A bromeliad species reveals invasive ant presence in urban areas of French Guiana

(1)

HAL Id: hal-02095569

https://hal.umontpellier.fr/hal-02095569

Submitted on 11 Jul 2019

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

A bromeliad species reveals invasive ant presence in urban areas of French Guiana

Stanislas Talaga, Jacques H.C. Delabie, Olivier Dézerald, Alex Salas-Lopez, Frédéric Petitclerc, Céline Leroy, Bruno Hérault, Régis Céréghino, Alain

Dejean

To cite this version:

Stanislas Talaga, Jacques H.C. Delabie, Olivier Dézerald, Alex Salas-Lopez, Frédéric Petitclerc, et

al.. A bromeliad species reveals invasive ant presence in urban areas of French Guiana. Ecological

Indicators, Elsevier, 2015, 58, pp.1-7. �10.1016/j.ecolind.2015.05.027�. �hal-02095569�

(2)

ContentslistsavailableatScienceDirect

Ecological Indicators

jou rn al h om ep a g e :w w w . e l s e v i e r . c o m / l o c a t e / e c o l i n d

A bromeliad species reveals invasive ant presence in urban areas of French Guiana

Stanislas Talaga

^a,∗

, Jacques H.C. Delabie

^b

, Olivier Dézerald

^a

, Alex Salas-Lopez

^a

,

Frédéric Petitclerc

^a

, Céline Leroy

^c

, Bruno Hérault

^d

, Régis Céréghino

^e,f

, Alain Dejean

^a,e,f

aCNRS,ÉcologiedesForêtsdeGuyane(UMR-CNRS8172),Campusagronomique,BP316,97387Kouroucedex,France

bU.P.A.LaboratóriodeMirmecologia,ConvênioUESC/CEPLAC,C.P.7,45600-970Itabuna,Bahia,Brazil

cIRD,UMRAMAP(botAniqueetbioinforMatiquedel’ArchitecturedesPlantes),BoulevarddelaLironde,TAA-51/PS2,F-34398MontpellierCedex5,France

dCIRAD,ÉcologiedesForêtsdeGuyane(UMR-CIRAD93),Campusagronomique,BP316,97387Kouroucedex,France

eUniversitédeToulouse,UPS,INP,LaboratoireEcologieFonctionnelleetEnvironnement,118routedeNarbonne,31062Toulouse,France

fCNRS,Ecolab(UMR-CNRS5245),118RoutedeNarbonne,31062Toulouse,France

a r t i c l e i n f o

Articlehistory:

Received29November2014

Receivedinrevisedform18March2015 Accepted11May2015

Keywords:

Aechmeaaquilega Bioindicators

Non-speciﬁcant–plantrelationships Surrogacy

Tankbromeliads Urbanecology

a b s t r a c t

Tankbromeliads,frequentlyassociatedwithants,areconsidered‘biodiversityampliﬁers’forbothaquatic andterrestrialorganisms,andthushaveahighecologicalvalue.Thefocalspeciesofthisstudy,Aechmea aquilega,shelteredthecoloniesof12antspeciesinaGuianeseruralhabitatwhereOdontomachushaema- todus,associatedwith60%oftheseplants,wasthemostfrequent.Unexpectedly,theantspeciesrichness washigherinacomparedurbanhabitatwith21species,buttwosynanthropicandfourinvasiveants werenotedamongthem.Consequently,weconductedbaitingsurveys(ontheground,ontreesandon treesbearingA.aquilega)aswellascomplementarysurveysusingdifferentsamplingmodesinurban areastotestifA.aquilegaisasurrogaterevealingthepresenceofcertaininvasiveants.Duringthebait- ingsurvey,werecordedfourNeotropicalandeightintroducedinvasiveantsoutofatotalof69species.

Ofthese12invasivespecies,fivewerenotedbybaitingA.aquilega(includingtwoonlynotedinthis way).AbootstrapsimulationpermittedustoconcludethatA.aquilegasignificantlyconcentratescertain speciesofinvasiveants.Thiswasconfirmedbycomplementarysurveys,wherewedidnotrecordfurther species.Weconcludethatbaitingontreesbearinglargeepiphytesinhuman-modified,Neotropicalareas isarelevantcomplementtotheearlydetectionofinvasiveants.

1. Introduction

Globaltradehasgreatlycontributedtothedispersalofplants andanimalswhoseintroductionintonewhabitatsgenerallyresults in their death or a low rate of survival as they are unable to adapt.Still,someofthemhavebecomeinvasive,constitutingan importantthreattobiodiversityintheirintroducedrangebecause theycaneliminatenativespeciesthroughresourcepre-emption and/ordirectcompetition,and,so,disruptecosystem functions.

Thishaseconomicrepercussionsduetothecostsofcontrolmea- sures(ClaveroandGarcia-Berthou,2005;ShogrenandTschirhart, 2005).

Ants are among the mostwidespread and harmful invasive taxa becausethey occupya central place in thefunctioning of

∗Correspondingauthor.Tel.:+33594594320643;fax:+33594594356522.

E-mailaddress:stantalaga@free.fr(S.Talaga).

ecosystems.Thisisduetotheirabundanceastheyconstituteoneof thelargestfractionsoftheanimalbiomassandplaydifferentroles infoodwebssincetheycanbeherbivores,generalists,scavengers orpredators.Innaturalconditions,antscoexistinwell-organized communitiesregulatedbycompetitionandpredationatboththe intra-andinter-speciﬁclevel.Yet,amonginvasiveants,so-called

‘unicolonialspecies’formspatiallyvastandcompetitivelydomi- nantsupercoloniesoverlargegeographicaldistances(Holwayetal., 2002;Moffett,2012).Thankstotheirhugecolonies,theylowerthe speciesrichnessandabundanceofnativeantsthroughexploitation andinterferencecompetitionsothattheycandisruptthearthro- podcommunitystructurewithsubsequentrepercussionsonthe entireecosystem(Holwayetal.,2002).

Therefore,amajorchallengeistodeveloppredictivemanage- mentstrategiesbasedonunderstandingtheprocessesbehindthese invasions.Theperiodjustfollowingtheintroductionofapoten- tially invasive species is central to that understanding. This is particularlytrue for insects which, due totheir small size,are http://dx.doi.org/10.1016/j.ecolind.2015.05.027

(3)

2 S.Talagaetal./EcologicalIndicators58(2015)1–7

difﬁculttodetect,sothattheirpresenceisgenerallydiscovered oncetheinvasiveprocessisalreadywellunderwayorcompleted.

Becausemostinvasivetaxa,includingants,areﬁrstintroducedinto urbanareasthroughmaritimeharborsorairportsandalongroads (Holwayetal.,2002;Lauranceetal.,2009),theearlydetectionof antsincitiesisacriticalcomponentofenvironmentalmanagement.

Yet,theecologicalrequirementsofsomepotentiallyinvasiveants meansthattheiroccurrencecanbemissedwhenresearchersusea conventionalsamplingtechnique,especiallywhentheseantsﬁrst congregateincertainhabitatswhere,infact,theycanbelikenedto the‘TrojanHorse’.

Because epiphytes abound in the trees that grow in some Neotropicalcitiesandbecauseepiphytesfavortheinstallationof numerousantspecies(DavidsonandEpstein,1989;Dejeanetal., 1995;Blüthgenetal.,2000),wehypothesizedthatlargeepiphytic plantswhichcommonlyformhabitatsforinsectscanbe“surro- gates”for theoccurrenceof invasiveantsprior totheirspread.

“Surrogacy”,orhoweasilyrecordedtaxapredictthepresenceof othertaxa,canconstituteausefultoolinconservationplanning (Warmanetal.,2004)andcouldalsobeusedtodetecttherecent introductionof potentially invasive ants (see also Addison and Samways,2006,forartiﬁcialsurrogatehabitats).

Tankbromeliads(Bromeliaceae)arefloweringplantscomprised of59generaandsome3140speciesnativemainlytotheNeotrop- ics(Givnishetal.,2011).Theinterlockingleavesoftankbromeliads formwellsthatcollectrainwater(fromafewmilliliterstoafew liters),leaflitterandotherorganicdetritus.Therosettesofthese plantspermitnumerousopportunisticantspeciestoprofitfrom themoisthabitat.Mostant-bromeliadassociationsarenotspecies- specific(Blüthgenetal.,2000),butspecializedassociationsdoexist (Dejeanet al.,1995).In this context,theaimofthis studywas toevaluatewhethertankbromeliadscanrevealantinvasionsin Neotropicalcities.Tankbromeliadsarefrequentlyfoundinhuman- modified,Neotropicalareas,particularlywhenthoseareasareclose totheseaside,rivers,orlakes,orsituatedathigh,humidaltitudes (Richardsonetal.,2000;SerramoLopez etal.,2009;Cach-Pérez etal.,2013;hereAechmeaaquilega).So,wefirstsoughttoassess whethertheypointtothepresenceofsynanthropicandinvasive antsinurbanizedareasofFrenchGuiana,allsituated alongthe coast.Second,wedeterminediftheseantspeciesareoutcompeted bynativespeciesinasurroundingruralarea.Third,welookedfor newapproachespermittingustoimprovetheconventionalbait- ingsamplingtechniquesoastodetectaseffectivelyaspossiblethe presenceofinvasiveantsbyextendingthesurveytoincludetrees, particularlythosebearinganA.aquilegacluster.Furthersurveys usingdifferentsamplingtechniquesroundedoutthisapproach.

2. Materialsandmethods

2.1. Thefocaltaxa

A.aquilega,foundfromCostaRicatoBrazil,usuallyformsmas- siveclumps of epiphytes on old trees, but can also grow as a geophyte.This largespecies (60–120-cm inheight; Mori et al., 1997)hastightlyinterlockingleavesthatformahighlycompart- mentedrosettecreatingatank(orphytotelm)thatcollectswater andorganicdetritusandprovidesahabitatforaquaticmicro-and macro-organisms(Carriasetal.,2014).

2.2. Studyareasandﬁeldsurveys 2.2.1. AntsassociatedwithA.aquilega

Between2011and2013,westudiedtheantsassociatedwith45 matureA.aquilegainanurbanarea(eachrandomlyselectedinacity blockofSinnamary,FrenchGuiana;05^◦2239N52^◦5735W),and

26othersinaruralhabitatsituated5kmaway,along6kmofadirt road(routedel’Anse)linedwith10–30m-talltrees.Inbothareas,A.

aquilegaaboundasepiphytes,withmostindividualsgrowingata heightof3–6mondifferenttreespecies.Thesamplingareaineach environmentextendedoverasurfaceofca.45ha.

WeusedaladdertoreachtheselectedA.aquilegaindividuals andremoved themfromtheirsubstratesusinga hacksawand thenplacedeachofthemintoaplasticbagwhichwassealedto avoidcontaminationduringtransporttothelaboratory.There,each individualwascarefullyinspectedandtakenapart;eachleafwas tornfromthebasestartingfromtheoutermostleafandworking inward,allowingustocollectentireantcoloniesinstalledbetween theleavesamongthedetritusaccumulatedbytheplant.

2.2.2. BaitingsurveytotestifA.aquilegaisasurrogatehelpfulin detectinginvasiveants

Todetect as effectively as possiblethepresence of invasive ants,weusedaconventionalsamplingtechniqueconsistingofbait- ingantswith,eachtime,aseriesoftwo2-mlEppendorfcolorless microtubes(onecontainingpiecesofcottonimbibedwithdiluted honeyandtheothercontainingpiecesofcannedsardinesinoil).

Eachtime,after30andthen60min,theantsoccupyingthebaits andthosepatrollingallaroundwerecollectedusinganaspirator.

The survey was conducted in Cayenne (4^◦5559N; 52^◦19 59W),Kourou(05^◦0930N;52^◦3834W)and Sinnamary,three citiesinthelittoralzoneofFrenchGuiana(whitesanddeposits) andsituatedalongariver.Lessthan95km(asthecrowﬂies)sepa- rateCayennefromSinnamary,withKouroulocatedalmostmidway.

First,weplaced30pairsofbaitsonthegroundseparatedbyan intervalofmorethan20minCayenne(theairport;theheliport;

thetarmacoftheharbor;themarina;andatthebaseofthebuild- ingoftheCustomsOfﬁceintheharbor),indifferentcityblocksof theVieuxBourgofKourou(areasnotfarfromthemarina)andSin- namary(indifferentcityblocks;morethan100mfromeachother).

Second,inKourouandSinnamary,weusedthesametypeofsam- plingtechnique,placingthebaitsatmorethan2minheighton thetrunksof30talltrees(>15m),mostlymango(Mangiferaindica) andmombin(Spondiasmombin),bothAnacardiaceae,andIngaspp.

(Mimosoideae).Thesetreeswerechosenhaphazardlyindifferent cityblocks.Third,toverifytheabilityofA.aquilegatoconcentrate invasiveantspecies,thesamesurveywasconductedinSinnamary on30otherlargetreesbearinganA.aquilegacluster(almostall weremangotrees)andsituatedindifferentcityblocksthanthe previoustreesorindifferentdistantparcelsofthesameblock(with adistanceofmorethan75mbetweentwotrees).Acomplemen- tarysurveywasconductedinthethreecities.Weﬁrstlyprospected bysightalongthegrassyroadsidesofdifferentcityblocks,totaliz- ingineachcitymorethan5km.Onlytheworkersofinfrequent speciesweregathered(usinganaspirator)forfurtheridentiﬁca- tion.Second,wesampledantsfrom20plots (6m×3m;18m²) situatedingrassyareasindifferentcityblocks.Samplingwasstan- dardizedbyspendingoneman-hourperplotcarefullysearchingfor antsinallsuitablemicrohabitats:theleaflitterincludingallhol- low,rottentwigs;deadwood;humusandthebareground.Third, weconductedabaitingsurveyontreessmallerthanintheprevious survey(i.e.,5–12minheight).Also,intheVieuxBourginKourou,we baitedtheonlyninereachablemangotreesbearinganA.aquilega cluster.

Voucherspecimensoftheantswereidentiﬁedanddepositedin theLaboratóriodeMirmecologiacollection(acronym:CPDC),Cocoa ResearchCentre(Ilhéus,Bahia,Brazil).

2.3. Statisticalcomparisons

WecomparedtheantdiversityinthetwohabitatsusingShan- non’sdiversityt-test(PaSTsoftware;diversitystatistics),andthe

(4)

Fig.1.ComparisonofthepercentagesofoccurrenceofantspeciesassociatedwithAechmeaaquilegainurbanversusruralhabitatsintheregionofSinnamary,FrenchGuiana.

*:synanthropicspecies;§:introducedinvasivespecies;**:Neotropicalinvasivespecies.

antcommunitiesusingthePermutationalMultivariateAnalysisof Variance(PERMANOVA;Rsoftware;RDevelopmentCoreTeam, 2013).

Theaimofthebaitingsurveywastotest therelativecontri- butionofeach of the10 baitingsituations tothetotal number of invasive antspecies recorded in thearea. Using Rsoftware, weconducteda bootstrapsimulation wherewe“sampled”with replacementsthe10baitingsituations.Fromthesesimulations,we countedthenumberofspecies,andtheprocedurewasrepeated 10,000times.Wethensplitthe10,000simulationsintotwogroups, onewheretheA.aquilegasituationwassampledatleastonceand asecondwhereitwasnotsampledatall.Finally,wetestedthe differencesinspeciesnumberbetweenthesetwogroupswitha non-parametricWilcoxonranksumtestwiththecontinuitycor- rection.

3. Results

3.1. ComparingtheantfaunaassociatedwithA.aquilega betweenaruralandanurbanarea

Inallcases,wenotedthepresenceofantsbetweenthebaseof theA.aquilegaandthebarkofthehosttrees.Atotalof56outof the71A.aquilegasampled(79%)sheltered,betweentheirleaves, antcoloniesbelongingto28speciesandﬁvesubfamilies(Fig.1and SupplementaryAppendixA).

A comparisonof thepercentages of occurrenceof these ant speciesillustratedthedifferenceintheirdistributionbetweenthe ruralandurbanareas(12versus21species,respectively;Fig.1).

Indeed,thedifferenceinantspeciesrichnesswassigniﬁcant(Shan- non’s diversity t-test: t=3.66; df=280; P=0.0003) and the ant assemblageswere signiﬁcantly differentbetween thetwo sites (PERMANOVA:N=54;F=15.05;P=0.0001).

Odontomachushaematoduswasbyfarthemostfrequentspecies intheruralareawhereitoccupied60%ofallofthesampledplants shelteringants.Wenotedthatelementarynests(thecoloniesare composedof multiplenests)ofthis specieswereinstalled both betweentheleavesofseveraladjacentA.aquilegaindividualsand betweentheirshootsandthebarkofthehosttrees.

3.2. BaitingsurveytotestifA.aquilegaisasurrogatehelpfulin detectinginvasiveants

Duringthebaitingsurvey,werecorded69antspecies,including 12wellknowninvasivespecies,fourofwhichareNeotropicaland

areinvasiveinhuman-disturbedareasoftheirnativerangeand elsewherewheretheywereintroducedthroughhumantransport;

the eight others were introduced (Table 1 and Supplementary Appendix A). Among theintroduced invasive ants, ﬁve species were recorded during baitingon theground (i.e., Monomorium ﬂoricola; Paratrechina longicornis; Tapinoma melanocephalum;

Tetramorium lanuginosum;and Tetramoriumsimillimum); twoof them,M.ﬂoricolaandTa.melanocephalum,werealsorecordedon trees.Cardiocondylaobscuriorwasonlydetectedthroughbaiting ontrees,andtwomorespecies,PheidolemegacephalaandTech- nomyrmexvitiensis,onlybybaitingontreesbearinganA.aquilega cluster;thelattertreesshelteredﬁveinvasivespecies,fourofthem introduced(Table1).

Furthermore,theprobabilityofdetectinganintroducedinvasive antspecieswasparticularlyhighontreesbearinganA.aquilega clustercomparedtotheothersituations.Eachtimethe“A.aquilega situation”wasincludedinthesimulatedsample,thenumberof invasivespecieswassigniﬁcantlyhigherthansimulationswhere the“A.aquilegasituation”wasnotincluded(Fig.2;Wilcoxonrank sumtest:W=2,251,583;N=10,000;P<2.2e−16).

ThecomplementarysurveypermittedustoagainnotePa.longi- cornisandTa.melanocephalumamongtheintroducedinvasiveants, plusatrampspecies,Cardiocondylawroughtoni(Seifert,2003).

4. Discussion

4.1. ComparingtheantfaunaassociatedwithA.aquilega betweenaruralandanurbanarea

Overall,onecannoteaconvergencebetweentheantassemblage foundintheruralareastudiedandthatofVenezuelantankbromeli- ads(Fig.1;Blüthgenetal.,2000)asO.haematoduswasthemost frequentspecies,Cyphomyrmexspp.abounded,andDolichoderus laminatus,Neoponeravillosa,andspeciesofthegeneraNylanderia, PheidoleandPlatythyreawerealsorecorded.ThedominanceofO.

haematodusisreminiscentoftheN.villosaassociationwithAechmea bracteatainMexico(Dejean,1990).Becausethesetwoponerineant speciescanalsonestinrottenlogsorbrancheslyingontheground, aconvergencelikelyexistswhereakindof“localtradition”causes thequeenstoselectthesebromeliadsasnestingsites.Thistypeof localtraditionhasalsobeennotedinOdontomachushastatusand socialwasps(Wenzel,1996;Dejeanetal.,1998;Gibernauetal., 2007), experimentalstudieshavingshownthat suchhostplant selectionis relatedtoanimprintingprocess(Djiéto-Lordonand Dejean,1999).

(5)

Table1

ComparisonofantspeciesrecordedinthreeFrenchGuianesecitiesusingbaits(30pairsofbaitseachtime).*:synanthropicspecies;

‡:introducedtrampspecies(notinvasive);§:introducedinvasivespecies;**:Neotropicalinvasivespecies;#:twoassociatedant species(parabiosis)whichbuildant-gardenscontainingseveralepiphytes,includingthetankbromeliadAechmeamertensii;†:

baitsontreesbearingAechmeaaquilegainKourou;total:69antspecies.

(6)

Fig.2. Comparisonofthesimulatednumberofinvasiveantspeciesbetweenthe10baitingsituations.Aresamplingwithreplacementsprocedurewasconductedresulting inthe10situationsrepresentedbytheviolinplots(Sarkar,2008)aroundboxplotswhichindicatethemedian(diamond),the25thand75thpercentiles(whitesquares),and theminimumandmaximumvalues(dottedlines).Foragivensituation,wethushadthespeciesnumberdistribution,eachvaluecorrespondingtoasimulationwherethis situationwasincludedinthe10baitingsituationssampled.Eachtimethe“Aechmeaaquilegasituation”wasincludedinthesimulatedsample,thenumberofinvasivespecies wassigniﬁcantlyhigherthansimulationswherethe“A.aquilegasituation”wasnotincluded(Wilcoxonranksumtest:W=2,251,583;N=10,000,P<2.2e−16).K.:Kourou;

C.:Cayenne;S.:Sinnamary;bromeliad:treesbearinganA.aquilegabromeliad:largetreesshelteringAechmeaaquilegaclusters.

The much lower level of O. haematodus occurrence in the urban area is likely due to anthropogenic disturbances which placednativeantsatadisadvantagewhilefavoringsynanthropic species,particularlyinvasive species(Fig.1 andSupplementary Appendix A).The latter likely lowerednative antrichness and abundance through exploitation and interference competition (see Holwayet al.,2002).Note that O. haematodus colonies,as wasobserved here, can develop in cities, even nesting in wall crevices or under ﬂower pots (Delabie et al., 1995). Also, this antspecies was recently introducedinto the GulfCoast of the USA (MacGown et al., 2014). Note that in the rural area only Tec. vitiensis was recorded as an A. aquilega associate among theknowninvasiveants,indicating thebeginningof apossible spread.

Threeofthespeciesrecorded,namelyWasmanniaauropunctata (Neotropical),Tec.vitiensisandPh.megacephala(bothofwhichare introducedspecies),areinvasiveantsparticularlyknownfortheir

negativeecologicalandagriculturalimpact(Holwayetal.,2002;

Bolton,2007;Delabieetal.,2011;Wetterer,2012).

4.2. Baitingsurvey

First, ground baiting permitted us to gather four invasive Neotropicalspecies.Wasmanniaauropunctataisinvasiveinmany tropicalareaswhereitwasintroduced;thisspecies,alsonoteddur- ingbaitingontreesincludingthosebearingA.aquilega,isknown tospreadinhuman-disturbedareasofitsnativerangewhereit can displaceotherants (Orivelet al.,2009).Nylanderia fulva, a Formicinae nativeto South America, has invadedthe southern USAwhere itis capableof detoxifyingthevenom oftheintro- ducedﬁre-ant,Solenopsisinvicta (LeBrun etal.,2014).Native to CentralandSouthAmerica,Solenopsisgeminataisnowwidespread in theTropics(Holway etal., 2002).Also, Solenopsissaevissima, with supercolonies extending over dozens of kilometers, has

(7)

negativeecologicalimpactsandhasbeennotedoutsideitsnative range(Taber,2000;Wetterer,2014a;Lenoiretal.,2015).

Second,duringgroundbaiting,werecordedthefivefollowing introducedinvasivespecies.Monomoriumfloricola,alsonotedon treesincludingthose bearingA. aquilega,isa widespreadarbo- realspecies nativeto tropical Asia which can bea house pest (Wetterer,2010a)oraffectthestructureofarborealantassem- blages (Conceiçãoet al., 2014).Paratrechina longicornis and Ta.

melanocephalum(thelatterwasalsonotedontreesbearingA.aquilega),whichhavespreadworldwideincludingintocoldregionsdue totheirabilitytoliveinhouses,arenativetotheOldWorldTrop- ics(Wetterer,2008,2009).Tetramoriumlanuginosum,whichwas alsonotedontrees,andTet.simillimumarewidespreadinvasive species,theformerisnativetotropicalEastAsia,thelattertoAfrica (WettererandWetterer,2004;Wetterer,2010b).Thisistheﬁrst reportofTet.lanuginosumincontinentalSouthAmericaalthough thisspecieshasbeenreportedinCentralandNorthAmerica,the GalapagosandseveralCaribbeanislands(Wetterer,2010b).

Third,Ca.obscuriorwasnotedonlyduringtreebaiting(Table1).

NativetoSoutheastAsia,thisspeciesiswidespreadintheTropics andSubtropics(Seifert,2003).NotethattheotherCardiocondyla speciesrecorded,Ca.minutiorandCa.wroughtoni(Table1andSup- plementaryAppendixB),bothnativetotheIndo-Malayanregion, aretrampspecieswidespreadintheTropicsandSubtropicsbutnot yetreportedasinvasive(Seifert,2003;Wetterer,2014b).

Fourth,Tec.vitiensisandPh.megacephalawereonlyfoundon treesbearingA.aquilega.AlreadynotedintheFrenchGuianesefor- est,Tc.vitiensisisawide-ranginginvasivespecieslikelynativeto SoutheastAsiawhosecolonygrowthisfavoredbycolonyﬁssion intomanynestsandbythepresenceofreproductiveworker-queen intercasts in addition to typical queens (Bolton, 2007; Delabie etal.,2011).NativetoAfrica,Ph.megacephalaisoneofthemost widespreadinvasiveants;itshugesupercolonieshaveamassively devastatingeffectonthenativeinsectfauna(Holwayetal.,2002;

Wetterer,2012).

Therefore,A.aquilegabaitingprovidesnewinformationonthe presenceofinvasiveantspeciesastwosuchspecieswerefound onlyinthiswayinthepresentstudy.Furthermore,ofthe12inva- sivespeciesnotedintotal,ﬁvewerecollectedbybaitingA.aquilega andwereatadensityhighenoughtoshowthatthereisasignif- icantconcentrationofthesespeciesinassociationwiththisplant (Fig.2).

Onecannotethatmanyinvasiveantsareground-nesters,sothat A.aquilegamightbeconsideredinadequateforshelteringcolonies ofsuchspecies.Yet,thisisnotthecase;forexample,Ph.megacephala,whichismostlyaground-nester,wasnotedonlybybaiting A. aquilega and not in the othersituations tested (Table 1 and AppendicesAandB).Thismightbeduetothefactthatthreenative invasivespecies,S.saevissima,S.geminataandW.auropunctata,are presentinallhuman-disturbedareas,renderingitdifﬁcultforan importedspecies(inthiscase,Ph.megacephala)toestablishitself.

Thefewnativespeciesabletonestintheirpresencehaveadapted tothesituationthroughtheuseofsubmissivebehaviorsanddefen- sivechemicals(Grangieretal.,2007;Rouxetal.,2013;Dejeanetal., 2015).Nevertheless,Pa.longicornisandTa.melanocephalaavoidthe pressurefromthesethreenative,invasivespeciesthankstotheir ecologicalpreferencesastheymostlynestindirectcontactorinside humanconstructions(Wetterer,2008,2009).

5. Conclusion

This survey, mostly conducted in cities where introduced invasiveantsareknowntonestbeforespreadingtoslightlyhuman- disturbedandthenundisturbednaturalareas(Holwayetal.,2002), hashighlightedthecomplementaritybetweenbaitingantsonthe

ground and onurban trees, particularly those bearing A. aquilega clusters. BecauseA. aquilega ‘concentrates’ certaininvasive ants,thisglobalstrategywilllikelypermitresearcherstorecord asexhaustivelyaspossibleintroducedinvasiveantsintheareas studiedbeforetheseantsspread.Itappears,therefore,thatbaiting surveysaimingtodetectinvasiveantpresenceinNeotropicalcities needtotakeintoaccountthecomplementaritybetweendetecting invasiveantspeciespresenceonthegroundandontrees,particu- larlythosebearinglargetankbromeliadsandotherepiphytes.

Acknowledgments

WearegratefultoAndreaYockey-Dejeanforproofreadingthe manuscript, Alexis Carteron and Frédéric Azémar for technical help, theLaboratoire Environnement dePetit Saut for furnishing logisticalassistance,andthemunicipalityofSinnamary(through theDepartmentof theEnvironment)forpermittingustowork inside thecitylimits. Financialsupport forthis studywaspro- videdbyan‘Investissementd’Avenir’grantmanagedbytheFrench AgenceNationaledelaRecherche(CEBA,ref.ANR-10-LABX-0025).

ST’sfinancialsupportwasprovidedbyaPhDfellowshipfromthe UniversitéAntilles-Guyane,ASLandOD’sfinancialsupportwaspro- videdbyaPhDfellowshipfromtheFrenchCentreNationaldela RechercheScientifique(CNRS)andtheFondSocialEuropéen(FSE).

JHCDacknowledgeshisresearchgrantfromCNPq.

AppendixA. Supplementarydata

Supplementarydataassociatedwiththisarticlecanbefound, intheonlineversion,athttp://dx.doi.org/10.1016/j.ecolind.2015.

05.027

References

Addison,P.,Samways,M.J.,2006.Surrogate habitatsdemonstratetheinvasion potentialoftheAfricanpugnaciousant.Biodivers.Conserv.15,411–428.

Blüthgen,N.,Verhaagh,M.,Goitía,W.,Blüthgen,N.,2000.Antnestsintankbromeli- ads.Anexampleofnon-speciﬁcinteraction.Insect.Soc.47,313–316.

Bolton,B.,2007.TaxonomyofthedolichoderineantgenusTechnomyrmexMayr (Hymenoptera,Formicidae)basedontheworkercaste.Contr.Am.Entomol.Inst.

35,1–150.

Cach-Pérez,M.J.,Andrade,J.L.,Chilpa-Galván,N.,Tamayo-Chim,M.,Orellana,R., Reyes-García,C.,2013.Climaticandstructuralfactorsinﬂuencingepiphytic bromeliadcommunityassemblagealongagradientofwater-limitedenviron- mentsintheYucatanPeninsula,Mexico.Trop.Conserv.Sci.6,283–302.

Carrias,J.F.,Céréghino,R.,Brouard,O.,Pélozuelo,L.,Dejean,A.,Couté,A.,Corbara,B., Leroy,C.,2014.Twocoexistingtank-bromeliadshostdistinctalgalcommunities onatropicalinselberg.PlantBiol.16,997–1004.

Clavero,M.,Garcia-Berthou,E.,2005.Invasivespeciesarealeadingcauseofanimal extinctions.TrendsEcol.Evol.20,110.

Conceic¸ão,E.S.,Delabie,J.H.C.,DellaLucia,T.M.C.,CostaNeto,A.O.,Majer,J.D.,2014.

Structuralchangesinarborealantassemblages(Hymenoptera:Formicidae)in anagesequenceofcocoaplantationsinthesouth-eastofBahia,Brazil.Aust.

Ecol.,http://dx.doi.org/10.1111/aen.12128

Davidson,D.W.,Epstein,W.W.,1989.Epiphyticassociationswithants.In:Lüttge, U.(Ed.),VascularPlantsasEpiphytes.Springer-Verlag,NewYork,pp.200–233, 270.

Dejean,A.,1990.Inﬂuencedel’environnementpréimaginaletprécocedanslechoix dusitedenidiﬁcationdePachycondyla(=Neoponera)villosaFabr.(Formicidae, Ponerinae).Behav.Proc.21,107–125.

Dejean,A.,Corbara,B.,Carpenter,J.M.,1998.Nestingsiteselectionbywaspsinthe Guianeserainforest.Insect.Soc.45,33–41.

Dejean,A.,Corbara,B.,Céréghino,R.,Leponce,M.,Roux,O.,Rossi,V.,Delabie,J.H.C., Compin,A.,2015.Traitsallowingsomeantspeciestonestsyntopicallywiththe ﬁreantSolenopsissaevissimainitsnativerange.InsectSci.,http://dx.doi.org/10.

1111/1744-7917.12078

Dejean,A.,Olmsted,I.,Snelling,R.R.,1995.Tree-epiphyte-antrelationshipsinthe lowinundatedforestofSianKa’anBiosphereReserve,QuintanaRoo,Mexico.

Biotropica,57–70.

Delabie,J.H.C.,Groc,S.,Dejean,A.,2011.ThetrampantTechnomyrmexvitiensis (Hymenoptera:Formicidae:Dolichoderinae)onSouthAmerica.Fla.Entomol.

94,688–689.

Delabie,J.H.C., Nascimento,I.C., Pacheco,P., Casimiro,A.B.,1995. Community structureof house-infestingants in southernBahia,Brazil (Hymenoptera;

Formicidae).Fla.Entomol.78,264–270.

(8)

Djiéto-Lordon,C.,Dejean,A.,1999.Tropicalarborealantmosaic:innateattraction andimprintingdeterminenestingsiteselectionindominantants.Behav.Ecol.

Sociobiol.45,219–225.

Gibernau,M.,Orivel,J.,Delabie,J.H.C.,Barabé,D.,Dejean,A.,2007.Anasymmetri- calrelationshipbetweenanarborealponerineantandatrash-basketepiphyte (Araceae).Biol.J.Linn.Soc.91,341–346.

Givnish,T.J.,Barfuss,M.H.J.,VanEe,B.,Riina,R.,Schulte,K.,Horres,R.,Gonsiska, P.A.,Jabaily,R.S.,Crayn,D.M.,Smith,J.A.C.,Winter,K.,Brown,G.K.,Evans,T.M., Holst,B.K.,Luther,H.,Till,W.,Zizka,G.,Berry,P.E.,Sytsma,K.J.,2011.Adaptive radiationanddiversiﬁcationinBromeliaceae:insightsfroma7-locusplastid phylogeny.Am.J.Bot.98,872–895.

Grangier,J.,LeBreton,J.,Dejean,A.,Orivel,J.,2007.Coexistencebetweendominant populationsofWasmanniaauropunctataandCyphomyrmexants.Behav.Proc.74, 93–96.

Holway,D.A.,Lach,L.,Suarez,A.V.,Tsutui,N.D.,Case,T.J.,2002.Thecausesand consequencesofantinvasions.Annu.Rev.Ecol.Syst.33,181–233.

Laurance,W.F.,Goosem,M.,Laurance,S.G.W.,2009.Impactsofroadsandlinear clearingsontropicalforests.TrendsEcol.Evol.24,659–669.

LeBrun,E.G.,Jones,N.T.,Gilber,L.E.,2014.Chemicalwarfareamonginvaders:a detoxiﬁcationinteractionfacilitatesanantinvasion.Science343,1014–1017.

Lenoir,A.,Devers,S.,Touchard,A.,Dejean,A.,2015.TheGuianesepopulationofthe ﬁreantSolenopsissaevissimaisunicolonial.InsectSci.(inpress).

MacGown,J.A.,Boudinot,B.,Deyrup,M.,Sorger,D.M.,2014.AreviewoftheNearctic Odontomachus(Hymenoptera:Formicidae:Ponerinae)withatreatmentofthe males.Zootaxa3802,515–552.

Moffett,M.W.,2012.Supercoloniesofbillionsinaninvasiveant:whatisasociety?

Behav.Ecol.23,925–933.

Mori,S.A.,Cremers,G.,Gracie,C.,deGranville,J.-J.,Hoff,M.,Mitchell,J.D.,1997.

GuidetothevascularplantsofcentralFrenchGuiana.Mem.N.Y.Bot.Gard.76, 1–442.

RCoreTeam,2013.R:ALanguageandEnvironmentforStatisticalComputing.

RFoundationforStatisticalComputing,Vienna,Austria,URL:http://www.R- project.org/

Orivel,J.,Grangier,J.,Foucaud,J.,LeBreton,J.,Andrès,F.X.,Jourdan,H.,Delabie, J.H.C.,Fournier,D.,Cerdan,P.,Facon,B.,Estoup,A.,Dejean,A.,2009.Ecologically heterogeneouspopulationsoftheinvasiveantWasmanniaauropunctatawithin itsnativeandintroducedranges.Ecol.Entomol.34,504–512.

Richardson,B.A.,Richardson,M.J.,Scatena,F.N.,McDowell,W.H.,2000.Effectsof nutrientavailabilityandotherelevationalchangesonbromeliadpopulations

andtheirinvertebratecommunitiesinahumidtropicalforestinPuertoRico.J.

Trop.Ecol.16,167–188.

Roux,O.,Rossi,V.,Céréghino,R.,Compin,A.,Martin,J.-M.,Dejean,A.,2013.Howto coexistwithﬁreants:therolesofbehaviourandcuticularcompounds.Behav.

Proc.98,51–57.

Sarkar,D.,2008.Lattice:MultivariateDataVisualizationwithR.Springer,Berlin, Heidelberg,pp.273.

Seifert,B.,2003.TheantgenusCardiocondyla(Insecta:Hymenoptera:Formicidae)– ataxonomicrevisionoftheC.elegans,C.bulgarica,C.batesii,C.nuda,C.shuckardi, C.stambulofﬁi,C.wroughtonii,C.emeryi,andC.minutiorspeciesgroups.Ann.

Naturhist.Mus.Wien104B,203–338.

SerramoLopez,L.C.,DaNóbregaAlves,R.R.,Rios,R.I.,2009.Micro-environmental factorsandtheendemism ofbromeliadaquaticfauna.Hydrobiologia625, 151–156.

Shogren,J.F.,Tschirhart,J.,2005.Integratingecologyandeconomicstoaddress bioinvasions.Ecol.Econ.52,267–271.

Taber,S.W.,2000.FireAnts.TexasA&MUniversityPress,CollegeStation,pp.308.

Warman,L.D.,Forsyth,D.M.,Sinclair,A.R.E.,Freemark,K.,Moore,H.D.,Barrett,T.W., Pressey,R.L.,White,D.,2004.Speciesdistributions,surrogacy,andimportant conservationregionsinCanada.Ecol.Lett.7,374–379.

Wenzel,J.W.,1996.Learning,behaviorprograms,andhigher-levelrulesinnestcon- structionofPolistes.In:Turillazzi,S.,West-Eberhard,M.J.(Eds.),NaturalHistory andEvolutionofPaper-Wasps.OxfordUniversityPress,Oxford,pp.58–74,414.

Wetterer,J.K.,2008.Worldwidespreadofthelonghorncrazyant,Paratrechinalongi- cornis(Hymenoptera:Formicidae).Myrmecol.News11,137–149.

Wetterer,J.K.,2009.Worldwidespreadoftheghostant,Tapinomamelanocephalum (Hymenoptera:Formicidae).Myrmecol.News12,23–33.

Wetterer,J.K.,2010a.Worldwidespreadoftheﬂowerant,Monomoriumﬂoricola (Hymenoptera:Formicidae).Myrmecol.News13,19–27.

Wetterer,J.K.,2010b.Worldwidespreadofthewoolyant,Tetramoriumlanuginosum (Hymenoptera:Formicidae).Myrmecol.News13,81–88.

Wetterer,J.K.,2012.WorldwidespreadoftheAfricanbig-headedant,Pheidolemega- cephala(Hymenoptera:Formicidae).Myrmecol.News17,51–62.

Wetterer,J.K.,2014a.ASouthAmericanﬁreant,Solenopsisnr.saevissima,inGuade- loupe,FrenchWestIndies.Biol.Invas.16,755–758.

Wetterer,J.K.,2014b.Worldwidespreadofthelessersneakingant,Cardiocondyla minutior(Hymenoptera:Formicidae).Fla.Entomol.97,567–574.

Wetterer,J.K.,Wetterer,A.L.,2004.Ants(Hymenoptera:Formicidae)ofBermuda.

Fla.Entomol.87,212–221.