almeida 2004

8/18/2019 Almeida 2004

1/19

Geographical distribution of Eupatorieae ( Asteraceae) in South-eastern

and South Brazilian Mountain Ranges

Adriana M. Almeida1,*, Paulo I. Prado2 and Thomas M. Lewinsohn1,21 Departamento de Zoologia, CP 6109, IB, Unicamp, Campinas, SP, Brazil, CEP 13083-970;2 NEPAM,Unicamp, Campinas, SP, Brazil; * Author for correspondence ([email protected])

Received 17 July 2002; accepted in revised form 18 August 2003

Key words: Biogeography, Brazilian mountain ranges, Mantiqueira range, Mountaintops, Species distribution,Species occurrences

Abstract

This study deals with the geographical distribution of Eupatorieae species in South-eastern and Southern Brazil-ian mountain ranges, with special emphasis on the Mantiqueira range Serra da Mantiqueira. In the Mantiqueirarange, five localities were sampled for two years during the flowering period of the plants. Species compositionin the Mantiqueira localities was then compared to data from a central range the Espinhaço and a southerncoastal range. Cluster and ordination analyses showed clear floristic differences among the three ranges. Nearerlocalities were more similar floristically only within the Espinhaço range and across all ranges, as shown byMantel tests. The dendrogram in the Mantiqueira range clustered Itatiaia and Campos do Jordão, two non-adja-cent localities with similar environment, landscape and topography. Different processes may be influencing thestructure and composition of the Eupatorieae flora at different geographical scales. Historical and biogeographi-cal processes may be more important in determining community composition of different mountain ranges,whereas factors such as climate and human impact may be determining differences among localities within each

mountain range.

Introduction

In recent years ecologists seeking for patterns of community structure have shifted their attention fromthe local to the regional scale e.g., Ricklefs andSchluter 1993; Brown 1995; Brown et al. 1996; Ca-ley and Schluter 1997. Local communities are notisolated ecological units. Regional and historical pro-cesses operating on larger temporal and spatial scalesare required to explain the local abundance, distribu-tion and diversity of species Ricklefs and Schluter1993; Brown 1995.

The inability derive global patterns of biodiversityfrom local environmental conditions alone has led re-searchers to search for other explanatory concepts.Following the reasoning of island biogeographyMacArthur and Wilson 1963; MacArthur and Wilson

1967, one can investigate the patterns of conver-gence of biodiversity among ecologically similar butgeographically separate localities, setting local as-semblages into the context of regional processesRicklefs and Schluter 1993.

Islands and mountaintops are outstanding systemsin which to address this problem, both showing theimprint of physical isolation. In some cases, such astropical East Africa, where some mountains reachmore than 4000m above sea level, mountaintops aremore effectively isolated than oceanic islands Hed-berg 1970; no diaspore of a mountain plant will nor-mally be carried across the savannah by an agentequivalent to oceanic currents, and no plant specieswill survive in the intervening savannah valleys.

Plant species occurring in mountain ranges facecertain problems that restrict the occurrence areas of

Plant Ecology 174: 163–181, 2004.© 2004 Kluwer Academic Publishers. Printed in the Netherlands.

163

8/18/2019 Almeida 2004

2/19

most species. Abiotic factors e.g., climate Bruijnzeeland Veneklaas 1998; Safford 1999a or soil conditionsSmith 1994; Sollins 1998; Tanner et al. 1998 changealong with altitude. Small areas with homogeneousconditions are limited colonization targets, often re-stricting species to a small altitudinal range of a par-

ticular mountain, sometimes even to a single slope.Thus, a species able to disperse through various alti-tudes is of necessity a habitat generalist. Furthermore,to disperse among different mountains, a species mustbe capable of colonizing intermediate valleys habitatgeneralists, have extraordinary dispersal abilitythrough air flotation Hedberg 1970 or long-distancedispersing agents. Suitable areas for high-altitudespecialists are often too small to support indefinitelya population, which will then depend on immigrationto persist the ‘mass effect’ of Shmida and Wilson1985.

The present study is centred in campos fields,

open physiognomies that occur above treeline c.a.1,000m in three Brazilian mountain ranges that ex-tend from the state of Bahia to northern Rio Grandedo Sul: the Espinhaço range in the states of Bahiaand Minas Gerais, but we studied only in the latter,the Mantiqueira range in Minas Gerais, Rio de Jan-eiro and São Paulo and the Southern Coastal rangestudied in Santa Catarina and Rio Grande do Sul.Within each mountain range the campos are separatedby deep and extensive river valleys in numerouspatches and so present a discontinuous distribution.

Botanists and phytogeographers distinguish “cam-

pos de altitude” – high altitude grasslands – from“campos rupestres” – rocky grasslands – from “cam-pos de cima da serra” – fields above Araucaria for-ests. The first term is reserved for plateaus mainly inthe Mantiqueira range above treeline Safford 1999a;the second is used for fields or higher slopes andmountaintops in central Brazilian mountain ranges,especially in the Espinhaço range Giulietti and Pirani1988; the last term is used for fields above Araucariaangustifolia Bertol. O.Kuntze forests in Santa Cata-rina and Rio Grande do Sul, with common frosts inwinter Menezes 2000.

Although recognized as different physiognomies,these montane campos have much in common. Theyare composed basically of herbaceous vegetation withxeromorphic perennials, mainly grasses and herba-ceous Asteraceae and Melastomataceae, with manyendemic species e.g., Safford 1999a, b. Accordingto Safford 1999a the campos de altitude or “Bra-zilian páramos” in the Mantiqueira range are among

the most diverse natural communities in extra-Ama-zonian Brazil, with very high numbers of endemicspecies Safford 1999a,b; yet, till now they havebeen relatively neglected by ecologists and conserva-tion biologists in contrast to the Atlantic rainforestand the cerrado savannah woodlands.

These areas have been visited and collected bymany botanists, which has resulted in an extensiveliterature comprised of local species lists and taxo-nomic descriptions of particular families. The distinc-tion of the campos in the southern Brazilian mountainranges into distinct vegetation formations is based ongeographical location, their underlying geology andtheir floristics, especially their unique or typical taxae.g., Martinelli 1989; Rizzini 1995. Brazilian mon-tane regions are thus still largely lacking in biogeo-graphic studies and, in the Neotropical domain, theyare far less known than several ranges in Mexico andwithin the Andes e.g., Rundel et al. 1994. This pa-

per, to our knowledge, presents the first comparisonamong areas based on a common taxon that is wellrepresented in all areas. Moreover, our data comefrom comparable samples that, although not strictlystandardised, can be investigated by multivariatemethods commonly used in vegetation ecology.

Asteraceae is the largest family of dicotyledonousplants with about 23,000 described species widelydispersed through all environments and continentsexcept Antarctica Bremer 1994, though they aremore abundant and diverse in open and /or non-for-ested areas. The tribes of Asteraceae are well estab-

lished natural subdivisions Heywood et al. 1977;Jansen et al. 1991; Bremer et al. 1992; Bremer 1994.The tribe Eupatorieae is pantropical and the most di-verse in Brazilian open areas, and its taxonomy hasrecently been reviewed and is now well establishedBarroso et al. 1986; King and Robinson 1987. Thetribe Eupatorieae is therefore well-suited to examinelocal assemblages and their variation in our region of interest.

Methods

Study areas

The Mantiqueira range is divided in two distinct geo-morphological units, the Campos do Jordão and Ita-tiaia massifs Gatto et al. 1983. The Campos doJordão massif encompasses the states of Minas Geraisand São Paulo; our sampling locality of Campos do

164

8/18/2019 Almeida 2004

3/19

Jordão and part of the sites of Passa Quatro are in-cluded in this unit. The Itatiaia massif reaches thestates of Minas Gerais and Rio de Janeiro; it includesthe sampling localities of Ibitipoca, Visconde deMauá, Itatiaia and part of Passa Quatro see samplingmethods for a description of collection sites.

The Serra da Mantiqueira extends roughly in theNE-SW direction, bordering on the states of Rio deJaneiro, Minas Gerais and São Paulo. The climate issub-humid to humid. The tree line in Itatiaia averagesbetween 2000m and 2200m, which is considered lowfor its latitude Körner 1998. Above treeline, annualprecipitation ranges from 1500-2000mm in the cam-

pos de altitude of São Paulo and Minas Gerais, to2000-3000mm in Rio de Janeiro. Maximum precipi-tation occurs in January mid-summer and minimumprecipitation in July winter. The dry season is char-acterized by less than 50mm precipitation and lastsfrom one to three months June-August in all the

southeastern mountains of Mantiqueira Safford1999a,b. Although it has been speculated that theItatiaia summits suffered glaciation during the Pleis-tocene, there is no solid evidence for this Safford1999a. Frost occurs on average 56 days per year at2200m in Itatiaia and at 1600m in Campos do Jordão,mostly during the dry winter July. Drought may bepartly offset by orographic fog, which in Itatiaia oc-curs 218 days per year at 2200m. In Itatiaia, monthlyaverage of air humidity ranges from 65% to 90%Safford 1999a.

Above the tree line we find the campos de altitude,

“a series of humid, subalpine grasslands restricted tothe highest peaks and plateaus of the South-easternBrazilian Highlands. Comprising a classic terrestrialarchipelago of isolated, mountaintop habitats, thesesystems form the highest, coldest orobiome in east-ern South America” Safford 1999a, b. The threelargest families in Itatiaia and presumably in the en-tire Mantiqueira are the Asteraceae, Polypodiaceaes.l. and Melastomataceae, summing about 40% of theplant species. In Itatiaia, the savannah-like associa-tions dominated by Asteraceae, especially the genus

Baccharis and tribes Eupatorieae and Vernonieae, areprobably the most species-rich formation in the cam-

pos. About a third of the vascular plant species in theItatiaia campos de altitude appear to be endemic tothis physiognomy Safford 1999a.

The Serra do Espinhaço Espinhaço Range ex-tends from Northern Bahia southward to the Serra doOuro Branco in the state of Minas Gerais. Its upperreaches are covered by campos rupestres lying

between 700 and 2,000m in elevation Giulietti andPirani 1988. The climate is mesothermic with mildsummers accompanied by a rainy season. The drywinter season lasts 3 to 4 months, in which frosts canoccur, but not as frequent as in Mantiqueira or theSouthern range. The average annual temperature var-

ies from 17° to 20 °C. The campos rupestres are in-terspersed with cerrado patches with denser shrubsand sparse treelets Davis et al. 1997. As in theMantiqueira, the campos rupestres vegetation is com-posed mainly of grasses and asteraceans.

In Southern Brazil the campos de cima da serravegetation occurs in montane areas above 1000m inelevation, above Araucaria angustifolia forests, ex-tending through the states of Santa Catarina andnorth-eastern Rio Grande do Sul. These regionspresent frosts very often during winter Menezes2000.

Sampling methods

For present purposes, we define region as one moun-tain range. Locality is a regional subdivision that in-cludes a group of sampling sites. A site is each pointwhere the Eupatorieae assemblage was sampled. Forinstance, Mantiqueira range is considered a regionand Ibitipoca is called a locality with severalsampling sites.

Five localities were sampled in each mountainrange see Figure 1 for localities’s codes andgeographical distribution and Table 1 for coordi-

nates. Since Mantiqueira flora will be described inmore detail, we detail here only the five localitiessampled within this mountain range:

IB – Ibitipoca State Park, in the State of MinasGerais. This is the northernmost part of the Man-tiqueira and also the locality closest to the Espinhaçorange. Our sampling sites mostly comprised camposde altitude.

VM The Visconde de Mauá, on the border be-tween the States of Minas Gerais and Rio de Janeiro,is an important sampling area on the slope oppositeto the Itatiaia National Park. Since we could not reachthe mountaintops, we sampled only in open siteswithin the highland forest belt, close to but not withinthe campos de altitude proper.

IT – Itatiaia National Park , which also lies on Stateborder between Minas Gerais and Rio de Janeiro, in-cludes the Agulhas Negras peak, the second highestMantiqueira summit at 2789m. On its southwesternslope and highland plateaus we had access to its

165

8/18/2019 Almeida 2004

4/19

campos de altitude, the most extensive in the Man-tiqueira.

PQ – Passa-Quatro in Minas Gerais. Samplingsites in this locality were spread out, along dirt roadsand tracks in the mountains. This area is more inhab-ited and the landscape is dominated by secondaryhighland forest and small-cultivated holdings.

CJ – The Campos do Jordão locality is situated inSão Paulo state, intruding slightly into Minas Geraisin the direction of Itajubá. Our sampling sites were incampos de altitude or in highland forest.

Whenever possible, we preferred to travel betweenlocalities on dirt roads within the mountains, alongwhich we could take additional samples. Linear dis-tances between pairs of localities varied from 26 kmbetween Visconde de Mauá and Itatiaia, to 203 kmbetween Ibitipoca and Campos do Jordão, the twoextreme sampled mountains.

Six field trips were carried out in the Mantiqueirarange, from January to June in 1998 and 1999, en-compassing the flowering and fruiting of all Eupato-rieae. In only one trip not all localities were surveyed.In each of the five localities a minimum of fifteensites were chosen for sampling plants, with a mini-mum of 1km exceptionally, 500m spacing amongthem. In each site, all flowering and/or fruiting Eupa-torieae species were sampled. Although each samplesite was chosen for the presence of Eupatorieae, allflowering and/or fruiting Asteraceae species presentin the site had at least one voucher specimencollected. Exact geographical coordinates of each sitewere obtained with a GPS receiver.

Data analysed for the Espinhaço range in centraland northern Minas Gerais are from 1995-1996samples in February, April, July and September, andfor south Brazilian ranges in February, April and Oc-

Figure 1. Sampling localities in the Espinhaço, Mantiqueira and Southern ranges. Espinhaço localities are: GMOG – Serra do Grão Mogol,MG, SCAB – Serra do Cabral, MG, DIA – Planalto de Diamantina, MG, SCIP – Serra do Cipó, MG and OB – Serra do Ouro Branco, MG.Mantiqueira localities are: IB Ibitipoca, MG; VM Visconde de Mauá, MG/RJ; IT Itatiaia, RJ/MG; PQ Passa Quatro, MG; CJ Campos do Jordão, MG/SP. Southern localities are: MC – Matos Costa SC; LG – Lages SC; CER – Cerrito SC; BJ – Bom Jardim SC;CAMB – Cambará do Sul RS. States are: MG – Minas Gerais, RJ – Rio de Janeiro, SP – São Paulo, SC – Santa Catarina, RS – Rio Grande

do Sul. See Table 1 for coordinates of reference points.

166

8/18/2019 Almeida 2004

5/19

tober in the same years. In these regions we used thesame sampling methods, but focused on the five ma-

jor Brazilian tribes of Asteraceae, including the Eu-patorieae e.g., Prado 1999; Prado and Lewinsohn2000.

Species assignment

Voucher specimens were first compared and assignedto morphospecies and later to species, with nomen-clature following King and Robinson 1987 and ref-erences therein. Although in the field we collected

separate vouchers for any difference noted, slightmorphological variants were later grouped and treatedas single species, so that our operational taxonomywas mostly conservative to avoid undue splitting andinflating differences among sites and localities. Spe-cialists see acknowledgements confirmed, correctedand completed our species identifications.

Six probably new Eupatorieae species, almost allfrom the Southern Brazilian range, were only identi-fied to tribe see Appendix. Since they are differentmorphospecies from all the other identified ones, theywere included in the analyses.

Data analysis

The floristic dissimilarity between all pairs of locali-ties was expressed by the relativized Euclidean dis-tance of the incidence of each plant species perlocality i.e., the number of sites where the species

was sampled per locality. Correlation of floristic dis-similarity and geographical distance among localitieswas evaluated with the Mantel test McCune andMefford 1999. The Mantel p value was obtained witha randomisation Monte Carlo test for the pool of thethree ranges, and with Mantel’s asymptotic approxi-mation for each range, since five localities is a smallsample size, resulting in a small number of possiblepermutations of the data. The geographical distancebetween two localities was calculated as a simplegeometric linear distance. The centroid of samplingsites for each locality was used as reference point for

that locality. Over the range of distances consideredand the precision of the data set, geodesic correctionwas not deemed necessary.

Localities were clustered using relativized Euclid-ean distances for species frequencies within localities,with the UPGMA aggregation method Digby andKempton 1987.

Localities were also subjected to ordination by de-trended correspondence analysis DCA, using theoccurrence of each plant species per locality as a sur-rogate measure of local abundance. DCA is animprovement of ordinary Correspondence Analysis in

which putative distortions are removed by rescalingof individual axis segments Gauch 1982; Digby andKempton 1987. Since DCA is sensitive to low val-ues, species were selectively weighted: abundances of all species rarer than 20% of the frequency of thecommonest species were downweighted in proportionto their frequency, thus reducing the effect of these

Table 1. Codes for the localities within three Brazilian mountain ranges, with their name, main county, state, and reference coordinate of thelocality centroid, used to measure distances among localities. States are: MG – Minas Gerais, RJ – Rio de Janeiro, SP – São Paulo, SC –Santa Catarina, RS – Rio Grande do Sul.

Mountain Range Code Name Municipality State Lat S Long W

Espinhaço GMOG Grão Mogol Grão Mogol MG 16° 32.64' 42° 54.87'SCAB Serra do Cabral Joaquim Felício MG 17° 42.95' 44° 14.51'DIA Diamantina Diamantina MG 18° 15.20' 43° 41.10'

SCIP Serra do Cipó Santana do Riacho MG 19° 17.29' 43° 35.45'OB Ouro Branco Ouro Branco MG 20° 30.05' 43° 39.71'

Mantiqueira IB Ibitipoca Lima Duarte MG 21° 41.58' 43° 52.73'VM Visconde de Mauá Bocaina de Minas MG 22° 14.38' 44° 29.21'IT Itatiaia Itatiaia RJ 22° 22.64' 44° 41.79'PQ Passa Quatro Virgínia MG 22° 25.66' 45° 04.51'CJ Campos do Jordão Campos do Jordão SP 22° 39.48' 45° 33.41'

South MC Matos Costa Matos Costa SC 26° 29.18' 51° 07.63'CER São José do Cerrito Lages SC 27° 47.09' 50° 29.61'LG Lages Lages SC 27° 55.30' 49° 59.30'BJ Bom Jardim Bom Jardim da Serra SC 28° 14.76' 49° 36.91'CAMB Cambará do Sul Cambará do Sul RS 29° 08.61' 50° 04.77'

167

8/18/2019 Almeida 2004

6/19

rarer species on the final configuration McCune andMefford 1999.

Analyses were performed with Systat© SPSS Inc.1997 and PC-Ord© McCune and Mefford 1999.

Results

Eupatorieae in the Mantiqueira range

In the Mantiqueira range we obtained a total of 596Eupatorieae collections from 56 species, within anoverall elevational range from 760m to 2460m. The

56 species belong to 9 subtribes and 17 genera. Themost speciose subtribe and genus in the area was Mi-kaniinae, with 16 Mikania species. The second sub-tribe in species number was Praxelinae, with 12species, 11 in the genus Chromolaena. At the otherextreme, the tribe Adenostemmatinae was represented

by a single species, Adenostemma brasilianum Cass.and seven genera belonging to various subtribes wererepresented each by a single species Appendix.

Not all subtribes and genera were present in all lo-calities Figure 2a,b. Adenostemma brasilianum wasfound only in Visconde de Mauá, while the subtribeAyapaninae, with three sampled species, was not

Figure 2. Occurrence of Eupatorieae subtribes a, genera b and species c in localities of the Mantiqueira range. Note that the number of species that are singletons is higher than the number of species recorded in four or five localities.

168

8/18/2019 Almeida 2004

7/19

sampled in Visconde de Mauá and Passa Quatro, andthe subtribe Critoniinae, with two species was notfound in Itatiaia and Passa Quatro. Eight genera47% occurred in all localities, whereas four generaoccurred in only one locality. Only five species 9%occurred in all localities throughout the Mantiqueira:

Ageratum fastigiatum Gardner R. King and H. Rob-inson , Austroeupatorium silphiipholium Martius R.King and H. Robinson, Campovassouria cruciataVell. Conc. R. King and H. Robinson, Campuloclin-ium purpurascens Schultz- Bip. ex Baker R. Kingand H. Robinson and Chromolaena laevigata Lam.R. King and H. Robinson see Appendix for a com-plete list of species and their localities. The five spe-cies are widespread in other Brazilian mountainsCabrera and Klein 1989; Lorenzi 1991; Prado 1999.

The richest localities were Ibitipoca, with 157 col-lections and 32 species, and Campos do Jordão, with129 collections and 32 species, followed by Visconde

de Mauá with 144 collections and 25 species. Sub-stantially lower species richness was found in Itatiaia,with 89 collections and 20 species, and in Passa Qua-tro, with 77 collections and 16 species. The plantspecies richness recorded in each locality tended tostabilize with greater sampling effort. The log of totalnumber of records per locality accounted for 78% of local recorded species richness r2 0.785;p 0.05.

Although we found 56 species altogether in theMantiqueira localities, the maximum in one localitywas 32 species in Campos do Jordão and Ibitipoca,

which suggests a fairly high species turnover amonglocalities. In fact, a high proportion of the species 22species, 39% was found in a single locality, eight of which were singletons – were collected only once –in the Mantiqueira range. Most species showed a re-stricted distribution and only 10 species 18% werefound in four or five localities Figure 2c.

It is worth noting that Praxelinae, the most widelydistributed and second most speciose subtribe in thestudied ranges, was represented in Itatiaia by only onespecies with three records Chromolaena laevigata,while on the opposite slope of the same mountain, inVisconde de Mauá, we obtained six species from thissubtribe Appendix.

Some common and widespread species were notfound in one or more localities in the Mantiqueirarange. Itatiaia was the only locality previouslysurveyed for Asteraceae, where Barroso 1957 listed42 Eupatorieae species. Sixteen of these species38% were not found in the localities we surveyed

from which more than half 56% belong to the ge-nus Mikania; 15 species were collected in Itatiaiaand 10 were collected in the Mantiqueira range butnot in Itatiaia itself. Mikania vitifolia DC., listed byBarroso 1957 in the Itatiaia, was recorded by usonly once in Bom Jardim in the Southern Brazil

mountain ranges. A more recent paper with a non-comprehensive floristic survey Barros et al. 1998cites eight Eupatorieae species occurring in Itatiaia,all of them previously cited by Barroso 1957. Onthe other hand, we collected six species in Itatiaia thatwere not cited by either of the preceding studies.

Barroso’s 1957 pioneering paper was based onspecimens amassed in a number of botanical excur-sions that covered the Itatiaia Park more widely andencompassing more habitats than we did; thus, sev-eral of the species we did not detect are vines in rain-forest gaps or edges at lower elevations. For instance,

Ageratum conyzoides L., a widespread weed in Bra-

zil Lorenzi 1991 listed by Barroso 1957 in the Ita-tiaia park, was not found in the sites sampled for thisstudy, although it was previously noted in roadsideclearings, camp sites and other disturbed areas in thelower part of Itatiaia Lewinsohn, unpublished. An-other widespread species, Trichogoniopsis adenanthaDC. R. King and H. Robinson =Trichogonia gard-neri, which in the Mantiqueira range was found onlyonce in Ibitipoca, is a common and fairly widespreadspecies Lewinsohn, unpublished.

The montane Eupatorieae flora in south-eastern and

southern Brazil

A total of 2351 records of 534 species in 15 Aster-aceae tribes were obtained in the three studied ranges.The tribe Eupatorieae represents almost half 1051vouchered collections or 45% of samples and morethan a quarter of the species 144 species or 27% re-corded. Both the Espinhaço and the Southern rangeshad a high proportion of singletons, respectively 40%and 41% of the Eupatorieae species recorded. Whenthe three ranges are considered, the Eupatorieaesingleton species in the Mantiqueira are reduced tosix species, or four percent of species from the threeranges.

In the Espinhaço range we obtained 1081 Aster-aceae collections belonging to 277 species, while inthe Southern mountains we obtained 442 collectionsand 139 species and in the Mantiqueira range we ob-tained 828 collections and 149 species Table 2.Since in the Mantiqueira we focused on Eupatorieae

169

8/18/2019 Almeida 2004

8/19

species, records for this tribe represented almost 75%of all collections in this range, while in both the Es-pinhaço and Southern ranges Eupatorieae collectionsrepresent about 30% of plant collections. Proportional

Eupatorieae species richness, on the other hand, var-ied less, from 24.5% of species in the Espinhaço to33% in the South and 37.6% in the Mantiqueiraranges Table 2.

The tribe Eupatorieae in the three mountain rangesis represented by 10 subtribes and 26 genera. Thesubtribe Alomiinae is the only one not represented inthe Mantiqueira, and is represented by one species,Pseudobrickellia brasiliensis Sprengel R. King andH. Robinson, present in four localities in the Espin-haço range. This genus is restricted to mountains inMinas Gerais and Goiás King and Robinson 1987.The most speciose subtribe is Mikaniinae with 43

Mikania species, followed by the subtribe Praxelinaewith 29 Chromolaena species. Conversely, eight gen-era from various subtribes are represented by singlespecies: Pseudobrickellia brasiliensis, Ophryosporus

freyreysii Thunb. and Dallm. Baker , Campovassou-ria cruciata, Hatschbachiella tweediana Hook andArn. R. King and H. Robinson , Stomatanthes

polycephalus Schultz-Bip. Ex B. Robinson H. Rob-inson , Gyptis crassipes Hieron. R. King and H.Robinson, Trichogoniopsis adenantha and Vittetia or-biculata DC. R. King and H. Robinson.

The Eupatorieae species turnover across the three

mountain ranges is fairly high. From the entire poolof 144 Eupatorieae species we found, 120 83% Eu-patorieae species were found in a single mountainrange, whereas only two species, Mikania micranthaH. B. K. and Chromolaena laevigata occurred in allthree ranges Figure 3. Only 13 species 9% areshared solely between the Mantiqueira and Espin-

haço, eight 5.5% co-occur only in the Mantiqueiraand Southern ranges, and a single species, Raulinor-eitzia tremula Hook and Arn. R. King and H. Rob-

inson occurred in the Espinhaço and Southern rangesbut was not found in the Mantiqueira Figure 3, andsee Appendix for details.

No subtribe was restricted to the Mantiqueira, butthree genera, each represented by one species werefound only in this mountain range: Ophryosporus

freyreysii, Adenostemma brasilianum and Trichogoni-opsis adenantha. The last two species are commonlyfound in other Brazilian open formations or in gapsand roadsides within midaltitude forests e.g., Cabreraand Klein 1989; Lorenzi 1991.

Few species were widespread throughout differentlocalities, within and among mountain ranges. Chro-molaena laevigata occurred in 13 localities but notin Bom Jardim and Cambará, the two southernmostlocalities, followed by Ageratum fastigiatum andCampovassouria cruciata, each of which occurred in10 localities. While A. fastigiatum was collected in alllocalities in the Espinhaço and Mantiqueira rangesbut was not found in the Southern range, C. cruciata

Table 2. Number of collections and species of the main Asteracean tribes in three Brazilian mountain ranges. Espinhaço and Southern rangeswere collected in 1995 and 1996, whereas Mantiqueira was collected in 1998 and 1999. Eupatorieae tribe is in bold. “Others” include thetribes Cardueae, Gnaphalieae, Inuleae, Lactuceae, Helenieae, Plucheae, Moquiniae, Tageteae and the subfamily Barnadesioideae.

Mantiqueira Espinhaço Southern Range

Tribe Collections Species Collections Species Collections Species

Astereae 70 33 57 17 88 24

Eupatorieae 601 56 315 68 135 46Heliantheae 15 7 89 39 20 9Mutisieae 10 7 93 21 38 14Senecioneae 23 15 31 7 67 15Vernonieae 97 19 462 113 61 14Others 12 12 34 12 33 17Total 828 149 1081 277 442 139

Figure 3. Venn diagram of the number of Eupatorieae species re-corded in three Brazilian mountain ranges, showing species sharedamong ranges.

170

8/18/2019 Almeida 2004

9/19

occurred in all localities in the Mantiqueira and

Southern ranges but not in the Espinhaço range.

Intra and interregional similarity of Eupatorieae

assemblages

From a total of 144 Eupatorieae species found, only24 17% are shared among regions. Floristicdissimilarities increased with geographical distancesamong localities in the Espinhaço and across allranges Figure 4. The Mantel test showed a signifi-cant positive correlation between floristic and geo-graphical distances, in the Espinhaço range r 0.906, p 0.05 and also for the pool of the threemountain ranges r 0.701; p 0.001. No corre-lation was found in the Mantiqueira r 0.111; p 0.70 and Southern r 0.041; p 0.80 rangesFigure 4. Note that the plot for pooled ranges inFigure 4 shows two discrete groups: a first more dis-persed group in relation to the relative Euclidean dis-

tance that encompasses pairs of localities that are less

than 700 km apart. The second group forms a moreconcentrated group of points in relation to floristicdistance, and includes all pair of localities with morethan 1,000 km geographical distance. The first groupshows mostly dissimilarities within mountain ranges,whereas the second group shows dissimilaritiesbetween localities belonging to different mountainranges.

In the cluster analysis, localities of the three moun-tain ranges were clearly separated Figure 5. In thecluster dendrogram, the Espinhaço clusters with theMantiqueira and then with the Southern range. All the

joinings found within single mountain ranges con-form well to geographic distance Figure 1, with theexception of Itatiaia – Campos do Jordão in the Man-tiqueira, non-adjacent localities with similar physiog-nomies and human activity, and Lages – Matos Costain the Southern range. In the Southern range, Cerritowas also displaced in relation to its geographical po-

Figure 4. Floristic dissimilarity determined by Relativized Euclidean distance of species composition for pooled or individual mountainranges plotted against geographical distance km. The significance levels reported were obtained with a Mantel test. Note the difference inscale in the abscissa.

171

8/18/2019 Almeida 2004

10/19

sition, probably because of the many exclusive Eupa-torieae species found in that locality.

An alternative cluster analysis with Euclidean Dis-tance on standardized data using the adjustment tostandard deviate, with b x ¯ * si

–1 and UPGMA

not shown clustered with the Mantiqueira andSouthern ranges and then to Espinhaço. However,within each mountain range, localities were groupedexactly as in Figure 5. This shows that the floristicdissimilarity among the three ranges is sensitive to thedistance measure used, and reinforces the intermedi-ate position of the Mantiqueira between the Espin-haço and Southern ranges.

Ordination by DCA also split localities clearlyamong the three mountain ranges according to theirEupatorieae flora Figure 6. The first axis had aneigenvalue of 0.791 and grouped the localities fromeach region.

Both cluster and ordination analysis clearly sepa-rated the campos of the three mountain rangesaccording to its Eupatorieae flora, reinforcing thesplitting of the campos in three physiognomies.

Discussion

Within the Asteraceae, the tribe Eupatorieae had thehighest number of records and species in the three re-gions, with exception of the Vernonieae in the Espin-

haço range Prado 1999. These two tribes, followedby the Heliantheae and Asteraeae, are indeed thelargest Asteraceae tribes in Brazil Barroso et al.1986.

Within the Mantiqueira range, only Itatiaia hadbeen previously studied for its Asteraceae flora Bar-roso 1957. In Itatiaia we obtained 19 Eupatorieaespecies of which six 32% were previously unre-corded, whereas 38% of the species previously listedin this locality were not observed in the present study.This is mainly due to the fact that Barroso 1957 alsocollected in lower altitudes occupied by humid for-ests; these probably harbour the many Mikania spe-

cies we did not observe. In comparison with the otherfour Mantiqueira localities, the Itatiaia higher slopeand plateau has the harshest climate, presenting lowertemperatures with recurrent winter frost and strongwinds Barros et al. 1998; Safford 1999a. For thatreason the flowering period for Asteraceae in this lo-cality is probably narrower than in other localities

Figure 5. Hierarchical cluster diagram for localities from Espinhaço, Mantiqueira and Southern ranges. The clustering method used UPGMAwith Relativized Euclidean Distance. Mountain ranges are assigned different symbols: Espinhaço range, triangles; Mantiqueira range, spheresand South ranges, squares. See Table 1 for locality codes.

172

8/18/2019 Almeida 2004

11/19

with milder climate, and therefore some Eupatorieaespecies may have eluded us by not flowering or fruit-ing during our sampling periods. Such a reducedflowering period was shown by Jonas and Geber1999 for Clarkia unguiculata Lindl. Onagraceaein the upper Sierra Nevada.

A high proportion of Eupatorieae records aresingletons, varying from 14% in the Mantiqueira to38% in the Espinhaço. Since in the Mantiqueira rangewe were sampling preferentially Eupatorieae specieswe always searched carefully for rare species in thesampled sites, which may have reduced the numberof recorded singletons in this range compared to the

other sites. However, since the sampling protocol wasconsistent among regions, we have no indication thatthis was the case.

Only two Eupatorieae were shared among the threestudied mountain ranges, and even the two closestranges, Mantiqueira and Espinhaço have only 15 spe-cies 14% in common. With few species sharedamong mountain ranges, the Eupatorieae flora wasclearly differentiated among the ranges, as shown byboth clustering and ordination analyses. In the clusterdendrogram, the Mantiqueira range grouped eitherwith the Espinhaço or with the Southern range, de-

pending on the distance measure used. This result re-inforces the intermediate position of this range, andmatches its geographical position as well. Althoughthe Mantiqueira is geographically much closer to theEspinhaço range than to the Southern range, its floraturned out to differ fairly equally from both ranges.

The campos vegetation present in the studiedmountain rages harbours discrete and distinct sets of Eupatorieae species, which are not directly related togeographical distance. This result suggests that eachmountain range is a discrete ecological unit. Althoughbiogeographers have considered the campos areas of each mountain range a distinct phytophysiognomy,their classification is somewhat controversial. Ac-cording to Safford 1999b, the Brazilian campos dealtitude proper are almost entirely restricted to an areaof less than 350 km2 in southeastern Brazil with anoutlier in the state of Santa Catarina. This outlier in-cludes what we consider here as campos de cima da

serra, and is not distinguished by Safford. The presentstudy helps to clarify this distinction.

Within each mountain range we find the same pat-tern of many species restricted to only one or two lo-calities. Species could be found in adjacent mountainsor not. For example, for the group of species occur-ring in only two localities in the Mantiqueira range,the proportion of species in adjacent localities is 14%,while the proportion of species occurring only in thetwo most extreme localities Ibitipoca and Campos doJordão is 50%. This is corroborated by the Manteltest, which detected no correlation between floristic

and geographical distance in either the Mantiqueira orSouthern range. In these two mountain ranges thefloristic dissimilarity between any pair of localitiesdid not depend on the geographical distance. Thecorrelation of geographical and floristic distances wasonly significant within the Espinhaço range. How-ever, the same analysis on the tribe Vernonieae in the

Figure 6. Axis 1 and 2 of a Detrended Correspondence Analysis DCA applied to Eupatorieae species collected in localities from threeBrazilian mountain ranges. Rare species were downweighted. Mountain ranges are assigned different symbols: Espinhaço range, triangles;Mantiqueira range, spheres; South ranges, squares. See Table 1 for locality codes.

173

8/18/2019 Almeida 2004

12/19

same Espinhaço localities was not significant P.Prado, unpublished, which shows that differentgroups respond differently to the same variables.Moreover, pooling localities among the three rangesshowed the Eupatorieae flora to be more similar innearer localities. The two distinct groups in Figure 4

for pooled localities shows that, on average, any pairof localities within a mountain range is floristicallymore similar than pairs of localities from differentmountain ranges. This shows up even more clearly inordination Figure 6.

The cluster dendrogram showed that geographicaldistance was a good predictor of floristic compositionboth for the Espinhaço range and for the pool of thethree ranges. In the Mantiqueira range, Itatiaia wasgrouped with Campos do Jordão, both localities withsampling sites in the campos de altitude proper,which did not occur in Visconde de Mauá and PassaQuatro. These two localities are most affected by hu-

man activity, which may reflect on its flora. In theSouthern range, Cerrito was a unique locality, withmany exclusive Eupatorieae species. This result sug-gests that environmental variables such as climate,altitude and human impact may be influencing thefloristic composition of localities at the more localscale.

At least three previous studies in tropical montaneforests observed adjacent localities have more similarfloras. Hedberg 1970 studied the alpine flora in Af-rica and concluded that the constituent speciesdispersed one by one, in stepping-stone “jumps” from

one mountain to another, and that the associationswere independently formed on each mountain. Smith1975, in the best studied New Guinea ranges,showed that adjacent mountains have a higher herba-ceous angiosperm similarity than more distant sites.Simpson and Todzia 1990 compared four localitiesin the high Andean flora in South America and twohigh-elevation floras in North America and detectedthat, at a generic level, the alpine flora is most simi-lar to that of the nearest locality within the same con-tinent. Two families in the high Andean flora,Asteraceae and Poaceae, constitute over one-third of all the species in the studied localities. Although bothfamilies characteristically have wind-dispersed fruits,successful colonization of the Andes by northtemperate genera depended on similarities in climateas much as, or more than, distance. The three studiesconcluded that, though distance between localities isof prime importance, climate could be equally influ-

ential in determining the composition of adjacentmontane floras.

Nekola and White 1999 studied the distance de-cay hypothesis for plant species divided into growthform and dispersal type classes in Northern NorthAmerica. They found that wind-dispersed species had

lower rates of distance decay than larger fruited spe-cies, being more widely dispersed and more variablein their occurrence. They also found the highest ratesof similarity loss with distance in herbs compared toother growth forms. These results may not apply inour case. The Eupatorieae have plumose wind-dispersed fruits, but encompass different growthforms, from herbs e.g., Ageratum conyzoides andvines e.g., Mikania spp. to trees e.g., Austroeupa-torium silphiifolium. The occurrence and distributionof Eupatorieae across the Brazilian mountain rangesseems to be more related to individual species char-acteristics and restrictions than to growth form.

In the Mantiqueira range another analysis using thesame data Almeida 2001; Almeida and Lewinsohn,unpublished produced two relevant results. First,both in Campos do Jordão and in Itatiaia the maxi-mum species richness was detected above the treeline,showing that for Eupatorieae the campos de altitudehave a richer flora, and a larger proportion of endem-ics, than the lower highland forest zone. The secondpoint is that more widespread Eupatorieae have awider altitudinal range, which means that species oc-curring in various localities are generalistic enough tooccur in a wide variety of altitudes. Campos do

Jordão and Itatiaia are not geographically closestamong the Mantiqueira localities, but the flora is mostsimilar between these two localities probably becauseof the similarity of suitable climatic and habitat con-ditions.

Processes that structure communities within a par-ticular locality may differ from those acting across anentire mountain range, and again from those actingamong mountain ranges. As already noted by severalauthors e.g., Ricklefs and Schluter 1993; Brown1995; Maurer 1999; Nekola and White 1999, thewider the study scale the more the chance that bio-geographical and historical factors will be influenc-ing the observed results. The high turnover rate of Eupatorieae in Brazilian mountain ranges, with fewspecies present in more than a region, suggests thatdifferent processes are occurring in the more localscale, within mountain ranges, and in the regionalmesoscale, among mountain ranges.

174

8/18/2019 Almeida 2004

13/19

Acknowledgements

Most of this work was conducted by Adriana MAlmeida in partial fulfilment of the PhD degree in theUnicamp Graduate Program in Ecology. We thanktwo anonymous reviewers for valuable comments on

the manuscript. We are very grateful to Drs. HaroldRobinson, Nelson I. Matzenbacher, Roberto L.Esteves and João Semir for help with plant identifi-cations. Antônio Carlos Macedo, Umberto Kubota,Érika P. Anseloni, Soraia de A. Ferreira, Antônio M.Rosa, Bruno D. Buys, Vinícius N. Mota, Flavia Q.Batista, Marcelo A. B. Lopes, Adalberto J. Santos and

José C. Silva for their invaluable help in field trips.Rafael L. G. Raimundo produced the locality map. C.Jacobi, A.V.L. Freitas, L.B. Klaczko read and com-mented an earlier version of the manuscript. IEFMGand IBAMA permitted collections at Ibitipoca andItatiaia parks, respectively. The staff of the Parque

Estadual de Ibitipoca and Parque Nacional de Itatiaiahelped with access and facilities within the parks.This project was supported by Fapesp grant 98/ 05085-2, as part of the Biota/FAPESP program, andby CAPES and Fapesp with graduate and post-doc-toral scholarships for A. Almeida.

175

8/18/2019 Almeida 2004

14/19

A p p e n d i x

T a b l e A 1

. N u m b e r o f c o l l e c t i o n s o f e a c h p l a n t s p e c i e s p e r l o c a l i t y a n d p e r m o u n t a i n r a n g e 1 . T a b l e 1 l i s t s l o c a l i t y c o d e s a n d c o o r d i n a t e s . A b b r e v i a t i o n s f o r s u b t r i b e s S T

a r e : A d –

A d e n o s t e m m a t i n a e ; A 1 –

A l o m i i n a e ; C r – C r i t o n i i n a e ; G y – G y p t i d i n a e ; M i – M i k a n i i n a e .

S T

S p e c i e s

M a n t i q u e i r a

E s p i n h a ç o

S o u t h

I B

V M

I T

P Q

C J

G M

O G

S C A B

D I A

S C I P

O B

M C

L G

C E R

B J

C A M B

A d e n o s t e m m a t i -

n a e

A d e n o

s t e m m a b r a s i l i a n u m C a s s .

1 2

A d e n o

s t e m m a v e r b e s i n a L . K u n t z e

1

A g e r a t i n a e

A c r i t o p a p p u s i n d e t .

1

A c r i t o p a p p u s i r w i n i i K & R

5

A c r i t o p a p p u s l o n g i f o l i u s G a r d n e r

K & R

2

2

A g e r a t u m c o n y z o i d e s L .

6

2 5

1 1

5

A g e r a t u m f a s t i g i a t u m G a r d n e r

K & R

3 0

2 4

2

1 1

7

7

2

8

3

1

S t e v i a

c o m m i x t a B . R o b i n s o n

1

S t e v i a

c r e n u l a t a B a k e r

1

1

1

S t e v i a

o p h r y o p h y l l a B . R o b i n s o n

1

A l o m i i n a e

P s e u d o b r i c k e l l i a b r a s i l i e n s i s S p r e n g e l

K & R

2

1

3

1

A y a p a n i n a e

A y a p a n a a m y g d a l i n a L a m . K & R

2

1

4

3

A y p a n a a f f . a m y g d a l i n a

1

H e t e r o

c o n d y l u s a l a t u s V e l l . C o n c .

K & R

1

4

1

4

3

H e t e r o

c o n d y l u s a m p h i d i c t y u s D C . K & R

5

2

2

4

H e t e r o

c o n d y l u s g r a n d i s S c h u l t z - B i p . e x B a k e r K & R

2

1

H e t e r o

c o n d y l u s j a r a g u e n s i s B . R o b i n s o n

K & R

2

9

K o a n o

p h y l l o n a d a m a n t i u m G a r d n e r

K & R

1

1

C r i t o n i i n a e

K o a n o

p h y l l o n t h y s a n o l e p i s B . R o b i n s o n

K & R

1

3

2

1

2

O p h y o

s p o r u s f r e y r e y s i i T h u n b . a n d D a l l m . B a k e r

1

D i s y n a p h i n a e

C a m p o v a s s o u r i a c r u c i a t a V e l l . C o n c .

K & R

1

3

4

1

4

1

2

2

1

5

G r a z i e l i a a f f . s e r r a t a

3

G r a z i e l i a g a u d i c h a u d e a n a D C . K & R

3

5

4

G r a z i e l i a g a u d i c h a u d e a n a v a r .

b i p i n n a t a D C . K &

R

2

G r a z i e l i a i n t e r m e d i a D C . K & R

8

1 5

3

1 6

2

3

G r a z i e l i a s e r r a t a D C . K & R

1

3

2

R a u l i n

o r e i t z i a c r e n u l a t a S p r e n g e l

K & R

1

1

R a u l i n

o r e i t z i a t r e m u l a H o o k a n d A r n . K & R

1

1

S y m p h

y o p a p p u s a f f

. r e t i c u l a t u s

1

S y m p h

y o p a p p u s a n g u s t i f o l i u s C a b r e r a

1

S y m p h

y o p a p p u s c o m p r e s s u s G a r d n e r

B . R o b i n s o n

4

1 1

4

S y m p h

y o p a p p u s c u n e a t u s D C . S c h u l t z - B i p . e x B a

k e r

3

2

3

2

1

S y m p h

y o p a p p u s d e c u s s a t u s T u r c z .

5

1

1

1

1

S y m p h

y o p a p p u s i n d e t .

1

S y m p h

y o p a p p u s i t a t i a y e n s i s H i e r o n . K & R

4

1 0

2

S y m p h

y o p a p p u s r e t i c u l a t u s B a k e r

1

176

8/18/2019 Almeida 2004

15/19

T a b l e A 1 . C o n t i n u e d .

S T

S p e c i e s


E s p i n h a ç o

S o u t h

I B

V M

I T

P Q

C J

G M

O G

S C A B

D I A

S C I P

O B

M C

L G

C E R

B J

C A M B

S y m p h

y o p a p p u s s p . 0 1

1

S y m p h

y o p a p p u s v i s c o s u s S c h u l t z - B i p .

1

2

E u p a t o r i i n a e

A u s t r o

e u p a t o r i u m i n u l a e f o l i u m H . B . K . K & R

1

1

A u s t r o

e u p a t o r i u m l a e t e - v i r e n s H o o k a n d A r n . K &

R

1

A u s t r o

e u p a t o r i u m n e g l e c t u m B . R o b i n s o n

K & R

1

2

2

A u s t r o

e u p a t o r i u m p a u l i n u m D C . K & R

1

7

3

A u s t r o

e u p a t o r i u m p i c t u r a t u m M a l m e K & R

1

1

1

1

A u s t r o

e u p a t o r i u m s i l p h i i f o l i u m M a r t i u s K & R

8

1 4

8

1 1

5

H a t s c h b a c h i e l l a t w e e d i a n a H o o k a n d A r n . K & R

1

2

S t o m a t h a n t e s p o l y c e p h a l u s S c h u l t z - B i p . e x B . R o b

i n s o n

H .

R o b i n

s o n

1

G y p t i d i n a e

B a r r o s o a b e t o n i c a e f o r m i s D C . K & R

1

5

1

2

1

5

7

B a r r o s o a c a n d o l l e a n a H o o k a n d A r n . K & R

4

C a m p u l o c l i n i u m c a m p u l o c l i n o i d e s B a k e r

K & R

2

C a m p u l o c l i n i u m c h l o r o l e p i s B a k e r

K & R

1

C a m p u l o c l i n i u m i n d e t .

1

C a m p u l o c l i n i u m m a c r o c e p h a l u m L e e s .

D C

2

5

1

C a m p u l o c l i n i u m m e g a c e p h a l u m M a r t i u s e x B a k e r

K & R

2

C a m p u l o c l i n i u m p a r v u l u m G l a z .

K & R

2

C a m p u l o c l i n i u m p u r p u r a s c e n s S c h u l t z - B i p . e x B a k e r

K & R

3

6

1

5

3

G y p t i s

c r a s s i p e s H i e r o n . K & R

1

T r i c h o

g o n i a h i r t i fl o r a D C . S c h u l t z - B i p . e x B a k e r

1 0

2

6

4

2

T r i c h o

g o n i a s a l v i a e f o l i a G a r d n e r

3

3

T r i c h o

g o n i a v i l l o s a S c h u l t z - B i p . e x B a k e r

9

3

5

1 2

3

3

T r i c h o

g o n i o p s i s a d e n a n t h a D C . K & R

2

V i t t e t i a o r b i c u l a t a D C . K & R

2

M i k a n i i n a e

M i k a n

i a a f f . l a s i a n d r a e

1

M i k a n

i a a f f . l e i o l a e n a

1

1

1

1

M i k a n

i a a f f . s e s s i l i f o l i a

1

M i k a n

i a b r a d e i B . R o b i n s o n

1

M i k a n

i a b u r c h e l l i i B a k e r

1

M i k a n

i a c a m p a n u l a t a G a r d n e r

5

1

1

M i k a n

i a c i p o e n s i s G B a r r o s o

3

M i k a n

i a c i t r i o d o r a n . s p e c i e s

1

M i k a n

i a c o r d i f o l i a L . f . W i l l d

3

1

M i k a n

i a d e c u m b e n s M a l m e

3

M i k a n

i a e l l i p t i c a D C .

8

1

1

M i k a n

i a e r i o s t r e p t a B . R o b i n s o n

2

M i k a n

i a g l a z i o v i i B a k e r

1

M i k a n

i a h e m i s p h a e r i c a S c h u l t z - B i p . e x B a k e r

1

M i k a n

i a i n v o l u c r a t a H o o k a n d A r n .

1

177

8/18/2019 Almeida 2004

16/19


S T

S p e c i e s


E s p i n h a ç o

S o u t h

I B

V M

I T

P Q

C J

G M

O G

S C A B

D I A

S C I P

O B

M C

L G

C E R

B J

C A M B

M i k a n

i a l a s i a n d r a e D C .

3

1

4

8

M i k a n

i a l e i o l a e n a D C .

5

3

2

2

M i k a n

i a l i n d b e r g i i B a k e r

2

3

M i k a n

i a l i n e a r i f o l i a D C .

1

M i k a n

i a m i c r a n t h a H . B . K .

1

1 3

3

4

2

1

1

1

M i k a n

i a m i c r o c e p h a l a D C .

1

5

M i k a n

i a m i c r o d o n t a D C .

1

1

M i k a n

i a m i c r o p h y l l a S c h u l t z - B i p . e x B a k e r

1

M i k a n

i a n e u r o c a u l a D C .

1

M i k a n

i a n i t i d u l a B a k e r

1

M i k a n

i a o b l o n g i f o l i a D C .

1

M i k a n

i a o f f ı c i n a l i s M a r t i u s

4

7

1

M i k a n

i a o r l e a n s e n s i s H i e r o n .

1

M i k a n

i a p a r a n e n s i s D u s e n

1

M i k a n

i a p h a e o c l a d o s M a r t i u s e x . B a k e r

2

1

M i k a n

i a p i n n a t i l o b a D C .

1

M i k a n

i a p s e u d o g r a c i l i s K & R

1

M i k a n

i a p u r p u r a s c e n s S c h u l t z - B i p e x . B a k e r

1

M i k a n

i a r a m o s i s s i m a G a r d n e r

1

1

1

M i k a n

i a r e t i f o l i a S c h u l t z - B i p e x . B a k e r

1

5

M i

M i k a n

i a s a l v i a e f o l i a G a r d n e r

2

2

M i k a n

i a s e s s i l i f o l i a D C .

2

1

2

1

M i k a n

i a v i t i f o l i a D C .

2

M i k a n

i a s p . 1

1

1

M i k a n

i a s p . 1 2

1

M i k a n

i a s p . 2

1

M i k a n

i a s p . 3 1

1

M i k a n

i a s p . 3 8

1

P r a x e l i n a e

C h r o m

o l a e n a a f f . a s c e n d e n s

1

C h r o m

o l a e n a a f f . c o n g e s t a

2

1

2

2

1

C h r o m

o l a e n a a f f . h i r s u t a

1

2

1

C h r o m

o l a e n a a f f . p e d u n c u l o s a

1

C h r o m

o l a e n a a f f . s q u a l i d a

1

C h r o m

o l a e n a a f f . s q u a l i d a s p . 1

2

1

C h r o m

o l a e n a a s c e n d e n s S c h u l t z - B i p . e x B a k e r K & R

2

7

2

3

C h r o m

o l a e n a b a r b a c e n s i s H i e r o n . K & R

1

1

C h r o m

o l a e n a c h a s e a e B . R o b i n s o n

K & R

1

3

3

2

1

C h r o m

o l a e n a c o n g e s t a A r n . a n d H o o k

K & R

1 1

4

6

2

C h r o m

o l a e n a c y l i n d r o c e p h a l a S c h u l t z - B i p . e x B a k

e r

K & R

2

3

6

2

C h r o m

o l a e n a d e c u m b e n s G a r d n e r K & R

1

2

C h r o m

o l a e n a h o o k e r i a n a G r i s e b . K & R

3

4

178

8/18/2019 Almeida 2004

17/19


S T

S p e c i e s


E s p i n h a ç o

S o u t h

I B

V M

I T

P Q

C J

G M

O G

S C A B

D I A

S C I P

O B

M C

L G

C E R

B J

C A M B

C h r o m

o l a e n a h o r m i n o i d e s D C . K & R

1

4

4

2

C h r o m

o l a e n a l a e v i g a t a L a m . K & R

7

5

3

7

9

1

1

1

1

1

1

1

2

C h r o m

o l a e n a l i n e a r i s M a l m e K & R

1

C h r o m

o l a e n a m a t t o g r o s s e n s i s H i e r o n . K & R

1

2

C h r o m

o l a e n a m a x i m i l i a n i i S c h r a d e r e x D C . K & R

4

3

4

2

C h r o m

o l a e n a m i n a s g e r a e s e n s i s H i e r o n . �

almeida 2004

Documents