,wecollectedcranialandpostcranialmeasurementsfromreconstructed backboneconstraint),byanalyzingtheextantcombinedmatrixinMrBayes
skeletonsusingX-raymicro-CT.Wecountedthenumberofsubdigitallamellae v.3.2.3(51).Weexecutedfourindependentrunsof50milliongenerations
Sherrattetal. PNASEarlyEdition | 5of6
witharandomstartingtree,asamplingfrequencyof1,000generations,and contype=allcompatcommand).Wecombinedthemajorityruletreeresultsof
remainingparametersleftatdefaultvalues.Wepartitioneddatabymor- allfossilsinasinglesummarytree(Fig.4).
phology,codonposition,andtRNAs(11partitions)andestimatedthemodel We used Bayesian topology tests (54) to evaluate support for the
foreachpartitionusingjModeltestv.2.1.4(52).Toconfirmanappropriate placement of individual fossils within extant clades that correspond to
sampling interval and that stationarity was achieved after discarding the thesameecomorph.Topologiesrepresentingeachofthehypothesesof
first25%ofthesampledtrees,weexaminedeffectivesamplesizevalues interest,e.g.,fossilDplacedassistertoinsolitus,wereusedasfiltersin
usingTRACERv1.5(53),comparedtheaverageSDofsplitfrequenciesbe- PAUP*toexaminetopologiescontainedinthe95%crediblesetoftrees
tweenchains,andexaminedthepotentialscalereductionfactor(PSRF)ofall resultingfromtheBayesiananalysis.
of the estimated parameters for the four runs combined. We estimated a
maximumcladecredibility(MCC)treeandcreatedaseriesofpartialconstraints ACKNOWLEDGMENTS. We thank those who allowed us access to amber
sothatwecouldenforcetheMCCtopologyofextantspeciesasastrictconstraint fossilsforstudy:G.Bechly,J.Calbeto,L.Costeur,M.Cusanovich,G.Greco,
M. Greco, D. Grimaldi, M. Halonen, and J. Work. We especially thank
indownstreamanalysesthatincludedfossiltaxa.Weconductedanindependent
E.Moroneforprovidingaccesstomorethanhalfofthefossilsexaminedfor
Bayesiananalysisforeachfossil,usingthesameprotocolandsettingsasthe
thisstudy.WethankJ.Klaczko,andG.Gartnerforprovidingsupportand
extanttopologyanalysis,butwithadurationof15milliongenerations.After discussionsandJ.Casartforassistingwithdatacollection.Scandatawere
discardingthefirst25%ofsampledtrees,weestimatedamajorityrulecon- alsoprovidedbyR.Boistel,M.Polcyn,L.Jacobs,M.Colbert,andR.Ketcham.
sensustreeforeachfossilspecimenincludingallcompatiblegroups(usingthe WethanktheNationalScienceFoundationforsupport.
1. WilliamsJW,ShumanBN,WebbT,III,BartleinPJ,LeducPL(2004)Late-Quaternary 30.RustJ,etal.(2010)Biogeographicandevolutionaryimplicationsofadiversepaleobiotain
vegetationdynamicsinNorthAmerica:Scalingfromtaxatobiomes.EcolMonogr amberfromtheearlyEoceneofIndia.ProcNatlAcadSciUSA107(43):18360–18365.
74(2):309–334. 31.DazaJD,BauerAM,WagnerP,BöhmeW(2013)AreconsiderationofSphaerodactylus
2. LyonsSK(2005)AquantitativemodelforassessingcommunitydynamicsofPleisto- dommeliBöhme,1984(Squamata:Gekkota:Sphaerodactylidae),aMiocenelizardin
cenemammals.AmNat165(6):E168–E185. amber.JZoologicalSystEvolRes51(1):55–63.
3. WilliamsJW,JacksonST(2007)Novelclimates,no-analogcommunities,andecological 32.EtheridgeR,deQueirozK(1988)AphylogenyofIguanidae.PhylogeneticRelationshipsof
surprises.PaleoecolRev5(9):475–482. theLizardFamilies,edsEstesR,PregillG(StanfordUnivPress,Stanford,CA),pp283–367.
4. BloisJL,HadlyEA(2009)MammalianresponsetoCenozoicclimaticchange.AnnuRev 33.SangerTJ,MahlerDL,AbzhanovA,LososJB(2012)Rolesformodularityandconstraintin
EarthPlanetSci37(8):8.1–8.28. theevolutionofcranialdiversityamongAnolislizards.Evolution66(5):1525–1542.
5. OlszewskiTD(2012)Persistenceofhighdiversityinnon-equilibriumecologicalcom- 34.HarmonLJ,KolbeJJ,CheverudJM,LososJB(2005)Convergenceandthemultidi-
munities:Implicationsformodernandfossilecosystems.ProceedingsoftheRoyal mensionalniche.Evolution59(2):409–421.
SocietyofLondonB:BiologicalSciences279(1727):230–236. 35.MaceyJR,LarsonA,AnanjevaNB,PapenfussTJ(1997)Evolutionaryshiftsinthree
6. BrettCE(2012)Coordinatedstasisreconsidered:Aperspectiveatfifteenyears.Earth majorstructuralfeaturesofthemitochondrialgenomeamongiguanianlizards.JMol
andLife,edTalentJA(Springer,NewYork),pp23–36. Evol44(6):660–674.
7. WilliamsEE(1983)Ecomorphs,faunas,islandsize,anddiverseendpointsinisland 36.JackmanTR,IrschickDJ,deQueirozK,LososJB,LarsonA(2002)Molecularphylo-
radiationsofAnolis.LizardEcology:StudiesofaModelOrganism,edsHueyRB, geneticperspectiveonevolutionoflizardsoftheAnolisgrahamiseries.JExpZool
PiankaER,SchoenerTW(HarvardUniversityPress,Cambridge,MA),pp326–370. 294(1):1–16.
8. LososJB,JackmanTR,LarsonA,deQueirozK,Rodríguez-SchettinoL(1998)Contingency 37.JackmanTR,LarsonA,deQueirozK,LososJB(1999)Phylogeneticrelationshipsand
anddeterminisminreplicatedadaptiveradiationsofislandlizards.Science279(5359): tempoofearlydiversificationinAnolislizards.SystBiol48(2):254–285.
2115–2118. 38.CreerDA,deQueirozK,JackmanTR,LososJB,LarsonA(2001)Systematicsofthe
9. Mahler DL, Ingram T, Revell LJ, Losos JB (2013) Exceptional convergence on the AnolisroquetseriesofthesouthernLesserAntilles.JHerpetol35(3):428–441.
macroevolutionarylandscapeinislandlizardradiations.Science341(6143):292–295. 39.GlorRE,KolbeJJ,PowellR,LarsonA,LososJ(2003)Phylogeneticanalysisofecological
10.PoinarGO(2010)PalaeoecologicalperspectivesinDominicanamber.AnnSocEn- andmorphologicaldiversificationinHispaniolantrunk-groundanoles(Anoliscybotes
tomolFr46(1-2):23–52. group).Evolution57(10):2383–2397.
11.PenneyD(2002)PaleoecologyofDominicanamberpreservation:Spider(Araneae)in- 40.HarmonLJ,SchulteJA,2nd,LarsonA,LososJB(2003)Tempoandmodeofevolu-
clusionsdemonstrateabiasforactive,trunk-dwellingfaunas.Paleobiology28(3):389–398. tionaryradiationiniguanianlizards.Science301(5635):961–964.
12.GrimaldiD(2000)StudiesonFossilsinAmber,WithParticularReferencetoCreta- 41.SchulteJA,ValladaresJP,LarsonA(2003)PhylogeneticrelationshipswithinIguanidae
ceousofNewJersey(BackhuysPublishers,Leiden). inferredusingmolecularandmorphologicaldataandaphylogenetictaxonomyof
13.MacPhee RDE, Grimaldi DA (1996) Mammal bones in Dominican amber. Nature iguanianlizards.Herpetologica59(3):399–419.
380(6574):489–490. 42.NicholsonKE,etal.(2005)Mainlandcolonizationbyislandlizards.JBiogeogr32(6):
14.PoinarGO,Jr,CannatellaDC(1987)AnUpperEocenefrogfromtheDominicanRe- 929–938.
publicanditsimplicationforCaribbeanbiogeography.Science237(4819):1215–1216. 43.Nicholson KE, Mijares-Urrutia A, Larson A (2006) Molecular phylogenetics ofthe
15.RieppelO(1980)GreenanoleinDominicanamber.Nature286(5772):486–487. Anolisoncaseries:Acasehistoryinretrogradeevolutionrevisited.JExpZoologBMol
16.DazaJD,BauerAM(2012)Anewamber-embeddedsphaerodactylgeckofromHis- DevEvol306(5):450–459.
paniola,withcommentsonmorphologicalsynapomorphiesoftheSphaerodactylidae. 44.CastigliaR,AnnesiF,BezerraA,GarciaA,Flores-VillelaO(2010)Cytotaxonomyand
Breviora529:1–28. DNA taxonomy of lizards (Squamata, Sauria) from a tropical dry forest in the
17.LazellJD,Jr(1965)AnAnolis(Sauria,Iguanidae)inamber.JPaleontol39(3):379–382. Chamela-Cuixmala Biosphere Reserve on the coast of Jalisco, Mexico. Zootaxa
18.deQueirozK,ChuLR,LososJB(1998)AsecondAnolislizardinDominicanamberand 2508:1–29.
thesystematicsandecologicalmorphologyofDominicanamberanoles.AmMus 45.CastañedaMdR,deQueirozK(2011)PhylogeneticrelationshipsoftheDactyloaclade
Novit3249:1–23. ofAnolislizardsbasedonnuclearandmitochondrialDNAsequencedata.MolPhy-
19.PolcynMJ,RogersJV,II,KobayashiY,JacobsLL(2002)Computedtomographyofan logenetEvol61(3):784–800.
AnolislizardinDominicanamber:systematic,taphonomic,biogeographic,andevo- 46.ThompsonJD,GibsonTJ,PlewniakF,JeanmouginF,HigginsDG(1997)TheCLUSTAL_X
lutionaryimplications.Paleontol.Electron.5(1):1–13. windowsinterface:Flexiblestrategiesformultiplesequencealignmentaidedby
20.Iturralde-VinentMA(2001)Geologyoftheamber-bearingdepositsoftheGreater qualityanalysistools.NucleicAcidsRes25(24):4876–4882.
Antilles.CaribbJSci37(3/4):141–167. 47.MaddisonDR,MaddisonWP(2001)MacClade:AnalysisofPhylogenyandCharacter
21.LososJB(2009)LizardsinanEvolutionaryTree:EcologyandAdaptiveRadiationof Evolution(SinauerAssociates,Sunderland,MA).
Anoles(UnivofCaliforniaPress,Berkeley,CA). 48.KumazawaY,NishidaM(1993)SequenceevolutionofmitochondrialtRNAgenesand
22.PoeS(2004)Phylogenyofanoles.HerpetologicalMonogr18(1):37–89. deep-branchanimalphylogenetics.JMolEvol37(4):380–398.
23.WilliamsEE(1965)ThespeciesofHispaniolangreenanoles(Sauria,Iguanidae).Bre- 49.Swofford DL (2002) PAUP*. Phylogenetic Analysis Using Parsimony (* and Other
viora227:1–16. Methods)(SinauerAssociates,Sunderland,MA).
24.BeuttellK,LososJB(1999)EcologicalmorphologyofCaribbeananoles.Herpetolog 50.FindenC,GordonA(1985)Obtainingcommonprunedtrees.JClassif2(1):255–276.
Monogr13:1–28. 51.RonquistF,etal.(2012)MrBayes3.2:EfficientBayesianphylogeneticinferenceand
25.BurbrinkFT,PyronRA(2010)Howdoesecologicalopportunityinfluenceratesof modelchoiceacrossalargemodelspace.SystBiol61(3):539–542.
speciation, extinction, and morphological diversification in New World ratsnakes 52.DarribaD,TaboadaGL,DoalloR,PosadaD(2012)jModelTest2:Moremodels,new
(tribeLampropeltini)?Evolution64(4):934–943. heuristicsandparallelcomputing.NatMethods9(8):772.
26.HertzPE,etal.(2013)Asynchronousevolutionofphysiologyandmorphologyin 53.RambautA,DrummondAJ(2007)Tracerv1.4.Availableattree.bio.ed.ac.uk/software/
Anolislizards.Evolution67(7):2101–2113. tracer/.AccessedJuly8,2015.
27.WegenerJE,GartnerGEA,LososJB(2014)Lizardscalesinanadaptiveradiation: 54.BrandleyMC,SchmitzA,ReederTW(2005)PartitionedBayesiananalyses,partition
VariationinscalenumberfollowsclimaticandstructuralhabitatdiversityinAnolis choice,andthephylogeneticrelationshipsofscincidlizards.SystBiol54(3):373–390.
lizards.BiolJLinnSocLond113(2):570–579. 55.Pyron RA, Burbrink FT, Wiens JJ (2013) A phylogeny and revised classification
28.WilsonEO(1985)InvasionandextinctionintheWestIndianantfauna:Evidencefrom ofSquamata,including4161speciesoflizardsandsnakes.BMCEvolBiol13(1):93.
theDominicanamber.Science229(4710):265–267. 56.MahlerDL,RevellLJ,GlorRE,LososJB(2010)Ecologicalopportunityandtherateof
29.PenneyD,LanganAM(2006)ComparingamberfossilassemblagesacrosstheCeno- morphologicalevolutioninthediversificationofGreaterAntilleananoles.Evolution
zoic.BiolLett2(2):266–270. 64(9):2731–2745.
6of6 | www.pnas.org/cgi/doi/10.1073/pnas.1506516112 Sherrattetal.
Supporting Information
Sherratt et al. 10.1073/pnas.1506516112
Fig.S1. Toepadmorphologyoffore-andhindfeetamongrecent(representingfiveecomorphs)andfossilanoles.Lamellanumbersforthemodernspeciesare
givenbesidedigitsIIIandIV.TheamberfossilsarecoloredbyecomorphaccordingtoDFAresultsifassignedwithaprobabilitygreaterthan0.90(TableS3).
ModernspecimensfromMuseumofComparativeZoologyReptilecollection,HarvardUniversity(prefixMCZR).
Sherrattetal.www.pnas.org/cgi/content/short/1506516112 1of12
)
s
n nu
wa
oy
rc
co
k-or
unhl
TrA. c
( T-0845 MCZ R-59788
)
s
u
h al F
sh
up
be
-c
so
sc
raoli
Gd
A. MCZ R-74865 MCZ R-124846
(
d )
ns
ue
oot
grb J
-y
kc
unA.
Tr(
T-0339 MCZ R-186752
)
s
u
olit
g s
Twis in
oli O
n
A
(
MCZ R-107016 MCZ R-107014
)
s
ri
st
k o
unvir
Trre
b
A.
(
K
MCZ R-155836 MCZ R-69274
Fig.S2. Cranialshapeoffiverecentanolesrepresentingfiveecomorphscomparedwithfouramberfossils.Craniaofmodernjuveniles(Left)andadults
(Center),aswellasamberfossilskulls(Right)areshowninlateralanddorsalviews.Theamberfossilsarecoloredbyecomorphifassignedwithaprobability
greaterthan0.90(TableS3).ModernspecimensfromMuseumofComparativeZoologyReptilecollection,HarvardUniversity(prefixMCZR),orFieldTagfor
T.J.Sanger(prefixT).
Sherrattetal.www.pnas.org/cgi/content/short/1506516112 2of12
Fig.S3. Variationamongthefossilsandhowitcorrespondswithvariationamongtheecomorphsforall10cranialmeasurements.Size-correctedresidualsare
plottedfortheamberfossilslabeledbyletters(A–T)andmodernspecimens(coloreddotsbyecomorph).AmberfossilsarecoloredaccordingtotheDFAresults
(detailedinTableS3)whenassignedwithaprobabilitygreaterthan0.90.AplotofthefirsttwocanonicalvariablesoftheDFAusingall10variables(Lower
Right)illustrateshowwellthecranialmeasurementsdiscriminatetheecomorphs(modernspecimensonly).
Sherrattetal.www.pnas.org/cgi/content/short/1506516112 3of12
Pelvis width Sacrum width Pubis length
0.15 0.1
0.1
uals)0.00.51 0.05 L 0.050 A
esid 0 A 0 A E -0.05 HJMR
Length (r--00-..010.551 BJ M -0-.00.51 H MR -0--.001..521
-0.2 -0.25
GB TRTCTGTW GB TRTCTGTW GB TRTCTGTW
Amber Ecomorph Amber Ecomorph Amber Ecomorph
Sternum width Humerus Ulna
0.2 0.15
0.15 0.1 0.1
ngth (residuals)--000-..0.1000.55.1510 A J -00-.0.00.5150 A C O J -00-.0.00.5150 ACFGHIJOR
Le -0.2 -0.15 -0.15
-0.25 -0.2
GB TRTCTGTW GB TRTCTGTW GB TRTCTGTW
Amber Ecomorph Amber Ecomorph Amber Ecomorph
Femur Tibia Metatarsal IV
0.1 0.1
h (residuals)-00..00550 ABCDEFGHIJLM --000-..0.100.55150 ABDCEFGHIJLMT -00..110 ABCDEFGHIJLMT
ngt -0.1 -0.2 -0.2
Le-0.15 -0.25
-0.3 -0.3
-0.2
GB TRTCTGTW GB TRTCTGTW GB TRTCTGTW
Amber Ecomorph Amber Ecomorph Amber Ecomorph
Hindtoe IV Hindtoe III Foretoe III
0.1 0.15 0.15
0.05 E M 0.1 0.1
esiduals)-0-.00.510 ABCDFGHIJ T -00..00550 ACDEFHIJLM -00..00550 C HJ O
h (r-0.15 -0.1 -0.1
gt -0.2 -0.15 -0.15
en-0.25 -0.2 -0.2
L
-0.3 -0.25 -0.25
GB TRTCTGTW GB TRTCTGTW GB TRTCTGTW
Amber Ecomorph Amber Ecomorph Amber Ecomorph
6
Foretoe IV DFA, All postcranial variables,
0.1 4 modern only
als) 0.050 CHGJ O 2
gth (residu--00-..010.551 Canonical 2--420
en -0.2
L -6
-0.25
GB TRTCTGTW -8
Amber Ecomorph
-10
-10 -5 0 5 10
Canonical 1
Fig.S4. Variationamongthefossilsandhowitcorrespondswithvariationamongtheecomorphsforall13postcranialmeasurements.Size-correctedresiduals
areplottedfortheamber(A–T)andmodernspecimens.Amberfossilsarecoloredaccordingtothediscriminantfunctionanalysis(DFA)results(TableS3)when
assignedwithaprobabilitygreaterthan0.90.AplotofthefirsttwocanonicalvariablesoftheDFAusingall13variables(LowerRight)illustrateshowwellthe
postcranialmeasurementsdiscriminatetheecomorphs(modernspecimensonly).
Sherrattetal.www.pnas.org/cgi/content/short/1506516112 4of12