, Rodriguèsia62(2):341-366.201I http.'/Zrodriguesia.jbrj.gov.br Floristics and life-forms along a topographic gradient, central-western Ceará, Brazil Florísticaeformasdevidaaolongodeumgradientetopográficonocentro-oeste doestadodoCeará,Brasil FranciscaSoaresdeAraújo' 2 RafaelCarvalhoda Costa1 Jacira RabeloLima' , , Sandra Freitasde Vasconcelos' Luciana CoeGirão', Melissa SouzaSobrinho', , Morgana MariaArcanjoBruno1 SarahSuedGomesdeSouza1 EdsonPaulaNunes1 , & , , MariaAngélicaFigueiredo',Luiz Wilson Lima-Verde' MariaIracemaBezerraLoiola' Abstract To tcst whether thc Hora is organized in discrete or continuous units along a topographic gradient, thrce physiognomicswcrcasscsscdondiffcrcntsoilclassesinasemi-aridregionofnortheastemBrazil:caatinga (xericshrubland)ataltitudesfrom300to500m,deciduousforestataltitudesfrom500to700mandcarrasco (dcciduousshrubland)at 700m. Ineachphysiognomyaspccies inventorywascarricdout,andplantswcrc classifiedaccordingtolifc-andgrowlh-forms.Spccicsrichncsswashighcrinthcdcciduousforest(250)than inthccarrasco(136)andcaatinga(137).Thccaatingasharedonlyafewspccicswitlithccarrasco(6spccies) andthcdcciduousforest(18spccics).Thchighestspccicsovcrlapwasbetwccnthcdcciduous forestandthc carrasco(62spccics).Onehundredandfourspcciesoccurredonlyinthccaatinga, 161 onlyinthcdeciduous forestand59onlyinthccarrasco.Woodyspccicspredominatedinphysiognomicsonsedimentarysoilswith latosolandarcnosol: 124spccicsoccurredinthcdcciduousforestand68inthccarrasco. Inthccaatingaon crystallinebasement rclicfwith predominanceofplanosol, herbs showcd thc highestspccics richncss(69). ComparingthcbiologicalspcctrumofBrazilianplantlife-forms,thccaatingastoodoutwithhighcrproportion ofthcrophytcsand chamacphytcs. Considering thc Hora ofthc threephytophysiognomics studicd herc, wc canaffirmlhatthccaatingaisadiscrctcfloristicunit. Kcywords: vegetationclassification,biological spcctrum,growth-form,phytoclimatc,plantcommunity. Resumo Paraverificarscacomposiçãoflorísticaconstituiunidadesdiscretasoucontínuasao longodcumgradiente topográfico foramanalisadastrês fitofisionomias(caatingasobrealtitudesdc300a 500m, florestadecídua sobrealtitudesdc500a700mccarrascosobreatitudesdc700m)sobreclassesdcsolosdistintasnosemi- áridosetentrional doNordestedo Brasil. Emcada fisionomia foi realizadoo levantamentodasespécies, as quais foram classificadas em formas dc vida cdccrescimento. A riquezadc espécies foi maiorna floresta decídua (250) do que no carrasco (136) c na caatinga (137). A caatinga apresentou poucas espécies cm comumcomasfitofisionomiasdccarrascooudcflorestadecídua(6c 18espécies).Amaiorsobreposiçãodc espécies ocorreu entre a floresta decídua c ocarrasco, 62 espécies. Foram exclusivas da caatinga, floresta decídua c do carrasco, 104, 161 c 59 espécies, respectivamente. Quanto ás formas dc crescimento, nas fisionomias sobre relevo sedimentarcom Latossolo c Arcnosolo predominaram espécies lenhosas: 124 na florestadecíduac68nocarrasco.Nacaatingasobrerelevodoembasamentocristalinocompredominânciadc Planossolo,amaiorriquezadcespécies(69)foidcervas.Naanálisecomparativadoespectrobiológicocom outrasformaçõesbrasileiras,odccaatingascdestacoudosdemais,constituindoumaunidadeindividualizada pela maiorproporçáodcterófitosccaméfitos. Em relaçãoà floradastrêsfisionomias, objetodesteestudo, pode-seafirmarqueadacaatingarepresentaumaunidadediscreta. Palavras-chave:classificaçãodcvegetação;espectrobiológico;forniadccrescimento;filoclima,comunidadevegetal. 'UniversidadeFederaldoCeará,Depio.Biologia.CentrodcOíncim,bloco906.CampusdoPiei.60455-760.Fortaleza.(P.Brazil. ;Corrcpondingauthor:tchcscawufc.br SciELO/JBRJ cm L2 13 14 15 16 17 0 1 1 342 Araújo. F.S. etaL Introduction indifferentdirectionsrefiectenvironmentaleffects, At a global scale, the main environmental especially related to climate, on plant adaptations variables used to classify vegetation are climate observedinacommunity(Raunkiaer1934).Hence, zones. A group of similar vegetation types that whereas the growth-form classification is used to occurinsimilarclimatezonesindifferentcontinents characterize community structure (because some isknownasavegetation-typeorbiome(Whittaker formsaredominantormoreconspicuous),thelife- 1975, 1978a,b;Box&Fujiwara2005). formspectrumdescribesenvironmentaladaptations Changes in topography or microclimate can of the species that compose that community affect the biology of the vegetation, leading to (Whittaker 1975;Raunkiaer 1934).Indirectly,this particularitiesthatcan be detectedonly ata local system provides information on local seasonality. scale (Spellerberg & Sawyer 1999). Gradual AccordingtoWhittaker(1975),life-formsarenota changes in climate related to topography or to structural attribute, but a floristic attribute: when distancefromtheocean,atasmall scale,resultin thenumberofspeciesisconvertedintopercentage continuous vegetation units, which makes a oflife-forms, this percentage would represent the classificationbasedonfloristicattributesdifficult. spectrum of life-forms in this community or However, when a climate variable is associated geographic area. The fact that agiven community withdifferentsoiltypes,theregionalfloramaybe is characterizcd by particular life-forms indicates discontinuously distributed, forming discrete speciesconvergencetowardcertainenvironmental communities, whose limits, along a topographic conditions; and this represents a functional gradient, can be determined by an analysis of attributeofthecommunity. floristic composition and ofthe main growth- or In the present study, weassessed life-forms, life-forms ofthe plant species (Whittaker 1975; growth-forms and floristic composition of three Box&Fujiwara2005). neighboring physiognomies that occur under Todescribccommunitytypes itisnecessary different climates, soils and topographies. These to characterize plant forms, since physiognomy community attributes weredetermined foranarea results from the dominant forms that compose a locatedinthesemi-aridregionofnortheastemBrazil, community (Whittaker 1975). Classes ortypes of which comprises two geomorphological units: plant forms are called growth-forms; this sedimentary basin andcrystallinebasement. classification usually does not correspond to the Basedon thesedata, wetestedthefollowing categories used by taxonomists to classify plants. predictions:i)theflorasofthetwogeomorphological Height,woodyorherbaceoushabit,stemform,leaf unitsaredifferent,andconstitutetwodiscreteunits; form and intensity of leaf deciduousness are ii)thelife-formspectrumvariesaccordingtoaltitude characteristics used to define the following types and soil type, probably as a consequence of of growth-forms (Whittaker 1975): trees, shrubs, differencesinwateravailability,resultingmainlyin lianas, epiphytes, herbs and thallophytes. theoccurrenceofphanerophytesinthesedimentary Instead of using a system of multiple basinandoftherophytesinthecrystallinebasement. characteristics such as the growth-form system proposedbyWhittaker(1975),thelife-formsystem Material and Methods of Raunkiaer (1934) is based on a single Location and environmental characteristic.therelationshipbetweentheposition ofthe perennial tissue (meristem), which remains characterization of the study area ginraocwtitvhe dsuurrifnagcet.heTwhientelrifoer-dfroyrmseoafsona,sapnedcitehse area oSfer5ra,6d4a6s Ahal,maasndNaitsurlaolcaRteesderbveetcwoeveenrstahne coordinates 5°15’-5°00’S and 40°15'-41°00'W crsleealpsersceitsfeiedendtbpsylaantthsseeitnetnoovffiirlvioefnelmiehfnei-tsft.oorrRmysa:ucpnhhkaairnaaecertroepr(hi1ys9tt3ei4sc),s kí()Fmicga.at1i).ngTahe(xsertiucdysharruebalahnads)twhirteheapnhyasrieoagonfom1i7.es1: achnadmtaheeprhoyptheyste,s.hemicryptophytes, cryptophytes an ar(e2a9.o1f92%7).,93fi)ksme-as(o4n7a.l6d4e%c)idaunoduisiif)ocraesrtrawsictho ctrheaeplrcweuoslTrealhndtte.eatwdFiorvrbeolydmsasRtpmhaeaputclntesrkauiommfap,leaoerlr,ln(ttohh1ree9m3pvaa4alt)stscepubrelnacasstrrerefdulcmoo,orrawndaeosdaf cga(reodeoemamcpoiolrfdeTpuxh1ho,e1ou.wls7io9tsgskhtihmcurfauid2alby(tl2uta0no.niatsd1rsl2)e:i%(agi)Rh).ottluihygseeucrnrlidyouesltcaaa1ltt9liei8ndn8ge)rbeiawlnsiieetfmhtaewnnaodnt Rodriguèsia62(2):34 -366.201 SciELO/JBRJ cm .. .2 13 14 15 16 17 343 Florísticsandlife-formsalonga topographicgradient IMSEAR; Biodiversity inventories - Caatinga (PROBIO-MMA)andEdital UniversaldoCNPq/ 476285/2003-8. In these studies, branchcs of angiosperms(fiveduplicates)inreproduetivephase (flowerbuds, flowcrsand/orfruits)wcrccollcctcd on trails and inside the best-conserved fragments ofeachphysiognomy.Vouchcrswcredepositedin thePriscoBezerraHcrbarium(EAC),ofUniversidade FederaldoCeará.Botanicalidcntificationwascarried out using analytical keys (Freire 1983; Barroso O£23|CDCcucumirtiimunigmiauifimnl tetheal.ma1t9e7r8i,al19p8r4e,se1n9t8i6n)tahnedEbAyCcoHmpearrbiasroinumwiotrh, Uiitiirlwltwtntul whcn necessary, by Consulting specialists. The Figure 1 - Location of Serra das Almas Natural cnlaamsseisfiwceartieonupudsaetdewdacsonAsPidGerIiIIng(2t0h0e9)s.ynSopencyimcys Reserve, Cratcús, StateofCeará, Brazil. of Missouri Botanical Garden (Tropicos.org 2009); names and/or abbreviations of spccies authors were written in accordance with low altitude (c. 400 m) and ii) the Meio Norte Brummitt& Powell(1992). sedimentary basin, on its eastern margin, which forms an asymmetric cuesta, known as Ibiapaba Growth- and life-forms Plateau(altitudesbctween500and700m). The caatinga occurs in the crystallinc Eachspecicswasclassiftcdintogrowth-forms basementcomplex, wherethedominantclassesof followingWhittaker(1975). soils are: Solodic Planosol, Solodized Solonetz Theclassificationofeachspecicsinlife-forms (natric Planosols)andLilhicsoils(LithicNcosols) wasdonebasedontheprotection leveiofgrowing ataltitudesthatvary from300to500nt. tips and on lhe reduction ofthe aerial part during In the Meio Norte sedimentary basin, on theunfavorableseason,followingRaunk&iacr(1934, IbiapabaPlateau,theLatosoloccursontheeastern seealsoCain 1950;Mueller-Dombois Ellenberg hogback and quartz sand (quartzarenic ncosols) 1974): therophytes (Th), cryptophytcs (Cr), tdoheneciptdlhuaeoteutasoup,fooraennsdtLaotbcoacsucorklss,ioadnletahl(etBiretaaussdtieeslrbn1ch9to7wg2e)be.ancTk5h0oe0f chpoheanmnsicicrdroeyprphetydotpehassyt(pePhhs)a.(nHcW)ro,oopcdhhyyatmleaiseanpaahsnydatncdnsocn(acC-thwi)owocadrnycd and700m.Thecarrascoispresentonthebackside lianas were classiftcd according to the levei of of the plateau, on quartz sand, at altitudes of ca. reductionoftheiraerial partduringlhedry season 700 m. Weemphasize thatthe Ibiapaba Plateau (accordingtoRaunkiacr 1934). is a ‘cuesta’, with highcr asymmetry in its Southern part, our study arca, where there is no Data analysis top, butan inverted V-shapcd topography where Floristic data were organized as a list with lhe lecward on the backside exhibits a smooth families,specics,vernacularnames,liWfeeandgrowth- forms,physiognomyandcollectors. calculutcd declivity. Climatedatawerenotavailable,bccausethere spcciesandfamilyrichnessforthewholedatasetand are no metcorological or pluviometric stations by physiognomy.Tocompare the richest families locatedon thecuesta, topand immediatebackside bctween physiognomics, weusedhistogramswith sites on lhe Southern part ofthe Ibiapaba Plateau, lhe tenFlroircihsteisctofvacmrillaipesbcintwdeeesncepnhdyisnigogorndoemri.cswas our study arca. analyzed by calculating the frcquency ofspecics and familiesinoverlappingclasses: occurrcnce in sweavsereaTFxllthopeerrnoisfjslietovciretaclsoy:ifnrsSevaesmererpnraltveodedarmsyafArnloammgaes1m9Ne9ant9turtpaolla2nR0;e0s4leo,rnvgie-n cpa(lfchllay/apsTthiUiyíoncsggoian/o/odgcmeanycroirmdaiucsocacusaos,t,ifnoicgrnaeaasptta)ici,narasgrnarod/aldsrcccpcoshitryodiruscoitducoedsgcntiIoodomrueeioascuctshs, CteNrPmqe/cPolEoLgDic;alInrsetsietaurtcohdporoMgirlaênmiso-dSoitSeeCmaiaátriindgoa-/ forest). Results(urc prese,nted in histograms. Rodríguésia62(2):341-366.2011 SciELO/JBRJ cm 12 13 14 15 16 17 1 344 Araújo,F.S. etal. Totestfordifferences in thecomposition of life-formsamongphysiognomies,wecalculatedthe life-form spectrum, which is the proportion of speciesofeachlife-form.Wedeterminedwhichlife- form characterized each physiognomy by comparing our results with the normal spectrum proposed by Raunkiaer (1934). This spectrum FAB Eli' CON MAL AST POA APO RUB MYR SAP represents the world flora and was used here as null hypothesis. At First, we tested fordifferences between the obtained and the normal spectrum usingax2test(Vieira2004).Whendifferenceswere significam,wecalculatedtherelativecontribution ofeach life-form’s deviation to the computed X2 statistic.Thelife-form withhighercontribution in FAB CON EUP MAL POA AST ACA APO PAS COM each test was considered as characteristic of the physiognomy where it occurs. Carrasco To test for similarities with other Brazilian vegetation types (in terms of life-forms), we compiled studies with spectra determined for Brazilian physiognomies (Tab. 1). We kept the namesusedbyeachauthorforthevegetationtypes ofeach study. To facilitate comparison, we used only the five main life-form classes ofRaunkiaer (1934). Hence, epiphytes and woody lianas were included in the class phanerophytes, saprophytes incryptophytes, andaerophytes in chamaephytes. Wecomparedthe life-form spectrafound in Serra das Almas Natural Reserve with those from other studies with a detrended correspondence analysis FAB EL!P mal CON AST MYR SAP RUB PCIA APO - DCA (Jongman etal. 1995; Batalha & Martins Families 2002); results were expressed in ordination ^Sure7~Species-richestfamiliesinthethreephysiognomies diagrams with scores of each study and of each ofSerradasAlmasNaturalReserve,Crateús,StateofCeará, life-form. Brazil. Abbreviations for families: FAB-Fabaccae, EUP- Euphorbiaceae,CON-Convolvulaceae,MAL-Malvaceae, Results AST- Asteraceae, POA-Poaceae, APO-Apocynaceae, — r7i2chfnaemWsislei(er2se5c0(o)Arnwdneedreex41h19i)g.shpFeearcmiiinelsit/hemesodi(ep5ch5io)sdapuenocduisesfspoefrcerisoetms. CBAoCRmAOm--eAlBicrnRaoaunmbcteiealhaceaie,acaceeeL,aaAeeM.M,-YLRa-PmMAiySar-ctPeaaacsees,aiefM,lAoPSr-AaMPca-elSapaiepg,ihnidCaaccOeeMaaee-,, Richness valuesofthe carrasco (46and 136) and caatinga (44 and 137) were similar to each other and lowerthan in the deciduous forest. Fabaceae (86 species), Euphorbiaceae (38 Familyoverlapwasaboutonethirdamongall rsipcehceisets)plaanndtCfoanmviolilevsuliancSeearera(2d2asspeAclimeas)swNearteurtahle pthheysdieocgindoumouisesf(oFriegs.t3s)h.aHroewdetvheerh,itghheesctarnruamsbceoranodf Reserve.Therichestfamiliesweredifferentamong tamilies,andhadthehighest(carrasco)andlowest pfspahemycisilieysoFgiannboalamlciteehasree,e(Fwpihhgyi.sci2ho).ghnaTodhmteiheesehx(icFgeihgpe.ts2it)o.nnHuowmwabesevretrho,ef cSs(dppaeeercccriiiadeesusscoouoocsvcfaeuornrrldersaetpdt)hnwienuamsadlbleelcporihwod,yfuseaoixsuocslgounnsfolioymvrieenefsisatnme(ihFloiaigued.st(3hoF)ifi.ggT4.hh13e)e9r. the representativeness ofsubfamilies varied, with floristicalfinity witheachother,sincethey shared higherrichness ofPapilionoidae in the deciduous morespecies(15%)andbothhadlowoverlapwith forest (25 species) and ofCaesalpinioidae in the thecaatinga(1.3%overlapwithcarrascoand4.2% caatinga (12 species) and carrasco (15 species). withdeciduousforest-Fig. 3). Rodriguésia62(2):341-366.201 SciELO/JBRJ. cm i 12 13 14 15 16 17 Rodrlguésla CTahb-lec1ha-mLaiefpeh-yftoersm,spPehct-raphuasneedrfooprhcyotmepsa.risonsinadetrendedcorrespondenceanalysis(DCA).Life-forms:Th-therophytes.Cr-cryptophytes,H-hemicryptophytes, Florístics and (622 Vegetation type Abbreviation Site Reference Th Cr H Ch Ph ): life-forms 341 caatinga caa Sa.dasAlmas,Crateús,CE This study 47,9 1.4 6,3 18,1 26,4 3-66 carrasco can Sa.dasAlmas,Crateús,CE This study 17,2 3,4 3,4 17,9 57,9 along . deciduousforest fldec Sa.dasAlmas,Crateús,CE This study 14,6 2,6 2,2 22,5 58,1 a 2011 caatinga caa Faz.NãomeDeixes,Quixadá,CE Costaetal.(2007) 42,9 2,3 12,8 15,8 26,3 cerradofechado cerfec Brasília.DF Ratter(1980) inBatalha& Martins(2002) 0,7 1,8 44,9 13,5 39,1 topographic cerradoaberto cerab PARNAdas Emas,GO Batalha& Martins(2002) 3,7 2 49,9 12,8 31,6 cerradoaberto cerab LagoaSanta,MG Warming(1892)inBatalha&Martins(2002) 4,6 5,4 55,1 6,1 28,8 cerradoaberto cerab Mojiguaçu,SP Mantovani(1983) in Batalha& Martins(2002) 7,8 2,1 47 12,2 30,9 gradient cerradofechado cerfec Pirassununga,SP Batalhaetal(1997)inBatalha&Martins(2002) 5,6 1,1 36,1 17,1 40,1 cerradofechado cerfec Sta.RitadoPassaQuatro,SP Batalha&Mantovani(2001) 6,7 0,8 30 17.2 45,3 & inBatalha Martins(2002) pluvial forest flpl AltodoPalmital,FozdoIguaçu,PR Cainetal.(1956) 0 3 11 6 80 pluvial forest flpl Caiobá.PR Cainetal.(1956) 0 3 3 7 87 pluvial forest flpl Mucambo,Belém,PA Cainetal.(1956) 0 0,9 2,8 0,9 95,4 temperate forest fl temp HortoBotânico.Pelotas.RS Cainetal.(1956) 5 5 16 4 70 cerradão cerradão AguasdeSta.Barbara,SP MeiraNetoetal. (2007) 0 0 4 1,3 94,7 cerradosensustrictu cerss ÁguasdeSta.Barbara,SP MeiraNetoetal. (2007) 0 2,8 10,7 9,6 77 campocerrado cpcer ÁguasdeSta.Barbara,SP MeiraNetoetal.(2007) 0 6,4 19,2 14,1 60.3 camposujo cpsj ÁguasdeSta.Barbara,SP MeiraNetoetal.(2007) 0 7.9 31,8 41,3 19,1 campolimpo cllp ÁguasdeSta.Barbara.SP MeiraNetoetal.(2007) 5 0 32 34 14 restinga res Itamaracá.PE AlmeidaJRetal.(2007) 16,8 5,3 8 19,5 50,4 inselbergvegetation inselb Quixadá.CE Araújoetal.(2008) 44,2 2,6 13 15.6 24,7 cerradosensustrictu cerss Itirapina, SP Batalha& Martins(2004) 1,8 1.8 18,6 11.5 66,4 caatinga caa Betãnia.PE Costaetal. (2009) 40,5 1,1 14.6 18 25.8 restinga res Caravela.BA MeiraNetoetal. (2005) 9 0 14,9 23,9 52.2 restinga res Mucurí. BA MeiraNetoetal. (2005) 7,5 0 28.3 24,5 39,6 Note:cüannga-xencshrubiand.carrasco—deoduousshrubLmd:cerradosensustriclo—sa\anna:cerrado<fechado=densesa\anna;cerradoaberto=opensavanna;campocerrado=grasslandwilhscateredshrubs;camposujo=grasslandwilh ^ scateredshrubs:campolimpo=grassland:cerradão=tallwoodlandsa\anna:restinga=sandvCoastalplains. SciELO/JBRJ 16 17 18 19 20 21 22 23 24 25 26 , : , 1 346 Araújo. F.S.etal. Th Ci H Ch O Ph CA CR FD CA/CR CATO CR/FD ALL Occurancyclasses o 20 40% 60 80 100 species Figure3 - Proportionoffamilies(white) and species Figure4-Life-formspectraofthethreephysiognomies ((bClAac-k)caoactciunrgraingCRin-ocnaer,rtaswcoo,orDiFn-aldlepchiydsuioougsnfoomreisets) (CA-caatinga CR-carrasco,DF-deciduousforest)of ofSerradasAlmasNatural Reserve, Crateús, State of SerradasAlmasNaturalReserve,Crateús,StateofCeará, Brazil, compared to Raunkiaer’s normal spectrum (N). Ceará, Brazil. Valuesovereachphysiognomybarindicatethenumber ofspeciesofeachlife-form.Speciespercentagesofeach woodyInstpheecipehsys(isohgrunbosmiaensdotnreseesd)ipmreendtoamriynraetlieedf,, tlihfeer-ofpohrymtaere(Tehx)p,rcersyspetdobpyhythteew(iCdr)t,hhoefmtihcerbyapr.toLpihfey-tfeor(mHs):, totaling 124 in the deciduous forest and 68 in the chamaephyte(Ch),phanerophyte(Ph). carrasco. In the caatinga, on the crystalline basement, the highest species richness (69) was represented by herbs. group, carrasco and deciduous forest exhibited The life-form spectra of the studied scoresclosetothoseofrestingaanddifferentfrom physiognomiesdifferedsignificantlyfromthenormal those ofcerrado, apparently because ofthe lower spectrum (caatinga %2= 159.33 p < 0.01 df= 4; proportion ofcryptophytes (Fig. 5). The caatinga carrascoX:=49.07p<0.01 df=4;deciduousforest composedawell-definedgroup,whichcomprised X"= 120, p<0.01 dl =4). In general, thecarrasco spectra of other caatinga studies, including andthedeciduousforestexhibitedsimilarproportions vegetation on inselbergs. This group is associated of species ofeach life-form, whereas the caatinga with higher proportion of chamaephytes and exhibitedadifferentspectrum(Fig.4).Therophytes, therophytes(Fig.5). hemicryptophytes and chamaephytes were the predominantlife-formsinthecaatinga(69%),carrasco Discussion (53%)anddeciduousforest(46%),respcctively;thus, In general, in the semi-arid region of they characterizeeach physiognomy. northeastemBrazil,areaswithhigherannualaverage Inthecomparisonsoflife-formspectraamong rainiall associated with higher altitudes exhibit pawxihetyshsioootfghtnehroemBiDreaCsziAolficaSonerrvrreeagsdepatosantAdielodnmattsoypoNevaset,urtrh6ael0t%RweoosfeFritrvhseet ahwli.ags2h0e0ra7lss;pFoeecriorebasszreeitrcavhl.ened1s9s9i8(n;LiGthmoeameeptshayl1.s982i00o)0.g9nT;hoAimrsiapeúasjtoteoemft totncohfertexaylltispfeitencot-eoporfntohdiryat.m:thIs4eenp9set.och6atner8neao%dsrsdotiinooinfoita)dhtepiooouhfftniar:sndctiei)haarsaxgopimrepsacahatemynrptdathehw1rysi3et,t.ehe3g0ssiric%o)oaurnopoendssf adNsleaettcdiuiitrdmuaeudlnoetusRasertsfyheoarraenrvseetat.sah.Beneedsvccieaadnarctrswia,instcgdhoae,rcaiibinodnftuaSholeullrsoircnavadetegdexeadtesaasttAshiiliomgnmihlaoeasnrr tothecaatingaarcaofthecrystallinebasement,have tcoAhraldermirranoaaspsthciyNootanetasusnrpd(aaFlcitegh,R.eei5snd)e.gerrcTvoiehudepunoe2lua,irsfwlehy-fifoocroverhsmetarsllpisaneopcpctSeoredmarproiranfisdtteahhssee Tberixhecceehensnpeeptsoeisixoncn(tesSepid(tliRvoooaundtesatiisnahleg.otewna2le0.tr0ha3a1l)t9,9abs8et;ihhnaoPgvueirsgneehigdrtihahmieeegrntheteaaalr.rrsey2p0sae0lco2oim)nee.se the restinga and cerrado spectra (Fig. 5). In this does not result in higher species richness; other Rodnguésiá62 2 341-366 201 ( ): . SciELO/JBRJ 2 13 14 15 16 17 18 1 347 Roristicsandlife-formsalonga topographicgradient inwoodygrowth-formsisrelatedtotheincreasein HHh +A OLtifhec-rfsotrumdies aridity (lowerrainfall and highertemperature). In o SerradasAlma* previous studies, the replacemcnt of non-woody ccrVfcc -cerab life-formsbywoodylife-formsandtheincrease in erradio &“A cpAsjA cplp richness along humidily gradients have bcen x observed inaridareas(Pavónetal.2000),tropical < Ph \ lltemcppwr1^ |+Cr ^ inselb savannas (Williams et al. 1996), forests and 11P* cmt temperategrasslands(Kovács-Langetal.2000). lldec O+C-li Considcring woody and herbaceous flora together, the deciduous forest on lhe sedimentary basin exhibited higherrichness than the caatinga o.o 0.5 1.0 1.5 2.0 2.5 located on the crystalline basement. Potentially, Axis 1 there must be higher humidity in the air and soil Figure 5 - Ordination diagram of lhe detrended resulting from the elevation; there must be also correspondenteanalysis(DCA),withscoresoflife-forms (solialtsowsiotlhspearnmdanqeunatrwlaztesraavnadisl)a,bilwihtyicinhdepeopsslaiybelrys andcompiledinventorics,includingthephysiognomiesol contributctothehigherfloristicrichnessobserved. Serra das Almas Natural Reserve. Abbreviations tor Comparing the carrasco and the deciduous (veTghe)l,aticornytpytpoepshfyotlelow(CTra)b,leh1e.mLiifcer-yfportmosp:htyhteerop(hHy)t,e forest located in the same sedimentary basin, the latter exhibited higher richness. In this case, chamacphyte(Ch),phanerophytc(Ph). humidity seems to be an important factor: lhe deciduous forest is located on lhe cuesta and the factorsmustalsobeconsidered,suchaslheposition carrasco on the immediate backside. On the of the hogback, levei of desiccation of the relief backsidetheairisprobablydrierandwindspeedis and physiochemical composition ofthe soil. The higher, which causes more desiccation. Besides, dReesceirdvueouisslofcoraetsetdoonftSheerwrianddawsarAdlsmiades, bNeattwucreanl ssoainldiseere(mAsratoújpola&yaMarrotleintsoo1,9s9i9n;ceArcaaúrjroasectoals.oi1l9s99a)r.e 500 m and 700 m. whereas the carrasco, though Despite the high species richness found in oloncattheedIatceawhairgdhesridaeltaitnuddeoanbosuatnd7i0e0r mso,iliss,lowchaitcehd thhieghrleiggihotn tohfeIbcioanptarbiabuPtliaotneauo,fitthies inmopnor-twaonotdtyo arensdulstlseinndearpphlyasntisogannodmlyoowferlorwiechrnheesisghtth,ansmianllteher clioamnpaos)nctnotth(ehertbost,alssupbeschireusbsriacnhdnehsesrboafceeaocuhs physiognomy. In lhe caatinga, on the crystalline deciduous forest. Conccmingtheherbaceouscomponcntoflhe basement. non-woody plants werc rcsponsible for Braziliansemi-arid Hora,studiescarriedoutinthe most of the floristic richness, that is expectcd in siinnetdneisrcu-aptlcsatttrehiaactutodtehipesrheiisngshiteohsnetohrfeitcrhhbenaeccrseyssotouafslltichneoemccpaoaomtnpiclnnegtxa oainrfittdhhacerncodpashreyrmtaiess-caoirniatdnhdcelsiienmcanttvhciesr,doedncumieendttuoso.tuhsOenpfortreehdsctoc.omniwtnoraoanrdcyye, (Sampaio 995:Rodaietal.2005;Costaetal.2007: plantswerercsponsibleforthehighestrichness,since Mamede &1 Araújo 2008). Comparatively, studies inmorehumidclimatesphanerophytcspredominatc. craircrhineedssouotfinheserdbiamceenotuasryflfoorram(utRioodnasireetcoarl.de1d9l9o9w; cstablHiisghmhcenrt owfaltiefre-faovrmasiltahbaitldiotynotfanveoerdslatrhgee reduetions of the aerial shool system during lhe FigueiIrnedSoercrtaal.d2a0s00A)l.mas Natural Reserve, the unfavorable season (phanerophytcs), which is a floristic richness of woody species incrcased at nccessarystrategy forlhesurvivalofmostspecies high altitudes in areas of deciduous forest and inaridandsemi-aridregions(sceRaunkiaer 1934; carrasco, whereas the richness of herbaceous van Rooyenetal. 1990;Kovács-Langetal.2000). speciesdecrcased.Theincrcascinrichnessof trees In the case of Serra das Almas Natural Reserve, and shrubs with altitude seems to bc a general whichisinsertedinasemi-aridclimaticdomain,the patternforvegetationofaridandsemi-aridregions. increaseinaltitudemaypotentiallyfavorhighwater In the Brazilian semi-arid region, the increase in availabilityonthewindwardside.Besides,soilmust richnessofherbaceousgrowth-formsanddecrease betakenintoaccounl,sincetherearetwodiffcrent Rodriguèsia62(2):341-366.201 ISciELO/ JBRJ cm 12 13 14 15 16 17 1 348 Araújo. FS.etal. geological units: lowlands of the crystalline deciduous forest occurs on the windward side bascmeHnetrbaancdeothuesMoeriosuNbor-tweoosdedyimpelnatnatrsy(bhaesribns., dwhoenroctasftohremcadrirscarsectoeocunciutrss.oTnhtehyelfeoerwmaradcsiodnel),itnhueusme subshrubsandherbaceouslianas)arethelife-forms represented by species overlap and by the same sthyastteexhmibdiutrtihenghigthheestdrreyduscteiaosnoonft(hteheaerroipalhystheoso,t biologiWchalesnpeacntarluymz,iansgepmhpyhsasiiozgendombiyeAsusotnindi(f2f0e0r5e)n.t c1r9y3p4t)o.phTyhteesb,ioalnodgichaelmiscpreycpttroupmhyotfest;heRacuanaktiiangear sgoeiolmocropmhpoolongeinctalmuaniyts,deaptaerrtmfirnoemdtihseccrleitmeatuen.itthse; studied was characterized mainly by therophytes, communities that, according toWhittaker(1975), a life-form characteristic of arid and semi-arid eanbedelimited by floristiccompositionand life- regions (Raunkiaer 1934; van Rooyenetal. 1990; forms, such is the case of the difference found Kovács-Langctal.2000). Indeed,amongthethree between the caatinga and the complex deciduous physiognomics studied, the caatinga occurs on forest+carrasco. shallow soils in the lowlands of the crystalline In the comparative analysis with the basement,wheretemperature ispotentiallyhigher biological spectra from other Brazilian seasonal andrainfall is potentially lowerthan in mountain- vegetationtypes,thediscriminationofthecaatinga range areas, resulting in lower water availability. by higher proportion of therophytes and The physiognomies on the Ibiapaba plateau chamaephytes shows that this vegetation is (carrascoanddeciduousforest) mustoccurunder composed of species whose life-forms represem lower water restrictions, since higher altitude better the semi-arid climatic pattern, since the craoinntfrailbluteasndtoltohweeprotteentmipaelraotcucruer,renwcheicohf hfiagvhoerr pofrevdeogemtiantainocnesooffatrhiedseanldifsee-mfio-ramrsiidsecnhavriarcotnermiesnttisc phanerophytes, a life-form characteristic of sites (Raunkiaer 1934; Cain 1950). The biological withlowerwaterrestriction. spectrumissimilarlothespectrumofaridandsemi- Inadditiontonumericdifferences in species aridclimatezonesoftheworld. richness, remarkable differences between the Insummary,thetwogeomorphologicalunits florislic complexes of each physiognomy were presentin thestudyareahavetwodistinct floristic observed in the present study. The two main complexes, characterized by the predominanceof complexes (caatinga and carrasco + deciduous therophytes on the crystalline basement and of forest)areconsistemwiththesoil typesthatoccur phanerophytes on the sedimentary basin. These in the area, resulting from the typeofsourcerock. results show that when implementing reserves in Allhough species overlap between deciduous Braziliansemi-aridareas,abioticlocalfactors,such forestandcarrascomaybeconsideredlow(15%), as soils and relief, must be taken into account, differences are even larger when compared with because these lactors seem to reflect regional caatinga,whoseoverlapisonly4%. Carrascoand floristicvariation.Theenvironmentalheterogeneity deciduous forest are floristically more similar may result not only in high species diversity, but becausebothhaveasetofspeciesthatprefersandy also in high functional diversity in the Brazilian soil with low pH, whercas caatinga differs from semi-ariddomain,which, inthepresentstudy,may that floristic group by the presence of species be observed in differences in life-form spectra typical of soils originated from the crystalline among the three physiognomies analyzed. basement of the inter-plateau depression. The crystalline and sedimentary floras ofnortheastem Acknowledgements oLABBlorbrrecaasamazzeúloiirjllsivoinea&veadntelnsRaistlonoe.odmar2daiini0-ieal0aslf52,rye0;isrc0deaL2sarri;treomiAgfaealiddcoebaonotrtfuaao(tolaA.rmidraa2neadt0srúor0ej-i9voFsc)eci.erealstalhlecao,lra.eertaae1staa9les.9di8to2faf0w,r0tao3hbsm;e; otrCheuaestaetarwrivieTneat,ghhaewtrnhaffsoruuonncnu-dadgrgierhnodigvewedthfmhoimreuecotnhm.mtLaatManhliteaenorigrf,sgelatosmnrétieiruszdnitaioittecisdopiewnlneAavCrsneiesnêotfcncooacirrriayraçtiãhoeeedof BrazilAiasnAsnedmria-adrei-dLriemgaio(n1,9w81h)eenmtphheaspirzeeddo,miinntahme -T(eCcNnPolqo/gPiEa,LDlo-nPge-stqeurimseacEocloolgóigciaclardeeseLaornchgaprDougrraaçmãso pvahryisaitoigonnoimsieisnsctluidmiaetde,inatsheoIbbsiearpvaebdaPilnattehaeu(ttwhoe C(NIsiMPtSeqEC(aAparRtoci-.nMgnCa°.)T,4/7IC6nN2st8Pi5qt/u)2t0,o0d3oo-f8M)EidlaiêntdnailPoRUdOnoiBvSIeerOmsi/aáMlriMddAoo Rodriguésia62(2):341-366.201 SciELO/JBRJ cm .. 2 13 14 1 349 Floristicsandlife-formsalonga topographicgradient (Biodiversity inventories — Caatinga). Marcelo Austin, M.P. 2005. Vcgctation and cnvironment: OliveiraTelesde Menezes helpedus make Figure discontinuitiesandcontinuities.In:Maarcl,E.V.D. (cd.). Vegctation ecology, Blackwcll Publishing, 1. Reviewers contribuled for improving the final Oxford. Pp. 52-84. version ofthe manuscript. Barroso,G.M.;Guimarães,E.F.; lehaso,C.L.F.;Costa. C.G.;Peixoto,A.L.&Lima,H.C. 1978.Sistemática References de angiospermas do Brasil. Vol. I. EdUSP, São Alcoforado-Filho, F.G.; Sampaio, E.V.S.B. & Rodai. Paulo. 255p. rMe.mJa.nNe.sc2e0n0t3e. dFelorvíesgteitcaaçãeofciatdouscsiofcóiloilaogeisapdinehousma BarroPseo,ixGo.tMo.,;AG.uLi.ma&rãeLsi,mEa..F.H;.lCeh.as1o9,8C4..L.SF.i;stCeomstáat,icCa.Gd.e; arbórea cm Caruaru, Pernambuco. Acta Botanica angiospermasdoBrasil.Vol.2.UFV,Viçosa.377p. Barroso,G.M.;Guimarães,E.F.;lehaso,C.L.F.;Costa,C.G.; AlmeFiBlilrdtaoaosrríJaarll.ei,cnEfao.or1Brt7;me:aP2sdi8me7dc-enP3tve0eir3ld.,naaRem.bmMu.cuMom.,a&BárrZaeisaciklde.el,RrCee.svSti.isnt2ga0a0nd7eo. BatalaoPhefnaig,BxiroMoat.szopiA,le.irAam&.naL.scMead&rrortaiBLdnrioasms,saiil,Ft..esRHV...olCF2.l.0o30r.12a9.U81F6L9V.i7,r:Sci4V-s5iftç2oeo-rms4máa6t.s0ip.3cc2ac6tdpre.a AndrGaedoeg-rLaifmiaa.24:D.19-13948.1. The caatingas dominium. Batalahna,dMl.iAf.e-&forMmarstpicncst,rFa.Ro.f2a00c4e.rrFalodroistsiitce,.frBerqauzeinlciya,n RevistaBrasileiradeBotânica4: 149-153. JournalofBiology64: 203-209. APGIII.2009.AnupdateoftheAngiospcrmPhylogeny Box, E.O. & Fujiwara, K. 2005. Vegctation typcs and Groupclassification fortheordersand familiesof their broad-scale distribution. In: Maarel, E.V.D. floweringplants: APG III.BotanicalJournalofthe (eds.). Vegetation ecology. Blackwell Publishing, LinneanSociety 161: 105-121. Oxford. Pp. 106-128. Araújo.F.S.&Martins,F.R. 1999.Fisionomiaeorganização Brasil. 1972. Mapa exploratório-reconhecimento de davegetaçãodocarrasconoplanaltodaIhiapaba,estado solos:estadodoCeará,escala 1:600.000.SUDENE. doCeará.ActaBotanicaBrasílica 13: 1-14. Brumilt,R.K.&Powell.C.E. 1992.Authorsoíplanlnames. Araújo, F.S.; Martins, F.R. & Shepherd, G.J. 1999. Richmond, KewRoyal BotanicGardens.732p. Variações estruturais e florísticas do carrasco no Cain.S.A. 1950.Lifeformsandphytoclimate.Botanical planalto da Ihiapaba, estado do Ceará. Revista Review 16: 1-32. BrasileiradeBiologia59:663-678. Cain, S.A.; Castro, G.M.O.; Pires, J.M. & Silva, N.T. Araújo, F.S.; Gomes, V.S.; Silveira, A.P.; Figueiredo, 1956. Application of some phytosociological M.A.;Olivcirra,R.F.;Bruno,M.M.A.:Lima-Verde. techniquesloBrazilianrainforest.AmericanJournal L.W.; Silva, E.F.; Otutumi, A.T. & Ribeiro, K.A. ofBotany43:911-941. 2(X)7.Efeitodavariaçãotopoclirnáticanafisionomia Costa, R.C.; Araújo, F.S. & Lima-Verde, L.W. 2007. eestruturadavegetaçãodaserradeBaiurité,Ceará. Floraandlife-formspeclruminanareaofdeeiduous In:Oliveira,T.S.&Araújo,F.S.(orgs.).Diversidade thorn woodland (caatinga) in northeastern, Bra/.il. econservaçãodabiotadaserrade Baturité,Ceará. JournalofArid Environments68: 237-247. Seri&AGráfica,Fortaleza.Pp.73-136. Costa, K.C.; Lima, A.L.A.; Fernandes, C.F.M.; Silva, Araújo,F.S.;Rodai.M.J.N.;Barbosa,M.R.V.&Martins, M.C.N.A.; Silva. A.C.B.L.& Rodai, M.J.N. 2009. F.R.2005. Repartiçãodafloralenhosanodomínio Flora vascularc formasde vidacm umhectarede & da caatinga. In: Araújo, F.S.; Rodai, M.J.N. caatinga no nordeste brasileiro. Revista Brasileira Barbosa M.R.V. (orgs.). Análise das variaçõesda deCiênciasAgrárias4:48-54. biodiversidade do bioma caatinga: suporte a Ferraz, E.M.N.; Rodai. M.J.N.; Sampaio, E.V.S.B. & estratégiasregionaisdeconservação.Ministériodo Pereira, R.C.A. 1998. Variação florísticaao longo MeioAmbiente, Brasília. Pp. 15-33. de um gradiente altiludinal no alto vale do Pujeú, AraújMo,.JF..SN..;S&ampFaeiron.anEd.eVs.S,.BA..;GF.igu1e9i9r8ead.o.CMo.mAp.;osRiodçaãio. FigucPierrêndao,mbLu.cSo..:RReovdiastia,BMr.asJi.lNe.ira&deMBeoltoâ,niAc.aL2.1:270-0103.. florísticada vegetaçãodocarrasco, NovoOriente, Florísticacfitossociologiadeumaáreadevegetação CE.RevistaBrasileiradeBotânica21: 105-116. arbustiva caducifólia espinhosa no município de Araújo, F.S.; Sampaio, E.V.S.B.; Rodai, M.J.N. & Buíquc-Pernambuco.Naturalia25:205-224. cFiogmupeoinreednot,eMl.eAn.hos19o9d8eb.trOêsrágraenaiszdaeçãcaorcroamsucnoietmárNioavdoo Freired,asC.fVa.míl1i9a8s3.daCshpalvaenstaasnaptlcírtiidcóafsitpaasr,agaidmentoesrpmeirnmaaçsãoc Oriente-CE.RevistaBrasileiradeBiologia58:85-95. angiospermas brasileiras ou exóticas cultivadas no AraújCoo,mFp.oSs.;içOãliov,eiersap.eRc.tFr.o&biLoilmóagi-cVoercdes,fnLd.rWo.me2s00d8e. GomeBsr,asiMl.,Av..FC.CC1.98C0o.leAçãvoeMgoestsaoçrãooednsoes,CMaorisrsiosróV.el3h6o6sp,. dispersãodavegetaçãodeumainsclbergucnodomínio no estado da Paraíba. Vcgctalia - escritos c dacaatinga.Ceará.Rodriguésia59:659-671. documentos I4(UNESP). Rodriguésia62(2):341-366.201 SciELO/JBRJ cm _2 13 14 15 16 17 . 1 350 Araújo. FS. etal. Jongman,R.H.G.;TerBraak,C.J.F.&vanTongeren,O.F.R. Rodai, M.J.N.; Lins e Silva. A.C.B.; Pessoa. L.M. & 1995. Data analysis in community and landscape Cavalcanti, A.D.C. 2005. Vegetação e flora ccology.CanibrigeUniversityPress,Canibridge.299p. fanerogâmicadaáreadeBelânia,Pernambuco.In: Kovács-Lang,E.;Kroel-Dulay,G.;Kertész,M.;Fekete, Araújo, F.S.; Rodai, M.J.N. & Barbosa, M.R.V. G.; Bartha, S.; Mika, J.; Dobi-Wantuch, I.; Redei, (orgs.).Análisedasvariaçõesdabiodiversidadedo T.; Rajkai, K. & Hahn, I. 2000. Changes in bioma caatinga: suporte a estratégias regionais compositionofsandgrasslandsalongagradientin de conservação. Ministério do Meio Ambiente, Hungary and implications for climate change. Brasília. Pp. 91-1 19. LemoPsh,ytJ.oRc.oe&noRloodgaiia,3M0.:J3.8N5.-420070.2. Fitossociologiado RodaiC,omMp.Jo.sNi.ç;ãNoasflcoirmísetnitcoa,deL.uMm. &treMcehlood,eAv.eLg.et1a9ç9ã9.o componentelenhosodeumtrechodavegetaçãode arbustivacaducifólia,nomunicípiodeIbimirim,PE, caatinganoparque nacional da SerradaCapivara, Brasil.ActaBotanicaBrasílica 13: 1: 15-28. Piauí,Brasil.ActaBotanicaBrasílica 16:23-22. Rougerie,G. 1988.Géographiedelabiosphére.Armand Lima,J.R.;Sampaio,E.V.S.B.;Rodai,M.J.N.&Araújo. Colin Editeur,Paris.288p. F.S.2009.Composiçãoflorísticadaflorestaestacionai Sampaio, E.V.S.B. 1995. OverView of the Brazilian decíduamontanadaSerradasAlmas,Ceará,Brasil. caatinga. In: Bullock. S.H.; Mooney, H.A. & MameAdcet,aBMo.tAa.ni&caABrraasúíjloi,caF2.S3.:725060-87.63E.ffects of slash UMneidvienras.itE.ySPeraessos.naClalmybdrriydtgreop.icPapl.fo3r5e-s6L3.Cambridge andbumpractices on a soil seedbank ofcaatinga Silva. R.A.; Santos, A.M.M. & Tabarelli. M. 2003. vegetation in Northeastem Brazil. Journal ofArid Riquezaediversidadedeplantaslenhosasemcinco Environments72:458-470. unidades de paisagem da Caatinga. In: Leal. I.R.; MeiraNeto,J.A.A.;SouzaA.L.;Lana,J.M. & Valente, Tabarelli, M. & Silva, J.M.C. (eds.). Ecologia e G.E.2005. Composiçãoflorísticaespectrobiológico conservação da caatinga. Ed. Universitária da e fitofisionomia da vegetação de muçununga nos UFRPE,Recife. Pp. 337-365. Ámrunviocríep2i9o:s1d3e9C-a1r5a0.velase Mucurí, Bahia. Revista Spellerberg,I.F.&Sawyer,J.W.D. 1999.Anintroduction to applied biogeography. Cambridge University MeiraNeto,J.A.A.;Martins,F.R.&Valente,G.E.2007. Press.Cambridge.243p. ComposiçãoflorísticaespectrobiológiconaEstação van Rooyen, M.W.; Theron, G.K. & Grobbelaar. N. Ecológicade Santa Bárbara, estado de São Paulo, 1990.LiteformsandspectraoffloraofNamaqualand. MuellBrears-iDl.omRbeoviisst,aDÁ.rv&orEell3e1n5b:er9g0,7-H9.221.974. Aimsand S1o3u3t-1h45Áf.rica. Journal of Arid Environments 19: Smoentsh,odNsewofYvoergke.ta5t4i7opn.ecology. John Willey and Vieira,S.2004.Bioestatística,tópicosavançados-lestes não-paramétricos,tabelasdecontingênciaeanálise Pavón,N.P.;Hemandez-Trejo,H.&Rico-Gray,V.2000. deregressão.2ed.Elsevier,RiodeJaneiro.216p. Distributionoflifeformsalonganaltitudinalgradient Whittaker,R.H. 1975.Communitiesandecosytems.2ed. PereiJirnoau,rtnhLaeMl.soe;fmViAe-ngadertriaadtdievoa,nllSLec.yiAe.on;fceBZaa1pr1ob:to3i9st-al4,á2n.M,.MRé.xVi.co&. WhittMaakcemri,llRa.nH.Pub1l9i7s8hai.ng,ANpperwoaYcorhke.s3t8o5pc.lassifying Sampaio, E.V.S.B. 2002. Composição florística e voefgpeltaanttiocno.mImnu:nWihtiitetsa.keDrr..RW..H.Ju(endk.)P.uCbllaissshiefrisc,átTihoen análise fitossociológicadocomponente arbustivo- Hague. Pp3-3 arbóreo de um remanescente florestal no agreste 1 Paraibano.ActaBotanicaBrasílica 16:241-369 Whittaker,R.H.1978b.Dominacc-types.In:Whittaker.R.H. Raunkiaer,C. 1934.Thelifeformsofplantsandstatistical P(eudb.l)i.sChlearsss.iTfhiceaúHoanguoef.plPapn.t6c5o-m7m9u.niúes.Dr.W.Junk Rodaip,laMn.tJg.eNo.;grAanpdhrya.dCel,aKr.eVn.dAo.n;PSraelsess,,OMx.fFo.rd&.6G3o2mpe.s, WilliGa.mDs..R1.J9.9;6D.ufVfa,rGi.aAt.i;onBoiwnmtahne,cDo.mMp.oJs.iSti&onCoaonkd. d5BA8.ue:Pí.quS5u.1me7,1-9rP59ee28fr6.ún.gFaiimtobousvcseoogc.ieoRtleaovcgiiisoatnadaoBlrcaonsmiolpeominrueannditeceíBlipeoinlohoogdsieoa aastnnrdducsstoouiirllettoeexfxtttuurrroeepiaicnlaoltnhsgeaaNvoalrnatnrhagees-rasnscaTaleefrurcinltciotmriayot,nicAoufgsrrtaaridanilfeianalt.l JournalofBiogeography23:747-756. Artigorecebidoem 09/05/2010. Aceitopara publicaçãoern 18/12/2010. Rodriguèsia62(2):341-366.201 SciELO/JBRJ cm 12 13 14 15 16 17